The Cambridge History of Later Medieval Philosophy

The Cambridge History of
Later Medieval Philosophy

FROM THE R E D I S C O V E R Y OF
ARISTOTLE TO THE D I S I N T E G R A T I O N
OF SCHOLASTICISM
ι100-1600
EDITORS
NORMAN KRETZMANN
ANTHONY KENNY JAN P I N B O R G
ASSOCIATE EDITOR
ELEONORE STUMP
CAMBRIDGE
UNIVERSITY PRESS
Published by the Press Syndicate of the University of Cambridge
The PiK Building, Trumpington Street, Cambridge CB2 IRP
40 West 20th Street, New York, NY 10011-4211, USA
10 Stamford Road, Oakleigh, Melbourne 3166, Australia
© Cambridge University Press 1982
First published 1982

Reprinted 1984
First paperback edition 1988
Reprinted 1989,1992,1996,1997
Printed in Great Britain at the University Press, Cambridge
Library of Congress catalogue card number: 81-10086
British Library Cataloguing in Publication Data

The Cambridge history of later medieval philosophy.
1. Philosophy, Medieval - Historiography
I. Kretzmann, Norman II. Kenny, Anthony
III. Pinborg,Jan
189B721
ISBN 0 52122605 8 hardback

ISBN 0 52136933 9 paperback
CONTENTS
Contributors ix
Preface xiii
Introduction ι
NORMAN KRETZMANN
I Medieval philosophical literature 9
1 ANTHONY KENNY, JAN PINBORG II
II A r i s t o t l e i n t h e m i d d l e a g e s 43
2 Aristoteles latinus 45
BERNARD G. DOD
3 T h e medieval interpretation o f Aristotle 80
C. H. LOHR
ΙΠ T h e o l d l o g i c 99
4 A n c i e n t scholastic l o g i c as t h e s o u r c e o f m e d i e v a l scholastic l o g i c 101

STEN EBBESEN
5 Predicables a n d categories 128
D. P. HENRY
6 Abelard a n d t h e c u l m i n a t i o n o f t h e old l o g i c 143
MARTIN M. TWEEDALE
IV L o g i c in the h i g h m i d d l e ages: s e m a n t i c t h e o r y 159

7 T h e origins o f the theory o f the properties o f terms 161
L. M. DE RIJK
8 T h e O x f o r d a n d Paris t r a d i t i o n s i n l o g i c 174
ALAIN DE LIBERA
9 T h e semantics o f t e r m s 188
PAUL VINCENT SPADE
10 T h e semantics o f p r o p o s i t i o n s 197
GABRIEL NUCHELMANS
11 S y n c a t e g o r e m a t a , e x p o n i b i l i a , s o p h i s m a t a 211
NORMAN KRET2MANN
vi Contents
12 Insolubilia 246
PAUL VINCENT SPADE
13 Speculative grammar 254
JAN PINBORG
V L o g i c in the high middle ages: propositions and modalities 271

14 Topics: their development and absorption into consequences 273
ELEONORE STUMP
15 Consequences 300
IVAN BOH
16 Obligations
A. From the beginning to the early fourteenth century 315
ELEONORE STUMP
B. Developments in the fourteenth century 335
PAUL VINCENT SPADE
17 Modal logic 3+ 2
SIMO KNUUTTILA
18 Future contingents 3S8
CALVIN NORMORE
VI Metaphysics and e p i s t e m o l o g y 383

19 Essence and existence 385
JOHN F. WIPPEL
20 Uni versais in the early fourteenth century 411
MARILYN MCCORD ADAMS
21 Faith, ideas, illumination, and experience 440
JOSEPH OWENS, C.SS.R.
22 Intuitive and abstractive cognition 460
JOHN F. BOLER
23 Intentions and impositions 479
CHRISTIAN KNUDSEN
24 Demonstrative science 496
EILEEN SERENE
VII Natural philosophy 519

25 The interpretation of Aristotle's Physics and the science of motion 521
JAMES A, WEISHEIPL, O.P.
Contents vu
26 The effect of the condemnation of 1277 537
EDWARD GRANT
27 The Oxford calculators 54°

EDITH DUDLEY SYLLA
28 Infinity and continuity 564

JOHN E. M U R D O C H
VIII Philosophy o f m i n d and action 593

29 The potential and the agent intellect 595
Z. KUKSEWICZ
30 Sense, intellect, and imagination in Albert, Thomas, and Siger 602

EDWARD P. M A H O N E Y
31 Criticisms of Aristotelian psychology and the Augustinian-

Aristotelian synthesis 623
Z. KUKSEWICZ
32 Free will and free choice 629
J. B. KOROLEC
33 Thomas Aquinas on human action 642
ALAN DONAGAN
IX Ethics 655
34 The reception and interpretation of Aristotle's Ethics 657
GEORG WIELAND
35 Happiness: the perfection of man 673
GEORG W I E L A N D
36 Conscience 687
T I M O T H Y C. POTTS
37 Natural morality and natural law 705

D . E. LUSCOMBE
X Politics 721
38 The reception and interpretation of Aristotle's Politics 723
JEAN D U N B A B I N
39 Rights, natural rights, and the philosophy of law 738

A. S. M C G R A D E
40 The state of nature and the origin of the state 757

D . E. LUSCOMBE
Vlll Contents
41 The just war 771

JONATHAN BARNES
XI The defeat, neglect, and revival of scholasticism 785

42 The eclipse of medieval logic 787
E. J. ASHWORTH
43 Humanism and the teaching of logic 797
LISA JARDINE
44 Changes in the approach to language 808
W. KEITH PERCIVAL
45 Scholasticism in the seventeenth century 818
JOHN A. TRENTMAN
46 Neoscholasticism 838
P. J. FITZPATRICK
Biographies »53
Bibliography 893
Index nominum 979
Index rerum 995
CONTRIBUTORS
PROFESSOR M A R I L Y N ADAMS
Department of Philosophy, University of California at Los Angeles.
PROFESSOR E. J. A S H W O R T H
Department of Philosophy, University of Waterloo.
DR J O N A T H A N B A R N E S
Balliol College, University of Oxford.
PROFESSOR IVAN BOH
Department of Philosophy, Ohio State University.
PROFESSOR J O H N BOLER
Department of Philosophy, University of Washington.
D R B E R N A R D G. D O D
189 Morrell Avenue, Oxford.
PROFESSOR ALAN D O N A G A N
Department of Philosophy, University of Chicago.
MRS JEAN D U N B A B I N
St Anne's College, University of Oxford.
LEKTOR STEN EBBESEN
Institut for Middelalderfilologi, University of Copenhagen.
D R P. J. FITZPATRICK
Department of Philosophy, University of Durham.
PROFESSOR E D W A R D G R A N T
Department of the History and Philosophy of Science, Indiana University.
D R D . P. H E N R Y
Department of Philosophy, University of Manchester.
DR LISA J A R D I N E
Jesus College, University of Cambridge.
DR A N T H O N Y K E N N Y
Balliol College, University of Oxford.
DR CHRISTIAN K N U D S E N
Ritterstrasse 3, D-4660 Gelsenkirchen-Buer, West Germany.
DR SIMO KNUUTT1LA
Institute ofPhilosophy, University of Helsinki.
χ Contributors
P R O F E S S O R J. Β . K O R O L E C
Pokkij Akademii Nauk, Instytut Filozofii i Socjologii, Patac Staszica, Nowy Swiat
72, oo-jjo Warszawa.
PROFESSOR N O R M A N K R E T Z M A N N
Sage School of Philosophy, Cornell University.
P R O F E S S O R ZDZISLAW KUKSEWICZ
Polskij Akademii Nauk, Instytut Filozofii i Socjologii, Paiac Staszica, Nowy Swiat
72, oo-jjo Warszawa.
D R ALAIN DE LIBERA
École Pratique des Hautes Études, Section des Sciences Religieuses, Sorbonne, 45-47
rue des Ecoles, Paris 5'·
D R C H A R L E S H . L O H R , S. J.
Raimundus-Lullus-Institut, Albert-Ludwigs-Universität, Freiburg i. Br.
P R O F E S S O R D . E. L U S C O M B E
Department of History, University of Sheffield.
P R O F E S S O R E D W A R D P. Μ Λ Η Ο Ν Ε Υ
Department of Philosophy, Duke University.
P R O F E S S O R A. S. M C G R A D E
Department of Philosophy, University of Connecticut.
PROFESSOR JOHN M U R D O C H
Department of the History of Science, Harvard University.
P R O F E S S O R CALVIN N O R M O R E
Department of Philosophy, Princeton University.
P R O F E S S O R GABRIEL N U C H E L M A N S
Filosofisch Instituut, University of Leiden.
FR JOSEPH O W E N S , C. S S . R.
Pontifical Institute of Mediaeval Studies, University of Toronto.
P R O F E S S O R W . KEITH PERCIVAL
Department of Linguistics, University of Kansas.
P R O F E S S O R JAN P I N B O R G
Institut for Middelalderfilologi, University of Copenhagen.
DR τ . c POTTS
Department of Philosophy, University of Leeds.
P R O F E S S O R L. M. DE RIJK
Filosofisch Instituut, University of Leiden.
P R O F E S S O R EILEEN S E R E N E
Department of Philosophy, Yale University.
P R O F E S S O R PAUL V I N C E N T SPADE
Department of Philosophy, Indiana University.
PROFESSOR ELEONORE STUMP
Department of Philosophy and Religion, Virginia Polytechnic Institute and State
University.
Contributors
PROFESSOR EDITH DUDLEY SYLLA

Department of History, North Carolina State University.
PROFESSOR JOHN TRENTMAN
Department of Philosophy, McGill University.
PROFESSOR MARTIN M. TWEEDALE
Department of Philosophy, University of Auckland.
FR JAMES A. WEISHEIPL, O. P.
Pontifical Institute of Mediaeval Studies, University of Toronto.
DR GEORG WIELAND
Philosophisches Seminar, University of Bonn.
PROFESSOR JOHN F. WIPPEL
School ofPhilosophy, Catholic University of America.
PREFACE
The idea of this History originated in the Cambridge University Press,

and the first discussions that led directly to its publication took place in
1975. Naturally it was conceived of as one of the Cambridge Histories of
philosophy. Its place in that series is described in the Introduction, where
its principles of organisation and its special purposes are also discussed.
Editorial decisions regarding the contents and the contributors were
made in 1976, and drafts of most of the contributions were received by
the end of 1978. During a two-week conference of the editorial staff in
the summer of 1979 the book was given very nearly its final form, and
the four editors put the finishing touches on the typescript during the
following summer. Proofreading and indexing were done during the
summer of 1981. The editorial work of the three last summers was
alleviated by the hospitality of Cornell University's Society for the
Humanities. The Directors of the Society during those years - first
Professor Michael Kämmen and then Professor Eric Blackall - cheerfully
provided excellent offices and work-space for the staff on all those
occasions. T w o Cornell undergraduates served as assistants to the staff:
Mary Tedeschi helped a great deal with the Bibliography, and William
Haines worked with admirable efficiency and intelligence on the Index
Nominum.
Editorial thanks can scarcely be offered to the contributors for the
substance of their chapters; they are, after all, the authors of this book.
But the editors would be remiss if they did not express their gratitude to
the contributors for trying to stay within irksome limits and for acqui-
escing in the editing necessitated by their occasionally failing to do so.
The resultant heavily edited typescript of an already complex volume
posed a special problem for the typesetters, and we are pleased to offer
our congratulations to the printers (the Asco Trade Typesetting company
in Hong Kong) to whom the Press entrusted this project; their remarkably
accurate work made the considerable task of proofreading and indexing
less burdensome than we had expected it to be.
xiv Preface
Long delays are evidently inevitable in the production of a work of

this size, and the contributors, almost all of whom submitted their drafts
on schedule in 1978, cannot be blamed if the Bibliography lacks
references to some recent important work within the fields of their
chapters. In the final phases of the book's preparation the editors did a
little to bring the Bibliography up to date, but our combined expertise is
not nearly broad enough to cover the range of all these chapters.
In editing the drafts we tried to reduce redundancy and inconsistency
among the forty-seven chapters, but a history organised topically rather
than chronologically and written by many expert authors is bound to
contain differing opinions and even discrepancies. We hope those that
remain prove to be stimulating.
The reader who wants to make effective use of this book should begin
by reading the Introduction and the introductory notes attached to the
Biographies, the Bibliography, and the Indices.
Norman Kretzmann
Ithaca, New York
August iç8i
INTRODUCTION
The Cambridge History of Later Medieval Philosophy finds its natural place
after The Cambridge History of Later Creek and Early Medieval Philosophy in
the sequence that begins with Guthrie's History of Greek Philosophy. The
sequence is not altogether smooth, however. At the beginning of The
Cambridge History of Later Greek and Early Medieval Philosophy its editor,
A. H. Armstrong, observes that although the volume 'was originally
planned in connexion with W. K. C. Guthrie's History of Greek Philosophy,
... [it] has developed on rather different lines, and is not exactly a continu-
ation ofthat work' (p. xii). Similarly, although The Cambridge History of
Later Medieval Philosophy was conceived of as the sequel to The Cambridge
History of Later Greek and Early Medieval Philosophy, the relationship be-
tween the two is not so simple as their titles suggest; in fact, the fit between
this volume and the Armstrong volume is less exact than that between the
Armstrong volume and Professor Guthrie's plan. Many reviewers noted
that the Armstrong volume seems misleadingly titled since it is really a
study of only the Platonist tradition in later Greek and early medieval
philosophy; but in concentrating in that way it does indeed complement
Professor Guthrie's plan, which includes the Stoics and Epicureans as well
as Aristotle while leaving out the Neoplatonists. O n the other hand, The
Cambridge History ofLater Medieval Philosophy cannot be put forward as the
full realisation of Professor Armstrong's expressed hope 'that the philos-
ophy of the thirteenth century and the later Middle Ages in the West, with
later Jewish, Moslem, and Byzantine developments, will some day be dealt
with in another Cambridge volume' (ibid.). We have of course undertaken
to deal with the philosophy of the thirteenth century and the later Middle
Ages in the West, but we have made no attempt to deal with later Jewish,
Moslem, and Byzantine developments.
In deciding to restrict our attention to the Latin Christian West, we were
motivated by two considerations. In the first place, we could scarcely hope
to do justice to even our chosen material in a single volume of this size; if
we had undertaken to deal with Arabic, Jewish, and Byzantine philosophy
AO
2 Introduction
as well, we surely could not have dealt adequately with later medieval
philosophy. And, in the second place, scholarship in those areas has not
kept pace with research on medieval Christian philosophy. When a scholar
with the authority of Richard Walzer acknowledges (on p. 643 of the
Armstrong volume) that 'It appears premature, at the present time, to
embark on a history of Islamic philosophy in the Middle Ages' because
'Too many of the basic facts are still unknown', no one else is likely to be
prepared, even twelve years afterwards, to undertake the task; and the cases
of medieval Jewish and Byzantine philosophy seem much the same. O f
course, Arabs, Jews, and Byzantine Greeks are among the philosophers
mentioned in this volume, but they figure in it only as contributors to the
development of Latin philosophy during the Middle Ages.
The Cambridge History of Later Greek and Early Medieval Philosophy is
described as covering the period 'from the fourth century B.c. to the
beginning of the twelfth century A.D., from the Old Academy to St
Anselm' (p. xii); but it encompasses those 1,500 years primarily in order to
trace the development of Platonism after Plato. The sense in which that
description is intended leaves ample room, of course, for Professor
Guthrie's volumes on Plato and Aristotle, on the Stoics and Epicureans.
Similarly, the fact that our predecessor volume reaches as far forward as the
beginning of the twelfth century is explained by the facts that the philos-
ophy of St Anselm may be thought of as the highwater mark of medieval
Platonism and that Anselm died in 1109. Our volume does indeed con-
centrate on philosophy after Anselm, beginning with Abelard, but because
it is part of our aim to present the medieval Aristotelian tradition and the
scholastic innovations that developed in that tradition, we must reach back
to consider many philosophers older than Anselm who were under-
standably left out of account in the Armstrong volume.
Like several other Cambridge Histories but unlike most histories of
philosophy, this volume is the work of many hands; forty-one scholars
from ten different countries contributed to it. We subdivided the material
and assigned the subdivisions to individual contributors with the intention
of providing a more faithful impression of the state of current research than
could have been provided by a smaller number of contributors to w h o m
larger areas had been assigned. Even with such a strategy we have naturally
had to emphasise some subjects at the expense of others that are equally
important, but we tried to make those difficult decisions in such a way that
our emphasis would fall on material that had been neglected in the
established literature on medieval philosophy and on material regarding
Introduction 3
which recent research had been making most progress. Thus the contribu-
tors have devoted relatively little attention to theological issues, even to the
philosophically outstanding medieval achievement in rational (or natural)
theology, for that side of medieval thought has not been neglected. And
because the areas of concentration in contemporary philosophical scholar-
ship on medieval thought naturally reflect the emphases in contemporary
philosophy, our editorial strategy has led to a concentration on those parts
of later medieval philosophy that are most readily recognisable as philo-
sophical to a student of twentieth-century philosophy.
By combining the highest standards of medieval scholarship with a
respect for the insights and interests of contemporary philosophers, par-
ticularly those working in the analytic tradition, we hope to have presented
medieval philosophy in a way that will help to end the era during which it
has been studied in a philosophical ghetto, with many of the major students
of medieval philosophy unfamiliar or unsympathetic with twentieth-
century philosophical developments, and with most contemporary work
in philosophy carried out in total ignorance of the achievements of the
medievals on the same topics. It is one of our aims to help make the activity
of contemporary philosophy intellectually continuous with medieval
philosophy to the extent to which it already is so with ancient philosophy.
Such a relationship has clearly benefited both philosophical scholarship on
ancient philosophy and contemporary work in philosophy, and we hope
to foster a similar mutually beneficial relationship between medieval philos-
ophy and contemporary philosophy.
The standard approach to the history of philosophy is, of course, by way
of the chronological study of the doctrines of individual philosophers. That
approach is not well-suited to the history of medieval philosophy, in which
the identity of individuals is sometimes uncertain, the attribution of doc-
trines or works to individual philosophers is often disputable and some-
times impossible, and even the chronological succession of men or of
works is often conjectural. We have organised our History around philo-
sophical topics or disciplines rather than around philosophers, but not only
because the standard approach is not well-suited to our period. Our
principal aims in this volume are, we believe, better served by the topical
approach than they would be by the standard approach. (We think of the
biographical sketches supplied at the end of the volume as an important
supplement to our topical approach.) In order to help the reader to discern
the plan of this History, which is to a large extent not organised histori-
cally, we provide the following synopsis of the contributions.
4 Introduction
The forty-six chapters that make up the text of this volume are arranged
in eleven parts. The first and shortest of those parts is the work of two
members of the editorial staff and is designed to introduce the reader to
some of the distinctively medieval forms of philosophical literature. Such
an introduction seems called for not only because most twentieth-century
philosophical readers are likely to be unfamiliar with the presentation of
philosophy in the form of quaestiones or sophismata, for instance, but also
because the literary forms of scholasticism are more influential on the
character of the philosophy presented or developed in those forms than are
the literary forms of any other period in the history of philosophy (with the
possible exception of Greek philosophy before Aristotle).
In the two chapters of Part II Bernard Dod and Charles Lohr provide
accounts of the transmission of Aristotle's works to the Latin Middle Ages
and of the changes effected in the form and content of thought as a result of
that legacy from antiquity. None of the succeeding chapters of the book
can be properly understood except against the historical background
delineated in Part II.
The fact that Parts III, IV, and V all contain the word 'logic' in their titles
may suggest an imbalance in the organisation of this History, and the fact
that three members of the editorial staff have contributed chapters to these
Parts might even suggest that editorial predilections account for the imbal-
ance. What medieval philosophers thought of as logic does indeed figure
very prominently in this book; several chapters in Parts VI, VII, and XI are
also principally concerned with aspects of medieval logic. But any history
of medieval philosophy which, like ours, leaves theology out of account is
bound to devote more space to logic than to any other branch of phi-
losophy. The imbalance, if there is one, is embedded in the nature of
medieval scholasticism, in which the unusual importance of logic is partly a
consequence of the fact that during the Middle Ages logic was conceived of
more broadly than in any other period of the history of philosophy. A
great deal of work that will strike a twentieth-century philosophical reader
as belonging to metaphysics, philosophy of language, linguistics, natural
philosophy, or philosophy of science was carried on during the Middle
Ages by men who thought of themselves as working in logic. Moreover,
the achievements of medieval logicians are historically more distinctive
and philosophically more valuable than anything else in medieval thought,
with the possible exception of rational theology; when Renaissance hu-
manists waged their successful battle against medieval scholasticism, it was,
understandably, scholastic logic against which they directed their fiercest
Introduction 5
attacks. After Christianity and Aristotelianism, the most important in-

fluence on the character of the philosophy of the Middle Ages is the
medieval conception of logic.
The dominance of logic is to some extent the result of an historical
accident: the fact that until the middle of the twelfth century the only
ancient philosophy directly accessible to the Latin medievals was contained
in two of Aristotle's works on logic, the Categories and De interpretatione.
These very short and very difficult books, along with a handful of as-
sociated treatises stemming from late antiquity, constituted the secular
philosophical library of the early Middle Ages and became known as the
Old Logic by contrast with the New Logic - the rest of Aristotle's
Organon - as it became available during the second half of the twelfth
century. To the extent to which the philosophy of the later Middle Ages is
a development of earlier medieval philosophy it rests on the accomplish-
ments of men who had been working out the implications and ramifi-
cations of the Old Logic, and that essential contribution to later medieval
philosophy is presented by Sten Ebbesen, D. P. Henry, and Martin
Tweedale in the three chapters of Part III.
The development of medieval logic during and after the advent of the
New Logic is explored in Parts IV and V. Several of the twelve chapters of
these Parts will help to show how far beyond Aristotelian logic medieval
logic eventually developed in various directions, but the non-Aristotelian
character of later medieval logic is most striking in its semantic theories,
different aspects of which are presented by L. M. de Rijk, Alain de Libera,
Paul Vincent Spade, Gabriel Nuchelmans, Norman Kretzmann, and Jan
Pinborg in Part IV.
The branches of medieval logic considered in Part V have not yet
received as much scholarly attention as has medieval semantic theory, but,
as the contributions of Eleonore Stump, Ivan Boh, Paul Vincent Spade
Simo Knuuttila, and Calvin Normore help to show, they are likely to
prove at least as rewarding to the further study they deserve. The first three
chapters of Part V are devoted to issues associated with logic in its central
role as theory of inference; the fourth and fifth chapters present medieval
contributions to inquiries that lie on the border between logic and
metaphysics.
Metaphysics and epistemology were very highly developed in later
medieval philosophy, and there are enormous quantities of relevant textual
material. The six chapters of Part VI sort out some of the more rewarding
issues and explore a few of them to considerable depth, but no one is more
6 Introduction
keenly aware than the authors of these chapters that they have had to
restrict themselves to merely alluding to developments that deserve de
tailed discussion. Fortunately, the secondary literature in these fields is
more extensive than in most of the fields dealt with in this History,
although a great deal of it is becoming obsolete as more texts become
available and traditional interpretations are revised in the light of new
evidence and changing philosophical perspectives. The first two chapters,
by John Wippel and Marilyn Adams, are concerned with topics at the core
of the subject-matter of metaphysics. Chapters 21 and 22, by Joseph Owens
and John Boler, deal with epistemological issues that arise in different
guises throughout the history of philosophy even though some of them
appear here in distinctively medieval trappings. In Chapters 23 and 24
Christian Knudsen and Eileen Serene deal with epistemological issues
adjacent to or included within medieval logic - semantic theory in Chapter
23, theory of inference in Chapter 24.
An important part of medieval natural philosophy, too, can be as
similated to medieval logic, as is clearly shown by Edith Sylla and John
Murdoch in Chapters 27 and 28 of Part VII. Aristotle's Physics informed
the developments in later medieval logic that look to us like speculative
physical theory or proto-mathematics, but it served also as an independent
source of developments in natural philosophy, especially those to be found
in the many commentaries on the Physics. In the first chapter of Part VII
James Weisheipl surveys these developments and the role of natural philos
ophy in the medieval university curriculum. The Condemnation of 1277,
often referred to in this History because of its apparent effect on the
character of later medieval thought, is summarised by Edward Grant in the
second chapter of Part VII, especially with regard to its probable influence
on the development of natural philosophy.
Part VIII begins with a full survey of the origins and development of
philosophy of mind in the Middle Ages, carried out in a series of three co
ordinated chapters by Edward Mahoney and Z. Kuksewicz in a way that
will help the reader understand not only medieval but also classical modern
theories of mind. Medieval accounts of the theoretical links between
philosophy of mind and moral philosophy are examined i n j . Β. Korolec's
chapter on freedom of the will and Alan Donagan's chapter on Aquinas'
theory of action.
Parts IX and X , on moral and political theory, are alike in beginning
with chapters, by Georg Wieland and Jean Dunbabin respectively, that
show how the reception and interpretation of Aristotle's treatises on those
subjects shaped their development during the later Middle Ages. The
Introduction 7
remaining chapters in each of these Parts deal with specific ethical or

political issues that were especially important to the medievals. In Part IX,
on ethics, Georg Wieland examines attempts to accommodate the
Aristotelian ideal of happiness within a Christian context, Timothy Potts
lays out the particularly subtle medieval theory of conscience, and, in
Chapter 37, D. E. Luscombe presents material that forms a natural tran-
sition between ethics and politics in his account of the natural foundations
of morality and law. In Part X, on politics, Chapter 39, by A. S. McGrade,
takes up the topics introduced in Part IX, Chapter 37, but in a more
specifically political context. D. E. Luscombe contributes a chapter to Part
X that is associated with his chapter in Part IX, this time pursuing the topic
of the role of nature in the foundations of social and political institutions as
the medievals saw it. Jonathan Barnes' chapter on justifications for war
illuminates medieval applications of Christian principles and theories of
international politics.
Because the humanist attack on medieval scholasticism aimed especially
at overthrowing late medieval logic and most of the linguistic theory and
educational practice associated with it, the first three chapters of Part XI, on
the end of the scholastic period, are in one way or another devoted to issues
of the sort that medieval logicians had concerned themselves with. In
Chapter 42 E. J. Ashworth details the loss or repudiation of medieval
accomplishments in logic, in Chapter 43 Lisa Jardine focuses on the educa-
tional reforms that may have constituted the primary motivation for the hu-
manists' anti-scholasticism, and in Chapter 44 W. Keith Percival describes
the new attitude towards languages and literature that saw them as subjects
in their own right and not merely as instruments. In the last two chapters of
Part XI and of the book John Trentman and P.J. FitzPatrick show us, first,
the survival of scholasticism in the era of classical modern philosophy and,
finally, the revival of scholasticism in the nineteenth century — a revival
without which, as Dr FitzPatrick observes, this History would hardly have
been written, however different its orientation may be from that of
neoscholasticism.
One of the special virtues of a work of philosophical scholarship pro-
duced by many specialists of different sorts is to be found in the treatment
of the same thinkers or closely related topics from different points of view.
No system of cross-referencing could present the connections among these
forty-six chapters adequately without becoming obtrusive; we urge the
reader to refer frequently to the Index Nominum and Index Rerum in
order to take full advantage of this History.
Limitations of space have naturally made it impossible for any of the
8 Introduction
contributing specialists to deal fully with his or her subject matter here, and
so the bibliographical references are important not merely as citations of
evidence but also, and especially, as guides to further study. The references
are presented in the footnotes to the chapters in forms that are brief without
being cryptic, and all such references are filled out in the general Bib-
liography. The Biographies, which are designed to help the reader make
convenient identifications of the more prominent figures in medieval
philosophy's enormous cast of characters, also contain many specialised
bibliographical references that do not appear in the Bibliography.
Ι
MEDIEVAL P H I L O S O P H I C A L
LITERATURE
I
MEDIEVAL PHILOSOPHICAL
LITERATURE*
Schools and universities1

Medieval philosophical literature is closely associated with medieval
schools and universities as well as with the material and psychological
conditions prevailing at these institutions. The economic prosperity of the
eleventh and twelfth centuries, which is manifest in the rise of towns and
the specialisation of labour, also had far-ranging repercussions in the world
of learning. Learning was no longer confined to monasteries and monastic
schools, but a new guild of professional intellectuals was created, which
developed new intellectual aspirations. Such men were no longer satisfied
with the traditional concept of Christian wisdom but wanted to pursue the
whole domain of human learning, and they resolutely set out to recover
and develop the intellectual heritage of antiquity.
Even in the very early Middle Ages, schools clustered around ecclesias
tical centres, especially the chapters of episcopal sees, which throughout the
Middle Ages provided the most important financial support for learning.
Scholars shared the privileges of clergy although they were not required to
take higher orders; in fact their clerical status was the best way of securing a
certain amount of protection and independence against local authorities in
surroundings which for the most part were hostile and brutal.
This description is especially appropriate for the situation in one of the
centres of the twelfth-century intellectual expansion, the central part of
France, between the Loire and the Rhine. Here schools flourished and
decayed within short spans of time, often due to the presence or absence of
one especially gifted teacher; Laon, Rheims, Melun, and Chartres are but a
few examples of such schools. This situation calls for caution in employing
* Anthony Kenny is responsible for the sections 'Disputations of the theologians', 'The origins of
disputation', and 'The later development of disputations'; the other sections are the work of Jan
Pinborg.
ι. This chapter is a very concise survey of complex material. The most recent studies are Verger 1973
and Cobban 1975, where further references can be found. The classic in the field is Rashdall 193Ö.
For twelfth-century schools see Delhaye 1947. Translated source material can be found in
Thorndike 1944.
12 Medieval philosophical literature
labels of the type 'the school of...', which often tend to confuse insti-
tutions and doctrinal trends. 2 Only at Paris was there a continuous flourish-
ing and growth of the schools, and the best teachers from most other
French schools eventually taught at one or more of the schools of Paris.
In northern Italy the situation was somewhat different. For instance,
there were secular schools there from the beginning, and the interest
centred more on legal training than on the liberal arts and theological
studies, thus attracting students who were generally more mature. The law
schools of Italy accordingly faced somewhat different problems, but even
there the impact of often hostile surroundings forced the students to
organise.
During the twelfth and thirteenth centuries the number of students and
masters in France, England, northern Spain and Italy increased continu-
ally. This growth soon led to the demand for more secure privileges and a
better way of organising, especially for means of controlling the granting
of a licence to teach. Scholars began to organise into corporations, follow-
ing normal medieval patterns of organisation. And so the latter decades of
the twelfth century saw the emergence of universities, one of the most per-
manent institutions created by the Middle Ages. The oldest universities,
such as those of Paris, Bologna, and Oxford, developed gradually and only
later received formal recognition of their privileges from the Pope; it is
therefore impossible to give a precise date for their foundation. But soon it
became customary for new universities to receive their privileges from an
international authority (the Pope, or in rare cases the Emperor) and from a
national or local authority.
By the middle of the thirteenth century there were flourishing univer-
sities in Paris, Oxford, and Bologna and smaller ones in, e.g., Toulouse,
Salamanca, and Cambridge. During the fourteenth century the university
trend spread to central and eastern Europe; the first university east of the
Rhine was the university of Prague, established in 1348. Before the end of
the fifteenth century Europe had more than seventy universities of greatly
varying size and importance. This of course does not mean that university
learning was not introduced east of the Rhine till the fourteenth century.
Scholars returning to their homelands from the university centres intro-
duced a higher level of education and, as a result, the teaching and research
in some town-schools of central Europe acquired almost university level.
An especially well-documented case is that of the schools of Erfurt before
2. See also Southern 1970, pp. 61-85.

Medieval philosophical literature 13
the foundation of the university in 1392.3 Moreover, as early as the

thirteenth century, the mendicant orders established studia generalia in most
of their provinces, where selected friars could receive an education in
philosophy and theology. 4
The spread of universities is indicative not only of a quantitative disper-
sion of learning, but also of a change in the nature of universities. In
principle, universities were equal, all of them having received the right to
bestow the ius ubique docendi on their graduates; but in fact some universities
could afford to be more critical than others regarding the acceptance of
teachers. It is significant that the ius ubique docendi granted to masters from
other universities did not normally gain them the right to teach in Paris.
Moreover, new universities were often founded at places where the finan-
cial support was hardly adequate, which tended to make these universities
smaller and more dependent on local authorities. Finally, some universities
came to be no more than professional schools for those who wanted a
career within civil or ecclesiastical administration, and this, too, had an
effect on the recruitment of students. At the end of the Middle Ages
universities were tending to become more and more aristocratic and
plutocratic. 5
Organisation of teaching
Already in the twelfth century schools tended to specialise in certain
studies. Paris and central France, for example, became famous for their
teaching in logic and theology, Bologna for civil and canon law. Still, the
curricula were not rigidly determined but continued to change in accor-
dance with the discovery of new material and the changing interests of the
masters. Furthermore, with the emergence of universities, knowledge
became organised into separable departments, differentiated and inter-
related in a genre of texts called Divisions of Science. 6
The fundamental division, which also affected the organisation of
universities, was the division between faculties. The basic faculty was the
faculty of Arts, which around 1250 and thereafter comprised the study of
grammar and logic (which together with rhetoric made up the Trivium),
and the whole field of Aristotelian philosophy, supplemented to some
extent with a number of more technical mathematical and astronomical
3. Kleineidam 1973; Pinborg 1967a, pp. 139-50; Pinborg 197} and I976d.
4. See Le Scuole Degli Ordini Mendicanti 1978.
5. Gabriel 1977; Ijsewijn & Paquet 1978; Mornet 1978.
6. A typical example of this genre was written by John of Dacia around 1280 (ed. A. Otto, 1955).
H Medieval philosophical literature
disciplines (the Quadrivium). Students in the faculties of Arts were normally

between 15 and 21 years old. Since there was no formal entrance exami-
nation (although of course the fact that all teaching was given in Latin
restricted the number of prospective students radically), everybody who
had the motivation could in principle embark upon an undergraduate
career. But the often extremely difficult problems of financing studies
meant that only a small fraction of the students ever completed their studies
with a degree. In the fourteenth and fifteenth centuries, when statistical
material is available, it can be made out that only 3 0 - 5 0 % of the students
ever attained the lowest degree possible, that of Bachelor of Arts, which
presupposed between one and a half and two and a half years of study, and
less that 15% the final degree of Master of Arts.
All the higher faculties of Law, Medicine, and Theology originally
presupposed a basic education in Arts. In the later Middle Ages, however,
this was required of theology students only, except for the friars, who had
their own centres for preparatory studies (studia particularia) and were never
required to receive a formal university training in arts. The students of
the higher faculties were always more mature men; frequently they had
already achieved some position or means. The number of students matric-
ulated with the higher faculties, was, of course, much lower than the
number of students studying in the Arts faculty; during the fourteenth and
fifteenth centuries the Law faculties increased from about 10% to almost
20% of the total of matriculated students, Theology stayed at about 5%,
and Medicine at even lower numbers (except perhaps at a few universities
specialising in medicine).
It is worth emphasising that the number of law students was much
higher than the number of theology students. Law studies were for the
ambitious who wanted an ecclesiastical or civil career. Theology was
almost a pure abstract science - its study was not required and was
sometimes not even an aid to a further career. 7
The students of these higher faculties were much more privileged than
their colleagues in the faculties of Arts. Most colleges, for example, were
open only to students or masters from the higher faculties. In these
circumstances, the Arts course tended to be a mere preliminary, not only
for theoretical reasons, because one should go on to more useful studies, but
also for practical reasons, because the higher studies were also plainly more
7. Roger Bacon complained about this state of affairs. Compendium studii philosophiae c. 4 (ed.
Brewer p. 418).
profitable. Thus most teachers in the faculties of Arts stayed for a few years
only (two or three years of teaching were required from every person
graduated as a master). Most of them were quite young, and they were
often pursuing more advanced studies while carrying out their teaching.
These general remarks may give some of the background necessary to
justify the following observations:
(a) The greatest quantities of medieval philosophical material as trans-
mitted in the manuscripts are connected with the teaching in the faculties
of Arts.
(b) Much of this work is of mediocre quality, even if the formal level of
knowledge attained and presupposed is often astounding to us. Never-
theless, some work of high quality was done in the faculties of Arts,
especially in grammar, logic, and astronomy.
(c) The most advanced scholarly research in philosophy, however, was
made by students or teachers in the faculty of Theology (especially in the
thirteenth and fourteenth centuries). Students of theology were more
mature and better educated than the Arts students, and they had more
leisure and better opportunities, whereas the 'Artists' were underpaid and
relatively unprivileged. That is why so much of the study of medieval
philosophy is concerned with theological texts. But this historical connec-
tion does not entail that philosophy and theology could not be studied
separately, or that theological goals determined philosophy and made it
unfree and unphilosophical. 8 There are large sections of pure philosophy in
theological texts, often to the extent that theological authorities thought it
necessary to intercede and demand a stricter limitation to theological
problems. 9
(d) Nevertheless, the Arts course, and especially the first two years of
studies in grammar and logic, are of fundamental importance for the
understanding of medieval philosophy. These studies provided the basic
knowledge medieval intellectuals shared, and they were taken for granted
in all other intellectual activity; they also formed the linguistic competence
of medieval intellectuals and established their idiom of highly technical and
precise Latin. 10
8. Van Steenberghen 1979.

9. See the section 'Question-commentaries' below.
10. We need a thorough study of the linguistic aspect of medieval philosophy with its important
consequences for modern European (scientific) languages. A modest beginning has been made in
Hubert 1949.
ιό Medieval philosophical literature
Teaching and books

The fact that the sources available to reconstruct the intellectual life of the
Middle Ages are written records tempts us to forget that university learn
ing was widely determined by oral teaching. That this had to be so is
immediately apparent as soon as we remember the scarcity of books, a
result of the great effort demanded for the production of a manuscript and
its consequent costliness. Although the centralisation of book-production,
beginning at the end of the thirteenth century, made book-prices much
lower and books more available, private libraries remained very small. The
use and the increased production of paper during the fourteenth century
also made books easier to acquire, but the situation did not change radically
until the invention of printed books in the late fifteenth century. ' l College
libraries certainly improved the situation, but frequently only students of
the higher faculties had access to them.
The pre-eminence of oral teaching as distinct from the study of books
was thus an unavoidable condition, especially for the younger students.
This necessity was turned into a virtue by a theory which we often find
expressed in the prolegomena to the courses or the texts studied. One
example will suffice (from Radulphus Brito, ca. 1300):
Ί rightly contend that we learn more by being taught than we find through our
own efforts, for one lesson heard is of more profit than ten lessons read privately.
That is why Pliny says "the living voice affects the intellect much more than the
reading of books". And he gives the following justification for his contention: the
teacher's pronunciation, facial expressions, gestures, and whole behaviour make
the pupil learn more and more effectively, and what you hear from another person
is situated deeper in your mind than what you learn by yourself.' 1 2
New arguments and new opinions, accordingly, did not spread only in
written form. However important the study of manuscripts and the
history of their availability to various medieval scholars is, we can never in
this way catch the whole range of the medieval exchange of ideas. Many
new thoughts spread rapidly to other universities by word of mouth,
11. Fink-Errera i960. Cf. Thorndike 1944, pp. 112—18.

12. Radulphus Brito, Prooemium in Parva mathanatiealia, MS Bruxelles, Β. Royale 3540—47, f. 1':
'Bene dieu quod plura seimus addiscendo ab alio quam per nosmet inveniendo, quia una lectio
audita plus proficit quam si decern studeantur per se. Et ideo dicit Plinius "multo magis enim viva
vox afficit intellectum quam lectio" id est quam inspectio librorum. Et reddit causam huius, quia
pronuntiatio dicentis, vultus, gestus, et habitus, ista faciunt audientem plura apprehendere et
magis firmiter, et altius audita ab alio in animo sedent quam si per se aliquis studeret.'
carried by traveling students and scholars. The exchange of ideas was not
seriously restricted by the scarcity of books.13
This preponderance of the oral aspects within the medieval transmission
of learning helps to elucidate the specific literary forms of medieval
philosophical literature. It also helps to explain the fact that the written
texts that were used for studies are often in a shockingly corrupt state. It is
frequently difficult to understand how such faulty texts could be of any use
to the students. But if we remember the facts associated with an oral
tradition - which imply that the students had a large number of formulas,
quotations, stock arguments, and standard moves stored in their memory-
it becomes much easier to understand. They used the texts not as their only
sources, but rather as abbreviations, reminders of what they had heard.
They used written sources mainly as a source of useful arguments or
distinctions, not as texts to be relied on for reconstructing the thoughts of
others. The written records as we have them are only a limited reflection of
a much richer oral culture.
The relations between oral and written sources changed gradually;
partly because of the increasing availability of books, medieval intellectuals
became more and more dependent on books. This change in the medium
of communication tended to make arguments and polemics much more
complex. Thus in fourteenth-century commentaries on the Sentences we
find long and exact quotations of contemporaries which can only have
been drawn from books. But, again, Sentence-commentaries were the
work of privileged students who could afford books or had easy access to
them. The average Arts teacher and student had to rely on memory and a
few selected course-books, frequently adorned with hastily produced notes
and commentaries.
Textbooks and university courses
I have just been emphasising the oral aspect of medieval philosophical

literature. In another sense, however, medieval teaching was very much
dependent on books. The cornerstone of any discipline was its authorita-
13. Maier 1964-67, II, pp. 317—34 ('Internationale Beziehungen an spätmittelalterlichen Univer-
sitäten'). The importance of personal relations is obvious - ideas spread more easily from master
to students, and between masters and students of the same nation. This might to some degree
account for the rapid acceptance of English ideas in fourteenth-century Germany, since both
English and German masters belonged to 'the English nation' at the University of Paris.
ι8 Medieval philosophical literature
tive course-book(s), the littera. Research and teaching was obliged to

explain the littera, to show its structure and contents, to prove its inner
consistency, and to harmonise any conflicting statements either in the
littera itself or arising from comparison with other authorities. The means
of doing so was a thorough reading of the textbook, explaining its argu
ments and presuppositions and analysing ambiguous expressions (terms or
phrases). Since teaching thus proceeds in an analytical way we can detect a
tendency towards ad-hoc solutions and distinctions which were sometimes
quickly forgotten or dismissed, even if they could have been of further use
in a more systematic approach to the discipline.
The textbooks determined the curriculum of the study. A student's
curriculum consisted of the list of books required for passing a degree,
together with an indication of the time required for the specific parts of the
curriculum. This is not the place to trace the development of the curriculum
in detail. 1 4 But an example from the end of the Middle Ages will reveal the
mechanics of the system:
Greifswald, statutes of 1456:

BACHELOR'S DEGREE
Lectures:
ars vetus, 3 months (minimum)
Analytica Priora et Posteriora, 3 months
Elenchi, 2 months
parva logicalia, 4 months
Labyrinthus, \\ month
Physica, 6 months
De anima, 3 months
Sphaera, 1 y month
Exercises (for this term see below):
, \ 2 J. , simultaneously
parva Icgicalta, y year
nova logica, \ year
Petrus Hispanus + sophismata simultaneously
or Sophistria f year
De anima + parva naturalia j year

simultaneously
Physica 1 year
MASTER S DEGREE
Lectures (without indication of duration):
Topics, De caelo et mundo. De generatione, Metheora, Parva naturalia,
14. See, e.g., Isaac 1953, pp. 61-85; Weisheipl 1964a.

Ethica Nie, Oeconomica, Polilica, Theoria planetarum, Perspectiva,

Arithmetica, Musica, Geometria, Metaphysica.
Exercises:
Physica, Nova logica. De caelo, De generatione, Metheora, Ethica,
Metaphysical
This standard curriculum was supplemented by readings of special texts

and disputations on freely chosen subjects.
In the faculty of Theology the curriculum consisted in principle of four
parts. After (1) eight years of preparatory studies, (2) the student had to act
for two years as a lecturer on the Bible [baccalaureus biblicus) and (3) two
years as a lecturer on dogmatics, using (from the thirteenth century
onwards) Peter Lombard's Sentences as the course-book (as baccalaureus
sententiarum). Following this (4) he was supposed to attend and participate
in disputations for four years. It is especially the two last phases of the study
which gave rise to works of importance for the history of philosophy.
Within this general framework it was possible to put the emphasis on
different parts of the text or on different problems. These special emphases
tended to change with time and place. For that reason the problems and
parts of the course-book treated in an otherwise anonymous text often may
serve as an indication of its date and provenance.
Forms of teaching: Lectures16

The interplay between oral and written teaching is best seen if we briefly
consider the forms of teaching. The first distinction we have to draw is
between lectures (lectiones) and disputations.
Lectures were either ordinary, extraordinary, or cursory. Ordinary
lectures were those delivered on the stipulated course-books and so were
repeated regularly. When the same teacher was to repeat his course he
could either repeat it verbatim or introduce some changes; he could even
choose to deliver lectures by somebody else and modify them. (This is one
cause of the vexed problem of different versions of a set of lectures, a
problem we will return to shortly.) Extraordinary lectures were not
restricted to the course-books but could also be on books not formally
required by the statutes. They were supposed to serve as reviews or
supplements of the ordinary courses. Cursory lectures were normally short
reviews of the main problems connected with a standard text.
15. Kosegarten 1857, pp. 309-10, cf. Piltz 1977, pp. 23ff.
16. Glorieux 1966; Classen i960.
The basic form of a lecture consisted of (a) the reading aloud of a section
of the littera (in written versions of lectures this part is often omitted or just
hinted at by quoting the beginning of the section of the text, the lemma,
succeeded by an 'etc'.); (b) an account of the disposition of the text (divisio
textus) which divides the text into always smaller parts (normally by
dichotomous divisions), until we arrive at the level of single propositions
(in this part of the lecture the text discussed is normally related to the
preceding parts of the littera, so as to keep the train of thought in mind); (c)
the exposition of each part, more or less extended according to the number
of difficulties acknowledged. In this part the text is sometimes merely
paraphrased, with an explanation of difficult words and distinctions among
their various uses, but often the exposition contains a thorough and very
precise interpretation of the text, occasionally summarised in the form of
rules. Necessary background information is supplied, often in the form of
Notabilia (rules or statements introduced by the words 'notandum' or
'nota'). Although such expositions can be quite tedious, they are often
indispensable, since they teach us how medieval authors understood and
read their sources, which is often quite different from what we think
obvious, (d) The final part of the lecture was dedicated to especially
important points, discussed in the form of real or fictitious disputations,
normally introduced by 'dubium est' or 'dubitandum est'. (This entire four-
part procedure could be repeated within the same lecture if there was not
enough material in a single part of the text.) This form of lectures has a long
history and for its main parts goes back to antiquity. 11
Another form of lectures which is specifically medieval was apparently
developed from the last section of the previously discussed form of lectures.
The dubia or disputation part of the lectures seems to have become in-
dependent from the other parts of the course, ' 8 and during the latter part of
the thirteenth century a new type of commentary developed. From about
1260 we find commentaries consisting only of a series ofquaestiones, each of
which has the basic form of a disputation. Presumably this reflects the
development of a new kind of lecture. We do not know to what extent, if
any, such quaestiones were staged as real disputations. Certainly from the
fourteenth century onwards we have testimony that they were only read
17. Peters 1968, pp. 10-17.

18. Already in the twelfth century we hear of independent disputations (see Landgraf 1935; Little &
Pelster 1934, pp. 2gf.; Grabmann 1911, II, pp. 497f, 53Sf. For the Arts courses see Hunt 1949-50,
II, PP· 19, 5 5 - 6 .
aloud by the master. Often he did not even have to compose his own
questions, but was allowed to read questions by renowned authors in
more or less rephrased forms. This was accepted explicitly in Prague (and
forbidden in Paris in 1452), but certainly reflects earlier and common
practice. 19 In the fourteenth and fifteenth centuries lectures of this kind
were termed exercitia, the 'exercises' mentioned in the Greifswald statutes
above.
This development is probably a result of the great interest of medieval
scholars in disputations. Since disputation is an element very characteristic
of medieval teaching, it will be helpful to discuss the nature of disputations
more fully.
Disputations of the theologians
In the thirteenth century the principal duties of a Master in Theology were

to lecture, to preach, and to dispute. The holding of academic disputations
many times in the year, perhaps even weekly, formed an integral part of
the academic curriculum, less frequent, but no less important, than the
giving of lecture courses. The records of these disputations, in the form of
Quaestiones disputatae, constitute a valuable part of the output of many
medieval philosophers and theologians.
A formal disputation in the mid-thirteenth century fell into two parts
occupying separate days. The time and topic of the disputation would be
announced by the professor well in advance; all bachelors in the faculty
were summoned to attend, and other masters and their students were
invited as well. O n the first day, after a brief introduction by a professor,
one of his bachelors was appointed to receive and reply to arguments
presented by members of the audience. The discussion followed the order
of topics suggested by the master in his introduction; the bachelor dealt
with the objections made to the master's thesis, assisted if necessary by the
master himself; a secretary recorded the arguments and replies. The session
continued for most of the morning, occupying perhaps three hours. O n
19. Monumente historica Universitatis Carolo-Ferdinandeae Pragensis, I. Prague 1830, p. 82. (Accepted,
but only if the masters presented the (abridged) versions under their own name!) - Chartularium
Parisiense IV, p. 727. Cf. Piltz 1977, pp. îoff. The scribe of Durandus' Commentary on the
Sentences, which is certainly not one of the least original medieval works, saw that large passages
were verbatim excerpts from Peter of Auvergne. He adopted the practice of underlining such
passages and added: 'Hoc oportui facerc in hac quaestione ne multotiens idem scriberetur, quia ille
Durandus est quidam latrunculus Petri dc Alvernia, sicut sunt communiter omnes Gallici, utpote
homines nullius inventionis existentes' (quoted from Decker 1967, pp. 84-5).
the next available day the company reassembled to hear the master sum-
marise the arguments pro and contra and to give his own overall solution
[determinatio) to the question in dispute. 20 The proceedings of the dispu-
tation were eventually published either in the form of notes taken at the
meeting (a reportatid) or in a revised and expanded version by the master
himself (an ordinatio).
Ordinary disputations throughout the academic year would concern
topics clustering around a single theme, such as are to be found in Aquinas'
De veritate, or disputed questions on truth. 2 1 Twice a year, in Advent and
Lent, there were special disputations, open to a much wider public, which
could be about any topic whatever {de quolibet) and could be initiated by
any member of the audience (a quolibet). Such quodlibetal questions had to
be answered impromptu by the bachelor and determined by the master in
the same way as the questions of an ordinary disputation. It is a tribute to
the intrepidity and resourcefulness of medieval professors that so many of
them were willing to undergo such a testing public ordeal: no less than 358
sets of quodlibetal questions have come down to us. 22 If the ordinary
disputation resembled a modern seminar, the quodlibetal disputation re-
sembled a public version of a modern professor's office hour or informal
instruction. Both types of disputation, in the words of P. Mandonnet, were
the academic equivalent of the medieval tournament-at-arms.
Quaestiones disputatae, such as Aquinas' De veritate, have come down to us
in texts divided into questions and articles. A question will be concerned
with a broad topic — e.g., Question T w o treats of the knowledge of God —
and will be divided into articles devoted to specific problems - e.g., the
twelfth of the fifteen articles on Question T w o asks whether God knows
future contingents. It is a matter of dispute whether it is the question or the
article which corresponds to a particular morning's debate: if the former,
the three hours appear to have been impossibly packed with rapid and
dense argumentation; if the latter, since the De veritate contains 253 dispu-
tations, the disputations of only three years, the academic calendar must
have been so full of disputations as to leave little time for lectures. 23
20. Classic accounts of the nature of disputations in the theological faculties are to be found in
Mandonnet 1928, Little and Pelster 1934, pp. 29-56, and in Glorieux, 1925-35.
21. Though the questions disputed by Aquinas in a single year commonly are linked fairly closely
with each other, the traditional groupings under the headings Df veritate. De potentia. De malo,
include several sub-groups rather tenuously linked with each other.
22. Glorieux 1925-35.
23. The controversy between Mandonnet and Dondaine on this topic is well summarised in
Weisheipl 1974a, pp. 123-6.
Disputations in the Arts faculties

We know less about disputations in the Arts faculties. From university
statutes we can gather that Arts disputations were normal and that atten-
dance at and participation in disputations were required before receiving a
degree. But we are not told much about the actual procedures, except that
it seems that some disputations were connected with the ordinary lectures,
whereas others were more or less independent. These latter disputations are
frequently called sophismata.2*
Information on the procedures of disputation must be gathered mainly
by induction. One type of evidence is the arrangement for final exami-
nations, which included a formal disputation that was supposed to prove
the candidate's ability to assume one of his main magisterial functions, that
of leading a disputation, and therefore presumably proceeded according to
the same rules as governed regular disputations. The structure of such
disputations follows the following pattern: first, the master in charge puts a
yes-no question, giving some arguments on each side. Next, the respondens
(sometimes called the promovendus) gives a short solution accompanied by a
refutation of the arguments leading to the opposite conclusion. Then the
presiding master in his role as opponens argues against the respondent's
solution and refutations, and the respondent is allowed to reply. Another
master could then argue against the respondent's new position. It is not
certain whether the respondent in his turn was obliged or at least allowed
to reply. This entire procedure is not very different from the pattern of
theological disputations. Also in the Arts faculties the disputation appar-
ently was followed by a determination by the master. We have no positive
evidence that the determination took place on the following day, but
analogies with theological disputations would support such an inference.25
Further evidence can be gathered from an analysis of actually recorded
disputations. The 'redacted' questions of the exercitia which retain only the
basic structure of the actual disputation, are not so valuable as are a number
of English manuscripts, which seem to be transmitted in a form closer to
original disputations. Preliminary answers and solutions (presumably by
the respondent) are subjected to further attacks and counter-arguments. It
is often difficult to locate the final answer to the original problem, since it is
mostly stated very briefly. Perhaps there was no formal determination in
these cases.26
24. Weisheipl 1964a; Gilbert 1976.

25. Cf. Roos 1963.
26. Pinborg 1979, p. 34; Little & Pelsrer 1934, p. 41.
Many sophismata, especially from Paris in the latter half of the thirteenth
century, are in the form of regular disputed questions. The formal occasion
of the disputation - the sophisma proposition - is often rapidly dismissed.
Then anywhere from one to four problems, with or without connection
with the sophisma proposition, are stated. For each problem a few principal
arguments are given on each side, followed by a brief solution by the
respondent (or bachelor), w h o also responds to the arguments against the
position he has chosen. Then numerous counter-arguments are raised
against this first solution, apparently by different opponents, and the
respondent replies to some or all of these. New counter-arguments are
raised, sometimes concentrated on a single point of the respondent's
argument, and this procedure can be repeated several times. Normally the
respondent gets the last word, although the copyist sometimes does not
care to report it. 2 7 In some sophismata we find a variant of this type of
disputation. Several respondents can participate successively, each being
subjected to the same treatment, but often maintaining opposed
opinions. 2 8 In most sophismata the final, and often the largest, part is
devoted to the master's determination of the problem.
That disputations were not always without danger can be seen from
statutes threatening to exclude students who demonstrate by 'clamoring,
hissing, making noise, stone-throwing by themselves or by their servants
and accomplices, or in any other way'. 2 9 Even the most technical reports of
medieval disputations can sometimes be relieved by glimpses of the tumult
of actual disputation. Thus Matthias of Gubbio, trying to give an orderly
account of his opposition against the opinions of Hervaeus Natalis con
cerning the nature of logical relations, is interrupted by someone: 'But
before I come to the fourth point, somebody shouts against me with a loud
voice: You deny such relations, I certainly deny y o u r s . ' 3 0 Certainly,
disputations did not proceed as solemnly as the written redactions might
make us believe.
The origins of disputation

Although scholars are unanimous in seeing the disputation as one of the
most important and influential features of scholastic method, they are less
27. Roos 1963; Pinborg 1975b.

28. E.g., the sophismata edited in CIMAGL 24, pp. 16—34; see esp. p. 19: 'Secundo quidam alius
sumpsit sibi quaestionem.' Cf. Little & Pelster 1934, P· 50.
29. Thorndike 1944, p. 237.
30. MS Erfurt 4° 276, f. Ι42 Λ .
agreed about the presuppositions and antecedents of the institution. A

disputation is an institution which excellently formalises the dialectical
procedure described by Aristotle: present a problem (aporia), set down the
conflicting opinions of philosophers (endoxa, phainomena), resolve the diffi-
culties and restate the endoxa in a muddle-free manner (e.g., N.E. VII I,
ii45 b 2-7). 31 But it is unlikely that the disputation was designed to
embody these procedures since it was in operation in medieval Europe
before Aristotelian ideas and methods were fully assimilated. Some
scholars regard the spread of disputations as the outcome of the study of
Aristotle's Analytics and Topics; but the eristic jousts for which the Topics
prescribe are a question-and-answer game quite different from the presen-
tation of conflicting arguments in a disputatio.32 Abelard's Sic et non was
long singled out by scholars as a progenitor of scholastic dialectic, because it
sets out contrasting Scriptural and theological statements in a manner
which highlights conflicts of doctrines in the same way as a Quaestio
disputata. Abelard's autobiography is one of the first records of school
disputations: but it is clear that the dialectical procedures of the Sic et non
were already practised in an earlier period.33 Recently it has been sug-
gested that the origin of the disputation is to be sought in the procedures
adopted to reconcile conflicting legal authorities by canonists, Roman
lawyers, and even Islamic jurisprudents.3*
Perhaps the disputatio simply grew out of the other and older vehicle of
professorial instruction: the lectio, or lecture. In the course of expounding a
text a commentator, from time to time, is bound to encounter difficult
passages which set special problems and need extended discussion. When
we are dealing with a sacred or authoritative text, the difficult passages will
have given rise to conflicting interpretations by different commentators,
and the expositor's duty will be to set out and resolve the disagreements of
previous authorities. Thus the quaestio arises naturally in the course of the
lectio, and the disputation and the lecture are the institutionalised counter-
parts of these two facets of a method of study oriented to the interpretation
of texts and the preservation of tradition.35
31. On this aspect of Aristotle's dialectic, see Owen 1967.
32. Grabmann 1911, II, p. 218, saw the influence of the Topics as decisive, but did not notice the
difference between the type of disputation discussed by Aristotle and that found in Aquinas.
33. Grabmann 1911,1, pp. 234-6, gives Bernold of Constance and Ivo of Chartres as practitioners of
the Sic-€t-non method prior to Abelard.
34. Makdisi 1974, pp. 640-1, suggests an Islamic origin for the disputation (practised in champion-
ship style in the munäzara) and refers especially to the dialectics of Ibn Aqïl (1040-1112).
35. This account of the development of the quaestio from the lectio is given by Chenu I9S4,
pp. 67-77.
Whatever its origin, the disputatio came to have a great influence on the
style of works written in other genres. The Summa theologiae of Aquinas,
though designed from the outset as a manual to be read by students, is
divided not into chapters but into questions and articles, and each article
has the form of a miniature disputation, with three or more arguments
against the position to be adopted, a brief citation of an authority in
favour of the preferred view (sed contra), a central section (respondeo)
corresponding to the Master's determination, and finally a set of answers to
the objections. In writers such as Scotus and Ockham commentaries on the
Sentences of Lombard follow the form of a Quaestio disputata.36
The later development of disputations

While commentaries and treatises written in and for the master's study
were written in the form of disputations, the disputations themselves
became more and more an exercise of dialectical skill for its own sake and
less and less a method of presenting and reconciling diverse opinions on a
topic of substantial import.
In post-reformation scholasticism the presentation of conflicting argu-
ments on a substantive topic of theology or philosophy was subject to
formal and codified rules analogous to those presented in an ars ob~
ligatoria.31 But there was a significant difference between the medieval and
the post-reformation disputations: whereas the former began with a quaes-
tio, the latter began with a thesis. It was the respondent's duty to enunciate
and explain a thesis which he was prepared to maintain against objectors.
The opponent must then produce arguments to contradict this thesis; the
respondent must react to each premiss of the opponent's argument by
granting, denying, or distinguishing (concedo, nego, distinguo). If he dis-
tinguished the opponent's premiss, that meant that he granted it in one
sense but denied it in another, he must then strive to show that in the sense
36. In Scotus in particular the quaestio-iorm can become very complicated. In the fifth part of the
prologue to his commentary on the Sentences he discusses theology as a practical science. Two
questions are posed, the first with five authorities on one side and three on the other, the second
with three authorities on each side. There follows a definition of praxis argued for clause by clause
through eleven paragraphs. Two further long preliminary sentences precede the solution of the
questions in reverse order. The solution of the first question is preceded by a long summary of the
opinions of previous thinkers, divided into five classes. This summary of previous thought,
corresponding to the first part of a magisterial determinatio, frequently plays a dominant role in the
Ordinations of Scotus and Ockham. Few quaestiones iisputatae of Scotus have survived, and fewer
have been published; but Scotus, like Ockham, left behind an important scries of Quaestiones
quodlibetales.
37. The two types of disputation are compared and contrasted in Angelelli 1970.
agreed to the premiss did not lead to the contradictory of his own original
position. The opponent must then try to prove the premiss in the sense in
which it is dangerous to the respondent; or he may turn to an independent
argument for the contradictory of the respondent's thesis. It is the distinguo
which is the heart of the post-reformation disputation: this means that the
interest of the debate turns on the disentangling of senses of ambiguous
words rather than on the more formal considerations which typically gave
excitement to medieval disputations.
Disputations of this kind have been in use in scholastic institutions
within living memory, and their form dictated the typical structure of neo-
scholastic textbooks in which, instead of chapters, we find the material
divided into theses.38 Each thesis is first explained, word by word (this is the
status quaestionis); there follows a list of adversarii (philosophers who have
taken a position contrary to the thesis); the thesis is proved by syllogistic
argument, and there follows a set of objections and replies, often with
scholia or appendices. The structure of a medieval quaestio disputata is more
lively: the adversaries are allowed to speak for themselves, and the argu-
ments pro and con are presented, as in a live debate, before the magisterial
resolution and not after.
The purpose of disputations

As a final consideration of medieval disputations I should like to sketch
some of the ideological and methodological impetus behind the medieval
preference for this method of teaching. T w o texts on the problem whether
the cognition of truth consists in the solution of dubitations (formally
expressed doubts) may serve as a starting point.
The first text is by Siger of Brabant. It states that the goal of teaching is to
find truth, and finding truth presupposes the ability to solve any objection
or dubitation against the proposition accepted as true. For if you do not
know how to solve objections that may arise, you are not in possession of
the truth, since in that case you have not assimilated the procedure of finding
truth and thus will not know whether or when you have arrived at truth. 3 9
38. A typical example of such a textbook is Stwek 1948.

39. Siger of Brabant, Quaestiones super librum Decausis, Prooemium, 1972, p. 35;'Sicut vult Aristoteles
in principio III. Metaphysicae, volcntes attingere ad cognitionem veritatis in aliquibus rebus
absque cognitione eorum, quae dubitationem inducunt in cognitionem veritatis illarum rcrum,
similes sunt incedentibus nescientibus tarnen ad quern locum ire debeant. Cuius ratio est, quia ab-
solutio dubitationis finis est tendentis ad vcritatem. Et ideo sicut qui nescit locum non venict ad
ipsum nisi casu, et cum ad ipsum vencrit ncsciet ipsum esse locum quo tendebat, et ideo ignorabit
utrum sit ibi quiesccndum vel ulterius procedendum, sic non praeconcipicns dubitationes ad
The other text, probably by Henry of Brussels (ca. 1300), alludes directly
to the three methods of teaching, stating that through lectures you arrive at
truth and so should be able to solve any objections. By a lecture in the form
of a (fictitious) disputation read aloud procedures for finding the truth are
presented to you, and by an actual disputation you learn to find truth by
actually evaluating and solving arguments.40
It was certainly not the case that this lofty purpose of disputations was
always in the mind of the master, not to mention that of the student. Many
disputations were obviously undertaken just as a means of intellectual
exercise or as unreflective repetition of normal academic usage.41 Still,
medieval disputations and their written counterparts show us scholars
partaking in what Chenu has aptly termed the 'recherche collective de la
vérité'.42 Medieval scholars were aware that any single disputation covered
only a tiny move within this vast enterprise, and that not all disputations
led to an equal degree of certainty.
There is a certain tension between the medieval ideal of'demonstrative'
science as a system of proofs deducing conclusions by ordered steps from
first principles, and the actual forms of doctrinal exposition. The argu-
ments adduced in a disputation have an almost fortuitous character and are
certainly not always demonstrative. Their aim is principally to persuade
cognitionem veritatis non dirigetur nisi casu, quia si veritatem attingerit nesciet utrum ibi
quiescendum vel ulterius procedendum. Et dubitans etiam similis est ligato vinculo corporali qui,
si ligamentum ignoraverit, ipsum dissolve« non valebit; dubitatio enim mentem tenet ne ulterius
per considerationem procedere possit, sicut vinculo corporali pedes tenentur. Et ideo dubitationes
non praeconsiderans non valet absolvere dubitationes, quare nee attingere ad veritatem. Cognitio
enim veritatis in aliqua rerum solutio est dubitatorum. Et sicut in iudiciis dicitur, quod melius
contingit iudicare audiendo rationes utriusque partis, similiter etiam praeconsideratis rationibus
ad utramque partem contradictionis dubitationem inducentibus melius contingit iudicare
veritatem.'
40. Grabmann 1944, p. 82: 'Secundo prenotandum est, quod cognitio veritatis generatur in nobis
dupliciter uno modo per inventionem, alio modo per doctrinam. Si per inventionem hoc fit sic,
quod aliquis proponit primo sibi aliquam conclusionem (f. 91*) quodam modo notam et per
consequens arguit ad utramque partem et tunc judicat, ad quam partem rationes sunt potiores
adducte illi consentiens et alias rationes dissolvens et per hoc patet, quod investigatio veritatis etc.
Alio modo generatur scientia sive cognitio veritatis per doctrinam et hoc dupliciter. Uno modo,
quod doctor proponat propositionem discipulo et arguât ad partem utramque et postea uni
consentiat et alia, quae sunt contra ipsa m partem quam tenet, dissolvit. [Alio modo ] Et hoc modo
patet etiam, quod cognitio veritatis est solutio dubitatorum. Alio modo per doctrinam fit
cognitio veritatis sic, quod doctor simpliciter sine omni arguitione proponit discipulo veritatem
et informat ipsum et sic item discipulus ista veritate cognita poterit argumenta solvere, que essent
contra istam veritatem et sic patet, quod cognitio veritatis etc'
41. An incidental remark in an anonymous series of questions on Priscian may illustrate this (MS
Nürnberg, Stadtbibl., Cent. V.21 f. 37rb): "Ista opinio tacta fuit in praecedenti quaestione et posset
satis probabiliter teneri. Tarnen exercitii causa aliam inquiramus.'
42. Chenu 1954, pp. I09ff.
the opponent, and because of that aim it is not necessary to start every
argument from first principles. One can instead begin with commonly
accepted presuppositions, either quotations from authorities or well-
known maxims. One can also formally agree on some accepted presup-
positions. 43 The arguments are developed in the form of syllogisms,
categorical or hypothetical, consequences, or dilemmas. Overt formal
errors are exceedingly rare, since such 'sophistical' moves could be detected
immediately and thus defeat their author.
In most questions or disputations the introductory arguments which set
up the question are almost stereotypical and transmitted from one author
to another. Masters and students chose the arguments which they could
remember having heard or could easily look up in accessible sources.
Accordingly, questions from the same environment, raising the same
problem, will tend to repeat the same stock-arguments, thus forming a
kind of'doctrinal family' which can be used with some caution for a rough
sorting of texts. In rare cases we can even see how a cluster of stock-
arguments arise from a specific historical situation. 44
The literary forms: literal commentaries

Most medieval philosophical literature reflects teaching practice and its
form; even writings that were never delivered as lectures or held as
disputations assume the traditional forms.
Accordingly, a large proportion of medieval philosophical literature is
in the form of commentaries. Commentaries range from more or less
complete marginal glosses (scholia), which in fact are often excerpted from
one or more complete commentaries, to full-scale literal commentaries,
following the standard form of a lecture as sketched above. It is important
to note that such commentaries do not necessarily reflect the personal views
of an author on the philosophical problems involved, but only what he
thinks is the correct interpretation of the littera. However, since new
doctrines were often disguised as merely new interpretations of the author-
ities, it is difficult to know where the borderlines between the exposition
of the authority and the author's personal philosophy may be drawn. It is
known that some of the strenuously attacked 'Latin Averroists' often used
as a defense that they did not develop their o w n opinions but only
43. There exists no analysis of the type of arguments which were actually used by the schoolmen.
Ocing-Hahnhoff 1963 discusses some of the aspects involved.
44. Thus the arguments ofJohn Aurifaber against the modi significandi soon became stock arguments
of the genre; see Pinborg 1967a, pp. 167-72.
interpreted Aristotle. 45 It is also interesting to follow the fourteenth-

century debate between Ockham and his followers on one side, who
brought their own doctrine into harmony with Aristotle by claiming that
Aristotle often spoke metaphorically, and more traditional Aristotelians on
the other side, who could not concede that the philosopher ever spoke
metaphorically, since he himself claimed to proceed demonstratively. 46
Question-commentaries
Besides the literal commentaries there are commentaries in the form of a
series of questions. Originally such questions formed only the latter part of
lectures, but apparently they gradually became independent from the
traditional lecture-form. From the latter half of the thirteenth century we
find such commentaries, consisting only of questions; sometimes short
paraphrases of passages of the littera, otherwise neglected, are still given; but
often only the opening words are left from the old structure. 47
Such questions retain the simplest possible structure of a disputation.
First a problem is stated in the titulus quaestionis which is always formed as a
question introduced by 'utrum'. The selection of tituli reflects current
interests, and so a mere list of questions is often indicative of the time and
place of origin of the commentary. This is perhaps most striking in the case
of commentaries on the Sentences. 48 Vastly different sections of the text
were selected for commentary, and the questions raised in connection with
the various distinctiones of the text varied greatly over time. Early
thirteenth-century commentaries for example, often have a dispropor-
tionate amount of commentary on the first Book, and various philo-
sophical problems are often introduced and discussed at great length in
45. Van Steenberghcn 1977, pp. 232C Cf. Wcisheipl 1974, p. 42, concerning Albert the Great.
46. Ockham, Expos. Phys. (MS Oxford Merton 293 f. 72r, ad 2o6bi6): "Verumtamen aliquando
Philosophus ponit unam pro alia non curans multum de verbis et supponens quod addiscentes
istam scientiam possunt esse sufhcienter exercitati in logica per quam sciant discernere inter
propositiones et advertere quando una ponitur pro alia et quando non. Quod tarnen multi
moderni ignorant ' Walter Burley, Expos. Phys. (Venice 1524, f 5"): 'Non est igitur
dicendum quod Philosophus et Commentator loquantur metaphorice vel singulariter ut isti
exponunt, reducendo totam philosophiam ad secundum modum amphibologiae. Quia Philo-
sophus reprehendit modum loquendi figurative vel metaphorice in doctrina demonstrativa. Sed
hie (seil, in Physicis) procedit demonstrative. Non est ergo dicendum quod ipse loquatur
metaphorice sicut loquuntur poetae in suis fabulis.'
47. Grabmann 1939, PP· 47-53·
48. The Sentence-commentaries were read, but in the fourteenth century the introduction to each
book (the principium or the quaestio collativa) was presented as an actual disputation and normally
published separately (Glorieux 1966, pp. 79-90). In the Arts faculties masters sometimes used
material from their disputations or sophismata in their redacted questions. (An example is
Boethius of Dacia; see Roos 1963, p. 379.)
connection with passages of the text which have only a very tenuous
connection with the problems discussed.
Thus we find an elementary discussion of light and the multiplication of species or
of the rainbow inserted into the context of creation, an examination of the
problem of the motion of gravia et levia in a similar context, an elaborate
consideration of terms of first and second intention in the context of Trinity
(specifically at the discussion of persona in dist. 23), an extended investigation of
astrology relative to the problem of whether creation occurs de necessitate, and
even a major question de creatione caeli dealing with whether one can prove that
there are nine spheres (for discussion of the ninth sphere, can, and does, lead to the
consideration of the precession of the equinoxes, the rising and setting of the signs,
the astronomy of eclipses and so on). And such phenomena can be found almost ad
infinitum.*9
It is hardly possible to give a catalogue of places where specific philosoph-
ical problems are discussed since it varies greatly with different authors
and periods. From this one can understand that the university authorities at
the end of the fourteenth century found it necessary to rule officially that
Sentence-commentaries should not deal with logical and philosophical
problems except to the extent justified by the text of the Sentences. 50 This
ruling apparently had some effect. Late fourteenth- and fifteenth-century
commentaries on the Sentences tend to be of a more purely theological
nature.
After the titulus quaestionis follows a short series of principal arguments
for one of the two possible answers to the problem stated, frequently
introduced by a formula such as 'et arguitur (videtur) quod sicInon . These
arguments normally defend the position eventually refuted. The normal
number of arguments is two or three. Then follow arguments on the
opposite side of the issue. They are often fewer in number (often only one),
and are frequently nothing but references to authority. This is justifiable
according to medieval tradition, even if an argument from authority was
held to be the weakest form of argument. Real arguments were normally
given in the solution of the question, where the author adopts the position
he himself means to defend.
The solution (or corpus quaestionis) is introduced by phrases such as 'ad hoc
a^cendum/dico1 and states the conclusions of the author, accompanied by
some arguments and the distinctions necessary to carry through the solu-
tion. These arguments are normally more carefully organised and artic-
49. Murdoch 1975, p. 278.

50. Murdoch 1975, p. 278.
ulated, but still may take as their major premisses propositions which have
not been or are not proved 'demonstratively' but are only regarded as
generally acceptable. Frequently several previous opinions on the subject
are summarised and refuted, before the author states his own opinion. 51
The structure of the solution of most of the Sentence-commentaries of the
fourteenth century is often very complicated along these lines. Originally
these references to earlier authors were anonymous (introduced by phrases
such as 'aliqui dicunt\ 'opinio cuiusdam viri' or even 'aliquis diceret'), either in
order to retain the fiction of disputation where non-participants could not
supply arguments or because nobody really cared about the authorship of
the ideas discussed. From the fourteenth century onwards there is an
increasing tendency to give exact references. It is interesting to note that
the cluster of such opinions on a given question often varies little from one
text to another, and that many authors are remembered only for their
solution to one specific question. In this way, e.g., Radulphus Brito's
opinion on first and second intentions is bound to appear in any discussion
on this subject although he is not quoted anywhere else, and the same holds
true for Peter of Auvergne on the verbum mentis.
Especially in earlier phases of the development, we find cases of several
questions being telescoped into one: first the arguments of a number of
different questions are given, then the questions are solved one after
another. This may reflect the practice of discussing more than one question
in the same disputation. Another way of structuring such sub-questions is
to divide a question into several articles. In the fourteenth and fifteenth
centuries the solution is often structured according to conclusions and
corollaries, each being defended and dubia being solved.
The last part of a question contains the refutations of the arguments
leading to the solution opposite to the one advocated by the author. They
often contain some distinctions which were thought not to be necessary to
the general solution of the problem but of importance only to solving one
of the counter-arguments.
The whole structure of such questions should make it immediately
apparent that not all parts of a question are of equal importance for
51. See, e.g., the Sophisma of Radulphus Brito edited in CIMAGL 24, p. 97: 'De ista quaestione
procedendum est sicut in aliis: primo tangendae sunt opiniones aliorum; secundo tangenda est
opinio probabilior; tertio tangendae sunt difFicultates circa istam opinionem et solvendae sunt;
quarto est solvendum rationes in oppositum.' Sometimes further complications were intro-
duced by expressions such as 'aliquis argueret contra me', 'aliquis diceret', which may sometimes
be reflections of the many-leveled disputation behind the redacted question, but often are nothing
but literary devices to treat a side-issue more thoroughly.
determining the author's own argument. For such purposes, the most
important part is, of course, the solution with its distinctions. The conclud-
ing responses to the counter-arguments are also important, but the distinc-
tions introduced are often only ad-hoc distinctions never used again, and
thus sometimes not equally well considered. This principle of estimating
the weight of a statement within the given context was already expressed
by the fifteenth-century Thomist Johannes Capreolus, but has often been
sinned against. 52
It has already been stated that some literary genres of medieval philosoph-
ical literature, such as the sophismata, the quaestiones disputatae, and the
quaestiones quodlibetales kept closer to the original form of a disputation.
This is already illustrated by small linguistic features of the text, such as the
frequent use of the past instead of the present tense (Sed contra hoc arguebatur,
etc.) or the frequent addition of 'per te\ which alludes to the opponent's
presuppositions or admissions, and above all by the far greater structural
complexity.
Other literary forms

Besides the commentaries with their detailed analysis of problems, a need
arose for more systematic expositions of doctrine. The standard title for
such expositions is summa, which originally meant a summary, or tractatus
(treatise). Such expositions were generally intended for the use of begin-
ners in order to facilitate their introduction into a discipline. 53 It is natural
that such expositions tended to be shortlived. They soon lagged behind the
doctrinal developments and needed to be replaced. 54
Such manuals are often conventional in their contents. Every teacher
adopting them felt free to revise his teaching manual, to change words or
whole sections, to add or to remove. This accounts both for the general
similarity of manuals within a given field and for their great variety in
details. Accordingly, it is often very difficult to see which specific manual
was used by an author, especially since we possess only a mere fraction of
52. Johannes Capreolus, Defensiones IV, d. 43, q. 1, a. 3, quoted from Grabmann 1926-56 III,
p. 379; 'Tarnen teneo cum sancto Thoma in Quodlibeto, unde sumptac sunt conclusioncs.
Qualitercumque enim sensit in Scriptis vel visus fuerit sensisse in Tertia Parte, determinatio
Quodlibcti vidctur mihi rationabilior, quia ibi solum tractavit istam materiam a proposito et in
forma; in aliis vcro locis incidcliter solum et cum suppositionc et respondendo magis ad homincm
quam ad rem.'
53. Cf. Grabmann 1939, pp. 54-103. See also Parkes 1976.
$4. Cf. John Wyclif, De veritate Saciae Scripturae I, 54: 'Aliac logicae (i.e., other than his logica
scripturae) sunt periodice et nimis multipliées; periodice quia, ut patet in Oxonia, vix durât una
aliéna logica per viginti annos sed saepissimc variantur.'
the material that once existed. In the same way it may often be very difficult
to trace the interrelations of such manuals.
Manuals are either summaries of the formal books required for a degree
or introductions to specific disciplines or aspects of disciplines for which no
authoritative textbook had been recognised. This last type is philosoph-
ically the most interesting. Within it are to be found the various gram-
matical and logical manuals which will be discussed frequently in the
following chapters of the present volume. 5 5 Also the manuals discussing
typical medieval contributions to natural philosophy, such as treatises De
proportionibus velocitatum, De intensionibus et remissionibusformarum, and De
primo et ultimo instanti belong to this category, even if they are often so
intricately argued that they defy characterisation as elementary textbooks.
Subjects generally accepted into the curriculum will often be incorpo-
rated at specific sections of the formal textbooks. We have already seen
examples of this in connection with the Sentence-commentaries, but
analogous developments can be found in the Arts faculties. Thus the modi
significandi are treated in connection with Priscian or with Aristotle's Peri
hermeneias, first and second intentions with Porphyry, consequences with
Aristotle's Sophistici elenchi or Prior Analytics, restrictions and ampliations
in connection with the Prior Analytics or Peri hermeneias, proportions of
velocities in connection with the twelfth book of the Metaphysics, and first
instants, intension and remission of forms, and impetus theory in connec-
tion with the Physics. This arrangement is found, for example, in some
series of questions prepared as a help for examinations. 56
Frequently certain manuals even came to be regarded as defacto authori-
ties and accordingly were made the object of commentaries or dis-
putations.
The nature of our sources: the channels of transmission

Since we have no direct access to the actual lectures and disputations we
have to rely on written sources. This is of course a truism, but it emphasises
the important fact that there is always an intermediary between the source
as we have it and the state of affairs we want to consider. In a technically
55. A survey of the types of logical manuals and the early history of their development can be found
in Dc Rijk 1962-7, 11. 1, pp. 593~6.
56. Pinborg njjôà. Other aids of this type can be found in Grabmann 1939, pp. 112-16 & 189-91,
and in MSS, e.g., Erfurt 4°24t:'Puncta matcriarum omnium que pro baccalariatu gradus Erfordie
leguntur ct cxaminatur.'
developed culture where it is possible for a large number of people to have

access to the same text directly, and in a form authenticated by the author,
this is less problematic, even if there is, of course, always the possibility of
misunderstanding. But when we are concerned with medieval philosoph-
ical literature such relatively unproblematical lines of communication are
very rare. The number of texts transmitted to us in the author's own
original version or in a copy directly authenticated by him is very small.
And even in these cases there are additional difficulties: the author's hand-
writing may be almost illegible, as is the case with Aquinas, or he may have
been remiss in checking the text. But in the normal case several written
intermediaries separate us from the original text of the author.
In order to gain a sketchy insight into the process of transmission let us
consider briefly the route of a medieval text from its author to us.
Let us suppose that a master has just been giving a course. The students
have been taking copious notes. The master has been reading either so
slowly that he has actually dictated the text, or, according to the recom-
mendations of the university, somewhat faster, so that even practised
stenographers among the students may have had difficulties in getting
everything straight. 57 N o w we have already two types of texts: a dictated
one and a reported one, both of which contain unavoidable errors: errors
of the master in reading, errors of the students who may have misheard,
miswritten, or misunderstood. The dictated text will of course be closer to
the intentions of the master, whereas the reported text will be more or less
abridged, deformed, and changed. From the students' notes other students
may make copies, which are totally exempt from the control of the master.
The master may want to have his text 'published' officially. He can take
either his own notes (eventually dictating from them to a secretary) or one
of the dictated or reported versions as the basis for such publication. 58 In
both cases he will certainly introduce some revisions such as additions,
reformulations, doctrinal adjustments, or references. Some errors may still
be overlooked and some may be newly introduced, e.g., if the revisions are
not consistent. Such a revised manuscript is called an ordinatio or a text
edited (editus) or provided for copying (in copia datus).59 Frequently several
57. Thorndikc 1944, p. 237Î

58. On reportations, see Pclzer 1964, pp. 422-9. On dictation: Dondaine 1956. See also Thorndikc
1944, p. 58, for a modest student-copyist who excuses himself for not being competent enough to
reproduce the lofty ideas of his master.
59. Fink-Errcra i960.
years pass between the course and the revision, which of course may add to
the differences between the original and the 'final' text of the author. 6 0
Disputations are reduced to written form along the same lines. But the
complexities of the sometimes rather chaotic disputations makes the
student-reporter's job much more difficult, and different reportata will tend
to show much greater differences. Misunderstandings arising from this
could sometimes be dangerous to the author, and he had a genuine interest
in seeking a more authentic form of his opinions. Accordingly we find at
least three different types of disputation-texts: reportations, which report
only from the disputation and not from the master's ordered determi
nation; more or less exact reportations from the determination; and the
revised edition of the master. 6 1
In this way we already have different versions of the 'same' text. Matters
may be still more complicated if the author introduces further changes in
his o w n copy, which is then recopied, perhaps even at different stages, or
when the master in repeating his course recasts the argument, omits some
problems and adds new ones. Here it is possible to talk about different
redactions of a text, but it will also be apparent that the concept is
somewhat fluid.62
The next stage in the transmission consists in the copies made from direct
reportations or from authenticated texts. In the latter case the texts pass
through official channels: the stationarii of the university, who could hire it
out for copying. In both cases copies were made by copyists of varying
competence and for varying purposes. Professional scribes could write
beautiful manuscripts, but they often did not understand the text very well.
60. To give an example: Hervaeus Natalis' Commentary on the Sentences were 'published' about ten
years after the course had first been given. In the meantime, reportations or private copies were
circulating haphazardly. So in the published edition Hervaeus removed some discussions
which he had published separately and in some other, probably more controversial, cases
emphasised that he still held the same opinion (Decker 1967, p. 73). The same move can be
recognised in some texts by Siger of Brabant, who says, e.g., 'quibus oretcnus tunc respondi-
musetadhucrcspondemus' (Van Stcenberghcn 1977, p. 193). Sometimes, however, a master just
changes his opinion without notifying us (Van Stcenberghcn 1977, pp. 402-3). Walter Burley
once even firmly rejects an opinion of'some' (aliqui) without mentioning that he previously held
it himself.
61. Glorieux 1925-35,1, pp. $1-5. A master defends himself by saying 'reportator meus non bene
conccpit" (ibid., p. 52). Discrepancies between the actual disputation and the written text are often
recorded, e.g., 'Multa alia fuerunt arguta quae tamen redeunt in idem' (CIMAGL 24,
p. 19). 'Hoc est sophisma nunc et alias propositum circa quod multa proponebantur inquirenda;
quibusdam tamen disputatis de uno quaeratur ad praesens' (CIMAGL 26, p. 93). 'Ad quartum
problema non respondebatur, propterea omitto ad praesens' (MS Worcester, 4Ί3, f. 33").
62. Macken 1973; Bocthius of Dacia 1969, p. xvii; Ockham ΟΤΙ, pp. 19*-21*. 26*-3I*; Scotus,
Opera omnia I (1950) pp. i66*-75*; Pinborg 1980,
Poor students who copied texts for more fortunate colleagues often
worked hurriedly. Only in some cases do we have the optimal circum-
stances of a copyist copying what he wants to have for himself in the best
possible form.
Copyists could be interested in the text they copied simply for its own
sake, which of course warrants their producing the best possible copy. But
often they were interested only in some aspects of it, some of the argu-
ments, the main trend of thought, or perhaps only in some definite
sections. Thus we find manuscripts excerpting discussions from different
sources, e.g., treating the same question, or having some degree of internal
relationship. Sometimes copyists got disappointed or tired and just quit
their work in the middle. Others might have been compiling notes which
were useful for their examination, etc. The study of such miscellaneous
manuscripts is still only beginning, but obviously it is important to have
some general idea of what a given manuscript was intended for in order to
judge its value as a source. 63
So far we have considered only linear transmission, that is, copies made
from one other copy. In the actual process of copying we meet with all
sorts of cross-currents. A copyist might want to get a better text by
adducing other versions for comparison. He may then substitute the
readings of the 'new' source(s) for those of his archetype, or he may add
alternate readings between the fines or in the margin, thus starting a
contaminated version. If in this way he is combining different versions
(reportations) or even different redactions of a text the resulting contami-
nation may give very confusing results. Add to this the fact that a large
number of the manuscripts are now no longer extant, and it becomes
obvious that it is hardly ever possible to reconstruct precisely the history of
the text which would be the necessary background for evaluating the
various sources.
Even the official publication through the stationarii, which at first sight
seems to warrant a high degree of authenticity, has been proven to intro-
duce by its very technique new sources of error. First the actual unit for the
transmission of a text is not the entire manuscript but the pecia, a quire
normally of 16 pages; since the stationarius normally has at least two sets of
peciae of a given text, more or less identical, and since the peciae are hired
one by one, any copyist may be combining peciae from two different
sources into his copy, thus making different parts of his text of different
63. Glorieux 1925-35, p. 54; Glorieux 1967.

critical value. Moreover, the/recta in itself is not a stable entity; it will suffer
wear and tear, so that words or even whole sentences may have become
difficult to read, corrections and marginal remarks (often totally irrelevant
to the text) may have been added by less conscientious borrowers, etc. The
pecia may even have become so worn that it had to be replaced by a new
copy, which even if carefully supervised by the university is certain to
introduce new errors. Thus even an official version does not offer the text
authenticated by the author, but a text authenticated by the university,
which might be a quite different matter. W e even have indications that
some texts were changed so as to offer more acceptable doctrines. 64
The last step in the transmission is the modern edition. Here it is always
important to recognise the character of the sources used and the way the
editor has chosen to render his material accessible. He may be offering a
mere transcript of one manuscript, or a contaminated version of his own,
or a genuinely critical edition based on all available material and structured
according to a specified hypothesis concerning the history of the text. In
any case the modern editor has several advantages over the medieval
copyist: he can make use of as much source material as he wants, thanks to
the technique of microfilming manuscripts; he has technical equipment at
his disposal in case the manuscript is difficult to read; he has many manuals
and much secondary literature composed to help him; and above all his aim
is to be faithful to the text being edited. Still in many respects his situation is
akin to that of the medieval copyist: he is apt to be guilty of the same types
of errors, misunderstandings and inexactitudes. As there is no manuscript
without errors, so there is no faultless modern edition. Even the almost
superhuman editors of the Leonine edition of Aquinas are not infallible.
Manuscripts and their errors

Manuscripts and their errors are not only a problem for the medieval user
and the modern editor of a medieval text, they are also a problem for any
user of modern editions. I am not referring to the fact that knowledge of
the material conditions of medieval scholars can be illuminating for
modern interpreters, but to the fact that any user of a critical edition must
use it critically. T o be able to do so, however, he must know the most
important principles which inflict errors upon medieval scribes and mo-
64. The classical study of the pecia-system is Destrez 1935. Very important issues are considered in
Fink-Errera 1962 and Brounts 1970. The prefaces of the recent volumes of the Leonine edition o(
Aquinas are veritable goldmines of information concerning university editions and paleograph-
ical and codicological matters generally.
dem editors. He must have a certain experience with medieval scripts in

order to know which letters can be mistaken for which; in particular, he
must have a working knowledge of the medieval system of abbreviations
which explain many errors (see further below). This knowledge is impor-
tant not only for using the critical apparatus, but also in order not to be
totally at a loss whenever a text appears misleading or false.
It is not possible within the limits imposed on this chapter to provide a
manual of paleography and textual criticism. N o r would it be appropriate,
since a number of important and useful surveys are already in existence. 65 I
propose to give a mere catalogue of typical faults with some examples
found in manuscripts or in modern editions. For our purpose here, it is of
little importance whether the error is medieval or modern.
All errors form, in point of principle, two different groups. In the first group,
the incorrectness of the reading can be established by means of various criteria:
morphological, syntactic, stylistic or contextual. (This means that a precondition
for applying such criteria is a thorough knowledge of scholastic terminology and
Latin usage). The second group includes cases in which the unauthentic text seems
to be correct in respect to the mentioned criteria: there is a divergence from the
original text, but the divergent text does not itself justify any suspicions as to its
authenticity, as far as language and content are concerned.66
Obviously the second group is very difficult to handle without access to a
great deal of primary material. But errors in the first group even the casual
user of medieval texts may be able to discover. And with some grasp of the
principles discussed in the following pages he may even be able to find
remedies.
One group of errors is connected with the omission of one or more
words, and the most common sort of omission occurs because ofhomoio-
teleuta. The copyist copies out a word, say 'Syllogismus', on line 3 and when
he turns back to the manuscript his eyes fall on 'Syllogismus' on line 7, and he
accidentally goes on from there. Since scholastic style always repeats the
same terms for the same concepts, and since it is replete with arguments,
which in order to work must repeat the same terms, this type of error is
extremely common in scholastic texts. There is no medieval manuscript
which does not commit this error over and over. In most cases, however, at
least when the omission is not too extensive, it is possible to reconstruct the
65. A good introduction to the history of texts is Reynolds & Wilson 1974, although it is mainly
concerned with the transmission and problems of classical authors. As an introduction to different
scripts Thomson 1969 can be recommended. See also CIMAGL 5.
66. Bergh 1978, p. 5.
gist of what is missing, since the structure of the argument can be used as a
control.
More difficult to deal with are omissions of one word or a few words,
either because the scribe judged them superfluous or because of sheer
negligence. Such omissions can in principle be detected only by com-
parison with other manuscripts or if the sense has become truncated or
illogical as a result of the omission. However, we must always allow for the
fact that both masters and students were thoroughly imbued with scholas-
tic procedure and terminology and so did not need as much redundancy as
a modern reader is likely to need.
Sometimes the scribe discovers his mistake and adds the missing words
in the margin or above the line, or he just repeats the whole passage in its
correct form, adding some sign such as 'va... cat' (written above the
words) to delete the incorrect version. This may cause the next copyist to
insert the words at a wrong place or to produce a confusing duplication of
material.
Deliberate changes are frequent and do not always reveal themselves.
They may be due to various motives. Purely stylistic changes, which are
not very serious from a philosophical point of view, are probably caused by
the speed at which the copying was done and the copyist's automatic
recognition of formulas. Especially noteworthy is the habit of shortening
an argument by just adding an 'etc' when the scribe thinks the rest is
obvious.
Among the more innocent changes which to a certain degree influence
the sense is the substitution of new personal or local names in examples. The
author or the copyist may substitute his own name or that of his own town
or country — or in other cases he may substitute a well-known for a less
well-known name. This is intriguing since such names may be a clue to the
identity of the author. Unfortunately, however, it is often difficult to
ascertain whether the name is the choice of the author, of an earlier copyist,
or of the copyist of the extant manuscript. Sometimes, especially when the
text is transmitted in more than one manuscript, it is possible to argue for
what the original wording must have been. But then the names ought to be
not too frequent and the context significant. 6 ''
Another motive for a change which is more uncomfortable for us may
be the copyist's disapproval of the wording of the original. A rather harsh
67. For geographical names see De Rijk 1976a, pp. 32—6. For examples of personal names introduced
by the author sec, e.g., Martin of Dacia içoi.pp. 12, 53; by the copyist, De Rijk 1977b, pp. 121-2.
example of this is quoted by Anneliese Maier. 68 A copyist of a work of

Burley's breaks off in the middle of a question, saying: 'From this question I
would not copy anything [more], for this fellow Burley fills up a whole
she«t with totally useless material, where he does nothing but give some
solutions and replications which he and some other master flung against
each other, and Burley presupposes many false things, so I did not copy it.'
Additions again include a whole range of more or less important errors.
The trivial form is the duplication of a word or a phrase; the interesting
ones are insertions of marginal material into the text, which again may
vary from irrelevant material, through comments by users of the text, to
afterthoughts by the master. 69
The last class of errors consists in substitution of wrong words or phrases.
Some are due to absentmindedness, when, e.g., the opposite term is
inserted into a complex argumentation. Such errors are often to be blamed
on the author himself (Aquinas is notorious for making such errors), but
most instances are probably the fault of the scribe(s). It is almost impossible
for anyone to avoid this type of error completely, but it is easily detectable
by a close scrutiny of the argument.
Most errors of substitution are connected with the medieval system of
abbreviations. 70 Because of the expensiveness of writing material and
because of the hurry often imposed upon the copyist, medieval schoolmen
and scribes developed a complex and effective system of abbreviations. The
system is not absolutely uniform; it changes with place, time, and disci-
pline, and even the individual scribe is not always consistent. The errors
connected with abbreviations can be categorised as (a) abbreviations
wrongly expanded, (b) abbreviations unrecognised as such, (c) imaginary
abbreviations expanded. 7 ! These three types are frequently combined with
misunderstandings of various sorts: 72 false reading or transposition of
individual letters, misunderstanding or ignorance of technical terms, false
divisions or combinations of words (reading, e.g., 'imagination!' for 'im-
aginatio i(n)' or 'syllogismus medisanus' for 'syllogismus in disamis'), or in-
68. Maier 1964-7,1, 223: 'De ista quaestione nihil volui scribere quoniam ille Burleus facit usque ad
finem quaestionis bene unum quaternum stationis de littera totaiiter inutili, unde non ponit nisi
solutiones ct replicationes quas ipse et quidam alius doctor sibi invicem facicbant, ct supponit
Burleus multa falsa, ideo non scripsi.'
69. See Macken 1973; Dondaine 1956.
70. Capelli 1961; Pelzcr 1966; Piltz 1977, pp. 31S-41.
71. Bcrgh 1978.
72. Very instructive examples can be found in Dondaine 1967, which discusses cases where the
original of the faulty copies is known.
sufficient knowledge of medieval lexical, grammatical, or orthographical

idiosyncrasies.
In order to give a faint impression of what can be expected I give a short
catalogue of errors actually found in modern editions or in manuscripts,
structured according to the three above-mentioned types.
(a) Abbreviations wrongly expanded: perjpraefpro; accidensjaccentus;
conclusiojcognitiolconstructiojquaestio (with confusion o f ' 9 ' = 'con and 'q');
in/id est; universale/virtuale; procediturjpraedicaturjprobatur; positiojratio (with
confusion o f ' p ' and long V ) ; dividiturjdicitur; homojnon (with confusion of
'h' and 'n'; conceptus\contemptus; sedjsecundum; enimjautem; etIvel.
(b) Abbreviations undiscovered: potojpotero; praedicatjpraedicatur; uno/
numéro; patiusfparatius; item\in tantum; antecedensjantecedens probatur (arts');
sicjsicut; item/iterum.
(c) Imaginary abbreviations expanded: tamen/tu; operejope; partitioj
positio.
One final sort of error which is often introduced by modern editors
consists in mistaken punctuation. Since the manuscripts normally have only a
very erratic punctuation which is not taken into consideration by most
modern editors, errors of this kind are the fault of modern editors, though
they have sometimes been misled by the manuscript. The most frequent,
and unfortunately least innocent form of this error is the separation of
adverbial phrases from the proposition to which they belong. But some
editors also unwittingly separate the main clause from its dependent
clauses. A general rule of thumb for users of medieval texts in modern
editions is simply not to trust the punctuation.
A catalogue of dangers and errors such as the one just presented might
discourage the reader by destroying his confidence in all manuscripts and
all editions. Fortunately, that would be an exaggerated reaction. Errors
normally infect only small sections of a text, and many of them are easily
recognisable. The main conclusion to be drawn is that since manuscripts
and modern editions are of varying quality they cannot be used indis-
criminately or uncritically. O n the other hand, this is also one of the charms
of the study of medieval philosophy: it is a study very much in progress and
it has not yet attained the level of stability. New finds may still change the
overall picture considerably, and closer scrutiny of the texts and the
arguments is certain to provide new insights. The present volume is an
interim report, and in no way a survey of established opinions and facts.
π
ARISTOTLE
IN THE MIDDLE AGES
2
ARISTOTELES LATINUS
Introduction to the medieval Latin Aristotle

All of Aristotle's works were translated into Latin in the Middle Ages and
nearly all were intensely studied. The exceptions are the Eudemian Ethics, of
which no complete translation survives, and the Poetics, which, although
translated by William of Moerbeke, remained unknown. Most of the
works were translated more than once, and two of them, the Physics and
Metaphysics, were translated or revised no fewer than five times. The
translations we are concerned with spanned a period of about 150 years;
some were made from the Arabic, but the majority directly from the
Greek. Some translations became popular and remained so; some became
popular but were then superseded by other translations; others barely
circulated at all.
An examination of the medieval Latin Aristotle cannot consider only
the genuine works of Aristotle, but must also deal with works credited to
Aristotle in the Middle Ages although now believed to be spurious. It is also
essential to consider translations of Greek and Arabic commentators on
Aristotle. (All these translations — of genuine and spurious works and of
commentaries - are listed for easy reference in a single table below.)
The basic source for our knowledge of medieval Latin translations of
Aristotle is a corpus of over 2,000 manuscripts dating from the ninth to the
sixteenth century, most of which are distributed among the major libraries
of Europe. They contain the texts of the translations and in some cases
constitute the only documentation we have. Other direct documentation
about translations and translators is sparse.
The scholarly study of the translations began in 1819 with the publi-
cation of Amable Jourdain's Recherches critiques sur l'âge et l'origine des
traductions latines d'Aristote. It was continued in the burgeoning of research
in the nineteenth and twentieth centuries, and culminated in 1930 when a
project to edit the complete corpus of the translations was launched under
the auspices of the Union Académique Internationale. The first achieve-
ment of this enterprise was a catalogue of medieval manuscripts containing
4
6 Aristotle in the middle ages
Aristotelian translations, Volume I of which appeared in 1939; Volume π

and a supplementary volume appeared in 1955 and 1962 respectively, and
these three volumes constitute the basic research tool for the study of the
Aristotle translations. The second achievement is the continuing series of
!
'Aristoteles Latinus' volumes containing critical editions of the translations.
As a result of all this scholarly and editorial activity, most of the basic
problems about the identity and dating of the translations have been
solved. 2
The history of the translations

The basic information about the translations is summarised in the table
below (pp. 74fr.). Only the more important ones are expressly considered
in this discussion, which attempts to outline the stages by which the
complete Aristotelian corpus slowly came into circulation.
At the beginning of our period only two of Aristotle's logical works, the
Categories and De interpretatione, were known in Latin, in Boethius' trans
lation; these two works, which together with Porphyry's Isagoge became
known as the 'logica vetus', had already become standard school texts in
logic. O n e of the results of the quickening interest in logic in the early
twelfth century was the recovery, from about 1120 onwards, of the rest of
Boethius' translations of the logic: the Prior Analytics, Topics and Sophislici
elenchi. How and where these translations, made some six centuries earlier,
were found is not known. The logical corpus was completed by James of
Venice's translation (from the Greek) of the Posterior Analytics; in 1159 John
of Salisbury in his Metalogicon shows a familiarity with all these works. (He
also quotes from a second translation of the Posterior Analytics, that of
Ioannes, which otherwise remained virtually unknown.)
James of Venice also translated the Physics, De anima, Metaphysics (at least
in part), five of the Parva naturalia treatises, and an anonymous introduc
tion to the Physics k n o w n as the De intelligentia. Fragments of his trans
lations of the Sophistici elenchi and of Greek commentaries on the Posterior
Analytics and Elenchi have also survived. Nearly all of these translations,
which were probably made before 1150, were widely circulated in the
1. The volumes of the catalogue (Codices) and the texts so far published are listed in the bibliography
under 'Aristoteles Latinus'; the text volumes are referred to in the footnotes by 'AL', the date, and
the volume number.
2. The AL catalogue and the prefaces to the published volumes of texts are the basic source both for
the history outlined below and for the table. The catalogue has substantial bibliographies. Of
fundamental importance also are L. Minio-Paluello's articles, which have been collected into a
single volume, Opuscula (1972). A valuable earlier survey ofthe translations is in De Wulf 1934-6,
I, pp. 64-80; 11, pp. 28-58.
Aristoteles latinus 47
thirteenth century and played an important role in the dissemination of

Aristotle's works. (The number of surviving manuscripts listed in the last
column of the table can be taken as a rough but reliable indication of the
popularity of the different translations.)
James of Venice's work was only part of a burst of translating activity.
Before η 62 Henricus Aristippus translated Book IV of the Meteorologica
from the Greek; whereupon Gerard of Cremona translated Books I—III
from the Arabic. The two translations later circulated as one text, with a
fragment translated from Avicenna (known as the De mineralibus) included
as a sort of appendix. There were also several translations made from the
Greek by unknown twelfth-century translators: one translated the De
generatione et corruptione; another the De sensu and De somno; another,
independently ofjames of Venice, the Physics (only a fragment survives) 3
and all the Metaphysics except Book XI (James' translation only went as far
as Book IV .4); another Books II and III of the Nicomachean Ethics (the so-
called 'Ethica vetus');* another, independently of Boethius, the Prior
Analytics5 and Topics.6 In Spain, Gerard of Cremona, who died in 1187,
translated Meteorologica I—III. Physics, De caelo, De generatione et corruptione
and Posterior Analytics, and also Themistius' paraphrase of the Posterior
Analytics - all from Arabic versions. O f these only the Meteorologica and De
caelo circulated widely.
In addition to genuine works, a number of important pseudo-Aristotelian
works were translated in the twelfth century, mostly from the Arabic.
These include the Deplantis (in fact a work by the first-century-A.D. Greek
philosopher Nicholas Damascenus), translated by Alfred of Sareshel before
1200; the De proprietatibus (an anonymous Arabic work) and De causis
(an Arabic paraphrase of Proclus, Elementatio theologica) by Gerard of
Cremona; the De differentia spiritus et animae (in fact by Costa ben Luca) by
John of Seville and by an anonymous translator. All these Arabic-Latin
translations were widely circulated, often under Aristotle's name. In ad
dition to these works, sections of the philosophical encyclopedia of
Avicenna, the Kitäb al-Shifâ, were translated into Latin in Toledo in the
second half of the twelfth century, as were works by Alkindi, Algazel,
Alfarabi, and Avencebrol (Ibn Gabirol). 7 All more or less Aristotelian
3. Published in AL 1957, vii.2.

4. AL t972-4, XXVI.1.
5. AL 1962, 111.3.
6. AL 1969, v.3.
7. For these works and their translators see Lemay 1963; D'AIvcrny 1952, 1961-72, and Van Riet
1968, 1972 (Avicenna); Nagy 1897 (Alkindi); Lohr 1965 (Algazel); Salman 1939, Langhade and
Grignaschi 1971, and Grignaschi 1972 (Alfarabi); Baeumker 1895 (Avencebrol).
4 8 Aristotle in the middle ages
in doctrine, these works made their contribution to the spread of Aristote-

lianism in the West. O n e other twelfth-century work should be men
tioned, although it was not a translation at all: this is the Liber sex
principiorum,8 a fragment of a work by an unknown twelfth-century Latin
author dealing with six of Aristotle's ten categories. Under this title it
became a regular part of the logical corpus, and it was often commented on
and accepted as containing genuine Aristotelian doctrine, although most
medieval authors recognised that it was not by Aristotle.
Thus by the end of the twelfth century the bulk of Aristotle's works had
been translated - all the logical works, all the works on natural philosophy
except the De animalibus, and part of the Ethics. They do not appear to have
been widely read, however; very few twelfth-century manuscripts sur
vive, and references to Aristotle's works are sparse. The works that were
studied most were the newly discovered or translated logical texts (the
'logica nova'), particularly the Sophistici elenchi.9
By the beginning of the thirteenth century Aristotle was obviously
gaining ground and being studied - as witness the edict issued in Paris in
I2I0 forbidding any lectures (public or private) on Aristotle's books of
natural philosophy. The ban, however, was only local and did not prevent
the work of translation from continuing. By 1220 Michael Scot had
translated three of the five treatises of the De animalibus (the Historia
animalium, De partibus animalium, and De generatione animalium). In the
1220S or 123OS (no dates are known), he accomplished the vast task of
translating Averroes' Arabic commentaries into Latin. Averroes' com
mentaries on Aristotle were of three types: short epitomes or compendia;
'middle' commentaries made up largely of paraphrase; and 'great' com
mentaries consisting of very detailed sentence-by-sentence exposition.
Michael translated the great commentaries on the Physics, De caelo, De
anima, and Metaphysics, at the same time translating the full Arabic text of
Aristotle in the form of lemmata interspersed with the sections of the
commentary. In the case of the Metaphysics the lemmata were then trans
cribed as a continuous text without the commentary, and this translation
(the 'Metaphysica nova') was widely circulated as the most complete text of
10
the Metaphysics available in Latin. (The anonymous translation men
tioned above, known as the 'Metaphysica media', although made in the
8. AL 1966,17.
9. See, for example, De Rijk 1962-7,1; Ebbesen 1973b.
10. However, it lacked all of Books XI, XIII and XIV (i.e. Κ,Μ,Ν), the beginning of I (up to 987^)
and the end of XII (from I075 b n).
twelfth century, appears to have remained unknown until the mid-

thirteenth century.) Michael also seems to be the translator of the surviving
Latin versions of Averroes' middle commentaries on the De generatione et
corruptione and Book IV of the Meteorologica and of his epitomes of the De
caelo, Parva naturalia and De animalibus. The story of Averroes in Latin can
be briefly rounded offby mentioning Hermannus Alemannus' translation
of the middle commentaries on the Nicomachean Ethics ( ι240) and Poetics
(1256), and William of Luna's translation of the middle commentaries o n
Porphyry's Isagoge, the Categories, De interpretatione, Prior Analytics, and
Posterior Analytics (thirteenth century).
T o return to translations from the Greek, early in the thirteenth century
the whole of the Ethics was translated. O f this translation, however, only
Book I, known as the 'Ethica nova',1 ' became known and circulated, and
the existence of a complete translation has been assumed from a few
surviving fragments from the other books. Between 1220 and 1230 James
of Venice's translation of Metaphysics I—IV.4 was incompletely revised;
subsequently scribes conflated the original and revised versions, so that the
texts of the revision which circulated (known as the 'Metaphysica vetus')12
contain varying proportions of revised and unrevised text. Towards the
middle of the century Robert Grosseteste translated the Nicomachean Ethics,
along with a great corpus of Greek commentary by Eustratius and others.
Grosseteste also translated at least part of De caelo along with Simplicius'
commentary (fragments survive in one manuscript), and also the pseudo-
Aristotelian De lineis indivisibilibus and De laudabilibus bonis. Between 1258
and 1266 Bartholomew of Messina, at the command of Manfred, king of
Sicily, translated a sizable group of pseudo-Aristotelian works: the
Problemata, Physionomia, De mirabilibus auscultationibus, Deprincipiis (in fact
by Aristotle's pupil Theophrastus), De signis aquarum, De mundo, Magna
moralia and De coloribus. Manfred himself may have translated the De porno,
a work describing the death of Aristotle. The De mundo was also translated
by Nicholas of Sicily.
The last major translator was William of Moerbeke. Between about
1260 and 1280 he translated anew or revised virtually the whole Aristo
telian corpus, including two works, the Politics and Poetics, that had not
been translated before. William's translations quickly established them
selves as the most popular versions, except in the case of the logical works.
11. AL 1972, xxvi.2. According to a note in one manuscript, the 'Ethica nova' was translated by
Michael Scot; see the preface to AL 1972, χχνι, pp. cxlii-cxlvii.
12. AL 1970, xxv. 1'.
Although the majority of Aristotle's works had been translated in the

twelfth century, the evidence of the manuscripts and other sources in-
dicates that they were not much studied and circulated until the thirteenth
century. From the early thirteenth century onwards numerous manuscripts
survive containing a collection of the logical works, all in Boethius'
translation except the Posterior Analytics (James of Venice), and almost
invariably with the addition of Porphyry's Isagoge and the Liber sex prin-
cipiorum. Very often two of Boethius' works, the De topicis differentiis and
the De divisione, are also included, and a typical medieval manuscript of
Aristotle's 'Organon' would thus contain the following ten works:
Porphyry's Isagoge, Categories, De interpretatione, Boethius' De divisione and
De topicis differentiis, Liber sex principiorum, Prior Analytics, Posterior
Analytics, Topics, and Sophistici elenchi.
Towards the middle of the thirteenth century a similar collection con-
taining works in natural philosophy was made, and those twelfth-century
translations that were included in it now became widely circulated. The
editors of the Aristoteles Latinus catalogue have christened this collection the
'corpus vetustius' and describe nearly one hundred manuscripts (mostly
from the thirteenth century) containing the same group of works (not
always complete, almost never in the same order, and with other non-
Aristotelian works often thrown in). A typical example of the 'corpus
vetustius' is found in a late thirteenth-century manuscript in the
Stiftsbibliothek at Admont (no. 126) which contains the following works.
Physics (James of Venice); De caelo (Gerard of Cremona); De generatione et cor-
ruptione (anonymous); De anima (James of Venice); De memoria (James of Venice);
De sensu (anonymous); De somno (anonymous); De longitudine (James of Venice);
De differentia spiritus et animae (John of Seville or anonymous); De planus (Alfred
of Sareshel); Meteorologica (Gerard of Cremona and Henricus Aristippus);
Metaphysics (James revised; Book I only); Metaphysics (Michael Scot); De causis
(Gerard of Cremona); Nicholas of Amiens, De articulisfidei.
These then, with the logic, are the Aristotelian translations (genuine and
spurious) that 'made the grade' in the thirteenth century, and in modern
terms the logical collection and the 'corpus vetustius' could be described as
the standard edition of the works of Aristotle.
It can be seen from the Admont manuscript that two versions of the
Metaphysics could appear side by side; in fact the history of the Metaphysics
is rather complicated. 13 James of Venice's version and its anonymous
13. See Diem 1967.

revisions (the 'Metaphysica vetus') covered only the first four books,
while the anonymous and nearly complete twelfth-century version (the
'Metaphysica media') seems to have remained unknown until the mid-thir-
teenth century; Michael Scot's version from the Arabic (the 'Metaphysica
nova') therefore represented the fullest available text and for this reason was
detached from its commentary and transcribed as a plain text. The lack of a
single authoritative text is reflected in the manuscripts of the 'corpus
vetustius'. About half of them have no text of the Metaphysics at all, while
the rest have sometimes the 'vetus', sometimes the 'nova', sometimes both,
sometimes a composite text; only a few have the 'media'.
In the second half of the thirteenth century William of Moerbeke's
versions were soon collected into a 'new edition' of the works of Aristotle.
The demand for Aristotle was high and William's versions represented a
more complete and in many cases obviously superior collection to the old
corpus. The new collection, labelled the 'corpus recentius' by the Aristoteles
Latinus editors, rapidly gained ascendancy in the late thirteenth century
and retained it until the Renaissance; over 170 manuscripts survive, al-
though many of these do not contain the complete corpus. O f the manu-
scripts containing a reasonably full collection the majority contain the
following sequence (or one very close to it):
Physics; De caelo; De generatione et corruptione; Meteorologica; De anima; De sensu; De

memoria; Desomno; De motu animalium; De longitudine; De iuventute; De respiratione;
De morte; Physionomia; De bonafortuna.
About half of these manuscripts also contain all or most of the following
works:
Metaphysics; De nilo; De coloribus; De planus; De progressu animalium; De porno; De

intelligentia; De causis; De proprietatibus; De lineis indivisibilibus; De mundo, Epistola
ad Alexandrum; Vita Aristotelis; De differentia spiritus et animae.
These works thus represent the new 'standard edition' of Aristotle's works
on natural philosophy; it had two forms, a basic collection containing the
first sequence, and a fuller collection incorporating the Metaphysics and
many pseudo-Aristotelian works. It is not obvious why the Metaphysics
should continue to be absent from many manuscripts of the 'corpus re-
centius', as it was from the 'corpus vetustius'; although William's translation
of the Metaphysics is often found as the sole work in a manuscript, and so
must have circulated independently. The logical collection continued to
circulate unchanged in Boethius' and James' versions, and William's new
translations or revisions of some of these works failed to challenge the

established versions.
We may finish this account of the history of the translations with a
résumé ofthose works that never gained a regular place in the collections
mentioned above. The De animalibus treatises translated by Michael Scot
were widely circulated, nearly always on their own, and continued to rival
William's version well into the fourteenth century, to judge from the
surviving manuscripts. William's version survives as a complete set in
relatively few manuscripts (40 against 79 of Michael Scot's), but his De motu
and De progressu achieved much wider circulation as part of the 'corpus
recentius'. The two incomplete versions of the Ethics, comprising the 'Ethica
nova' (Book I) and the 'Ethica vetus' (Books II and III), achieved modest
popularity in the thirteenth century, and often appear as additional
texts in manuscripts of the logical collection. They were superseded
by Grosseteste's complete version, which achieved immense popularity.
Known as the Liber ethicorum, and subjected to various revisions, some of
which may have been by William of Moerbeke, it survives in almost 300
manuscripts. It was often transcribed alone, but it also regularly appears
along with the Politics, Rhetoric, Oeconomica or Magna moralia. One or
more of these five 'ethical' treatises are often found together in manuscripts
from the late thirteenth century onwards, but not in any regular combi-
nation or sequence.
T w o major observations may be made on the basis of this short account
of the translations. The first is the overwhelming importance of trans-
lations made directly from the Greek. There is a tenacious legend that the
West learnt its Aristotle via translations from the Arabic, but the fact is that
the West turned to Arabic-Latin translations only in default of the more
intelligible Greek-Latin ones. The only translations from the Arabic to
achieve wide circulation were the De caelo, Meteorologica I—III, De anim-
alibus and Metaphysics, and all of these except the De animalibus were
quickly displaced by William of Moerbeke's versions. The legend has more
basis, however, when one considers Aristotelian doctrine in a vaguer sense.
The twelfth-century translations of Avicenna, Alfarabi, and Algazel, for
example, helped to disseminate Aristotelian doctrine, albeit in a not-very-
pure form. And of course the commentaries of Averroes in the thirteenth
century made a powerful impact on the West. Nevertheless, when the
Latin schoolmen came to writing their own commentaries, with few
exceptions they used the Greek—Latin and not the Arabic-Latin texts.
The second observation concerns the slowness with which the texts of
Aristotle came into circulation. Although much translation was done in the
twelfth century, it was not until well into the thirteenth that manuscripts
survive in large numbers. Many early manuscripts have undoubtedly
perished, so the full story will never be known, but from the surviving
evidence it seems plain that Aristotle did not become really important in
the academic world until the middle of the thirteenth century.
The translators
Biographical information about most medieval authors is scarce, and this is
particularly true of the translators. Even the famous figures such as Robert
Grosseteste and William of Moerbeke are in fact very poorly documented,
and few personal details have survived. Furthermore, for the most part the
translators cannot be linked with one another; there were no schools
of translators, 14 and the work was done by a handful of individuals.
Seventeen translators are known by name, and about fifteen more are
anonymous. Their work spans the period from about the mid-twelfth
century to 1295 with the exception of Boethius, who was of course very
much earlier.
Boethius
A Roman senator and minister under Thcodoric, Boethius 15 falls far
outside the period of this book, his translations of Aristotle being made ca.
A.D. 510-22. He intended, apparently, to translate the whole of Aristotle,
but the only works he is known to have translated are the Categories,16 De
interpretatione,11 Prior Analytics,18 Topics,19 and Sophistici elenchi20 - all of
the 'Organon' except the Posterior Analytics. He also translated Porphyry's
Isagoge,21 a short work preparatory to the study of logic which became a
regular part of the medieval 'Organon'. Boethius also wrote commentaries
on the Isagoge, Categories, and De interpretatione22 in addition to his in-
dependent works on arithmetic, music and logic.
14. The nearest approach to a school is the group of translators in Toledo in the second half of the
twelfth century; see Lemay 1963.
15. For a good general account of Boethius with extensive bibliography, see Minio-Paluello 1970b;
see also Armstrong 1967, pp. 53S—64.
16. AL 1961b, 1.1-2.
17. AL 1965, 11.1.
18. AL 1962, 111.1-2.
19. AL 1969, v. 1-2.
20. AL 1975, vi 1.
21. AL 1966, 1.6.
22. For a possible commentary on the Posterior Analytics see Ebbesen 1973a.
Boethius' authorship of the translations cannot be established with

absolute certainty, but there is strong documentary and internal evidence
(based on quotations in his commentaries) that the surviving translations of
the Isagoge, Categories, and De interpretatione are by him. This being ac-
cepted, his authorship of the other three follows from a stylistic analysis, for
all six are consistent among themselves in their method of translating, and
different from all the other translations. Furthermore, in the oldest and best
manuscript of the Latin version of the Topics there is a note attributing the
translation to Boethius.
All the Boethian translations except the Sophistici elenchi have compli-
cations in their textual history indicating that the texts were revised at some
stage. The revisions may be Boethius' own, or they may be the work of an
unknown editor, possibly working in Constantinople where Boethius'
works are known to have been transcribed (and perhaps edited) already in
the sixth century.
fames of Venice
The next translator, James of Venice,21 does not appear until the second
quarter of the twelfth century. He was long known to scholars through a
passage in Robert of Torigny's Chronicle: 'James, a cleric from Venice,
translated from the Greek into Latin several books of Aristotle and com-
mented on them, viz. the Topics, Prior and Posterior Analytics, and Elenchi,
although an earlier translation of these same books was already in ex-
istence.'24 Other documentation about James is sparse and can be summed
up as follows: he called himself a Venetian Greek and a philosopher; in 1136
he was present at a theological debate in Constantinople between Anselm
of Havelberg and the archbishop of Nicomedia; in 1148 he presented some
advice to the archbishop of Ravenna on the subject of the precedence of
Ravenna over other archbishoprics; he may have been in Bologna in the
114OS, disputing with Magister Albericus over the interpretation of the
Sophistici elenchi; his commentary (or translation of a commentary) on the
Elenchi is mentioned in a twelfth-century grammatical 'quaestio' and an
early-thirteenth-century author mentions his commentary on the Posterior
Analytics.25 In addition to Robert of Torigny two other sources men-
23. For James see Minio-Paluello 1952 and the prefaces to AL 1968a, iv and AL 1975, vr.
24. 'Iacobus clericus de Venetia transtulit de Greco in Latinum quosdam libros Aristotilis et com-
mentatus est scilicet Topica, Analiticos Priores et Posteriores, et Elencos, quamvis antiquior
translatio super eosdem libros haberetur.' Minio-Paluello 1952, p. 267.
25. Ebbesen 1977b, pp. 1-3.
tion James' translation of the Posterior Analytics: the translator Ioannes (see
below) and the author of a note in a thirteenth-century Oxford
manuscript.
Robert of Torigny's note led to considerable debate among scholars as
to whether the surviving translations of the 'logica nova' (preserved anony-
mously in the manuscripts) should be ascribed to Boethius or to James. The
debate was resolved in 1952 by L. Minio-Paluello in an important article
which not only established James as an important translator but also
demonstrated the validity of stylistic analysis as a means of distinguishing
translators. Minio-Paluello's analysis revealed the following facts: the
common versions of the Prior Analytics, Topics and Sophistici elenchi,
surviving in numerous manuscripts, are consistent in style with each other
and with the known Boethian translations of the 'logica vetus'; the Posterior
Analytics26 is in a different style from the rest of the 'logica nova' but has the
same stylistic features as a passage translated from the Greek in James'
advice to the archbishop of Ravenna; the twelfth-century translations of
the Physics, De anima, Metaphysics,21 De memoria, De longitudine, De iuven-
tute. De respiratione, De morte, and De intelligentia reveal the same stylistic
features as the Posterior Analytics (as do fragments of translations of
the Sophistici elenchi and of the more recently discovered commentaries
on the Elenchi and Posterior Analytics, purportedly by Alexander of
Aphrodisias). 28
The conclusion is inescapable: James translated the Posterior Analytics,
Sophistici elenchi, Metaphysics and several important works of natural philo-
sophy as well as most of the Parva naturalia. Moreover, recent research has
revealed more and more of his activity as a translator. Most of his trans-
lations achieved wide circulation and make him the most important of the
twelfth-century translators.
Henricus Aristippus
The opportunity and the impetus for James' translating work arose out of
contact between Italy and Constantinople in the twelfth century, and in
Constantinople with James in 1136 were two other distinguished trans-
lators, Burgundio of Pisa and Moses of Bergamo, neither of whom,
26. AL 1968a, iv. 1.

27. AL 1970, xxv.i. The text now ends abruptly at Book iv. 4 (ioo7'31), butjames possibly translated
the whole work; see Minio-Paluello 1972, pp. 98-102, AL 1976, xxv.2, p. xi.
28. Minio-Paluello 1954, 1962; AL 1975, VI. 2; Ebbesen 1972.
however, translated any Aristotle. 2 9 Some twenty years later, in 1158,

another Italian, Henricus Aristippus, 3 0 was in Constantinople on a diplo
matic mission for the King of Sicily. At that time the Sicilian court was an
important centre of translation and scholarly contact with both Arab and
31
Byzantine worlds. Henricus Aristippus was archdeacon of Catania in
1156 and became chief minister of the kingdom in 1160; in 1162, however,
he was imprisoned by the king and died in prison. His only Aristotelian
translation was of Book IV of the Meteorologica. His choice of this work
reflected his interest in natural phenomena - one witness describes him
investigating the marvels of Mount Etna. Henricus also translated Plato's
Phaedo32 and Meno.33 He began work on the Phaedo in 1156, but the date of
the Meteorologica translation is not known.
loannes
Another translator of this period is loannes, a shadowy figure. His trans
lation of the Posterior Analytics3* survives complete in only one manuscript,
discovered in 1913 by C. H. Haskins in the cathedral library at Toledo. 3 5
The translation is anonymous in this manuscript, but the name 'loannes' is
suggested by three citations in other sources: in a Paris manuscript a phrase
quoted from this translation has the rubric 'translatio loannis'; in another
manuscript fragments of this translation bear the rubric 'translatio Ιο.' ('Io.'
being a regular abbreviation for 'loannes' or 'loannis'); and Albert the Great
mentions a 'translatio loannis' which could well be this version. loannes'
translation has an engaging and interesting prologue:
Although hindered by many duties, my love for you compels me to translate the
Posterior Analytics from Greek into Latin, which task I have undertaken the more
readily as I know that the book contains many fruits of science. I am equally sure
that knowledge of it is not widespread among the Latin-speakers of our genera
tion, for Boethius' translation is not to be found complete among us, and what has
been discovered of it is obscured by corruption. James' translation, on the other
hand, is known to the masters of France, as are translations of commentaries made
by the same James, but they by their silence bear witness that James' version is
wrapped up in the shadows of obscurity, and do not dare to proclaim their
knowledge of it. Wherefore, if Latinity is able to procure any benefit from my
29. Haskins 1927, pp. 197-209.

30. See Haskins 1927, pp. 1 s 96e.
31. Haskins 1927, pp. 155-7.
32. Plato Latinus 1950,11.
33. Plato Latinus 1940,1.
34. AL 1968a, IV.2.
35. Haskins 1927, pp. 228IF.
translation, the credit for this is due to your request. For I undertook the task of
translating not for money or empty fame but to please you and to impart
something of value to Latinity. Moreover, if in any matter I shall be found to have
strayed from the path of reason, I shall not be ashamed to correct it with help from
you or other learned men.36
The references to Boethius' translation (now lost, if it ever existed), to
James of Venice's translation and to the deplorable state of affairs in France
are intriguing. Unfortunately the pusillanimous masters of France cannot
be identified; nor can loannes and his learned patron.
That loannes' translation was made before 1159 is proved by a reference
in John of Salisbury's Metahgicon which was completed in that year. The
Metalogicon is the first medieval Latin work to show a knowledge of all the
books of the O r g a n o n ', and in the course of a short account of the Posterior
Analytics }ohn of Salisbury quotes en passant a phrase from what he calls the
'new translation', which is in fact loannes'; the rest of his quotations are
from James' version. 37 Apart from Albert the Great, whose reference is
uncertain, John of Salisbury is the only medieval author who is known to
have quoted loannes' translation, and this coincidence is the basis for a
conjecture. In the Polycraticon38 John describes a banquet in Apulia in
southern Italy where he met John Belmeis, treasurer of York, w h o m he
describes as excelling in his knowledge of three languages. Could John
Belmeis and our loannes be the same person? It is perhaps more likely,
however, as Haskins suggests, that both the translator and his patron were
south-Italian or Sicilian.39
James of Venice, Henricus Aristippus and loannes are the only twelfth-
century translators of Aristotle from the Greek known by name. They do
not constitute a school, but they are definitely part of a movement. To this
same movement belong the five anonymous twelfth-century translators
mentioned above (p. 47).
36. 'Vallatum multis occupationibus me dilcctio vestra compulit ut Posteriores Analeticos Aristotelis
dc Greco in Latinum transferrcm. Quod eo affectuosius aggressus sum quod cognoscebam librum
ilium multos in sc sciencie fructus continere et certum erat noticiam eius nostris temporibus
Latinis non patere. Nam translatio Boecii apud nos intégra non invcnitur, ct id ipsum quod de ea
reperitur vitio corruptionis obfuscatur. Translationem vero lacobi obscuritatis tcnebris involvi
silentio suo peribcnt Francie magistri, qui quamvis illam translationem et commenrarios ab
eodem lacobo translatos habeant, tamen noticiam illius libri non audent profiteri. Eapropter
siquid utilitatis ex mea translationc sibi novcrit Latinitas provenire, postulationi vestre debebit
imputare. N o n enim spe lucri aut inanis gloric ad transferendum accessi, sed ut aliquid conferens
Latinitati vestre morem gererem voluntati. Ceterum si in aliquo visus fuero rationis tramitem
excessisse, vestra vel aliorum ammunition? non erubcscam emcndarc.' Haskins 1927, p. 229.
37. John ofSalisbury 1929, pp. 111-12, 170-2.
38. John ofSalisbury 1909, II, p. 271.
39. Haskins 1927, p. 235.
5° Aristotle in the middle ages
Gerard of Cremona
Translations from the Arabic constituted a second major movement. Its
most important twelfth-century representative was Gerard of Cremona,
who lived in Toledo and died there in 1187 at the age of 73. A brief eulogy
and a catalogue of Gerard's works were produced by his pupils, 40 and
among the 71 translations of Arabic texts listed are Aristotle's Posterior
Analytics,*1 De caelo, De generatione et corruptione, Meteorologica, and Physics,
and Themistius' paraphrase of the Posterior Analytics.*2 All of these trans-
lations have survived and two of them, the De caelo and Meteorologica
(Books I—III), were widely circulated as part of the 'corpus vetustius'.
Gerard's translations of the pseudo-Aristotelian De proprietatibus (or De
causis proprietatum elementorum) and of the De causis*3 were also popular.
The dates of the translations are not known.
fohn of Seville and Alfred of Sareshel

T w o other twelfth-century translators from the Arabic were John of
Seville (Iohannes Hispalensis) and Alfred of Sareshel, who were both in
Spain, John around the middle of the century, Alfred towards the end.
John of Seville 44 was active ca. 1130-40, and is credited with many
translations, among them the De differentia, dedicated to Raymond,
archbishop of Toledo, and the De regimine sanitatis (a fragment of the
Secretum secretorum).*5 Alfred 46 translated the De plant is before about 1200
and wrote a commentary on it; he also translated the fragment of Avicenna
known as the De mineralibus. He is best known, however, for his work De
motu cordis,*7 written ca. 1210 and dedicated to Alexander Neckham, and
his commentary on the Meteorologica.*8
Michael Scot
Translation from the Arabic was continued in the thirteenth century by
Michael Scot. 49 He is first heard of in the entourage of the bishop of Toledo
40. Sudhoffi9i4; Sarton 1931,11.2, pp. 338-44.

41. AL 1968a, IV.3.
42. Ed. in O'Donnell 1958.
43. Ed. in Steele 1935.
44. See Thorndike 1959.
45. See Manzalaoni 1977, p. xiv.
46. See Otte 1972.
47. Baeumker 1913, 1923.
48. Otte 1976.
49. See Haskins 1927, pp. 272-98; Thorndike 1965.
in 1215, 50 and was in Toledo in 1217 when he completed his translation of

the Arabic astronomer Al-bitrogi's On the Sphere. He appears to have been
still there when he translated Aristotle's De animalibus from the Arabic
version containing only three of the five treatises; the evidence for this is a
colophon which appears in several manuscripts: 'Here ends Aristotle's
book on animals, translated in Toledo from Arabic into Latin by the
master Michael.'
In October 1220 Michael was in Bologna; this is known from a mem-
orandum, signed and dated, which he wrote there and inserted into a copy
of his translation of the De animalibus. From 1224 to 1227 there is evidence
that he was under the patronage of popes Honorius III and Gregory IX,
and from about 1227 he was in the service of King Frederick II of Sicily,
where he was court astrologer. He died ca. 1236.
Michael's great achievement was his translation of many of the Arabic
commentaries of Averroes: the great commentaries on the De caelo, De
anima,51 Metaphysics, and Physics, along with complete translations of the
texts; the middle commentaries on the De generatione et corruptione52 and
Book IV of the Meteorologica; and the epitomes of the De caelo, Parva
naturalia53 and De animalibus. Of these translations only the De caelo is
incontestably by Michael, for Michael's dedication of the work to Stephen
of Provins is found in several of the manuscripts. That Michael also
translated the other commentaries is a reasonable inference based on
stylistic analysis and general likelihood. Further analysis, however, is
needed before it can be accepted with complete confidence. The dates of
these translations are not known, but scholars assume they were made in
the 1220S and 1230s, probably at Frederick's court.
Hermannus Alemannus
Four minor thirteenth-century translators from the Arabic are Hermannus
Alemannus, Philip of Tripoli, William of Luna and Petrus Gallegus.
Hermannus ('Herman the German') 5 4 worked in Toledo around the
middle of the thirteenth century (he was there in 1240 and 1256) and is
almost certainly to be identified with the Hermannus who was bishop of
Astorga in Léon from 1266 until his death in 1272. His translations have
50. Pvivero Recio 1951. I am indebted to Mlle M.-T. d'Alverny for this reference.
51. Ed. Crawford 1953.
52. Ed. Fobes and Kurland 1956.
53. Ed. Shields 1949.
54. Sec Luquct 1901.
6ο Aristotle in the middle ages
been identified from prologues and colophons in the manuscripts, three of

which are dated. They are the Rhetoric, comprising the almost complete
text of Aristotle interspersed with portions of Averroes' middle com
mentary and short fragments from Avicenna and Alfarabi; the intro
55
ductory section of Alfarabi's commentary on the Rhetoric; Averroes'
middle commentary on the Ethics, the 'Liber Nicomachie' (Toledo, 1240);
an Arabic epitome of the Ethics known as the Summa Alexandrinorum (1243
or 1244); and the middle commentary on the Poetics (Toledo, 1256), S6 this
last being known as the 'Poetria'.
Philip of Tripoli
Philip of Tripoli, 5 1 a cleric in the Crusader kingdom of Syria, found at
Antioch and translated into Latin an Arabic text of the Secretum secre-
torum,58 a work on occult science and kingship allegedly written by
Aristotle for Alexander the Great. Philip of Tripoli is probably identical
with the Philip whose ecclesiastical career in Syria between 1227 and 1259
can partly be traced in papal and other registers. The translation was
probably made ca. 1243 and became very popular, some 350 manuscripts
surviving.
William of Luna
'Here ends the work of Averroes on the Categories of Aristotle translated by
William of Luna at Naples.' This colophon in a fourteenth-century manu
script and a similar colophon to Averroes' epitome of the Isagoge in the
same manuscript constitute our sole evidence of William of Luna (a town
in Spain) as a translator of Averroes. That he was also responsible for
translating the epitomes of the De interpretatione and the Prior and Posterior
Analytics in the same manuscripts is a reasonable inference. Nothing else is
known about him except that he may also have translated a work of
algebra. 5 9
Petrus Gallegus
60
Petrus Gallegus, a Spanish Franciscan who was a confidant of Alfonso
X and became Bishop of Cartagena in Spain 1250-67, translated Averroes'
55- Boggess 1971.

56. AL 1968, XXXIII.2.
57. See Haskins 1927, pp. 137-40.
58. Ed. Steele 1920. See also Manzalaoni 1977, pp. xvff.
59. Thorndike 1931, 11.2, p. 563.
60. See Pelzer 1924.
Aristoteles latinus 6ι
epitome on the De partibus animalium. This version has survived in a

fragmentary state in one manuscript. His only other known work is a
translation of a similar compendium on household management
(Oeconomica) which, however, was not Aristotelian.
Robert Grosseteste
After the spate of translators from the Greek in the twelfth century, the first
name to figure in the thirteenth century is that of Robert Grosseteste, 6 '
first Chancellor of Oxford University, bishop of Lincoln from 1235 until
his death in 1253, and a major political, ecclesiastical, scientific, and philo
sophical figure as well as a translator of Aristotelian and other works.
Documentary evidence about Grosseteste's scholarly career and the chro
nology of his writings is almost entirely lacking. R o g e r Bacon says that he
learnt Greek late in life, and scholars have assumed that his translating work
was undertaken during his episcopate or at least after about 1230. He
procured several Greek-speaking assistants, one of w h o m was Nicholas of
Sicily (see below), and on at least one occasion he sent agents to Greece to
acquire manuscripts. 6 2
Grosseteste's most important Aristotelian translation was of the
Nicomachean Ethics.63 With characteristic thoroughness he also translated a
large corpus of Greek commentaries by Eustratius and others, 6 4 to which
he added numerous notes of his own explaining Greek terms and points of
grammar. The translation of the Ethics, which makes use of the previously
existing translations, is thought to have been made around 1246—7, and
became in its original or in a revised form the standard version in the
Middle Ages. Grosseteste also translated the first two books of the De
caelo65 along with Simplicius' commentary, and two pseudo-Aristotelian
works, the De lineis indivisibilibus and De laudabilibus bonis (or De virtute) as
well as several important non-Aristotelian works including the works of
66
Pseudo-Dionysius. The De lineis indivisibilibus was circulated widely as
part of the 'corpus vetustius' but the De laudabilibus bonis was not widely
known and the De caelo remained unknown and survives incomplete in
only one manuscript. A small fragment of Simplicius on the Physics found
in an Oxford manuscript may have been translated by Grosseteste.
61. On Grosseteste and his works see Baur 1912, 1917; Franccschini 1933; Russell 1933; Thomson
1940; Callus 1955a.
62. Callus 1955a, p. 40.
63. AL 1972-4, XXV1.3A.
64. Ed. Mcrckcn 1973.
65. Allan 1950.
66. Sec Callus 1955a, pp. 44ff.
Nicholas of Sicily
Master Nicholas the Greek 6 7 was a member of the household of Robert
Grosseteste and was no doubt one of the 'adiutores' (helpers) who assisted
him in his translating. He came from Sicily, and his name first appears in
ecclesiastical documents of Grosseteste's episcopacy in 1237. He was a
canon by 1246, and in 1256 he was sent to R o m e to try to procure
Grosseteste's canonisation. H e died in 1279. Only one translation by
Nicholas is known: that of the pseudo-Aristotelian De mundo.68
Bartholomew of Messina
Another Sicilian was Bartholomew of Messina. His translation of six
pseudo-Aristotelian treatises - Problemata, Physionomia, De mirabilibus aus-
cultationibus, De principiis, De signis, and Magna moralia - is testified by a
series of rubrics, almost all found in a single Padua manuscript containing a
collection of these and other works. A typical rubric reads as follows: 'Here
begins the book by Aristotle on Physiognomy, translated from Greek into
Latin by master Bartholomew of Messina, in the court and at the command
of the most illustrious Manfred, most blessed king of Sicily and lover of
science.' 6 9 From these rubrics we know that Bartholomew had some
position at Manfred's court; the translations must have been made between
1258 and 1266, the dates of Manfred's reign. Bartholomew also translated
from the Greek a veterinary treatise by Hierocles, 7 0 and two further
pseudo-Aristotelian works, the De mundo71 and De coloribus,12 his author
ship of the last two translations being established by stylistic evidence.
Nothing else is known about Bartholomew.
According to some sources, Bartholomew's patron, Manfred himself,
translated the De porno. Scholars tend to doubt this, and it seems more likely
that he was merely responsible for having it translated.
William of Moerbeke
The second half of the thirteenth century is dominated by the Dominican
William of Moerbeke, 7 3 the most famous and most prolific of the medieval
67. See Callus 1955a, p. 229; Russell 1933.
68. AL 1965b, xi.2.
69. 'Incipit liber physiognomonomie liir] Aris. translatus de Greco in Latinum a magistro Bartho-
[lomeo] de Messana in curia illustrissimi Manfredi serenissimi regis Sicilie scientie amatoris de
mandato suo. MS. Padua Ant. XVII.370, f. 72'.
70. Haskins 1927, p. 269.
71. AL 1965b, XI. l.
72. AL 1965b, χι, p. xvii.
73. See Grabmann 1946; an excellent short account is in Minio-Paluello 1974, with a good biblio
graphy.
translators of Aristotle, said to be a friend and collaborator of Thomas

Aquinas. Yet very little is known of his life. He was born ca. 1215 or later in
the village of Moerbeke, now in Belgium, and probably entered the
Dominican convent at Louvain as a young man. He is known to have
gone to Greece, for a manuscript colophon records that the translation of
Alexander's commentary on the Meteorologica was completed at Nicea in
April 1260, and a similar colophon records the completion of the De
animalibus at Thebes in December of the same year; William was pre
sumably a member of the Dominican convent established at Thebes at least
since 1253. By 1271 William was at the papal court at Viterbo as chaplain
and confessor to the pope, a position he had probably held for several years,
for two of his translations were made in Viterbo in 1267 and 1268, as is
shown by manuscript colophons. He remained at the papal court until
1278, when he was made Archbishop of Corinth in Greece. He died there
in 1286. A later medieval tradition records William as translator of'all the
books of Aristotle on natural and moral philosophy' at the request of his
friend Thomas Aquinas. However, there is no contemporary documen
tation of either friendship or collaboration, and William's relations with
Thomas are a matter of conjecture.
The traditional ascription to William of all the 'new translations' dating
from the second half of the thirteenth century is supported by colophons in
several manuscripts. Thus the translations of the De animalibus, Metaphysics,
Meteorologica, Politics, Rhetoric, and several of the Greek commentaries are
known definitely to be by William. These and other colophons also give a
place and/or a precise date for some of the translations, as noted in the table.
Stylistic analysis shows that the other 'new translations' must be written by
the same author, and modern research thus confirms the medieval
tradition.
William was the first to translate into Latin the De motu animalium, De
1 15
progressu animalium, Politics, * and Poetics and he made new translations
16 11
of the Categories, De interpretatione, De caelo, Meteorologica, Rhetoric, and
the rest of the De animalibus.18 He also revised the Posterior Analytics,19
60
Sophistici elenchi, Physics, De anima, Parva naturalia, and Metaphy-
74. Ed. Susemihl 1872. William's translation is an important witness to the Greek text. What is
apparently an early draft by William of Book I and part of Book II is in AL 1961, XXIX. I.
75. AL 1968b, xxxiii. 1 (2nd cdn., cd. L. Minio-Paluello, 1968).
76. AL 1961b, ιj.
77. AL 1965a, 11.2.
78. The De generatione animalium is in AL 1966b, xvii.2.
79. AL 1968a, IV.4.
80. AL 1975, V13. The apparent connections with a twelfth-century text noted by Ebbesen 1972,
pp. 16-18, are not significant.
sics,81 and possibly also the De generatione et corruptione and Ethics.82 He

appears to have translated one pseudo-Aristotelian work, the De coloribus.
Of the Greek commentators he translated Simplicius on the Categories83
and De caelo, Ammonius on the Perihermenias,8* Alexander on the Meteoro-
logica85 and De sensu, Philoponus on the De anima (Book III) 86 and
Themistius on the De anima.81 His most important non-Aristotelian trans-
lation was of the Elementatio theologica of Proclus, 88 an Arabic-Latin
version of which had long been circulating under the title De causis and was
sometimes ascribed to Aristotle.
With the exception of the logical texts, where the existing translations
were firmly established, all of William's revisions and new translations
became the standard texts of Aristotle up to and beyond the Renaissance.
Durandus de Alvernia
The series of medieval translations of Aristotle ends with a revised version
of the anonymous translation of the Oeconomica. Several manuscripts have
a colophon recording that this translation was made at Anagni (Italy) in
1295 by 'an archbishop and a bishop from Greece and Durandus de
Alvernia, a Latin-speaker, proctor of the University of Paris, at that time
in the R o m a n Curia'. 8 9 Apart from his authorship of three Aristotelian
commentaries, virtually nothing else is known of Durandus de Alvernia; 90
the Greek archbishop and bishop are also unknown. Durandus is the last of
the medieval translators of Aristotle; the next wave of 'humanist' trans-
lations begins in the fifteenth century. 91
The language and method of the translators

All the medieval translators of Aristotle (and medieval translators in
general) adopted a literal style of translating, in which the Greek was
81. William was the first to translate Book XI, which was omitted in both the anonymous and
Michael Scot's versions.
82. AL 1972-4, XXVI.3B.
83. Ed. Pattin 1971. I97S-
84. Ed. Vcrbeke 1961.
85. Ed. Smet 1968.
86. Ed. Vcrbckc 1966.
87 Ed. Verbeke 1957·
88. Ed. Vansteenkiste 1951.
89. '...per unum archiepiscopum et unum episcopum de Grecia, et magistrum Durandum de
Alvernia latinum procuratorcm Universitatis Parisicnsis, tunc temporis in curia Romana ...'
90. See Lohr 1967-74, Authors A - F, p. 402.
91. For a summary of fifteenth-century translators and translations see Minio-Paluello 1970a,
pp. 273—4; Garin 1951. For printed editions of the fifteenth and sixteenth centuries see the
Cesatntkatalog der Wiegendrucke (1925fr.) and Cranz 1971 respectively; also Risse 1965.
rendered into Latin more or less word for word, with the Greek word
order being retained where possible, and with conscious consistency in
vocabulary. Some short examples taken at random from four of the most
important translators will give the flavour of the medieval translations and
show how closely they all adhere to the literal method.
1. Boethius, Topi« 134*5-7:

Deinde destruentem quidcm si quod naturaliter inest
Έπειτ' άνασκευάζοντα μεν ει τό φύσει υπάρχον
volens assignare hoc modo ponit secundum locutionem,
βουλόμενος άποδοΟναι τούτον τόν τρόπον τίθησι τη λέξει,
ut quod semper inest signified.92
ώστε ιό άεί υπάρχον σημαίνειν.
ι. James of Venice, Metaphysics 989*26-30:

Et omnino alterationem destrui necesse est sic
"όλως τε άλλοίωσιν άναιρεΐσθαι ανάγκη τοις οίίτω
dicentibus; non enim ex calido frigidum neque ex
λέγουσιν ού γαρ εκ θερμού ψυχΡ° ν ουδέ έκ
frigido calidum erit. Quid enim ipsa utique patietur
ψυχρού θερμόν δσται. τϊ γαρ αυτά αν πάσχοι
contraria, et que erit una natura que fit ignis et
τάναντία, και τις εϊη αν μία φύσις ή γιγνομένη πυρ και
93
aqua, quod ille non dicit.
ϋδωρ, ο εκείνος ou φησιν.
3 Grosseteste, Ethics 1127*13""16:

>
Circa eadem autem fere est et iaetantie
περί τα αυτά δε σχεδόν έστι και ή της αλαζονείας
medietas. Innominata autem et ipsa. Non malum autem
μεσάτης· ανώνυμος δέ και αϋτη. ού χείρον δέ
et tales pertransire. Magis enim utique sciemus
και τάς τοιαύτας έπελθεΐν μάλλον τε γαρ αν εϊδείημεν
92. 'Next, for destructive purposes, see if, while intending to render an attribute that naturally
belongs, he states it in his language in such a way as to indicate one that invariably belongs.'
(Oxford translation.)
93. 'And in general, change of quality is necessarily done away with for those who speak thus, for on
their view cold will not come from hot nor hot from cold. For if it did there would be something
that accepted the contraries themselves, and there would be some one entity that became fire and
water, which Empcdoclcs denies.' (Oxford translation.)
que circa mores singulos pertranseuntes.94

τα περί τό ήθος καθ' δκαστον διελθόντες.
4 William of Moerbeke, Politics i33i b io-22:
Sed immorari nunc diligenter exquirentes et dicentes
Άλλα τό διατρίβειν νυν άκριβολογουμένους και λέγοντας
de talibus inutile est. Non enim difficile est
περί των τοιούτων άργόν εστίν où γαρ χαλεπόν έστι
talia intelligere, sed facere magis. Dicere quidem
τα τοιαύτα νοήσαι άλλα ποιήσαι μάλλον το μεν γαρ
enim voti opus est, evenire autem fortunae.95
λέγειν ευχής έργον εστί, τό δέ συμβήναι τύχης.
This method of translating makes for somewhat stilted and occasionally

very curious Latin, but one should not conclude that the translators were
therefore crude and unsophisticated, or that their knowledge of either or
both languages was inadequate. The method of translating was born out of
respect for authority; as with sacred texts, so with Aristotle it was impor
tant to preserve the actual words of an authoritative work. Hence the ideal
was to present to the reader Aristotle's actual words, put together in just the
way Aristotle had put them together, with minimum 'interference' from
the translator. In this respect the translator's aim matched the reader's
expectation; for as the medieval scholars read and commented on the
works they did not worry much about authenticity and the problems of
expressing thoughts in different languages. They simply assumed that they
were dealing with Aristotle's actual words - an assumption that was largely
justified, even down to fine detail.
The literal method of translating was made possible by the basic simi
larity of the two languages, and it served its purpose well enough. O n the
whole the Latin of the translations is readily intelligible; at any rate it is no
less intelligible than the original Greek, for awkward and odd passages in
Latin often do no more than reflect faithfully the awkwardness and
oddness of the original Greek. Much of Aristotle is difficult to read in
Greek, and one should be careful not to blame the translators for difficulties
they did not create.
94. "The mean opposed to boastfulness is found in almost the same sphere; and this also is without a
name. It will be no bad plan to describe these states as well; for we shall know the facts about
character better if we go through them in detail...' (Oxford translation.)
95. "But it would be a waste of time for us to linger over details like these. The difficulty is not in
imagining but in carrying them out. We may talk about them as much as we like, but the
execution of them will depend on fortune.' (Oxford translation.)
In practice there are various reasons why the ideal of perfect literalness
was never fully achieved. In the first place there are features of Greek
grammar and idiom that cannot be translated literally into Latin. Take for
example the phrase έν τω είναι ('in being'). The Greek uses a definite article
(absent in Latin) and combines a preposition with an infinitive verb (not
Latin idiom). O n e Latin translator might translate this boldly by 'in esse',
ignoring Latin grammar and the Greek article,while another might respect
Latin grammar and express the article in a laborious circumlocution, 'in eo
quod est esse'. Secondly, some Greek words may have many Latin equi
valents. Thus the Greek λόγος ('word', 'account', 'definition') may be
expressed by different translators as 'ratio', 'oratio', 'definitio', 'ratiocinatio',
'sermo', 'disputatio', 'argumentatio'', 'verbum', or 'proportio'. In this case the
different renderings reflect both systematic preferences by different trans
lators and conscious variation to accommodate different meanings of the
word. Thirdly, no translator is perfectly consistent with himself, and all
make mistakes and have their idiosyncrasies.
Idiosyncrasies among the translations are of particular interest to scholars
since they provide a means of distinguishing translators. 9 6 There are many
small words (particles) in Greek, which can be translated by more than one
Latin word. For example δέ ('but') can be rendered by 'autem', 'vero' or
'sed'; γάρ ('for') by 'nam', 'namque', or 'enim'; οΰτως ('thus') by 'sic', 'itaque',
or 'ita'. If one analyses the manner in which all these particles are translated,
distinct patterns emerge, and it becomes plain that each translator adheres
to his own preferred renderings of the particles. Each combination of
preferred renderings is an unconscious signature by the author. Thus the
translations now assigned to James of Venice were identified by a distinc
tive set of particle translations (including a strong preference for 'autem',
'enim' and 'sic') which were found in those translations and no others.
Some similar clear-cut differences among the translations emerge also in
choice of general vocabulary, in ways of handling grammatical and
idiomatic differences between the two languages, and in some general
features. Boethius, for example, exercised some ingenuity in devising Latin
versions of arguments based on ambiguities or plays on words in Greek;
James of Venice had a habit of leaving some untranslated words in Greek
letters in his translations, to the great confusion of subsequent scribes who
tried to copy them; Grosseteste was extremely literal, and helped the reader
by adding explanatory notes, particularly where the Latin might be
96. See esp. Minio-Paluello 1952.

ambiguous; William of Moerbeke was also very literal, sometimes to the

point of simply transcribing Greek words in Roman letters.
Translators from the Arabic adopted the same literal method as their
Graeco-Latin counterparts, but the result was very different. The Arabic
translations themselves were made not directly from the Greek but
through intermediate Syriac versions, and so the translators into Latin were
working at two removes from the original through the distorting medium
of Semitic languages that did not lend themselves readily to literal trans-
lation either out of Greek or into Latin. The result is that the Arabic-Latin
translations of Aristotle are much more difficult to read and understand
than the Greek-Latin ones, a fact which explains why the latter were
preferred when they were available.
Medieval manuscripts were of course transcribed by hand, a process that
introduces a complicating factor into any assessment of the language of the
translations. All the texts, and particularly the most popular ones, were
subject to unintended variation at the hands of scribes, who could and did
make numerous mistakes as they copied them. Errors accumulated as new
copies were made, and most surviving manuscripts of Aristotle are very
faulty —words and whole sentences are sometimes omitted, transposed, or
garbled, similar-sounding words are interchanged, conjectures and 'im-
provements' are deliberately introduced, glosses are accidentally incorpo-
rated into the text, and many absurd and seemingly random errors are
made simply through carelessness. Although numerous, most such mis-
takes are minor ones which would not have seriously impeded the reader;
indeed it is surprising how often the general sense of a passage will remain
clear in spite of serious errors in the text. Nevertheless, one should always
bear in mind that the texts used by medieval readers were very variable,
and that textual error can sometimes be a source of real misunder-
standing.
For the historian of philosophy the chief interest of the language of the
translations lies in its contribution to the philosophical vocabulary of
medieval and later times. Aristotle became 'the Philosopher' and the study
of philosophy became practically synonymous with the study of Aristotle,
and in this way the Latin philosophical and technical terms chosen by the
translators entered into a living philosophical tradition and became part of
the language of philosophical discussion. Detailed analysis of the voca-
bulary of the translations is now being made possible as the full indexes of
the Aristoteles Latinus volumes are published.
The reception of the translations91

The evidence on which our knowledge of the reception of Aristotle's
works is based is of several kinds. First, there are the manuscripts them-
selves; second, there are the glosses in the manuscripts, which are direct
evidence of their being read and studied; third, there are university docu-
ments, at first banning Aristotle's works but later prescribing them; fourth,
there are hundreds of surviving commentaries, quaestiones, and other aids
to study such as compendia and collections of extracts; 98 fifth, there are the
numerous references to Aristotle by medieval authors.
Using all this evidence-manuscripts, university decrees, commentaries,
quotations - we can sketch an outline of Aristotle's reception by the Latin
West which will complement the history of the translations already given.
What emerges very clearly from the evidence is the slowness of the
assimilation, even of those works, such as the logic, which never came
under doctrinal suspicion. Up to about the middle of the thirteenth century
the surviving material with which to document Aristotle's progress is
somewhat meagre; after that point it becomes a flood.
For the twelfth century," evidence of interest in and study of the 'logica
vetus' and the Sophistici elenchi100 is relatively abundant. The Prior
Analytics,101 Posterior Analytics,102 and Topics,103 on the other hand,
appear to have been known to a few scholars, but not to have been widely
studied; no commentaries on these works have survived from the twelfth
century, and references to them by authors of the period are few. In
particular the Posterior Analytics was regarded as difficult - the comment by
the translator loannes quoted above is paralleled by John ofSalisbury, who
complained that the work has 'as many stumbling-blocks as it has chapters'
and reported that people were inclined to blame the translator for this. 1 0 4
As for the rest of the works of Aristotle (i.e., the non-logical works, known
as the 'new Aristotle'), a few twelfth-century manuscripts (some with
glosses) and a handful of references are our only surviving witness to their
97. For general accounts see Grabmann 1939,1950; De Wulf 1934—6; VanSteenbergh.cn 1966; Callus
1943 (for Oxford); and many detailed studies in Grabmann 1926-56.
98. For compendia etc. sec Grabmann 1939, pp. 54-104.
99. See especially Grabmann 1950.
100. De Rijk 1962-7, 1; Ebbcscn 1972, 1973b, 1976b.
101. AL 1962, ill. pp. 429-42.
102. Dod 1970, pp. 59-80.
103. Grabmann 1938b; Grccn-Pedersen 1977a, pp. 41-2.
104. John ofSalisbury 1929, p. 171.
study. 105 Minio-Paluello has persuasively argued that northern France,

and in particular the abbey of Mont-Saint-Michel, was an important
centre for the dissemination of the new Aristotle in the third quarter of the
twelfth century. 1 0 6
The oldest surviving commentaries on the new Aristotle are those by
Alfred of Sareshel on the Meteorologica and De plantis, dating from around
the turn of the thirteenth century, and probably written at Oxford. Alfred
also wrote a commenEary on the De generatione et corruptione, and an old
catalogue of Beauvais Cathedral Library listed commentaries by him on
the De caelo, De anima, De somno and De morte as well. This collection long
antedates any other known commentaries on the new Aristotle. 107
Considerable knowledge of the new Aristotle is shown in the Tractatus de
anima by John Blund, probably dating also from the turn of the century,
and probably also from Oxford. 1 0 8
Aristotle in Paris and Oxford before 1210
It seems likely that the 'logica nova' and the new Aristotle were being
lectured on in Oxford and Paris in the first decade of the thirteenth
century. 1 0 9 Roger Bacon says that Saint Edmund of Canterbury was the
first to lecture on the Sophistici elenchi at Oxford, and that a 'magister
Hugo' (otherwise unknown) was the first to lecture there on the Posterior
Analytics; both these courses probably took place ca. 1200-10. 1 1 0 A list of
textbooks compiled by Alexander Neckham probably ca. 1200-10 in-
cludes all the 'logica nova' among the books on which the student should
attend lectures, and goes on to recommend that he look at (inspiciat) the
Metaphysics, De generatione et corruptione, and De anima.111 Alexander was
familiar with Paris and Oxford, and one can plausibly suggest that the three
last-mentioned works as well as the 'logica nova' had some place in the
curriculum at either or both universities. At any rate, we can be sure that
some at least of the works on natural philosophy were being lectured on in
Paris by 1210, for otherwise a ban on them would have had no point.
105. Grabmann 1950, pp. 159-62.

106. Minio-Paluello 19S2, pp. 291-5.
107. See Otte 1972, 1976.
108. Callus 1943, pp. 241-51; Callus and Hunt 1970, p. xi.
109. See Callus 1943; Grabmann 1950, p. 162.
no. Callus 1943, pp. 238-41; Ebbesen 1977b, pp. 4-9; Lawrence i960, pp. H2ff. It can be argued that
the dates should be as late as r<j. 1230 - an extreme example of the general uncertainty over the
correct dating of many works of this period.
i n . Haskins 1927, pp. 356-76.
The proscription of 1210

The proscription of Aristotle's works in Paris in 1210 is probably the most
famous event in the history of Aristotle's reception by the West. · ! 2 In that
year a council of the ecclesiastical province of Sens, presided over by the
archbishop, issued an edict condemning (among others) the heretics
Amaury of Bene and David of Dinant and forbidding any lectures (public
or private) in Paris on Aristotle's works of natural philosophy. 1 1 3 The ban
appears to have been in part a move by conservative theologians against
the turbulent spirits of the Arts Faculty, where Aristotle was lectured upon,
and behind it was the fear that Aristotle's natural philosophy was a threat to
the Christian faith. O f the two men condemned, David of Dinant is
known to have been an Aristotelian, and Albert the Great attacks his
materialistic and pantheistic philosophy. 114 The ban was renewed in 1215
by Robert of Courçon, the papal legate charged with reorganising studies
in Paris. Of Aristotle's works Robert permitted all the logic and the Ethics
to be lectured on, but 'the books of Aristotle on metaphysics and natural
philosophy may not be lectured on, nor may any commentaries or sum-
maries of them'. 1 ' 5 (The 'commentaries and summaries', mentioned here
may be the works of Avicenna and Alfarabi translated from the Arabic.) ' 1 6
The ban remained in force for over 20 years, and seems to have been
effective, for in 1229 the newly founded University of Toulouse, boasting
about 'academic freedom' (libertas scholastica), issued a prospectus advertis-
ing lectures on 'all the books of natural philosophy proscribed in Paris'. ' i 7
In 1231 the pope reaffirmed the ban 'until such time as the books shall be
examined and purged of all errors', for which purpose he set up a three-
man commission. 118 What happened to the commission is not known, and
eventually the ban was forgotten. 119
From 1210 to 1250

From around 1210 to the middle of the century the evidence regarding the
study of Aristotle continues to be very thin. In Paris, for example, nothing
is known about any lectures before about 1230, not even in logic, a subject
112. See Grabmann 1941; Van Steenberghen 1966, pp. 88-110.
113. Denifle and Châtelain 1889-97 (CUP), 1, p. 70.
114. Van Steenberghen 1966, p. 91.
115. CUP, 1, pp. 78-9·
116. The reference cannot be to Averroes, whose works did not come into circulation before the
1230S. See De Vaux 1933. See also Hunt 1977, pp. 195, 199.
117. CUP, I, p. iji.
118. CUP, 1, pp. 138, 143-4.
119. For the documents apparently imposing the ban at Toulouse in 1245 and renewing it in Paris as
late as 1263 (CUP, 1, pp. 185, 427) see Van Steenberghen 1966, pp. 109, 146-8.
that was not banned. 1 2 0 An interesting document, probably dating from

the 1230S, throws some light on the teaching in the Arts Faculty in this
period. 121 It is a 'crib' of quaestiones that examination candidates would be
likely to encounter in disputations, and it gives overwhelming preponder-
ance to logic (60 columns of the manuscript) and grammar (23 columns).
The Ethics122 is definitely on the curriculum, with five columns devoted to
it, but the Metaphysics and natural philosophy are summarily treated in one
and a half columns. It is thus clear that the promulgations of 1215 still held
in the 1230s. However, the new Aristotle was certainly being read, for it is
quoted in works by masters in the Theological Faculty: William of
Auxerre in his Summa aurea (ca. 1215—20) quotes the Ethics, De anima, and
Physics; Philip the Chancellor in his Summa de bono (ca. 1230-6) quotes these
works and the De animalibus, Metaphysics, De caelo, and De generatione et
corruptione; William of Auvergne (writing ca. 1231-6) and Roland of
Cremona (ca. 1229—32) embrace a similar range of quotation. 1 2 3
In Oxford (presumably) Robert Grosseteste wrote the earliest surviving
medieval Latin commentary on the Posterior Analytics, a work of great
intelligence and maturity; guesses about its date range from before 1209 to
the late 1220s. 124 Grosseteste also wrote glosses on the Sophistici elenchi and
notes on and summaries of the Physics and the Ethics,125 the last of these
presumably dating from the time he was translating the Ethics (probably
1247—8). The date of the work on the Physics and Elenchi is not known, but
one may reasonably assume that, like the commentary on the Posterior
Analytics, it reflects his teaching career at Oxford (i.e., before 1235). One
may equally assume that Grosseteste lectured on Aristotle, although there is
no evidence to prove it. In fact, as at Paris, there is very little evidence
surviving at all about the study of Aristotle at Oxford before the mid-
thirteenth century, 1 2 6 and the same is true ofother centres of learning. One
should remember that the total number of people in the whole of Europe
who had an active interest in the new Aristotle was extremely small - a
handful of masters and students in a few places.
120. Grabmann 1950, pp. 143-7. f ° r t n e more general history of logic in the twelfth and thirteenth
centuries see Grabmann 1937, 1938a; for masters in the Arts Faculty see De Wulf 1936, pp. 85-6;
Van Stcenberghcn 1966, pp. 132-51.
121. Grabmann 1934a; De Wulf 1936, pp. 8 3 - 5 ; Van Steenberghen 1966, pp. I [ 8 - 3 2 .
122. Sec also Grabmann 1940b.
123. Callus 1943, p. 231; for masters in the Theological Faculty sec De Wulf 1936, pp. 7 0 - 8 3 ; Van
Steenberghen 1966, pp. 151-70.
124. See Rossi 1975.
125. Thomson 1940, pp. 8 1 - 2 (the authority o f o t h e r works listed by Thomson is less assured);
Dales 1963.
126. For a review of the k n o w n personalities see Callus 1943; Van Steenberghen 1066, pp. 171— 6.
The mid-thirteenth-centuryfloweringof the study of Aristotle

It may be partly chance that has preserved so little evidence for the study of
Aristotle in the first half of the thirteenth century. There can be no doubt,
however, of the burgeoning of studies around the middle of the century.
This is shown by the greatly increased number of surviving manuscripts,
by a plethora of commentaries, and by the statutes of the University of
Paris. To take the last first: in 1252 the statutes of 'the English Nation'
prescribe only the De anima in addition to the logic. ' 2 1 Three years later the
statutes of the Arts Faculty (the oldest to survive) prescribe practically the
whole corpus (and several spurious works): Ethics, Physics, Metaphysics, De
animalibus, De caelo, Meteorologica (Books I and IV), De anima, De gene
ratione et corruptione, De causis, De sensu, De somno, De planus, De memoria,
De differentia, De morte, and of course the logic. 1 2 8 Thus by 1255 Aristotle
was firmly established at Paris, as he was indeed at Oxford, where docum
entation about commentaries and lectures becomes abundant as early as the
1240S.129
The importance of the glosses

It is of course no coincidence that manuscripts containing the 'corpus
vetustius' began to be written in large numbers around this time; they were
written to meet a demand. Many of these manuscripts (and of course others
written both earlier and later) are heavily glossed, and thus provide a rich
store of material about the study of Aristotle which has so far been little
explored by scholars. Often manuscripts were prepared with wide margins
ruled specially for writing glosses, and glosses were copied from one manu
script to another as an aid to study, with increasing elaboration as successive
masters and students transcribed sets of glosses and then added their own
notes, which would then themselves be incorporated in the next round. In
this way some sets of glosses eventually became so elaborate that they
amounted to full-scale commentaries.
Glosses are interesting for several reasons. The very simple sets of glosses,
mostly giving elementary explanations of individual words and phrases,
represent the first halting attempts at interpretations by readers new to
difficult texts, and enable us to see the beginnings of the assimilation of
Aristotle. In the increasing elaboration of the glosses one can trace the
growing self-confidence and sophistication of subsequent generations of
127. CUP, 1, p. 228.

128. CUP, 1, p. 278.
129. Callus 1943, pp. 255ΓΓ; De Wulf 1936, pp. 878".; Van Steenberghen 1966, pp. 176-81.
readers. Many of the glosses are probably the work of humble masters and
scholars and reflect the level of understanding reached by ordinary stud-
ents. Many probably also represent the contents of lecture courses. The
technique of the lecturer was to read the text out and expound it word for
word as he went along, and one can well imagine many sets of glosses
originating in lecturers' or students' notes.130
Glosses reflect the humble side of Aristotelian study, which has its
interest and importance; of far greater importance, however, is the less
humble side, represented by those teachers who felt that their contribution
to the elucidation of Aristotle was sufficiently valuable to deserve publi-
cation, and whose names and works now begin to survive in large num-
bers. Aristotle has become established at the centre of the philosophical
curriculum, and from the mid-thirteenth century onwards the roll-call of
important Aristotelian commentators is identical with that of the impor-
tant philosophers — to teach philosophy was to teach Aristotle. Albert the
Great, Thomas Aquinas, Giles of Rome, Walter Burley, William Ockham
and the other major figures all wrote commentaries or quaestiones on
Aristotle, and for every major figure there are dozens of minor ones.
Hundreds of commentators and surviving commentaries are known, pro-
viding a massive body of evidence with which to document Aristotle's
domination of later medieval philosophy.131
A Table of Medieval Latin Translations of Aristotle's Works and of Greek

and Arabic Commentaries
No. of
Work Translator Date surviving MSS.132
Categories Boethius ca. 510-22 306

William of Moerbeke 1266 10
Simplicius William of Moerbeke 1266 10
Averroes (middle William of Luna 13 th century 4*
commentary)
De interpretatione Boethius ca. J10-22 297

130. See Grabmann 1939, pp. 12-13; Callus 1943, pp. 2658".; Dod 1070, pp. 81-97.
131. C. H. Lohr's catalogue (1967-74), with its bibliographical summaries and lists of manuscripts,
provides an invaluable starting-point for the analysis of this material.
132. These statistics are taken from the AL catalogue, and are valuable as an approximate guide to the
relative popularity of the translations. For pseudo-Aristotelian and spurious works the catalogue
is not comprehensive and the statistics in the second section of the table are likely to be
underestimates; those for the Secretum secretorum and the De regimine sanitatis are from Manzalaoni
1977. An asterisk indicates a translation from the Arabic; all the rest are from the Greek.
Table (cont.)
N o . of
Work Translator Date surviving MSS.
Ammonius William of Moerbeke 1268 4

Averroes (middle William o f Luna 13 th century 3*
commentary)
Prior Analytics Boethius ca. 510-22 273
Anonymous 12 th century 2
Averroes (middle William of Luna 13 th century I*
commentary)
Posterior Analytics James of Venice ? U25-JO 275

loannes before 1159 I
Gerard of Cremona before 1187 3*
William of Moerbeke ca. 1269 or 4
earlier
'Alexander' James o f Venice ? 1125-50 Fragments
Themistius Gerard o f Cremona before 1187 3*
Averroes (middle William of Luna 13 th century 1*
commentary)
Topics Boethius <<j. 510-22 268

Anonymous 12th century I
Sophistici elenchi Boethius ca. 510-22 271
James of Venice ca. 1125-50 Fragments
William of Moerbeke ca. 1269 or 1
earlier
'Alexander' James o f Venice ? 1125-50 Fragments
Physics James o f Venice ? 1125-50 139
Anonymous ('Physica ? mid-i2th 1 (fragment)
vaticana') century
Gerard of Cremona before 1187 7*
Michael Scot ca. 1220-35 65*
William of Moerbeke ? ca. 1260-70 230
Simplicius ? Robert Grosseteste ? after 1235 I (fragment)
Averroes (great Michael Scot ? ca. 1220-35 62*
commentary)
De caelo Gerard of Cremona before 1187 101*
Robert Grosseteste ? after 1247 1 (fragment)
William of Moerbeke ? 1260-70 185
Simplicius Robert Grosseteste ? after 1247 1 (fragment)
(Viterbo)
Averroes (great Michael Scot ca. 1220-35 36*
commentary)
Table (cont.)
N o . of
De generatione et Anonymous (vetus) 12th century 118

corruptione Gerard of Cremona before 1187 8*
? William of Moerbeke before 1274 190
Averroes (middle Michael Scot ca. 1220-35 40*
commentary)
Meteorologica Henricus Aristippus before 1162
(Book IV)
n3(*)
Gerard of Cremona before 1187
(Books I—III)
William of Moerbeke ? ca. 1260 175
Alexander William of Moerbeke 1260 (Nicea) 9
Averroes (middle Michael Scot (Book IV) ca. 1220-35 18*
commentary)
De anima James of Venice ? 1125-50 144
William of Moerbeke ? before 1268 268
Philoponus
(Book III) William of Moerbeke 1268 3
(Viterbo)
Themistius William of Moerbeke 1267 8
(Viterbo)
commentary)
De sensu Anonymous 12th century 94

Alexander William of Moerbeke ? 1260-70 4
Averroes (epitome) ? Michael Scot ca. 1220-35 48*
De memoria James of Venice ? 1125-50 Hi

De somno Anonymous 12th century 102

De longitudine James of Venice ? 1125-50 101

De iuventute James of Venice ? 1125-50 4

William of Moerbeke ca. 1260-70 157
De respiratione James of Venice ? 1125-50 4
Table (cont.)
N o . of
De morte James of Venice ? 1125-50 5

De animalibus Michael Scot (Hist., before 1220 69*
(comprising Historia, part., gener. only)
De progressu. De motu. William of Moerbeke 1260 (Thebes) 237'
De partibus, De Anonymous (part, only) ? 13th century 1
generatione)
Avicenna Michael Scot ca. 1220-35 29*
Averroes (epitome) Petrus Gallegus ? 1250-67 I*
? Michael Scot ca. 1220-35
8*
Metaphysics James of Venice ? 1125-50
5
('vetustissima')' 34
Anonymous ('media') ' " 12th century 24
Michael Scot ('nova') 1 3 6 ca. 1220-35 126*
Anonymous (revision of ca. 1220-30 41
James; 'vetus')
William of Moerbeke before 1272 217
('novae translationis')
commentary)
Nicomachean Ethics Anonymous (Books 12th century 48

II—III; 'vetus')
Anonymous (Book t early 13 th 40
and fragments of century
I l - X ; 'nova*)
Robert Grosseteste ? 1246-47 33
Anonymous (? William 1250-60 246
of Moerbeke) revision
of Grosseteste
Eustratius and others Robert Grosseteste ? 1246-47 22
Averroes (middle Hermannus Alemannus ? 1240 9*
commentary)
Eudemian Ethics
(De bona Anonymous ? 13th century 139
fortuna)131
(Fragments) Anonymous ? 13th century 3
"33 Of which only 40 contain all five treatises.

134 Comprising Books I-IV.4 (1007*31).
135 Lacks Book XI.
136 Comprising Books II (= a), I.5-10 (from 987*9), III-X, XII.1-10 (up to I07j b n).
137 The work circulating under this title consists of a single chapter of the Eudemian Ethics (VII.14)
combined with a chapter from the Magna moralia (II.8).
Table (cont.)
No. of
Polities William o f Moerbeke 1260-4 3

(Books I—II; ? early
draft)
William o f Moerbeke ? 1260 107
(complete)
Oeconomica Anonymous ?late 13th 15

century
Durandus de Alvernia 1295 79
Rhetoric Anonymous ? mid 13th 5
century
Hermannus Alemannus ca. 1256 3*
William of Moerbeke before 1270 100
Rhetorica ad
Alexandrum Anonymous ? 14th century I
Anonymous ? 14th century I
Poetics William of Moerbeke 1278 2

Averroes (middle Hermannus Alemannus 1256 24*
commentary)
Translations of Pseudo-Aristotelian and Related Works
N o . of
Problemata Bartholomew of Messina 1258-66 55

Physionomie Bartholomew of Messina 1258-66 119
De mirabilibus Bartholomew of Messina 1258-66 1
auscultationibus
De principiis Bartholomew of Messina 1258-66 I
(Theophrastus,
Metaphysics)
De signis aquarum Bartholomew of Messina 1258-66 10
Anonymous ? 13th or 1:4th I
century
De lineis indivisibilibus Robert Grosseteste ? ca. 1240-- 5 0 68
De mundo Bartholomew of Messina 1258-66 6
Nicholas of Sicily ? before i:240 50
Magna moralia Bartholomew of Messina 1258-66 56
(De bona
fortuna)1"
Translations of Pseudo-Aristotelian and Related Works (cont.)
N o . of
Anonymous ? 13th century 139

De coloribus Bartholomew of Messina 1258-66 78
William of Moerbeke ? 1260-70 I
De inundatione Nili Anonymous ? 13th century 82
De plantis Alfred of Sareshel before 1200 159*
(by Nicholaus
Damascenus)
De proprietatibus (or De Gerard of Cremona before 1187 119*
causis proprietatum
eltntentorum)
De mineralibus Alfred of Sareshel ? before 1200 32*(+U3*)
(Avicenna)
De itttelligentia James of Venice ? 1125-50 45
(anonymous
introduction to the
Physics)
Secretum secretorum Philip of Tripoli ? ca. 1243 350*
De regimine sanitatis John of Seville (?) mid-i2th 150*
(a fragment of the century
Secretum secretorum)
De causis (Proclus) Gerard of Cremona before 1187 202*
De differentia spiritus John of Seville mid-12 th 82*
et animae century
Anonymous 12th century 71*
Enigmata Aristotelis Anonymous 12th century 2
Liber sex principicrum 12th century 23I
De porno (De morte ? Manfred, King o f Sicily ca. 1258-66 70*
Aristotelis)
Vita Aristotelis Anonymous 12th-13th 65
century
Porphyry's Isagoge Boethius ca. 510-22 295
De laudabilibus bonis Robert Grosseteste ? 1240-53 14
Epistola ad Alexandrum Anonymous 13th century 52
(an anonymous
introduction to the
Rhetorica ad
Alexandrum)
Summa Alexandrinorum Hermannus Alemannus 1243-4 14*
(an Arabic epitome
ofthe£f/ii«)
3
THE MEDIEVAL I N T E R P R E T A T I O N OF
ARISTOTLE
Clerical and Aristotelian science

The question of the interpretation of Aristotle in the Middle Ages must be
dealt with within the context of the medieval conception of science.
Medieval learning was characterised by an attitude which was dominant -
though in varying degrees and varying circumstances — from the time of
Alcuin to that of Bellarmine. For the Middle Ages it was not the individual
who taught, but the Church, through the clergy. Clerical science was
accordingly the corporate transmission of traditional wisdom. The cleric
was a 'master' chosen by God to teach his people the way of salvation, as
Rhabanus Maurus put it at the beginning of his De institutione clericorum.1
His authority as a teacher was guaranteed by a divine call within the
ecclesiastical hierarchy; the authority of his teaching was guaranteed by
Scripture and the Church Fathers. But his authority extended even beyond
the sacred sciences, in a way which reveals the relationships between this
clerical attitude towards knowledge and the structure of medieval society.
The relationship between the clergy and the laity is clearly symbolised
in Alcuin's Dialectica, a dialogue in which Alcuin as magister instructs
Charlemagne himself as discipulus, in one of the profane sciences of the
trivium and quadrivium.2
Within this conception of the scientific enterprise a standard method of
interpretation was developed based on the presumed concordance of the
fundamental authorities, and schools evolved whose function was the
training of masters w h o should transmit traditional learning to God's
people. In the thirteenth century Thomas Aquinas in his inaugural lecture
as master of theology in Paris could still cite the Psalmist, 'From thy lofty
abode thou waterest the mountains, and the earth is satisfied', and interpret
the lofty abode as the divine wisdom which waters the minds of the masters
ι. Rhabanus Maurus, De institutione clericorum 1,2: 'doceantque populum Dei omnia légitima eius et
praecepta quae mandaverat ad eos' (PL 107, 297f.).
2. Alcuin, Dialectica (PL 101, 949-76).
The medieval interpretation of Aristotle 81
(signified by the mountains), so that by their ministry that wisdom might

be channelled to the laity (signified by the earth). Mixing his metaphors,
Thomas tells us that the masters are, like the mountains, elevated above the
earth and thus first illumined by the rays of the sun. 3
With the introduction of the Aristotelian encyclopaedia this clerical
conception encountered a radically different notion of science, a notion
which would require a new method of interpretation, a new form of
school, and a new type of teacher. It is clear that the gradual replacement of
the clerical attitude to learning by this new conception is also related to the
slow emergence of new social structures in the early modern period. But
not much attention has been paid to the fact that the way in which the
change took place was decisively conditioned by the various stages in the
reception of Aristotelian science.
The reception of Aristotelian science

The works of Aristotle were made available in the Latin West in three
clearly distinguishable stages. The first began in the sixth century with
Boethius' translations of Aristotle's treatises on logic and his adaptations of
various other works on logic and rhetoric. The second stage began in the
twelfth century with the gradual translation of the entire corpus of
Aristotle's works. At this stage the reception of Aristotle was part of a vast
effort to absorb the philosophical, medical, astrological, and natural science
not only of ancient Greece, but also of past and contemporary Judaism and
Islam. The Aristotelian encyclopaedia provided the framework for all this
new material. The third stage in the pre-modern study of Aristotle began
in the late fifteenth century and concentrated rather on the text of
Aristotle's works than on the co-ordination of the sciences. This stage
produced above all new editions of the Greek text, new Latin and verna-
cular translations and commentaries, Greek editions and Latin translations
of practically the whole corpus of the ancient Greek commentaries, and
Latin versions of hitherto untranslated commentaries of Averroes.
Boethius' translations
The first wave of translations broke on the late R o m a n world. This world
knew little of Greek philosophy and science, and, apart from some rheto-
rical notions transmitted by Cicero, very little of Aristotle. Boethius'
translations were, so to speak, an historical accident and could have but
}. Thomas Aquinas 1954a, Breve principium de commendatione sacrae scripturae, 1, 441—3.

little influence, not only in the final phase of classical civilisation, but also in
the monastic schools of the early Middle Ages. Although the Aristotelian
logic fitted neatly into the scheme of the liberal arts and although its
introduction into the trivium would later have a profound effect on the
Augustinian vision of arts and theology united in one comprehensive
system of knowledge, the monastic teacher, for whom the role ofmagister
was only a part-time, fugitive aspect of his existence, could find little he
could use in the predicables and predicaments of the ars vetus. The rudi-
mentary treatment of these subjects in the various works De institutione
clericorum and the very fact that half the translations of Aristotle's treatises
which Boethius had made were lost in this period shows that intellectual
work was only one small part of the monk's concern. The monastic
vocation did not regard study as an end in itself. The task of the monastic
teacher was rather ordered to the service of God and centred on the
understanding of God's word as recorded in the sacred writings and
interpreted by the Fathers.
The twelfth-century translations and the quest of the new masters

The interest in Aristotle which appeared with a second wave of translations
in the twelfth century presupposed another type of teacher, and the
birthplace of this type of teacher was the medieval town. From the eleventh
century knowledge was no longer confined to remote monasteries. With
the rise of the towns new interests appeared and the specialisation of labour
led some to concentrate on the production of goods, others on their
transport, still others on financing their purchase. Within this matrix arose
a new type of teacher, a new type ofmagister. Although in accordance with
the traditional division of society he still belonged to the clerical class, he
was no longer a teacher like Alcuin or even like Notker Labeo or Scotus
Eriugena. Like the men of trade who established themselves in the towns,
like the carpenters and masons who organised themselves in guilds, this
new master had the consciousness of belonging to a profession. His trade
was learning and teaching, personal reflection and its diffusion in the
classroom. The liberal arts were his speciality.4
It was this type of master who turned avidly to Aristotle. From about the
beginning of the eleventh century the masters of arts slowly pieced to-
gether the original fabric of the Aristotelian logic with the exception of the
theory of demonstration as it is found in the Posterior Analytics. The vague
4. LeGoffi957.
references to the various Aristotelian treatises on proof which we find in

Alcuin's Dialectica were gradually replaced by a treatment of categorical
and hypothetical propositions and syllogisms drawn from the works of
Boethius. The Aristotelian method of the topics and the treatment of the
fallacies were reconstructed from hints in the available works of Aristotle
and Cicero and from the treatises surrounding them. Abelard's Dialectica is
worlds away from Alcuin's, and it shows that the full range of the
Aristotelian logic which became known in the latter half of the twelfth
century was not used because the treatises were translated, but the treatises
were translated because this new generation wanted to use them. 5 By the
middle of the twelfth century these younger masters had come to realise
that there were whole areas of knowledge of which they knew only the
names. It was only natural that they should try to learn more of their
subject matter.
Their seeking was already a kind of interpretation of the texts. In
looking for the lost books of the Aristotelian logic they already knew what
they were expecting to find. And what they were expecting to find
matched their o w n understanding of their role in society. This is why the
masters' study of the Aristotelian logic did not proceed without opposition
from the representatives of the traditional conception of the cleric's task.
The polemics of Peter Damian against the dialecticians, of Lanfranc against
Berengarius, of Bernard of Clairvaux against Abelard represent the reac
tion of the older, monastic idea to the new, urban conception of the teacher's
role. The new generation's search for hitherto unknown Aristotelian
works is the expression of its o w n new self-image.
Parallel to this effort to forge a new tool for the sciences, a novum
Organum, ran an awakening interest in the subjects of the old quadrivium.
The new interest in the study of nature which is such a prominent feature
of twelfth-century thought entailed another conflict with the traditional
understanding of the clerical vocation. Although in accordance with the
structure of society the masters whose trade it was to teach were fitted into
the clerical class, their function was no longer simply that of transmitting
traditional wisdom. They could no longer simply proclaim the word of
God. They themselves had to learn; they had to master the knowledge of
the ancients, they had to go to school as the discipuli of Greeks, Arabs, and
Jews, they had to confront the sayings of the Scriptures and the Fathers
with the sciences of the ancient pagans and the religious teachings of infidel
5. De Rijk in Peter Abelard I95 fi , ΡΡ- xvi-xix.

nations, they had to concern themselves not only with the relationship
between God and his people, but also with man's relationship to the world
in which he finds himself. They learned the names of many new and
strange sciences from the merchants who brought reports from Spain and
the Orient, from the Jews whose knowledge of Hebrew and Arabic and
whose international contacts had given them access to the ancient sciences
of the Greeks, from scholars like the Englishmen Adelard of Bath and
Daniel of Morlay whose travels in Sicily and to Toledo had brought them
into direct contact with the philosophy and science which R o m a n civilisa-
tion had failed to hand down.
As the masters learned the names of these new sciences, they were like a
modern librarian who finds a lacuna of several volumes in one of his
library's periodicals; they could not rest until they had found the means to
fill the gap. Thus they turned to the translators. The additions which these
interpreters of the classical tradition made to medieval knowledge was
immense: in geometry Euclid, in astronomy Ptolemy, in medicine Hip-
pocrates and Galen, and above all - for method, for system, for wholly new
and undreamt-of sciences — the works of Aristotle, the Philosopher par
excellence, together with his Arabic commentators.
This quest of the twelfth-century masters of arts to regain the ancient
sciences was the first step in the interpretation of the works they recovered.
The works of Aristotle which were thus made available by about the year
1200 did not gain the influence they had because they were fortuitously
translated, but they were translated because the masters wanted no longer
simply to transmit, because they wanted to learn themselves. The spirit of
reason, of curiosity, of criticism which they found in Aristotle matched
their own spirit and helped to crystallise their self-image. The system, the
encyclopaedia of the sciences they found gave them a sense of autonomy
and freedom with respect to the traditional understanding of the clerical
function. In fact, though the working out of the implications of this second
wave in the reception of Aristotle would take three hundred years, the fate
of the clerical conception of knowledge was already sealed with the
openness of the twelfth-century physici to the sciences of ancient Greece.
A thirteenth-century student's guide
in Barcelona, in the Archives of the Crown of Aragon, there is a

thirteenth-century manuscript which contains a manual or guidebook for
students in the arts faculty in Paris. 6 This text, which was apparently based
on early thirteenth-century practice, was composed about 1230-40 by an
unknown master of the faculty for the benefit of students who had to
prepare for examinations. This students' guide is important not only
because the standard texts prescribed for each branch of the programme
correspond exactly to the newly translated works in logic, mathematics,
and the natural sciences, but also because it reveals the direction which the
development of the arts faculty would have to take. For the author of the
guide-book the arts are no longer simply the seven liberal arts of the trivium
and quadrivium; they comprise rather all the philosophical and scientific
disciplines newly recovered at his time. And because the author attempts to
situate the plan of studies in the arts faculty within the context of a
complete classification of the sciences, these arts include some disciplines as
yet unknown to him.
After some reflections on the nature of philosophy, the author divides his
subject into three branches: rational, natural, and practical or moral philo-
sophy. Under rational philosophy he takes up the subjects of the trivium,
assigning to grammar the works of Priscian and Donatus, to rhetoric
Cicero's De inventione, and to dialectic Aristotle's Organon together with
the Isagoge of Porphyry and the logical treatises of Boethius.
Natural philosophy he divides into metaphysics, mathematics, and
physics. For metaphysics the standard texts are Aristotle's Metaphysics and
the pseudo-Aristotelian Liber de causis. Under mathematics he takes up the
subjects of the quadrivium, but assigns to some of its branches works which
were unknown in the earlier Middle Ages. To astronomy he assigns Pto-
lemy's Almagest, to geometry Euclid's Elements, to arithmetic Boethius'
Institutio arithmetica, and to music Boethius' Institutio musica. Physics, being
at a lower degree of abstraction than metaphysics and mathematics, is
described as scientia naturalis inferior. Here are taken up all the works
ascribed to Aristotle on natural philosophy: Physics, dealing with the
general principles of change; De caelo, dealing with the eternal motion of
the celestial bodies; De generatione et corruptione, dealing with the four sub-
lunary elements which explain generation and corruption; Meteora, dealing
with a great variety of natural phenomena; De planus. De animalibus, De
anima, Parva naturalia, and De motu cordis, which deal with the whole range
of animate nature.
6. MS Ripoll 109 f. 134'- 158". See Grabmann 1936, and Van Stcenberghcn 1966, pp. 119-32.
Moral philosophy the author divides into the treatment of the life of the
soul, first in its relation to God, then in its relation to others, and finally in
itself. Here the author's assignment of texts to the different branches lacks
the clarity we have found in the other sections. The study of the life of the
soul in God he identifies with theology, but he indicates no standard text.
The other divisions reflect Aristotle's classification of the practical sciences
into those concerning the individual, the family, and the state. But the
author does not yet know the Oeconomica and the Politics, and so assigns
Cicero's De offrais to the consideration of the life of the soul in the family,
and the study of R o m a n and canon law to the consideration of the life of
the soul in the state. He assigns Aristotle's Ethics only to the treatment of the
life of the soul in itself. After the treatment of ethics the author adds the
note that two other books are also read in the faculty of arts: Plato's
Timaeus and Boethius' De consolatione philosophiae.
The influence of the Aristotelian classification of the sciences
This students' guide marks a definite stage in the evolution of the

medieval arts faculty, the final stage in the formation of a new type of
school, a school representing the interests of the new, urban type of master
and his basically unclerical conception of the scientific enterprise.
Although the author attempts to assign theology a place among the
practical sciences, his concern, which certainly reflects the gradual in-
stitutionalisation of instruction in Paris, is rather with the Aristotelian
system of the sciences. This system will lead the masters of arts inevitably to
Aristotle's division of the practical sciences. We have observed that this
section is the least clear in the author's treatment, because he does not yet
know the Oeconomica or the Politics. But he does know the names of the
sciences, and no doubt his colleagues were searching the libraries of Europe
for copies of the works to be translated. From this point the new Latin
Aristotelians went beyond their Arabic forerunners. They began to turn
increasingly to the Greek tradition of Aristotle's works and came even-
tually to regard the Arabic contribution as alien to their own self-image as
the successors of the Greeks and Romans.
The Aristotelian classification of the sciences was thus instrumental in
the recovery of Aristotle's o w n works. It also supplied the framework for
the vast amount of new scientific material, for the Greek, Arabic, and
Hebrew works on mathematics, astrology, medicine, and natural science
which the translators of the late twelfth and early thirteenth centuries had
made available. But more than this, the Aristotelian system of the sciences
was decisive for the formation of the medieval university.
The arts faculty as a philosophical faculty

On 19 March 1255 Aristotelianism was officially adopted in the University
of Paris as the arts faculty proclaimed a new syllabus which imposed the
study of all the known works of Aristotle. 7 O n that day the arts faculty
became what we might call a philosophical faculty, with a new importance
in its own eyes and a tendency to develop a teaching independent of the
theological faculty. Such a development was bound to arouse violent
reactions and a growing rivalry between the two faculties.
The conflict had broken out even earlier and concerned at first moral
philosophy, the third major division of the sciences as we have found them
in the students' guide. Although the author knows only Ethics I—III, he
knows of the existence of the remaining books and assigns to the subject an
importance second only to logic, if we may judge by the length of his
treatment. It was certainly from this point that the conflict took its
departure. The author brings the philosophical and theological concep-
tions of beatitude into direct opposition with each other. The resurrection
of the body is a miracle which does not answer to natural laws and
therefore does not belong to the subject matter of philosophy. In answer to
the question whether we are the causes of our good actions as we are the
causes of our bad actions our author distinguishes between the point of
view of a philosopher and that of a theologian: 'To which we reply that
speaking philosophically we are the entire cause of both; speaking theolo-
gically however, we are not capable of good actions, but it is necessary that
God pour grace into us.' 8
In a few decades questions concerning the eternity of the world and the
immortality of the human soul were added to the questions regarding
which philosophy and theology were thus expressly opposed. But far more
profound than these particular differences was the difference in attitude
which our master of arts here reveals. Basically the issue concerned the
definition of the role of the master of arts confronted with the still
dominant conception of the clerical mission. The mendicant controversy
and the history of the Franciscan order bring out one side of this issue; the
7. CUP 1, pp. 277-9.

8. 'Ad quod dicimus quod loquendo philosophice sumus tota causa utriusque; loquendo tamen
theologice, non sumus sufficientes ad bonum, sed oportet gratiam in nobis a Deo infundi' (f. 136';
Grabmann 1936, p. 196).
view that attendance at a university and the possession of books were

contrary to the practice of poverty was the allegorical expression of the
anxiety aroused by the secular sciences introduced with Aristotle.
The effect of the Averroistic controversy

What came to be known as the Averroistic controversy in the 1260s and
70s led to some of the most intransigent formulations of the masters' own
understanding of their role. Again Aristotle's Ethics furnished the arsenal.
The battle was joined over the ideal of humility and its ethical antithesis,
magnanimity. The victory to be won was the theoretical grounding of the
dignity, the superiority of the university status. In his Quaestiones morales
Siger of Brabant discusses the opinion that humility is not a virtue, for it is
opposed to a virtue like magnanimity which tends to great things. In his
own answer he makes both virtues depend on recta ratio and accordingly
subordinates the moral to the intellectual virtues. In the same way Boethius
of Dacia in his brief tract De summo bono sive De vita philosophi maintains
that it is in the operation of the intellectual virtues that the most perfect
condition possible to man is to be found. This, he proclaims, is the status of
the philosopher who dedicates his life to the pursuit of wisdom. 9
One recognises here the road which led to certain of the propositions
which were condemned in 1277. Prop. 40: That there is no more excellent
way of life than the philosophical way. Prop. 144: That the highest good of
which man is capable consists in the intellectual virtues. Prop. 154: That the
philosophers alone are the wise men of this world. 1 0 Throughout all of this
the catchword is 'philosophus'. The philosopher has succeeded the cleric.
Abelard had already described himself as a philosophus. The designation
was championed also by Siger of Brabant and Boethius of Dacia. For the
thirteenth century Aristotle was the philosophus par excellence. It was in his
train that the masters of arts thought of themselves as following.
Philosophical and theological discourse

The prescription of the Aristotelian corpus as the basis of instruction in the
arts faculty brought with it for the masters the obligation of interpreting
the texts they had so eagerly sought after. Their commentaries on the
works of the Philosopher open a new epoch in the history of medieval
exegesis. The masters themselves were fully conscious of the revolution
9. Siger of Brabant 1974, Quaestiones morales qq. 1 and 4, pp. 98f. and I02f. Boethius of Dacia 1976.
10. CUP I, pp. 543-55-
their interpretation of the newly recovered texts involved. As early as our

students' guide we find the author, in the text cited above, distinguishing
between philosophical and theological discourse (loquendo secundum philo-
sophos; loquendo secundum theologos et secundum veritatem). Siger of Brabant
explains his purpose even more explicitly: 'We seek what the philosophers
meant in this matter, their intention rather than the truth, because we
proceed philosophically.' ' '
The tradition of biblical exegesis

Medieval exegesis had been concerned with the Bible. Its premiss was that
the exegete was already in possession of a truth revealed by God himself.
His task was accordingly not the discovery of new truths, but rather the
unveiling of the truth concealed in the words of the sacred text. In
accomplishing this task he not only turned to the councils and Church
Fathers as authorities to lead him, but also felt himself, as a living link in a
corporate undertaking, endowed with the same authority to teach. In the
twelfth century, as discrepancies among his authorities became increas-
ingly obtrusive, his conviction that the tradition of which he was custodian
was at bottom coherent guided his efforts to penetrate more deeply into the
truth of God's word as a sort of Concordia discordantium. Even the great
Summae of the thirteenth century which arose out of this effort are in this
sense exegesis. Their point of departure was the articles of faith which God
had revealed in the Bible. The purpose of the summist was to try to make
the res, the transient things of this world, shine in the light of the voces,
the divine words as the bearers of immutable truth. That is why Thomas
could say that theologians are like the mountains, elevated above the earth
and first illumined by the rays of the sun.
The new philosophical exegesis

The point of departure of the masters of arts was radically different. Siger of
Brabant and his fellow masters were the first to want to interpret philoso-
phical texts 'philosophically', that is, in a way which we might perhaps call
philological, or at least in the very unclerical way of abstracting from the
question of the truth of the teaching. The very fact that they identified
the viewpoint of the theologians with the truth shows in a paradoxical
way that for them the text they had to comment upon was not a unique
11. Siger of Brabant 1972a, DP anima intellective, cap. 7, p. 101, 'quaerendo intentionem philoso-
phorum in hoc, magis quam veritatem, cum philosophice procedamus'.
authority, but rather one source among many. Their task was not the
unveiling of a truth already possessed but hidden; it was rather the discus-
sion of the opinion of a most distinguished colleague. For this reason Siger
gave the following rule for the interpretation of Aristotle: 'It should be
noted by those who undertake to comment upon the books of the Philo-
sopher that his opinion is not to be concealed, even though it be contrary to
the truth.' ' 2 A further consequence of this 'philosophical procedure' was
that the interpreter need make no effort at a concordia discordantium. In
rejecting Thomas Aquinas' interpretation of Aristotle's discussion of the
eternity of the world one of Siger's colleagues - possibly the very Peter of
Auvergne who completed so many of Aquinas' Aristotle-commentaries-
explicitly opposed the method of concord:
But as is clear, Aristotle proves that motion is eternal and this is apparent from the
reasons he gives. Some, however, wanting to harmonise Aristotle's intention with
the teachings of faith, say that Aristotle was not for these reasons of the opinion
that the world is eternal, or that he did not hold them to be demonstrations
necessarily concluding what is true, but that he only adduced these reasons
hypothetically and for no other purpose. But this is manifestly false, for it would
follow then that Aristotle presented the greater part of his philosophy as an
hypothesis. ' 3
In these two rules we can see clearly the revolution which has taken
place. The theologian sought to unveil a truth concealed; the philosopher
need not seek to conceal the errors in his sources. Since the work of
Aristotle, the primary source for a member of the arts faculty, was for him
neither a new dogma nor an infallible guide, he need make no clerical
attempt at harmonising science and the Bible. The interpreter, having
abandoned the notion of truth possessed for the notion of truth to be
sought, could approach the text of the Philosopher in a critical, questioning
12. Siger of Brabant 1948, Quaestiones in Metaphysicam ill, q. i5Comm.,p. 140: 'Propter hoc sciendum
quod sententia Philosophi ab his qui eius libros suscipiunt exponendos, non est celanda, licet sit
contraria veritati. Nee debet aliquis conari per rationcm inquircre quae supra rationem sunt, vel
rationes in contrarium dissol verc. Sed cum philosophus quantumcumque magnus in multis possit
errare, non debet aliquis ncgare veritatem catholicam propter aliquam rationcm philosophicam,
licet illam dissolvere nesciat.*
[3. The Pseudo-Siger of Brabant, Quaestiones super libros Physicorum vill, q. 6; in Siger of Brabant
1941, p. 199: 'Aristoteles autem, ut manifestum est, probat motum esse aeternum, et hoc apparet
ex rationibus quas ponit. Quidam tamen volentes concordare intentionem Aristotelis fidei dicunt
quod Aristoteles non fuit opinatus ex istis rationibus mutidum esse aeternum, nee tenuit eas
demonstrationes concludentes verum de necessitate sed solum adduxit istas rationes propter
dubitare et non propter aliquid aliud. Istud tamen est manifeste falsum, quia sic sequeretur quod
Aristoteles dubitaret in majori parte philosophiae suae, et maxime ubi loquitur de substantiis
separatist ex aeternitate enim motus probat quod sunt substantiae separatae, sicut patet Libro "caeli
el mundi".'
way. He could explain Aristotle's words by reducing them to their prin

ciples, confident that even where Aristotle's conclusions might conflict
with the faith, Aristotle's spirit would be an invitation to go beyond
himself in a search for new truths. It is this conception of truth that makes
the exegesis of the masters of arts so novel in the development of medieval
thought. With the distinction between philosophical and theological dis
course the 'truth' has rolled out of its centre within the text to become an
unknown V .
The effect of conflicts between Aristotle and Christian doctrine
Behind this revolution lay no doubt the de facto conflicts between

Aristotle's teachings and the doctrines of faith. The masters of arts were
confronted with a vast literature opposing various interpretations of
Aristotle: Albert the Great, De XV problematibus; Thomas Aquinas, De
unitate intellectus contra Averroistas; Giles of Rome, De erroribus philoso-
phorum; the condemnations of 1270 and 1277. In the face of such opposition
it was difficult to maintain that Aristotle had spoken the whole truth. But
on a deeper level this revolution in the theory of interpretation represents
the beginning of the end of the clerical paradigm for the scientific enter
prise. The masters of arts could recognise the deficiencies in Aristotle's
teaching. But in him they found a new paradigm, a new model not only
for interpretation, not only for science, but also for the vocation of the
university man. In Aristotle, the Philosopher, they found the researcher,
the questioner, - or to use Aristotle's own words, the hunter, the dis
coverer, the seeker14 - one who subjected the teachings of his predecessors
to a relentless critique, who was subservient to no authority and free of all
dogmatism. The masters of arts wished their commentaries also to be
philosophical, seeking, hunting, critical, and in this way different from the
clerical commentaries of the theologians. By their own self-image they
were precluded from wanting to raise Aristotle's teaching to the level of a
new dogma. They claimed an authority for Aristotle's teaching, an author
ity not guaranteed by a divine call or a sacred text, however, but one based
solely on reason. Their 'philosophical procedure' in the interpretation of
Aristotle reflected their consciousness of their own corporate position in
society - an elite which owes its dignity not to privilege or hierarchical
status, but to intellectual superiority.
14. For example, thheuein. Anal, prior. 1,30 (46Ί1); heurêsis, Eth. Nie. Ill, 5 (1112 b l0); zêtêsis, Metaph.
1.2(983*23)·
The reactions of the theologians

In the face of such an attitude the theological faculty had - since at least the
time of our students' guide - been on the defensive. The theologians had
traditionally attempted to solve problems arising out of divergent author-
ities by seeking a standpoint from which all the relevant texts could be
brought into harmony. But in the thirteenth century the newly translated
philosophical and scientific sources rendered questionable the simple con-
cordances which the twelfth century had made between authorities limited
to the Latin ecclesiastical tradition. In this new situation some rejected the
new literature and attempted by ecclesiastical condemnations to prevent its
being read; others, like Bonaventure and Olivi, saw in Aristotle the
apocalyptic beast of the last days and took refuge in the historical specu-
lations of Joachim of Fiore; still other theologians, like Albert the Great and
Thomas Aquinas, showed themselves receptive to the new sources and
tried in a new and very subtle way to continue the clerical enterprise of a
concordia discordantium.
Thomas Aquinas' solution

Thomas went furthest in the attempt to answer the challenge posed by the
approach of the masters of arts to the new literature. As a theologian he had
to maintain the existence of truths revealed in the Bible which transcended
human understanding. At the same time, the encounter with the religious
teachings of Judaism and Islam had constrained Latin theologians to
attempt the construction of an apologetic based on arguments acceptable
to the three faiths. Because such arguments could be based only on rational
demonstration, Thomas sought to justify the inclusion of philosophical
questions in the subject matter of theology. Because theology is the science
of revelation, he maintained that God had revealed not only strictly
supernatural truths, but also some truths which are philosophically de-
monstrable. 15 For example, God revealed his existence, for otherwise but
few men would have attained certain knowledge of this truth. Never-
theless, Thomas argued, God's existence can be also demonstrated, and he
proposed five ways of doing so. The refrain which we find at the con-
clusion of each of them - 'Et hoc dicimus Deum' ('And this is what we call
God') - i s the clearest example of Thomas' solution of the problem of the
concord between philosophy and revelation. The first cause whose ex-
istence has been rationally demonstrated on the basis of the principles of the
15. Thomas Aquinas, ST, 1, q. 1, a. 1. See also Lang 1962 and Lang 1964.
philosophers is that very being which the Christian by revelation knows as

God. 16
The concord between philosophy and revelation which Thomas in-
tended involved not only the demonstration of rationally accessible truths,
but also the discovery of natural analogies to transcendent truths and the
ordering of both natural and supernatural truths in a scientific way. 1 7
Thomas' theologian had therefore to turn to nature and could employ in
this effort the works of'the master of them that know'. In the Aristotelian
logic Thomas found prescriptions for the ordering of theological doctrine
as a strict science. In the Aristotelian metaphysics he found the principles
for the demonstration of truths such as the existence, infinity, and omni-
potence of God. In the Aristotelian natural philosophy he found natural
analogies to the hierarchical view of the world which the clerical tradition
had handed down. This conception of the Aristotelian encyclopaedia
implied, on the one hand, a reinterpretation of Aristotle's principles in the
light of conclusions already known by revelation. This was the reason why
the theologians could not admit the Averroist interpretation of Aristotle's
doctrine. On the other hand, because the Aristotelian philosophy claimed
to be an explanation of nature in all its aspects, Thomas' conception meant
tying the whole of Aristotle's physics to dogma in a way which would be
disastrous for both.
It was in dealing with the Aristotelian astronomy that Thomas en-
countered a type of discord different from that between dissenting
theological authorities. In explaining the account of creation in Genesis,
the Latin theologians prior to the thirteenth century had generally adopted
Rhabanus Maurus' discussion of the number of the heavens, but later
commentators were confronted with radically different conceptions. The
translators from Arabic and Greek had made available two far more
advanced, but mutually opposed discussions of this problem: Ptolemy's
Almagest and Aristotle's De caelo, together with Averroes' commentary. As
we have seen in the systématisation of the sciences undertaken by the
author of our students' guide, the former was subsumed under the
mathematical sciences, the latter under scientia naturalis inferior. While the
professional astronomers of the period adopted Ptolemy's theory of eccen-
trics and epicycles and paid little attention to Aristotle's theory of homo-
centric spheres, the theologians, who had to treat the question in their
16. Thomas Aquinas, ST, I, q. 2, a. 3.

17. Thomas Aquinas 1948c, prooem. q. 2, a. 3.
commentaries on Book II, dist. 14 of Lombard's Sentences, were very

disturbed by the contradiction between Ptolemy's mathematical as-
tronomy which claimed to save the phenomena and Aristotle's physical
theory which was presented as a deduction from first principles.
Thomas' attempt at a solution of this problem - a solution which we
have already seen rejected by one of the colleagues of Siger of Brabant —
shows clearly the difference between his theological interpretation of
Aristotle and what we may call the philosophical interpretation of the
masters of arts. For Thomas the harmonious order which he found in
Aristotle's physical theory was based on absolutely certain, metaphysical
principles. To the argument that Ptolemy's hypotheses are supported by
experience, Thomas rejoined that the experimental verification of a hypo-
thesis does not demonstrate the hypothesis:
The [mathematical] hypotheses which [Eudoxus and later astronomers] invented

are not necessarily true. For even though such hypotheses should save the pheno-
mena, it is not right to say that they are true, because the astronomical phenomena
can perhaps be saved in some other way not yet understood by men.18
Although Thomas thus formulated explicitly one of the most important

principles in the theory of science, he employed it to render harmless the
objections to his theological interpretation of Aristotle's astronomy - in
the hope that some day a way might be found to make Aristotle's theory
agree with experience. Saving Aristotle's physics was for Thomas more
important than saving the phenomena. His appeal to the principle that
verification does not demonstrate an hypothesis meant only that his con-
ception of the concordance between philosophy and revelation need not be
disturbed by the contrary data of experience. Armed with Thomas' prin-
ciple, the clerical world-view was able to maintain itself until the time of
Bellarmine and disappeared only with the new astronomical discoveries
and the new conception of science of the sixteenth and seventeenth
centuries. The falsification of the Aristotelian physics then implied for
many the falsification of Thomas' approach.
18. Thomas Aquinas 1889, II lect. 17; pp. i86f: 'Illorum autem suppositiones quas adin venerum, non
est necessarium esse veras; licet enim, talibus suppositionibus factis, apparentia salvarentur, non
tarnen oportet dicere has suppositiones esse veras, quia forte secundum aliquem alium modum,
nondum ab hominibus comprehensum, apparentia circa Stellas salvantur. Aristoteles tamen utitur
huiusmodi suppositionibus quantum adqualitatem motuum tamquam veris.' Cf. STi, q. 32, a. I,
ad 2. See also Mintclstrass 1962, pp. 173-8.
Aristotelian demonstration and discovery

Thomas Aquinas' answer to the challenge posed by the new literature was
the last speculative attempt to save the clerical conception of science as the
corporate transmission of traditional wisdom. Thomas gave the clerical
teacher the consciousness of belonging to a profession and remade theo-
logy on the model of an Aristotelian deductive science. But he did not alter
the basic clerical notion of the scientific enterprise. In emphasising the
deductive strand in science, he emphasised only one side of the Aristotelian
philosophy. It is true that Aristotle himself spoke of the philosopher as the
arranger, the co-ordinator, 19 but his desire for order was not directed
simply toward a deductive presentation of the results of the sciences, for
Aristotle was also concerned with the organisation of research. He seems to
have conceived of his school as a sort of university in which research and
teaching in the most diverse fields would find their place. If the method he
prescribed for teaching in his Posterior Analytics was strictly deductive, the
practice he followed in his scientific treatises ignored formalisation and
concentrated on research. His practice was aimed at the acquisition of new
knowledge, at discovery, at the via inventionis. That is what the clerical
approach could not admit, and that is what the Parisian masters of arts in
the thirteenth century rediscovered.
Philosophical inquiry in the thirteenth-century arts faculty

The rejection by these masters of the method of concordance, their rejec-
tion of the notion of a prior truth known by faith to which philosophical
truth must conform, their conception of truth as something to be sought
rather than as something possessed conformed to the image they had of
themselves as the successors of the critical, curious, and searching
Philosopher. Because their own status was not based on an appeal to
authority, they could admit that Aristotle - whose works they had sought
and made the first efforts to interpret - made mistakes. The authority of
their teaching was guaranteed only by reason. Since they claimed no
authority in the sacred sciences, they enjoyed a new liberty in their
research, a liberty which brought with it the many new, un-Aristotelian
developments of the fourteenth and fifteenth centuries.
The masters of arts regarded their work as philosophy, but in order to
write a history of their contributions today it is necessary to go beyond the
19· Metaph. 1,2 (982MH).

standard histories of philosophy and to consult histories of logic and

linguistic theory, of mathematics, mechanics, and astronomy, of ethics and
political theory. We have here considered only the distinction between
philosophical and theological discourse, which we first encountered in our
students' guide. This distinction is usually associated with what is known
as Averroism in the latter half of the thirteenth century, but in a deeper
sense it laid the foundation for the great scientific achievements of the later
Middle Ages. The distinction between philosophical and theological (not
truth, but) discourse represents not only a break with the clerical com-
mentatory tradition and the beginning of the end of the clerical paradigm in
science, but also a new autonomy for the medieval arts faculty. N o longer
simply the gateway to theology, the arts faculty became an institution on
an equal footing with the faculties of law, medicine, and theology. The
dignity thus gained for the philosophical vocation made it possible for a
Buridan to devote his entire life to the problems of logic and natural and
moral philosophy. The distinction from theology also made possible the
shift from the realm of the deductive rriiio to the realm of experience and
contingency, the whole new epistemological orientation which the four-
teenth century itself described as a transition from cognitio propter quid to
cognitio quia. The 'philosophical procedure' made it possible for the masters
of arts to turn increasingly from the exposition to the question-form of
commentary, to criticise the Philosopher, to ask the new logical and
mathematical questions with which Ockham and the Merton school led
science in the early fourteenth century into new paths; and it made it
possible for Oresme to fuse Mertonian mathematics with Parisian physics
in the late fourteenth century, and for Paul of Venice and others in Padua in
the fifteenth century to bring these developments together with the
Averroist attitude to form the secular Aristotelianism of the sixteenth-
century Italian universities.
Forms of Aristotelianism from 1500 to 1650

A third wave of editions of, translations of, and commentaries on the
works of Aristotle began in the late fifteenth century and lasted until about
the middle of the seventeenth. 20 The Aristotelianism of the period
1500-1650 presents, however, a picture which differs radically from the
university philosophy of the Middle Ages. Despite the many late medieval
developments in logic and physics which would eventually contribute to
20. See Lohr 1974-.

the breakdown of Aristotelian science, the Aristotelianism of the earlier

period remained predominantly clerical and offered an essentially unified
world-view. But in the sixteenth century this unity broke down, so that we
must speak not of one, but of several Aristotelianisms in the Renaissance.
Within the Catholic Church the Jesuits and the other religious orders
attempted to maintain Thomas' interpretation of Aristotle's Metaphysics in
the service of Catholic theology. In Protestant Germany Melanchthon
constructed a new Aristotelianism - without the Metaphysics - for the new
schools which should serve Luther's gospel. In France scholars concerned
with constitutional reform searched the logical works of the Greek
Aristotle for new ways to interpret legal doctrine. In Italy humanists
turned to Aristotle's moral philosophy, literary critics to the teachings of
the Poetics, university professors to works either unknown or ignored in
the Middle Ages, like the Problemata and the Mechanica, and to the Greek
commentators on the natural philosophy.
This great variety of Renaissance Aristotelianisms, each concerned with
a different aspect of Aristotle's encyclopaedia, certainly contributed to the
disintegration of the medieval world-view. But beneath them lay a new
conception of knowledge and science, a conception which was born with
the Parisian masters of the thirteenth century and could still be shared by
Descartes and Galileo, by Bacon and Hobbes, a conception of science no
longer bound by traditional authority. The supplanting of the clerical
paradigm for science is of course to be associated with the social changes
which characterised this period. But the way in which the change took
place was decisively conditioned by the Aristotelian philosophy. Aristotle
had never really fitted into the clerical mould which formed medieval
science. It was indeed the Aristotelian spirit of free research which eventu-
ally led to the breakdown of the theological syntheses of the medieval
period. This was a process lasting centuries, but after some four hundred
years the original fears of popes and theologians proved justified. In the
thirteenth century 'the philosophical interpretation' of Aristotle appeared,
in the fourteenth the arts faculty acheived institutional independence, in
the fifteenth the Aristotelian encyclopaedia constrained theologians to deal
with medical problems and professors of medicine to concern themselves
with the problem of the immortality of the soul. In the sixteenth century
new scientific interests, new classes of students, new geographical divisions
led such groups of scholars to attend to various works of Aristotle without
reference to his organisation of science. Although these developments took
different forms in different contexts, they were all alike in discarding the
clerical Aristotle of the medieval schools. The expression varied from land
to land, but all had in common the rejection of traditional authority in
scientific method. It was in this emancipated and pluralistic atmosphere, an
atmosphere charged with the spirit of authentic Aristotelianism, that the
new science was born.
Ill
THE OLD LOGIC
4
ANCIENT SCHOLASTIC LOGIC AS THE
S O U R C E OF MEDIEVAL SCHOLASTIC
LOGIC
Scholasticism and its means of expression

Medieval logic grew out of the school (university) curriculum; con
sequently, one characteristic vehicle of it was the commentary on a school-
book. Medieval philosophers were not, in general, people who believed
that the authoritative authors of the text-books were infallible or had said
all that could be said about the relevant subjects, but they shared some
convictions that can lead to that misimpression. In general they believed
that (i) the auctores had laid down the right principles of the several
disciplines and did not normally disagree over fundamental issues; (2) they
had divided logic into its sub-disciplines in a reasonable way and taken care
to provide posterity with treatises on all the main subjects; (3) therefore the
right way to do logic was to reach a full understanding of those books and
then proceed further in the footsteps of the auctores, remembering never to
contradict them without fully explaining the necessity of doing so or -
even better - showing that their text could be interpreted so as to make
them say what they ought to have said; (4) Aristotle was the greatest of the
auctores.
'Scholastic' properly characterises philosophers who approach their task
in the way men did in medieval Western Europe, but scholasticism in this
sense was neither a medieval nor a Western invention. It had flourished in
the Greek-speaking part of the world between ca. A.D. 150 and 550, and
medieval Latin scholasticism is not just a phenomenon comparable with its
1
Greek predecessor, it is directly descended from it.
The ancient world had not only an Aristotelian scholasticism (in logic)
but also a Platonic (in ontology). Similarly, in the Middle Ages the
I. As one illustration of the continuity between ancient scholasticism and that of the high Middle
Ages, see the very similar arguments offered by Radulphus Brito in Paris around 1300 and by
Philoponus in Alexandria around 525 to prove that Aristotle's Organon is a well-organised and
exhaustive course in the essentials of logic: John Philoponus 1909, pp. 1-2; Radulphus Brito
Prooemium Quaeslionum super artem veterem, MS Bruxelles Β. R. 3540-47, ff. 33r—3Sr; id.,
Prooemium Quaeslionum super librum Elenchorum, same MS, ff. 480' - 482'.
102 The old logic
scholastic method was not peculiar to logic, but the twelfth-century

renaissance of philosophical studies concentrated on this elementary dis
cipline rather than on any of the more advanced ones (metaphysics,
physics, ethics).
Late ancient and medieval logicians had several literary forms at their
disposal, but the scholastic attitude favoured the cultivation of the scholium,
a note on some particular passage in a book. Its length may vary from a line
to several pages, and long scholia often contain digressions and discussions
of problems (Greek aporiai, Latin dubia) to which the text gives rise. Most
scholia contain some amount of paraphrase of the text. A paraphrase of a
few words is commonly called ζ gloss. In medieval times short glosses were
generally written above the relevant passage of the text while the longer
notes, the scholia, were placed in the margins. The typical Greek com
mentary is nothing but a collection of scholia provided with a preface in
which some general aspects of the text are treated. Sometimes a certain
inner structure is imposed on the commentary by means of survey para
graphs inserted into the chain of scholia. The normal way to produce a
commentary was to cull earlier works and subject the excerpted scholia to
some usually very superficial revision.
The Latin Middle Ages continued the tradition of writing glosses,
scholia and literal commentaries (so called because they follow the wording
of the text, the 'littera1 in medieval terminology). Twelfth-century litera
ture of these kinds often follows the Greek patterns rather closely, even in
details of phrasing, as when paraphrases are regularly introduced by the
words 'quasi diceret' = 'has ei elegen = 'as if he were to say'. Thirteenth-
century university teaching produced a new kind of commentaries, the
Quaestiones — collections of formal disputations of problems raised by the
text. There is nothing quite like the quaestiones in the Greek tradition,
although, of course, the rules of disputation were inspired and greatly
influenced by the description Aristotle gives of'the dialectical game' in the
Topics.
Scholia and commentaries, especially the literal ones, contain a good
deal of trivial material; and the fact that they follow the arrangement of
Aristotle's text means that if the commentator has some theory of his own
we rarely get a coherent exposition of it, but must piece it together from
fragments occurring in unconnected scholia. Nevertheless, they merit close
attention because they played a major role in shaping the late ancient and
medieval students' conceptions of Aristotle's logic. Few, if any, would try
to read the bare text of the Organon, unaided by any commentary. At the
very least, they would use an annotated copy with some scholia and glosses.
The source of medieval scholastic logic 103
A survey of the history of Greek scholasticism

The Greeks had written commentaries on classical authors before the
second century A.D., but scholasticism did not really conquer philosophy
till then. The fourth and third centuries B.c. were the age of the great
innovative thinkers, Plato, Aristotle, Epicurus, Zeno, Chrysippus. The
period from about 250 B.c to A.D. 100 was characterised by a split between
higher philosophical education, which was supposed to be instruction in
the dogmatic system of one of the traditional 'sects' (Epicurean, Peripatetic,
etc.), and a more elementary level of instruction, where doctrines of
different origin coalesced. The sects had their last days of glory in the
second century when they even won a kind of official recognition through
the institution of public chairs of Platonic, Peripatetic, Epicurean and Stoic
philosophy. 2 But during the same period the foundations of the sects were
undermined by a public demand for non-sectarian instruction in all branches
of philosophy and on the basis of the classics of each branch. In practice this
meant that Plato should reign in metaphysics with Aristotle providing the
principles of such auxiliary disciplines as logic. Such pieces of Stoic doctrine
as had become the common property of educated people were to be pre-
served, but the system as a whole was discarded, as was Epicureanism.
Within a relatively short time the rich fathers who would pay a philoso-
pher to read central Aristotelian and Platonic texts with their sons must
have become more numerous than those who would pay for traditional
courses of Hellenistic Peripateticism or Stoicism. The old sects were almost
completely wiped out before the end of the third century. Whenever later
writers polemicise against Stoics, for instance, they are fighting the
shadows of men long since dead. Even the books of Stoic philosophers
disappeared. Simplicius informs us that in the early sixth century most of
them were no longer available. 3
The logic adopted within this educational reform was Aristotelian and
scholastic. Its Bible was the Organon, the received interpretation of which
was masterfully developed in the commentaries of Porphyry (ca. 234-304).
Porphyry had few, if any, original opinions in logic, but he knew how to
combine apparently disparate pieces of commonly accepted doctrine in an
intelligent way. His successors appreciated this achievement, but not the
length of some of his commentaries, and so they soon boiled them down to
what they thought were the essentials. N o later philosopher undertook a
radical revision of the Porphyrian interpretation, which still looms large in
2. Cf. Lynch 1972, pp. 170-1.

3. See Simplicius 1907, 334.1—3.
104 The old logic
the works of Ammonius (ca. 440—ca. 520), Boethius (ca. 480—524), his near-
contemporaries Philoponus and Simplicius, and 'Alexander' (a shadowy
figure, much later than Alexander of Aphrodisias (ca. 200), whose name he
borrowed; see below), all of whom were to influence the Latin Middle
Ages.
After about 550 the study of logic suffered a decline in the East (as in the
West). O n e of the few memorable events was external: the tradition of
Aristotelian scholasticism was taken up by the Arabs, a development that
was to be important for the West in later times when Arabic books could
be used, because Arabic philosophy had sprung from the same Greek
scholasticism as had given rise to Latin scholasticism. There were periods of
revival, as in the early twelfth century when a rising interest in the
Organon may have helped James of Venice find commentaries he could
translate into Latin. But Byzantine 'logicians' were more nearly philol-
ogists than logicians and never rose above rehashing earlier scholia and
compendia. With the possible exception of 'Alexander's' commentaries,
which cannot be securely dated, no Byzantine works influenced Latin
logic.
The programs of ancient courses of logic
Scholasticism had swept the old sects away but not the old traditions of
elementary instruction, in which a course like the following appears to
have been standard: I 1. Theory of description: division, definition, pre-
dicables; 2. Semantics of uncombined expressions; 3. Structure of the
proposition, quantity, quality, modality, square of opposition, conversion
etc.; 4. Syllogistic: 4.1 Categorical syllogisms; 4.2 Hypothetical syllogisms;
5. Fallacies/ The commentators tried to demonstrate that the Organon
provided a full course as follows: II 2. Categories, 3. De interpretatione,
4. Prior Analytics, 5. Sophistici elenchi. It was difficult to fit the Posterior
Analytics and the Topics into the programme; almost all commentaries on
the Organon contain a discussion of the right 'order of reading' (taxis tes
anagnöseös),5 probably reflecting a serious uncertainty in early scholasticism
about where to put those two works; however, most people agreed to
4. Cf Cicero, Brutus 41.152; id.. De finibus 1.7.22; id.. Orator 32.115; id., Tusculatiae 5.25.72;
Maximus Tyrius, Phiiosophumena 1.7.8; Sextus Empiricus, Pyrrhoneae hypotyposes 2.213 & 229.
Galen's institutio logica shows that syllogistic was commonly divided into categorical and hypo-
thetical. Albinus' Didascalkus (Albinus 1858, Ch. 3, p. 153, and Ch. 5, p. 156) and other sources
indicate that induction was also a standard topic, though hardly an important one.
5. E.g., John Philoponus 1909, pp. 2-3.
insert them after 4. as 4.a and 4M respectively. The instant success of

Porphyry's Isagoge must have been partly due to the simple fact that it
could (almost) fill the gap before the Categories. Porphyry was not the first
to try to stop the gap. In the first century B.C. Andronicus of Rhodes, the
6
first editor of the Organon, had composed a book On Division but, for
whatever reason, his work did not become a classic, and so it was still felt
that something was missing until Porphyry supplied it.
Course II was for advanced students. After the second century the
contents of course I became more Aristotelian, but the skeleton remained
and was intact in Byzantium a thousand years later.7 It looks as if a
specifically Western variant of courses I and II arose in the fourth century, a
variant which not only omitted items 4.a-b but also 5, while introducing
Cicero's Topica and its doctrine as item 4.C8 The literary activity of
Boethius testifies both to a will to supply the needs of this Roman course
and to a wish to expand it by adding 4.a-b and 5.
The books that transmitted ancient logic

Boethius' efforts to expand the course met with no immediate success.
With a few exceptions it was the 'Roman course' that determined which
Latin books on logic were handed down to the Middle Ages. They were
these:
A Basic books Β Commentaries C Compendia and monographs
I. Porph. Isagoge Two, by Boethius (a) Marius Victorinus De definitionibui
(b) Boethius De divisione
2. Arist. Categories One, by Boethius (a) Augustine De dialectica
(b) Ps.-Augustine Decern categoriae
3· Arist. De interpret. Two, by Boethius Apuleius Peri hermeneias
4 Arist. Prior Analyt. None 9 (a) Apuleius Peri hermeneias
(b) Boethius De syllogismis categoricis
(c) Boethius De hypotheticis syllogismis
4.a None None None
4b Arist. Topics None None
4.C Cicero Topica One, by Boethius Boethius De topicis differentiis
5· Arist. Sophistici el. None None
6. Cf. Moraux 1973, pp. 120-32.

7. See Nicephorus Blemmydcs' Epitome logica (from ca. 1260) in Migne's Patrologia graeca 142,
Anonymous 1929 (from 1007), plus the remarks of Michael Psellus 1926 (eleventh century) in his
Chronogtaphia III.3 and of the anonymous hagiographer in Vita beati loannis Psichaitae (ninth
century) (Anonymous 1902) §4.
8. Cf. Jerome Ep. 50 in PL 22:513 (the course he refers to may have included the Posterior Analytics as
well). For Roman traditions cf. Hadot 1971 and Pfligersdorffer 1953.
9. Except for the Philoponean scholia (translated by Boethius?) in Aristoteles Latinus 1962. III.4. Cf.
Ebbesen 1981a.
ιο6 The old logic
To make the C-list complete we must add Book IV of Martianus Capella's

De nuptiis Philologiae et Mercurii, which covers ι - 4 .
Apuleius' Peri hermeneias10 is the oldest of the C-books, dating from the
infancy of scholasticism (second century) and containing much that smacks
of Stoicism. It starts with a consideration of the kinds of utterances (orders,
wishes, etc.) and picks out the indicative propositions as the one kind
relevant to logic. Having divided that sort of proposition into the species
categorical and hypothetical, Apuleius goes on to deal with quantity, quality,
etc., and ends with a chapter on categorical syllogisms; a concluding
chapter on hypothetical syllogisms may have been lost. The sources are
unknown but large parts look like adaptations of a Greek text.
Victorinus' De definitionibus1 ' is a mid-fourth-century treatise which
is really more closely associated with rhetoric than with logic. The
'definitions' dealt with are all kinds of identificatory descriptions. The
opuscule is studded with Greek technical terms, and other considerations
also point to a Greek main source, 1 2 but it cannot be identified.
The Pseudo-Augustine's Decern categoriae13 is a late-fourth-century (ad
apted) translation of some dull Greek compendium of Aristotle's Categories
emanating from the environment of Themistius (317-88).
Augustine's De dialectica1* is a fragment of an unfinished survey of the
liberal arts begun in 387. It is a most interesting study of semantics with
particular attention paid to ambiguity. It owes much to the Stoic tradition,
which may be one reason why it ceased to be important in the twelfth
century when it became clear that it was difficult to produce a synthesis of
De dialectica and Aristotle's Sophistici elenchi.
15
Martianus Capella's De nuptiis Philologiae et Mercurii is a compendium
of the liberal arts, written in an extremely precious style and dating from
the late fifth century. Book IV deals with logic, starting with the pre-
dicables, definition, and division and ending with the hypothetical syllo
gisms. Martianus, like Apuleius and Augustine, preserves several features
of pre-scholastic systematics and terminology.
We have no Greek compendia of logic from the fourth or fifth centuries,
but it can hardly be doubted that such as existed were much more similar to
10. About this work and its fate see Sullivan 1967.
11. For a detailed discussion of this work see Hadot 1971.
12. Thus De definitionibus, ed. Stangl p. 29 ( = 359 Hadot) is to be compared with Blemmydes
Epitome logica, PC 142:6930.
13. On this work and its fate, see Minio-Paluello's preface to Aristoteles Latinus 1961b, 1.5.
14. On the work and its fate, see the introduction to the 1975 edition.
1 J. On this work and its fate, see Stahl 1971; Hadot 1971, p. 196 and elsewhere.
the Aristotle commentaries in systematics and terminology than are the

Latin compendia. The conservative character of the latter is presumably
due to the fact that most teaching of philosophy was done in Greek, even in
the West. Latin treatises were composed at long intervals only, and so such
old books as Varro's Disciplinae (first century B.C.) continued to be influen-
tial long after they had become antiquated by Greek standards.
Boethius, who was thoroughly acquainted with contemporary Greek
philosophy, made an astonishing effort to bring the Latin philosophical
library up to date. His works 1 6 all date from the first quarter of the sixth
century. De divisione deals with various ways of drawing systematic
distinctions. This short treatise is firmly rooted in the Greek tradition and is
probably an adaptation of a Greek treatise on the subject.
The fragmentary Introductio ad syllogismes categoricos appears to be the
first half of a scheduled revised and enlarged edition of De syllogismis
catégoriels which together with De hypotheticis syllegismis-ixaûà_j3iû%iy
almost any student's hunger for syllogistic. These are very thorough text-
books, starting with the elementary notions (terms, propositions, quality,
quantity, conversion, etc.) and moving on to lengthy reviews of syllogistic
forms, questions of validity, etc. The material is all traditional, and it is
probable, though not provable, that Boethius used Porphyrian treatises as
his models.
The treatise on hypothetical syllogisms is perhaps the most interesting.
A hypothetical syllogism must contain at least one hypothetical pre-
miss, a molecular proposition made up of categorical propositions. The
basic connectives are 'si' ('if'), 'cum' ('when/since'), and 'auf ('or').
Hypothetical syllogisms are said to be secondary in relation to categorical
syllogisms because a hypothetical premiss has categorical propositions for
its constituent parts and because, if the truth of a hypothetical premiss is
questioned, it must be established by means of a categorical syllogism. Thus
the 'inferential power' (vis consequentiae) of the hypothetical major premiss
and the 'power of the conclusion' (vis conclusionis) of the hypothetical
syllogism depend on the categorical syllogism (Hyp. Syll. I. II. 3 - 5 , p. 212
Obertello).
Boethius' syllogistic treatises were very influential in the twelfth cen-
tury, but after the early thirteenth they were neglected. Peter of Spain and
later writers of compendia do not even summarise Boethius' doctrine of
the hypothetical syllogism. Hypothetical reasoning later became a subject
16. See Obertello 1074, which includes a rich bibliography that can be supplemented with the
bibliography in Stump 1978.
ιο8 The old logic
of discussion in treatises on consequences (for instance, the chapter 'De

consequents' in Albert of Saxony's Perutilis logica), but the authors of such
treatises did not take their cue from Boethius' work.
De topicis differentiis is a monograph on proof by means of universal
axioms. It expounds the theory in the light of which Boethius had inter
preted Cicero's Topica. Boethius' book, which, though dependent on
Greek theorising (see below), is not an adaptation from the Greek, was the
subject of intense study in the twelfth century and still much read in the
thirteenth. A chapter on the subject matter of this treatise ('De lotis') was a
stock item of compendia of logic till the end of the Middle Ages.
As for the B-books, Boethius' commentary on Cicero's Topica contains
nothing of interest to logic that is not in De topicis differentiis. His other
commentaries follow Greek patterns quite closely without being mere
translations; they depend on the same sources as their contemporary Greek
counterparts, to which they are in no way inferior, and in particular on
Porphyry, whose views Boethius embraces even more emphatically than
many Greeks had done.
About 1130 James of Venice augmented the list of B-books by translat
ing the commentaries of 'Alexander' on the Posterior Analytics and the
Elenchi. His translations are not extant today, but quotations and reports of
their scholia occur in many extant books. 1 7 The medievals thought both
commentaries were by Alexander of Aphrodisias (ca. 200), but they were
wrong. Neither of the two works can have had the Aphrodisian for its
author, and they are even likely to have had two different authors. The
commentary on Posterior Analytics I (it is uncertain if there was one on
Book II) was virtually or completely identical with Philoponus'. If it was
not simply his, it was a later compilation consisting mainly of extracts from
his. The Elenchi commentary resembled most of all the one by James'
contemporary (and acquaintance?) Michael of Ephesus, 1 8 but it was not
simply identical with it, nor with any other extant Greek collection of
19
scholia, though it had similarities with some of t h e m . It may have been
James' own compilation from Michael and other sources. If not, it was
17. See Ebbesen 1976a, where the fragments relating to the Posteriora are edited. The fragments of the
Elenchi commentary are edited in Ebbesen 1981b, vol. II.
18. The first version of Michael's commentary is contained in the MSS of Wallics' 'second class' (see
CAQ II.3, pp. XVIII-XXII); the final (revised and enlarged) version has been edited by Wallies in
CAG II.3; extensive excerpts from the commentary on an intermediate stage occur in MSS Paris
gr. 1917 ff. 539'-5SOv and Paris gr. 2019 ff. i87 v -203 v . Cf. Ebbesen 1981b, vol. I, ch. V.14.
19. So in particular with Commentarium II, contained in Ms Vat. Barb. gr. 164 ff. 235"—254* and other
MSS. This collection antedates Michael, who culled many scholia from it.
probably written between about 900 and iioo. At any rate it consisted of
Greek scholia, several of which had originally been extracted from com-
mentaries on other books of Aristotle's. Thus, whichever the date of the
compilation, a large part of the materials dated back to the sixth century
and some items even further back. The commentaries of'Alexander' had
the important role of initiating the twelfth century into the lore of the
Posterior Analytics and the Elenchi, and their influence - especially that of the
Elenchi commentary - was great and lasting although mostly indirect. The
last copies of James' translations may have been destroyed in the early
fourteenth century or even earlier; in fact, most of the twelfth- or
thirteenth-century people who quote, cite, paraphrase, or echo 'Alex-
ander' betray in one way or another that they never held his books in
their hands. But the works had been quarried so soon after publication of
the translations that more or less precise accounts of 'Alexander's' tenets
and examples, with or without indication of source, began to circulate in
compendia and scholia and then to be handed down from generation to
generation together with other traditional material.
The influence of ancient grammar and rhetoric

The second century A.D. wanted non-sectarian instruction on the basis of
classical texts not only in logic but in other disciplines as well. Grammar
(linguistics) was one of them, but since it was only a few centuries since
grammar had become emancipated from its parent, Stoic logic, 20 there was
no early writer qualified to become an auctor.21 Consequently second-
century grammarians such as Apollonius Dyscolus and Herodian them-
selves became classics. Apollonius was the author Priscian chose for his
guide in the early sixth century when he wrote that monumental grammar
of Latin that was to mean so much to medieval Latin grammar and logic.
Ancient rhetoric, too, had close ties with logic. Surviving manuals of the
art and Victorinus' commentary on Cicero's De inventione exercised some
relatively unimportant influence on twelfth-century logicians.
Ancient theories that influenced the medievals

Such mediators as Boethius, 'Alexander' and Priscian let the Middle Ages
catch a glimpse of a wealth of ancient theories. Each of these theories made
its contribution to the development of Latin logic and it is not possible even
20. Cf. Pinborg I975d; Frede 1977 and 1978.
21. I have to except Dionysius Thrax, but his Arsgrammatica (ed. Uhlig, Grammaticigraeci 1.1) is a very
jejune work, and most of it may be spurious (cf. Pinborg I975d).
no The old logic
to list them all. For the sake of illustration I select three that I consider
particularly important, viz. Apollonian grammatical semantics, the proof-
procedure of Galen and Themistius, and Porphyrian logical semantics.
Apollonian grammatical semantics

Priscian's grammatical theory is Apollonian, as he himself admits. 22 Like
all ancient grammars, that of Apollonius Dyscolus was not historical. Of
course, some historical observations would be included, but the general
attitude was that the corpus of normative authors should be treated as evi-
dence of one and the same language - Greek or Latin, as the case might be.
The elements of the discipline were ordered according to complexity.
The minimal units were called elements, their symbols, letters. Elements
combine to form syllables, syllables combine to form (minimal) expres-
sions (Greek 'lexeis', Latin 'dictiones'; in the following I shall use the loose
rendering 'words'). Words are signs of intelligible significata which may
combine to form complete meanings signified by grammatical sentences.
Rules determined the constructional possibilities on each level of com-
plexity, and some effort was made to make the rules on different levels
conform to the same patterns. Considered as a constituent of sentences, a
word is a part of speech (Greek meros tou logou, Latin pars orationis) and it
belongs to a word-class (also called 'part of speech') the members of which
share certain constructional 'attitudes', as it were, some of which regard the
surface form of the sentences only, while others have to do with semantical
components, both significates and consignificates. The Greek name for
some, and perhaps all, such attitudes was 'scheseis'; Priscian seems to have
had no technical term for them. Compatibility of attitudes is required for
grammatical construction. For instance, verbal forms other than the in-
finitive not only signify an 'activity' (Greek 'pragma', mistranslated into
Latin as res) but also consignify some definite mood, number, and the like,
these 'accidents' (Gk. 'parhepomena', Lat. 'accidentia') being specified by
morphological devices. Grammaticality (Gk. 'katallëlotës', Lat. 'congruitas')
requires, for instance, that the subject term and the verb have compatible
attitudes, which implies the same or at least compatible semantic com-
22. Priscian 1855-9, XVII.1.1; cf. Schneider in Grammatici Graeci II.}, pp. Ill—IV. Some crucial
passages for the facets of Apollonian theory expounded below; Apollonius, De syntaxi I, §§1-2,
pp. 1-3; III, §§24-6, pp. 290-2; §§54ff, pp. 3i9ff.; IV, §15, p. 448. More may be found via the
index in Gramm. Gr. II.3 under 'semaine', 'sëmainomenon, 'sussêmainô', 'sundeloô', 'schéma', 'schë-
matismos', and the terms mentioned in the text of this chapter. Cf. Ebbesen 1981b, vol. I, ch. IV.
*.I7.
ponents. The complex of semantical components underlying a sentence

is not just an intelligible duplicate of the surface sentence, for several
distinct surface sentences may be equivalent, which means they share a
common semantical base. For example, Ί wish that you would write' -
that is, wishing mood + second person + singular number + present
tense + active disposition + the activity 'write' - may be expressed by
means of three different Greek sentences, namely 'euchomai se graphein',
'graphois' and 'eithe graphois'. In the first sentence the wishing mood is
expressed by means of the modal verb 'euchomai' = Ί wish', the second
person singular by means of the pronoun 'se' = 'you', the remaining
semantical components by means of the infinitive 'graphein = 'to write'.
In the second and the third sentences all the information is contained in the
second person singular present optative of the active voice 'graphois'; the
modal particle 'eithe' in the last sentence does not provide any information
not contained in 'graphois', it just serves to emphasise the wishing mood.
The relations between the semantical components and the surface sen
tences may be illustrated as follows:
1. euchomai se graphein
(I wish) (you) (to write)
MOOD PERSON NUMBER TENSE DISPOSITION ACTIVITY
wishing second singular present active write
2.-3 . (eithe) graphois

( would that ((woul d that) you write)
The theory of inference in the literature on the topics

From Boethius' two works on topics the medievals inherited a theory of
inference which is perhaps best called 'axiomatic topics', since in this theory
the notions of topos and axioma had coalesced.
In Boethius' Latin the Greek terms topos and axioma had become locus
and maxima (propositio), respectively, but as there is no doubt which terms
underlie his Latin,2 3 we may retain the original nomenclature and say topos
23. That 'maxima propositio' translates axioma is evident, not only because Boethius adorns the
'maximae propositiones' with epithets usually given to Aristotelian axioms, but also because he
actually translates axiomata at Arist. Top. VIII, c. 1, I55 b i5 as 'maximae propositiones' (see
Aristoteles Latinus 1969, V. 1-3, p. 156). Cf. Ebbesen 1981b, vol. I, ch. IV. 3.5.
112 The old logic
and 'axiom' (leaving 'topos' untranslated in order to be able to reserve the

term 'topics' for the discipline or theory that deals with topoi). Notice that
in the following 'axiom' is used in the sense in which the authors discussed
used it, not in any modern sense.
In the theory Boethius presents, every inference owes its cogency to an
axiom. Arguments of the form 'non-axiomatic premiss(es) .'. conclusion'
are abbreviations of 'axiom, non-axiomatic premiss(es) .'. conclusion'.
Boethius describes an axiom as a self-evident, primitive, general pro-
position. His axioms are, in fact, laws of inference, whose only constants,
besides 'if... then', 'and', etc., are relations (be predicated of, be the genus
of, be the opposite of, etc.); the relata are variables (though there may be
restrictions of domain; for instance, they may all have to be good things).
The non-axiomatic premisses and the conclusion are assertions of the
antecedent and the consequent, respectively, of the axiom, with proper
constants substituted for the variables.
The implication of the Boethian theory would seem to be that all proof
proceeds, implicitly or explicitly, by instantiation and detachment and, as
some medievals saw, 24 that a categorical syllogism is not anything sui
generis, as it depends on a law of inference of the same type as the ones that
licence inferences involving other relations than plain predication.
Boethius himself may not have seen that far. In fact, he turns most axioms
into superficially categorical propositions by putting 'whatever' and
metalinguistic terms, such as 'the opposite', 'the genus', in name-positions.
According to Boethius, the topoi of Aristotle's Topics are axioms.
This is a most startling claim in view of the description of the axioms as
self-evident, etc. How do dialectical proofs differ from demonstrative ones
if they have such strong backing? Boethius does distinguish between dem-
onstrative, dialectical, rhetorical, and sophistical arguments, but he does
not solve the problem of differentiation by reserving the term 'axiom' for
the propositions which demonstrative proof rests on; instead he extends the
notion of axiom to cover not only demonstrative axioms, which are
inherently necessary, but also the propositions on which weaker proofs (or
pseudo-proofs) rest; thus a dialectical axiom need only be generally known
and inherently probable, not self-evident and inherently necessary.
Boethius further notes that the loci of Cicero's Topica are not topoi in the
sense of axioms; they are classes or labels of groups of axioms. (This
distinction between two sorts of topoi is reflected in medieval Latin ter-
24. See Pinborg 1969.

. minology where an axiom-iopos is called locus maxima, whereas a

'Ciceronian' topos is called locus differentia (maximae).
Boethius stands at the beginning of a tradition because much medieval
thought about inferences was inspired by his works on topics. But he also
had a long tradition behind him. Not all stages in the development are
clear, but the main ones seem to have been these: Boethius produced a
compromise between Cicero's and Themistius' conceptions of topics. The
axiomatic topics came from Themistius, who had produced (or inherited) a
combination of Galenic ideas about axiomatic proof with a Theophrastean
conception of what an Aristotelian topos is. Galen had identified axioms in
the sense intended by the Stoic Posidonius with Aristotelian axioms.
But Aristotle left no clear definition of a topos, and his statements about
axioms create difficulties of interpretation. One of the most serious is, to
borrow Jonathan Barnes' words,25 that 'Aristotle is clear that principles' -
including axioms - 'function as premisses of demonstrations; but it is not
easy to see how they can do so. A typical axiom is the Law of Excluded
Middle; and that is not expressible in syllogistic form.' Aristotle never
compares the functions of topoi and axioms so as to make similarities (if
there were any in his view) and differences clear.
Theophrastus, Aristotle's successor as head of the Lyceum, made an
effort to clarify some aspects of his predecessor's doctrine. He provided a
quasi-definition of an axiom: 'An axiom is a tenet, either about things in
related domains, such as "if equals [are subtracted] from equals, [the
remainders will be equals]", or about everything -without restrictions, such
as "the affirmation or the negation".' 26
He identified the topoi of Aristotle's Topics as those propositions that state
laws of inference, and defined a topos as a 'starting point (arche) or element
from which we get the particular starting points [i.e., which suggests to us
the premisses we need], definite in outline [because it states certain re-
lations], indefinite as far as the particulars are concerned [because it does not
identify the relata]'.27
He simplified Aristotle's classification of topical problems and showed
how Aristotle's arrangement of the topoi could in some places be made
more systematic.28
Alexander of Aphrodisias (ca. A.D. 200) has an explanation of the func-
25. Barnes 197}, p. 104.

26 Theophrastus 1973, frgm. 33.
27. Theophrastus 1973, frgm. 38.
28. Theophrastus 1973, frgm. 40; Thcophr. apud Themistius Top. in Averroes 15fi2-74c, ff. 101-13.
114 The old logic
tion of a topos that is probably Theophrastean. If, he says, 29 you are

inquiring whether the good is beneficial, the topos 'if something has a
property, its opposite will have the opposite property' gives you the
premiss 'if evil is harmful, the good is beneficial', which is potentially
included in the topos and derives its credibility from it. But Alexander is
clear that the topos itself is not part of the dialectical argument, which
consists of the premiss 'if evil etc.', an assertion of its antecedent, and has its
consequent for its conclusion.
There is no evidence that Theophrastus called any topos an axiom or any
axiom a topos. There is good reason to believe he never did so. 30
Theophrastus was succeeded by Strato (d. 269/8). He too wrote treatises
connected with topics, but the extant remains are negligible. 31 To all
appearances he was the last philosopher to show any interest in the subject
before the first-century B.C. revival of Aristotelian studies. Aristotle's
Topics was completely or almost completely forgotten for a couple of
centuries. 32 O n e of the first to have a look at it after the interval was Cicero.
He wrote a Topica which pretends to be a sketch of the Aristotelian
discipline. In Cicero there is no trace of the Theophrastean conception of a
topos. T o Cicero a topos is a notion with which certain standard methods of
inference are connected; the study of the topics is useful, because consider-
ation of the proposition one wants to prove may suggest such a notion and
hence the way to prove the proposition, but Cicero does not state the
methods of inference in propositions; they are implicit in examples of
arguments depending on each topos.
Cicero also presents a systematic list of topoi. The fact that the list, in spite
of similarities with the one in Aristotle's Rhetoric II.23, cannot have been
simply extracted from the Rhetoric, or from the Topics, suggests a
Hellenistic origin. The idea is supported by the occurrence of a similar list
in Themistius' paraphrase of the Topics (see below), and by its structure.
The list may have had some sort of precursor in Theophrastus' works, but
29. Alexander of Aphrodisias 1891, pp. 126-7.
30. Alexander of Aphrodisias seems to have had first-hand acquaintance with Theophrastus' works.
Yet, he never mentions the idea that topoi are axioms. It is scarcely credible that he would have
passed over such a remarkable doctrine in silence. Galen also had first-hand acquaintance with
Theophrastus* works; but he never suggests an elevation of topical propositions to the rank of
scientific axioms. Averroes 1562-74C, ff. 27-8, explicitly contrasts Themistius' axiomatic inter-
pretation of the topoi with Alexander's and Theophrastus'. Hence, at least, Themistius did not
invoke the authority of Theophrastus for his interpretation.
31. Strato, 1967-9, frgm. 20-31.
32. Of course, much is unknown about the period 2SO-50 B.c., but it is indicative that Sextus
Empiricus, who is usually a copious source for philosophical discussions in that period, does not
waste words on topoi.
Theophrastus was hardly Cicero's direct source. One possibility is that

Cicero had his list and his conception of topics from his rhetorical training
(Themistius was also a rhetor). A superficial inspection of Aristotle's Topics
then gave him the impression it was a longer treatise on the sort of topics he
knew from rhetoric. Another possibility is that Andronicus of Rhodes, the
first-century editor of Aristotle, had written a kind of companion or
introduction to his edition of the Topics, giving in it a systematic list of the
topoi inspired, perhaps, by early Peripatetic treatises and/or rhetorical
tradition. Cicero then took Andronicus' opuscule to be a summary of
Aristotle's Topics. Wallies suggested in 1878 that Cicero's source was his
old teacher Antiochus of Ascalon. None of the three hypotheses has any
real support in the sources.33 In any event, Cicero's Topica is not the sketch
of the doctrine of Aristotle's Topics it purports to be.
Aristotle probably took his term 'axiom' from geometry, and it con
tinued to be used in geometry. The propositions which in modern editions
of Euclid are called 'common notions' were generally referred to as
axioms. Geometry was the science par excellence in antiquity, the one all
dogmatic philosophers admired for its strict methodology and which
sceptically inclined thinkers were fond of attacking (for if geometry was a
fraud, then surely all dogmatic teaching was so). The Stoic Posidonius (ca.
33. Cicero sketches the doctrine of loci ( = topoi) both in Deoratore II.38.ioosqq. and in his Topica.
His division into technical and non-technical loci comes from rhetoric. But much smacks of
Peripatetic logic, such as the division of the technical loci into ( ι ) such as are derived from the thing
itself, and (2) such as are derived from its concomitants (cf. Theophrastus' classification of topical
problems into definitional and accidental in Theophrastus 1973, frgm. 40). The subdivision of (1)
into loci a definiliotte, a partium enumeratione, a notatione, makes the general scheme similar to the
Peripatetic classification of division known from Boethius' De divis.; John Damascene (see
Moraux 1973, pp. 129-30); and Blemmydes' Epitome logica, Ch. II, PG vol. 142, coll. 701-8
(cf. Sextus Empiricus, Pyrrhoneae hypotyposes 2.213). Themistius' classification of topoi is known
from Boethius, De top. diff., II—III. The common source of Cicero and Themistius might be
Theophrastus whose works on topics Themistius certainly drew on, directly or indirectly, as is
evident from the frequent collocation of the two men's names in Averroes' Expositio media
Topkorum (see, in particular, Averroes 1562-74C, ff. 101 -13). But then the silence of Alexander of
Aphrodisias about such a salient feature of Theophrastus' topics is hard to explain. The theory of a
rhetorical source receives little support from extant manuals of the art. There is no evidence that
Andronicus wrote any companion to the Topics, and Alexander of Aphrodisias' silence is a
weighty argument against assuming he did so. Wallies' proposal is, in my estimation, the least
probable of all. He considered Cicero's system a concoction of Arist. Rhet. II.23 and Stoic
doctrine, and thought such a mixture could only stem from Antiochus. But (1) he ignored
Themistius; (2) his reasons for saying that the system underlying Cicero's list is Stoic are, in my
opinion, insufficient; (3) there is no need to postulate with Wallies that all the logical matter in
Cicero has one source; if Cicero had the systematic list of topoi, he could easily supply comments
on various logical questions drawing on his own knowledge of the subjects, and what he knew
would often be of Stoic origin; (4) there is no independent evidence that Antiochus ever discussed
topics.
ιι6 The old logic
135-50 B.c.) seems to have been worried about the fact that neither
traditional Stoic nor Peripatetic syllogistic could be used to explain the
validity of proof procedures characteristic of geometry. T o defend them he
proposed that relational syllogisms, i.e., such as contain propositions with
polyadic predicates, are valid 'by the force of an [implicit] axiom'; thus Ά
and C are each equal to B; therefore A and C are equal to each other' is valid
by force of the axiom 'those that are equal to the same are equal to each
other'. He probably also held that Peripatetic categorical syllogisms are
elliptic arguments because they do not state the principle which permits the
move from premisses to conclusion. 3 4
Admiration for geometry and inspiration from Posidonius and Aristotle
met in Galen, who was convinced that traditional Hellenistic syllogistic,
Stoic and Peripatetic alike, was an insufficient and largely useless tool for
practical men. He concluded that the thing to do was to elaborate a general
theory of proof modelled on the practice of geometers and meeting, in
general, the requirements for scientific procedures set forth in Aristotle's
Posterior Analytics. He therefore stressed the importance of having explicit
scientific and demonstrative axioms, capable of serving as premisses in
arguments, and having the quality of self-evidence; and he insisted that
such axioms be strictly separated from the similarly functioning, but less
reliable (or even deceptive) propositions that belong to the spheres of
dialectic, rhetoric, and sophistic. He also stressed the importance of having
a method by which to discover the premisses needed to prove a conclusion,
a method analogous to the 'analysis' of geometry. If a geometer wanted to
prove the equality of A and C, he would take from his stock of axioms one
about equality, viz. 'those that are equal to the same are equal to each
other'; he would see that he needed a magnitude to which both A and C are
equal; he would find it and reason, 'Those that are equal etc.; A and C are
each equal to B; therefore A and C are equal to each other.' Galen all his life
wanted a method of proof along these lines, an axiomatic super-syllogistic
embracing the best of traditional Peripatetic and Stoic syllogistic, relational
syllogistic and all. He propounded his ideas in a large number of treatises,
but apparently it was only near the end of his life that he found a pregnant
formulation of his own solution. While writing a short manual of logic for
some friend (the Institutio logica) he consulted some work of Posidonius',
34. For Posidonius' views, see Kidd 1978. T h e attack o n Peripatetic syllogistic which I tentatively
connect with the name of Posidonius is evidenced by the defence and counter-attacks of
Alexander of Aphrodisias 1883, pp. 21—2 and 334—5; and 1891, p. 14. For ancient polemics about
syllogistic, cf. Frede 1974.
which suggested to him the idea that the principle justifying relational
syllogisms could be extended, so that all syllogistic validity might be said to
derive from the force of an axiom (implied or stated).35
Galen's theory had qualities the second century could appreciate: com-
prehensiveness, explicitness, simplicity. It was a unifying theory, not one
that would force people to discard procedures that had won general
approval. But for the theory to survive in a scholastic age one more thing
was needed; a place must be found for it in the Aristotelian curriculum. The
solution was to develop it in commentaries on the Topics. Galen had not
thought of his axioms as topoi or vice versa (nor of his discovery procedure
as topical); yet, it was fairly obvious that his axioms resembled topoi in the
Theophrastean sense; somebody must have seen this, and he may even have
been guided by Galen himself, for in the Institutio logica Galen has a remark
to the effect that comparative syllogisms (hoi kata to mallon) are not really
different from the ones he calls relational.36 Reasoning about comparisons
is treated in Aristotle's Topics.
It is not known who first described the topoi as Galenic axioms. There is
still no trace of the doctrine in Alexander of Aphrodisias' commentary on
the Topics, written while Galen was still alive or shortly afterwards. But
Themistius (317-88), probably following an older commentator, pro-
pounded it in the initial chapter of his paraphrase of Topics II, though
apparently without acknowledging any debt to Galen. Themistius also
introduced the Hellenistic classification of topoi.31
35. Galen's logical works were many and some, in particular the monumental On Proof, continued to
be read for centuries. Now they are all lost except Df captionibus (which contains an interesting
demonstration of Galen's discovery procedure) and Institutio logica. The information they give
about his views must be supplemented with casual statements about logical matters in his non-
logical treatises. Much of the material has been gathered by Müller 1897. Important sources are:
De animi passionibus II; Deplacitis Hippocratis et Piatonis, ed. Müller, II & VII, pp. s88f; De propria
libris Ch. XI. The following titles of lost writings arc revealing of his interests: 'On Proof-
discovery'; 'On the Premisses that are Omitted in the Verbal Formulations of the Proofs'. In the
Institutio, chh. 16-18 are particularly important. For Galen's dependence on Posidonius, see Kidd
1978.1 think, however, that Kidd tends to overrate Posidonius' influence on the early Galen. It is
evident from the earlier works that the thoughts developed in chh. 16-18 of the Institutio mark no
radical change of attitude on Galen's part, but it looks as if he got a bright idea while writing the
work. He starts with a fairly traditional exposition of traditional doctrines; then, in Ch. 16, he
adds as an afterthought that there are also relational syllogisms and goes on to deal with them; in
Ch. 17 he announces it as a recent discovery of his that all demonstrative syllogisms depend on
axioms in the same way as the relational syllogism (notice that in 17.7 he is careful to point out
that he uses 'axiom' in the sense of'self-evident proposition*) and starts to show that this is so; at
the end of Ch. 18 he reveals that he owes the inspiration to Posidonius.
36. Galen 1896, Ch. 16.12, p. 41; Ch. 18.1, p. 45.
37. Themistius' paraphrase of the Topics is not extant, but Boethius and Averroes refer to and echo it
(see Stump 1974, pp. 89-91). It is particularly important that Boethius' definition of the topoi as
ιι8 The old logic
Boethius saw it as his task to update the Latin philosophical library.

Cicero's Topica was already a classic and Boethius would not try to replace
it. It could be modernised through interpretation. Accepting the view that
Cicero reviews the doctrine of Aristotle's Topics, but realising that Cicero's
topoi were not axioms (in any sense), he introduced the Themistean
conception of a topos and established harmony between Cicero and the
Aristotle of Themistius by viewing Ciceronian topoi as notions that group
related axioms (topoi in the sense he attributed to Aristotle) together. He
then supplemented Cicero by stating explicitly which axioms Cicero
implicitly referred to. Finally, he endeavoured, not without success, to
show that there is not a world of difference between Cicero's and
Themistius* fists of topoi. He did not however, manage to conceal the fact
that the axiomatic proof-procedure had no original connection with
dialectic. When he introduces the notion of an axiom, it gets the epithets of
an Aristotelian-Galenic demonstrative axiom; only secondarily does he
make clear that not all dialectical axioms are that strong.
Thanks to Boethius, the doctrine of axiomatic topics had a future in the
West. In the Greek world it did not survive the end of antiquity. Even
Themistius' paraphrase of the Topics disappeared.
Porphyrian logical semantics
Everyone agrees that Porphyry influenced posterity very much, but little
has been done in modern times by way of reconstructing Porphyry the
logician. The work of Lloyd (1956) is a notable exception. My reconstruc
tion, which is in many ways similar to his, is based on the assumption that
Porphyry was a coherent thinker; on that assumption we may supplement
statements in his extant concise Isagoge (Introduction to the Categories) and his
equally concise commentary on the Categories (extant except for the last
part, o n Chapters 10-15) with doctrines he expounded in his lost major
works insofar as they can be recovered from later authors. The lost works
had a decisive influence on later interpretations of Aristotle, although
axioms and his description of their function in argumentation can be seen to stem from
Themistius (InCic. Top. I, PL 64:1051c-1032A, cf. Themistius in Averroes 1562-74C, f. 28A).
Boethius* examples of arguments with explicit and implicit axioms also stem from Themistius
(they are elaborations of Arist. Top. Ill, c. 1, 116*13-14 and II, c. 2, i09 b 35-8). Given that
Themistius said the topoi are axioms, one should expect him to say that some axioms, at least, are
topoi. In Themistius 1900 (pp. 18-19, cf. pp. 24-5) he does, in fact, seem to say that the most
general axioms ('common axioms', which can serve as axioms of an Aristotelian science only
when restrictions of domain are added) are topoi; but his text is not very clear.
The source of medieval scholastic logic ιίο
people soon ceased to read any but the ones we still have, and few would
know how much of what they read in later authors had a Porphyrian
origin. Thus when late ancient and medieval scholars asked 'What does
Porphyry think?' they would often disagree because they sought the
answer in the concise works only, where he had deliberately simplified
matters. T o mention just one important case, he had avoided introducing
the word 'concept' ('noëma') because he could drive home his main point
(that the Categories is not a treatise ' O n What There Is') without enlarging
on the role of concepts.
Porphyry clearly viewed the Organon as a systematic course of logic, 38
moving from simple expressions in the Categories to compound expres-
sions (propositions) in De interpretatione and to compounds of propositions
(syllogisms) in the Prior Analytics, and thence to the several kinds of
syllogisms in the Posterior Analytics, the Topics, and the Elenchi. He held that
the Categories is about simple expressions qua significative, 39 i.e., about the
structure of reality our language presupposes, which may not be the
structure we would posit as metaphysicians. Language is fundamentally a
tool for communication about the world of experience, logic studies the
way this communication functions and it would be wrong to blame
Aristotelian logic for being bad metaphysics as it was never intended to be
metaphysics at all. Therefore, Porphyry holds, 4 0 Aristotle was right in
considering individuals prior to universals when he established the cate-
gory of substance, though a metaphysician would have to reverse the
order. Similarly, it is irrelevant whether or not there are metaphysical
realities corresponding to the universals of logic. Porphyry believed there
are, but he also thought that logic works in either case. This radical
separation of logic and ontology was not always completely grasped or
accepted by later scholastic authors. Ironically, the medieval preoccupation
with the ontological status of universals arose out of Boethius' comments
(In Isagogen, éd. î", ed. Brandt, I, c. 10, pp. 23S*., ed. 2" I, c. 10, pp. isoff.) on a
passage (Isagoge, ed. Busse, 1 .off.) where Porphyry discards the question as
unimportant in a logical context.
A public language presupposes the identifiability of the concepts one
individual has with the concepts other individuals have, 4 1 and the forma-
38. Cf. Porphyry 1887, pp. 56-8.

39. Porphyry 1887, 58.3fr. & 91.7π".; Porphyry, Cat. ad Gedalium, in Simplicius 1907,10.20-13.18.1
believe that Simplicius' interpretation of Porphyry is substantially correct.
40. Porphyry 1887, p. 91.
41. Porphyry, In De int. in Boethius 1880, I, c. 1, pp. 39-40.
120 The old logic
tion of concepts ultimately presupposes sensation. 42 Porphyry accounted

for the relation between sensation, concepts, and language as follows: Men
perceive the sensible world. The perceptions give rise to structured con-
cepts, such as one with the ordered components A-corporeal, + animate,
+ sensitive, —rational, -f-quadrupedal, A-able to bark. It starts as a concept of
an individual but becomes universal when men discover that several
individuals give rise to it. Somehow the convention is established that to
externalise the concept and cause it to be present in another man's mind
you just have to say 'dog'. And as surely as people will have a rfö^-concept,
they will also acquire one unmarked for the last three features (rationality,
means of moving, way of making sound — in short, the concept animal —
and they will associate a name with it. Marking for certain secondary
(accidental) features, such as having floppy ears or a protruding stomach,
creates fully individualised concepts with which names like 'Snuffy' are
associated. Dogs happen to be conceptualised as substances, but concepts of
and words for qualities, quantities, etc., are arrived at in the same way; in
short, people acquire the vocabulary Aristotle treats of in his Categories.
The words of this vocabulary are called 'names of first imposition'.
Some concepts overlap, and some are temporarily combined. T o express
such facts people create a new kind of words which are not names and do
not externalise concepts but convey information about their 'syntax'.
Armed with these 'consignificative' words (copula, prepositions, etc.)
people can frame such sentences as 'The dog is an animal' and 'This dog is
barking down a hole.' But men also reflect on and want to be able to talk
about their own words and concepts and their 'syntactical' properties. For
this purpose they create a new series of names by a second 'imposition' —
names such as 'noun', 'disyllabic', 'species'.
The fact that language arises from an occupation with particulars is
responsible for an inherent ambiguity: names of individuals and universals
42. Full documentation of the following would require extensive discussions of philological details. I
list here some important passages: Imposition: Porphyry 1887, pp. 57-8; Dexippus 1888, pp. 11-
16; Ammonius 1895, p. II; Simplicius 1907, p. 15; Boethius i86og, col. 159. Concepts, internal and
external discourse, truth,falsity, andrtmiigiiil)/:Porphyry 1887,64.29,101.24ff.-106.] iff; Porphyry,
In De int. in Boethius 1880, I, c. 1, pp. 26-40, and II, c. 5, pp. io6ff.; Porphyry 1886, III.3;
Dexippus 1888, pp. 7-10; Ammonius 1897, pp. 18, 21, 72-5; Boethius 1880, I, c. 1, pp. 20-1
(notice how close 21.1 is to Porphyry 1887,57.20) and p. 49. /I&srracrion and predication: Porphyry,
Isagoge passim; Porphyry 1887, pp. 00-1; Porphyry, Cat. ad Gedalium in Simplicius 1907, pp. 53
and 79. (Concepts of) individuals, proper names: Porphyry 1887. p. 7; Dexippus 1888, 15.27;
Boethius 1880,1,c. 1, 28.i8ff. Consignificative words: Porphyry, In De int. in Boethius 1880,1, c. 3,
p. 77; Dexippus 1888, pp. 32-3; Pseudo-Augustinus 1961, §2; Ammonius 1897, pp. 12-13;
Boethius 1880, I, c. 1, p. 6; Simplicius 1907, pp. 64-5. For a more thorough discussion of
Porphyrian semantics see Ebbesen 1981b, vol. I, ch. IV. 4.
on various levels are morphologically indiscernible, and no special kind of

names exists to signify the concept of a particular name or of a particular
concept. It is this fact that makes sentences like 'Socrates is (a) man', 'man is
(an) animal', 'man is (a) species' and 'man is monosyllabic' potentially
deceptive. The Stoics had dealt profusely with such problems, often as a
means of combatting Platonist and Aristotelian ontology, and it is charac-
teristic of Porphyry's writings that he incorporated many of the Stoic
examples and some of their discussion of those examples.
Rational 'activity' takes place in that 'part' of the mind that is called
dianoia and is reflected primarily in speech, secondarily in writing.
Dianoetic 'activities' and their externalised reflections in speech and writ-
ing are referred to by the same names, viz. 'proposition,' 'argument',
'syllogism', etc. W h e n the need arises, one can distinguish by saying
'vocalised discourse signifies (carries messages about) dianoetic discourse,
which in turn signifies (carries messages about) things'. In normal circum-
stances this may be shortened to 'expresssions (vocalised discourse) signify
(carry messages about) things'. Truth and falsity are primarily properties of
dianoetic propositions, ambiguity is exclusively a property of external
propositions.
Porphyry's account of the genesis of concepts and language can be read
as an historical account implying that at some definite time men intro-
duced all names of the first imposition and at some later time they decided
to add the names of the second imposition. Post-Porphyrian commen-
tators (Greeks and Westerners alike) sometimes took this to be the idea, and
some of them would even have people believe in two different congresses
at which the convened wise men created the respective vocabularies. But
this was hardly what Porphyry had intended. What he meant was rather
that there are certain natural steps in the development of our stock of
concepts and words. Roughly speaking, it may be true that the first-
imposition vocabulary of Greek is older than the second-imposition
vocabulary, but what is important is that anybody must develop his stock
of concepts and words in the order indicated. You simply cannot use a
meta-language if you have no object-language, and you cannot master an
object-language unless you have sensed and conceptualised objects.
Stages in the reception of ancient thought

The ancient materials were of great importance for the Middle Ages, but
they were not received and assimilated all at once. Boethius had presented
the West with an almost complete series of the A- and B-books (see above),
122 The old logic
but the fruits of his labour were not reaped till centuries after his death. In
the early Middle Ages when higher education was virtually extinct most
people read the C-books only. Until the twelfth century the Analytics, the
Topi«, and the Elenchi (4-4M.A and 5.A) were neglected. This neglect may
have caused the irrevocable loss of Boethius' commentary on the Topics;
references to it in medieval authors are so rare that it is very doubtful if they
can be trusted; 4 3 to all appearances it disappeared before the twelfth
century. In the course of the ninth to eleventh centuries 1-3. A +Β (Isagoge,
Categories, and De interpretatione with Boethius' commentaries) took prece
dence over 1-3.C, only De divisione retaining some importance. In the
twelfth century 4.A and 4M.A (Prior Analytics and Topics) began to be
studied again, and they gradually replaced all other books in group 4
except De topicis differentiis, but the process was a slow one because of the
lack of commentaries. Before 1150 James of Venice supplied the missing
4.a.A+B - viz. the Posterior Analytics with 'Alexander's' commentary -
and the forgotten 5.A (Elenchi) became at once a central book when James
provided 5.B by translating 'Alexander's' commentary. The immediate
effects of late arrivals such as Themistius' paraphrase of the Posterior
Analytics (1180s), Simplicius on the Categories (1266), and Ammonius on
De interpretatione (1268) were less important because they arrived at a time
when Latin scholasticism was finding or had already found its own way of
doing things.
In the early twelfth century, then, half of the C-books had ceased to be
important. A century later all of them except Boethius' De divisione and De
topicis differentiis had been superseded. The most surprising feature of this
development is the completeness of the oblivion into which the C-books
fell. It is doubtful if any important feature of later medieval logic can be
traced back to the use of them during the early phases, even if the
idiosyncratic Roger Bacon did dig up Augustine's Dialectica and find
some inspiration in it. The role of these books had been to keep the ancient
tradition alive and to prepare the way for a teaching based on A + B.
Medieval theorising on ancient foundations

The medieval schoolmen did not only read about the ancient theories, they
also used them and developed them. They read about proof-procedure in
Boethius' De topicis differentiis and were led to inquire into the foundations
43· Cf. Green-Pedersen 1973, p. 28.

of syllogistic and other proof. They read fragmentary accounts of

Porphyrian semantics in Porphyry's own Isagoge, in the commentaries of
Boethius and 'Alexander', and from the second half of the thirteenth
century also in Ammonius and Simplicius. They did not always know how
much of the material in the secondary sources had a Porphyrian origin, but
they were impressed by what they read. The question of universals became
the subject of a lively debate; the theory of the imposition of names came to
hold a central position in thirteenth-century modistic semantics; considera-
tion of passages about metalinguistic statements and the like contributed
decisively to the formulation of the theory of supposition; and Porphyrian
passages in Ammonius encouraged the fourteenth-century creators of
'nominalist' semantics.
The medievals read Greek grammatical theory in Priscian. It was tempt-
ing to combine grammar and logic, and, in fact, the two had a common
origin. But — partly because logic was Aristotelian and grammar owed
more to the Stoa - the same terms were not always used in the same way in
the two disciplines. Thus both logic and grammar deal with 'parts of
speech', but the logician operates with two parts only, viz. the name and
the verb as the signifiers of the subject and the predicate concepts
('noëmata') of his dianoetic propositions. The grammarian has eight parts of
speech and must posit some intelligible content ('noêton') even of con-
significative words such as prepositions because otherwise he cannot for-
mulate rules of syntax. Ancient writers were well aware of the differences
between logical and grammatical analysis, so grammatical theory was only
rarely imported into works on logic. In the Middle Ages much was done to
bridge the gap between the two disciplines. The Apollonian grammar
inherited via Priscian was developed into the theory of modi significandi; it
was joined to a developed version of Porphyry's theory of imposition, and
it was thought applicable to logical questions, particularly in the thirteenth
century. At first blush it seemed an immense advance, for now it became
possible to discuss many problems, such as syntactical ambiguity of sen-
tences (amphiboly), which had been very difficult to formulate precisely in
ancient logical terminology. But when the weaknesses of this approach
became apparent and a more truly Porphyrian logic came to the fore, the
gap between logic and grammar widened again, as we see in the writings of
John Buridan.44
44. See, e.g. Buridan, Summulae 7.3.10, in Ebbesen 1976b, pp. 157-8.
124 The old logic
An example of the ubiquity of ancient doctrine in medieval books

The penetrating influence of the Greeks and Boethius on the Latin Middle
Ages may perhaps best be illustrated by examining some quite trivial books
that a late thirteenth-century student at Paris might read to learn about
fallacies. Perhaps the first thing the student would read would be a com-
pendium such as (?Pseudo-) Aquinas' Defallaciis, Chapters 3-4 of which
would teach him that (1) just as a dialectical argument derives its strength
from a true locus, so a sophistical argument derives its semblance of strength
from an apparent locus and is based on a false maximal proposition such as
'whatever is true of the accident is true of the subject'; (2) a sophistical locus
requires a 'reason for the semblance' (causa apparentiae), the plausible
appearance that makes people assent to the argument, and a 'reason for the
failure' (causa defectus); (3) there are six sophistical loci or fallacies 'in speech',
and seven Outside speech'; (4) 'in speech' means 'having a linguistic reason
for the semblance', Outside speech' means 'having a real [extra-linguistic]
reason for the semblance'; (5) fallacies in speech depend on actual, poten-
tial, or fantastical ambiguity (multiplicitas).
We may stop with these. Only (3) derives from Aristotle's Sophistici
elenchi. (1) stems from Boethius' doctrine of loci, applied to fallacies.
Boethius had indicated that such an application is possible (De top. diff. I, PL
64:1182c) without elaborating the point. (2), the theory of causae ap-
parentiae et defectus, is a medieval invention, though with some foundation
in Aristotle's text. (4) is an adaptation of 'Alexander's' adaptation of an
ancient, and probably Stoic, classification of fallacies.45 (5) stems from
'Alexander's' report of Galen's proof (in On the Sophisms in Speech) that
Aristotle's list of fallacies 'in speech' in exhaustive. 46
After De fallaciis the student might try to read Aristotle's Sophistici
elenchi. He would use the standard translation, by Boethius, and his copy
would contain interlinear glosses and marginal scholia. The book begins:
'De sophisticis autem elenchis et de his qui videntur quidem elenchi, sunt autem
paralogismi sed non elenchi, dicemus.' O n e of the first glosses he might find
would be 'id est' or 'pro id est' written above 'et' to indicate that the sense is:
'We shall speak about sophistical refutations, though in reality they are
paralogisms and not refutations.' 4? He might also find a marginal scholium
45. Cf. Ebbesen 1976a, p. 108; Anon. Summa SE in De Rijk 1962-7, I, p. 353; Michael of Ephesus
1898, pp. 20-1 (shows awareness of Stoic origin; cf. Philoponus 1905,241.1 - 4 ; Diogenes Laertius
1925, 7.43; Albinus 1858, Ch. 6, p. 159).
46. Cf., e.g., Peter of Spain, 1972, VII.24-5; Michael of Ephesus 1898, pp. 22-3.
47. E.g., in these MSS of various origins and dates: Assisi B. Com. 286: 98'; Clermont-Ferrand B. M.
explaining that some hold the view that this is the sense of the passage,
while others think it means 'We shall speak of sophistical refutations, that
is, syllogisms that are materially defective (having a false premiss), and of
apparent refutations, that is, formally defective syllogisms.' 4 8 Both inter
9
pretations can be traced back to 'Alexander's' commentary,* and the
distinction between materially and formally defective syllogisms goes back
further, at least to the second century A.D., since it appears in Alexander of
Aphrodisias' commentary on the Topics (ed. Wallies, pp. 20-1). It is almost
certainly an 'Aristotelianising' of a Stoic distinction between false argu
ments (pseudeis logoi) that are valid but contain a false premiss, and false
arguments that are so because they are invalid, not being reducible to any
of the Stoic syllogistic figures.
A few lines later, the student might find an explanation of the strange
phrase 'tribualiter infantes' (164*27, literally 'tribally inflating') to the effect
that in days of old the Athenian tribes competed to see which could supply
the fattest animal for sacrifice at the great festivals, and some would
fraudulently try to make a meagre victim appear fat by blowing air into
it. 5 0 The scholium might also contain the information that the source of
this explanation was 'Alexander', as indeed it was. 5 1
Thus, in small matters and in great ones, the student would be offered an
interpretation that reflected 'Alexander's' and through him the work of
earlier Greek logicians and scholars. He might also read a literal com
mentary, which would contain much the same material plus some dubia,
169:3'; Erlangen U. Β. 194:1'; Escorial g. III.28 ; 156", Napoli Β. Ν. C. VIII.E.I4:77'i Oxford

Balliol 253:24s': Oxford Bodl. Canon. Class, lat. 188:87'; Paris Maz. 3470:66'; Paris Β. N . lat.
7766:63'; Vat. Pal.lat. 092 ;4'; Vendôme Β. M. 171 :</ι'; Wien VPL 2377:49'.
48. E.g.. in MS London B.L. Harlcy 3272:12i'(cf Fircnze Laur. St. Croce 2 sin., 9:86·): 'Nota quod
ista littera secundum quosdam divcrsas habet expositioncs. Quidam enim legunt li "et" exposi
tive, sic: "De sophisticis elenchis ct — pro 'id est' - ". Alii sic: "De sopbisticis etc., id est de elenchis
peccantibus in materia, el de his etc., id est dc peccantibus in forma".' Similar scholia occur in
numerous other MSS of various dates and origins. In MS Vendôme 171:9i' an early fourteenth-
century hand writes: 'De sophisticis autem etc. Ista autem pars posset sic exponi: U n o modo: "De
sophisticis etc., i.e. de locis peccantibus in materia, et de his etc., i.e. de peccantibus in forma". Vel
aliter: "De sophisticis etc., i.e. dc fallacies in dictionc, ff de his etc., i.e. de fallaciis extra dictionem".
Vel aliter, secundum Alexandrum, quod li et tencatur expositive.'
49. Cf. Michael of Ephesus 1898, 5.4-7; 5.24-7; 7.4-11; Ebbesen 1972, p. 25.
50. E.g., MS Paris, B. N. lat. 6289:84': 'Dicit tamen Alexander quod hoc dictum tribualiter inflames
tract[at]um est a consuctudine Athcnicnsium, qui solebam sacrificare diis secundum differentiam
tribuum. Uli enim qui divitcs cram opimas adducebant victimas et pingtiissimas, ill· vero qui
paupercs erant, quod divitiis suis non potuerunt faccre arte volucrunt, scilicet macras victimas
inflames calamis, ct apparebant pinguissimac.' The same scholium with minor variations appears
in MSS Avignon 101» :.i'; Cambridge Gonv. & Caius 468/575:113; Cambridge Pembroke
193 :7s'; Paris B. N. lat. 17806:69'. Similar scholia may be found in many other MSS.
51. Cf. Michael of Ephesus 1898, 8.28ff.; Ebbesen 1972, p. 28.
126 The old logic
i.e. discussions of problems to which the text gave rise. For a fuller
elucidation of such problems he would read or take part in class-room
disputations of Quaestiones super librum Elenchorum ('Questions on the
Elenchi'). The questions would be roughly the same, whoever directed the
course. The following all stem from an anonymous work from the 1270s
(edited in Ebbesen 1977):
(1) Is a materially defective syllogism a dialectical syllogism? (op. cit.
qu. 14)
(2) Does a sophistical argument derive strength from a locus? (op. cit.
qu· 33)
(3) Is the essential form of any dictio its capacity to signify (its ratio
significandi)? (op. cit. qu. 40 ( + 41-42))
(4) D o different conditions of verification (causae veritatis) cause a term to
be equivocal? (op. cit. qu. 48)
(5) Which sort of ambiguity (multiplicitas) is most genuinely so, the
actual, the potential, or the fantastical? (op. cit. qu. 50)
(6) Is an equivocal term such as 'dog' one or more than one constituent of
a sentence (partes orationis), and which construction is grammatical: 'The
dog runs' or 'The dog run'? (op. cit. qu. 54, cf. qu. 817)
The Greek and Boethian background for asking (1), (2) and (5) should
be clear from what has already been said. The discussion of (3) may have
originated in an analysis of the Galenic proof of exhaustiveness mentioned
above, because the proof proceeds from a search for the essential virtue or
perfection of a dictio ('lexis'). Galen considers the possibility that it is to
signify, but decides that it is to signify well, i.e., precisely, wherefore the
gravest vice in a dictio is for it to signify badly, i.e. ambiguously. However
the quaestio first arose, in the 1270s it had grown into an inquiry into the
foundations of semantics, involving the whole machinery of imposition,
rationes significandi, etc., that had developed from Priscian's version of
Apollonius' semantics combined with Boethius' version of Porphyry's
version of what seems to have been originally a Stoic model of the rise and
acquisition of language. 52 (4) deals with the difference between 'man' in
'(A) man is running' and 'man' in 'Man is a species'. The medieval
discussion of such problems, which the theory of supposition was used to
solve, was sparked off by passages in Boethius' and 'Alexander's' com-
mentaries. They in turn depended on Porphyry, whose ultimate sources
were Stoic. The reason why the question is raised in this particular place, in
52. Cf. Frede 1978, p. 69.

a commentary on the Elenchi, appears to be that 'Alexander' used

'Disyllabic is tetrasyllable, Dio is disyllabic, therefore Dio is tetrasyllable' to
exemplify a sophism resulting from equivocation.53 The example can be
traced back to an ancient scholium (probably by Porphyry) on Categories c.
5
ι, * and it may be of ultimately Stoic origin. (6) became a problem because
'Alexander', quoting Galen's On the Sophisms in Speech, had taught the
West that an equivocal term 'actually' signifies two things, and in the
framework of Priscian's Apollonian grammar this might seem to imply
that construing such a term with a verb in the singular would produce an
ungrammatical result.
Many more examples could be added, but these should suffice to show
that ancient scholasticism meant much to its medieval heir. One should
not, however, infer that the medievals just repeated what their predecessors
had said and thought. The ancients were a source of inspiration. Often a
casual remark or an example used by an ancient author would start a long
debate that could lead to rediscovery of forgotten insights, but also to new
discoveries and new theories. As often as not the result would be that the
medieval pupils surpassed their ancient teachers.
53. Cf. Anonymus Aureliancnsis I (twelfth century), Comment, in SE, MS Orléans B. M. 283 :114A (ad
i65"4-6j: 'Referendum est ergo quod dicit unus locus ad aequivocationem. Alexander tamen hie
exemplum ponit de illo vilissimo univocationis genere, quando scilicet modo agitur de nomine
per nomen ipsum, modo de re, ut "homo est nomen, omne risibile est homo, ergo quoddam
risibile est nomen". [- · ·] Vet arbitramut accidere in rebus quod accidit in nominibus, nam transferimus a
nominibus ad res, ut hie: "disyllabum est tetrasyllabum, sed musca (?) est disyllabum, ergo est
tetrasyllabum".' Cf. also Anonymus Pragensis, Quaestiones super SE, qu. 9 (late thirteenth
century), MS Praha M. K. L. 66:82"A: 'Consequenter quaeritur utrum terminus significans se et
suum significatum sit aequivocus ad ilia quae significat, verbi gratia "homo" significat sc cum
dicimus "homo est disyllabum", sed significat suum significatum cum dicimus "homo est
substantia".' - For the Greek counterpart see, e.g., Michael of Ephesus 1898, 12.32-3.
54. Cf. Simplicius 1907,31.3-4.
5
PREDICABLES AND CATEGORIES
Aristotle, Porphyry, and the old logic

Prominent among the antecedents of medieval philosophy, particularly of
logic and philosophy of language, are two logical works of Aristotle's. His
De interpretatione and Categories, as transmitted by Boethius 1 and coloured
by Porphyry's 2 introduction (Isagoge) to the Categories,3 endowed the
logica vetus with a substantial inheritance of technical terms and the meta-
physical puzzles that go with them. An explanation of this inheritance
requires a brief excursion into the history of the terminology central to the
Isagoge and the Categories.
The terminology of the Isagoge and the Categories

Aristotle developed two loosely linked doctrines regarding the structure of
propositions; 4 one is that of the predicables, the other that of the categories
(or predicaments). Chapters 4, j , and 8 of Book I of Aristotle's Topics con-
tain an account of the predicables definition, property, genus, differentia, and
accident. A definition is a phrase signifying a thing's essence. A property is a
predicate not indicating a thing's essence but predicable convertibly of it
since it belongs to that thing alone; e.g., the capacity for literacy belongs
exclusively to man, so that anything that has that capacity is a man, and
only a man has that capacity. A genus is predicated essentially of diverse
sorts of things; e.g., animal of man, ox, horse, and so forth. A differentia in
combination with the genus produces the definition, as when mortal rational is
added to animal to produce the definition of man. Finally, an accident may
be specified either as any predicate other than any of the predicates
mentioned above or as that which may or may not belong to a given
individual; e.g., the predicate being seated with respect to the individual
1. Liebschütz 1967, pp. 538-43. The originals, in translation, and with commentary, are available in
Ackrill 1963.
2. Pinborg 1972, pp. 32-42.
3. Lloyd 1967, pp. 283-93, cf. pp. 319-22.
4. Bochenski 1961, p. 51; cf. Kneale 1962, pp. 25-32.
Predicables and categories 129
Socrates. By the time of the Isagoge, definition had been replaced by species,
so that the Porphyrian list of predicables runs: species, property, genus,
differentia, and accident.5
The Categories, to which the Isagoge purports to be an introduction,
reveals certain connections with the scheme described above. Apart from
some allusions to genus and differentia in Chapter 3, the significant links
occur after Aristotle has listed and exemplified, in Chapter 4, his ten
categories: substance (e.g., man), quantity (e.g., two cubits long), quality
(e.g., literate), relation (e.g., bigger), place (e.g., in the Lyceum), time (e.g.,
yesterday), position (e.g., sitting), state (e.g., armed), action (e.g., cutting),
and undergoing (e.g., being cut). In Chapter 5, after exemplifying primary
substances as the individual man, the individual horse, and so on, he
identifies species and genera, within which such primary substances are
contained, as secondary substances. This link is one of several seen as
significant by Boethius. 6
The Porphyrian tree
In his commentary on Porphyry, Boethius discusses the Porphyrian 'tree' 7

which takes the category substance as genus generalissimum (most general
sort) and uses the predicables to divide substance into a hierarchically
ordered series of genera and species. These genera and species, then, may be
viewed as secondary substances. For example, when one obtains the species
man in the development of the tree by the addition of the differentiae
rational and mortal to the genus animal, the terms 'animal' and 'man' may be
seen as lying at that secondary substance level, the precise semantic status of
which is problematical. Categories other than substance constitute further
genera generalissima beneath which schemes of classification may also be
constructed in accordance with the predicables. For instance, the prop-
osition ' White is a colour' is an example drawn from the classifications under
the category of quality used as a genus generalissimum, and it displays a
semantic complexity analogous to that of'Man is animal'8 in which the
genus (i.e., animal) is predicated of the species (i.e., man).9
5. On the consequent medieval distinction between the Aristotelian and Porphyrian treatments, see
Green-Pedersen 1973, pp. 9-11. Kluge 1973-4 gives Ockham's interpretation of the predicables
in translation.
6. PL 64, col. 16B,
7. PL 64, cols. 42, 103.
8. PL 64, col. 28C.
9. Henry 1974, pp. 30-1; p. 241.
130 The old logic
The semantic source of the problem of universals

In the absence of any competent logical analysis of propositions such as
those just mentioned, pre-medieval authors were still at the comparatively
naive stage of searching for referents of the names occurring in such
propositions. This search, and the assumption that words, things, or quasi-
things must be the referents to be found, led to the formulation of the
problem of universals. For instance, Boethius opts for words as the
referents when he claims that 'species' and 'genus' are names of names
(nominum nomina):10 they name the subject terms of the sentences in
question and indicate something about those terms. Thus in 'Man is a
species', the word 'species' names the word 'man' and indicates that that
word is predicated only of individuals belonging to that species. Likewise
'genus' in 'Animal is zgenus' names the word 'animal' and indicates that it is
predicated of its various species.1 ' The full force of such semantic naivete is
plainly visible in the often-quoted Porphyrian passage which sets out a
purportedly exhaustive list of possible referents of terms for species and
genera: do they subsist outside the mind or are they merely mental? If
outside the mind, are they corporeal or incorporeal; and are they joined to
things perceptible by the senses or not thus joined? 1 2 Answers to these
questions were taken to settle the question of the status of the universal -
that which, in Aristotle's terms, is naturally fitted to be predicated of many
things. 13 Since the five predicables fall under this Aristotelian description,
they were sometimes known as 'the five universals'. Furthermore, since
some of the alternatives in the Porphyrian passage carry implications
regarding what things there are, we are here thrown into the area of
metaphysics, the science of being.
Important ambiguities in the doctrine of the predicables and categories

The ambiguities and obscurities 14 permeating the predicables and cate-
gories have been deliberately maintained in the account given above; they
may be heightened by considering at least two further factors. First, a term
such as 'rational' is classified as a differentia and hence excluded from being
an accident according to the five-fold Porphyrian classification of pre-
dicables; nevertheless, it is predicated as a quality (in eo quod quale).15 But
io. PL 64, col. 176D; cf. also cols. 159—60.

11. Cf. Green-Pedersen 1977a, pp. 51-3 and 71—2 on some later medieval developments.
12. PL 64, col. 82; cf. Tweedale 1976, pp. 63-88 for detailed discussion.
13. PL 64, col. 170B.
14. Cf. Tweedale 1976, p. 59.
15. PL 64, col. 52.
the categories other than that of substance, including the category of

quality, are collectively classified as 'accidents'. So it seems that the dif-
ferentia rational should after all be categorised as an accident. Boethius'
answer to this puzzle involves separating the qualities that are differentiae
for the category substance from the qualities that are accidents in one of the
other categories. To accomplish this he uses various criteria that push the
former sort of qualities in the direction of secondary substances rather than
accidents. Differentiae are not merely otherwise-making (alteratum facien-
tes) but also other-making (alterumfacientes); i.e., they affect the species and
not merely the accidents of their subjects.16 They are not susceptible of
degree, a feature which they have in common with (secondary) sub-
stances,17 and which distinguishes them from accidents that do admit of
degree.
It is here that the second of the two complicating factors intervenes. In
his commentary on the Categories Boethius associates the problem of
differentiae with Aristotle's contrast between being predicated of a subject and
being in a subject. Since the contrast between the linguistic (being predicated
o f . . . ) and the ontological (being in . . . ) which is apparently being de-
veloped here is in fact misleading, it will be convenient to discuss this
contrast in terms of two sorts of predication, i.e., de subiecto (of a subject)
and in subiecto (in a subject), respectively.18 Roughly speaking, de subiecto
predications are those that would be embraced by a theory or science of the
subject of the predication: the theory of man, for example, would contain
'Man is animal' as a thesis. It is in this sense that de subiecto predications may
be termed substantial or essential relative to their subjects. The connection
of de subiecto predication with definition19 emphasises this sort of align-
ment.20 In subiecto predications, on the other hand, do not constitute theses
belonging to the theory of their subjects. As far as the problem of dis-
tinguishing between a quality that is a differentia and a quality that is an
accident is concerned, it turns out that a differentia is predicated de subiecto
and is hence akin to (secondary) substance; predications of qualities that are
accidents, on the other hand, are in subiecto predications. So a differentia is a
kind of half-way (quoddam medium) between substance and quality, having
some of the nature of each.2 '
16. Cf. Tweedale 1976, p. 61; PL 64, col. 49A-B; Henry 1974, pp. 190-1.
17. PL 64, col. 50.
18. Ackrill [963, pp. 75-6; Henry 1974, n. 4.101a; Henry 1967, §3.122.
19. PL 64, cols. 176A, 185B.
20. Henry 1967, pp. 43-50.
21. PL 64, col. 191-2; Ackrill 1963, pp. 85-7; De Rjjk 1956, p. 51.
132 The old logic
A final distinction which must be added in order to understand

Boethius' own summary of the predicables is that between predications
that are in eo quod quid (in respect of what-ness), which answer the question
'What is it?' ('Quid sit? '), and those that are in eo quod quale, or qualifying,
predications:
h genus is that which is predicated in respect of what-ness of many things different
in species. A species is that which is predicated in respect of what-ness of many
things not different in species. A differentia is that which is predicated substantially
and in respect of quality of many things different in species. A property is that which
is predicated of one species only in respect of quality and not substantially. An
accident is that which is predicated in respect of quality and not substantially of
many things different in species.22
Both genus and species are predicated in eo quod quid, and hence both those
predications are called quidditative. In that respect they differ most from
accident, the common definition of which (drawn from Porphyry) is 'that
which comes and goes without the subject's perishing' (quod adest et abest
praeter subiecti corruptionem).23
It is difficult to distinguish the various topical threads running through
the predicables and the categories. One way to trace those threads lies in
seeing as central the distinction between theory-relevant ('substantial') and
theory-irrelevant ('accidental') predicates which Boethius stresses both in
his Dialogue on Porphyry2* and when he explains the distinction between de
subiecto and in subiecto predications in his commentary on the Categories.25
Hence the predicables may be construed as providing a general rationale
for the construction of theoretical descriptions, which naturally calls forth
questions of classification, class-inclusion (genus, species), and class-
exclusion (differentia); these are sometimes expressed in terms of class-
extensions in Boethius' commentaries both on the Isagoge26 and on the
Categories.21 A full and satisfactory expression of these class-relations and
22. 'Genus est quod de pluribus specie difFerentibus in eo quod quid sit praedicatur. Species est quod
de pluribus minime specie difiercntibus in eo quod quid sit praedicatur. Differentia est quod de
pluribus specie differentibus in eo quod quale sit in substantia praedicatur. Proprium est quod de
una tantum specie in eo quod quale sit non in substantia praedicatur. Accidens est quod de
pluribus specie differentibus in eo quod quale sit non in substantia praedicatur.' PL 64, col. 94C;
Henry 1974, pp. 231 and 236.
23. PL 64, col. 132C; cf. Henry 1974, p. 231, Pinborg 1973a, pp. 59—60, 63^4 for examples and
discussions.
24. PL 64, 1 2 B - C .
25. PL 64, col. 176A.
26. PL 64, col. 29B.
27. PL 64, cols. 177-8.
their counterparts in the theory of collective classes sometimes adumbrated

here 2 8 requires an exact modern artificial language, incorporating all the
apparatus of quantifiers, variables, and many-link functors. 2 9 The same
applies to some of the examples, with their recondite sentential structures,
used to illustrate the various categories: 'Man is a substance', 'Literate is a
quality', and so on. The concentration of discourse at the level of'second
ary substances' (i.e., of genera and species), is a device enabling semanti-
cally complex material to be expressed in comparatively simple forms of
speech such as 'Man is a species', 'Animal is zgenus', 'Man is animal'.30 It is
from lack of appreciation of the complex deep structures underlying these
forms that the metaphysical disputes concerning universals can all too
naturally arise. Both the problems and the disputes concerning universals
are symptomatic of struggles to unravel that complexity, which was to
have a part in inspiring the development of the medieval linguistic doctrine
of suppositio.31
Meaning, logic, and grammar

It has long been customary 3 2 to allocate words to diverse 'parts of speech'
or 'semantic categories', e.g., noun, verb, preposition, participle, and so on.
By the eleventh and twelfth centuries the divergence between the logical
and the grammatical traditions regarding the nature and number of parts
of speech was already of long standing. Boethius was of the party of the
logicians who, following Aristotle, held that the parts of speech important
to the logician were merely two in number, these being the name (com
prising nouns and adjectives) and the verb; the alleged other parts were said
to be completions of, or connections between, these two, much in the same
33
way as the brakes of a chariot supplement its main parts. In contrast, the
grammarian Priscian admits as essential for his purposes parts other than
the name and verb, i.e., participle, pronoun, preposition, adverb, interjec
tion, and conjunction. 3 4 The differences between these two opposing
views were prolonged into points of detail. The most fruitful difference in
the period before Abelard arises from the fact that Priscian assigned to
28. PL 64, cols. 44-5, 112 elaborated in the De divisione of PL 64; cf. Henry 1972, pp. 121-2.
29. Cf. Henry 1972, pp. 42, S3.
30. Henry 1974, pp. 30-1; Henry 1972, p. 43.
31. Detailed in De Rijk 1962-7; sec Dc Rijk 1062-7, 11(1), ΡΡ- 27, 51, 139, 357; Fredborg 1973,
p. 20; and Nielsen 1976, pp. 51, 66 for recognition of the equivocation of'homo' ('man') when
'Man is a species' is introduced.
32. Robins 1951.
33. PL 64, cols. 79SC-797A: cf. 766A-B.
34. Keil 1855, 11, pp. 551.18-552.14; cf. p. 54.5-26; and Henry 1974, pp. 262-3.
134 The old logic
names the specific function of signifying substance and quality, 3 5 whereas

Aristotle had claimed that paronyms (a sub-class of names) signify a quality
and only a quality. 3 6 Superficially, of course, there need be no incompati
bility here if Priscian is understood to be saying that some names signify
substances and others signify qualities, a natural interpretation of his
doctrine. But twelfth- and thirteenth-century versions of Priscian's doc
trine prefer the use of 'cum' (substance along with quality) instead of the
mere 'et' ('and') that Priscian himself used, thereby allocating to names the
signification of substance along with quality. 3 7 Examples on both sides of
this dispute will serve to fill out this rather abstract description.
Paronyms or denominative names

O n the logicians' side there are the well-known illustrations of the various
categories provided by Aristotle in Categories 4. There literate is given as an
example of a quality, and that example was to be used as a ground not only
for saying that 'literate' signifies quality (rather than substance) but also
that literateis a quality, i.e., literacy. Again, the name 'white', categorised as
a quality-word, signifies a quality and nothing but a quality, according to
Aristotle. 3 8 Both 'literate' and 'white' were classified as paronyms, or
denominative names - i.e., alleged derivatives from the corresponding
abstract forms: 'white' from 'whiteness', 'literate' from 'literacy', 'just'
from 'justice', and so o n . 3 9 Anselm's practice in his Degrammatico indicates
that he used the term 'denominative (name)' to cover only accidental
qualitative derivatives, hence preserving not only the contrast (mentioned
in the foregoing section) between the qualities that are accidents and the
qualities that are differentiae but also that between accident and substance.
The fact of being thus 'derived' can distinguish quality-names from
substance-names only if the language in which the terms in question are
embedded lacks abstract nouns corresponding to substance-names such as
the 'man' and 'horse' used as examples in Aristotle's list of categories.
Although that prerequisite was already not fulfilled in Boethius' logical
language, 4 0 Anselm still presupposes it; 4 1 and the problem of explaining
35. Keil 1855, 11, p. 55.6.

36. PL 64, col. 194C.
37. E.g. De Rijk 1956, p. 113. 15-24; Henry 1974. pp. 146-7, 213; Henry 1958, p. 176; Fredborg
1973, Ρ- 47; Nielsen 1976, p. 53; Fredborg et al. 1975, p. 111. See also note 45 below.
38. PL 64, col. 194C.
39. Categories 1; cf. PL 64, cols. 167-8.
40. Henry 1964, p. 87; Henry 1958, p. 174.
41. Henry 1967, §3.131.
the meaning of such paronyms or denominative names is a persistent

theme of medieval42 and modern 43 logic. Their separation as names
signifying only a quality remains a standing challenge to the grammarians'
generalisation that all names signify both substance and quality.
Adjectives and other non-proper names

On the grammarians' side a concrete example which served to exacerbate
the rift between them and the logicians may be seen in Priscian's attempt to
sort out adjectives from other non-proper names. For him name is a genus
having the proper and the common names as its species. The common
name (nomen appellativum) in its turn has the nomen adiectivum or adjectival
name as a further sub-species. Such adjectives, Priscian explains, are cus-
tomarily joined to proper or common names, both of which signify
substance, in order to make clear the quality or quantity of their referents.
Among the examples he gives of such substance-signifying names is that
very 'literate' ('grammaticus') that had been enshrined in the master-
logician's Categories as a quality-word.44 Hence to the general counter-
logical doctrine that all names signify both substance and quality, Priscian
adds what could be seen as a provocatively counter-logical example. And
that example could (but of course need not) have inspired Anselm's choice
of 'grammaticus' as the central example in his dialogue on paronymous
meaning (De grammatico).
Grammarians' and logicians' approaches to meaning

The resultant clash is well in evidence in the early medieval period.4* And
in spite of the ancient terminology used for its presentation the problem
here at issue still has implications that reflect two possible approaches to
studies of meaning. On the one hand, one may base such studies on an
inspection of the way in which an approved group of users of a given
language actually use that language. Priscian's refusal to deviate from the
usage of those whom he regards as the best authors is an index of this
adherence to a descriptive basis for his account of language.46 On the other
hand, a way of accounting for the meaning of words which need not wait
upon usage came from logic, in connection with the discussion of par-
42. Jolivet 1975.
43. Henry 1967, pp. 68-9; Henry 1958, pp. 169, 173; Henry 1964, pp. 83, 86,
44. Keil 1855, it, p. 58.20-4; Henry 1967, pp. 66-71; Pinborg 1974, pp. 43-4.
45. De Rijk 1962-7, n(i), pp. 98, 183-6, 222-3, 226, 230-4, 241-3, 259-60, 521-2; Colish 1968,
pp. 93-108 and note 37 above.
46. Keil 1855,11, p. 371.18-22; cf. Henry 1967, pp. 16-17; Robins 1951, p. 71.
136 The old logic
onymy. Thus, to take an example from Chapter 14 of Anselm's De

grammatico, suppose some white horses are concealed within a building, and
suppose a speaker truly asserts that some whites are within that building.
The hearer, who knows nothing of the horses' presence in the building,
would still be at a loss to know the denotation of the paronymous subject
term of the assertion. The openness of the range of possible denotata arises
simply because, other things being equal - e.g., there are no race-riots in
the vicinity — any of Aristotle's physical substances might be assumed to be
the referents. (This range of possible referents would not, according to
Anselm, be strictly comprised within the meaning of'white', but would be
contingently connected with 'white' because of an association grounded in
experience.) Both in their use as subjects of declarative sentences and in the
role of denoters of the objects of commands, such paronyms must be seen as
theoretically available for novel applications. Anselm goes on to exemplify
this availability by turning to a command situation in which a hearer is
faced with a white horse and a black bull and receives the order to strike; in
reply to his question 'Which one?' he is told '(The) white'. In the context he
can see that the horse rather than the bull (both of them substances,
however) has been indeterminately referred to by means of the paronym
'white'. By claiming that the strict meaning of the paronym 'white' is
merely ' . . . h a v i n g whiteness' (and similarly with other paronyms), he
shows, in opposition to Priscian's general claim that a name signifies both
substance and quality, that some names at least (i.e., paronymous adjectival
names) signify 'a quality and only a quality', exactly as Aristotle had
claimed. 47
Anselm recognises that pressure from the side of common usage (usus
loquendi), on which grammar bases its descriptive account, can push one in
the direction of Priscian's claim. Inspection of usage shows that at least
certain paronyms, such as the already-noted 'literate', have always been
used to refer to the same sort of object (i.e., human beings), a fact that seems
to confirm the grammarians' claim that reference to a substance should be
incorporated in any account of the meaning of 'literate'. Thus 'man
displaying literacy' (substance and quality) rather than just the indeter-
minate ' . . . displaying literacy' (quality only) would be the appropriate
account. Anselm resists this grammarians' claim by means of arguments 48
47. De grammatico Ch. 12, 18; PL 64, col. 194C; Henry 1967, pp. 86-91; Pinborg 1972, pp. 43-7.
48. De grammatico Ch. 12-14.
that chisel away the proposed 'man' component of the meaning of'hterate'.
These involve a terminology which, while not novel, was to become
widespread in the next century.
Appellatio and signiflcatio

Appellatio (calling) is for Anselm that facet of meaning whereby a name in
actual use points to its referents; in the case of paronyms these may be
constant in kind (as with 'literate') or varied (as with 'white'). In contrast
significatio (meaning) strictly or properly (perse) so-called conveys only the
understanding or concept (intellectus) in the definition associated with a
name; for paronyms the definition must leave place for changes in future
applications by remaining open to referents not hitherto encountered in
actual usage. In this sense 'literate' no more signifies man than does
'white', 49 and hence must, like 'white', be said to signify per if just the
appropriate quality, i.e., literacy. Thus is justice done to the logician's
contention and the methodological presuppositions which support it. At
the same time, and so as to do justice to the grammarian's view, it can be
said that 'literate' also signifies in an improper or oblique sense (per aliud) its
referents; or, in the terminology mentioned above, 'literate' is appellative of
those referents. 50 The distinction between significatio (meaning) and ap-
pellatio (calling) or impositio (denoting) was henceforth to be the subject of
extensive discussion.5 ' Thus the twelfth-century Ars Meliduna persists in
maintaining the distinction, while also continuing to stress the connection
between appellatio and the actual history of speech usage. 52 Abelard is
alluding to the same distinction when he insists that the intellectual import
rather than the denotation (omnia quibus est impositum) is the proper
Aristotelian sense of'signify'. 5 3
This association of significatio with understanding served to separate the
abstract account of meaning given by the logician from the descriptive
49. De grammatico Ch. 13.

50. De grammatico Ch. 12; cf. Kretzmann 1967, pp. 368-9.
51. De Rijk 1962-7,11(1), pp. 213,228-9,536-48; Nielsen 1976, pp. 42-3; Fredborg 1973, pp. 14,16,
20-3,31.
52. 'Notandum tamen quod institucio vocum non fuit facta ad significandum, sed tantum ad
appcllandum, quippe cum appcllacio vocum magis sit necessaria ad loquendum de rebus subiectis
quam significacio'; De Rijk 1962-7,11(1), p. 294; cf. Pinborg 1972, pp. 55-8.
53. De Rijk 1956, p. 562.20-5: 'est attendendum quare "significare" divcrsis modis accipitur. Modo
enim circa ca tantum de quibus intellectum générât, modo vero circa omnia quibus est im-
positum, solet accipi et secundum quidem prima m et propriam significationem ita Aristoteles in
Libro Periermenias "significare" descripsit '.
138 The old logic
account given by the grammarian. Aristotle had spoken of a word's

'settling the understanding' (constituere intellectum),5* and the whole early
medieval tradition habitually used that expression for indicating exacdy
what a word conveys. 5 5 'Understanding' thereby has its sense linked with
meaning, so that William of Conches' maxim 'Even as it can be under
stood, so also can it be signified' (ita ut potest intelligi potest significari)56 could
well stand as the motto of those complex inferences based on 'intelligere' (to
understand) which adorn the early chapters of Anselm's De grammatico. His
countenancing of the ungrammatical 'grammaticus est grammatica' (literate is
literacy) as the de re (thing-centred) correlate of the de voce (word-centred)
assertion that 'literate' strictly signifies literacy 5 7 hence corresponds exactly
to William of Conches' later assertion that the two names 'album' ('white')
and 'albedo' ('whiteness') do not differ in respect of what is signified ('in re
significata').58 William also appreciates that when we talk of a name's
signifying substance or quality, or of its content (intelligibile), we are
working at the same level as that of the semantically complex talk of species
alluded to above, 5 9 which was later to be subsumed under the umbrella of
suppositio simplex.60
The usefulness of linguistic reformulation

The work of Anselm in logical and linguistic theory has, until recently,
been completely overshadowed by his acknowledged accomplishments in
theology and metaphysics (which have been discussed in The Cambridge
History of Later Greek and Early Medieval Philosophy).61 His concern with
questions of nominal meaning was all of a piece with that of his later
contemporaries, however. He is likewise at one with them in observing
that for logical purposes, such as the statement of the de re counterparts of de
62
voce meaning-statements, violation of the rules of the grammar of cur-
54. PL 64, col. 309B-C, 430C; cf. Henry 1974, p. 275.

55. E.g. De grammatico Ch. II, 14; cf. Henry 1974, pp 93-7; William of Conches (De Rijk 1962-7,1,
p. 222); Abelard (De Rijk 1962-7,11(1), p. 195; De Rijk 1956, p. 112.6-8, p. 128. 9-16, p. 147.30,
Ρ- 15333. P- 15538. Ρ 502.25)·
56. De Rijk 1962-7 11(1), p. 223.
57. De grammatico Ch. 5, 12, 16, 18.
58. De Rijk 1962-7,11(1), p. 224. Cf. Gilbert of Poitiers as mentioned in Nielsen 1976, p. 65 and, in
general, Pinborg 1972, pp. 43-50.
59. De Rijk 1962-7, 11(1), p. 224.
60. Henry 1967, §3.4.
61. Liebschütz 1967, pp. 611 -39.
62. See De grammatico, Ch. 18, for this distinction. Anselm's triple sense of the word 'word' is given in
Monologion, Ch. 10. Evans 1978 and Liebschütz 1967 deal with the theological applications
thereof, and Fredborg et al. 1975, pp. 52-3 may be consulted as an example of the use of such a
rent speech may be a useful reminder of semantic profundities. Thus

technical considerations63 relating to precisive (per se) meaning may call
forth deviations from the norms of that ordinary usage (usus loquendi)
which is the province of the descriptive grammarian. The latter is in no
better position when it comes to the relation between his technical asser-
tions and the way things are: his grammatical classifications of certain verbs
as active or passive and of names as masculine, feminine, or neuter, may
well be at odds with the nature of things.64
Anselm goes beyond his contemporaries, however, in claiming that a
like deviation is sometimes apparent when verbs (as distinct from the
names with which De grammatico is concerned) are in question. For
example, usage allows one to assume that a correct sense of the verb 'can' is
embodied in 'The universe, before it actually exists, can exist.' In a stricter
language embodying the precisive (per se) sense of that verb, however,
sentences of the form 'X can .,.' require, according to Anselm, that the
object named by the subject term 'X' in this context should in fact possess
the capacity which the use of this sentence-form properly implies. A
nonexistent object has no capacities,65 and the form of the sentence con-
cerning the universe is hence incorrect. It should be reformulated with the
correct subject term 'God', since it is God who, before the universe is, can
make it to be. 66 Similarly, 'Hector can be conquered by Achilles' fails to
display the proper, precisive sense of'can', and rewriting it as 'Achilles can
conquer Hector' reveals Achilles as the one who truly possesses the capacity
implied by 'can'.67 In De grammatico 15 the distinction between the precis-
ive (per se) and improper (per aliud) senses of 'meaning' is said to be thus
generally extendable to verbs as well as to names.
Anselm's general scheme of linguistic analysis

Before 1936 the exact import of this extension would have been rather
mysterious. Research on the many examples of sentence-analysis that are
scattered throughout Anselm's works could have produced a hypothesis
triple sense in speculative grammar. The 'mental' sense of 'word' was to become the 'mental
word' prominent in later medieval logic.
63. On Anselm's attitude to technical language in general see Henry 1967, §a; Evans 1977; Hopkins
1976, Ch. 1 and pp. 66 and 92.
64. De grammatico, Ch. 18.
65. For Anselm's development of the consequences of this principle in modal logic see Schmitt 1936,
pp. 23-4; cf. Henry 1967, pp. 158-64.
66. De casu diaboii, Ch. 12.
67. De veritate; Ch. 8, De libertate arbitrii, Ch. 5; Cur Deus homo 11, Ch. 10; cf. Henry 1967, pp. 154-5,
166-73.
140 The old logic
that the allusion was to cases such as those just described; but one might also
have formed the further, bolder conjecture that a unified systematic pro-
cedure lay behind such analyses. That such a conjecture may well be
justified is suggested by Anselm's 'Philosophical Fragments', first brought
to light by F. S. Schmitt in 1936. 68 The Fragments constitute an incomplete
draft of what could be construed as a general scheme for sentence-analysis,
designed to bring out the contrast between the looseness of ordinary usage
and the strictness of logically precise language.
In order to assemble a set of general sentence-forms Anselm argues in the
Fragments that the Latin verb facere' ('to do', 'to make', 'to bring about')
can stand in the place of any verb (including even 'not to do'). 6 9 He
concentrates on the four general forms facere esse...' ('to do so t h a t . . . ' ) ,
'facere non esse ...' ('to do so that n o t . . . ' ) , 'non facere esse ...' ('not to do so
t h a t . . . ' ) , and 'non facere non esse ...' ('not to do so that n o t . . . ' ) . The
relations of contrariety and contradiction described as holding among
these general forms parallel those holding among the modal expressions
'posse esse ...' ('to be possible to be . . . ' ) , 'posse non esse ...' ('to be possible
not to be . . . ' ) , 'non posse esse ...' ('to be not possible to be . . . ' ) , and 'non
posse non esse ...' ('to be not possible not to be . . . ' , i.e., to be necessary), as
described by Aristotle in De interpretatione and by Boethius in his commen-
taries on it. 70
Anselm claims that the first of the four general sentential forms (i.e.,
facere esse . . . ' , 'to do so t h a t . . . ' ) has a proper sense of its own. In that sense,
one does so that so-and-so if one does so that so-and-so, which was not the
case, becomes the case.7 ' But it is sometimes used improperly in a context
that is strictly a case of not doing so that not-p (using 'p' as a propositional
variable instead of the 'so-and-so' of the preceding sentence). For example,
the looseness of ordinary usage allows us to say that one does so that (brings
it about that) someone is dead when in fact the context is such that one has
not done so that it is not the case that the person is dead, as when one refrains
from using one's (perhaps miraculous) power to restore the dead to life.72
And Anselm carefully distinguishes other contexts in which the same
general form 'one does so that p' is improperly used. 73
68. Schmitt 1936; Southern and Schmitt 1969, pp. 334-51; translated Hopkins 1972, pp. 215-42.
69. Schmitt 1936, pp. 25-7; Southern and Schmitt 1969, pp. 337-8, cf. Henry 1967, pp. 121-3.
70. Henry 1967, pp. 136-8, 124—5.
71. Schmitt 1936, p. 32; Henry 1967, p. 126.
72. Schmitt 1936, pp. 29, 31; Henry 1967, p. 126.
73. Schmitt 1936, pp. 29-32; Henry 1967, pp. 126-7.
Anselm's Fragments attain a high degree of systematic generality.74 His

many examples suggest that while his general project demands exploi
tation of the widest sense of facere', he is trading on the fact that this verb
has 'to bring about' as one of its particular senses. It is this which has led
contemporary commentators to credit Anselm with having in the Frag
ments the foundation of action theory (praxiology).75 He provides com
plex rules of transformation by means of which modes and examples
relating to the negative forms ('not to do so t h a t . . . ' and 'not to do so that
not...') may be generated from those relating to the affirmative forms ('to
do so that...' and 'to do so that n o t . . . ' ) . 7 6 He also discusses the extent to
which the patterns thus evolved apply to other verbs considered as
instances of facere' in its most general sense ('to do'), 7 7 and he stresses his
awareness that further research and elaboration of his proposals are
needed.78 Perhaps the best example of such possible developments is to be
found in his extended remarks on 'velle' ('to will'), which he assimilates to
'facere' in several respects.79
The utilisation of this architectonic draft for an analytic scheme (if that is
really what it is) extends into nearly all of Anselm's works. Some of the
material on facere' recurs elsewhere,80 and the system is used to solve
puzzles concerning verbs of capacity and liability, as well as some involv
ing the verbs 'to give' 8 1 and 'to be obliged to .. ,'. 8 2
Anselm on 'nothing'
A final example of the all-pervasive distinction between the proper (log

ical) and improper (merely usage-approved) forms of an utterance may be
drawn from Anselm's treatment of the pseudo-name 'nothing'. The lin
guistic puzzles centring around this and kindred words are manifest and
numerous. Lewis Carroll's 'Nobody walks much faster than I do' requires
(if inferential oddities are to be avoided) that the 'nobody' be recognised as
74. E.g., 'Dicimus namque rem quamlibet facere aliquid esse aut quia facit idipsum esse, quod facere
dicitur, aut quia non facit idipsum non esse, aut quia facit aliud esse, aut quia non facit aliud esse,
aut quia facit aliud non esse, aut quia non facit aliud non esse' (Schmitt 1936, pp. 29, 30).
75. Danto 1973; Walton 1976.
76. Schmitt 1936, pp. 30-3; Henry 1967, pp. 126 and 128-9.
77. Schmitt 193Θ, pp. 33-7; Henry 1967, p. 129.
78. Schmitt 1936, p. 29; Henry 1967, p. 124.
79. Schmitt 1936, pp. 37-9: Henry 1967. pp. 130-3.
80. De veritate, Ch. 5; De casu diaboli, Ch. 1; cf. Henry 1967, pp. 181-5,
81. Decasu diaboli. Ch. 18, 20; Cwr Dens homo, 11, Ch. 10; cf. Henry 1967, pp. 185-91.
82. Schmitt 1936, pp. 35—7; Cur Deus homo 11 18; Dc veritate 8; cf. Henry 1967, pp. 191—201.
142 The old logic
its merely grammatical (but not logical) subject; it has its medieval counter
part in Anselm's 'Nothing taught me to fly' ('Nihil me docuit volare'). This
too could lead to odd inferences (e.g., that I have been taught to fly), were
'nothing' taken to be a name. Modern logicians would analyse 'nobody' or
'nothing' out of the proposition by recourse to propositional negation and
the quantifier 'there exists an χ such t h a t . . . ' . Thus Anselm's sentence
would become 'It is not the case that there exists an χ such that χ taught me
to fly', i.e., 'It is not the case that something taught me to fly.' This is the
counterpart of Anselm's Latin analysis 'NOM me docuit aliquid volare', in
which the initial 'non' plainly performs the office of propositional nega
tion. 8 3 By means of various examples Anselm underlines the analogy of
grammatically permissible structure holding between sentences in which
the nominal positions are occupied by empty and negative names, and
those in which such positions are occupied by names that do refer. The
name 'nihil' is then seen as signifying something 'secundumformam loquendi'
(i.e., from the viewpoint of grammatical structure) but not 'secundum rem'
(i.e., from the viewpoint of reality). 8 4
Apparent form and real form of propositions

These developments may be construed as further natural consequences of
that rivalry between logic and grammar which was a pervasive feature of
Anselm's intellectual context, a rivalry that helps to account for the
sometimes striking resemblance between his theses and the present-day
contrast between the 'real' (or 'logical') form of an utterance and its
'apparent logical' (or 'merely grammatical') form. It is not only 'Russell's
merit' (as Wittgenstein put it) but also Anselm's 'to have shown that the
apparent logical form of the proposition need not be its real form'. As far as
the medievals were concerned, this was by no means the end of the matter.
Equally trenchant obverse statements of this same contrast were to emerge
from the side of autonomy-seeking grammarians. 8 5
83. Monologion, Ch. 19; Henry 1967, pp. 218-19.

84. De casu diaboli, Ch. 11; Henry 1967, pp. 211-18.
85. Pinborg 1972, pp. 111-12.
6
ABELARD AND THE C U L M I N A T I O N OF
THE OLD LOGIC
Abelard's conception of logic

Of all the scholastic logicians writing while the old logic (logica vetus) was
still virtually the whole of the logical curriculum in the schools, Abelard is
generally conceded to have been the most profound and original. 1 He
himself was keenly aware of the subtlety required of the logician and in one
place says it depends on a divinely bestowed talent, rather than anything
that can be developed by mere practice. 2 Abelard treats dialectic ( = logic)
as an ars sermocinalis, i.e., like grammar a linguistic science. Its peculiar
subject matter is arguments as expressed in language, whose validity it tries
to judge in a scientific way. 3 This linguistically oriented conception of the
subject means that dialectic will overlap to some extent with grammar. In
the first section below I shall selectively explore this overlap; in the second
section I shall consider some of Abelard's views on more purely logical
topics.
For Abelard logic also had a close relation to physica* i.e., the sciences of
nature, since in explaining the 'uses of words' the logician must investigate
in a general way the 'properties of things' which the mind uses words to
signify. This relationship leads to a concern with the psychology of
signification, to be explored in the third section below, and with ontology,
the topic of the fourth section. This discussion is necessarily very selective
and must omit consideration of many of Abelard's philosophical insights
on relevant topics.
Dialectic and grammar

Throughout his logical works Abelard frequently deploys terms and
analyses borrowed from grammar and generally views his own dialectical
ι. For example, Jolivet 1969, p. 33 8: 'Logicien, Abélard doit être mis très haut. On ne connaît avant
lui aucun médiéval qui soit de sa taille.'
2. See Abelard 1956, p. 471.4-9.
3. Sec Abelard 1954, p. 209.
4. See Abelard 1956, pp. 286.31-287.1.
144 The old logic
enterprise as deepening and to some extent correcting what grammar has

already begun. The main contribution of dialectic is its subtler inquiry into
what and how words signify, a study that sheds much light on many
grammatical categories and constructions. It is clearly Abelard's assump
tion that the two disciplines should work hand in hand, composing what
would in effect be a single science of language. 5 This is illustrated in the
following summary of Abelard's work on a few crucial questions.
The distinction of nouns or names (a category that includes adjectives)
from verbs was as much argued over in Abelard's day as in ancient times,
and it provided Abelard with an entry to matters that lie at the heart of his
philosophy. He rejects Aristotle's view that verbs differ from nouns in
having a consignification of time, for he sees no reason not to ascribe a
similar consignification to nouns:
For just as 'run' or 'running' indicate running in connection with a person as

presently inhering in him, so 'white' determines whiteness in connection with a
substance as presently inhering, for it is called white only because of present
whiteness.6
Abelard is equally unhappy with the idea that verbs, as distinct from other
parts of speech, signify only actions and passions, or what might better be
called 'receptions' (passiones, i.e. the passive correlates of actions). This
view runs afoul of the copula which although accepted as a verb can be used
to 'join' (copulare) to its subject any sort of entity whatsoever. 7
What distinguishes verbs, in Abelard's view, is that they provide the
'completeness of sense' (sensus perfectio) characteristic of whole sentences
(orationes perfectae) as distinct from mere phrases (orationes imperfectae).
What is the difference, he asks, between Ά man runs' and 'a running man'?
Of constructions such as the latter he remarks:
But a completeness of sense has not yet been brought about in them; for when this
expression has been uttered the mind of the hearer is suspended and desires to hear
more in order to arrive at completeness of sense, for example, 'is' or some other
8
acceptable verb. For without a verb there is no completeness of sense.
5. Jolivet 1969, p. 55: ' . . . préparée par les Closes, une science unifiée du langage sous-tend les traites
de la Dialectique . . . il a fondu en un seul corps deux arts qu'il avait reçus séparés.' Abelard did
compose a grammatical treatise, but it is lost. See Van den Eynde 1962, pp. 473-6.
6. Abelard 1956, p. 122.22—5: 'Sicut enim "curre" vel "currens" cursum circa personam tamquamei
praesentialiter inhaerentem demonstrat, ita "album" circa substantiam albedincm tamquam
praesentialiter inhaerentem déterminât; non enim album nisi ex praesenti albedine dicitur.'
7. ft«., pp. 130.32-131.7.
8. Ibid., p. 148.26—30: 'Sed nondum in eis compléta est sensus perfectio. Adhuc enim praemissa
oratione prolata suspensus audicntis animus aliquid amplius audirc dcsidcrat, ut ad perfectionem
Abelard and the culmination of the old logic 145
Verbs can perform this function because they propose the 'inherence' of
what they signify in the subject.
Thus we see that this completeness of sense depends mainly on verbs, since only by
them is inherence of something in something indicated in a manner expressive of
different mental states; without this inherence there is no completeness of sense.
When I say 'Come to me' or 'If only you would come to me', in a way I propound
the inherence of coming to me in a manner expressive of my order or my desire; in
the one case I order that coming should belong to him, in the other I have a desire,
namely, that he come.9
The talk of'inherence', however, must be treated delicately, for it is not

Abelard's view that any verb, even the copula, signifies some relational
property of inherence. Rather verbs generally signify that which 'inheres',
while the copula, according to one of Abelard's accounts of it, signifies
nothing at all. If any verb were to signify 'inherence', then it would be
unable to perform the 'linking' function, i.e., the function of the copula,
which is to 'link' what the predicate signifies to the subject. ' R u n s ' signifies
running and 'links' it as well. The copula 'is' really expresses just this
linking function implicit in all verbs; if it were to signify anything on its
own it could no more take a predicate noun or adjective and link its
significate to a subject than can 'runs'. 1 0
This view is taken still further when Abelard separates 'to be' used as a
copula from 'to be' used to mean 'to exist'. If the separation is not made we
have problems with sentences such as 'Homer is a poet' and Ά chimera is
conceivable', where the subjects are non-existent or even impossible.
Abelard's solution is to treat the whole phrase consisting of copula plus
predicate noun or adjective as a single verb-phrase and in this way elimi
nate any idea that 'to be' on its own is predicated of the subject.
Thus it seems to me if I may dare to speak freely, that it would be more rational and
satisfying to reason t h a t . . . we understand as a single verb 'to be a man' or 'to be
white' or 'to be conceivable'. Aristotle indeed says that in 'Homer is a poet' 'to be' is
predicated per accidens, i.e., 'to be' is predicated accidentally of Homer in that the
sensus perveniat, veluti "est", aut aliquod aliud competcns vcrbum. Praeter verbum namque nulla
est sensus perfectio.'
9. Ibid., p. 149.20-6: 'Perfectio itaque sensus maxime penderc dinoscitur in verbis, quibus solis
alicuius ad aliquid inhaerentia secundum varios affectus animi demonstratur; praeter quam
quidem inhaerentiam orationis perfectio non subsistit. Cum enim dico: "Vent adme" vel "utinam
venires adme", quodammodo inhaerentiam veniendi ad me propono secundum iussum meum vel
desiderium meum, in eo scilicet quod iubeo illi ut venire ei cohaereat, vel desidero, idest ut ipse
veniat.'
10. See Abelard 1919-27, p. 362.25-9.
146 The old logic
poem belongs to him, but it is not predicated per se of Homer that he is. But since
'to be', as was said, is not a verbal unit, to be predicated per accidens is not to be
predicated; rather 'to be' is part of the predicate."
Abelard in effect wants us to treat the copula as what a modern grammarian
would call an auxiliary, and indeed Abelard draws support for his view
from the implausibility of dividing up 'eritsedens' (will be sitting) into two
parts, because of the conflict of tenses between auxiliary and participle. 12
The copula, then, turns out to be a verb-phrase-maker, taking as com-
plements nouns, participles, and whole clauses and turning them into verb-
phrases. But the process should not be thought to leave the complement
with the same meaning it has in isolation, for this leads to logical ab-
surdities. As we have seen, the noun following the copula has in isolation a
tense of its own (generally the present), and this can conflict with the tense
of the copula. For example, 'This old man was a boy' will be necessarily
false if we treat 'boy' as retaining its signification of present time, for then
the sentence is equivalent to 'This old man was one of those who is
presently a boy.' 1 3 What we must do, Abelard says, is treat the whole
copula plus predicate noun as a single verb having the tense of the copula.
He shows how only in this way can the rules of conversion and syllogistic
inference be made to apply to sentences with verbs in tenses other than the
present. ' *
Having gone this far, Abelard sees that nouns in isolation really have the
same signification as verb phrases, but just lack the ability to 'link'. He puts
it this way:
It seems that the signification of a substantive verb [i.e. a copula], to which time is
joined as well, is associated with nouns just as it is with verbs. For just as 'runs'
amounts to saying 'is running', so 'man' amounts to 'is a mortal rational animal'. ' 5
11. Abelard 1956, p. i38.ii-22:'Undemihi,siprofiteriaudeam,illudrationabilius videtur ut rationi
sufficere valeamus, ut scilicet, qucmadmodum oppositionem in adiecto secundum oppositionem
magis quam secundum appositionem sumimus, ita accidentalem praedicationem accipiamus, ac
cum dicitur: "est homo" vel "est opinabile" vel "est album" pro uno verbo "esse hominem" vel
"esse album" vel "esse opinabile" intelligamus. Quod vero Aristoteles, cum dicitur: "Homerus
est poeta", dirit per accidens "esse" pracdicari hoc modo: "secundum accidens enim praedicatur
'esse' de Homero, quoniam inest ci poema, sed non secundum se praedicatur de Homero
quoniam est", cum non sit "esse", ut dictum est, una dictio, praedicari per accidens non est
pracdicari, immo pars est "esse" praedicati.'
12. Ibid., pp. 138.26-139.il.
13. Sec Abelard 1919-27, pp. 348.28-349.17.
14. See Abelard 1956, pp. 139.12-140.22.
15. Ibid., p. 122.28-31 with the correction suggested by Tweedale 1976, p. 288: 'Sicut enim sub-
stantivi verbi significatio, cui quoque tempus adiunctum est, verbis adiungitur, sic et nominibus
videtur. Sicut enim "currit" quantum "est currens" dicit, ita "homo" tantumdem quantum "est
animal rationale mortale".'
It is not, then, that what is basic from a logical point of view is two items,
the copula and the noun, and that the verb phrase which grammatically
combines these is logically derivative from them. Rather the noun in
isolation, or whenever it is used to denote things (i.e. appellative), derives its
meaning from the verb phrase. What is logically basic, i.e. basic for the
analysis of meaning, is the combination of copula and predicate noun
considered as an indivisible unit.
Abelard is not unaware that his inquiry has led him to treat lightly many
canons of the grammarians, but he urges his reader to consider dialectic the
subtler discipline:
Do not recoil in horror when in order to open up the correct meanings of
sentences we put together many verbs or substantives, such as 'to be a man', or
verbs derived from things other than actions or passions, such as 'to be white', and
thus seem to go against the rules of grammarians. For those who are in the first
rank of the discipline, out of a consideration for the capacities of beginners, have
left to the advanced the inquiry into and correction of many matters in which the
subtlety of dialectic must be employed.' 6
Abelard's logical inquiries have led him to find an underlying similarity
where the mere grammarian sees only basic divergence. The implication is
that the grammarian, at least in large part, sets out on an inquiry which can
really be completed only by the more difficult art of dialectic. And this in
turn implies that the study of meaning, or semantics is more fundamental
to the study of language than mere syntax.
O n the other hand, Abelard is fond of showing how certain grammatical
constructions do not admit of logical analysis in any straightforward way.
He does this in order to correct pseudo-dialecticians who naively apply
standard Aristotelian logic where it is in fact quite out of place.
A very important instance of this use of grammar concerns the notion of
an 'impersonal' construction, an idea central to some of Abelard's most
puzzling remarks in ontology. Impersonal constructions in grammar are
sentences without any subject, such as 'It is raining' or 'It is going well'. In
such cases English supplies the grammatical subject 'it', but from a logical
point of view the sentence lacks a subject altogether.
This phenomenon of language encourages Abelard to treat sentences
16. Ibid., p. 140.23-9: 'Quod autem grammaticorum regulis contrarii videmur, quod multa compo-
nimus verba vel substantiva, ut "esse hominem", vel ab aliis quam ab actionibus vel passionibus
sumpta, ut "esse album", propter rectam enuntiationum sentcntiam aperiendam, non abhorrcas.
Uli enim qui primum disciplinae gradum tenent, pro capacitate tenerorum multa provectis
inquirere aut corrigenda reliquerunt in quibus dialccticac subtilitatem opertet laborare.'
148 The old logic
whose grammatical subjects are nominalisations of sentences or verb

phrases as logically subjectless. *7 Thus modal propositions such as 'That a
chimera does not exist is true' or 'It is impossible that a chimera exist' can be
treated as subjectless. Consequently, Aristotelian term logic is utterly
inapplicable to these sentences. Abelard says bluntly of modal words such
as 'true' or 'possible':
But where impersonal nouns are predicated impersonally no conversion is
admitted . . . and neither can they properly be called universals when they do not
cover anything personally [i.e. do not denote]. Nor can they be used universally or
particularly .. . 18
For the moment it suffices to note that in these cases Abelard does not
find the grammatical fact of impersonality just another occasion on which
dialectic can correct the grammarian. Rather he turns this point of gram-
mar into a weapon against a pseudo-dialectic that feels impelled to confine
all of language to the standard term logic that the logica vetus took over
from Aristotle.
Pure dialectic
The conception of dialectic and grammar as overlapping in interests would
be difficult if not impossible had Abelard not taken words to be the subject
matter of logic. But Abelard is very aware that this approach, one shared by
most of his immediate precursors, is easily misunderstood. He reminds us
that words are of interest to the logician because they signify, not because
of their physical characteristics. Dialectic is not interested in phonology.
And yet Abelard does not allow this obvious point to draw him into
postulating words (sermones) as an entirely separate class of things from
physical vocal utterances (voces).
The problem is to allow that the logician uses many predicates such as 'is
a universal' or 'is a genus' which apply only to words and not to utterances,
while maintaining that words and utterances are the same things. In Logica
'Nostrorum petitioni sociorutn l 9 Abelard tentatively suggests a solution. It
requires treating these predicates of logic much like the predicate 'was
made by Socrates', which could be true of a statue but not of a piece of
17. See Abelard 1919—27, p. 361.26-36. Also discussion in Tweedale 1976, pp. 244-72.
18. Minio-Paluello 1956-8,11, p. 19.17—21: 'At vcro, ubi inpersonalia nomina prcdicantur inpcr-
sonaliter, nulla (ut dictum est) conversio fit, nee ipsa, (cum aliquid personaliter non contineant),
universalia proprie dici possum, nee universaliter nee particulariter profcrri possunt, sed semper
ad infinitivum modum construuntur.'
19. See Abelard 1919-27, pp. 522.10-524.24. Also discussion in Tweedale 1976, pp. 142-62.
stone even when the statue is the piece of stone. This can happen because
'This statue was made by Socrates' really means 'This statue was made by
Socrates to be a statue', and likewise 'This piece of stone was made by
Socrates' means "This piece of stone was made by Socrates to be a piece of
20
stone'. Now, the predicates of logic, e.g. 'is a universal', can similarly be
affected by their subjects, if we allow that they implicitly assert of some
thing that it is made or 'established' by the conventions of language. Then
since a word is made to be a word by linguistic conventions but an
utterance is not made to be an utterance by linguistic conventions (it is an
utterance by nature), we can explain why words but not utterances are
universals, although words and utterances are not different things.
When we turn to the meanings of words, we find that in the case of
nouns Abelard distinguishes significatio (signification) from appellatio (de
noting) or nominatio (naming). A noun appellat or nominal the things a
modern logician would say it denotes or is true of, while it signifies
(significat) some property belonging to whatever it denotes. Abelard is
keen to distinguish denotative from non-denotative uses of nouns. The
former, it seems, can be replaced by an indefinite pronoun plus relative
clause without changing the meaning of the whole sentence. 2 ' For ex
ample, Ά boy is running' means 'Something which is a boy is running'.
Furthermore, in this latter sentence the occurrence of'a boy' is not denota
tive since, according to Abelard, no predicate noun is denotative. His
reasons here are closely allied to his arguments for treating copula plus
predicate noun as a single verb. 2 2 If'a boy' in 'This old man was a boy'
were denotative, the whole sentence would mean 'This old man was
something which is a boy' and would consequently be absurd. We see then
that Abelard has provided himself with a way of replacing all denotative
occurrences of nouns with quantifier pronouns, such as 'something' or
'anything', accompanied by a relative clause with a verb constructed from
the copula plus the noun in question used non-denotatively, as when
'something which is a boy' was substituted above for 'a boy'.
This idea is closely related to a remark Abelard makes about 'is' or 'exists'
(i.e., 'est') in his Theologia Christiana. What he proposes there is that we not
treat it as a predicate taking a denotative noun as subject. For example, in
the sentence Ά father exists' we are not saying 'Something which is a father
20. See the account of Abelard's view in Introductiones Montane minores as edited in De Rijk 1962-7,
11(2), pp. 15-16.
21. See Tweedale 1976, pp. 166-9.
22. See nn. 13 and 14 above.
150 The old logic
exists' but simply 'Something is a father'. In other words, 'exists' disappears

as a predicate on its own to be replaced by a quantifier plus a verb
constructed from a copula plus the noun which was the subject in the
original. 23 Abelard is innovating here, and he has moved very far in the
direction of letting quantificational pronouns operating on verbs carry the
burden of both reference and existential import, a move whose full
possibilities were not realised until Frege formulated it mathematically in
the nineteenth century.
Central to dialectic is the notion of validity as applied to arguments and
inferences. Abelard says of inference that it 'consists in a necessity of
entailment — namely, in the fact that the meaning of the consequent is
demanded by the sense of the antecedent.' 2 4 The antecedent and con-
sequent must be so related that: 'It cannot happen that it is as the first says
but it is not as the second says.' 2 5 In other words, the truth of the
antecedent requires the truth of the consequent. Abelard asks on what this
'necessity of entailment' rests. Obviously it is dependent not on the
physical character of the sentences as utterances but on their meanings.
Abelard rejects the view which would treat psychological realities, i.e.
thoughts (intellectus), as meanings, and then claim the entailment here is
one between these thoughts. The thought expressed by the antecedent, he
says, can exist entirely without that expressed by the consequent. 26 Nor is it
possible, on his view, to see the entailment as one between extra-mental
things, for then there would be no entailment when the things in question
were non-existent. But there can be true conditionals (or consequences:
consequentiae) about the non-existent:
For example, since we admit that the consequence 'If there is a rose, there is a
flower' is always true and necessary, even when the things are destroyed, we have
to see why the signification isjudged to be necessary. None of the necessity is in the
things, however, since even when they have been completely destroyed, that
which is said by the consequence — i.e., that if this is that is - is no less necessary.27
23. See Theologia Christiana (PL vol. 178), 1312D—1313B; Abelard 1956, pp. 132.36—133.23; and
Tweedale 1976, pp. 195-9.
24. Abelard 1956, p. 253.28-9: 'Infcrentia itaque in necessitate consecutionis consistit, in eo scilicet
quod ex sensu antecedentis sententia exigitur consequentis...' See discussion in Jolivet 1969,
pp. 148—66.
25. Abelard 1919-27, p. 367.8-9: 'Sed non potest contingere ita ut prior dicit quin sit ita ut dicit
posterior.'
26. See Abelard 1956, p. 154.30-8.
27. Abelard 1919—27, p. 366.6—12: ' . . . veluti cum istam consequentiam "Si est rosa, est flos", veram
semper etiam destruens rebus et necessariam concedamus, oportet videri, pro quo significatio
necessaria iudicetur. At vero in rebus nihil est necessitatis, quibus etiam omnino destructis non
minus necessarium est, quod a consequentia dicitur, id est, si hoc est, illud esse.'
Rather it is from what the sentences say, their dicta, that we judge whether
or not the necessity of connection is present. But these dicta, or significates
of propositions, are neither mental nor extra-mental things.
Abelard is very aware that the necessity of entailment may arise from the
28
very form of the sentences involved, as with standard syllogisms. On the
other hand, this necessity may rest on the significates of the terms, as in the
inference 'This is a man, therefore it is an animal.' These latter give rise to
the doctrine of loci (topics) and the development of certain rules, maximae,
that express the principles involved.29 For example, the cited consequence
holds because whatever a species is predicable of, the genus ofthat species is
also predicable of.
In general Abelard will not allow that affirmative categorical prop
ositions can be necessarily true, for their truth requires that the subject
term denote some existent thing, but all things other than God have a
contingent existence. Natural science, of course, asserts many categorical
propositions which must be true as long as there do exist the things it is
talking about, but this qualified necessity always rests on a 'law of nature'
(lex naturae), i.e. a hypothetical that expresses a necessity of entailment. Ά
man is an animal' must be true as long as there are men, and this conditional
necessity depends on the 'law of nature' that the nature of man cannot exist
without the nature of animal being in it, 3 0 i.e., the law that finds expression
in 'If there is a man, there is an animal.' Science, for Abelard, is the search
for these necessary entailments holding between the natures of the things
that actually exist.
These entailments, or their absence, also determine the truth of modal
propositions - i.e., those that use the terms 'necessary', 'possible', 'im
possible', or expressions equivalent to these. Here Abelard nicely dis
tinguishes two senses of modal propositions, one per divisionem and one per
31
compositionem. If we say it is possible for one standing to be sitting, we
may mean to attribute, so to speak, possible truth to the proposition 'One
standing is sitting', and the result is absurd since that proposition cannot be
true. This is the sense per compositionem or de sensu propositionis. On the other
hand we may mean to attribute to someone standing the possibility of
sitting, and then the assertion is very likely true. This is the sense per
28. See Abelard 1956, p. 256.20-4.

29. See Jolivet 1969, pp. 153-68, and Bird 1959.
30. See Abelard 1956, p. 280.12-4.
31. Sec Minio-Paluello 1956-8, 11, pp. 13.15—14.4. Abelard says he draws the distinction from
Aristotle's Sophistici elenchi (i65b26, i66"23-8); which shows he had some acquaintance with at
least that portion of the logica nova.
152 The old logic
divisionem or de re. Under this latter interpretation Abelard can treat the
modal proposition as really categorical with a modal predicate, and all the
rules of Aristotelian term-logic apply. But when the modal proposition is
interpreted de sensu proposition is, Abelard prefers not to treat the nominalisa-
tion of the sentence as a logical subject naming something of which a
modal property is predicated. Consequently it falls totally outside the
categories of Aristotelian term-logic, 3 2 and is, as noted above, to be treated
as impersonal, i.e., subjectless.
The psychology of signification

Abelard seems to take the view that the basic role of language is to express
overtly certain activities of the mind of the speaker which are prior to and
independent of language itself. 33 But while allowing that language makes
thoughts overt, Abelard is very keen to deny that language is used solely or
even primarily to describe thoughts. Words are not used, except when we
do psychology, to talk about the thoughts themselves; rather they are used
to talk about the same things the thoughts are about. In the following
passage Abelard talks of words signifying both ideas and things, but it is
clear that this amounts to noting two quite different kinds of signifying.
For just as nouns and verbs have two significations, one of a thing, another of an
idea, so also the propositions which are made out of them draw from them two
significations, one of ideas, the other of things. Like nouns and verbs, propositions
also deal with things, and like them they generate certain ideas. When we say, Ά
man runs', we deal with the very things, a man and running, and we join running
to a man; we do not link the ideas of them to each other. We are not saying
anything about ideas; we simply put them in the mind of the hearer while dealing
3
solely with things. *
Words signify ideas in the sense that they produce them in the minds of
hearers; they signify things in that they are used to talk about them.
It is easy to misunderstand Abelard here as saying that for each word
there is some thing it signifies. But that interpretation directly contradicts
what he says about how universals, i.e. common nouns, signify.
32. Minio-Paluello 1956-8, II, p. 14.14-7.

33. He even says that in some cases 'the soul grasps an idea better than language can express it'. See
Abelard 1956, pp. 118.29—119.2.
34. Ibid., p. 154.20-9: 'Sicut enim nominum et verborum duplex ad rem et ad intcllectum signi
ficatio, ita etiam propositiones quae ex ipsis componuntur, duplicem ex ipsis significationem
contrahunt, unam quidem de intellectibus, aliam vero de rebus. De rebus enim, sicut ilia,
propositiones quoque agunt ac de ipsis quoque, sicut ilia, quosdam intellectus générant. Cum
enim dicimus: "homo currit", de homine ac cursu rebus ipsis agimus cursumque homini
coniungimus, non intellectus eorum ad invicem copulamus; nee quicquam de intellectibus
dicimus, sed de rebus sous agentes eos in animo audientis constituimus.'
In a common noun such as 'man' neither Socrates nor some other man nor the
whole collection of men is rationally understood via the force of the utterance
Thus it seems that neither 'man' nor any other universal word signifies anything,
since there is no thing the idea of which it introduces.35
This leads directly to one formulation of the problem of universals: what is

the idea expressed by a universal noun an idea of? Abelard notes the close
association of ideas with mental images or likenesses made up in our
imaginations. It is clear he does not identify the idea, which is an act of
thinking of something, with its associated image, for there can be ideas
without images, 36 and in any event the image is not a thing at all, not even
a mental thing, whereas an idea clearly is. 37 Might it be, though, that the
image is what the idea is an idea of? Certainly Abelard says the idea 'grasps'
(concipit) an image, 38 but in the end it is clear that his view is more subtle
than this, for he allows that ideas of different things might be associated
with the same image:
With one and the same image before the mind's eye I can consider both the nature
of quality and the nature of whiteness. Although the image is the same, there are
many ways of conceiving it, as, for example, when I consider it at one time as
being a quality and at another as being white. Thus the force of an idea does not
consist in that mode of signifying which even the beasts bring about with images;
rather it consists in the distinction of attention, which pertains only to reason and
which the beasts lack entirely.39
Human reason uses images to think of any of the characteristics they

exhibit, and indeed it is this ability to vary what the mind is attending to
without necessarily changing the contents of the imagination that Abelard
thinks distinguishes human from sub-human intelligence.
Must we then say that the things we have ideas of are these natures or
properties Abelard speaks of? We shall see in the next section that this too is
a partial misunderstanding of his view. Here we may simply note that in
35. Abelard 1919-27, pp. 18.23-19.6: 'In nomine vero communi, quod "homo" est, nee ipse
Socrates nee alius nee tota hominum collectio rationabiliter ex vi vocis intelligitur nee etiam in
quantum homo est, ipse Socrates per hoc nomen, ut quidam volunt, certificatur ... Nullum
itaquc significare videtur vel "homo" vel aliud universale vocabulum, cum de nulla re constituât
intellectum.'
36. IM., p. 21.18—22.
37. /*/'</., pp. 314.z5-315.17-
38. Ibid., p. 21.27-36.
39. Ibid., p. 329.14-22: 'Similiter eadem imagine ante mentis oculos constituta in ipsa et qualitatis et
albedinis naturam considero et licet sit eadem imago, plurcs sunt dc ea concipiendi modi, quod
modo in eo quod qualitas est, modo in eo quod est album. Vis itaque intellectus non in modo
significandi consistit, quod et bestiae faciunt per imagincm, sed in discretione attendendi, quod ad
rationcm solum pertinet, cuius omnino bestiae expertes sunt.'
154 The old logic
Logica 'Nostrorum petitioni sociorum' Abelard suggests that the whole search
for things our ideas are of is misguided. There he compares the question
'What are you thinking of?' with 'What do you want?', in order to note
that just as Ί want a hood' does not entail that there is some hood I want, so
Ί am thinking of a man' may not entail there is some man I am thinking
of. 40 The careful logician, then, can note that some nouns serving as direct
objects of verbs of mental attitude are non-denoting, i.e. not logical objects
at all. He can then allow that an idea is of something while denying that
there is something which it is an idea of. Basically it is the same point
Abelard made about impersonal sentences, only where they were subject
less these are objectless.
Ontology
What most associate Abelard with nominalists of the later Middle Ages
such as Ockham, and even of our o w n time, such as N . Goodman and
W . V. O. Quine, are his incessant efforts to show that dialectic and the artes
sermocinales in general can be developed without their requiring us to
believe in the existence of things other than those more or less ordinary
ones described by physica. This comes out most clearly in his lengthy
remarks on status and dicta, i.e. the significates of verb phrases (and hence of
nouns, see above) and sentences respectively. T w o similar passages estab
lish this line of approach clearly enough:
But it seems absurd for us to understand the agreement of things in such a way that
it is not some thing, as though we were uniting in nothing things which exist when
we say that they agree in the status of man, that is in this: that they are men. But we
mean merely that they are men and do not differ at all in this regard, i.e. not in as
much as they are men, although we call on no essence.41 We call the status of man
his being a man, which is not a thing. We also say that this is the common cause of
2
the application of the noun to singulars in as much as they agree with each other.*
But it is objected: Since the dicta of propositions are nothing, how is it that
propositions happen to be true on account of them, for how can that be called a
cause which is completely nothing and which cannot be? But a man is hanged on
40. Abelard 1919-27, pp. 532.30-533.9.

41. 'Essence' ( = essentia) has for Abelard, in contrast to later scholastics influenced by Avicenna, the
same meaning as 'thing' ( = res).
42. Abelard 1919-27, p. 20.1-9: 'Abhorrendum autem videtur, quod convenicntiam rcrum
secundum id aecipiamus, quod non est res aliqua, tamquam in nihilo ca quae sunt, uniamus, cum
scilicet hune et ilium in statu hominis, id est in eo quod sunt homines convenire dicimus. Sed nihil
aliud sentimus, nisi eos homines esse, et secundum hoc nullatcnus differre, secundum hoc,
inquam, quod homines sunt, licet ad nullam vocemus essentiam. Statum autem hominis ipsum
esse hominem, quod non est res, vocamus quod etiam diximus communem causam impositionis
nominis ad singulos, secundum quod ipsi ad invicem conveniunt.'
account of a theft he performed which is now nothing; and a man dies because he
does not eat and is damned because he does not act rightly, yet not eating and not
acting rightly are not things.43
This 'dereification' 44 of status and dicta is not to be confused with treating
them either as unreal or subjective. This emerges most clearly in Abelard's
ill-received effort to explain the doctrine of the Holy Trinity. Here he says
we should treat the distinction of the Father, Son, and Holy Spirit not as a
difference in essentia, i.e. as among three distinct things, but as a difference
of three properties or status.*5
The Persons, i.e. the Father, Son, and Holy Spirit, are different from each other in
something like the way that things different by definition or property are different,
i.e., although the very same essence which is God the Son is God the Father or God
the Holy Spirit, nevertheless the property distinctive of God the Father inasmuch
as He is the Father is other than that distinctive of the Son and that distinctive of the
Holy Spirit.*6
Abelard certainly did not want either to deny the objective reality of the
divine persons or treat those persons as some sort of'forms' existing in God
yet distinct in essence from him. Both he considered obvious heresies. The
latter in particular destroys the divine simplicity:
For if the paternity which is in God is an essence other than God, is it not true that
God the Father consists of two things, i.e. of God and paternity, and that He relates
as a whole to these two from which He is made up?*7
Various lines of thought reviewed in the preceding sections meet when
Abelard tries to disentangle logic from its supposed ontological impli-
cations. We recall that verbs must signify in such a way that the whole
sentence says something, rather than just being a name of something. But
43. Ibid., pp. 368.40-369.6: 'Sed opponitur, cum dicta propositionum nil sint, quomodo propter ea
contingat propositiones esse veras, quia haec quae nil omni no sunt vel esse possunt, quomodo did
causa possunt? Sed propter patratum furtum homo suspcnditur, quod tarnen furtum iam nil est, et
moritur homo quid non comedit, et damnatur quia non bene agit. Non comedere tamen vel non
bene agere non sunt essentiac aliquac.'
44. I have taken this term from Jolivet 1069, p. 355.
45. See Jolivet 1969, pp. 286-96.
46. Theologia Christiana (PL, vol. 178), 1253D-1254A: 'Sunt autem ab invicem diversae personae, id
est Pater, et Filius, et Spiritus Sanctus ad similitudinem eorum quae diftinitione diversa sunt seu
proprietate, eo videlicet quod quamvis eadem penitus essentia sit Deus Pater quae est Deus Filius,
seu Deus Spiritus Sanctus, aliud tamen proprium est Dei Patris, in eo scilicet quod Pater est, et
aliud Filii et aliud Spiritus Sancti.'
47. Ibid., 1255B: 'Nunquid enim si paternitas, quae inest Deo, alia essentia sit ab ipso Deo, verum est
Deum Patrem ex duobus consistent, hoc est ex Deo et paternitate, ipsumque esse totum ad haec
duo ex quibus consistit?"
156 The old logic
this is just what prevents the verb from being itself a name and signifying
some thing as does a name. Also, common nouns really have the same
signification as verbs, so they too are not to be conceived as signifying by
naming what they signify. And whole sentences, of course, do not signify
by naming either, because they contain the verb which imports the force of
saying. Where we do have apparent names of what verbs and sentences
signify, as with nominalisations, we can treat these as 'impersonal' and the
sentences in which they appear as 'subjectless'. And just as we find no
necessity either in ordinary discourse or in logic to employ names of what
verbs and sentences signify, so in talking of the objects of thought (i.e. what
our ideas are ideas of) we discover that there is no need to treat psycho-
logical discourse as containing names of things which are these objects. In
the end all uses of noun phrases apparently denoting these status and dicta
can be seen as non-denoting occurrences of nouns, or as parts of sentences
whose assertion is quite unnecessary to any legitimate inquiry. 48
Further we see that the 'necessity of entailment' which Abelard claims is
involved in all valid inference and in the 'laws of nature' on which science
relies, really concerns connections between these natures or status and the
dicta of whole propositions. If we were to treat the status and dicta as things,
they would be the eternal, necessary realities beloved by all Platonists,
which provide whatever intelligibility the world may have. Abelard is not
entirely repelled by such a vision. 49 He takes note of Priscian's view that the
world is constructed on the basis of forms in the mind of God, and these
forms are what we think of and what our words signify, albeit in a very
confused fashion. Certainly Abelard is quite willing to acknowledge that
we do not have correct conceptions of the very natures our ideas are of.50
But in the end he avoids any commitment to this sort of realism. It was left
to logicians at Melun, a school where Abelard had taught, to carry the
master's ideas in this Platonist direction. 51
Abelard's program of 'dereification' is directly connected with his effort
to unify grammar and dialectic in such a way that while the latter eluci-
dates at a deeper level much of what the former treats, grammar forces
dialectic out of a simplistic reliance on Aristotle's term-logic as it is handed
down by the logica vetus. Abelard had rethought the contents of the logica
vetus, and in his analyses he frequently moved far beyond anything that
48. See Tweedale 1976, pp. 2 7 3 - 8 .

49. See Jolivet 1969, p. 353, where Abelard is described as a sort of Platonist.
50. See Abelard 1919—27, p. 23.18-24.
51. See excerpts from the Ars Meliduna, in De Rijk 1962-7,11(1), pp. 264-390.
could be extracted from that limited heritage. Yet his style is often tenta-
tive, rarely claiming to say the last word and frequently inviting further
reflection. His work should have formed the basis of a highly original
development of the artes sermocinales and of philosophy in general in
Western Europe. But this was not to be. Partly because of Abelard's poor
reputation with the church authorities, but mostly because of the influx of
hitherto unavailable works by Aristotle and his Islamic commentators,
Abelard's work was to be largely ignored in favour of the more com-
prehensive and systematic philosophy these new texts provided.
IV
L O G I C IN T H E
HIGH M I D D L E A G E S : S E M A N T I C
THEORY
7
THE ORIGINS OF THE
T H E O R Y OF
THE P R O P E R T I E S OF TERMS
Language, thought, and reality

Beginning as early as the eleventh century, the relationship between
thought and language was a focal point of medieval thought. This does not
amount to saying that the basic nature of that relationship was being
studied; rather it was accepted without discussion, as it had been in
antiquity. Thought was considered to be linguistically constrained by its
very nature; thought and language were taken to be related both to each
other and to reality in their elements and their structure. In the final
analysis, language, thought, and reality were considered to be of the same
logical coherence. Language was taken to be not only an instrument of
thought, expression, and communication but also in itself an important
source of information regarding the nature of reality. In medieval thought,
logico-semantic and metaphysical points of view are, as a result of their
perceived interdependence, entirely interwoven.
The first medieval scholars to have a professional interest in language as
such were the grammarians. Their interest was focused on what we would
call logico-semantical and syntactical questions; this is especially true of the
School of Chartres as early as the 1030s. N o longer were words studied as
separate units quite apart from their linguistic context; rather it was that
context itself that attracted the most intense interest. I have labelled this
concentration of attention the 'contextual approach' (De Rijk 1967,
pp. 113-17; 123-5). The statement (propositio), not isolated words, was
taken to be the fundamental unit of meaning. The meaning of a word in its
actual use in a propositional context was considered so important, indeed,
as to lead to the introduction of a special terminology, centred around
terms (i.e., words as parts of actual propositions) and their properties.
The development of the doctrine of the properties of terms was in fact
nothing but the growing dominance ofjust one of the properties of terms,
suppositio, i.e., actually 'standing for'. On the other hand, throughout that
development, there remained some basic presuppositions which not only
162 The high middle ages: semantic theory
established the framework of the development but were, unfortunately, to

determine later developments as well. Among them the most obstinate was
the doctrine of significatio.
Signification and the contextual approach
In spite of the fact that the medieval logicians were aware that the actual
context in which a term occurs is of paramount importance to its actual
meaning, and that, accordingly, in investigating the meaning(s) of a term
they practiced the contextual approach, all their investigations kept on
being mortgaged by the ineradicable doctrine of significatio as upheld in
ancient and medieval philosophy. This doctrine may be best expressed this
way: a word's actual meaning (its meaning on a particular occasion of its
use) ultimately is, or can be reduced to, its fundamental 'significance'
(significatio), which as the word's natural property constitutes its essence or
form (essentia, forma), in virtue of which it is at the root of every actual
meaning ofthat word.
If we distinguish the contextual approach and the basic doctrine of
signification as the two focal points of the theory of the properties of terms,
the different aspects of the theory seem to be in close correspondence with
the preponderance either of the contextual approach or of the doctrine of
signification.
Medieval logicians regularly classified meaningful words into such as
have meaning in their own right (termini significativi or significantes) and
such as are meaningful only when joined to words of the first kind (termini
consignificativi or consignificantes). The former are also called categorematic
terms (termini categorematici or categoremata), the latter, syncategorematic
terms (termini syncategorematici or syncategoremata). The quasi-physical
character ascribed to those words by medieval logicians clearly appears
from a thirteenth-century treatise (called Syncategoreumata) by master
Nicholas (of Paris?), which opens as follows:
As the Philosopher remarks, the things that belong to art and reason are considered
in relation to and in imitation of things that belong to nature. Now as regards
natural things, we see that there are some that are naturally suited to accomplish
something without the assistance of anything else, but others that are not suited to
move unless they have been moved . . . The situation is similar as regards things
belonging to reason, especially as regards words, because some perform their
function —i.e. they signify —without the aid of anything else . . . , and words of that
sort are called categoremata — i.e. significant. There are others that are not significant
Origins of the theory of the properties of terms 163
in themselves, but in conjunction with the others, and those are called
syncategoremata.1
The syncategorematic words are dealt with elsewhere in this volume. Here
the categorematic words are under discussion, the only words that have
meaning in the strict sense (significatio).
Medieval semantics is focused on the meaning of the noun (nomen). The
function or property of the noun is commonly defined in the Middle Ages
(following Priscian) as signifying substance (substantia) together with
quality (qualitas).2 Substantia, according to the medieval interpretation, is
nothing but the individual thing, and the qualitas meant here is the uni-
versal nature in which the particular thing participates. The anonymous
author of the grammatical gloss Promisimus (not later than the last quarter
of the twelfth century) explicitly says that, seeing that all things have
existence in common, the older grammarians (antiqui) used 'substantia' as a
word with which to speak about things generally. So to be a substantia is to
have the property of being subsistent (proprietas subsistendi), to be that
which receives a thing's essential, specific nature (quodsuscipitformam). The
same author explains qualitas as the property which is designated by a
common or appellative noun such as 'man' or the one designated by a
proper noun such as 'John'. 3 An anonymous commentary on Priscian
dating from about the same time says that to signify the qualitas is to
designate to what class of things something belongs or of what nature a
thing is. 4 Indeed, the prevailing view in twelfth- and thirteenth-century
grammar and logic is that a substantia is an individual and qualitas its nature,
picking out the class of things to which the individual belongs. As early as
the second half of the twelfth century substantia is further identified with
the subject of discourse (id de quo sermo habetur). In the Quaestiones Victorinae
1. 'Ut dicit Philosophus [Physics II, 2; 194a 21-27], ea que sunt in arte et ratione sumuntur ad
proportionem et imitationem eorum que sunt in natura. In naturalibus vero ita videmus quod
sunt quedam que per naturam nata sunt in se aliquid agere sine alieno suffragio, alia vero sunt que
non sunt nata movere nisi mota, sicut homo a se motus et non ab alio protrahit litteras, calamus
vero non a se sed ab homine motus. Similiter se habet in rebus rationis, maxime in voribus, quod
quedam faciunt id ad quod sunt sine auxilio alterius, scilicet significant, quia omnis vox est ad
significandum, quoniam, ut dicit Aristotiles [De interpretatione 1:16a 3-4J, voces sunt notae earum
que sunt in anima passionum, idest significant intellectus, qui sunt signa rerum; ct ita voces
significant res; et tales voces dicuntur categoreumata, idest: significantes: alie sunt que per se non
significant sed in coniunctione ad alias; et tales dicuntur sincategoreumata' (Braakhuis 1979,
1.2-15).
2. Priscian 1855, II, 18, 55.6: 'proprium est nominis significare substantiam cum qualitate'.
3. MS Oxford, Bodleian, Laud Lat. 67, f. 4o ,k .
4. MS Vienna, V. P. L. 2486, f. 24vb: ' . . . dc quo genere rerum aliquid sit vel de qua manerie'.
(presumably written in the school of William of Champeaux) an interest-

ing note is found on the signification of a noun. The author paraphrases
Priscian's definition of the noun ('The noun signifies substance together
with quality'), that is, supposits (supponit) a thing together with the thing's
status, the latter term being in common use to designate a thing's universal
nature. So we find as early as the twelfth century 'supponere' (to supposit) as
an equivalent for 'significare substantiam', i.e. to signify the individual
thing. 5
Signification, univocation, and appellation

As we have seen, signification was taken by medieval logicians to be a
word's natural property, the formal constituent of every kind of meaning.
From the viewpoint of the interpreter of the word, it is the presentation of
some universal nature to the mind, to use William of Sherwood's defi-
nition. 6 A word's signification depends on its imposition (impositio), that is,
the word's original application. A single imposition implies a single signifi-
cation. Whenever a word is used two or more times with the same
signification (that is, signifying the same universal nature), the use is called
univocation and the word is called univocal. (A univocal word is con-
trasted with an equivocal word, one that has had more than one imposition
and so has different significations on different occasions of its use.) But
although a univocal word retains a single signification, it stands for dif-
ferent things when used in different propositions. That is, appellative
nouns such as 'man' or 'horse' used univocally may stand for different
things. This standing for ('suppositing for') is taken to be the result of
different nominationes or appellationes. Consequently, a fundamental dis-
tinction between signification and appellation arose as early as the first half
of the twelfth century. Indeed, when appellative nouns are used in a
proposition, they sometimes stand for the universal nature (as in 'Man is a
species'), and sometimes for themselves (as in 'Man is a noun'); more
importantly, their actual meanings are specified by the tenses of the verbs
of the propositions of which they are subject terms. In this context appel-
lation is taken to be the reference of the noun to some existent particular
thing or things, regardless of whether it or they exist in the present or in the
past or in the future.
When taken in its primary sense, however, the notion of appellation is
5. De Rijk 1962-7,11(1), pp. 261-2 and 242; 11(2), p. 739. 31 ff.

6. William of Sherwood 1937, p. 74. 16-17.
always connected with the present tense of the substantive verb 'est' ('is') or
some other word implicitly containing it. So we find appellation defined
by William of Sherwood as 'the present correct application of a term'. 7 But
an appellation may be 'ampliated' or 'restricted' within the proposition for
some reason, e.g., because of the past or future tense of the verb ofthat
proposition, or as a result of the use of such words as 'potest' ('can') which
ampliate the appellation to include merely possible individuals. The notion
of appellation certainly derives from the grammatical term 'nomen appel-
lativum' (appellative noun). The anonymous Fallacie Parvipontane says that
the appellative noun was invented in order to bring together all the things
denoted by it (its appellata) under one and the same name. 8 However,
which appellata are actually referred to in a proposition depends upon the
verb ofthat proposition.
So the first stage of the development of the theory of the properties of
terms is characterised by the central position of appellation, not sup-
position, as was to be the case in later stages. During this first period the
word 'suppositio' is found principally as an equivalent for subiectio ('putting
as a grammatical subject'). Likewise the word 'suppositum' does not yet
mean the thing supposited for (or referred to) but primarily means the
grammatical subject of the proposition, as may appear from such phrases as
'res destgnata per suppositum' (the thing designated by the suppositum).9
At this stage univocation is regularly defined as 'the altered appellation
of a name' (variata nominis appellatio) rather than as 'the altered sup-
position'. 10 Another striking fact is that ampliation and restriction are of
appellations, not of suppositions, as became standard in the fully developed
form of the theory of the properties of terms. Appellation seems to have
primacy even over signification. So in the two 'modern views' mentioned
by the anonymous author of the Ars Meliduna, a work on logic of the
utmost importance dating from about the middle of the twelfth century,
appellation turns out to play the central role. The first view holds that the
appellative noun 'appellates' (appellat) each thing comprehended under it
but signifies (significat) it only in an indeterminate way; thus the word
'man' signifies the species (universal nature) man indeterminately, not as
this or that man, and it may appellate a man who actually exists as well as
one who does not exist. The other view is closely connected with the status-
7. William of Sherwood 1937, p. 74. 26-8.

8. De Rijk 1962-67,1, p. 563. 27-8.
9. Fallacie Parvipontane in De Rijk 1962-67,1, p. 566. 36.
10. E.g., Tractatus de univocatione Monacensis in De Rijk 1962-7,11(2), p. 337. 4-7.
doctrine concerning universals, which was prominent as early as in the

days of Peter Abelard: all appellative nouns 'signify' either a common or a
proper status. Again, not a single word about supposition. 1 '
The introduction of supposition

The second stage of the development of the theory is characterised by the
fact that the notion of the significative use of the appellative noun was
supplemented without losing its pre-eminence. To catch the non-
significative uses of nouns, too, (such as are found in 'man is a noun' or
'man is a species') the appellation theory had to be extended into a more
general theory of the use of a noun as the subject term in a proposition.
Thus arose a general theory of the properties of terms in which all the uses
of terms in propositions were brought under the general viewpoint of
suppositio (or subiectio), i.e., the way in which a word is put as the subject of a
proposition, and, accordingly, as the subject of discourse (id de quosermofit).
Although at this stage the different kinds of supposition do not yet appear
under their later technical labels, the first beginnings of the technical
terminology can already be found.
The climax of the development

The third stage of the development shows the theory of the properties of
terms clearly focused on the notion of supposition, while the notions of the
other, already recognised properties of terms (appellatio, ampliatio, restrictio)
are subordinated to it. All the uses of a term in a proposition are covered by
the doctrine of supposition, and the different kinds of supposition have
received their own technical labels. Some authors of logical treatises (e.g.,
the anonymous author of the Introductiones Parisienses) attribute suppo-
sition to both the subject and the predicate term. 1 2
The contextual approach plays a most fundamental role in all three stages
of the development, sometimes to such an extent that it is no longer the
appellative noun but the 'terminus' (defined as a part of speech which may
be used as the subject term or the predicate term of a proposition) that has
become the special linguistic element in the focus of the theory.
Accordingly, it is commonly spoken of as a theory of the properties of terms
(rather than of appellative nouns or words).
ii. Ars Meliduna in De Rijk 1962-7,11(1), p. 537.

12. In De Rijk 1962-7,11(2), 372. 15-24.
The pre-eminence of reference over signification

From the preceding sketch of the development of the theory of the
properties of terms it will be apparent that both in the earliest stage, where
appellation was the prevailing notion, and in the later stages a term's actual
meaning was the focus of interest and reference or denotation was far more
important than the more abstract notion of signification. What is primarily
meant by a term is the concrete individual objects the term can be correctly
applied to; that the term may also be taken to mean what those things have
in common is of interest in a secondary way only.
I offer just two examples of this situation. First, the Tractatus Anagnini
holds that the signification of a noun is its propria confusio, that is, the wide
and often vague meaning proper to a word, by means of which the whole
range of individuals covered by that word is designated. 13 Secondly, as has
already been remarked, the Ars Meliduna mentions the view that an
appellative noun appellates concrete individual things, but presents its
significate only in an indeterminate way. So the species (universal nature)
man that is the significate of the word 'man* is taken to be presented by the
word in a way that involves no existential commitment (nullam recipiunt
relationem pertinentem ad essentiam).1*
No doubt the impulse to this development of the theory came from the
practice of dialectical disputation. It is not surprising at all that many a
treatise begins its exposition by stressing the basic importance of the theory
of meaning for logical dispute. The grammatical-logical subject of a
proposition (id de quo, that about which) is, at the same time, the subject
matter of discussion or disputation (id de quo sermofit, that which the talk is
about), and, furthermore, 'that for which speech is true' (idpro quo vera est
locutio). From this time onwards ' supponere pro' has the broad sense of'stand
for'.
The persistent importance of the doctrine of signification

What has been said above about the preponderance of the contextual
approach over the doctrine of signification is, however, only one side of the
development of the theory of the properties of terms. As a matter of fact,
throughout the whole of the Middle Ages the theory betrayed a radical
inconsistency, which can probably be best characterised as a persistent
13. In De Rijk 1962-7, 11(2), 274. 26-9.

14. De Rijk, 1962-7,11(1), p. 295.
r68 The high middle ages: semantic theory
hesitation of medieval logicians between the domains of connotation

(universals) and denotation (individuals). We have seen that they were
committed to the individual thing as that which is primarily meant by a
term; we shall now see how important a role was assigned to signification
by some authors, especially those who were somehow influenced by
Neoplatonic views.
William of Conches gives as his view of the meaning of substantive
nouns that they do not signify substantia and qualitas, as Priscian had
thought, 1S but only that which is intelligible, the universal nature
(intelligibile), not this or that existing thing (actuate). For it is the specific
function of the intellect to signify things as universal natures: 'We say that
it signifies the species and not the individual, for it signifies man in such
a way as not to signify any [particular man], and man understood in this
way is the species and not the individual.' Therefore he accepts all such
phrases as 'signifies substance', 'signifies quality', 'signifies that which is
intelligible', and 'signifies species' as if they meant the same. The
picking out of extra-mental individual things he assigns to 'nomination' or
'appellation'. 16
This aspect of the theory of the properties of terms may be clarified by
the history of one specific type of supposition, viz. natural supposition
(suppositio naturalis).
Signification and natural supposition

For Peter of Spain natural supposition is the acceptance of a common
substantival term for all those individuals that are of such a nature as to
participate in the universal form signified by it. So the term 'man' when it
is taken by itself (per se sumptus) naturally supposits for all men, those who
exist, those who have existed, and those who (possibly) will exist. 17
Accidental supposition, the complement of natural supposition, is the
acceptance of the same common term for only those individuals deter-
mined by what is adjoined to the term. So in the proposition 'man is' (homo
est) the term 'man' accidentally supposits for (denotes) all men who exist at
the present time (at the time of speaking), whereas in 'man has been' (homo
fuit) it accidentally supposits for all men who have existed in the past, and in
'man will be' (homo erit) for all men who will ever exist. 18
15. See n. 2 above.

16. See De Rijk 1962-7,11(1), pp. 223-4.
17. Peter of Spain 1972, p. 81. 2 - 5 .
18. Peter of Spain 1972, p. 81. 5-10.
But how is natural supposition related to signification? For Peter signifi-

cation is merely the conventional representation of a res by a word, the res
being either a universal nature or any individual partaking in this universal
nature. So for Peter the notion of signification covers both connotation (of
the universal nature) and denotation (of the individuals participating in this
universal nature). But the introduction of natural supposition seems to blur
the clear-cut distinction commonly made between signification and sup-
position: signification arises by imposing a vocal sound (vox) on some-
thing, while supposition is the acceptance of a term already significant as
standing for something or other. 19 Natural supposition, however, being
defined in contradistinction with accidental supposition, which is deter-
mined by a propositional context, seems to occur apart from such context,
since it is said to be the supposition of a term taken by itself.
Different interpretations have been given of Peter's natural supposi-
tion. 20 A careful exegesis of the texts involved seems to lead to the
conclusion that, for Peter of Spain, natural supposition is the natural
capacity of a significative word for supposition, whether within or without
a proposition. When a vocal sound is endowed with signification in such a
way as to become a term, it signifies a universal nature or essence (its
significate), and it acquires a natural capacity to stand for (supponerepro) all
actual and possible individuals partaking in this universal nature; it is said,
then, to have natural supposition. This natural capacity, which the term has
in virtue of its signification alone, may be limited in its function by another
word adjoined to the first one. This adjunct may be the predicate term of
the proposition of which the first term is the subject term; and in such a case
the propositional context is an essential condition of the resultant
accidental supposition. But the limitation, or restriction, can also come
from a non-propositional adjunct, such as an adjective (e.g., 'white man');
whenever in such cases the actual context is left out of account for a
moment and the term involved is taken by itself, the term has its whole,
unrestricted extension: all particular men, present, past, and future.
So Peter's natural supposition is really the denotative counterpart of
signification. Its resemblance to signification lies in the fact that in natural
supposition the capacity of standing for all particulars involved, a capacity
which the significative term has by its very nature, is completely exploited
- that is, its extension is taken exhaustively - by leaving aside the actual
19. Peter of Spain 1972, p. 80. 8 - n .

20. See De Rijk 1971, pp. 74-6.
context in which the term occurs. But natural supposition is unlike signifi-
cation in that there is a context, an actual linguistic framework, which is
left out of account for a moment.
Signification and 'habitual' (dispositional) supposition

William of Sherwood's 'habitual' (dispositional) or 'virtual' supposition
bears some resemblance to Peter's natural supposition. However, what is
striking in William's defintion of it is that he considers it to be a kind of
signification. In doing so he goes even further than Peter seems to do in
distinguishing between signification and supposition. It should be noticed,
further, that for William signification, as well as supposition, is a property
of a term. 21 As a matter of fact, William's habitual supposition is the
signification of subsistent things (forms), that is, those things or forms that
are suited to be underlying (hypokeimena: subiecta or supposita), i.e., entities
that serve as substrata for adjacent entities and are signified by substantival
terms. For William, too, the context required for supposition is not
necessarily propositional; but how shall we imagine a context as required
for habitual supposition, which is really a kind of signification? That
question leads us to an important side of Sherwood's semantics. O n the level
of signification there is no distinction among the species man, an individual
man, humanness, etc., no more than among running, the running,
runner(s), running as such, etc.: the significate of such terms is always the
same. It is the actual context in which the term occurs that divides
signification into substantival signification (which he calls habitual sup-
position) and adjectival or verbal signification (which he calls copulatio). It
is only after that division has been established that we get the further
division into the different kinds of supposition.
So Sherwood's signification seems to be a depth-structure, whereas
habitual supposition is found in the surface-structure where functions are
determinant not only of the way of suppositing, but, more fundamentally,
of the linguistic differentiation into substantival, adjectival, or verbal
forms. The functions realised in the surface-structure, including habitual
supposition, are all determined by some sort of context. William's habitual
supposition is not different from Peter's natural supposition in that respect.
The real dissimilarity occurs at a far more fundamental level, viz. in their
different views of signification. Unlike Peter and the majority of
thirteenth-century logicians, Sherwood (along with Roger Bacon) 22
2i. William of Sherwood 1937, p. 74.
22. See Braakhuis 1977.
identifies a term's significative character with its referring solely to actually

existing things. For Sherwood and Bacon signification is denotative in
character, and a term's extension can cover only presently existent
particulars.
So with Sherwood, too, this peculiar dispositional or habitual sup-
position, so closely related to signification, gives us some insight into what
signification meant to these thirteenth-century terminist logicians, and,
especially, how central the notion of signification continued to be in the
theory of supposition. 23
Copulation, appellation, ampliation, and restriction

As we have seen already, in considering Sherwood's position, the counter-
part of supposition in his theory is copulation. Copulation (or linkage)
arises from the fact that an adjective is linked with a substantive (e.g. in
'white man' or 'the man is white'). It concerns the significative function
not only of predicates (that is, all those terms that are not subject terms in a
proposition) including verbs and participles, but also of adjectives when
joined attributively to substantives. Copulation therefore is the kind of
signification modifiers have. Most authors do not discuss copulation as
extensively as Sherwood does. 24
The notions of appellation, ampliation, and restriction are of major
importance. In treatises dating from the thirteenth century onwards appel-
lation is taken as a counterpart of supposition. The divergencies of the
views regarding it run parallel with those concerning signification. The
common view holds that the difference between supposition and appel-
lation is that appellation concerns only things actually existing at the time
of the utterance, whereas supposition, like signification, may concern non-
existing things as well. 25 According to this view, then, a term has by itself
(de se) the power of referring to all the things covered by the term,
regardless of whether they are present, past, or future. For a term to denote
in fact only present things, special conditions are required which are laid
down in the well-known rule of appellation found already in the twelfth-
century treatises: an unrestricted common term, having sufficient appellata
and being the subject of a present-tense verb that has no ampliating force,
23. Another specimen of what is here meant by the preponderant role of signification as compared
with the denotation of individual things can be found in the controversy over whether a word
can lose its signification (ulrum vox possil cadere a sua significatione). See Braakhuis 1977, pp. 1338".
24. See, e.g., Peter of Spain 1972, p. 197. 4—14.
25. See, e.g., Peter of Spain 1972, p. 209. 4-10.
stands only for those things that do exist. O n this view, then, the range of
supposition is to be narrowed (restricted) in order to become the range of
appellation, referring to actual things only.
Restriction accordingly is defined as the contraction (coartatio) of a
common term from its wider supposition to a narrower one. Ampliation is
defined reciprocally as an extension of a common term from a narrower
supposition to a wider one. It is effected by such words as 'potest' (can),
which cause the terms to which they are adjoined to stand for things that
exist possibly but not actually.
William of Sherwood and Roger Bacon defend a view contrary to this.
They say that a term stands on its own (de se) for present things only, and
whenever it stands for other things, it will do so because of what is adjoined
to it, that is an ampliating verb. Sherwood provides an example: 'Thus,
when I say "a man is running" " m a n " stands on its own for present men
and is not drawn away from that supposition by the predicate. But if I say
"a man ran or can run" the supposition of " m a n " is actually drawn away to
men who do not exist.' 2 6 Thus, on this view the proper range of sup-
position basically coincides with that of appellation, with the result that
supposition must be ampliated to reach beyond appellation, and no possi-
bility is left for restriction to be effected by the verb (as is the case in the
former view).
In his Summule dialectices Roger Bacon clearly distinguishes between the
two views of appellation and is eager to defend the view that a term on its
own stands only for existing things and that only as a result of some special
predicates can a term stand for non-existing things. 27 The reason is that a
noun is given or imposed upon present things. It is not surprising that the
adherents of this second view, including those later logicians who were
influenced by them, do not pay much attention to the notions of ampli-
ation and restriction.
Conclusion
Concluding this sketch of the early development of the theory of the
properties of terms, it may be said that in the very beginning of the
26. See William of Sherwood 1937, p. 85. 15-24. Revised text (see Braakhuis 1977, p. 166): 'Vel
aliter, si proprie velimus loqui, dicamus quod terminus de sc supponit pro presentibus; et si
supponat pro aliis, hoc erit ratione sui adiuncti, scilicet verbi ampliandi vel verbi preteriti vel
futuri temporis. Et erit hec ampliatio non solum ratione significati nee consignificati, sed per
virtutem amborum. Unde cum dico "homo currit", supponit ly "homo" de se pro presentibus et
ab hac suppositione non trahitur per predicatum. Si autem dicam "homo cucurrit" vel "potest
currere", iam trahitur hec suppositio ad non existentes.'
27. See Roger Bacon 1940, pp. 277.28-283.19.
development some confusion is found regarding the relative status and

function of signification in this semantic theory. The sound basis of the
contextual approach seems to be undermined (more or less with various
authors) by the implicit presupposition of the natural priority of signifi-
cation. Even the most sagacious among medieval logicians, such as
William Ockham, were not able to trace that troublesome presupposition,
let alone to expose and to cut it out. They would have done a better job, if,
instead of rejecting such notions as natural or simple supposition, they had
abandoned their notion of signification itself. The most critical logicians of
the Middle Ages used a sharp knife, but amputated the wrong leg.
8
THE O X F O R D AND PARIS
T R A D I T I O N S IN LOGIC
The hypothesis of an Oxford-Paris split

Terminist logic grew to maturity in the period 1175-1250, a period that
was also crucially important in the development of the universities of Paris
and of Oxford. Scholars have recently focused their attention on diver-
gences in the early development of the logical and semantic theories that
constitute terminist logic, divergences suggesting that one cluster of doc-
trines is to be associated with Oxford, another with Paris. 1 In view of the
very marked differences between British and continental logic in the early
fourteenth century, it seems important to investigate whether such dif-
ferences can be traced backwards into the thirteenth and late twelfth
centuries. As I hope to show, such divergences do in fact exist even if they
are not so great as to make the traditions of Oxford and of Paris entirely
independent.
Although by the turn of the thirteenth century terminist logic was
acknowledged by all logicians as a common frame of reference, various
interpretations of important issues were still being put forward. If diver-
gences between the traditions of Oxford and Paris are to be established, the
evidence is likely to be found in the discussions concerning the various
properties of terms, such as supposition, appellation, ampliation, and
restriction.
The school of the Parvipontani

Twelfth-century logicians seem to have basically agreed in claiming that
an appellative (or common) name may vary its reference (appellatio)
according to changes in the tense of the main verb of the proposition; they
agreed further in describing this variation as either 'restriction' or 'ampli-
ation' of the reference or appellation, 2 and in providing rules associated
1. De Rijk 1972, pp. LXIX-LXXX; Kretzmann et al. 1975, pp. 563-5; Braakhuis 1977; Kretzmann
et al., 1978; Murdoch 1978b; Pinborg 1979; Ebbesen 1979·
2. Roger Bacon's later interpretation of this variation as a case of equivocation is also evidenced as
early as in the twelfth-century Summa Soph. Elench.; see De Rijk 1962-7, 1, p. 365.21-4: 'sunt
tamen alii qui dicunt quod hoc nomen "homo" habet plura significare, quia accipitur "homo" qui
The Oxford and Paris traditions in logic 175
with the three main tenses of the verb: present, past, and future. They did,
however, differ over, for instance, the question whether accidental (or
adjectival) names, too, may vary their appellation, 3 or whether a name
loses its signification rather than its appellation when every thing it denotes
4
has ceased to exist. The antagonists of the twelfth century belonged to
various (French) schools - those, for example, of Melun or of the Petit-
Pont - and their divergences are thus not yet evidence of any doctrinal
differences between the universities. 5 It has been argued, however, that the
early Oxford tradition of logic should be traced back to the Parisian school
of the Petit-Pont, the school of the Parvipontani,6 and so it will prove
helpful to say a few words about the most characteristic claims in semantic
theory developed within this school.
In the works which have been preserved, appellation (or supposition -
no clear distinction was yet drawn between these two terms) is defined
against the background of the doctrine of univocation. The Fallacie
Parvipontane offers a division of univocation that obviously underlies the
division of supposition which was subsequently introduced in the early
treatises from Oxford. 7 Univocation is defined as a 'variation in the
supposition of a name the signification of which remains the same', 8 a
variation which must therefore be effected in different contexts. This
definition is followed by a set of rules designed to exhibit cases in which the
range of reference of the terms is either narrowed (restriction) or extended
(ampliation). 9 In the Parvipontanean tradition 'restriction' and 'ampli
ation' are considered as relative and reciprocal terms, without either notion
being more emphasized than the other; but the roles of restriction and
ampliation became the focus of some significant divergences to be ob
served in Oxford and Paris thirteenth-century logic.
The early Oxford treatises

10
In the early Oxford treatises supposition is clearly distinguished from
appellation (the reference to none but present, existent individuals), but the
fuit homo vel qui est homo vel qui erit homo, undc secundum hoc erit in predicto paralogismo
sophisma secundum cquivocationem'.
3. Ars Meliduna (1170s. sec Hunt 1975, pp. 18-19, n. K), quoted by De Rijk 1962-7,11(1), p. 302.
4. Dc Rijk 1962-7,11(1), p. 316.
5. Dc Rijk 1962-7,11(1), pp. 288-9.
6. De Rijk 1976a, p. 35, 11. 47.
7. Dc Rijk 1962-7,1, pp. 562-71. especially pp. 562-3.
8. De Rijk 19Λ2-7, i, pp. 561-2.
9. De Rijk 1962-7,1, pp. 563-4.
10. Logica 'Cum sit nostra' in De Rijk 1962-7, 11(2), pp. 418-51, Logica 'Ut dicit' ibid. pp. 379-411,
Introductions Parisienses ibid. pp. 357-73.
general lines of the Parvipontanean doctrine are preserved. In the Cum sit
nostra (which probably originated in the late twelfth century but was
variously adapted during the whole thirteenth century) 11 restricted and
ampliated suppositions are significantly presented as subdivisions of uni-
vocal supposition. 12 In addition, a variety of univocal supposition is
acknowledged that introduces the divisions of metaphorical supposition
(gemination, metonymy, antonomasia, synecdoche) that became standard
in Oxford logic of the fourteenth century. ' 3 A striking difference between
the two groups of texts is that the Oxford treatises ascribe supposition to
subject and predicate terms alike, while the Fallacie Parvipontane provides
rules of appellation for subject terms alone. 14 Nevertheless, the Oxford
texts present restricted and ampliated suppositions in a set of rules that do
not differ significantly (as regards subject terms) from the Parvipontanean
rules of appellation. The Oxford rules specify the kinds of propositions in
which terms may have univocal supposition either for existent entities only
(by restriction through a present-tense verb) or for existent or non-existent
entities (by ampliation through a past- or future-tense verb), as a con-
sequence of the syntactical definition of supposition as 'the positioning of
the referent (res) of the subject under the predicate' by means of the verb. 1 5
The Oxford theory of appellation is significantly connected with the
doctrine and division of suppositions, ampliation being associated with
distributive confused supposition and restriction with merely confused
supposition - in a way reminiscent of the Parvipontanean texts. 16
One might sum up the early Oxford doctrine by saying that the Oxford
logicians consider supposition as a general property of terms - whether
subject or predicate — occurring in propositions; that they view all the
different kinds of supposition as varieties of univocal supposition; that they
11. De Rijk 1962-7, 11(1), pp. 4 4 2 - 6 .

12. De Rijk 1962-7. "(2). PP· 446.34-447-32.
13. De Rijk 1962-7, 11(2), pp. 447.33-448.3; R o g e r Bacon 1940, pp. 287.33-288.36. It should be
noted, however, that Buridan has a division quite different from the one used by the English
logicians O c k h a m , Burley, and Lavcnham.
14. Cum sit nostra in De Rijk 1962—7, 11(2), p. 451.11-13: 'Terminus non restrictus ad présentes
appositus verbo dc prcterito vel de futuro, vel vcrbo habenti vim ampliandi supponit confuse
tantum'; Introductions Parisienses in De Rijk ibid. p. 372.1 j - 2 4 : 'Terminus communis positus ex
parte subiecti eque supponit pro presentibus et pro p r e t e r i t i s . . . si sit positus ex parte prcdicati,
solummodo supponit pro preteritis . . . Similiter si supponat verbo de futuro (De Rijk: prcterito)
eque supponit pro presentibus et pro futuris; si autem prédicat, pro futuris solum.'
15. D e Rijk 1962—711(2), p. 446.29-33 :'Supponere est rem suam sub apposito p o n e r e . . . . Terminus
supponit quando p o n i t u r i n oratione.'
16. Compare Fallacie Parvipontane, De Rijk 1062-7,1, pp. 565.28,566.28, 569.7-9, and Cum sit nostra.
De Rijk 1962-711(2), p . 451.11-13.
maintain that no term is univocal by itself but only in context; and that
they acknowledge that in virtue of the tensed forms of propositions terms
may stand univocally either for present things (restriction) or for present or
past (or present or future) things (ampliation).
Contemporary developments on the Continent

The situation on the Continent during the same period is quite different.
Instead of the syntactically determined univocal supposition, Parisian
logicians in the early thirteenth century 1 7 consider 'natural supposition',
defined as the acceptance of a term taken by itself for all of its possible
referents, classifying as 'accidental' all sorts of supposition affected by
context. 18 This doctrine is not only concerned with the question of
whether or not terms supposit only in propositions; it is further connected
with a doctrine of appellation that does not consider restriction and
ampliation as reciprocal in the description of the variations of the truth-
value of a proposition over time. In fact, Parisian texts explain all problems
connected with this variation by restriction from the original omnitem-
poral supposition of the term, and relegate ampliation to the solution of
modal problems. Both the Fallacie Parvipontane and the early Oxford texts,
on the other hand, seem to have viewed the influence of past- and future-
tense verbs as quite similar to that of ampliating verbs such as 'can' — i.e.,
they seem to have assimilated tense and modality. 19
TTie Paris tradition in Peter of Spain

During the Summulist period (1230-45), the contrast between Oxford and
Continental doctrines increased. The Parisian doctrine essentially rep-
resented by Peter of Spain's Tractatus (written probably during the 1230s)
definitely sets the notion of natural supposition in the forefront, 20 while
that of appellation is relegated to the background. Peter focuses on sup-
position and restriction, treating appellation as no more than the particular
kind of restricted supposition effected by a present-tense verb. All the rules
of appellation are presented as rules of restriction, basically designed to
17. Summule antiquorum in De Rijk 1968, pp. 9-24, Introductiones antique, ibid. pp. 24-33. For some
parallel doctrines in the Dialectica Monacensis see De Rijk 1962-7, 11(2), pp. 616.31-617.8;
619.6-7; 623.24-7; 626.5-8.
18. De Rijk 1968, p. 9, ibid. p. 30. The Introductiones antique do not provide rules of appellation.
19. In the later summulist period William of Sherwood seems to have focused on the syntactical
resemblance between tensed and modal propositions by means of the doctrine of the com-
pounded and divided senses of propositions (Introductiones, p. 84.14-30; cf. p. 90.13-31).
20. Peter of Spain 1972, p. 8 i . i - i o .
accommodate natural supposition to the actual use of words in prop-

ositions with differently tensed forms in the predicate. 21 Accordingly,
ampliation becomes a notion of distinctly secondary importance, quickly
disposed of in a separate, very short treatise in Peter's Tractatus.22 Peter's
rules of restriction are the same as those of the early Parisian textbooks.
The Oxford tradition in William of Sherwood

At the same time, in Oxford, 2 3 William of Sherwood was developing a
doctrine of supposition and appellation founded on a syntactical definition
of supposition as 'the ordering of some thought under some other
thought'. 2 4 Sherwood provides two different accounts of the doctrine of
appellation: one based on restriction and one based on ampliation, and
he calls the latter the 'proper' interpretation. 25 The 'ampliative' interpreta-
tion is based on the two notions of dispositional supposition (suppositio
habitualis, defined as the 'signification of something as subsisting') 26 and
dispositional copulation (copulatio habitualis, defined as the 'signification of
something as adjoining') 27 that designate the semantic nature of words that
may actually be used in propositions as subject or as predicate terms
respectively. Against that background, Sherwood's basic claim is that the
appellation of substantive terms is entailed by their dispositional sup-
position. 28 The effect of this is that a substantive term occurring as the
subject of a present-tense verb by itself (de se) supposits and appellates for
present things only, since 'signification of something as subsisting' entails
'standing for present things'. When a substantive word occurs as a predi-
cate term it appellates only insofar as the significate of the predicate is
21. Peter of Spain 1972, pp. 202—4.

22. Peterof Spain 1972, pp. 194-6.
23. There is unequivocal documentary evidence that Sherwood was a master at Oxford; there is a
now discounted tradition and no comparable evidence that Sherwood was at Paris also. See
Kretzmann 1966, pp. 8-12, and De Rijk 1976, pp. 31-42.
24. Sherwood 1937, p. 7417-18.
25. The restrictive doctrine is found on pp. 82.20—85.15 of Sherwood 1937, the ampliative on
p. 85.15-31. Though it is difficult to unravel the thought in the whole passage, it may be assumed
that from pp. 82.23 to 85.15 Sherwood presents a very basic and simple doctrine which he
supplements with more complicated considerations on p. 84.14-30. This last passage is no doubt a
restrictive interpretation of the set of rules provided by Cum sit nostra. Compare especially Dc
Rijk 1962-7, 11(2), p. 451.II —13, or p. 372.15—24, with Sherwood 1937, p. 84.27-30: 'Et sic
intelligcndum est de verbo de futuro scilicet quod terminus communis ex parte ante supponit pro
presentibus vel futuris per compositionem et divisionem, ex parte autem post solum pro futuris.*
See also Braakhuis 1977, pp. 114fr.
26. William of Sherwood 1937, p. 74.22-3.
27. Ibid., p. 74.24-5.
28. See Maierù 1972, p. 91 and Braakhuis 1977, especially p. 130 n. 57; Braakhuis' interpretation is
opposed both to Maierù's and to mine.
actually related to the subject's significate through the subsistent thing(s).

Adjectives obviously appellate only as predicates. O n the one hand, having
dispositional copulation they signify an accidental form which intrinsically
requires the existence of a subject; on the other hand, being predicated,
their form is extrinsically taken in a syntactical relationship with that of the
subject. Hence adjectives cannot appellate without having actual copu-
lation, defined as the 'ordering of some thought above another thought'. 2 9
In presenting this theory, Sherwood undoubtedly offers a more sophisti-
cated interpretation of appellation than that provided by the earlier
Oxford treatises. But I think he significantly agrees with Cum sit nostra in
developing the theory of the entailment of appellation by dispositional
supposition, insofar as this particular claim seems to be primarily meant to
preserve both the old syntactical definition of supposition (in cases where
the supposition of a term has a more extended range than its appellation)
and the new characterisation of signification as the persisting factor that
underlies appellation (in cases where the range of supposition and appel-
lation coincides).
But even in his 'improper' account of appellation, based on restriction,
Sherwood's views are much closer to the doctrine found in Cum sit nostra
than to any text associated with Paris. For instance, Sherwood's presen-
tation of the 'improper' account is in substantial agreement with Cum sit
nostra on the question of the supposition of predicate terms in propositions
about past and future events (except for Sherwood's concern with com-
pounded and divided senses of propositions, an innovation that partially
develops from his treatment of modalities). In both doctrines the temporal
reference of the predicate is the same as that indicated by the verb, whereas
the Parisian texts apply the same rules to subject and predicate terms, thus
allowing predicate terms after a past-tense verb to stand for present or past
things, rather than for past things only. 3 0
It is surprising that Sherwood's account based on restriction may be said
to be closer to the Oxford tradition in which restriction and ampliation
were considered as reciprocal to each other, and less close to the Paris
tradition which shows the same emphasis on restriction as Sherwood's
improper account, But perhaps the approach based on restriction had been
29. William of Sherwood 1937, p. 7418.

30. Summule Antiquorum in De Rijk 1968, p. 15; Peter of Spain 1972, pp. 203-4, especially
p. 203.24-7: 'Ut cum dicitur "homo fuit animal", iste terminus "homo" supponit pro hiis qui
sunt vol fuerunt homines, si illi qui sunt homines fucrunt in prcterito, et "animal" pro hiis qui sunt
vel fucrunt ammalia.'
filtered through the Oxford tradition before finding the form presented by
Sherwood. The 'proper' account may then be seen as an attempt to re-
instate the genuine Oxford theory of univocally ampliated and restricted
supposition and re-establish it on a more solid semantic basis. But even if
this guess is incorrect, it is safe to say that before the 1250s the Oxford and
Paris traditions of logic followed divergent lines of development, their
characteristic disagreements being concerned with the very fundamentals
of terminist logic, that is: the syntactical definition of supposition, the role
and significance of restriction and ampliation, the definition of appellation
as an instance of restricted supposition (Paris) or as a basic feature of terms
having dispositional supposition (Oxford). From about 1250 onwards those
divergences changed their character to a certain degree.
Opposition between the two traditions in Roger Bacon

In his Oxford treatise Summule dialectices (ca. 1250) Roger Bacon mentions
two opposed views on the nature of appellation:
But there are two theories of appellation, since some people say that a term
appellates by itself present, past, and future appellata, and that it is common to
beings and non-beings. Other people say that a term is only the name of present
things and that nothing is common to being and non-being, or to past, present,
and future, in keeping with what Aristotle says in the first book of the
Metaphysics.31
The second opinion, which he favours, sounds very much like Sherwood's
'proper' account. The first one, which he rejects, is accompanied by a set of
rules that are closer to those in Cum sit nostra or to Sherwood's 'improper'
account of appellation than to Peter of Spain's. 32
That Bacon was digging in the Oxford tradition of logic is well attested
by his borrowing from Cum sit nostra the divisions of metaphorical sup-
position (which are not found in any other thirteenth-century textbook of
logic) and in his upholding a strictly syntactical definition of supposition. 33
When he calls the first account of appellation, based on restriction, 'the
Common Doctrine' it is accordingly tempting to think of Cum sit nostra
and Sherwood's 'improper' account. But there are special features in
31. Roger Bacon 1940, p. 277.28-34: 'duplex tamen est scntencia dc appellacionibus, quia quidam
dicunt quod terminus appellat de se appellata presencia, preterita et futura, et est communis
entibus et non-entibus. Alii dicunt quod terminus est solum nomen presencium et nichil est
commune enti et non enti, sive preterito, present! et futuro, secundum quod dicit Aristoteles in
primo Metaphysice.'
32. Ibid., pp. 277.35-278.13 and 279.8-15 (for the predicate terms).
33. Ibid., p. 268.33-4: 'suppositio non est proprietas nisi termini actualiter ordinati in oratione'.
Bacon's presentation of the C o m m o n Doctrine that do not tally with such

an assumption. As a matter of fact, Bacon not only discusses such familiar
notions as that of verbal restriction in the way Sherwood had done, but he
also discusses doctrines that are not found in Sherwood, such as the unusual
notion of the appellation of a term taken by itself for all its possible
referents. Moreover, the various arguments adduced in Bacon's text both
against the Common Doctrine and in support of his own doctrine are quite
different from those that are found in the usual discussions concerning
supposition and appellation before the 1250s.34 In this respect the use of the
word 'appellare' deserves special consideration. Since Bacon's most basic
claims are that a term cannot be common to being and non-being without
ampliation, and that when it is ampliated it has equivocal appellation, 35 he
is not only criticising the doctrine of univocal supposition as found for
instance in the Cum sit nostra, but he also attacks the concept of'appellation
for all possible referents', which is associated with the Parisian notion of
'natural supposition'. In doing so, he is engaged in 'modern' controversies.
In his last work, the Compendium studii theologiae (1292), dealing with the
two much-debated questions whether a word can signify anything that is
univocally common to being and non-being, and whether it can lose its
meaning, 36 Bacon tells us that as early as the 1250s some people in Oxford
began to answer the former question affirmatively, including Richard
Rufus of Cornwall. 3 7 This suggests that the C o m m o n Doctrine as de-
scribed in the Summule is already influenced by these new tendencies,
which first developed within the Paris tradition.
In his later works, the De signis (1267) and the Compendium studii
theologiae (1292), Bacon restates the whole doctrine of supposition, ex-
plicitly discarding univocal supposition and re-introducing the twelfth-
34. Arguments (in Roger Bacon 1940) supporting the restrictive doctrine of appellation (Sentencia
communis): (1) 285.5-9 (Nomen significat sine tempore), (2) 283.20-283.37 (Terminus in ne-
gativa stat pro ente et non ente), (3) 284.1-5 (Ad propositionem de predicate privato vel infinito
sequitur propositio de predicato negato et non convertitur), (4) 284.6-13 (Propositiones in-
definitae possunt esse simul verae). Bacon's reply: ad(i) 283.9-19; general determination of (2),
(3), (4) 284.15-24; ad(2) 184.25-37; ad(3) 285.1-286.16; ad(4) 286.17-287.10. Arguments sup-
porting Bacon's own ampliative doctrine of appellation: (a) 279.19-279.28 (Homo mortuus-
homo vivus); (b) 279.29-34 (Positio-privatio); (c) 279.34—280.1 (In omni nomine intelligitur
ens), 280.2-9: Objection (ens nomen, ens participium) and its solution; (d) 280.10-15 (Nature of
ampliation), 280.16—281.14 Determination, general exposition of the ampliative doctrine of
appellation including the discussion of the Ps.-Boethian: 'Talia sunt subiecta qualia permiserunt
predicata' and an important passage on the Determinatio equivocationis pp. 281.5—283.4.
35. Roger Bacon 1940, pp. 275.5-10,287.5-10, 332-28-334-I9-
36. Roger Bacon 1911, pp. 52-64.
37. Ibid, pp. 52-3.
l82 The high middle ages: semantic theory
century notions of equivocation and transumption, as if he wanted to make

a new start.38 In order to understand what kind of doctrines he was facing
in 1250, it may help to examine the way in which he presented the
Common Doctrine more than forty years later. In the Compendium the
precise wording is that 'a name may signify something that is univocally
common to being and non-being, or to present, past, and future'.39 Thus I
infer that already in the Summule he is thinking of a doctrine that holds that
a term may appellate or supposit for present, past, and future things, insofar
as it signifies something common to being and non-being. Now, whoever
may have been the actual proponents of the Common Doctrine, those
words do not tally with the doctrine of natural supposition as formulated
by the early thirteenth-century Parisian terminist logicians. The use of
'appellare' instead of 'supponere pro' is not found in Peter of Spain's
Tractatus, where appellation is introduced only as a kind of restricted
supposition. Moreover, there is no Paris logician of the period before 1250
who maintains that positive terms like 'man' signify something that is
common to being and non-being: Peter of Spain clearly states that sub-
stantive terms like 'man' are not common to the existent and the non-
existent.40 Hence, the Common Doctrine was probably influenced by new
theories that are first found in texts from about 1250 onwards. A text
representative of this new approach, accepting the kind of signification
attacked by Bacon, is the Lectura Tractatuum, one of the earliest com-
mentaries on Peter of Spain, by William Arnaud, master of arts at
Toulouse.41
From this I infer that from about 1250 onwards the dispute which took
place in Oxford concerning the nature of supposition and appellation had
to some degree shifted ground. It can no longer be described alone in terms
of two conflicting interpretations of univocal supposition; new doctrines
made themselves felt, doctrines that derive from a new approach to the
problem of meaning, connected with Aristotle's more sophisticated dis-
cussion on equivocation, univocation, and analogy as found in his
Metaphysics. This new approach may safely be described as Parisian, and its
38. See Roger Bacon 1978, III.3, §81-8; pp. 109-10 (theory of equivocal and metaphorical sup-
position); III.4 §89—99; pp. 110-15 (ampliation and restriction); and IV.I, §134—42; pp. 125-7
(exposition of the main arguments supporting Bacon's doctrine).
39. Roger Bacon 1911, p. 53.
40. Peter of Spain 1972, p. 207.9fr.
41. De Rijk 1969b, pp. 120—62. For example p. 146: 'Sic igitur patet quod quilibet terminus
communis, ut "homo", subponit de natura in subpositione pro forma communi que salvatur in
suis subpositis, sive ilia subposita sunt actu existentia sive non, solum quod in significando existât'
(italics added).
principal opponent at Oxford was the former Parisian master and regent of
arts, Roger Bacon.
The influence ofmodism in the Paris tradition

In the second half of the thirteenth century, logic at Paris was dominated
by the 'Modist' approach. 42 The basic claims of Modist logicians - most of
which were already to be found in William Arnaud's work 4 3 - are these:
the new division of logic according to the three operations of the soul taken
over from Aristotle's De anima, the reformulation of the terminists' 'na-
tural' and 'accidental' suppositions as respectively 'supposition by virtue of
discourse' (de virtute sermonis or locutionis) and 'supposition for truth's sake*,
the gradual accentuation of the problem of empty classes, the generali-
sation of the question whether terms may lose their signification, the
elaboration of a doctrine of univocation firmly rooted in ontology and
epistemology, and the logical use of the so-called 'modes of intellection,
signification, and being' (modi intelligendi, significandi, essendi).**
A novel theory of univocal signification and supposition

The problem of univocation seems to have evoked a number of con-
troversies between Modist logicians of the late thirteenth century such as
Siger of Brabant and Peter of Auvergne. 4 5 But the prevailing opinion was
that a term, insofar as it signified something univocally common to being
and non-being, could supposit (or appellate) for present, past, and future
things by virtue of discourse - i.e., in any proposition as well as taken by
itself (outside of any context). Against such a background the notions of
restriction and ampliation no longer played a prominent role. Once subject
terms of propositions were said to supposit by virtue of discourse for all
their possible referents, their standing for only present appellata or supposita
was explained by referring not to the requirement of the verb but to the
verification-procedure for the proposition. Thus, in the second half of the
thirteenth century some Parisian masters begin to discard the notion of
42. Pinborg 1975a, esp. p. 69, n. 105; pp. 70-1; p. 70, n. n o .

43. De Rijk 1969b, pp. 143-6. Some of the characteristics of'Modistic' logic are prefigured in the
earlier Parisian terminist tradition. One might think, for instance, of Peter of Spain's concept of
'acceptance' (as the definiens of 'supposition') or of his distinction between supponere and redden
locutionem veram (Peter of Spain 1972. P- 82; cf. esp. Robert Anglicus' Modist presentation of the
supposition of subject-terms indefinite sumpti in Dc Rijk 1969a, pp. 55-6).
44* Jolivet 1969.
45. Siger of Brabant 1974, pp. 44-52 (esp. p. 49.4-11) and pp. 53-9 (esp. p. 56.15-24, where a
'Baconian' doctrine of appellation is discussed); see also Siger of Brabant 1948, pp. 225-7. For
Peter of Auvergne sec Ebbesen & Pinborg 1970 and Pinborg 1973a; 1975a.
restriction, at least insofar as it is executed by predicate terms (leaving only

to modifiers the possibility of restricting the reference of terms). 46 In doing
so, they use a new terminology that shows the point of this new doctrine of
reference: they prefer to speak of the various acceptations (acceptiones) of a
word rather than of its various 'suppositions' or 'appellations', and they
focus on the various causes of truth for propositions (causae veritatis), since
these causes determine the various acceptations of the subject terms. For
example, the two propositions 'Man is an animal' and 'Man is a species'
have different verification - procedures or different 'causes of truth', which
in turn influence the acceptation of the term 'man' in the two propositions.
Of course, those different acceptations of the subject terms correspond to
the basic varieties of suppositions, but against a differently interpreted
background.
One of the characteristic features of logic at Paris from 1250 to 1300 is
that this new doctrine of univocal signification and reference embraces the
semantic question of whether or not the different contextual acceptations
(the former 'accidental suppositions') constitute equivocation. This use of
the concepts of equivocation and univocation is different from that of early
Oxford terminism; it is linked with the discussions on equivocation,
univocation, and analogy that derive from reflexions on Aristotle's theory
of being as found in his Metaphysics and developed by the Arabic
commentators.
Roger Bacon's later views on supposition and equivocation

At the same time in Oxford, Roger Bacon was deepening his criticism of
the Common Doctrine. In the De signis he is obviously grappling with the
Parisian question of whether or not the various acceptations of a term in
different propositions are equivocal. And his answer - contrary to the usual
Parisian o n e - is clearly affirmative. Furthermore, he states that all divisions
of supposition, not only those of restricted and ampliated suppositions,
reflect various cases of equivocation. In doing so, he generalises to all kinds
of supposition the doctrine he had outlined in 1250 solely for the variations
brought about by the change of tense in the verb of the proposition. He still
46. Radulphus Brito, Anal, priora I. qu. 46; in Pinborg 1976a, pp. 272-5.
47. For a general survey of Bacon's argumentation throughout the late thirteenth century, compare
(1) Bacon 1940, 283.5-19; Bacon 1978 §139; Bacon 1911, p. 57.11-19; (3) 1940, 285.1-286.16;
1978,§141; 1911,pp. 57.19-58.17; (4) 1940,284.6-14 and 286.17-287.10; I9ii,pp. 58fr Further:
(a) 1940,279.20-8; 1978, §140. (The numeration of the arguments is that of note 34 above.) For
the determinate equivocationis compare 1940, 281.34-282.22; and 1978, §93; for the Ps.-Boethian
adage see 1940, 281.15-283.4; and 1978, §§95 & 99.
lays particular emphasis on the problems of time and tense, but his
argumentation is from now on unmistakably linked with Parisian claims
and concerns. The fact that he still speaks of supposition, restriction, and
ampliation - against the Parisian fashion - indicates that his Oxford
contemporaries are probably remaining closer to terminist logic than the
Parisian masters. But his treatment of the material is nevertheless re-
juvenated if compared to his earlier Summule dialectices. His basic claims are
that a name does not name present, past, and future things except equivo-
cally, since nothing is common to being and non-being (in opposition to
the prevalent premodistic and modistic doctrine in Paris); all the divisions
of supposition must be referred to a specific sort of equivocation (contrary
to Oxford's doctrine of univocal supposition and Paris' doctrine of the
various acceptations). In short, terms are imposed on existing things only,
they signify only what is common to beings, and they cannot stand for other
things except equivocally, ampliation being an instance of equivocal sup-
position. Significantly, all the main arguments of the Summule are adapted
for use in the Designis, but recast in a Parisian mould. But it is striking that
the role he now assigns to restriction and ampliation remains Oxonian in
character: ampliation is associated with the problem of tense and modality,
while restriction is confined to what was earlier described as the univo-
cation of simple and personal suppositions.
Opposition and influence between Paris and Oxford

How do these facts tally with the hypothesis of divergent traditions of
logical doctrine at Paris and Oxford universities? Since what Bacon de-
scribes as the Common Doctrine at Oxford is mainly known from Parisian
sources, and since even some of Bacon's own theses are also found in a
Parisian scholar such as Radulphus Brito, it is obviously not possible to talk
about two independent traditions. The very persistence of Bacon's attacks
throughout the second half of the thirteenth century seems to indicate a
steady influence of Parisian doctrines at Oxford. At present there is not
much textual evidence of such an influence, but it has been argued convin-
cingly that the various theses considered as 'foolish' by Bacon in his
Compendium could not lead one to think only of Richard of Cornwall as a
promoter of the Parisian 'invasion'. 48 Further evidence of the Parisian
influence might be drawn from the anonymous commentary on
Sherwood's treatise on suppositions preserved in the Worcester manu-
48. Ebbesen and Pinborg 1970, p. 43.

script Cath. Q 13 (f. 59 vb -62 va ). As a matter of fact, this work, probably to

be dated about 1270 49 uses the doctrine of modes of conceiving and of
signifying in redefining Sherwood's four properties of terms, and it restates
Sherwood's claims in typical Parisian style. But most obvious, it mentions
a solution to the problem of the persistence of signification after the thing
signified has ceased to exist, a solution that had been eagerly discussed in
Parisian circles since William Arnauld. This solution - significantly op-
posed to Bacon's - is ascribed to Geoffrey of Hasphall, a prominent figure
at Oxford around 1270, but a master and regent of arts at Paris before
1265. 50 Thus, whoever was the anonymous commentator on Sherwood, I
think that he, together with Geoffrey of Hasphall, was among the con-
temporary Oxford logicians criticised by Bacon. In this respect Bacon's
praise of Sherwood in 1267 in comparison with Albert the Great (whose
doctrine of appellation is typically Parisian) 51 and his mentioning Sher-
wood as an antiquus in 1271 could take on a particular significance, 52 as if
Bacon wanted to draw a line between the old Oxford way of doing logic
and the new one, markedly influenced by contemporary Parisian masters,
that had begun to 'invade* Oxford after 1250. In any case, when in 1292
Bacon returns for the last time to the problem of appellation, he no longer
uses the terminists' terminology of supposition. This does not necessarily
mean that at that time terminism had been superseded by modistic logic in
Oxford — as a matter of fact the premodistic and modistic theory of
meaning never attained the dominant position in Oxford which it held in
Paris, and terminism was never entirely neglected 53 - but it might mean
that Bacon could not find anything in later Oxford terminism that would
help avoid the Parisian errors. It seems that Bacon no longer trusted the
terminist approach, but thought that the opposition between Oxford
terminism (univocal supposition) and Parisian modistic logic (doctrine of
univocal meaning) was actually too imprecise to prevent what Bacon
49. See Pinborg 1979, p. 26.

50. On Hasphall see Bazân, in Siger of Brabant 1974, p. 38. The text is found on f. 62" of the MS
quoted: 'Alia opinio est magistri G. Haspale et est satis bona. Ponit quod diversimode ad
representandum significatum suum imponitur terminus communis et terminus discretus,
quoniam terminus communis imponitur prêter omnem differcntiam temporis, terminus dis-
cretus imponitur ad tempus. Et quia in tcrmino discrcto idem est suppositum et significatum,
corrupto supposito corrumpitur et significatum. Tamen suppositum et significatum in tcrmino
discreto sunt diversa secundum rationem quia significatum dicitur unde intellectui representatur,
suppositum unde actui substat.'
51. Albert the Great 18yO-9a, I. 10, p. 474 a-b.
52. Kretzmann 1966, pp. 5-7.
53. For some evidence of the continuation of terminist logic in England, see Lewry (forthcoming);
Pinborg 1979; Ebbesen 1979.
considered misunderstandings and misleading claims. He preferred to

attempt a new start, based on his doctrine of signs and of equivocation.
Conclusion
I hope to have shown that there is evidence of substantial divergences
between Oxford and Paris logic during the whole of the thirteenth cen-
tury. Of course, these divergences changed character during that period,
especially when, after about 1250, logicians shifted their attention from
explaining the variations of the truth-value of a proposition over time to
justifying predication regarding non-existent individuals and empty
classes. This does not mean, however, that the two traditions are uniform:
each tradition experienced strong internal disputes. After 1250 at the latest,
a strong Parisian influence was felt in Oxford, associated with premodistic
and modistic semantic theories. But Oxford, perhaps partly due to the zeal
of Roger Bacon, never succumbed entirely to the Parisian doctrine.
9
THE SEMANTICS OF TERMS
Two senses of'term'

Medieval philosophers and logicians used the word 'term' (terminus) in
several senses, two of which are especially pertinent to this discussion.
Strictly speaking, a term is what is subjected to the predicate or predicated
of the subject in an ordinary categorical proposition — the subject term or
the predicate term, the two ends (termini) of the proposition. In this sense
whole phrases may be terms, but only certain sorts of words — nouns,
adjectives, and verbs - can serve by themselves as terms. Less strictly, and in
the later Middle Ages more prevalently, a term is any word at all, regardless
of propositional context. In this discussion 'term' will be used in the less
strict sense unless otherwise noted.
Signification as a psychological and causal property of terms

There are two basic properties for the medieval semantics of terms: signifi-
cation and supposition. Signification is a psychologico-causal property of
terms - a fact responsible for many disagreements and tensions in medieval
semantics. The main source for the notion of signification was Boethius'
translation of De interpretatione 3, i 6 b 19: '[Verbs] spoken in isolation are
names and signify something. For he who speaks [them] establishes an
understanding and he who hears [them] rests'. 1 Hence 'to signify' some-
thing was 'to establish an understanding' of it. 2 The psychological over-
tones of 'to signify' are similar to those of the modern 'to mean'.
Nevertheless, signification is not meaning. A term signifies that of which it
makes a person think, so that, unlike meaning, signification is a species of
the causal relation. Some authors explicitly drew the conclusion that the
relation of signification is transitive. Thus Lambert of Auxerre argued 'For
just as it is said that whatever is the cause of a cause is a cause of the caused, so
1. 'Ipsa quidem secundum se dicta nomina sunt et significant aliquid. Constituit enim qui dicit
intellectum et qui audit quiescit.' Boethius 1877, p. 5.5-7.
2. See De Rijk 1962—7, I, Ch. 4. For the role of Augustine and for further texts, see Spade 1975c,
pp. 214-17.
The semantics of terms 189
it can be said in its own way that whatever is the sign of a sign is a sign of the
significate.'3
Signification in written, spoken, and mental terms
Taking their cue from De interpretatione 1, i6 a 3-8 and from Augustine's
De trinitate, XV, 10-11, most logicians held that there are three kinds of
terms: written, spoken, and mental (or conceptual). Concepts or mental
terms are the most basic; they signify 'naturally'. Spoken terms signify only
derivatively, by a conventional (ad placitum) correlation with concepts;
written terms are related to spoken terms in the same way.4
On the authority of Boethius, many authors held that these conven-
tional correlations are signification relations. Hence written terms directly
or immediately signify spoken ones, which in turn directly signify con-
cepts. Only by the transitivity of signification - ultimately by means of the
signification of concepts - do written or spoken terms signify anything
further.5 Hence the terminology of 'immediate' and 'ultimate' signifi-
cation in some authors.6
This theory fits the view that language is for social communication, to
make others understand our thoughts. But it conflicts with the view that
language, whether spoken or written, is thoroughly conventional, so that
we can 'impose' terms to signify immediately whatever we want to discuss,
not just our own thoughts. Some logicians, therefore, rejected the view
that the conventional correlations among the three kinds of terms are
signification relations; Ockham, for instance, called them relations of
'subordination'.7
3. 'Sicut enim dicitur quod quidquid est causa cause est causa causale, sic potest dici suo modo quod
quidquid est signum signi est signum significati.' (Lambert of Auxerre 1971, pp. 205-6.) See also
Duns Scotus 1639, In I Perihermeneias quaestiones, q. 2, Vol. 1, pp. 186-9; ar| d Walter Burley in
Brown 1973, p. 55 par. 1.15 and in Brown 1974, P· 208 par. 1.3. (Burley is not speaking for
himself in these places.)
4. Peter of Ailly is the only author I know to have held that written language is not inferior to
spoken language. He says this explicitly only for sentences, but seems to have held it for terms too.
Peter of Ailly 1980, par. 93: 'Hence the spoken sentence and the written [sentence] are sub-
ordinated to the mental one. But it is not necessary that the spoken [sentence] and the written one
be subordinated among themselves, as many put it.' ('Unde propositio vocalis et scripta sub-
ordinantur mcntali. Sed non oportet quod vocalis et scripta subordinentur sibi invicem inter se
sicut multi ponunt.')
5. Boethius 1877, p. 40.15-22, and Boethius 1880, p. 25.15-p. 29.29; Lambert of Auxerre 1971,
pp. 20S-6; Aquinas In Peri Hermeneias, 1, lect. 2, §5; Aquinas Summa theologiae, I, q. 13, a. 1, resp.;
Aquinas De potemia, q. 8, a. 1, resp.; Duns Scotus 1639, In I Perihermenias quaestiones, q. 2, Vol. 1,
pp. 186-9; John Buridan 1977, 1, concls. 1-2; translated in John Buridan 1966, pp. 70—1.
6. Duns Scotus 1639, In I Perihermenias quaestiones, q. 2, §4, Vol. 1, p. 188; Buridan 1977, 1, concls.
8—9; translated in Buridan 1966, pp. 74-5. See also John Buridan 1957, p. 20.34-41.
7. Ockham 1974a, 1,3; Commentarium in Perihermeneias ca. 1 (1603-4), cd- in Boehner 1946, p. 320.
It seems to follow also from the psychologico-causal notion of signifi

cation that a speaker's or writer's terms are not significant to him but only
to others. If they were significant to him, their signification would consist
of 'establishing' the very concepts from which they derive their signifi
cation, which seems circular. This feature, too, fits the view that language is
for social communication although, as far as I know, no medieval logician
explicitly recognized the consequence.
There was another problem. The standard notion of signification did
not fit mental terms. They do not causally 'establish' an act of understand
ing; they are concepts, and so are either themselves such an act or else, on
another theory, the intentional objects of such an act, 8 but in any case they
do not cause it. Some authors, therefore, refused altogether to accept the
notion of a naturally significant mental language. 9
Signification in syncategorematic and categorematic terms

Terms were divided into categorematic words - those that can serve by
themselves as terms in the strictest sense - and syncategorematic words -
those, such as conjunctions and prepositions, that enter into propositions
only along with categorematic words. 1 0 There were several theories, not
always clearly distinguished, of the signification of syncategorematic
words:
(1) They have an 'indefinite' or 'unfixed' signification, whereas categorematic

words signify definitely.1 '
(2) They 'consignify' — i.e., signify only when properly combined with categore
matic words. 12
(3) They do not signify at all, but only determine truth conditions. 13
14
(4) They signify mental attitudes. (Recall the claim that language is for talking
See also Duns Scotus 1639. Opus oxoniense, 1, d. 27, a. 1. §§1 & 19, Vol. 5.2, pp. 1135 and 1146
(contrast the text from Scotus cited in n. 5 above); Albert of Saxony 1522 Perutilis logica, 1, ca. 2,
f. 2*; Albert of Saxony 1496, Quaestiones in Perihermeneias, q. unica in proem.
8. See Bochner 1946 and Adams 1977.
9. See William of Crathorn in Schepers 1972, pp. 115-18, discussed in Nuchelmans 1973, p. 212.
Buridan too tends not to speak of concepts as 'signifying' but rather as 'conceiving' their objects.
(Reina 1959, pp. 382-7.) Nevertheless, contrary to Nuchelmans 1973, p. 243, Buridan does
sometimes speak of concepts as signifying. (Buridan 1977,1, concls. 2 and 7; translated in Buridan
1966, pp. 70-1 and 74·)
10. Priscian 1855—9, Institutionesgrammatkae, 11, iv, 15. Sec Nuchelmans 1973, p. 124.
n. Abelard 1919-27, Logica 'Ingredientibus'', p. 13711-32; Abelard 1970a, Dialectica, p. 118.4-25;
Ockham 1974a, I, 4; Burley 1955, De puritate, p. 220.10-11.
12. Boethius 1880, p. 14.30-2; Abelard 1919-27, p. 337.33-40; Burley 1955, p. 220.8-11; Buridan
'957, Ρ· I88.III-I4.
13. Ockham 1974a, I, 4.
14. Augustine 1877, De magistro, 2, 3, PL 32, col. 1196. See Priscian 1855-9, 18,9, 76.
about our thoughts, and that the relation of spoken terms to thoughts is a kind
of signification.)
(5) They 'generate an understanding' but 'have no subject thing'. 15 That is, they
do not produce an understanding 'of something'. In effect, 'to signify' is being
used without a direct object. The objection that every understanding is 'of
something' •' is in part a terminological matter and in part depends on one's
theory of mental acts.17
(6) They signify ways of conceiving things.18
(7) They do not signify 'things' but rather 'modes' or 'characteristics' of things.19
There was general agreement that categorematic words (except for such
as 'concept' or 'knowledge') signify something extramental - 'ultimately'
if not 'immediately' - and that a proper name so signifies its bearer.
Signification here coincides with 'being truly predicable of'. 20 But there
was controversy over the extramental significate of a general term. Realists
of various degrees said it is a universal or common nature; when I hear
'man' I think only of man in general. 21 For nominalists, there are no
universals or common natures in the realists' sense; they did, however,
recognise universal concepts. But one cannot hold that general terms are as
a rule imposed to signify concepts without committing oneself to the view
that language is for talking only, or very largely, about thoughts. The only
plausible alternative was to make 'man' signify individual men, and since it
would be arbitrary to single out some at the expense of others, it must on
that alternative be said to signify each man. For the nominalists signification
and predication are thus linked not just for proper names but for all
categorematic words. 2 2
The realist theory has the disadvantage that the sentence 'Some man is a
philosopher', for instance, cannot be said to make one think of individual
men or philosophers, but of the universal natures 'man' and 'philosopher';
the hearer of the sentence is thus left ignorant of what it is about. The
nominalist theory, however, entails that that same sentence makes the
15. Abelard 1970a, p. 119.3-16.

16. Ibid.
17. Buridan likewise considers and rejects the claim that the verb 'to signify' need not take a direct
object (Buridan 1977, 1, sophisma 4 and concl. 3; translated in Buridan 1966, pp. 67 and 71-2).
18. Buridan 1977,4, translated in Buridan 1966, p. 116.
19. Abelard 1970a, p. 120.3-20; Nicholas of Paris, Summe Metenses, quoted in De Rijk 1962-7, I,
p. 481; Albert of Saxony 1496, Quaestiones in Perihermeneias, q. 2 de nomine, concl. 3.
20. See the discussion in Abelard 1919-27, Logica 'Ingredientibus,' pp. 21.27-22.6.
21. Bacon 1940, Sumule dialectices, p. 272.4-8; Duns Scotus 1639, Opus secundum in Perihermenias.q. 1,
§§4-5, Vol. 1, pp. 212-13; Burley 1955, pp. 7.26-8.33; Ferrer 1909, De suppositionibus, p. 3; also
1977. p. 87.
22. Ockham 1974a, I, 33.
hearer think of all men and all philosophers, even those unknown to him —
which is just as great a disadvantage. 23
Connotation
The theory of categorematic words was complicated by the notion of
'secondary signification' or 'connotation', a notion closely related to
Anselm's theory of paronymy. 2 4 Briefly, connotation-theory attempted
to account for the fact that words such as 'blind' in an oblique way make
one think of, and so signify, sight. The theory is too rich and complex to
rehearse in detail here. But it should be mentioned that it bears directly on
the notion of'nominal definition', 25 on the theory of demonstration, 26
and on the program of reducing the number of ontological categories. 27
Supposition proper and 'descent to singulars'

Supposition is a property of categorematic words only when they serve as
terms (in the strict sense), or extremes, of sentences. 28 (Disputed portions of
this claim will be discussed below.) Supposition-theory is best viewed as
two theories: the theory of supposition proper and the theory of'descent to
singulars'. 29
Divisions of supposition proper

The theory of supposition proper is a theory of reference. Normally an
extreme of a sentence refers to or 'supposits for' everything of which it is
truly predicable; 30 in that case it was said to have 'personal' supposition.
But there are other cases. Some of them — figurative or metaphorical uses —
were classed as 'improper supposition'. 31 Others were given intermediate
status as proper but not personal. W e may distinguish two classes of such
23. See Spade 1975c.

24. See Henry 1964, 1967, 1974.
25. See Spade 1975b and Loux's introduction to Ockham 1974c, pp. 1—24.
26. Moody 1935, Ch. 6.
27. Moody 1935, Ch. 4, and Loux's introduction to Ockham 1974e, pp. 1-24.
28. Ockham 1974a, I, 63; Buridan 1977, Sophismata, 2. remark 1 ; translated in Buridan 1966, pp. 99—
100; Buridan 1957, Dr suppositionibus, p. 180.10—11; Burley De suppositionibus in Brown 1972,
p. 34, par. 2.01; Lavenham Suppositiones in Spade 1974a, p. 93. This claim was fairly standard in
the fourteenth century. For the earlier period sec De Rijk 1962-7 and the discussion in Brown
1972, pp. 19-21.
29. See Scott's introduction to Buridan 1966, pp. 29-42.
30. Ockham 1974a, I, 63; Buridan 1977, Sophismata, 4; translated in Buridan 1966, pp. 99—
IOO.
31. Ockham 1974a, I, 77; Burley 1955, De puritate, pp. 46—7; Buridan 1957, De suppositionibus,
pp. 2 0 0 - 1 ; Lavenham Suppositiones in Spade 1974a, pp. 9 3 - 4 .
cases, although some theories combined them: 3 2 'material' supposition, in

which an extreme that does not have personal supposition supposits for a
spoken or written expression, e.g., 'man' in 'Man is a monosyllable', and
'simple' supposition, in which an extreme that lacks personal supposition
supposits for a universal, e.g., 'man' in 'Man is a species' — whether it is a
universal of the realists' variety or the nominalists' universal concept. 33
Predication and the divisions of supposition

Reversing a remark in Boethius' De trinitate,3* many authors held that the
predicate of a sentence determines which of the three kinds of supposition
the subject may have. Most held that a subject can always have personal
supposition, but one of the other kinds only if the predicate is semantically
appropriate. Such 'rules of supposition' suggest that predicates always have
personal supposition. 35 If so, the laws of'conversion' must be restricted to
sentences with subjects in personal supposition; otherwise the conversion
'Man is a monosyllable; therefore, a monosyllable is a man' would be
sanctioned. 36
Truth-conditions and supposition

On the basis of the theory of supposition proper, some authors constructed
a theory of truth-conditions for categorical sentences. Thus a universal
affirmative is true if and only if its predicate supposits for everything for
which the subject supposits; other cases were handled analogously. 37
Affirmative categoricals have 'existential import'; negatives do not. Thus
'Some man is not a Greek' - read as the contradictory of 'Every man is a
Greek' — is true if and only if either there is a man who is not a Greek or else
there are no men at all. This odd reading is required in order to preserve
traditional relations of opposition. 38
32. Buridan 1977, Sophismata, 3, remark 2; translated in Buridan 1966, p. 100; Buridan 1957, De
suppositionibus, p. 201.34-41; Paul of Pergula 1961, Logica, 2.1, pp. 24-6.
33. On these divisions sec Bochncr 1952, Pt. 2, Ch. 2; Kncalc 1962, pp. 246-74.
34. Boethius 1918, §4, lines 4—5. See Spade 1974c.
35. Spade 1974c, p. 72 n. 38.
36. Swiniarski 1970, pp. 189-91.
37. Ockham 1974a, 2, 1-4; Buridan 1977, 2, concls. 10-14; translated in Buridan 1966, pp. 90-4.
Such an account is by no means a regular part of supposition theory. It is not to be found, for
instance, in William of Sherwood Introductiones in logicam, ed. Grabmann 1937 and tr. Kretzmann
1966, Peter of Spain Tractatus (Summule logicales), ed. De Rijk 1972, Burley De ptiriiate, cd.
Boehner 1955, Burley Dc suppositionibus in Brown 1972, or Lavenham Suppositiones in Spade
1974a.
38. Moody 1953, pp. 51-2.
Ampliation
In order to accommodate tense and modality, the theory of supposition
proper included a theory of 'ampliation'. This theory explains how the
supposition of a subject is extended or 'ampliated' by past- and future-
tensed or modal copulas, and how the semantics of predicates, too, must be
adjusted in those cases. Details varied, but the doctrine was relatively settled
by the fourteenth century. 39 Some authors conjoined this with a theory of
'restriction' to explain how the supposition of a term may be narrowed by,
for instance, an adjective ('tall man') or a relative clause ('man who
runs'). 40
Descent to singulars or modes of personal supposition

The second main part of supposition theory is the theory of descent to
singulars or 'modes of personal supposition'.41 The standard modes are:
'discrete', 'confused and distributive', 'determinate', and 'merely con
fused'. Discrete supposition is the supposition of proper names, demonstra
tives, and demonstrative phrases. There was wide divergence regarding the
definitions of the other modes. Ockham's definitions in Summa logicae, I,
70, are perhaps the most complete. They may be put thus:
Let φ(Τ) be a noncompound sentence containing an occurrence Τ of a
general term in personal supposition, and let i,, i 2 , . . . , f,·... be all the
singular terms that supposit for a suppositum of T. Then Τ is
(a) confused and distributive if and only if φ(Τ) implies φ(ίγ) αηάφ(ί2) and.... and
no φ(ί,) implies φ(Τ);
(b) determinate if and only if φ(Τ) implies φ(ίλ) or φ(ί2) or ..., and each φ(ί·)
implies φ(Τ);
(c) merely confused if and only if φ{Τ) does not imply <*((,) or φ(ί2) or ..., but
does imply φ(ίι or t2 or ...), and each 0(f,) implies φ(Τ).
Although many scholars treat descent to singulars as a theory of analysis

or of truth-conditions, there are reasons to doubt this. First, no medieval
author seems to have made such a claim. Second, some authors, as noted
above, did have an explicit theory of truth-conditions based on sup-
39. Moody 1953, pp. 53-63; Scott's introduction to Buridan 1966, pp. 32-4; Loux's introduction to
Ockham 1974e, pp. 37-44.
40. See Maierù 1972, Ch. 2.
41. Boehner 1952, Pt. 2, Ch. 2; Moody 1953, pp. 43-53; Kneale 1962, pp. 246-74; Scott's intro-
duction to Buridan 1966, pp. 35—42; Swiniarski 1970; Loux's introduction to Ockham 1974e,
ΡΡ· 23-37-
position, and it is quite different. Third, if descent to singulars was intended

to provide such a theory, it will not work, as has been pointed out by
Swiniarski 1970, pp. 210-13, and Matthews 1973.
It is not clear, therefore, what the doctrine was intended to accomplish.
One possibility is that it was meant to provide a technique for checking
inferences. Some authors gave syntactic criteria for deciding which mode
of supposition a term has, 42 and bridge-rules to link the modes to the
theory of consequences.43 As a result, the theory of modes of personal
supposition could in principle be applied to the evaluation of a wide range
of inferences. Thus the inference from 'Twice you ate some bread' to
'Some bread you ate twice' is rejected because on the syntactic criteria
'bread' is merely confused in the former and determinate in the latter (as
the definitions will verify), and there is a rule 'Whenever one argues from a
term suppositing merely confusedly to the term suppositing determinately
with respect to the same multitude, there is a fallacy of figure of speech.' 44
This technique has an advantage over modern quantification theory in that
it handles inferences in natural language directly, and does not require
them to be first translated, more or less by feel, into symbolic notation. On
the other hand, if this was the purpose of the doctrine, it must be said that
the medievals never succeeded in formulating adequate criteria and rules to
accomplish this purpose.
Some controversial points
Finally, some controversial points should be mentioned briefly. First, some

authors held that only subjects supposit, not predicates.45 Again, although
in order to simplify certain rules many authors held that only whole
46
extremes supposit, not their parts, few observed this principle con
sistently. The theory of restriction violates this principle, as does the
analysis of'Twice you ate some bread' by which 'bread' supposits although
'ate some bread' is the predicate. Again, although most authors held that
terms supposit only in sentences, a few explicitly allowed them to have
42. E.g., Ockham 1974a, 1, 71-4.

43. E.g., the rules in William of Sherwood 1937, Introductiones in logicam, pp. 80-1; 1966, pp. 118-19.
44. Burley 1955, De puritate, pp. 21.30—23.35.
45. Ferrer 1909, pp. 10-12; 1977, pp. 95-7.
46. Ockham 1974a, I, 69 and 72 ad 3; Burley De suppositionibus in Brown 1972, pp. 31-4 pars. 1.Σ
Ι.22; Ferrer 1909, p. 12; 1977, pp. 97-8. Sec Buridan 1977, 2, remark 1; translated in Buridan
1966, p. 100.
natural supposition outside sentences.47 But the phrase 'natural suppo

sition' does not always refer to this obscure doctrine. 48
VARIETIES OF SUPPOSITION
Supposition
(a) improper (metaphorical) proper (literal)
(b) material formal
(c) discrete common
(d) simple personal
(e) determinate confused
(f) merely confused distributive confused
(g) immobile (h) mobile
EXAMPLES
Each variety of supposition is exemplified by the occurrence of the word 'man' in each of
these propositions, (a) After six moves the Russian chess player was a man down, (b) Man is
a monosyllable, (c) That man is my brother, (d) Man is a species, (e) A man is at the door, (f )
Every masseur is a man. (g) Every man other than an employee is eligible. [Immobile
because the logical descent under 'man' is blocked by 'other than an employee'; one cannot
infer, e.g., 'John other than an employee is eligible' or even (without further information)
'John is eligible'.] (h) Every man is an animal.
NOTE. The schema presented above is intended to include all the most important varieties
of supposition recognised by medieval logicians, but neither this particular set of varieties
nor their organisation can be attributed to a particular logician.
47. E.g., Lambert of Auxerre I97i,p. 206, and Robertus Anglicus quoted in Brown 1972, p. 21 n. 25.
See the discussion ibid.
48. See Buridan 1957, ΡΡ· 206-8, And Ferrer 1909, pp. 14-43; 1977, PP· 100-34. On the notion of
natural supposition, see De Rijk 1971-73.
ΙΟ
THE SEMANTICS OF P R O P O S I T I O N S
Terms and propositions

Separate treatments of the semantics of terms and the semantics of prop
ositions are justified by the Aristotelian distinction between two levels of
speech and thought (Categories i a i6,2*4; De interpretatione i6 a io): the level
of names and verbs and the thoughts corresponding to them, which do not
yet involve any combination (symploke, complexio) that makes the notions
of truth and falsity applicable, and the level of expressions and thoughts
formed by a kind of combination that has to do with truth and falsity. Just
as Aristotle had made the applicability of the notions of truth and falsity the
criterion for the relevant kind of combination, the medieval semantics of
complex units of speech and thought (complexa) concentrated on sentences
that are used for making statements and are thus either true or false - the
sort of sentences with which logic is primarily concerned. A combination
of words that is used to make known something that is either true or false
(oratio verum falsumve significans)1 was called an enuntiatio or propositio. The
Latin word 'propositio' practically always designates a declarative sentence;
accordingly, in this chapter 'proposition' is used in this medieval sense and
never in the modern sense of that which is expressed by a declarative
sentence. Most medievals were aware of a distinction between a complexio
in the sense of mere predication, without any assertive (or other) force, and
a complexio which is accompanied by an act of judging or asserting that it is
so. Abelard considered mere predication as the common element in dif
ferent speech acts;2 and from the beginning of the twelfth century it was
customary to distinguish between an act of merely putting a predicative
combination before the mind and an act ofjudging that that combination
is the case in reality. The author of the Ars Meliduna3 (between 1154 and
1180) even employed a terminological distinction to mark the difference in
1. Cf. Boethius, De topicis differentiis, PL 64, 1174B, 1177C.

2. Peter Abelard 1970a, Dialectica, pp. 148-53.
3. De Rijk 1962-7, n(i), p. 342.
assertive force between propositions uttered by themselves and propo-

sitions in so far as they are part of compound statements. Uttered by itself, a
categorical proposition both signifies an inherence and asserts that the
predicate inheres in the subject (significat et enuntiat); as part of a compound
statement it only signifies an inherence (tantum significat), without asserting
that it is so. For many purposes, however, the words 'complexio' and
'complexum' could be used in such a way that they included both the
predicative and the assertive aspects.
Written, spoken, and mental propositions
As in the case of incomplexa, or terms, it was held that there are three kinds
of propositions: written, spoken, and mental. Written and spoken de-
clarative sentences and their mental images were contrasted with the
corresponding thoughts, which were seen as belonging to a sort of uni-
versal mental language. N o w the first question of the semantics of prop-
ositions concerns the relation between conventionally signifying written
and spoken propositions and their mental counterparts. This question,
which we might call the problem of the meaning of a declarative sentence,
received a relatively uncontroversial answer.
The meaning of a declarative sentence

In accordance with Aristotle's conception in De interpretatione i6 a 3, it was
held that a written proposition or its mental image conventionally signifies
the corresponding spoken proposition and that a spoken proposition or its
mental image conventionally signifies the corresponding mental prop-
osition or act of thought. Some authors, for instance Ockham 4 and
Gregory of Rimini, 5 were of the opinion that written and spoken prop-
ositions have the same significate as the mental propositions, the first two as
conventional signs and the latter as natural signs, but even they restored
something of the Aristotelian hierarchy by considering the written proposi-
tion as a secondary conventional sign compared to the spoken proposition,
and the spoken proposition, in its turn, as secondary and subordinate to the
mental proposition. The written and spoken propositions discharge their
signifying function in subordination to the mental acts of apprehending
and judging which always precede them. In the last instance, therefore, it is
the mental proposition which is immediately directed towards the outside
world.
4. William Ockham 1974a, Summa logicae, I, 12.
5. Gregory of Rimini 1522, Prologus, q. 1, art. 3; cf. 11, Dist. 9-10, q. 2, art. 1.
The semantics of propositions 199
The mental proposition

The mental proposition is an act of thought which consists of an act of
combining the predicate with the subject in an affirmative or negative way
and of an act of judging that what is thereby conceived is so in reality.
Considered as a unit of a universal mental language, it is said to signify in a
natural manner; at the same time - and this, as will appear, is a point of
crucial importance - it contains a mental copula, and its total signification
is thus determined by a syncategorematic element.
The significate of a mental proposition
In order to understand clearly the second and most controversial question

of the medieval semantics of propositions, namely the problem of the
significate of a mental proposition, one should realise that the word
'significare' has a special sense in this context. O n the one hand, it is used of a
person who performs an act of expressing a belief or making known an
opinion, and of the expression involved in that act; in such cases it has the
same meaning as the verbs 'enuntiare', 'dicere', 'proponere'. O n the other
hand, in keeping with the interpretation of thought as mental speech the
terminology of the speech act of asserting could easily be transferred to the
purely mental acts of judging, assenting, dissenting, believing, and know-
ing; judging is then conceived of as a kind of asserting in the mind
(enuntiare mentaliter quod sic est). The question of the total or adequate
significate of a mental proposition, as contrasted with the significates of
written and spoken propositions and with the significates of the categore-
matic terms, thus becomes the problem of the specific object of an act of
judging or an act of asserting, or, in other words, the question of the nature
of 'the thing' (pragma, res) which, according to Aristotle, underlies an
affirmation or negation (Categories I 2 b 6 - i 5 ) . In so far as such an object is
postulated for both true and false assertions or judgements, it must have the
character of something intermediate between the acts of asserting or
judging and the outside world; if only true assertions and judgements are
considered, it may be asked what kind of place such an object occupies in
the actual world. This problem concerning the nature of the object of what
we would call the propositional attitudes naturally leads to the further
question of what are the primary bearers of truth values, the modalities,
and the logical relations. Again, part of this further question may be seen as
the problem of deciding what Aristotle meant by 'the thing' in his formula
'It is because the thing is or is not that the statement is said to be true or false'
(Categories 4 b 8, I4 b 2i).
The breadth of medieval semantics

From the foregoing it will be clear that the medieval semantics of prop-
ositions included, apart from problems which we would call semantic in a
strict sense, certain issues which are at least very closely connected with
such disciplines as philosophical psychology, epistemology, and ontology.
The questions treated, which still belong to the core of what is now called
philosophy of language, were often conditioned by such typically medi-
eval interests as God's knowledge, the articles of faith, and the object of
theology. But the ways in which those questions were asked and answered
can usually be sufficiently dissociated from these narrow contexts to remain
instructive for the modern philosopher of language.
From Abelard through the thirteenth century

Both in the Logica 'Ingredientibus' and in the Dialectica Abelard draws a
sharp distinction between mere predication and the act of asserting. Since
the copula is a syncategorematic sign, predication is essentially a manner of
conceiving (modus concipiendi), an operation oflinking subject and predi-
cate in an affirmative or negative way. In order to get a full statement-
making utterance, however, it is not enough to perform such a formal act
of compounding or dividing; one must also assert, by means of a finite
verb, that the constructed inherence is as a matter of fact the case. N o w the
dictum, or that which is asserted to be the case, although it is about things
and not about ideas or words, cannot itself be a thing in the sense of things
which are denoted by nouns; it is rather an asserted state of affairs or
manner in which the things denoted by the categorematic subject-term
and predicate-term are related (rerum modus habendi se).6 Because of the
syncategorematic character of the copula a proposition can signify or assert
only the way in which things are connected; the significate of the prop-
osition, therefore, does not belong to the category of things, but is a way
of being which can be made more explicit by such impersonal or non-
denoting expressions as 'ita (or sic) est', 'accidit', 'contingit', 'evenit' when
they are combined with an expression in indirect discourse (accusative plus
infinitive). This way of being can be signified only in a complex manner,
namely by adding a certain mode of conceiving to the conceptions of the
things which by this very complexio become the subject and the predicate of
the proposition. Abelard considers the dictum, or asserted predicative con-
nection, as the primary bearer of the truth values; the written or spoken
6. Peter Abelard 1970a, p. 160.

The semantics ofpropositions 201
propositions and their mental counterparts are true or false in a derivative

sense. As a dictum can be true or false, it must be something intermediate
between the act of asserting and the actual world, but apart from the
emphatic categorisation as a non-thing its ontological status is left very
much in the dark.
Some of the treatises on logic which date from the period between the
middle of the twelfth century and the beginning of the thirteenth century
clearly show the influence of Abelard's dictum theory. Besides 'dictum' and
'significatum propositionis' they often use the word 'enuntiabile' for that
which is asserted to be the case; in accordance with its form, this term also
occasionally stands for an assertible, in the sense of that which can be
asserted although it is not actually asserted. The author of the Ars Meliduna1
mentions several opinions concerning the nature of assertibles and prefers
the view adhered to by his teacher, according to which enuntiabilia are
neither substances nor qualities but have a peculiar being of their own: they
are grasped only by reason and thought and are inaccessible to the senses.
This Platonising doctrine is also found in the Ars Burana,8 where an
assertible is called an extrapraedicamentale, not, however, because it does not
belong to any category at all, but because it belongs to a separate category
of being which is not one of the ten distinguished by Aristotle.
Whereas Abelard and his followers limited their attention almost exclu-
sively to the object of an act of asserting, theologians of the same period saw
themselves forced to deal with the object of acts of believing and knowing
(thecrerfi'iKMi, the scitum) by certain difficulties concerning the identity of the
articles of faith and the immutability of God's knowledge. The problem of
the identity of the object of faith arose from some remarks made by
Augustine in his exegesis of the Gospel according to Saint John 10. 8,9 and
in his De nuptiis et concupiscentia:10 although what is believed about the
events in Christ's life is worded in different ways, notably by different
tenses of the verbs, by those who express their belief before and after his
coming, the content must remain exactly the same.
One answer to the question of what the object of faith actually is came to
be called the res theory: the object of faith which in spite of different ways
of expressing it remains the same is the actual thing or event, for instance
Christ's birth (nativitas Christi), considered as an incomplexum, as something
7. De Rijk 1962-7, 11(1), pp. 357-9·

8. De Rijk 1962-7,11(2), p. 208.
9. In Joltanms Evangelium tractatus, XLV, 9, PL 35, 1722.
10. II, 11, PL 44, 450.
which in any case does not have the complexity of a proposition. This view
was supported by pointing to the fact that God, the prime object of faith, is
absolutely simple. Moreover, the faith of the believers who live in this
world has the same object as the future vision which they will enjoy in
heaven; and the object of that vision, which is a simple intuition of the
highest light, is certainly not complex or determined by qualifications of
time. 11
Against this res theory it was urged by others that the object of an act of
believing is something to which the notions of truth and falsity are
applicable and which must therefore be the product of a complexio: a
complexum or proposition. The adherents of this complexum theory often
use the word ' enuntiabile' for the proposition which is the object of belief;
it should be noted that in this context 'enuntiabile' has practically the
same meaning as 'enuntialio' and designates a declarative sentence or the
accusative-plus-infinitive expression corresponding to it. N o w in order to
save the sameness of the object of faith which they identified with a
proposition, the complexum theorists tried to replace the declarative sen-
tences which have an indexical character because of the different tenses of
verbs by some kind of eternal sentence in which the indexical elements
have been neutralised or eliminated and whose truth value cannot change.
The simplest device was to make the proposition which is the object of faith
consist of the disjunction of three differently tensed verbs, for example:
'that Christ was born or is born or will be born' ('Christum esse natum, vel
nasci, vel nasciturum esse').12 Another solution was suggested by the so-
called nominales, who made a distinction between the principal signifi-
cation and the accidental signification of words. Words which belong to
different word-classes, such as 'albedo', 'albet', 'album', have a common
principal signification and different accidental significations, just as 'albus',
'alba', 'album', are one and the same nomen, with the same principal
signification and different accidental significations. In the same way, it was
argued, declarative sentences with different tenses of the verbs and con-
sequently different accidental significations may be taken as varying forms
of one and the same proposition, whose identity is determined, not by the
sounds or the accidental modes of signifying, but by the unity of the thing
signified. 11 Others, for instance Bonaventure 1 4 and Peter of Tarantasia, 15
n. Cf. Bonaventure i882-ioo2a, Dist. 24, art. i,q. 3; Peter of Tarantasia 1652, Dist. 24, art. 3.
12. Cf. Chenu 1934, p. 131.
13. Some of the relevant texts are quoted in Chenu 1934.
14. Bonaventure 1882—19023, Dist. 24, art. 1, q. 3.
15. Peter of Tarantasia 1652, Dist. 24, art. 3.
who had objections against certain aspects of the theory of the nominales,
distinguished two kinds of propositions, one with tenses of the verb which
are in accordance with the contingent position occupied by the believer in
the course of time, and one that is neutral and indifferent to all time, with a
verb that has been made tenseless. The substance of faith does not depend
upon superficial changes in the position of the believer and may thus be
best expressed by means of tenseless propositions. Finally, there were also
mixed views. Thomas Aquinas,16 for example, was of the opinion that
according as the question is considered from the standpoint of the believer
or from the standpoint ofthat which is believed both the complexum theory
and the res theory may claim relative truth.
Indexical expressions were also discussed in connection with the immu-
tability of God's knowledge. Peter Lombard17 mentions the following
objection against the doctrine that if God knows something at a certain
time, he has always known it and will always know it. God formerly knew
that the world would be created; but he does not know now that the world
will be created; therefore he knew something which he does not know
now. From a consideration of the proper function of such indexical
expressions as 'yesterday', 'today', 'tomorrow' and the tenses of the verb,
Peter concludes, however, that in the same way as those who believed that
Christ would be born and those who believe that he was born have the
same faith in spite of the differences in the wording, God, who knew before
the creation of the world that it would be created and who now knows that
it has been created, has exactly the same knowledge about the creation of
the world.
The fourteenth century

The theories concerning the significate of a proposition which were put
forward in the course of the fourteenth century show a marked continuity
with the views that had been developed in the past. But while one gets the
impression that Abelard's dictum theory and, on the other hand, the res
theory and the complexum theory were not really rivals, in the later period
theories of all three types were commonly seen as in sharp conflict with
each other. The theory which comes closest to Abelard's dictum theory is
the theory of the complexe signiflcabile, usually associated with Gregory of
Rimini. It is not unlikely that a similar view had already been defended by
William of Crathorn, but from the few sources which are available at
16. ST, llallae, q. I, art. 2.

17. Sent., I, Dist. 41.
present it is hard to tell exactly what William's doctrine was. 1 8 According

to Gregory, then, a mental proposition in the proper sense is either an act of
assenting or an act of knowing. At the very beginning of his commentary
on the Sentences, which he completed shortly before 1345 in Paris (Prologus,
q. 1, art. 1), he raises the question of what exactly the object of theology is
and in that connection discusses the more general problem of the nature of
the object of knowledge that is acquired by scientific proof. Rejecting the
view that the object of knowledge is a proposition or complexum and also
the view that it is a res in the outside world, he defends the thesis that what is
known and assented to or believed is that of which the mental proposition
is a natural sign and of which the written and spoken propositions or their
mental images are conventional signs. This total and adequate significate of
the proposition he also calls the enuntiabile (in the sense in which that word
was used in such treatises as the Ars Meliduna and the Ars Burana) and the
complexe signiflcabile: something that can be signified only by a proposition
which contains an act of complexio and never by a single word or a combi-
nation of words that lacks affirmative or negative force. It should be noted
that Gregory does not use 'dictum in Abelard's sense; when he makes use of
that term, it means the words uttered and not that which is asserted by
those words. In trying to answer the question of the ontological status of
this object of assent and knowledge Gregory distinguishes three nuances of
meaning in such words as 'aliquid', 'res', 'ens'. Although it cannot be held
that an assertible is something in the sense of a substance or accident as
denoted by categorematic terms, it is a thing in the sense in which Aristotle
speaks of the thing which underlies an affirmation or negation and in
which he says that a statement is true or false because the thing is or is not;
and it may also be called a thing, in case the proposition is true, in so far as it
is part of the actual world. In other words, it is either that which is asserted
or believed (or could be asserted or believed) to be the case and can thus be
true or false, or that which is actually the case, but never a thing in the
narrowest sense. For Gregory the bearers of truth and falsity (and of the
modalities) are not only actually existing propositions and the significates
of actually existing propositions, but also states of affairs that are capable of
being signified by true or false propositions even if these corresponding
propositions do not in fact exist. In that case the true assertibles are always
actually signified by God as the uncreated sign of all truth. 1 9
18. Cf. Nuchelmans 1973, pp. 212-9. Recent research seems to prove that Adam Wodeham held a
similar theory before Gregory of Rimini; see Gàl 1977.
19. Gregory of Rimini 1522,1, Dist. 39, q. i,art.2.
Among those who followed Gregory in postulating some kind of

complexe signiflcabile - Bonsembiante Beduarius of Padua, 20 John of
Ripa, 2 1 and Albert of Saxony 22 - there were also some, notably Ugolino
of Orvieto 2 3 and perhaps Paul of Venice, 2 * who held that in certain cases
the complexe significabile may be identified with a thing in the strictest sense.
If, for example, the complexe significabile is the content of a proposition that
asserts the existence of one actual thing, such as that there is a God (deum
esse), then the significate is nothing but the thing in so far as it actually
exists. The only difference between the significate that God exists and the
incomplex thing, the existing God, is that they are signified and under-
stood in different ways: in a verbal way (deum esse) and in a nominal way
(deus ens). Only if an actually existing thing is signified in a verbal way, by
means of a complexio which makes the notions of truth and falsity ap-
plicable, is it a proper object of acts of assenting and knowing.
The Ars Meliduna25 had already mentioned a theory according to which
an assertible is nothing but an act of thinking that something is the case. If
'complexum' is taken in the sense of a mental proposition as a particular act
of apprehending things in a manner characteristic of a complexio, then such
an act theory of the assertible may also be called a complexum theory. In
the fourteenth century this type of theory was put forward by William
Ockham 2 6 and Robert Holkot. 2 7 In their view the immediate object of an
act of believing or knowing is nothing but that act itself; the act of naturally
signifying and the thing signified are one and the same. The mental
propositions or particular acts of believing and knowing are also the
primary bearers of the truth values; truth and falsity are not qualities of such
propositions, but those propositions themselves. More or less the same
thesis was maintained by John Buridan and Marsilius of Inghen. 28
Buridan 29 holds that such accusative-plus-infinitive expressions as
'Socratem currere' ('that Socrates runs') can often be interpreted according
20. Quattuorprincipia; cf. Élie 1937, pp. 140-5.

zi. Cf. Vignaux 1977, pp. 327-9.
22. Albert of Saxony 1497b, Quaestiones super Analytica posleriora, I, q. 2, q. 7, q. 33.
23. Prologus to the commentary on the Sentences, q. 1, art. 2 (Zumkcller 1941, pp. 290-301). Cf. also
Eckermann 1972.
24. Cf. Nuchelmans 1973, pp. 268-9.
25. De Rijk 1962-7,11(1), p. 357.
26. William Ockham 1967, Prologus; William Ockham 1491, QuodUbeta septem, especially III, 6; IV,
17; V, 6; V, 24.
27. Robert Holkot 1510b, I; also QuodUbeta 1,6, in Courtenay 1971.
28. Marsilius of Inghen 1501, Prooemium to Book I, q. 2, art. 3; Marsilius of Inghen 1516,1, q. 1.
29. John Buridan 1588, esp. IV, q. 10, q. 14; V, q. 7; VI, q. 10., q. 11.
2θ6 The high middle ages: semantic theory
to material supposition, in such a way that they stand for the proposition
'Socrates currit'. If, on the other hand, they have to be taken in a signifying
function (significative or personaliter), then they refer to exactly the same
thing in the outside world as does a combination of words or concepts
which does not contain a copula. Just as the phrase 'Socrates currens'
('running Socrates') has an application if in the outside world there is an
individual named 'Socrates' in the state indicated, the expression 'Socratem
currere' is true if the same individual in that state exists. If there is any one
significate of a complexum as such, it is a thing in a certain state (res sic se
habens); but exactly the same thing can also be signified in a nominal way.
Peter of Ailly, 3 0 on the other hand, who was also an adherent of the
complexum theory, rejected the doctrine that basic affirmative propositions,
if true, name one entity in a certain state as determined by the categore
matic and syncategorematic parts of the proposition. According to him
such propositions refer to all the entities referred to by the categorematic
parts and they do so in a certain way (aliqualiter), but asking what kind of
thing the significate of the proposition as a whole is betrays a lack of insight
into the proper workings of language.
Finally, mention should be made of some variants of a res theory. In the
first place, some authors agreed with the complexum theorists that the
mental proposition is the primary bearer of truth and falsity, but at the
same time rejected the view that the proposition is the object of acts
of assenting or knowing. A res theory in this sense was probably held
by Walter Chatton 3 1 and quite clearly by André de Neufchâteau. 32
According to the latter it is not the proposition 'God is three and one' that is
the object of assent, but rather that which is signified by the proposition,
namely that God is three and one. Before we judge that the proposition is
true, we must give our assent to this significate. But when he comes to
explain what this significate is, André does not follow Gregory of Rimini,
but sides with Buridan: that God is three and one is nothing but God being
three and one, the thing that is such and is in such a state as is required for
the truth of the judgement. T o a certain extent this theory resembles the
view taken by the Thomistic school in the debate concerning the object of
assent and knowledge. John Capreolus, 33 for instance, is of the opinion
that the proximate object of belief and knowledge is the proposition
30. Peter of Ailly 1490—5b, esp, the first part of the Insolubilia.
31. Cf. Nuchelmans 1973, pp. 210-12,217-18.
32. André de Neufchâteau 1514, Prologus, q. 1—3; Dist. 2, q. 1—2; Dist. 33—4.
33. John Capreolus 1900-8,1, pp. 51-7.
formed by the mind; this mental proposition is the product of an act of

compounding or dividing and is the matter to which the act of judging is
related. This internal object, however, is not that to which the act of
believing or knowing is actually directed, but only the medium through
which the mind tries to establish contact with the outside world. The
ultimate object of belief and knowledge is, therefore, the thing in the
outside world or rather the way in which things in the outside world may
be connected, namely either as a combination of matter and form or as a
combination of substance and accident. Such a proposition as 'Socrates is
white' does not, according to Capreolus, have any one significate, but
rather has many significates: it signifies Socrates, whiteness, that Socrates
falls under the concept of whiteness, and the present. All these things,
however, do not constitute a genuine unity. The Thomistic view thus
seems to differ from the extreme realism maintained by Walter Burley, 34
who argued for the existence of propositiones in re which are compounded
of things outside the mind and form the adequate and ultimate significate
of true mental propositions, which consist of concepts. In a proposition
which exists in the world the things that serve as subject and predicate are
the matter, and the copula, which consists in the compounding or separat-
ing activity of the mind, is the form. Just as an act of attention unites the
seeing in the eye and the visible object outside the eye so that we can speak
of an object seen, the intellectual act of compounding or separating
together with the subject-thing and the predicate-thing forms a unity
which is a copulated entity (ens copulatum). This copulated entity pre-
sumably consists of the subject-thing and the predicate-thing in so far as
they are truly judged to be identical or separate.
Sentence-tokens and sentence-types

This survey of the main currents in the medieval semantics of propositions
may be concluded by briefly describing at least one further important issue
connected with determining the nature of the proposition and its signi-
ficate. When the proposition was considered as a bearer of truth values,
either primary or secondary, it was generally taken to be a sentence-token.
That at least some medievals were aware of a distinction between sentence-
tokens and sentence-types is evident from Abelard, 35 who maintained that
one cannot make sense of the objection that the same utterance, for
34. Walter Burley 1497, esp. the prooemium; cf. Perer Tartaretus 1514a, f. 5 R; 1514b, f. 41 R;
Domingo de Soto 1 587, Prologus Praedicamentorum, pp. 108-9.
35. Peter Abelard 1970a, p. 54; cf. also p. 71.
2θ8 The high middle ages: semantic theory
instance 'Socrates is seated', seems to be both true and false, unless the
utterance is taken in the not unusual acceptation of those who call different
utterances the same sentence on account of a similarity in form. This
notion of a sentence-type is, however, mentioned only to be rejected. The
interpretation of a proposition as a sentence-token was supported by the
Boethian definition of a proposition as a combination of words which
signifies (significans) something true or false. This definition was held not to
apply to an ambiguous sentence (multiplex propositio). In the same vein
sentences containing such indexical expressions as demonstrative or per
sonal pronouns were thought to signify nothing if they are uttered without
reference to a thing present. By themselves, without context or situation,
they are not sufficient to assert something, although they are capable of
being used to make a statement which is either true or false. 36 When Paul of
Venice 3 7 felt that not only sentence-tokens but also sentence-types should
be counted as propositions, he accordingly changed the Boethian defi
nition and spoke of a combination of words which is capable of signifying
(signiflcativa) something true or false. That sentence-types were occasion
ally taken as signifying the true or the false is obvious from those cases in
which the truth value is said to change. Bonaventure, 3 8 for example, thinks
it wrong to hold — as the nominales did — that a verb in the present tense
signifies different present things at different times. Just as the noun 'man'
has exactly the same signification when it is used to refer to Peter and when
it is used to refer to Paul, so a verb in the present tense has exactly the same
signification whether it is uttered today or tomorrow; it signifies the pre
sent in general. Similarly, a sentence to the effect that you will be born re
mains the same at whatever time it is uttered; consequently, the same sentence
which was formerly true is now false. And Peter of Ailly 3 9 says that such a
proposition as 'The Antichrist will exist at the future moment c' is true
before c if the Antichrist exists at r, but false (rather than inapplicable) after
c. It may cease to be true and is therefore mutably true; but once it has
become false it remains immutably false. In this connection it is worth
mentioning that the author of the Ars Meliduna*0 considers the assertible
which belongs to the sentence 'Socrates loves his son' as nugatory (nuga-
torium) when Socrates ceases to have a son. The context makes it clear that
36. Introductiones Montanae minores, in Dc Rijk 1962-7,11(2), p. 19.

37. Paul of Venice 1499, Logica magna, f. ΙΟΙ.
38. Bonaventure 1882-19023, Dist. 41, art. 2, q. 2.
39. Peter of Ailly 1490, q. II, art. 1, C - D .
40. In De Rijk 1962-7,11(1), pp. 362-3.
'nugatory' means the same as 'neither-true-nor-false'. There are, therefore,

assertibles which, according to the circumstances in which they are asserted
(iuxta rei variationem), can repeatedly begin and cease to be either-true-or-
false. Apparently the assertible is here taken as belonging to the sentence-
type.
The bearers of logical relations

As to the question of the bearers of logical relations, an interesting distinc-
tion is found in the Tractatus Anagnini.*1 Such a logical law as the law of the
conversion of negative universal statements may be formulated about
things (No man is a stone; therefore no stone is a man), about propositions
(The proposition 'No man is a stone' is true; therefore the proposition 'No
stone is a man' is true) or about assertibles (It is true that no man is a stone;
therefore it is true that no stone is a man). In practice, though, logic was
done mainly in the first way. The relevance to logic of the question of the
nature of the bearers of truth and falsity becomes clear when one considers
the consequences of such a complexum theory as was held by Robert
Holkot.42 According to him a proposition which is not actually formed
cannot be true or false. Consequently, such logical rules as 'Some prop-
osition is true; therefore its contradictory is false' apply only to those cases
in which the proposition concerned really exists. Without this presup-
position of existence the rules do not hold, for it is quite conceivable that,
for example, the true proposition 'You run' is the only proposition in
existence, and then it does not follow that its contradictory is false, since
that contradictory has not been formulated. Walter Burley 43 pointed out
that if this view were correct, it would be impossible for a debater to get
involved in a contradiction. For given that the affirmative proposition and
the negative proposition do not exist at the same time, the first is not true
when the second is false and the second is not false when the first is true.
Moreover, every disputation becomes pointless, since the respondent
cannot react to the propositions that are propounded by his opponent. As
an answer to this difficulty Burley maintains that the mind can understand
a proposition which is abstracted from particular utterances, just as it can
think of a lion or an elephant in general without conceiving of a particular
lion or elephant. In the same way as the word 'homo' uttered by me and the
word 'homo' uttered by you have something in common - the universal
41. In De Rijk 1962-7,11(2), p. 236.

42. Courtenay 1971. p. 15.
43. Walter Burley 1497, ad d l . I4bi2.
nature of the word that exists in both particular utterances and is dis-
tinguishable from them - so the mind can understand a proposition which
is the universal nature common to different particular utterances. It is this
kind of abstract proposition which both the respondent and the opponent
in a disputation have before their minds; by means of this universal it can be
explained that a debater contradicts himself, because the universal is some-
thing that remains the same throughout the debate and can be first con-
ceded and then denied.
II
SYNCATEGOREMATA, EXPONIBILIA,
SOPHISMATA
A grammatical distinction between categoremata and syncategoremata

The paradigm of the categorical proposition with which medieval lo
gicians were primarily concerned is a sentence of two words that serve as
the subject term and the predicate term - e.g., 'Socrates currit.' ' Any word
that can be used alone as a subject term or as a predicate term is classifiable as
a categorematic word; all other words are classifiable as syncategorematic
words, those that can occur in a proposition, whether categorical or
hypothetical,2 only along with at least one properly matched pair of
categorematic words - e.g., 'Solus Socrates currit', 'Socrates currit contingen
ter', 'Socrates non currit', 'Si Socrates currit, Socrates movetur.' Drawing the
distinction between categoremata and syncategoremata along this fine,
which seems to have been the original line of distinction,3 produces
mutually exclusive and jointly exhaustive classes that coincide almost
I. In standard twentieth-century philosophical usage a proposition is not a sentence but the content
of a sentence, so that the two sentences 'Socrates currit' and 'Socrates is running* express one and
the same proposition. But when the medievals spoke of a propositio they were speaking not of the
content of a sentence but of a sentence, a propositional vehicle or sign, written or spoken or
mental. The medieval logicians* enuntiabile or dictum of a propositio corresponds most closely to a
proposition considered as the content of a sentence. (See Kretzmann 1970 and esp. Nuchelmans
1973·) Nevertheless, in this discussion I will use 'proposition' as the English equivalent of the
medieval 'propositio'.
In Latin a complete subject-predicate proposition may consist of only one word - e.g., 'Curro'
(Ί am running', Ί run') — and Latin, like English, has impersonal verbs, such as 'pluere' ('to rain'),
which take only dummy subjects in order to form complete propositions. For purposes of this
analysis, however, such understood personal pronouns and dummy subjects also count as subject
terms.
z. The medieval categorical/hypothetical distinction is broader than the twentieth-century use of
'hypothetical' might suggest, closely approximating the modem distinction between atomic and
molecular propositions (or sentences). Thus both these propositions are hypothetical in the
relevant sense: 'If Socrates is running, he is moving'; 'Socrates is running, and Plato is walking.'
3. The source of the medieval distinction seems to have been this well-known passage from
Priscian's Institutionesgrammatical. 'Partes igitur orationis sunt secundum dialecticos duae, nomen
et verbum, quia hae solae per se coniunctae plenam faciunt orationem; alias autem partes
"syncategoremata", hoc est, consignificantia, appellabant' (Priscian 1855-9,2,54-5)· The distinc
tion as Priscian presents it is grammatical in character despite his attribution of it to 'the
dialecticians', (On the identity of these dialecticians, see Nuchelmans 1973, p. 124.)
perfectly with certain groupings of the parts of speech (partes orationis)

recognised by medieval grammarians: the categoremata are the names
(both substantival and adjectival),* the personal and demonstrative pro
nouns, and the verbs (excluding auxiliary verbs); 5 the syncategoremata are
all the others - e.g., the conjunctions, adverbs, and prepositions.
The logicians' notion of syncategoremata

The notion of syncategoremata that became important in medieval logic
was, however, both narrower and broader than that comparatively orderly
classification in terms of the parts of speech. Although more than fifty
different words were considered in one or another medieval logician's
treatment of syncategoremata, by no means all non-categorematic words
in even the relatively small classes, such as conjunctions, were of enough
interest to the logicians to be treated expressly among their syncatego
remata. O n the other hand, several words that might have been classified as
categorematic on grammatical grounds became prominent members of the
logicians' syncategoremata. The inclusion of certain grammatically categore
matic adjectives, for instance, such as 'omnis' and 'infinita', may be among
the reasons that led to a further distinction between categorematic and
syncategorematic uses of a single word. 6 Certain pronouns, such as 'quic-
quid' and 'uterque',1 and at least four verbs — 'differ*', 'vult', and especially
4. Substantival names are either common ('homo') or ptoper ('Socrates'), either concrete ('homo') or
abstract ('humanitas'). Adjectival names are adjectives, but a Latin adjective in the neuter case can
be used as a substantival name — e.g., 'Album currit' ('What is white is running', Ά white thing is
running').
5. The verb 'esse' is a special case. While it is not uncommon for medieval logicians to distinguish
categorematic and syncategorematic uses of a single word, such uses arc particularly obvious and
important in the case of'esse', which can be used either existentially. and hence categorematically,
as in 'Socrates est', or copulatively, and hence syncaiegorcmaticalty, as in 'Socrates est homo',
'Socrates est albus', 'Socrates est currens'. (For William of Sherwood's treatment o f this distinction
see Kretzmann 1068, pp. 90—3.) The distinction (and close relationship) between these two uses of
'esse' was recognised and made much of in Abelard's logic, before the development of a branch of
logic devoted to syncategorematic words (see Kretzmann forthcoming). Obviously, intransitive
verbs such as 'currere' ('to run*) constitute the paradigm of verbs considered as categorematic
words in this classification, but even such clearly transitive verbs as Omare' can occur in a complete
two-word proposition — as in 'Socrates amal' ('Socrates loves') - without taking an object.
6. For example, in 'Omnes currunt' ('All are running'), 'omnes' is used pronominally and hence
categorematically; and in 'Mundus est omne' ('The world is everything*) 'omne' is used nominally
and hence categorematically. But in 'Omnis homo currit' ('Every man is running*) "omnis" is, despite
its grammatical character as an adjective, being used as a quantifier, a signum (sign |of distri
bution]), and hence syncategorematically. (See Kretzmann 1968, pp. 17-18.) For distinctions
between the categorematic and syncategorematic uses of 'infinita' ('infinitely many'), see
Kretzmann 1968, pp. 41—3.
7. 'Quicquid' ('whatever*) and 'uterque' ('both') can and frequently do stand alone as subject terms,
just as their English counterparts do, but they are also signs o f distribution.
Syncategoremata, sophismata, exponibilia 213
'incipit' and 'desinit'8 - were also discussed more or less regularly as syn
categoremata. But those pronouns and verbs found places among the
syncategoremata not so much because of their syntactic roles or in virtue of
special syncategorematic uses to which they can be put as because of the
9
meanings they have in their standard uses. The logicians' notion of
syncategoremata, then, was less precisely defined than the grammatical
notion from which it had been derived. Most logicians who dealt with
syncategoremata explicitly offered some sort of general account of them,
and a few attempted to order them systematically,10 but the notion
persisted and evolved because of its usefulness and not because it picked out
a clearly recognisable category of linguistic or logical entities. Perhaps the
most persistent theme in general accounts of the nature of the syncatego
remata is that they are words whose signification is incomplete in a special
respect, different from the respect in which, as Aristotle had pointed out,
the signification of (categorematic) verbs is incomplete. 11 Here, for in
stance, is the account offered by Henry of Ghent: 'And they are called
syncategorematic as if to say "consignificant" - i.e., significant together
with others, namely, with categoremata - not because they signify nothing
on their own, but because they have a signification that is not definite but
indefinite, a signification whose definiteness they derive from those
[words] that are adjoined to them. For they do not signify any thing, but
8. All these verbs ordinarily call for complements, either infinitive or prepositional phrases. They
can, however, stand alone as predicate terms, as in the crusaders' motto 'Deus vuW. ('KM//', by the
way, seems very rare as a syncategorematic word; I have found it treated explicitly as such only by
Nicholas of Paris.) 'Differt', 'incipit', and 'desinit' arc likely to have found their way into the
syncategoremata originally because each of them involves covert negation (brought out ex
plicitly in the analyses (expositiones) of propositions in which they occur), and negating devices are
among the most important and universally recognised syncategoremata. (See Kretzmann 1976.)
9. See, however, William of Sherwood's attempt to distinguish categorematic and syncategore
matic uses of 'incipit' (Kretzmann 1968, pp. 106-8).
10. See, for example, the reasonably systematic organisation of William of Sherwood's treatment of
syncategoremata (Kretzmann 1968, p. 8). The notion of syncategoremata might be sorted less
systematically, but perhaps more recognisably from a twentieth-century point of view, into the
following topics, each of which is associated with one or more than one logical or semantic
relationship: Distribution (or quantification), e.g., 'omnis', 'totus'; Negation, e.g., 'non', 'nihil';
Exclusion, e.g., 'solus', 'tantum'; Exception, e.g., 'praeter', 'nisi'; Composition (or predication),
e.g., 'est', 'incipit'; Modality, e.g., 'necessario', 'contingenter'; Conditionality, e.g., 'si', 'quin';
Copulation (or conjunction), e.g., 'et'; Disjunction, e.g., 'vel', 'utrum'; Comparison, e.g., 'quam';
Reduplication, e.g., 'inquantum', 'secundum quod'.
11. De interpretatione, Ch. 3: Ά verb . . . is a sign of things said of something else ... And it is always a
sign of what holds, that is, holds of a subject... When uttered just by itself a verb is a name and
signifies something - the speaker arrests his thought and the hearer pauses - but it does not yet
signify whether it is or not. For not even "to be" or "not to be" is a sign of the actual thing (nor if
you say simply "that which is"); for by itself it is nothing, but it additionally signifies some
combination, which cannot be thought of without the components.' (Ackrill translation.)
they signify rather in the manner of a disposition of a thing and of terms

signifying things. Every disposition, however, is indefinite in itself and is
made definite by that which it disposes.'12
Syncategoremata in the logica moderna

The logicians' interest in syncategoremata began to flourish, naturally
enough, in the rise of the logica moderna, stimulated by the recovery of
Aristotle's treatise on fallacies around the middle of the twelfth century.13
The analysis of linguistic fallacies seems to have alerted the early terminist
logicians to the possibility that the logical relationships central to syllogistic
theory can be altered or blocked when certain expressions (common in
ordinary discourse but excluded from the standard forms of Aristotelian
logic) occur in the propositions whose logical relationships are in question.
Some of the words that were later to be treated systematically as the
syncategoremata are cited more or less casually in the early literature of the
logica moderna just because of their disruptive effect on standard logical
relationships or operations,14 or because of the changes they bring about in
the standard interpretation ofother expressions important to logicians.15
12. 'Et dicuntur sincathegoreumatice, quasi: consignificative. idest: cum aliis significative, scilicet cum
cathegoreumaticis; non quia de se nichil significant, sed quia habcnt significationem non finitam
sed infinitam, cuius finitationem trahunt ab adjunctis. Non enim significant aliquam rem, sed
significant per modum dispositionis rei et terminorum significantium res. Dispositio autem
omnis est in se infinita et finitatur per illud quod disponit' (ed. Braakhuis 1978, 1.7-2.2).
13. On this development see especially L. M. de Rijk's indispensable Logica Modernorum (De Rijk
1962-7).
14. An example from the anonymous Ars Emmerana: 'It must also be noted that there are sophistical
terms by means of which the conversion of propositions is blocked. Such terms are "alone",
"only", "besides", "except". Thus this proposition does not convert: "Something is one alone**;
"Something is only an animal"; "Something besides Socrates is a man"' (although without
'alone', 'only', and 'besides Socrates' each of those particular affirmative propositions would
convert). 'Hoc etiam notandum est quod sunt sophistici termini quibus propositionum impeditur
conversio. Tales sunt: "solum", "tantum", "praeter", "nisi". Unde haec propositio non con-
vertitur: "aliquid est unum solum", "aliquid est tantum animal", "aliquidpraeter Socratem est homo" '
(ed. De Rijk 1962-7,11(2), 157.30-158.2).
15. An example from the anonymous Tractatus de univocatione Monacensis: 'Again, there are people
who say that when the word "besides" or the word "other" are placed between the name of an
existent thing and the name of a non-existent thing, the reference of the name is expanded, even
though a present-tense verb is used in (the proposition). Thus they say that these are true and
grammatically correct: "Something besides Antichrist is not other than Caesar." It is better,
however, that these be called worthless, because they say falsely that something is not, and every
assertion of what is false is worthless.' 'Item sunt qui dicunt quod quando hec dictio "praeter" sive
hec dictio "aliud" ponuntur inter nomen rei existentis e t . . . nomen rei non existentis, nominis
appellatio ampliatur, licet in ea ponatur verbum presentis temporis. Unde dicunt has esse veras et
congruas: "aliquid prêter Antichristum non est aliud quam Cesar". Melius tamen est ut hec dicantur
nugatorie, quia ^dicunt") aliquid non esse quod falsum est, et omnis positio falsi nugatoria est' (ed.
De Rijk 1962-7, 11(2), 340.23-30).
Presumably as a result of being noticed in such connections, the syn-

categoremata emerged as occasions for refinements and extensions of
logical or semantic theory, the role in which they naturally became and
remained important for medieval logicians.
With all the warnings and disclaimers appropriate to historical generalis-
ations of vast scope and meagre detail, I suggest that the career of the
syncategoremata within the logica moderna falls into three stages, the third
of which is divided into two contemporaneous lines of development:
[1 ] their emergence as the focal points of certain logical or semantic relationships
or special problems of interpretation (in the twelfth century, especially the
latter half);
[2] their identification as a distinguishable set of topics worthy of development in
separate treatises called, typically, Syncategoremata (from the last quarter of the
twelfth century to the last quarter of the thirteenth);
[3a] their assimilation into general treatises on logic, sometimes as a group, but
sometimes dispersed in ways designed to associate particular syncategoremata
with more general topics in logic to which they are appropriate; and
[3b] their absorption into the sophisma-literature, where a particular syncate-
gorema may serve as the germ of a paradox the interest of which is often
associated with metaphysics or natural philosophy more than with logic or
semantics proper (from the first quarter of the fourteenth century to the
disintegration of scholastic logic).16
It might be an exaggeration to say that concern with the syncategoremata
constituted the stimulus for the development of medieval techniques of
linguistic analysis (of which expositio is the most important) · 7 and the basis
for the development of the sophisma-literature, but there can be no doubt
16. These historical stages may be exemplified by texts drawn from the list of relevant texts in
Appendix I as follows. Stage [l]: Texts (2/3), (3/2), (7); Stage [2]: Texts (5), (9), (21); Stage [3a]:
Texts (22), (23), (29); Stage (3b): Texts (24), (25/26), (28). For an earlier development in the
sophisma-literature, much less closely associated with the syncategoremata, sec Pinborg 1975a,
PP· 44-5-
17. The Tractatus exponibilium formerly attributed to Peter of Spain but certainly later (see De Rijk
1972, pp. LIV-LV) defines an exponible proposition in terms of syncategoremata: 'An exponible
proposition is a proposition that has an obscure sense requiring exposition in virtue of some
syncategorema occurring either explicitly or included within some word.' 'Propositio ex-
ponibilis est propositio habens obscurum sensum expositione indigentem propter aliquod syn-
categorema explicite positum vel in aliqua dictione inclusum' (Pseudo-Petrus 1489, f, jS"1').
Perhaps the systematic classification and organized explicit discussion of exponible propositions
(exponibilia) in treatises devoted to them as such (which seem to have begun around the middle of
the fourteenth century and to have flourished in the fifteenth and sixteenth centuries) should be
considered a fourth stage in the career of the syncategoremata. See Ashworth 1973. Professor
Ashworth's article has more than a general relevance to the topics of this investigation: Section
Three (pp. 153-9) 's devoted to exceptive propositions (which she inadvertently calls 'exclusive
propositions' several times in the Section) and thus constitutes a historical extension of the study
of medieval treatments of exception that will be presented in this discussion.
2i6 The high middle ages: semantic theory
that both those developments owed a great deal of their form and content
to problems associated with syncategoremata and to techniques developed
first in order to cope with those problems. Without too much distortion it
can be said that the inclusion of at least one syncategorematic word is the
defining characteristic of an exponible proposition, and that the vast
majority of sophismata focus on exponible propositions.18
The strategy of this investigation

At the present stage of the scholarly investigation of syncategoremata, any
study such as this that attempts to present the topic in a strictly limited
format can only scratch the surface in one way or another.19 Tracing the
treatment of one or two of the standard syncategoremata through many
texts20 reveals historical connections and doctrinal developments in more
detail than would be accessible in a broad encyclopedic survey of com-
parable length. The remainder of this discussion is therefore based on a
fairly detailed examination of two of the most frequently discussed syncate-
gorematic words, 'praeter ('but', 'besides', 'except') and 'totus' ('whole'),
representing exception and distribution, respectively.21
18. O n the development of the techniques of linguistic analysis, see Pinborg 1972. O n the develop-
ment of the sophisma-literature, see Grabmann 1940.
19. Research in this area will be stimulated and aided by the publication of H. A. G. Braakhuis'
monumental study of the treatment of the syncategoremata in the thirteenth century, ac-
companied by his editions of the central texts. Braakhuis 1979, the only portion of his work
published to date, has appeared in two volumes: Dcel I, Inleidende Studie (Hoofdstuk I, De
ontwikkeling van de théorie van de syncategoremata tot aan de ijde eeuw; Hoofdstuk II, De lyde eeuwse
syncategoremata-tractaten; Hoofdstuk III, Enkele aspecten van de ontwikkeling van de théorie van de
syncategoremata in de ijde eeuw); Deel II, Uitgave van Nicolaas van Parijs' Sincategoreumata.
Eventually Dr Braakhuis plans to publish his historical and critical studies in English. I am grateful
to Dr Braakhuis for his extraordinary generosity in allowing me to make use of every thing he had
completed as of November 1978 (Braakhuis 1978 and 1978a): Deel I (Hoofdstuk I and parts of
Hoofdstuk II), Deel II (preliminary version), and preliminary editions of Robert Bacon (?) and
Henry of Ghent, with excerpts from John le Page and Peter of Spain. This study could never have
been attempted without his help. Braakhuis 1979 appeared too late for me to take full account of
it, but my references to Nicholas o f Paris are to Braakhuis 1979, Deel II.
20. In one way or another I have made use o f thirty-three texts, dating from the first half of the
twelfth century to the first half of the fifteenth. The texts are listed in Appendix I attached to this
discussion. Some o f those texts are not treatises independent of one another, and in a few cases the
relationships among them are not yet dear. The confusion surrounding Peter of Spain's treatise
on syncategoremata is particularly great, partly because of the misleading impression made by
J. P. Mullally's translation (Mullally 1964), the only form in which this work has been available to
non-specialists, and partly because 'the text of the Syncategoremata as presented in the Cologne
incunabular e d i t i o n s . . . is quite different from Peter's o w n text' (Braakhuis 1977, p. 122, n. 33).
(Mullally's translation is based on the Cologne editions. The relationship of the text of the
Cologne editions to the genuine text of Peter's Syncategoremata may be determinable after the
publication of Braakhuis' edition based on the manuscripts. Meanwhile it is safe to say that
neither Mullally's translation nor the Cologne editions o f 1489 and 1494 can be considered
reliable witnesses to Peter's views on syncategoremata. See n. 48 below.)
21. Other syncategorematic words are likely to seem better choices for such single-track explor-
Standard elements in the treatment of a syncategorema

For all the variation in the ways in which different authors treated a single
syncategorema, there is a marked tendency to organise the discussion
around certain common expository and critical elements which thus pro-
vide the most promising lines along which to pursue an investigation of
this sort. In tracing the treatments of 'praeter' and 'totus' it is useful to sort
the material in terms of these seven elements:
I. Definitions and analyses (of exception or of the sort of distribution effected

by 'totus'),
II. Classifications of the uses of the words ('praeter' and 'totus'),
III. Grammatical, semantic, and logical rules governing propositions (involving
'praeter' or 'totus'),
IV. Examples (associated with II or III),
V. Sophismata (involving 'praeter' or 'totus'),
VI. Expositions of propositions (involving 'praeter' or 'totus'),
VII. Questions (arising from I-VI). 22
Focusing on any one of these elements through all the texts that contain it is
instructive, 23 but the fullest development of the most interesting material is
to be found in the sophismata. 24 It is there that the rules are applied in order
to resolve apparent paradoxes or tested by being confronted with apparent
counter-instances. And because different medieval logicians tend to take up
the same logical or semantic issues in connection with the same sophismata,
their discussions of those puzzles can provide a rich, historically continuous
arions. From the standpoint of the history of logic, 'omnis' ('every') or 'si' ('if), for instance, look
more promising; they are however, likely to receive more elaborate treatments than 'praeter' or
'loins', which makes them less manageable for present purposes, and just because of their wider
logical interest they are less distinctive of the medieval literature on syncategoremata. Similarly,
the pairs 'necessario'/'contingenter' ('necessarily'/'contingently') and 'incipit'l'desinit' ('begins'/
'ceases') have many more points of connection with other philosophical inquiries than do 'praeter'
and 'lotus', but for just that reason they would give rise to issues that would be distracting in this
context. Obviously there is a great deal of interesting work to be done in this area.
22. For samples of all these elements drawn from various texts, see Appendix 11 attached to this
discussion.
23. It would be natural to suppose that the most philosophically interesting material would be found
in VII, the questions; but VII is the rarest of these elements in the literature I have examined.
Although any author may occasionally organise some of his discussion of a syncategorema in the
form of a scholastic question, I have found only Nicholas of Paris making regular and extensive
use of this form of exposition and criticism in connection with 'praeter' and 'totus'.
24. On sophismata generally, see Kretzmann 1977, esp. p. 6: '[A sophisma is J a sentence puzzling in its
own right or on the basis of a certain assumption, designed to bring some abstract issue into
sharper focus - the medieval ancestor of "The morning star is the evening star" or "George IV
wished to know whether Scott was the author of Waverley" '; also n. 9: 'Because sophismata are
sentences rather than arguments and intended to be illuminating and instructive rather than
specious and misleading, it is misleading to call them sophisms', as many writers on this material
still do.
line of doctrinal development or dispute. Accordingly, a consideration of a

few of the relevant sophismata is perhaps the most efficient means of
acquiring an initial understanding of philosophical and historical aspects of
the medieval treatment of the syncategoremata.
A typical analysis of exception
I will present two sophismata involving 'praeter', which it will be helpful to

view against the background of this fairly typical analysis of exception: 'It
must be noted that four things are required for exception: [i] the part that is
excepted, [2] the whole from which the exception is made, [3] something in
respect of which the exception is made, and [4] the act of excepting, which
is conveyed by means of the exceptive word as by an instrument.' 2 5 Thus
in the standard example 'Every man besides Socrates is running', Socrates is
[1], all men constitute [2], the act of running is [3], and 'besides' ('praeter') is
the instrument of the act of excepting.
The sophisma 'Socrates bis videt omnem hominem praeter Platonem'
The first of these two sophismata is treated in nine of the texts listed in
Appendix I, and I have had access to the full text of eight of those
treatments. 26 The interesting differences among those treatments occur
not in the presentation of the problem but in its resolution, and so I will
present the problem in a homogenised version of the sophisma sentence,
the hypothesis, the proof, and the disproof before considering the various
resolutions.
SOCRATES TWICE SEES EVERY MAN BESIDES PLATO 27
Suppose that on one occasion Socrates sees every man and that on another occasion
he sees every man other than Plato and does not see Plato.28
25. Walter Burley (ed. Boehner 1955, 165.16—22): 'Notandum, quod ad exceptionem requiruntur
quatuor: Pars quae excipitur; totum, a quo fit exceptio; et aliquid, respectu cuius fiat exceptio; et
actus excipiendi, qui importatur per dictionem exceptivam sicut per instrumentum. Verbi gratia
sic dicendo: "Omnis homo praeter Sortem currit". Sortes est pars, quae excipitur, omnis homo
est totum, a quo fit exceptio, currere est illud, respectu cuius fit exceptio, "praeter" est dictio
exceptiva.' (I have madesome changes in Boehner's punctuation.) See Appendix II, I, Definitions
and Analyses, for other examples of such analyses of exception.
26. It is found in texts (7), (9), (11 /10), * ( i 3 ) , (15), ( i 8 ) , * ( 2 o / i 9 ) , (25/26), and (33/32).
27. This present-tense form is found in texts (7), * ( i 3 ) , (18), and * (20/19); texts (9), (1 i / i o ) , and (15)
have it in the future tense; (25/26) and (33/32) in the perfect.
28. Text (7) has no hypothesis — indeed, nothing but the sophisma sentence and the resolution (see n.
31 below). Text (11/10) also lacks a hypothesis, but in rather different circumstances, as the
Proof: The proposition 'Socrates twice sees every man' is false, and Plato is the
only counter-instance; therefore when Plato is excepted, the resultant proposition
is true. 29
Disproof: Socrates twice sees every man besides Plato; therefore on one occasion
he sees every man besides Plato, and on another occasion he sees every man besides
20
Plato - which is false ex hypothesi.
With one possible exception to be noted later, each of the eight texts
under consideration offers the same basic appraisal of the sophisma sen
tence: it is ambiguous, being true in one of its senses and false in the other.
But there are significant differences among the ways in which the am
biguity is characterised and resolved.
In Dialectica Monacensis, where the sophisma is introduced in connection
with the fallacy of division, the entire resolution consists in this one, highly
compressed observation: 'For it can be judged on the basis of " t w i c e " or on
the basis of "besides".' 3 1 The point ofthat observation and the theory that
may be implicit in it can be understood most easily in the light of the fuller
treatments to be discussed below.
Bacon clearly sees the sophisma as presenting a scope-ambiguity: is the
logical operation of distribution (over two occasions) performed by 'twice'
included by or inclusive of the logical operation of exception performed by
'besides'? That is, is the sophisma sentence to be read in this way '(Twice
Socrates sees every every man) but Plato' or in this way 'Twice (Socrates
sees every man but Plato)'? In the first way it is true; in the second, false.
Bacon thinks that the ambiguity might also be expressed in terms of a
choice between predicates in respect of which the exception is made -
either 'twice sees' or 'sees' - but he prefers the resolution in terms of the
inclusion of one operation by another as more generally applicable. 3 2
discussion o f it will show. In the other seven texts the hypothesis is virtually the same, but for
purposes of tracing historical associations it may be worth noting that in texts (9), (15), and
* (20/19) the situation is presented in terms of two times, designated Ά ' and 'B', while in texts
* ( ' 3 ) , (18), (25/26), and (33/32) the presentation is in terms o f ' o n e occasion' (una vice) and
another, with no special designation for either occasion.
29. Text (7) has no proof, and (9) has merely 'Probatio, etc.'. I have not seen anything beyond the
hypothesis in * (20/19), and so it will not figure in the rest of this discussion. In the other six texts
the proof is essentially the one presented here.
30. Text (7) has no disproof. In the other seven texts the disproof is essentially the one presented here,
but it is given most succinctly in (9) and most elaborately in (25/26).
31. 'Item. Secundum has fallacias accidit multiplicitas in omni oratione in qua sunt duo sincate-
goreumata sic sehabentia quod locutio potest iudicari penes unum illorum vel penes reliquum . . .
Eodem modo solvitur hoc sophisma: "5or bis videt omnem hominem prêter Platonem." Potest enim
iudicari per "bis" vel per "prêter"' (ed. De Rijk 1962-7,11(2), 572.38—573.2; 573.7—8).
32. 'Similiter contingit aptare omne sophisma quod contingit in exceptivis ex includere vel includi...
Solutio: prima duplex est ex eo quod potest fieri exceptio respectu eius quod est "videre bis" vel
The author of the Abstractiones33 takes a position that is at least super-

ficially quite different from the positions taken by the other seven authors.
Having provided no hypothesis at all,34 he appraises the sophisma sentence
as 'unconditionally true' (vera simpliciter). But he shows that he recognises
its ambiguity when he rejects the disproof because it requires one to derive
from the hypothesis the premiss that on neither occasion does Socrates see
Plato, and 'we know that from a proposition that can have more than one
cause of its truth we cannot infer one of those causes'.35 In the absence of a
hypothesis stipulating one cause of the truth of the sophisma sentence the
author cannot say in which sense the sentence is true, and his observation
regarding the invalid inference implicit in the disproof depends on his
preserving the ambiguity of the sophisma sentence - an intelligent resolu-
tion but one that suggests a lack of interest or of sophistication regarding
the peculiar properties of the syncategoremata.36
Peter of Spain locates the ambiguity in the possibility of assigning two
different scopes (or determinations) to 'twice': either it determines the
entire expression 'sees every man besides Plato' - in which case the
sophisma sentence is false — or it determines only 'sees every man' — in
which case the sophisma sentence is true. But Peter presents this basically
respectu eius quod est "videre". Primo modo vera, quia tunc significat vere (?) quod Sortes non
videbit Platonem bis, sed omnes alios; secundo modo falsa; non enim erit vera hec bis "Sortes
videbit ortmem hominem prêter Pktonem", sed tantum in B. Hoc autem nichil aliud est quam quod
exceptio potest includere distributionem de li "iiiî" vel econverso. Si enim includat: excipit
respectu eius quod est "videre bis"; si includatur: excipit ab eo quod est "videre". Et multiplicand
ilia exceptio per distributionem deli "bis"' (ed. Braakhuis 1978,171.12-13; 171.18-172.2). Ihave
made one significant change in Braakhuis* punctuation. O n the uncertain authorship and
relative dating of this treatise, see n. 41 below.
33. The author is identified in MSS Digby 24 and Bruges 497 as a master Richard, sometimes referred
to as 'Ricardus sophista'. De Rijk has suggested that he may be Richard Fishacre, a student of
Robert Bacon's, w h o died in 1248, when Bacon also died (De Rijk 1962-7,11(1), pp. 71-2). More
recently Pinborg has suggested Richard Rufus o f Cornwall, a slightly younger contemporary of
Richard Fishacre's (Pinborg 1976c). The editors o f the forthcoming edition of the Abstractiones
(Calvin Normore, Mary Sirridge, Paul Streveler, and Katherine Tachau) have so far not
committed themselves on the question of authorship. I am grateful to those editors, and to Tachau
in particular, for allowing me to see rough drafts of their transcriptions.
34. The text is obviously defective at this point (f. 83"" b in MS Digby 24, transcribed Tachau et al.),
and it is conceivable that the hypothesis has simply dropped out; but the structure of the resolution
leads me to think that the author deliberately omitted any hypothesis.
35. 'Et scimus quod ex propositione quae potest habere plures causas veritatis non sequitur una
illarum' (MS Digby 24, f. 83*; transcribed Tachau et al.).
36. The treatment o f this sophisma and the nature of the Abstractiones generally suggest, however
weakly, that the treatise is somewhat anachronistic if it is indeed to be located chronologically
between the treatise attributed to Robert Bacon and Peter of Spain's Syncategoremata; it seems
earlier than either of those. I think that the source o f the apparent anachronism is more likely to be
in the proposed dating (and authorship) of the Bacon treatise, however, about which I will say
more in n. 41 below.
simple, sensible analysis in a way that suggests that it was still inchoate
when he wrote. His exposition of it is labored and unduly complex, he
confuses that in respect of which the exception is made with that from
which it is made,37 and he introduces the peculiar doctrine of different
arrival times of expressions in a proposition in order to explain the two
possible determinations ascribable to 'twice': 38 in the formation of the first
sense 'besides' arrives before (prius advenit) 'twice'; in the formation of the
second (true) sense that order of arrival is reversed.39
William of Sherwood uses this sophisma to conclude his development of
the rule that 'If there is more than one division [i.e., distribution] and an
exception is made from one in respect of another, then the one in respect of
which the exception is made is immobilised'.40 The two distributions are
those effected by 'twice' and by 'every', and since Plato is excepted^»! all
the men in respect ofthe two occasions, the distribution effected by 'twice' is
'immobilised'; that is, the inference from 'twice' to 'on the one occasion...
and on the other occasion ...' is blocked. In presenting his resolution,
however, William leaves the application of the rule implicit and sorts out
the ambiguity as we have seen it done in the treatise attributed to Robert
Bacon,41 in terms of the one operation's including or being included by the
37. For Peter's confusion, see n. 39 below. This particular confusion is not uncommon among writers
on the syncategoremata. however.
38. William of Sherwood also uses this doctrine in connection with his discussion of exposition,
although not in his discussion of this sophisma. See Kretzmann 1968, p. 62, p. 62 n. 22, pp. 66-7,
p. 67 n. 34; also Gcach 1962, pp. 102-4.
39. 'Solutio. Prima est duplex, eoquod hec dictio "bis" potest determinare hoc verbum "videre" prout
transit supra accusativum sequentcm cum sua dcterminationc facta per exceptioncm. Et tunc est
sensus "Sortes v'tdet omnem hominem prêter Platonem et hoc bis": et tunc sequitur quod nulla vice
vident Platonem, et sic est falsa. Et sic haec dictio "prêter" excipit ab hoc quod est "videre omnem
hominem", et sic prius advenit hec dictio "prêter" in oratione. Alio autem modo hec dictio "bis"
potest determinare tantum hoc quod est "videre omnem hominem". Et tunc hec dictio "prêter"
excipit ab eo quod est "bis videre omnem hominem". Et sic est vera, quia non bis videt Platonem. Et
sic hec dictio "his" prius intelligitur advenire in oratione quam hec dictio "prêter" (ed. in
Braakhuis 1978, pp. 162-3). (I have made some changes in Braakhuis' punctuation.)
40. 'Si fuerint plures divisiones et excipiatur ab una et respectu alterius, etiam ilia respectu cujus fit
exceptio îmmobilitatur' (ed. O'Donnell 1941, 62.38-40; cf. Kretzmann 1968, p. 66).
41. On grounds that deserve careful consideration of a sort they cannot be given here, Braakhuis
argues that it is Robert Bacon to whom we ought to attribute the Syncategoremata that has been
attributed to Roger Bacon, and that it should be dated as early as thefirstdecade of the thirteenth
century. Robert certainly wrote earlier than William of Sherwood: Roger was William's
younger contemporary and outspoken admirer (see Kretzmann 1966, pp. 5-7). Everything I
have seen in the Bacon treatise and in the others I have consulted leads me to think that it belongs
after William of Sherwood and Peter of Spain rather than before them or, even harder to credit,
before the Abstractiones. Its advanced level of sophistication, its casual treatment of proofs and
disproofs, and its casual treatment ofother resolutions are strongly suggestive of the middle rather
than thefirstdecades of the thirteenth century; and in this particular instance Bacon's use of the
letter-designations in che hypothesis, his casual introduction of inclusion as the leading idea in his
other. But he adds the observation that the familiar distinction between the
compounded and divided senses is involved in the resolution in terms of
inclusion. 42
Nicholas of Paris, having put forward his rule that 'An argument from
an inferior to its superior with an exceptive word is unacceptable', 43 takes
up two apparent counter-instances. It is easy to see how the first of them
serves his purpose, 44 but the second counter-instance is supposed to be this
sophisma, and I have not seen any good reason for his treating it in that
way. 4 3 Nicholas' resolution of the sophisma consists first in an observation
strikingly like the one made in Dialectica Monacensis: the proposition 'can
be judged on the basis of the numerical adverb', in which case it is false. He
notes, on the other hand, that the disproof may be considered unacceptable
because of treating 'Socrates twice sees' as if it were compounded rather
than divided. O n this particular sophisma Nicholas is less good than his
English predecessors. 46
Albert of Saxony, after quickly sorting out the ambiguity, focuses on the
semantic issue in the sophisma: 'But which of those senses is more proper?'
His analysis of the issue makes intelligent use of the concepts of the
resolution, and his restatement of the type of resolution employed by Peter of Spain in his own
preferred terms of inclusion could even be taken to suggest that the author was aware of William's
work. Braakhuis' formidable considerations Co the contrary notwithstanding, I am still inclined
to think that this Bacon is Roger, not Robert, or at any ra te that the treatise was written later than
1200-10.
42. 'Et dicendum quod exceptio potest indudere ly bis vel e converse Si ly bis includat, significat
quod "videre omnem hominem praeter Platonem" conveniat Sorti bis et falsa est; si ly bis
includatur, significat quod "videre o m n e m alium" conveniat Sorti bis; sed "videre Platonem"
non, et est vera et sic probatur. Primo modo debet hoc totum "videre omnem hominem praeter
Platonem" componi, ut significetur quod hoc totum multiplicatur per ly bis; secundo modo
debet ly praeter dividi a residuo, ut significetur quod ly bis non extendit se ad ly praeter, sed e
converso' (ed. O'Donnell 1941, 63.33-40; cf. Kretzmann 1968, pp. 6 8 - 9 ) .
43. 'Regula est quod non valet argumentatio ab inferiori ad superius cum dictione excepliva' (ed. Braakhuis
1979, «, 167.5-7)·
44. 'Sed quod hoc sit falsum videtur, quia sequitur: "omnis homo prêter Sortem currit; ergo omnis
homo prêter unum currit"; sed unum superius est ad Sortem; ergo' (ed. Braakhuis 1979, II,
167.16-18).
45. The only relevant unacceptable inference is the one that derives 'on the one occasion . . . and on
the other occasion' from 'twice', and, as is clear in William of Sherwood's treatment of this
sophisma, that inference is a case of logical descent (from what is logically superior to what is
logically inferior).
46. 'Ad illud vero "Sortes bis videt omnem hominem prêter Platonem", dicendum quod est duplex
secundum regulam generalem: quia potest iudicari per adverbium numeri, et sic falsa, et sic
improbat et sensus patet <in> improbatione; vel potest iudicari per dictionem exceptivam, et sic
vera et patet sensus in probatione et est instantia huius "Sortes (.bis) videt omnem". Vel potest dici
quod non sequitur: "Sortes bis videt omnem hominem prêter Platonem; ergo una vice videt
omnem hominem prêter Platonem et alia", quia fit exceptio ab hoc toto "Sortes bis videt"
coniunctim, et arguitur ac si fieret divisim, unde est ibi fallacia consequentis" (ed. Braakhuis 1979,
Π, 168.8-17).
exposition of propositions and the supposition of terms: 'Now in "Socrates

did not see Plato twice" the whole predicate ["did see Plato twice"] is
denied of the whole subject; but in "Socrates twice did not see Plato" not
the whole predicate but a part of it is denied of the subject, because the term
"twice" remains affirmed.'47
The Pseudo-Petrus48 uses the sophisma to illustrate the widely recog-
nised rule that 'If a proposition is in part false, it can be made true by means
of an exception, but not if it is completely false.'49 The proposition
'Socrates twice saw every man' is in part false because it is false only as
regards Plato on one occasion, not completely false as it would be if
Socrates had never seen any man. The Pseudo-Petrus is like Peter himself in
developing the resolution in terms of different determinations, but here
'twice' is said to determine 'besides' in the false sense of the proposition,
47. 'Ad sophisma rcspondetur quod in ipso exceptio potest denotare vel quod Socrates non vidit bis
Platonem — et sic sophisma est verum — vel quod Socrates bis non vidit Platonem — et sic est
falsum. Sed quis istorum sensuum sit magis proprius? Dico quod primus videtur esse magis
proprius quam secundus, ex eo quod dictae propositionis exceptivae expositio est magis propria
secundum quam totum praedicatum affirmatur de subiecto toto quantum ad supposita non
excepta et negatur totum de toto quantum ad suppositum exceptum quam in qua non totum sed
pars negatur de toto. Modo in ista "Socrates non vidit bis Platonem" negatur totum praedicatum
de toto subiecto; sed in ista "Socrates bis non vidit Platonem" negatur non totum praedicatum sed
pars eius de subiecto, quia iste terminus "bis" remanet affirmatus' (Albert of Saxony 1502, Part II,
Sophisma Ixvii (actually Ixv], f. 57").
48. I am using this designation to refer to the author or authors of the presumably fifteenth-century
material included in the Cologne 1489 edition of the logical works of Peter of Spain in such a way
that it has been or can be confused with the genuine text of Peter's Syncategoremata. The Cologne
Dominicans brought out their edition of Peter's logical works 'cum copulatis secundum doctrinam
divi Thomae Aquinatis, iuxta processum magistrorum Coloniae in bursa Montis regentium' (from the
title-page of the 1489 edition). In the 1489 edition the text of what purports to be Peter's treatise
on the syncategoremata (there entitled Tractatus syncategorematum) is printed in larger type; that
text has been designated (14) in Appendix I attaching to this study. In the same edition smaller
type is regularly used for the commentary on Peter's text. The material in smaller type printed
with the Tractatus syncategorematum is not a commentary, but it is explicitly distinguished from the
text putforwardas Peter's; for instance, at the end of the material in smaller type concerned with
exception and before the beginning of the next section of larger type the words 'Sequitur texlus'
appear (f. 46vb). This materia] contains sophismata that occur also in Peter's genuine Syncate-
goremata, but they are treated differently here. The resolution of this tangled textual relationship
must await the publication of Braakhuis' edition. Meanwhile I am considering the smaller-type
material accompanying the Tractatus syncategorematum as a fifteenth-century supplement to
Peter's treatment of syncategoremata; this material is designated as text (33/32) in Appendix I.
The presentation of this sophisma through the disproof is very much the same in texts * ( 13) and
(isls1) - it does not appear at all in (14) - but the resolutions are too divergent to be considered
merely two versions of the same text (compare nn. 39 and 50), and the divergence between
resolutions is even more marked in other sophismata, as we shall see. Moreover, the contexts in
which the same sophismata are introduced are quite different in texts * (13) and (33/32). (See n. 20
above.)
49. 'Si enim aliqua propositio exceptiva [!] est in parte falsa, potest per execptionem fieri vera, non
autem si fuerit in tota falsa, ut prius visum est' (Pseudo-Petrus 1489. f- 46*). Versions of this rule
are found also in texts (15), (18), (21), (23), (25/26), and (26/25).
while 'besides' is said to determine 'twice' in the true sense 'in such a way
that the adverb "twice" determines the verb itself and not everything
following it'. There is nothing novel in this resolution except, perhaps, for
the characterisations of the two senses as 'a duality of exception . . . [and] an
50
exception from duality'.
The sophisma 'Omnis homo praeter Socratem excipitur'

Of the many explicit rules governing exceptives, 5 1 the one perhaps most
widely recognised was the rule that the universal proposition on which a
true exceptive proposition is founded — the 'prejacent' of the exceptive — is
false without the exception. 5 2 The second sophisma I will present to
illustrate the treatment of exception was designed primarily to challenge
that fundamental rule and is, accordingly, more important and more
complicated than the previous example. This sophisma is found in nine of
the texts listed in Appendix I, two of which are by a single author. 5 3 I will
again present the problem portion of the sophisma in a homogenised (and
in this case modernised) version.
(S) EVERY MAN BESIDES SOCRATES IS EXCEPTED
Suppose that every man other than Socrates is excepted from the provisions of
some law - i.e., every man is subject to this law except for those whose names are
other than 'Socrates'.5*
Proof ι : The universal proposition 'Every man is excepted' is false, and the only
counter-instance to it is Socrates; therefore (S) is true. 55
50. 'Solutio. Prima est duplex co quod haec determinatio "bis" potest determinare hanc dictionem
"praeter" et praesupponere earn. Et sic est falsa, quia sensus est "Socrates vidit omnem hominem
praeter Platonem ct hoc bis"; ct tunc sequitur quod nulla vice viderit Platonem. Alio modo ilia
determinatio "praeter" potest determinare illud adverbium "bis" ita quod illud adverbium "bis"
determinct verbum secundum se et non totum sequens. Et sic est vera, quia sensus est quod
Socrates bis vidit ( o m n e m ) hominem alium a Platone sed non bis Platonem. Ideo communiter
fit exceptio pro Platone. Primo modo ponitur dualitas exceptionis; secundo modo ponitur
exceptio a dualitatc' (Pseudo-Petrus 1489, ff. 4 6 r b " , a ) .
51. I have picked out more than fifty in the texts I have consulted.
52. Some version of this rule is found in texts (5), (9), (15). (18), (21), (22), (23), (25/26), (26/25), (29),
and (33/32).
53. The sophisma is found in texts (11/10), *(13), (18), (21), (23), (25/26), (26/25), (27), and (33/32).
54. Text (18) mentions four men, texts (25/26) and (26/25) mention three men and provide them
with names, and all the other texts refer to all men. In (11/10), (18), (23), (25/26), and (26/25) me
exception is said t o b e in respect of (or, mistakenly, from) some action; in (23) Walter Burley uses
the (unfortunate) example o f the act of being; in both (25/26) and (26/25) Albert of Saxony uses
running. In * ( i 3 ) and (33/32) the exception is in respect of some (unspecified) predicate, and in
(21) it is simply 'from something'. M y example of a law is simply intended to render the notion of
being excepted less unnatural in English.
55. Some version of this first proof is found in seven of the nine texts under consideration; notice that
it is structurallyjust like the proof in the previously considered sophisma. Neither (11/10) nor (27)
Proof 2: The correct exposition of a proposition of the form 'Every χ besides A

is F ' is (i) 'There is at least and at most one χ such that χ = A' and (ii) 'Every χ such
that χ φ A is F' and (iii) Ά is not F'. The instances of the exponents (i), (ii), and
(iii) are all true in this case ex hypothesi; therefore (S) is true. 5 6
Disproof 1: If Socrates is running and all the other men are running, then
'Every man besides Socrates is running* is false. In that case 'Every man is running'
is true, and a universal proposition that is true without an exception is false with
one. Analogously, (S) is false if Socrates is excepted and all the other men are
excepted. But Socrates is excepted in (S). Therefore 'Every man is excepted' is true;
therefore (S) is false.57
Disproof 2: If (S) is true, then Socrates is not excepted. But Socrates is excepted
in (S). Therefore if (S) is true, (S) is false.58
The resolutions can be understood best as attempts to deal with the

apparent paradox (brought out most clearly in Disproof 2) that the excep
tive proposition 'Every man besides Socrates is excepted' and its prejacent
'Every man is excepted' are true together. Seven of the eight authors under
consideration are out to dispel the paradox and to accept the exceptive (S)
as true; only Albert of Saxony thinks (S) is 'false and impossible'. Almost all
the resolutions offered or mentioned fall into one or another of three
types. 5 9
The first and most important type of resolution seems to grow out of a
fundamental observation to be found in the Abstractiones: to be excepted in
respect of being excepted is not to be excepted at all; and so even though
Socrates is excepted in that respect, he is not excepted; therefore it is not
has this proof explicitly, but those two texts differ from the others in their non-standard
presentations of the sophisma, and the omission has no discernible historical or doctrinal
significance.
56. Proof 2 is a modernised version of the proof from the exponents offered in texts (23), (25/26), and
(26/25); a proof from the standard exponents is expressly rejected by Paul of Venice in (27), as we
shall see. Walter Burley presents it schematically: 'the exponents are true; therefore the exceptive
is true'; Albert of Saxony spells it out a little more fully: 'Socrates is nor excepted from the act of
running, and every man other than Socrates is excepted from the act of running; therefore every
man other than Socrates is excepted from the act of running', omitting the conclusion in (25/26).
Albert also offers a third proof in each of his texts: 'Every man besides Socrates is excepted from
the act of running; therefore every man besides Socrates is excepted', but it is only technical
reasons of his own that prompt him to present what is in any case an obviously trivial proof.
57. Some version of Disproof 1 is found in (11/10) (18), (23), (25/26), and (26/25); die version
presented here is most like the one in (18).
58. Only Albert of Saxony offers Disproof 2 in this form (the versions of the disproofs in (26/25) a r e
more succinct than those in (25/26)). But* (13), (21), and (33/32) offer as their only disproof a very
simple argument that seems intended to take the same line as this Disproof 2. As Peter of Spain
puts it, 'In this proposition the exceptive word is adjoined to "Socrates"; therefore it excepts him
in respect of the predicate. Therefore Socrates is excepted; therefore the first [i.e., (S)] is false.'
59. The unusual resolutions offered by the Pseudo-Petrus and by Paul of Venice will be considered
after the consideration of the three main types.
true in general that if Socrates is excepted in this or that respect, he is

excepted. 60 Albert of Saxony, acknowledging that others base an affirma-
tive reply on this observation, interprets it as an instance of the even more
fundamental observation that an inference from an inferior to its superior
with negation is invalid. Thus just as 'Socrates is not running; therefore
Socrates is not moving' is invalid, so 'Every man besides Socrates is
excepted in respect of some action; therefore every man besides Socrates is
excepted' is invalid; and so (S) is not supported by the hypothesis, for 'is
excepted' includes negation. 61 But the more usual development of the
observation regarding exception in respect of being excepted lies in the
direction of accepting (S) and rejecting the negative side of the sophisma as
an instance of the fallacy of secundum quid et simpliciter. Peter of Spain
provides a full version of this type of resolution: 'The first [(S)] is true
unconditionally (simpliciter), and the disproof commits the fallacy of secun-
dum quid et simpliciter; for to except from exception is not to except
unconditionally but in a certain respect, just as to be deprived of a privation
is not to be deprived unconditionally but in a certain respect. Indeed,
it is rather to give possession, as to deprive of blindness is to give
60. "... Socrates non excipitur, quamvis excipiatur respectu eius quod est exceptum, quia excipi
respectu istius non est <esse) exceptum . . . Et non valet "Socrates excipitur in hac; igitur
excipitur", sed sequitur eius oppositum' (MS Digby 24, f. 83"'; transcribed Tachau ef al).
61. 'Ad sophisma respondctur quod ipsum est falsum ct impossibile, sicut probatum est. Ad rationes.
Ad primam, quando dicitur "Omnis homo praeter Socratem [ab actu currendi excipitur; ergo
omnis homo praeter Socratem excipitur]", negatur consequentia propter negationem inclusam in
hoc verbo "excipitur". Unde sicut non sequitur ab inferiori ad superius cum negatione, ita nee
cum termino includente negationem. Et ideo non sequitur "Hoc excipitur a currere; ergo
excipitur ab agere"; nee sequitur "Omnis homo praeter Socratem ab agere excipitur; ergo
omnis homo praeter Socratem excipitur." Ex hoc patet ad alias duas rationes sequcntes. / Aliter
alii dicunt ad sophisma quod ipsum est verum. Et Socrates non excipitur, quia excipi ab
excçptione non est excipi; modo in proposito Socrates excipitur ab exceptione* (Albert of Saxony
1502, f. 56*). The version found in Albert's Perutilis logica differs in some interesting details:
'Respondctut quod sophisma est falsum et impossibile, sicut patet per improbationem. / Sed
contra sicbreviter. Ista propositio sit A: "Omnis homo praeter Platonem et Ciceronem currit."
[He is supposing that Socrates, Plato, and Cicero constitute all men.) Tunc sic: "Omnis homo
praeter Socratem ab actu currendi excipitur; ergo omnis homo praeter Socratem excipitur." /
Breviter respondctur negando conscquentiam; immo, ratione negationis indusae in hoc verbo
"excipitur" est fallacia consequcntis - ab inferiori ad superius negative. Unde non sequitur
"Socrates excipitur a currere; ergo Socrates excipitur ab agere"; sic etiam non sequitur "Omnis
homo excipitur praeter Socratem in B; ergo omnis homo praeter Socratem excipitur." / Ad
rationes dico quod non sequitur "Omnis homo praeter Socratem excipitur ab illo actu - scilicet,
ab actu currendi; ergo omnis homo praeter Socratem excipitur." / Ad secundam dico quod haec
est vera: "Omnis homo excipitur", quia quilibet homo excipitur in sophismate vel in A
propositione. Et propter hoc dico quod exponentes non sunt verae, quia ista est falsa: "Socrates
non excipitur." /Tertiam [!] soluta est per primam solutionem' (Albert o f Saxony 1522, f. 22").
The fact that the differences between Albert's Sophismata and his Perutilis logica in this case and
others leave Perutilis logica looking generally like an amplification or improvement of Sophismata
has led me to think, contrary to my original assumption, that his Sophismata is earlier than his
Perutilis logica; but see n. 94 below.
sight.. ,' 62 Henry of Ghent offers a version very like Peter's as one of his
two alternate resolutions,63 and Walter Burley offers it as his preferred
resolution.64 Nicholas of Paris mentions it, attributing it to others and
oddly reversing the assignments of secundum quid and simpliciter.65 The
Pseudo-Petrus mentions Peter's only resolution merely as one that might be
offered.66
The type of resolution evidently preferred by Nicholas of Paris is based
on a distinction that was frequently drawn between actually effecting
exception and merely signifying it,67 for he introduces this sophisma as
pertaining to the rule that 'an argument from a word signifying the
operation [of exception - e.g., "besides"] to one that signifies the concept
[of exception -e.g., "is excepted"] is not acceptable'.68 Nicholas leaves the
application of the rule tacit; the type of resolution thought to be available
under it is more accessible in the version offered later by Henry of Ghent as
the second of his two alternate resolutions.69
62. 'Solutio. Prima est vera simpliciter, et improbatio peccat secundum quid et simpliciter, quia excipere
ab exceptione non est excipere simpliciter, sed secundum quid, sicut privari a privatione non est
privari simpliciter, sed secundum quid, immo potius est dare habitum, ut privare a cecitate est dare
visum; ergo excipi ab exceptione non est excipi simpliciter, sed secundum quid. Et ideo, licet Sortes
excipiatur respectu huius predicati "excipi ", sicut est in prima, non tamen excipitur; et ideo est ibi
fallacia secundum quid et simpliciter' (ed. Braakhuis 1978a, p. 153).
63. 'Solutio: dicendum quod prima est vera simpliciter. Et inprobatio peccat per fallaciam secundum
quid et simpliciter, quia excipi ab exceptione est excipi secundum quid, sicut privari a privatione,
ut a cecitate, est privari secundum quid' (ed. Braakhuis 1978a, 47.19-22).
64. 'Posset tamen probabiliter dici, quod illo casu supposito haec est falsa: "Omnis homo excipitur."
Et cum dicitur: Sortes excipitur in ista: "Omnis homo praeter Sortem excipitur", dico, quod
Sortem excipi in ista: "Omnis homo praeter Sortem excipitur", est Sortem excipi ab exceptione,
et Sortem excipi ab exceptione est Sortem non excipi, et ideo Sortem excipi respectu excipi est
Sortem non excipi. Et ideo dicitur, quod haec est fallacia secundum quid et simpliciter. quia Sortes
excipitur in ista: "Omnis homo praeter Sortem excipitur", quia Sortem excipi in ista est Sortem
excipi secundum quid, et simpliciter non excipi' (ed. Boehner 1955, 170.13-22).
65. 'Aliter dicunt alii quod hie est fallacia secundum quid el simpliciter, quia excipi ab hac actione est
excipi secundum quid; sed. cum dicitur "Sortes excipitur", est ibi excipi simpliciter; unde
proceditur ab eo quod est secundum quid ad id quod est simpliciter' (ed. Braakhuis 1979, 11,
170.6-9).
66. 'Vel potest dici quod ibi Socrates excipitur ab exceptione, sed excipi ab exceptione solum est
excipi secundum quid et non simpliciter, sicut privari (a> privatione est privari secundum quid et
non simpliciter, etc' (Pseudo-Petrus 1489, f. 46™).
67. See, for example, Bacon's introductory remark that 'praeter' 'significat exceptionem, sed non per
modum conceptus, sicut hoc nomen "exceptio" et hoc verbum "excipio", sed per modum affectus'
(ed. Braakhuis 1978a; 163.18-20); and compare this passage from Abstractiones: 'Et solet dici quod
non valet "Omnis homo excipitur praeter Socratem; ergo omnis homo excipitur, Socrate
excepto" eo (quod) per hanc dictionem "excepto" non exerceretur exceptio sed significatur; per
hanc dictionem "praeter" <autem) exerceretur et non significatur' (MS Digby 24, f. 83*";
transcribed Tachau et al.).
68. 'Regula est quod non valet argumentatio a dictione significante affectum ad earn que significat
conceptum.' And Nicholas goes on at once to exemplify this rule in a way strongly reminiscent of
the Abstractiones: 'Secundum hoc dicitur quod non sequitur: 'omnis homo prêter Sortem currit;
ergo omnis homo Sorte excepto currit"' (ed. Braakhuis 1979,11, 168.19-169.1).
69. 'Vel: prima potest solvi per cquivocationem exceptionis, quod quedam est excrcita, quedam
Three authors—Ricardus sophista, Nicholas of Paris, and Walter Burley —

allude to a type of resolution that none of them prefers. It seems right to
describe this type generally as based on the concept of transcasus even
though only Nicholas uses that technical term in this connection:
' Transcasus is the change of the truth of a statement in accordance with the
change of time. For example, if with my hand closed I say "My hand is
closed" and go on to say "therefore my hand is not open", it is said that this
does not follow in virtue of transcasus.'70 But how is transcasus supposed to
apply in the resolution of this sophisma? Nicholas offers very little help:
'Likewise this does not follow: "Every man besides Socrates is excepted;
therefore Socrates is excepted".' Drawing on hints in the other two
discussions, we can, however, fill in at least the essential details. If all one
knows is that every man other than Socrates and not Socrates is excepted in
respect of some predicate, on that basis alone the prejacent (P) 'Every man
is excepted' seems clearly false. And just because (P) is false on that basis,
one would be led to affirm (S) 'Every man besides Socrates is excepted' and
to prove it as in Proof i. But then, realising that if (S) is true, Socrates, too,
is excepted, one would feel entitled to affirm (P). Thus before (S) is affirmed
and proved, (P) seems false and (S) seems true; but after the affirmation and
proof of (S), (P) seems true and (S) seems false. Ricardus sophista flatly rejects
a resolution of this type: 'But . . . I say that the prejacent is false and the
exceptive true after the proof just as before it', because to be excepted in
respect of being excepted is not to be excepted at all. 71
The resolution based on transcasus suggests that this sophisma might be
significata. Et loquendo dc exceptione significata: sic prima est vera, ut probatur; loquendo autem
de exceptione exercita: sic est falsa, ut inprobatur. Et sic loquendo de eadem exceptione, non sunt
simul vere cum exceptione et sine exceptione' (ed. Braakhuis 1978a; 47.22—48.4).
70. 'Ad hoc sophisma solvunt quidam per transcasum. Est autem transcasus transmutatio veritatis
enuntiationis secundum transmutationem temporis, ut manu mea existente clausa si dicam
"manus mea est clausa", et proferendo hanc "ergo manus mea non est aperta", dicitur quod non
sequitur propter transcasum. Similiter non sequitur: "omnis homo prêter Sortem excipitur, ergo
Sortes excipitur"' (cd. Braakhuis 1979, H, 169.22-170.5).
71. Ricardus sophista introduces this type of resolution as one likely to be familiar to his readers: 'Et
solct dici...'. The portion of the text in which he presents this resolution is obviously corrupt in
the transcription from which I have worked, but his rejoinder is clear enough: 'sed quia Socrates
non excipitur, quamvis excipiatur respectu eius quod est exceptum, quia excipi respectu istius non
est (esse) exceptum, dico quod praeiacens est falsa et exceptiva vera post probationem sicut et
ante' (MS Digby 24, f. 83"»; transcribed Tachau et al.; cf. η. 6ο above). Walter Burley's presentation
of a resolution of this type: 'Solutio huius sophismatis secundum aliquos est. quod ante pro-
lationem huius: "Omnis homo praeter Sortem excipitur" est praeiacens falsa et exceptiva vera, et
post prolationem huius accidit econtrario, et ita non sunt simul verae' (ed. Boehner 1955,
170.9-12). Four of the six MSS on which Boehner based his edition have 'probationem' where his
edition has 'prolationem'.
viewed as presenting a special problem of self-reference. The Pseudo-

Petrus seems to have taken such a view of it, although the resolution as he
offers it is too succinct to be altogether clear: 'The first [i.e., (S)] is true on
the hypothesis, as is evident from the proof. And the reply to the disproofis
that although in "Every man besides Socrates is excepted" Socrates is
excepted in respect of the predicate "is excepted", in another proposition
only those other than Socrates are excepted in respect ofthat predicate and
not Socrates, and it is ofthat [proposition] that [this] proposition is speak-
ing, and not of itself.'72
Paul of Venice's treatment of this sophisma is very different from that of
the other seven authors, but to a large extent the unusual features of his
discussion are a function of the rigid, artificial structure of his Quadratura.13
As one of the fifty principalia of the first of the four dubia that comprise the
treatise, Paul's discussion of the sophisma centres around thatfirstdubium:
whether one and the same inference can be both acceptable and unac-
ceptable at once. Accordingly he presents the sophisma as the conclusion of
an inference regarding which that question might be raised: 'Socrates is not
excepted, and every man other than Socrates is excepted; therefore every
man besides Socrates is excepted.' The acceptability of the inference seems
obvious since its premisses appear to be the standard exponents of the
exceptive proposition that is its conclusion. And yet it is easy to lay down a
hypothesis on which those premisses are true even though the conclusion
appears to entail a contradiction - both that Socrates is not excepted and
that Socrates is excepted.74 After developing the obligatory four con-
clusiones relevant to this issue, in which he raises putative counter-instances
72. 'Solutio. Prima est vera in casu posito, ut patet per probationem. Et ad improbationem dicitur
quod licet in ista "Omnis homo praeter Socratem excipitur" Socrates excipiatur respectu illius
praedicati "excipitur", tamen in alia propositione solum alii a Socrate excipiuntur respectu illius
praedicati et Socrates non; et de ilia loquitur propositio, et non de se ipsa' (Pseudo-Petrus 1489,
f. 46"). Conceivably the reference to 'another proposition' is based on considerations like those
raised in text (26/25); see n. 61.
73. Each of the four dubia has fifty principalia (or chapters), and in the course of developing each of
those principalia he defends four conclusiones. A printed marginal note in the 1522 edition ofAlbert
of Saxony's Perutilis logica alongside Albert's discussion of this sophisma reads: 'Vide de hoc
sophismatc Petrum Mantuam, capitulo proprio, et Paulum Venerum, octavo principal! primi
dubii Quadraturae' (t. 22" marg.). I am grateful to Edith Syllaforlending me her photocopy of the
Venice 1493 edition of the Quadratura and Sophismata.
74. 'Consequcntia tenet cum prima parte antecedentis, quia arguitur ab exponentibus ad expositum.
Et secundam partem [i.e., quod eadem non valet] probo: casu possibili posito, est antecedens
verum et consequens falsum; igitur consequentia non valet... Et quod consequens sit falsum
probatur: ex ipso enim sequitur quod Socrates non excipitur et quod Socrates excipitur; igitur
ipsum est falsum et implicans contradictionem' (Paul of Venice 1493a, f. 5").
to the standard exposition of exceptives, 75 Paul offers new exponents for

(S): ' N o Socrates is excepted, and every man not Socrates is excepted.' And
on that basis he can reject the original grounds for accepting the inference:
its premisses were, after all, not the genuine exponents of its conclusion. As
he admits, however, that move does not resolve the difficulty, since we can
simply reformulate the inference with those refurbished premisses. His
final resolution of the sophisma consists in claiming that it involves a
confusion between Socrates and the name 'Socrates'; from the fact that
Socrates is a part extracted from the quantitative whole that is all men we
can rightly infer only that 'Socrates' is excepted, and not that Socrates is
excepted 76 — a resolution that is neither credible nor creditable.
Wholes and parts

Exception as conceived of by medieval logicians is invariably the exception
of some sort of part from the appropriate whole; thus the familiar excep-
tion of Socrates from all men is the exception of a 'subjective' part from its
'universal* whole. But other sorts of parts and wholes provide different
contexts for exception. The most important of those contexts after the
universal/subjective context in which we have been considering exception
75. In the first conclusio, which turns out to be most important for his purposes, he makes this
implausible and, as far as I know, idiosyncratic claim: 'Unde hie notandum quod iste terminus
"pars extracapta" est terminus secundae intentionis vel impositionis limitans ad suppositionem
materialem, et iste terminus "excipitur" est terminus primae intentonis vel impositionis. Et ideo
non mirum si ilia consequentia non valet: "Socrates est pars extracapta; igitur Socrates excipitur",
quia ly "Socrates" in antécédente supponit materialiter et in conséquente personaliter. Sed
deberet inferri ilia: ly "Socrates" excipitur — quod est verum' (Paul of Venice 1493a, f. 5"). The
second conclusio raises the possibility that women as well as men are referred to in the conclusion,
which contains no inflectional barrier to such an interpretation, but only men in the second
premiss, where 'homo' is restricted to such an interpretation by the masculine ending of'alius'. The
third conclusio presents a variation on that theme, and the fourth proceeds on the possibility that
there are two men named 'Socrates', so that "Socrates non currit, et omnis homo non Socrates
currit, et tamen non omnis homo praeter Socratem currit' (Paul of Venice 1493a, f. 5*).
76. 'Pro tanto igitur dico quod ilia "Omnis homo praeter Socratem currit" sic exponitur: "Nullus
Socrates currit, et omnis homo non Socrates currit"; et ita ilia "Omnis homo praeter Socratem
excipitur: 'Nullus Socrates excipitur, et omnis homo non Socrates excipitur'. Ergo, etc." Ex
quibus sequitur manifeste rcsponsio ad rationem principalem, negando consequential!. Nee
arguitur ab exponentibus ad expositum, quia illae non sunt suae exponentes, ut dictum est. Sed
quoniam per hoc non solvitur difficultas argumenti, ideo proponatur consequentia ilia sub hac
forma: "Nullus Socrates excipitur, et omnis homo non Socrates excipitur; igitur omnis homo
praeter Socratem excipitur", et redit eadem difficultas. Proptcrea dico quod consequentia est
bona. Et admisso casu illo, dico quod consequens est verum, sicut et antecedens. Et ad im-
probationem nego quod sequitur quod Socrates excipitur. Et cum dicitur "Socrates est pars
extracapta; igitur Socrates excipitur", negatur consequentia, ut docuit prima conclusio. Sed bene
sequitur quod ly "Socrates" excipitur - sicut etiam non sequitur "Chimacra est terminus; ergo
chimaera est", sed bene sequitur "igitur ly 'chimaera' est", reservando continue eandem sup-
positionem. Quare, e t c ' (Paul of Venice 1493a, f. 5 ,b ).
is that provided by the notion of'integral' wholes and parts,77 and 'totus',
the second syncategorema to be presented in this sampler, is associated pri-
marily with an integral whole as 'omnis' is with a universal whole. There is,
therefore, a special connection between the two syncategorematic words
'praeter' and 'totus',18 one that is brought out in the fact that the same
examples can sometimes be found associated with either or with both -
e.g., 'Tota domus est alba praeter parietem' ('The whole house is white except
the wall'), ' Totus Socrates est albus praeter pedem' ('All of Socrates is white
except his foot').79 Grammatically the preposition 'praeter' and the adjec-
tive 'totus' are quite distinct, and that distinction is worth mentioning
because it means that 'totus', unlike 'praeter, can be used categorematically
as well as syncategorematically,80 an ambiguity made much of in the
discussions of 'totus', as we shall see.
Distribution effected by 'totus'

' Totus' typicallyfindsits place among the syncategoremata just after 'omnis'
since it seems clear that it is like 'omnis' in being adjoined to names and
77. Broadly speaking, it is only composite entities that are integral wholes, and integral parts are those
that can be produced by the physical division of such a whole. The development of the discussions
of 'totus' revealed difficulties in that simple notion, however, and eventually a more refined
account was called for. The one offered by Albert of Saxony is a good example: 'pars integralis
dicitur pars quantitativa - id est, habens quantitatem quae cum alia parte quantitativa constituit
aliquod totum quantum, nee una illarum est potentia ad aliam, nee perfectibilis per earn' (Albert
of Saxony 1502, f. 25rb). The reasons for Albert's final restrictions will become clear in the
remainder of this discussion.
78. Because the notion of exception depends on the notion of a whole, there is also a general
connection in the use of'totus' in definitions of exception - e.g., in texts (5): 'Exceptio vero nichil
aliud est quam captatio partis a toto' (ed. Braakhuis 1979,1,97.3), and (9): 'Quoniam autem illud a
quo fit exceptio non solum est totum sed sumptum per modum totius, si deficiat in modo totius,
tunc tenebitur hec dictio "prêter" additive vel remotive, ut: "homo currit prêter Sortem", idest:
aliquis homo et Sortes, et: homo currit sine Sorte' (ed. Braakhuis 1978, 164.19-23).
79. Thefirstexample is found in both (18) and (33/32) associated with 'praeter'; the second is found in
(18) associated with 'totus', and several highly similar examples are associated with 'praeter' in
other texts. But the most interesting link of this sort is the sophisma 'Animal est pars animaiis,
which occurs in texts (15) and *(2o/i9) associated with 'praeter' and in texts (9) and (11/10)
associated with 'lotus'. The sophisma deserves separate study, but its problem, briefly, is this.
Suppose that we designate a whole man Ά ' and the same man except for his foot 'B'. In that case
Β is now a part of an animal. But if the foot in question were amputated tomorrow, Β would be
an animal. Now if Β will be an animal tomorrow, surely it is also an animal today; therefore an
animal is a part of an animal. And since there are infinitely many parts of the B-type in any whole
animal, every animal contains infinitely many animals. (Problems of this sort eventually became
associated with the sophisma 'Totus Socrates est minor Socrate', as we shall see.)
80. See p. 212 above. In text (29) Paul of Venice devotes an entire treatise to the categorematic and
syncategorematic uses of 'totus' (Part I, Tr. 14), which he introduces as the first of a group of
words 'qui quandoque categorematice, quandoque syncategorematice tenentur - ut sunt hi:
"totus"/"tota"/"totum", "semper" et "ab aeterno", quibus annectitur "infinitus"/"infinita"/
"infinitum"' (Paul of Venice 1499, f. 56").
thereby effecting distribution; but it is equally clear that 'totus' and 'omnis'
operate differently. 81 The significations of 'omnis homo' ('every man') and
'totus homo' ('the whole man') are so different that some authors suggest that
'totus' is suited only to uniquely referring expressions as 'omnis' is suited to
common names, 82 and others think that at any rate some syntactic refine-
ment is called for when 'totus' is adjoined to a common name. 8 3 Their
apparently grammatical scruples were no doubt inspired by logical con-
siderations: one cannot infer from 'totus homo' as from 'omnis homo', and
even if the logical descent under the former is restricted to integral (rather
than subjective) parts, one is obviously in danger of committing the fallacy
of division. Thus if the whole house is white, its wall is white; but it is not
the case that if the whole house is worth one hundred pounds its wall is
worth one hundred pounds. 8 4 'Whole' is distributive in the antecedent of
the first conditional, but not in the antecedent of the second; and the
standard way of registering that difference is to say that in the first case
'whole' ('totus') is being used syncategorematically and in the second
categorematically.
But there are special problems of interpretation surrounding 'totus' even
in its syncategorematic use alone. Since 'totus' does not distribute the term
to which it is adjoined as 'omnis' does — compare 'whole man' and 'every
81. For instance, adjoining 'omnis' to the indefinite proposition 'Homo est albus' transforms it into a
universal proposition; but what is the corresponding effect of adjoining 'totus' to that proposition,
or to any other? This difference between the distributive signs was recognised very early in the
discussion of 'lotus' and led the author of Ars Emmerana to say that ' Totus lapis est substantia' is a
'mixed proposition' that, like 'Omnis homo et quidam asinus currunt', is not of any quantity at all (ed.
De Rijk 1962-7.11(2), 154.23-30).
82. E.g., text (9): 'Totus' 'distribuit enim inter partes integrales per se et non per accidens; ergo non
potest distribuere in tcrmino communi. Ergo terminus subiectus [cui "totus" adiungitur] aut erit
discrctus aut equipollens discreto, ut "totus Sortes" ... Omnia autem que predicta sunt intel-
ligenda sunt dc "toto" quando tenetur proprie, lmproprie enim potest adiungi termine communi,
secundum quod est commune, eo modo quo dicit Boethius in libro Divisionum quod particulares
homines sunt partes integrales hominis simpliciter . . . ' (ed. Braakhuis 1978, 14.10-13; 15.3-6).
83. E.g., text (8): 'Quando ["totus"] déterminât dictionem confuse significantem et non articulariter,
tantum potest postponi, ut "homo totus est albus". Si autem preponatur, incongrua est locutio, ut
"totus homo est albus". . . . Si autem determinct dictionem discrete significantem vel articulariter,
indifferenter potest preponi et postponi, ut "Socrates totus", "totus Socrates", "rex totus", "totus
rex'" (cd. De Rijk 1962-7,11(2), 305.25-7, 29-32).
84. Discussions o f examples of this sort are easy to find. In text (5) such a discussion constitutes almost
the entire treatment of 'totus': 'Sequitur de hac dictione "lotus". Que quandoque tenetur collec-
tive, quandoque distributive. Quando tenetur collective, tunc non exigit predicatum convenue
cuilibet sue parti divisim sed coniunctim. Et secundum hoc hec est vera "tota domus valet centum
libras", sensus enim est: domus valet centum Vibras et non quelibet eius pars valet centum libras. Quando
tenetur distributive, tunc exigitur quod predicatum conveniat cuilibet sue parti divisim. Unde
hec est falsa: "tota domus ista valet centum libras; ergo quelibet eius pars valet centum libras" ' (ed.
Braakhuis 1979,1, 102.14-22).
man' - what does it distribute and how does it do so? The short, standard
answer to the first question is that it distributes the integral parts of the
whole named by the term to which it is adjoined, but, as we shall see, that
standard answer ran into difficulties as the literature on 'totus' developed.
There seems to be no standard answer to the second question, and some of
the earlier answers are best applied to the two questions taken together -
e.g., 'totus' 'distributes for the integral parts', 'divides the whole adjoined to
it into integral parts', 'indicates the distribution of the integral parts of the
term to which it is adjoined'.85 But the long-term tendency of these
answers seems to point in the direction of Albert of Saxony's careful
formulation: 'There are some syncategoremata that do not distribute the
term outside themselves to which they are added - i.e., a term that is not
part of the expression to which that syncategorema is equivalent in signifi-
cation - but do nevertheless distribute a term within themselves - i.e., a
term that is part of the expression to which they are equivalent in signifi-
cation - if that term is distributable. For example, the syncategorema
"totus" does not distribute the term outside itself to which it is added, such
as the term "Socrates", but rather the term within itself, such as the term
"part".' 86
A typical analysis of'whole'

As Albert's account of the distribution effected by 'totus' indicates, much of
what a medieval logician maintains regarding 'totus' depends on his initial
analysis of it, his choice of'the expression to which that syncategorema is
equivalent in signification'. The most common analysis of 'whole' in its
syncategorematic use is simply 'each part', so that 'the whole man' ('totus
homo') is typically said to be equipollent to 'each part of the man' ('quaelibet
pars hominis'). Obviously the categorematic use of 'whole' calls for a
different analysis, and in this case the standard equipollent expression is
'made up of its parts', so that 'the whole man' is equipollent to 'the man
made up of his parts' ('homo perfectus ex partibus'). Neither of these common
85. 'Totus' 'distribuit pro partibus integralibus' (text (5), ed. Braakhuis 1979, 1, 102.25); 'totus' et
'singuli' sunt signa universale 'que totum sibi adiunctum dividunt in partes integrales' (text (7),
cd. De Rijk 1962-7, 11(2), 469.32-470.2); 'totus' 'notât distributionem partium intcgralium
termini cui adiungitur, et non appellatorum' (text (8), ed. De Rijk 1962-7,11(2), 305.16-18).
86. '... aliqua sunt syncategoremata quae non distribuunt terminum extra se cui adduntur - id est,
terminum qui non est pars orationis cui ilia syncategoremata aequivalent in significando - sed
tamen distribuunt terminum intra se - hoc est, terminum qui est pars orationis cui aequivalent in
significando - et hoc si ille terminus sit distribuibilis. Verbi gratia, hoc syncategorema "totus"
non distribuit terminum extra se cui additur, sicut est iste terminus "Socrates", sed bene intra se,
sicut iste terminus "pars" ' (Albert of Saxony 1502, f. 2$"~b).
analyses is entirely unproblematic, but the first is both more important and
more troublesome, as can best be seen in connection with a final sophisma.
Like the two sophismata considered above in connection with 'praeter', this
one will be presented in a homogenised version up to the resolution. 87
The sophisma 'Totus Socrates est minor Socrate'

THE WHOLE SOCRATES IS LESS THAN SOCRATES88
Proof. Socrates's foot is less than Socrates, Socrates's head is less than Socrates, and
so on as regards his integral parts; therefore each part of Socrates is less than
Socrates, and so the whole Socrates is less than Socrates.89
Disproof. The whole Socrates is less than Socrates, but the whole Socrates is
Socrates; therefore Socrates is less than Socrates — which is absurd.90
Despite its unpromising appearance, this sophisma generated resolutions
of at least three basic types, and those of the third type show that it contains
hidden depths. But the resolutions of the first type are just what we might
expect: they take the sophisma sentence to be ambiguous and resolve its
ambiguity in terms of two uses of'totus'. Full versions of this first type of
resolution are offered by Ricardus sophista, Henry of Ghent, and Walter
Burley. Burley's resolution is representative of the type: 'The first [i.e., the
sophisma sentence] is ambiguous in respect of equivocation, because
"whole" can be taken categorematically; and in that case it is false, because
it denotes that Socrates made up of his parts is less than Socrates. But if it is
understood syncategorematically, it is true, because it denotes that each
part of Socrates is less than Socrates.' 91
87. In connection with 'praeter' the treatises I have taken into account provide many sophismata of
several different types; in connection with 'totus' they provide very few, and of those the only one
that receives a great deal of attention and provides historical continuity is 'Totus Socrates est minor
Socrate', which is discussed more or less thoroughly in sixteen of the texts: (11/10), (12), (i5),*(ioj,
(21), (22), (23), (24), (25/26), (26/25), (27). (28), (29). (30), (31), and (32/33).
88 The problem this sophisma presents, at least initially, is the fundamental issue of the correct
analysis of 'totus', and the sophisma sentence raises that issue clearly without any specified context.
Consequently this sophisma needs no hypothesis.
89. Most authors use the proposition appearing as the subconclusion in this version of the proof as its
only premiss. This slightly more elaborate, inductive version is employed by Henry of Ghent,
Walter Burley, and Albert of Saxony (25/26).
90. Although the proof and disproof offered here are to be found in most presentations of this
sophisma, texts (24), (25/26), (28), and (29) involve more elaborate arguments in addition to these.
91. 'Solutio. Prima est multiplex secundum aequivocationem, eo quod li "totus" potest tencri
categoremacice, et sic est falsa, quia denotatur, quod perfectus Sortes ex suis partibus est minor
Sorte. Si autem intelligatur syncategorematice, sic est verum, quia denotatur, quod quaelibet pars
Sortis est minor Sorte' (ed. Boehner 1955,256.23-7). (The 'lotus' in line 21 of Boehner's edition of
this sophisma should be deleted.) Ricardus sophista offers a slightly more primitive resolution of
this type: 'Solutio. "Totus Socrates" est aequivocum, sicut dixi; et illo sumpto pro eo quod est
The simple strategy of resolutions such as Burley's is what leads me to

group them as a type, but they are alike also in their use of 'quaelibet pars
Socratis' as the analysis of'totus Socrates' in its syncategorematic use.92 Three
further texts also use that analysis in dealing with the sophisma sentence
although they do not develop the sophisma fully, and on that basis they
may be loosely associated with texts of the first type. 93 Albert of Saxony's
Perutilis logica illustrates such partial resolutions: 'Thus when ["whole"] is
taken syncategorematically it is equivalent to "each part." And it is on that
basis that the sophisma "The whole Socrates is less than Socrates" is usually
resolved; for it is equivalent to "Each part of Socrates is less than Socrates",
and that is true.' 94
A second type of resolution of this sophisma is identifiable by its
centering around a non-standard analysis of the syncategorematic use of
'totus'. Peter of Spain and William of Sherwood are the principal pro-
ponents of resolutions of the second type, but in their rejection of the
"quaelibet pars", maior vera et minor falsa. Si dicatur "totus" "perfectum ex partibus", accidit
econverso' (MS Digby 24, f. 68vb; transcribed Tachau el at). The version offered by Henry is
somewhat more refined than Walter's: 'Solutio: dicendum quod prima est duplex ex eo quod hoc
signum "totus" teneri potest collective vel distributive. Si teneatur collective: sic est falsa, ut
inprobatur; et est sensus: "totus Sortes est minor Sorte", idest: Sortes ex omnibus suis partibus conpositus
simul sumptis est minor Sorte. Si teneatur distributive: sic est vera, ut probatur; et est sensus: "totus
Sortes etc.", idest: quelibet pars Sortis per se sumpta est minor Sorte' (ed. Braakhuis 1978,9.23-10.6).
Henry also offers an alternate resolution; see n. 97 below.
92. Henry of Ghent refines the categorematic and syncategorematic analyses in order to bring out
more clearly their collective and distributive functions respectively, adding 'taken together'
('simul sumptis') to the former and 'taken individually' ('per se sumpta') to the latter; but for reasons
to be brought out in connection with resolutions of the second type, these refinements do not
affect the essential characteristics of the analyses.
93. In his Summa logicae William Ockham uses the 'quaelibet pats' analysis in this way at least twice:
'Sic enim concedunt in logica eruditi quod hoc signum 'totus' includit suum distribuibile, ut
aequivaleat isti "quaelibet pars", quando sumitur syncategorematice. Unde ista "totus Sortes est
minor Sorte" aequivalet isti "quaelibet pars Sortis est minor Sorte"' (William Ockham 1974a,
OP I, 32.95-8). (Gedcon Gâl, the principal editor of OP I, suggests William of Sherwood and
Peter of Spain as the 'eruditi' Ockham refers to (p. 32, n. 5), but they both reject this analysis, as we
shall see.) See also Summa logicae, Pars II, cap. 6 (Ibid., 267.1-269.50, esp. 268.22-7, where
Ockham's uniform application of the 'quaelibet pars' analysis leads him to accept 'Totus Sortes est
minor Sorte' as true and to reject 'Totus Sortes currit' as false on the syncategorematic interpre-
tation). Paul of Venice treats this sophisma quite fully in other works, and with different results, as
we shall see, but in his Quadratura he says simply 'non totus Socrates est minor Socrate,... quia
non quaelibet pars Socratis est minor Socrate, ut patet de anima sua ...' (Paul of Venice 1493a,
f. 27 (bis)*).
94. 'Unde cum tenetur synactegorematice valet istam "quaelibet pars". Et secundum hoc solet
concedi hoc sophisma "Totus Socrates est minor Socrate"; valet enim istam "Quaelibet pan
Socratis est minor Socrate", et hoc verum est' (Albert of Saxony 1522, f. I7vb). The contrast
between this treatment of ' Totus Socrates est minor Socrate' and the treatment Albert gives it in his
Sophismata is so great and of such a nature as to suggest that the Sophismata is the later work, just as
other comparisons suggest the converse (see n. 61 above). Albert's other treatment of the
sophisma will be considered below along with other resolutions of the third type.
standard analysis they are joined by Nicholas of Paris (who happens not to
have considered this sophisma). 95 The standard 'each part' analysis has the
advantages of simplicity and of making explicit the distribution implicit in
'whole', but it achieves those advantages at the cost of changing the subject
term of the proposition from 'Socrates' to 'part of Socrates'. Peter's analysis
brings out the distribution while retaining the original subject term, and he
explains the importance of doing so: 'For in the original proposition "The
whole Socrates is white", Socrates is subjected to whiteness in respect of
himself, and the parts not in respect of themselves but insofar as they are in
the whole of him. But insofar as they are in the whole of him, they are
under the form of the whole. Therefore they are subjected to whiteness
only in virtue of the whole. Therefore what follows first is "Socrates in
respect of each part of him is white", and afterwards "Each part of Socrates
is white".' 9 6 O n the basis ofthat analysis Peter feels entitled to take the
position that the sophisma sentence is true and that the disproof is guilty of
the fallacy of secundum quid et simpliciter; inferring the unconditional
'Socrates is less than Socrates' from 'The whole Socrates (i.e., Socrates in
respect of each part of him) is less than Socrates' is as illegitimate as inferring
it from 'Socrates in respect of his foot is less than Socrates.' 97
95. At least three other texts, all o f them from the fifteenth century, use Peter of Spain's analysis of
'totus' in resolving this sophisma, and so their resolutions are of the second type. But one of those
texts is (31), a commentary on Peter's resolution (Peter of Spain 1489, f. s a ' ^ ^ . a n d the other two
are at least strongly influenced by Peter of Spain: (30) (Ibid., f. 40"'*) and (32/33) (Ibid., f. 40 rb ).
Furthermore, texts (30) and (32/33) offer only partial treatments of the sophisma.
96. 'In hac enim propositione: "totus Sortes est albus" Sortes subicitur albedini secundum se, et partes
non secundum se, sed prout sunt in suo toto. Sed prout sunt in suo toto, sunt sub forma totius.
Ergo non subiciuntur albedini nisi per totum. Ergo per prius sequitur hec: "Sortes secundum
quamlibet sui partem est albus", et per posterius ilia: "quelibet pars Sortis est alba" ' (Tractatus, ed. De
Rijk 1972,226.12-17). William is more emphatic in his rejection of the standard analysis, but his
preferred analysis is more awkward than Peter's (I have not seen it in any other text), and his
explanation is cryptic: 'Et dicendum quod "totus Sortes" non acquipollet huic "quaelibet pars
Sortis", sed huic "Sortes, ita quod quaelibet pars". Cum enim sit signum universale affirmati vum,
supponit praedicatum inesse subjecto, et est in probatione fallacia consequentis. Ad contra [i.e., ad
improbadonem] sciendum quod ipsum proccdit secundum quod ly "totus" est syncategorema'
(ed. O'Donnell 1941, 54.23—6; some punctuation added). Nicholas, like William, flatly rejects the
'quaelibet pars' analysis, offering an explanation different from Peter's and much fuller than
William's: 'Primo videndum est utrum hec dictio "totus" aliquid signified aut nichil... Si
aliquid: sed nonnisi totalitatem. / Ex hoc videtur quod non sequatur: "totus Sortes est albus; ergo
quelibet pars Sortis est alba", quia Sortes non est aliqua pars sui; unde ille terminus "lotus Sortes"
non supponit pro aliqua parte Sortis; ergo ilia "totus Sortes est albus" non debet sic exponi: quelibet
pars Sortis est alba.. . unde sicut "partim" sic se habet resolvi: "partim" idest: secundum partem, eodem
modo "totaliter", idest: secundum quamlibet partem. Eodem modo dicimus de hac dictione "totus",
quia "totus" et "totaliter" non differunt nisi in casu' (ed. Braakhuis 1979, 11, 432.7-8; 432.11-
4331; 434-15-18)·
97. 'Peccat etiam improbatio secundum quid et simpliciter, quia ista "totus Sortes est minor Sorte" non
ponit Sortem esse minorem Sorte simpliciter sed secundum suas partes; et ita ponit Sortem
It seems fair to say that resolutions of the second type mark an advance
over those of the first type, but the qualitative difference between those first
two types is negligible compared with the difference between the third
type and either of those first two; for what I am designating resolutions of
the third type are characterised by a radically new approach to the so
phisma. 'Totus Socrates est minor Socrate', which had seemed moribund in
Ockham's and Burley's discussions, became the occasion for considering
new and more important issues in William Heytesbury's complicated,
impressive treatment of it. Heytesbury's work may well have been what
inspired Albert of Saxony and Paul of Venice to develop philosophically
interesting discussions of a sophisma that for a hundred years had been the
occasion for only a few relatively simple observations; at any rate, it is in
works of those three authors that I have found resolutions of the third
type. 98 Of course, Heytesbury may not have been the first to discern the
greater potential in this sophisma, but until another candidate for the
honour emerges we may think of most of the distinguishing characteristics
of this third type of resolution as Heytesbury's innovations. First, he
supplements the standard proof with another, in which 'Totus Socrates est
minor Socrate' is justifiably assimilated to the subtler 'Animal est pars ani-
malis',99 thereby enhancing the philosophical interest of this sophisma. 100
Second, after beginning his reply to the sophisma with the standard
observation regarding the two interpretations of 'totus', he points out that
there is a further and equally important ambiguity in the 'pars' of the
secundum quid esse minorem Sorte. Et ita, cum simpliciter infcrt sic: "ergo Sortes est minor Sorte",
peccat secundum quid et simpliciter. Sicut hic: "Sortes est minor Sorte secundum pedem; ergo Sortes est
minor Sorte"' (Tractatus, ed. De Rijk 1972, 227.15-21). Peter also attacks the disproof as an
instance of the fallacy of accident (227.8-15), a line taken up by Henry of Ghent as an alternative
to his preferred resolution (ed. Braakhuis 1978, 10.7-13; see n. 91 above).
98. The relevant texts are (24), (25/26), (28), and (29). This sophisma is treated in Heytesbury's
Sophismata on ff. Ι47 Α -Ι48*", in Albert's Sophismata on ff. 24 vl> -25 , ", in Paul's Sophismata on
ff. 40 ,a -4i' t> , and in Paul's Logica magna on f. 56'*""*. I am grateful to Edith Sylla for calling my
attention to the fact that Heytesbury treats this sophisma. In his list of Heytesbury's sophismata,
Curtis Wilson mentions Albert's and Paul's Sophismata as containing treatments of this sophisma
(Wilson 1956, p. 162).
99. See n. 79 above.
100. The material Heytesbury introduces in the second proof and his rejoinder to it seems clearly to
have been drawn from discussions of 'Animal est pars animalis', but Heytesbury does not explicitly
allude to it: 'Antecedens probatur in casu communi signando totum residuum Socratis praeter
digitum illius; et ponatur quod ille digitus incipiat non esse pars Socratis, ita quod totum residuum
maneat continue secundum quamlibet sui partem. Quo posito, sit A illud residuum. Tunc
arguitur sic. A incipit esse Socrates, et A est pars Socratis; ergo pars Socratis incipit esse Socrates...
Si enim A foret compositum huiusmodi, sequitur quod A foret homo; quia omne compositum ex
materia et forma ultima quae est anima intellectiva est homo. Et sic sequitur quod unus homo
foret infiniti homines - quod est impossibile' (William Heytesbury 1494c, f. Ι47 Α ; f. I47 vb ).
'quaelibet pars' analysis; for not only quantitative parts may be at issue, but
also qualitative parts, such as form and matter. 1 0 1 Third, after completing
an admirably thorough discussion of the sophisma as enhanced by those
first two innovations, concluding that the sophisma sentence is false on any
legitimate interpretation, he undertakes a second consideration of it, taking
'Socrates est minor Socrate', the putative absurdity to which the sophisma
sentence is reduced in the standard disproof, as a premiss from which to
infer 'Totus Socrates est minor Socrate' on radically different grounds. 1 0 2
Finally, and most significantly, he organises this second, even fuller con-
sideration of the sophisma around the concepts and lines of argument that
are characteristic of the kind of work for which he and the other Oxford
Calculators are most famous. 103
Albert of Saxony offers some arguments that may be innovations of his
101. 'Ibidem etiam distinguitur de isto termine "pars", quoniam "pars" dicitur dupliciter. Quaedam
enim est pars quantitativa, quaedam est pars qualitativa. Pars quantitativa est talis quae est in actu
quantam minorem habens magnitudinem quam totum cuius est pars; qualitativa est maxime
[reading 'maxime' for 'maxima'] materia rei, propinqua vel remota, sive sit quanta in actu sive in
potentia, et forma, sive quanta sive non quanta' (f. I47 rb ). Conceivably Heytesbury could in this
case be taking his cue from Ockham: 'Verumtamen sciendum quod aliquando, sive de virtute
sermonis sive ex usu vel placito alicuius utentis, non euro, "totus" tantum distribuit pro partibus
integralibus, non pro partibus essentialibus. cuiusmodi vocantur materia et forma, quandoque
autem distribuit pro omnibus partibus, sive sint integrales sive esscntiales sive qualescumque'
(William Ockham 1974a, OP I, 269.37-41). Albert of Saxony later expressly identified qualita-
tive and essential parts; see n. 105 below.
102. 'Ad sophisma arguitur adhuc sic. Socrates est minor Socrate; ergo e t c ' (William Heytesbury
1494e, f. I47 vb ). Heytesbury's second consideration of the sophisma, stemming from this first,
novel proof of it, is about three times as long as andfar more complex than his first consideration
of it. Paul of Venice's second, third, and fourth proofs appear to be drawn from Heytesbury's
second consideration of this sophisma as his first appears to derive from the first consideration
(Paul of Venice 1493 b, f. 4o"- b ).
103. See. e.g., the passage immediately following the one quoted in n. 102: 'Antecedens arguitur.
Socrates erit minor Platone ante A instans, et continue ante A instans Socrates erit aequalis
Platoni; ergo, etc. Assumptum arguitur sic. Ponatur quod Socrates et Plato iam sint aequales, et
quod quandocumque erunt, erunt aequales, et augeatur uterque illorum aequevelociter usque ad
A instans. Et ponatur quod Plato erit in A instanti, et quod Socrates tunc primo erit corruptus.
Quo posito, probatur quod Socrates erit minor Platone ante A instans; quia Socrates erit minor
quam erit Plato, quia Plato erit maior quam erit Socrates. Ergo Socrates erit minor quam erit
Plato, sed numquam nisi ante A instans; ergo Socrates erit minor Platone ante A instans, casus
enim ponit quod Socrates numquam erit nisi ante A instans" (William Heytesbury 1494c,
f. I47 vk ). This hypothesis and the lines of argument that derive from it are so typical of the
concerns of the Calculators and so indirectly attached to the original problems of the sophisma
that in the intricacies of this second consideration Heytesbury seems sometimes to be addressing
only the concerns of the Calculators. For example, he refers at one point to views held by others,
and they are surely views that can be found among the earlier Calculators, such as Thomas
Bradwardine and Richard Kilvington, but they are very unlikely to have had anything to do
with the sophisma that is ostensibly still at issue: 'Ideo dicitur a multis quod Plato habebit ante A
quamcumque quantitatem sibi acquiret ante A et etiam ipse Socrates, sed ipse in nullo instanti
habebit maximam quantitatem quam habebit vel acquiret sibi ante A. Unde dicunt ipsi quod A
erit primum instans in quo Plato habebit maximam quantitatem quam ipse acquiret ante A,
tamen Plato tunc non primo habebit totam quantitatem quam acquiret ante A; in toto enim illo
own, 104 but the most valuable features of his treatment of the sophisma are
to be found in the three notes in which he lays the foundation for his
resolution of it; and in the third note he provides a detailed account of the
notion of an integral part in order to set aside the sort of ambiguity pointed
out in Heytesbury's second innovation.105
In his Sophismata Paul of Venice apparently borrows heavily from
Heytesbury and perhaps also from Albert, but his use of the material is
intelligent, and he seems to make some contributions of his own in his
rejoinders to the four proofs he offers.106 In his Logica magna, however,
Paul seems to present the sophisma in a way designed to avoid some of the
problems to which he devotes his attention in the Sophismata; here again he
tempore terminato ad illud A instans Plato habebit totam quantitatem quam habebit in A. sed in
nullo instanti illius temporis habebit illam. Et dicunt quod ilia quantitas prius erit acquisita vel
habita a Platone, sed in nullo instanti prius erit acquisita alicui' (Ibid., f. 148™).
104. E.g., his second disproof and his rejoinder to it. 'Si totus Socrates est minor Socrate, tunc totus
homo est minor Socrate - quod est falsum. Consequentia videtur valere eo quod arguitur ab
inferiori ad superius. Falsitas consequentis probatur, nam si totus homo est minor Socrate, tunc
quaelibet pars hominis est minor Socrate. Modo hoc est falsum, posito quod Socrates sit parvus et
Plato sit magnus; tunc non quaelibet pars hominis est minor Socrate. Nam sequitur "Quaelibet
pars hominis est minor Socrate; ergo quaelibet pars Platonis est minor Socrate"; sed hoc est falsum
ex casu. Consequentia tenet: a superiori distributo ad inferius distnbutum; nam in ista "Quaelibet
pars hominis est minor Socrate" hoc aggregatum "pars hominis" distribuitur per unam regulam
prius dictam - scilicet, Quotienscumque signum distributivum addilur aggregate ex recto el obliquo, recto
praecedente obliquum, tunc aggregatum ex recto et obliquo distribuitur. Et ideo ly "hominis" ita bene
distribuitur sicut ly "pan" ' (Albert of Saxony 1502, f. 25'*). Albert's rejoinder consists essentially
in rejecting the main inference: 'Unde sicut non sequitur "Quelibet pars integralis Socratis est
minor Socrate; ergo quaelibet pars integralis hominis est minor Socrate", ita non sequitur "Totus
Socrates est minor Socrate; ergo totus homo est minor Socrate" ... in ista oratione "Quaelibet
pars integralis hominis est minor Socrate" non solum distribuitur ibi "pars integralis" sed hoc
aggregatum "pars integralis hominis"' (Ibid., f. 25"").
105. 'Tertio notandum est quod pars integralis dicitur pars quantitativa - id est, habens quantitatem
quae cum alia parte quantitativa constituit aliquod totum quantum, nee una illarum est potentia
ad aliam, nee perfectibilis per earn. Ex hoc patet quod anima Socratis vel corpus Socratis non
dicitur pars integralis Socratis, quia licet anima Socratis constituât Socratem quantum, et etiam
corpus Socratis sic habens quantitatem, tamen anima Socratis non est pars quantitativa; et etiam
materia Socratis est perfectibilis per animam Socratis... Et taies partes quarum una est sic
perfectibilis per aliam non dicuntur partes integrales sed cssentiales vel qualitativae' (Albert of
Saxony 1502, f. 25**).
106. Here, for instance, is a passage in which Paul again shows his concern with the details of
exposition as a means of resolving difficulties involving syncategoremata (see also n. 76 above):
'Non est ergo aliqua propositio exponibilis ratione alicuius syncategorematis praecedentis totam
propositioucm quin ab ilia ad suum praeiacens sit bonum argumentum. Sed ly "Socrates est pars
Socratis" est praeiacens illius "Totus Socrates est pars Socratis", quia praeiacens est illud quod
remanet, dempto signo. Ergo ilia consequentia est bona: "Totus Socrates est pars Socratis; ergo
Socrates est pars Socratis."/Dicatur quod ilia propositio "Totus Socrates est pars Socratis" sic
exponitur: "Socrates est pars Socratis, et quaelibet pars Socratis est pars Socratis"; ergo, etc. Et hoc
patet secundum communem modum loquendi; si enim totus Socrates est albus, intelligitur quod
Socrates secundum se et quamlibet sui partem est albus. Et notanter dico secundam exponentem
universalem affirmativam et non illam universalem negativam: "et nulla est pars Socratis quin illa
sit pars Socratis", quia aliter iliac essent verae: "Totus punetus est punetus", "Tota anima
intellectiva est anima", quia ambae exponentes essent verae. Sed consequens est falsum, quia ly
seems to have learned from Heytesbury and perhaps even more markedly
from Albert.107
Conclusion
As I have indicated at several points in this discussion, my historical and
philosophical sampling has turned up doctrinal or methodological similar-
ities among some of the texts considered in this exploration, but the
historically significant groupings of texts that may eventually emerge from
comparative studies of this sort are not yet clear enough to be presented
definitively. At the conclusion of this necessarily restricted exploration of
the medieval literature on the syncategoremata it is essential to remember
that it has been no more than the shallow, narrow trenching of an
immeasurably rich site, intended primarily to stimulate and orient further
exploration.108
APPENDIX I
A SELECTION OF T E X T S IN A P P R O X I M A T E
CHRONOLOGICAL ORDER
The numerical order of this list indicates either the received opinion or my
best guess regarding the chronological order of the texts listed. Divided
numerical designations, such as '(2/3)' and '(3/2)', indicate particular un-
"totum" syncategorematice tantum distribuit implicite pro partibus et importât significatum
subiecti habere partes' (Paul o f Venice 1493b, f. 40vl>). Paul's concern about such expressions as
'totus punetus' was also felt by several earlier authors: see, e.g., Nicholas of Paris in Summe Metenses,
exc. Dc Rijk 1962-7,11(1), 488.16-21. The account of the distributive effect of'tolHj' offered by
Paul at the end of the passage quoted above is reminiscent of the one developed by Albert of
Saxony; see n. 86 above.
107. See, e.g., his disproof and his rejoinder to it: ' . . . arguitur probando illam esse falsam: "Totus
Socrates est minor Socrate", quia non quaelibet pars quantitativa Socratis est minor Socrate . . .
nam materia Socratis, quae est altera pars compositi, est pars quantitativa . . . et tamen non est
minor Socrati; igitur, etc. / . . . nego quod quaelibet pars quancitativa Socratis est minor Socrate,
et etiam nego quod materia quae est altera pars compositi est pars quantitativa, ut huiusmodi
responsio intelligatur. Est notandum quod in Socrates est duplex pars - scilicet, qualitativa, quae
est altera pars compositi, ut patet tota ilia materia cum qua componitur anima, et quantitativa,
quae est minor toto minorem continens materiam quam totum, quae cum alia quantitate
constituit aliquod totum quantum, quarum nulla est in potentia ad aliam. Ex hoc patet quod
materia Socratis vel eius corpus non dicitur pars quantitativa vel integralis, quia perfectibilis est
per animam et in potentia respectu eiusdem, non obstante quod ambo constituant corpus habens
quantitatem' (Paul o f Venice 1499, f. s6 r b ; ff. $6*-$6Ύ'). Compare Albert of Saxony on
quantitative and qualitative parts, n. 105 above.
108. I am very grateful to Eleonore Stump for her helpful criticisms of an earlier draft and to Jan
Pinborg and Barbara Ensign Kretzmann for many suggestions that enabled me to improve my
presentation o f this material.
certainty on my part regarding the order of the two texts thus designated
relative to each other; my best guess is represented by the first numeral in
such divided designations. Designations preceded by asterisks indicate texts
that I have seen only in excerpts. Titles in brackets have been supplied by
me.
W Quaestiones Victorinae (ed. De Rijk] ist half XII

(2/3) Tractatus de univocatione Mcmacensis [ed. De Rijk] 3rd qu. XII
(3/2) Ars Emmerana [ed. De Rijk) 3rd qu. XII
*(4) Ars Meliduna [exe. De RijkJ 1170/1180
(5) Sincategoreumata Monacensia [ed. Braakhuis] 4th qu. XII
(6) Tractatus implicitarum [ed. Giusberti] turn o f century
(7) Dialectica Monacensis (ed. De Rijk] 4th qu. XII
(8) Tractatus Anagnini [ed. De Rijk] ist decades XIII
(9) Robert Bacon (?): Sincategoreumata [ed. Braakhuis] 1200/1210
*(IO/II) John le Page: Syncategoremata [exc. Braakhuis] 1220/1230
(n/10) Ricardus sophista: Abstractiones [ed. Tachau et al] 1220/1230 (?)
(12) Peter o f Spain: Tractatus [ed. De Rijk] 1230/1240
*(I3) Peter o f Spain: Syncategoremata [exc. Braakhuis] 1230/1240
(14) Pseudo-Petrus: Tractatus syncategorematum [1489 edn.] 1230/1240 (?)
(15) William of Sherwood: Syncategoremata [ed. O'Donnell] 1230/1240
*(I6) ['Sophismata logicalia'][exc. Grabmann] after 1230
*(17) Nicholas of Paris: Summe Metenses [exc. De Rijk] ca. 1250
(18) Nicholas of Paris: Sincategoreumata [ed. Braakhuis] ca. 1250
* (19/20) 'Quoniam ignoratis communibus' [exc. Grabmann] mid-XIIl
* (20/19) ['Sophismata Parisius determinata'] [exc. Grabmann/Braakhuis] mid-XIII
(21) Henry of Ghent: Sincathegoreumata [ed. Braakhuis] 3rd qu. XIII
(22) William Ockham: Summa logicae [ed. Boehner/Gâl] 1323
(23) Walter Burley: De puritate artis logicae [ed. Boehner] before 1329
(24) William Heytesbury: Sophismata [1494 edn.] ca. 1340 (?)
Os/26) Albert o f Saxony: Sophismata [1S02 edn.] ca. 1350 (?)
(26/25) Albert of Saxony: Perutilis logica [1522 edn.] ca. 1350 (?)
(27) Paul of Venice: Quadratura [1493 edn.] after 1390 (?)
(28) Paul of Venice: Sophismata [1493 edn.] after 1390 (?)
(29) Paul of Venice: Logica magna [1499 edn.] after 1393
(30) Pseudo-Petrus: Tractatus exponibilium [1489 edn.] X V (?)
(31) Cologne Dominicans: Comm. on Tractatus (12) XV
(32/33) Cologne Domincans: Comm. on Tr. exponibilium (30) XV
(33/32) Pseudo-Petrus: Tr. syncategorematum (supp.) [1489 edn.] XV
A P P E N D I X II
EXAMPLES OF ELEMENTS IN THE T R E A T M E N T OF ' P R A E T E R '

('BESIDES') AND ' T O T U S ' ('WHOLE')
/. Definitions and analyses (of exception or of the sort of

distribution effected by 'totus')
'Exception is nothing other than the taking of a part from a whole.' (5)
'To except is to take one thing from another in respect of a third.' (18)
'For exception, properly so-called, four things are required - namely, that
by which it is excepted, and that in respect of which, and that which is
excepted, and that from which it is excepted.' (9)
'Five things are required for exception. One is the excepting thing, and that
is the soul; and another is the instrument of excepting, and that is an
exceptive word, such as "besides" and the like. Moreover, these three,
regarding which it occurred, are required — namely, that which is excepted,
and that from which it is excepted, and that in respect of which it is
excepted.'*(i3)
'Notice that the signs "every" and "whole" are different, for the sign
"every" distributes for subjective parts, and the word "whole" distributes
for integral parts.' (5)
'Its signification is evident in this way. " W h o l e " is either a categorema or a
syncategorema. If it is a categorema, it signifies the wholeness of a thing;
and in that case one usually speaks of a formal, or collective, or integral
whole, and "whole" is the same as "complete" and is explained in this way:
a whole is that outside of which there is nothing [belonging to itself]. If it is
a syncategorema, it signifies the wholeness of the subject insofar as it is the
subject; and in that case it is distributive and one usually speaks of a material
whole.' (9)
//. Classfications of the uses of the words ('praeter' and 'totus')

'Exceptively, Inclusively, Exclusively, Distributively, Collectively' (6)
'Additively, Exceptively: Diminutionally, Counter-instantively' (15)
'Exceptively, Diminutionally' (22)
'Properly, Improperly: Negatively, Additively, Subtractively (or Diminu-
tionally)' (25/26)
'Collectively, Distributively' (5)
'It sometimes determines a word signifying confusedly and not articu-
lately, sometimes [a word signifying] discretely or articulately.' (8)
'It is, however, equivocal relative to these two: "each part" and "made up
of parts". (11/10)
'The name "whole" can be considered in two ways: one in which it is a
distributive sign and one in which it is the same as "complete".'*(l7)
'The word "whole" is sometimes taken distributively, sometimes collec
tively; thus sometimes it signifies universally and divisively, sometimes
universally and conjunctively.' (18)
HI. Grammatical, semantic, and logical rules governing propositions

(involving 'praeter' and 'totus')
'If a sentence in which the word "besides" occurs is true, it is false without
it.' (5)
'Even though [a proposition involving] exception has a copulative pro
position within itself implicitly [i.e., the conjunction of its exponents], it is
called not hypothetical but categorical.' (9)
'Whenever as many things are excepted in a given expression as are
supposited, the sentence is false or ungrammatical.' (21)
'One must not make a logical descent together with exception under the
subject of an exceptive proposition.' (25/26)
'When ["whole"] is taken distributively, it is required that the predicate go
together with each part of the subject dividedly.' (5)
'The word "whole" cannot determine a word that discretely signifies
something that is not composite.' (8)
'The sign "whole" can be added to a term to which the sign "every"
cannot be added, because something can be divided into integral parts that
cannot be divided into subjective parts.' *(i7)
'An inference from a term taken together with "whole" to the term taken
without it does not follow.' (22)
IV. Examples (associated with II or III)

'Every man besides Socrates is running.' (5)
'Ten besides five are five.' * (10/11)
'The whole house besides the wall is white.' (18)
'Every animal besides man is irrational.' (22)
'The whole stone is a substance.' (3/2)
Ά genus is a whole relative to its species.' (18)
'If the whole man is an animal, the man is an animal.' (22)
Ά whole [or the whole] that is in the world is in your eye or in your purse.'
(29)
V. Sophismata (involving 'praeter* or 'totus')

'Every animal is rational besides the irrational.' (i)
'Everything is everything with the exception of everything besides every
thing.' (9)
'Ten besides one know themselves to be nine.' * ( i o / n )
'Every man sees every man besides himself (11/10)
'Nothing is true except (nisi) at this instant.' (18)
Ά whole disjunctive proposition is true of which either part is true.' (ι 1/10)

'Every whole is greater than its part.' (25/26)
VI. Expositions of propositions (involving 'praeter' or 'totus')
' N o man besides Socrates is running — i.e., Socrates is running, and no one
else is running.' (6)
' N o man besides Socrates is running — i.e., Socrates is running, and no man
other than Socrates is running.' (22)
' N o man besides Socrates is running — i.e., Socrates is running, and every
man other than Socrates is not running.' (25/26)
'Some man besides Socrates is risible - i.e., N o t only Socrates is risible, but
also some other man.' (23)
'Ten besides five are five—i.e., If from ten five are taken away, five remain.'
(29)
'The whole house is worth one hundred pounds - i.e., The house is worth
one hundred pounds, and not each part of it is worth one hundred pounds.'
(5)
'The whole man is white - i.e., The man is white, and each part of him is
white.' (8)
'This whole proposition is true - i.e, Each quantitative part of this pro
position is true.' (29)
VII. Questions (arising from I- VI)
'Can an exception be made from the predicate as from the subject?' (18)
'Can an exception be made from a relative pronoun related to a distributed
term?' (18)
'Since everything that signifies signifies either by way of a concept or by
way of an operation, in which of those ways does the word "besides"
signify exception?' (18)
'In some cases this follows: "the whole Socrates; therefore Socrates' -e.g.,
"The whole Socrates is white; therefore Socrates is white" — and in some
cases it does not follow; therefore in which does it follow?' (13)
'Can the sign "whole" be added to a term signifying something that is not
divisible into integral parts -i.e., can one say "whole God", "whole soul",
"whole point", "whole unity"?' * (17)
12
INSOLUBILIA
The nature and history of insolubilia

Medieval literature on 'insolubles' began to appear by the early thirteenth
century at the latest and continued to the end of the Middle Ages. 1
Insolubles were primarily certain sorts of self-referential sentences, seman-
tic paradoxes like the 'liar paradox' ('What I am now saying is false'). But
few authors tried to give a rigorous definition, 2 so that other more or less
unrelated kinds of paradoxes were also treated under this heading. 3
Three periods may be distinguished in the medieval insolubilia literature:
(i) from the beginnings to ca. 1320; (2) the period of the most original
work, from ca. 1320 to the time of the Black Death (1347-50); (3) after ca.
1350, a period of refinement and elaboration but, with a few exceptions,
little that was new.
Resolutions in terms of cassation
Several approaches may be distinguished during the first period. One was
called 'cassation' - i.e., nullification. O n this theory, he who utters an
insoluble 'says nothing'. 4 The earliest known text 5 adopts this view, and by
ca. 1225 it was said to be 'according to the common judgement'. 6
Nevertheless, it soon died out and seems not to have been revived until
David Derodon in the seventeenth century. 7 It is not clear how the theory
is to be taken. In the middle thirteenth century a text attributed to William
1. Spade 1973 and Spade 1975a, Bottin 1976. For the later period see Ashworth 1972, Ashworth
1974, Ch. 2, §4, Prantl 1855-67, Vol. 4, §22, and Roure 1962.
2. William Heytesbury did. See the texts in Spade 1975a, pp. 117-18.
3. E.g., Buridan 1977, Sophismata, 8, sophismata 1—2 and 17—20, translated in Buridan 1966,
pp. 180-5 and 219-23.
4. Spade 1975a, pp. 32 and 44.
5. Edited in De Rijk 1966b. See the discussion in Spade 1975a, p. 32, and Bottin 1976, pp. 47-54.
6. Spade 1975a, p. 43. This might mean only that it is the way the common man would respond to
insolubles, not that it was the general view of those who had considered the question theoreti-
cally. See the anonymous author's remarks in Spade 1975a, p. 44.
7. Ashworth 1974, p. 115.
Insolubilia 247
of Sherwood8 discussed several views that might be considered versions of

cassation, including one theory that insolubles for semantic reasons fail to
be sentences, so that he who utters them 'says nothing'. 9 This theory has
modern parallels, e.g., in Fitch 1970. Nevertheless, the author of the text
10
reserved the word 'cassation' for only one of those views, which he
refuted merely by observing that it 'denies the senses'.11 Thomas
Bradwardine made the same point ca. 132412: he who says the words Ί am
saying a falsehood' is saying at least those words, and so not 'nothing'. 1 3
Through Bradwardine, this argument became standard. ' 4 The theory was
not so facile as this refutation implies, however. In the earliest known
version of the theory, distinctions were already made that suffice to disarm
this refutation.15 Nevertheless, those who later held the theory seem to
have been remarkably unable to defend themselves against such objections,
so that their view soon died out.
Resolutions in terms of'secundum quid et simpliciter'
A second approach, the most common early one, tried to treat insolubles as
fallacies of confusing what is true only in a certain respect with what is true
absolutely (secundum quid et simpliciter). Aristotle discussed such fallacies
in Sophistici elenchi, 25, i8o a 27- b 7, and briefly alluded to 'the problem
whether the same man can at the same time say what is both false and
true' (Oxford translation). The almost universal testimony of the extant
thirteenth-century treatises indicates that the medieval literature arose out
16 17
of speculation on this passage. All but two of those treatises adopt
variants of this approach, sometimes combining it with other views.
Nevertheless, insolubles do not fit very well the pattern of the fallacy
secundum quid et simpliciter, so that such approaches were always strained. ,8
8. Edited in Roure 1970. See Spade 1975a, pp. 26-8. On the doubtful attribution, see De Rijk
1966b, p. 93.
9. Roure 1970, pp. 257-60, pars. 8.01-ι 1.01. Note the phrase'says nothing' -'nihil dicit'- in 11.01.
10. Roure 1970, pp. 249-50, pars. 1.01 -2.06.
11. Ibid., p. 259, par. 2.01. At least two other thirteenth-century authors discussed cassation. See
Spade I97sa, pp. 43-4 (anonymous) and pp. 94-5 (Richard of Sherwood).
12. On the date see Weisheipl 1968, p. 190.
13. Insolubilia, edited in Roure 1970, p. 295, pars. 5.05-5.06.
14. E.g., Paul of Venice 1499, Logica magna, 2,15, 4th and 5th previous opinions, f 192 "*•; also Ralph
Strode cited in Spade 1975a, p. 87.
15. See the discussion in Bottin 1976, pp. 51-2.
16. Spade 1973.
17. Edited in De Rijk 1966b and Braakhuis 1967.
18. Spade 1973.
They seem to have died out around 1330 with Richard Kilvington, 1 9
although a certain Henry of England held this view, perhaps much later. 2 0
Resolutions in terms of transcasus
A third early theory perhaps arose out of attempts to make insolubles fit
what Aristotle says about fallacies secundum quid et simpliciter.2 ' According
to this view, present-tensed verbs in insolubles refer to the time just before
the insoluble utterance, so that Ί am speaking a falsehood', despite its
grammar, means Ί spoke a falsehood a moment ago'. Insolubles are true or
false depending on whether one did in fact utter a falsehood at that earlier
time. Walter Burley refers to such a view as 'transcasus', and rejects it. 2 2
T h e anonymous text in Braakhuis 1967 adopts this position and links it
with a rejection of self-reference. Bradwardine, t o o , 2 3 treats the view as a
variant of the theory of the 'restricters' (restringentes), who denied self-
reference. The restricters argued that 'a part cannot supposit for the whole
of which it is a part' - e.g., a term for the sentence in which it occurs. Some
ruled out more general kinds of referential cycles as well. T w o groups of
restricters may be distinguished. Some ruled out self-reference in all cases,
mostly on the basis of very bad arguments. 2 4 It was quickly realised that
this strong view prevented innocuous as well as vicious self-reference. 25
Others allowed self-reference in some cases, but not in insolubles; 2 6 they
did not provide any very informative way to distinguish such cases. 27
19. See Spade I97sa, pp. 92-3, and the anonymous text, probably from the fourteenth century,
discussed ibid., pp. 41 —2.
20. Spade 1975a, pp. 5S-6.
21. Spade 1973, p. 307, n. 64.
22. Insolubilia, edited in Roure 1970, pp. 270-1, pars. 2.07-2.08. The exact sense of the word
'transcasus' in this context is obscure, and Burley's own explanation is of little help: 'And there is a
transcasus when some sentence is changed from truth to falsehood, or conversely' ('Et est transcasus
quando aliqua propositio mutatur a veritate infalsitatem vel e conversa'; ibid., p. 270 par. 2.07). Perhaps
the best translation is 'passage of time', i.e., the time between the occurrence of the events
described and the description of them. Thus 'The time in which I say something is other than the
time for which I say something' ('[aliud] est tempus in quo dico aliquid et pro quo dico aliquid'; ibid.).
Perhaps 'transcasus' is a corruption of transcursus', which had the sense 'passage of time' in the late
Middle Ages; see Latham 1965.
23. In Roure 1970, pp. 294-5, pars· 5·°4-5·°42·
24. E.g., the text in Braakhuis 1967 and the anonymous text discussed in Spade 1975a, pp. 33-4.
25. Thus Pseudo-Sherwood in Roure 1970, p. 252, par. 4.02.
26. E.g., Burley in Roure 1970; Ockham 1974a, 3 -3, 46, discussed in Spade 1974b; Walter Sexgrave,
discussed in Spade 1975a, pp. 113-16; Roger Roseth, discussed in Spade 1975a, pp. 101-
2.
27. See Spade 1974b.
Insolubilia 249
Resolutions in terms of truth-conditions

Bradwardine presents a survey of these and other early views.28 With him,
the «K5o/MÎ>i7iij-literature entered its second and most productive phase.29
Bradwardine seems to have been the first to formulate carefully and take
seriously the theory that insolubles not only 'signify' or 'imply' that they
are false, but also that they are true. 30 Some went further and said that all
sentences signify or imply that they are true.31 Since in order that a
sentence be true, everything that it signifies or implies must be so, it follows
that insolubles are true only if they are both true and false. On this approach
insolubles are false and no paradox can be derived. Variations of this
approach were adopted by many authors; indeed, it became one of the
main traditions in the fourteenth century.32 The phrases 'direct' and
'consecutive signification' that one sometimes sees in the insolubilia-
literature belong to this tradition.33 Although the claim that sentences
signify or imply that they are true may be traced back at least to
Bonaventure,34 Bradwardine was among the first to apply it to insolubles.
Note that if only insolubles signify or imply their own truth, the theory
seems ad hoc. But if all sentences do so, then the theory amounts to a
wholesale abandonment of half the so-called 'Tarski biconditionals',35
which seems too heavy a price.
28. Roure 1970, pp. 286-96, pars. 2.01-5.08.

29. Ralph Strode, after paraphrasing Bradwardine's survey, says, 'After them there arose the prince
of modern natural philosophers, namely. Master Thomas Bradwardine, w h o first came upon
something o f value concerning insolubles.' ("Post quos surrexit princeps modcrnorum physi-
corum, videlicet, magister Thomas Bradwardine, qui aliquid quod valuerit de insolubilibus
primitus adinvenit.' Quoted in Spade 1975a, p. 88.)
30. In Roure 1970, p. 298, par. 6.05, conclusion 2. See Spade 1975a, p. 109, n. 200 for an importantly
different reading of this passage. Not all who held this view were willing to put it in terms of
signification. E.g., Buridan 1977, 8, sophisma 7, translated in Buridan 1966, pp. 194-6. The
theory is mentioned very briefly, before Bradwardine, in Burley's Insolubilia, ed. Roure 1970,
p. 272, par. 3.02.
31. Burley 1955, p. 25.37-8; Burley in Roure 1970, p. 272, par. 3.02; Buridan 1977, 8, sophisma 7,
translated in Buridan 1966, pp. 195-6; Albert of Saxony 1518, Perutilis logica, 6, I, concls. I - 3 ,
f. 43"*-
32. See, besides Bradwardine, the texts cited in Spade 1975a as items IV, VIII, XII (all anonymous),
XXIV (Albert of Saxony), Χ Χ Χ Ι (Henry Hopton), X X X I V (John Buridan), XXXVIII (John of
Holland), X X X I X (John Hunter), XLIX (Paul of Pergula), LIU (Ralph Strode), LVI (Richard
Lavenham), LVIII (Robert Fland).
33. See Spade 1975a, pp. 47 and 60.
34. Quaestionesdispulatae de mysterio trinitatis, q. i,a. 1, ad 5; in Bonaventure 1882-1902, Vol. 5, p. 50.
See Spade 1975a, p. 53.
35. See Spade 1971, pp. 1-3.
Resolutions in terms of self-falsification

Roger Swineshead, writing probably shortly after Bradwardine, took a
different approach. For him, a true sentence is one that not only cor-
responds to reality ('signifies principally as is the case') but also does not
'falsify itself'.36 Conversely, a false sentence is one that either misrepresents
reality ('signifies principally otherwise than is the case') or else 'falsifies
itself'.37 A sentence 'falsifies itself just in case it is 'relevant' (pertinens) to
inferring that it is false.38 This notion of'relevance' is a complex one, and
needs further study.39 Insolubles, in any case, turn out to 'falsify them-
selves'.40 Thus the insoluble 'This sentence is false' is simply false, and no
paradox arises. One cannot argue that, since that is exactly what the
sentence signifies, it must be true after all. For Swineshead, truth requires
more than that.
Swineshead's conclusions
Swineshead drew three famous conclusions from his approach; they were
the topic of much controversy after him.41 First, 'some false sentence
signifies principally as is the case';42 wehavejustseenone. Second, 'in some
good formal consequence, the false follows from the true'. 43 For example,
36. 'Propositio vera est propositio non falsificans se principaliter sicut est significans naturaliter aut ex
impositione vel impositionibus qua vel quibus ultimo fuit imposita ad significandum.'
(Insolubilia, ed. Spade 1979a, par. 14.) See also Spade I975a» PP- 102—5.
37. 'Tertia definitio: Propositio falsa est oratio falsificans se vel oratio non falsificans se principaliter
aliter quam est significans naturaliter, ex impositione, vel impositionibus qua vel quibus ultimo
fuit imposita ad significandum.' (Ed. Spade 1979a, par. 15.)
38. 'Omnis propositio pertinens ad inferendum se ipsam fore falsam est falsificans se.' (Ed. Spade
1979a, par. 17.)
39 It is not the same as the notion of relevance in the obligationes-hterature, where a sentence S is said
to be 'relevant' (pertinens) to another sentence S' if and only if S either follows from or is
inconsistent with S'. See Swineshead's Obligationes, ed. Spade 1977, par. 4; 'Tertia divisio est haec:
Propositionum alia est pertinens obligato, alia est impertinens obligate Et pertincnfium obligato
alia est sequens ex obligato, alia repugnans obligato.' See also n. 43 below.
40. Strictly speaking, Swineshead includes certain epistemic paradoxes under the heading 'insoluble',
so that his official definition provides for them too: 'Insolubile ad propositum est propositio
significans principaliter sicut est vel aliter quam est pertinens ad inferendum se ipsam fore falsam
vel nescitam vel creditam, et sic de singulis.' (Insolubilia, ed. Spade 1979a, par. 16.)
41. Seee.g., Spade 1975a, items HI (anonymous), XXVI (Anthony de Monte), XLII (John of Wesel),
XLIX (Paul of Pergula), L (Paul of Venice), LVIII (Robert Fland), LXI1 (Roger Roseth), and
LXIX (William Heytesbury).
42. 'Aliqua propositio falsa significat principaliter sicut est.' (Ed. Spade 1979a, par. 25.)
43. 'In aliqua consequentia bona formal·' ex vero sequitur falsum.'(Ed. Spade 1979a, par. 26.) For the
example, see ibid. Perhaps this feature of Swineshead's position was prompted by reflection on the
end of Sophisma 48 in Richard Kilvington's Sophismata. Kilvington allows that 'in some good
consequence, some sentence that is false in a certain respect follows from [ones that are) true
Insolubilia 251
'The consequent of this consequence is false; therefore, the consequent of

this consequence is false'; although both antecedent and consequent cor-
respond to reality, the latter falsifies itself and the former does not. (For
many medievals, it was the sentence token that was the bearer of truth-
value.) Swineshead explicitly says that although valid consequences need
not preserve truth, they do preserve the property of corresponding to
reality.44 Third, in the case of insolubles, 'two mutual contradictories are
false at the same time'. 45 Thus the insoluble 'This sentence is false', al-
though false, corresponds to reality. Its contradictory, 'That sentence is not
false', fails to correspond and so is likewise false.
Swineshead distinguished the notion of truth from the notion of cor-
responding to reality. If, when the two were identified, there were para-
doxes involving truth and falsehood, one may well suspect that the para-
doxes are still there in the notion of correspondence, even though they will
no longer be expressed in terms of truth and falsehood. Thus, consider the
sentence 'This sentence signifies principally otherwise than is the case.'
Swineshead discusses these paradoxes too, 46 and seems to say that they are
neither true nor false.47 But the details are obscure and it is not clear that he
can handle them adequately. Swineshead's view was sharply attacked by
William Heytesbury in his Regulae solvendi sophismata, c. 1. Nevertheless it
was adopted in an amplified and sometimes modified form by Paul of
Venice48 and John of Celaya,49 and by several authors in the sixteenth
century.50
absolutely' ('in aliqua consequentia bona aliqua propositio quae est falsa secundum quid sequitur
ex veris simpliciter'). In Swineshead, of course, the distinction between truth or falsehood 'in a
certain respect' and truth or falsehood 'absolutely' is dropped. This possible source for
Swineshead's conclusion is all the more interesting because in the immediately following lines
Kilvington goes on to say, 'But nevertheless this sentence [that is) false in a certain respect is not
false because it follows from truths, but because from it, together with some truth or other truths,
there follows its contradictory' ('Sed tamen ista propositio falsa secundum quid non est falsa quia
sequitur ex veris, sed quia ex ea cum aliquo vero vel aliis veris sequitur suum contradictorium').
This looks suggestively like Swineshead's notion of a 'sentence relevant to inferring that it is false'.
44. 'Si ex aliquibus propositionibus quarum quaelibet significat principaliter sicut est sequitur aliqua
propositio, ipsa significat sicut est.' (Ed. Spade 1979a, par. 35.)
45. 'Duo contradictoria sibi mutuo contradicentia sunt simul falsa.' (Ed. Spade 1979a, par. 27.) For the
example below see ibid.
46. Ed. Spade 1979a, pars. 2-3, 93, 99.
47. This is explicit in MS Vat. lat. 2154, f. 6"*. Other MSS omit the claim, but it seems to follow from
Swineshead's doctrine.
48. Logica magna, 2,15. See Spade 1975a, pp. 82-4.
49. Edited in Roure 1962.
50. Ashworth 1974, pp. 112-13.
Resolutions in terms of obligationes

Heytesbury's own theory was even more influential. 51 According to him,
insolubles should be treated within the context of obligationes, the codified
conditions of formal scholastic disputation. The upshot of his approach is
that no sentence, under a set of circumstances or hypothesis (casus) that
makes it insoluble, can signify exactly as it ordinarily does; the assumption
that it can do so is what is responsible for the paradoxes. If in such a case the
sentence signifies as it ordinarily does, it must signify more as well.
Depending on what that additional signification is, and how it is related to
the ordinary one, different responses to the insoluble are appropriate.
Heytesbury formulates his position in five rules governing a respondent's
replies when insolubles emerge in a disputation de obligationibus.52
One of Heytesbury's claims proved to be especially controversial: when
the opponent in a disputation does not specify an insoluble's additional
signification, the respondent does not have to specify it either. Heytesbury
was in effect refusing to say how the signification of a sentence in insoluble
circumstances differs from its ordinary signification. While by the strict
rules of obligationes, his silence is justified, many authors felt Heytesbury
had simply failed to answer the most obvious question his theory raised. 5 3
Some tried to answer it for him by saying that insolubles, in addition to
their ordinary signification, signify that they are true. 5 4 This move brings
together the two traditions stemming from Heytesbury and Bradwardine.
Heytesbury's claim, that sentences in insoluble circumstances do not
signify what they ordinarily do, has important consequences. For the
signification of spoken and written sentences, like that of terms, depends
on the voluntary adoption of conventions. If sentences signify differently
in insoluble circumstances, then it must be because a new convention has
been voluntarily adopted. But in fact that seems rarely if ever to be what
happens. Heytesbury considers this objection, and admits he has no satis
factory answer. 5 5 Again, for most authors, mental language signifies by
51. See e.g., Spade 1975a, items V, VII, VIII, XII, XIII, ΧΧΙΙΙ (all anonymous), XXV (Angelo of
Fossombrone),XXVIII (Gaetano diThiene),XXXV (John ofConstance).XXXVI (John Dum-
bleton), XXXVIII (John of Holland), XXXIX (John Hunter), XLII, (John of Wesel), XLIII
(John Wyclif), XLVIII (Paul of Pergula), L (Paul of Venice), LIII (Ralph Strode), LVIII
(Robert Fland).
52. See Spade 1975a, p. 118.
53. See. e.g., Spade 1975a, pp. 24-5.
54. See, e.g.. Spade 1971 and Spade 1975a, items XII (anonymous), XXXVIII (John of Holland), and
XXXIX (John Hunter).
55. Heytesbury 1494b, Regulae, f. 6*: 'Multae possent fieri hujusmodi objectiones contra istam
responsionem quibus esset difficile vel impossibile in toto satisfacere.'
Insolubilia 253
nature, not by convention. Hence mental sentences always signify the

same way, never other than as they 'ordinarily' do. It follows from
Heytesbury's theory, then, that there are no insolubles in mental language.
Resolutions in terms of the notion of mental language

It was perhaps by speculating on this fact that Gregory of Rimini and Peter
of Ailly were led to base their own theories of insolubles on the notion of
mental language. Little is known of Gregory's view,56 but Peter in his
Conceptus et insolubilia held that there are no insolubles in mental language,
and that spoken or written insolubles are ambiguous sentences insofar as
they correspond not to one but to two distinct mental sentences, one true
and the other false.57 Peter's view was subtle and complex; it had some
influence in the sixteenth century.58
With the exception of Peter of Ailly's and related views, the major
theories of insolubles after ca. 1350 were elaborations or modified forms of
the theories of Bradwardine, Swineshead, and Heytesbury. There were
other views, of course, throughout the history of the insolubilia-literiture;
but these were the main ones.
Conclusion
Certain general features characterised this literature. First, there was a
tendency to multiply examples. Albert of Saxony, for instance, in Perutilis
logica, 6, 1, discusses no fewer than nineteen insolubles. Contrast the
modern tendency to look for a paradigm case that shows the structure of
the paradox with a minimum of inessentials. Second, the medievals did not
seem 10 have had any 'crisis mentality' about these paradoxes. Although
they wrote a great deal about them, there is no hint that they thought the
paradoxes were crucial test cases against which their whole logic and
semantics might fail. Again, contrast the modern attitude. Third, the
medievals did not draw great theoretical lessons from the insolubles. They
did not seem to think the paradoxes showed anything very deep or
important about the nature of language or its expressive capacity. Once
again, contrast modern attitudes. One might do well to speculate on the
reasons for these differences between medieval and modern semantic
theory.
56. Spade 1975a, pp. 54-5. But see Peter of Ailly 1980, pp. 6-7 and 11-12 for a tentative
reconstruction.
57. See Spade 1975a, pp. 84-5 and Peter of Ailly 1980, pp. 11-12 and Chs. 3-4.
58. Ashworth 1974, pp. 108-10.
13
SPECULATIVE G R A M M A R
Grammar in the early Middle Ages

Medieval speculative grammar grew out of the schoolmens' work with
ancient Latin grammar as it had been transmitted in the canonical works of
Donatus and Priscian. The efforts of early medieval glossators were directed
towards explaining the authoritative texts, towards systematising the de-
scriptional apparatus used by the authors, and towards harmonising the
apparent or real contradictions which arose in a comparison of the gram-
matical and logical traditions.' The results of their combined efforts were
summarised in the famous Summa super Priscianum compiled about 1140 by
Peter Helias. 2 The grammarians' discussions, which had been influenced by
the logicians, in their turn influenced and refined logical doctrine and
played an important role in the emergence of the specifically medieval
logical doctrines known collectively as terminist logic. The grammatical
discussions about the meaning of substantive words, for instance, were
crucial to the development of the theory of supposition. 3 The twelfth-
century grammarians emphasised the importance of explaining linguistic
features causally, instead of just describing them as Priscian had done, 4 and
in this way attained a high degree of linguistic sophistication. But it would
perhaps be too much to say that their efforts already inaugurated a new
paradigm of linguistic description.
The general nature of'modistic' grammar

Around 1270, however, a new theoretical framework was established. The
phases of the development which brought this about are not yet known in
1. See Ebbesen's and Henry's contributions to this volume.

2. For further discussion of the early glosses see Thurot 1868; Hunt 1949-50; De Rijk 1962-7,11(1),
PP- 95-I25; Fredborg 1973; Kneepkens 1977 and 1978.
3. See De Rijk 1962-7.
4. See William of Conches, De philesophia mundi (Thurot 1868, p. 17): 'Quoniam in omni doctrina
grammatica praecedit, de ea dicere proposuimus, quoniam, etsi Priscianus inde satis dicat, tamen
obscuras dat definitiones nee exponit, causas vero inventionis diversarum partium et diversorum
accidentium in unaquaque pretermittit.'
Speculative grammar 255
detail, but the first representatives of the new doctrine seem to be Boethius
of Dacia and Martin of Dacia. 5 In their works we find a coherent linguistic
theory, in which every grammatical feature treated is fitted into a single
descriptional framework, based on expressly formulated premisses. The
attitude governing this endeavour has several points of resemblance with
later types of rationalistic or universal grammar, including some
twentieth-century developments, but it is often more explicit. 6 The
medieval theory is based on the concept of meaning (significatio), inherited
from ancient tradition, and develops its consequences in a comprehensive
and pertinent way. The theory has been labelled 'modistic' grammar from
the concept of modus significandi, which is a central concept of the theory
since it is the very term used to describe the changes of meaning imported
by grammatical features. The term 'Modistae' is used accordingly to
denote the (mostly Parisian) masters of the late thirteenth and early four-
teenth century who wrote on grammar, logic, and metaphysics within this
tradition. 7
The influence of the new Aristotle and the Arabic commentators

The most important factor for the development of modistic theory is the
recovery of the whole Aristotelian corpus, especially the Posterior Analytics,
the Metaphysics, and the De anima, with their strong requirements for the
construction of a scientific theory and their more complex semantic doc-
trines based on an elaborate epistemological foundation. These features of
the Aristotelian writings were emphasised by the Arabic commentators,
and it is no accident that the writings of the Modistae are filled with
quotations from Avicenna and Averroes. The claim — read out of
Aristotle's Posterior Analytics — that any discipline worthy to be called a
science would have to treat features of reality which are universal and
immutable had especially far-ranging repercussions.
Grammar as a speculative and auxiliary science

Such reflections determined the course of grammar when grammarians
wanted to raise it to the status of a science and situate it within the medieval
system of sciences. It was accordingly determined to be a speculative and
auxiliary science: 'speculative' (i.e., theoretical) because its goal was not to
teach language but to describe and explain the nature and organisation of
5. Pinborg 1967a, pp. 19-21 and 6off.

6. See Trcntman 1975.
7. For a list of authors and information on MSS see Pinborg 1975a.
language (in this case Latin) as the most important and convenient vehicle
of communication; 8 'auxiliary' because grammar, like logic, was not
direcdy concerned with the world, but with the reflection of it in our
descriptions. 9
The modi significandi

Since vocal expressions obviously differ from one language to another,
they cannot constitute the true objects of grammar. The obvious place to
look for universal features of language is in the semantic component, but it
is not the meanings of individual words which prove to be relevant to the
grammarian. Grammar is traditionally concerned with more general
matters, namely, types of words and their constructions. Accordingly, the
interest of the Modistae was concentrated on what was sometimes called
the general meanings of words (significata generatia). These comprise all
such components of meaning as constitute grammatical categories - e.g.,
nouns, verbs, cases, or tenses. All such components were described as modi
significandi.
The career of speculative grammar

Among the later authors of modistic grammar the most important is
perhaps Radulphus Brito, who provided the most comprehensive discus-
sion on the principles of speculative grammar after Boethius of Dacia, one
which took into account all the doctrinal refinements introduced in the
meantime. Shortly after 1300 Thomas of Erfurt wrote a manual that
included the new theoretical developments and replaced the work of
Martin of Dacia. This text, long attributed to Duns Scotus, remains the best
known treatise of modistic grammar.
After 1300 no original contribution to modistic theory was made,
although modistic terminology continued to govern grammatical descrip-
tion. The theoretical framework of the theory was made the target of sharp
attacks, directed especially against the epistemological presuppositions of
the theory, the proliferating subtleties of which could not conceal the
theory's serious intrinsic difficulties, particularly in connection with the
analysis of the relationship between language and reality. The critics
included both nominalists like Ockham and Buridan and conservative
8. Martin of Dacia, Modi significandi, 10 (1961, p. 7): 'Et sciendum quod vox per accidens con-
sideratur a grammatico. Quia omne quod potest esse signum rei significatae etiam potest esse de
consideratione grammatici. Sed quia vox est habilius signum quam aliquid aliud, utpote nutus
corporeus et conniventia oculorum et huiusmodi, ideo plus consideratur a grammatico; et
intelligendum quod hoc est per accidens.'
9. John of Dacia, Divisio scientiae (1955, pp. 34-5).
Averroists like John of Jandun and John Aurifaber; their fundamental

charge against the Modistae was that they had confused linguistic distinc-
tions with real ones. So medieval speculative grammar gradually lost its
position of dominance, even if parts of the theory were never wholly
abandoned, even by the Humanists. 10
The linguistic doctrine of the Modistae1 '
According to modistic analysis words consist of a phonological element

(vox) and two levels of semantic components, one concerned with specific
or lexical meanings (significata specialia), the other with more general
meanings, called modi significandi, on which in turn the syntactical com-
ponent depends. Since the coupling of expression and meaning is arbitrary,
it presupposes a deliberate act by which it is brought about, an impositio
associating an expression with an object or content. A distinct act cor-
responds to each level of meaning. By a first imposition the expression is
connected with a referent, insofar as a name is instituted to refer to a
definite object or attribute of an object. How this happens is almost never
discussed in any detail. 12 The relation holding between the expression and
the object referred to is called the ratio significandi. It is often described as the
'form' which turns a mere sound into a lexeme (dictio). (The term 'lexeme'
seems to be the nearest modern equivalent to the modistic dictio, which is a
highly abstract term, including under one head not only all flexional forms
and all occurrences of a word but even derivational forms.)
In a secondary imposition the lexeme receives a number of modi signifi-
candi which determine the grammatical categories of the word. 1 3 They are
not bound to any specific phonological element. The modus significandi of
grammatical case can be expressed by different terminations, and the same
termination can express more than one modus (e.g., in the second declension
'-us' represents nominative case, singular number, and masculine gender).
The relation of the modi to the lexical meaning is not wholly overt. A given
lexeme can be associated with different modi, so that the same lexeme may
be realised as different parts of speech and as different grammatical forms. A
favorite example of the Modistae is the lexeme connected with pain, which
according to the general feature consignified may be realised as a noun
10. See Percival's contribution to this volume.

11. For further particulars sec Pinborg 1967a and 1972; also Bursill-Hall 1971. In rhc following
sections I have tried to use a terminology fairly consistent with the one advocated in Lyons 1977.
II. A notable exception is John of Dacia 1955, pp. I77ff·, who is, however, dependent on a pre-
modistic source, viz. Robert Kilwardby 1975.
13. Sec Boethius of Dacia, Modi sign., (1969) qu. 114
(dolor) — this again in various cases and numbers — as a verb (doleo), a

participle (dolens), an adverb (dolenter), or even as an interjection (riew).14
Obviously then, the modi are a kind of semantical modifiers, further
determining the lexical meaning of the dictio, thus preparing it for various
syntactical functions. The ontological counterparts of the modi are some
general features of the object signified, which do not belong to the
definitional properties of the object, and accordingly are not signified
directly, but only implied. Thus when I use the form 'homo' to talk about
man I signify a human being and imply that he is an agent (because of the
nominative case). According to modistic terminology the general features
are 'consignified', i.e., signified along with the specific, lexical meaning.
Therefore, the relation between the expression and the ontological
counterparts of the modi is called the ratio consignificatidi, a term which is
used synonymously with modus significandi; it is described as the 'form'
which turns a lexeme into a part of speech (pars orationis), i.e., prepares it to
perform a particular function as a segment of a linguistic string.
Of the modi significandi attributed to a lexeme one is essential (essentialis),
namely the mode which determines the subsumption of the word under
the most fundamental grammatical categories: the eight parts of speech
taken over from ancient grammar. Some Modistae divide this mode
further into a modus generalis 'which is common to several parts of speech
and a modus specificus which defines one and only one part of speech. Thus
they arrive at a paradigm like the following, where R o m a n numerals
indicate general modes, Arabic numerals specific modes:
I. -f- stability (modus habitus et quietis or modus substantiae)

1. + definite reference (modus determinatae apprehensionis or modus qualitatis)
NOUN (substantival and adjectival)
2. — definite reference (modus indeterminatae apprehensionis) PRONOUN
II. -f- change (modusfiuxuset fieri)
3. -f- predicability or separability (modus distantis) VERB
4.. — predicability or separability (modus indistantis) PARTICIPLE
III. + modification (modus disponents)
5. + act-determinant (modus determinantis actum) ADVERB
6. + referring object to an act (modus retorquendi substantiam ad actum)
PREPOSITION
7. + uniting (modus uniendi) CONJUNCTION
5
8. + emotionally affecting the mind (modus afficientis animam) INTERJECTION1
14. See ibid. qu. 14.

15. See Pinborg 1967a, pp. 125-6.
Besides the essential mode, lexemes are endowed with a variant number
of further modes. Some of them (the modi speciales) sort the part of speech
into semantic subcategories (e.g., proper names), others endow it with less
basic grammatical features, such as case, number, or tense. These are called
modi accidentâtes, and their number and to some extent their definitions
correspond to the accidents of traditional grammar. ' 6
Thus each wordform is a bundle of semantic and syntactical features.
Two forms can be combined in speech only when at least one feature or
mode of one form (now called a constructibile) is related to a mode of the
other form in either of two ways: The two forms may have identical
modes, but such that the one possesses it only because of the other, as e.g.
the adjective receives gender, case, and number from the noun to which it
is attributed (this case corresponds to 'concord' in traditional grammar); or
the mode of one may be proportional (proportionalis) to a mode of the
other, such that the one specifies a relation, by which the other is specified
(this case corresponds roughly to 'government' in traditional grammar). 1 7
In both cases, however, the relation between the two forms is one of
dependency. One form has a semantic feature which depends on or
presupposes a feature in the other form, which is thus viewed as completing
the dependency (being the 'terminant'). 1 8 Unfortunately, the notion of
dependency seems to have been intuitional for the Modistae: no purely
linguistic rules are formulated in modistic grammar which help in deciding
which of two semantic features is dependent on the other. A look at the
examples of dependency which we actually find stated may help to eluci-
date the nature of modistic 'dependency': (the arrows denote dependency
and its direction) Socrates «- currit (Socrates *- runs); percutit —• Platonem
([he] strikes —• Plato); Socrates «- albus (the white —• Socrates);
misereor —> Socratis (I pity —> Socrates); currit «- bene ([he] runs «- well); est
«- homo ([he] is «- [a] man) ; cappa *- Socratis (Socrates' -*• cape). It may be
difficult for us to discover one and the same relation in all instances, but the
main intuition seems to have been that of an accident or a relation
dependent on or being predicated of a substance (something per se stans).
This interpretation covers the first four examples without difficulty; for
16. For details see Bursill-Hall 1971, pp. I33ÎF. and 391.
17. For further discussion of the syntactical theory of the Modistae see Pinborg 1972, pp. 120-6;
Pinborg 1973a and 1973b; Siger of Courtrai 1978, pp. xxixff. - For an intcresring comparison of
modistic doctrine with the extended standard model of transformational grammar, especially
based on the analogous role of'syntactical features', see Coleman 1971, pp. I05ff.
18. See Martin of Dacia, Modi significandi 203—11 (1961, pp. 90-4); Anon., Quuesf. in Prise, minor, (MS
Nürnberg, Stadtbibl. Ccnt.V.21, f. 36"): 'Omne determinans sive specificans praesupponit suum
detcrminabile sive spccificabilc et dependet ad ipsum.'
2öo The high middle ages: semantic theory
the last three further explanation is required. Currit <— bene is anomalous in
the sense that both forms signify an accident; the adverb, however, de-
termines the action signified by the verb and can thus be said to be
dependent on it. In the same way homo in est homo specifies the kind of
being signified by est. Cappa <— Socratis is difficult since both wordforms
signify a substance and since Socratis would seem to determine cappa rather
than the other way around. However the construction is interpreted in the
same way as misereor Socratis and the genitive is seen as completing one
semantic feature of cappa, viz., the feature of belonging-to.
This model of description as presented so far does not permit us to
distinguish different types of constructions. Medieval grammarians, how-
ever, did distinguish two main types of constructions: transitive (e.g., the
Latin examples above in which the arrow points to the right) and intransi-
tive (e.g., the examples in which the arrow points to the left). In order to
establish this distinction the Modistae had to introduce a further syntactic
relation. They came to distinguish a dependency ex parte ante from a
dependency ex parte post. This distinction implies a relation of priority
within the sentence. If the dependent wordform is prior to the terminant
we have a transitive construction; if the terminant is prior we have an
intransitive one. But what does 'prior' mean here? Obviously it has
something to do with word-order, but not necessarily with the arbitrary
word-order of the surface-structure of a Latin sentence. Radulphus Brito
makes this point explicitly when discussing the construction quern video:
Even if quern is expressed first in the surface structure, it is nevertheless
posterior when we consider the nature of its modus significandi (the accusa-
tive case). 19 Here a natural (or logical) word-order seems to be presupposed
which somehow reflects the order of nature: in this order the agent must
precede the action, the substance its accidents, the action its objects, etc. 20
By means of the two concepts of dependency and natural word-order
the Modistae succeed in deducing the whole system of possible Latin
constructions. It is fundamentally a system based on wordforms and their
potential constructions, not an analysis of constituent structures of sen-
tences, and accordingly it is not very efficient in analysing propositional
relations. 21 But despite these shortcomings it is a major achievement: the
first systematic syntax developed in Western linguistics.
19. See Radulphus Brito 1980, I, qu. 44: 'Dico quod [quern] construitur cum verbo a parte post
realiter, tamen vocetenus construitur a parte ante.'
20. See Priscian, Inst, gramm. XVII, 105 (1855, HI, p. i64.l6ff.). Here as in the Modistae the ordo
naturalis is presupposed and exemplified, but not described exhaustively.
21. According to Boethius of Dacia 1969, qu. 132, propositions as such have no modi significandi and
accordingly cannot be construed in the strict sense. This conclusion is modified by Radulphus
We have seen that logical and ontological considerations play a con-

siderable role for the grammatical analysis of the Modistae. But they did
not forget that they were dealing with language, not reality. They were
not so naive as to think that every sentence corresponded to the structure of
reality. They recognised for instance that the subject of the process de-
scribed and the subject intimated by the proposition describing it could
very well be different. 22 What mattered to the grammarian was the way
reality was described, not reality itself. O r in the terminology of the
Modistae: the grammarian was concerned with congruity of speech, not
with truth.
The discovery-procedures of the Modistae

How, according to the Modistae, do we discover the existence of semantic
features in speech? Since the answer to this question may help us in assessing
the nature of speculative grammar, it deserves closer scrutiny. Radulphus
Brito distinguishes four discovery-procedures:
(a) a priori from the nature of the object signified: if the object has a certain modus
essendi the word may have the corresponding modus significandi.
(b) a posteriori from the constructions into which the word actually enters: the
wordform must possess the modus significandi necessary for the construction.
(c) from the modi significandi of the wordform with which it is construed: if the one
wordform has certain modes, the other form must have proportional modes.
(d) by the authority of the classical grammarians.23
These discovery procedures tell us something about the nature of modis-
tic theory. It was based on observation of linguistic facts, often as inter-
preted in the grammatical tradition, which were then fitted into a descrip-
tive framework of great consistency. The observations were not limited to
the surface-structure of words and sentences, but involved consideration of
'deeper' semantic features. The aim of speculative grammar was to describe
intra-linguistic relationships, but the Modistae could not accomplish what
they wanted without invoking to some degree the structure of reality. This
actually was their warrant for believing that their results were scientific and
universal. Accordingly it becomes crucial to investigate the epistemolo-
gical presuppositions of their doctrine.
Brito 1980,1, qu. 70, and Siger of Courtrai 1978, pp. 60-1. Butin their view, too, propositions
are connected only in virtue of the semantical relations holding between their terms. This view
makes problems for the theory of consequences, and especially for the development of a
propositional logic.
22. See Radulphus Brito 1980, II, qu. 4; Pinborg 1972, pp. n - 1 2 .
23. See Radulphus Brito 1980,1, qu. 30.
The semantics of the Modistae

The semantic status of the modi significandi was described within a triadic
system: the modi significandi correspond to modi intelligendi, which in turn
correspond to modi essendi. This is consonant with the traditional interpre-
tation of Aristotelian semantics - to be found, e.g., in Boethius, Avicenna,
Averroes, Albert the Great, and Thomas Aquinas - according to which
words signify concepts which in turn represent the objects signified.
We find no definition of modus essendi in modistic literature, although it
obviously is a term essential to their doctrine. This may result from the fact
that the theory was centred around the modes of signifying, whereas the
modi essendi were simply introduced as the necessary ontological counter-
parts of the former. A description can, however, be constructed from stray
remarks in the logical writings of the Modistae. 24 The modi essendi (or
proprietates rerum) are accidental properties of the objects, as distinct from
their substantial forms. It is, however, such properties which give rise to
our concepts. The Modistae agreed with Aquinas in assuming that we can
attain knowledge even of the essence of objects through their accidental
properties - a doctrine which was to be challenged by Peter Aureoli and
John of Jandun. 2 5
According to the Modistae two kinds of modi essendi must be taken into
account: proper and common modes. Proper modes are those which belong
to the object considered in itself and give rise to first-order descriptions;
common modes are those which serve as the source of a description in terms
that are general and sometimes are second-order predicates: the properties
involved are all such that they do not characterise items of one category
only, but can be used indiscriminately of items of all categories. From such
common properties the logicians derive their second intentions and the
grammarians their modes of signifying. It is tempting, then, to identify
these properties with second-order features, but this interpretation is too
narrow: according to the logician genus and species are not to be explained
only as concepts of concepts (and thus as second-order terms): they reflect
features of the object to which they refer. Not every item in the world can
be described as a genus, but only such items the nature of which can be
24. Most logical texts of the Modistae are so far unpublished, the only exceptions being the
commentaries on the Organon by John Duns Scotus, Boethius of Dacia's questions on the Topics
(1976), the questions on Logica vetus by Simon of Faversham (1957), and two sets of anonymous
questions on the Sophistici elenchi (Ebbesen 1977a). Further fragments of texts are to be found
scattered in various publications, especially Ebbesen & Pinborg 1970; Ebbesen 1977c, 1979;
Pinborg 1971, 1973a, 1973b, 1975a, 1975b, 1976a; Roos 1977.
25. See Pinborg 1975c.
predicated of different species.26 In a similar way the modes of signifying

are not purely arbitrary; their relation to the objects is nevertheless less
precise than that of the second intentions. Speaking in terms of the modi
significandi, one may even be allowed to describe objects as having pro-
perties which they do not, strictly speaking, possess. Even non-existent
objects may be denoted by modi significandi, which of course only reflect
our way of imagining such objects. Hence we can draw an important
corollary: the fact that an object is signified by a noun does not imply that
the object signified belongs to the category of substance; it does, however,
imply that we conceive of it as if it were a distinct object, somehow
analogous to a primary substance. Thus a mode of signifying need not have
an ontological counterpart in the object signified, as long as the combi-
nation of the object and the general feature signified by the mode is not
inconceivable.27
This association of the modes of signifying with modes of conceiving of
things explains the necessity of positing modi intelligendi between the modi
essendi and significandi. Even if words are taken to signify extramental
objects they do so sub modo intelligendi. The following example may help to
make this more precise. Conceiving of man we conceive of something
which possesses the proper mode of being human (humanitas), which can in
turn be analysed into being alive and self-moving (animalitas) and being
rational (rationalitas); besides we find in man some common properties,
e.g., that of being predicable of different individuals (specialitas) and that of
being an autonomous entity (persestantid). Each of these modes can of
course be conceived of separately and abstractly; they can also be conceived
of and signified concretely, giving rise to concrete first-order concepts and
terms such as 'man', 'animal' (or 'living being'), 'rational', and - relative to
the common modes - such second-order concepts and terms as 'species'
and 'noun'. 28 The mode which is directly signified is said to be the mode
under which the term is imposed to signify. A first-order term such as
'man' thus signifies under the mode of'being human' (humanitas) and refers
to (denominat) all human beings who participate in this nature. The
common modes observed in man are not signified directly, but they are
nevertheless modes under which the object is also understood or consigni-
fied. In this way the term has both a direct signification 'under' which it is
26. See Pinborg 1975b; Boethius of Dacia 1976, pp. 10 and 183-95.
27. See Boethius of Dacia 1969, 17-20; Radulphus Brito 1980, I, qu. 21.
28. See Radulphus Brito's sophisma 'Aliquis homo est species' in Pinborg 1975b; see also Pinborg
1975a.
imposed (ratio significandi), implied meaning (ratio consigniflcandi), and

reference (denominatio).29
The status of meanings is, however, still not exactly determined. Even if
they are obviously bound to words since they are what make words signify
what and as they do, and thus are not properties of the objects, they can
nevertheless not be identified with any specific part of the vocal expression.
Neither can they be said to be a property of the intellect; for the expressions
as such are real entities, and how could something existing in the intellect
determine real entities? The later modistic solution to this problem was to
distinguish between active and passive aspects of meaning. The active
aspect consisted of properties of words, imposed by the intellect, whereas
the passive aspect consisted of properties of the objects as conceived of or
signified (i.e., sub modo). Hence, according to Radulphus Brito the active
and passive aspects are formally identical, since they are distinguished from
all other beings by the same relational form, namely, the relation holding
between the word and the property actually signified or consignified. But
they differ materially since the active aspect concerns attributes of words,
while the passive aspect is concerned with properties of objects referred to
(denominated) by words. Sometimes Radulphus even talks about 'aggre-
gates' of the extramental property and the ratio significandi.30
The entire modistic theory of semantics obviously belongs to a type of
semantics in which sense, not reference, is the focal point. This distinction
is described as the distinction between 'formal' and 'material meaning'.
'Formal meaning', often also called virtus sermonis, is associated with
imposition and comprises all that belongs to the word in virtue of its ratio
significandi and modi significandi; 'material meaning' covers also the denoted
or implied referents (the passive aspect), and unlike formal meaning intro-
duces questions of applicability, and, in propositional contexts, of truth and
falsity.31 Such questions are considered accidental to the study of formal
29. See Radulphus Brito, Quaest. Peri hermeneias in Pinborg 1971, p. 276: 'Essentia rei est quod
intelligitur et per consequens est illud quod significatur per vocem primae impositionis. Est tamen
notandum quod istac voces significant illas res sub aliquibus conceptions, sicut " h o m o " et
" a n i m a l " significant sub aliqua ratione intelligendi. Et iliac rationes intelligendi sunt conccptus
quidam existentes in anima. Et ex hoc sequitur quod consignificant istos conceptus. U n d c dato
quod non significantur per istas voces, tamen dantur intelligi.'
30. See Pinborg 1967a, pp. 85ff., 109fr.; 1975a. In his logical writings Radulphus Brito talks in terms
of abstract and concrete intentions instead of acrive and passive modi intelligendi. In both cases,
however, a modus essendi conceived under a concept (active or abstract) is called a passive mode or
a concrete intention.
31. The terms 'formal' and 'material meaning' are used by Peter of Auvergne (Pinborg 1973a, p. 52).
' Virtus sermonis' is c o m m o n in all texts. Cf., e.g., Simon of Faversham (Pinborg 1975a, p. 69): 'In
locutione sunt duo consideranda, scilicet virtus et Veritas locutionis' or the anonymous author of
the questions o n the Elenchi (Ebbesen 1977a, p. 107): 'Ilia autem quae de virtute sermonis sunt in
termino sunt sua significata et modi significandi.'
meaning. This of course reflects the correct assumption that the meaning
(or sense) of a term determines its range of application (or reference).32
Endeavouring to find invariables of language, the Modistae tended to
establish a rather static interpretation of formal meaning. Even if the
imposition through which words originally receive their meaning is ar-
bitrary, once this imposition has occurred the meaning cannot change.
This holds true also for the accessory meanings conveyed by the modes of
signifying. As a result, terms and clauses which may be interchanged
without affecting the truth-value of a proposition, such as 'Brunellus' and
'this donkey', assuming 'Brunellus' to be the name of the donkey in
question, may nevertheless have different formal meanings.33
This notion of meaning has some obvious advantages for the construc-
tion of a linguistic theory. For this reason the speculative grammar of the
Modistae is a valuable attempt to systematise a universal semantic approach
to language, leading to a high degree of sophistication and adequacy in
linguistic description. The success of this doctrine as regards problems of
logical analysis depends very much upon its ability to account for 'material
meaning' and the various acceptations of words in different contexts. In a
manner not easily describable in modistic terms, the material meaning of
the term, or its reference in a specific context, can influence the acceptation
of the term in a way which changes the conditions for the verification of
the proposition. Thus even if'man' retains the same formal meaning in the
two propositions 'man is a species' and 'every man is an animal', the
propositions nevertheless have genetically different presuppositions or
causae veritatis. The type and range of referents to be investigated before the
truth or falsity of the proposition is determined are accordingly restricted
or ampliated in comparison with the normal range of the term. In this way
potential acceptations or functions of the term, included somehow in its
formal meaning, become actualised. This actualisation is effected by the
addition of attributes, or even to a certain degree, by the addition of
predicates.34
Curiously enough, the Modistae found no use for the terminists' theory
32. The usual way of phrasing this assumption is: 'Terminus communis suppoint significatum pro
suppositis' (Sec Pinborg 1975a).
3J. See e.g. the anonymous Quaest. in Prise, minor, (MS Nürnberg, Stadtbibl., Cent.V.21, f. 39*): 'Dico
quod possunt aequipollere [sc. Brunellus et iste asinus] quo ad significatum [used in a loose way,
probably equivalent to significatum materiale], sed non quo ad modum significandi. Quantum-
cumquc enim pronomen demonstrativum contrahat ly "asinus" ad determinatum suppositum,
hoc tamen non aufert sibi modum significandi. Voces enim non cadunt a suis signifîcatis nee
consignificatis.' See also Pinborg 1972, pp. 111-12.
34. See Boethius of Dacia 1976, II, qu. 11. Radulphus Brito, Qu<j«f. Priora analyt. 1, qu. 46 in Pinborg
1976a, pp. 272-5; Anon., Quaest. Soph, elenchi, qu. 48, in Ebbesen 1977a.
of supposition to solve problems of this type. The fourteenth century

would prove the superiority of the terminists' approach, especially when it
was combined with the doctrine of exposition, which could handle presup-
position and explicate the causae veritatis with a detailed terminology and
within a system of rules. It is difficult to explain why the terminist approach
was neglected by the Modistae, since it does not necessarily contradict
modistic doctrine. Perhaps the elaborate semantic system inherited from
Aristotle and the Arabs at first looked so impressive in itself that the
previous form of logical interpretation was thought to be dispensable.
The fundamental epistemological difficulty of modistic semantics is the
same encountered by any picture theory of cognition or of meaning,
namely: how can properties of spoken words 'correspond' to mental acts
and through them to external objects? N o division into active and passive
aspects will ever provide the necessary tertium comparationis. At the high
price of accepting as mediators the curious aggregates of real and mental
properties termed the passive aspect, the Modistae avoid committing
themselves to a third and independent 'level' of meaning. Others (e.g.
Walter Burley, Hervaeus Natalis and Peter Aureoli) were more prone to
accept the existence of 'objectively' existing conceptual contents, to be
distinguished both from the objects and from actual mental acts. To them
this assumption was the only means of warranting intersubjective know-
ledge. But apparently this departure did not meet with any greater success,
except within the discussions of propositional meaning. 35
Alternative approaches: Roger Bacon

Modistic grammar and semantics did not meet with unanimous approval.
Roger Bacon is often alluded to as a proponent of speculative grammar,
and we owe to him one of the best known formulations of the basic
principle of universal grammar. 3 6 Nevertheless, he was not influenced by
the fully developed modistic doctrine and actually he reacted strongly
against some of its main tenets. He had a strong predilection for empirical
grammar, as is well known, but he also supplied a theoretical framework
for language studies, partly adapted from Augustine's doctrine of signs.
This is best known from his Compendium studii theologiae (1292), but it was
already developed in 1267 when he wrote the treatise De signis.31
35. See Kretzmann 1970; Nuchelmans 1973; see also Nuchelmans' contribution to this volume.
36. See Roger Bacon 1902, p. 27: 'Grammatics una et eadem est secundum substantiam in omnibus
unguis, licet accidentaliter varietur.'
37. See Fredborg et al, 1978. See also A. de Libera, forthcoming.
Bacon emphasises that human utterances always belong to a context.

The meaning of words is not unchanging; a new imposition can always
supersede earlier ones. If all new impositions were explicit, there would be
no difficulty; but most of them are tacit and not even the result of a
conscious act on the part of the speaker. Moreover, they are often con-
cealed by the fact that the new sense of a word is related to an earlier sense.
These facts emphasise the necessity of determining the particular sense in
which a word is used in a particular context. In consequence of this theory
of new impositions, Bacon can explain all meaning-differences which
influence the truth-conditions of statements as cases of homonymy
(equivocation), without involving himself in any of the trouble arising
from the rigidity of'formal meaning'.38
fohn offandun and John Aurifaber

Bacon's approach differs significantly from the standard modistic ap-
proach, but thefirstdirect attack on the theoretical foundations of modistic
theory was launched from another direction. Sometime around 13 30 John
Aurifaber, an Averroist master at Erfurt, gave a public determination in
which he undertook to prove the non-existence of the modi significandi.39
Drawing his inspiration from the Parisian master John of Jandun, he
propagated a purified Aristotelianism which defied all superfluous innova-
tions of the schoolmen. According to Aurifaber, it suffices to know the
distinctions of objects and concepts in order to describe how language
works. Words are just secondary signs used by the intellect to com-
municate its concepts and express the distinctions already formed. There is
no need to posit any intermediate entities such as the modes of signifying in
order to explain the different ways in which words signify. Nouns and
verbs are distinguished, not in virtue of any inherent semantical features,
but because they are used to signify different classes of objects. Some
changes in the vocal substratum may occur (such as inflection and deriva-
tion), but either they have no semantical relevance, or they substitute a
different, though related concept. Similarly, words are construed with
other words because they are subordinated to relational concepts. On this
view every conceptual distinction has an ontological counterpart, a doc-
trine that would seem to commit Aurifaber to a great many real, abstract
38. Roger Bacon, De signis (ed. Fredborg et al. 1978), especially §§143-61.
39. Aurifaber's text is edited in Pinborg 1967a, pp. 215-32. For the interpretation of Aurifaber as an
Averroist see Pinborg 1975a, 1975c.
entities; but, like John of Jandun, he would probably not be deterred by

such a consideration.
An important corollary of this approach to language is that any specula-
tive grammar founded on inherent semantic features is made impossible.
Only one task remains for the grammarian: to assert the positive rules of
actual usage in the different languages; the task of constructing a universal
grammar is reserved for the logician or metaphysician. 40
Walter Burley and William Ockham

A similar evaluation of the modi significandi is hinted at by Walter Burley. 41
The nominalists also agreed with Aurifaber in an interesting way, although
of course without committing themselves to his strong realist assumptions.
To William Ockham the term 'modus significandi' is nothing but a meta-
phor, the literal basis of which is the fact that words signify or connote
different classes of objects. 42 Since words in signifying and connoting
objects are subordinated to concepts, we need not bother with the modi at
all; it is enough to know the objects and the mental language of concepts
used in describing them. The use of language, emphasised in Ockham's
account, differs from the modi in that it does not change the properties of
the objects signified.*3 Language, in his view, establishes many distinctions
which have no semantic relevance at all. Such distinctions are not reflected
in mental language. But the disparity between language and concepts goes
even deeper: Where spoken or written language uses one word - e.g. a
verbal noun - mental language sometimes has to use a proposition. The
true interpretation of a spoken or written proposition consists in reducing
it to the mental proposition or set of propositions which explicates the
conditions for verifying it. 4 4
Peter of Ailly
Peter of Ailly applied this approach to language in an efficient attack on the
grammar of the Modistae, in which he tried to reduce all grammatical
40. See Aurifaber (Pinborg 1967a, p. 231.271!.): 'Unde nunc dico quod [grammaticus et logicus]
considérant easdem partes diversimode: logicus quidem ut ilia consideratio videtur esse com-
munis omni linguae; grammaticus autem non sic universaliter, quia non considérât ut sint
communia omni linguae, cum aliqua considérât grammaticus (congruentia quae) lingua latina
tantum invenitur, et iste grammaticus graecus congruentia linguae graecae.'
41. See Pinborg 1975a, p. 60 (Burley, In Peri herm. ad i6'28); Brown 1973, p. 254.
42. William Ockham 1974a, OP I, III, 4, 10.
43. William Ockham 1957b, pp. 14-15; 1977a, OT III, I, d. 4, q. 1, pp. 9-11.
44. William Ockham 1974a, OP I, I.3, et passim.
distinctions to the relevant distinctions within mental language.45 But

even if the nominalists in this way renounced the theoretical framework of
modistic grammar, they accepted the fact that grammatical analysis was
carried on with modistic terminology, except for the proviso that 'modus
significandi' had to be understood in the right way, i.e. as a metaphor that
had to be interpreted in terms of mental language.46
Modistic theory in the revived via antiqua

The revival of the via antiqua (the revived Thomist, Albertist, and Scotist
schools) in the late fourteenth and the fifteenth century also entailed an
increased interest in modistic theory, even if no theoretical innovations
have been recorded. Thus central parts of the linguistic theory of modistic
grammar remained in use and could be repossessed by later universalist
grammarians, even if the new approach to grammar advocated by the
humanists dislodged the doctrines from their context within medieval
philosophy.47
45. See Pinborg 1967a, pp. 202-7.

46. See Pinborg 1967a, pp. 208-9; Heath 1971, pp. 56-7.
47. See Heath 1971, pp. 40-1.
ν
LOGIC IN THE
HIGH MIDDLE AGES:
P R O P O S I T I O N S A N D MODALITIES
TOPICS: THEIR DEVELOPMENT AND
A B S O R P T I O N INTO C O N S E Q U E N C E S
From Aristotle through Boethius to scholasticism

'Topic' is the infelicitous but by now standard translation for the Latin
technical term 'locus', designating a logical concept variously understood
throughout ancient and medieval philosophy. The medieval tradition of
the Topics has its roots in Aristotle's Topics.1 In that book, Aristotle's
purpose is to present an art of arguing, more precisely the art of dialectical
disputation or Socratic arguing; and most of the book is devoted to a
method for the discovery of arguments. The main instrument of this
method is a Topic, by which Aristotle understands primarily a strategy of
argumentation (such as, 'If the species is a relative, [one must] examine
whether the genus is also a relative') and secondarily a principle confirming
the Üne of argument produced by the strategy (for example, 'If the species is
a relative, the genus is also').2 Six of the eight books of the Topics consist
largely in a loosely ordered compilation of such strategies and principles.
Aristotle considers these Topics part of dialectic and distinguishes them
from two different but analogous sorts of Topics, rhetorical Topics (which
aid in the construction of rhetorical arguments)3 and mnemonic Topics
(which aid in recalling things committed to memory). 4
Topics received considerable attention in later antiquity from the Greek
commentators on Aristotle5 and from Latin rhetoricians,6 including
Cicero, who wrote his own treatise (Topica) on dialectical Topics.7 In the
ι. For discussion of Aristotle's Topics, see, for example, De Pater 1965 and 1968, pp. 164-88, and
Stump 1978, pp. 159-78.
i. Topics I24 b i5-i6.
3. See Stump 1978, pp. 170-2.
4. Cf. Yates 1966 and Sorabji 1972. For mnemonic topics in the Latin rhetorical tradition, see, e.g.,
the anonymous Rhetorica ad Herennium 1954, m, xvii-xxiv, pp. 208-24.
5. See esp. Alexander of Aphrodisias 1891.
6. See, e.g., the following: Tacitus 1949, p. 31; Quintilian 1920-2, V. x. 2off., V. x. iooff., V. xii.
151T.; Victorinus, Explanationum in Ciceronis rhetoricam libri II, in Halm 1863, pp. 213fr.; Martianus
Capella, Liber de arte rhetorica, in Halm 1863, pp. 465fr.; Fortunatianus, Artis rhetoricae libri III, in
Halm 1863, pp. I05ff.; Cassiodorus 1937, pp. 125fr.
7. For a brief historical survey of Cicero's Topica and the literature on it, see Stump 1978, pp. 20-3.
274 The high middle ages: propositions and modalities
course of their work, the discipline of the Topics changed until by Boe-
thius' time it had become very different from Aristotle's art of Topics,
particularly in its understanding of the nature of a Topic. Boethius himself
wrote two treatises on Topics, one a commentary on Cicero's Topica, In
Ciceronis Topica, and one a definitive summary of dialectical and rhetorical
Topics, De topicis differentiü. His work on the Topics, especially Book II of
De topicis differentiis, is the most important (though by no means the sole)
source for the scholastic discussion of the Topics.
According to Boethius, who is dependent on both the Greek and Latin
traditions, 8 two different sorts of things are Topics: a Topic is both a
maximal proposition and the Differentia 9 of a maximal proposition. O n
Boethius' view, a maximal proposition is a self-evidently true, universal
generalisation, such as 'Things whose definitions are different are them-
selves also different.' Boethian Topics of this sort probably have as their
ancestors the Aristotelian Topics that are principles. Their official function,
on Boethius' account, is to aid in the discovery of arguments; but in
practice Boethius tends to use them to confirm arguments. 1 0 Differentiae
are theoretically the differentiae dividing the genus maximal proposition into
its subaltern genera and species, and in that capacity they do serve to classify
maximal propositions into groups. Some maximal propositions have to do
with definition, for example, and others with genus; so from definition and
from genus are Differentiae. Much more important, however, is the role
Differentiae play in Boethius's method for the discovery of dialectical
arguments. For the most part, Boethius thinks of dialectical arguments as
having categorical rather than conditional conclusions, and he conceives of
the discovery of an argument as the discovery of a middle term capable of
linking the two terms of the desired conclusion. Boethian Differentiae are,
for the most part, the genera of such middle terms. To find an argument,
using Boethius' method, one first chooses an appropriate Differentia
(criteria for appropriateness are left to the arguer's intuition). The genus of
middle terms, determined by the Differentia chosen, and the two terms of
the desired conclusion then indicate the specific middle term of the argu-
ment and so indicate a dialectical argument supporting the conclusion. In
Book II of De topicis differentiis, Boethius gives what he claims is an
8. For a summary of the controversy over Boethius' sources, see Stump 1074.
9. I am capitalising 'Differentia' here to distinguish this technical use of the word from its more
ordinary use designating one of the predicables.
10. For a detailed analysis of Boethius' use and understanding of Topics, see Stump 1978, especially
pp. 179-204.
Topics: their development into consequences 275
exhaustive list of Differentiae, taken from Themistius. These Themistian

Differentiae or some subset of them constitute the core of most scholastic
discussion of the Topics (except, of course, for scholastic commentaries on
Aristotle's Topics).
In one form or another, Topics were treated by Isidore of Seville11 and
various writers in the Carolingian Renaissance,12 and they were discussed
continuously throughout the scholastic period, beginning in the first half
of the eleventh century. The culmination of the scholastic tradition of the
Topics is the absorption of the Topics into the theories of consequences, or
conditional inferences, in the fourteenth century. ' 3 A definitive history of
the scholastic treatment and transformation of the Topics cannot yet be
written; much of the important material is unresearched, and many of the
relevant treatises are unedited. What follows is no more than a preliminary,
tentative sketch of the development of the Topics in the scholastic period.
From Garlandus Compotista through the early logica moderna

The earliest known scholastic discussion of Topics which is still extant
occurs in Garlandus Compotista's Dialectica,1* one chapter of which is
devoted to Topics. It is plain from Garlandus' references to other discus-
sions of Topics15 and from the considerable but apparently not original
divergence between his work on Topics and Boethius' that scholastics
contemporary with and immediately preceding Garlandus also discussed
the Topics, but as yet we know too little about this early period to say
anything definite about Garlandus' sources for his treatment of the Topics.
In many respects—in language, subject matter, order of material presented
- Garlandus' discussion of the Topics resembles Boethius' closely, but his
conception of the function and purpose of a Topic is very different from
that of Boethius.
On Garlandus' account, Topics provide and confirm the conditional
premiss of a simple hypothetical syllogism (consisting in a conditional
premiss, a categorical premiss, and categorical conclusion), in roughly this
way.16 Garlandus lists and discusses almost all the Boethian Differentiae,
11. Isidore 1911, II. xxx.

• 2. See e.g., Alcuin 1941, pp. ii2tf. and 120 ff.; Rhabanus Maurus 1901, pp. 227—30; for Notker's
later work, see also Piper 1882, vol. 1, pp. 593fr., 6i8ff., and 623fr. (esp. pp. 632-3).
13. This view is now well established. See e.g., Boehner 1952, p. 54, Bird i960, 1961, and 1962a;
Pinborg 1969, p. 157, and 1972, pp. 168-9.
14. Garlandus Compotista 1959.
15. Garlandus Compotista 1959; see, e.g., pp. 87.168"., 88.10-21, and 88.28-36.
16. For a detailed analysis of Garlandus' treatment of the Topics, see Stump 1980a.
and for a number of these he gives modes of arguing. For example, one
mode of arguing given for the Differentia/rom an integral whole is 'univer-
sally attributing something to an integral whole'. 17 This mode and its
Differentia together spell out a certain kind of true conditional proposition.
The mode gives the quantity and quality of the antecedent and, implicitly,
of the consequent as well (though frequently the quantity of the con-
sequent is simply indefinite). Furthermore, the mode indicates whether the
term specified by the Differentia from an integral whole will be subject or
predicate in the antecedent, and implicitly determines the subject or predi-
cate of the consequent also. Suppose, for example, that the integral whole
under discussion is house and that we attribute, say, whiteness to this integral
whole. Following the suggestions of the mode gives us a conditional in
which whiteness is universally attributed to the integral whole in the
antecedent and indefinitely attributed to the integral part in the con-
sequent: 'If the whole house is white, the wall also is white.' i8 The function
of the Differentiae on Garlandus' account, then, is to aid in the discovery or
construction of true conditionals. The function of the maximal prop-
ositions is to demonstrate or confirm the truth of these conditionals,
which turn out to be instances covered by the generalisations that are the
maximal propositions.
Garlandus is interested in the Topics because he thinks they are useful in
the study of hypothetical syllogisms, which appear to be his main interest
in the Dialectica. His chapter on hypothetical syllogisms is more than five
times as long as his chapter on categorical syllogisms and more than twice
as long as the next longest chapter in the book. Garlandus considers the
study of Topics propaedeutic to the study of hypothetical syllogisms,19 and
he says that all Topics, not just some subset of them, are an aid to
hypothetical syllogisms.20 He holds this view of the Topics apparendy
because he conceives of the study of hypothetical syllogisms almost exclu-
sively as the investigation of the acceptable forms of hypothetical syllo-
gisms; but the discovery and confirmation of true conditional premisses
used in hypothetical syllogisms are provided, on his account, by the Topics.
Garlandus ranks the various Differentiae and maximal propositions in
two historically and philosophically important ways. First, he says that
17. Garlandus 1959, p. 103.10-11.

18. Garlandus 1959, p. 103.11.
19. Garlandus 1959, p. 86.18-23.
20. Garlandus 1959, p. 114.17.
categorical syllogisms are aided only by the Topicsfrom a whole, from a part,
and from an equal.21 And he uses two of the maximal propositions from the
Topic from a whole - 'What is universally attributed to the whole is
attributed also to the part' 2 2 and 'What is universally removed from the
23
whole is removed also from the part' - as if they were rules governing
the first-figure syllogistic moods Barbara and Celarent, on which the other
moods of the syllogism depend.2* Secondly, he says that all Topics can be
subsumed under the Topics/rom the antecedent and from the consequent. The
maximal propositions Garlandus gives for these Differentiae amount to the
rules for modus ponendo ponens and modus tollendo tollens respectively; and
these are, of course, basic principles for all hypothetical syllogisms. These
two claims Garlandus makes strongly suggest that on his view all inferences
are dependent on the Topics.
There is a marked divergence between Garlandus' treatment of the
Topics and that provided by the major terminist logicians in thefirsthalf of
the thirteenth century; and Garlandus in the eleventh century and the
terminists in the thirteenth mark the poles between which the abundant
twelfth-century work on the Topics lies. Much of this work, mostly from
the second and third quarters of the twelfth century, bears some resem
blance to the sort of Topical theory represented by Garlandus' work,
though whether and to what extent Garlandus' own treatment of the
Topics was itself a source for this twelfth-century material is not clear.
An early twelfth-century example of Topical treatment resembling
Garlandus' occurs in an Introductiones dialectice which may come from the
school of William of Champeaux. 25 There, in connection with inferences
in which one categorical is inferred from another, the author singles out the
three Topics which Garlandus described as aids to categorical syllogisms:
from a whole, from a part, ana from an equal. (When the Differentia or its
instance is a predicate in the inference, the anonymous author also adduces
21. Garlandus 1959, p. 114.18.
22. Garlandus 1959, p. 88.18.
23. Garlandus 1959, p. 103.1-6.
24. Garlandus 1959, pp. 118.3 5η*. What connection Garlandus thinks there is among the Topicsfrom a
whole,froma part, mdfrom an equal is not clear; but in twelfth-century treatises the Topic from an
(fünf is frequently cited as confirming a logical conversion. See, e.g.. De Rijk 1962-7, n(i), p. 143,
where this Topic is cited as support for contraposition; and in Introductiones Montane minores.
Abbreviate Montana, and Tractatus Anagnini, conversion of all sorts is associated with this Topic
(De Rijk 1962-7,11(2), pp. 34,96-7, and 235; cf. also 11(1), pp. 393-4). Cf. also Logica 'Ut dicit' in
De Rijk 1962-7, 11(2), p. 402.12-14.
25. Introductiones dialectice secundum Wilgelmum in De Rijk 1962-7, 11(1), pp. 135-6. For further
information on William of Champeaux's theories of Topics, see Green-Pedersen 1974.
the Topics from opposites and from immédiates.) At least thirteen twelfth-
century commentaries on Boethius' De topicis differentiis have survived, 26
including one by Abelard; 27 and several of these also show points of
contact with the views represented in Garlandus' work, for example, in
their concern over the nature and ontological status of Topics and in their
emphasis on the two Topics governing hypothetical syllogisms, the Topics
from the antecedent and from the consequent.2*
A large number of the (mostly twelfth-century) treatises edited or
partially transcribed by De Rijk in Logica Modernorum discuss Topics:
Introductiones Montane minores, Abbreviatio Montana, Excerpta Norimber-
gensia, Tractatus Anagnini, Ars Meliduna, Introductiones Parisienses, Logica
'Ut dicit', Logica 'Cum sit nostra', Dialectica Monacensis, and Summe Me-
tenses. These tend to fall into two groups, each of which has its own
pattern of contents and particular handling of the Topics. Introductiones
Montane minores and Abbreviatio Montana belong to one group, which is
represented also by the Ars Emmerana and the Ars Burana. All four of these
treatises show, very roughly, this pattern in the table of contents: De sono, de
voce, de oratione, de propositione, de propositione categorica (including a section
on conversion), de propositione hypothetica, Topics (de locis), and hypo-
thetical and categorical syllogisms (usually in that order). (Both Ars
Emmerana and Ars Burana have a section on dialectical questions in place of
the section on Topics in the other two treatises. Introductiones Montane
minores lacks the section on categorical and Ars Burana the section on
hypothetical syllogisms, and Ars Emmerana concludes with a section on the
properties of terms.) The Tractatus Anagnini seem idiosyncratic; but the
treatment of Topics found there places them in this group, and the order of
materials in the second tractate shows some resemblance to the order of the
contents of the treatises of this group: conversion, Topics, categorical
syllogisms, and hypothetical syllogisms. It is worth noticing that the order
of subjects in Garlandus' Dialectica also resembles this pattern. After a first
Book on the predicables and categories, Garlandus' work is arranged,
roughly, in this way: de oratione, de propositione, de categorica propositione
(including a section on conversion), single and composite propositions,
Topics, categorical syllogisms, and hypothetical syllogisms.
The treatments of Topics in the three treatises of this group that discuss
26. See Green-Pedersen 1977b.

27. Peter Abelard 1969, pp. 205-330.
28. For a very brief but detailed comparison of these commentaries with Garlandus' views, see Stump
(forthcoming).
Topics at any length - Introductiones Montane minores, Abbreviatio Montana,

and Tractatus Anagnini - resemble the views in Garlandus' work much
more than the views of a terminist such as Peter of Spain. All three treatises
devote considerable attention to hypothetical propositions or hypothetical
syllogisms or both; and Introductiones Montane minores and Abbreviatio
Montana include their treatment of Topics in their sections on hypothetical
propositions. All three concentrate almost exclusively on the three Topics
Garlandus associated with categorical syllogisms, from a whole, from a part,
and from an equal (from opposites and from similars also receive attention).29
Finally, like Garlandus, all three have detailed discussions of conversion,
which they associate with the Topics.30 All these treatises date from the
second or third quarter of the twelfth century, except the Tractatus
Anagnini, which stems from the end of the century.31
In contrast with these are the treatises of the second group: Introductiones
Parisienses, Logica ' Ut dicit', Logica 'Cum sit nostra', and Dialectica Monacensis.
These, too, tend to show a rough pattern in the order of their contents:
introduction, propositions, syllogisms, Topics, and properties of terms
(supposition and sometimes others as well). All but Introductiones Parisienses
have a section on categories or predicables or both, variously placed, and
Dialectica Monacensis includes a section on sophistical arguments between
the chapters on Topics and properties of terms. These treatises tend to have
little or nothing on hypothetical propositions and syllogisms; with the
exception of Dialectica Monacensis, they also have little or nothing on
conversion. They all have lengthy discussions of categorical syllogisms,
however. Furthermore, all of them discuss the entire list of Boethian
(Themistian) Differentiae;32 and all tend to present the Topics in the same
29. In emphasising the Topicsfrom a whole, from a part, and from an equal, Excerpta Norimbergensia and
Ars Meliduna are like these treatises, and both discuss Topics in their section on hypothetical
propositions; see De Rijk 1962—7 11(2), p. n6ff., and n(i), pp. 272—3 and 347B*.
30. See n. 24 above.
31. Braakhuis 1979,1, pp. 407—8, n. 89.
32. And in this respect Summe Metenses (now believed to be by Nicholas of Paris, ca. 1250; see
Braakhuis 1979, pp. 317-26) apparently resembles them; see De Rijk 1962-7,11(1), pp. 472-3. In
their discussion of Topics that Boethius classifies as intrinsic and intermediate, all the treatises
discussed in this paragraph of the text follow the Boethian order as we now have it in the PL
edition, except for Dialectica Monacensis, which misplaces the Differentia from uses. For the
extrinsic Differentiae, however, each treatise has its own particular order not reflected in any of
the others. The order in the Dialectica Monacensis is the same as the Boethian order in the PL
edition; it is identical with the order in Roger Bacon's Sumule dialectices and very similar to that in
William of Sherwood's Introductiones in logicam (only the Topicfrom opposites is in a different place
in the list). The order in Introductiones Parisienses is very similar to the order in Peter of Spain's
Tractatus; again only the Topicfrom opposites occurs in a different place. And the order in Logica
'Ut dicit' is the same as the order in Lambert of Auxerrc's Logica. The orders of the extrinsic Topics
28o The high middle ages: propositions and modalities
way: one Differentia, for which generally more than one maximal prop-
osition (each with corresponding example) is given. In all these respects,
and others as well,33 these treatises are like terminist treatises, such as Peter
of Spain's Tractatus, and unlike Garlandus' Dialectica. For present purposes
their theories of the Topics can be considered to be represented by those of
the terminists, discussed below. The twelfth-century treatises in this group
stem from the last decade of the century, except for Dialectica Monacensis
(whose account of Topics is in certain respects more sophisticated than even
Peter of Spain's), which De Rijk thinks may be as early as 1160-70.34
Abelard's treatment of the Topics is voluminous and cannot be handled
in detail here.35 In general, it is highly original and represents (or, perhaps,
introduces) a transitional stage in twelfth-century discussions of Topics and
syllogisms. He distinguishes inferences into those that are perfect and those
that are imperfect. The former he identifies with syllogisms, whose form
alone guarantees their validity. Imperfect inferences he identifies with
Topical arguments, which need to be confirmed by a Differentia and
maximal proposition.36 In these respects, his work resembles thirteenth-
century discussions of Topics, such as Peter of Spain's. On the other hand,
he claims that the truth or falsity of hypothetical propositions is known by
means of Topics, and he identifies Topical inferences with hypothetical
propositions.37 And in this respect, his work is remniscent of treatises from
in Roger, William, and Lambert resemble one another much more than they resemble that in
Peter of Spain:
Peter William Roger Lambert

opposites authority authority authority
greater similars similars greater
lesser greater greater lesser
similars lesser lesser similars
proportion opposites proportion proportion
transumption proportion opposites opposites
authority transumption transumption transumption
33. To take just one example, Introductiones Parisienses and Dialectica Monacensis define a Differentia as
Peter of Spain does, as a relationship of a certain sort.
34. De Rijk 1962-7, II (1), pp. 410-14. Grabmann had suggested a date later than Peter of Spain's
Tractatus fat it (Grabmann 1937, pp. 48-51), but Dc Rijk argues at length against Grabmann's
daring.
35. Dialectica, ed. De Rijk 1970, pp. 253-466 and Super Topicaglossae in Scriiti di Logica, ed. Dal Pra
1969, pp. 205-330. For analyses of Abelard's account of Topics, see Bird 1959, pp. 53-7, and
i960, pp. 141 -5; Pinborg 1969, pp. 160-2, and 1972, pp. 69-71. For a detailed discussion of the
originality of Abelard's work on Topics and its place in twelfth-century logic, see 'Abelard on
Topics' and 'Topics and Formal Logic in the Twelfth Century' in Stump forthcoming c.
36. Dialectica, pp. 253 and 256-7.
37. Ibid., p. 253.
the earlier part of the twelfth century. The extent to which Abelard's
treatment of Topics influenced subsequent discussion of the subject is not
clear, but it is not unlikely that he is responsible for turning twelfth-
century treatments of Topics away from the tradition represented by
Garlandus Compotista and towards the approach taken by the terminists.
Terminist logicians
Peter of Spain, William of Sherwood, and Lambert of Auxerre all included
a chapter on the Topics in their logic texts. 38 Their treatments of the Topics
differ from one another in significant ways: Peter's treatise is straight-
forward and rather elementary, William's includes a laborious effort to
show that Topical arguments can be reduced to syllogisms, and Lambert's
is more detailed and sophisticated than either of the other two. They are
enough alike, however, that for present purposes they can be treated
together; Peter of Spain may be taken as representative of the group.
In contrast with the first group of twelfth-century logic texts discussed
above, Peter's Tractatus has very little or nothing on hypothetical prop-
ositions and syllogisms and only a short section on conversion. (The same
can be said of William and Lambert with this exception, that Lambert has a
lengthy section on conversion, especially the conversion of modal prop-
ositions.) Instead, the Tractatus has a long exposition of the categorical
syllogism, and Topical (as well as sophistical) 39 arguments are explained in
terms of the categorical syllogism. O n Peter's view, Topical arguments are
dialectical syllogisms, and he makes a sharp distinction between dialectical
and demonstrative syllogisms. The premisses of dialectical syllogisms are
not necessary but probable. 40 Consequently, they produce only opinion,
not knowledge, and the study of Topics is just an art, not a science. 41 So
38. Cf. Peter of Spain 1972, Tractatus, pp. 55-77; Lambert of Auxerre 1971, Logica, pp. 121-40; and
William of Sherwood 1937, Introductiones in logicam, pp. 56-74.
39. Cf. Peter of Spain, op. cit., pp. 90-3.
40. See Peter of Spain 1972, p. 90.18-24; cf. William of Sherwood 1937, p. s6; and Lambert of
Auxerre 1971, pp. 105C and 141fr.
41. There is a sense in which dialectic is a science. Dialectic can be thought of as the use of Topics to
construct and evaluate probable arguments (this is dialectica Mens), or it can be thought of as
reflection on and analysis of such use ofTopics (dialectica docens). Dialectica utens is only an art, not a
science; its arguments arc Topical and its conclusions only probable. Dialectica docens may be
thought of as the study ofdialectica utens. It uses demonstrative arguments about dialectica utens; it
produces knowledge, rather than opinion; and it is a science. For examples of this distinction,
widespread among scholastics, see, for example, Lambert of Auxerre 1971, pp. 5-6; Simon of
Faversham's commentary on Peter of Spain's Tractatus (De Rijk 1968b, p. 81); Boethius of Dacia
1976, pp. 12-13. ^°' m e distinction in the unedited commentaries on Aristotle's Topics by
Kilwardby, Albert the Great, Adenulph of Anagni, Simon of Faversham, Radulphus Brito, and
Angelo of Camerino, see Green-Pedersen 1973, pp. 14-15. Green-Pedersen suggests that the
distinction may go back to Abelard; see Abelard 1969, p. 315.
dialectical or Topical arguments are plainly given a subsidiary status,

secondary to demonstration and demonstrative syllogisms.
According to Peter, a Topical argument is theoretically an enthymeme,
an incomplete syllogism missing a premiss, 4 2 and all enthymemes can be
43
reduced to syllogisms by supplying the missing premiss. But the validity
of an enthymeme that is a Topical argument is also shown and confirmed
by a Differentia and a maxim ('maxima', Peter's abbreviated version of
'propositio maxima'). O n Peter's view, a Differentia is a relationship of a
certain sort 4 4 (for example, the relationship of a definition to its definitum)
and a maxim is a rule governing inferences dependent on that relationship
(for example, 'What is predicated of a definition is predicated also of its
definitum').*5 Any Differentia may (and usually does) have more than one
maxim corresponding to it; the Differentia/rom definition, for example, has
three maxims besides the one just given. Together, a Differentia and a
maxim confirm an enthymeme in this way. Take the enthymeme Ά
mortal rational animal is running; therefore a man is running.' The
relationship of the subject in the premiss to the subject in the conclusion is
that of the Differentia from definition; that is, mortal rational animal is the
definition of man, the definitum. The maxim quoted above gives us an
inference rule for such a relationship; and the maxim, the statement of the
relationship between mortal rational animal and man, and the enthymeme's
premiss together entail the enthymeme's conclusion. Peter gives twenty-
five Differentiae and fifty-seven maxims, which are meant to cover all the
kinds of Topical enthymemes. 4 6
This special method for coping with a special group of enthymemes is an
odd excrescence on the logical theory of the terminists. First, Topical
arguments are taken to be enthymemes, which are simply incomplete
syllogisms, and the syllogism is taken as the foundation for and the
guarantor of valid inferences; but in practice the validity of a Topical
argument is shown and confirmed not by reducing it to a syllogism but by
42. See Stump 1978, pp. 218—21; cf. Lambert of Auxerre 1971, pp. 139—40.
43. See Peter of Spain 1972, pp. 57-8.
44. The character of the relata, as well as the nature and ontological status of a Differentia, were the
subject of dispute among scholastics. See Green-Pedersen 1977a and 1977b. Peter's views on the
subject are not unambiguous, but he seems to understand a Differentia as a relationship between
two terms; see Peter of Spain 1972, p. 59.11-16 and p. 61.21-5. This was a popular view in the
thirteenth century. See, for example, Radulphus Brito 1978a, pp. 25-6.
45. Peter of Spain 1972, p. 60.17-19.
46. The numbers depend on what is counted as one Differentia or one maxim. For instance, I have
counted the Topicfrom definition and from the definitum as one Topic, and I have counted all four
varieties of opposites as one Topicfrom opposites.
adducing for it a Differentia and a maxim. Secondly, no reason is given to

explain why just these kinds of enthymemes can be confirmed without
reduction to syllogism or why just twenty-five relationships indicated by
the Differentiae should confirm arguments. And thirdly, all Topical argu-
ments are said to be dialectical, but the reasons why Topical enthymemes
should always reduce only to dialectical and never to demonstrative
syllogisms are obscure. That the first of these perplexities, at least, was
something of an embarrassment to the terminists themselves can be seen
most clearly in William of Sherwood's case. He treats Topical confirma-
tion of enthymemes as if it were only a kind of abbreviated alternative to
real confirmation by reduction to syllogism. 47 For many of the Topics, he
gives a Topical confirmation of a particular enthymeme and then im-
mediately appends a syllogistic reduction, which often appears laboured
and contrived. For example, William reduces the enthymeme 'Socrates is
not running; therefore he is not running well' to this syllogism, which he
claims is a syllogism in Ferio: 'Not running is Socrates, running well is
running; therefore not running well is Socrates.' 48
In the work of the terminists there are occasional hints of metaphysical
theories which might go some way towards resolving these perplexities,
but the hints are slight. Peter says, for example, that a good dialectical
argument occurs where something is naturally suited to be proved in one
way of knowing (by means of the intellect or by means of the senses) and is
proved by what is prior to it in that way (that is, intellectually prior or prior
in sense experience). 49 And Lambert says that the relationships indicated
by the Differentiae are natural relationships between a term in the premiss
of an enthymeme and a term in the conclusion. He explains the relation-
ship of a Topic such as from definition by saying that a definition and its
definitum are convertible terms because each expresses the whole substance
of the other. 50 But the meagreness of such suggestions and the labored
attempt to force Topical arguments into syllogistic moods indicate that the
terminists took up the Topics as a part of logic inherited from antiquity
without being altogether sure, and perhaps without being much con-
cerned, about the way in which the Topics fit into the rest of their logical
theory.
47. Cf. also Roger Bacon 1940, p. 315.7-13.

48. William of Sherwood 1937. p. 63; cf. Kretzmann 1966, p. 83.
49. Peter of Spain 1972, p. 169.1-4.
50. Lambert of Auxerre 1971, pp. 123-4.
The shift towards the consequences exemplified in

Robert Kilwardby and Boethius of Dacia
The sort of metaphysical theory suggested in the terminists can be found
more fully developed in other thirteenth-century logicians, particularly in
the modists but also, for example, in Robert Kilwardby. Kilwardby takes
this not uncommon thirteenth-century view of logic. The subject of logic
is reasoning, and all reasoning derives its force from the syllogism and is
reducible to the syllogism. 51 A syllogism has both form and matter; the
form consists in the arrangement of the three terms comprised in two
premisses according to the various syllogistic moods and figures, and the
matter is the necessity or probability of the premisses (which depends on
the actual terms employed in the premisses). 52 Both dialectical and dem-
onstrative arguments are syllogisms. The two sorts of arguments do not
differ from one another with respect to the form of the syllogism but only
with respect to its matter; the matter of demonstrative syllogisms consists
in necessary premisses, that of dialectical syllogisms in probable pre-
misses. 53 Furthermore, a demonstrative syllogism takes as its middle term
only the cause or the definition of one of the terms in the conclusion, but a
dialectical syllogism uses a great variety of middle terms. Although the
Topics from definition and from cause may produce a dialectical syllogism
with a cause or definition as middle term, these middle terms are not the
same in dialectical and demonstrative syllogisms because they are used in
dialectic only insofar as they are considered probable and contingently
related to a term in the conclusion, while in demonstration they are always
essentially and necessarily related to a term in the conclusion. Con-
sequently, a dialectical syllogism can produce only opinion, while a dem-
onstrative one produces knowledge; 5 4 and dialectical syllogisms, unlike
demonstrative ones, depend on Topics to confirm them. 5 5 The meta-
physical theory Kilwardby gives to support the status and function of the
Topics in this account is only briefly sketched. Individual things have
certain common characteristics (communes rationes) which we can in
51. Robert Kilwardby 1976, LHI.523.

52. Robert Kilwardby 1976, LIII.500-2.
53. Robert Kilwardby 1976, LIII.503 and 506. For discussion of this view of Kilwardby's and
comparison with other thirteenth-century views, see Pinborg 1969, pp. 164-74, and Green-
Pedersen 1977a, pp. 55-60.
54. Robert Kilwardby 1976, LIII.5S7, 558, and 561.
55. Cf. Thomas 1953, also 1954, pp. 132-3; and Pinborg 1969, p. 167. Kilwardby's commentary was
printed in Venice in 1516 under the name of Giles of Rome (Aegidius Romanus). This edition
was reprinted by Minerva in 1968 but was not available to me during the writing of this chapter.
thought abstract from them - e.g., whole, part, genus, species, and similars
-and the Topics are drawn from these common characteristics of things. 56
A more fully developed metaphysical theory along these lines can be
found in Boethius of Dacia. According to him, things have certain modes
of being; and from these modes of being, common concepts (intentiones
communes) are drawn. For example, a thing may have a mode of being
according to which it can be divided by several differentiae into various
species. From this mode of being, the common concept of genus is
drawn; 57 and from the common concepts, the Topics are drawn. 5 8
Common concepts are relatives (respectivae) because they are concepts not
of a thing considered absolutely and by itself but of a thing considered in
relation to something else. A thing is thought of as a genus, for instance,
only when considered in relation to the things that are its species. Con
sequently, the Topics, which are based on common concepts, are also
relatives and consist in the relation of one thing to another. Hence, a
dialectical argument, which is confirmed by a Topic, depends on a rela
tionship.59
Because the common concepts are relatives, however, they are only acci
dents 60 of the things of which they are concepts, and not part of their
substance.61 And apparently because a common concept is only acciden
tally related to the thing of which it is a concept, the Topic drawn from it is
not a cause of the conclusion of a dialectical syllogism. But a Topic is a sign
of various dialectical consequences,62 because (since it is founded on a
relative concept of a thing) it is a sign of a thing's following from
something else or of something else's following from that thing. For
example, 'this thing which is signified by the word "colour", by means of
the relation which it has to the thing which is signified by the word
56. Robert Kilwardby 1976, XLVIII.454 and LIII.498.
57. Second intentions, which are similar to but not identical with Boethius' common concepts
(communes intentiones), receive a great deal of attention in connection with the Topics. See. for
example, Simon of Favcrsham's commentary on Peter of Spain's Tractatus (De Rijk 1968b,
pp. 94-5); Buridan's commentary on the same work (Green-Pedersen 1976, p. 137); and Albert
of Saxony 1522, f. 33™. They receive especially detailed treatment by Radulphus Brito; cf., e.g.,
Radulphus Brito 1978, pp. 100-1. and 'Aliquis homo est species* in Pinborg 1975b. See also
Grecn-Pcdcrsen 1977a, pp. 51-5 and 58-9; and Pinborg 1974, pp. 49-59, especially p. 54.
58. For the connection between Topics and modes of being and the properties of things, cf. also
Radulphus Brito 1978a, pp. 58-9; 1978b, pp. 93/Γ.
59. Boethius of Dacia 1976, pp. 10-11; cf. also pp. 209-10. For the difference between a common
concept and a Topic, see pp. 217—18.
60. Relatives are, of course, one of the nine categories that are accidents with respect to the subjects
they are predicated of.
61. Boethius of Dacia 1976, p. 14.
62. Boethius of Dacia 1976, p. 15.
"whiteness", has the characteristic (ratio) of genus and the Topical relation-
ship of genus to species'. 63 A dialectical inference such as 'Socrates is a man;
therefore Socrates is an animal' holds by means of the Topical relationship
of species to genus. It is a good inference because of the nature of the things
signified by the words 'man' and 'animal': man includes animal in its
substance, name, and definition, so that whatever 'man' is predicated of,
'animal' is predicated of also. The Topical relationship of species to genus is
not the cause of this connection between man and animal, and so it is not a
cause of the validity of the inference either, but it is a sign of both the
connection between the things and the validity of the inference. All
dialectical inferences (considered as such) hold by means of such signs,
rather than by means of causes of the conclusion; and so dialectical argu-
ments, in contrast with demonstrative ones, produce only opinion, not
knowledge. 6 4
Seen apart from metaphysical theories, the logical strand of the accounts
of Topics found in Kilwardby and Boethius of Dacia represents a kind of
Aristotelianism 65 in the tradition of the Topics, characterised by emphasis
on the categorical syllogism as the foundation of all inference, a sharp
distinction between dialectic and demonstration, and relegation of dia-
lectical or Topical arguments to a secondary epistemological status. 66 The
transformation of the Topics into the consequences seems to have been
accomplished by the convergence of two lines of development: a gradual
erosion of this Aristotelianism, and an increasing concentration on the
nature of and the rules for consequences. In what follows, I will con-
centrate on the first line of development, the change in attitudes towards
dialectic and the Topics, but I will also say a little about the rise of interest in
the consequences. Plainly, the two lines of development are not uncon-
nected. As long as the categorical syllogism is conceived of as the ultimate
guarantor of validity, interest in non-syllogistic inferences will focus on
attempts to reduce them to syllogisms; and study of non-syllogistic in-
63. Boethius of Dacia 1976, p. 28: 'haec res, quae significatur per hoc nomen "color", per relationem,
quam habet ad rem, quae significatur per hoc nomen "albedo", habet rationem generis et
habitudinem localem generis ad speciem'.
64. Boethius of Dacia 1976, pp. 20-3.
65. By 'Aristotelianism' I mean to suggest only that their views were heavily influenced by their
understanding of Aristotle's writings, not that these views represent Aristotle's own theories,
which are very unlike anything that can be found in scholastic treatments of the Topics.
66. Cf. Roger Bacon 1940, pp. 303-5; Simon of Faversham's commentary on Peter of Spain (De
Rijk 1968b), p. 80; Boethius of Dacia 1976, pp. 30-9, esp. pp. 32.55-33.59; and Simon of
Faversham (forthcoming), Quaestiones novae super librum Elenchorum; cf. Question 6, 'Utrum sit
ponere syllogismum peccantem in materia absque syllogismo peccante in forma', and Question
30, 'Utrum locus sophisticus accidat in demonstrativis".
ferences will not be a central concern as long as all knowledge is thought to

be produced only by demonstrative syllogisms. When the distinction
between dialectic and demonstration weakens and the special status of the
categorical syllogism is undermined, interest in consequences is free to
develop.
The decline in the status of the syllogism

The undermining of Aristotelianism in the treatment of the Topics can
itself be seen as a two-pronged development: first, a growing inclination to
see all syllogisms as dependent on Topics, and, second, a tendency to
consider not only demonstrative but also Topical arguments necessary.
Both developments at least blur the distinction between dialectic and
demonstration, and the first drastically undermines the special status of the
syllogism as well.
A sign of the coming decline in the status of the syllogism can be found
in the work of the terminist logicians themselves, in their account of the
Topicfrom a quantitative whole,61 which they take in a way not intended, I
think, by Boethius. 68 Lambert of Auxerre's account is the most sugges-
tive. 69 O n his view, a quantitative whole is a common term with a universal
sign, and the Topical relationship is that of a quantitative whole to its part.
It validates two sorts of inferences, represented by these examples: 'Every
man is running; therefore Socrates is running', and ' N o man is running;
therefore Socrates is not running*. The description of the Topic and the
examples given are remniscent of first-figure syllogisms in Barbara and
Celarent; and the maxims for this Topic — 'Whatever is posited of a
quantitative whole is posited also of its part' and 'Whatever is denied of a
quantitative whole is denied also of its part' - resemble the principles dici de
omni and dici de nullo, associated with the foundation of the syllogism.
Discussion of this Topic must have provoked inquiry into the relation
between this Topic and the principles dici de omni et nullo and given impetus
to the notion that syllogisms are Topically dependent, as will be more
apparent after a brief discussion of dici de omni et nullo.
The connection between the principles dici de omni et nullo and the
syllogism is variously expressed by thirteenth- and fourteenth-century
logicians. For instance, Roger Bacon says simply that every syllogism
67. Peter of Spain 1972, pp. 64-5; William of Sherwood 1937, p. 61; and Lambert of Auxerre 1971,
p. 127.
68. De topicis differentiis, 1189B-C.
69. See also Roger Bacon 1940, p. 319.
depends on (decurrit super) these principles. 7 0 Siger of Courtrai and the

Pseudo-Scotus claim that every first-figure syllogism (to which all the
other syllogistic moods can be reduced) holds by means of these prin
ciples; 71 by this, the Pseudo-Scotus apparently means that the validity of
the syllogism depends o n these principles. And Giles of R o m e goes so far as
to say that the principles dici de omni et nullo prove the form of the syllogism
and include within themselves the syllogistic order. 7 2 These views are all
consistent with the scholastic conception of these principles. The ab
breviated phrase 'dici de omni' is generally said to mean that in a universal
proposition nothing is to be subsumed under the subject which is not also
subsumed under the predicate, 7 3 so that this principle amounts to a descrip
tion or explanation of the nature of a universal proposition. Its connection
with the syllogism derives from its connection with the first-figure syllo
gism in Barbara. This syllogistic mood has a universal affirmative prop
osition for the major premiss ('Every Β is C ) , and the minor premiss
('Every A is B') in effect indicates that a certain group is to be subsumed
under the subject of the major premiss. By the nature of a universal
proposition, that is, by the principle dici de omni, the conclusion of this
mood ('Every A is C ) follows: every Β is C (the major premiss); a universal
proposition is one in which nothing is to be subsumed under the subject
which is not subsumed under the predicate (dici de omni); every A is
subsumed under the subject of the universal proposition 'Every Β is C
(minor premiss); so every A is also subsumed under the predicate ofthat
universal proposition, and hence every A is C (conclusion). So the validity
of the first-figure syllogism in Barbara appears to be dependent on the
nature of a universal (affirmative) proposition, expressed in the principle
dici de omni; and, mutatis mutandis, the same things can be said about the
principle dici de nullo and the first-figure syllogism in Celarent.
This explanation makes plainer the resemblance between the principles
1
dici de omni et nullo and the Topicfrom the whole. * The subject of the minor
premiss in a first-figure syllogism in Barbara is included in the subject of the
universal major premiss; and so it appears that the subject of the minor
premiss is a quantitative part of the subject in the major premiss, which is its
70. Roger Bacon 1940, p. 290.

71. Cf. Siger of Courtrai 1913, Ars Priorum, pp. 10 and 22; cf. also the Pseudo-Scotus 1891, qu. VII,
P- 97-
72. Giles of Rome, In 'Rhetoricam Aristotelis', in Pinborg 1972, pp. 200-3.
73. Cf., e.g., Peter of Spain 1972, p. 43.6-12.
74. The connection may have been made easier by Aristotle's identification of dici de omni and being
in the whole (esse in toto). Prior Analytics ztpz6-3o, and Boethius' Desyll. cat. 809C10-810C4. Cf.
also Dialectica Monacensis, De Rijk 1962-7, 11(2), p. 490, and Lambert of Auxerre 1971,
pp. 112-13.
quantitative whole. It is reasonable, then, to ask whether the first-figure

syllogism in Barbara is dependent on a Topical relationship between its
terms and whether the principles dici de omni et nullo are equivalent to the
maxims for this Topic.
That the scholastics did ask these questions and frequently enough
answered them affirmatively is evident from a number of texts. An anon-
ymous commentary on the Prior Analytics from around 1270 attacks the
view that every syllogism considered apart from any particular material
proves its conclusion, and in doing so it has to argue against the objection
that every syllogism holds by the Topical relationship/rem a quantitative
whole, whose maxims are in every syllogism constituted by the principles
dici de omni et nullo, and that every Topical relationship proves some-
thing.75 Radulphus Brito in his commentary on Boethius' De topicis
differentiis devotes a question to the issue whether there is such a Topic as
from a quantitative whole and if so, whether it is a good one. He answers both
questions affirmatively. On his view, this Topic is good whenever a
predicate is more known to inhere in a term taken universally than in any
of the supposita for that term, and he identifies predicates (perse) in the first
and second moods of the syllogism as of this type. 76 Simon of Faversham,
commenting on the Sophistici elenchi, specifies as one of the general con-
ditions for all syllogisms that one of the two premisses is related to the other
as whole to part.77 And in an anonymous commentary on the Sophistici
elenchi from the 1270s it is said that 'Every syllogism holds by means of a
Topical relationship, for every syllogism holds by means of the Topic/row
a quantitative whole to its part, because the minor is a part of the major . . . ' 7 8
What results from such considerations, and no doubt from other, very
different ones as well, is the view that all syllogisms are dependent on the
Topics for their validity. In addition to its implicit or explicit occurrence
in the passages just discussed, mention of this view or something very simi-
lar can be found, for example, in Boethius of Dacia's commentary on
Aristotle's Topics where he apparently accepts the view that there is a
Topical relationship in every demonstrative syllogism,79 or in Simon of
Faversham's commentary on the Sophistici elenchi, where he says that every
75. Pinborg 1969, pp. 166-7.
76. Radulphus Brito 1978a, p. 47; 'in toto in quantitate universalis distribuitur in omnia sua inferiora,
et quodlibet eius suppositum suscipit eius pracdicationem ...'
77. Simon of Faversham (forthcoming), Quaestiones novae super librum Elenchorum, Question 5,
"Utrum Syllogismus peccans in forma sit Syllogismus'.
78. Ebbesen 1977a, pp. 34-5 (qu. 16, lines 21—3): 'Omnis Syllogismus est tenens per habitudinem
localem; omnis enim Syllogismus tenet per locum a toto in quantitate ad partem suam, quia minor
est pars maioris . . . '
79. Boethius of Dacia 1976, pp. 39-40.
good consequence is founded on a Topical relationship, and the ensuing

explanation strongly suggests that syllogisms in general are included under
the heading of'consequence'. 8 0
The acceptance of Topical inferences as necessary

A second change in the treatment of the Topics, which is complementary
to the view of syllogisms as Topically dependent, is the increased accep-
tance of the claim that Topical inferences are necessary. This claim is not
unconnected to metaphysical theories about the ontological status of
Topics and their relations to real things, and it is well represented in
Boethius of Dacia. He retains the traditional view that dialectical inferences
produce only probable conclusions and opinion rather than knowledge,
but he makes a subtle distinction which in effect demolishes the traditional
view. In the case of the dialectical argument 'Socrates is a man; therefore
Socrates is an animal', the things signified by 'man' and 'animal' are the
causes of the inference. Because man includes animal in its substance, name,
and definition, the connection between these two things is necessary, and
consequently the argument is necessary too. But considered in this way, in
respect of its relation to a necessary connection between two things, the
argument is not dialectical. If we think of the argument as based on the
Topical relation of species to genus, however, it is a dialectical argument;
and in that case, its conclusion is only probable. And similarly in all other
cases: the very same argument considered in respect of its relation to the
natures of things, which are the causes of consequences, is a necessary
argument; considered as based on a Topical relationship, it is a dialectical
argument and its conclusion is only probable. 81
Finally, it is worth pointing out in connection with changes in the
tradition of the Topics that Topical arguments are increasingly referred to
as consequences in the thirteenth century. Use of the term 'consequentia' to
refer to dialectical inferences can be found, for example, in Kilwardby and
Bacon, 82 and it is very evident in scholastics such as Radulphus Brito and
Simon of Faversham. 83
80. Simon of Faversham (forthcoming), Question 32, 'Utrum ista consequentia sit bona: omnis
homo currit; ergo omnis homo currit'.
81. Boethius of Dacia 1976, pp. 20-3. Cf. also Radulphus Brito 1978a, pp. 58-60.
82. Cf. Robert Kilwardby 1976, pp. 174 and 185, and Roger Bacon 1940, p. 323.
83. Cf., e.g., Radulphus Brito 1978a, pp. 25, 26, 28-9, 33, 40, and 42ff; and Simon of Faversham,
Commentary on Peter ofSpain's Tractatus in De Rijk 1968b, p. 98, and Quaestiones novae super librum
Elenchorum (forthcoming). Question 29, 'Utrum petitio principii sit locus sophisticus' and
Question 32, 'Utrum ista consequentia sit bona: omnis homo currit; ergo omnis homo currit'.
Discussions of consequences in connection with 'si'

Apart from the commentaries on Aristotle's and Boethius' works on
Topics,84 there are three other places in thirteenth-century logic, relevant
to the purposes of this chapter, where consequences are discussed.85 The
first is in the treatment of the syncategorematic word 'si'. In their treatises
on syncategorematic words, William of Sherwood and Peter of Spain
include some detailed discussion of consequences in the section οη'ίί'. 8 6
William distinguishes absolute and as-of-now (ut nunc) consequences and
natural and accidental consequences. The first is a basic distinction occur
ring very frequently in fourteenth-century accounts of consequences, and
something like the second can be found in Walter Burley's work. 8 7
William also discusses very briefly different sorts of difficult consequences,
such as those involving exclusives and modal operators, the sorts of con
sequences typically discussed at length in the fourteenth century. Peter's
basic division of consequences into simple and composite can also be found
in Burley's work, 88 and four of the six rules of consequences Peter discusses
occur frequently in one form or another in fourteenth-century accounts of
consequences. Two other rules of consequences which Peter considers and
rejects - 'From the impossible anything follows' and 'The necessary fol
lows from anything' - are widely accepted in the fourteenth century.
Many of the consequences Peter discusses in connection with the syn
categorematic word 'si' are standard Topical arguments, as he himself
points out. In Burley's own discussion of's»", he says that since the rules for
consequences have already been given in the section on consequences,
89
there is not much left to be said about 'si'.
Discussions of consequences in connection with the fallacy of the consequent

The second place in which consequences tend to be discussed at some
length is in analyses of the fallacy of the consequent. As early as the twelfth-
century treatises Dialectica Monacensis, Fallacie Londinenses, and Fallacie
90
Magistri Willelmi, there is some work on consequences in connection
with this fallacy, for which Fallacie Londinenses makes use of the Topics/rowi
84. Cf., e.g., Boethius of Dacia 1976, pp. 22off. and Radulphus Brito 1978a, pp. 38ff.
85. For a fuller discussion of the following material, see the contributions to this volume by Boh,
Kretzmann, and De Rijk.
86. See William of Sherwood 1941, pp. 78-82; and Peter of Spain 1489, ff. 47 ri, -48' ,, .
87. Walter Burley 1955, Tractatus longior p. 61.
88. Ibid.
89. Walter Burley 1955, Tractatus brevior p. 248.
90. De Rijk 1962—7, 11(2), pp. 588-92, 675-6, and 699-700 respectively.
a similar andfrom proportion. Peter of Spain and Lambert of Auxerre in their

discussions of this fallacy give detailed accounts of certain kinds of con-
sequences, including Topical inferences, in which the fallacy of the con-
sequent can or does occur. 91 Peter begins his account by describing basic
sorts of consequences, in a discussion similar to that in his treatment of the
syncategorematic word 'si". Some of the rules of consequences Peter and
Lambert give in their discussions can also be found in fourteenth-century
treatises on consequences; for example, Lambert's rule that an inference
from a proposition having many causes of truth to one having only one of
those causes of truth is fallacious is one of the ten basic rules of consequences
in Burley's Tractatus brevior.92 Peter, Lambert, William of Sherwood, and
Roger Bacon all give, in one form or another, Aristotle's explanation for
the origin of the fallacy of the consequent: the fallacy of the consequent
occurs when we are deceived and think that a consequence converts when
it does not. 9 3 The same view is represented also in Simon of Faversham's
commentary on the Sophistici elenchi, for example, where three questions
are devoted to detailed discussion of consequences and rules for con-
sequences. 94 It is not unreasonable to suppose that this explanation of the
origin of the fallacy gave impetus to the investigation of rules distinguish-
ing valid from invalid consequences.
Discussions of consequences in connection with conversion

The third and most important place, however, is the discussion of conver-
sion in commentaries on the Prior Analytics. Concern about the logical
status of conversion can be found, for example, in Albert the Great, Robert
Kilwardby, Lambert of Auxerre, and texts associated with Boethius of
Dacia. 95 A common response, found in one form or another in all these
authors, is to say that conversion is an inference (sometimes, a con-
sequence) but not an argument. The concern generated by the problem of
conversion and the importance of this solution to it both stem from the fact
91. Peter of Spain 1972, pp. 160-73; and Lambert of Auxerre 1971, pp. 195-9.
92. Lambert of Auxerre 1971, p. 197 and Burley 1955, Tractatus brevior, p. 212.29-31.
93. Soph.el. i67 b i-2, Peter of Spain 1972,p. 170.15-17; Lambert of Auxerre I97i,p. 195; William
of Sherwood 1937, p. 100; and Roger Bacon 1940, p. 350.15—19. Cf. also Dialectica Monacensis,
De Rijk 1962-7,11(2), p. 589.
94. Simon of Faversham (forthcoming), Quaestiones novae super librum Elenchorum, Questions 35—7,
'Utrum ubicumque est fallacia consequentis necesse sit consequentiam conversam esse bonam',
'Utrum arguendo a positione consequentis ad positionem antecedentis sit bona consequentia',
and 'Utrum arguendo a destructione antecedentis ad destructionem consequentis sit bona
consequentia'.
95. Cf. Pinborg 1972, pp. 86-7; Kilwardby's commentary on the Prior Analytics in Thomas 1953,
pp. 65fr.; and Lambert of Auxerre 1971, pp. 24fr.
that syllogisms in the second and third figures are reduced to (and so
confirmed by) first-figure syllogisms by conversion. When conversion is
explicitly referred to as a consequence, the implication is that second- and
third-figure syllogisms are at least partially dependent on certain sorts of
consequences for their validity. This notion works together with the
contemporaneous discussion of the principles dici de omni et nullo to under-
mine the primacy of the syllogism: second- and third-figure syllogisms
depend on consequences (conversion) and first-figure syllogisms; first-
figure syllogisms depend on the principles dici de omni et nullo, which look
very much like Topical maxims. Lambert adds, with some justice,96 that
reduction to the impossible, also used to reduce second- and third-figure
syllogisms to those of the first figure, is equivalent to or dependent on (his
views are not quite clear) the Topic from division.91 In effect, such views
shift the emphasis from the syllogism to consequences as the ultimate
guarantor of validity, and they must have stimulated study of the nature
and kinds of valid consequences. Detailed work on assertoric and modal
consequences can be found in the treatments of conversion by, for exam-
ple, Roger Bacon, Siger of Courtrai, Richard of Campsall, and the
Pseudo-Scotus;98 in fact, Richard of Campsall's Questiones super librum
Priorum analeticorum is devoted mostly to conversion and consequences.
The logical culmination of such treatment of conversion can be seen, for
instance, in Albert of Saxony's Perutilis logica where conversion, as well as
syllogism, is among the various subjects contained within his account of
consequences.99
The incorporation of the Topics into the consequences

Among the earliest full-fledged accounts of consequences now known to
be extant are those by Walter Burley in his De consequents, Depuritate artis
logicae (Tractatus brevior and Tractatus longior) and by William Ockham
in his Summa logicae. The relation between Burley's Tractatus brevior and
Tractatus longior, as well as the relation between either of these and
Ockham's Summa logicae is not clear, but the currently prevailing view is
96. Cf. Boethius, De topicis differentiis 1193A-C and Stump 1978, p. 125, notes 112 and 113.
97. Lambert of Auxerre 1971, pp. 138-9.
98. Cf. Roben Kilwardby's commentary on the Prior Analytics in Thomas 1953, pp. 56ff and see also
Thomas 1954, pp. 129fr. Roger Bacon 1940, pp. 290-7, cf. also pp. 322-3. Siger of Courtrai 1913,
pp. 10-20, cf. also pp. 25fr. Richard of Campsall 1068, pp. 69fr. The Pseudo-Scotus 1891, qu. to,
pp. 103fr. De Rijk mentions Simon of Faversham's discussion of conversion in his commentary
on Peter of Spain's Tractatus as another example; De Rijk 1968b, p. 89.
99. Albert of Saxony 1522, f. 26" ff.
that the Tractatus brevior was written before Ockham's Summa logicae and
that the Tractatus longior is an expanded revision of the Tractatus brevior,
written in answer to Ockham's w o r k . 1 0 0
Topics and consequences in Burley's Tractatus brevior

We know very little about Burley's immediate sources for his work on
consequences. The section on consequences in the Tractatus brevior gives
the impression of being a summary compilation, for pedagogical purposes,
of already existing material. 1 0 1 It consists in the presentation and discussion
often principal rules of consequences (with some corollaries), very few of
which, if any, originate with Burley even in their formulation. If there is
anything new here, it is the facile, unexplained but explicit inclusion of
syllogisms and rules for syllogisms within a treatment of consequences.
There is very little discussion of Topics in the Tractatus brevior. The most
notable discussion occurs not in the section on consequences but in the
section on the syncategorematic word 'si'. There Burley cites the Topic
from the lesser as support for two rules for consequences: the necessary
follows from anything, and from the impossible anything follows. 1 0 2
Topics and consequences in Ockham's Summa logicae

The third treatise of Ockham's Summa logicae is devoted to arguments, and
the first chapter of the treatise is given over to the syllogism because of the
syllogism's special status among arguments, as Ockham explains. O n his
view, all syllogisms are divided into those that are demonstrative, those
that are Topical, and those that are neither (this group seems at least to
103
include and perhaps to coincide with sophistical syllogisms). Demon
strative syllogisms are those whose premisses are necessary propositions;
Topical syllogisms, on the other hand, consist of probable propositions. 1 0 4
This account of Ockham's is reminiscent of views from the first half of the
thirteenth century, but what Ockham goes on to say is a repudiation of the
spirit of those views. First of all, he takes the probable to be what is true,
readily believable, and necessary, only not self-evident or derivable from
ioo. See Boehner 1955, pp. VI-XIV.

ιοί. Walter Burley 1955, p. 199, "Ut iuvenes in quolibet problemate disputantes possint esse exercitati
et velociter obviantes, quemdam tractatum de puritate artis logicae propono, concedente Deo,
compila«.'
102. Walter Burley 1955, pp. 248.24-249.3.
103. William Ockham 1974a, p. 360.43—6.
104. William Ockham 1974a, p. 359.
what is self-evident. So a Topical syllogism, consisting of probable pre-

misses, is deficient in neither the matter nor the form of a syllogism.105
Secondly, he says that all the syllogistic moods of the first figure are directly
dependent on (regulari per) the principles dici de omni et nullo for their
validity. The other syllogistic moods are reducible to those of the first
figure by conversion, reduction to the impossible, or transposition of
premisses;106 and he thinks of cases of conversion (at least) as conse-
quences.107 Furthermore, in his chapter on demonstrative syllogisms, he
says that maxims are necessary for all demonstration, because it is in virtue
of such maximal propositions that the premisses of a demonstration are
known in some way or other (aliquo modo).108 And finally, in the section
on consequences, he claims that all syllogisms hold in virtue of extrinsic
means, by which he understands a general rule of consequences, and the
extrinsic means he gives are often traditional Topical maxims.109 So
though Ockham casts his discussion of arguments and inferences in Aris-
totelian terminology, his views differ widely from the scholastic Aristote-
lianism found in the earlier part of the thirteenth century.
After the chapters on the syllogism in general and on the demonstrative
syllogism comes Ockham's first chapter on Topics and consequences. He
introduces it as his chapter on enthymemes and claims that what follows
will teach the diligent student correct views about all non-demonstrative
(i.e., Topical and sophistical) syllogisms.110 The chapter begins with three
basic and overlapping divisions of consequences. (1) Some consequences
are absolute and some are as-of-now (Mî nunc). For those of the first sort,
there is no time at which the antecedent can be true without the con-
sequent; but for those of the second sort, there is some time, only not this
time, at which the antecedent can be true without the consequent. (2) Some
consequences hold by an intrinsic and some by an extrinsic means. The
former are consequences valid in virtue of a proposition composed of the
same terms as the consequence, while the latter are those valid in virtue of a
general proposition not composed of the same terms as the consequence.
'Socrates is not running; therefore a man is not running' is a consequence
holding by an intrinsic means, because it is validated by the addition of the
105. William Ockham 1974a, p. 360.20-4,360.35-6.

106. William Ockham 1974a, pp. 362-4.
107. William Ockham 1974a, pp. 322.18-23, 323.40-50, and 323.67-324.86. The following twenty
pages contain many such examples.
108. William Ockham 1974a, pp. 509.3-510.10.
109. William Ockham 1974a, p. 588.23-35.
no. William Ockham 1974a, p. 587.4-9.
proposition 'Socrates is a man'. 'Only a man is a donkey; therefore every

donkey is a man' holds through an extrinsic means, because it is validated
by the rule 'An exclusive and a universal, with the terms transposed, signify
the same thing and convert.' (3) Some consequences are formal and some
material. A formal consequence is one holding by an extrinsic or intrinsic
means. A material consequence is one holding just in virtue of the terms
composing it; by this Ockham apparently intends a consequence whose
antecedent is impossible or whose consequent is necessary.1 ' '
Most of the rest of Ockham's treatise on arguments is devoted to
presenting and discussing rules of consequences, and the succeeding chap
ters are divided roughly in this way. First he deals with assertoric con
sequences, beginning with those that hold in virtue of an intrinsic means,
whose terms supposit significatively and personally. Then he turns to
assertoric consequences that hold by an extrinsic means. Though Ockham
does not use the technical terminology generally associated with the
Topics, Topical maxims and arguments traditionally associated with one
of the Differentiae are scattered throughout these parts of his treatise.
Consequences depending on definition, description, interpretation of a
name, genus and species (superior and inferior), integral whole and part,
whole in mode, and relative opposites, for example, can be found recur
rently throughout these chapters (chapters 2—9). Chapters 10—16 deal with
modal consequences; if there is Topical lore in these chapters, it is much less
evident than in the preceding chapters. All the consequences in chapters
2—16 are those in which the terms supposit significatively and person
ally; 1 1 2 beginning in chapter 17, Ockham discusses consequences whose
terms supposit materially. Under this heading, he deals in detail with
Aristotelian Topics in roughly the order in which they occur in Aristotle's
Topics (consequences dealing with accident, genus, property, and defi
nition), and then he discusses the things annexed to one of these four, such
as questions of sameness and difference, annexed to definition. After several
chapters on induction and equivocation, Ockham concludes the material
on consequences with a brief chapter on eight general rules (and some
corollaries) of consequences; Topics play no part in this chapter. Numerous
Boethian as well as Aristotelian Topics are woven throughout Ockham's
chapters on consequences, then; and the Boethian Topics are most in
i n . William Ockham 1974a, pp. 587-9. For discussion of Ockham's material consequences and the
relation between his views of consequences and contemporary logic, sec, e.g., Boehner 1952,
ΡΡ- 53-70, Moody 1953, pp. 64-80, Mullick 1971, pp. 117-24, and Adams 1973, pp. 5-37.
112. William Ockham 1974a, pp. 649.4-650.6.
evidence in the analyses of consequences holding by intrinsic or extrinsic

means, that is, the formal consequences.113
Topics and consequences in Burley's Tractatus longior
There is not so much theorising about logic in the Tractatus longior as there
is in the Summa logicae. Burley divides the section of his book on con-
sequences into three main parts. The first deals with enthymematic con-
sequences, and the second concerns conditional syllogistic consequences.
The third - which is fully half the entire book - investigates hypothetical
consequences which do not involve conditionals (i.e., those involving
conjunctions, disjunctions, exclusives, exceptives, reduplicatives, or the
verbs 'incipit' and 'desinit'); syllogistic consequences having such hypo-
theticals among the premisses are also discussed in this section. According
to Burley, every good consequence holds by a Topic, and a Topical maxim
is nothing other than a rule in virtue of which certain consequences hold.
He distinguishes, however, between logical and dialectical Topics.
Although every good consequence holds in virtue of a logical Topic which
is a maxim, it is not the case that every good consequence holds in virtue of
a dialectical maxim. By 'logical Topic' or 'logical maxim', he seems to
understand simply a rule of logic; he says that Aristotle presents his logical
Topics in the Prior Analytics.11*
In the Tractatus brevior, Burley divided consequences just into absolute
and as-of-now consequences. In the Tractatus longior, he gives a much more
elaborate division, reminiscent of Ockham's, though cast in different terms
and differendy ordered. Like Ockham, he begins with the division into
absolute and as-of-now consequences, though his explanation of these
divisions is slightly different from Ockham's. Absolute consequences he
divides into two sorts; natural and accidental. Natural consequences are
those in which the antecedent includes the consequent, and these hold by
an intrinsic Topic; accidental consequences are those in which the ante-
cedent does not include the consequent, and these hold by an extrinsic
Topic. He makes three other divisions of consequences, ambiguously
related to the preceding ones. First, he says that all consequences are either
simple (simplex) or composite; the former consists of two categoricals, the
113. For Ockham's theory of Topics, see Bird 1961, pp. 65—78 and 1962a, pp. 307—23.
114. Walter Burley 1955, pp. 75-7. 'Dico quod omnis consequentia bona tenet per aliquem locum
logicum, non tamen oportet quod omnis consequentia bona tenet per aliquem locum dialecticum,
nisi extcndo dialccticam ad totam logicam' (p. 75.35-8).
latter of two hypotheticals or of a hypothetical and a categorical. 115 And

secondly, some consequences hold by reason of a whole complex (conver-
sion and syllogism are examples); some hold by reason of the terms. This
second distinction overlaps with the third. Some consequences that hold by
reason of the terms hold materially or accidentally; these are not formal
consequences. Formal consequences are those that hold formally by reason
of the terms and those that hold by reason of a whole complex. 1 1 6 He says
little to explain this complicated division, but the distinction between
consequences that hold by reason of the terms materially, on the one hand,
and both sorts of formal consequences, on the other, bears some resem-
blance to Ockham's distinction between material and formal consequences.
Topics, as either logical or dialectical maxims, occur throughout all three
parts of the section on consequences in the Tractatus longior. Burley ex-
plicitly identifies two of his general rules for consequences as logical
maxims, 1 1 7 and he shows how the use of logical Topics can be cast in the
traditional form for the presentation of dialectical Topics. 1 1 8 He cites the
Topics/row the antecedent und from the destruction of the consequent as support
for inferences of modus ponendo ponens and modus tollendo tollens.119 And he
gives the Topic from division as the warrant for the principle of the
disjunctive syllogism. 120 Theoretically, then, Topics are basic to Burley's
theory of consequences, but there are very few recognisable dialectical
Topics in his treatise, far fewer than in Ockham's Summa logicae.121
Conclusion
In Ockham's and Burley's work, the slow process of the absorption of the
Topics into theories of consequences has been completed. Throughout the
fourteenth century, Topics continue to be discussed, in treatises on con-
sequences but also (and frequently more fully) in, for example, com-
mentaries on Peter of Spain's Tractatus and on Aristotle's Topics.122 Certain
of the fourteenth-century logicians writing on consequences understand
115. Walter Burley 1955, p. 61.

116. Walter Burley 1955, pp. 84, 86.
117. Walter Burley 1955, cf. p. 76.7-10 and p. 62.9-13.
118. Walter Burley 195s, pp. 76.33-77.23.
119. Walter Burley 1955, pp. 103.24-104.7.
120. Walter Burley I95S, pp. 119.32-121.18.
121. For a detailed discussion of Burley's treatment of consequences, see, for example. Prior 195 3 ; Boh
1962, 1963a, 1963b,and 1964.
122. For example, John Buridan, whose work on consequences was very influential, wrote commen-
taries on both works and discussed Topics in both commentaries. Cf. Markowski 1968, pp. 3-7,
and Green-Pedersen 1976, pp. 121-38.
material consequences as Ockham does, namely, as those in which the

antecedent is impossible or the consequent necessary. Robert Fland, Ralph
Strode, Richard Lavenham, Paul of Venice, and Paul of Pergula appear to
fall into this group. 123 Whether they, like Ockham, include Topical
inferences among the formal consequences or even whether Topics play
any part at all in their theories of consequences is as yet uninvestigated.
Buridan, Albert of Saxony, and the Pseudo-Scotus, on the other hand, are
examples of logicians who take material consequences in a different
way. *24 For them, a formal consequence is one which holds solely in virtue
of the form of the consequence, and a material consequence is a good
consequence that does not hold in this way. 125 In the work of these
logicians, Topical inferences are equated with a subdivision of material
consequences, sometimes labeled 'simple material consequences'.126 In
these authors, (simple) material consequences tend to be taken as en-
thymematic consequences, reducible to formal consequences by the ad-
dition of a premiss.127 Very little of the theories of Topics among these
authors has been investigated.128 But as far as we now know, Topics
continue in this way to eke out a meagre existence throughout the four-
teenth and early fifteenth centuries until they experience their own rebirth
in the Renaissance in the work of logicians such as Lorenzo Valla and
Rudolph Agricola,129 who elevate Cicero's Topica and Boethius' two
treatises on the Topics to the very centre of logical studies.130
123. Robert Fland, Consequential; in Spade 1976, cf. pp. 63-4. Ralph Strode ng),Consequentiae, f.2'*.
Richard Lavenham, Consequential, in Spade 1974a, pp. 76-81 and 99-101. Paul of Pergula 1961,
Logica, pp. 87-8. Cf. Pinborg 1972, p. 176, Boh 19ÎS5, pp. 33-4, and Bottin 1976, pp. 305fr.
124. For a thorough examination of Albert's theories of consequences with some useful comparisons
with Ockham and Buridan, see Gonzalez 1958 and 19S9. For a study of consequences in the
Pseudo-Scotus, see McDermott 1972.
125. Cf. Buridan 1976, Tractatus de consequentiis, pp. 22-3; Albert of Saxony 1522, ff. 24™"*; the
Pseudo-Scotus 1891, pp. 105 and 184. Cf. also Bos 1976, pp. 62-9, and Pinborg 1972, pp. 174-6.
and 1969, pp. 176-7.
126. Cf. Green-Pedersen 1976, pp. 125-6 and pp. 136-7. Albert of Saxony 1522,6". 36,fc-37'* and 37*
and the Pseudo-Scotus 1891, p. 105. Cf. also Bendiek 1952, p. 220. This division may explain why
Buridan, for example, has little if anything to say about Topics in his treatise on consequences,
because in that book he concentrates on formal consequences; cf. Pinborg 1969, pp. 175-6.
127. See, e.g., Buridan 1976, p. 23.10-23, and Bendiek 1952.
lai. Cf. Green-Pedersen 1976. pp. 121-8.
129. See Jardine 1977, pp. 143-64.
130. I am glad to acknowledge my debt to John Crossen, whose excellence at the theory and practice
of dialectic first aroused my interest in the subject, and f am grateful to Norman Kretzmann and
Jan Pinborg for their numerous helpful comments and suggestions.
15
CONSEQUENCES
The notion of consequentia
Etymologically, 'consequentia' suggests a following along. In medieval

philosophical literature it was apparently quite proper to say that one
concept follows another - e.g., that animal follows man - but more
generally consequence was thought of as involving entire propositions.
There are, of course, many different relationships in which propositions
can stand to each other. For instance, in a conditional proposition of the
form 'if ρ then q the proposition taking the place of 'p' is the antecedent of
the conditional proposition, and the proposition taking the place of 'q' is
the consequent. The relationship between that antecedent and the con
sequent in a true conditional is called implication by modern logicians.
Again, two propositions may be related to each other in such a way that the
first cannot be true unless the second is true also; and the relationship
between the two propositions in that case is called entailment. Again, two
propositions may constitute an argument. In the argument-form 'p; there
fore q' the proposition taking the place of 'p' is the premiss and the
proposition taking the place of 'q' is the conclusion. To employ an argu
ment is to derive or infer the conclusion from the premiss (or premisses),
and so the relationship between those two propositions is called derivation
or inference.
Implication, entailment, and inference are all distinct from one another.
For instance, a conditional proposition, like other propositions, is accepted
or rejected by being classified as true or as false; an argument is neither true
nor false but is accepted or rejected as valid or as invalid. Again, to assert 'p;
therefore q' is to assert both 'p' and 'q', but to assert 'if ρ then q' is to assert
neither 'p' nor 'q'. Nevertheless, medieval logicians disconcertingly use the
single notion of consequence to cover all three of these relationships
between propositions. A consequentia may be a conditional proposition or
the relationship between the antecedent and the consequent in a con
ditional proposition. It may be an argument or the relationship between
Consequences 301
the premiss and conclusion of an argument, which may be called, confus

ingly, 'a rational proposition'. A consequentia may be an immediate in
ference - e.g., 'No S is P; therefore no Ρ is S' - or an enthymeme - e.g.,
'Socrates is a man; therefore Socrates is an animal' — or a fully expressed
syllogism in the object language - e.g., 'If every M is Ρ and every S is M,
then every S is P' — or, finally, as a disconnected series of propositions
arranged as premisses and conclusion expressed in the meta-language -
e.g., Ά , B; therefore C*.
The relationship between propositions in a consequence is a second
intention, ' is in fact the second intention of most interest to logicians of the
Middle Ages (or, indeed, of any other period). The terms 'antecedent',
'consequent', and 'consequence' are terms of second intention, which stand
for such expressions of first intention as 'Socrates is a man', 'Socrates is an
animal', and 'If Socrates is a man, then Socrates is an animal.'
Consequences, conditional propositions, and rational propositions

When reading the medievals it is frequently most natural to think of a
consequentia as a conditional proposition, though it often becomes clear
from the context that what is being discussed is an argument. This circum
stance leads to difficulties: is the consequence to be evaluated as true or false
or as valid or invalid?2 If medieval logicians wanted to classify con
sequences as propositions, why should they not have classified them as
rational rather than as conditional propositions?3 Logic, on the medieval
view of it, was supposed to teach men to speak truly, and so it is perhaps
understandable that arguments were often evaluated as true; but the
parallel use of such expressions as 'consequentia est bona' or 'consequentia
valet', or simply '[consequentia] bene sequitur' indicates also a recognition that
the transition from the one proposition in the consequence to the other is to
be evaluated in terms other than those appropriate to affirmations and
denials.
1. See Knudsen's contribution to this volume.

2. Both consequentia est vera (falsa) and consequentia valet (non valet) were considered proper charac
terisations by authors who flatly stated that a consequentia is a hypothetical proposition (Buridan
1976,1, 3), by those who said that it is an inference or illatio (Strode 1484, f. l v and by those who
wavered (cf. Burley 1955, pp. 65 and 208).
3. The rational proposition - 'p; therefore q' - was counted among the six or seven basic hypo
thetical propositions (conjunctive, disjunctive, conditional; causal, local, temporal, rational).
Buridan treats it as an alternative to the conditional proposition: 'if* and 'therefore' both
designate that one of the two propositions connected by them follows from the other; they differ
in that 'if designates the proposition immediately following it as the antecedent while 'therefore'
designates just the opposite. Cf. Buridan 1976,1, 3.
There is, of course, a connection between the validity of an argument

and the truth of a corresponding conditional proposition. Whenever ' A;
therefore B* is a valid argument, there is a true conditional proposition of
the form 'if A, then B'. It is this connection that is exploited in the medieval
notion of consequentia: the aim is to use the truth of the conditional
proposition as a criterion for the validity of the corresponding argument.
But the matter is not simple, because the connection is not a symmetric
relationship. We cannot correctly say that whenever 'if A, then B' is true
there is a valid argument of the form Ά ; therefore B*. It is only if the 'if...
then ...' is taken as representing entailment that this is so.
This fact about the relationship between arguments and conditional
propositions seems to have been recognised by the medievals in their
insistence that the conditional corresponding to an argument is to be
understood in a strong sense, as a hypothetical proposition that expresses a
close tie between its antecedent and its consequent. Thus we find in the
early twelfth century Abelard's clear predilection for a consequence which
is based on the 'necessity of consecution'.4 Peter of Spain in the middle of
the thirteenth century states that the antecedent in a true conditional
cannot be true without the consequent being true also, adding that 'every
true conditional is necessary and every false conditional impossible', a view
reaffirmed in the fourteenth century by Albert of Saxony and in the
fifteenth by Paul of Pergula.5 There was occasionally even some reluctance
to accept as a consequence a consequence which does not hold. The
distinction in question here is that between a genuine consequence, in
which the appropriate consequential connections are present, and a
counterfeit consequence, in which we have merely an external semblance
of consequence. While Buridan recognises that the issue is basically a
terminological one, he says that he will understand by 'consequence' a
4. See Abelard 1970a, pp. 283.37-284.3: note that even the broader sense of the necessity of
consecution is expressed by him in modal terms: 'Videntur autem due consecutionis necessitates:
una quidem largior, cum videlicet id quod dicit antecedens non potest esse absque eo quod dicit
consequens; altera vero strictior, cum scilicet non solum antecedens absque consequenti non potest
esse verum, ([sed etiam^ ex se ipsum exigit; que quidem nécessitas in propria consecutionis
sententia consistit et veritatem tenet incommutabilem ...'
5. See Peter of Spain 1972, p. 9.15-18: 'Ad veritatem conditionalis exigitur quod antecedens non
possit esse verum sine consequenti, ut "si homo est, animal est". Unde omnis conditionalis vera est
necessaria, et omnis conditionalis falsa est impossibilis.' Cf. Gonzalez 1958, p. 333, η. 48, where he
quotes from Albert's Perutilis logica, t. 3, c. 5, f. 19": 'Omnis conditionalis vera est necessaria, et
omnis falsa est impossibilis'. Cf. Paul of Pergula 1961, p. 17: 'Nota quod omnis conditionalis vera
est possibilis et necessaria, et omnis falsa est impossibilis et nulla est quae sit contingens.'
Consequences 303
hypothetical proposition which is true,6 even though in the course of his

discussion he shows no hesitation to use the notion of a consequence that
does not hold, is invalid, is bad, etc.
The origins of the theory of consequences

Scholars disagree about the origins of the theory of consequences. The
word 'consequentia' can be found in Boethius, who found its Greek equi-
valent ('akolouthësis') in Aristotle, even though it does not there have the
technical sense of a relation among propositions.7 If we understand the
theory of consequences as basically a theory of propositions, and if we
accept the view that Megaric-Stoic logic was the first systematic logic of
unanalysed propositions,8 we might expect to find some influence, direct
or indirect, of Stoic logic upon the medieval theories of consequences.9 We
do in fact find Stoic elements in Boethius' hypothetical syllogistic, even if
they are not the result of direct influence,10 and Boethius' work on the
subject made a significant contribution to scholastic logic. He introduced
the notion of consequentia into medieval logical discussion, he distinguished
the temporally-conditioned accidental connection from the natural con-
nection between an antecedent and its consequent, and he classified and
discussed the Topics, which were influential in the development of scholas-
tic theories of consequences. ' (
Consequences in Garlandus Compotista

Garlandus Compotista and Peter Abelard inherited much from Boethius,
yet in many ways the two were rethinking his doctrine very carefully. In
his Dialectica Garlandus is fairly detailed in his discussion of Boethian
6. Buridan 1976, p. 21: 'Alii dicunt si [consequentia) sit falsa non debet dici "consequentia", sed
solum si sit uera. Et de hoc non est disputandum, quia nomina significant ad placitum; et siue sic
siue non sic, ergo in hoc tractatu per hoc nomen "consequentia" uolo intelligere consequentiam
ueram, et per "antecedens" et "consequens" uolo intelligere propositiones quarum una sequitur
ad aliam uera seu bona consequentia.'
7. See Bochenski 1961, p. 189.
8. The agreement here is unanimous. Cf. i-ukasiewicz 1935, Boehner 1952, Bochenski 1961,
Dumitriu 1977.
9. Everyone agrees that there are striking resemblances between the theory of consequential and the
Megaric-Stoic propositional logic; e.g. Lukasiewicz 1935, Bochenski 1938,1961, Dumitriu 1977,
Mates 1965b, Moody 1953. However, the actual historical influence, if it indeed took place, has
not been demonstrated and the opinions of the foremost scholars differ. Cf., e.g.. Moody 1953,
1967; Dürr 1951.
10. An extensive study of Boethius' sources is available in Obertello 1969.
11. See Stump's preceding contribution to this volume.
hypothetical syllogistic and, interestingly enough, he places the chapter on

Topics just before the one on syllogisms. What is interesting from our
point of view here is that Garlandus finds himself discussing not only
Topics but also the four Aristotelian types of argument - categorical
syllogism, induction, enthymeme, and example - within the chapter on
Topics; furthermore, some Topical maxims are so general as to pass for
rules of consequences - for example, 'If the antecedent is posited [i.e.,
affirmed], so is the consequent'; 'If the consequent is removed [i.e., denied],
so is the antecedent.' 12
Much of what Garlandus says about hypothetical syllogism obviously has
its roots in Boethius, both in content and in terminology. He discusses
negation and allows for contradictory opposition between 'P' and ' ~ P'
and also, at least implicitly, for the equipollence of 'P' and '~ ~ P'.
Although he seems to recognise both term-conjunction and propositional
conjunction, he does not have much to say about either. He compares 'If (si)
a man is, an animal is' and 'When (cum) a man is, an animal is'. If both
imply 'Because (quia) a man is, an animal is', then 'cum' is not significantly
different from 'si'. But he finds that there is a great difference between these
two conjunctions, as indicated by the following example: 'When (cum) fire
is hot, the sky is round.' Here, he says, Ί say not that becausefireis hot, the
sky is round, but rather that the sky is round at a time when fire is hot.' 1 3 He
even goes so far as to suggest a division of consequences based on this
distinction, into consequences per accidens and natural consequences. The
recognition of the difference between 'si' and 'cum' is echoed in the later
distinction between consequences that are absolute and those that hold as of
now (utnunc).1*
Of the remaining hypothetical propositions Garlandus recognises dis-
12. Garlandus 1959, p. 114. He ends his chapter 'De topicis differentiis' with the remark that: 'While
all the Topics are in the service of hypothetical syllogisms, only the Topics from the whole and
part and from the equals are of service to categorical syllogisms.' ('Sciendum est quod omnes loci
scrviunt hipoteticis sillogismis; cathcgoricis vero tantum scrviunt locus a toto et a parte et a pari.')
13. Garlandus 1959, p. 141: 'Propositio hipotetica connexa in qua preponitur "cum", aliquando
candcm vim optinct cum ea in qua preponitur "si", ut cum dico: "si homo est, animal est": utraque
enim dicit: "quia homo est, animal est". Aliquando autem aliam vim optinet, ut hie "cum ignis calidus
est, celum rotundum est": non enim hie dico quia ignis calidus sit, celum rotundum esse, sed dico
celum esse rotundum ex tempore quo ignis calidus est.' Similar remarks on the ambiguity of 'cum'
ate found in Boethius, although the latter does not explicitly stress the causal connection between
the antecedent and the consequent, but rather the omnitemporal character of 'si' and the
temporally-restrictive character of'cum'. Sometimes the logical force of 'cum' is that of 'et'. Cf.
Dürr 1951, pp. 56f. in conjunction with Prantl 1855-70,1, p. 715, n. 162.
14. Garlandus 1959, p. 141. This division of consequentia was adopted but greatly refined by Abelard
1970a, pp. 4738"., and accepted by most later medieval writers on consequentiae; see, for example,
Ockham 1974a, III, 3,1. '"'
Consequences 305
junctions, but he undermines the role of disjunctions in syllogisms. In fact,

he reduces the disjunctive proposition formed by 'aut aut' to a con-
ditional; conversely, he equates 'Either it is not a man or it is an animal' with
'If it is a man, it is an animal.' l s The questions whether or not 'aut' is truth-
functional and if it is, whether it is inclusive or exclusive, are undetermined
then, unless we know what the truth-conditions of the associated con-
ditional proposition are. Garlandus says that a conditional proposition is a
sentence composed of'if ('si') and two or more categorical propositions.
He recognises that the component categorical propositions can be both
true, both false, one true and one false, or - surprisingly - both neither true
nor false. By the last kind we should, it seems, understand a conditional
composed of subjunctive propositions. It is the indeterminacy of subjunc-
tive propositions that makes him talk about four ways in which a con-
sequence can be true and five ways in which it can be false.16 (The examples
ofboth true and false consequences given by Garlandus make it implausible
that he considered Philonian implication at all.) The basis for the connec-
tion between the antecedent and the consequent is provided by the
Topical maxims that serve as the foundation for natural consequences in
which 'si' has more than a temporal force. In fact, Garlandus' talk about
propositions which are neither true nor false suggests that his natural
consequences are supportable by counterfactual conditionals.
Consequences in Peter Abelard

Although Abelard does not define 'consequentia', he does characterise 'in-
ferentia', of which consequentia is, in his view a sub-species.17 Inference
consists, he says, 'in the necessity of the consecution, that is, in the fact that
from the meaning of the antecedent the thought of the consequent fol-
15. See Garlandus 1959, pp. 13 if-, where he not only considers the equivalence of disjunctives and
conjunctives but also relates both of them to a universal affirmative proposition: 'Equipollent
etiam quedam cathegorice quibusdam hipotcticis propositionibus tarn connexis quam disiunctis.
Ut hie patet: "si est homo, est animal", "aut non est homo, aut est animal", "omnis homo est animal"; he
enim eandem veritatem tenent et per unam alie probantur. Ideoque notandum est quod con-
iuncte atque disiuncte simplices per universales cathegoricas sibi equipollentes probantur: "si est
homo, est animal", "aut non est homo, aut est animal" verificantur per istam in veritate eis
consimilcm: "oimn'j homo est animal".'
16. Garlandus 1959, pp. I36f: 'Nunc sciendum est item quod consequentia quatuor modis fit vera,
alia ex utrisque veris . . . , alia ex utrisque falsis.... alia ex falso antecedenti et vero con-
sequenti . . . , alia ex utrisque terminis neque veris neque falsis, quemadmodum in ista potest
dinosci: "si esset homo, esset animal ": neuter namque verus est neque falsus. / Verum etiam quinque
modis fit falsa consequentia: alia ex utrisque veris.... alia ex utrisque falsis . . . , alia ex falso
antecedenti et vero consequenti..., alia ex utrisque neque veris neque falsis, ut hie: "si Socrates
esset animal, esset homo", alia iterum fit falsa ex vero antecedenti et falso consequent!.'
17. De Rijk in Abelard 1970a, p. XXXIV.
3θ6 The high middle ages: propositions and modalities
lows'. 1 8 It is perfect if the necessity of the consecution is based on the

arrangement of terms regardless of their meaning; it is imperfect if the
meaning of terms must be taken into account in order to have the necessary
connection. This is the most general division, which comprises all the
traditional forms of argumentation. Abelard understands a consequence as
fundamentally a coniunctio between antecedent and consequent, and the
basic division of such a conjunctive implication which he rightly attributes
to Boethius, is into natural and temporal. 1 9 The difference between the
two is perhaps best seen in Abelard's examples: 'If a man is, an animal is'
and 'When it rains, it thunders.' In the former, we have a natural consecutio,
in the latter we have only a temporal comitatio. For the most part Abelard
uses something like modern strict implication as the paradigm of impli
cation, 2 0 but he is also aware of implications of a weaker sort. 2 1
Since Abelard follows Boethius in associating disjunction with some
thing like strict implication, it is obvious that disjunction in his view is not a
mere truth-function. But while Abelard would reject the view that for the
truth of a disjunction it is sufficient that one of its components be true, he
also wanted to relax the (Stoic) requirement that the disjuncts must be
incompatible; in this way he left room for further developments which led
eventually to the medieval discovery of inclusive truth-functional disjunc
tion, the so-called De Morgan's Laws. 2 2
The period immediately following Abelard is still not adequately ex
plored, although recent scholarship has shown that there was intense
logical activity during that t i m e . 2 3
Consequences in the terminist logicians

As for the terminists, it is significant that in none of the three best-known
24
textbooks of the thirteenth century - viz., those of Peter of Spain,
18. Abelard 1970a, p. 253.28-31: 'Inferentia itaquein necessitate consecutionis consistit, in eo scilicet
quod ex sensu antecedentis sententia exigitur consequentis, sicut in ipotetica propositione dicitur,
ut in scqucntibus monstrabitur. Hec autem inferentia alias perfecta est, alias imperfecta.'
19. Abelard 1970a, p. 472.
20. Abelard 1970a, pp. 283f. (Cf. η. 4 above.)
21. Abelard 1970a, p. 293.13—17. '[Aliquaeconsequentiaejliberiusexpediunturet assignantur secun
dum sensus totarum enuntiationum et laxius accipiuntur, ut videlicet ita exponantur: posito
antecedenti, idest existente eo quod dicit antecedens propositio, existit illud quoque quod
consequens proponit'; p. 330.28-30 'Nichil itaque (aliud) in "antecedere hoc ad aliud" vel "inferre
hoc illud" accipimus, quam "si hoc est, illud esse".'
22. For the scholastic origins and tradition of these laws see Boehner 1951a.
23. Sce,c.g.,Minio-Palluello 1956, De Rijk 1962-7, and Giusberti 1977. The survey in Pinborg 1972
gives an impression of a very lively interest in logic, semantics, and language generally during this
period. Grabmann 1937 is still worth reading in this connection.
24. I.e., William of Sherwood 1937, 1966; Lambert of Auxerre 1971; and Peter of Spain 1972.
Consequences 307
William of Sherwood, and Lambert of Auxerre - do we find a special

chapter or treatise on consequences. But even in these introductory texts
the authors supply us with information on the notion of consequence
generally in the chapters on propositions, on Topics, and on fallacies.25
And further details are provided in the treatises on syncategoremata by
William and by Peter. 26
Consequences in the Pseudo-Scotus
We can consider only a few features of the theory of consequences and its
rise to prominence during the late thirteenth, fourteenth, and early fif
teenth centuries.27 For a sample of the difficulties encountered in later
medieval attempts to formulate a satisfactory definition of a consequence,
we can turn to the Pseudo-Scotus. In his commentary on the Prior
Analytics, he proposes the following definition: Ά consequence is a hypo
thetical proposition composed of an antecedent and a consequent by means
of a conditional or rational conjunction, which denotes that it is impossible
for them - namely, for the antecedent and the consequent - that the
antecedent be true and the consequent false when they are formed at the
same time.' 2 8 The apparently undiscriminating acceptance of two ways of
25. Cf. William of Sherwood 1066, Ch. IV, esp. pp. 34f. where he speaks of consecutive conjunction
which he equates with conditional proposition, and where he states the truth conditions of a
conditional: 'In order that a conditional be true the truth of its parts is not required, but only that
whenever the antecedent is [true] the consequent is [true]. ' Cf. ibid., Ch. VI, esp. the sections on the
Fallacy of Begging the Original Issue (pp. 157f.) and the section on the Fallacy of Consequent
(p. 160). Cf. Lambert of Auxerre 1971, p. 16, for his definition of conditional; p. 120 for his
recognition of locus maxima as an inferential rule, and esp. pp. I95f. where he explicitly uses
'consequentia' in a technical sense and distinguishes two types: "Duplex est consequentia: una
scilicet in qua, posito antécédente, de necessitate ponitur consequens, et hec potest dici naturalis
vel necessaria; alia vero est consequentia in qua, posito antécédente, non propter hoc de necessitate
ponitur consequens, sed ut frequentius concomitatur antecedens consequens et hoc potest dici
consequentia probabilis vel consequentia ut in pluribus.' Cf. Peter of Spain 1972, pp. 8f. and esp.
pp. i69f. where he devotes several brief paragraphs to the notion of consequence and its several
types before discussing the Fallacy of the Consequent.
26. See William of Sherwood 1968, Chapter XVII on 'IF' (SI) and also the following chapter on
'UNLESS' (NISI), which present a meticulous discussion of basic consecutive conjunctions. In
Peter of Spain 1964, pp. so-65 on 'consecutive words' are instructive.
27. The most important texts from the early fourteenth century arc by Burley (written in the 1320s
cd. Boehner 1955), Ockham (written in the late 1320s, ed. Boehner and Gâl 1974). Buridan
(written in the 1330s, ed. Hubien 1976), and the Pseudo-Scotus (written around 1350, in Duns
Scotus ed. Vives 1891).
28. Pseudo-Scotus 1891, I. 287B: 'Consequentia est propositio hypothetica, composita ex antécé-
dente, et conséquente, mcdiante coniunctione conditional!, vel rationali, quae dénotât, quod
impossibile est ipsis, scilicet antécédente, et conséquente simul formatis, quod antecedens sit
verum, et consequens falsum'. The reconstruction of the difficulties connected with formulating
conditions of the truth or validity of consequences is based on q. 10 of the Pseudo-Scotus'
commentary on Book I of the Prior Analytics, but it also takes into account the discussions by
Bendiek 1952, Mates 1965b, Kncalc 1962, pp. 286-8, and McDcrmott 1972.
3o8 The high middle ages: propositions and modalities
forming a consequence — 'ifp, then q', and 'p; therefore q' — is typical of the
medievals and does not present an important difficulty. Much more inter-
esting is the fact that plausible counter-examples can be offered to the
definitions typically given. The Pseudo-Scotus, among others, 29 reviews
and criticises accepted or possible formulations of a definition.
(i) 'For the validity of a consequence it is necessary and sufficient that it
be impossible for the antecedent to be true and the consequent false.' 30 To
this, according to the Pseudo-Scotus, there is the following counter-
example: 'Every proposition is affirmative, therefore no proposition is
negative.' The objector points out that 'Every proposition is affirmative'
would be true if all negative propositions were destroyed, while ' N o
proposition is negative' could not possibly be true since it falsifies itself by
its very existence. Thus the intuitively acceptable consequence. 'Every
proposition is affirmative, therefore no proposition is negative' turns out to
be unacceptable if we apply definition (i). 31 T o avoid such undesirable
results, the proponent offers two other criteria, but they also are subjected
to powerful objections.
(ii) 'For the validity of a consequence it is necessary and sufficient that it
be impossible for things to be as signified by the antecedent without also
being as signified by the consequent.' 3 2 The Pseudo-Scotus' counter-
example here is: ' N o chimaera is a goat-stag; therefore a man is a jack-ass.'
To appreciate the force of this, one should note that for the Pseudo-Scotus
negative propositions require for their truth only that things should not be
as signified by the corresponding affirmative proposition; on that assump-
tion, the antecedent is clearly true and the consequent false. Hence, since
formulation (ii) does not rule out such counter-examples, it is inadequate.
The Pseudo-Scotus therefore considers another definition.
(iii) 'For a consequence to be valid it is necessary and sufficient that it be
impossible that if the antecedent and the consequent are formed at the same
time, the antecedent be true and the consequent false.' 33 But again there is
a troublesome counter-example: 'God exists; therefore this consequence is
not valid.' Although the counter-example fulfills the conditions of (iii), it is
not valid. For if it were a valid consequence, its consequent would be false,
29. E.g., Buridan 1976,1, 3.
30. Pseudo-Scotus 1891, p. 286B: 'Ad bonitatem consequcntiac requiritur, ct sufficit, quod im-
possibile est antecedens esse verum, et consequens falsum.'
31. Pseudo-Scotus 1891, p. 287A.
32. Pseudo-Scotus 1891, p. 287A: 'Ad bonitatem consequentiae requiritur, et sufficit, quod im-
possibile est sic esse, sicut significatur per antecedens, quin sic sit, sicut significatur per consequens.'
33. Pseudo-Scotus 1891, p. 287A: 'Ad bonitatem consequentiae requiritur, et sufficit. quod im-
possibile est antécédente, et conséquente simul formatis, antecedens esse verum, et consequens
falsum.'
Consequences 309
since it asserts precisely that the consequence is not valid. But since the
antecedent, 'God exists', is true if the consequent, 'this consequence is not
valid', is false, the consequence could not be valid. Thus the assumption that
this consequence is valid refutes itself. Yet the conditions of (iii) are
fulfilled, since the antecedent and the consequent are both necessary and so
the case of the antecedent being true while the consequent is false could
never arise. That the consequent is necessary is shown by the fact that the
consequence 'God exists; therefore this consequence is not valid' could not
possibly be valid.
In spite of the difficulty encountered by such an unusual counter-
example, the Pseudo-Scotus endorses (iii) but allows for an exception of the
case 'where the meaning of the consequent is incompatible with the
meaning of the sign of consequence as the very connective which specifies
that there is a consequence'.34
According to the Pseudo-Scotus, the principles 'From the impossible
anything follows' and 'The necessary follows from anything' are generally
accepted but need explication. First, one should determine what is meant
by 'impossible'. Everyone will agree that a formal contradiction, 'p. ~ p',
is a paradigm case of impossibility. The Pseudo-Scotus easily proves 'q' -
any proposition at all - from the conjunction of 'p' and its denial by an
appeal to the intuitively obvious formal rules of simplification-of-conjunc-
tion, logical addition, and disjunctive syllogism.35 (John Buridan had
already offered much the same proof.)36
Attempts were made to provide an adequate classification of con-
sequences. One important step in this direction was the general division
into formal and material consequences.37 In the Pseudo-Scotus' theory,
formal consequences comprise all the logical principles of traditional syl-
logism, the laws of conversion of propositions, the laws of the square of
opposition, and other principles of Aristotelian logic; but they also include
the laws of propositional logic determined by the most general rules of
consequences governing propositions taken as unanalysed units. Material
consequences are subdivided into those that hold absolutely (simpliciter) and
those that hold only as-of-now (ut nunc). Both are enthymematic as they
34. Pseudo-Scotus 1891, p. 287A-B: 'Dico igitur, quod ad bonitatem consequentiae, requiritur et
sufficit vltimus modus, scilicet impossibile est, antécédente, et conséquente simul formatis, quod
antecedens sit verum, et consequens falsum, excepto vno casu, scilicet vbi significatum con-
sequentis répugnât signification! notae consequentiae, sicut coniunctionis, quae dénotât con-
sequentiam esse...'
35. Pseudo-Scotus 1891, p. 288A-B.
36. Buridan 1976, I, 8, 7* conclusio: 'Ad omnem propositionem copulativam ex duabus inuicem
contradictoriis constitutam sequi quamlibet aliam . . . consequentia formali.'
37. Pseudo-Scotus 1891, p. 287B. On this division cf. Stump's preceding contribution to this volume.
stand, but a material consequence is said to be absolute if it is reducible to a

formal consequence by addition of a necessary premiss; and it is said to be
as-of-now if it is reducible to a formal consequence by the addition of a
factually true premise which could conceivably be false at some time. Both
absolute and as-of-now consequences are such that if they do hold at all, it is
impossible for the antecedent to be true and the consequent false. The
apparently paradoxical principles 'From the impossible anything follows'
and 'The necessary follows from anything' in the sense of ' ~ Ο ρ -»
(p-*q)' and ' ~ Ο ~ q —* (p —* <?)' were considered to be 'merely material'
consequences, presumably on the ground that it is their content, their
modality, rather than their propositional form which determines their
validity.
Consequences in William Ockham

William Ockham presents a somewhat different, threefold division. 3 8 (i)
A consequence is valid either absolutely or as-of-now. (2) A consequence
holds per medium extrinsecum or per medium intrinsecum. A consequence of the
former kind — one that holds in virtue of an extrinsic medium - is valid in
virtue of a logical rule. For example, 'Only an animal is a man; therefore
every man is an animal' holds in virtue of the rule 'From an affirmative
exclusive proposition to the universal affirmative proposition with the
terms transposed there is a valid consequence (and conversely).' Con
sequences that hold in virtue of an intrinsic medium are valid in virtue of an
additional premiss which transforms the original enthymeme into a fully
stated argument. For example, 'Every man is an animal; therefore every
man is mortal' is turned into an explicit argument by the addition of the
premiss 'Every animal is mortal'. The added premiss may, of course, be
either necessary or contingent. (3) A consequence is either formal or
material. A formal consequence can be one of two sorts: either it holds (i) by
an extrinsic medium, that is, by some meta-rule such as is operative in the
above example of logical relation between an exclusive and a universal
affirmative proposition; or (ii) if it holds directly by an intrinsic medium
and indirecdy by an extrinsic medium or a rule 'which concerns general
features of propositions and not their truth and falsity or possibility and
impossibility' — as in the case of'Socrates is not running, therefore a man is
not running' (which holds directly by the additional premisses 'Socrates is a
man' and indirectly by some syllogistic or Topical rule).
38. Ockham 1974a, III, 3, 1.

Consequences 311
A consequence is material39 if it holds precisely in virtue of the meanings

of its terms - that is, not in virtue of some extrinsic principle determining
the inferential connection on the basis of the general conditions of prop
ositions but on the basis of special conditions such as their truth-value or
modal status. The examples given are the two paradoxical consequences:
Ά man is running; therefore God exists' and Ά man is a jack-ass; therefore
God does not exist.'
Consequences in Walter Burley

For a good sample of a medieval theory of consequences we can turn to
Walter Burley.40 He was perhaps the earliest medieval logician who fully
understood the logical priority of propositional logic and also developed it
to a considerable degree. Although his contemporary William Ockham
also wrote lengthy sections on consequences and on hypothetical prop
ositions, Ockham's Summa logicae is organised around terms, propositions,
and arguments. Burley on the other hand, in his two versions of De puritate
artis logicae,*1 starts with the most general rules of consequences and then
subsumes hypothetical and categorical syllogistics under that primary
logic. Furthermore, his treatment of the properties of terms and of various
sophismata throughout his work presupposes an understanding of con
sequential rules.
Of the two versions of the chapter 'De regulis generalibus consequentiarum',
the one contained in the Tractatus brevior is the more revealing. I will base
my presentation on it and invoke the other version only for comparison. In
the prologue Burley remarks that he will state 'certain general rules which
will have to be used in what follows'. Then, without offering any general
definition of a consequence, he draws a distinction between absolute and
39. There is still disagreement about whether medieval logicians, and Ockham in particular, recog
nised material implication. 'Material' in the sense in question precludes any modal notions or any
semantic connection between the antecedent and the consequent. Ρ materially implies Q if and
only if it is not the case that Pis true and Q false. Boehner 1951b argued with some force on the
basis of selected textual evidence that Ockham indeed recognised material implication of a purely
truth-functional sense. Adams 1973, on the other hand, makes so many careful distinctions
regarding what could be meant by knowing material (and strict) implication that one prefers to
stay within Ockham's own conceptual framework and try to do one's best to determine precisely
how/ his three divisions of consequentiae organise the whole field.
40. See n. 54 below.
41. Boehner argued persuasively that the first version (the Tractatus brevior) was begun independently
of Ockham's Summa logicae. But when Ockham's tteatise appeared, Burley wrote his second
version, the Tractatus longior, starting with the tract on supposition in which Burley argues against
Ockham's nominalistic interpretation of supposilio simplex. This second version was finished
before 1329. All the references are to Boehner's edition of both versions; the sections on
«mjfijMfMiae are on pp. 60-6 and 199-219.
as-of-now consequences. Ά consequence is absolute', he says, 'if it holds

always; for example, "A man is running, therefore an animal is running."
A consequence is as-of-now if it holds for a certain time and not always; for
example, "Every man is running, therefore Socrates is running"; for
this consequence does not hold always but only while there is a man
Socrates. ' 4 2 Then he states his ten principal rules, discussing each of them as
he goes along.
Using the schematic letters P, Q, R, etc., the symbol '—>' as a sign of a
consequence, and the usual symbols for logical constants, we can schema
tise Burley's rules and proofs, preserving as much as possible of his natural
deduction technique. (PR — Principal Rule, DR = derived rule.) Only
those rules which can reasonably be considered as propositional rules are
given.
PR ι Ρ -> Q/.'. ~ 0 ( P . ~Q)> and the corollary:
0(P. ~Q)/.\ ~ ( P - Q ) 4 3
DR i. I OP, ~ O Q / - ' - ~ ( P - > Q )
DR ι..2 ~ 0 ~ P , 0 ~ Q / · ' · ~ ( P - > Q ) 4 4
PR 2 Ρ -> Q / ·'· (Q -• R) -*• (P -> R) and the corollary
Ρ -> Q/-'. (R --> P) -*• (R -* Q) 4 5
D R 2. I P^Q,Q->R...T-> UI.-.P-* U*b
D R 2.2 P-yQ, (P.Q) -> R/.\ Ρ -** R
D R 2 •3 P-*Q, (Q-R) -*SI.\ (P.R) -* S*1
,48
PR 3 P-+QI-'· ~Q-*> ~ P "
42. Burley 1955, Ρ-199-" 'Consequcntiarum quaedam est simplex, quaedam est ut nunc. Consequentia
simplex est ista, quae tenet pro omni tempore, ut: "Homo currit, igitur animal currit."
Consequentia ut nunc tenet pro tempore determinato et non semper, ut: "Omnis homo currit,
igitur Sortes currit"; ilia enim consequentia non tenet semper, sed solum dum Sortes est homo.'
43. Burley 1955, pp. I99f.: 'In omni consequentia bona simplici antecedens non potest esse verum
sine conséquente. Et ideo, si in aliquo casu possibili posito posset antecedens esse verum sine
conséquente, tunc non fuit consequentia bona. In consequentia autem ut nunc non potest
antecedens ut nunc, scilicet pro tempore, pro quo consequentia tenet, esse verum sine
conséquente.'
44. Burley 1955, p. 200: 'Ex contingenti non sequitur impossibile in consequentia simplici... Ex
necessario non sequitur contingens.'
45. Burley 1955, p. 200: 'Quidquid sequitur ad consequens, sequitur ad antecedens... Quidquid
antecedit ad antecedens, antecedit ad consequens.'
46. Burley 1955, p. 200: 'Per hanc rcgulam: Quidquid sequitur ad consequens, sequitur ad ante-
cedens, tenet consequentia a primo ad ultimum, quando arguitur per multas consequentias
intermedias.'
47. Burley 1955. p. 203: 'Quidquid sequitur ex conséquente et antécédente, sequitur ex antécédente
per se . . . Quidquid sequitur ad consequens cum aliquo addito, sequitur ad antecedens cum
eodem addito.'
48. Burley 1955, P· 207: 'In omni consequentia bona, quae non est syllogistica, ex opposito con-
sequentis contradictorie sequitur oppositum antecedentis.'
Consequences 313
For syllogistic consequence Burley states the following important rule(s):

DR 3.1 (Ρ, Q)-y Rj:. (~R, P) -• ~ Q
DR 3.2 (Ρ, Q)-> R / : . (~R, Q) -> ~ P 4 '
This insight into the link between the indirect syllogistic reductions and
the principles of propositional logic is remarkable.
To clarify the nature of the denial of propositions, especially of com
pound propositions, Burley gives the following rule:
PR 4 'The basic formal element (formale) affirmed in one of the
contradictories must be denied in the other.' 5 0
Among particular instances we find
DR 4.1 ~(P.Q)<^(~P ν ~Q)
DR4.2 ~ ( P V Q ) H ( ~ P , « V Q)
DR 4.3 ~ (P -» Q) -^ (P. ~ Q) 5 1
PR 5-10 go beyond propositional logic, since they govern propositions
whose internal structure needs to be taken into account.
To illustrate Burley's method of deriving a rule from other rules, we can
present his proofs for the two rules DR 2.2 and D R 2.3 in a schematic way,
justifying each step on the basis of Burley's text.
Proof (ratio) for DR 2.2 'Every proposition implies itself along with its
consequent... Since, then, the antecedent implies both the antecedent
and the consequent, and since whatever follows from the consequent
follows from the antecedent, it follows that whatever is implied by the
antecedent and the consequent follows from the antecedent alone.' 5 2
ι. Ρ —» Q 1 Assumption of a consequence
2. (P.Q) -» R Assumption that something follows from the con
sequent and the antecedent conjoined
3. (P -* Q) -+ [P -» (P.Q)] A rule tacitly assumed by Burley (see the
first sentence of his proof)
4. Ρ —• (P.Q) From (3) and (1) by modus ponens (stated by Burley in his
hypothetical syllogistic, but apparently too obvious to be
explicitly stated here)
49. Burley 1955, p. 208: 'In consequentia syllogistica ex opposito conclusionis cum altera
praemissarum sequitur oppositum alterius praemissae.'
50. Burley 1955, p. 208: 'Formale amrmatum in uno contradictorio debet negari in reliquo ...'
51. Burley 1955, p. 209: 'Contradictorium copulativae valet unam disiuncrivam habentem partes
contradicentes partibus copulativae... Contradictorium disiunctivae aequipollet copulativae
factae ex contradictoriis partium disiunctivae. Contradictorium conditionis valet unam pro-
positionem, quae significat oppositum sui consequentis stare cum suo antécédente...'
52. Burley 1955, p. 203: 'Quaelibet propositio infert seipsam cum suo conséquente ... Cum igitur
antecedens inferat antecedens et consequens, et quidquid sequitur ad consequens, sequitur ad
antecedens, sequitur quod quidquid sequitur ex antécédente et conséquente, sequitur ex antécéd-
ente per se.'
5. Ρ -»• R From (4) and (2) by PR 2

6. {(P -> Q). [(P.Q) -» R]} - > ( P - » R ) From the conjunction of as
sumptions (1) and (2) the con
sequent (5) has been shown to
follow.
Proof for D R 2.3 'An antecedent with something added implies the
consequent with the same thing added . . . Since, therefore, whatever is
implied by the consequent is implied by the antecedent, it follows that
whatever is implied by the consequent with a proposition added is
implied by the antecedent with the same proposition added.' 5 3
1. P - > Q
2. (Q.R) - S J A s s u m P t i o n s
3. (P -» Q) -> [(P.R) -» (Q.R)] A rule expressed by Burley's first
sentence
4. (P.R) -* (Q-R) From (3) and (1) by modus ponens
5. (P.R) -> S From (4) and (2) by PR 2
6· {(P -* 0 ) . [(Q-R) -> S]} -> [(P.R) -* S] From (1) and (2) and (5)
Q.E.D.
The main shortcoming of these and similar proofs is obvious: Burley
ought to have stated all the rules he needs before constructing the proof. In
any case his fundamental idea of proof is sound, and his work paved the
way for the subsequent logical treatises on consequences.54
Conclusion
The century of Burley, Ockham, Buridan, and others was indeed a golden
age of logic, in which the theory of consequences attained its mature form.
It was followed by a slow but steady decline. While the theory of con
sequences continued to be cultivated, it grew by way of commentaries and
extensive discussions of details, rather than by strikingly new discoveries.
However, the post-fourteenth century period is certainly not uninterest
ing. 5 5 Examinations of the immense literature which came to be available
in printed form might be very revealing regarding the cause of the eventual
eclipse of medieval logic in general and the theory of consequences in
particular.
53. Burley 1955, p. 203: 'Antecedens cum atiquo addito infert consequens cum eodem
addito ... Cum igitur, quidquid sequitur ad consequens, sequitur ad antecedens, oportet quod
quidquid sequitur ad consequens cum aliquo addito, sequitur ad antecedens cum eodem addito.'
54. For a comprehensive study of Burley see Una Juarez 1978.
55. Cf. Ashworth 1974, esp. pp. 118—186. For activities in this field in Spain see Munoz Dclgado 1964
and his numerous other works. See also Ashworth's contribution to this volume.
ι6
OBLIGATIONS:
A. FROM THE BEGINNING TO
THE EARLY F O U R T E E N T H C E N T U R Y
Introduction
Perhaps one of the last really obscure areas of medieval logic is contained in
the scholastic work on 'obligations'. W e know something about the
authors and the contents of scholastic treatises on obligations (De ob-
ligationibus), and we think that these treatises came to be a standard, perhaps
even an important part of medieval logic; but we do not yet fully under
stand the nature of the material contained in them. We are unclear about
the function and purpose of obligations and its significance for other parts
of medieval logic; and we have only a sketchy notion of the rich and
complicated development of obligations from its beginnings in the late
twelfth or early thirteenth centuries to the end of the scholastic period.
There are as many guesses about the purpose and function of obligations as
there are scholars who have written on the subject: it has been described
variously as anything from ingenious schoolboy exercises' to primitive
axiomatised logic. 2 My own account of obligations will emerge in the
course of this chapter.
Historical survey
Even in the twelfth century, there is some use of terminology associated
3
with obligations in discussion of disputation, and in some treatises from
this period either disputation or obligations themselves are connected with
fallacies or sophismata. 4 In the first half of the thirteenth century, in the
i. Weisheipl 1956, p. 150. For a brief survey of contemporary literature on obligations, see Spade
1977.
2. Boehner 1952, pp. 14-15.
3. Excerpta Norimbergensia, for example, says that a disputation consists of opposition and response
(De Rijk 1962-7, II.2, p. 127; cf. also pp. 125-6 and II. 1, pp. 155-60); and Ars Emmerana explains
that disputation has throe parts: positio, opposition, and response (De Rijk 1962—7, II.2, p. 148).
The thirteenth-century treatises edited by De Rijk tend to make the same point (cf. De Rijk 1974,
p. 117.8-9 and 1975, pp. 26.1-27.26).
4. Excerpta Norimbergensia discusses disputation in its section on the sophistical art, where fallacies are
also discussed. It divides disputation into two kinds, the science of argument and the science of
redargutio, which is opposing and responding; and it claims that all species of sophismata fall under
3i6 The high middle ages: propositions and modalities
work of the terminists, 5 the investigation of fallacies tends to contain a long

discussion of disputation, and one of the species of disputation discussed
there has as its goal redargutio: the forced denial of something previously
granted or the granting of something previously denied in one and the
same disputation — very similar to the stated goal of obligations in, for
example, Walter Burley. 6 So there seems to have been some early connec-
tion between interest in fallacies or sophismata and interest in obligations;
but we are not yet in a position to trace the early history of this material,
and anything definite about the origin of treatises on obligations must
await further identification and analysis of the relevant early texts.
It is clear that obligations existed in some form in the thirteenth century.
For the second half of the thirteenth century, L. M. de Rijk mentions three
authors in whose work we find evidence of obligations: Godfrey of
Fontaines, John Duns Scotus, and Roger Marston; 7 and we can add
Boethius of Dacia to this list.8 With some reservations, Romuald Green has
argued that one of the treatises on obligations edited in his dissertation An
Introduction to the Logical Treatise "De Obligationibus"9 should be attributed
to William of Sherwood; ' ° De Rijk, dismissing Green's reservations, has
claimed that the treatise in question is indisputably by Sherwood. 1 ' If this
attribution is correct, then we have at least one complete, interesting
treatise on obligations from the first half of the thirteenth century. But
apart from Green's own reservations, which De Rijk seems correct to set
aside, there are other serious worries about the attribution of this treatise to
Sherwood, worries to which De Rijk has not addressed himself in claiming
that the treatise is definitely by Sherwood. Careful consideration of these
this science of opposing (De Rijk 1962-7. II.2, pp. 130-2). Tractatus Sorbonnensis de petitionibus
contrariorum divides all sophismata into those which depend on a sophistical locus (fallacies proper)
and those that depend on a hidden incompatiblity, which is how this treatise seems to understand
obligations (De Rijk 1976, p. 43).
5. See Peter of Spain's Tractatus (De Rijk 1972, pp. 94fr.) and Lambert of Auxerre's Logica (1971,
pp. 142a".); cf. also William of Sherwood's Introductiones in logicam (i937, p. 86).
6. Walter Burley, Tractatus de obligationibus (Green forthcoming), 0.01. Because the published
version of Green's work will not be available until sometime in 1981,1 have based my citations on
the texts as edited in the revised version of Green's dissertation, made available by the Franciscan
Institute. References to the introductory chapters of Green's book are given as 'Intro.: [plus
chapter number]'; references to the edition of Burley use Green's paragraph numbers. I am
grateful to Father Gedeon Gâl and to Professor Paul Spade for helping me obtain a copy of the
typescript.
7. De Rijk 1974, p. 95, n. 11.
8. Cf. Boethius of Dacia 1976, pp. 329.4-331.46.
9. Green forthcoming; see n. 6 above.
10. Green forthcoming, Intro.: Chapter V.
11. De Rijk 1976.
Obligations 317
worries, together with the manuscript evidence, makes it seem altogether

possible that what we really have in the putative Sherwood treatise is an
early treatise on obligations by Walter Burley. 12 If that is so, then Burley
wrote two treatises on obligations, just as he wrote two versions of De
puritate artis logicae: a first, brief treatise, which is now associated with
Sherwood in the literature, and a second, more developed treatise, prob-
ably written around 1302. 13
De Rijk has edited four short anonymous treatises on obligations from
the thirteenth century: (a) Tractatus Emmeranus de falsi positione, (b)
Tractatus Emmeranus de impossibili positione,1* (c) Obligationes Parisienses,15
and (d) Tractatus Sorbonnensis de petitionibus contrariorum.16 De Rijk ven-
tures no guess concerning the date of treatise (d); it may stem from any
time in the thirteenth century. Treatise (c) De Rijk dates by internal
evidence, on the basis of a comparison with the putative Sherwood obliga-
tions; because it is considerably more primitive than the work ascribed to
Sherwood, he assigns (c) to the beginning of the thirteenth century. ' 7 If the
treatise ascribed to Sherwood is not in fact his, however, then the dating of
treatise (c) will have to be reconsidered also. Treatises (a) and (b) De Rijk
dates on the basis of the similarity in content between the two and a
reference in (b) to a thesis of the 'Adamites' (namely, that from the
impossible anything follows). De Rijk identifies the Adamites as followers
of Adam Parvipontanus; and he claims that since their thesis was still in
dispute in treatise (b), that treatise could not have been written later than
the 1220s.' 8 He is mistaken in thinking that the thesis is in dispute in treatise
(b); *9 but perhaps all he needs for his claim is the attribution in treatise (b) of
this thesis to the Adamites (provided, of course, that his identification of the
Adamites as followers of Adam Parvipontanus is correct). The content of
treatises (a), (b), and (c) is much less advanced than that of the putative
Sherwood obligations; but none of these anonymous treatises has been
carefully studied yet.
From the fourteenth century onwards, scholastic work on obligations
proliferates and diversifies. In what follows, I will concentrate on the first
12. Spade and Stump forthcoming.

13. Green forthcoming, Intro.: Chapter V.
14. De Rijk 1974.
15. De Rijk 1975.
16. De Rijk 1976.
17. De Rijk 1975, p. 26.
18. De Rijk 1974, p. 102.
19. Stump 1980, n. 12.
decades of the fourteenth century; a discussion of obligations in the suc

ceeding period is contained in Part Β of this chapter.
Walter Burley
The putative Sherwood and the authentic Burley treatises on obligations

differ only in the fulness of discussion, in the number of species of obli
gations considered, and in minor details of organisation or of content. In
general, the organisation, the approach taken, and the theory of obli
gations are the same in the two treatises; and even the wording, the
examples, and the order of their discussion are in many cases the same. I will
look in detail at Burley's treatise, then, leaving the putative Sherwood
obligations to one side; but much of what is said here about Burley applies
to the putative Sherwood also. Burley's treatise, stemming from the
beginning of the fourteenth century, constitutes a representative account
of obligations in an early stage of their development, before the subtle shifts
of emphasis in the work of Richard Kilvington, 2 0 one of the earliest of the
Oxford Calculators, and before the enormous changes in obligations
introduced by or represented in later Oxford Calculators such as Roger
Swineshead 2 1 and other later fourteenth-century logicians. Consequently,
Burley's treatise is fundamental for an understanding of obligations in the
fourteenth century.
As is true in general of obligations treatises, Burley's work is set in the
context of dialectical disputation, in the highly structured and stylised form
22
the scholastics inherited from Aristotle's Topics via Boethius. The scho
lastics themselves tend to attribute the source for their work on the
obligations to one or another or both of two Aristotelian passages. 2 3 The
first, from Topics VIII 3 (159*15-24), says that the primary j o b of the
respondent in a disputation is to answer in such a way that anything
impossible which the respondent is compelled to defend is the fault of the
position he maintains rather than the fault of his defence ofthat position.
The second, from Prior Analytics 113 (32*18-20), says that from the possible
nothing impossible follows. These two quotations together provide the
theoretical foundation for obligations. If the respondent in a disputation
20. Richard Kilvington forthcoming. I am grateful to Norman Kretzmann for calling my attention
to the material on obligations in Kilvington's Sophismata.
21. Spade 1977; see also Pan Β of this Chapter.
22. Cf. Stump 1978, pp. 159-204.
23. Green forthcoming, Intro.: Chapter II.
Obligations 319
adopts a position which is possible and is subsequently compelled to

maintain something impossible (something logically incompatible with
the position adopted or something impossible in its own right), then he has
failed in his job as respondent. Nothing impossible follows from the
possible; so, because it is possible, the position adopted.by the respondent
does not entail anything impossible. If the répondent is then logically
compelled to maintain contradictory propositions, it must be because he
has made logical mistakes in responding, so that the impossible, which is
not entailed by the original position the respondent adopts, is entailed by
the respondent's faulty defence ofthat position. The job of the interlocu-
tor, called the 'opponent' in obligations disputations, is to trap a respon-
dent into maintaining contradictories, and the job of the respondent is
to avoid such traps. Burley's own main interest is in what might be called
paradoxes of disputation. A paradox of disputation occurs when, after a
seemingly unimpeachable exchange between an opponent and his re-
spondent, the opponent proposes a proposition which the respondent can
neither grant nor deny because either the granting or the denial leads
immediately to a contradiction. The heart of Burley's treatise is a long
series of such paradoxes, each accompanied by his solution to it.
Burley's treatise begins with an ordered division of obligations into six
species.24 In every obligations disputation, the disputation opens with
something put forward by the opponent, which the respondent obligates
himself to take a certain attitude towards. The variations on what the
opponent can propose and on the attitudes the respondent can adopt
constitute the species of obligations. One is obligated to an act or to a
disposition, Burley says;25 and in either case, the obligation covers either
what is complex (that is, a proposition), or what is non-complex. If one is
obligated to an act and the obligation covers what is non-complex, the
species of obligation is petitio. If one is obligated to an act and the obligation
covers what is complex, we have the species sit verum. Ifone is obligated to a
disposition with respect to what is non-complex, we have institutio. And if
one is obligated to a disposition with respect to what is complex, there are
three possibilities: one can maintain the proposition at issue as true, as false,
or as uncertain. These three give us the last three species of obligation,
namely, positio, depositio, and dubitatio. We can represent Burley's schema
in the following way, then:
24. Green forthcoming, Burley 0.02.

25. 'Obligatio sic dividitur: aut obligat ad actum aut ad habitum* (ibid.).
Act Disposition
Non-complex petitio institutio
Complex sit verum positio

depositio
dubitatio
Burley discusses these six species of obligations one by one, beginning

with the two types in which the obligation covers what is non-complex,
and taking first each time the case in which one is obligated to a disposition.
So he considers first institutio, then petitio, then positio, depositio, and dubi-
tatio, and finally sit verum. His main interest, however, is clearly in positio,
and his examination of positio alone constitutes more than three-fifths of
the entire treatise. I will say something briefly about these various species of
obligations and then follow Burley in concentrating on positio.
Institutio is a new imposition for some utterance, giving it a new
signification. 26 Burley's first example is this: 'let " A " signify a donkey in a
true proposition, a man in a false proposition, and the disjunction a-man-
or-not-a-man in an uncertain proposition [i.e., a proposition whose truth-
value is not known to the respondent].' 2 7 Many cases of institutio, like this
one, resemble or just are insolubilia. Petitio is the species of obligation in
which the opponent asks or obligates the respondent to do something. 28
The first proposition an opponent in petitio puts forward and obligates his
respondent to grant is not an ordinary third-person proposition such as
'Socrates is running', but rather a second-person proposition regarding an
ordinary proposition - for example, 'You respond affirmatively to the
proposition "Socrates is running" ' or 'You deny that Socrates is running',
or something of the sort which makes reference to an action on the part of
the respondent with a proposition as its object. Sit verum, on the other
hand, is the species of obligations in which the main proposition at issue
always includes mention of the respondent's knowing, his not knowing, or
his being in doubt. 2 9 So, for example, the first proposition put forward by
the opponent in a case of iii verum is not 'Socrates is running' or even 'You
grant that Socrates is running', but 'You know that Socrates is running', or
'You do not know that Socrates is running', or 'You are in doubt whether
26. Ibid., Burley 0.02 and 1.01.

27. Ibid., Burley 1.03.
28. Ibid., Burley 2.0I and 2.02.
29. Ibid., Burley 6.01-6.03.
Obligations 321
Socrates is running'. Burley gives this variety of obligations short shrift.

But it may be that it played an important role in stimulating the investiga-
tion of epistemic verbs in the later fourteenth-century treatises De scire et
dubitare and in the sections of the treatises De sensu composito et diviso that
deal with epistemic verbs, where there are sometimes strong, explicit
connections with obligations. For example, Paul of Pergula in his treatise
on the compounded and divided sense concludes with a discussion of
epistemic verbs, which he ends by saying that this material cannot be
perfectly understood except after a study of obligations, which is a neces-
sary prerequisite, for 'otherwise no one will understand what is said in this
material [on the compounded and divided sense in the context of epistemic
verbs]'. 30 Depositio and dubitatio are just like positio, except that while the
respondent is obligated to maintain the proposition that is the positum as
true, he is obligated to maintain the depositum as false and the dubitatum as
uncertain (of unknown truth-value). 31 Consequently, the basic rules of
obligations for positio, which apply to petitio, institutio, and sit verum as well,
do not apply to depositio and dubitatio, each of which has its own set of basic
rules.32
Burley begins his discussion of positio, as he does his discussion of
obligations at the opening of the treatise, with a division into species. 33 The
various species of positio are determined by considering the sorts of pro-
positions which can be the positum. Such propositions can be possible or
impossible, simple or composite. Those that are composite can consist in
two propositions joined by a conjunction (conjoined positio) or a disjunc-
tion (indeterminate positio), or they can consist in propositions with at-
tached conditions - for example, 'If you respond affirmatively to the first
propositum, let the positum be "You run", and otherwise not' 3 4 (dependent
positio). This last species of positio is divided into two further sub-groups,
depending on the nature of the condition attached to the positum. If the
condition specifies that after a certain time or a certain point in the
disputation, the positum ceases to be the positum, we have positio cadens; and
if the condition specifies that after an interval or at a certain point it
30. Paul of Pergula 1961, p. 158.307-15. I am grateful to Norman Krct2mann for calling my
attention to this material.
31. The rules for dubitatio bear at least some resemblance to three-valued logic; and scholastic attempts
to solve the problem of foreknowledge and free will by using three-valued logic arc sometimes
couched in obligations-terminology. See, for example, Ferdinand of Cordoba, in Baudry 1950,
p. 145. I am grateful to Anthony Kenny for the reference.
32. Green forthcoming, Burley 4.01fr. and 5.oiff.
becomes the positum again, then we have positio renascens. So Burley has in
mind something like this division of positio:
POSITIO
I
Possible Impossible
I I
Simple Composite
r 1 •
Conjoined Indeterminate Dependent
. « !
[Simple) Cadens
. ' 1
[Simple] Renascens
For positio, Burley gives three fundamental rules of obligations: 35
(i) Everything which follows from (a) the positum, with (b) a granted proposition
or propositions, or with (c) the opposite(s) of a correctly denied proposition or
propositions, known to be such, must be granted.
(2) Everything which is incompatible with (a) the positum, with (b) a granted
proposition or propositions, or with (c) the opposite(s) of a correctly denied
proposition or propositions, known to be such, must be denied.
(3) Everything which is irrelevant (impertinens) [that is, every proposition to
which neither rule (1) nor rule (2) applies] must be granted or denied or
doubted according to its own quality, that is, according to the quality it has in
relation to us [i.e., if we know it to be true, we grant it; if we know it to be false,
we deny it; if we do not know it to be true or do not know it to be false, we doubt
it).
These rules, together with the original definitions of the basic species of
obligations, are to obligations what the rules of tennis are to tennis: they
constitute the enterprise, making it what it is and differentiating it from its
near relatives. The way in which the rules and definitions give structure to
obligations can be seen in the examples which follow.
In general, in discussing positio, Burley proceeds by a successive con-
sideration of paradoxes. A paradox begins with what Burley calls a casus,36
which is a stipulation (frequently altogether fictional) about the nature of
reality outside the obligational disputation. The edit« is followed by the
positum, which is in some way or other at odds with the casus. For example,
a typical casus might stipulate that Socrates and Plato are black, and the
positum will be that Socrates is white. After the positum a series of pro-
positions are put forward by the opponent (the proposita), to each of which
the respondent must respond in a way governed by the three basic rules of
35. Ibid., Burley 3.14-16.
Obligations 323
obligations. At some point, the opponent puts forward a propositum to

which the respondent can give no correct answer, because any answer he
gives involves him in contradiction, because it commits him either to
assigning contradictory truth-values to the same proposition or to as
signing the same truth-value to contradictory propositions. At that point
the disputation is over; the respondent has lost. Burley then offers a solution
to the respondent's difficulty.
With that background, I want to look at some typical examples of
positio. Consider the following case:
And let this be posited: 'You are in Rome or that you are in Rome is to be
granted.' Then let this be put forward: 'That you are in Rome is to be granted.'
This is false and irrelevant, and therefore it must be denied. Then let this be put
forward: 'That you are in Rome follows from the positum and the opposite of
something correctly denied.' This is necessary because this conditional is necessary:
'If it is the case either that you are in Rome or that that you are in Rome is to be
granted, and that you are in Rome is not to be granted, then you are in Rome.'
Once this is granted, [namely], 'That you are in Rome follows from the positum
and the opposite of something correctly denied', then let this be put forward: 'That
you are in Rome is to be granted.' If you grant [this], you have granted and denied
the same thing, and therefore [you have responded] badly. If you deny it, the
disputation is over; you have denied something which follows by a rule [of
obligations]. Because if the rule is good, then this follows: 'That you are in Rome
follows from the positum and the opposite of something correctly denied; there
fore, that you are in Rome is to be granted.' 3 7
We can show the structure of this example in this way.
Opponent Respondent Reason

(1) You are in R o m e or that you are in (ia) Τ (lb) (1) is the positum
R o m e is to be granted.
(2) That you are in R o m e is to be (2a) F (2b) (2) is irrelevant and false
granted.
(3) That you are in R o m e follows from (3a) Τ (3b) (3) follows from the truth
the positum and the opposite of some- of (1) and the falsity of
thing correctly denied. (2)
(4) That you are in R o m e is to be (4a) —
granted.
37. 'Item, probatur quod non omne sequens ex posito et opposito bene negati sit concedendum. Et
ponatur ista: tu es Romae vel "te esse Romae" est concedendum. Deinde, proponatur: "te esse
Romae" est concedendum. Haec est falsa et impertinens, igitur neganda. Deinde, proponatur:
"te esse Romae" sequitur ex posito et opposito bene negati. Haec est necessaria, quia haec
conditionalis est necessaria: si tu es Romae vel "te esse Romae" est concedendum, sed "te esse
Romae" non est concedendum, igitur tu es Romae. Concessa ista '"te esse Romae" sequitur ex
posito et opposito bene negati', proponatur ista: "te esse Romae" est concedendum. Si concédas,
idem concessisti et negasti, igitur male. Si ncges, cedat tempus, negasti sequens per regulam. Quia,
si régula sit bona, tunc sequitur: "te esse Romae" sequitur ex posito et opposito bene negati,igitur
"te esse Romae" est concedendum.' Ibid., Burley 3.21.
324. The high middle ages: propositions and modalities
Trouble arises for the respondent at (4a), on Burley's view, in this way.
The respondent cannot grant (4) since he has already denied the same
proposition at (2a). O n the other hand, he cannot deny (4) either. From (3)
it follows that you are in R o m e ; and if it follows, it must be granted, and so
(4) is true.
Burley's solution to this paradox looks bizarre. O n his view, 3 8
It is said that this should be denied: 'That you are in Rome follows from the positum
and the opposite of something correctly denied.' Nor is this [(3) ] necessary. Even if
it were necessary that it follows from the posited disjunction with the opposite of
one of the disjuncts that you are in Rome, nonetheless it is not necessary that this
disjunction be posited.
O n the face of it, then, Burley is saying that (3) is to be denied because it is
not necessary, and his reason for claiming that (3) is not necessary is that
one of the premisses it is derived from, namely, (1), is not necessary. But in
the first place, there is no obligations rule to the effect that we must deny
any propositions which are not necessary. And in the second place, if we
had to deny as not necessary any proposition which followed from the
positum on the fallacious grounds that it was not necessary to posit this
particular positum and therefore the positum was not necessary, we would
have demolished rule (1) of obligations and with it the whole precarious
structure of obligations.
But I think w e do not have to read Burley's solution as quite so bizarre. If
we look again at (3) and the reasons given for granting (3), we should see
that there is an important confusion there. Let ρ stand for 'You are in
R o m e ' and q for 'That you are in R o m e is to be granted'. Then what we
have in steps (1)— (3) of this argument is this:
Opponent Respondent Reason
(Γ) ρ V q (l'a) Τ
(2') q (a'a) F
(3') ρ follows from the positum and the (3'a) Τ (3'b) (3) follows from the
opposite of something correctly truth of (1) and the
denied. falsity of (2)
N o w (3') does not follow from the truth of (1/) and the falsity of (2'), as
(3'b) alleges; ρ follows (as Burley acknowledges), but (3') does not.
Suppose we then consider (3) irrelevant and judge it on its own merits.
38. "Dicitur quod haec est neganda: ' "te esse Romae" sequitur ex posito et opposito bene negati '. Nee
haec est necessaria. Et, si sit necessarium "te esse Romae*' sequi ex disiunctiva, quae ponitur, cum
opposito alterius partis, tamen non est necessarium istam disiunctivam poni.' Ibid., Burley 3.22.
Obligations 325
According to Burley, we ought to judge it false. If (3) had read 'That you
are in R o m e follows from the conjunction of this disjunction - namely,
"You are in R o m e or that you are in R o m e is to be granted" with the
denial of the second disjunct', then Burley would, I think, agree that (3)
had to be granted. But (3) does not read that way; and if we are considering
(3) as it stands, on its o w n merits, then Burley seems to think we should
deny it because the phrase 'from the positum' need not refer to the particular
disjunction which just happens to be the positum in this particular case of
obligations. It could refer to some other positum; but unless it refers to this
one, (3) is false. Therefore, on Burley's view, it is a mistake to grant (3).
This second and, I hope, more charitable reading of Burley turns what
was an unintelligible solution into an intelligible red herring, I think.
Because suppose that we deny (3), on the rather dubious grounds that it is
irrelevant and false in virtue of the ambiguous phrase 'from the positum'.
Then let (4) be put forward again. Remember that the reasoning for
granting (4) in the paradox went like this: (a) that you are in R o m e follows
from the positum and the opposite of something correctly denied; therefore
(c) that you are in R o m e is to be granted. And what Burley attacks is not the
inference from (a) to (c), but the truth of premiss (a). He seems to have in
mind the mistaken notion that if (a) is false, (c) is false — hence we can deny
(4) and the paradox is solved. But, of course, he has not solved the paradox
by such means since (c) can be true even if (a) is false. Furthermore, if
Burley is willing to accept the inference from (a) to (c), we can show him
that on his own views (c) must be true. The inference from (a) to (c)
depends on accepting as valid the inference from (b) 'You are in R o m e ' to
(c) "That you are in R o m e must be granted'; and there is some reason for
accepting the inference from (b) to (c) in an obligations disputation. If the
proposition 'You are in R o m e ' is true in an obligations disputation, it is to
be granted. And so in an obligations disputation, the inference from ρ to 'p
is to be granted' is a good one. Or, to put it another way, if the respondent
must grant p, he must also grant that he must grant p.
N o w consider step (4) of the argument, supposing for the moment that
we have accepted Burley's argument and have denied (3) for the reasons he
gives. Even with the denial of (3), however, from the truth of (1) and the
falsity of (2), it follows that you are in R o m e . And so, given the validity of
the inference from (b) to (c), from 'You are in R o m e ' it follows that 'You
are in R o m e ' is to be granted. Hence even with the denial of (3), we are
committed, on Burley's own views, to the truth of (4). And so, even if we
give Burley all he wants, the falsity of (3) and the validity of the inference
from (a) to (c), we can demonstrate that the paradox remains. The re
spondent is committed to the truth of (4) although he has correctly denied
the same proposition at (2a).
The real solution to this paradox depends on considering at the outset
the relations between the two propositions 'p', 'Youare in R o m e ' , and 'q',
"That you are in R o m e is to be granted'. It will help m this discussion to
take 'is true' as meaning 'is true in an obligational disputation' or 'must be
granted by a respondent', and analogously for 'is false'. If ρ is true, q is true,
as I have already argued. And we can also show the converse, that if q is
true, ρ is true. If you grant q, you are granting that ρ is to be granted. But if
you grant that ρ must be granted, you cannot consistently deny p, and so
you must also grant p. Hence, if q is true, so is p. Consequently, ρ and q are
logically equivalent. N o w in the obligations paradox we are considering,
the positum, which we are obligated to maintain as true, is a disjunction of ρ
and q. But since ρ and q are equivalent, the disjunction can be true only in
case both ρ and q are true, because
[(p<->q) Λ (p V q)] -> (ρ Λ q)

Hence, in this case, it is a mistake to consider (2) in the schématisation as
irrelevant. In fact, it follows from the positum and the implicit logical
relationship between ρ and q. And so (2) ought to be granted. In this way, I
think, the paradox is really solved; or, more accurately, in this way the
paradox fails to arise in the first place.
Finally, for historical purposes, 3 9 it is worthwhile considering again
Burley's solution to the paradox on my second interpretation of his
39. It is also worth pointing out that Green takes this paradox to be an example of disagreement
between Burley's obligations and the putative Sherwood obligations because, he says. (Intro.:
Chapter V), Burley denies what 'Sherwood' concedes, namely, that the proposition 'You must
grant that you are in Rome' entails the proposition 'You are in Rome'. 'Sherwood' in the passage
Green has in mind (Ibid., Sherwood ι .56 and 1.58) does address just that issue, and he does grant
the validity ofthat inference. Burley is, of course, not addressing that issue at all in this paradox of
his. He is, as we have seen, concerned instead with the proposition 'That you are in Rome follows
from the positum and the opposite of something correctly denied." And his concern with this
proposition and his overall purpose in considering this paradox stem from a worry over the
acceptability of the obligations rule that whatever follows from the positum with the opposite of
something correctly denied must be granted (cf. Burley 3.21). But in so far as we have any
evidence at all from this paradox about Burley's attitude toward the inference from (c) tcr(b) -
'(c) That you are in Rome is to be granted; therefore, (b) you are in Rome' - it seems plain that he
would accept it, because the sorts of reasons for accepting the converse inference from (b) to (c) -
'(b) You are in Rome; therefore, (c) that you are in Rome is to be granted' - also support the
inference from (c) to (b). And, as we saw, Burley clearly does accept the inference from (b) to (c).
So this paradox of Burley's in fact shows us a similarity between the two treatises, rather than a
difference, as Green thinks.
Obligations 327
solution. On that interpretation, Burley is taking (3) as irrelevant; and so,

by the obligations rule for irrelevant premisses, he considers (3) on its own
merits. But the way in which he does so suggests that he is giving the rule
for irrelevant premisses an extreme interpretation (although his interpreta-
tion of it may well have been typical for the period). Burley's decision to
judge (3) false stems apparently from his feeling that, after all, the phrase
'from the positum' which occurs in (3) could refer to just any positum. But,
of course, it cannot be taken to refer indifferently in that way if we
remember that (3) is occurring within this obligations disputation. To feel
the force of the ambiguity of the phrase 'from the positum', it is necessary to
wrench (3) out of its context in this disputation and to consider it altogether
apart from the disputation in which it occurs. It is as if Burley took the
obligations rule for irrelevant premisses to read in this way: If a proposition
does not follow from or is not compatible with what has been previously
maintained, it is irrelevant; and if it is irrelevant, then ignore everything
said in the disputation so far - in fact, ignore the fact that you are in an
obligations disputation - and judge the irrelevant proposition as you
would if you had never even heard of obligations disputation. That
extreme interpretation of the rule for irrelevant propositions has signifi-
cant implications for the whole function of obligations, as we will see more
clearly when we consider the changes Richard Kilvington makes in this
rule.
But as for Burley, the function of obligations at the stage of develop-
ment represented by his treatise and that attributed to Sherwood can be
summarised in this way. 40 Obligations in these treatises is a complicated set
of rules for consequences or inferences which are set in a disputational
context, where the disputational context makes a difference to the evalu-
ation of the inferences. Frequently, complications for the evaluation of an
obligations consequence arise because there is a reference in the premisses
themselves to the evaluator of those premisses or to an action by the
evaluator (such as granting or denying something) within the disputation.
In the example above, for instance, the validity of the crucial inference
from (b) 'You are in Rome' to (c) 'That you are in Rome must be granted
[by you]' depends on the disputational context and the references in the
inference to the evaluator of it. The inference 'p; therefore, Socrates must
grant p', or even 'Socrates is in Rome; therefore Socrates must grant that
40. For detailed argument for this conclusion, see Stump 1980 and Obligations according to Walter
Burley" in Stump forthcoming c.
he is in R o m e ' is invalid. Socrates may have been brought to R o m e while

he was in a coma, for example, or while in the hands of kidnappers who
deceive him about his location, and so on; and so it can be true that Socrates
is in R o m e without its being true that Socrates must grant that he is in
R o m e . But we need not have a reference in the premisses to the evaluator
of those premisses in order to generate the sort of difficulties of interest to
medieval logicians working on obligations. Other characteristics of dispu-
tation, such as the passage of time, will do just as well. Consider, for
example, this very simple disputation argument, which Burley presents in
connection with a different issue (namely, to show that we must sometimes
and without penalty grant an impossible proposition if it has become
impossible in the course of the disputation): 41
'For example, let A be the instant at which the positum is posited. And let it be
posited that it is A. Then let [this] be put forward: "It is A". This should be granted
since it is the positum and yet it is impossible per accidens. ...'
Because A designates the instant at which the positum is posited, the prop-
osition 'It is A' is true only at the instant at which the positum is posited. At
any step of the argument after the first, the proposition 'It is A' is not only
false but impossible, since it maintains in effect that an instant which is past
(namely, instant A at any time after step (i) of the argument) is present.
Nonetheless, the proposition 'It is A' is the positum, and so by the basic rules
of obligations it must be granted whenever it is put forward in the
disputation. And so this paradox arises solely in virtue of the passage of
time, which must be taken serious account of in an obligations disputation,
as it need not be or even should not be in a straightforward evaluation of
inferences outside the context of disputation.
What we have in Burley's obligations, then, as in the putative Sherwood
obligations, is a concern with special sorts of difficulties in evaluating
consequences or inferences as a result of the disputational context in which
the inferences occur. Given the enormous scholastic emphasis on dispu-
tation, it is not surprising that scholastics would develop an interest in
coping with inferences made tricky by the peculiarities of the disputational
context. But any practical concerns they may have had in connection with
obligations seem to have been quickly swamped by theoretical interest in
the complexities of obligations inferences for their own sake.
41. 'Verbi gratia: sit A: instans in quo ponitur positum. Et ponatur A esse. Et proponatur: A est. Hoc
debet concedi, quia est positum. Et tamen est impossibile per accidens, nee est hoc inconveniens.'
Ibid., Burley 3.59.
Obligations 329
Richard Kilvington
A clear and historically significant example of theoretical, philosophical
interest in obligations can be seen in the forty-seventh of Richard
Kilvington's Sophismata,*2 written around 1325, more than twenty years
after Burley's obligations. Kilvington's S47 is a sophisma and not a stand-
ard paradox of disputation, and so it has a somewhat different form from
the problems presented by Burley. It consists of a hypothesis and the
sophisma sentence, which is to be proved or disproved. The hypothesis is a
conjunction of conditionals:43
(H) If the king is seated, you know that the king is seated; if the king is not seated,
you know that the king is not seated.
The sophisma sentence is 'You know that the king is seated'. The
hypothesis, together with the tautologous premiss 'Either the king is
seated or the king is not seated' entails this disjunction:
(HD) (i) You know that the king is seated, or (ii) you know that the king is not
seated.
Now infact, because you are in Oxford and the King is in Westminster, this
conjunction is true:
(C) (i) Y o u d o not k n o w that the king is seated, and (ii) y o u d o not k n o w that the
king is not seated.
The entire conjunction (C) is plainly incompatible with (HD), the disjunc-
tion entailed by the stipulated hypothesis in S47 (the hypothesis here plays
the same role as the positum in Burley's obligations). But neither conjunct
of (C) taken separately is incompatible with (HD); each conjunct is in fact
irrelevant in respect of (HD). And the basic problem of the sophisma arises
from these relationships between (C) and its conjuncts, on the one hand,
and (HD), on the other. (HD) entails that (C) cannot be true and, con-
sequently, that one or another of the conjuncts of (C) must be false; but it
does not entail that one particular conjunct must be false. Both the Proof
and the Disproof for this sophisma depend on that fact.
The Proof proceeds in this way. We posit (HD), which must be ac-
cepted. Then we put forward (C) (ii); this is irrelevant, and judged on its
42. For a full exposition of this sophisma of Kilvington's, see Kretzmann's commentary on it in
Richard Kilvington forthcoming.
43. 'Supposito isto casu, quod si rex sedcat, tu scias regem sedere; et si rex non sedeat, tu scias regem
non sedere.' Ail citations of Kilvington's text are from Kretzmann's edition in Richard
Kilvington forthcoming.
own merits, outside the context of the sophisma, it is true. But (HD) and
(C) (ii) together entail (HD) (i), the sophisma sentence; and so the sophisma
sentence is proved. The Disproof works in precisely the same way, except
that the second step is (C) (i), instead of (C) (ii). The conclusion derived is
(HD) (ii), and this entails the denial of (HD) (i), which is the sophisma
sentence. 44 From a possible proposition (namely, (HD)), then, Kilvington
has by the rules of obligations derived contradictory propositions. He
considers a variety of attempts to solve this problem, and he shows
effectively that none of them succeeds. Consequently, what his Proof and
Disproof show us is an incoherence in the basic rules of obligations, an
incoherence apparently unrecognised during the stage of obligations
represented by Burley. The trouble in Kilvington's example is brought
about by the basic rule of obligations which has to do with irrelevant
propositions. The rule brings it about that a second and different set of
criteria for determining the truth-value of propositions is sometimes
brought into play in assessing the premisses of obligations arguments. In
general, in obligations disputations we assess the truth or falsity of prop-
ositions proposed by considering them with regard to the false positum and
other previously granted or denied propositions, if there are any. But when
a proposition is irrelevant, we assess it instead by considering only the state
of affairs actually obtaining in the world as it now is (or the state of affairs
stipulated in the casus). If we construct a conjunction (or disjunction) in just
the right way, we can ensure that the individual conjuncts (or disjuncts) are
assessed by criteria different from those used to assess the truth-value of the
whole conjunction (or disjunction) if the conjuncts (or disjuncts) are put
forward separately at the same step of the argument as the conjunction (or
disjunction). So, in S47, for example, either conjunct of (C), put forward as
the second step of the argument, is irrelevant; consequently, assessed
outside the sophisma context, either conjunct is true. But the whole
conjunction, put forward at step (2), is not irrelevant; consequently, as-
sessed within the sophisma context, it is false, which entails that at least one
of the conjuncts is false.
44. 'Tunc probatur sophisma sic. Tu scis regem sedere vel tu scis regem non sedere, sed tu non scis
regem non sedere; igitur tu scis regem sedere. Maior patet per casum, et minor patet quia est vera
non repugnans. Quod patet, nam ista non repugnant: "Si rex sedet, tu scis regem sedere; et si rex
non sedet, tu scis regem non sedere" et "Tu non scis regem non sedere".
Ad oppositum arguitur sic. Tu scis regem non sedere; igitur tu non scis regem sedere.
Antecedens patet, quia tu scis regem sedere vel tu scis regem non sedere, sed tu non scis regem
sedere; igitur tu scis regem non sedere. Et per consequens sophisma est falsum. Et minor patet ut
prius, quia est vera et impertinens.'
Obligations 331
Faced with the apparently insoluble paradox of S47, and the incoherence
in the rules for obligations which S47 brings to light, Kilvington revises the
rule for irrelevant propositions. The Kilvingtonian revision in effect trans-
forms the traditional rule into this:
(K) If a proposition is neither entailed by nor incompatible with what has been
previously maintained in the disputation, it is irrelevant; and an irrelevant
proposition is to be replied to on the basis of the way the world would be if
things were the way the positum says they are.45
The traditional rule for irrelevant propositions found in Burley's treatise
interrupted the obligations disputation and obliterated all its preceding
steps every time an irrelevant proposition was to be evaluated. (K) forbids
such a disconnected train of reasoning and ensures that the positum will
exert its influence over all succeeding propositions, including those prop-
ositions neither entailed by nor incompatible with the positum.
By using this new rule for irrelevant propositions, Kilvington solves the
original difficulty of S47. In accordance with (K), the second step of both
the Proof and the Disproof must now be evaluated on the supposition that
the hypothesis is true. O n this supposition, what we must conclude is not
that the proposition (C) (ii) 'You do not know that the king is not seated'
(the second step of the Proof) is plainly true. Rather, if the world were the
way the hypothesis says it is, you would not know whether (C) (ii) is true
or false; that is, you would not know whether or not you know that the
king is not seated. By the same token, it is not the case that the second step
of the Disproof, (C) (i) 'You do not know that the king is seated', is plainly
true; given the hypothesis, by (K), you do not know whether or not you
know that the king is seated. Kilvington here is apparently reasoning in this
way:
(1) If the king is seated, you know that the king is seated; if the king is not seated,
you know that the king is not seated. [This is the hypothesis (H).]
(2) You do not know that the king is seated, and you do not know that the king is
not seated.
(3) Therefore, you do not know whether you know that the king is seated or
whether you know that the king is not seated.
45. Kilvington's formulation of the rule in fact takes into account only those irrelevant propositions
which are true outside the context of the sophisma or obligations disputation, but there is no
reason to suppose that he would object to spelling out the rule for irrelevant propositions which
are false outside the context of the sophisma. 'Si tamen accipiatur iste terminus "impertinens" pro
propositione quae nunc est vera et quae non foret vera ex hoc quod ita foret ex parte rei sicut
significatur per positum, tunc dico quod ista propositio '"Tu es Romae" et "Tu es episcopus"
sunt similia' est impertinens huic posito, quod est "Tu es Romae". Quia si tu esses Romae et non
esses episcopus, haec foret neganda: ' "Tu es Romae" et "Tu es episcopus" sunt similia.' '
Since the hypothesis ascribes to you infallibility regarding the king's

position, it must be the case that you know he is seated or you know he is
not seated. In view of premiss (2) above, then, what we must conclude is
that you do not know which of those alternatives you know; that is, it is
possible to know something and yet not know that you know it. The
disjunction (HD) can be true, and you can grant it as true, without your
thereby being in a position to grant or deny either of the disjuncts; and that
is the case the example in S47 illustrates. Consequently, for both the Proof
and the Disproof of S47, instead of granting the second step of the
argument as irrelevant and true, the respondent should take it as irrelevant
in Kilvington's sense and doubt it. 4 6 Hence, we can neither prove nor
disprove the sophisma sentence; and the original paradox is resolved.
What Kilvington has done in his work on S47, by his change in the rule
for irrelevant propositions, is to shift the whole purpose of obligations. The
new rule (K), when added to the rest of the traditional structure of
obligations, shifts the emphasis of obligations away from a consideration of
scattered paradoxes arising from difficulties in evaluating certain inferences
in disputational contexts and towards a logic of counterfactuals. 47
William Ockham
Something of the same shift of emphasis, though much less dramatic or
historically significant, I think, can be seen in Ockham's work on obliga-
tions in his Summa logicae, written only a year or so before Kilvington's
Sophismata.*8 One of the things that apparently interested Ockham most in
the study of obligations is the case when the proposition that is the positum is
an impossible proposition.
Ockham begins his discussion of an impossible proposition in the/><wi»M»i
in a way that seems puzzling at first glance, by claiming that there is a great
difference between an impossible proposition in the positum and a con-
sequence composed of impossible propositions. 49 The consequence 'God is
46. 'Ad argumentum concedendum est quod tu scis regem sedere vel tu scis regem non sedere. Sed
minor coassumpta est dubitanda - scilicet, haec: "Tu non scis regem non sedere". Quia si rex
sedet, tu non scis regem non sedere — per casum — et antecedens est dubitandum ; igitur consequens
est dubitandum.'
47. For a discussion of counterfactual reasoning in connection with this sophisma, see Kretzmann's
commentary on it in Richard Kilvington forthcoming.
48. For a fuller discussion of Ockham on obligations, see 'Topics, Consequences, and Obligations in
Ockham's Summa logicae' in Stump forthcoming c.
49. Ockham 1974a, pp. 739ff. 'Similiter multum refert ponere istam propositionem "Deus non est
Deus" et inferre istam consequentiam "si Deus non est, Deus non est Deus"; propter quod ista
condicionalis sive consequentia concedenda est ct positio non est recipienda, et quandoque
positio est recipienda tamquam impossibilis ct sustincnda est in positione impossibili. Et tarnen
Obligations 333
not three persons; therefore God is not God' is a good one, on Ockham's
view; it is, in fact, a material consequence, warranted by the seventh of
Ockham's general rules of consequences, namely, that anything follows
from an impossible proposition. Nonetheless, according to Ockham, if the
antecedent of this consequence - 'God is not three persons' - is put in the
positum a respondent ought to deny the consequent, 'God is not God".
On the basis of this puzzling claim, Ockham makes two further sets of
stipulations about the positum when the proposition posited is impossible.
In the first place, he maintains that not every impossible proposition can be
posited, but only those which do not manifestly or obviously entail a
contradiction. In the second place, not all the rules for positio where the
proposition posited is possible hold for the positing of an impossible
proposition; in particular, not all the rules of consequences hold for cases in
which the proposition posited is impossible. No as-of-now consequences
or material consequences are acceptable in any obligations disputation
where the positum is impossible.
This second stipulation makes sense out of the original claim, namely,
that although the consequence 'God is not three persons; therefore God is
not God' is a good consequence, the consequent is to be denied if the
antecedent is posited. The consequence in this case is a material con-
sequence; and if all such consequences are to be rejected in cases where the
proposition posited is impossible, then, of course, the respondent cannot
maintain that 'God is not God' follows from 'God is not three persons'. But
why would Ockham stipulate that material consequences are to be rejected
in such cases? There are, I think, two reasons - the first practical and the
second philosophical - for his stipulation. If we accept the rule that
anything follows from an impossible proposition in an obligations dispu-
tation where the positum is impossible, it will not be possible to have
anything but a trivial disputation, because no matter what proposition is
put to the respondent, it is clear from the outset that the respondent will
have to grant it as following from the proposition posited. Hence if we
accept this rule of consequences in the context of an obligations dispu-
tation, we cannot have a disputation which is of any philosophical interest.
Secondly, if we reject this rule of consequences in such obligations and
allow only simple, formal consequences, the result will be philosophically
interesting, because what we will have in such cases then is something very
consequentia seu condicionalis bona est; sicut ista consequentia bona est "Deus non est très
personae, igitur Deus non est Deus"; et tamen antecedens potest recipi in positione impossibili et
debet ncgari consequens' (p. 739.9-15).
roughly similar to a logic of counterfactuals for a restricted class of prop

ositions. The purpose of obligations disputations where the positum is
impossible is apparently to see what is entailed if one takes as true such
propositions as 'God does not exist', 'The Holy Spirit does not proceed
from the Son', or 'Man is not capable of laughter*. This enables us to
understand Ockham's other stipulation, namely, that impossible prop
ositions which can be used in the positum must not manifestly and ob
viously entail a contradiction. If they did obviously entail a contradiction,
the disputation would end quickly in an uninteresting way. The opponent
would derive the contradictory propositions entailed by the posited prop
osition, which the respondent would be forced to grant; and the dispu
tation would be over because the respondent has granted contradictory
propositions. Furthermore, even if the opponent agreed to forego such a
cheap victory, interesting and rational results are still not likely to follow
since all sorts of incompatible propositions are quickly derivable from the
impossible proposition in virtue of its obviously entailing contradictories.
Conclusion
We are not yet in a position to assess either the significance of this shift of
emphasis in the work of Ockham and Kilvington for later developments in
obligations or the historical importance and influence of Kilvington's
treatment on subsequent discussions of obligations, though there is some
reason to suppose that both were great. 5 0 But an understanding of
Kilvington's influence, as well as of the subsequent changes in scholastic
theories of obligations, must wait until a good deal more work has been
done on obligations in the fourteenth century. 5 1
50. It seems likely that Kilvington's work had some influence on Roger Swineshead's obligations,
which was an important source for subsequent changes and developments in obligations (see Part
Β of this Chapter). I hope to investigate Swineshead's obligations in detail in a forthcoming paper.
Si. I am grateful to Norman Kretzmann for numerous helpful comments and suggestions on an
earlier draft of this chapter and to John Crossett, whose efforts on my behalf made this work
possible.
OBLIGATIONS:
Β. DEVELOPMENTS IN
THE F O U R T E E N T H C E N T U R Y
The old and the new responses

The obligations-literature appears to have entered a new phase with the
Oxford Calculators, centred at Merton College in the 1320s and 1330s.
Although the Mertonian Thomas Bradwardine seems to have contributed
litde to the development of obligations,1 his contemporary Richard
Kilvington was more innovative in this regard, in ways described in Part A
of this Chapter.
We are in a somewhat better position to assess the contribution of Roger
Swineshead to obligations. Swineshead certainly appears to have been part
of the intellectual circle with which Kilvington and Bradwardine are
associated, and he may well have studied with them. 2 Probably sometime
after 1330 and before 1335,3 Swineshead wrote his pair of treatises on
obligations and insolubles.4 There is reason to speculate that some of the
most characteristic features of Swineshead's Insolubilia grew out of reflec
tion on Kilvington's Sophismata.s It is possible that this is true of
Swineshead's Obligationes as well, but that remains to be established.
Swineshead's Obligationes is markedly different from earlier treatises in
the genre. So true is this that Robert Fland, writing some time between
6
1335 and 1370, distinguishes two separate traditions in the obligations-
literature. One of these traditions he calls the 'old response' (antiqua re-
sponsio)1; it conforms to the views of Burley, to those of the treatise
attributed to William of Sherwood, and to those found in most if not all of
8
the other early treatises. The second tradition Fland calls the 'new re-
1. See the list of his works in Weisheipl 1969, pp. 177-83.

2. See Weisheipl 1964, especially p. 237. Roger Swineshead is not to be confused with Richard
Swineshead, another of the Oxford calculators.
3. Ibid., pp. 244-7.
4. Edited in Spade 1977 and Spade 1979a.
5. See Chapter 12 above, n. 43.
6. See Spade 1978c, pp. 56-02.
7. Ed. Spade 1980, pars. 13 and 20.
8. See the texts edited in De Rijk 1974-6, and in Green forthcoming.
sponse' (nova responsio)9; it appears to have originated with Swineshead.

Fland lists two rules as characteristic of the new response. 1 0 First, it is
possible that a copulative (conjunctive) sentence is to be denied, even
though both conjuncts have been granted. Second, it is possible that a
disjunctive sentence is to be granted, even though both disjuncts have been
denied. These rules are in fact to be found combined in Swineshead's
second 'conclusion'. 1 1
The basis for this remarkable conclusion was a fundamental change in
the rules governing a respondent's replies to proposed sentences. The
change centred on the notion of relevance (pertinentia). For simplicity, let
us confine our remarks to the species of obligation called positio.12
According to the old response, a relevant sentence was one that either
followed from (pertinens sequens) or was inconsistent with (pertinens re-
pugnans) the conjunction of (a) the posited sentence, together with (b) all
subsequendy proposed sentences that had been granted by the respondent,
and together with (c) the contradictory opposites of all subsequently
proposed sentences that had been correctly denied by the respondent. 1 3
For the new response, however, a relevant sentence was one that either
followed from or was inconsistent with the posited sentence alone. The
subsequent steps of the disputation played no role in assessing relevance or
irrelevance. 1 4
Hence if the false sentence P i s posited, it must be granted whenever it is
subsequently proposed, since it (trivially) follows from the posited sen
tence, namely from itself. And if the true but irrelevant sentence Q is
proposed anywhere in the same disputation, it must likewise be granted.
But if the conjunction Ρ & Qis proposed anywhere in the disputation, it
must be denied, according to the new response, because that conjunction as
a whole neither follows from nor is inconsistent with P, and is moreover
false since Ρ is false. Similarly, if not-P is proposed anywhere in the
disputation, it is to be denied since it is inconsistent with the posited
9. Ed. Spade 1980, pars. 14 and 20.

10. Ibid., par. 17.
11. Ed. Spade 1977, p. 257, par. 32: 'Secunda conclusio: Propter concessionem partium copulativae
non est copulativa concedenda, nee propter concessionem disjunctivae est aliqua pars ejus
concedenda.'
12. For a description of this species and its relations with the others, see the discussion in Part A of this
Chapter.
13. See, for example, the treatise attributed to William o f Sherwood, ed. Green forthcoming, par.
1.04.
14. See Swineshead in Spade 1977, especially pars. 4 and 32, and Fland in Spade 1980, especially pars.
10-20.
Obligations 337
sentence P. And if not-Q is proposed anywhere in the disputation, it is to be

denied since it is irrelevant and false. But if not-P or not-Q is proposed
anywhere in the disputation, it is to be granted since it neither follows from
nor is inconsistent with Ρ and is moreover true since not-P is true. Hence,
15
Swineshead's second 'conclusion'.
One of the most characteristic features of the old tradition of obligations
was lost in this new response. For according to the old tradition, the order
in which the opponent proposed his sentences in the subsequent steps of the
disputation was an important factor in determining the respondent's
correct reply. 16 This fact was the basis for many complicated and tricky
examples in the old tradition. 17 But, according to the new response, the
order of the proposed sentences is not a factor, since relevance or irrele
vance is not affected by any steps of the disputation after the initial
'positing'.
The rationale behind Swineshead's new response is not at all clear.
Certainly, if the purpose of the obligations was to provide students with
logical exercises to sharpen their skills, the old response was much better
suited to that purpose than was the new response. (Nevertheless, perhaps
students were exercised by shifting from the one style of disputation to the
other.) It is possible that Swineshead's innovations were meant to take
account of the fact that a disputation de obligationibus takes place in practice
during an interval of time, and that during that interval there may be
changes in the world that are irrelevant to the posited sentence. ' 8 Given the
present state of research on obligations, this is no more than speculation.
But if it is true, it means that Swineshead also departed from his pred
ecessors insofar as the earlier treatises maintained that all the steps in a
disputation de obligationibus are to be taken to refer to the same moment of
time. 19
Swineshead's new response generated a certain amount of controversy.
His view was accepted by Richard Lavenham 20 in the second half of the
15. See Spade 1977, par. 32.

16. See, for example, Burley in Greenforthcoming,par. 3.34: 'Una régula de arte obligatoria est ista:
ordo est maxime attendendus. Et ratio hujus regulae est ista: quod uno loco est concedendum, alio
loco non est concedendum. Et ideo, quando aliquid proponitur, videndum est quern ordinem
illud habeat ad praeconcessa.'
17. See, for example, the texts edited in Green forthcoming, and some of the examples excerpted
from those texts and discussed in Part A of this Chapter.
18. See Spade 1977, p. 274 n. 97.
19. See, for example, the text attributed to William of Sherwood in Green forthcoming, par. 1.36,
and Burley in Green forthcoming, par. 3.84.
20. Ed. Spade 1978b.
fourteenth century, and was discussed by Robert Fland as apparently no

less plausible than the old response. Indeed, Fland appears generally
favourable to the new response. Ralph Strode, on the other hand, a
contemporary of John Wyclif, rejected Swineshead's second conclusion,
but observed that 'not a few sophisters' held it. 2 ' 'Many people', he says, 22
thought that they demonstrated conclusively from the common rules that a
copulative was to be denied, such as the one just proposed, each part of which was
nevertheless to be granted. And likewise they grant the disjunctive opposite of
such a copulative, while denying each of its parts. By this means not a few
sophisters are given courage to admit cases like 'Everything standing in this house
is a donkey'. Along with this, when 'You are standing in this house' is proposed,
they grant it. But when from these things it is concluded that you are a donkey,
they grant the consequence and deny the antecedent. And when it is asked 'In
virtue of which part?', they say that it is neither in virtue of the major nor in virtue
of the minor, but in virtue of the whole copulative composed of them, which is the
antecedent. They say that this is false and irrelevant since it is composed of a posited
falsehood and an irrelevant truth, as was said in the beginning. Therefore, al-
though this view is now upheld by many people, it could deservedly be called
sophistical, since it appears to be valid but is not valid.
Peter of Candia also rejected Swineshead's second conclusion in the later

fourteenth century, 2 3 as did Paul of Venice in the early fifteenth century. 24
By the time Paul wrote, he was able to say that those who held Swines-
21. Strode 1493, f. 95vl>. See n. 22 below. Bodleian Ms Canon. Misc. 219, f. 37**, explicitly mentions
the 'opinio Suyset* in this context. ('Suyset' is a common medieval spelling of Swineshead's
name.)
22. Strode 1493, f. 95 v b (corrected in a few places according to the Bodleian MS Canon. Misc. 219,
f. 39"): '... putavcrunt sc multi vclut inevitabiliter ex rcgulis communibus [omnibus ed. 1493]
demonstrare quod copulativa esset neganda cujus tamen quaelibet pars esset concedenda. Et
consimiliter concedunt disjunctivam oppositam tali copulativae negantes quamlibet ejus partem.
Per quod medium non pauci sophistae sunt animosi facti ad admittendum tales casus "Omnes
stans in ista domo est asinus". Et cum hoc cum proponitur "Tu stas in ista domo" concedunt. Sed
cum concluditur ex illis quod tu es asinus concedunt consequential!! et negant antecedens. Et
quando quaeritur pro qua parte dicunt quod [om. ed. 1493] nee pro majori nee pro minori sed pro
tota copulativa quae est antecedens ex illis composita. Quam dicunt esse falsam et impertinentem
cum sit ex posito falso et impertinenti vero, sicut primo dicebatur. Licet ergo modo a multis
sustineatur ilia via, merito tamen sophistica potent appellari cum sit apparens valida sed non
valet.'
23. Peter of Candia, Obligationes: 'Décima tertia régula est haec: Quod concessa qualibet parte
copulativae copulativa est concedenda... Et quia aliqui ponunt oppositum istius regulae pro
régula summa, ideo pro régula maiori declaratione arguo contra ipsos.' (I am grateful to Stephen
F. Brown for allowing me to see his partially completed edition of this work.) On Peter of
Candia, see Ehrle 1925 and Gray 1967.
24. Paul of Venice 1499, II, 14, f. I79 v b (on 'positio'): 'Ultima régula est ista: qualibet parte copulativa
concessa, concedenda est copulativa cujus illae vel consimiles sunt partes principales . . . Sed quia
aliqui tenent oppositum hujus regulae quasi omnes antiqui. pro majori declaratione arguo contra
COS.*
Obligations 339
head's view were 'quasi omnes antiqui'.25 Swineshead's new response, which
Strode in the above passage regarded as rather widespread in his own day,
appears to have been only a historical reference by the time of Paul of
Venice; it seems never to have been the dominant view. The old response
was apparently the norm throughout the history of the obligations-
literature, as far as present research has determined.26
There were several other features of Swineshead's view that were also
disputed. At present, it is impossible to say whether these features orig-
inated with Swineshead or came from some other source. Indeed, until
further research is done, it would be premature to think that there were
only two 'responses' in the obligations-literature, the standard one and
Swineshead's. The situation was almost surely much more complex.
Nevertheless, Swineshead did maintain, as the eighth assumption or
'supposition' in his treatise, the characteristic thesis that 'the response to a
sentence is not to be varied because of an imposition'.27 The idea seems
to have been to regard 'imposition', one of the main species of obligation,
as a kind of metalinguistic positing, so that once the respondent admits,
for instance, that the sentence 'God exists' is newly imposed to signify
that man is a donkey, he has to grant the sentence '"God exists" is
false' whenever it is proposed in the disputation, since it follows from the
admitted imposition. And yet the respondent must also grant the sentence
'God exists' whenever it is proposed in the disputation. To think that he has
to deny the latter sentence on the basis of the admitted new imposition is to
make an illegitimate move from metalanguage to object-language.
Not everyone accepted this thesis. Fland felt the need to discuss it at some
length,28 although he did not reject it. Richard Brinkley, however, did
reject it sometime during the second half of the fourteenth century.29 And
15. See n. 24 above. The most natural reading of this phrase is 'almost all the ancients*, implying that
at one time Swineshead's view was almost universally held. No other evidence has been found so
far that Swineshead's new response was ever widely accepted. On rhe other hand, the phrase
might also be read 'almost all [of them were] ancients'. This admittedly less natural reading would
not imply that Swineshead's view was ever the dominant one, but it would suggest what in any
case appears to be true, that the view had died out by the time Paul of Venice was writing.
26. In addition to the texts before Swineshead, in De Rijk 1974—6 and Green forthcoming, see Albert
of Saxony 1522, VI, 2, ff. 46""—siïb; the Tractatus de arte obligandi attributed to Peter of Ailly in
Peter of Ailly 1489 (this tract is perhaps in fact by Marsilius of Inghen); Paul of Venice 1472, tract.
V; Paul of Venice 1499, II. 14, ff. I77ta-192*; and Paul of Pergula 1961, tract. V.
27. Ed. Spade 1977, p. 254, par. 21: 'Propter impositionem alicujus propositionis ad illam non est
responsio varianda.'
28. Ed. Spade 1980, pars. 69-86.
29. Richard Brinkley, Obligationes, Prague, Bibl. univ., MS 396, f. 98ra: 'Octava suppositio: Propter
impositionem novam factam circa propositioncm est responsio ad earn varianda. Nam propter
impositionem significat propositio aliquid, aliqua vel aliqualiter, quod, quae vel qualiter prius
non significavit; igitur.'
30
John Buridan remarks,
Still another case frequently occurs in obligational disputations, namely, that in the
schools the master posits that for the time of those disputations the term 'donkey'
signifies to them precisely the same as the term 'animal' signifies to us according to
its usual signification. And the respondent and the others agree. Then the sentence
Ά man is a donkey' is true for them, and to be granted. Yet a vocally similar one
would be false and impossible if, without such an obligation, it should be posited in
the church of the Blessed Mary to those who were there.
This violates Swineshead's eighth assumption. Note that Buridan seems to

suggest that what he says here agrees with more or less standard practice in
the schools.
Fland points to yet another rule that he calls 'new' and seems to associate
with Swineshead. 3 1 His formulation of the rule is slightly different from
Swineshead's 3 2 and also from Lavenham's, whose own text is based on
Swineshead's. 3 3 But the idea is this. If a posited sentence is inconsistent with
the claim that it is posited, the respondent must reply to it, if it is proposed,
as though it were irrelevant. For instance, if the opponent says, Ί posit that
nothing is posited to you', the respondent must admit the position since the
posited sentence 'Nothing is posited to you' is a possible one. But if that
posited sentence should be proposed at a later step of the disputation, the
respondent must respond as though it were irrelevant. For although
'Nothing is posited to you' is quite possible in itself, it is inconsistent with
the claim Ί posit that nothing is posited to you', since if nothing is posited
to you, then in particular, 'Nothing is posited to you' is not posited to you.
The exact purpose of this suggestive rule and the details of how it works
are not yet well understood. Neither are its possible antecedents in the
earlier literature or its subsequent history in the later. Fland treats the rule as
a novelty. The rule is obviously meant to avoid certain kinds of paradoxes.
But exactly h o w those paradoxes would otherwise arise and exactly how
this rule disarms them are not yet clear.
30. Buridan 1977, VI, soph. 1, p. 105: 'Adhuc alius casus accidit communitcr in disputationibus
obligatoriis, videlicet quod in scholis magister ponit quod tempore illarum disputationum iste
terminus "asinus" significet eis praecisc idem quod significat iste terminus "animal" nobis,
secundum comtnunem eius significationem. Et respondens et alii consentiunt. Tunc ista pro
positio "Homo est asinus" est illis vera et concedenda. Et tarnen similis secundum vocem esset
falsa et impossibilis, si sine tali obligatione poneretur in ecclesia beatae Mariae illis qui ibi essent.'
31. Ed. Spade 1980.par.23: 'Alia régula est nova, scilicet, eodem modo respondendum est ad positum
et propter repugnans posito infra tempus obligationis sicut extra.'
32. Ed. Spade 1977, p. 265, par. 64: 'Si lamen positum répugnât position! respondendum est ad illud
sicut ad impertinens.'
33. Ed. Spade 1978b, p. 229, par. 8: 'Unde quandocumque propositio ponitur cum pertinemia
obligationis vel cum repugnantia positionis homo non tenetur concedere propositionem ob-
Iigatam in eo quod est obligata, sed si est vera debet concedere earn, si est falsa debet negare earn."
Obligations 341
Conclusion
The study of the obligations-literature is only just beginning to get under
way in earnest. But already it is obvious that such common and seemingly
innocuous expressions as Ί grant' (Concedo), Ί deny' (Nego) and 'when the
case has been admitted' (admisso casu) are highly rule-governed expressions,
rule-governed in ways that are codified in the obligations-literature itself,
but that spill over into other, more concretely philosophical contexts
as well. This had happened already by the time of Richard Kilvington's
Sophismata, and it became increasingly obvious in later literature, for
instance, in treatises De scire et dubitare.3* As the study of obligations
progresses, it may be expected to shed light on many now obscure argu
ments in such texts.
34. See, for instance, the influential Ch. 2 ('De scire et dubitare') of William Heytesbury's Regulae
solvendi sophismata, in Heytesbury 1494, ff. I2 va -ii v '.
17
MODAL LOGIC
Two notions of possibility

For Aristotle the term 'possibility' is homonymous (Pr. An. I, 3,25*37-40):
on some occasions the possible and the impossible are contradictories (e.g.,
De int. 12, 22*11-13; 13, 22*32-38), while on others possibility is incom-
patible not only with impossibility but also with necessity (e.g., An. pr. I,
13, 32*18-21).* Whenever the distinction is relevant, I shall call possibility
in the first sense 'possibility proper' and possibility in the second sense
'contingency'.
In the Latin translation of De interpretatione by Marius Victorinus which
was used by Boethius Aristotle's two terms for 'possible' were translated by
the Latin terms 'possibile' and 'contingens',2 which Boethius understood to
be synonyms. 3 This was the usual view in early medieval logic, 4 and it can
still be found in the squares of opposition for modalities presented by
Wdliam of Sherwood and Peter of Spain in the middle of the thirteenth
century. 5 Already in the twelfth century, however, there were attempts to
give separate meanings to the two words. For instance, John of Salisbury
criticised those who used the terms as synonyms; according to usage in his
time a mere absence of impossibility did not warrant calling something
contingent. 6 Even in those works in which the terms are used as synonyms
there often is a remark referring to a related distinction, according to which
'contingens' is opposed to 'necessarium' in the sense that some possible
sentences are necessary and others contingent. 7 This became the dominant
1. For Aristode's different sorts of possibility see Hintikka 1973, pp. 27-40.
2. Although Aristotle has two words for 'possible' - 'dunaton and 'endechomenon - he appears not to
use them systematically in order to mark a distinction. See Becker-Freyseng 1938, pp. 20-4.
3. Boethius 1880,11, 382.17-18; 383.4-5; 384.6-7 et passim. See, however, 11, 392.17-393.9, where
Boethius draws a distinction between enuntiatio possibilis and enuntiatio contingens.
4. See, e.g., Abelard 1970a, p. 193.31; De Rijk 1962-7 11(2), 39°·35~6; 3 9 i i 7 - i 8 ; 393~4i 43°-i;
481.14-17.
5. William of Sherwood 1937, p. 45; Peter of Spain 1972, p. 16.
6. John ofSalisbury 1929, III, 4.
7. De Rijk 1962-7,11(2), 481.9—13. For the whole question see Schepers 1963.
Modal logic 343
use of the words in later medieval logic. It was further encouraged by the
need of having available different terms for possibility proper and con-
tingency in dealing with the Aristotelian modal syllogistic.8 'Possibile' is
then reserved for possibility proper. Sometimes a subdivision is made
within the field of what we are calling contingency so that 'contingens'
describes what is true but can be false and 'possibile' describes what is false
but can be true.9
Because of the two senses of 'possibility' and the different fields of
application of possibility proper, Aristotle's treatments of the equipollence
and opposition of various modalities are rather complicated. His accounts
of them in De interpretatione, Chapters 12 and 13, are in fact incompatible
because different senses of'possibility' are presupposed in them. 10
In order to present a square of opposition for modal sentences (analogous
to that for sentences de inesse) early medieval logicians took possibility
proper as the basic notion. ' ' The rules of equipollence and opposition were
accordingly presented by many authors as in the following square:12
non possibile est non esse non possibile est esse

non contingens est non esse contrariae non contingens est esse
impossibile est non esse IMPOSSIBILE EST ESSE
NECESSE EST ESSE necesse est non esse
subalternae subalternae
POSSIBILE EST ESSE possibile est non esse

CONTINGENS EST ESSE contingens est non esse
sub- non impossibile est non esse
non impossibile est esse
contrariae non necesse est esse
non necesse est non esse
8. So Albert the Great in commenting on Aristotle's modal syllogistic uses the term 'contingens pro
possibili' for possibility proper; see Albert the Great 1890—93, tr. IV, cap. 4, 5, pp. 545—9.
9. See De Rijk 1962-7,11(2), 391.18-19; Roger Bacon 1940, p. 267.
io.Jaakko Hinrikka has described and diagrammed the different patterns of analysis used in De
interpretatione; see Hintikka 1973, p. 48.
11. In this respect they were following the Theophrastian and Eudemian modal logic rather than
Aristotle's, in which contingency is basic. See Bochenski 1947, pp. 73-87; 1951, pp. 73-5; 1956,
pp. 116-18; Graeser 1973, pp. 72-5.
12. De Rijk 1962-7,11(2), 393.6-394.5; 431.19-26; 484.5-6; William of Sherwood 1937, p. 45.11;
Peter of Spain 1972,1, 25; p. 16.12-13.
The principle of plenitude

Aristotle's modal theory evidently implies the so-called Principle of
Plenitude: 13
(P) No genuine possibility can remain forever unrealised.
Although Aristotle does not directly define modal terms with the help of
(P), it seems that (P) was in some way natural for him. The principle can be
found in its most explicit form in that part of Aristotle's modal theory
which has been described as the 'statistical' interpretation of modal con
cepts as applied to temporally indefinite sentences.14 Aristotle regarded a
temporally unqualified sentence as the typical form of a singular declara
tive statement (see, e.g., Cat. 4*23-b2, Met. I05i b i3-i8). Such sentences
contain implicitly or explicitly a reference to the time of utterance as a part
of their meaning. On the statistical interpretation of modal statements the
use of such sentences is connected by Aristotle with the principle (P): if a
sentence which is true now is true whenever uttered, it is necessarily true. If
its truth-value can change in the course of time, it is possible. And if a
sentence is false whenever uttered, it is impossible.15
Potentiality as a power striving to realise itself
Another important Aristotelian paradigm is the idea of potentiality as a

power that strives to realise itself The principle (P) is explicitly connected
to this doctrine, too. When Aristotle says that the absence of external
hindrances must be built into the definition of such a potentiality (see Met.
1048*23-4), it follows that a potentiality which is prevented from being
realised is only a partial potentiality, because while it is impeded it cannot
be realised. It is trivially true that all kinds of partial potentialities can
remain unrealised according to Aristotle. But when he is speaking of total
possibilities, which are the only ones that can be actualised, he is confronted
by the difficulty that if a singular total possibility is present, it seems to be
13. For the history of the principle in general see Lovejoy 1936. According to Lovejoy, the principle
originated in the ancient Platonic tradition and was not adopted by Aristotle. Hintikka has
disproved the second part ofthat view in Hintikka 1973, pp. 93-113.
14. Hintikka 1973, pp. 63-70, 84-6, 102-3, ι+9~5ΐ· See also Becker 1952, pp. 16-19.
15. For the link between time and necessity see, e.g., Nie. Eth. 1 i39 b i8-25, time and possibility Met.
i°47 b 3-9 and De caelo 28i b i 5-21, and time and impossibility Met. 1047Ί2-14; cf. also Hintikka,
Remes, and Knuuttila 1977, pp. 15-22, 31-2, 43-5. See also Bosley 1978, pp. 29-40.
Modal logic 345
realised immediately (see, e.g., Met. 1048*6-21, 1049*5-14, Phys. VIII, 4,

255 b 3-i2). 16
Aristotle's 'logical' definition of possibility
Aristotle's 'logical' definition of possibility might be recognised as another
important element of his modal theory. According to it, what is possible
can be assumed to be actual without any impossibility resulting from this
assumption (Pr. An. I, 13, 32*18-21). In many places he relies on the
definition outside strictly logical contexts, and in De caelo 28i b i5-2i he
explicitly says that if a possibility is never realised, something impossible
will follow from supposing it to be realised.17
The medieval use of the statistical interpretation of modality

Several of the concepts and principles characteristic of Aristotle's modal
theory remained important in post-Aristotelian philosophy,18 which is
marked by an implicit or explicit acceptance of the principle (P). The same
is true, with qualifications, of medieval thought until the late thirteenth
century. I shallfirstgive some examples of the medieval use of the statistical
interpretation of modality, because it seems to have been the dominant
model in logical contexts.
According to the twelfth-century Dialectica Monacensis,19 an assertoric
statement is changed into a modal statement by increasing or decreasing
16. Hintikka, Remes, and Knuuttila 1977, pp. 25-8, 35-9, 50-6. There is in Aristotle a more
promising way to find room for unactualised potentialities qua potentialities. It is based on a
distinction between potentialities of energeiai instantaneously realised through energeiai and
potential outcomes of kinesis realised gradually through a kinesis. See Hintikka, Remes, and
Knuuttila 1977, pp. 59-76.
17. Hintikka, Remes, and Knuuttila 1977, pp. 22-5, 32-3; Pape 1965, pp. 39-43.
18. The modal paradigms of later antiquity seem to be related to those employed by Aristode. The
Master Argument of Diodorus Kronos is designed to establish the statistical interpretation of
modality; see Epictetus 1898, II, 19, 1; for literature concerning the argument see Döring 1972,
pp. 132-5. Philo of Mcgara's idea of possibility seems to have resembled Aristotelian partial
possibility (Boethius 1880,11,234.10-22; 235.6-9; John Philoponus 1905,169.19-21; Alexander
of Aphrodisias 1883,184.6-10). According to Alexander of Aphrodisias 1892,177.7fr., the Stoics
said that the statement 'There will be a sea-battle tomorrow' is possible and not necessary because,
even if it is true, it does not remain true after the battle has taken place (for the Stoic doctrine of
changing truth-values see Knealeand Kneale 1962, pp. 144-6,153-4; Frede 1974, pp. 44-8; cf. the
rather cryptic characterisation of Stoic modal theory in Diogenes Laertius VII, 75). On modalities
in the Stoic system of causal determinism see Sambursky 1959, pp. 76-9. In Neoplatonism,
Plotinus is the principal witness for the metaphysical form of the Principle of Plenitude discussed
in Lovejoy 1936; for Plotinus' modal notions see also Büchner 1968.
19. For an introduction to the work see De Rijk 1962-7 11(1), 408-15; the text is edited in 11(2},
435-638.
(addendo vel diminuendo), as in 'It is necessary that Socrates is running' or 'It

is impossible that Socrates is running.' True and false are not modal
concepts because they do not change the assertoric statement in a similar
way. 2 0 Modal concepts are evidently thought of as qualifying the temporal
domain of a token-reflexive assertoric statement, a view that involves an
acceptance of the statistical interpretation of modal statements as applied to
temporally indefinite sentences. The same treatise states that what is neces-
sarium per se is infallibly true in all times - i.e., in the past, present, and
future - and what is necessarium per accidens cannot be false either now or in
the future, although it might have been false in the past - e.g., 'This man
has studied in Paris.' 21 If the criterion of the necessity of a sentence is the
immutability of its truth-value, it follows that every true sentence is
necessarily true when it is uttered, although not necessarily per se. This is in
effect precisely the Aristotelian principle that 'what is, necessarily is when it
is'. It follows, of course, that a possible sentence must be true at some time,
because otherwise its negation would be immutably true and the sentence
itself impossible.
These ideas are repeated in identical terms in many logical treatises from
the twelfth and thirteenth centuries. 22 Boethius' commentaries on De
interpretatione present them in an early form. He makes a distinction be-
tween necessarium temporale and necessarium simplex. The former refers to
the necessary truth of a temporally definite true sentence. This necessity
disappears, however, when the sentence is taken in a temporally indefinite
sense, sine temporis praesentis descriptione.23 In the fifth book of his
Consolation of Philosophy Boethius tries to use the distinction mentioned
above to show that the purported inference from divine providence to
determinism is fallacious. 24 It is easy to collect other examples in which this
solution is offered to the problems of divine foreknowledge, determin-
ism, and related questions; 25 it, too, is based on the statistical interpretation
of modal notions.
20. De Rijk 1962-7,11(2), 478.27-4792
21. De Rijk 1962-7,11(2), 481.22-482.9.
22. See De Rijk 1962-7, 11(2), 390.18-31; 429.1-10; William of Sherwood 1937, p. 41.13-16.
23. Boethius 1877,1,121.25-122.4; 1880, II, 241.29-242.4.
24. Boethius, Philos, cons., V, 6. There is another example of an analogous change of viewpoint in
Boethius 1880, II, 387.2-4.
25. See Peter Abelard 1859, pp. 142-3, 146; 1970a, 221.3-13; 1919-27, 229.34-6; 274.10-12;
429.18-25;Thomas Aquinas, 1949b, I, q. 2, a. 11, ad4; In quatuor libros Sententiarum, I, d. 38, q. i,a.
5, ad 3. In thirteenth-century physics every effect, when it occurs, is necessary with respect to its
cause, but the same effect can be taken to be contingent ifat another rime the corresponding cause
is impeded; see Maier 1949, pp. 226—34. This interpretation of the Avicennian principle of
causation was first presented by Averroes, who is one of the most explicit adherents of the
Modal logic 347
Modal sentences in the compounded or the divided sense

Medieval Aristotelian solutions to certain kinds of problems containing
modal notions are often based on an analysis according to which modal
sentences can be taken either in the compounded or in the divided sense (in
sensu composito or in sensu diviso). This was also called the distinction
between readings de diclo (or de sensu) and de re.26 It owes its origin to
Aristotle's Sophistici elenchi, 166*22-30. There Aristotle shows how the
meaning of a statement changes according to whether its elements are
understood in a conjoined or a divided meaning. The sentences 'a man
who is not writing can write' and 'a sitting man can walk' are false when
the possibility is understood to qualify the conjunction of two mutually
exclusive predicates with the same subject at the same time. But although it
is true that a man who is not writing cannot write while not writing, and
that while not writing he is necessarily not writing, the same proposition
can also be taken in the divided sense. It is then true that someone-not-
writing can write. In Aristotle's opinion a temporally unqualified modal
sentence which includes mutually exclusive predicates is structurally am-
biguous, because the possibility can refer to a supposed actualisation of
predicates at the same time (in sensu composito) or at different times (in sensu
diviso). The distinction is, in the last analysis, reduced to a temporal
distinction between the simultaneity and non-simultaneity of the actualisa-
tion of two predicates. In fact Aristotle says that in the divided sense the
possibility must be actual at some time, because otherwise the distinction
cannot be made (De caelo 28i b i5—21).
This analysis of modal sentences was widely accepted in the Middle
Ages. While discussing the fallacy secundum diversum tempus Boethius main-
tains that both of two apparently contradictory sentences can be true if
they refer to different points of time. He is thus articulating precisely the
semantical principle by means of which Aristotle tried to overcome the
difficulties of statistical modalities in connection with temporally qualified
sentences.27 Abelard seems to have been the first medieval author to use the
crucial passage from the Sophistici elenchi quoted above in connection with
the strategy of moving from temporally definite to temporally indefinite
sentences.28 After Abelard, this analysis of modal statements by means of
statistical interpretation of modal notions. See, e.g., his Quaesitum II in Pr. an., pp. 78-80; Rescher
1963, pp. 9l-i05;Jalbert 1961, pp. 37-40; Knuuttila 1978, pp. 79-87.
26. Abelard 1919-27, 489.1-14; Peter of Spain 1972, Vll, 68, 71, pp. 122-4.
27. Boethius 1880, II, 133.9-29; 186«, PL 64, 778-9.
28. Abelard 1919-27, 489.1-14.
the compositiofdivisio distinction became a commonplace in logical text

books. 29 The sophism stans potest sedere (or nigrum possibile est esse album)
was solved in the Aristotelian manner by saying that it is false in sensu
composito but true in sensu diviso, because things can be otherwise at another
time. The usual formulation of the sophism is as follows:
W h a t is possible w i l l be true
A black thing can be w h i t e
Thus it will be true that a black thing is white.
According to the solution proposed in very similar terms by William of

Sherwood, Peter of Spain, Lambert of Auxerre, and others, the first
premiss (which entails (P)) is not taken to be problematic. In the second
premiss the compounded and the divided senses are to be distinguished
from each other, and the distinction is typically interpreted in the texts as a
temporal distinction between the possibilities of simultaneous actuality and
actuality at different times. The problematic conclusion is said to rely only
on the composite meaning. 30 Thomas Aquinas believed that by means of
this strategy he could defend indeterminism in the world while accepting
(P), without resorting to the concept of the absolute power of God. 31
Modality and consequences

In their doctrine of consequentiae ut nunc some writers explicitly used the
interrelated ideas of statistically interpreted modal concepts and changing
truth-values.32 The distinction between as-of-now consequences and ab
solute consequences is then characterised so that as-of-now consequences
do not hold at every time but only at a determinate time, and absolute
29. In the oldest extant medieval commentary on Aristotle's Soph. el. the following gloss is made
upon the passage i66a22-3o: 'Secundum conpositionem autem huiusmodi, ut posse sedentem
anbulare, idest possibile est quod sedens simul anbulet, vel possibile est quod sedens alio tempore
anbulet. Si quis didicit litteras, possibile est nunc discere Htteras quos seit. Hec oratio significat
aliquem didicissc litteras et eum nunc posse illas litteras discere, que coniunctim sunt falsa, quia
aliquis non potest didicisse litteras et simul discere; sed divisim potest aliquis didicisse et discere.
Sicut si Deus providit Socratem legere, possibile est Socratem {legere).' See De Rijk 1962-7,1,
210.10-17. I shall supply 'non legere instead of De Rijk's Xlegerey in the last line; for 'didicisse'
and 'discere', like 'sedere' and 'ambulare', are mutually exclusive predicates. For similar analyses, see
De Rijk 1962-7, 1, 311.6—15; 316.1-7; 613.23-8; 11(2), 687.28-688.1; 570.18-29. In the last-
mentioned text the writer does not interpret the possibility as an alternative state of affairs,
because he believes that every true statement about the present is necessarily true: 'omne dictum
de praesenti verum et aflirmativum est necessarium'. 11(2), 481.32-3.
30. William ofSherwood 1937, 90.11-24; Peter of Spain 1972, VII, 68, 70-1; Lambert of Auxerre
1971.Ρ· 15»·
31. See especially De veritate I, q. 2, a. 12, ad 4.
32. William of Sherwood 1941, pp. 79-80; Walter Burley 1955, p. 61; William Ockham (?) 1964, p.
77-
Modal logic 349
consequences hold at every time. When the validity of the absolute ma-
terial consequence is based on the impossibility of the conjunction of the
antecedent and the denial of the consequent, it seems that the modalities are
to be interpreted in terms of time, so that a proposition is necessary if and
only if it is always true, impossible if and only if it is never true, and possible
if and only if it is true at some time. 33 The necessity of an 'as-of-now'
consequence is then a kind of temporal necessity - the truth-value cannot
change at the time when a contingent condition is actual.
Medieval acceptance of the Aristotelian conception of potentiality

The Aristotelian paradigm of potentiality as a power which strives to
manifest itself was unqualifiedly accepted in Aristotelian scholasticism as a
characterisation of natural reality.34 Although according to this model
there can be potentialities which are unrealised for some time, as partial
ingredients of genuine possibilities they must be actualised at some time. So
Moses Maimonides formulates the following principle, which he describes
as Aristotelian: 'It is possible for whatever is in potentia and in whose essence
there is a certain possibility, not to exist in actu at a certain time.' According
to Maimonides, it follows from this that if there were anything potential in
the essence of a perpetually existing being, that being would at some time
become nonexistent. This is the basis for a proof of God's existence: if there
were only beings which have a possibility of not being, all things would
have ceased to exist and there would be no world. 35 The same idea, based
on (P), is used in Thomas Aquinas' Third Way, and in other contexts,
too.36
Medieval attitudes towards the principle of plenitude

Modal paradigms containing (P) were not accepted by most schoolmen
without qualification because the principle was thought to restrict God's
freedom and power. In the twelfth century it was usual to make a distinc-
tion between natural possibilities and God's possibilities. By this distinction
the schoolmen wanted to show that possibilities according to the superior
cause (God) may be impossibilities according to inferior (natural) causes.
The criterion of natural possibility and impossibility was the accustomed
33. Adams 1973, pp. 13-14; see also Moody 1953, pp. 73 - 4-
34. See, for example, Thomas Aquinas, In De caelo I, cap. 12, led. 26,6. In John ofJandun's question
Utrum omne generabile de necessitate generabilur (MS Firenze BN, Conv. Soppr. I III 6, ff.
109™-! iovb) John disproves fifteen different counterexamples to the Principle of Plenitude.
35. Moses Maimonides 1974, introduction to the second part, premiss 23, and Ch. 1, pp. 238-9, 249.
36. Thomas Aquinas, ST, I, q. 2, a. 3c; sec also In IX Metaph., led. 9; n. 1870. ST, I, q. 48, a. 2C.
3 50 The high middle ages: propositions and modalities
course of nature (consuetudo).31 God's possibilities were sometimes taken to

be an exception to this criterion, such that (P) does not hold of them
although it holds with respect to natural possibilities.
Thomas Aquinas put the theological denial of (P) on a firmer theoretical
basis. Accepting (P) as far as natural possibilities are concerned, he denied it
for God's possibilities. But, following Aristotelian epistemology, Aquinas
thought that, although we can prove that there are unrealised Divine
possibilities, all knowledge of what they are is beyond human capacity. 3 8
Divine possibilities are characterised simply by the absence of contradic
tion, a characterisation that Thomas also gives to Aristotle's possibilities ou
kata dynamin.39 In spite of its epistemic limitations, this theory was a big
step towards a logical interpretation of modality in which (P) is explicitly
denied.
Problems in the interpretation of Aristotelian modal syllogistic

In Book I, Chapters 8-22 of his Prior Analytics Aristotle presents his modal
syllogistic in which the following combinations of modal premisses are
considered: D " D , D ~ A , A - Q <>-<>, 0 ~ A , A-<>, 0 ~ D , D ~ 0
( ' • ' stands for a necessary premiss, ' O ' for a possible premiss, and Ά ' for
an assertoric premiss). It has offered frustrating difficulties to commentators
beginning with Aristotle's immediate Peripatetic followers. After Albert
Becker's dissertation Die Aristotelische Theorie der Möglichkeitsschlüsse
(1933) many attempts have been made to reconstruct it as a system in terms
of modern symbolic logic. 4 0 The results have been scanty, and the reason
for this lack of success is almost certainly the fact that Aristotle's modal
syllogistic is put together from elements which contain different and
partially independent insights into the logic of modal notions. 41 It is
therefore not surprising that the first medieval comments on Aristotle's
modal syllogistic are often confused.
Aristotle's choice of the two-edged possibility (incompatible with both
impossibility and necessity) as the principal notion of possibility in his
modal syllogistic suggests that he understood it to be a theory about the
structure of the actual world - it discusses the logic of necessary and
37. See for example Alan of Lille, Regulae de sacra theologia, 57-8; PL 210, 648; Simon of Tournai,
Expositio super Simbolum, ed. prima, 36-43; ed. secunda, 48-54.
3 8. Thomas Aquinas, ST, I, q. 12, a. 8, ad4; De unione verbi incarnait, q. un., a. le; cf. In Post. an. II, 6, n.
461.
39. ST, I, q. 25, a. 3c; Pape 1966, pp. 45-9.
40. See, e.g., McCall 1963; Lukasiewicz 1957, Chs. 6^-8; Rescher 1963, pp. 152-77.
41. Hintikka 1973. PP· 135-46.
Modal logic 351
accidental properties.42 This assumption may solve a general difficulty in

the interpretation of Aristotle's theory. Since Becker it has often been
pointed out that the premisses of Aristotle's modal syllogisms are usually to
be read de re, but the Aristotelian rules of conversion essential to his theory
in many cases seem to imply that modal premisses are to be interpreted de
dicto.*3 Now, according to Aristotle, natural kinds are eternal. If necessity is
interpreted as expressing a genus-species relationship between eternal and
hence necessary kinds, then the rules of conversion are similar no matter
whether the statements are read de dicto or de re. In the conversion of the
contingent premisses interpreted de re the statistical interpretation of mo
dality is accepted, but this was natural for Aristotle, as we have seen. 44
Since the premisses of modal syllogisms are read de re by the author of
Dialectica Monacensis, Averroes, and Albert the Great, 45 but no separate
treatments are offered for the conversion of modal statements interpreted
de dicto or de re,46 it seems that philosophical assumptions similar to those
just mentioned were held by them. The acceptance of the statistical inter
pretation of modality was quite natural for Averroes and the author of
Dialectica Monacensis; the same holds good of Albert the Great as well. He
proves, e.g., the rule 'if q follows from p, then possibly-q follows from
possibly-p' by stating that what is possible is at some time actual. If q were
impossible it would follow that ρ is at some time true without q. 47
The problems caused by the philosophical assumptions may be il
lustrated by the difficulties in the mixed apodictic and assertoric syllogisms.
According to the statistical interpretation of modality, what is always true
is necessarily true. In Pr. An. I, 15, 34b7~i8 Aristotle seems to say that the
premisses in syllogisms must be taken without any limitations with respect
to time. What happens then to the distinction between apodictic and
assertoric premisses? On the other hand, Aristotle says in Pr. An. 30b3i—40
that if the conclusion is necessary in a valid mixed syllogism, it must be
necessary haplös and not only teuton onion, i.e., when the assertoric premiss
is true. It thus seems that according to Aristotle the truth value of an
assertoric premiss can be changed and the conclusion may nevertheless be
42. Ross 1949, pp. 43, 46.
43. See Becker 1933, pp. 19, 39; Bochenski 1956, pp. 98-101; Hintikka 1973, p. 139.
44. In Prior Analytics I, 39'3>-5 it is explicitly stated that if some Χ is possibly Y, then some Y is
possibly X. The rule is often applied in the succeeding chapters; cf. also Becker 1933, pp. 60, 64.
45. See De Rijk 1962-7 11(2), 480.10-16; Averroes 1562-74b, f.84; Albert the Great 1890-98, tract.
IV, cap. 2, pp. 540-3·
46. De Rijk 1962 7,11(2), 501.21 -6; Averroes 1562-748, f.4-5; Albert the Great 1890~9a, tract. I, cap.
11—IS. pp. 474-8S
47. Albert the Great 1890-98, tract. IV, cap. 6, p. 550.
necessary, because in the preceding chapter, he insists that the predicate of

an assertoric premiss is such that it can fail to belong to the subject
(30*27-8). An inference of the form
A necessarily applies to all Β

Β applies to all C
A necessarily applies to all C
then means that because every Β has A as its necessary property, this
property is transferred to every C. Even if Cs cease to be B, they still are A
because A is a necessary property which they cannot lose. When the
corresponding valid modes in mixed necessary and contingent syllogisms
are such that the conclusion is possible in the sense of possibility proper (Pr.
An. 35 b 23~36), this can perhaps be understood so that in contingent uni
versal and particular premisses the predicate may actually belong only to
some of those beings to which the subject term refers.
Medieval solutions of the difficulties in mixed syllogisms

Albert the Great takes the first way out of the difficulty. He says that the
assertoric premiss in valid syllogisms must be de inesse simpliciter and not de
inesse ut nunc. The former type of predication is substantial, and as such it is
necessary secundum rem.*6 When he tries to explain how the necessity of one
premiss determines the second premiss to be de inesse simpliciter and not
only de inesse ut nunc, he seeks arguments from the categorical structure of
the real world. If, e.g., in the first figure the first premiss is necessary, then,
in order to save the structure of the figure the predicate term of the second
premiss must be either a proprium of the subject term or a term which is
superior to the subject in the same category, as for example colour to white
things. 4 9 The difficulties of this approach can be seen in the fact that Albert
defines de inesse simpliciter differently while discussing mixed contingent
50
and assertoric syllogisms.
Averroes' solution is based on the statistical interpretation of modality.
The terms of the necessary premiss are necessary (semper ens). In a mixed
apodictic and assertoric syllogism of the first figure the predicate term of
the second premiss must be necessary, too, because a term de inesse is not
necessarily semper ens, and such a term does not follow from a necessary
term. The subject of the second premiss is de inesse. Such a premiss is de inesse
48. Albert the Great 1890-93, tract. Ill, cap. 3, p. 523.

49. Albert the Great 1890-93, tract. Ill, cap. 4, p. 525; tract. IV, cap. 4, p. 545.
50. Albert the Great 1890-93, tract. IV, cap. 10, pp. 558-9; see also Mignucci 1969, pp. 907-9.
Modal logic 353
per se and necessarium per accidens if the predicate always belongs to the
subject when it is actual, as for example 'Everybody who writes is a man.' If
the necessary premiss is 'Every man is an animal' the conclusion 'Every
body who is writing is necessarily an animal' is de inesse per se and necessary
51
per accidens.
Duns Scotus' theory of modality

These examples illustrate the difficulties of an Aristotelian modal logic in
which modal notions are understood as classifying what happens in the
actual world at different moments of time. A less troublesome treatment of
modal syllogistic was offered in the late thirteenth and early fourteenth
centuries in accordance with the new understanding of modality in which
the apparent difficulties of the statistical model were realised. We have seen
signs of a new kind of modal theory in the ideas of divine possibilities and,
perhaps, in the ideas of the nominales.52 It seems that Duns Scotus had a
dominant role in creating the principles of modal thought which are no
longer bound by the limitations of the statistical interpretation of
modality.
In the fourth chapter of his Tractatus de primo principio Scotus criticises
Aristotelian cosmology, in which the contingency of causally determined
events is saved by the statistical interpretation of modality. He does not
accept the Aristotelian line of thought as a proof of contingency, because it
implies that at any given moment of time everything happens by necessity.
For an indeterministic theory it is necessary that something (something
that is or takes place) could have been or could have happened otherwise at
the same moment of time. The crucial sentence in Scotus' argument runs as
follows: Ί do not call contingent that which is not necessary or not always,
but that the opposite of which could have happened at the very same time it
actually did.' " The notion of contingency is thus understood to involve a
consideration of several alternative states of affairs with respect to the same
time.
It is instructive to see how Scotus applies this new way of understanding
modality in his discussion of the freedom of the will. For Scotus, the will is
not free in the sense that it could at the same time will a thing and its
opposite. But the will is free with respect to the act of willing at any given
51. Averroes 1562-74K ff. 84 ff.

52. For the ideas of this twelfth-century group see Knuuttila 1981, pp. 195-8.
53. See also Ord. I, d. 2, pars i,q. 1-2, n. 79-88. (All references to Duns Scotus'Commentaries on the
Sentences are to John Duns Scotus 1950-.)
time.54 He refers to earlier attempts to explain the freedom of the will by

means of the statistical interpretation of possibility - i . e . , by asserting that
the will can successively intend opposite objects. 5 5 This kind of con
tingency is not a sufficient basis for the freedom of will, however. It says
nothing about the question whether the will while willing something
could have not willed it then. That is w h y a satisfactory account of the
freedom of the will presupposes a different concept of possibility. Scotus
calls it a logical potency (potentia logica) or logical possibility (possibilitas
logica). It characterises what is not contradictory regardless of whether or
not it ever is actual. 5 6 Its function for Scotus in this connection is to show
that contradictory sentences can be possible with respect to the same time.
When Scotus says that there are real potencies which correspond to logical
possibilities, 57 he means that realisation in the actual world is no longer the
criterion of real possibility. The distinction between logical and real possi
bilities takes account of the fact that not all logical possibilities are real
alternatives of the actual world. In Scotus' opinion the general structure of
nature puts certain limits on real possibilities. But the possibility of willing
otherwise is a real alternative because it does not demand that the structure
of the actual world should be different. 58 The sentence 'He who is willing
something at instant A can be not willing it at instant A' is then analysed by
Scotus in accordance with the traditional distinction secundum compo
sitionem and secundum divisionem: on the first interpretation it is false and on
the second interpretation it is true. But it is historically important that
Scotus goes on to point out that on the latter interpretation it is true not
because the mutually exclusive predicates refer to different moments of
time. Instead, the sentence must be interpreted as a conjunction of two
sentences, in one of which an act of willing is said to occur, and in the other
59
an opposite act of willing is said to be possible.
The starting-point of Duns Scotus' modal theory is the concept of
logical possibility. Every sentence which does not contain a contradiction
60
is logically possible. Each possibility can be thought of as being ac-
54. Lect. I, d. 39, q. 1-5, n. 45, 47.

55. Ibid. n. 48.
56. Ibid. n. 49-50,
57. Ibid.
58. Ibid. n. 50-4; see also Balic 1931, pp. 193-99, Roberts 1973 pp. iff. For examples of different
logical possibilities see Sharp 1930, pp. 362-3. For real possibilities and the structure of the world
see Ord. I, d. 3, pars 1, q. 4, n. 235-7, d. 8, pars 2, q.u., n. 306..
59. Lect. I, d. 39, q. 1-5, n. si-2.
60. Orii. I,d.2, pars 2,q. 1-4, n. 262, Ord. I,d.7,q. i,n. 27, Ord. I,d. 20,q. u.,n. u - 1 2 , Lect. I, d. 5,
pars 2. q.u., η. 118. The concept possibile logicum does not occur in the literature before Scotus; see
Faust 1931-2, II, pp. 247-59. Deku 1956, p. 15.
Modal logic 355
tualised. This does not mean, however, that every possibility could be
actualised in the world, because not all possibilities are compossible.61
In Duns Scotus' modal theory modal notions are no longer understood
as classifying the states of the actual world at different times. The domain of
possibility is accepted as an a priori area of conceptual consistency which is
then divided into different classes of compossible states of affairs, of which
the actual world is one. Thus it was not Leibniz who invented the idea of
possible worlds; the idea is present in Duns Scotus' modal theory, and this
new view of modal notions constitutes the general basis of fourteenth-
century modal logic.
Buridan's theory of modal consequences

As an example of the new modal logic I present the main lines of the theory
of modal consequences in John Buridan. He first discusses the modal
statements in sensu diviso, claiming that the domain of the subject term of
both necessary and possible sentences must be ampliated so that in addition
to the actual objects it contains those that are possible.62 First the relation
ships between modal statements and de inesse statements are regulated. In
the following formulations Ί - ' indicates validity and 'rf' invalidity, ' • '
means 'necessarily' and Ό ' 'possibly', while '(x)' and '3x' mean 'all' and
'some', respectively. The domain over which the variable 'x' ranges in
modal sentences is that of possible objects; they may be actual or merely
possible.
The rules for necessary and de inesse sentences are as follows:
(i) rf(x)(OAx 3 fJBx) => (x)(Ax => Bx)
(2) \+ 3x(0 Ax & QBx) z, 3x(Ax & Bx)
(3) rf 3x(0 Ax & D ~ Bx) => 3x(Ax & ~ Bx)
(4) r-(x)(OAx => D ~ B x ) => (x)(Ax => - B x )
63
None of the converse consequences are accepted. (i) and (2) are not valid
because the domain of the relevant individuals may be empty. In medieval
logic it is usually supposed that in such a case affirmative sentences are not
true although negative sentences can be true even if the subject is an empty
64
term. Therefore (4) is valid even if the extension of A in the actual world
is empty. By contrast, (3) is not valid because it is compatible with the
61. Ord. I,d. 43, q.u., η. 16; Leer. I, d. 39, q. l-5,n.72.Theideaofcompossibilityin the model entails
that possibilities are classified into equivalence classes on the basis of the relations of «impossibility
(Ord. I, d. 43, q.u., n. 14,16). One of the classes into which logical possibilities are partitioned is the
actual world (Lect. I, d. 39, q. 1-5, n. 62-3).
62. Buridan 1976, lib. II, cap. 4, pp. 58-60.
63. Ibid. lib. II, cap. 6, concl. 3, pp. 64.106-65.138.
64. See, e.g., Ockham 1974a, I, cap. ηζ, p. zic>.iz4~6.
antecedent that every actual A should be B. The corresponding rules for

possible and de inesse sentences are:
(5) ^ (x)(OAx => O B x ) z> (x)(Ax => Bx)
(6) hf (x)(OAx 3 0 ~ B x ) ^ (x)(Ax z> ~ B x )
(7) \+ 3 x ( O A x & OBx) => 3x(Ax & Bx)
(8) rf 3 x ( O A x & Ο ~ B x ) z> 3x(Ax & ~ B x )
(9) h (x)(Ax r> Bx) -3 3 x ( O A x & O B x )
(10) h 3x(Ax & Bx) => 3 x ( 0 Ax & O B x )
(11) rf (x)(Ax ^ ~ B x ) => (x)(OAx 3 0 ~ B x )
(12) rf 3x(Ax& ~ B x ) => 3 x ( O A x & 0 ~ B x ) 6 5
Of these, (11) and (12) are not accepted because the subject of a true
negative de inesse statement can be empty - it can even be impossible, and
then the subject of a possible predicate is not one which actually is or can be.
The conversion rules for modal sentences in sensu diviso are such that
from an affirmative statement de possibili a particular affirmative statement
depossibili follows by the conversion of terms. But a negative sentence does
not imply any sentence de possibili. O f necessary statements only the
universal negative statement is convertible. Every statement de contingenti
ad utrumlibet can be converted into one of an opposite quality, but no
conversion of terms yields a statement de contingenti ad utrumlibet.66
After the rules for divided modal statements Buridan discusses those for
modal statements in sensu composito. If the dictum of the proposition is taken
to be the subject and the modal term the predicate, then there is a
conversion of the terms, except that a universal affirmative is converted
only into a particular affirmative, i.e., 'Every proposition " B is A " is
67
possible' is converted into 'Some possible proposition is " B is A " ' . As far
as the terms of the dictum are concerned, they are converted in the same way
as statements de inesse. This is based, according to Buridan, on the two rules
of inference which are basic in the modal logic of statements in sensu
composito:
p ^q
(13)
O p =3 O q
and
(Η)
Dp = Dq 6 8
65. Buridan 1976, lib, II, cap. 6, concl. 4, pp. 65.140—66.170.
66. Ibid. cap. 6, concl. 5-8, pp. 66.172-69.254.
67. Ibid. cap. 7, concl. 10, pp. 71.68-72.88.
68. Ibid. cap. 7, concl. 12, pp. 72.98-73.129.
Modal logic 357
The rules governing the relationships between modal statements in sensu

composito and in sensu diviso are similar to those between de inesse and in sensu
diviso. Thus the following statements are accepted:
(15) h 0(x)(Ax => Bx) => 3x(OAx & OBx)
(16) h 03x(Ax & Bx) =. 3x(OAx & OBx)
(17) r- (x)(OAx ^ D ~ B x ) z> D(x)(Ax => ~ B x ) 6 9
These rules are then applied by Buridan to modal syllogistic. If the
conjunction of premisses in sensu composito is the antecedent and the
conclusion is the consequent, then by rules (13) and (14) it follows that in
every valid assertoric mode a necessary conclusion follows if the premisses
together are necessary, and the same holds of possible premisses. Because
the compossibility of two possible premisses is not assured, however, only
conclusions from necessary premisses are acceptable. This is based on the
following rules:
(18) h (Gp & Dq) 3 D(P & q)
(19) rf ( O p & O q ) = > 0 ( p & q ) 7 0
While speaking of modal syllogisms in which the premisses are in sensu
diviso, Buridan separately discusses the modes in which the domain of
subject terms of modal premisses is restricted to actual beings and the
modes in which it is ampliated to include possible beings as well. 71
I have discussed Buridan as an example of the new fourteenth-century
approach to modal logic. Similar treatments can be found, e.g., in Ockham
and the Pseudo-Scotus.72
The formulations introduced above show that something like quantifi
cation into modal contexts was usual in fourteenth-century modal logic. It
is interesting that certain recent problems connected with this type of
73
quantification were discussed in the fourteenth century. This side of
medieval philosophical logic is, ut nunc, largely unexplored; the same holds
of fourteenth-century epistemic logic and other branches of modal logic as
well.74
69. Ibid. cap. 7, concl. 17-18, pp. 76.214-78.256.

70. Ibid. lib. IV, cap. 1, rowri. 1, p. 113.69-90. Buridan does not discuss mixed modes here.
71. Ibid. cap. 2, pp. 114-28.
72. Ockham 1974a, II, cap. 24-9, pp. 327-45; III—1. cap. 20-64, ΡΡ- 4H _ 97; III—3, cap. 10-16,
pp. 631-49; Pseudo-Scotus 163g, qq. 25-33, PP- 3°9- 2 J; McDermott 1972, pp. 273—306.
73. For the problem of reaching the reference of a term in different states of affairs see Ockham 1974a,
I, cap. 72, pp. 216.58-it7.80; Loux 1974, pp. 38-42. In Chapter IV of his Sophismata Buridan
discusses extensively the principles of substitutivity of identity and existential generalisation in
cognitive contexts. See Moody 1975, pp. 353-70.
74. See, e.g., Pseudo-Scotus 1639, q. 36, pp. 328-30; Buridan 1976. /16. IV, cap. 1, concl. 3,
p. 114.107—12; Ockham 1974a, III—1, cap. 30, pp. 433-9·
ι8
FUTURE CONTINGENTS
Three problems involving future contingents

There is more than one problem of future contingents. There is first the
problem raised by Aristotle 1 - that of reconciling the principle of bivalence
(the principle that for any sentence Ρ either Ρ is true or not-P is true) with
the view that some claims about the future are contingent, are such that
neither the claim nor its denial is necessarily true. 2 Medieval discussions of
this problem often rely on our intuitions that the past and the present are
'fixed' in some way in which the future is not, and so these discussions often
illuminate medieval views on tense and modality.
A second problem has to do with the possibility of foreknowledge. Can
one hold both that some future event is contingent and that it is
foreknown?
A third problem is specifically theological. Can complete knowledge of
the future by an immutable, infallible, impassible God be reconciled with
the contingency of some aspects of the future?
These are distinct problems. Theories which solve the problem of
contingent truth may fail to account for foreknowledge, and theories
which account for both future contingent truth and foreknowledge may
yet fail to explain how contingent future events, e.g. sins, can be known by
a knower who cannot be causally acted upon.
Yet all three problems are variations on a single theme. We are inclined
to think that there is an objective difference between the past and the
future. What has happened happened as a result of earlier events including
perhaps actions of our o w n but now that it has happened it does not depend
ι. Aristotle discusses the contingency of the future in at least two places: De interpretatione, Ch. 9, and
Metaphysics, Book VI, Ch. 3. This chapter will focus on the medieval response to the first
discussion. The discussion in Metaphysics VI was typically understood by medieval writers to be of
'causal' rather than 'logical' determinism.
2. It is useful to distinguish what I here call the principle of bivalence from another principle, namely
that for any sentence Ρ, 'P or not-P' is true. The two are equivalent within the usual logical
frameworks but can be distinguished. It seems to me, for example, that Ockham's reconstruction
of Aristotle's account in De interpretatione, Ch. 9,-separates them. For a discussion of the logical
points involved, see Van Fraassen 1971, Ch. 5.3.
Future contingents 359
on anything agents will do. An adult may regret his misspent youth but
cannot prevent it. On the other hand what will happen is dependent on
what is happening and is, we think, in part dependent on what we and
other agents will do. We think that some of what we will do we could
avoid. The problems of future contingents arise because there seem to be
various principles which connect every statement about the future with a
corresponding statement about the past in such a way that it is impossible
for one of the statements to be true and the other false. If any such principle
is accepted, our intuitions about the objective difference between past and
future may have to be revised.
Future contingents in antiquity and the patristic period
The later Middle Ages inherited the various problems of future con-
tingents from antiquity and from the Church Fathers. The earliest extant
discussion, in Chapter 9 of Aristotle's De interpretatione, was available
throughout the medieval period, and this work, together with Boethius'
two commentaries on it, formed part of the core of medieval curricula. The
special problems raised by prophecy and divine omniscience were known
through a number of sources, notably Augustine's comments on Cicero's
views in De civitate Dei, Book V, Ch. 9, and Boethius' remarkable discus-
sion in De consolatione philosophiae Book V. Although apparently little
utilised, neo-Platonic conceptions of fate were available in Calcidius'
commentary on Plato's Timaeus.
Anselm on future contingents, foreknowledge, and freedom
We can obtain a good picture of the state of the problem at the beginning
of the later Middle Ages by considering the position defended by Anselm
of Canterbury. Anselm discusses future contingents at length in his De
concordia praescientiae et praedestinationis etgratiae Dei cum libero arbitrio, Q.i,
and provides short discussions of particular aspects of the matter in Cur
Deus homo II, c. 17, and in De casu diaboli XXI. His discussion centres on the
compatibility of contingency and divine foreknowledge.
Anselm holds that God exists in eternity, and so events which occur at
different times are 'simultaneous' for him - the doctrine of God's eternality
on which Boethius had founded his rejection of the concept of divine
/oreknowledge.3 Attempts to clarify the notion of eternity and show its
3. 'Habet enim aetemitas suum simul, in quo sunt omnia quae simul sunt loco vel tempore, ct quae
sunt diversis in locis vel tcmporibus.' 'For eternity has its own "simultaneity" wherein exist all
things that occur at the same time and place and that occur at different times and places.' Anselm
I946d, p. 254.13-14. Cf. Boethius, Df consolatione philosophiae. Book V, Prose 6.
relevance to the problem constitute one strand of the later medieval

development.
The key to Anselm's treatment is his distinction between 'antecedent'
and 'subsequent' necessity. The nature of 'antecedent necessity' is not
entirely clear. Sometimes, notably in Cur Deus homo II, 17, Anselm speaks
as if necessity is the basic notion and as if some situation is antecedently
necessary just in case some force compels or constrains it to obtain. At other
times, particularly in the drafts which have been published as the Incomplete
Work (or Philosophical Fragments) and in De libero arbitrii, it seems that the
basic notion is power or ability, and so it seems that some situation is
necessary just in case there is no power which can prevent or undo it. These
notions are not trivially equivalent. For example, if there were nothing at
all, that there be nothing at all would be necessary on the second concep
tion but not on the first. It is clear that some notion of what can and cannot
be done underlies Anselm's picture of antecedent necessity. T o illustrate this
type of necessity he uses an example from Boethius, claiming that it is
antecedently necessary that the sun will rise tomorrow and moreover that
the sun will rise tomorrow because it is necessary that it rise (De concordia I,
3). 'Subsequent necessity', on the other hand, is pretty clearly relative
necessity. If A then (subsequently) necessarily Β if and only if it is im
possible that A obtain and Β not obtain (where 'impossible' is the correlate
ofantecedently necessary). It is in this sense of subsequent necessity that what
is past is necessarily past, what is present is necessarily so, and what will be
necessarily will be; 4 and it is in this sense that what God foreknows
necessarily will come to pass. As Anselm insists, this sort of necessity is
innocuous. It is only if we can be persuaded that what will be necessarily
will be in some stronger sense of'necessarily' that our belief in contingency
is endangered. The problem of future contingents arises only if the future is
antecedently necessary.
Anselm holds that God and the angels choose freely, but he also holds
that one can choose freely even where what one chooses is the only
5
available option. This feature of Anselm's account of free choice is of
4. 'Paritcr autem verum est quia fuit et est et erit aliquid non ex necessitate, et quia necesse est fuisse
omne quod fuit, et esse quod est, et futurum esse quod erit.' 'However, it is equally true that there
was and is and will be something which is not by necessity, and that all that was necessarily has
been and what is [necessarily! is and what will be [necessarily] is going to be.' Anselm I946d.
p. 249, 10—12.
5. I take this to be implied by Anselm's argument that the ability to sin is not essential for freedom of
the will. 'Magister. Libertärem arbitrii non puto esse potentiam peccandi et non peccandi. Quippe
si haec eius esset definitio; nee deus nee angeli qui peccare nequcunt liberum haberent arbitrium;
quod nefas est dicere.' Anselm 1946a, p. 207.
considerable importance for his account of future contingents. Because free

choice requires absence of constraint but does not always require more than
one alternative, many of the problems about the relation between pro-
phecy and free action which beset later writers do not arise.
The one context in Anselm's work within which some of those prob-
lems do arise is his consideration of the problem whether one can foreknow
one's own sin. In De casu diaboli, c. 21, Anselm argues that one could not do
so, and at one place in his argument he seems to suggest that one cannot
simultaneously will not to commit a sin and believe that one will commit
it.6 This principle would in particular raise puzzles about Peter's threefold
denial of Christ and Christ's prediction to Peter of this denial. Not surpris-
ingly it plays a role in fourteenth- and fifteenth-century discussions.
The interpretation of Aristotle's view of truth

In the period after Anselm much of the discussion of future contingents
centred around the interpretation of Aristotle's view of truth, the view that
a sentence is true if things are as the sentence says they are. On one
interpretation of this view a sentence is true only if the way things now are
verifies it. If one holds that a sentence is false only if the way things now are
verifies its negation then this interpretation might lead one to abandon the
principle of bivalence.
Abelard on future contingents

The first towering figure in this discussion after Anselm is Peter Abelard.
Both the problem of future contingent truths and the problem of fore-
knowledge arise naturally in his discussions of Aristotle's De interpretatione,
Ch. 9, and the problem of foreknowledge arises in many of his theological
works as well. Abelard understands De interpretatione Ch. 9 as a defence of
the claim made in De interpretatione Ch. 6 that every sentence has a con-
tradictory. He considers two arguments against this which he extracts from
De interpretatione, and in the course of replying to them he develops an
unusual account of his own.
The second argument is sketched in his gloss Editio super Aristotelem De
6. 'M. Est et aliud quod mihi satis videtur ostendere nullo modo eum suam ante putavisse futuram
prevaricationem. Nempe aut coactum putavisset aut spontaneam. Sed nee ullatenus erat unde se
aliquando cogi suspicaretur, nee quamdiu voluit in veritate pcrseverare, ullo modo putare potuit
se sola voluntate illam deserturum. Supra namque iam monstratum est eum, quamdiu rectam
voluntatem habuit, in hac ipsa voluntate voluissc perseverare. Quapropter volendo tencre
perseveranter quod tenebat, nullo modo video unde potuisset vel suspicari nulla alia antécédente
causa sola se illud deserturum voluntate.' Anselm 1946c, c. 21, p. 268.
interpretatione.1 It begins with the premiss that some contingent sentence Ρ

is true and proceeds in this way. If Ρ is true then 'P will be true' was true,
and since this is a sentence about the past, ' "P will be true" was true' is
determinately true, or determinate. But then what makes it true, namely
that P, is determinate, and hence Ρ is determinate. Abelard's discussion of
this argument naturally focuses on the sense of 'determinate'. In both his
Dialectica and the commentary on De interpretatione in his Logica Ίη-
gredientibus' Abelard explains 'determinate' in epistemological terms, claim
ing that it applies primarily to states of affairs and secondarily to sentences
which report them. In the Dialectica he explains that a state of affairs is
determinate if it can be known either from the fact of its occurrence (ex
existentia sui) or from its very nature (ex natura rei).6
In the Logica ' Ingredientibus' Abelard claims that future contingent sen
tences are not determinate; though they are true or false, they are in
determinately true or false. But he points out that since (i) ' "Socrates will
eat" is true' entails (2) "Socrates will eat", we must either deny that (1) is
determinate and so admit that not all sentences 'about' the past or present
are determinate, or assert that valid inference does not preserve deter-
minateness.9 His reason for adopting the first alternative seems to be that
7. 'Et amplius probo quod perit casus et utrumlibet; tua sententia est quod si est album nunc in re
verum erat diccrc primo quoniam album est, id est vera erat propositio determinate quae prius
enuntiavit; et si quia vera fuit determinate propositio quae hoc enuntiavit, ideo quia postea ita
evenit in re: quare semper verum fuit dicere determinate quoniam erit quod li bet eorum quae facta
sunt, id est omnes illae propositiones fuerunt verae determinate quarum res evenerunt.' Abelard
1969a, p. 103.
8. 'Determinates autem eos rerum eventus dicimus quicumque vel ex existentia sui cognosci possunt
vel ex natura rei certi sunt. Ex existentia quidem sui cognosci potuerunt tarn presentia quam
preterita. Ipsa namque preterita, quia quandoque presentia fuerunt, [quare] ex presentia sui
cognosci valuerunt. Futura vero necessaria sive etiam naturalia ex proprietate nature determinata
sunt.' Abelard 1970a, p. 211.5fr.
9. 'De quibusdam tamen praesentibus sive praeteritis restât quaestio, utrum définira sint vel in-
definita, cum indefinita per futura indeterminata sciri queant. Ita cum dicitur "Socrates come-
det", haec propositio iam profecto praesentaliter est vera vel falsa et haec propositio quae dicit
'"Socrates comedet' vera est" de praesenti est et ideo iam determinate vera est vel determinate
falsa . . . Sed nee istae de praesenti "Socrates est nomen hominis comesturi in crastino" vel "Homo
comesturus in crastino est Socrates" . . . determinari possunt nisi per futurum. . . . Non itaque
omnes de praesenti vel praeterito propositiones verae vel falsae determinate esse videntur, quando
videlicet veritatis earum vel falsitatis discretio ex futuro pendet Sunt autem qui dicunt, quod
"Socrates comedet" quae fortasse se praesentaliter habet, quippe determinatus est eventus pro-
prietatis praesentaliter ei inhaerentis et ipsam iam determinate veritatem habet. Sed licet de-
terminate vera dicatur propositio quantum ad praesentem et determinatam inhaerentiam veritatis
quantum tamen ad eventum, quern loquitur indeterminatum, indeterminate vera e s t . . . . Aliud
etiam "determinatum" sonare videtur quam "certum" quia determinatus dicitur eventus, qui ex
se cognoscibilis nobis, ut paritas et imparitas astrorum ex ipsa praesentia, quam habent, de se
cognitionem dare potest; certa vero sunt, quae quoquomodo actualiter cognita sunt, ut si quid de
futuro contingenti testimonio angeti ad discretionem <venerit), certum quidem esset mihi non
there are some present-tensed truths which we cannot know and which are
therefore not determinate as defined. In the Dialectica, however, he opts for
the second alternative.10
Abelard on foreknowledge
Abelard's views on foreknowledge are straightforward by comparison. He
thinks that God knows all future contingent truths and that these can be
revealed to us by God. l ' But most of his discussion centres on reconciling
the contingency of parts of the future with the view that God's foreknowl-
edge is infallible. In particular he aims to show that 'It is possible for
something other than what God foreknows to happen' has two senses.
Read as claiming that 'Something other than what God foreknows hap-
pens' could be true it is false, but determinism is not thereby supported.
Read as claiming of some particular event that it will not happen and so is
not foreknown although it could happen, it is true, but divine infallibility is
not thereby threatened.
Peter Lombard
The most influential twelfth-century work on future contingents is re-
corded in Book I, d. 38-40, of the Sentences of Peter Lombard. Lombard
begins Book I, d. 38, c. 1, by asking 'whether the knowledge or fore-
knowledge of God is the cause of things or conversely?' Unlike Augustine
and Anselm who seem to favour the view that God knows the future
because he in some sense makes it, Lombard denies any causal connection
between divine knowledge as such and its objects. He does not explain how
knowledge without causal connection might work, however. He thus sets
the stage for one of the most persistent later medieval controversies.
In distinction 39 Lombard is concerned largely with whether God can
know what he does not know. His problem is thus to reconcile the
contingency of the future with divine infallibility and immutability. His
solution to the problem of divine infallibility is very like Abelard's. He
ex se, sed ex auctoritate angeli, sed determinatum non esset, quia ex se ipso cognoscibilis non
esset.. . Unde etsi talis consequentia recipiatur: "Si 'Socrates comedet' est vera, Socrates com-
edet" et certum sit antecedens, certum est consequens. Sed non fortasse, si determinatum sit
antecedens, et consequens, quia praesens eventus ex se cognoscibilis est, sed non ita futurus
contingens.' Abelard 1919-27, pp. 427ff.
10. Abelard 1970a, p. 212.
11. In Abelard 1970a, p. 212.151T., Abelard supposes that God can reveal future contingents: 'Futura
vero ex se cognosci non valent, sed si aliquam certitudinem per reponsum divinum vel per signa
aliqua de futuris habere contingeret, certa quidem esse possent quocumque modo cognita, sed
non determinata, nisi ex proprio eventu cognoscerentur.' Cf. Abelard 1919-27, pp. 426-9.
distinguishes two senses of'Things cannot be other than as God foreknows

them', a compounded sense in which it is false and a divided sense in which
it is true. The truth of the latter guards divine infallibility and the falsity of
the former preserves the contingency of the future.
But Lombard's discussion of the immutability of divine knowledge
differs markedly from Abelard's. In his Logica 'Ingredientibus' Abelard allows
that God's knowledge can change as the world changes but insists that such
change is not real change in God. 1 2 Lombard's view is that although God's
knowledge is immutable, God can know what he does not know because
he creates freely and hence can create what he does not create (a position
disputed by Abelard), and because some creatures can do what they will
not do. 1 3
By insisting on God's freedom Lombard focuses attention on a sense of
'possible' which is sharply distinguished from mutability. God can know
what he does not know, but his knowledge cannot change. For Lombard,
then, there are claims which can be true but which never are true.
Robert Grosseteste
The distinction between necessity and immutability is taken up by Robert
Grosseteste and used as the basis of one of the most penetrating and
influential medieval approaches to the question. Grosseteste begins his
treatise De libero arbitrio by distinguishing what is simply necessary from
what is merely immutable. 1 4 The simply necessary is what could not be
otherwise no matter how the history of the world had gone. (His example
is ' T w o and three is five.') The immutable is what cannot be one way after
12. 'Similiter cum modo me sedcnte, modo non sedente dicitur scire vel intelligere me sedentem et
non intelligere vel scire me sedentem, nulla est in ipso variatio, sed in me; quocumque modo me
habeam ipsum latere non possum. Cum itaque omnino in se Deus invariabilis subsistât, secundum
variationem creaturum vel varios effectus ipsius quasi distenditur'. Abelard 1919-27, p. 428.
13. Lombard writes: 'Ex hac auctoritate clare ostenditur, scientiam Dei omnino invariabilem esse,
sicut ipsa essentia Dei omnino invariabilis e s t . . . Et tamen conceditur posse scire quod non seit et
posse non scire quod seit, quia posset aliquid esse subiectum eius scientiac, quod non est, et posset
non esse subiectum aliquid quod est, sine permutatione ipsius scientiae.' Peter Lombard 1916, vol.
1, p. 246.
14. 'Dico igitur quod est necessarium duplex: uno modo quod non habet posse aliquo modo ad eius
oppositum vel cum initio vel fine, cuiusmodi est hoc: "Duo et tria esse quinquc". Istud enim posse
non habuit neque ante tempus. neque in tempore ad non esse verum. Et tale est necessarium
simpliciter. Et est aliud necessarium, quod neque secundum piaeteritum, neque secundum
praesens, neque secundum futurum habet posse ad eius oppositum, sine tamen initio fuit posse ad
hoc et fuit posse ad eius oppositum et tale est "Antichristum fore futurum" et omnium eorum
quae sunt de futuro, quod eorum Veritas, cum est, non potest habere non-esse post esse, ut supra
ostensum est. Et tarnen posse ad hoc, ut ab aeterno et sine initio fuerint falsa.' Grossereste 1912a,
Ch.6.
being another, and so the truth of sentences about the future is immutable.
For if 'The Antichrist will come' ever was true then (until the Antichrist
comes) it will be true. 1 5 Grosseteste thus distinguishes three modal
strengths: the simply contingent, - as in 'Socrates is white', which can be
true at one time and false at another - the immutable, and the simply
necessary. Grosseteste uses this distinction to explain how God can know
what he does not know. God's knowledge is immutable but not simply
necessary. Even though it is immutably true that the Antichrist will come,
it is not simply necessary. Were its (simply contingent) contradictory to be
true, God would always have known it.
Both claims about the past and claims about the future are immutable,
and there is no indication that the past any more than the future is simply or
absolutely necessary.16 Yet Grosseteste does want to claim that an un
realised possibility can be realised only in the future.17 He thus must still
explain how God can know what he never has known even though his
knowledge is immutable. The solution is that for God the act of knowing
A is the very same act as that by which he would know not-A were not-A
true.
Grosseteste's distinction between simple necessity and immutability
enables him to provide a sense in which the future is contingent, but it does
not explain how the past is necessary in a way in which the future is not.
There are true past-tensed sentences which are immutably true whose true
future-tensed analogues are not immutably true: 'Adam existed' is im
mutably true but 'The Antichrist will exist' will become false when the
Antichrist dies. But there are also immutable future-tensed sentences
15. One might object that sentences such as 'The Antichrist will come' will cease 10 be true after being
true; once the Antichrist comes it is false to say that he will come. Grosseteste is not unaware o f
this. Indeed, he writes: 'Sicut enim Veritas huius "Deus seit A " vel "Antichristus fuit futurus" vel
"lsias praedixit hoc verum" non potest desinere, sic Veritas vel futuritio huius "Antichristus est
futurus vel erit" non potest desinere »«ι per exhibitionem esse Anlichristi' (emphasis added).
Grosseteste 1912a, Ch. 6, p. 170.23-8.
16. 'Unde illud simpliciter est necessarium, ad cuius aliquo modo non-esse nullo modo est posse. Est
alio modo necessarium, scilicet quod cum est, non est posse ad ipsum habere non-esse post esse,
quod habet. Sic vera de praeterito patenter sunt necessaria. Et hoc necessarium sequitur ad
necessarium prius dictum. Unde necessarium simpliciter dividitur in necessarium supra dictum et
in necessarium quod cum non-posse ad habere non esse post esse habet posse ad sine initio
numquam esse vel fuisse vel fore. Talis est Veritas dictorum de futuro, quia eorum Veritas, cum est,
non potest habere non-esse post esse, ut supra ostensum est, et tamen posse ad hoc, ut sine initio et
ab aeterno fuerint vera; et posse ad hoc ut ab aeterno et sine initio fuerint falsa est posse ad
falsitatem non initiabilem et ad veritatem non initiabilem in his. Sed cum habet veritatem et
falsitatem, non est posse ad oppositum eius quod habeat post id, quod habet.* Grosseteste 1912a,
p. 169.
17. 'Et tamen verum est, quod possibilitas sine actu solum est respectu futuri.' Grosseteste 1912a,
p. 183.22-4.
whose past-tensed analogues are mutable. Thus Grosseteste's solution dis

solves the problem by abandoning any deep difference between the modal
status of the past and of the future. It is with this difference that much of the
succeeding century is concerned.
Thomas Aquinas
Thomas Aquinas discusses the problem of future contingents in each of his
comprehensive works, focusing on whether and how God can have fore
knowledge of contingents. His position sometimes is that God can have
such foreknowledge and that he has it through his knowledge of his own
creative power: 'The knowledge of God is related to all created things just
as the knowledge of a craftsman is related to his p r o d u c t . ' 1 8 This raises a
problem, however. A craftsman knows what he will make because his is
the only free choice involved, but God is not always the only free agent
involved in a contingent future. Either God's creative activity is a sufficient
condition for m y choosing to sin, in which case he can foreknow it but it is
not clear that my choice is both free and efficacious, or God's creative
activity is a necessary but not a sufficient condition, in which case it is not
easy to see how knowledge of it alone constitutes foreknowledge. Aquinas
seems to take the first horn of this dilemma. Like Anselm he denies that
being able not to choose A is a necessary condition for choosing A freely,
and so he can hold that a particular outcome is freely chosen by me even if
God's activity guarantees its coming to pass. 1 9
There is however another strand in Aquinas' thought - the view that
God is able to know the future because all things, past, present, and future,
are eternally present to him. This suggests a picture of the whole of time
laid out before God, as does Aquinas' famous image of God as a man high
up on a mountain watching others move along a road below. 2 0
As interesting and as puzzling as Aquinas' account of God's mode of
knowing the future is his discussion of the problems raised by the fact that it
was true to say that God knew what is yet future, and that the past is in some
18. 'Dicendum quod scientia Dei est causa rerum. Sic enim scientia Dei se habet ad omnes res creatas.
sicut scientia artificis sc habet ad artificialia.' ST", I, Q. 14, art. 8.
19. At S Τ, I, Q. 62, art. 8, Aquinas argues that the good angels have free will though they cannot sin.
Cf. De veritate, Q. 6.
20. The image of the road appears in ST, I, Q. 14, art. 13 ad 3. Following out this strain of Aquinas'
thought one might conclude that he thinks that the past and the future in some sense are. But he
denies this in De veritate Q. I, art. 5, ad 6 and 7: 'Ad sextum et septimum dicendum quod id quod
est futurum, in quantum est futurum non est, et similiter praeteritum, in quantum huiusmodi.
Unde eadem est ratio de veritate praeteriti et futuri, sicut et de veritate non entis, ex qua non
potest concludi aeternitas alicuius veritatis, nisi primae, ut dictum est' (Aquinas 1953a).
sense necessary. 21 The sentence 'If God knew that some contingent thing
was going to be, then it will be' is true, and its antecedent is if true
necessarily true because it is about the past. Hence if the antecedent is true
the consequent is necessarily true. Aquinas agrees that the conditional is
true, and that if the antecedent of a true conditional is necessary so is the
consequent, and that a true sentence about the past is necessarily true. But
he apparently doubts that the antecedent of this conditional really is about
the past. Aquinas, following Boethius, argues that since God sees every-
thing in an eternal present and since 'all that is, when it is, is necessarily', the
antecedent and so the consequent of the conditional under discussion are
necessary but with the relative (and harmless) necessity which attaches to
present truths. 22
fohn Duns Scotus

When considered against the background of Aquinas's position, Scotus'
discussions of God's knowledge signal the clash of two fundamentally
different ways of conceiving the nature of time. The first, which seems to
have been Boethius' and may have been Aquinas', conceives the difference
between past and future as perspectival rather than ontological. Human
beings have relations to past events which differ from those we have to
future events, but these differences are grounded in us rather than in the
world outside us. The second view, the one for which Scotus argues and
the one which seems to be taken more or less for granted in the first quarter
of the fourteenth century, sees the difference between past and future as an
objective difference, one that exists for God as well as for us.
Having rejected the possibility that God knows the contingent future by
inspection, Scotus seeks another account. The one he adopts grounds God's
infallible foreknowledge in the determinations of the divine will. As Scotus
develops the account both in his Lectura and later in the Ordinatio, the truth-
value of contingent claims about the future is not first apprehended by
21. 'Ad primum ergo dicendum quod, licet praeterita non fuisse sit impossibile per accidens, si
consideretur id quod est praeteritum, idest cursus Socratis; tamen si considerctur praeteritum sub
ratione praeteriti, ipsum non fuisse est impossibile non solum per se, sed absolute, contradic-
tioncm implicans.' ST, I, Q. 25, art. 4.
22. 'Et ideo aliter dicendum est quod quando in antécédente ponitur aliquid pertinens ad actum
animae, consequens est accipiendum non secundum quod in se est, sed secundum quod est in
anima. Aliud enim est esse rei in se ipsa, et esse rei in anima ut puta si dicam "si anima intelligit
aliquid, illud est immateriale", intelligendum est quod illud est immarenale secundum quod est in
intellectu, non secundum quod est in se ipso. Et similiter si dicam: "Si Deus scivit aliquid, illud
erit", consequens intelligendum est prout subest divinae scientiae, scilicet prout est in sua
praesentialitate. Et sic necessarium est, sicut et antecedens. Quia omne quod est, dum est, necesse
est esse, ut dicitur in 1 Perih.' ST, 1, Q. 14, art. 13, ad 2.
God. 23 Rather the divine intellect first considers such unasserted complexa
as that the Antichrist will come. The divine will then makes these complexa
true or false by asserting or denying them, and by inspecting the divine will
the divine intellect then sees how things will turn out.24
As an account designed to reconcile foreknowledge with contingency
this seems to leave too little to the created agent. If the divine will chooses to
make that Peter will deny Christ true, then unless this choice is itself de-
pendent on Peter's choice it does not seem that it is Peter who makes the
sentence true.
But whether or not Scotus* own account of divine foreknowledge is
successful, the theory of necessity and possibility with which he buttresses it
is of great importance for the subsequent history of the problem. Unlike
Aquinas, Scotus holds that if any effect is to be contingent the activity of
the first cause on which that effect depends must be contingent. But all of
God's activity is a single eternal act which is God himself. Hence if there is
to be contingency God must be able to will the contrary of whatever he
wills. Scotus insists that this is so and moreover that this is a power of
created free agents as well.2S Of course Scotus does not mean that an agent
can at the same time will that a state of affairs and its contrary obtain.
Rather he is proposing replacing the doctrine of the necessity of the present
with a new criterion: a claim is possible if there is no 'repugnantia ter-
minorum', no semantical inconsistency, in supposing it to obtain.26
Scotus uses his abandonment of the necessity of the present to explain
23. According to Balic 1965, Scotus lectured on the Sentences at least three times, first in Cambridge
and then, in a complex sequence, at Oxford and Paris. The Lectura is a transcription of the notes
Scotus used for his Oxford lectures. The Ordinatio or Opus Oxoniense contains, in Balic's words
(p. 20), 'the text that Scotus himselfeither wrote or dictated from his various lectures and writings
and ordered in such a way as to present the material in the form of a Summa of his entire doctrine'.
The material with which this section is chiefly concerned is found in distinctions 39 and 40 of
Book I of the Lectura and the Ordinatio. The most detailed discussion ofthe problem ofthe truth of
contingent claims about the future to be found in the Scotist corpus is presented in Scotus 1891,
QQ. 7 and 8. This work is usually considered authentic, but there are features of the discussion -
notably the apparent denial of a simultaneous capacity for opposites in Q. 8, 8 — that lead me to
distrust it.
24. 'Unde quando intellectus divinus apprehendit "hoc esse faciendum" ante voluntatis actum,
apprehendit ut neutram, sicut cum apprehendo "astra esse paria"; sed quando per actum
voluntatis producitur in esse, tunc est apprehensum ab intellectu divino ut obiectum verum
secundum alteram partem contradictionis. Oportet igitur assignare causam contingendae in
rebus ex parte voluntatis divinae.' Scotus 1960b, d. 39, q. 1-5, n. 44; p. 493.5 11.
25. Scotus 1960b, d. 39, q. 1-5, n. 53-4; p. 496.15-497.30.
26. 'Sed adhuc illam libertärem voluntatis consequitur alia potentia, quae est logica (cui etiam
correspond« potentia realis). Potentia logica non est aliqua nisi quando extrema sic sunt possi-
bilia quod non sibi invicem repugnant sed uniri possunt.' Scotus 1960b, d. 39, q. 1—5, n. 49;
p. 494-20-5.
how it is that we can accept conditionals such as 'If God foresees X, X will
happen' and accept that God from eternity foresees X without committing
ourselves to the necessity of'X will happen'. God's knowledge is a single
act which never 'passes into the past* and so is as contingent as any present
act. God knows what will happen, but he can know otherwise than he
knows.27
Although Scotus claims a simultaneous capacity for opposites and with it
the contingency of the present he clings tenaciously to the necessity of the
past, but it is his own concept of possibility as semantic consistency which
grounds fourteenth-century attempts to show the past contingent.
Peter Aureoli
At the beginning of the fourteenth century a series of conflicts about the
nature of necessity and the development by Ockham and others of a theory
of what it is for a claim to be about a time threw the problem of future
contingents into sharp focus. The story begins with Peter Aureoli's attempt
to defend the conflation of immutability and necessity.28 Aureoli's view
centres around a principle linking possibility and mutability: 'If a thing is A
and can be not-A, then it can change from being A to being not-A.' From
this principle it follows that if a thing is A and cannot change to being not-
A, then it is necessarily A. In other words, what is immutably in a certain
state is necessarily in that state. Given this identification, Aureoli sets out to
show that if a sentence about the future is true it is immutably and hence
necessarily true. If a contingent proposition about the future such as 'The
Antichrist will come' is now true, then it cannot become false before the
Antichrist does in fact come. For when could it be false? Neither now when
27. 'Ad primum argumentum, quando arguitur quod illud quod transit in praeteritum, est neces-
sarium. concedatur. Et quando arguitur quod "istum praedestinari" transiit in praeteritum,
dicendum quod hoc falsum est; si enim voluntas nostra semper haberet eandem volitionem in
eodem instanti immobili non esset sua volitio praetcrita sed semper in actu. Et sic est de volitione
divina, quae semper eadem est; unde, sicut dictum est, eadem volitione qua vult aliquem
praedestinare, potest velle eundem damnari pro eodem instanti aeternitatis.' Scotus 1960b, d. 40
q. 1, n. 5; pp. 512.27-513.7·
28. Aureoli discusses the relation between immutability and necessity at two places in his Commen-
tatium in primum librum Sententiarum. Thefirstis D. 8, Pars II, art. 1, where he argues: '... definitio
necessitatis et immutabilitatis est eadem ilia namque sunt unum et idem quae communicant in
definitiva ratione nam definitio indicat quid est esse rei, sed definitio necessarii est quod est
impossibile aliter se habere. Incommutabilis autem quod est impossibile mutari; mutari autem est
aliter nunc se habere quam prius. Ergo "immutabile" et "necessarium" sunt idem.' Aureoli
1596-1605, D. 8. pars ll.arr. i;p. 274. Aureoli advances other arguments here and in D. 39, art.
1; p. 893; but all of them seem to presuppose that''aliter impossibile se habere' and 'aliter impossibile
nunc se habere quam prius' are equivalent. It is precisely this which Aureoli's opponents are not
likely to grant.
37° The high middle ages: propositions and modalities
it is true nor later; for if it were false at some time later than now, that could
only be because the Antichrist will not come after all; but in that case 'The
Antichrist will come' is now false, contrary to our hypothesis. 29
T o save the contingency of the future Aureoli proposes abandoning the
view that contingent sentences about the future are now true or false -
apparently the first serious medieval attempt to argue that bivalence is not a
fundamental logical principle. Aureoli did not work out the consequences
of his view for logic, but he did ascribe the view to Aristotle; Ockham and
Gregory of Rimini among others accepted the ascription.
William Ockham
Aureoli's view was never popular, but it had some influence in the fifteenth
century, and it or another very like it was defended both in Oxford and at
the papal court, according to Bradwardine. 3 0 The position which rapidly
became the opinio communis of the early fourteenth century was rather that
worked out and defended by Ockham. It is not entirely novel; indeed, it
contains clear echoes of views expressed by Grosseteste and by Peter
Lombard and of a view attacked by Aquinas, but Ockham's development
of it is especially vigorous and sophisticated - one of his finest philosophical
achievements.
29. 'Primum ergo assumptum patet ex multis, scilicet quod si haec propositio vera est "Antichristus
erit" immutabilitcr et inevitabiliter est vera. Si enim mutari potest, ne fit vera, aut mutabitur in
illo instanti quo vera est, aut in instanti praecedenti, aut in instanti subsequenti usque ad instans
quo res fiet aut in instanti quo res non fiet. Sed manifestum est quod non potest mutari in illo
instanti quo est vera quia pro eodem instanti esset vera et non vera quod impossible est; nee in
instanti praecedenti: turn quia, si in instanti dato, est vera et in omni praecedenti fuit vera, quia si
hodie verum est quod Sortes erit eras, et heri verum fuit quod Sortes esset eras at ita si tunc poterat
mutari in falsitatem, mutaretur in illo eodem tunc quo esset falsa et per consequens simul esset vera
et non vera . . . Nee potest dici quod mutari possit in aliquo instanti subsequenti in tempore
signabili inter instans propositionis et instans quo res fiet quia si in aliquo instanti ut pote eras, falsa
sit haec propositio "Sortes erit", impossibile est quod fuit heri vera et hodie falsa, hoc erit propter
aliquam mutationem factam in re. Nulla autem mutatio facta est quia nondum res e s t . . . nee
potest dici quod in illo instanti in quo res fiet vel non fiet mutabitur ilia Veritas turn quia Veritas ilia
transit in praeteritum . . . quod autem transit in praeteritum immutabile e s t . . . ergo cum non
inveniatur instans in quo possit mutari propositio de futuro a veritate in falsitatem, necessario
immutabiliter vera, si aliquo modo ponatur vera.' Aureoli 1596-1605, D . 38; pp. 883-4.
30. O p i n i o sexta fingit quod aliquid est futurum ad utrumlibet, vel non futurum in sensu composito,
non autem in sensu diviso, quam sic astruere moliuntur. Nulla propositio simplex de futuro in
materia contingente aequaliter est vera vel falsa. Philosophe I Peri Hermenias ultimo attestante,
quare nee aliquid est futurum ad utrumlibet vel non futurum divisim. Hanc autem opinionem
audivi in Curia Romana a quodam famoso Philosophe Tolosano in quadam disputatione
solemne de contingentia futurorum, secundum earn totaliter publice respondente quam et
Oxoniae similiter audicbam.' Bradwardine 1618, Book III, Ch. 17, p. 6 9 2 A - B . In Baudry 1950
Peter of Candia and Pierre de Nugent are listed as supporters o f Aureoli's view, and, as Baudry
shows, it is Aureoli's position which was defended at Louvain in the fifteenth century by Petrus de
Rivo.
Unlike Aureoli, Ockham is committed to bivalence and, more clearly

than most of the major figures in the thirteenth century, to the view that
the past is absolutely necessary. In fact, where A is a real quality he agrees
with Aureoli that if something is immutably A it is necessarily A.
Moreover unlike Scotus, Ockham accepts the doctrine of the necessity of
the present. 31 But Ockham does not think that 'true' and 'false' signify real
qualities inhering in the propositions of which those terms are predi-
cated. 32 Thus while "The Antichrist will come' is true, its being true is not a
fact different from the fact that the Antichrist will come. So while ' " T h e
Antichrist will come" is true' is present-tensed and seems to be about the
present, it is, Ockham thinks, really about the future; and so it is just as
contingent as 'The Antichrist will come.' He distinguishes past- and
present-tensed sentences which are merely verbally (vocaliter) about the
past or present from those which are really (secundum rem) about the past or
present. The latter are if true necessarily true; the former may be true and
yet contingently true. A true sentence verbally about the past or present
and yet really about the future will be contingently true if it is equivalent to
or depends on a sentence which is uncontroversially about the future and is
contingent.
For the Ockhamist programme to work, every present- or past-tensed
sentence which is entailed by a future-tensed contingent sentence must be
only verbally about the past or present. T o make the programme credible
Ockham must show us how to treat a variety of such past- and present-
tensed sentences. This is relatively easily done for claims about the truth or
31. Indeed Ockham argues against Scotus that the necessity of the past entails the necessity of the
present. 'Omnis propositio mere de praesenti, si sit vera, habet aliquam de praeterito necessariam;
sed haec "voluntas vult hoc in a instanti" est vera per positum et est mere dc praesenti; igitur haec
erit semper postea necessaria "voluntas voluit hoc in a instanti"; igitur post a instans ista non
potest esse vera "voluntas non voluit hoc in a instanti". Confirmatur: si post a haec semper fuit
necessaria "voluntas voluit hoc pro a instanti", igitur post a sua opposita semper fuit impossibilis;
et ultra: igitur post a semper fuit et erit verum dicere quod haec propositio non potuit esse vera in a
"voluntas non vult hoc oppositum," quia tunc sua opposita fuit vera, haec scilicet "voluntas
voluit hoc pro a instanti". Responsio igitur consistit in hoc, quia si voluntas vult hoc in a, post a erit
semper haec necessaria "voluntas voluit hoc in a", et tunc si sua potentia non manifesta posset
reduci ad actum in a instanti, vel contradictoria erunt simul vera post a instans, vel post a ilia
propositio quae est necessaria de praeterito, quia habuit aliquam mere de praesenti veram, erit
falsa, quia sua opposita erit vera.' Ockham 1978, Depraedestinatione, Q. Ill; QTII, p. 534.36-52.
It seems that Ockham's argument requires a principle equivalent to 'If it was the case that
possibly P, then possibly it was the case that P.' But that is what is in dispute.
32. 'Sed si Veritas propositionis sit aliqua parva res distincta a propositione sive sit absoluta sive
respectiva ille res essent subjective inhérentes proposition! successive et per consequens ilia
propositio que aliquando est vera aliquando falsa realiter in se susciperet contraria quod est
manifeste contra Philosophi determinationem. Praeterea si sic sequerentur quaedam absurda."
Ockham 1491, Quodlibet V, Q. 24.
falsity of contingent sentences and for attributions of predestination and

reprobation, but how is the account to be extended to sentences such as
'God knew that the Antichrist will come' and 'Christ prophesied that Peter
would deny him'? Ockham considers such cases in Quodlibet IV, Q. 4;
Oxford theologians quickly took up the task of refining and extending his
account.
Both in Quodlibet IV, Q. 4, and in his De praedestinatione33 Ockham
argues that sentences of the form 'X knew (or knows) that P' where Ρ is a
future contingent sentence depend on P, because 'X knew (or knows) that
P' entails 'It is true that P' and this in turn entails ' P ' . Hence if ' P ' is
contingent, so is the knowledge-claim. This seems plausible where X is a
fallible creature but less plausible where X is God. God is infallible and
omniscient, and if God knows that Ρ only if he is in what we might
metaphorically call a certain belief-state, (believing that P) then Ρ is true
just in case God is in that state. But whether God is in such a state is not a
contingent matter; hence Ρ is not contingent either.
Ockham solves this problem by denying that for God to know some
thing is for God to be in a certain state. 3 4 Ockham is here echoing
Grosseteste's doctrine that knowing A and knowing not-A is just the same
in God. But however it is with God, with us there seems to be a (formal)
difference between knowing A and knowing not-A. In one case we are in
one mental state and in the other case in another. Again, if Christ reveals
that Peter will deny him three times, the world seems really different from
the way it would be if such a revelation were not made.
Ockham actually has two very different accounts of prophecy. In his De
35
praedestinatione , he treats prophecies as disguised conditionals, usually as
conditionals asserting what God would do under certain conditions. O n
Ockham's usual account a conditional sentence is true just in case the
33. 'Et sic non est necessaria, nee debet concedi quod Deus habeat sciemiam necessariam de futuris
contingentibus sed potius contingentem, quia sicut hoc futurum contingens contingenter erit, ita
Deus seit ipsum contingenter fore, quia potest non scire ipsum fore, si ipsum seit.' Ockham 1978,
Q. II, art. 4; p. 530.267-71.
34. 'Dico quod illud quod est in Deo vel potest esse in eo formaliter necessario est Deus; sed scire a non
est sic in D e o sed tantum per praedicationem, quia est quidam conceptus vel nomen quod
[aliquando] praedicatur de Deo et aliquando non.' Ockham 1978, Q. II, art. 4, p. 531.302-5.
35. 'Et concedo quod non fuit revelatum tamquam falsum, sed tamquam verum contingens et non
tamquam verum necessarium, et per consequens tale potuit et potest esse falsum. Et tamen
Prophetae non dixerunt falsum, quia omnes prophetiae de quibuscumque futuris contingenribus
fuerunt condicionales, quamvis non semper exprimebatur conditio. Sed aliquando fuit expressa,
sicut patet de David et throno suo; aliquando subintellecta, sicut patet de Ninive destructione a
lona prophetata: "Adhuc post quadraginta dies et Ninive subvertetur", nisi scilicet poeniterent; et
_8
quia poenituerunt, ideo non fuit destructa." Ockham 1978, Q- ', "d *. Ρ· 5 ΐ 3 · ΐ 7 3 2 ·
36
corresponding inference holds; there is then no need to treat prophetic
conditionals as contingent at all. But when Christ asserts 'Peter, before the
cock crows you will thrice deny me', he seems to be making a categorical
and not a conditional assertion. Indeed it is not easy to see what could be the
suppressed antecedent in such a case.
In Quodlibet IV, Q. 4, Ockham proposes a different account of prophecy,
one on which prophecies do turn out to be categorical. Here he argues that
after God has revealed a future contingency it is necessary that the physical
or mental things he used to reveal it have existed, but what is revealed is not
necessary; for while 'God says that P; therefore P' is a valid inference 'God
utters "P"; therefore God reveals that P' is not. If Peter had not denied
Christ three times, Christ's utterance 'Peter, you will deny me three times'
would not have been a revelation that Peter would deny him three times.37
Robert Holkot
Robert Holkot, one of Ockham's followers, devotes the last part of Book
II, Q. II of his Seniencei-commentary to a discussion of whether God is able
to reveal a future contingent. Holkot begins by distinguishing two senses
of'reveal'. To reveal is to cause a new assent in the mind of some creature
(1) to a true complex, or (2) to a complex whether true or not. 38 Holkot
claims that in thefirstsense of'reveal' 'God revealed that there will be a day
of judgement' is contingent while in the second sense it is a sentence about
the past involving no future-tensed sentence and hence necessary.39 But it
36. '... condicionalis aequivalet uni consequentiae, ita quod tunc condicionalis est vera quando
antecedens infert consequens et non aliter.' Ockham 1974a, Pars II, c. 31; p. 347.1-4.
37- 'Et si dices quod hec aliquando fuit vera "Deus creat hoc" demonstrata propositione de futuro,
ergo semper postea haec erit necessaria: "Deus creavit hoc." Respondeo et dico "Deus creavit
hoc" si per ly hoc demonstratur ista propositio de futuro vel ista qualitas quae est propositio post
instans creationis ista de praeterito erit necessaria quia sua de praesenti non dependet a futuro. Sed
si demonstratur per ly hoc revelatum vel notitia evidens tunc iste de praeterito est contingens quia
ista de praesenti dependet ex futuro. Nam ista propositio "Deus creat revelatum vel notitiam
evidentem" importât quod hoc revelatum erit quia falsum nee evidenter sciri nee revelari potest.
Sed ista "Deus creat hanc propositionem vel qualitatem" non importât quod ista propositio sit
vera vel falsa.' Ockham 1491, Quodl. IV, Q. 4.
38. 'Tertia distinctio est de isto termino "rcvclare" quia isto possumus uti dupliciter. Primo modo pro
causatione novi assensus in intellectu alterius creaturae alicui complexo vero ... Secundo modo
accipitur "revelare" pro causare assensum alicui complexo in intellectu alterius de novo.' Holkot
1518, Lib. II, Q. II, art. 8 AA.
39. 'Primo modo accipiendo hoc nomen "revelare" haec est contingens: "Deus revelavit Sorti dies
iudicii fore" et postquam hoc revelavit Sorti haec est contingens: "Iste assensus fuit revelatio" -
demonstrate assensu qui in rei veritate fuit revelatio quia sic iste terminus "revelatio" est unus
terminus connotativus . . . Si autem iste terminus "revelatio" exponatur secundo modo pro co
quod est simpliciter causare assensum alicui complexo, sic est necessaria haec: "Deus revelavit
Sorti quod dies iudicii erit" et non dependet ab aliquo futuro quia sic potest Deus revelare falsum
seems that if God asserted that there would be a day ofjudgement and it is
still contingent whether there will be one, then either God can be deceived
or he can lie. Holkot denies that God can lie, for to lie is to speak a falsehood
with an improper intention to deceive, and God can have no improper
intentions. He also denies that God can be deceived; but he admits that God
can utter a falsehood and can fail to fulfill a promise. God has deceived evil
men; Holkot claims that God never will deceive good men, but not that he
cannot do so.40
Thomas Bradwardine and the reaction to Ockhamism

The Achilles' heel of the Ockhamist theory appears to have been its in-
ability to give a plausible, orthodox treatment of prophecy, but that was not
the only feature likely to provoke a conservative theologian.41 The strong
reaction to the Ockhamist position began with Thomas Bradwardine.
Bradwardine's views on future contingents are intimately connected
with his views about time and modality. Drawing to some extent on the
De libero arbitrio of Grosseteste (which he quotes extensively in De causa
Dei, e.g., Ill, c. 50) and to some extent on Anselm, Bradwardine insists that
God is immutable and hence that the divine power is not affected by the
passage of time.42 Whenever it is correct to say that God was or will be able
to do something it is correct to say that God is able to do it. Thus if God was
able to bring it about that Adam never would exist, he is able to bring it
about that Adam never did exist. His insistence on the immutability of the
si velit decipere et fallere creaturas et nullum inconveniens sequitur ex hoc quia nullum incon-
veniens est quod illo qui est Veritas causet in aliquo falsitatem.' Holkot 1518, Lib. II, Q- II, art. 8
A A (but CC follows immediately).
40. Holkot takes up this complex of questions in Holkot 1518, Lib. II, Q. II, art. 9 and in
his quodlibetal question 'Utrum generalis resurrectio necessaria sit futura' (MS Cambridge
Pembroke Coll. Cod. 236f, i86 v b -i87 n ). W.J. Courtenay and Paul Streveler have edited this
question and two others of Holkot's on future contingents and kindly allowed me to examine
their as yet unpublished edition.
41. We have a collection of epigrams from early in the fourteenth century - the Notabilia of Richard
of Campsall (in Synan 1962) - which seem almost designed to startle theological conservatives.
Several very similar epigrams appear in Holkot 1518, Lib. II, Q. II, art. 7, and in Peter of Ailly
1490, Q. XI, f. qa*. Campsall's Notabilia suggest the impact of the Ockhamist approach.
Campsall has the skeleton of a very different theory of future contingents in his (presumably
earlier) Questiones super librum Priorum analeticorum (in Campsall 1968).
42. Divine immutability is a major theme of De causa Dei; see, e.g., Lib. I, c. 23-5. That time does not
affect the divine power follows from Bradwardine's view that the past, the present, and the future
all depend on God's power: "Si Deus esse desineret, nihil esset praeteritum, nee futurum, verum
nee falsum, possibile vel impossibile, necessarium vel contingens, nee etiam posset esse.' Brad-
wardine 1618, Lib. I, c. 14, p. 209.
divine power provides Bradwardine with the core of his solution to the
problem of future contingents: the future and the past are contingent in
exactly the same sense - both depend upon the free will of God.
But to suggest that he sacrifices the necessity of the past to safeguard the
contingency of the future would be to mistake Bradwardine's project.
Both past and future are contingent on God's will, but given God's
antecedent will, which is, Bradwardine claims, naturally prior to what
God foresees, the whole history of the world is determined. 4 3 He dis-
tinguishes first between antecedent and consequent necessity (consciously
echoing Anselm) and distinguishes antecedent necessity into two sorts: one
'wholly absolute' and another 'in some way relative'. This second sort in
turn is divided into necessity relative to secondary causes and necessity
relative to the first cause. Everything that happens is necessary relative to
the first cause, the divine will. 4 4 Bradwardine is consciously reviving the
view that God predestines everything. His forceful presentation of this
view is central to its popularity in the fifteenth and sixteenth centuries.
Given his views about God's determination of the world, Bradwardine
could have claimed that God knows the world perfectly in knowing his
own intentions; but in fact his doctrine is more complex. In De causa Dei I,
c. 24, he argues that God knows everything at once, reviving the image of a
circle with God at the centre and the instants of time on the circumference.
But in his use of the image God sees something first as future, then as
present, and finally as past, much as an all-seeing eye located at the centre of
a revolving circle would see a point first in one quarter and then in another.
What God 'sees* changes with time (though God does not), and so God can
know temporally indexical truths; God 'sees' by means of his causal
power. 45
43. 'Secundum hanc ergo distinctioncm videtur mihi dicendum quod non omnia quae evenient,
evenicnt de necessitate penitus absoluta sicut duodecimo huius probat, nee etiam respectiva,
respectu scilicet aliquorum aut omnium inferiorum sive secundarum causarum sicut 5, 8 & 9
huius ostendunt, sed omnia quae evenient, evenient de necessitate respectiva, respectu scilicet
superiorum sive primarum causarum, quae sunt voluntas et potentia summi Dei.' Bradwardine
1618, Lib. Ill, c. 27, p. 704 C & D.
44. 'Possunt autem haec omnia in unam syllogisticam rationem breviter sic reduci: Omnia quae
evenient de necessitate naturaliter praecedente, de aliqua necessitate evenient. Omnia quae
evenient, evenient de necessitate naturaliter praecedente quia voluntas divina quae omnibus suis
volutis est nécessitas naturaliter antecedens, ergo omnia quae evenient de ista necessitate evenient.'
Bradwardine 1618, Lib. Ill, c. 27, p. 705B.
45. 'Aliter autem fortassis clarius et brevius dici potest, quod omnes sunt in Deo non realiter, sed
causaliter, sicut secundum Philosophum 4 Phys. 23 res dicitur esse in sua causa'. Bradwardine
1618, Lib. Ill, c. 14, p. 690B-C.
The fourteenth-century controversy

Bradwardine's work is seminal in a conceptual revolution which was
taking place in the fourteenth century. During the preceding millenium
the concepts of possibility and necessity derived their content largely from
theories about the powers of things. The Condemnation of 1277 had
clearly established that God's power extended beyond the actual course of
nature, but even this had not produced a fundamental change. One still
looked to God's causal power for insight into possibility, and so intuitions
about causality and power (for example the intuition that an effect could
not precede its efficient cause) were still crucial to modal theories. Duns
Scotus' insistence on a simultaneous power for opposites and his argument
that this was possible simply because there was no repugnantia terminorum in
the supposition that a being who in fact willed one contrary willed the
other signalled a new focus. Bradwardine accepted Scotus' claim that only
where there was a repugnantia terminorum in one contradictory was the
other necessary and used it to argue that the past was no more necessary
absolutely speaking than the future. In doing so he attacked a premiss on
which the whole Ockhamist theory of future contingents had been con-
structed and so set the stage for a struggle which exercised the universities
in the middle of the fourteenth century.
The nature and some of the details of this struggle have become clearer
within the last decade. 46 It seems to have begun as a dispute centred at
Oxford in the 1330s between Bradwardine and a number of representa-
tives of the new Ockhamist opinio communis, notably Adam Wodeham and
Robert Holkot. The issues involved include problems about foreknowl-
edge, especially prophecy, but centre mainly on problems of predestination
and the relative roles of merit and grace in salvation. At some point around
1340 the focus of the debate shifts to Paris, where, it seems, Thomas
Buckingham champions the one side and Bradwardine (represented by
writings earlier than the final version of the De causa Dei) the other. In 1342
Gregory of Rimini devotes a major part of his Sentences commentary to the
questions involved in this debate. Gregory is a major figure in the con-
troversy and seems to have been the architect of a position which reflects
46. The foundation for this recent activity was laid in part in Trapp 1956. Much recent work has been
done by W.J. Courtenay; see, e.g., Courtenay 1972-3 and 1978. Bartholomew de la Torre has
been working on a study of Thomas Buckingham's views on future contingents, and I have
benefited from conversation with him.
Ockham's account of future contingent truth but accepts Bradwardine's

criticisms of the view that the past is necessary.47
Peter of Ailly
The results of the debate may be seen in the work of the late fourteenth-
century figure Peter of Ailly. His Quaestiones super libros Sententiarum were
read at Paris probably in 1378 and so give us a picture of the situation after
the initial controversy had somewhat subsided. Much of the difficulty of
the problem of future contingent truth is generated by the tension between
the supposed necessity of the past and the supposed contingency of the
future. Hence it is rather startling to find d'Ailly arguing that the con-
tingency of the future 'is only to be held on faith',48 and maintaining that
the necessity of the past is only probabile, not evidens.*9 His major argu-
ments against the necessity of the past rest on his inability to find a relevant
difference between past and future which would justify us in claiming that
one was contingent and the other not. If God could once prevent the world
ever having been but can no longer, it must be either because the divine
power has changed or because something outside of God constrains his
will, both of which are absurd.
Peter of Ailly's discussion of revelation50 is heavily dependent on Adam
Wodeham and Gregory of Rimini and is a good summing up of the state of
the discussion near the end of the fourteenth century. He operates from
within the framework of the Ockhamist approach and considers a number
of ways of reconciling revelation with the approach. The first is the one
proposed by Ockham in his De praedestinatione - namely that all revelations
are disguised conditionals. This is rejected as too much limiting God's
power to reveal. A second is a proposal made by Gregory of Rimini and
47. No philosophical consensus was produced in this controversy, and the position worked out by
Gregory of Rimini and hisfollowersby no means dominated later discussion. For example John
Buridan, Albert of Saxony, and Marsilius of Inghen all adhered to the Opinio communis of the
early fourteenth century. (See, e.g., Buridan 1588 and Marsilius of Inghen 1501.) And as Baudry
(1950) has shown, several thirteenth- and fourteenth-century viewpoints were being debated at
Louvain in the late fifteenth century.
48. 'Tertia est quod non est nobis evidens nee naturaliter demonstrabile quod aliquid futurum potest
non fore. Patet quia in naturali lumine ita faciliter vel non magis difficiliter sustineretur quod
praeteritum potest non fuisse sicut quod futurum potest non fore.' Peter of Ailly 1490, Q. XI,
f.q4*C.
49. 'Quarto sequitur quod probabile est quod nullus de facto sic obligatur de peccato suo penitere
quod teneatur velle peccatum suum non fuisse. Patet ex dictis suppositio quod impossibile sit non
fuisse est probabile quamvis non evidens.' Peter of Ailly 1490, Principium, f. a4r.
50. Peter of Ailly 1490, Q. 12, art. 3.
designed to save the possibility of infallible revelation. Gregory suggests

that while the physical or mental sign by which God reveals something is
part of the past, and hence it is necessary that it have existed in whatever
sense of'necessary' past things are necessary, yet the significance of the sign
is no more part of the past than is the truth of a proposition. Hence just as
Peter could make a true sentence about the future never to have been true,
so Peter could make Christ's utterance 'Peter, you will deny me three
times' never to have meant that Peter would deny Christ three times. 51
Ailly rejects this proposal as well, arguing that a sign's having a certain
sense can have real effects over and above the effects the existence of the
physical object which is the sign would have. Hence either Gregory is
appealing to God's absolute power to change the past (in which case he
need not stop with meanings) or his proposal fails. Peter finally settles for
the view that Wodeham, Holkot, and Ockham all seem finally to have
accepted: there can be no revelation of a future contingent to a creature
which precludes the possibility of the creature having been deceived. This
of course does not mean that there can be no certain revelation. W e can
know that God will not deceive us; what we cannot know (because it is
false) is that God cannot deceive us.
Knowledge without causal connection: scientia media

Perhaps the least well-developed aspect of thirteenth- and fourteenth-
century theories about contingent truth and foreknowledge is their ac-
count of how a contingent truth about the future can be known. The
stumbling-block is the role causality ordinarily plays in knowledge.
Ordinarily we know things because they (indirectly) causally affect us.
51. See Gregory of Rimini 1955, Lib. I, D. 42-4, Q. 3, art. 1, f. i68"A: 'Unde si teneatur quod res
praeterita potest per divinam potentiam non fuisse diceretur possibile esse ut Deus numquam illud
dictum vel ilium assensum creaverit et per consequens non dixit "a fore" et ille alius etiam non
credidit. Si hoc vero non teneatur sicut aliqui antiqui doctores dixerunt quod possibile est illud
dictum non fuisse significativum, aut non illius significati quid est "a fore" ct similiter assensum
alterius non fuisse assensum seu iudicium de a fore. Sicut etiam dicendum esset quod iudicium quo
iudico Petrum sedentem sederc, si conscrvaretur in mc a Deo postquam Petrus surrcxit non esset
iudicium quo iudicaret Petrum sedere. Vel adhuc aliter dici potest. Nam ut dicendum est ad hoc
quod aliquis dicat aliquid proprie loquendo requiritur intentio significandi illud per suum dictum.
Cum ergo Deus quid voluit possit non voluisse alias nullus praedestinatus posset non fuisse
praedestinatus cuius oppositum communiter tenetur secundum sensum divisum. Possibile est
Deum numquam voluisse significare per suum dictum "a fore" et per consequens numquam
dixisse "a fore". Si vero nulla praedictarum viarum placet inveniatur alia melior. Et si invenire
quis non possit potius iudico confitendum esse ignorantia quam praecipitandum se in tarn
horrendam auribus fidelium . . . qualis est si dicatur Deum posse mentiri aut dicere falsum aut
decipere cum omnis scriptura catholica canonis et sanctorum clamet oppositum.' Note that
Gregory does not endorse any of the views he suggests.
Occasionally we know things because we bring them about. But future

contingents do not exist and so cannot have effects, and attempts to ground
knowledge of them in their causes seem always to threaten their con-
tingency. What is needed is a via media, a way of having knowledge
without causal connection. Many writers, Ockham and Gregory of
Rimini among them, were content to say that God knows the future
intuitively and that it is a mystery how he knows it. It is not until the late
sixteenth century that we find a well-developed and promising proposal.
The architect of the proposal seems to have been the Spanish Jesuit Luis
de Molina,52 and the proposal is that besides God's knowledge of all
necessary truths and his knowledge of his own intentions, God has a scientia
media — a knowledge of all true subjunctive conditionals regarding con-
tingent matters. Molina couples this doctrine with the view that of any
pair of conditionals 'If it were to be that A, it would be that B' and 'If it
were to be that A, it would be that not-B' (where A is a complete
specification of a context) exactly one is true. This is sometimes called the
Principle of the Conditional Excluded Middle. The principle is far from
obvious and, indeed, Dominican theologians contemporary with Molina
argued against it. Their debate is important, and not only for the history of
the problem of foreknowledge, for it is a lively issue today in the theory of
counter-factual conditionals.53
The major argument against the Principle of the Conditional Excluded
Middle is just that it seems false. There are cases in which it seems that if
someone were in situation X he might do Y and might not. Indeed, one
might take this as a partial account of what it is to be free in situation X. The
arguments in favour of the principle are more subtle. The best of them,
developed by Molina's contemporary Francisco Suârez,54 exploits a puta-
tive connection between subjunctive and indicative conditionals. Many
52. The authoritative statement of Molina's views is in Molina 1876. Other works of Molina dealing
with the question and an account of his life and work may be found in Stegmüller 1935. Perhaps
the most central passage in Molina's work is this from Molina 1876, p. 317: 'Tertiam denique
mediam scientiam, qua ex altissima et inscrutabili comprehensione cuius liberi arbitrii in sua
essentia intuitus est, quid pro sua innata libertate, si in hoc, vel illo, vel etiam infinitis rerum
ordinibus collocaretur, acturum esset, cum tamen posset, si vellet, facere re seipsa oppositum.'
53. The sixteenth- and seventeenth-century debate about the scientia media (and so about the Principle
of the Conditional Excluded Middle) ranged Molina and his supporters against such opponents as
Diego Alvarez and Domingo Baiiez. On Bellarmine's work in support of Molina see Tromp
•933· For an enlightening discussion of the sixteenth-century debate in the light of twentieth-
century issues, see R. Adams 1977.
54. 'Si Deus (verbi gratia) postquam voluit Petrum creare, et in tali opportunitatc constituere aut
permittere, vidit ilium peccaturum, iam de illo conditionalis est vera: "Si Petrus creatur in tali
occasione, pcccabit"; ergo etiam si intclligcrctur ilia oratio proferri prius ratione, quam Deus
have supposed that one can test a conditional of the form 'If X were to
happen Y would happen' by bringing it about that X does happen. If Y
ensues, the original conditional was true; if Y does not, it was false, but in
that case 'If X were to happen Y would not happen' is verified. Thus it
seems that in a situation in which a subjunctive conditional has a true
antecedent the Principle of the Conditional Excluded Middle holds. But
how can it make a difference to the truth of the conditional whether one in
fact brings about the antecedent or not? Suârez and Molina think it cannot
make a difference.
Molina's position is that not only is the Principle of the Conditional
Excluded Middle true but God knows which of each pair of conditionals is
true and he knows this even though such conditionals are themselves
contingent. If the doctrine of the scientia media is to help, it must not merely
explain God's knowledge of one kind of contingent truth in terms of
another more mysterious, but Molina himself appears to do just this. When
pressed as to how God knows which of a pair of subjunctive conditionals
(whose subject terms may even denote only non-existents) is true, Molina
replies that we should not look for an explanation on the side of the known
but on the side of the knower. It is because God's intellect is so keen that he
has an especially acute comprehension of things and hence knows what
they would do under any possible circumstance. 55
This is not much of an advance beyond Ockham and Gregory of
Rimini, and the view is mercilessly attacked by Molina's opponents,
particularly by Domingo Bafiez, as incoherent. 36 The theory of the scientia
media is put on a surer footing by Suârez, 57 who argues that what a
particular free agent would do in given circumstances is grounded in a sui
generis property of that agent. On Suârez' view this is a contingent
vellet Petrum creare, etiam ilia propositio esset vera determinate. Probatur consequentia, quia iam
habuisset ex tunc conformitatem cum re significata; hanc enim non fecit, sed ostendit potius et
manifêstavit effectus subsecutus, postquam Deus voluit Petrum creare etc.; ergo eadem ratione
omnis conditionalis similis intellecta vel cogitata in illo signo, antequam Deus aliquid decrcvcrit
aut voluerit de illo antécédente sub conditione sumpto, est determinate vera, aut determinate
falsa, quia statim ac Deus ponere vult inesse illud antecedens, absolute determinatum est quid ex
illo sequendum sit vel non sit; ergo illud ipsum antea sub conditione prolatum erat determinate
verum, aut determinate falsum, iuxta exigcntium obiecta.' Suârez, 'De scientia quam Deus habet
de futuris contingentibus'. Lib. II, c. 5, n. 13; in Suârez 1856-77, vol. 11, p. 359.
55. 'Neque enim ad intuendum in re libera in quam partem se inflectet satis est illius comprehensio,
neque quaecumque major comprehensio quam sit res comprehensa, sed necessaria est altissima
atque eminentissima comprehensio, qualis in solo Deo comparatione creaturarum reperitur.'
Molina 1876, Q. 14, art. 13, disp. 52; p. 319.
56. See Baiiez' Apologia en Defensa de la Doctrina Antigua y Catholica (in Hercdia 1968), Pars Prima,
c. 16; p. 185.
57. See his De gratia, prol. 2, c. 7; in Suârez 1856-77, vol. 7.
property, and so there remains some question about how God could know
that a non-existent agent (for example a man before his birth) 'possesses' it.
But on Suârez' view it is also a timeless property and so there is no special
problem of/oreknowledge. By knowing how he starts the world and how
things would evolve were they to begin in that way God knows the
world's entire history. This suggestion on the one hand echoes theories of
complete individual divine ideas, and on the other hand suggests the theory
of complete concepts which was to be proposed by Leibniz. But it differs
from both these in that it does not claim that the property which grounds
the conditional is an essential property of the individual who has it. On the
contrary, since the conditional is contingendy true the property which
grounds it must be contingent. But how can even God know the con-
tingent properties of non-existents? We seem to have come full circle.
VI
METAPHYSICS A N D EPISTEMOLOGY
19
ESSENCE AND EXISTENCE
First philosophy, divine science, and the science of'being as being'

The recovery of Aristotle's Metaphysics by medieval Western thinkers
prepared the way for them to concentrate on the science of'being as being'
in the high Middle Ages. This work was enhanced by the translation into
Latin of Avicenna's Metaphysics in the twelfth century and of Averroes'
Commentary on Aristotle's Metaphysics in the early thirteenth century.
But as medieval Latin thinkers began to examine Aristotle's text more
closely, they encountered a problem of interpretation relating to the very
nature of metaphysics.
In Metaphysics IV, c. ι (ioo3 a 2i-32), Aristotle speaks of a science which
studies being as being and contrasts it with more particular sciences which
restrict themselves to investigating the attributes of a portion of being. But
in Metaphysics VI, c. ι (i026a23-32), after referring to his investigation of
'beings as beings' and presumably, therefore, to his science of being as
being, Aristotle distinguishes three theoretical sciences — physics, mathe
matics, and first philosophy or 'divine science' — and then seems to justify
the viability of the last-mentioned one only insofar as it concerns itself with
separate and immobile entities. One might wonder whether this first
philosophy or divine science can be identified with Aristotle's general
science of being as being, a difficulty which he himself recognises. He
concludes the discussion by asserting that if there were no separate and
immobile entity, then physics would be the first science. But if there is
some immobile substance, then the science that considers it will 'befirstand
will be universal insofar as it is first. And it will pertain to this same science
a
to study being as being' (i026 30—i). Aristotelian scholars continue to
debate his justification for this move and the degree to which he did or did
not in fact succeed in bringing together his science of being as being and his
first philosophy;1 and this same problem exercised the philosophical skills
I. For reviews of recent discussions of this see During 1966, pp. 594-9; König 1970, pp. 226ff.;
Owens 1978, pp. xiii-xxvii, 35-67.
386 Metaphysics and epistemology
of many of his followers during the high Middle Ages, although they
tended to formulate it in terms of their efforts to identify the precise
'subject' of metaphysics.2
Avicenna and Averroes on the subject of metaphysics
The divergent positions of Avicenna and Averroes on this issue were well-
known to thirteenth- and fourteenth-century Latin scholastics.3 In his
Metaphysics Avicenna examines in some detail the possibility that meta
physics might have as its subject God, but he rejects it. For him its subject
can only be being as being. Since he is convinced that no science can
demonstrate the existence of its own subject and since he also maintains
that God's existence can be demonstrated in metaphysics and only in
metaphysics, he seems to have no other choice.4 But since God's existence
will be established in metaphysics, for Avicenna the philosophical science
that studies the divine is a part of the science of being as being.5 And it
would seem to follow that God, too, will be included under the notion of
being as being.6
Averroes flatly rejected Avicenna's position on this question. While he
agrees that no science can demonstrate the existence of its subject, he insists
that it is in physics rather than in metaphysics that one demonstrates the
existence of God, the first mover. 7 Hence while he does not reject 'being as
2. See Zimmermann 1965; Doig 1965, pp. 41—96; Doig 1972, pp. 172—213.
3. See for instance, Siger of Brabant 1948, Inlrod., q. 1, pp. 3-4; and especially Duns Scotus,
Quaestiones super libros Metaphysicorum, I. 1 (1891 -5, v. 7, p. 11): 'Utrum subiectum Metaphysicae
sit ens inquantum ens, sicut posuit Avicenna? vel Deus, et Intelligentiae, sicut posuit commen
tator Averroes?' In referring to Avicenna and Averroes in what follows 1 will be taking account of
only the medieval Latin translations of their works.
4. In his Opera (1508, repr. Minerva, 1961), see I. 1-2. See f. 70™'': God is not the subject of
metaphysics. Note in particular: 'Postquam autem inquiritur in hac scientia an sit, tunc non potest
essesubiectum huius seien tiae. Nulla enim scientiarum debet stabilire esse suum subiectum' (70**).
Here too he rejects the suggestion that the causes might be its subject (ff. 70 Λ -7ΐ "*)· On being as
being as the subject of metaphysics see f. 70 ν Λ . For the above in the recently published critical
edition (Avicenna Latinus. Liber de philosophia prima) see pp. 4-8, 10-13. On this in Avicenna see
Zimmermann 1965, pp. 108-16; Doig 1965, pp. 73-82; Brown 1965, pp. 117-19; Gilson 1927,
pp. 91-9: Goichon 1937. pp. 3-5.
5. See f. 71 **, PP· 14-15 (erit. ed.). Here Avicenna divides metaphysics into different parts including
one which investigates the ultimate causes insofar as they are causes of every caused being insofar
as it is being ('omnis esse causati inquantum est esse'); others which study the first cause from
which flows every caused being insofar as it is caused being ('et aliae inquirunt causam primam ex
qua nuit omne esse causatum inquantum est esse causatum, non inquantum est esse mobile vel
quantitativum'); others which inquire after the dispositions which 'happen' to being; and finally
others which study the principles of the particular sciences.
6. Unfortunately, Avicenna does not make this point as clearly as one might wish. For discussion see
Zimmermann 1965, pp. 112-14.
7. See Averroes 1562~74d, I, com. 83, ff. 47"*—48": 'Sed notandum est quod istud genus entium, esse
scilicet separatum a materia, non declaratur nisi in hac scientia naturali. Et qui dicit quod prima
Essence and existence 387
being' as a description of the subject of metaphysics, he maintains that what

is signified thereby is really substance.8 And it is his view that it pertains to
metaphysics to study substance in terms of its prime instance, that is, that
separate substance which is the first form and the end or final cause of all
other substance.9 But such a study presupposes the demonstration in
physics that such a substance, there established as thefirstmover, does in fact
exist.10 Averroes would apparently safeguard the general character of
metaphysics by reasoning that in studying the substance that is the first
form and ultimate end one also studies all other substance and thereby all
other being as well. ' '
Thirteenth- and fourteenth-century thinkers tended to side with Avi-
cenna against Averroes in their discussions of this issue. But although
most were in agreement that being as being rather than divine being is the
subject of metaphysics, the precise relationship between being as being and
divine being continued to divide them.
Siger of Brabant andfohn Duns Scotus on metaphysics and God

First of all, one might reason that insofar as God is a being, he will be
included under that notion of being which serves as the subject of meta-
Philosophia nititur declarare entia separabilia esse peccat. Haec enim entia sunt subiecta primae
Philosophiae, et declaratum est in posterioribus Analyticis [76bi -23] quod impossibile est aliquam
sciential» declarare suum subiectum esse.' See also II, com. 22, ff. 56v,,-57ra (where he interprets
Avicenna as saying that no science can demonstrate the causes of its subject and counters that if
this is true as regards demonstration simpliciter and demonstration propter quid, it is not true as
regards demonstration quia. Thus Aristotle's demonstration of the first mover is proper to and
peculiar to physics, and is a demonstration quia or, as he also refers to it, per signum); Averroes
I562-74f, XII, com. 5, f. 293".
8. Averroes 1562—74c, I, com. 83, f. 47 ,,, ': 'Omne enim de quo loquitur in hoc libro principaliter est
propter illud principium; et iste est primus locus in quo naturalis inspicit ahum modum essendi ab
illo de quo considérât, et apud ilium cessât, et dimisit considerationem de eo usque ad scientiam
nobiliorem, quae considérât de ente secundum quod est ens.' See also 1562-74C IV, com. 1,
f. 64rab (on Aristotle's effort to define the subject of this science, and on 'being as being'); com. 2,
ff. 6i*-66* (on the meaning of being and its manifold predication by way of reference or
attribution to substance).
9. Averroes 1562-74^ IV, com. 2, f. 66" (the philosopher, i.e., the metaphysician, must study
Λ
the principles and causes of substance); XII, com. 5, f. 293 ("Et dicemus nos quidem quod
Philosophus inquirit quae sunt principia substantiae secundum quod substantiae et déclarât quod
substantia abstracta est principium substantiae naturalis: sed hoc ponendo accepit pro constant!
hoc quod declaratum est in naturalibus de principiis substantiae generabilis et corruptible,... et
quod declaratum est in Octavo, scilicet, quod movens aeternam substantiam est abstractum a
materia ...) ; f. 293 " ('quod principium primae substantiae abstractae etiam est substantia et forma
et finis et quod movet utroque modo ...').
10. Aveiroes i562-74f, XII, com. 5, f. 293".
11. His commentary on the final lines of Aristotle's Metaphysics VI, c. 1, is not very enlightening on
this point. But see Doig 1965, pp. 53-60; also Zimmermann 1965, pp. 116-17; Gilson 1927,
pp. 93ff.; Gilson 1948b, pp. 66-7; Gilson 1952a, pp. 77-8.
physics. Siger of Brabant in the 1270s and Duns Scotus around the turn of
the fourteenth century may be regarded as representative of this position.
Siger concludes that since the subject of metaphysics must be universal,
God is not its subject. 12 He attempts to mediate between Avicenna and
Averroes on the proof of God's existence by suggesting that the middle
term whereby one establishes a conclusion in a given science need not be
proper to that science. Thus the middle term employed by Aristotle to
prove God's existence in Metaphysics XII is in fact taken from the Physics.13
The implication seems to be that for Siger one may establish God's
existence either in physics or in metaphysics, granted that the middle term
for doing so is physical. And in other contexts he offers more properly
metaphysical argumentation to show that this first mover is also the
efficient cause of all caused beings, and that it is unique. 1 4
Since Siger evidently regards as metaphysical the last-mentioned kind of
effort to arrive at more precise knowledge of the divine, it seems that he
does include God within the notion of being that serves as the subject of
metaphysics. This inference is reinforced by his refusal (along with
Avicenna) to admit that there are causes and principles for being as such.
Granted that it pertains to metaphysics to investigate general causes and
principles, these can be causes only of all caused being, not of being as
being. ' s If God were not included under the being that serves as the subject
of metaphysics, Siger could assign to metaphysics the study of the prin-
ciples and causes of being as such. 16
Duns Scotus touches on this problem more than once. By 'the subject of
a science' he usually means that adequate object of study which implicitly
contains the conclusions that may be drawn in a given science and which
has sufficient unity to account for the unity of that science itself.11 As he
12. Siger of Brabant 1948, Intr., q. 1, p. 4.

13. Ibid. Also see Zimmermann 1965, pp. 181-2.
14. For discussion of these and appropriate references see Van Steenberghen 1977, pp. 296-300.
15. Siger of Brabant 1948, Intr., q. 2, p. 5: 'Tamen entis, secundum quod ens, non est principium quia
tunc omne ens haberet principium.' Also see his Commentary on Book IV, p. 185.60-80; also on
Book VI, q. 1, p. 364.4-365.8.
16. Zimmermann 1965, pp. 185-6.
17. Ordinatio, ProL, pars 3, qq. 1-3, nn. 142-4 (Duns Scotus 1950-, v. I, pp. 96-7); Lectura in I Sent.,
Pro!., pats 2, qq. 1-3, n. 66 (Duns Scotus 1950-, v. 16, p. 26); and especially, his Reportalio
Parisiensis examinata (Reportatio I-A according to the classification of the Scotus commission, for
which see Duns Scotus 1950-, v. 1, i25*-6*), Prol., q. 1, a. 2 (MS Vienna bibl. nal. 1453, f. I,
pp. 1 - 2 of the transcription prepared by the late Victor Doucet). In light of recent findings by the
Scotus Commission, one should no longer cite as Scotus' own work the version of Book One of
the Reportatio Parisiensis contained in rhe Wadding and Vives editions. On this see Duns Scotus
1950-, v. 7, pp. 4*-6*.
explains in the Prologue to his Reportatio Parisiensis examinata, the conclu

sions in any science are known only by reason of their principles. And these
principles, if they are immediate, are themselves known from their terms.
But the terms are known by reason of the subject ofthat science. Therefore,
'that is the first object (subject) of a science which contains virtually and
potentially knowledge of all the truths ofthat science'.18
It is clear that when Scotus so defines the subject or object of a science he
will not admit that God is the subject of metaphysics.19 Hence in his
frequent contrasts between the positions of Avicenna and Averroes, he
usually defends Avicenna. The subject of metaphysics can only be being as
being. God cannot be the subject of metaphysics since the subject of any
science must be universal, not particular. Moreover, Scotus contends that it
is in metaphysics rather than in physics that one most effectively establishes
the existence of God, and so has another reason for not admitting God as
the subject of metaphysics.20 On the other hand, since God is one of the
things investigated by metaphysics, it follows that for Scotus God, too, is in
some way included within that general notion of being that serves as its
18. '... dico quod illud est per se primum obiectum alicuius scientiae quod continet virtualiter et
potemialiter notitiam omnium veritatum illius scientiae... quia conciusiones cognoscuntur ex
pnncipiis, ut patet ex dictis; et principia tandem, si sint immediate, cognoscuntur ex terminis,
sicut dictum est. Terminus etiam praedicatus principii cognoscitur ex ratione subiecti, quia
principia communiter sunt per se primo modo In isto ergo ordine tandem oportet devenire ad
aliquod subiectum simplex, quod est subiectum principii, vel principium principiorum, ex cuius
cognitione cognoscuntur omnia pertinentia ad scientiam." (MS Vienna 1453, f. 1, p. 2 of the
Doucet transcription).
19. Zimmermann 1965, pp. 267-9.
20. Ordinatio, Pro/., pars 3, q. 2, ad 2, nn. 189, 193-4 (Duns Scotus 1950-, v. I, pp. 127, 129-31);
Lectura, Piol. pars 2, q. 2, ad arg., η. 97 (Duns Scotus 1950-, v. 16, pp. 34-5); Report. Paris,
examinata, Prot., q. 3 (f. 8, p. 27 of Doucet transcription): 'De primo est controversia inter
Avicennam et Averroem. Posuit enim Avicenna quod Deus non est subiectum in metaphysica,
sed aliquid aliud ut ens, quia nulla scientia probat suum subiectum esse; metaphysial: probat
Deum esse et substantias separatas esse; ergo etc. Averroes reprehendit Avicennam in commento
ultimo 1 Physicorum: supposita maiori Avicennae, quod nulla scientia probat suum subiectum
esse, quae est communis utrique, capit quod Deus est subiectum in metaphysica et quod Deum esse
non probatur in metaphysica, sed in physica Sed Avicenna bene dicit et Averroes valde male.
Et accipio propositionem utriusque communem, scilicet quod nulla scientia probat suum ob
iectum esse, quae vera est propter primitatem subiecti ad scientiam, quia si esset posterius, posset
ipsum probari esse in ilia scientia in qua habet rationcm posterioris et non obiecti adaequati. Sed
maiorem primitatem habet subiectum respectu scientiae posterioris quam prions. Ergo, si scientia
prima non potest probare suum subiectum esse, quia est subiectum primum, ergo multo magis
nee scientia posterior... Dico ergo ad quacstioncm . . . quod Deus non est subiectum in meta
physica.' In this same context Scotus again defends the superiority of the metaphysical demon
stration of God. As is well known, Scotus spared no efforts in his concern to work out a
metaphysical demonstration of God's existence, one which has been described as 'what is, no
doubt, the most perfectly-elaborated proof of God's existence in the Middle Ages' (Maurer 1976,
p. 177)- For full discussion of this argumentation and for appropriate references sec AUuntis 1965,
pp. 133-70.
subject.21 This can be seen from the fact that he denies that there are
principles or causes of being as being, the subject of metaphysics, since this
would mean that every being, God included, has a principle or cause.22
Thomas Aquinas on metaphysics and God
The position of Thomas Aquinas on this issue is almost unique. He agrees

with Avicenna, Siger, and Duns Scotus in maintaining that the subject of
metaphysics is being as being or being in general (ens commune),23 but he
denies that God is included under this notion of being in general.24 In his
view it belongs to a science to seek after knowledge of the principles or
causes of its subject. And since Thomas regards God as the cause of ens
commune (meaning thereby the cause of all that falls under ens commune), he
proposes knowledge of God as the goal at which metaphysical investiga-
tion aims.25 This would seem to imply that argumentation for the ex-
istence of God is a part of metaphysics. But there are some passages in his
Commentary on Aristotle's Metaphysics which suggest that it is in physics
that one establishes the existence of an immobile, separate entity, and that
this conclusion is a necessary condition for beginning metaphysics.26
Hence contemporary interpreters remain divided over two issues. Accord-
ing to Thomas, does it pertain to physics, or to metaphysics, or perhaps to
both to establish the existence of God?2 7 And does the very possibility of
21. Zimmermann 1965, pp. 271—3. 'Deus autem etsi non est primum subiectum in metaphysica, est
tamen consideratum in ilia scientia nobilissimo modo quo potest considerari in aliqua scientia
naturaliter acquisita' (Ord., Prol-, pars 3, q. 2, n. 193, Duns Scotus 1950—, v. 1, p. 130).
22. ' . . . quia ends in quantum ens, quod ponitur subiectum metaphysicae, nulla sunt principia, quia
tunc essenc cuiuslibct ends' (op. cit., q. I, n. 191 [p. 128J).
23. Aquinas 195s,Q.5, a. i,nd6,p. i 7 i ; Q . 5,a.4,p. i94:'EtideopertractanturinilIadoctrina,inqua
ponuntur ea quae sunt communia omnibus entibus, quae habet subiectum ens in quantum est
ens'; In III Sent., d. 27. q. 2, a. 4. sol. 2; Aquinas 1971b, Prooemium, p. 2. On Aquinas'
understanding of the subject of a science see Zimmermann 1965, pp. 160—5.
24. Aquinas 1950b, v. 2, no. 660, p. 245: '... omnia existentia continentur sub ipso esse communi,
non autem Deus, sed magis esse commune continetur sub eius virtute'. ST, la Hae, q. 66, a. 5, ad4:
' . . . quia ens commune est proprius effectus causae altissimae, scilicet Dei".
25. Aquinas 1955, Q. 5, a. 4 (p. 192.21-5; p. 195.6-24); Aquinas 1971b, Prooemium, p. 2: 'Unde
oportet quod ad eamdem scientiam pertineat considerare substantias separatas, et ens commune,
quod est genus, cuius sunt praedictae substantiae communes et universales causae... Nam
cognitio causarum alicuius generis, est finis ad quern consideratio scientiae pertingit'; cf. SCG III,
c. 25. See Zimmermann 1965, pp. 174-5, especially on the different relationships of God and of
other separate substances to ens commune.
26. See especially Aquinas 1971b, IV.v, no. 593; Vl.i.nn. 1169-70; Xl.vii, n. 2267, also III. vi, no. 398.
On these see Wippel 1978, pp. 447-52.
27. On this see Owens 1953, pp. 109-21. For differing assessments as to whether in Thomas' eyes
Aristotle's argumentation in the Physics leads merely to a sphere soul or to God, see Paulus I933,
pp. 259-94, 394-424; Owens 1966, pp. 119-50; Pegis 1973, pp. 67-117.
metaphysics presuppose that one has already demonstrated the existence of

God or at least of some spiritual entity in physics?28
As regards the second of these disputed points, it seems that when
Thomas is clearly writing in his own name rather than merely as the
expositor of Aristotle, he does not maintain that one must base one's
discovery of being as being, the subject of metaphysics, upon a conclusion
already established in physics. But he does recommend that one study
physics (as well as logic, mathematics, and ethics) before taking up meta-
physics.29 And as regards the first disputed point, it is fairly evident that
Thomas does allow for strictly metaphysical argumentation for the ex-
istence of God, as one would expect from his discussions of the subject of
metaphysics.30 But when he is commenting on the Physics, he at least
appears to regard purely physical argumentation as also sufficient for
reaching this same conclusion.31
William Ockham on the subject of metaphysics

Quite different from any of these positions regarding the subject of meta-
physics is the one taken by William Ockham,32 who distinguished be-
28. For the view that for Aquinas the possibility of metaphysics does rest on such a conclusion from
physics see, for instance. Smith 1954, pp. 78—94; Smith 1958, p. 382; O'Brien i960, p. 160; Doig
1972, p. 243, n. 1; p. 303, n. 1; Weisheipl 1976, pp. 194-6. For a different interpretation see
Klubertanz 1954a, pp. 13, 17; 1954b, pp. 196-8; Renard 1956.
29. See Wippel 1978, pp. 452-68. For texts in Thomas on the proper learning order see Aquinas
1964b, Vl.vii, nn. 1209-11; Aquinas I954d, Prooemium, p. 2; Aquinas 1955, Q. 5, a. 1, ad 9
(p. 172:3-11). For fuller discussion of this see Wippel 1973; Owens 1966, pp. 131-2. On the
difficulty in determining when in his commentaries on Aristotle Thomas is simply interpreting
Aristotle, when he is also accepting such an interpretation as his own position, and when he is
going beyond the text to express his personal views, see Doig 1972, pp. ix-xiv, and the references
given there; Wippel 1978, pp. 452fr. It seems evident enough that both in his Commentary on the
De Trinitate and in the Prooemium to his Commentary on the Metaphysics Thomas is writing in
his own name. If, as appears to be the case, it is difficult to reconcile certain statements taken from
his Commentary proper (see n. 26 above) with those made when he is writing in his own name,
preference should be given to the latter in any attempt to determine Thomas' own mind.
30. See, for instance, the argumentation presented in his De ente et essentia, c. 4, granted that in certain
later writings Gilson has denied that this was really intended by Thomas to be a demonstration of
God's existence. See Gilson 1950, pp. 257—60; Gilson 1961a, pp. 26—8. For another interpretation
of this chapter together with references to others see Wippel 1979, pp. 279-95. Of Thomas' well
known 'Five Ways' in ST I, 2, 3, the first argument, based on motion (along with its parallel
version in SCG) has been interpreted by many as being physical rather than metaphysical. But for
a different reading sec Owens 1953, pp. 109-21.
31. See the conclusion of his Commentary on Physics VIII: 'Et sic terminât Philosophus con-
sidcrationcm communcm de rebus naturalibus, in primo principio totius naturae, qui est super
omnia Deus benedktus in saecula. Amen' (VHI.xxiii, η. 1172). One might argue, however, that
this concluding remark is really not germane to Thomas' opinion about the physical or meta
physical nature of Aristotle's reasoning here; see above, n. 27.
32. On this see Guclluy 1947, pp. 277-93; Maurer 1958, pp. 98-112; Zimmermann 1965, pp. 330-38.
tween the object and the subject of a science. The object of a science is
simply any proposition that is demonstrated within that science, but the
subject of a science is the subject of such a proposition. 33 Since a given
science, such as metaphysics, includes many propositions with different
subjects, it will have many different subjects. 34 One of the many subjects of
metaphysics is, of course, being, and it even enjoys a certain primacy; for
the metaphysician is especially concerned with arriving at conclusions
about being. From another vantage point God, also one of the subjects of
metaphysics, may be regarded as primary by reason of his pre-eminent
perfection. 35 But in Ockham's view there is no single unifying subject for
metaphysics as envisaged by Avicenna, Siger, Thomas, or Scotus. 36
Essence and existence in Boethius

Philosophers of the high Middle Ages felt a need to account for the
metaphysical structure not only of corporeal beings but also of immaterial
entities such as angels and separated souls. It was necessary to explain the
caused, contingent, and composite character of such entities while dis-
tinguishing it from the absolutely uncaused, necessary, and simple nature
of God. As a consequence, controversy arose concerning the relationship
between essence and existence in creatures. 37 Although Boethius himself
should not be regarded as a proponent of real distinction and composition
of essence and existence in such entities, he had provided considerable
impetus for those later discussions through a series of axioms at the
beginning of his De Hebdomadibus. There he had written that 'being (esse)
and that which exists are different'. He had also maintained that although in
a simple entity 'its being and that which it is are one and the same', in any
composite 'its being and that which it is are not one and the same'. 3 8 Such
33. William Ockham 1957b, Prologue, p. 9; see also Ockham 1967, Prologue, q. 9, p. 266.17-22.
34. Ockham 1957b, Prol., pp. 8-9; Ockham 1967, Prol., q. 9, pp. 247-9.
35. Ockham 1957b, Prol, p. 10; Ockham 1967, Prol., q. 9, pp. 255.12-257.2, 258.14-259.13.
36. See, for instance Ockham 1967, Prol., q. 9, p. 255.13-15: 'Ad tcrtium dico quod metaphysicae ct
similiter mathematicae non est unum subiectum, loquendo de virtute sermonis, sed quot sunt
subiecta condusionum tot sunt subiecta scientiarum.' See the immediately following context for
his defence of some order between these various subjects. For some reservations as to whether this
approach can really safeguard the unity of a science such as physics or metaphysics see Maurer
1958, pp. 109—10.
37. A complete history of this controversy concerning the relationship between essence and existence
remains to be written. For helpful introductions see Grabmann 1924, pp. 131-90; Roland-
Gosselin 1948, pp. 142-205; Gilson 1955, pp. 420-7; Paulus 1938, pp. 260-91.
38. Often referred to as the De Hebdomadibus by medieval authors, this work appears in the Rand
edition under the title: 'Quomodo substantiae in eo quod sint bonae sint cum non sint sub-
stantialia bona' (p. 38). See Axioms II: 'Diversum est esse et id quod est...'; VII: 'Omne simplex
esse suum et id quod est unum habet'; VIII: 'Omni composito aliud est esse, aliud ipsum est'
Boethian texts were frequently cited by thirteenth-century participants in

the controversy over the relationship between essence and existence.39
Essence and existence in Avicenna and Averroes

Avicenna was thought by many scholastics to have defended an extreme
form of distinction between essence and existence. For instance, according
to Aquinas, Siger of Brabant, and James of Viterbo, he had not only
distinguished between them but had regarded existence as a kind of
accident that is superadded to essence.40 Averroes had already taxed
Avicenna with having made this mistake both with respect to being and
with respect to 'unity' or 'oneness', and it was an easy step for Latin critics
to take up the Averroistic critique and sometimes also to apply it to
thirteenth-century defenses of real distinction between essence and ex-
istence.41 They would argue that if an entity enjoys real being only by
reason of something that is superadded to its essence, then why not posit
something else by reason of which that superadded thing (existence) itself
enjoys reality, and so on ad infinitum?*2
(pp. 40, 42). For different interpretations of these and the other axioms in Boethius himself see
Bruder 1928, pp. 72ff. (who does see real distinction between essence and existence implied there);
Brosch 1931, pp. 1-12I; Roland-Gosselin 1948, pp. 142-5; Fabro 1950, pp. 98-105, 25—33;
Fabro i960, pp. 204-13; Fabro 1961. pp. 267-79; (in each of the above Fabro also considers
Thomas' commentary on the Boethian text); Schrimpf 1966, pp. 5-26 (pp. I2iff. for a brief
survey ofother treatments of Thomas' commentary and for that commentary itself); Hadot
1970, pp. 143-56; Mclnerny 1974, pp. 227-45.
39. In his Commentary on the De Hebdomadibus, Thomas interprets Axiom II as cited above as
implying only logical or intentional distinction. But he interprets Axiom VIII in the sense of real
distinction between essence and existence (esse). See In De Hebdomadibus, L. II, n. 22: 'Quae
quidem diversitas non est hie referenda ad res, de quibus adhuc non loquitur, sed ad ipsas rationes
seu intentiones' (p. 396); n. 32: '... sicut esse et quod est différant in simplicibus secundum
intentiones, ita in compositis différant realiter' (p. 398). For other passages in which Aquinas uses
the Boethian terminology to express his own theory of composition of essence and esse in
creatures see: In I Sent., d. 8, q. 5, a. 1; De ente, c. 4; STI.q. 50, a. 2, ad y. Aquinas 1956, Quodl. 2,
q. 2, a. 1 (p. 24); Quodl. 3, q. 8, a. 1 (p. 61). For Giles of Rome's citation of the Boethian axioms in
support of his theory of real distinction of essence and existence see Giles of Rome 1503, q. 9,
f. i8,a"b. There Giles also contests Henry of Ghent's 'theological' interpretation of the same, for
which see Henry's Quodlibet 1, q. 9; Henry of Ghent 1518, ff. 7"-8r.
40. For Aquinas see Aquinas 1971b, IV.ii, nn. 556,558; for Siger see Siger of Brabant 1948, /nlr.,q. 7,
pp, 16, 18, 20; for James see James of Viterbo 1968, Quodlibet 1, q. 4, p. 46.102-7. On the
Avicennian position see Roland-Gosselin 1948, pp. 150-6; Goichon 1937. But for some recent
attempts to defend the Arabic Avicenna against this interpretation see Rahman 1958, pp. 1-16;
Rahman 1963, pp. 483-6; Morewedge 1972, pp. 425-35. But see also Verbeke 1977, pp. 34*-6*,
76*~9*.
41. For Averroes see Averroes 1562—746 IV, com. 3, f. 67"*. On this see Forest 1956, pp. 142fr
42. See, e.g., Siger of Brabant 1948. Intr. q. 7, sed contra, p. 14 (the last two arguments, with which
Siger apparently agrees); Godfrey of Fontaines 1904—37, Quodlibet 3, q. 1 (v. 2, p. 303) (short
version), 163-4 (long version). Godfrey explicitly attributes this argument to Averroes.
Aquinas' theory of real composition and distinction of essence and existence

Among thirteenth- and fourteenth-century scholastics, a variety of positions
developed with respect to this issue. On the one side there is Aquinas' well-
known theory of real composition and distinction of essence and existence
(esse) in all creatures.43 He appears to defend this position already in his
youthful De ente et essentia.** There in Ch. 4 he introduces this doctrine
within the general context of his refutation of universal hylemorphism.
According to that position, which Aquinas traces back to Avicebron, all
beings other than God, including purely spiritual ones, are composed of
matter and form. While Aquinas denies matter to purely spiritual entities,
he proposes to establish their composite character by postulating another
kind of composition for them, that of essence and esse. And he explicitly
makes the point that because the quiddity or essence of a separate entity
receives its existence from God, essence and existence are related to each
other as potency and act.45
Frequently enough in writings dating from his more mature period
Aquinas returns to and deepens his understanding of essence-ewe com-
position in beings other than God.46 It is clear that for Aquinas existence
43. For some studies by those who do find this position in Aquinas see De Finance i960, pp. 9 4 - i n ;
Fabro 1950, pp. 212—44; Gilson 1952a, pp. 171—8; Gilson 1955, pp. 420—7; Grabmann 1924,
pp. 131-90; Owens 1965, pp. 19-22, Sweeney 1963, pp. 97-131. But for some who deny that
Thomas ever defended this position see Chossat 1939, col. 1180; Chossat 1932, pp. 129.465—
177.513; Cunningham 1962, pp. 279-312; Cunningham 1964, pp. 283-306; Cunningham 1970,
pp. 9-28. As will be clear from what follows, I agree with the former rather than the latter.
44. For references to some recent disputes over the interpretation of De «lie, see Wippel 1979, p. 279,
n. 1. For a fuller discussion of the interpretation of De ente, c. 4, presented here, see Wippel 1979,
pp. 279-95.
45. De ente (1948b) p. 35.10—25. Central to his reasoning is the assumption that whatever is not
included in one's understanding of an essence or quiddity can only come to it from without and
unite in composition with it (p. 34.7—10). But even more important is his claim that there can be
at most one being in which essence and existence are identical. Only after making that claim does
he introduce argumentation for God's existence (pp. 34.15-35.9).
46. See 5CG II, 52, 2nd arg., (ca. 1261 or thereafter); Q«. disp. de spiritualibus cteaturis (1267-8),
Aquinas 1953c, pp. 370b—371a. In contrast with the De ente, in each of these he accepts God's
existence as given (for which he has already argued in SCG I, q. 13). In the latter text he then
reasons that there can only be one being which is its esse, and hence that in every other being, since
it is not its esse, it must have an esse that is received in something (essence) whereby it is limited. He
then applies his metaphysics of participation: ' . . . et sic in quolibet creato aliud est natura rei quae
participât esse, et aliud ipsum esse participatum', and then correlates the participated esse and the
participating nature in any such being as act and potency. For another text where he reasons from
the factual existence of God as the only being in which essence and esse are identical to their
distinction in others see his Tractatus de substantias separatis (1271 -3), Aquinas 1963, p. 79. For the
increasing importance in Thomas' mature writings of arguments based on the participated
character of beings other than God see Fabro 1950, pp. 222-44. Central to this reasoning is
Thomas' conviction that: 'Quandocumque autem aliquid praedicatur de altero per partici-
pationem, oportet ibi aliquid esse praeter id quod participatur.' See Aquinas 1956, Quodl. 2, q. 2,
(esse) is not to be regarded as a thing (res) or as another kind of essence, or as

a predicamental accident that would be superadded to essence; existence
has no quidditative content in addition to that of the essence which it
actualises.47 Interpreters of Aquinas sometimes express this by saying that
existence is a principle of being rather than a being in itself, that by reason
of which an entity actually exists. Essence is another ontological principle
of being, that by reason of which an entity is what it is and enjoys
quidditative content. But neither essence nor existence can exist indepen
dently from the other. 4 8
In his De potentia Aquinas describes esse as the 'actuality of all acts and the
perfection of all perfections'.49 It is not, then, a principle that simply
accounts for a given being's existing. It is taken to account for this, to be
sure, but it is also the ultimate intrinsic ontological principle of perfection
in any existing entity. 5 0 And if existence actualises a distinct essence
principle in every finite being, essence receives and limits existence.51
This same doctrine plays a central role in Aquinas' metaphysics of
participation and in his solution to the classical problem of'the one and the
many'. 52 A finite being may be said to participate in being because of its
a. ι (p. 24b). There he immediately concludes from this that in every creature: '... est aliud ipsa
creatura quae habet esse, et ipsum esse eius', and cites Boethius' De Hebdomadibus.
47. See his critique of Avicenna in Aquinas 1971b, IV. ii, n. 558 (p. 155b). See also Aquinas 1956,
Quodl. 12, q. 5, a. 1: '... dico quod esse substantiate rei non est accidens, sed actualitas cuiuslibet
formae existentis, sive sine materia sive cum materia*. But he will admit that one can refer to
existence (esse) as an 'accident' if one simply means thereby that it is not a part of the essence:
'... accidens dicitur large omne quod non est pars essentiae' (p. 227a). For the same see Quodl. 2,
q. 2, a. I, and ad 2 (p. 24b).
48. See Gilson 1952a, pp. 172-6; De Raeymaeker 1954, pp. 106-7. See ST, I, q. 50, a. 2, ad 3: 'ipsum
autem esse est quo substantia est, sicut cursus est quo currens currit' (where Thomas is again
commenting on the Boethian couplet esse and quod est).
49. Aquinas 1953b, q. 7 a. 2, ad 9 (p. 192b): '... hoc quod dico esse est inter omnia perfectissimum...
Unde patet quod hoc quod dico esse est actualitas omnium actuum, et propter hoc est perfectio
omnium perfectionum.' For parallels see ST, I, q. 3, a. 4; I, q. 4, a. 1, ad 3.
50. See Gilson 1952, pp. 172-86; De Finance i960, pp. 111-19.
51. Aquinas 1954c, c. 18 (p. 18b): Nullus enim actus invcnitur finiri nisi per potentiam, quae est vis
receptiva.' As in the present context, Thomas frequently appeals to this notion in order to
establish divine infinity. See, for instance, De spirituatibus creaturis,q. i.art i;ST,I,q. 7, a. ic; In I
Sent., d. 8, q. 2, a. 1. In the following text he appeals to it to establish essence-;»; composition in
creatures: 'Practerea, omnis creatura habet esse finit urn. Sed esse non reccptum in aliquo, non est
fini turn, immo absolutum. Ergo omnis creatura habet esse receptum in aliqua; et ita oportet quod
habeat duo ad minus, scilicet esse, et id quod esse recipit' In I Sent., d. 8, q. 5, a. 1. Though this
appears in the sed contra, it reflects Thomas' own thinking.
52. Recognition of the major role played by Thomas' doctrine of participation in his overall
metaphysics is one of the major achievements of twentieth-century Thomistic scholarship. Here
greatest credit must be given to the groundbreaking studies in Fabro 1950 (ist ed., 1939), i960,
1961; and in Geiger 1953 (ist ed., 1942). See also Little 1949 (to be used with caution); De Finance
i960, pp. 120-49; Clarke 1952a, 1952b.
essence—existence composition. As Aquinas frequently phrases it, no such

being is identical with esse; it simply has or participates in esse.53 In every
such entity essence receives and limits eise and therefore accounts for its
presence therein to this degree rather than any other. In such beings,
therefore, the nature or essence that participates in esse must be distinct
from the esse in which it participates.54 In other words, real composition
and distinction of essence and existence in any particular being is necessary
if one is to account for its limitation, according to Aquinas.55 He also
maintains that any participated being is efficiently caused by God, pure and
unparticipated esse.56
Giles of Rome on the real distinction
Shortly after the death of Aquinas in 1274, Giles of Rome emerged as a

leading advocate of real distinction between essence and existence in
creatures. By 1276 he appears to have been already identified as a defender
of this position, and soon afterwards he devoted a systematic treatise to it,
his Theoremata de esse et essentia.51 Shortly after his return from his 'exile'
from the theology faculty at Paris in 1285, he defended it again at great
53. Sec, for instance, Aquinas 1956, Quodl, 2. q. 2, a. 1 (p. 24a):'... ens . . . de qualibet autem creatura
praedicatur per participationem: nulla enim creatura'est suum esse, sed est habens esse.' See SCG I,
22; II, 52; Aquinas 1953b, q. 3, a. 5: 'Unde oportet quod ab uno illo ente omnia alia sint,
quaecumque non sunt suum esse, sed habent esse per modum participationis' (p. 49b).
54. See Aquinas 1956, Quodl, 2, a. 1 ; Aquinas 1953c, a. 1 (p. 371a): 'ct sic in quolibet creato aliud est
natura rei quae participât esse, et aliud ipsum esse participatum'.
55. Here Fabro and Nicolas appear to be correct in their criticisms of Geiger for having assigned
priority to 'participation by similitude' in his account of Aquinas, and for having denied that
Thomas appeals to composition of essence and existence in order to account for limitation and
multiplicity of finite beings. See Fabro 1950, pp. 20-2; Fabro I961, pp. 63-73; Nicolas I948.
pp. 561-4-
56. Seeforinstance.ST,I,q.6i,a. 1: 'Omne autem quod est per participationem, causatur ab eo quod
est per cssentiam.' For many texts and fuller discussion of this sec Fabro 1061, passim.
57. It seems clear that Giles was Henry of Ghent's target in his attack against the real distinction in his
Quodlibet 1, q. 9 of 1276. Giles may even have participated personally in this Quodlibetal debate.
SeeHocedez 1928, pp. 100-1,104; Paulus 1938, pp. 280-2. For early formulations of this doctrine
by Giles see Giles of Rome 1521, d. 8, p. 2,pr. i,q. 2, f. 52"; pr. 2, q. I, ff. 53 vb -54' b . On this see
Nash 1950, pp. 66-8; Nash 1950-1, pp. 15—18; Suârez 1948, pp. 94.96; Pattin 1953, p. 84*, Trapé
1964, pp. 330-7. For the same in his Theoremata de corpore Christi sec Giles of Rome 1481,
ff. ι Ι9Λ—120". Note in particular: 'Ex his clare patet quod esse accidit cuilibet creaturae et dicit
aliquid receptum in natura cuiuslibet creati (et dicit additum cuiuslibet entis creati: omitted in the
Venice 1502. ed., cited by Pattin 1953, p. 87*) et facit realem différentiam in rebus creatis'
(f. U9 ,b ). On this doctrine in this work see Pattin 1953, pp. 85*~7*; Suârez 1948, pp. 96-7;
Trapé 1969, pp. 452-3. Both of these works appear to date ca. 1275-6, and before Henry's
Quodlibet 1 of Advent, 1276. See Paulus 1938, p.281; Trapé 1969, p. 455; Pattin 1953, pp. 82*,
85*. For other studies of Giles' general position see Trapé 1966, pp. 49-86; 1967, pp. 170-93;
1968, pp. 351-68.
length against the criticisms of Henry of Ghent.58 Gdes insists that one
must appeal to the real distinction in order to account for the contin-
gent character of creatures.59 He also draws a close parallel between
matter—form composition and that of essence and existence: existence
actualises essence even as form actualises matter. If substantial change
points to matter-form composition of material beings, the fact of creation
requires essence—existence composition of all creatures.60
A major difficulty in interpreting Giles' doctrine arises from his ref-
erence to the distinction between essence and existence as between thing
(res) and thing (res).61 His intent in using such language continues to be
debated by scholars today, but his terminology is at least unfortunate.62
For in addition to suggesting that a distinctive quidditative content might
be assigned to esse, it leaves him open to charges similar to those raised by
Averroes against Avicenna.63 Although the Thomism or non-Thomism of
58. For the controversy between Giles and Henry see in addition to Hocedez 1928, and Paulus 1938,
pp. 280-2; Hocedez 1927, pp. 358-84; 1929, pp. 365-86; 1930, pp. (8i)-(84); Paulus 1940-42;
pp. 323-58.
59. Giles' Theoremata de esse et essentia probably dates from 1278-1280. See Suârez 1948, p. 80; Pattin
1953. P· 91*; Paulus 1940-2. p. 328 (between 1278 and 1286); Hocedez 1930, p. (12), who also
places it between 1278 and 1286; Sicmatkowska i960, pp. 4-5, 31, 48 (before 1276). For the
present point see Giles of Rome 1930. th. 19, p. 129; Giles of Rome 1503, q. 9: '... sic creatio facit
scire quod essentia esset (read: est) aliud ab esse quia ex hoc est creatio inquantum essentia acquirit
esse'(f. 21 ™~b). These Quaestiones date from 1285-7 (see Pattin 1953. Ρ· 90*, n. 37). Qq. 9 and 11
are placed in 1286 by Paulus and before Henry's reply in Quodlibet ΙΟ, q. 7 (Christmas, 1286). See
Paulus 1940-2, pp. 328, 334.
60. Giles of Rome 1930, th. 5, pp. 19-20; Giles of Rome 1503. q. 9, ff. 2o' b -2i ta :'Dicemus ergo sicut
generatio facit scire materiam aliud esse a forma, sic creatio facit nos scire essentiam esse aliud ab
esse.' For some differences between matter - form and essence - existence composition see Giles of
Rome 1930, th. 6, pp. 26-30.
61. In the Theoremata see Giles of Rome 1930, th. 16, p. 101; th. 19, p. 127: '... quod esse et essentia
sunt duae res'; th. 19, p. 134: '... sicut materia et quantitas sunt duae res, sic essentia et esse sunt
duae res realiter différentes'; Giles of Rome 1503, q. 9 (f. 20vb): 'Res ergo ipsa quae est esse est in
genere substantiae'; q. 11 (f. 24"b): "...ct per consequens intelligitur quod esse sit alia res ab
essentia'.
62. For some who are highly critical of Giles and charge that he has in effect 'reified' or turned into
things principles of being, that is to say, essence and existence, see Hocedez 1930, pp. (62)-(/>5),
(117); Paulus 1938, pp. 283-4. For even sharper criticism see Nash 1950, pp. 57-91 ; 1957. pp. 103-
15; 1967. pp. 484-5; Carlo 1966, pp. 14-19, 31,66ff, 83. For much more sympathetic treatments
see Suârez 1948, pp. 66-9, 251-4, 262-8, 270-1; Pattin 1953. pp. 90*, i02*-6*; Trapé 1969,
pp. 445fr., 467-8.
63. See, for example, Henry of Ghent, Quodlibet ι, q. 9, f. T; as well as some of Godfrey's arguments
against the theory of real distinction as presented below. See Giles of Rome 1503, q. 9 (f. 2o ,b ), for
his refusal to reduce esse to the status of an accident. Esse rather belongs to the genus of substance,
but only by way of reduction: 'Bene igitur dictum est quod res ipsa quae est esse [est] in genere
substantiae, sicut res ipsa quae est punetus est in genere quantitatis.' See also Giles of Rome 1930,
th. 22, pp. 155-9. While admitting that essence and existence are separable from one another, he
denies that either can ever exist in separation from the other. See, th. 12 (pp. 67-70); th. 5
(pp. 21-2,29); th. 7 (p. 37). On this see Pattin 1953, pp. i04*-5*; Suârez 1948, pp. 252-4,270. See
Giles' position continues to be discussed, it is certain that his terminology is

not that of Aquinas.64 And it was Giles' terminology, not Thomas', that
was to become standard in subsequent discussions of this issue in the late
thirteenth and early fourteenth centuries.
Siger's rejection of real composition and real distinction
In reviewing some major positions on this problem in the early 1270s, Siger
of Brabant presents Aquinas as defending an intermediary position.65 He
then cites from Thomas' Commentary on Metaphysics IV, but acknowl-
edges that while the conclusion is correct, he does not understand Thomas'
way of formulating it. 66 Siger eventually concludes by rejecting any kind
of real composition or distinction of essence and existence in creatures.67
Like Thomas, Siger was concerned with safeguarding the nonsimple or
composite character of purely spiritual creatures such as intelligences, even
though he, too, rejects matter-form composition of such entities.68 Rather
also Giles of Rome 1481, ff. 119 vb -i 20™, and especially: "... videtur enim absurdum quod essentia
et esse dicantur duae essentiae'.
64. See the studies cited above in n. 62.
65. Siger of Brabant 1948, Intr. q. 7 (p. 16.21-4; and p. 16.14-17). On the dating see Van
Steenberghen 1977, p. 218. Siger here presents Thomas' position as intermediary between lhat
defended by Avicenna (and apparently by Albert the Great) on the one hand, and that of
Aristotle, as interpreted by Averroes. See p. 16.17-20. For confirmation of this see Venne-
busch 1966, p. 168, and in the Questiones melaphysice 1res of Siger there edited by him,
pp. 179.133-180.154. On Albert and Avicenna as defending the view that esse is a disposition
superadded to essence see Siger of Brabant 1948, pp. 14-15; Vennebusch 1966, p. 179, In the latter
context Siger explicitly refers to Albert's exposition on the Liber de causis and even comments: 'et
hoc eum dicentem viva voceaudivi'(p. i79.no-ii);for which see Albert 1890-9C, 1,1,8; p. 377.
In brief as he reports Albert's reasoning, because caused beings receive their esse from the first
principle and not of themselves, and have their essence or that which they are of themselves, their
essence and esse are not the same. On this see Maurer 1946, pp. 75-6. On the difficulty of
determining Albert's definitive position on the essence-«.!« relationship see Roland-Gosselin
1948, pp. 172-84; Geyer 1963, p. 9.
66. See Siger of Brabant 1948, Intr. q. 7 (p. 16.21-32); Siger of Brabant 1066, p. 180.139-54. For
Thomas' text sec n. 47 above. According to Siger Thomas seems to be saying that though esse is
added to a thing's essence, it is neither that essence (res) itself, not a part of the essence (such as
matter and form), nor an accident. This appears to lead to the untenable conclusion that esse is
some kind of fourth nature in reality, Siger reasons. For more on Siger's discussion of Thomas'
text see Van Steenberghen 1977, pp. 284-9; Maurer 1946, pp. 76-7.
67. As Siger puts it, essence (res) and existence (ens) signify the same thing or essence and even the
same intelligible content (intentionem), but in different ways: 'unum ... ut est per modum actus ut
hoc quod dico ens, aliud per modum habitus ut res'. Siger of Brabant 1948, Intr. q. 7; p. 17.60-1.
See Siger of Brabant 1966, pp. 180-1.
68. Siger of brabant 1948, Intr. q. 7; p. 13.50-4 (for an argument for the real distinction based on the
fact that certain things apart from God are not composed of matter and form); p. 20.24-5 ('Ista ct
ultima ratio movit fratrem Thomam'). For rhc same argument see Siger of Brabant 1966, p. 177
(arg· 7)·
than conclude to their essence-existence composition, however, he pro-

poses two alternatives, the first of which he does not present as definitive.
According to this first solution, one might maintain that all such entities
fall short of the divine simplicity in that they recede from the actuality of
the first being and approach potentiality.69 In his second solution Siger
appeals to a substance-accident composition of created separate entities.70
In his only recently discovered and edited Commentary on the Liber de
causis, which seems to be his final word on this topic, Siger's thinking shifts
so much that there he at least approaches Thomas' theory of real com-
position of essence and esse in creatures.71
During the decades following Siger's work many others rejected any
kind of real composition and real distinction of essence and existence.
While some of these defended nothing more than a distinction of reason
(a purely mental distinction), others proposed intermediate positions.72
Already in 1286 and continuing thereafter throughout his career as a
Master in the theology faculty at Paris, Godfrey of Fontaines was un-
wavering in his refusal to admit anything more than a distinction of reason
between essence and existence.73 In his discussion of the theory that
defends their real distinction, Godfrey assumes that according to that
position existence (esse) is a thing (res) that is added to essence from
69. Siger of Brabant 1966, q. 7; p. 21. He expresses some doubt as to whether beings other than God
must be composite, and as to whether Aristotle would admit this (p. 182), and stresses the point
that insofar as they are not God, they are not pure act and approach potentiality. But, as he
comments in the first passage: 'Hoc tamen non concludit quod habeant diversas cssentias'
(p. 21.33). Note that only this solution appears in the short version of the same (p. 21.60—3). This
statement again indicates his failure to understand Thomas' position, since it implies that for him
essence and esse are distinct essences!
70. Siger of Brabant 1948, p. 22.50-2; Siger of Brabant 1966, p. 182.229-31. On this sec Van
Steenberghen 1977, pp. 289-91.
71. See Siger 1972b, q. 53, pp. 183-4, and especially p. 184.35-8. On this sec Van Steenberghen 1977,
p. 292: Marlasca 1972, p. 21, n. 20.
72. By 'mental distinction' I here have in mind the scholastics' distinctio secundum rationem, which is
to say, the kind of distinction that is imposed by the mind or intellect without implying any
distinct realities in the thing. As developed by Siger with application to the present issue, it docs
not even seem to imply distinct concepts (intentiones), but only distinct ways of signifying one and
the same thing (modi significandi).
73. For Godfrey's differentiation of the three basic positions known in his time see Quodlibet 2, q. 2
(Les Philosophes Belges 1904 (PB) 2, p. 60). Here he refers to the theory of real distinction as
maintaining that esse is a distinctive res which is the actuality of essence. In other words, he uses
Giles' language. Also see Quodlibet 4, q. 2 (PB 2.235) for another brief description of these same
three opinions, namely, real distinction, intentional distinction (Henry of Ghent), and real
identity with only mental distinction and distinction in terms of their modi intelligendi el
significandi; 'omnino sint idem secundum rem et différant solum secundum rationem et modum
intelligendi et significandi...' For fuller discussion of Godfrey's knowledge of and reaction to the
theory of real distinction see Wippel 1964. For the dates of Godfrey's Quodlibets see Glorieux
1925, pp. 149-68.
without.74 Godfrey is evidently heavily indebted to the terminology of

Giles of Rome for his understanding of the theory of the real distinction.
He follows Averroes' lead in reducing that position to an absurdity.7S
Against the contention that one must appeal to essence-existence com-
position in order to account for the contingent character of creatures,
Godfrey appeals to the notion of creation ex nihilo.16 And against the claim
that one must admit such composition in purely spiritual creatures in order
to distinguish them from God, Godfrey develops more fully the first of the
solutions proposed by Siger, while rejecting the second, the solution
favored by Siger.77 A less perfect essence may be regarded as potential in
comparison with one that is more perfect, and as actual viewed in itself or
in comparison with something else that is even less perfect. Godfrey offers
as his source for this unusual doctrine of'composition' of act and potency
in separate intelligences Proclus' Elementatio theologica (prop. 2): 'That
which participates in the One is both one and not one.' 78 And Godfrey
would safeguard the participated character of any creature not by appeal-
ing to an intrinsic composition of a participating principle (essence) and
74. See Quodlibet 3, q. 1 (PB 2.158): 'esse existéntiae est aliquid differens realiter ab essentia sive etiam
ab esse essentiae'; and in the shorter version of the same (PB 2.301): 'dicunt quidam quod esse et
essentia sunt diversae res...'. Frequently enough in Giles, Godfrey, Henry, and others, the
expressions 'esse exislentiae' (existential being) and 'esse essentiae' (essential being) appear. For all
practical purposes Godfrey takes them as synonyms and easily shifts from 'essentia' to 'esse
essentiae', and from 'existentia' or 'esse' to 'esse exislentiae'.
75. Quodlibet 3, q. ι (PB 2.163-4, for the long version; PB 2.303 for the shorter version (see arg. 3)).
In the shorter version, Godfrey concludes this argument with the observation: 'Relinquitur ergo
quod unumquodque sit ens per se et non per aliquam rem additam...' (italics added). In both the
longer and shorter versions Godfrey explicitly attributes this reasoning to Averroes.
76. Quodlibet 3,q. 1. See PB 2.160 for Godfrey's recognition of the importance of this argument for
defenders (Giles) of the real distinction. For his reply see pp. 166-9, 171—3· In sum, Godfrey
rejects the close parallel drawn by Giles between matter-form composition as entailed by
generation of material entities and alleged essence - existence composition as implied by creation
of any creature.
77. See Quodlibet 3, q. 1, PB 2.159; and p. 306 (shorter version). Also Quodlibet 3, q. 4 where, in the
course of rejecting matter — form composition in angels, he again finds the appeal to substance —
accident composition therein insufficient to account for the composite or nonsimple character of
their essences. It is not by holding that such essences are composed of distinct res (essence and
existence or matter and form) or by holding that the essence unites with something else (an
accident) that one meets this difficulty. See PB 2.186; 309 (shorter version).
78. See Quodlibet 7, q. 7, where Godfrey is attempting to show that there is sufficient potentiality in
angelic entities to allow for their being included in a logical genus, though not in a natural one
(PB 3.354-5). For the point that all such entities are potential insofar as they are not identical with
God, pure actuality, see pp. 355, 357-9. See in particular, p. 360: '... ita etiam in natura angeli,
recedendo ab actualitate primi et accedendo ad potentialitatem simpliciter habet quodammodo
compositionem, non rei, sed rationis ex potentia et actu'. For Proclus see Proclus 1951, p. 265:
'Omne quod participât uno est unum et non unum' (cited by Godfrey, p. 359). Both Godfrey and
Siger (in his first reply) liken the different degrees of being to the different kinds of number. See
Godfrey pp. 359-60; Siger of Brabant 1948, Intr. q. 7, pp. 21-2.
that in which it participates (esse) as both Thomas and Giles had proposed,
but by noting that prior to its actual creation any such being is to be
regarded as having potential being insofar as God has the capacity to bring
it into being. 79
Against Giles' reasoning based upon the possibility of conceiving an
essence as not existing, Godfrey counters that an essence cannot be under
stood as actual unless its existence is also viewed as actual, that is, as actually
existing. If one considers the essence as only potential, then its existence
likewise can be understood only as potential. This follows for Godfrey, of
course, because of his defense of the real identity of essence and existence.
As he sees it, whatever is true of essence is true of existence, and vice versa.80
According to Godfrey, then, essence and existence are related as that which
is signified abstractly is related to the same signified concretely. Just as there
is no real distinction between that which is signified by the abstract term
'light' and the concrete expression 'to give light', so too, there is no real
distinction between that which is signified by the abstract term 'essence'
and the concrete term 'existence' (esse). There is, he concedes, diversity in
the way in which these terms signify. Essence and existence are neither
really nor even 'intentionally' distinct (as Henry would have it), but only
rationally (secundum rationem).*1
William Ockham against the real distinction
William Ockham addresses himself to this same issue in some detail both in
his Summa logicae and in his Quodlibet 2. 82 In each of these discussions he
reacts critically to the theory according to which essence and existence are
really distinct, and presents that theory according to the terminology
introduced by Giles of Rome. In one of his arguments against this position
he reasons that if existence is really distinct from essence, then it must either
be a substance or an accident. Since it cannot be an accident (which for
Ockham would amount to reducing it to a quality or a quantity), it can
79. For Godfrey's presentation of this argument for the real distinction see Q. 3, q. 1 (PB 2.158-9;
302 (shorter version)). For his refutation see pp. 169-71; 305 (shorter version).
80. For Godfrey's presentation of this argument, see Quodlibet 3, q. 1 (PB 2.158 and 302 (shorter
, k
version)). For this in Giles, see Giles of Rome 1503. q. 11 ; f. 24 '~ ; see also q. 9; f. 20"; also Giles
of Rome 1930, th. 12; pp. 67-70. For Godfrey's reply, see pp. 171; 305 (shorter version). Also cf.
Quodlibet 13, q. 3 (PB 5.208-9).
r
81. See Quodlibet 3, q. ι (PB 2.164-5; 303-4 f° the shorter version). Cf. Quodlibet 13, q. 3 (PB
5.207-8).
82. Sec Ockham 1974a ParsIII-2,c. 27, pp. 553-5; Ockham 1491, Quodl. 2,q. 7. On this see Boehner
1958, pp. 388-97; Mcngcs 1952, pp. 102-3; Lcff 1975, pp. 165-6.
only be a substance. But this alternative will not do since every substance is
either matter, or form, or a composite of the two, or a separate entity. 83
Again, if essence and existence are distinct things (res), they will unite either
to constitute something that enjoys essential (per se) unity, or else only
accidental unity. The first suggestion is unacceptable because then, one of
these - existence, presumably - would be actuality and the other poten-
tiality, and to admit this would be to identify one with form and the other
with matter. But to hold that they unite to form an accidental aggregate is
no more satisfactory. For this would again result in reducing one of them -
existence, presumably — to the status of an accident. 84
In another line of attack Ockham reasons that if essence and existence are
distinct things (res), it would not be contradictory for God to preserve the
essence (entitas) of a thing in being without its existence, or its existence
without its essence. 85 In a somewhat similar vein, Ockham insists that one
has no more right to conclude from the previous nonexistence of a given
essence to a real distinction between that essence and its existence than to
conclude to a real distinction between that essence and itself. For if there
was ever a time when that essence did not enjoy existence, it was not then
an essence either, but only nothingness. 86
Ockham concludes, therefore, that 'essence' and 'existence' signify one
and the same thing. Nonetheless, the Latin term 'esse' can be taken either as
a noun or as a verb. When used in the first way ('being'), it signifies the
same thing as essence and even in the same grammatical and logical mode.
When taken in the second way ('to be'), 'esse' signifies as a verb that which
'essence' signifies as a noun. 8 7
83. Ockham 1974a, p. 553.6-12. Note his final comment in this argument: '...nullum istorum
potest dici "esse", si "esse" sit alia res ab entitate rei' (ital. mine). Cf. Siger's criticism of Thomas'
argumentation in n. 69 above. See also Peter John Olivi 1922-6, qu. 8 (v. I, pp. I47ff).
84. Ockham 1974a, p. 553.13—18. This argument seems to move too quickly, in that it assumes that to
correlate essence and existence as potentiality and actuality is to identify them with matter and
form. But Giles of Rome had insisted that the essence—existence composition differs from that of
matter and form in certain ways, granted that both are potency-act compositions. See above,
n. 60. But for a similar refutation of Giles' position, see Godfrey of Fontaines, Quodlibet 3, q. 1
(PB 2.167-9; 304 (shorter version)).
85. Ockham 1974a, p. 5S3.19-21. See Olivi 1922-6, p. 149.
86. Ockham 1974a, p. 554.32-43; Ockham 1491, Quodl. 2, q. 7. Cf. Godfrey, Quodlibet 3, q. 1 (PB
2.171).
87. Ockham 1974a, p. 554.23-32; Ockham 1491, Quodl 2, q. 7. There see in particular: 'Tamen esse
quandoque est nomen, et sic tunc significat omnimode grammaticaliter et logicaliter idem cum
essentia. Aliquando vero est verbum: tunc idem significat verbauter quod essentia significat
nominaliter.'
Henry of Ghent's intermediate position

Quite different from any of the positions considered above is that de
veloped by Henry of Ghent. From the time of his first Quodlibet of 1276 he
had been unyielding in his opposition to Giles of Rome's defense of the real
distinction between essence and existence. But he was not content to
conclude that there is only a distinction of reason between them. 8 8 One of
his major concerns is to account for the possibility of there being meaning
ful knowledge of nonexistent possibles, on the one hand, though not of
merely imaginary entities such as chimeras, on the other. 8 9 In his solution
to this issue he is heavily indebted to Avicenna's doctrine of the threefold
way in which a nature or essence may be viewed: (1) absolutely or simply
in itself; (2) as realised in singular things; (3) as present in the intellect.
Avicenna had admitted that a nature can in fact exist only in the second or
in the third way, even though it may be considered in the first way. 9 0
Henry appears to go somewhat farther in that he assigns a special kind of
being, essential being (esse essentiae), to an essence when it is considered in
this first way. 91
Possible essences, prior to their realisation in individual existents, enjoy
essential being from all eternity insofar as they are objects of God's
knowledge. This essential being provides them with sufficient ontological
consistency in themselves for them to be objects of knowledge prior to
their realisation as individual existents in time. Because any such essential
being is lacking to merely imaginary entities, true science concerning them
is impossible. Creation of individual existents - that is to say, creation as it is
normally understood - requires the added intervention of the divine will.
God, now acting as an efficient cause, communicates actual existence (esse
existentiae) to certain essences in time. 9 2
88. See Henry of Ghent 1518a, Quodl. i,q. 9; v. 1, ff. 6»-8r. On the dates of Henry's Quodlibets see
Glorieux 1925, pp. 87-93 and 177—99; Gomez Caffarcna 1957, pp. 116—33. On this particular text
in Henry see Hocedez 1928, pp. 92-117; Paulus 1940-2, pp. 324-7; Gomez Caffarcna 1958,
pp. 72-3·
89. See, e.g., Henry of Ghent 1518a, Quodl. 3, q. 9, f. όι"; Paulus 1938, p. 124. For a contemporary
who recognises this concern on Henry's part see Godfrey, Quodlibet 2, q. 2 (PB 2.53-9). On this
see Wippel 1974b, pp. 294-8.
90. For Henry see esp. Henry of Ghent 1518a, Quodl. 3,q. 9, (ff. 60*—61'), where he cites Avicenna's
Metaphysica I. c. 6 and V, c. 1. On this in Avicenna see Paulus 1938, pp. 69-74; Gomez Caffarena
'95 8 · PP· 26-7; Hoeres 1965, pp. 122-3.
91. See Henry of Ghent 1518a, Quodl. 3, q. 1 (f. 6i r ): 'Triplicem quidem habet intellectum verum
sicut et très modos habet in esse. Unum enim habet esse naturae extra in rebus; alteram vero habet
esse rationis; tertium vero habet esse essentiae.'
92. Ibid., Quodl. I, q. 9 (f· Th Quodl. 3, q. 9 (ff. 6i'-62 v ); Quodl. 9, q. 2, v. 2, (f. )4S'~H5*)
Henry also tries to determine more precisely the relationship between

essence (esse essentiae) and existence (esse exislentiae) in actually existing
creatures. While existence does not add any new and distinct thing (res) to
essence, it does add to it a new and distinct 'intention', a relation of actual
dependence upon God as its efficient cause. 93 Hence Henry introduces a
new kind of distinction, his 'intentional distinction', which falls between
the real distinction and the mere distinction of reason. In order to illustrate
this he observes that there is only a distinction of reason between a
definition and that which is defined, for instance, between man and
rational animal. Substance and accident, on the other hand, are really
distinct from each other. But the distinction between a genus and its
differentia - for instance, between animal and rational - must be more than
a mere distinction of reason but less than real. Hence, contends Henry,
these are intentionally distinct, and so, too, are essence and existence in any
actually existing creature. 94 Given this, Henry is reluctant to say of any
creature that its essence is its existence; this he would reserve for God
alone. 95
Although Henry's views on essential being and his defence of an inten-
tional distinction between essence and existence were well known in his
own time and immediately thereafter, they do not seem to have gained
wide acceptance. Giles of R o m e and Godfrey of Fontaines, to mention but
two, rejected his intentional distinction as unintelligible. And both
Godfrey and Duns Scotus submitted Henry's notion of essential being to
sharp criticism. 96
fames of Viterbo's intermediate position

Shortly after Henry's retirement from active teaching at the University of
Paris in 1292, James of Viterbo reviewed the three major positions on the
essence-existence question which had been developed by then: Giles'
93. Ibid., Quodl. 1, q. 9 (f. 7 r -7 v ); Quodl. 10, q. 7 (ff. 4i7'-4i8'). Note in particular: 'Ut secundum
hoc esse existcntiae non addat super cssentiam nisi respectum ad cfficientem' (f. 417'). See also
Quodl. 11, q. 3 (f. 44i'-44i»).
94. Ibid., Quodl. 10, q. 7 (ff. 4i7 v -4i8')· For discussion of Henry's intentional distinction and
its application to essence and existence see Paulus 1938, pp. 220-36, 284-91; Hoeres 1965,
pp. 129-50; Gomez Caffarena 1958, pp. 65-92.
95. Ibid., Quodl. i,q. 9 (f. 7'—7V); Quodl. I l . q . 3 (f. 441'-441*). For another succinct presentation of
Henry's overall theory see Henry of Ghent 1520, a. 28, q. 4 (vol. 1, ff. i67 v -i68").
96. For Giles see Giles of Rome 1503, q. 9 (f. 19"), and Paulus 1940-2 for Giles' running controversy
with Henry. For Godfrey's explicit rejection of the intentional distinction see Godfrey of
Fontaines 1973, p. 368: ' . . . quia non est aliud differre ratione et intentione'. For Godfrey's general
critique of Henry's position see Wippel, 1974b, passim. For Scotus see Ord. I, d. 36, q. 1,
nn. 13-17; Scotus 1950-, VI, pp. 276-7. On this see Hoeres 1965, pp. 161-3; Paulus 1938,
pp. Uiff.
theory of real distinction, Henry's intentional distinction, and the view

that defends only a distinction of reason between them (probably known
to James as defended by Godfrey).97 Although James tries to find some
thing good in each of these, he ends by differing with them all. While Giles'
position strikes him as reasonable, he has difficulty with the way in which it
is expressed, and especially with Giles' description of esse as a superadded
actuality which is neither matter, nor form, nor the composite, nor an
accident. With Henry he agrees that a creature is related to God as both its
exemplar and its efficient cause, but doubts that the term 'existence' itself
expresses this relationship, at least when taken in its primary meaning.
Although James' own position seems closer to that which identifies essence
and existence, he finds that solution unacceptable unless it be qualified as he
himself proposes.98
According to James, existence (esse) and essence are related to one
another as the concrete and the abstract. But a concrete term signifies
something more than does its corresponding abstract term. While both
signify a given form, the concrete term also signifies the subject in which
that form is realised, although not with equal immediacy. First and fore
most (primo et principaliter) it signifies the form, and the subject only in a
secondary way. Essence and existence as realised in creatures are one and
the same as regards their primary meaning: that is to say, 'essence' and
'existence' signify the same. But that which is signified by 'existence' in its
secondary meaning really differs from that which is signified by 'essence'.
For while the term 'existence' signifies essence first and foremost, in a
secondary way it also signifies that which must be conjoined with essence
when it is realised in a concrete existing subject.99
Duns Scotus' intermediate position

Duns Scotus also rejects any kind of real distinction or real composition of
100
essence and existence. Like his predecessors, Scotus is concerned with the
97. See James of Viterbo 1968, Quodl. 1, q. 4; pp. 45-7. For confirmation that James has Giles in mind
as the representative of the theory of real distinction see his reference to 'pulcris theorematibus'
(p. 54.383). James' Quodlibet 1 appears to date from 1293. For discussion as to whether it could be
placed in the 1292-3 academic year or only thatof 1293-4 see Wippel 1974a; Ypma 1975. p. 274,
n. 147.
98. James of Viterbo 1968, pp. 54-6.
99. Ibid., pp. 47.120-49.214. For further discussion see pp. 50-6. One might wonder whether James'
position really differs from that of Godfrey, since he, too distinguishes between existence and
essence as between the concrete and the abstract (see n. 81 above). James goes farther, however, in
stating that essence and existence are really distinct when existence is considered in this secondary
way. This Godfrey will not do. See PB 5.208.
100. Op. Oxon. 4, d. n,q. 3, n.46 (Scotus 1891—5, vol. 17, p. 429a); d. 13, q. ι,η. 38 (Scotus 1891—5,
vol. 17, p. 692b); Op. Oxon. 2, d. 16, q. 1, n. 10 (Scotus 1891—5, vol. 13, p. 28a).
4θ6 Metaphysics and epistemology
need to distinguish every creature, including simple intelligences or angels,

from God. But he does not do so by postulating composition of essence and
existence in created spiritual entities. In reasoning that reminds one in part
of Godfrey of Fontaines and in part of Siger of Brabant, he argues that if
such an essence is not composed of distinct things, it is still in some way
both composite and capable of entering into composition with something
else (componibilis). It is composite insofar as it possesses only a given degree
of being and at the same time lacks any greater degree. But he then observes
that this composition of something positive and of privation is not intrinsic
to the thing's essence, a criticism that Godfrey had already directed at
Siger's preferred solution to this difficulty - substance-accident com
position of such intelligences, which composition Scotus also defends. 1 0 1
Although it is clear enough that Scotus rejects real distinction between
essence and existence in creatures, it is more difficult to determine precisely
how he does distinguish between them. Real distinction will not do
because for him this can be realised only between things which are sep
arable in fact, and, according to Scotus, essence and existence are not
separable. 1 0 2 A number of modern commentators have stated that Scotus
here applies the intermediate kind of distinction that is so often associated
with his name, the 'formal distinction'. 1 0 3 But this interpretation has been
challenged by others. With somewhat greater reason it has also been
suggested that Scotus appeals to another intermediate kind of distinction, a
modal distinction, which obtains here between a given essence and its
intrinsic mode, existence. 1 0 4 In the absence of more positive indications by
ιοί. Ord. i,d. 8, pars I, q. 2, n. 32 (Scotus 1950—, vol.4, pp. 165—6). Note in particular:'Componitur
ergo non ex re et re positiva, sed ex re positiva et privatione . . . Nee tamen ista compositio "ex
positivo et privativo" est in essentia rei, quia privatio non est de essentia alicuius positivi.' See also
Lectura i,d. 8, pars i,q. 2,nn. 31,37 (Scotus 1950-, vol. 17, pp. 10,12). For Godfrey see above,
n. 77. For discussion of Scotus' theory of 'composition' see Gilson 1968, pp. 189-98. On
substance-accident composition in angels, see Ord., loc. cit., η. 34 (pp. 166-7), also Lectura, loc.
cit., η. 30 (pp. 9-10). Gilson connects the diversity between Thomas and Scotus on the essence-
existence question with their different understandings of being. Cf. Wolter 1946, pp. 66-71.
On Scotus' doctrine of univocity see especially Barth 1939. pp. 181-206, 277-98, 373-92;
Barth 1953, pp. 89-94; Wolter 1946, pp. 31-57. The literature on analogy in Thomas is vast. For
an excellent study and for much of this literature see Montagnes 1963. As these authorities righdy
indicate, Scotus' primary target in his critique of analogy was the unusual theory developed by
Henry of Ghent.
102. For the point that for Scotus (and Ockham) real distinction implies separability see Wolter 1965b,
p. 46 and n. 3. On the nonseparability of essence and existence for Scotus see Op. Oxon. 2, d. 1,
q. 2, n. 7 (Scotus 1891-5, vol. 11, p. 63).
103. For references to some of these see O'Brien 1964, pp. 62-4. To these Wolter adds Day 1947, p. 63 ;
Weinberg 1964, p. 218. See Wolter 1965b, p. 54, n. 26, and pp. 45-60 for a helpful discussion of
Scotus' formal distinction.
104. See O'Brien 1964, pp. 65-77; also see Hoeres 1965, pp. 171-9; Gilson 1952a, pp. 202, n. 2, 235,
549, n. 2.
Scotus himself, however, this interpretation is also open to question.105

Another way of attempting to convey his thought is the suggestion that for
him the distinction between essence and existence simply expresses the
aspect of an existing thing which is grasped by abstractive cognition on the
one hand (its essence), and by intuitive cognition on the other (its exis-
tence).106 In any event, for him this distinction is something more than a
mere distinction of reason, and something less than a real distinction.
Potentiality and actuality, matter and form

Many of the thinkers discussed in the preceding sections, who did not
accept matter-form composition of purely spiritual entities, were con-
cerned with showing that even in these beings potentiality and com-
position are in some way realised.107 According to those who defended
real distinction of essence and existence, there is a real composition of
potentiality (essence) and actuality (existence) in every actually existing
creature. And insofar as the potency or essence principle receives and limits
the existence or act principle, it came to be referred to by some participants
in the discussion as 'subjective' potency. For those who rejected real
composition and distinction of essence and existence, only what some of
them termed 'objective' potency would be realised in created spiritual
beings, meaning thereby that any such being was a potential object or
terminus of God's creative activity. According to them essence is not a
'subjective' potency, since it does not receive a really distinct existence as its
actuality.108 As we have seen many would account for the act-potency
105. On the modal distinction see Wolter 1946, pp. 24-7. For more on this and for discussion of its
applicability to essence and existence see Wolter 1965, pp. 54-60. (In brief, while the formal
distinction obtains between two formalities, the modal distinction is rather that between a given
formality and its intrinsic mode, as between wisdom and infinite in God, or between being and
finite in a creature. Also see Alluntis-Woltcr 1975, pp. 505-7, 508-9, 518-19.)
106. See Wolter 1965, pp. 58-60.
107. Because the De rerum principio, now attributed to Vital du Four, was formerly thought to be by
Duns Scotus, and because it defends universal hylemorphism. this view was mistakenly ascribed
to Scotus in the past. For James of Viterbo's appeal to substance-accident composition in order to
assure the presence of act-potency even in simple creatures sec James of Viterbo 1968, Quodl. 1,
q. 4 (pp. 57-61). James adjusts this to his own views on essence and existence.
108. For the distinction between 'subjective' and 'objective' potency see Henry of Ghent 1518,
Quodl. 1, q. 9 (f. 6"), where he distinguishes between two ways of understanding participation.
The first (which he here rejects) would view essence as a subject and the existence in which it
participates as received in it after the manner of a form. Here and in his subsequent discussion of
this in Quodl. 10, q. 7 (f. 4i8'-4i8 v ), Henry is refuting Giles of Rome's understanding of essence
as potency vis à vis existence taken as actuality. According to Henry, the essence ofa creature may
be described as a potential object (Objectively potential') insofar as it is an object or terminus of
the divine creative action which produces it from nothing. And existence may be viewed as the
same essence insofar as it is regarded as an actual object and terminus of God's creative activity. He
dubs as a phantastica imaginatio Giles' position according to which essence would be a subject
'composition' of creatures in other ways, whether by appealing to the

union of being and nonbeing (absence or privation of greater being)
therein or by having recourse to substance-accident composition even in
spiritual entities insofar as they are capable of intellection and volition.
Universal hylemorphism
Reference has been made in passing to another tradition according to
which there is matter-form composition in all beings other than God,
including human souls and created separate intelligences or angels. Aqui
nas and other thirteenth-century thinkers traced this doctrine back to the
Fons vitae of Avicebron, but certain defenders of this position attempted
to show that it owed its origin to Augustine. 109 Perhaps best known
among those who espoused this view in the thirteenth century was
Bonaventure, but even before him Roger Bacon had developed it while
teaching in the Arts Faculty at Paris in the 1240s.110 It continued to be
defended by other Franciscan thinkers later in the century such as John
Peckham, William of La Mare, and Richard of Middleton, and early in the
fourteenth century by Gonsalvus of Spain.1 ' 1
This doctrine easily enabled its defenders to hold that all beings apart
from God are composed and that they include potentiality - viz., matter.
It also implies that in man two instances of matter are present, one which is
intrinsic to his soul and with which the form of his soul is inseparably
which receives existence. Also see ff. 4i9 , -420 r . See Giles of Rome 1503, q. 12 (f. 29 ra " Tb ). Giles
concedes the distinction between the potential (possible) taken subjectively and objectively
(teminalive), but insists that the latter presupposes theformer.For Henry's reaction see Quodl. 11,
q. 3 (f. 443"). For Godfrey of Fontaines' use of this distinction in refuting argumentation
(apparently Giles*) for the real distinction based on participation see Quodlibet 3, q. I (PB
2.169—71). See Wippel 1964, pp. 404—5. For Scotus' use of this see Lectura 1, d. 8, pars I, q. 2, nn.
31, 38; Ord., i,d. 8, pars 1, q. 2, n. 33.
109. For Thomas' attribution of this to Avicebron see his De ente, c. 4 (Aquinas 1948b, p. 30); In Π
Sent.,d.},q. i,a. 1; Treatise on Separate Substances, c. 5 (Aquinas 1963, p. 56). On this doctrine in
Avicebron see Forest 1956 (2nd edn.), pp. 109-10. Henry of Ghent also attributes it to Avicebron.
See Henry of Ghent 1518, Quodl. 4, q. 16 (f. 130»). But Thomas of York and then Gonsalvus of
Spain ascribe it to Augustine. See Zavalloni 1951, pp. 442-3 (on Thomas of York); and
Gonsalvus' Quaestiones disputatae ..., q. II (Gonsalvus 1935, p- 221). For modern scholars who
stress Avicebron's influence upon medieval discussions of this see Crowley 1950, pp. 82,90; Van
Steenberghen 1966, pp. 46—7, 150, 245—6, 249. Zavalloni rather tends to stress the Augustinian
influence (p. 422).
110. For a general discussion see Kleineidam 1930; Lottin 1932, pp. 21 -4. For this in Bonaventure see
Gilson 1953 (3rd edn.), pp. 198-201 (in angels), 255-6 (in the human soul); Forest 1956,
pp. 116-19; Quinn 1973. pp. 139-50; Macken 1976. On Bacon see Crowley 1950, pp. 81-91.
m . For Peckham see his Tractatus de anima, John Peckham 1948, pp. 47-8,61-3 (for discussion of his
text). See William of La Mare 1927, pp. 49-52, 118-21. On this in Richard see Hocedez 1925,
pp. 190-9; Sharp 1930, pp. 262-3. See Gonsalvus 1935, q. 11, pp. 204, 213-21.
united, a spiritual matter, and another which is extrinsic to the soul, the
corporeal matter of the body. 112 But Bonaventure and other proponents
of universal hylemorphism also attempt to determine whether the matter
of spiritual and of corporeal entities is essentially the same in kind, or
different. He acknowledges the difficulty of this question and attributes the
diverse answers proposed by others to the different ways in which matter
may be viewed. Thus one might simply consider it as it is in itself and as if it
were devoid of all forms, or one might view it in relationship to the
different kinds of forms that may actualise it. Ultimately, however, he sides
with the view that matter is essentially the same and even one in number in
all created substances.1 x 3 Some decades later Gonsalvus of Spain is still
troubled by this problem but also concludes that it is preferable to hold that
matter as found in spirits and in corporeal entities is the same in kind. 114
Critics of universal hylemorphism

Earlier in the thirteenth century universal hylemorphism had been criti-
cised by William of Auvergne.115 It was later rejected by Albert the Great,
and then by Aquinas and other thinkers discussed above. 116 Central to
Thomas's refutation in his De ente, for instance, is his contention that the
presence of matter within an intelligence is incompatible with the latter's
capacity to perform intellectual operations.1 ' 7 Godfrey of Fontaines main-
tains that if the soul or a separate intelligence were composed of matter and
form, it would then be corruptible.118
Whether one wishes to credit Augustine or Avicebron with having
inspired the medieval doctrine of universal hylemorphism, it is surely not
112. See Bonaventure i882-loo2a, II, d. 17, a 1, q. 2 (vol. 2, pp. 413-15); Quinn 1973, pp. 139-42. For
Bonaventure's defense of matter-form composition in angels sec II, d. 3, pars 1, a. 1, q. 1 (vol. 2,
pp. 89-91).
113. Ibid., II, d. 3, pars 1, a. 1, q. 2 (vol. 2, pp. 94-8). On the numerical sameness of matter in all
substances see loc. cit., q. 3 (pp. 100-1). On this see Quinn 1973, p. 148.
114. Quaestiones disputatae, q. 11 (pp. 204, 219-21).
115. For William see his De universo I.2, cc. 2-12. On this see Roland-Gosselin 1948, pp. 71-4; Forest
'95*. PP· 121-3.
116. For Albert see in particular In II Sent., d. 3., a. 4. On Albert's position see Forest 1956. pp. 123-6;
Kleineidam 1930, pp. 51 -7. On this in Siger see Van Steenberghen 1977, pp. 282-92. See Henry
of Ghent 1518, Quodl 4, q. 16 (ff. îyP-iji*); Paulus 1938, p. 216. See Giles of Rome 1930, th. 19
(pp. 128-9), Quodl. I, q. 8 (pp. 17-19). For James of Viterbo see Quodl. 3, q. 18, and for the same
by implication see Quodl. I, q. 4. For Ockham's rejection of matter—form composition in angels
see his Ord. I, d. 8, q. 1; OT, III, p. 176.1-2. Also see η. 110 above.
117. Aquinas 1948b, pp. 31—2. For more on Thomas' critique of universal hylemorphism see Collins
1947. Ch. 2, pp. 50-74·
118. See Quodl. 3, q. 3 (PB 2.183-4, 3°8 (shorter version)). For his detailed refutation of Gonsalvus of
Spain's theory insofar as it applies to the human soul see Quodl. 15, q. 10 (PB 14.50-56).
an Aristotelian view. In his Metaphysics Aristotle had described matter as

'that which in itself is neither a something nor a quantity nor any of those
other things by which being is determined'.119 This description led many
scholastics such as Albert, Thomas, Siger of Brabant, Giles of Rome, and
Godfrey, to conclude that prime matter is pure potentiality, that is, that it is
completely devoid of any actuality in and of itself and apart from its
corresponding substantial form. 120 To assign any degree of actuality to
matter in itself would, they feared, make of it something substantial in itself
rather than a mere constituent or principle of an existing substance. And if
matter were a substance in itself, then any superadded form or actuality
acquired through change could only be accidental, not substantial.121
Not all of the schoolmen understood the nature of prime matter in that
way. Many assigned some degree of actuality to matter in and of itself, and
some even contended that it could be sustained in existence by God
through his absolute power apart from any substantial form. Versions of
this were defended especially by Franciscans such as John Peckham,
Richard of Middleton, and, somewhat later, Duns Scotus and Ockham;
but the secular Master, Henry of Ghent, also championed this position.122
In sum, it numbered among its supporters not only defenders of universal
hylemorphism but others who had rejected that position, such as Henry,
Duns Scotus, and Ockham.
119. See Metaphysics VII, iii (i029 a 20-2i).

120. Onthisin Albert sec Weisheipl 1965a, pp. i5i-2.AsWeisheiplindicates, Albertdoes allow for an
'incipient actuality' or inchoatioformae in prime matter, something which Thomas rejects. For
Thomas on the purely potential character of matter see Weisheipl 1965, pp. I52ff; Forest 1956,
pp. 210-16. Also see ST, 1, q. 115, a. 1, "d 2; SCG II, 43. For Siger see Van Steenberghen 1977,
pp. 327-8. For Giles see especially Giles of Rome 1503, qu. 8, f. 98vb; also 1930, th. 10 (p. 58). For
Godfrey see Quodl. 1, q. 4 (PB 2.7-9); Quodl. 2, q. 4 (PB 2-83); Quodl. 10, q. 9 (PB 4.336-41).
121. For this sec, for instance, Aquinas 1971b, VIII.1, n. 1689; Godfrey, Quodl. 2, q. 4 (PB 2.83);
Quodl. io, q. 9 (PB 4.337-9).
122. See Sharp 1930, pp. 178-82 (on Peckham), 220-1 (Richard of Middleton); Zavalloni 1951,
pp. 303-9. On Scotus and especially on Ockham see Wolter 1965a, pp. 131-46. Interestingly,
precisely because matter is not merely in objective potency but also in subjective potency (to its
form) Scotus concludes that it must enjoy some degree of actuality (see Wolter 1965a, p. 132; and
Ord. 2, d. 12, qq. 1,2 (Scotus 1891-5, vol. 12, pp. 558, 577)). For this in Henry see Henry of Ghent
1518,1, q. 10 (ff. 8 r -9 v ). On this see Macken 1976a, pp. 107-13. For an excellent introduction to
the related question which was heatedly debated during the closing decades of the thirteenth
century - viz., unicity vs. plurality of substantial form in any given material substance, and
especially in man, see Zavalloni 1951; for further texts see Roos 1977.
20
UNIVERSALS IN THE EARLY
FOURTEENTH CENTURY
Two sorts of questions for moderate realism

The vigorous early-fourteenth-century debate about universals was based
on a rejection of Platonism, the theory that universal natures really exist
independently of the particulars whose natures they are and independently
of every mind. Fourteenth-century 'moderate' realists agreed that natures
must be somehow common to particulars in reality, but Aristotle had
convinced them that no one in his right mind could hold that the nature of
a thing exists separated from it as Platonic forms were supposed to do. 1
They insisted instead that the natures really exist in the things whose
natures they are, as metaphysical constituents of them. But this contention
had its own problems. Since there can be more than one particular in a
given genus or species, natures cannot be the only metaphysical con-
stituents of particulars; there must also be individuating principles that
serve to distinguish one particular from another.
But what are these individuating principles? William of Champeaux's
position that accidental properties individuate2 was denied by virtually
everyone on the Aristotelian ground that substance is naturally prior to
accidents but particular substances are not naturally prior to what in-
dividuates them. Thomas Aquinas held that prime matter, the ultimate
property-bearer in composite substances, combines with quantitative di-
mensions to individuate.3 But Duns Scotus found this tantamount to
conceding that accidents individuate after all.4 Besides, he argued, neither
matter, quantitative dimensions, nor their combination was distinct and
determinate in itself. Taking it as axiomatic that only what is distinct and
determinate in itself can individuate, Scotus concluded that neither matter
by itself, existence, nor any combination of accidents can do the job
1. William Ockham 1967, Ordinatio I, d. 2, q. 4;OTII, p. 117. Cf. Walter Burley iyyj,Superartem
veterem Porphyrii et Aristotelis,ff.5'*-6".
2. Peter Abelard 1919-27, Clossae secundum magistrum Petrum Abaelardum super Porphyrium, p. 13.
3. Aquinas, De ente et essentia, c. 2; ST, I, q. 75, a. 7, c.
4. Ordinatio II. d. j , q . 4 , n . i n ; Duns Scotus 1950-, VII, p. 446.
(Ordinatio II, d. 3, qq. 5, 3, and 4, respectively). Since the numerical unity

that belongs to particulars is more perfect than the specific unity of natures,
and since what is merely potential cannot increase the perfection of what is
actual, Scotus concluded that the individuating principles must be some-
thing positive and have an actuality of their own and not fall themselves
under any of the ten categories. 5
What is the relationship between the nature and the so-called individu-
ating principles? There must be some distinction between them, or else the
nature will not be in any way the same in many particulars. But what sort
of distinction? Further, do the individuating principles individuate the
nature, or can one nature numerically the same, be a member of many
different collections of metaphysical constituents?
I will examine two of the various answers given to the second sort of
questions by moderate realists of the early fourteenth century — viz., those
of Duns Scotus and Walter Burley. Their positions were criticised, as we
shall see, by the father of late fourteenth-century nominalism, William
Ockham, and his older contemporary, Henry of Harclay, whose views
occupy a middle ground.
Duns Scotus' theory of universals

Scotus' theory of universals develops his conviction that the nature must be
somehow common in reafity, even though it cannot exist apart from any
and every particular. He defends the first part of his thesis with 'almost
infinitely many arguments': 6 For example, he holds that 'the unity that is
required to found a relation of similarity is real. And it is not numerical,
since no one single thing is similar or equal to itself.'7 Conversely, real
opposition between numerically distinct things requires real extremes of
opposition. 8 Again, if Socrates is to be more similar to Plato than to a Une,
it must be that some real thing is common to the former two that is not
common to Socrates and a line. 9 Similarly, even 'if no intellect existed, fire
5. Ordinatio II, d. 3, q. 6, nn. 169-70; Duns Scotus 1950-, VII, pp. 474~5-
6. Henry of Harclay, 'Utrum universale significet aliquam rem extra animam, aliam a singulari vel
supposito' in Gâl 1971, sec. 4, p. 186.
7. Ordinatio II, d. 2, p. 1, q. 1, n. 18; Duns Scotus 1950—, VII, p. 398: 'Secundum Philosophum V
Metaphysicae cap. de "Ad aliquid", idem, simile et aequale fundantur super "unum", ita quod licet
similitudo habeat pro fundamento rem de genere qualitatis talis, tamen relatio non est realis nisi
habeat fundamentum reale et rationem proximam fundandi realem; igitur unitas quae requiritur
in fundamento relationis similitudinis, est realis: non est autem unitas numeralis, quia nihil unum
et idem est simile vel aequale sibi ipsi.'
8. Ibid., η. 19; Duns Scotus 1950-, VII, pp. 398-9.
9. Ibid., η. 23; Duns Scotus 1950-, VII, pp. 400-1.
Universals in the early fourteenth century 413
would still produce fire and destroy water. And there would be some real
unity of form between the producer and the product, because of which
unity the production would be univocal...' 1 0
Yet, Scotus argues that if the nature is essentially something that can
really exist in and be common to numerically distinct real things, it cannot
of itself be numerically one. For whatever pertains to a thing of itself,
pertains to it in whatever it is in. 'Therefore, if the nature of stone were of
itself this, the nature of stone, whatever it is in, would be this stone.' "
Likewise, 'if one of a pair of opposites pertains to a thing of itself, the other
of the pair of opposites is incompatible with that thing of itself. Therefore,
if the nature is of itself numerically one, numerical multiplicity is incom
patible with it.' 1 2 But Scotus contends that human nature is numerically
one in Scorates and numerically many in numerically many distinct
particulars, and concludes that there must be some individuating prin
ciples, or thisnesses, that are numerically one and particular of themselves
and that contract the nature, which is common of itself, rendering the
nature numerically one and particular as well. Since the nature is numeri
cally one and particular only through the individual difference, it is said to
be one denominatively as Socrates is said to be white denominatively by
virtue of the inherence of whiteness.13
Just as the nature is not numerically one and particular of itself, so the
nature is not of itself completely universal.14 For as completely universal,
the nature is in fact truly predicable of each and every particular in the
species. But nothing that is essentially predicable of many could be numeri
cally one and particular at all. Nor is the nature completely universal
insofar as it exists in reality, since actually being truly predicable of many is
15
incompatible with being numerically one and particular. Rather, the
nature is completely universal only insofar as it exists in the intellect as an
object of thought.
10. Ibid., η. 28; Duns Scotus 1950-, VII, pp. 401-2: 'Nullo cxsistcnte intellectu ignis gcneraret ignem
et corrumperet aquam, et aliqua unitas realis esset "generantis ad genitum" secundum formam,
propter quam cssct gcncratio univoca. Intellectus enim considcrans non facit gencrationem
esse univocam, sed cognoscit earn esse univocam.'
11. Ibid., η. 3; Duns Scotus 1950, VII, p. 392: 'Quidquid incst alicui ex ratione sua per se, inest ei in
quocumque; igitur si natura lapidis de se essent "haec", in quocumque esset natura lapidis,
natura ilia esset "hic lapis".'
12. Ibid., η. 4; Duns Scotus 1950-, Vil. p. 393: 'Praeterea, illi cui de se convenit unum oppositum, ei
de se répugnât aliud oppositum; igitur si natura de sc sit una numéro, répugnât ci multitudo
numeralis.'
13. Ordinatio II, d. 3, p. 1, q. 6, nn. 172-5; Duns Scotus 1950-, VII, pp. 476-8.
14. Ordinatio II, d. 3, p. 1, q. 1, nn. 33—4; Duns Scotus 1950—, VII, pp. 403—5.
15. Ibid., n. 37; Duns Scotus 1950-, Vil, pp. 406-7, n. 38; pp. 407-8.
So far, then, Scotus' view could be summarised in the following five

theses: (Ti) the nature is c o m m o n of itself and is also common in reality;
(T2) the principle of individuation or contracting difference is numerically
one and particular of itself and cannot be common to numerically distinct
particulars; (T3) both the nature and the contracting difference exist in
reality as constituents of a particular and they can exist in reality only as
such; (T4) as a result of combination with the contracting difference, the
nature is numerically one denominatively and is numerically many in
numerically distinct particulars; and (T5) the nature is completely universal
only insofar as it exists in the intellect.
The most striking feature of Scotus' theory of universals, however, is his
contention that the common nature and individuating principles are nei
ther really distinct — in the sense of being distinct real things - nor distinct
only in reason - in the sense of being thought of by means of distinct
concepts. Rather Scotus holds that
(T6) The nature and contracting difference are formally distinct, or not formally
the same. 16
But Scotus' works contain at least two importantly different accounts of

this alternative sort of non-identity or distinction, which he employs so
often in his philosophy and theology.
(1.1) The first and earlier version stipulates that within what is really one
and the same thing (res) there often is a plurality of entities or property-
bearers whose non-identity or distinction in no way depends upon the
activity of any intellect, created or divine. Scotus had two motives for
making this stipulation. (1.1.1 ) The first is epistemological. Scotus thought,
as Ockham was later to think, that if in reality, prior to every act of
intellect, χ and y are in every way the same, the intellect could not make χ
and y distinct. But, contrary to Ockham, he did not see how distinct
concepts could genuinely signify a real thing without there being some
non-identity or distinction in the thing corresponding to the distinction in
conceived objects. Since it sometimes happens that, prior to every act of
intellect, what is really one and the same thing is simultaneously apt to fall
under or be signified by distinct concepts - e.g., a species falls under
concepts of genus and differentia simultaneously - Scotus concluded that
prior to every act of intellect there must be some sort of non-identity or
16. Ordinatio II, d. 3, p. 1, q. 6, nn. 187-8; Duns Scotus 1950-, VII, pp. 483-4.
distinction among entities within the real thing. 1 7 (1.1.2) Scotus' other
motive is logical or metaphysical. He realized that often in philosophy and
theology there is reason to deny that χ and y are really distinct things (res)
and yet apparent cause to affirm that χ is F and y is not F. But according to
the principle of the Indiscernibility of Identicals, which clearly applies to
everything that exists in reality, nothing that is in every way the same can
be both F and not F at once. Distinguishing non-identical or distinct
property-bearers within what is really one and the same thing might seem
to open the way for a solution to such problems.
Scotus labels the non-identical entities within what is really one and the
same thing (res) 'realities' (realitates), 'formalities' (formalitates), 'aspects'
(rationes), 'formal aspects' (rationes formales), 'intentions' (intentiones), or
'real aspects* (rationes reales). Likewise, because he claims that such entities
are not formally the same, or are formally distinct, formally different, or
formally diverse, the relation between them is best known as formal non-
identity or distinction.
Scotus restricts the relations of formal identity and distinction to entities
that are or are in what is really one and the same thing and understands that
(A) χ and y are formally distinct or not formally the same, if and only if (a) χ and y
are or are in what is really one and the same thing (res); and (b) if ac and y are
capable of definition (in the strict Aristotelian sense, in terms of genus and
differentia), the definition of χ does not include y and the definition of y does
not include x; and (c) if χ and γ are not capable of definition, then if they were
capable of definition, the definition of χ would not include γ and the definition
of y would not include x.18
Thus, Scotus' first account of formal non-identity or distinction presup

poses an ontology that begins with formalities or realities that can have a
double mode of existence: they can exist in reality as constituents of real
things or they can have a non-real mode of existence in the intellect as
objects of thought or concepts. Some properties belong to formalities of
themselves - viz., their essential properties; others belong to them only
insofar as they have one mode of existence or the other. Accordingly,
Scotus says that man is of itself rational, mortal, and animal; and that man
of itself is neither universal nor particular, but indifferent to each. Man is
17. Quaestiones subtilissimae super libros Metaphysicorum Aristotelis VII, q. 19, n. 5; Duns Scotus 1639c,
p. 727.
18. Lectura I, d. 2, p. 2, q. 1-4, n. 275; Duns Scotus 1950-, XVI, p. 216. Ordinatio I, d. 2, p. 2, q. 1-4,
n, 403; Duns Scotus 1950-, II, pp. 356-7.
4i6 Metaphysics and epistemology
completely universal only insofar as it exists in the intellect, and numeri

cally one and particular only insofar as it exists in reality.
Some of Scotus' Parisian opponents charged that his application of
formal identity and distinction to God compromised divine simplicity.
Perhaps it was for this reason that Scotus took a different position in the
Reportata Parisiensia I, d. 33, q. 2-3 and d. 34, q. 1. " In these works he still
argues for the existence of some distinction in reality (ex natura rei) and
prior to every act of intellect, but alternative to the real distinction between
one thing and another. Yet he now denies that this involves distinguishing
a plurality of property-bearers within what is really one and the same
thing. Scotus elaborates this idea by contrasting absolute distinction (dis-
tinctio simpliciter) with distinction secundum quid. Distinct real things such as
Socrates and Brownie the donkey are absolutely distinct. But a distinction
between χ and y may be secundum quid for one of two reasons: it may be that
the being of χ and y is somehow diminished (as it is when a: and y have only
a non-real mode of existence as objects of thought, or when χ and y exist
only virtually in their causes, or when χ and y are elements in a mixture); or
it may be that although χ and y are both fully real and actual, they are
not absolutely non-identical, but only non-identical secundum quid.20
According to Scotus, χ and y are non-identical secundum quid if and only if
they lack either formal identity or adequate identity. Criterion (A) remains
the criterion of the former, but it is no longer taken to signal the existence
of any pluarlity of property-bearers within a single thing. Scotus says that χ
and y lack adequate identity when one of them exceeds the other according
to predication or according to perfection. For example, animal is not
adequately identical with man, because the former is predicated of more
things than the latter is, while the latter is more perfect than the former is
(presumably because the genus is in potentiality with respect to the speci
21
fic difference, but the species is not). And Scotus insists that the
Indiscernibility of Identicals does not necessarily hold where χ and y are
19. Cf. Gelber 1974, pp. Soff. For some reactions to her interpretation and to the interpretation in
Henry 1965, and Henry 1972, pp. 88-95. see Adams 1976.
20. Reportatio Parisiensis I, d. 33, q. 2; MS Civitas Vatican.! bibl. apost., cod. Borgh. lat. 325,
ff. tt2yb-%}n: Quaestionesmiscellaneaedeformalitatibus, q. i.n. 12: Duns Scotus 1639b, p. 444. Note
that the Scotus commission has now identified the Reportato Parisiensia printed in Volume xi of
Duns Scotus 1639a as the Additiones magnae written by Scotus' pupil William of Alnwick (cf.
Modric 1978, p. 83). While Alnwick had the intention of representing Scotus' teaching at Paris,
he also drew heavily on Scotus' Oxford lectures and has heavily edited the text in places. I refer,
therefore, to a manuscript of Reportatio I A, which the Scotus commission regards ;is the Reportatio
examinata which was revised by Scotus himself.
21. Reportatio I A, d. 33, q. 2; MS Vat. Borgh. 325, f. 83™.
really but not formally the same nor do the Transitivity and Symmetry of
Identity hold where χ and y are really but not adequately the same. 2 2
Ockham's attack on Scotus' formal distinction

Ockham attacked Scotus' theory on two independent grounds: first that
the notion of formal non-identity or distinction presupposed by (T6) leads
to ontological paradox or contradiction; and second, that in any case the
conjunction of (Ti)-fT4) and (T6) is contradictory.
Ockham charges that no creatures can be formally distinct without
being really distinct.
. . . I argue this way: (i) Wherever there is any distinction or non-identity [between
beings], there some contradictories can be truly asserted of those [beings], (ii) But it
is impossible that contradictories should be truly asserted of any [beings] unless
they or those for which they supposit are (a) distinct real things or (b) distinct
concepts or beings of reason or (c) a thing and a concept. But if (iii) they all exist in
reality, (iv) they are not distinct concepts, (v) Nor are they a thing and a concept.
Therefore, (vi) they will be distinct things. 23
Premiss (i) is trivially true, and (iii) is maintained by Scotus. The crucial
premiss (ii) combines an assertion of the Indiscernibility of Identicals, with
the assumption that (a)-(c) constitutes an exhaustive list of the alternatives
under which non-identity or distinction might obtain. But Scotus would
insist on adding a fourth alternative: (d) χ is formally distinct from y; or
(e) χ and y are distinct secundum quid.
The list of (a)—(c) reflects Ockham's ontology, according to which real
things and beings of reason or concepts are the only beings there are, and
there are no real beings that are not real things. But Scotus' first account of
formal distinction is premissed on a wider ontology that allows for distinct
formalities within one and the same real thing. He can, therefore, say that
propositions of the forms 'x is F' and 'y is not F' are sometimes true about
such mind-independent formalities.
Ockham's complaint is that the ontology that lies behind (d) leaves us
with no resources for proving a real distinction among really existent
22. Ibid., MS Vat. Borgh. 325, f. 83 Λ .

23. Ockham 1970, Ordinatio I, d. 2, q. 1; OTII, p. 14: "Contra istam opinionem arguo per unum
argumentum quod est aequaliter contra distinctionem vel non-identi tatem formalem ubicumque
ponatur. Et arguo sic: ubicumque est aliqua distinctio vel non-identitas, ibi possunt aliqua
contradictoria de illis vcrifican; sed impossibile est contradictoria verificari de quibuscumque nisi
ilia vel ilia pro quibus supponunt sint disti nctae res vel distinctae rationes sive entia rationis vel res
et ratio; sed si omnia ilia sint ex natura rei, non sunt distinctae rationes nee res et ratio; igitur erunt
distinctae res.' Cf. ibid., à. 2, q. 6; OT H, p. 173.
beings. He reasons that since all contradictories are equally contradictory, if

some propositions of the forms 'x is F' and 'y is not F ' can be true about
really existent property-bearers that are only formally distinct, then any
pair of contradictories can. For example, if 'Human nature is formally
human nature and Socrateity is not formally human nature' entails only
'Human nature and Socrateity are formally distinct', what reason is there
to suppose that Ά man is rational and a donkey is not rational' entails Ά
man and a donkey are really distinct' rather than Ά man and a donkey are
formally distinct'? Ockham rejects the suggestion that a proposition of the
form 'x is F and y is not F ' entails 'x and y are really distinct' while a
proposition of the form 'x is formally F and y is not formally F' entails 'x
and y are formally distinct'. For since 'formally F ' and 'not formally F ' are
just as contradictory as ' F ' and 'not F ' , there is no more reason why the
latter should entail a real distinction than the former. 2 4
Scotus' account of distinction secundum quid abandons the ontology that
assumes a plurality of property-bearers within one and the same thing, but
continues to insist that propositions of the forms 'x is F ' and 'y is not F '
might be simultaneously true even though χ and γ are distinct only
secundum quid, i.e., where χ and y are really the same but lack formal or
adequate identity. But, Ockham observes, when Scotus says that χ and y
lack formal identity if and only if neither falls (or would fall) under the
definition of the other, he is talking about real, not nominal, definitions.
And while Ockham allows - contrary to Scotus's epistemological argument
— that what is in every way the same in reality may simultaneously fall
under distinct concepts with distinct nominal definitions, he insists that it is
impossible that any χ and y should be in every way the same in reality and
have distinct real definitions; for there would be only one reality there to be
defined. 2 5 Hence, it is impossible for any χ and y to be really the same and
lack formal identity.
Again, for χ and y to lack adequate identity is for χ and y to be really the
same and yet for one to exceed the other in predication or perfection. For
example, human nature and the contracting difference Socrateity would
lack adequate identity; for while they are really the same, man is predicable
of more than Socrateity is. This is tantamount to saying that χ and y lack
adequate identity where χ and y are really the same, but for some F, rv is F
and y is not F. Where Scotus abandons the ontology according to which χ
24- Ordinatio I, d. 2, q. i; Ockham 1970, OT II, pp. 16-17.

25. Ordinatio I, d. 2, q. 1; Ockham 1970, OT II, pp. 30-1.
and y are distinct formalities within the same real thing, he is left asserting
contradictories about what is in reality one and the same thing. And
Ockham will again object that to relinquish the Indiscernibility of
Identicals as a criterion of real distinction is to lose every way of proving
distinction among real things.
It would be obviously circular to reply that Scotus never endorses
contradictories about what is in every way the same in reality, but only about
those that lack formal or adequate identity. For the lack of formal or
adequate identity itself partially consists in the assertion of contradictories
about what is really one and the same thing, and so it cannot be invoked to
explain how those very predications are possible.
Ockham's claim that Scotus' theses are mutually inconsistent

Granting the formal distinction for the sake of argument, Ockham tries
with varying success to show that Scotus' theory of universals is
contradictory.
First, Ockham maintains that (Ti) is incompatible with (T3) because
together they entail the absurdity that something common is a constituent
of a particular. Following Aristotle, Ockham insists that necessarily every
particular is homogeneously particular in the sense of having no con-
stituents that are not particular. 26 Scotus could reply, however, that he does
not claim that any particulars have constituents that are not particular. For
while (Ti) asserts that the nature is common of itself and in reality, (T4)
says that it is numerically one and particular denominatively, by virtue of
its combination with the individuating principles.
Ockham would object, however, that the conjunction of (Ti) and (T4)
entails the absurdity 'that there are as many species as there are individuals'.
For by (Ti) this commonness or unity less than numerical unity pertains to
the nature itself and so will accompany it wherever it exists. But by (T4)
human nature existing in reality in Socrates is numerically distinct from
human nature existing in reality in Plato. Hence 'there are two common
entities in Socrates and Plato, and consequently two species'. 27 Scotus
26. Ordinatio I, d. 2, q. s; Ockham 1970, OT II, pp. 158-9.
27. Ordinatio I, d. 2, q. 6; Ockham 1970, OT II, p. 181: 'Confirmarur. quia ad multiplicationem
subiecti proximi sequitur multiplicatio passionis; sed secundum istum ista unitas minor est passio
naturae; igitur sicut natura realiter multiplicatur, ita passio - cum sit realis - realiter multiplica-
bitur. Et per consequens sicut realiter sunt duae naturae in Sorte et Platone, ita erunt realiter duae
unitates minores: sed ista unitas minor vel est communitas vel inseparabilis a communitate, et per
consequens inseparabilis a communi; igitur sunt duo communia in Sorte et Platone, et per
consequens duae species. Et per consequens Sortes esset sub uno communi et Plato sub alio, et ita
tot essent communia — etiam generalissima — quot sunt individua, quae videntur absurda.'
would have to grant the first of Ockham's conclusions, but not the second.
For he holds (T5) that the nature is incompletely universal in reality and
completely universal only insofar as it exists in the intellect as an abstract
general concept predicable of many. And the nature counts as a species
only insofar as it is completely universal.
Ockham sees Scotus' theory of universals as running counter to a
fundamental axiom of identity:
(B) Nothing is individuated through anything extrinsic to it; rather, being iden-
tical with itself and being distinct from everything else are properties that a
thing has in and of itself.
A thing can be white because a really distinct quality of whiteness inheres in
it or thought of because a really distinct intellect attends to it, but being the
same as itself or distinct from others are properties that a thing has of itself
or by virtue of something intrinsic to it. 28 It follows from (T4) and (T6),
however, that the nature is individuated and numerically multiplied by
combining with individuating principles formally distinct from it, so that
neither is an essential constituent of the other. Scotus would doubtless have
agreed that the nature cannot be individuated by anything really distinct
from it, but would have rejected (B) and replaced it with his own axiom:
(C) Only what is distinct and determinate of itself can individuate something
formally distinct from it.29
Nothing in this argument of Ockham's would compel Scotus to do
otherwise.
Ockham's fourth argument - that (Ti) is inconsistent with the conjunc-
tion of (T3) and (T4) — goes to the heart of Scotus' theory of universals in
challenging the intelhgibihty of his contention that the nature is common
or incompletely universal in reality. As noted above, Scotus argues that the
nature is not 'this' of itself on the basis of the principle that
(D) 'If one of a pair of opposites pertains to something of itself, the other of the
pair of opposites is incompatible with that thing of itself...'
Ockham insists that Scotus should equally grant the following:
(E) ' . . . Whenever one of a pair of opposites really pertains to something in such a
way that that thing is truly and really denominated from it, whether they
pertain to it of itself or through something else — this fact remaining un-

29. Quaestiones in Metaphysicam Aristotelis VII, q. 13; Duns Scotus 1639a, VII, pp. 417-18.
changed - the other of the opposites will not really pertain to it, but will be
absolutely denied of i t . . . ' i 0
But, by (T3), the nature does not exist otherwise than together with some con
tracting difference or other; and, by (T4), the nature is numerically one and
particular as a result ofits combination with a contracting difference. Thus,
by (E), no nature can simultaneously be common, as (Ti) asserts it to be.
Ockham notes the objection that 'the two unities are not really op
posed'.31 But this seems inconsistent with Scotus' own claims. For he says
that 'Multiplicity that cannot stand together with the greater unity, be
cause it is opposed to it, can without contradiction stand together with the
lesser unity.' 3 2 And he has appealed to this fact to justify regarding
'commonness' and 'numerical unity' or 'particularity' as opposites and
proceeding to infer (by (D)) that if the nature is common of itself, it cannot
be numerically one and particular of itself. Ockham argues analogously on
the basis of (E) that if the nature is truly numerically one and particular -
whether of itself or only denominatively - it cannot also be common, and
vice versa.33 Alternatively, a defender of Scotus might say that while (D)
holds good- i.e., that if* of itself is F, not-Fis incompatible with χ of itself
- it is not the case that if χ is F - no matter whether of itself, per se, or
denominatively — not-F is incompatible with x. For a man is white
denominatively, and a white is white per se. But ' . . . blackness goes toge
ther with man and does not go together with what is white, and yet a man is
white, and α is a man and white'. 3 4 Likewise, being Plato's may be
compatible with the nature, even though being Plato's is not compatible
with being Socrates'; and numerical multiplicity may be compatible with
the nature of itself, even though numerical multiplicity is not compatible
with the particular.
30. Ordinatio I, d. 2, q. 6; Ockham 1970, ΟΓΙ1, p. 177: 'Primo sic: quandocumque convenu alicui
realiter unum oppositorum, ita quod vere et realiter denominatur ab illo, sive conveniat sibi ex se
sive per aliud, — hoc stante et non mutato —, reliquum oppositorum sibi non conveniet realiter,
immo simpliciter ab eo negabitur. Sed per te omnis res extra animam est realiter singularis et una
numéro, quamvis aliqua de sc sit singularis et aliqua tantum per aliquid additum; igitur nulla res
extra animam est realiter communis nee una unitate opposita unitati singularitatis, igitur realiter
non est aliqua unitas nisi unitas singularitatis.'
31. Ordinatio I, d. 2, q. 6; Ockham 1970, OT II, p. 177: "Si dicatur quod istae duae imitates non sunt
oppositae realiter, et eodem modo singularitas et communitas non opponuntur realiter.'
32. Ordinatio II, d. 3, p. 1, q. 1, n. 9; Duns Scotus 1950-, VII, p. 395: 'quia cum unitate minore sine
contradictione potest stare multitudo opposita maiori unitati, quae multitudo non potest stare
cum unitate maiore, quia sibi répugnât...'
33. Ordinatio I, d. 2, q. 6; Ockham 1970, OT II, p. 178.
34. Ibid., p. 178: 'Si dicatur quod ista forma arguendi non valet, quia cum homine stat nigredo et cum
albo non stat nigredo, et tamen homo est albus, et a est homo, ec est album.'
Ockham thinks that this response is likewise mistaken. For 'to be

compatible* can be taken actually or potentially. If Socrates is white,
blackness is not actually compatible with Socrates, because 'Socrates is
white and Socrates is black' is a contradiction. But blackness is potentially
compatible with Socrates, because even if Socrates is in fact white, it is
logically possible that whiteness not inhere in him and that blackness inhere
in him instead. Likewise, if humanity in Socrates is Socrates' denomina-
tively, it is not actually, but at most potentially compatible with Plato's
contracting difference. And if humanity in Socrates is numerically one and
particular, it is not actually, but at most potentially compatible with
numerical multiplicity.
Scotus might reply, however, that this is exactly what he was claiming
when he said that the nature in reality is common and incompletely un-
iversal. Just as Socrates is not black while he is white, so humanity is not
Plato's while it is Socrates' or numerically many while it is numerically
one. Nevertheless, just as it is possible that Socrates should exist without
whiteness inhering in him and have blackness inhering in him instead; so it
is possible that the humanity in Socrates should not be combined with
Socrates' contracting difference and possible that it should be combined
with Plato's contracting difference instead; and possible that humanity
should not be numerically one but should be numerically many instead.
Ockham also thinks Scotus' theory is unsatisfactory in virtue of his
assertion of (T2) and (T6). For if humanity in Socrates is formally but not
really distinct from Socrateity, it follows that it is not logically possible for
the former to exist in a thing without the latter existing in the same thing,
as it is logically possible for Socrates to exist without whiteness inhering in
him and to have blackness inhering in him instead. Consequently, it is not
even logically possible for humanity in Socrates to exist without being
determinately Socrates' and so not even logically possible that it should be
determinately Plato's instead. Again, it is not even logically possible that it
should actually exist in numerically many. 3 5 Thus, (Ti) is inconsistent
with the conjunction of (T2), (T3), (T4), and (T6); and Scotus' theory of
universals is thereby shown to be unacceptable.
Walter Burley's moderate realism

Another version of moderate realism current in Ockham's time and
apparently endorsed by his rival Walter Burley, 36 in effect concurs in
35. Ibid., p. 179.
36. Burley 1507, f. 4'
Scotus' theses (Ti)-(T3) but rejects the troublesome (T4)-(T6). Invoking

Aristotle, Burley maintains, with Ockham, that the Indiscernibility of
Identicals is our chief criterion of distinction among real things. He then
observes that 'every universal exists in many' but 'no particular exists in
37 38
many'; that 'the universal is defined, but no particular is defined'; that
universals but not particulars are the subject of demonstrative science;39
and that 'the universal - e.g., genus - is divided by contrary differentiae,
but no particular is divided by such differentiae'.40 His conclusion is that
(T7) Universals are really distinct from particulars.
Analogous arguments would show that different universals - e.g., genus
and differentia - are really distinct from one another. Again, like Scotus,
Burley thinks that similarity and difference must have a real foundation in
something common, but he holds that the common thing is really distinct
from the particulars in which it exists.41
Further, while the particular is a composite of such universals and some
individuating principles, the latter do not individuate or particularise the
former. Rather
(T8) The whole universal (secundum se totum) exists in each of its particulars and
is not numerically multiplied by its existence in numerically distinct
particulars. 4 2
Henry of Harclay and William Ockham both found many reasons for
rejecting this position.43
37. Ibid., f. 4": 'Omne universale est in multis. Nullum singulare est in multis, ut patet ex libro
Periermenias. Ergo aliquid vere affirmatur de universale quod negatur a quolibet suo singulari.'
38. Ibid., f. 4"": 'Secundo sic: Universale diffinitur. Sed nullum singulare diffinitur, ut patet ex septimo
Metaphysicae. Ergo etc.'
39. Ibid., f. 4"- b .
40. Ibid., f. 4 v a : 'Item universale ut genus dividitur per differentias contrarias, ut animal per rationale
et irrationale. Sed nullum singulare dividitur per huiusmodi differentias. Ergo etc'
41. AM., f-4,b·
42. Ibid., f. 4 Α ·
43. See Harclay in Gil 1971. In this question, it is Scotus' arguments that are presented in behalf of an
affirmative answer, but the counter-arguments really attack the position Burley held. For this
reason, Gâl speculates that the question may be a record of an actual debate in which the students
supplied the objections and Harclay is speaking only at the end when he gives and defends his own
opinion (Gâl 1971, pp. 184—5). Sec also Ockham 1970, Ordinatio I, d. 2, q. 4, who presents the
position as one mistakenly attributed to Scotus. Gâl shows that Ockham is drawing on Harclay's
question; but Ockham never quotes Burley, and there is no positive evidence - other than the
similarity of views under consideration and the spatio-temporal proximity of the two philoso-
phers - that Ockham had Burley in mind. Burley's treatise docs not quote from Harclay or
Ockham either. But many of Harclay's and Ockham's arguments are directed against a position
like Burley's, and many of the latter's remarks constitute replies to their objections.
Two attacks by Ockham on Burley's position

Ockham argues that the conjunction of (T3) and (T7) entails the absurd
conclusion that the universal can exist without the particular or individuat
ing principles, and vice versa. First, it was widely agreed that if χ is
naturally prior to and really distinct from y, then it is logically possible for χ
to exist without y. By (T7) the universal and particular or individuating
principles are really distinct; and by (T3) the universal is a part of the
particulars in which it exists. And the part is naturally prior to the whole.
Consequently, the universal can exist without the particular or its in
dividuating principles. 4 4 The latter conclusion is scarcely surprising by
itself, since everyone would agree that human nature can exist without
Socrates or his individuating principles; for human nature continues to
exist in Plato after Socrates' death. But Ockham mistakenly reasons that
when some real thing really distinct from other things can exist without
each of them taken one by one, and can so exist by its nature, and does not
essentially depend on any of them, it can exist without each of them taken
in conjunction. And he concludes that universal human nature could exist
without any and every particular human being. 4 5
Conversely, Ockham observes that ' . . . according to them [supporters of
Burley's sort of moderate realism], the individual adds something to the
nature, and this something combines with the universal real thing to make
what is one per se', and he concludes that 'therefore, it does not seem
contradictory that what is added should be preserved by God in the absence
of any universal nature - which seems absurd'. 4 6 T o be sure, moderate
realists did not envisage either of these consequences; but it is far from clear
why they should have found them impossible rather than merely unusual.
Harclay and Ockham vs. Burley
Harclay and Ockham agree that the conjunction of (T3) and (T8) entails
the falsehood that numerically one thing exists wholly in numerically
distinct particulars at the same time. (T8) does not, however, explicitly say
that the universal is numerically one, but only that it is not numerically many in
44. Ordinatio I, d. 2, q. 4; Ockham 1970, O T II, p. 115.

45. Ibid., 115.
46. Ibid., 115: 'Confirmatur ista ratio, quia Individuum aliquid addit supra naturam, secundum istos,
et hoc aliquid faciens per se unum cum ilia re universal], quia si non, tunc esset aliquid quod nee
esset substantia nee accidens; igitur non videtur includere contradictionem quod illud additum
conservetur a Deo sine omni natura universal! advenicntc, quod videtur absurdum.'
numerically distinct particulars. Harclay simply assumes47 that Burley's

position has this further consequence, but Ockham offers two arguments to
prove it. In the first of these arguments Ockham claims that it follows from
(T3) that the universal is a part of the particulars in which it exists. Now if
each of a pair of really distinct things is equally simple and either is
numerically one, the other will be numerically one as well; likewise if the
more complex of a pair is numerically one. But a part is at least as simple as
the whole of which it is a part. It follows that since the particular is
numerically one, the universal that is a part of it will be numerically one.
The second argument claims that the position under attack implies not
only that universals and particulars are really distinct, but that universals
are really distinct from one another. Thus, the genus substance, the genus
animal, and the species man would be three really distinct things. Ockham
plausibly supposes that 'every thing that combines with another thing
really distinct from it to make a number [of things] is either numerically
one or numerically many'. 48 But the view under attack would deny that
the universal is numerically many.49 Consequently, it follows, given (T3),
that "humanity, which is numerically one and undivided in itself, exists
in Socrates and Plato" simultaneously. When Harclay condemns this as
impossible50 and Ockham rejects it as false51 they are only echoing Boe-
thius' objection that genus or species cannot be numerically one in numeri-
cally distinct particulars.52
Responding to a similar objection, Burley maintains that 'numerically
the same' or 'numerical unity' can be understood two ways: Commonly, it
signifies 'what can combine with another to make or has constituted a
number [of things], in such a way that it is true to say ofit and the other that
they are two'. But taken strictly, numerical unity is contrasted with specific
and generic unity. Burley then, in effect, concedes to Ockham's second
argument the conclusion that a universal is numerically one in the first
47. Harclay, in Gâl 1971, sec. 41—2, p. 197.
48. Ordinatio I, d. 2, q. 4; Ockham 1970, OT II, p. 113: 'Tou ratio praecedens confirmatur sic: omnis
res faciens numerum cum alia re realiter distincta est una res numéro vel plures res numéro; sed
talis res universalis, si ponatur, vere facit numerum cum resingulari; igitur ipsa est una res numéro
vel plures res numéro; sed non est plures tes numéro, quia tunc esset plura singularis, quia
secundum istos et secundum veritatem omnis res una numéro est singularis; igitur plures res
numéro sunt plura singularia; sed nulla res universalis est plura singularia secundum istos, quia
secundum istos distinguitur realiter ab omnibus singularibus, igitur est una numéro.' Cf. ibid.,
pp. 112,114.
49. Ordinatio I, d. 2, q. 4; Ockham 1970, OTII, pp. 113-14.
jo. Harclay, in Gâl 1971, sec. 41, p. 197.
$1. Ordinatio I, d. 2, q. 4; Ockham 1970, OTII, pp. 108-12.
52. Boethius 1860h, Commentaria in Porphyrium Lib. I; PL 64, col. 83B.
sense, and further, that the species man and the species donkey are two
really distinct things. But he denies that any universal is numerically one in
the strict sense, the only sense in which, he contends, existing simul-
taneously in numerically distinct subjects is problematic.53 Harclay and
Ockham argue, on the contrary, that this conclusion has unacceptable
consequences, even if 'numerically one' and 'numerically the same' are
understood in this broader sense.
They argue first from creation and annihilation: for 'creation is ab-
solutely out of nothing in such a way that nothing essential and intrinsic to
the thing absolutely precedes it in real existence';54 and 'when something is
annihilated, nothing intrinsic to the real thing will remain in real existence
-either in it or in anything else'.55 Thus, if numerically the same universal
pertained to the essence of numerically distinct particulars of a given
species, no individual ofthat species could be created at a time when some
individual of that species already existed, or be annihilated unless all
members of that species were simultaneously destroyed56 — which is
absurd. Burley replies that 'the annihilation of something is the destruction
of all its parts. But the species is not part of an individual.' For it follows
from Aristotle's teachings that an individual or particular that is numeri-
cally one in the strict sense can have as parts only those things that are
numerically one in the strict sense.57 'Therefore, it is not necessary that
when the individual is annihilated, the species is annihilated.'58
Ockham accepts the thesis that particulars have only particulars as parts,
and he has used it as a weapon against moderate realists. But he insisted that
the position under attack implies the opposite: for by (T3) the universal
53. Burley 1507, f. $"~h; see esp. f. 5™: 'Dicendum est ad primam rationem, quando dicitur quod idem
numéro esset in caelo et in terra et in inferno, dicendum est quod "idem numéro" accipitur
dupliciter - scilicet communiter et stricte. "Idem numéro" acceptum communiter est illud quod
cum alio potest facere numerum seu constituit numerum, ita quod de illo et de alio est verum
dicere quod ista sunt duo. Et isto modo concedo quod natura hominis est una numéro, quia ilia et
natura spedfica asini sunt duae naturae. Sed "unum numéro" stricte acceptum est solum illud
quod distinguitur contra unum specie et unum genere, quomodo loquitur Philosophus de uno
numéro quinte et septimo Metaphyskae. Et eadem est divisio de "eodem numéro". Nam idem
numéro uno modo est illud quod cum alio ponit in numerum et alio modo est illud quod
distinguitur contra idem specie et idem genere, ut patet ex primo Topicorum.'
54. Ordinatio I, d. 2, q. 4; Ockham 1970, O T II, p. 115: "... sed creatio est simpliciter de nihilo, ita
quod nihil esscntiale ct intrinsecum rei simpliciter praeccdat in esse reali...'
55. Ibid., p. 116; 'Sed in adnihilatione nihil intrinsecum rei remanet nee in se nee in alio quocumque in
esse reali...'
56. Ibid., pp. 115—16; for annihilation only in Harclay in Gâl 1971, sec. 43, p. 197.
57. Burley 1507, f. 5™; cf. f. 5".
58. Ibid., f. s": 'Et si dicatur annihilatio alieuius est destruetio quantum ad omnes partes eius. Sed
species non est pars individui. Ideo non oportet quod annihilato individuo annihiletur species.'
exists in particulars as a constituent of them; and by (T7) it is really distinct

from the particular and its individuating principles, as parts are really
distinct from their wholes. Given these claims, Burley's denial would seem
to amount to little more than the stipulation that only particular con-
stituents of a particular count as its 'parts'. In any case, Ockham's creation
and annihilation arguments depend, not on whether universals are parts of
particulars, but on whether they are essential constituents of them; and
surely Burley would not deny that human nature is essential to Socrates.
Even if it were granted that universals cannot be parts of particulars and
that the annihilation of a particular involves only the destruction of its
parts, an analogous difficulty would arise at the level of genera and species.
And since genus and species are both universals, Burley cannot raise the
previous objection to saying that the genus is part of its species. But if
numerically the same genus exists in, as an essential part of, all its species, it
would follow that 'God could not destroy the species of man without
destroying every species of the genus of substance - which is absurd.' 59 To
this line of argument Burley replies that 'annihilation is the destruction of
all those things that are proper to the thing, not of all those things that are
common to it and to others. Therefore, the species can be annihilated
without the annihilation of the genus, because the genus is not a part proper
to the species but a part common to it and to others.' 60
Considering a similar move,61 Ockham rightly insists that it does not
accord with traditional definitions. For Socrates is not said to be created if
his matter exists before he does. Yet the matter is not 'proper to' Socrates
since it can exist as a part of something else, such as Socrates' corpse.62
Surely genus is at least as essential to species and the species at least as
essential to Socrates as Socrates' matter is to him.
Ockham and Harclay also argue that Burley's position will lead to
contradictories, when it comes to accounting for the relation between
59. Ibid., f. 5«* 'Contra: quicquid est de specie est et de individuo secundum istam opinionem est
genus. Sed genus est pars speriei. Ergo sequitur quod non potest annihilare speciem hominis nisi
destruat hoc genus animal, et etiam genus generalissimum substantiae. Sed destructo genere
destruuntur omnes illius generis. Ergo Deus non posset destruere speciem hominis nisi destrucret
omnem speciem de genere substantiae - quod est inconvénient.'
60. Ibid., f. 5V* 'Et quando arguitur in contranum, quando dicitur quod ad minus sequitur istud
inconveniens: quod Deus non posset annihilare speciem hominis nisi annihilaret totum genus
substantiae, dicendum quod illud non sequitur, quia annihilatio est destructio rei quantum ad
omnia quae sunt ei propria, non quantum ad omnia quae sunt communia sibi et aliis. Et ideo
species potest annihilari absque annihilatione generis, quia genus non est pars propria speciei, sed
communis sibi ct aliis.'
61. Ordinatio I, d. 2, q. 4; Ockham 1970. OT II, p. 116.
62. Ibid., p. i n .
substances and accidents. For it follows from the view as discussed so far
that
(Tç>) A particular substance exists only if the real things that are its genera and
species in the category of substance really exist in that particular.
But Aristotle says that substances are susceptible of contraries, and Ockham
infers that the position would endorse the analogous theses for contraries as
well:
(Tio) Particular contraries do not really exist unless the universal contrary real
things exist;
and
(Ti i) A particular accident inheres in a particular substance only by virtue of the

universal accident's inhering in the universal substance.
Ockham then observes that frequently one of a pair of contraries inheres in

one member of a given species and the other in another member of the
same species at the same time. For example, Socrates may be white all over
while an Ethiopian is black all over; 6 3 Christ is happy in heaven while Judas
is damned and miserable in hell. 64 It follows from the conjunction of (T8),
(TQ), (Tio), and (Τι i) that in such cases contraries simultaneously inhere in
one and the same real thing - which is impossible. 6 5
Burley's reply to a similar objection is, in effect, to deny that (Τι i) is true
in general. Instead, the universal is the first subject of some accidents, such
as spatial location and being in motion or at rest. But other accidents
pertain primarily to the particular and pertain to the universal per accidens,
insofar as it is a constituent of the particular to which they pertain per se.
Thus Burley insists that where we are speaking about universals that are
numerically one in the broad sense only, 'it is not absurd that numerically
the same thing is in heaven and in hell and that it is simultaneously in
motion and at rest'. 6 6 Universals exist in many subjects and hence in many
places at once. Experiencing joy and sorrow pertain primarily to par
ticulars, however, and so simultaneously pertain to the universal man only
per accidens, insofar as it is a constituent of Christ and Judas respectively. 67
63. Ibid., pp. 121-2.

64. Ibid., p. 121. Cf. Harclay in Gâl 1971, sec. 43, p. 198.
65. Ordinatio I, d. 2, q. 4; Ockham 1970, O T II, pp. 120-2.
66. Burley 1507, f. 5": 'Dico igitur quod accipiendo idem numéro communiter pro omni eo quod
constituit cum alio numerum, sic non est inconveniens quod idem numéro sit simul in caelo et in
inferno et quod simul moveatur et quiescat.'
67. Ibid., f. s""b-
One may wonder how the latter reply is supposed to avoid a violation of
the Indiscernibility of Identicals, however. For if happiness pertains to the
composite of the universal man with Christ's individuating principles, and
misery to the composite of the universal man with Judas' individuation
principles, it still follows that happiness and misery pertain to the universal
man simultaneously. Burley's idea may be that just as apparent contradic-
tions can be avoided by relativising the predications of contraries to
different times - e.g., the table is white all over at one time and black all
over at another - or to different parts of the particular - e.g., the table has a
white top and black legs - so general ascriptions of the second sort of
accidents to universal substances must be relativised to the particulars in
which it has them - e.g., the universal man is happy in Christ and miserable
in Judas. Ockham, for one, could not consistently object to this move, since
he insists that no contradiction is involved even in the same particular's
existing simultaneously in discontinuous places and in being simul-
taneously hot at one place and not hot at the other.68
Thus, whereas Ockham seems eventually to succeed in convicting
Scotus' theory of contradiction, he is able to find Burley's theory guilty
only of the odd or unexpected.
Henry of Harclay's position

While Scotus held, in (T2), that the individuating principles of things are
logically incapable of existing in, as constituents of, numerically distinct
things simultaneously, Harclay insists that
(T12) Everything that exists in reality is essentially singular - i.e., logically
incapable of existing in, as a constituent of, numerically many simul-
taneously.69
Further, distinguishing singularity from universality and particularity,
Harclay maintains that
(T13) Extra-mental (singular) things are neither universal nor particular of them-
selves or essentially.70
His thought is that essentially singular things such as Socrates are capable of
acting on the intellect to produce two sorts of concepts of themselves: a
distinct concept, by which the intellect conceives of that thing in such a
way as to be able to distinguish it from others of the same kind; and a
68. Ockham 1494-6, Reportatio IV, q. 4, N ; q. 5 J; cf. Quaestiones in libros Physicorum, q. 32.
69. Harclay in Gâl 1971, sec. 67, p. 211.
70. Ibid., p. 211.
confused (or generalised) concept by which the intellect cannot so dis-

tinguish it. 7 '
(T14) The universal is a thing confusedly conceived, and a particular is the same
thing distinctly conceived.72
Hence it follows that a universal and its particular are the same real thing,
but distinct in reason. And Harclay says that 'what is superior to Socrates,
such as " m a n " and "animal", signifies no real thing other than the thing
that is Socrates, but insofar as he is confusedly conceived; that is, insofar as
he moves the intellect to conceive of him in a confused way. And so I say
that "Socrates is a m a n " is a case of the superior being predicated of the
inferior, which is nothing other than that Socrates is Socrates, i.e., Socrates
absolutely is Socrates, but insofar as he is confusedly thought of; and
"Socrates is an animal" or "Socratesis a m a n " are the same.' 7 3 But Socrates
is particular insofar as he is distinctly conceived. As for the claim that
something essentially common or universal must exist in reality to be the
object of definition or of the demonstrative sciences, Harclay replies
that their objects are essentially singular extra-mental things considered
not absolutely, but insofar as they are confusedly conceived. 74 Further,
Harclay simply rejects the contention by Scotus and Burley that real
similarity or agreement between things must be founded on the real unity
of a constituent common to them both. On the contrary, 'if they agree,
then they are not one in any third real thing'. 7 5
Burley's attack on Harclay

Burley devotes considerable space to refuting this position, and Ockham
declares Harclay's view to be 'absolutely false and unintelligible'. 76
71. Ibid., sec. 79, p. 216.

72. Ibid., sec. 95, p. 222.
73. Ibid., sec. 79, p. 216; 'Modo ego dico quod superius ad Sortem, puta "homo" et "animal", non
significat rem aliam nisi rem quae est Sortes, ut tamen concipitur confuse, id est movet
intellectum ad concipiendum ipsum modo confuso. Et sic dico quod hic "Sortes est homo" est
praedicatio superioris de inferiori; quod non est aliud nisi quod Sortes est Sortes, id est Sortes
absolute est Sortes, ut tamen confuse intellectus; et hoc est idem "Sortes est animal" vel "Sortes est
homo"...'
74. Ibid., sec. 100, p. 224.
75. Ibid., sec. 78, p. 215: '... quia concedo quod est unitas et convenientia realis inter mensuram et
mensuratum inter comparabilia; pro secundo argumento: inter similia. Sed propter hoc non
habent aliquam unam rem communem eis, distinctam ab omnibus. Nee convenientia aliqua
realis, quantumcumque magna, hoc arguit, immo oppositum. Si enim sunt convenientia, ergo
non sunt unum in aliquo reali tertio.'
76. Ordinatio I, d. 2, q. 7: Ockham 1970, OT II, p. 241.
First, Burley mounts a double-pronged attack on the claim that what is

defined is a singular thing confusedly conceived. On the one hand,
Burley appeals to the Aristotelian claim that the definition is a means of
arriving at a distinct and perfect cognition of the object defined. Thus, if
definition were of singular things, it would be no more of one singular
thing of a given kind than another. Therefore no one would be able to
know the definition mortal rational animal without having distinct and
perfect knowledge of each and every individual human being - which is
absurd.77 To this Harclay could reply that, on the contrary, knowledge of
the definition gives one at most confused knowledge of the singulars
defined. For singular things are the objects of definition only insofar as they
are confusedly conceived; and the confused concept caused by Socrates is
indistinguishable from that caused by Plato.
More fundamentally, Burley charges that the definition is interchange-
able with what is defined, while the definition is not interchangeable with
any singular thing - e.g., mortal rational animal is not interchangeable with
Socrates confusedly conceived.78 Harclay considers the similar objection
that, as superior to Socrates, animal will be predicated of more than
Socrates. But if animal is Socrates confusedly conceived, and Plato is an
animal, it will follow that Plato is Socrates confusedly conceived - which
is absurd.79 Harclay replies that that argument involves a fallacy of
accident,80 taking a position that Ockham rejects as 'neither true nor
logical'.81
Again, Burley reasons, if universals were singulars confusedly con-
ceived, then the most generic genus of substance would be a singular
77. Burley 1507, f. 4*.

78. Ibid., f. 4".
79. Harclay in Gâl 1971, sec. 81, p. 217.
80. Ibid., sec. 85, pp. 218-19.
81. Ordinatio I, d. 2, q. 7; Ockham 1970, OT II, pp. 242-3: 'Sed istud non est vere nee logice dictum,
quia quando dicitur sic "animal est Sortes confuse conccptus" aut "animal" supponit simpliciter,
aut personaliter. Si personaliter, "animal" non tantum est Sortes confuse conceptus, sed animal est
Sortes distincte conceptus, quia tunc est una indefinita habens unam singularem veram, scilicet
istam "hoc animal est Sortes distincte conceptus", demonstrando Sortem. Si autem "animal"
supponat simpliciter, aut supponit pro aliqua vera re, aut pro ente tantum in anima, aut pro
aggregato. Si pro re, igitur aliqua vera res est communis, et per consequens vera res praedicatur
vere de alia re. Et ita, sicut haec est vera simpliciter "Sortes est animal" vel "asinus est animal", ita
erit haec vera "asinus est Sortes conceptus", quia per te illud commune pro quo supponit
"animal" in ista "animal est Sortes confuse conceptus" est Sortes conceptus, et nullo modo
distinguitur a Sorte concepto; igitur de quocumque praedicatur unum et reliquum. Si utrumque
supponat personaliter, igitur sicut haec est vera "asinus est animal", secundum quod "animal"
supponit personaliter, ita haec erit vera "asinus est Sortes conceptus", secundum quod praedic-
atum supponit personaliter.'
substance confusedly conceived. But there is no more reason why it should

be one singular substance than another; hence, if it is one of them con-
fusedly conceived, it is all of them confusedly conceived, and there would
be as many most generic genera of substance as there are individual
substances — which is absurd.82
Burley, following Scotus, assumes that where concepts have a genuine
application to reality, there is a distinction in things corresponding to the
distinction among concepts; but where Scotus allows that a formal distinc-
tion will do, Burley insists that there must be a real distinction. Thus he
reasons that 'the identity between universal and individual in extramental
things is not greater than that between universal and individual in con-
cepts'. But Harclay admits that the distinct and confused concepts, the
inferior and superior concepts are different concepts. Therefore, the things
corresponding to them must be distinct.83 Again, Burley remarks,
'everyone agrees that the universal is abstracted from singulars and from
material conditions by the intellect, but what is abstracted from such
material conditions really exists under them; otherwise abstraction would
deceive us. Therefore, the universal is really distinct from the singular from
which it is abstracted.'84 But Harclay would reply with Ockham by
rejecting as one of the most fundamental errors in philosophy the notion
that distinctions in concepts must be mirrored by distinctions in reality.
Ockham's attack on Harclay
Unlike Burley, Ockham concedes Harclay's (T12), but argues that (T12)
entails not only (T13), but also the denial of (T14). For to say that a thing is
universal is to say that it is logically possible that it should be simultaneously
predicated of many, whereas to say that it is singular is to say that this is
logically impossible. Thus, it is a contradiction to say that one and the same
real thing is simultaneously singular and universal. As for Harclay's claim
82. Burley 1507, f. 4,1~l>·
83. Ibid., f. 4rb: 'Item non est maior identitas inter universale et Individuum in rebus extra quam inter
universale et Individuum in conccptum, quia ita essentialiter et in quid praedicatur universale de
individuo in conceptibus sicut in rebus extra. Sed in conceptibus conceptus universalis et
conceptus individualis sive conceptus superior et conceptus inferior non sunt idem. Ergo ncc in
rebus erunt universale et Individuum sive superius et inferius eadem.'
84. Ibid., f. 4,b: 'Item in hoc conveniunt omnes quod universale abstrahitur a singularibus per
intellectum et a conditionibus materialibus. Nam Commentator primo Di anima dicit quod
intellectus facit universalitatem in rebus. Et ex hoc sic arguitur: Illud quod per intellectum
abstrahitur a conditionibus et accidentibus materialibus realiter est sub illis; aliter enim ab-
strahentium esset mendacium - quod est contra Philosophum secundo Physicorum. Universale
ergo abstractum ab accidentibus realiter existens sub eisdem est aliqua res a singular!, et sic in rebus
universale et singulare sive Individuum non sunt eadem res.'
that what is essentially singular is made universal through the consideration

of the intellect, Ockham replies that the intellect is extrinsic to the thing.
And nothing extrinsic to a thing can bring it about that the thing has a
property that is incompatible with one of its essential properties.85 Conse-
quently, no matter how the intellect conceives of essentially singular things
such as Socrates and Brownie the donkey, it will not be able to bring it
about that either is universal, any more than it could bring it about that
Socrates is a donkey or that Brownie is a human being.86
Harclay might grant that whereas 'Socrates is singular' and 'Socrates is
universal' are contradictory, 'Socrates of himself is singular', 'Socrates
distinctly conceived is particular', and 'Socrates confusedly conceived is
universal' are not; for the modifiers 'of himself', 'distinctly conceived', and
'confusedly conceived' function to remove the contradiction just as much
as 'on top' and 'on the bottom' do in 'The table is white on top and black on
the bottom'.
Ockham replies that such modifiers could remove the contradictions
only if they served to alter the supposition of the subject term, so that the
resultant statements asserted the contradictory properties about different
property-bearers. But Ockham believes that this can happen only (a) when
the modifiers mention some part of the whole denominated by the subject,
in which case they diminish the supposition from the whole to the part -
e.g., in 'An Ethiopian is white with respect to his teeth', or (b) when the
added determination is incompatible with the existence of the thing to
which it is added, in which case it distracts the supposition to something
else - e.g., adding 'dead' to 'man' distracts the supposition from the man to
his corpse.87 The above-mentioned modifers - 'of himself, 'distinctly
conceived', and 'confusedly conceived' - neither mention part of Socrates
nor signify anything incompatible with his existence, however, and so
cannot alter the supposition of the subject term in Harclay's propositions.
Thus, 'Socrates of himself is singular; therefore Socrates is singular',
'Socrates distinctly conceived is particular; therefore Socrates is particular',
'Socrates confusedly conceived is universal; therefore Socrates is universal'
are all formal inferences. And if the last conclusion is incompatible with
each of the first two conclusions, the last premiss is incompatible with each
of the first two premisses as well.88
85. Ordinatio I, d. 2, q. 7; Ockham 1970, O T II, p. 241; cf. pp. 243, 249.
86. Ibid., pp. 2 4 8 - 9 : 2 5 1 .
87. Ibid., pp. 244—5.
88. Ibid., pp. 244, 248.
Harclay might insist that he was not trying to argue that while 'uni-
versal' and 'singular' are genuinely incompatible predicates, they are not
really asserted about the same thing. Rather he meant to contend that
although the predications 'Socrates is singular', 'Socrates distinctly con-
ceived is particular', and 'Socrates confusedly conceived is universal' are
made about one and the same real thing - viz., Socrates - the predicates
involved are not genuinely incompatible. In effect, Harclay has distin-
guished the one-place predicate '—is singular' from the two-place predicates
'— is universal relative to — ' and '— is particular relative to —'. The first
blank in those two-place predicates is to be filled by a proper name of an
essential real singular thing, such as Socrates, the second blank by an
expression designating a concept whose existence in the mind is caused by
the real thing picked out by the expression filling the first blank. Thus
'Socrates is universal relative to a confused concept caused by Socrates' and
'Socrates is particular relative to a distinct concept caused by Socrates' are
no more mutually contradictory than are 'Plato is older relative to Aris-
totle' and 'Plato is younger relative to Socrates'. Moreover, without
further argument (which Ockham does not supply) there is no reason to
think that 'Socrates is singular' and 'Socrates is universal relative to a
confused concept caused by Socrates' are mutually contradictory either.
Ockham's own view

Throughout his career, Ockham insisted as Harclay did that
(T1.2) Everything that exists in reality is essentially singular - i.e., logically
incapable of existing in, as a constituent of, numerically many simul-
taneously,
and concurred with Harclay against Scotus that one and the same singular
thing can both cause and genuinely fall under both confused and distinct
concepts, even though this distinction in concepts is in no way reflected by
a distinction between nature and individuating principles in the singular
thing. Unlike Harclay, however, Ockham maintained that universals are
nothing other than names - naturally significant general concepts pri-
marily, and secondarily the conventional signs corresponding to them.
Ockham's nominalism is thus less misleadingly classified as a form of
conceptualism. 89
89. As noted in Boehner 1946.

Replies to opposing views

Replying to arguments in favour of moderate realism, Ockham says that
Aristotle used the phrase 'primary object of definition' in several ways:
When he says that the primary object of definition is interchangeable with
the definition, he is referring to the fact that the terms 'man' and 'rational
animal' are interchangeable, 90 contrary to Burley's contention. 9 1 O n the
other hand, when Aristotle says that definition is primarily of the thing
whose parts it signifies, he takes the primary object of definition to be
something other than a name. Ockham agrees with Burley that no par-
ticular man can be the primary object of the definition 'rational animal', in
the sense that its parts are signified by 'rational animal' prior to those of
anything else.92 And he concludes that since only particulars can be real,
nothing is the primary object of definition in this sense. Yet, each particular
man is primarily defined by 'rational animal' in the sense that there is
nothing whose parts are signified prior to the parts of it. 93 Thus Ockham
insists with Harclay that knowledge of the definition cannot give one
distinct and perfect knowledge, but only confused knowledge of the
particulars thus primarily defined; 94 so that one can know, contrary to
Burley, that every man is risible without having a distinct cognition of
every peasant in India. 95 Similarly, in one sense the object of knowledge is
mental, spoken, or written propositions and not any substance, universal
or particular. Substances are known only in the sense that such propositions
contain terms that supposit for them. And since only particular substances
can really exist, only particular substances are known in this way. 9 6 Finally,
Ockham joins Harclay in denying that real similarity between things
requires the existence of something common to and existing in both. 9 7
If Ockham was constant in his insistence that universals are in the first
instance general concepts, he seems to have held two different views about
the ontological status of concepts. 98 Early in his career, 99 Ockham main-

91. Burley 1507, f. 4".
92. Ordinatio I, d. 2, q. 4; Ockham 1970, OT II, p. 132.
93- Ibid., pp. 127-9.
94. Ibid., p. 131.
95. Burley 1507, f. 4rt.
98. For a fuller discussion of these issues see Adams 1977.
99. Ordinatio I, d. 2, q. 8; Ockham 1970, OT II, pp. 273-4. Expositio in librum Perihermenias Aristotelis,
Prooemium §§7, 9; Ockham 1978, OP II, pp. 359-60, 364-5.
tained that whatever is thought of must have some sort of existence. Yet,
we sometimes think of objects that do not or cannot really exist: e.g.,
impossibles such as chimeras, abstract objects such as universals, and un-
actualised possibles. He concludes that as objects of thought they must have
100
some non-real mode of existence, which Ockham labels 'objective' or
101
'cognized' existence. Having analysed thoughts of such objects into a
really existent mental act and an objectively existent object, Ockham seems
to extend his conclusion to thoughts about real things as well. 1 0 2 He seems
to have held further that what directs a really existent mental act towards
one objectively existent entity rather than another is that the former bears
an appropriate relation of comparative similarity or causality to the
latter. 1 0 3
Walter Chatton's objections and Ockham's replies

Ockham gave up the objective-existence theory, partly because of his own
developing reservations and partly because of criticisms raised by his
contemporary and fellow-Franciscan, Walter Chatton. Three objections
challenged the application of the theory to our thoughts of particulars.
(5.2.1) First, Chatton 1 0 4 charged that the objective-existence theory com
promises direct realism in epistemology, because it identifies the immediate
object of thought or awareness with the objectively existent entity and not
with the mind-independent real particular. Although Ockham accepts this
criticism and directs it against a similar theory held by Peter Aureoli, 105 he
seems to have been mistaken in doing so. For his early theory claims that
one and the same particular can simultaneously have both real and objec
tive existence, and hence that the immediate object of awareness in intui
tive cognitions is identical with the real particular.
Ockham himself criticises Aureoli's similar theory on the ground that it
leads to ontological paradox: ' "The trees appear to move; therefore some
motion has objective existence" no more follows than that "The trees
100. Ordinatio I, d. 2, q. 8; Ockham 1970, OT II, pp. 272—4, 283.

101. Expositio in librum Perihermenias Aristotelis, Prooemium §7: Ockham 1978, OP II, p. 360.
102. The evidence here is indirect but seems to be implied in Ockham's discussion of divine ideas in
Ockham 1979b, Ordinatio I, d. 35, q. 5 Κ and G, and in his suggestion that the objects of intuitive
cognitions are the terms of propositions (Ockham 1494—6, Reportatio II, q. 15E; see Boeh
ner 1943)-
103. Ockham 1494-6, Reportatio II, q. 15 EE.
104. Chatton, Lectura I, d. 3, q. 2; quoted in Gâl 1967, p. 203, n. 26; raised by Ockham against Aureoli
in Ockham 1979b, Ordinatio I, d. 27, q. 3 H, and against his own objective-existence theory in
Ockham 1491, Quodl. IV, q. 19.
105. Peter Aureoli 1952-6, p. 156; cf. I, d. 3, sec. 14, a. 1; II, pp. 696-700.
appear to move in reality; therefore a real motion appears" follows.' 106 It is

as if Ockham assumes that Aureoli were operating with a system whose
universe of discourse includes everything that exists either in reality or in a
non-real mode. Within this system, statements of the form 'x is F' do not in
general entail statements of the form 'x has objective existence' or state-
ments of the form 'x has real existence', but only the disjunction of such
statements. Further, this system is not one in which the predications are
relativised to modes of existence, but one in which predicates attach
absolutely. It will follow that things really have all of the properties that
they appear really to have - which is absurd. This difficulty could be easily
circumvented by relati vising the predications to one mode of existence or
another. Thus, 'Real motion appears' would become 'In the mode of
objective existence, there is some motion and it really exists' which no
more entails 'Motion really exists' than 'It is possible that motion really
exists' does.
Since God is essentially omniscient and necessarily existent, He neces-
sarily conceives of everything - actuals, possibles, and impossibles - from
eternity. On the objective-existence theory, it will follow that all actuals,
possibles, and impossibles necessarily have objective existence from eter-
nity. And while this existence is not independent of the divine intellect, it is
independent of the divine will, since God could no more destroy such
objectively existent entities than He could alter His own nature.107 But
such a limitation on God's power is unacceptable.
Chatton objected to the application of the objective-existence theory
even to our thoughts of universals, on the ground that it 'does with more
what could be done with fewer'. While Ockham's early expositions of his
theory make objective existence mind-dependent, they do not rule out the
possibility that two people might simultaneously have the same objectively
existent entity before their minds. Chatton judges this impossible on the
apparently mistaken ground that if the objective existence of an entity does
not logicaDy presuppose the real existence of any act in particular, then it
cannot logically presuppose the real existence of some act or other
either.108 But if objectively existent entity a logically presupposes the real
existence of my act of intellect and objectively existent entity b logically
106. Ockham 1979b, Ordinatio I, d. 27, q. 3 K.

107. Raised by Chatton against Aureoli in Lecturae Chaton Anglici in Sententias, Prologi Quaestio
Secunda; edited by O'Callaghan in O'Donnell 1955, p. 241; cited by Ockham against his own
objective-existence theory in Ockham 1491, Quodl. IV, q. 19.
108. Chatton quoted in Gâl 1967, p. 202.
presupposes the real existence of your act of intellect, then since my mental
act can continue to exist when yours ceases and vice versa, it follows that a
can continue to exist when f> does not and vice versa. Hence, there are two
objectively existent entities, just as there are two acts of intellect. 109
Such multiplication of objectively existent entities would be futile,
however, only if there were no theoretical role for them to play. But on the
objective-existence theory, they are necessary, because they prevent our
thoughts of chimeras, abstract objects, and impossibles from being
thoughts of nothing. And in general, it is the fact that a really existent
mental act is directed to an objectively existent entity and to one rather
than another that explains why the thought has an object and has one
object rather than another.
Chatton argues that such objectively existent entities cannot possibly fill
the latter theoretical role, while the real things assumed by the theory can
do so without them. He reasons that if one real thing can exist without any
other really distinct (created thing), it can a fortiori exist without any and
every objectively existent entity. But it is contradictory to suppose that an
act of thought exists without being a thought of something. Chatton
concludes that the direction of a thought towards an objectively existent
entity and towards one rather than another cannot be what makes the
thought have an object or have one object rather than another. 11 ° Further,
Chatton rejects Ockham's contention that whatever is thought of must
have some sort of existence. In order for us to think of a chimera or a
golden mountain, it is not necessary that any existent entity be a chimera or
a golden mountain; it is enough if some real thing has the property of
being-of a-chimera or being-of a-golden-mountain. , 1 x And even on the
objective-existence theory, Ockham has, in effect, allowed that really
existent mental acts have such properties; for he claims that what directs a
really existent mental act towards one objectively existent entity rather
than another is appropriate relations of comparative similarity and cau-
sality. Ockham accepts Chatton's conclusion 112 and, abandoning the dis-
tinction between objective and real existence, identifies concepts with
really existent acts of intellect and universals with really existent abstract
general concepts.
Ockham's nominalism contrasts with modern nominalisms in identify-
109. Ibid., p. 202.

no. Ibid., p. 202.
in. Ibid., p. 207.
112. Ockham 1491, Quodl. IV, q. 19.
ing universals primarily with naturally significant names (or concepts) and
not with conventional names. But he does not devote much attention to
explaining what this relation of natural signification is. His basic thought
seems to be that a general concept C naturally signifies χ if and only if there
does not exist a y such that C resembles y more than C resembles x. Thus,
the general concept man naturally signifies Socrates and Plato and not a
gorilla, because there is nothing it resembles more than it resembles
Socrates or more than it resembles Plato, but it resembles each of them
more than it resembles a gorilla. 1 1 3 Many objections may be raised against
this formulation, however. Indeed, it seems doubtful whether (on the
objective-existence theory) the objectively existent universal man will be
more similar in the relevant respects to the fully determinate particulars it is
supposed to signify than to the objectively existent universal Australoid
man or Negroid man respectively; or (on the mental-act theory) whether
there is anything about an unextended mental quality by virtue of which it
could resemble Socrates and Plato more than a particular gorilla. 1 1 4
Conclusion
Although both the moderate realist and nominalist positions were de
veloped with more variety and subtlety in the early fourteenth than they
had been in the twelfth century debate over universals, the fundamental
disagreements over identity, distinction, and similarity remained the same.
Nominalists tried to show that any version of moderate realism was bound
to be contradictory, while realists usually had more resources for preserv
ing consistency than nominalists reckoned with. Ockham is no more
successful in convicting Burley's position of contradiction than is Abelard
in his attack on the first realist theory of William of Champeaux.
Surprisingly, it is against his most distinguished opponent, Duns Scotus,
that Ockham's onslaughts really do succeed. Even if Henry of Harclay's
theory were consistent, however, his attempt to combine a nominalist
ontology of singulars and concepts with a realist vocabulary is, as Ockham
observed, obscurantist. In this respect, at least, Ockham's more straightfor
ward nominalism enjoys a definite advantage.
113. Expositio in libmm Perihermenias Aristotelis, Prooemium §6; Ockham 1978, OP II, 354-5.
114. For a fuller discussion of these topics, see Adams 1978.
21
FAITH, IDEAS, ILLUMINATION,
AND EXPERIENCE
Medieval Platonism
In the later Middle Ages philosophical questions on the ordinary sources of
human knowledge attracted continuous though uneven attention. The
fundamental problem for the discussions involved taking account of two
commonly recognised extremes.
O n the one hand, Augustine had in summary fashion heralded a unified
philosophy. For him the best in all preceding Greek thought had been
assimilated into the Platonism current in his epoch, a type now con-
veniently designated by the nineteenth-century term 'Neoplatonism'.
Within its own competence the perfected philosophy, as Augustine saw it,
paralleled revealed biblical truth. His view set the framework for Christian
intellectual tradition among the Latins for the eight ensuing centuries.
Medieval Aristotelianism
O n the other hand, Boethius, from whose translations and commentaries
medieval students learned their logic and received their general introduc-
tion to philosophy, had handed down an acquaintance with certain facets
of Aristotle that resisted absorption into the Neoplatonic stream. By the
mid twelfth century Aristotle had attained the status of the Philosopher par
excellence.1 His thought, as enhanced by Islamic writers translated during
the latter half of that century, deepened medieval inquiry into subjects
significant for problems of cognition. During the thirteenth century
Aristotle's major works became available in direct translation and were
read with the commentaries of Averroes. They rapidly imposed their
philosophical techniques upon the intellectual training in the newly es-
tablished universities and guided it for the rest of the medieval period.
i. 'Nam et antonomastice, id est excellenter, Philosophus appellatur.'John ofSalisbury 1929, II, 16;
p. 90.18-19.
Faith, ideas, illumination, and experience 441
Augustine and the ideal of Christian philosophy

Augustine's mature thinking had been entirely dominated by the Christian
faith. Scriptural passages prompted him to reason:
Hence, because there is the one Word of God, through which all things were
made, which is the unchangeable truth, all things in it are originally and un-
changeably simultaneous... In it, however, these things have not been, nor are
they to be, but they only are; and all things are life,... that life which was 'the light
of men', the light certainly of rational minds, through which men differ from
beasts and are, therefore, men. ... nor has it been placed far away from any of us,
for in it we live and move and have our being.2
All things were within God, and in God men lived and thought. Philos-
ophy, Augustine further reasoned, had come into harmony with those
revealed tenets of faith:
... after many generations and many conflicts there is strained out at last, I should
say, one system of really true philosophy. For that philosophy is not of this world-
such a philosophy our sacred mysteries most justly detest - but of the other,
intelligible, world.3
Philosophy and faith

Obviously this attitude ascribed to the Judeo-Christian revelation, as
rationally developed through sacred theology, the primacy in one's overall
world outlook. But it gladly recognised the autonomous even though not
dominant role of philosophy. Philosophy by itself had come to know the
intelligible world that had been revealed in the gospel as the eternal Word
of God, but without recognising it as such. For Augustine the Word was in
fact Christ, the second person of the Trinity. In that W o r d all things,
including the human mind, really existed and were truly known.
2. De Trin., IV, I, 3 (tr. McKenna, pp. 132—3). 'Quia igitur unum Verbum Dei est, per quod facta
sunt omnia, quod est incommutabilis Veritas, ibi principaliter atque incommutabiliter sunt omnia
simul;... Ibi autem nee fuerunt, nee futura sunt, sed tantummodo sunt; et omnia vita sunt,... sed
vita erat lux hominum: lux utique rationalium mentium, per quas homines a pecoribus differunt, et
ideo sunt homines... nee longe posita ab unoquoque nostrum: in ilia enim vivimus, et mo-
vemur. et sumus.' PL, 42, cols. 888-9. Cf. Augustine 1961, 40, 1; p. 492-5: 'docetur enim non
verbis meis, sed ipsis rebus deo intus pandente manifestis'.
3. Contra acad.. Ill, 19, 42 (tr. O'Meara, p. 149). '...multis quidem saeculis multisque conten-
tionibus, sed tamen eliquata est, ut opinor, una uerissimae philosophiae disciplina. Non enim
est ista huius mundi philosophia, quam sacra nostra meritissime detestantur, sed alterius intel-
lcgibilis.,.' Augustine 1950, p. 79.15-19- Cf. 'Lumen autem mentium esse dixerunt ad discenda
omnia eundem ipsum Deum, a quo facta sunt omnia.' Deciv. Dei. VIII, 7; Augustine 1899-1900,
p. 366.21-2.
Religious faith, accordingly, was one distinct source of human knowl-

edge. Through theology it was to maintain its dominating and judging
role during later medieval times. From that viewpoint medieval thinking
remained within the pattern set by Augustine, no matter how details varied
in different writers.
Intellection and illumination

Philosophical knowledge, however, had to flow from other sources. For
Augustine, sharing the Platonic tradition, these were intellection and sense
experience. 4 Intellection was the interior vision of things in their true
reality, reality that was vitally and luminously present within the mind in
Neoplatonic fashion (see Plotinus, En., V, 9, 6.1-10) as an intelligible
object that was identified with each particular intelligible and yet was one
intelligible. Augustine's use of this intelligible world to explain human
thought is generally called today, as it was occasionally in the Middle Ages,
the doctrine of'divine illumination'. s That doctrine obviated any need to
explain how external things could get into the mind in order to be known.
They were already there, since they really existed in the mind's interior
light, God himself; indeed, they had a higher type of being in the divine
Word than in the external world.
Ideas and concepts

With things spread before the mind in that interior and living light, and
with the mind's own vital response, explanation through metaphors of
vision and birth seemed sufficient for Augustine. The things known were
actually in front of the mind's eye. The mind saw them, thereby giving
birth to human knowledge. 6 In this setting, ideas (which he called ideae,
rationes, formae, or species)1 were what the mind objectively confronted, not
subjective mental features. Understandably, in this setting the 'idea* of a
4. ' . . . duos esse mundos, unum intellegibilem, in quo ipsa Veritas habitaret, istum autem sensibilem,
quem manifestum est nos visu tactuque sentire'. Augustine 1950, III, 17, 37; p. 76.8-10. Cf.
Augustine 1899-1900, VIII, 7; p. 366.18-21.
5. E.g. Roger Bacon 1897-1900, II, 5; I, p. 39.9. Bonaventure 1882- 1902a, III Sent.,2s, 2,2, Resp.;
Ill, p. 491b (for faith). Henry of Ghent 1520 Summa quaest. ord., I, 2;ff. 3V-4V ('illustratio divina',
used repeatedly). Cf. Augustine 1894, De Cen. ad lit., XII, 31: 'ipsum lumen, quo inlustratur
anima,... nam illud iam ipse Deus est', p. 425.22—4. 'Jam superior ilia lux, qua mens humana
illuminatur, Deus est', Infoan. ev., XV, 4,19; PL, 3 s, col. 1517. Augustine 1961,40,1; p. 48.25-6:
'... inferiore luce veritatis, qua ipse, qui dicitur homo interior, inlustratur
6. ' . . . omnis res quamcumque cognoscimus, congenerat in nobis notitiam sui. Ab utroque enim
notitia paritur, a cognoscente et cognito." De Trin., IX, 12, i8; PL, 42, col. 970.
7. 'Quas rationes, ut dictum est, sive ideas, sive formas, sive species, sive rationes licet vocare, et
multis conceditur appellare quod libet...' Dediv. quaest. LXXX1II, 42, 2; PL, 40, col. 31.
sensible object had much more reality than its external embodiment; in
more exact phraseology, the same thing had greater reality in its idea than
in its sensible existence. Yet in accord with the birth metaphor, Augustine
could designate the mind's own representation of an idea as a 'word' that
was conceived.8 This use of 'word' was modelled on the way the term in
theology helped to signify the difference between the divine persons. The
mental word was an 'image' or 'likeness' of the thing, what medieval as
well as modern writers call a 'concept'. Its presence was acknowledged by
Augustine, but it remained peripheral in his account of cognition.
The role of the senses

Universal knowledge of a thing through its presence in the soul's interior
light remained a persistent Augustinian heritage in medieval thought
about cognition. But factual (quantitative and qualitative) acquaintance
with corporeal things was not given in illumination only. For such cog-
nition, the human mind had to be 'admonished'9 by the senses. Yet even
here the experiencing was by the mind, though through the senses.10 The
senses could not affect the soul, for nothing corporeal could act upon
something incorporeal.1 ' Such an account of experience of course avoids
the difficulties of explaining transition from sense to intellect, since from
the start the entire cognitive activity was in the mind.
The Aristotelian viewpoint transmitted by Boethius

Although there is no mention of illumination in Boethius' logical works,
they exercised an influence that managed by and large to function within
the Augustinian framework even while handing down in clear outline the
basis of an Aristotelian theory of knowledge. Human cognition attained
8. '... visu mentis aspicimus: atque indc conccptam rerum veracem notitiam, tanquam verbum,
apud nos habemus, et dicendo intus gignimus'. De Trin., IX, 7,12; PL, 42, col. 067.
9. Cf. 'aut admonitione a sensibus,... cum intellegimus esse corpus', Augustine 1895, Epist., XIII,
4; p. 31.21-2. '... necidonea est ipsa mens noster, in ipsis rationibus, quibus facta sunt, ea uidere
apud deum, ut per hoc sciamus, quot et quanta qualiaque sint, etiamsi non ea uideamus per
corporis sensus', Augustine 1894, V, 16; p. 159.24-7. '... ad intelligibilia sua videnda a sensu
admonetur potius quam aliquid accipit', Augustine 1895, VI, 2; p. 12.23-4. '• • · verbis fortasse ut
consulamus admoniti', Augustine 1961,11, 38; p. 47.9-10. '... admonetur. cum de istis partibus
intcrrogatur,' ibid., 12, 40; p. 49.10-11.
10. 'Mens itaque humana prius haec quae facta sunt, per sensus corporis experitur.'Augustine 1894,
IV, 32; pp. 129.27-130.2. '... anima cum sentit in corpore, non ab illo aliquid pad, sed in ejus
passionibus attentius agere, et has actiones... non earn latere; et hoc totum est quod sentire
dicitur'. De musica, VI, 10; PL, 32, col. 1169.
11. 'Nullo modo igitur anima fabricatori corpori est subjecta materies.' De musica, VI, 5, 8; PL, 32,
cols. 1167-8.
one and the same corporeal thing in sensation (as singular) and in thought
(as universal). But only singulars existed. Universals, which from this
viewpoint were held to coincide with Platonic Ideas, did not exist at all.
They were only thought about. The meaning was plain, though Boethius
explicitly made the reservation that this Aristotelian view did not find
particular favour with him. 1 2 In spite of his own Neoplatonic leanings, he
transmitted a doctrine that gave ideas no real existence, and in the case of
secondary substances far less substantiality than singulars. This was the
opposite of the Neoplatonic gradation in which the ideas had greater
reality than sensible things. Boethius, with Porphyry before him, had said
enough to leave the question controversial, with the Neoplatonic position
as the establishment and the Aristotelian viewpoint as the troublemaker.
The Neoplatonic - Aristotelian controversy of the early Middle Ages

The eleventh and twelfth centuries, lacking Aristotle's Metaphysics and De
anima, were unable to resolve the controversy. At one extreme there arose a
nominalism that is known today only through the rebuttals by its en-
trenched opponents. It accorded universal substances no reality other than
the air emitted by the voice speaking about them. 1 3 O n the other hand
Anselm upheld the Augustinian tradition in a mildly expressed form, with
God as the light of minds, and sense-experience as a requisite for the image
from which the mind gave birth to the human concept (verbum).1* With
Abelard and others there was further concentration on the concept, with
universals assessed as mental expressions (sermones, conceptus) abstracted
from singulars. 15 The influential theologians of the school of Chartres,
12. ' . . . singularitati et uniuersalitati, unum quidem subiectum est, sed alio modo uniuersalc est, cum
cogitatur, alio singulare, cum sentitur in rebus in quibus esse suum habit.... sed Plato genera et
species cctcraque non modo intellegi uniuersalia, ucrum etiam esse atque praeter corpora sub-
sisted putat, Aristoteles uero intellegi quidem incorporalia et uniuersalia, sed subsistere in
sensibilibus putat; . . . studiosius Aristotelis sententiam executi sumus, non quod earn maxime
probaremus, sed quod hie liber ad Praedicamenta conscriptus est, quorum Aristoteles est auctor'.
Boethius 1896, In Isag. Porph., editio secunda, II, 11; p. 167.4-20.
13. ' . . . dialecticae haeretici, qui non nisi flatum vocis putant universales esse substantias'. Anselm
1946e, De incarn. verbi, 2; p. 9.21-2. Cf. ibid., prior rccensio, 4; Anselm 1946,1, p. 285.4-5. 'Alius
ergo consistit in uocibus; licet hec opinio cum Rocelino suo fere omnino iam euanuerit.'John of
Salisbury 1929, II, 17; p. 92.1-2.
14. 'Quanta namque est lux ilia, de qua micat omne verum quod rationali menti lucet!' Proslog., 14;
Anselm 1946,1, p. 112.5-6. 'Vere, domine, haec est lux inaccessibilis, in qua habitas et tamen
quidquid video, per illam video.' Ibid., 16; p. 112.20-2. "... non nascitur verbum cogitatae rei ex
ipsa re ... sed ex rei aliqua similitudine vel imagine ...' Monol., 62; Anselm 1946,1, p. 72.15-16.
15. 'Eadem namque res ab universali nomine et particular! continetur et hoc loco hoc verbum
"subsistit" de rebus ad sermonem transfertur per adiunctionem horum nominum: genus et
species, quae sermonibus data sunt.' Peter Abelard 1919—27, p. 525.33—6. Cf. John ofSalisbury
1929.Il, 17; pp. 92.2-93.2.
though immersed in the Neoplatonism of Augustine and Dionysius the

pseudo-Areopagite (fifth century A.D. ?), tended in their commentaries on
Boethius to describe universals as somehow abstracted from singulars by
the activity of the intellect.16 John ofSalisbury, himself inclined towards
considering universals as human concepts, recorded with facile humour the
protracted efforts at reconciling the long-dead Plato and Aristotle, who
had not been able to agree when alive.17
William of Auxerre and the material intellect

In the first quarter of the thirteenth century William of Auxerre was in full
accord with the Aristotelian doctrines as transmitted by Boethius.18 Yet he
could neither accept sense experience as the sole source of philosophical
knowledge, nor believe that Aristotle had done so. 19 Illumination,
William reasoned, was necessary for religious faith, as well as for the first
principles of the natural law and of the intellectual disciplines.
Accordingly, he retained the doctrine of the existence of all things in the
soul's interior light, as outlined in Augustine's De magistro, in connection
with the notion of study in general.20 In parallel fashion he accepted
Aristotle's assertion (De an., Ill 8,43 i b 2l) that the soul is in a way all things.
In both cases, William claimed, the manner of expression was meta-
phorical. As for Augustine, we existed in God from eternity only poten-
tially, insofar as we exist and live in virtue of the divine causality. And the
meaning of the Aristotelian dictum is that the thing known is potentially in
16. ' . . . substantiarum indiuiduarum uniuersalia quedam sunt, que ab ipsis indiuiduis humana ratio
quodammodo abstrahlt... Id est: res uniuersales intellectus ex quibuslibet particularibus sumit.'
Gilbert of Poitiers 1966, Contra Eut., Ill, 31; p. 278.96—2. 'Sed intellectus uniuersalium rerum
sumptus est ex particularibus.' Thierry of Chartres 1971, Contra Eut., Ill, 17; p. 239.5-6.
'... inductio est ex multis similibus particulatim inductis ad universalia progressio'. Clarembald
of Arras 1965, In Boeth. de Trin., II, 41; p. 123.25-6.
17. '... ut componerent inter Aristotilem et Platonem, sed eos tarde uenisse arbitror et laborassc in
uanum ut reconciliarent mortuos qui, quamdiu in uita licuit, dissenserunt'. John of Salisbury
1929, H, 17; P- 94-23-6.
18. 'Dicimus ergo quod sensibile et intelligibile tantummodo différant ratione . . . genera et species
non sunt ideae rerum, sicut dicebat Plato, sed sunt idem quod et ipsae res, sicut bene dicit
Boethius.' William of Auxerre 1500, Summa aurea, I, q. add., a. 2; f. 33*1.
19. 'Quod objicitur quod omnis scientia inchoatur a sensu. Dicimus quod hoc est falsum secundum
Augustinum. nee Aristoteles dicit illud.' William of Auxerre 1500, III, 7, c. 1, q. 4; f. i54"2.
20. '... sicut dicit Augustinus, fides est illuminatio mentis, quia illuminatur a prima luce sive a vera
luce ad videndum bona spiritualia.' William of Auxerre 1500, III, 3, c. 2, q. 3; f. 135Ί. '... in
intellectu angeli et animae impressa est notitia boni et ibi relucet prima Veritas. In luce primae
veritatis vident animae et angeli principia juris naturalis, sicut dicit beatus Augustinus in libro
Soliloquiorum, et etiam principia facultatum.' Ibid., II, 5, c. 1, q. 3; f. 47 v i. '...sicut dicit
Augustinus in libro De magistro; et quidam philosophus ait, quod nihil aliud est Studium quam
supernae et internae illustrationis expeoatio.' Ibid., II, 12, q. 2; f. 66'2.
the knower. 2 ' But in accord with the Aristotelian setting of William's
position, the material intellect had to receive its species from the species of
corporeal things, that is, from their sensible representations. 22
The expression 'material intellect' goes back to about A.D. 200, when
Alexander of Aphrodisias (De an., 81.24-5) used it to designate the Aristo-
telian potential (or passible) intellect (pathëtikos nous - Aristotle, De an., Ill
5, 430*24; Latin, intellectus possibilis) and justified the phrasing on the
ground that what receives anything can be said to function as matter for it.
The expression had been transmitted through the Islamic writers. The
material intellect received the forms abstracted from sensible things, forms
that were by William's time called by the Augustinian term 'species' (see
n. 7 above). The existence of things in the Augustinian intelligible world
was being aligned with their potential existence in the soul's material
intellect, and in each case 'existence' was regarded as metaphorical. In this
way Augustine and the Boethian Aristotle were finally reconciled. Yet
thinkers still felt a theoretical need for the doctrine of divine illumination,
an illumination that was by then understood as the divine causation of
existence and life.
William of Auvergne and the active intellect

William of Auxerre's interpretation of the material intellect leads naturally
to the Aristotelian doctrine of the intellect that was causal and substantially
actual (De an., Ill s, 430*10-23). It had been illustrated by the simile of
light, and had the function of actualising the potentialities of the material or
potential intellect. To the medievals it became known as the active (or
agent) intellect. William of Auvergne, the outstanding theologian of the
second quarter of the thirteenth century, claimed that Aristotle was forced
to invent it in trying — unsuccessfully — to accommodate the Platonic Ideal
world within his own theory of knowledge. It was rejected outright by
William. It would destroy moral responsibility as well as one's own
21. 'Secundum autem quod res dicuntur esse in Deo per cognitionem fuerunt res in Deo ab aeterno,
et secundum hunc modum dicuntur unum esse in Deo, propter unam ideam qua omnia
representantur. Secundum quem tropum dicit Aristoteles quod anima quodammodo est omnia.'
Ibid., II, 5, c. 1, q. i; f. 47r2. Cf. 'non aliter in eo sumus nisi quod id opcratur unde vivimus et
sumus. Secundum hunc modum non fuimus in Deo ab aeterno, scilicet per causam, nisi
potentialiter.' Ibid. 'Amor enim et notitia sunt in anima cognoscente tanquam cognitum in
cognoscente, propter assimilationem cognoscentis ad cognitum, sicut dicit Aristoteles quod
anima quodammodo est omnia.' Ibid., II, 10, c. I, q. 6; f. 59Ί. But in Augustine: 'haec in anima
existere.' Augustine, De Trin., IX, 4, 5; PL, 42, col. 963.
22. ' . . . intellectus enim materialis recipit species specierum rerum corporalium'. William of Auxerre
1500, IV, tr. ult., de suffragiis, q. 4; f. 305*2.
initiative in study and invention. The material intellect is the essence of the
soul and leaves no need for any other intellect. It knows singulars as well as
universals and makes judgements about them, and it enjoys freedom in
individual actions. 23 Yet it does not become a changing subject for material
forms, even though the sensible things have the role of occasions on which
it is aroused. The material intellect was equipped by nature to do its own
work of thinking and of thereby producing the species in which it knows
sensible things, the species being only similitudes by which the things
themselves are known. 2 4 The notion of a species produced by thinking
itself instead of being impressed on the intellect by a distinct agent was a
significant insight that would have its history. For the production of those
species the material intellect needed no further illumination from on high,
nor action upon itself by the sensible things that merely occasion its
knowledge. But besides being joined by sense experience to the corporeal
world, man in his highest aspect is joined to the divine order, with
illumination from above in order to know the first principles in the moral
and intellectual orders. 25
Albert the Great and the reception of forms

Albert the Great, whose writings date from at least 1245 onwards, bene-
fited from closer interpretation of Aristotle's De anima. Sense experience,
according to Albert, was the origin of all universal knowledge. But to
abstract the universal from the image of the sensible particular an active
23. 'Causa autem, quae coegit ipsum hanc intclligentiam ponere, fuit positio Piatonis de formis, sive
de mundo specieruni.' William of Auvergne 1674, De universe, II-II, 14; I, p. 821b (A).
'... figmentum igitur est tantum, et vanissima positio intellectus agentis'. Ibid., De an., VII, 4;
Suppl. p. 209b. 'Quod si dixerit illam fieri ab intelligentia agente, quod est illuminatrix . . . neque
Studium, neque inventio aliquid erit.' Ibid., V, 7; p. 122a. '... alioquin ut pracdixi nihil est in libera
potestateejusdem ex operationibus'. Ibid., VU, 4; p. 208a. '... res ipsas sibi exhibet, et praesentat,
et earum species ipse sibi in scmetipso format'. Ibid., II—III» 4; p. ioi8b(EF). 'Quapropter ipsa
anima non erit nisi intellectus matcrialis solus Manifestum igitur est ipsi quod essentia sua non
est nisi intellectus matcrialis; et propter hoc intellectus agens, sive formalis, nee ipsa essentia eius
est, nee de ipsa.' Ibid., VII, 3; p. 206b. '... destruxi errorem eorum qui dixerunt intellectivam
virtutem esse apprehensivam univcrsalium tantum'. Ibid., VII, 1; p. 203a.
24. 'Quemadmodum audivisti in praecedentibus de intellectu, qui cum nihil patitur a formis ma-
terialibus, et sensibilibus, per occasionem tamen alienac passionis, scilicet quae fit in aliquo ex
organis sensuum, exeunt ab ipso novi cogitatus...' Ibid., II-II, 76; p. 929b (C). '... quia non ipsae
imaginantur aut intelligantur, sed magis res, quarum sigillationes et similitudines sunt'. William
of Auvergne 1674, De retrib. sand.; I, 318a (H).
25. '... ponendum est animam humanam velut in horizonte duorum mundorum naturaliter esse
constitutam et ordinatam'. Ibid., VII, 6; p. 211b. '... creator ipse est liber naturalis et proprius
intellectus humani. Ab illo igitur fiunt impressiones de quibus agitur, et inscriptiones signorum
antedictorum in virtute nostra intellectiva, et ipse est lumen ...' Ibid. Cf. 'lucem veritatis per
ipsum lumen eius, quod est vcrus fides ac salutaris'. William of Auvergne 1976, De Trin., 26;
p. 150.39-40·
intellect was indeed required, for a material thing could not bring about a
form in the soul. 2 6 Since the soul itself performed the act of intellection, the
active intellect as an intrinsic faculty of the soul had to act within the
potential (material) intellect. It could not be something extrinsic. It
possessed n o intelligibles of its own; it had to abstract them from sin
gulars. 2 7 Correspondingly the potential intellect was a tabula rasa for all
intelligible objects. The species it received were in it not as in a subject but
rather as light from the active intellect. 2 8 Accordingly n o third thing was
constituted by the cognitional union of knower and thing known, as
would have been the case in a material reception of form. 2 9 The light of the
agent intellect, however, was not self-sufficient; it had to be supplemented
by the uncreated intelligible light. 3 0 Moreover, knowledge of God and self
was naturally present in the soul, apart from sensible images. 3 1
Albert's insistence on the reception of forms for the cognition of sensible
things, and his understanding of it as different in kind from the reception of
forms into a subject, are profoundly Aristotelian. But his interpretation of
this reception in terms of light, as well as his emphasis on the dependence
of the active intellect's light upon a higher light and his exemption of
knowledge of God and self from its activity, show fingering traces of the
Augustinian illumination. Albert recognises that cognition gives a new
20. ' . . . unumquodque phantasms est particulare determinatum: et ideo necesse est ponere agens
universale esse in intellectu.' Albert the Great 1890-9, Summae de creat., II, 55,1, adzm; XXXV,
p. 456a. ' . . . ita forma non habet a materia quod efficiatur in anima, sed ab actu intelligentiae
agentis'. De intellectu et intellig., II, tr. un., 1; 1890-9, IX, p. 504b.
27. 'Similiter dicimus intellectum agentem humanum esse conjuncture! animae humanae, et esse
simplicem, et non habere intelligibilia, sed agere ipsa in intellectu possibili ex phantasmatibus,
sicut expresse dicit Averroes in commento libri de Anima.' Summae de creat., II, 55, 3, Solut.;
1890-99, X X X V , p. 466b. ' . . . intellectus vero agens non est extrinsecus animae intellectivae, sed
est de constitutione ipsius'. Ibid., 2, ad i m ; p. 460a.
28. ' . . . se habet ad omnia intelligibilia sicut tabula rasa'. De intellectu et intellig., II, tr. un., 4; 1890-9,
IX, p. 508b. ' . . . hoc modo mover lux agentis species intelligibiles ad possibilem, et sunt in ipso
non sicut in subiecto, sed potius sunt in ipso sicut lumen agentis intentionatum intentione rei...'
De unit, int.. Ill, z, ad 27m; 1890-9, XVII, ι, p. 29.38-41.
29. 'Et ideo cum fiat unum tertium a componentibus, quando aliquid componitur ex materia et
forma, non fit sic unum; quando componitur intentio universalis cum intellectu possibili, sed fit
unum actu, quia idem est actus intelligibilis qui est actus possibilis intellectus.' De unit, int.. Ill, 1 ;
1890-9, XVII, 1, p. 23.15-20. Cf. Averroes 1953, Comm. magn. in De an.. Ill, 5; p. 404.506-7.
30. 'Lux intellectus agentis non sufficit per se, nisi per applicationem lucis intellectus increati.' / Sent.,
B, 5, Solut; 1890-9, XXV, p. 60a.
31. 'Unde cum non possit sine phantasmate secundum condition«« hujus vitae intelligere nisi ea
quae sunt per cssentiam in ipso, ut Deum et se,..." De coel. hierarch-, II, 2, dub. 1, Solut.; 1890-9,
XIV, 27a. ' . . . quorum utriusque notitia insita est animae naturaliter, ut dicit Augustinus'. I Sent.,
3, G, 20, ad 2m; 1890-9, XXV, p. 120a. Yet Albert can maintain the sensible origin o f all our
knowledge: ' . . . omnis nostra scientia oritur ex sensibilibus'. De an.. Ill, 2, 19; 1890—9, VII, 1,
p. 206.70.
and spiritual existence to the thing known. He mentions Avicenna's

pervasive doctrine that one and the same essence has material or natural
being in the singular thing, spiritual being in the mind, and simple being in
itself.32 Yet he does not develop this insight into a full-fledged explanation
of cognition in terms of being rather than of illumination.
Roger Bacon's Aristotelian explanation of Augustine

In the view of Roger Bacon, the soul's active intellect thought by means of
innate exemplar ideas or species without being served by sense cognition,
while the potential intellect, directed towards lower things, depended
upon what was given it by the senses.33 Bacon, however, maintained the
doctrine of the overall illumination described by Augustine. From that
viewpoint he could deny that the active intellect was part of the soul, since
an agent had to be substantially other than the patient.34 He opposed the
doctrine that the same form can be the ground of both the being and the
knowledge of the thing,35 seeming content with the stand that the illumi-
nation itself allowed both intellects to see their objects. But he explicitly
ranked the individual higher than the universal.36
In Bacon the notions of active and potential intellect are found accom-
modated to the Augustinian higher and lower reason, on the one hand, and
on the other hand to the framework of the Augustinian illumination. The
aim is still to explain Augustine by means of Aristotle. The Augustinian
strain in Bacon appears also in his view that philosophy is not only in
32. 'Et ideo una et eadem est essentia in se et in anima et in singulari: sed in anima secundum esse
spirituale, in singulari secundum esse materiale et naturale, in se autem in esse simplici.' De
praedicabiiibus, II, 6; 1890—9,1, 35a. Cf. ' . . . prout est essentia quaedam absoluta in seipsa, et sic
vocatur essentia, et est unum quid in se existens, nee habet esse nisi talis essentiae'. De intellectu el
inteilig., 1,2, 2; 1890-9, IX, 493a. See Avicenna: 'Et est sicut esse proprium rei.' Avicenna 1508,
Metaph., I, 6; f. 72*1. 'Essentiae vero rerum aut sunt in ipsis rebus: aut sunt in intellectu.' Ibid.,
Logica, I; f. 2'2.
33. '... et hec vocatur intellectus agens, et hec non intclligit rem per administrationem sensuum, set
per exempla sibi innata, confusa tamen; . . . Alter est intellectus possibtlis ... et hic intclligit per
administrationem sensuum, de quo dicitur "nichil est in intellectu quin prius fucrit in sensu."'
Roger Bacon 1905-40, Quaest. XI. Metaph., VII, p. 110.4-13. '...quedam est principium
cognoscendi tantum, ut ydea vel species alicujus rei; . . . tertia est principium operandi ct
cognoscendi, et hec est exemplar.' Ibid., p. 111.22-7. Cf. Roger Bacon 1897-1900, Op. maj.,11,};
I, pp. 38-9.
34. '... agens semper est aliud a materia et extra earn secundum substantiam . . . et sic nullo modo
sequitur quod intellectus agens sit pars animae'. Op. maj., II. 5; 1897-1900, 1, pp. 40-1.
'Augustinus... vult in pluribus locis quod non cognoscimus aliquam veritatem nisi in veritate
increata et in regulis aetcrnis.' Ibid., p. 41.
35. ',.. licet sit aliquo modo principium cognoscendi ipsam rem, non oportet quod sit principium
essendi tamquam vera forma realis'. Quaest. super Lib. de causis, 11; 1905-40, XII, p. 63.9-11.
36. 'Item, cum probatum est quod Individuum est prius secundum naturam, melius est sine com-
paracione quam universale.' Comm. nat., I, 2, 2, 7; 1905-40, II, p. 95.29-31.
harmony with sacred theology, but necessary for it and engaged in on its
account.37
Bonaventure's Aristotelian Augustinianism

Bonaventure likewise maintained that philosophy was necessary for the
pursuit of theology, yet care had to be taken to avoid deception by it.
Without faith philosophy had in fact held as impossible some things that
are true; accordingly philosophy was not self-sufficient. As with Augustine
and Anselm, understanding requires faith.38
On its own level according to Bonaventure, human reason travels the
route of sense experience, with the soul impressing upon itself the species
abstracted from material things.39 Yet, as with Albert, knowledge of God
and soul were exempt from that dependence. Knowledge of God did
require a species, but impressed in a way different from abstraction.40 In
every case, however, the human intellect required a concept that emanated
from itself41 Both the impressed species and the engendered concept were
likenesses of the thing known. Things had existence in the divine exemplar
ideas, in the created intellect, and in the world. In the divine exemplar they
had being only as likenesses. In the world they exist in their own entity, a
way of existence that was 'truer' and better for them. 42 But in their
37. 'Omnia quae tracto sunt propter thcologiam.' Comm. not., I, 2, 2, 7; 1905—40, II, p. 95.9—10.
' . . . manifestum est quod philosophia necessaria est legi divinacetfidelibusincagloriantibus'. Op.
maj., II, 7; 1897-1900,1, p. 43.
38. 'Sed quia ista scripta adducunt philosophorum verba, necesse est, quod homo sciat vel supponat
ipsa.' Bonaventure 1882-1902, In Hex., XIX, 10; V, pp. 421-2. "Philosophi autem habent pro
impossibili quae sunt summe vera.' Ibid., III. 4; p. 343b. ' . . . magistri cavere debent, ne nimis
commendent vel apprctientur dicta philosophorum. . . . ad aquas philosophorum, in quibus est
aeterna deceptio'. Ibid., XIX. 12; V, p. 422a. 'Nisi credideritis, non intelligetis. Hunc ordinem
ignoraverunt philosophi.' Serine IV('Christus Unus Omnium Magister'), no. 15; 1882-1902, V,
p. 571.
39. 'Ratio . . . prout iudicio proprio relicta est, et sic procedit inspiciendo ad naturas et causas
inferiores; acquirit enim scientiam per viam sensus et experientiae. ' Bonaventure 1882-1902a, / /
Sent., 3 0 , 1 , 1 , Concl.; II p. 716a. 'Ideo anima non cognoscit rem, nisi speciem eius et formam sibi
imprimat; et hoc non potest esse, nisi ilia abstrahatur a materia.' Ibid., 17, 1, 2, ad 4m; p. 415b.
40. 'Ncccssario enim oportet ponerc, quod anima novit Deum et se ipsam ct quae sunt in sc ipsa, sine
adminiculo sensuum exteriorum.' II Sent., 39, 1, 2; 1882-1902a, II. p. 904b. 'Deus est praesens ipsi
animae et omni intcllectui per veritatem; . . . dum cognoscitur ab intellectu, intellectus in-
formatur quadam notitia, quae est velut similitudo quaedam non abstracta, sed impressa.' / Sent.,
3, 1, art. un., ad 5m; 1882-19023,1, p. 70a.
41. 'Mens autem concipit intclligendo, et intelligcndo aliud concipit simile alii, intelligendo se
concipit simile sibi, quia intelligcntia assimilatur intellecto.' / Sent., 27, 2, art. un., q. 1, Resp.;
i882-i902a, V, p. 482b. Cf. ' . . . verbum est quod émanât a mente per modum conceptionis'.
Ibid., arg. 4; p. 481a.
42. ' . . . triplex est existentia rerum scilicet in exemplari aeterno, et in intellectu creato, et in ipso mundo.
In exemplari aeterno et in intellectu creato sunt res secundum similitudinem; in ipso mundo
existence in the divine light they were able to illuminate the human mind
in the way described by Augustine in De magistro. That illumination was
required for every truth. 43 Though rejecting an explanation of cognition
solely in terms of causal influx, Bonaventure tended to align it with the
divine concurrence in human actions.44 The illumination of faith is dif-
ferent in kind from the illumination of philosophy, and can accordingly
allow simultaneous belief and knowledge about the same thing.45
In Bonaventure, then, the doctrine of illumination continued in its full
Augustinian range and power. But Bonaventure had integrated it into an
Aristotelian framework of active and potential intellect and paralleled it
with the causality through which being is imparted by the first cause. Yet it
remained basically an account of cognition in terms of light rather than of
existence. The existence of things in their cognitional likeness was ex-
istence only from a particular viewpoint, and was not made to carry the
burden of explaining cognition considered as such.
secundum entitatem proprium. . . . verius est unaquaeque res in proprio genere quam in
Deo . . . similitude rei verius et nobili us esse habet in Deo, quam ipsa res in mundo ratione eius quod
est; quia est ipse Deus". / Sent., 26,2, 2, Resp.; 1882-19023, V, 625b. ' . . . eadem principia, quae
sunt principia essendi, sunt principia cognoscendi; sed tamen principia essendi conferunt esse per
se ipsa, sed cognitionem non conferunt per se, sed per suas similitudines'. Ibid., p. 626b. Cf. "... in
Deo solum dicitur esse secundum quid... sed in proprio genere simpliciter'. Ibid., arg. I contra;
p. 625a. 'Item, verius est res, ubi est secundum propriam entitatem, quam ubi solum secundum
similitudinem.' Ibid., arg. 3 contra; p. 6253b.
43. ' . . . ct constat secundum Augustinum et alios sanctos, quod "Christus habens cathedram in caelo
docet interius"; nee aliquo modo aliqua Veritas sciri potest nisi per illam veritatem.' In Hex-, 1,13;
1882-1902, V. p. 33 lb. 'Lux ergo intellectus creati sibi non sufficit ad certam comprehensionem
rei cuiuscumque absque luce Verbi aeterni.' Sermo IV, 10; 1882-1902, V, pp. 569-70.
44. "... in opcre vero, quod est a creatura per modum imaginis, cooperatur Deus per modum talionis
moventis; et tale est opus certitudinalis cognitionis, quod quidem non est a ratione inferiori sine
superiori'. De scientia Christi, IV, Resp.; 1882-1902, V, p. 24a. Cf. ' . . . sed Deus sic est causa
essendi, quod nihil potest ab aliqua causa effici, quin ipse se ipso et sua aetcrna virtute moveat
opcrantem: ergo nihil potest intelligi, quin ipse sua acterna veritate immediate illustrer intel-
ligentem'. Ibid., arg. 24; p. 19b. 'Item omne ens in potentia reducitur ad actum per aliquid existens
in illo genere... restât igitur, quod quidquid anima intelligens apprehendit, per aliquid quod est
supra animam appréhendât. Sed supra animam non est nisi Deus.' Ibid., arg. 32; p. 20b. 'Alio modo,
ut intelligatur, quod ad cognitionem ccrtitudinalem necessario concur rit ratio aeterna quantum
ad suam influentiam, ita quod cognoscens in cognoscendo non ipsam rationem aetemam attingit,
sed influentiam eius solum. - Et hie quidem modus est insufficiens secundum verba bcati
Augustini.' Ibid., Resp.; p. 23a.
45. '... sie nihil impedit, unum et idem secundum alium et alium cognoscendi modum esse infra et
supra; ex ita scitum et creditum.' Ill Sent., 24, 3,1, ad 4m; 1882- 1902a, III, p. 524a. Cf. ibid., ad 3m.
See also: 'unde aliquis credens, Deum esse unum, creatorem omnium, si ex rationibus necessariis
ineipiat idem nosse, non propter hoc desinit fidem habere; vel si etiam prius nosset, fides
superveniens talem cognitionem non expelleret, sicut per cxpericntiam patet'. Ibid., Resp.;
p. 523a. 'Tertium lumen, quod illuminât ad veritates intelligibiles perscrutandas, est lumen
cognitionisphilosophicae.' De reduet. art., no. 4; 1882-1002, V, p. 320b. 'Quartum autem lumen,
quod illuminât ad veritatem salutarem, est lumen sacrae Scripturae.' Ibid., no. 5; p. 321b.
Aquinas and the rejection of illumination

In Thomas Aquinas, on the other hand, cognition meant the existence of a
thing in the knower. 4 6 In both sensation and intellection there was
thorough identity of thing and cognitive agent, even though they were
different in real existence. 47 This meant, in a recognisably Avicennian
framework (see n. 32 above), that one and the same thing could have
existence in reality and existence in cognition. Just in itself, however, the
thing's nature had no being whatever. N e w in this regard was Aquinas'
insight that a thing becomes known intellectually through two different
but always concomitant acts of the human mind. It is apprehended in
respect of its nature through an incomplex concept expressed by a single
word (e.g., 'Socrates', 'man') and, simultaneously, in respect of its being
through a synthesising act of apprehension expressed in a proposition (e.g.,
'Socrates exists', 'Socrates is a man', 'Socrates is himself'). 48 If the nature
common to both ways of existing and known through the incomplex
concept had any being at all of its own, it could not remain the same thing
substantially under the different existential actualisations. Yet the common
nature was what was predicated of all the singulars. With no being of its
own, it could not be apprehended immediately; it had to be inferred. In
that setting, obviously, absolute primacy was accorded to existence. When
existing in the human mind, the nature was universal, even though the
concept expressing it was something individual that was associated with a
plurality of singulars. 49 Like other medieval writers, however, Aquinas
46. ' . . . cognitio non dicit effluxum a cognoscente in cognitum, sicut est in acribus naturalibus, sed
magis dicit existcntiam cogniti in cognoscente'. Thomas Aquinas 1882-, De ver., II, 5, ad 15m;
XXII, p. 64b. '...secundum hoc cognitio perficitur quod cognitum est in cognoscente non
quidem materialiter sed formaliter". Aquinas I954d, In Lib. de causis, 18; p. 101.14-16.
47. 'Secundum autem quod intelligit res alias, intellectum in actu fit unum cum intellectu in actu,
inquantum forma intellecti fit forma intellectus, inquantum est intellectus in actu, non quod sit
ipsamet essentia intellectus... quia essentia intellectus manct una sub duabus formis, secundum
quod intelligit res duas successive' Aquinas 1874-89, IV Sent., 49. 2, I, ad 10m; XI, 486a.
48. 'Cum in re duo sunt, quidditas rei et esse ejus, his duobus respondet duplex opcratio intellectus.' /
Sent., 38,1, 3, Resp.; 1874-89, VII, p. 468b. Cf. 'Sed intellectus noster, cujus cognitio a rebus
oritur, quae esse compositum habent, non apprehendit illud esse nisi componendo et dividende.'
Ibid., ad im; p. 469a.
49. ' . . . sed uerum est dicere quod homo, non in quantum est homo, habet quod sit in hoc singulari
uel in illo aut in anima. Ergo patet quod natura hominis absolute considerata abstrahlt a quolibet
esse, ita tamen quod non fiat precisio alieuius eorum. Et hec natura sie considerata est que
predicatur de indiuiduis omnibus.' De ente et essentia, III; 1882-, XLIII, p. 374.65-72. 'Et quamuis
hec natura intellects habeat rationem uniuersalis secundum quod comparatur ad res extra
animam, quia est una similitudo omnium, tamen secundum quod habet esse in hoc intellectu uel
in illo est quedam species intellects particularisa Ibid., p. 375.102-7.
allowed the nature itself to be called a universal in a derived way, since it

grounded the genuine universal that existed in the mind. 50
On this view, existence in cognition was neither metaphorical nor
dependent upon a particular viewpoint. Rather, both cognitional and real
being were authentic ways of existing. Existence was undertood in various
senses, all genuine, but in the order of primary and secondary instances.
Primary was the thing's existence in God, secondary its existence in itself;
dependent on both was its existence in human cognition.51 But the thing
confronted human intellection in the existence it had in itself, and not in
the existence it had in the divine light. In consequence the primary
existence in God did not ground a basic account of cognition in terms of
illumination.52 Though Aquinas like all his contemporaries made use of
the traditional terminology of light, he developed his explanation of
cognition in thoroughgoing terms of being. Philosophically God was
reached only through reasoning based on the existence of sensible things in
themselves and not on the basis of any Anselmian argument. 53 In this
framework the sharp contrast between the species received by the intellect
and the further species engendered by it became explicit.54 The species was
indeed impressed by a really existent thing, yet the thing's nature had to be
expressed in an incomplex concept that did not attain the being of the
thing.
For Aquinas theology was the highest science. Its needs were served by
the other sciences. It acknowledged the independent principles and meth-
ods of those sciences, yet had the role of judging through its own principles
the validity of their findings. One could not have simultaneous faith and
50. ' Uno modo potest dici universale ipsa natura communis, prout subjacet intentioni universalitatis.
Alio modo secundum se.' Aquinas 1959, In IIDean., lect. 12, no. 378; p. 132b.
51. 'Unde in Deo est per esse increatum, in se autem per esse creatum, in quo est minus de veritate
essendi quam in esse increato.' / Sent., 36,1, 3, ad2m; 1874-89, VII, 434ab. 'Unde uniuscuiusque
naturae causatae prima considerario est secundum quod est in intellectu divino; secunda vero
consideratio est ipsius naturae absolute; tenia secundum quod habet esse in rebus ipsis, vel in mente
angelica; quarta secundum esse quod habet in intellectu nostro In his ergo illud quod est prius,
semper est ratio posterions.' Aquinas 1956, Quodl., VIII, 1,1, Resp.; pp. 158-9.
52. 'Quia tamen praeter lumen mtellectuale in nobis, exiguntur species intelligibiles a rebus acceptae,
ad scientiam de rebus materialibus habendam; ideo non solum per participationem rationum
aeternarum de rebus materialibus notitiam habemus.' ST, I, q. 84, a. 5, Resp.; 1882-, V, p. 322b.
53. 'Nee potest argui quod sit in re, nisi daretur quod sit in re aliquid quo maius cogitari non potest:
quod non est datum a ponentibus Deum non esse.' ST, I, q. 2, a. I, ad 2m; 1882-, IV, p. 28b.
54- 'Unde necesse est quod species intelligibilis, quae est principium opcrationis intellectualis, différât
a verbo cordis, quod est per opcrationcm intellectus formatum; quamvis ipsum verbum possit dici
forma vel species intclligibilis, sicut per intellectum constituta, prout forma artis quam intellectus
adinvenit, dicitur quaedam species intclligibilis.' Quodl., V, 5, 2, Resp.; 1956, p. 103b.
knowledge about the same thing, since faith involved assent without
evidence of truth, while knowledge required that evidence. 5 5 The nature
of the acts of faith and of knowledge themselves was the only reason
offered by Aquinas for the conclusion. There was no further doctrine of
illumination to make possible simultaneous presence of the two acts.
Henry of Ghent and the survival of illumination
In the final quarter of the century Henry of Ghent faced the traditional
sceptical attacks - as transmitted through Cicero and Augustine - on the
reliability of sense experience. He addressed himself to the Augustinian
question whether any pure (sincera) truth, in the sense of truth unmixed
with error, could be had from the senses. His answer was that it could be
had by looking at the eternal exemplar ideas, attainable by men in the
present state at the will of God. Divine illumination was accordingly
required. 5 6 But the one species received in sensation sufficed to provide an
object in the imagination from which the universal could be abstracted. 5 7
The intelligible species so abstracted was but the means of knowing, and not
the object of the cognition. 5 8 In the object the essence of the thing enjoyed
a being of its own (esse essentiae) that was distinguished intentionally from
55. "... theologia debet omnibus aliis scientiis imperarc et uti his quae in eis traduntur'. / Sent., Prot.,
1, 1, Solut.; 1874-89, VII, p. 5b. 'Et ideo non pertinet ad earn probarc principia aliarum
scientiarum, sed solum iudicare de eis.' ST I, q. ι, a. 6, ad2m; 1882-, IV, p. 18a.'Ilia autem videri
dicuntur quae per seipsa movent intellectum nostrum vel sensum ad sui cognitionem. Unde
manifestum est quod ncc fides nee opinio potest esse de visis aut secundum sensum aut secundum
intellectum.'ST, Ilallae.q. 1. a. 4,Resp.; 1882-, VIII, pp. 13-14. Cf. 'Sedsacra doctrina procedit
ex articulis fidei, qui non sunt per se nota.' ST, I, q. 1, a. 2, arg. 1; 1882-, IV, p. 8a.
56. ' . . . bene a sensibus sincera Veritas expetenda est, et hoc quantum ex puris naturalibus judicio
rationis in lumine puro naturali potest conspici: vel simpliciter judicio intellectus in claritate lucis
aeternae.'Henry of Ghent, 1520 Summa, 1,1, ai 2m; f, 2V, F. 'Sincera igitur Veritas ut dictum est
non nisi ad exemplar aeternum conspici potest.' Ibid., I, 2; f. 6Γ, Η. 'Ex puris igitur naturalibus
exclusa omni divina illustratione nullo modo contingit hominem scire liquidam veritatem.
... Nunc autem ita est quod homo ex puris naturalibus attingere non potest ad régulas lucis
aeternae, ut in eis videat rerum sinceram veritatem.... Sed illas Deus offert quibus vult: et quibus
vult substrahit.' Ibid., I, 2; f. η", LM.
57. ' . . . ex parte autem intellectus nostri est ab ipso intelligibili universali, quod seipso est praesens
intellectui in phantasmate actione intellectus agentis, propter quod seipso inclinât intellectum non
mediante specie, quam solum ponimus in sensu.' Henry of Ghent 1518, Quodl., XI, 5, Resp.; f.
45i r ,S.
58. 'Abstractio tamen non fit neque a specie impressa: quia intellectus speciei materialis impressionem
non recipit, quia vere esset alterabilis et transmutabilis sicut sensus . . . Neque fit abstractio ab actu
imaginandi eadem ratione, sed solum ab objecto imaginato, ut illud quod est sicut cognitum in
imaginativa cognoscente uno modo, sit ut cognitum in intellecto cognoscente alio modo.'
Quodl, IV, 21, Resp.; 1518, f. 137', I.
the being of its actual existence (esse actualis existentiae).59 This notion of
essential being allowed ontological reasoning to God's existence. 60
Though Henry utilized the distinction made by Aristotle (De an., Ill, 6,
43o 3 26- b 3i) between simple and complex intellection, and could readily
say that the thing known is in the knower and is one with the knower, he
did not show any tendency to explain cognition through existence as
Aquinas had done. 6 1 His effort was a continuation of the doctrine of
illumination as a requisite for explaining the truth of human knowledge
in the present state. Illumination was likewise strongly defended by
Franciscan theologians such as Matthew of Aquasparta (d. 1302) and Vital
du Four (d. 1327).
Duns Scotus and the common nature

With John Duns Scotus the human intellect and its object, under the divine
causality ordinarily required for the functioning of creatures, suffice for the
attainment of truth. Consequently no special illumination was required. 62
But the intellect's basic object had to be sufficiently common to ground the
universal, since the intellect could not be held to produce the universal
from the singular alone without disastrous results. The widely accepted
tenet that only singulars exist, while what is common springs just from the
intellect, would make everything distinct from everything else in exactly
the same utterly alienating way. It would provide no more ground for
abstracting something common from Socrates and Plato than from
Socrates and a line; it would make the universal a pure fiction. Prior to any
work of the human intellect the specific nature of the sensible thing had to
have its own type of unity and commonness outside the mind. In the thing
59. 'Et est hie distinguendum de esse secundum quod distinguit Avicenna in quinto in fine Meta-
physicae suae, quod quoddam est esse rei quod habet essentialiter de sc: quod appellatur esse
essentiae. Quoddam vero quod recipit ab alio: quod appellatur esse actualis existentiae.' Quodl., I,
9; 1518, f. 7', Y.
60. 'Ideo ex talibus conceptibus propositionum universalium contingit secundum Avicennam et
Augustinum intelligere et scire Deum esse, non ex via testificationis sensibilium.... iste modus
ortum sumit a cognitione essentiae creaturae.' Summa, XXII, 5; 1520, f. 134", DE.
61. 'Cognitione igitur intellectiva de re creata potest haberi duplex cognitio. Una qua praecise scitur
sive cognoscitur simplici intelligentia id quod res est. Alia qua scitur et cognoscitur intelligentia
componente ct dividente Veritas ipsius rei.' Summa, I, 2; 1520, f. 4", C.
62. 'Et ex isto apparet qualiter non est necessaria specialis illustratio ad videndum in regulis aeternis,
quia Augustinus non ponit in eis videri nisi "vera" quae sunt necessaria ex vi terminorum. Et in
talibus est maxima naturalitas - tarn causae remotae quam proximac - respectu effectus, puta ram
intellectus divini ad obiecta moventia, quam illorum obicctorum ad veritatem complexionis de
eis. ...quia termini apprchensi et compositi, sunt nati naturaliter causare evidentiam con-
formitatiscompositionis ad terminos.' Scotus 1950-, Ord., I, 3,1,4, no. 269; III, p. 164.13-165.2.
itself the nature had to be formally distinct from the individuating entity
(haecceitas). The common nature, understood in this way, was in fact the
first object of the human intellect. 63 Having its own distinct entity, it was
not identified with the singular in the manner required for predication. But
as grasped by the intellect it was actualised into the universal that was
predicated with the requisite identity. 6 4 Socrates is not the common
nature, humanity, but he is a man. This way of regarding the specific
nature as common in reality to the singulars was a remarkable innovation
with Duns Scotus. It did not involve a plurality of forms, for the common
nature pervades the forms themselves as well as the matter and the com-
posite; nor did it make possible the real existence of a non-individuated
nature. It meant, rather, that if, per impossibile, the individuating entities
could be removed, there would be nothing to separate the real humanity of
Plato from that ofother men. 6 5
For Scotus the nature could be known either as existent by intuitive
cognition, or without regard for existence by abstractive cognition. 66
Having its own proper entity it permitted valid reasoning to the existence
of God in a version of the Anselmian argument, and it furnished a concept
of being that was with requisite qualifications univocal to God and crea-
63. ' . . . ita omnia cssent acque distincta; et tunc sequitur quod non plus posset intellectus a Socrate et
Platone abstraherc aliquid c o m m u n e , quam a Socrate et linea, et esset quodlibet universale purum
figmentum intellectus'. Ord., II, 3, I, 1, no. 23; 1950-, VII, pp. 400.20-401.2. ' . . . et secundum
prioritatem naturalem est "quod quid est" per sc obiectum intellectus, et per sc, ut sic, con-
sideratur a metaphysico et cxprimitur per definitioncm'. Ibid., n o . 32; p. 403.8-10. Cf. ' P r i m u m
actualiter cognitum confuse, est species specialissima, cuius singulare efficacius et fortius primo
movet sensum.' Ord., I, 3, 1—2, no. 73; 1950-, III, 50.8-9.
64. 'Est ergo in re " c o m m u n e , " quod non est de se hoc, et per consequens ei de se non répugnât n o n -
hoc. Sed talc c o m m u n e non est univcrsalein actu, quia deficit ei ilia indifferentia secundum quam
completive universale est universale, secundum quam scilicet ipsum idem aliqua identitate est
praedicabilc de quolibet individuo, ita quod quodlibet sit ipsum.' Ord., II, 3, 1, 1, n o . 38;
1950- VII, pp. 407.20-408.3. Cf. 'Aliter dicitur. quod intellectus agens non causât universale, sed
intellectus possibilis considérai« illam quidditatem illimitatam, causât in eo universale.' Scotus
1891-5, Metaph., I. 6. n o . 7; VII, p . 74ab.
65. 'Et sicut compositum non includit suam entitatem (qua formaliter est "hoc") in q u a n t u m natura,
ita nee materia "in quantum natura" includit suam entitatem (qua est "haec materia"), nee forma
"in quantum natura" includit suam." Ord., II, 3, 1, 5-6, no. 187; 1950-, VII, p. 483.14-17.
'B-espondeo, si loquamur realiter, humanitas quae est in Socrate, non est humanitas quae est in
Platone, ct est realis differentia ex differentiis individualibus unitive contentis, inscparabilibus hinc
indc. Si autem circumscribamus differentiam hinc indc, sic ut nec natura intelligitur una maxima
unitate in se, sed tantum ilia unitate minori, quae est communis; sic nec est divisa ab humanitate
Piatonis divisione numerali, nec aliqua, quia non specifica,...' Metaph., VII, 13, no. 21; 189I—5,
VII, p. 421b.
66. 'Primam voco "abstractivam". quae est ipsius quiditatis, secundum quod abstrahitur ab
existentia actuali ct non-existcntia. Secundam, scilicet quae est quiditatis rei secundum eius
existentiam actualcm (vel quae est rei praesentis secundum talem existentiam), voco intel-
Icctionem intuitivam.' Ord., II, 3, 2, 2, no. 321; 1950-, VII, p. 553.6—11.
tures.6' As an object present in its common status before the mind's

intuitive or abstractive gaze, the nature continued to allow intellection to
be described in visual terms in keeping with the doctrine of illumination, a
version of which Scotus regarded as the 'common opinion' 6S of his day on
divine causality. The Scotistic common nature diverges radically from
that of Aquinas (De ente et essentia, c. II; pp. 20.2—23.7), for whom the
nature taken just in itself had no being at all and though when abstracted
precisively (i.e., in such a way as to be signified by an abstract term such as
'humanity') was not predicable of the singulars, nevertheless when ab-
stracted non-precisively (i.e., in such a way as to be signified by a concrete
term such as 'man') was exactly what was predicated, along with existence.
William Ockham and intuitive cognition

In the first part of the fourteenth century Durandus of Saint-Pourçain
(d. 1334) developed the view that the acts of sensation and intellection are
produced basically by the causes that created and engendered the knower,
rather than by the object.69 William Ockham followed this lead to the
extreme in concluding that by way of miracle there could be intuitive
cognition of something that did not exist. With special divine intervention
all that was absolutely required to produce the intuitive cognition was the
act on the part of the intellect. Accordingly the difference between ab-
stractive cognition (not naturally requiring the real existence of its object)
and intuitive cognition did not arise from the object but from the nature of
the acts themselves.70 Intelligible species in this setting were obviously
67. 'Per illud potest colorari ilia ratio Anselmi...'Onf., 1,2, 1.12, no. 137; 1950-, II, p. 208.16. 'Sed
ponendo illam positioncm quam posui ... de univocationc ends, potest aliquo modo salvari
aliquod esse primum obiectum intellectus nostri.' Ibid., I, 3, I, 3, no. 129; III, pp. 80.21-81.2.
'... dico quod primum obiectum intellectus nostri est ens,... nam omne per se intelligibile aut
includit essentialiter rationem entis, vel continetur virtualiter vel essentialiter in includente
essentialiter rationem ends: omnia enim genera et species et individua, et omnes partes essentiales
generum, et ens increatum includunt ens quiditative'. Ibid., I, 3, I, 3, no. 137; p. 85.12—18.
68. 'Si dicas quod lux increata cum intellectu et obiecto causât istam veritatem sinceram, haec est
opinio communis, quae ponit lucem aeternam sicut "causam remotam" causare omnem certam
veritatem.' Ord., I, 3, 1, 4, no. 260; 1950-, III, p. 159.7-10.
69. "... quod sentire et intelligere non dicunt aliquid reale additum super sensum et intellectum,
facicns cum eis realem compositionem; et ulterius, quod talcs actus sunt in nobis per se a dantc
sensum et intellectum, quod est creans et generans, ab obiecto autem sicut a causa sine qua non'.
Quae st. de nal. cog.; in Koch 1930, p. 18.24-30. Cf. '... et ideo ad hoc, quod reducatur in actum,
non indiget agente dantc novam formam'. Ibid., p. 23.6-8.
70. 'Ideo dico quod notitia intuitiva et abstractiva se ipsis différant et non penes obiecta nec penes
causas quascumquc, quamvis naturaliter notitia intuitiva non possit esse sine existentia rei...
Notitia autem abstractiva potest esse naturaliter ipsa re nota simpliciter destructa. . . . Ex istis
sequitur quod notitia intuitiva, tarn sensitiva quam intellectiva, potest esse de re non existente.'
William Ockham 1967, Sent. I, Prol., 1; OT, I, p. 38.5-16.
superfluous, as was likewise an agent intellect. 7 1 Further, intuitive cog

nition, both sensible and intellectual, was only of singulars. A universal,
remaining something singular in the mind, was a sign for a plurality of
individuals. It was a concept only, and formed no part of any substance. 7 2
As something singular the concept was a real being, whether in the first
intention it signified things or in the second intention it signified con
cepts. 7 3 What is predicated universally is consequently a term referring
indeterminately to a singular. 7 4 Since there is nothing common in either
things or concepts, only the word is univocal. 7 5 Unlike his predecessors,
Ockham does not regard theology as a single scientific habitus, though he
was in accord with them in acknowledging its supremacy. 7 6
From Ockham to Descartes

Ockham's nominalistic way of philosophising became widespread during
the fourteenth and fifteenth centuries. Though divine illumination drop
ped out of the discussions, the deep Neoplatonic heritage continued
through Meister Eckhart (1260-1327) and Nicholas of Cusa (1401-64).
Followers of Aquinas and of Scotus engaged in continuous controversies,
in the course of which the doctrine of essential being (esse essentiae) in the
object of the incomplex concept came to be accepted by the Thomistic
71. ' . . . talis species non est necessaria.' Ockham 1970, Sent., I, 2, 8; OT, II, p. 269.7. 'Dico quod
species neutro modo dicta est ponenda in intellectu, quia numquam ponenda est pluralitas sine
necessitate. Sed sicut alias ostendetur, quidquid potest salvari per talem speciem, potest salvari sine
ea aeque faciliter. Ergo talis species non est ponenda.' Ockham 1495-6, Sent., I, 27, 2, K; III.
'Intellectus agens nullo modo distinguitur ab intellectu possibili, sed idem intellectus habet
diversas denominationes.' Sent., I, 3, 6; 1495-6, II, p. 520.1 [-13.
72. 'Conceptus et quodlibet universale est aliqua qualitas existens subiective in mente, quae ex natura
sua ita est signum rei extra sicut vox est signum rei ad placitum instituentis.' Sent., I, 2, 8; 1970,
OT, II, p. 289.13 — 15. 'Hoc tamen teneo, quod nullum universale, nisi forte sit universale per
voluntariam institutionem, est aliquid existens quocumque modo extra a n i m a m . . . et quod
nullum universale est de essentia seu quidditate cuiuscumque substantiae. ' Ibid., pp. 291.17—292.1.
73. ' . . . tarn intentiones primae quam secundae sunt vere entia realia, et sunt vere qualitates subiective
existentes in anima'. Ockham 1491, Quodl., IV, 19; f. 58*1.
74. 'Stat confuse tantum, hoc est semper in universali affirmativa praedicatum supponit confuse
tantum.' Ockham 1974a Summa logicae, I, 73; OP, I, 204.4-6.
75. ' . . . nihil a parte rei est univocum quibuscumque individuis, et tamen est aliquid praedicabile in
quid de individuis'. Sent,, 1,2,7; 1970, OT, II, p. 256.9-10. ' " U n i v o c u m " proprie accipitur pro
voce univoca.' Ibid., 9; p. 306.18-19.
76. 'Ex istis sequitur quod divcrsarum partium theologiae sunt diversa subiecta, et quod theologiae
non est unum subiectum.' Sent.. Prol., 9; 1967, OT, I, p. 269.17-18. 'Ideo aliter dico ad
quacstionem quod theologia non est una notitia vel scientia, sed habet vel continct plures notifias
realiter distinctas quarum aliquae sunt practicae simpliciter et aliquae speculativae.' Ibid., 12;
Ρ' )37·Ι7~20- '. ·. dico quod aliae artes dicuntur eius ancillae, et quod de aliis habet iudicare
propter maiorem veritatem in cognitis et propter firmiorem adhaesionem'. Ibid., 7; p. 200.9— 11 ·
Cf. Aid., p. 185.6-7.
participants. Against that background the 'objective concept' 77 in the

meaning of what was known through the formal concept became es-
tablished in Suârez (1548-1617) and Vasquez (1551-1604). This set the
stage for the notion of 'idea' found in Descartes, with its long legacy of
problems regarding its relation to its object.
Conclusion
For all representative thinkers in the later Middle Ages religious faith,
meaning acceptance of truths on the authority of divine revelation,78 was a
source of knowledge necessary for man in his present state. They differed
widely and radically on the roles played by divine illumination and sense
experience in the formation of human ideas, with the doctrine of divine
illumination becoming attenuated and finally disappearing for centuries
till the seeing of 'all things in God' was revitalised by Malebranche. The
notion of 'idea' gradually metamorphosed from the Neoplatonic forms
through intelligible species into the 'formal' and 'objective' concepts famil-
iar to students of Descartes. 'Innate' (innata, inserta, insita, indita) was used
on occasion to designate knowledge possessed naturally or developed
spontaneously by the mind.
In the medieval treatment of cognition, Aristotelian influence became
dominant without the balance of a corresponding access to Plato.79 The
discussion of cognition, moreover, was carried on apart from any inkling
that epistemology might be looked on as a distinct science, and its develop-
ment was rendered still more uneven through its location in the context of
various theological discussions. Its history is instructive, however, for
understanding developments in early modern philosophy.
77. 'Conceptus objcctivus dicitur res ilia, vel ratio, quae proprie et immediate per conccptum
formalem cognoscitur seu repraesentatur;... ut objectum et materia circa quam versatur formalis
conception Suârez 1856-77, Disp. metaph., II, 1, 1; XXV, p. 65a.
78. E.g. 'Non enim fides, de qua loquimur. assentit alicui, nisi quia est a Deo revelatum.' Aquinas,
ST, llallae, q. 1, a. 1, Resp.; 1882-, VIII, p. 7b.
79. E.g. 'Quae fuerunt rationes, vel probationes, Platonis non pervenit ad me. Ponam igitur rationes,
quas vel habuisse videtur, vel habere potuisset.' William of Auvergne 1674, De uni verso, I-II, 14; I,
p. 82lb.
22
INTUITIVE AND ABSTRACTIVE
COGNITION
Scotus and Ockham as the focal points of the discussion

The fourteenth century is especially rich in controversies about knowl-
edge, but our understanding of them, while improving, is still limited. The
relevant texts are not widely available, and as a result the analysis that has
been produced is isolated and sketchy. Consequently, while we can frame
tempting hypotheses about developments in the period and their influence
on subsequent thought, it is still the familiar landmarks that best serve to
present the themes of the time and the orientations of recent commentary.
Especially notable among those landmarks are the theories of intuitive
cognition in Duns Scotus and William Ockham. Nearly all the medieval
discussions of intuition that follow them are an attack on or defence of one
or the other. Consequently, a presentation of the notion of intuition that
focuses around Scotus and Ockham will provide a useful picture of the
terrain on which subsequent battles have been fought.
The problem of the cognition of individuals

Around 1 2 5 0 - the position of William of Auvergne suggests things may
not have been so neat in the immediately preceding period 1 — writers of
both Aristotelian and Augustinian persuasions could maintain as a matter
of course that the province of the human intellect is the immaterial, so that
with respect to the physical world our cognitive experience of existent
individuals comes through sensation while the intellect contributes only
the universal. Orthodox belief, of course, required that God's knowledge
extend, as his providence does, to individuals. But, as is clear from disputes
about what angels could know of material things, it had been traditional to
locate the problem about an intellectual cognitive grasp of contingent and
material individuals in the peculiarity of divine knowing. 2
i. See Moody 1975, pp. 5S (notes), 59ff., 75 and 79.

2. Beruhe 1964 - a refreshingly non-polemical account of the controversy down to Ockham - lists
among early proponents of'non-intellection' of individuals Alexander of Hales, Robert Grosse-
teste,John of la Rochelle, Albert the Great, Bonaventure, and John Peckham.
Intuitive and abstractive cognition 461
Medieval philosophers were generally not inclined to scepticism, but the

contrast with divine knowing provided an alternative base from which to
project the limitations of human ways of knowing. It was to be expected,
consequently, that the accommodation typical of the mid thirteenth cen-
tury would eventually direct attention to the adequacy of its picture of
human cognitive capabilities. And within a few years, for example with
Thomas Aquinas, the emphasis was on making it clear that there is
knowledge of material singulars at the level of intellect while preserving a
special dependency on sensory experience.3 Such a moderate liberalisation
of the accepted opinion might have been expected to escape the full brunt
of the anti-Aristotelian reaction that was productive of and stimulated by
the Condemnation of 1277. But one feature of that reaction was an
unprecedented sensitivity to any qualification on the immediacy of
the intellectual awareness of individuals.4 In 1282, the doctrine of the
direct intellectual cognition of material singulars was officially adopted
by Franciscan theologian-philosophers.5 The emotionally charged
atmosphere managed to obscure even some of the careful statements it
occasioned.
The Augustinian background

The remote background of the issue is of more than ordinary interest, for
the dispute offers a good example of modern-seeming developments
carried by tradition-oriented attitudes. The tradition, of course, is
Augustinian.6 Augustine's world is exclusively individual. The phenom-
enon that interests him is necessity rather than generality and he locates its
source in the mind of God.7 What results is a familiar tripartite division of
3. The influential Augustianian Henry of Ghent also adopted 'indirect intellection', noting that it is
not Aristotelian. Siger of Brabant, for that very reason, at first launched an Averroist attack
against the theory, but seems eventually to have come around to it (Berube 1964, pp. 78-81).
4. Roger Bacon must be credited at least with an early emphasis on the cognition of singulars, but
his polemical style makes it difficult to place his position accurately. Bacon's 'suspicious novelties'
(Wolter 1967, p. 240) were a scandal to Franciscan authorities, and his direct influence on later
theories of cognitio singularis is not as clear as is that of Matthew of Aquasparta and Peter John
Olivi.
5. The philosophically undistinguished Correctorium fratris Thomae by William de la Mare achieved
notoriety for being adopted officially by the order in 1282 (Lynch 1972, pp. 28—9).JohnQuidort
is one of the better known authors of the Dominican rebuttals under the title CorrecrdriM»!
corruptorii (Berube 1964, pp. 89-91.)
6. Robert Grosseteste is an early example of the effort to accommodate Aristotle within an Augus-
tinian scheme (cf. McEvoy 1977). For the later period, see note 61 below,
7. 'Plato is known as the first to have named ideas ... In fact, ideas are the primary forms or the
permanent and immutable reasons [rationes] of real things... It is denied that the soul can look
upon [iniueti] them unless it be rational, in that part whereby it excels, that is, in its mind and
knowledge, where only the soul's knowledge of itself is direct and un-
problematic. Its knowledge of the physical world consists in organising the
flux of sensory experience in the light of intelligible forms. Because those
forms play an essential role in human knowing and yet can be located only
in God's mind, problems arise regarding their accessibility and our objec
tivity, problems that Augustine tried to solve with his doctrine of divine
illumination.
The influence of Aristotle

The interest medieval philosophers showed in the doctrine of illumination 8
is testimony equally to Augustine's authority and to the concern, as old as
philosophy itself, with problems about knowledge that is necessary, ab
stract or universal. By the early thirteenth century there had been so much
discussion of illumination that when an emphasis was placed on texts of
Aristotle 9 in which the intellect is said to contribute the universal, the idea
was accepted easily. But the problems involved in such an account of
knowledge give rise to discussions in which claims about how we know
things are not easily distinguished from claims about what we k n o w . I 0 For
example, must one choose between the Aristotelian doctrine that the
essential mode of intellectual cognition is universality and a straight
forward application to the case of material singulars of the idea that the
objects of knowledge must be real?
Cognition of individuals and Aristotelian demonstrative science

One might expect the emerging concern with the problem of an intel
lectual cognition of singulars to have generated explicit dissatisfaction with
reason, as it were in its face or interior and intellectual eye . . . What religious man, infused with
the true religion, even though not yet able to contemplate these objects, would nevertheless dare
to deny and even refuse to confess that all things that a r e . . . were created by God as their
source . . . ? N o w where would w e think these reasons are, if not in the mind of the Creator?'
Eighty-three Different Questions q. 46, 1-2 (transi, in Bourke 1964, pp. 62-3).
8. Besides explicit philosophical discussion of the doctrine o f illumination, there was also a special
appeal to illumination in the mystical traditions of the Middle Ages and renaissance, and the
doctrine is also thought to have inspired certain 'light theories' in medieval science (Weinberg
1964, pp. 163-4).
9. E.g., Posterior Analytics}, 18 ( 8 ι ^ 5 , 8 7 b 3 j - 4 o ) , Physics I, 5 (i89 a 7). De anima III, 48t 8 ( 4 2 9 b i o - i i ,
432'3-U).
10. At least as early as Matthew o f Aquasparta, there had developed the helpful distinction between
the fact o f our cognition o f singulars and the manner of our knowing them (Lynch 1972, pp. 31 -
2). It is Vital du Four, however, who distinguished further between our knowledge of 'the
singular' as existent and in its individuality (ibid.. p. 39), and showed some sensitivity to the
difference between what the intellect 'knows' and -what the human person does (sec note 77
below).
Intuitive and abstractive cognition 4°3
the Aristotelian concept of a science, which we know to have been firmly

entrenched on other grounds. But even Ockham, a staunch proponent
of the importance of our knowledge of singulars, subscribed to the
Aristotelian scientific ideal of demonstration from necessary premisses, and
he required that science be 'of terms' because it is 'of the universal'.11 But
even if sciences are only of the universal and necessary, knowledge is not.
Some account is needed not only of how we might arrive at suitably
scientific premisses but also of the knowledge we obviously have of
contingent facts about individuals and their existence.12 And while it was
the issue of individuality that raised the emotional temperature of the
controversies, it was the problem of the knowledge of existents that
determined the course of the major theories of intuition.
Duns Scotus and haecceitas
Neither the theory itself nor the term 'intuitive' which was used to mark it
appeared without precedent; there were important and interesting transi-
tional figures.13 It was with Scotus, however, that a distinction between
the knowledge of individuality and the knowledge of existent individuals
was systematically developed, and the contrast between intuitive and
abstract cognition was applied to the whole range of human knowing. His
Franciscan heritage is evident in his advocacy of a positive principle of
11. Expositiophysicorum, Prol. (in Boehner 1957, p. 11); and see Ordinatio, I, d.2, q. 4 (Ockham 1970,
OT., II, p. 34). Robert Holkot criticises Ockham for being insufficiently thorough regarding the
reduction of terms to individuals: Moody 1975, pp. 345 and 3 52. But that issue is not usually taken
as part of the dispute about cognitio singularis.
12. See Scotus, Otdinatio. I, prol., pt. 4. qq. 1 and 2; Scotus 1950-, pp. 142fr. (esp. p. 145), and Opus
Oxon., II, d. 3,q. n , n. 11; and Bettoni 1961, pp. 123-5. A direct connection between intuitive
cognition and the formation of scientific concepts is proposed by Scott 1969, p. 48.
13. The verb 'intueri* has its non-technical uses (see note 7 above) but appears, perhaps for the first
time, in Matthew of Aquasparta in connection with cognitio singularis (Lynch 1972, p. 124). The
adverbial form, which became standard in Scotus and Ockham, was used by Matthew and by
Vital du Four in a special triad: intuitive, arguitive. and speculative (ibid.. pp. 124-5). Vital says that
wc have knowledge intuitive of our own acts of the soul and allows for an extraordinary intuitive
knowledge of the soul itself (Aid., p. 144) - the latter providing an unusual use of'intuitive' to
mark the end rather than the beginning of a complex cognitive process. He objects that the theory
of indirect intellection is not adequate but refrains from calling our intellectual cognition of
material singulars 'intuitive': only the senses 'experience' material singulars so that the intellect
knows them "in the senses' (ibid., pp. 40, 52,92; Lynch thinks this amounts to intuitive cognition:
p. 34, n. 25). This 'cognising in the senses' is reminiscent of Augustine's theory of vital attention,
as is the account of another Augustinian, John Peckham (Aid., pp. 98ff.). Vital's account is an
elaborate and important description (see also note 10 above), but Lynch may be unduly
enthusiastic in claiming that Vital had a theory to rival the developed accounts of Scotus and
Ockham. (For the link between experience and intuition, see Berube 1964, pp. 127-8.)
individuation — the haecceitas,1 * as it came to be called — and in his use of the

principle that the individual as 'primary being' must be intelligible if
'being' is. i s Moreover, Scotus held that the human intellect, while less
powerful than that of an angel, is not less capable of grasping the individual
in its singularity. 16
What set Scotus off from the more radical proponents of cognitio singu-
laris was his contention that we lack an intuition of individuality since in its.
present condition (pro statu isto) the human intellect does not realise its
inherent capabilities. 17 For him, an intuition of individuality would in-
volve a simple and formal understanding of merely numerically different
individuals of the same kind. Beyond knowing that this table is an in-
dividual, 18 intuition as a grasp of haecceitas would as easily distinguish it
from an otherwise similar one put in its place as if an altogether different
kind of object had been substituted. 19 As Scotus saw it, such formal
understanding as human beings do have of things derives from a process of
abstraction of the sort Aristotle proposed. 20
While the haecceitas of any actual thing will figure in a metaphysical
description of the causality an object exercises in our experience of it, that
alone does not imply that our experience involves a cognitive grasp of its
individuality. Scotus did use the argument, characteristic of medieval
proponents of intuition, that the intellect as a superior power is capable of
whatever sensation is capable of.21 But he denied to sensation the capacity
for cognising the haecceitas,22 so that argument supports no claim about our
understanding of individuality, but only the more moderate conclusion
that the intellect can cognise the individual as the senses can.
14. '[The term] designates the unique formal principle of individuation that makes the nature, which
all individuals of the same species have in common, to bejust this or that individual and no other.'
AUuntis and Wolter 1975, p. 511.
15. In Metaph., VII. q. 15, n. 4; and see Gilson 1952b, pp. 543—55. Walter Burley agrees with Scotus
and Ockham on this point: Baudry 1943, p. 164.
16. Quodl., q. 6 art. 3 (Alluntis and Wolter 1975, 6.19).
17. Opus Oxon., II, d. 3, q. 6, n. 16; Quodl., q. 13, a. 2 (Alluntis and Wolter 1975, 13.27-32). See also
Bettoni 1961, p. 122, and Berube 1964, pp. 196 and 284fr. Peter Aureoli held the same position (II
Sent., 11,4, 2), as did William of Auvergne (Moody 1975, p. 55, n. 38).
18. T o know the universal in the singular is to 'know' the individual: Opus Oxon., IV, d. 42, q. 4, n. 6;
cf. also Aquinas, ST, I, q. 84, a. 7.
19. The De anima (22), whose authenticity is disputed, argues that a grasp of haecceitas would imply
that we could recognise an individual substance even if all its accidents were removed. See
Copleston 1963, II, p. 493.
20. But see Bettoni 1961, p. 100.
21. Opus Oxon., IV, d. 45, q. 3, n. 11; Quodl., q. 6, a. 1 (Alluntis and Wolter 1975, 6.19). The same
argument appears in Ockham (Ordinatio I, d. 3, q. 6; Ockham 1967, O T I, p. 492) and in Burley
(see Baudry 1943, p. 164). For the contrasting position of Aquinas, see below.
22. Rep. Par., II, d. 3,q. 3,n. 15; d. I 2 , q . 8 , n . 10; In Metaph., VII, q. 13, n. 26 (all cited in Gilson 1952b,
p. 546, n. 1).
Scotus' distinction between abstractive and intuitive cognition

Once the issue of a grasp of individuality is set aside - something that many
of Scotus' disciples found hard to do - the orientation of the distinction
Scotus drew between abstractive and intuitive cognition can be clearly
seen in the analogy he frequently made with sensation. In the distinction of
visio and imaginatio at the level of sensory awareness, perception of things is
differentiated from a form of representation that can take place in their
absence.23 Accordingly Scotus posited pre-judgmental24 acts at the intel
lectual level which are intuitive or abstractive depending on whether or
not they are indifferent to the existence or non-existence of the objects they
signify.25
Scotus sometimes described intuitive cognition within an Augustinian
setting as an initial, confused awareness of what will come to be refracted
into multiple, distinct (cognitive) elements.26 The emphasis, however, is
still on our experience of something real. 27 He extended that emphasis
under the label 'imperfect intuition' to cover both the memory of ex
perience and even a vision of the future - the latter to accommodate
prophetic visions and, in particular, the foreknowledge of Christ in his
human intelligence.28 While such extensions put a strain on the definition
of intuitive cognition as cognition of what is present and existent, one can
understand their place in Scotus' scheme if, undistracted by the more
familiar epistemological concern with certainty, one realises that for Scotus
the fundamental distinction between intuitive and abstractive cognition is
the difference between knowing what is actual and knowing what is
merely possible or necessary.29
The same perspective helps to explain what might otherwise seem
23. Quodl., q. 13, a. 2; q. 6, a. 1 (Alluntis and Wolter 1975, 13.28-9; 6.19).

24. Quodl., q. 6, a. 1 (Alluntis and Wolter 1975, 6.17). See Bettoni 1961, p. 123, and Alluntis and
Wolter 1975, p. 499.
25. As 'present and existing': Opus Oxon., HI, d. 14, q. 3, n. 14; and see II, d. 3, q. 6, n. 16. He says he
means to contrast intuitive cognition with abstractive '... eo modo quo dicimur in tueri rem, sicut
est in se'. /6i<i., II, d. 3, q. 9, n. 6.
i6. Bettoni 1961, p. 123; Berube 1964, pp. 172fr.; Alluntis and Wolter 1975, p. 500.
27. Scotus occasionally uses the term 'experimur' (e.g., at Opus Oxon., IV, d. 43, q. 2, nn. 9-10).
28. Opus Oxon., IV, d. 10, q. 5, n. 4. The case of memory, of course, is not extraordinary. Sec Opus
Oxon., Ill, d. 14. q. 3. nn. 4-7; and ibid., II, d. 9, q. 2, n. 19 and Rep. Par., II, d. 3, q. 2, η. 11, where
Scotus suggests one might postulate a special habit as an alternative to imperfect intuition. For
Ockham on 'imperfect intuition', see In Sent., II. q. 15 and the discussion in Baudry 1958,
pp. 177-8.
29. Quodl., q. 6, a. 1; q. 7. a. 2; q. Il, a. 2 (Alluntis and Wolter 1975, 6.18; 7.24-5; 3.33). In Scotus'
scheme, experience serves the purpose of picking out (in a passive way) from the cognitive
possibilities open to the human intellect what the Divine Will does (in a creative way) by
determining one set of compossibles to actual existence. (For Scotus' theory of creative will and
possibility, see Gilson 1955, pp. 460-1 and notes pp. 765-6 and the references given there.)
strange: Scotus found it obvious that we have abstractive cognition but

thought an argument had to be given to show that there is an independent
intuitive cognition. 30 Since what is possible can be non-actual, although
what is actual cannot be impossible, it is to be expected that any case in
which our thinking involves no claims about existence provides an ex-
ample of the non-intuitive (abstractive); while every judgment of existence
involves, so to speak, an admixture of what is possible. The difficulty is the
same as that in trying to determine what it is that differentiates a drawing of
an existent horse from one that is meant simply to illustrate the concept, as
in a dictionary or biology text. It is also a difficulty that motivates in part
Ockham's adjustments to the definition of intuitive and abstractive
cognition. 31
Ockham on concepts and individuals as objects of cognition
Although Ockham acknowledged a debt to Scotus' theory of intuition, it is
characteristic of him to have taken up the whole issue of singular cognition
in association with the question of what concepts signify. Convinced as he
was that concepts, the mental correlates of terms, function as the basic
components of our thought and that they are somehow caused by existent
things, Ockham maintained that knowledge is either of concepts or of
individuals outside the mind (extra animam) and that the latter is prior. 32
Ockham's penchant for parsimony put a heavy burden on this simple
scheme, but one can imagine him insisting that whatever adjustments
might have to be made to his account of the cognitive process, there is at
least no need to expand one's ontology to admit non-individuals.
Ockham on evident cognition

Against this familiar background, Ockham proposed to demonstrate the
need for a distinction among ways in which concepts can occur in the
cognitive process. 33 There are cases, he says, where the same proposition
can be known to be true in certain circumstances and not known to be true
in others: for example, 'The wall is white' when I am looking at the white
wall and when I am not. This knowledge-in-experience is one of three
30. Quodl., q. 6, a. r; cf., q. 7, a. 2 (Alluntis and Wolter 1975, 6.18- 19; 7.22).
31. Peter Aureoli thinks one should define intuitive cognition only in terms of purporting to be about
existents; see note 71 below.
32. See the extended discussion at Ordinatio, d. 3, qq. 5-8 (Ockham 1970, OTII, pp. 442-542); also
Quodl., I, q. 13 (translated in Boehner 1957, pp. 27—32). Walter Burley, who more often opposes
Ockham, agrees about the intellectual cognition of material singulars: Baudry 1934, p. 164 and
note 4-
33. Ordinatio I, Prol., q. 1 (Ockham 1967, O T I, pp. 22-3): translated in Boehner 1957, p. 20.
types of knowledge Ockham brings under the heading of what is 'cognised

evidently'.34 The others are conclusions seen to follow demonstratively
from necessary premisses (that is, science in the strict sense), and per se notae
propositions recognised as true without either an appeal to experience or
any inference from other propositions.
Ockham defines evident cognition as ' . . . cognition of some true com
plex [or proposition], the nature of which is to be adequately caused,
immediately or mediately, by non-complex cognition of the t e r m s . . . ' 3 3
The truth of the cognised proposition is an obvious requirement if
Ockham meant to introduce more than psychological certainty. But to put
knowledge-in-experience in the class of knowledge that derives from
apprehensions (i.e., terms) will strike a modern reader as an unusual
handling of'matters of fact'. At least part of Ockham's motivation here,
however, was to insulate all evident cognition from the action of the will (a
free agent) in favour of the unimpeded action of the intellect (a natural
agent). 36
Ockham on intuitive cognition

Intuitive cognition is precisely that incomplex apprehension in virtue of
which contingent propositions are evidently cognised; abstractive cog
nition is non-intuitive.37 Ockham argued his case in an extended account,
34. Ordinatio, I, Prol., q. 1 (Ockham 1967, ΟΤΙ, pp. 5-7).

35. "... so that when a non-complex cognition of some terms, whether they arc terms of that
proposition or of another or of different propositions, adequately causes or is of a nature to cause,
either mediately or immediately, cognition of a complex in any intellect having such a cognition,
then that complex is cognised evidently'.
'... notitia evidens est cognitio alicuius veri complexi ex notitia terminorum incomplcxa im
mediate vel mediate nata sufficienter causari. Ita scilicet quod quando notitia incomplexa
aliquorum terminorum sive sint termini illius propositionis sive alterius sive diversarum pro-
positionum in quocumque intellectu habente talem notitiam sufficienter causât vel est nata
causare mediate vel immediate notitiam complexi tunc illud complexum evidenter cognoscitur.'
Ordinatio, Prol., q. 1 (Ockham 1967, OT I. pp. s-6).
36. 'Praeterea, quicumque seit evidenter aliquod complexum, non potest dissentire illi complexo solo
imperio voluntatis, sed oportet quod persuadeatur per rationem fortius moventem intellectum
suum ad dissentiendum, vel oportet quod obliviscatur alicuius evidenter noti.' Ordinatio, Prol.,
q. 7 (Ockham 1967, OT I. p. 192)·
37. 'Et universaliter omnis notitia incomplexa termini vel terminorum scu rei vel rerum virtute cuius
potest evidenter cognosci aliqua Veritas contingens, maxime de praesenti, est notitia intuitiva.
... Et omnis notitia complexa terminorum vel return significatarum ultimate reducitur ad
notitiam incomplexam terminorum. Igitur isti termini, vel res, una alia notitia possunt cognosci
quam sit ilia virtute cuius non possunt cognosci tales veritates contingentes, et ilia erit intuitiva. Et
ista est notitia a qua incipit notitia expérimentales, quia universaliter ille qui potest accipere
experimentum de aliqua veritate contingente, et mediante ilia de veritate necessaria, habet
aliquam notitiam incomplexam de aliquo termine vel re quam non habet ille qui non potest sic
experiri." Ordinatio I, Prol., q. 1 (Ockham 1967, OT I, pp. 31-3)·
the crucial premiss of which is that the proximate cause of assent to a

proposition is the apprehension of its terms. 3 8 That premiss is supported by
a less peculiar but no less carefully supported claim that assent to a pro
position presupposes apprehension of the proposition, which in turn pre
supposes apprehension of the terms. Finally, if evident assent to the same
contingent proposition is possible in one case and not in the other, the
apprehension of terms in virtue of which the assent is made must be
different in the two instances.
The argument supposes that the representative content of the differing
apprehensions is in fact the same, and so Ockham's theory provides no
more support than Scotus' for proponents of the intellectual grasp of
singularity. Intuitive cognition pertains to an individual, Ockham says,
because it is 'caused by this one rather than that'. 3 9 The character of
intuitive cognition, therefore, is not captured by any logical or gram
matical classification (e.g., the use of indexicals or proper names). Intuitive
and abstractive cognition differ 'in themselves', but not by any mark other
then having or lacking the capacity to cause evident assent with respect to
contingent fact.
The structure of the argument in Ockham's explicit and detailed ac
count of intuitive cognition in the Prologue to his commentary on the
Sentences exhibits intuitive cognitions as a presupposition of our knowing
contingent facts to be true. Ockham begins with a full-fledged case of
knowledge and makes no special effort to hedge it against the possibilities
of doubt. It is, moreover, knowledge expressible in the form of a pro
40
position: 'The wall is white', Ί am sad'. Ockham then proceeds to isolate
the components of the proposition in terms of logical form, marking the
steps and elements of a cognitive process in keeping with that structure.
That one can truly say 'The wall is white' cannot, for Ockham, be
explained directly as a matching of a proposition with some extra-mental
38. Ordinatio I, Prol., q. 1 (Ockham 1967, ΟΤΙ, pp. 16-22). The extended argument is analysed in
Boler 1976. The premiss that the proximate causes of assent must be intellective was attacked by
Adam Wodeham on the grounds that it depends on a theory of plural substantial forms:
Quaestiones in librum Sententiarum I, Prol.,q. 1 (§20'... quod visio sensitiva immediate accipitur in
intellectu'.) The text was brought to my attention by Fr. Gedeon Gâl, whose help in surveying
the literature of the fourteenth century was invaluable to me.
39. Quodl., I, q. 13; translated in Boehner 1957, p. 30 (see also pp. 28—9).
40. Scotus (Opus Oxon., IV, d. 45, q. 3, n. 17) and Ockham (Ockham 1967, ΟΤΙ, pp. 39?) hold that
the intellect has intuitive cognition of mental acts. Walter Chatton argues that intuitive cognition
is not needed to account for that knowledge: O'Callaghan 1955, article 5, pp. 255-61. The
question of the soul's knowledge of itself is more complex and received various answers.
state of affairs. He has argued at length (in another context) that there is no
extra-mental composition onto which predication is mapped. 41 But his
definition of intuitive cognition very carefully reflects only the argument
which produced it. 4 2
Ockham on intuitive cognition of non-existent objects

On the other hand, Ockham was not fully in control of the implications of
his having defined intuitive cognition in terms of true propositions, and he
seems unaware of the troubles that arise because of his havingjeJ intuitive
cognition in a causal sequence as well. The form of his definition is no
accident, however, as is clear from the effort he takes to distinguish it from
Scotus'.43 For Scotus, it is impossible for there to be intuitive cognition of
an object that is not present and existent. For Ockham, the necessary
connection is between intuitive cognition and the true assent made in
virtue of it. 4 4 That the two definitions are not equivalent is shown in
Ockham's favourite if unfortunate device for displaying necessary connec
tion. On Scotus' account, God cannot cause an intuitive cognition of a
non-existent object; but on Ockham's definition, although the case would
be extraordinary, God could cause an intuitive cognition so long as the
ensuing judgement is true: namely, 'That object does not exist.' 45
The most likely objection Ockham could think of is that his definition
will seem faulty for implying that God could never deceive us. In reply, he
simply denied the consequence. God can cause in us a 'creditative act' that,
for example, something exists when it does not; but Ockham's definition
requires only that the deception not be accomplished by means of an
intuitive cognition. 46 The argument succeeds in what it is intended to
accomplish. And Ockham's handling of the issue should dissolve any
inclination to charge him with scepticism: one is hard pressed to imagine a
41. Summa logicae II, Chs. 2-20.

42. The definition is given in note 37 above.
43. Ockham considers a number of ways his position differs from that of Scotus: Ordinatio I, Prol-, q. 1
(Ockham 1967, Ο Τ Ι , pp. 33-9). The requirement of an existent object is treated at pp. 38-9;
a more detailed account is in QuodUbeta V, q. 5 and VI, q. 6 (tr. in McKeon 1929—30, II, pp. 368-
75)·
44. In an early text, Ockham does at one point characterise intuitive cognition in terms of the
existence of its object (Rep., II, q. 15: in Boehnet 1943b, p. 248). Fot his later, more careful
definition, see note 37 above.
45. Walter Chatton thinks Ockham's account is incoherent on this point (O'Callaghan 1955, articles
3-6: pp. 246-69). Of course, he uses, as Scotus does, the analogy of visio and imaginatio to
distinguish intuitive and abstractive cognition (cf., ibid., p. 248); see below.
46. Quodl., V, q. 5 (McKeon 1929-30, II, p. 371).
sceptic who could leave the argument where Ockham leaves it. 47 But the
issue is also something of a distraction, for the troubles in Ockham's
account of intuition are not linked with the possibility of extraordinary
exercise of God's absolute power.48
Present objects and unconditioned beginnings

The positions adopted by Scotus and Ockham, according to one interest-
ing suggestion,49 can be viewed as diverging along classical lines for
intuition theories according to whether the emphasis is on the presence of
the object or on an unconditioned beginning: 'a cognition not determined
by previous cognitions'. Considering the way in which Scotus draws an
analogy to vision, he is to be classed with those who adopt theories of the
former sort. That such theories are probably more common, perhaps
because they are less technical and more natural, may account for the
continuing influence Scotus' theory had even after the publication of
Ockham's criticisms. Ockham's weakening of the link between intuitive
cognition and its object, along with other indications in the way he argued
for intuitive cognition, favors his being classed with proponents of the
unconditioned-beginnings sort of intuition theories. Since the question of
where intuitions 'come from' is incidental for such theories, however,
Ockham's concern for the causal origins of intuitive cognitions will appear
from that perspective to be misplaced. But even if the causal fink to objects
were entirely dissolved, there would remain problems with Ockham's
having defined intuitive cognition in terms of assent to true propositions. It
is hard to see how he could have met the difficulties without adopting a
rather sophisticated coherence theory of truth. 50
Intuition and deception

One can, of course, argue that, despite the form Ockham's presentation
takes in the Prologue, he was really engaged in a more straight-forwardly
epistemological project where intuitive cognitions are introspectively re-
47. Sebastian Day (1947) brought to the boil a minor controversy about Ockham and scepticism
which is not worth recounting (see Baudry 1958, p. 177 and Adams 1970, pp. 389-93). Day's
book provides a useful collection of texts, but his interpretations of Scotus and Ockham must.be
read with caution; his account of Aquinas is altogether unreliable.
48. See Adams 1970, pp. 393fr., and Scott 1969, pp. 43-8.
49· The point is made by Robert Wengert in an unpublished manuscript concerning Hervaeus
Natalis, who in Quodl. IV takes intuition as an unconditioned beginning.
50. Scott (1971, pp. 28ff.) is willing to claim that Nicholas of Autrecourt is only developing
Ockham's position in maintaining that one must adopt the principle that what we are certain
about is true.
coverable elements available for appeal in justifying knowledge claims

about contingent facts.51 Such a point of view puts into even starker relief
the oddity of defining intuitive cognition in terms of true propositions and
makes it impossible not to see a flaw in Ockham's having nowhere told us
how to distinguish intuitive cognitions from deceptive creditative acts.
Mental language
Either approach to interpreting Ockham's theory will eventually uncover
more radical anomalies. The basic structure and components of 'mental
language', Ockham tells us, are determined by finding out what elements
of a natural language are necessary for its statement-making functions. 52 If
only for the appeal to the notorious 'razor', which Ockham himself points
out is relevant only outside observational contexts, 53 the determination of
intuitive cognitions - or on the theory that they should be introspectible,
the correlation of intuitive cognitions with concepts (which are the basic
components of mental language) - will be a complex and indirect pro-
cedure. It has even been persuasively argued that, when combined with
Ockham's account of the derivation of scientific knowledge of fact, there
results the paradox that we can have knowledge only if we do not know
that we have it. 54
Intuition, scepticism, and the absolute power of God

From a modern point of view, the project Ockham described, with its
concern for first cognitions within a quasi-causal sequence from existent
things through intuitive cognition to assent is so unmistakably pregnant
with epistemological problems that it tends to foreshorten our perspective
on philosophical developments between his time and that of Descartes. 55
An emphasis on scepticism, however, may distort our understanding of the
actual development of fourteenth-century epistemology.
The sceptical proclivities of certain so-called 'Ockhamists' have been
well advertised if not always well analysed. 56 The distinction characteristic
51. Adams 1970 gives a good account of this.

52. Summa logicae I, Ch. 3.
53. Reportatio II, q. 150 (quoted in Boehner 1957, p. xx, n. 2). Since the world results from a free act of
creation, it contains more than only necessary things.
54. Scott 1969. pp. 43 and 46.
55. Intuitive cognition was still important for Suârez (Coplcston 1963, III, p. 375), but the exact
influence on Descartes' thought is not definitively established.
56. See Moody 1975, pp. 127-60. That such figures as Nicholas of Autrecourt and John of Mirecourt
were called 'Ockhamists' tells us more about their social attitudes - or the attitudes of those who so
labelled them - than about their philosophical positions.
of them is one between absolute certainty and ordinary certainty.57 The

former consists of what can be 'reduced' 58 to the principle of non-
contradiction plus the knowledge of one's own existence and mental states
which Augustine had established in arguments against the Academics.59 As
evidence that ordinary perceptual claims cannot meet this higher standard
of certainty, these writers customarily alluded to God's ability to cause in us
deceptive creditative acts about the existence of things.
The keystone of the argument, of course, is the doctrine of the absolute
power of God, and deception with regard to intuitive or pseudo-intuitive
cognition is introduced only to illustrate the range of that power. Ock-
ham's theories were not likely to have been more than a mere occasion
for this argument and his account of intuitive cognition served only the
form in which it came to be expressed.60 Like the accusation of scepticism,
the adoption of scepticism usually stems from mixed motives, and the most
powerful in this instance may have been social and political. The history of
scepticism in the Middle Ages, the use of sceptical arguments and the efforts
to avoid and counter them, is a complex story that deserves to be told. But
its intersection with the history of the notion of intuition, while dramatic,
may not be as important for either of them as has been supposed.
In general, throughout the thirteenth and fourteenth centuries, discus-
sion of the knowledge of material singulars was oriented to the topics that
gave rise to the original dispute: the domains of and relationship between
sense and intellect, the limits of human knowing (in contrast with angelic
and divine knowledge), and the knowledge of existence and singularity.
With the appearance of explicit theories of intuition, attention was in-
evitably focused on the beginnings of the cognitive process. It is not easy to
determine, however, for the individual studies which develop, whether
their intentions were to inquire into the justification of knowledge claims
or to provide a description of our ways of knowing. Moreover, the most
prominent appeals to the possibilities of scepticism do not seem to function
57. Labelled 'special' and 'natural' by John of Mirecourt (Copleston 1063, III, p. 128). Peter of Ailly
contrasts 'natural light' with 'reason' as concerned with the two types of certainty (Leff 1967c,
p. 61). Buridan had proposed that science needed only practical or natural certainty (In
Melaphysicen Aristotelis quaestiones, II, q. I: tr. in Hyman and Walsh 1967, pp. 704-5). And
Nicholas of Autrecourt consequently charges Buridan with scepticism! (Copleston 1963, III,
pp. 137-8). Nicholas, however, denies that there are degrees of certainty; he allows a distinct type
of certainty for matters of faith, however (ibid.).
58. Scott (1971, pp. 21-2, esp. n. 17) argues that Nicholas, at least, does not mean by this that we can
only be certain of truths whose denial is a contradiction; he was, instead, talking of the relation of
evidence to conclusion. For the more familiar interpretation see Weinberg 1964, p. 268.
59. See note 61 below.
60. See the discussion of Ockhamism in Moody 1975, pp. 127-60.
either as a base or as a foil for the development of explicit discussions of the

perceptual process.61
From the perspective of the early modern era, it is striking that the
intersection of theories of intuition and the principle of God's omnipotence
could produce appeals to the possibility of deception but seem to have
occasioned no developed accounts of the justification of perceptual
claims.62 Of course, in the absence of some reason to think that God is
interfering with some or all of our perceptions, the mere possibihty of
deception is insufficient to render perceptual claims doubtful. As it hap-
pens, however, the target of the better known 'sceptics' of the post-
Ockham period seems not to have been perception at all but rather the ideal
of Aristotelian demonstration which was thought to provide an avenue to
certainty that would rival or even undermine the faith.63 In this connec-
tion, an appeal to God's omnipotence is a simple (if excessive) means for
showing contingent claims to be contingent. And if the theory of intuitive
cognition is supposed to provide the essential link (causal or otherwise)
between the actual world and our perceptual claims, from which further
transformations can produce scientific premisses, the possibility of divine
intervention is a legitimate consideration. For here, unlike the case for
perceptual claims, the necessity of the demonstration is at stake, and the
otherwise plausible rider: 'if God is not deceiving us', defeats its purpose.64
Intuition and intelligible species

The mixture of descriptive and justificatory purposes is a problem in the
history of another theme which, during the period, was as prominent as
61. The model for this practice is again Augustine: see, for example, his Contra academicos and De libera
arbitrio. Book II. The importance of Augustine for the epistemological developments of the later
Middle Ages is easily overlooked: '... it is worthy of note that Gregory [of Rimini] should have
been able to confirm by so many quotations from Augustine some theses one would otherwise
feel tempted to explain by the spreading influence of Ockham.' (Gilson 1955, p. 502.)
62. The problem of scepticism is not ignored however. Scotus tried to counter what he took to be
sceptical consequences in Henry of Ghent's theory of illumination (Gilson 1952b, p. 558);
Ockham and Chatton are concerned with the sceptical elements in Peter Aureoli's idiosyncratic
development of an esse apparent (see note 71 below); and Nicholas of Autrecourt works at
articulating a rule for the inference from what appears to us to what is the case (discussed in both
Scott 1971 and Weinberg 1948).
63. This was the primary reason at least of Peter John Olivi, Gregory of Rimini and Peter of Ailly.
The picture developed in such classics as Michalski 1969 needs to be modified: cf. Copleston 1963,
III, p. 148, and Gilson 1955, p. 759, n. 36.
64. For a discussion of some of these issues, see Scott 1971 ; Buridan's suggestion that science — and he
may mean even Aristotelian science - does not need that sort of certainty could be either
disingenuous or especially sophisticated: In Metaph. Arist. quaes!., II, q. 1 (Hyman and Walsh 1967,
pp. 704-5). Scott takes the former alternative: 1971, p. 34.
and often linked with intuitive cognition: the role of the 'intelligible
species'. Its history is too contorted to be straightened out in the present
context, but it is a source of such confusion both in medieval writers and in
the interpretation of them that the epistemological issues deserve a brief
mention.
There is little enough in Aristotle's texts 65 to insure that a single meaning
would be attached to the label. One is well advised to treat 'intelligible
species' as equivocal in different theories and to be especially cautious when
one writer presents the theory of another. There were two general types:
causal and image theories.
Because of the analogy with visiojimaginatio, it is tempting to think
Scotus' hesitations about an intelligible species in intuitive cognition derive
from problems about images; but his theory is causal. 66 And so is Aquinas',
for whom the intelligible species is like 'the form by which an agent acts'. 67
That Ockham was sensitive to the general lines of Scotus' account is clear
from his conclusion that a mental habit will adequately serve the required
purposes. 68 The change reflects the greater autonomy or independence in
its activity which Ockham attributes to the human intellect. 69 But it should
be kept separate from his criticism of the species as an image.
Ockham's complaint against images, echoed by opponents of an intel-
ligible species in the period following him, was that the introduction of a
medium or 'third thing' between concept and object is an opening to
scepticism. 70 That critics of the species did not carry the day in the post-
Ockham period was due not to a return to the causal theory — or a taste for
65. Primarily De anima III (e.g., 429*27-8, 43i b 2). For different 'species' in Aquinas, see Lonergan
1967, pp. 163-8.
66. See Alluntis and Wolter 1975, pp. 515-17 and the references there.
67. ST, I, q. 55, art. 1 and q. 85, art. 2; see also Lonergan 1967, pp. 82 and 155.
68. In Sent., II, q. 15 R.
69. The rejection of intelligible species by Henry o f Ghent and John Peckham is clearly connected to
the Augustinian rejection of anything that suggests the passivity of the intellect with respect to
material objects (see Lynch 1972, pp. 6 7 - 8 and 98ff.).
70. In Sent., I, d. 27, q. 3. Even Peter Aureoli's strange-sounding claim that intuitive cognition is, like
any concept, an esse apparens (cf. Gilson 1955, pp. 479—80; In Sent., 1, proem., sec. 2, a. 3; and see
14a.1: Peter Aureoli 1952-6,1, pp. 197fr., and II, pp. 696ff.) is motivated in part by the need to
eliminate any 'medium' in knowledge: a thought o f the object, for Aureoli, simply is the object in
its intentional existence. However, when Aureoli says that an intuitive cognition can exist
without the object, the sense is altogether different from that given the analogous claim in
Ockham's theory. For a discussion of Ockham's criticism of the sceptical import o f Aureoli's esse
apparens, see Adams 1977, pp. 154—63. Walter Chatton's criticism of Aureoli can be found in
O'Callaghan 1955, pp. 241-6.
scepticism - but to the resilience of image theories. 71 There were pre-

cedents in the older tradition for adopting a triadic scheme in which
thinking of an object is taking something as a sign of the object. 72 It is
substantive concerns, of course, and not the history of labels that determine
the philosophical value of various accounts; and students of the fourteenth
century have to be especially sensitive to the difference.
Thomas Aquinas on the intellect and the cognitive process

These sketchy remarks on the intelligible species suggest the possibility of a
more radical hypothesis that even 'intellect' may label significantly dif-
ferent functions with the result that epistemological issues — at the very least
their formulation - may be relative to the model of intellect in which they
are embedded. It is a theme that allows us to return to the broader context
introduced at the beginning of this section.
Compared to his Franciscan successors - and to Augustine — Aquinas
offered a most unusual picture of radical incompleteness in the intellect's
operation within the context of human knowing. It is a direct analogue to
his account of the soul as incompletely human; and it seems to create a
similar sort of scandal. 73 As the human being, for Aquinas, is a body and
has a soul - so that the co-principle of soul is not body but matter 7 4 - so the
human cognitive process, for him, is basically a sensory one that is or-
ganised and structured by the intellect. 75 The intellect of an angel (an
incorporeal being), he says, is not simply a more powerful version of our
own; it corresponds in fact to the whole human cognitive apparatus
including both sensation and understanding. 76 Consequently, Aquinas
71. Most adherents to a species theory in the thirteenth and fourteenth centuries treated the species as
an image: a particularly clear case is Giles of Rome (see Lynch 1972, p. 87 and n. 68). See alsoJohn
of Reading's defense of Scotus in Super Senlentias, I, d. 3, q. 3, presented in Gâl 1969. Fr. Gil's
introductory remarks give a brief account of the dispute about the intelligible species from
Aquinas to Reading.
72. A triadic definition of the sign relation was used by Augustine (for example in Dr magistro). For a
discussion of Abelard's critical and complex account of ideas and images, see Tweedale 1976,
pp. 169fr. and 2ioff.
73. 'If [thej general theorem [that knowledge is by immateriality] is taken out of its historical context
and made the premise of merely dialectical deductions, endless difficulties arise' (Lonergan 1967,
p. 150). For Lonergan's account of individuation, see p. 153; for knowledge of individuals, see
pp. 168-77 (which covers Aquinas' early and late theories), 179-80, and 184.
74. Anscombe and Geach 1961, p. 98.
75. That is, our experience is bodily: De veritate, q. 10 a. 5, and q. 8, a. 11. Contrast the experience of
angels: ST, I, q. 54, a. 5; q. 57, a. 2; q. 75, a. I.
76. ST, I, q. 57, a. 2. It is the human knower rather than the intellect which cognises the singular: De
veritate, q. 2, a. 6, ad 3; De anima III, lect. 8, §§712-16.
uses the principle that a higher power can do what a lower power can to
infer from the fact that sensation is a cognition of material singulars the
conclusion that angels can cognise them intellectually.77 But he denies that
a similar conclusion can be drawn for the human intellect.78
It is sensory presentation, the 'matter' of our experience, that carries the
presence of things for us within Aquinas' scheme.79 That in itself was not
unusual. But where the focal point of knowledge of singulars for Ockham
was the application of concept to individual and for Scotus was the unity
and reafity of the individual thing represented by a complex of characterisa-
tions, the important polarity for Aquinas was not thought and thing but
thought and data.80 Our knowledge of a thing is a construct of sensory data
organised by the intellect. The knowledge of angels is not; and were they
to operate by abstraction from sensory experience, their more powerful
intellects could not transform this 'artificial' process.81
The limited condition of physical objects, Aquinas thinks, is a match for
our 'composite' or bodily mode of knowing. 82 They are individuals
because the kind of things they are is differentiated only for being exempli-
fied in this 'stuff' rather than that. And when he says that 'matter is not
intelligible', it is not a self-defeating claim that we cannot think of matter,
but that, as distinct from some structure which accounts for what some-
thing is, matter adds no property or specification. And consequently, if
angels have a more complete, direct, and simple intellectual grasp of
material singulars than we do, it does not result from their penetrating to
something 'in' things which we fail to see but is due to their using a
principle of intelligibility of a higher sort than the forms of material
singulars themselves.83
The world of singular things then, according to Aquinas, is only poten-
tially intelligible as it is sometimes only potentially sensible. It can be made
actually sensible, however, by a non-cognitive process: a coloured body in
the dark becomes actually visible if a light is lit;84 and sensation follows
unmediatedly when a well-disposed sense apparatus is in the presence of an
77. De veritate, q. 8, a. 11, sed contra §4.

78. ST, I, q. 86, a. i,ad4.
79. Cognitio singularis has precedence simply because sensation does: ST, I, q. 85, a. 3.
80. Cf. ST, I, q. 84, aa. 7-8; q. 86, a. 1; De veritate, q. 13, a. 3, ad2 and 3.
81. ST, I, q. 57, a. 1,4.13.
82. The forms of physical things do not even exhaust the potentialities of the matter which they
inform: ST, I, q. 55, a. 2; q. 84, a. 3, ad 1. This limitation plays an important role in moral
contexts: ST, lallae, q. 94, a. 4; Ilallae, q. 120, aa. 1-2.
83. ST, I, q. J5, aa. 2-3; q. 36, a. 2; q. 57, a. I, ad 3; q. 85, a. 5.
84. SCG, II, Ch. 59, §14.
actually sensible object. In contrast, the step from potential to actual

intelligibility involves a cognitive operation.85 We do not intellectually
'see' physical objects, according to Aquinas; our intellectual grasp of them
is accomplished by structuring a sensory manifold, and the intellect's
contribution to that construction is in providing the 'form'. This is a
complex and perhaps unusual model for cognitive operations, but
Aquinas' claims for the directness of sensation and the indirectness of
intellection - that material singulars are actually sensible but not actually
intelligible - take their significance from it.
Given the autonomy86 that Scotus and Ockham accorded the intellect,
they had to allow it a cognition of singulars in order to account for the
knowledge we quite obviously have of existent individuals. The question
about a cognitive grasp of singularity is independent and involves different
philosophical issues precisely because our knowledge of singularity is
already problematic at the level of ordinary claims about what and how we
know. Even when that distinction is made, however, the approaches taken
by Scotus and Ockham are likely to seem more familiar to a modern reader
than does Aquinas' version of an 'incomplete' intellect because an auto-
nomous intellect fits more readily our customary rhetoric of'the mind'. 87
A similar fit obtains, significantly enough, if we focus the comparison and
contrast on the writings of Augustine. And it was no small matter during
the medieval period that the rhetoric in classics of Christian spirituality -
the journey of the soul to God, the search within, and so on - was more
easily accommodated in the model of an autonomous intellect, as were
problems about the status and activities of the separated soul.
Conclusion
In sum, from the fact that Scotus and Ockham, in their differing ways,
meant to be talking about the same thing as Aquinas was talking about —
namely, our knowledge of reality - it does not follow that they must have
meant the same thing even by a pivotal concept such as the intellect. And
whether Aquinas' scheme was a deviant one for the medieval period or the
norm for thirteenth-century theories, the pattern of change in models of
intellect from Aquinas through Scotus to Ockham and of relative con-
85. ST, I, q. 84, a. 4, ad 2; De veritate, X, 6.
86. The autonomy of the intellect is a natural outcome of the Augustinian emphasis on the activity
rather than passivity of the intellect: Gilson 1952b, p. 524 n. 2. See also Belmond 1928,
pp. 463-87.
87. At least one more step is needed to transform the medieval idiom of'extra animam' into 'outside
the mind'; see Matthews 1977.
stancy after that is actually clearer than the pattern of epistemological

development.
Commentators on the later Middle Ages who are otherwise quite at
odds in their approaches have long felt that something happened with
Ockham which set the character of theories of knowledge until the time of
Descartes. The growing interest in the epistemology ofthat period is sure
to overthrow any too simple account of its development and will alter,
perhaps radically, the now familiar picture of its sceptical tendencies. But
the old hypothesis may prove to be right after all when directed not at the
justification of knowledge but at its description.
23
I N T E N T I O N S AND I M P O S I T I O N S
Sources of the concept of intention

The concept of intention played a key role in the discussions of epi
stemological, logical, and semantic questions in later medieval philo
sophy. 1 The significantly different use of'intention' in other fields such as
ethics or natural philosophy is not at issue here. 2
'Intention' in the relevant sense is associated with two concepts that
occur already in the writings of Al-farabi and Avicenna, where they are
associated with the words "ma'qul" and "ma'na", both of which were
translated into Latin as 'intentio'.3 In his commentary on the first chapter of
Aristotle's De interpretatione Al-farabi understands by "ma'qul" — his trans
lation of the Greek word 'noema' * — a concept or a thought that has to be
examined by the logician in two respects: in its relation to things outside
the soul and in its relation to words. 5 "Ma'qul" means nearly the same as
"ma'na", which appears already in Al-farabi's De intellectu et intellecto6 and
was later used by Avicenna to signify the reality of the k n o w n considered
as known. Thus ma'qul, ma'na, or intentio is that which is immediately
before the mind, whether the object of the intention is outside the mind (in
which case the intention is ά first intention) or itself an intention (in which
case the intention is a second intention). The distinction between first and
second intentions was prefigured in Al-farabi's theory of abstraction. 7
Second intentions associated with first or primarily k n o w n intentions
1. I have taken into consideration only such texts as have already been edited or partly edited and
discussed by scholars of medieval philosophy. Too few texts have been studied, especially of the
thirteenth and fourteenth centuries, to give more than a rather episodic history of the concept of
intentiones.
2. For this use of 'intentio' see Engelhardt 1976 with bibliography; and also articles on 'intentio' in
dictionaries of individual philosophers such as Schütz 1895, Baudry 1958, and Garcia 1910.
3. For the root of the concept of intentio in Arabic philosophy see Gyekye 1971.
4. Aristotle, De interpretatione, 16*3-18.
5. Al-farabi, De interpretatione, quoted by Gyekye 1971, p. 35, n. 16.
6. Al-farabi, De intellectu et intellecto in Gilson 1929-30, pp. 118, 119, 144.
7. See Madkour 1934, p. 140.
48ο Metaphysics and epistemology
are designated by Avicenna as the subject of logic. 8 This association of logic

with intentions considered as epistemological entities marks the starting
point of the development of an 'intentionalistic' logic or logic considered
as a scientia rationalis, parallel to and sometimes connected with the de
velopment of terminist logic, according to which logic was primarily
concerned not with concepts but with language, a scientia sermocinalis.9
According to Avicenna an intention is what the inner sense finds in things
over and above the phantasms found by the exterior senses. 10 Thus an
intention is nearly the same as a concept as well as the foundation of the
concept's content - o n e reason w h y the ontological status of intentions was
ambivalent from the beginning. 1 '
Roger Bacon on intentions

Roger Bacon considered intentions as 'determinations' of the things acting
upon the highest inner senses and leading to the cognitive operations of the
mind. Ontological problems aside, however, Bacon's threefold distinction
among modes of abstraction seems important for the epistemological and
logical application of the notion of intentions. The first mode is the
abstraction of a thing from a thing. This belongs to the domain of
mathematics, where mathematical entities (res mathematicae) are abstracted
from real things (sensibilia). Second is the abstraction of an intention from a
thing. This belongs to the domain of physics, where such intentions as man
and horse are abstracted from singulars. Finally, there is the abstraction of an
intention from an intention. This belongs to the domain of logic, which
deals with such second intentions as species in their association with first
intentions. The second intention species may, for instance, be founded on
the first intention man. The same is true of genus and other universal
2
concepts with which logic deals. ' (Genus, like species, is a second intention;
third and higher intentions seem not to have been recognised.)
8. Avicenna 1508, Metaphysica, I, ii; f. 70": 'Subiectum vero logicae, sicut scisti, sunt intentiones
intellectae secundo quae apponuntur intentionibus primo intellectis.'
9. For a characterisation of terminist logic see Kneale 1971, pp. 198-297, Boehner 1965, p. 228 and
Kretzmann 1967, pp. 37off.
10. Avicenna 1508, De anima. Ι, ν; f. 5": 'Intentio autem est id quod apprehendit anima de sensibili
quamvis non prius apprehendit illud sensus exterior."... 'Differentia autem inter apprehendere
formas et apprehendere intcntioncm hec est quod forma est ilia quam apprehendit sensus interior
et sensus exterior simul.' ( = Avicenna Latinus 1972, p. 86).
11. Avicenna 1508, says that an intention has a lesser being than its object has; Metaphysica, IX, iii; f.
I03 ,l> : 'Omnis enim intentio est propter id quod intenditur et est minoris esse quam id quod
intenditur.' On further complications in the ontological status of intentions, see Vescovini 1965,
pp. 64fr. and 8off.
12. The basic text is Bacon's Quaestiones super IV Metaphysicae, IV, i (Bacon 1905-40, XI, pp. 89-90).
Intentions and impositions 481
Thomas Aquinas on intentions and operations of the mind

Thomas Aquinas provided a much more detailed and systematic treatment
of intentions than Bacon had done. 13 In epistemological contexts he uses
the expression 'intentio intellecta' and defines such an 'understood intention'
as a likeness of the thing known, a likeness conceived within the intellect.
The intention is not the thing itself in its physical reality nor is it the
substance of the intellect itself; the intention is rather a sort of accident of
the intellect, a likeness of the thing signified by extra mental words. It is to
be identified with the inner word (verbum interius, or mentis), which is the
significate of the outer word. The being of the understood intention
consists in its being known. 14
Thomas distinguishes between what is primarily known — i.e., the
extramental reality to which the intellect is first directed as its objects -
which is the object offirstintentions; and the secondarily known - i.e., the
intellect itself in various modes of cognition, the domain of second inten-
tions.15 Second intentions (or secunda intellecta, as Thomas sometimes calls
them) are indeed the subject of logic, as Avicenna had said; but logic is
concerned with second intentions considered not as epistemological or
psychological entities but only as representations of the objects from which
they derive. Logic considers the concept or intention not in itself, but only
in the relation of likeness it has to the object conceived of. What is
represented in the intellect by a first intention is the intelligible nature (or
essence) of the thing known apprehended absolutely - i.e., before it has
been confined to either of the two possible modes of existence: in reality or
in the soul. Since intentions are related both to the thing known and to the
knower, two ways of studying their nature are possible: epistemology is
concerned with the correspondence of the understood nature to the nature
existing in reality, while logic studies the accidents that follow upon the
existence of this nature in the soul. Logic is concerned with what happens
to the absolute nature as a result of its being in the intellect. It is not the
13. The following remarks concerning Aquinas derive from Schmidt 1966, pp. 94fr.
14. Aquinas, SCG, IV, xi: 'Dico autem intentionem intellectam id quod intellectus in seipso concipit
de re intellecta. Quae quidem in nobis neque est ipsa res, quae intelligit ur; neque est ipsa substantia
intellectus; sed est quaedam similitudo concepts in intellectu de re intellecta, quam voces
exteriores significant; unde et ipsa intentio verbum interius nominatur, quod est exteriori verbo
significatum.'
i j . Aquinas, De potentia VII, ix: 'Prima enim intellecta sunt res extra animam, in quae primo
intellectus intelligenda fertur. Secunda autem intellecta dicuntur intentiones conséquentes
modum intelligendi: hoc enim secundo intellectus intelligit in quantum rcflectitur supra seipsum,
intelligens se intelligere et modum quo intelligit.'
nature itself represented in the intentio intellecta that is the subject of the
science of logic, but rather the accidents which that nature acquires from
the manner (or mode) in which it exists in the intellect. 16
According to Thomas' general division of the three operations of the
mind (operationes animae, or mentales), the first operation, the abstractive
apprehension of quiddities, gives rise to the intention of universality; the
second operation, the composition of judgement, gives rise to the intention
of attribution or predication; and the third operation, the discursive process
from one thing to another, gives rise to the intention of consequence
(consequentia).11 Although they exist in the intellect, all these intentions are
founded on natures that exist in the things themselves, and they are
attributed by the intellect to those natures. For example, the intention genus
does not exist in a donkey, but it is the nature of animal, which does exist in
the donkey, to which that intention is attributed. 18
Henry of Ghent on intentions and words

Henry of Ghent, referring to Thomas' doctrine of the 'intentio intellecta',19
took intentions to have being only within the actual consideration of the
intellect. Thus the intellect is able to form two intentions of one and the
same thing, although this duality must not be purely fictive. Just as leaves as
well as fruits are drawn from one root, so different intentions may be
drawn from one thing. An intention is the result of the relation or relations
between the knower and the thing known. An extramentally existing
thing is composed of intentions only virtually (not actually, as it is com-
posed of parts). The universal 'in potentia' becomes an intention 'in actu' as
soon as the thing is known in some way by the intellect. 20
In his Summa quaestionum ordinarium Henry draws distinctions among
three semantic levels: 21 (i) words signifying pure realities, (2) words
signifying first intentions, (3) words signifying pure (second) intentions.
16. See Schmidt 1966, pp. 124fr., for further evidence.

17. See Schmidt 1966, p. 127.
18. Aquinas, In I Sent., XXXIII, i, 1, ad 3: 'In omnibus autem intentionibus hoc communiter verum
est, quod intentiones ipsae non sunt in rebus, sed in anima tantum: sed habent aliquid in re
respondens, scilicet naturam, cui intellectus huiusmodi intentiones attribuât; sicut intentio generis
non est in asino, sed natura animalis, cui per intellectum haec intentio attribuitur.'
19. See Brown 1971, p. 253.
20. Cf. Henry of Ghent 1518, Quodl. V, xii, c; 171 Χ - Υ : 'Unde et intentio non dicitur esse aliquid in
re ut est extra, sed solum ut cadit in intellectus actuali consideratione considerantis unum in re ut
duo intentione, quod vere non fictive duo est intentione, quia in natura illius rei ut in fundamento
et quasi in radice est utraque intentio educenda de ea opere intellectus tamquam res rationis et
intellectus, quemadmodum in ligno velut in radice sunt folia et fructus ut diversa educenda opere
naturae tamquam res naturae.'
21. The following passages are based upon Henry of Ghent 1520, LIII, v. H-K; f. 64".
Words of level (1), such as 'this man' or 'Peter', signify singulars. The being
of these singulars in no way depends on their being considered by the
intellect. Words of level (2), such as 'man' or 'animal', are ambivalent in
their signification. On the one hand they signify a universal considered as
the product of abstraction from singulars; on the other hand they signify
the reality itself considered as actually formed by that universal nature.
Since the 'real sciences' (as distinct from the rational or sermocinal sciences)
are concerned with things only in their universal natures, it is words of the
second level rather than of the first that are used in the propositions of the
'real sciences'. Words of level (3) signify the relations between things that
have already been taken into consideration by the intellect. But among
words of the third level a further distinction is to be drawn between words
of logical intentions, which are founded on things, and words ofgrammatical
intentions, which have to do with words associated with things. Thus such
words for words as 'name', 'substantive', and 'verb' are words of gram-
matical intentions. Third-level words of logical intentions, such as 'genus',
'species', and 'differentia', are used in logic, the science that studies reality
insofar as it is expressible in words. Words for words are used in grammar,
where words for things are considered only in their status as words. Logic
thus has a stronger connection with reality than does grammar, which is
truly a scientia sermocinalis.22
The Pseudo-Kilwardby on intentions and words

Another three-part division of word-levels appears in a commentary on
the 'Priscianus maior' which until recently has been wrongly ascribed to
Robert Kilwardby.23 We are to distinguish among words for things,
words for intentions, and words for words. The imposition of words on
things takes place in metaphysics, the imposition of words on intentions
takes place in logic, and the imposition of words on words takes place in
grammar.24 The principles of grammar are the 'modi significandi'; the
22. Cf. Henry of Ghent 1520, LIII, v. I; f. 64": 'Consideratio primarum intentionum, quae est rerum
secundum se, pertinet ad scientias reales. Consideratio vero secundarum intentionum, quae vel est
circa res ut sunt exprcssibiles vocibus, et hoc quo ad intentiones logicales, vel est circa ipsas voces,
et hoc quo ad intentiones grammaticales, pertinet ad scientias sermocinales; et tamen logica minus
sermocinalis et magis realis est quam grammatica et quasi media inter scientias reales et
gramma ticam.'
23. For the text with an introduction including a discussion of the authorship see Pinborg et al. 1975.
24. Commentary on 'Priscianus Maior', 2.1.11, in Pinborg et al. 1975, p. 77: 'Primorum nominum
impositio pertinet ad metaphysicum, cuius est res generaliter et per se considerare. Secundorum
nominum impositio pertinet ad logicum, cuius est per se intentiones considerare; est enim logica
de sccundis intentionibus adiunctis primis, ut dicit Avicenna. Tertiorum nominum impositio
pertinet ad grammaticum, cuius est considerare nomina partium orationis secundum eorum
proprias rationes.'
principles of logic are the common intentions that are founded on things.
Since the principles of the two sciences have different causes, logic is not to
be subordinated to grammar.25
Robert Kilwardby on intentions and words

In the undoubtedly authentic De ortu scientiarum Robert Kilwardby iden-
tifies first intentions with the things themselves.26 Words of first intention,
such as 'substance' or 'quantity', signify the things themselves. Second
intentions are the 'rationes' of the things, such as universal or particular. The
relation between logic and grammar is not reflected within this context,
however, as it is in the divisions considered above.
The connection between intentions and impositions

During the thirteenth and fourteenth centuries two distinctions that had
been parallel distinctions in separate domains became increasingly inter-
connected: the distinction between first and second intentions (which has
been discussed above) and the distinction between first and second im-
positions.27 The latter distinction can be traced back to Porphyry 28 and
Boethius29 and is based on the observation that whereas some signs have
been imposed in order to signify non-signs, others are signs of signs.
Accordingly, words of first imposition are (conventional) signs of ex-
tralinguistic entities, and words of second imposition are (conventional)
signs of linguistic entities. In terms of this distinction Aristotle's Categories
was seen as a discussion of words of first imposition, while the subject
matter of De interpretatione involved words of second imposition, such as
25. Commentary on 'Priscianus Maior" 1.3.4, in Pinborg et al. 1975, pp. 25-6: 'Sed principia gram-
maticae non habent ordinem ad principia logicae, quia principia grammaticae sunt modi
significandi vel consignificandi, générales vel speciales dictionum. Principia per quae procedit
logica et quae considérât sunt communes intentiones fundatae in rebus, sicut sunt universale,
particulare, genus, species, causa, causatum et sic de aliis. Modi autem significandi res aut
consignificandi et communes rerum intentiones non habent ordinem sed potius disparationem,
cum a diversis causentur. Et ideo clarum est quod non subalternabit grammatica logicam.'
26. Robert Kilwardby 1976, 459, p. 157.24-32: 'Res enim ipsae sunt primae intentiones, et nomina
eas significantia, cuiusmodi sunt substantia, quantitas et huiusmodi, sunt nomina primarum
intentionum; sed rationes rerum, cuiusmodi sunt universale, particulare, antecedens, consequens
et huiusmodi, sunt secundae intentiones, et nomina eas significantia nomina secundarum in-
tentionum. Et dicuntur iliac primae et istae secundae, quia primo comprehend un tur res et deinde
ex consideratione et collatione rerum ad invicem colliguntur rationes earum.'
27. In the following passages I depend on Kretzmann 1967, pp. 369-71. For the history of the
doctrine of first and second imposition see Dal Pra 1954.
28. Cf. Porphyry 1887, pp. 57-8.
29. Boethius 1860g, PL 64, c. 159 B-C: 'Ergo prima positio nominis secundum significationem
vocabuli facta est, secunda vero secundum figuram: et est prima positio, ut nomina rebus
imponerentur, secunda vero ut aliis nominibus ipsa nomina designarentur.'
'name', 'verb', 'proposition', and the like. The distinction between im-
positions is obviously a close parallel to the distinction between first and
second intentions, since first intentions are (natural) signs of extramental
entities and second intentions are (natural) signs of first intentions and so
can be considered as signs of signs.
The two distinctions became complicated and confused for two main
reasons. First, there were, of course, vocal or written terms imposed on first
and second intentions - terms such as 'humanity' and 'genus'. Such terms
were all of first imposition, since the intentions they were imposed upon
were not conventional signs, but they could also be further described as
names of first intention - 'humanity' - or of second intention — 'genus'.
The second complicating factor was that the first and second intentions
themselves were considered to be the terms in mental propositions. Thus
whereas in the spoken or written proposition 'Animal is a genus' the
subject and predicate terms are both of first imposition, in the mental
proposition that animal is a genus the subject term is a first intention and
the predicate term is a second intention. The relationship between the two
distinctions was systematically elaborated by William Ockham. 30
fohn Duns Scotus on intentions and knowledge

Scotus' theory of intentions31 is characterised by a twofold understanding
of the intention in an epistemological context: the intention as a formal
structure in the thing itself and the intention as a concept.32 Scotus defines
the second intention as a rational relation (relatio rationis) of which one
relatum is an act ofjudging on the part of the intellect.33 Correspondingly,
a first intention is to be understood as a rational relation of which one
relatum is a simple act on the part of the intellect. First intentions are
30. See below.
31. The theory has been carefully examined in Swiczawski 1934. In the following passages concern-
ing Scotus I draw on this article, but I do not present S wiezawski's conclusions, since they are true
only of the Scotists, not of Scotus himself.
32. Reportata Parisiensia, II, d. 13, q. I, n. 4; Scotus 1891-5, XXIII, p. 440: '...nomen intentio
aequivocum uno modo dicitur actus voluntatis; secundo ratio formalis in re, sicut intentio rei, a
qua accipitur genus, differt ab intentione, a qua accipitur differentia; tertio modo dicitur con-
ceptus; quarto ratio tendcndi in obiectum, sicut similitude dicitur ratio tendendi in illud cuius est;
et isto modo dicitur lumen intentio, vel species lucis.' For present purposes the definitions given in
'secundo' and 'tertio' are essential.
33. Scotus, Ordinatio, I, d. 23, q. unica, 10; Scotus 1950-, V, p. 352.12-18: '... omnis intentio secunda
est relatio rationis, non quaecumque, sed pertinens ad extremum actus intellectus componentis et
dividends vel saltern conferentis unum ad alterum (hoc patet, quia intentio secunda - secundum
omnes - causatur per actum intellectus negotiantis circa rem primae intentionis, qui non potest
causare circa obiectum nisi tantum relationem vel relationes rationis).' Cf. also Lectura, I, d. 23, q.
unica, 12; Scotus 1950-, XVII, p. 306.10-13*.
directed towards the absolute forms that determine the objects of simple
intellectual acts. A second intention is produced by a complex act that
compares two objects previously apprehended by simple acts and formed
into first intentions. 34 Every object in its being known can be a first
intention. Even a second intention considered simply as being known can
become a first intention. Likewise, second intentions can be based on other
second intentions. The second intention genus considered as a formal
structure in the thing that is animal - i.e., the universal structure by which
animal is a genus of the species man - can be comprehended by a simple act
and as such can be a first intention. Put in relation to another first intention
it can constitute a further second intention. Every intention exists to the
degree to which it is effectively and concretely thought. In comparison
with extramental reality, an intention has diminished being, an 'esse
diminutum' or 'esse cognitum'.35 T h e logician has to study intentions as such,
while the metaphysician is concerned with them insofar as they are things.
The proper subject of logic is the second intention founded on the first, the
prime example of such a combination being the syllogism. 36
The role of intentions in speculative grammar
In the period between Aquinas and Scotus, the second half of the thirteenth
century, intentions were very eagerly discussed by an unusual group of
philosophers working in linguistic theory, the Modistae. 37 The Modistae
differ from the terminist logicians in that they all give a special ontological
interpretation of the modi significandi, one that determines the ontological
interpretation of logic. The key concept of terminist logic, supposition, is
seldom used by the Modistae, and then only in a rudimentary way. Their
interests tend more to metalogical questions, to the character of logical laws
and logical concepts, than to the development of logic itself. The logic of
the Modistae is even further from Aristotelian logic than is terminist logic,
although it, too, finds its source in Aristotle's logical writings. And al-
though its non-Aristotelian developments are quite different from those of
terminism, it is also influenced by some theories of terminist logic. The
34. See Swiezawski 1934, pp. 22iff.

35. See Swiezawski 1934, p. 236.
36. See Swiezawski 1934, pp. 2388".
37. To these belong, among others, Thomas of Erfurt, Peter of Auvergne, and Simon of Faversham
as well as Duns Scotus (as a commentator on Aristotle's logical works), and Radulphus Brito. For
the history of the 'modi significandi' see Pinborg 1967a; 1972, pp. I02ff; Bursill-Hall 1971, pp. 42fr.
For manuscripts and editions of philosophers mentioned above see Pinborg 1975, pp. 4iff.
Modistae chiefly taught at Paris, while the Oxford tradition remained

more closely associated with terminism.
According to the Modistae, logic is to be divided on the basis of the three
operations of the mind: apprehension, judgement, and ratiocination. Each
of these operations represents an act from which special logical concepts or
second intentions result. These are the main subject of logic. T o the first
operation, the simple apprehension of quiddities, belong simple intentions
- i.e., such non-complex terms as are treated in Porphyry's Isagoge and
Aristotle's Categories: 'genus', 'species', 'differentia', and so on. Intentions
referring to propositions belong to the second operation, predication and
judgement. They are treated in Aristotle's De interpretatione and in the first
book of the Posterior Analytics. Intentions referring to a set of propositions
belong to the third operation, ratiocination - e.g., the intention syllogism -
and these are treated in the Prior Analytics, the second book of the Posterior
Analytics, the Topics, and the Sophistici elenchi.38
All the intentional entities, first intentions as well as second intentions,
are drawn from the properties of the objects signified. A proper modus
essendi corresponds to each first intention, a common modus essendi to each
second intention.
Intentions and speculative grammar in Radulphus Brito

In the writings of Radulphus Brito 3 9 the modistic analyses are refined and
extended. Partly in connection with grammatical analysis Brito develops
the following theory of intentions, based on the triad of thing, concrete
intention, and abstract intention. Whenever a thing is known, a cognition
arises in respect of an abstract intention. The act of cognition and that
intention are identical. 40 The intentions originate from the fact that the
object and the active intellect act together on the potential intellect, the
active intellect's role being to abstract from all the particular and material
conditions of the object. 41 The resulting abstract intention is signified by
such abstract words as 'humanity' or 'animality'; it is the essence or pure
form of the object or of the feature conceived. This abstract intention has
38. Sec, e.g. Boethius of Dacia 1976, pp. 3ff.; Radulphus Brito, Prooemium to his Quaestiones super
Porphyrium (forthcoming).
39. The important texts are to be found in Brito's commentary on Porphyry, in his Quaestiones on
Aristotle's De anima, in his commentary on the Sentences and in various Sophismata, especially in
the sophisma 'Aliquis homo est species'. For a survey of Brito's life and works, manuscripts and
editions, see Pinborg 1975a, pp. 71-97, Fauser 1973, pp. 3-36, and the forthcoming edition of
Brito's Quaestiones super Porphyrium.
40. Brito, De anima III, q. 25; in Fauser 1973, pp. 296-7.
41. Brito, De anima III, q. 16; in Fauser 1973, p. 236.
being only in the intellect and is a psychological entity. 42 What is known is

not the intention, however, but rather the thing as it stands under that
abstract intention. 4 3 The complex aggregate consisting of the thing and
the abstract intention is called the concrete intention. The abstract inten-
tion is the same as the cognition, and the concrete intention is the same as
the thing known. In its material part - i.e., the thing - the concrete
intention exists extramentally ; in its formal part - i.e., the intention per se -
it exists in the intellect. Thus the two intentions, abstract and concrete, are
formally identical but materially distinct. In an analogy with real accidents,
it might be said that there is a relation of denomination between the
intention and the thing. Just as real accidents denominate their subjects, so
intentions denominate their objects. For example, the whiteness inherent
in a man is the reason why the man can be correctly called white; that same
whiteness is also the reason why the intellect can form the intention
whiteness and associate it with that man in a concrete intention. Objects
thus cause intentions and are in turn denominated by them; but this must
not be taken to mean that the intention thus formed is a part of the object. 44
Only a concrete intention is predicable of an object, not qua concept (its
formal aspect) but only qua thing (its material aspect). The abstract in-
tention represents the formal aspect under which the predication is accom-
plished. The real man is predicated under the aspect of his humanity.*5
Whenever the cognition of an object occurs under the aspect of a proper
modus essendi of the thing, an absolute and primary cognition arises, a first
intention. Whenever the cognition occurs under the aspect of a modus
essendi which the thing has in common with other things, a relative,
secondary cognition arises. The common modi essendi, from which the
second intentions are drawn, do not belong to the essence of the object but
are rather the foundation of an external relation. 46 The foundation of the
cognition of a second intention — e.g., as expressed in the proposition 'Man
is a species' — is the same man as constitutes the foundation of the cognition
of a first intention, as expressed in the proposition 'Man is an animal'. The
42. Brito, In Porpkyrium, q. 11; in Pinborg 1974, p. 52, n. 18: 'intentio in abstracto est de con-
sideratione naturalis'.
43. Brito, De anima. III, q. 7; in Fauser 1973, p. 176: 'Illud quod intelligitur, de se est quiditas rei
secundum se, cui accidit et esse signatum et esse abstractum. Tamen intelligitur sub esse quod
habet in anima, ita quod illud esse quod habet in anima, non est illud quod intelligitur, sed illud,
sub quo res intelligitur.'
44. Brito, In Porphyrium, ct. 8; in Pinborg 1974, pp. 51-2.
45. Brito, In Porphyrium, q. 13 and q. 8; in Pinborg 1974, p. 53.
46. Brito, In Porphyrium, q. 5 and q. 7; in Pinborg 1974, pp. 52-3.
modus essendi is different in each case, but in both cases the real man is the
subject.47
The predicates of second intention are the logician's concern. He has to
work out the accidental relations of those predicates and to estabhsh the
logical rules that are accidental in relation to the quidditative cognition of
the things. The second intentions belong to the category of quality. They
are passiones animae caused by the real modi essendi. The logician does not
consider them in this respect, however, but rather in their accidental
relations when they are considered as the likenesses of the things.48
In the domain of the first mental operation, the apprehension of quid-
dities, the connection between the corresponding second intentions, such
as the universals genus or species, and the thing itself is guaranteed in the
sense mentioned above. On the level of the second and third mental
operations, judgement and ratiocination, however, it becomes difficult to
preserve the connection between second intentions and the things. Within
these latter operations it is not really the things that are the immediate
objects, but rather certain constructs of the intellect - complex objects, as
Brito calls them - and the relations of these objects to other 'complexa'
existing only on a mental level. Thus it seems that not all second intentions
share the same degree of reality.49
According to a revised version of Brito's theory,50 even the intentions of
the higher mental operations are materially connected with the thing,
because the relation between the terms (concepts) in a proposition or in a
syllogism corresponds to a modus essendi of the thing in virtue of the fact
that all logical operations have to be interpreted and verified of the things.
This interpretation and verification is not the task of the logician, however.
He need not concern himself with the specific nature of the objects. Instead,
the logician considers them as genus, as species, and so on. In this way he is
able to draw conclusions, but not in his role as logician to verify them.51
The bond between logic and the real sciences is not broken in principle,
however. According to Brito and the Modistae, logic is the speculative
47. Brito, In Porphyrium, q. 8; in Pinborg 1974, p. 53, n. 19.

48. See Pinborg 1974, p. 54. and Pinborg 1969. p. 174. where Pinborg gives an extract from the text
of Brito's sophisma 'Aliquis homo est species'.
49. In this sense Brito discusses the intentions which belong to the higher operations in the sophisma
mentioned above. Sec Pinborg 1972, pp. 91-2. For another extract from the text see ibid. 197-9
and in Pinborg 197$, p. 58, n. 69.
50. Given in Brito's sophisma 'Omnis homo est homojomnis homo de necessitate est animal'. Cf. Pinborg
1975b. p· S*, and Pinborg 1969, pp. 170-4.
51. See Pinborg 1972, pp. 8off., and Pinborg 1975a, pp. 58-9.
science of'the conceptual order that parallels the synthetic structure inter
nal to the things'. 5 2
By his treatment of second intentions as the subject of logic Brito
secured the foundations of a logic that was developed by various thinkers,
especially by the Thomists and the Scotists. Treatises on this sort of logic
continued to be written in the sixteenth century. 5 3 Reflections on the
intentions found their natural place in the commentaries on Aristotle and
on the Sentences (especially in the context of the question concerning the
theology of the Trinity, whether 'person' is a word of first or of second
intention: I, d. 23) as well as in special treatises De intentionibus.
Reactions to Brito
In the writings of Hervaeus Natalis 5 4 and Peter Aureoli the theory of
intentions developed by Brito was more or less modified. But it was also
attacked rather forcefully both by nominalists (especially Ockham) and by
realists (especially Walter Burley).
Peter Aureoli on intentions

According to Peter Aureoli 5 5 an intention is not to be identified with the
act of cognition. What is being predicated (both in first and in second order
predications) is neither acts of cognition nor extramental things, but
rather the objective concept of the thing. ('Objective' here is taken in the
medieval sense in which it indicates the conceptual content that appears to
the intellect but is strictly speaking not a component [subiective] part of the
intellect in the way actually formed concepts considered as psychological
entities are component parts of the intellect.) Thus second intentions are
not founded on the primary cognition of the thing but rather on the
56
objective content of the first order concept that is formed. The cognition
establishes not only a relation of denomination between intention and
thing but also a special 'intentional being' (esse intentionale). It is in its
52. Moody 1935, p. 303·

53. For the history of this development see Hickmann 1971.
54. See Pinborg 1974, pp. 54—5, who gives some information about Hervaeus' theory in his treatise
'Dc secundis intentionibus' (MS Basel UB Β III, 22, especially f. 143), and Domanski 1966.
55. For the discussion of Aureoli I depend on Pinborg 1974, pp. 56-9. The basic text is Peter Aureoli,
In Sententias, I, d. 23, art. 2 (MS Vat. Borgh. lat. 329, ff. 26or-264'). A short survey of Aureoli's
arguments is provided by the list of 'Capitula' in Aureoli 1952-6, pp. 59-61.
56. Aureoli, In Sententias, I, d. 23, art. 2 (f. 260"): 'Actus autem intellectus quo universalitas
apprehenditur non inest tamquam subiecto actui quo intelligitur animal. Quam vis enim inter eos
sit ordo praesuppositionis, non tamen inhaerentiae habitudo. Ergo prima et secunda intentio per
prius et formalius se tenent ex parte obiectivorum conceptuum quam ex parte actus intellectus.'
intentional being that the thing appears to the intellect, but the appearance
takes on different forms according to the intellect's settling on one or
another objective concept. For example, the concepts man and animal,
which are both predicable of Socrates, are distinct intentional entities of
this kind, not to be found in Socrates himself.57 Aureoli himself gives the
following concise account of his theory of intentions:
An intention is the objective concept formed by the intellect containing in itself

indistinguishably the passive conception and the thing conceived. 'Intention' is the
same as 'concept', and 'first intention' the same as 'concept of the first order',
which the intellect forms of the things without reflecting on its own concepts.
Second intentions are concepts of the second order, which the intellect constructs
by reflection on the first concepts: universality, predicability, and the like concerning
the simple mental operation of apprehension; affirmation or negation concerning the
second mental operation of judgement; and the connection of the extremes of a
syllogism concerning the third mental operation, the discursive and ratiocinative
act.58
Walter Chatton and William Ockham on intentions and impositions

Aureoli's separation of the content of the concept from the act of cognition
and his association of intentions with a special esse obiective brought onto-
logical problems to the fore again. These problems were clearly discerned
by Walter Chatton, a contemporary of Aureoli and Ockham. Chatton's
criticism was directed against the esse obiectivum of the concept, which he
saw as an unnecessary mediator between object and cognition. Ockham
accepted Chatton's view after having propagated the so-called fictum-
theory for some time. 5 9 For Ockham, however, this question was a
secondary problem for his theory of concepts or intentions. 6 0 According
to Ockham the primary problem is whether the intentions are not simply
to be identified with the mental terms of propositions and arguments.
During an earlier period, in his commentary on the Sentences, Ockham
had defined the first intention as the really existing thing itself, the second
intention as something in the soul that bears a relation to the things and can
be predicated by means of words which in a proposition have simple rather
than personal supposition. 61 In his later Summa logicae Ockham provided
57. Sec for the text Pinborg 1974, pp. 56-7.

58. See for the text Pinborg 1974, pp. 58-9.
59. For the controversy between Ockham and Chatton see Gâl 1967c, pp. 191-212.
60. See Knudsen 1975, pp. 10-14.
61. Ockham 1495-6, In Sententias, I, d. 23 D: '... intentio prima vocatur res realiter existens. Intentio
autem secunda vocatur aliquid in anima rebus applicabile praedicabile de nominibus rerum,
quando non habent suppositionem personalem, sed simplicem'.
definitions of intentions that are suited to the intellectio-thcory propagated

by Chatton, according to which a concept is the act of cognition itself.62
According to Ockham there are no intentiones ex parte rei corresponding
to the intentions properly so-called, as in Brito's theory. There is no need
for entities intermediate between intentions and objects. An intention is
something in the soul which is a natural sign signifying something for
which it can stand, or which can be a part of a mental proposition. 63 A first
intention is a natural sign of something that is not itself a sign. A second
intention is a natural sign that signifies other natural signs or first inten-
tions. The first intention or concept man naturally signifies all men and
each individual man; the second intention species is a natural sign of the
natural sign man and of other species-intentions. The signs of spoken
language correspond in a special way to the intentions considered as natural
language signs: they are arbitrarily (ad placitum) connected with certain
significates- i.e., they have been imposed by men on certain objects in the
relation of signification, the intellectual side of which involves the
intentions.
Ockham makes use of the traditional distinction between words of first
and of second imposition in the following way. Spoken names (nouns or
adjectives) are either of first or of second imposition, and within this
general distinction of names the distinction between first and second
intentions is a sub-distinction. Names of second imposition are names of
n a m e s - i.e., they are conventional signs that signify conventional signs as
such. For instance, the spoken word 'name' signifies every name such as
'man' or 'animal'. In this connection Ockham distinguishes between
names of second imposition in the strict sense — those that have no
corresponding elements in mental language, e.g., the name 'conjugation' -
and names of second imposition in the broad sense - those that do have
such corresponding elements, e.g., 'verb' and 'name'. 6 4 Names of first
imposition are all those that are not names of second imposition either in
the strict sense or in the broad sense; but syncategorematic words such as
'not' or 'only' do not belong to names of first imposition in the strict sense;
indeed, they are not names at all.
Ockham introduced the distinction between names of first and second
62. In the following discussion I depend on Boehner 1958, pp. 201-32; the basic text is Summa logicae
I, Chapter 12; Ockham 1974a, pp. 41—4. An English translation of Summa logicae I is in Loux 1974;
for our text see pp. 73-4.
63. Ockham 1974a, I, c 12; p. 43.41-4.
64. Ockham 1974a, I, c. 11; pp. 38.4-39.31.
intention (a distinction applied here of course to spoken signs and not to
intentions themselves or mental signs) within the class of names of first
imposition in the strict sense.65 Names of second intention in the strict
sense are imposed precisely in order to signify intentions of the soul, which
are natural signs. Names of second intention in the broad sense are imposed
on conventional signs in their capacity as signs. In this broad sense names of
second intention can be names either of first or of second imposition.
Names offirstintention are imposed in order to signify things or objects
that are neither signs nor derived from signs — e.g., 'Plato', 'whiteness'.
Some names, however, signify both things and signs; these are the so-called
transcendentals; 'unum', 'verum', 'bonum', etc. The transcendentals are
nevertheless considered names of first intention.
Names of second imposition belong to the domain of grammar, names
of first imposition and second intention belong to the domain of logic,
names of first imposition and first intention belong to the domain of the
real sciences, and names imposed on the transcendentals belong to the
domain of metaphysics.66
A controversy arose between Ockham and Chatton concerning the
distinction between first and second intentions and their respective sup-
position in a proposition,67 For Ockham a first intention is what has
personal supposition in a proposition, suppositing for really existing sin-
gulars. Words of first imposition and first intention signify singulars
directly, and so in their case what is supposited and what is signified
coincide. A second intention is a concept that is predicated of a subject that
has simple supposition. Thus in the proposition 'Man is a species', 'species'
is a second intention and supposits for the same as the word 'man' supposits
for in simple supposition - viz., the concept man. Words of first imposition
and second intention signify concepts as concepts.68 A first intention
extensionally intends singulars; a second intention intends concepts of
singulars insofar as those concepts are used for themselves rather than for
the singulars - i.e., insofar as they have simple rather than personal
supposition. Logic deals with second intentions.
For Chatton a first intention is a natural sign of singulars in their
quiddity (or essence). A word of first intention -e.g., 'man' - supposits for
65. Ockham 1974a, I, c. 11; pp. 39.36-40.55.

66. Ockham 1974a, 1, c. 11; pp. 40.56-41.78.
67. For Ockham's theory of supposition sec Boehner 1958, pp. 232-67; the main text is Ockham
1974a, I. cc. 63-7: PP· 193-238-
68. Ockham 1495-6,1, d. 12, C; see also Knudsen 1975, pp. 7-8.
the thing signified in its quiddity. What is supposited and what is signified
coincide only de facto, not per se. A word of second intention stands in a
proposition of logic for a concept suppositing simply, but not only for the
concept as concept, also for the concept in its function as a sign of many
things. The word 'species' signifies the concept man not insofar as it is a
concept but insofar as it is predicable of many. A second intention is a
connotative concept that signifies a concept as a sign of singulars. 69
According to Chatton a concept is signified by a word of second intention
not only in its extensionality but also in its intensionality — i.e., in its
character as a sign, which derives its applicability to many singulars from
the inner structure of the things. Logic deals with the relation between the
sign (second intention) and its significate (first intention), which is not a
real relation. Thus logic is not a 'real science' - i.e., science of reality. 70
Walter Burley on intentions and the nature of logic

Ockham criticised the logic of intentions introduced by the Modistae
particularly because he rejected any sort of reification of the relations
between extramental things and concepts or intentions. Realists such as
Walter Burley also opposed the Modistic theory, however. 7 ' According to
Burley a word can exercise its function of signification without a special
form mediating the act of signification. Just as a corporeal sign, which is a
natural thing, can be introduced (or imposed) to signify something - e.g.,
the hanging barrel-hoop to signify the presence of wine in a tavern - so a
natural vocal sound can be introduced to signify something without a new
artificial form being added to the sound. 72 Second intentions inhere in the
69. Chatton, /11 Sententias, I, d. 23, art. 2, in Knudsen 197s, p. 24: 'Dico, quod difficile est assignare
bonam differentiam. Dico tamen, quod intentio prima est ilia, quae significat rem esse talem,
qualis est in essendo, secunda, quae significat rem esse signum alterius rei, cuiusmodi sunt
pracdicari de pluribus vel pracdicari de uno solo. Nam conceptum hominis praedicati de pluribus
est ipsum esse signum plurium, non quidem significans conceptum esse talem naturam, sed esse
signum. Est igitur intentio secunda conceptus relativus significans conceptum ilium esse signum
plurium vel unius.'
70. Ibid, in Knudsen 1975a, pp. 25-6: 'Sed contra . . . tunc logica esset scientia realis, quia relatio signi
ad significatum est realis, de qua tractat logica . . . dico, quod supponit, quod relatio signi ad
significatum esset realis, et non est sic'
71. For this passage see Pinborg 1975a, p. 60.
72. Walter Burley, In Perihermenias, cap. 'de nomine' (for the text and the manuscripts see Pinborg
1975a, p. 60, n. 83): ' . . . quamvis nomen sit nomen per institutionem imponentis tamen non est
res artificialis, quia per hoc quod instituitur ad significandum nulla forma accidentalis sibi
inhaerens acquiritur. Res enim praesupposita omni impositione imponitur ad significandum.
Unde signum, quod est res naturalis, puta circulus, potest institui ad significandum vinum in
taberna, ita vox naturalis potest institui ad significandum absque hoc quod aliqua forma nova sibi
acquiratur.'
things; they are parts of extramental reality. Whereas Ockham located the
intentions strictly within the mind, Burley locates them strictly outside the
mind, each man in his different way opposing the Modistae. Since accord-
ing to Burley the intentions are features of reality, logic, which deals with
the intentions, is indeed a 'real science'.73 This extraordinary interpretation
of the nature of the intentions and of the nature of logic was carried on by
the so-called Averroists of Bologna.74
73. Walter Burley 1478, f. 14"): '... liber praedicamentorum est de rebus secundum quod eis insunt
intentiones secundae, scilicet intentio generis generalissimi et generis subalterni, intentio speciei et
sic de aliis'.
74. See Pinborg 1975a.
24
D E M O N S T R A T I V E SCIENCE
Scientia demonstrativa in the Middle Ages

From the thirteenth century down to the Renaissance, philosophers at-
tempted to forge plausible accounts of Aristotelian 'demonstrative science'
and its basis, the 'knowledge-producing syllogism' (syllogismusfaciensscire).
The term 'scientia demonstrativa' is ambiguous, referring both to the knowl-
edge a demonstrative syllogism effects in someone who understands it and
to a system of syllogisms comprising propositions which satisfy the re-
quirements for demonstration stipulated in Aristotle's Posterior Analytics.
In expounding Aristotle's theory, medieval authors typically interpret and
criticise it in the light of their own conceptions and doctrines; for example,
their treatments of the requirements that premisses of demonstrative syl-
logisms be true, necessary, and certain invoke various views of truth,
necessity, and certainty. * So while it is true that almost all the major figures
of medieval philosophy in some sense endorse what is traditionally called
'the Aristotelian ideal of demonstrative science', this appearance of un-
animity can be misleading. The generalisation that this ideal dominated
medieval thinking regarding scientific knowledge obscures or ignores the
variety in philosophical accounts of its foundation and scope.2
Much of our current understanding of medieval epistemology is based
on doctrines concerning acquaintance with and knowledge of particular
entities or states of affairs, and the subsequent formation of general con-
cepts. Because discussions of more elaborate cognitive activities, those
involving relatively complex judgements and inferences, have until re-
cently received less attention, the full significance of theories of demon-
strative science in late medieval epistemology remains to be determined.
Similarly, the relationship of those highly abstract theories to the actual
1. For reports of the medieval Latin commentaries on the Posterior Analytics, see Lohr 1967-74.
Discussions relevant to demonstrative science are also regularly found in comments on Book VI,
Chapter 6 of the Nicomachean Ethics and on Book I, Dist. 39 of the Sentences.
2. Barnes summarises a frequently held view: '... the Analytics imposed on the learned world a
narrow, blinkered logic, and a stultifying theory of science' (Barnes 1975, ix).
Demonstrative science 497
practice of natural philosophy and other medieval 'sciences' remains to be

established. These topics are closely connected, since the question whether
it is possible to attain scientia demonstrativa of the natural world was itself at
issue. Mathematics - particularly geometry - was often thought to come
closest to satisfying Aristotle's requirements for demonstrative science.3
But the status of natural sciences with respect to the demonstrative ideal
was suspect insofar as their premisses fell short of the truth, necessity, or
certainty of geometrical axioms. Thus the import of a philosopher's allegi-
ance to the ideal of demonstrative science varies according to his position
on at least three topics: (i) the interpretation of the requirements for a
demonstrative syllogism; (2) the relationship between demonstrative
science and other sorts of knowledge; and (3) the possibility of attaining a
demonstrative science of nature.
The theory of demonstrative science in the Posterior Analytics
The basic tenets of the theory of demonstrative science Aristotle presents in

the Posterior Analytics are simple to identify but difficult to understand and
assess. According to Aristotle, full-fledged scientific knowledge of some-
thing requires understanding its necessitating causes; this knowledge is
produced or best manifested by a demonstrative syllogism.4 A demonstra-
tive syllogism yields knowledge of a 'reasoned fact' by showing not just that
something is so, but why it always must be so; the medievals marked this
distinction by the phrases 'demonstratio quia' (demonstration of the fact) and
'demonstratio propter quid' (demonstration of the reason why). 5 Such ex-
planatory demonstrations must be syllogisms in Barbara whose premisses
are true, necessary, certain, immediate, and appropriate to the phenom-
3. Many, though certainly not all, of Aristotle's examples arc mathematical, and in particular,
geometrical; for a discussion of ancient geometrical method and its subsequent history, see
Hintikka and Remes 1974.
4. 'Hence that of which there is understanding simpliciter cannot be otherwise... [It is what we]
know through demonstration. By demonstration I mean a scientific deduction; and by scientific I
mean one in virtue of which, by having it, we understand something' (Post. An. 7l b i6-l9; see
also 73*21-5). (All quotations from the Posterior Analytics follow the translation in Barnes 1975.)
In part because Aristotle provides no examples which clearly meet all his requirements for
demonstrative syllogisms, commentators often emphasise the potential utility of demonstrative
syllogisms for displaying or teaching scientific explanations, rather than their efficacy for the
discovery of scientific knowledge. See, for example, Barnes et al. 1975, pp. 77-87. An example
which appears to meet the requirements for demonstration is this: 'Having incisors belongs
necessarily to every carnivore. Being carnivorous belongs necessarily to every dog. Therefore:
having incisors belongs necessarily to every dog.' See Barnes et al. 1975, p. 66.
5. Post. An. \, Ch. 13.
enon to be explained. 6 As a proximate cause of what is explained, the

middle must be simultaneous with its effect.7 If a syllogism which meets
these requirements is to manifest the nature or necessity of its 'reasoned
fact' to a particular person, he must realise that it satisfies all these require-
ments and thereby explains the fact.8 While the logical structure of dem-
onstrative or 'scientific' syllogisms is simple, the additional requirements
severely limit the number of full-fledged 'scientific' syllogisms. Perhaps
Aristotle was content to posit demonstrative syllogisms as an ideal for
scientific explanation which could be only approximated in the practice of
most sciences.
The early reception of the Posterior Analytics

From John of Salisbury's Metalogicon, we know that the Posterior Analytics
was translated by 1159,9 but the first full Latin commentary now known
was composed by Robert Grosseteste only towards the end of the first
quarter of the thirteenth century. ' ° The slow reception of the Posterior
Analytics by twelfth- and even thirteenth-century philosophers is not
surprising in view of the difficulty of the text and the differences between
its doctrine and the Augustinian assumptions about truth and knowledge
which pervaded early medieval thought.
The Augustinian background

While the details of Augustine's theory of knowledge fluctuate somewhat,
several fundamental tenets are fixed. God is identified with Truth or
Supreme Truth, and the forms or eternal exemplars of things are identified
6. ' . . . it is necessary for demonstrative understanding in particular to depend on things which are
true and primitive and immediate and more familiar than and prior to and explanatory of the
conclusion (for in this way the principles will also be appropriate to what is being proved)' (Post.
An. 7i b 20—3). 'Demonstration is deduction from what is necessary' (73*24—5).
7. 'For the middle term must be coeval - something that came about for what came about,
something that will be for what will b e , . . . something that is for what is; but it is not possible for
anything to be coeval with "it has come about" and "it will b e " ' (Post. An. 95*38-42).
8. 'Similarly, (there is) both knowledge and opinion of the same thing. For the one is of animal in
such a way that it cannot be an animal, and the other in such a way that it can be - e.g. if the one is
ofjust what is man, and the other of man but not of just what is man' (Post. An. 89*33—8).
9. The summary occurs in Book IV, Chs. 6 - 8 . It is interesting to note that John emphasises the
connection between demonstrative science and pedagogy; one of the merits of having dem-
onstrative knowledge of a subject, he says in C h . 7, is that it enables one to teach the subject well.
Although there were several subsequent translations, the fust translation, by James of Venice,
remained the standard one. These texts are contained in Minio-Paluello 1968; for a discussion of
the quality of James' text, see Dod 1971.
10. For a discussion of the dating of this w o r k , see Dod 1971, pp. 46fT.
with ideas in the divine mind.11 Insofar as human knowledge involves

access to these exemplars, it requires divine aid of some sort.12 Augustine's
most frequent metaphor for knowledge is vision which provides a clear
and direct acquaintance with its object; hence the divine aid requisite for
knowledge came to be known as divine illumination.13 Since full under-
standing or comprehension involves direct acquaintance with eternal ex-
emplars, the highest epistemic state is reached only in a vision of Truth
itself.14
But this ideal does not exclude the possibility of our having knowledge
of temporal objects or of truths given in propositions.15 Augustine relates
his epistemic ideal to his account of mundane knowledge by explaining
that in its proper sense the term 'scientia' refers to knowledge of temporal
things, while in its improper or extended sense it refers to knowledge of
eternal things, which is more properly called wisdom. 16 Because he be-
lieves that wisdom supersedes knowledge, he accords primarily instru-
11. 'And Truth isour God who liberates us...'; 'Haec est libertas nostra, cum isti subdimur veritati; et
ipse est deus noster qui nos libérât...' (De libero arbitrio, Book II, sec. 37). 'But he does not,
therefore, know all his creatures, both spiritual and corporeal, because they are, but they are
because he knows them.... He created, therefore, because he knew... this knowledge, therefore,
is quite unlike our knowledge.'
'Universas autem creaturas suas, et spirituales et corporales, non quia sunt ideo novit; sed ideo sunt
quia novit. ... Quia ergo scivit, creavit... longe est ergo huic scientiae scientia nostra dissimilis'
(De Trinitate Book XV, sec. 13).
12. 'Concerning universals of which we have knowledge, we do not listen to anyone speaking and
making sounds outside ourselves. We listen to the truth within us which presides over our
minds... our real teacher is Christ.'
'De universis autem, quae intelligimus, non loquentem, qui personal foris, sed intus ipsi menti
praesidentem consulimus veritatem ... Ille autem, qui consultitur, docct... est Christus...' (De
magistro, sec. 38).
13. 'But when we have to do . . . with the intelligence and with reason we speak of things which we
look upon directly in the inner light of truth which illumines the inner man and is inwardly
enjoyed.'
'Cum vero de his agitur ... intellectu atque ratione, ea quidem loquimur. quae praesentia
contuemur in ilia interiore luce veritatis, qua ipse, qui dicitur homo interior, inlustratur et
fruitur...' (De magistro, sec. 40).
14. 'In the preceding book ... we were occupied in distinguishing the function of the rational mind
with respect to temporal things... from the more excellent function of this same mind when it is
employed in the contemplation of eternal things.'
'In libro supcriore . . . egimus discernere rationalis mentis officium in temporaiibus rebus... ab
excellentiorc eiusdem mentis officio, quod contcmplandis aeternis rebus impenditur . . . ' (De
Trinitate Book XIII, sec. 1).
15. Ά human mind, therefore, knows all these things which it has acquired through itself, through
the senses of its body, and through the reports of others.'
'Haec igitur omnia, et quae per se ipsum, et quae per sensus sui corporis, ct quae testimoniis
aliorum percepta seit animus humanus* (De Trinitate Book XV, sec. 12, 22).
16 'If this is the correct distinction between wisdom and knowledge, that intellectual cognition of
eternal things should pertain to wisdom, but rational cognition of temporal things to knowledge,
then it is not hard to judge which is to be ranked above and which below.'
mental value to knowledge of the natural world, which is a source of signs

of the attributes of its creator. 17 And so it is not surprising that Augustine's
writings fail to provide extensive analyses of such topics as cognition,
induction, and judgement.
Augustine's view of truth and knowledge left several tasks for sub-
sequent Christian thinkers. They had to show how any additional or
alternative account of knowledge relates to the putative fact of divine
illumination, and then provide a fuller treatment of our acquisition of
knowledge of the natural world and of the truth of our judgements about
its states of affairs. Augustine's characterisation of God as Truth led to a
conception of truths as relationships between something in the world and
the relevant divine idea. O n the. one hand, things in the world are true if
they are genuine tokens of the types which God conceives; on the other
hand, our thoughts are true when they correspond to what God thinks. 18
The relationship of states of affairs in the natural world to concepts, and the
nature of the truth of concepts, judgements, or propositions required
considerable clarification.
'Si ergo haec est sapientiae et scientiae recta distinctio, ut ad sapicntiam pertineat aeternarum
rerum cognitio intellectualis; ad scientiam vero, temporalium rerum cognitio rationalis: quid cui
praeponendum sive postponendum sit, non est difficile judicare' (Ibid., Book XII, sec. 15, 25).
'But now w e speak o f knowledge, later of wisdom . . . And we do not take these terms so strictly
that one cannot speak either of wisdom in human affairs or knowledge in divine ones. When
speaking in broader fashion, one can talk in both cases of wisdom or o f knowledge.'
'Sed nunc de scientia loquimur, post de sapientia . . . Nec ista duo sic accipiamus, quasi non licet
dicere, vel istam sapicntiam quae in rebus humanis est, vel illam scientiam quae in divinis.
Loquendi enim latiore consuetudine, utraque sapientia, utraque scientia dici potest' (Ibid., Book
XIII, sec. 19, 24).
17. 'For without knowledge we cannot even possess the very virtues by which we live rightly . . .
and arrive at that eternal life which is truly blessed.'
'Sine scientia quippe nec virtutes ipsae, quibus recte vivitur . . . ut ad illam quae vere beata est,
perveniatur aeternam' (Ibid., Book XII, sec. 14, 21).
' . . . w h o e v e r . . . does not inquire further to learn the source of truths... can be seen to be in no
way wise.'
' . . . quisquis . . . non potius quaercre, unde sint vera . . . videri potest, esse autem sapiens nullo
modo' (De doctrina Christiana Book II, sec. 38, 57).
18. Augustine characterises the foundation of thought as an inner, prelinguistic word: *... a true
word is produced when w e say what we know . . . the word is most similar to the thing known,
and is that word of no language from which its image is also produced.'
' . . . gignitur verbum verum, quando quod scimus loqui mur . . . est verbum simillimum rei notae,
de qua gignitur et imago eius . . . quod est verbum linguae nullius' (De Trinitate Book X V , sec. 12,
22).
The 'inner word' resembles either a temporal state of affairs, in the case of sensible truths, or an
eternal exemplar, in the case o f intelligible truths. See also Book XV, sec. 14, 24. Later Augustine
describes the inner word as 'knowledge of knowledge, vision of vision' 'scientia de scientia, et
visio de visione' (Ibid., Book XV, sec. 21, 40).
Anselm's Augustinianism
Anselm of Canterbury followed Augustine in identifying God as Truth,
but he constructed a more explicit account of the relationship between
Truth and particular truths, one that rests on his general definition of truth
as 'tightness perceptible by the mind alone'.19 The standard of rightness for
particular things remains God's idea of the essence of things of that type,
while the standard of rightness in judgements is their correspondence to
what God knows or wills to be the case.20 In the first sense, someone is a
'true friend' or a 'true artist' when his nature corresponds to the divine idea
of a friend or an artist. Anselm explains the truth of propositions or
judgements in terms of the second sense of rightness. Here he relies on a
view of truth as the correspondence between a proposition and what is the
case, but he depicts this correspondence as a species of rightness by reason-
ing that if a proposition is true, it is doing what it ought to do, i.e. affirming
what is the case or denying what is not the case.21
Augustinianism in Grosseteste's interpretation of the Posterior Analytics
The first known commentator on the entire Posterior Analytics in the Latin
West, Robert Grosseteste, in effect endorsed Anselm's Augustinian treat-
ment of the nature of truth. He agrees that the truth of things is their
adequate conformity to their 'reasons' in God's mind, while the truth of
propositions is a function of the adequate conformity between what they
19. 'Possumus igitur, nisi fallor, defmire quia Veritas est rectitudo mente sola perceptibilis'
(Anselm 1946b, p. 191.19-20).
20. 'If there is no rectitude in those things which ought to have rectitude, except when they are as they
ought to be, and this alone amounts to their being right, it is clear that there is one sole rectitude
for all these things.'
'Si rectitudo non est in rebus illis quae debent rectitudincm, nisi cum sunt secundum quod debent,
et hoc solum est illis rectas esse: manifestum est earum omnium unam solam esse rectitudinem'
(Ibid., p. 199.7-9)·
'Thus there is truth in the being of all things that exist, because they are what they are in the
highest truth.'
'Est igitur Veritas in omnium quae sunt essentia, quia hoc sunt quod in summa veritate sunt'
(/f>irf.,p. 185.18-19).
21. 'Thus it is the same for it to be right and to be true, that is, to signify to be the case what is the
case ... It is similar, when an utterance signifies not to be the case what is not so.'
'Idem igitur est illi et rectam et veram esse, id est significare esse quod est... Similiter est, cum
enuntiatio significat non esse quod non est' (Ibid., p. 178.22-3 and 27). 'Proposition' is used in its
medieval sense here, in which it refers primarily to a spoken sentence. For discussions of the
medieval senses of this term, see Kretzmann 1970 and Nuchelmans 1973.
assert and what is the case. 22 He also follows Anselm's account of the unity
of truth in asserting that the truths of things and of propositions are simply
manifestations of the same exemplar, Truth. But Grosseteste adds a meta-
physical explanation of the participation of instances of truth in their
exemplar. He maintains the view that things are created and conserved
through emanations of divine light; according to this theory, the forms of
things which give them their truth are not entirely distinct from the divine
emanation, or Truth itself.23 Drawing an epistemic moral from his meta-
physics, Grosseteste infers that whether one understands something as a
particular or as a participant in its divine idea, he understands it by means of
indirect or direct divine illumination. Since divine light gives things the
forms through which they are intelligible, it always plays a role in knowl-
edge, regardless of the knower's awareness of it. 24
Grosseteste's Augustinian assumptions about truth and knowledge in-
fluence his interpretation of the Posterior Analytics. Although he generally
agrees with what he takes to be Aristotle's account of demonstrative
science, he places it within the framework of his metaphysical and epi-
stemological theory of divine illumination. His theocentric theory of truth
provides a cosmic context for understanding the requirement that de-
monstrative premisses be true. The commitment to divine illumination
also affects his treatment of the requirement that the premisses of scientific
syllogisms which are indemonstrable first principles or definitions must be
more certain than their conclusion. Aristotle indicates that experience and
22. ' . . . the truth of each thing is its conformity to its reason in the eternal word ...'
'... Veritas cuiuscunque est eius conformitas rationi suae in aeterno Verbo .. .'(Grosseteste 1912c,
p. I37-I-2).
' . . . the truth [of a declarative sentence] will be an adequate conformity between interior speech
and a thing ...'
' . . . erit Veritas [orationis enuntiativae] adaequatio sermonis interioris et rei...' (Ibid.,
p. 134.23-4).
23. 'And the species and perfection of all bodies is light... the lower bodies participate in the form of
the higher bodies..."
'Et species et perfectio corporum omnium est l u x . . . corpora inferiora participant formam
superiorum corporum ...' (Grosseteste 1912b, p. 56.36 and p. 57.9-10).
24. "... no truth is perceived except in the light of the highest truth . . . many of the impure of heart
see the highest truth [in conjunction with true things] and many of them do not realise that they
see it in any way . . . But the pure of heart and the perfectly purified mentally perceive the light
itself... Therefore, there is no one who knows any truth, who does not also know the highest
truth in some way, either knowingly or ignorantly.'
"... nulla conspicitur Veritas nisi in luce summae veritatis... immundi multi summam veritatem
vident [in conjunctione . . . rebus veris] et multi eorum nec percipiunt se videre earn aliquo
modo ... Mundicores vero et perfecte purgati ipsam lucem in se conspiciunt... Nemo est igitur,
qui verum aliquid novit, qui non aut scienter aut ignoranter etiam ipsam summam veritatem
aliquo modo novit' (Grosseteste 1912c, p. 138.3-21).
induction lead to an intuitive grasp of such premisses, but he hardly

considers the question of providing criteria to determine the correctness of
our apprehension of indemonstrable premisses.25 So to a reader who raises
this question, Aristotle's account of indemonstrable premisses appears
implicitly conditional: If one has understood a first principle or definition
correctly, one grasps something which is in some sense more certain than
its consequences. But Grosseteste's assumptions allow him to make an
explicit case for the possibility of certainty in demonstrative science. He
characterises a person's intuitive apprehension of an indemonstrable pre
miss as a direct or indirect irradiation of his mind by divine light, and insists
that no certainty is possible without such illumination. 26
Grosseteste is optimistic about the possibility of attaining a demonstra
tive science of nature. He realises that the stipulation that demonstrative
syllogisms identify the necessitating causes of phenomena creates a pre
sumption against the applicability of scientific demonstration to nature, on
the grounds that what is necessary is always the case, while many natural
phenomena occur frequently rather than always. Even the lunar eclipse -
Aristotle's favorite example - appears to be only an intermittent pheno
menon. But Grosseteste emphasises that one can have scientific knowledge
of such frequently occurring events, since whenever their necessitating
causes obtain, the effect occurs. 27 In his summary of the virtues of
Aristotle's theory of demonstrative science, he says that it enables us
25. Post. An. Book II, Ch. 19; Serene 1979, pp. 99-101.
26. 'Pure intellects receive direct irradiation from the divine light; generally human intellects are not
directly irradiated by the light of the divine ideas, but by the created light of their minds.'
'Et apud intellectum purum et separatum a phantasmatibus possibile est contemplari lucem
primam quae est causa prima . . . et intellectus humanus... multoticns recipit irradiationem a luce
creata quae est intelligentia' (Grosseteste 1514, I, 7, f. 8").
In any case, we know first principles by a spiritual light: ' . . . that by which something is an
immediately accepted first principle has the same nature as intellectual capacity, because this
nature is a spiritual light directly visible to the mind's eye, as was said above.'
' . . . eadem natura quae est virtus intellectiva est id secundum quod aliquid est principium et sine
medio acceptum. Quia ut superius dictum est lux spiritualis per sc visibilis a mentis aspectu est
haec natura1 (Ibid., II, 6, f. 40 Λ ).
27. 'For the eclipse simpliciter always exists in its causal reasons, although no particular eclipse always
exists in its causal reasons... whenever the moon falls into the earth's shadow it is eclipsed.'
'Eclipsis enim simpliciter semper est in rationibus suis causalibus nulla tamen eclipsis particularis
semper est in ratione sua causait... quotienscumque luna cadit in umbram terrae edipsatur' (Ibid.,
I, 8, f. 9"). A recent interpretation of this passage suggests that according to Grosseteste: "The
eclipse demonstration is just one step away from being verified at any time, est proximo habens
veritatem in omni bora, in this sense: it specifies the exact conditions under which an eclipse will
occur at any time, and the moment such conditions are verified, the demonstration itself is true'
(Wallace 1972, pp. 32-3). But the more plausible reading of 'probatio' for 'proximo' yields the
simpler point that the demonstration is always true, provided that its premisses are qualified by
the key term 'whenever'.
to judge easily when an explanation produces scientific knowledge.28

Lacking the concept of a divine guarantee, Aristotle had not made such a
strong chim for his theory.
Aquinas on the truth, necessity, and certainty of scientific premisses

Thomas Aquinas is better known for treating theology as a demonstrative
science than for contributing to the theory of science. But his consideration
of demonstrative science is interesting just because he seems so sympathetic
to the details and spirit of Aristotle's enterprise, as is clear from his
exposition of the requirements that demonstrative premisses be true,
necessary, and certain.
Augustinian accounts of truth are relevant primarily to the hereafter,
though they are not denied; the end of man is a vision of God which will
include acquaintance with the eternal exemplars, but even now the specu-
lative intellect can work toward its end, to attain truths which are expressed
in judgements.29 Truth, 'the conformity of the understanding with reality,
such that the understanding says that what is the case is so, and that what is
not is not', is a correspondence between a judgement and the reality which
is its object.30 Although it is not easy to say exactly what Aquinas means by
'reality' or 'thing' (res) in the definition of truth, it is clear that he does not
think the relevant extra-mental relatum for human knowledge of the
natural world is a divine idea. In his treatment of the requirement that
demonstrative premisses be necessary, Aquinas assumes that no propo-
sition which is ever false can be absolutely necessary.31 When he discusses
the demonstrability of frequently occurring phenomena, he follows
Grosseteste's justification for considering the lunar eclipse demonstrable:
whenever the causes of an eclipse occur, the effect occurs.32 But he refuses
28. 'And by this science which has been exhibited it is easy to recognise whether a proposed syllogism
is demonstrative.'
'Et hac scientia habita facile est cognoscere de syllogismo proposito an sit demonstrative'
(Grosseteste 1514, II, 5, f. 39*)- For discussions of Grosseteste's view of the utility of demonstra-
tive science, see Crombie 1953.
29. '...and therefore the intellect does not recognise truth except in composing and dividing
through its judgement.'
' . . . et ideo intellectus non cognoscit veritatem, nisi componendo vel dividendo per suum
judicium'(Comm. De Trin.W, t). See also In Periherm. I, lect. 3, n. 9; ST, 1,16, i-2,Quaest. disp. de
ver. I, 3. For a discussion of the nature ofjudgement, see Maritain 1959, pp. 84-100.
30. 'Cum enim Veritas intellectus sit adaequatio intellectus et rei, secundum quod intellectus dicit esse
quod est vel non esse quod non est...' (SCG, I, 59, n. 2).
31. In Periherm. I, lect. 15, n. 2; see also Comm. Post. an. I, lect., 42. For interpretations of Thomas'
conception of modality, seejalbert 1961 and Knuuttila 1975.
32. 'But there cannot be demonstration of particulars, as we have shown, but only of universals.
Hence it is clear that such [frequently occurring] things are always so, insofar as there is a
to extend the mantle of genuine demonstrability to frequent phenomena

whose causes can obtain without bringing about their usual effect.
Syllogisms which explain what we now call statistical phenomena, such as
the generation of trees from some but not all seeds under the same general
conditions, differ in kind from demonstrative syllogisms because they have
only qualified necessity, do not yield demonstrative knowledge of what is
true absolutely, and are less certain than fully demonstrative syllogisms. 33
In his characterisation of the certainty of demonstrative premisses, Aquinas
recapitulates Aristotle's views, holding that we naturally grasp first prin-
ciples and definitions through experience without needing a special learn-
ing process. 34
Aquinas on knowledge and opinion

When Aquinas speaks of science or demonstrative knowledge in the strict
sense, he follows the firm version of Aristotle's distinction between knowl-
edge and opinion. The objects of knowledge include only what can be
demonstrated by syllogisms meeting all the proper criteria; anything else is
an object of opinion. 3 5 Thus one has only opinion with respect to truths
which have not been demonstrated, including all contingent truths, and
even with respect to the conclusions of demonstrative syllogisms in the
demonstration of them. As is the case with the lunar eclipse, so it is with all other similar things...
For some things are not always so with respect to time, but are always so in relation to a cause.'
'De particularibus autem non potest esse demonstratio, ut ostensum est, sed solum de univcr-
salibus. Unde patet huiusmodi, secundum quod de eis est demonstratio, sunt semper. Et sicut est
dc defectu lunae, ita est dc omnibus aliis similibus , . . Quaedam enim non sunt semper secundum
tempus, sunt autem semper per comparationem ad causam' (Comm. Post. an. I, lect. 16, n. 8).
33. 'Yet such demonstrations do not make one know that what is concluded is true simpliciter, but
only in a qualified sense, namely, that it is true in most cases... Hence sciences of this kind fall
short of sciences which concern absolutely necessary things with respect to the certitude of the
demonstration.'
'Huiusmodi tamen demonstrationes non fàciunt simpliciter scire verum esse quod conduditur,
sed secundum quid, scilicet quod sit verum ut in pluribus... Unde huiusmodi scientiae deficiunt a
scientiis, quae sunt de neccssariis absolute, quantum ad certitudinem demonstrationis' (Comm.
Post. an. II, lect. 12, n. 5).
34. 'Hence every aspect of the speculative sciences is reduced to some first principle, which someone
should not necessarily have to learn or find out... but should recognise naturally.'
'Unde omnis consideratio scientiarum speculativarum reducitur in aliqua prima, quae quidam
homo non habet necesse addiscere aut invenire . . . sed eorum notitiam naturaliter habet' (Comm.
De Trin. VI, 4). Thomas agrees with Aristotle that one cannot err in the apprehension of a first
principle; see In Metaph. IV, lect. 6.
35. For the view that all propositions not seen as necessarily connected to first principles yield only
opinion, see Quaest. disp. de ver. II, ι; Comm. De Trin. VI, ι; Comm. Post. an. I, lect. 44; and ST, II,
II, 116. In particular, Aquinas stresses that we have only opinion and neither understanding nor
knowledge with respect to contingents: 'Sic enim se habet [opinio] circa contingentia, sicut
intellectus et scientia circa necessaria.' See Byrne 1968, p. 184.
absence of an apprehension of their premisses. Demonstrated truths, then,

are a small subset of truths. And since Aquinas' definition of certitude
parallels his definition of truth, it follows that one can be certain of many
more truths than one knows by demonstration. 3 6 When Aquinas considers
the work of the natural and moral sciences, however, he relaxes the
distinction between knowledge and opinion, using the terms 'scientia' and
'demonstratio' broadly to refer to these areas of inquiry and their arguments.
His concession that the sciences which deal with singulars and so fall short
of the necessity found in geometry can nevertheless produce some certi
tude is as important as his caveat that they cannot produce absolute
knowledge or certitude. 3 7 Aquinas does not expect investigators to con
struct a stricdy demonstrative science of the physical world, but he is
sanguine about the development of sciences in the broad sense, although
they can only approximate the ideal of producing scientific knowledge
through fully demonstrative syllogisms. 3 8
Aquinas on the importance of final causation

This is a natural attitude for Aquinas to take, since it squares with his special
interest in the explanatory power of final causes, whose metaphysical
priority he thought Aristotle had overlooked. 3 9 Aquinas extends the
36. 'This, indeed, is certitude of cognition: when the cognition does not deviate at all from what was
found in the thing, but rather considers the thing in this way: just as it is.'
'Cognitionis quidem certitudo est, quando cognitio non déclinât in aliquo ab eo quod in re
invenitur, sed hoc modo existimat de re sicut est' (Quaest. disp. de ver. VI, 3). Byrne 1968 notes:
"... certainty as a characteristic of human science differs from opinion or probable knowledge not
in kind but only in degree' (p. 269).
37. 'And so also to the extent to which any science has more to do with singulars, such
a s . . . medicine, alchemy, and ethics, they can have less certitude because of the great number of
singulars that must be considered in such sciences - if any of them is omitted, error will follow -
and because of the variability of singulars.'
'Et ideo etiam quanto aliqua scientia magis appropinquat ad singularia, sicut... medicina,
alchimia et moralis, minus possunt habere de certitudine propter multitudinem eorum, quae
consideranda sunt in talibus scientiis, quorum quodlibet si omittatur, sequetur error, et propter
eorum variabilitatem' (Comm. De Trin. VI, 1). See also In I Eth., lect. 3, n. 36 and In Metaph. XI.
lect. 7, n. 2249.
38. 'But the necessary . . . is one way in natural things which are true for the most part and fail to be
true in a few cases; and it is another way in disciplines such as mathematics which are always true.'
'Necessarium autem . . . aliter est in naturalibus, quae sunt vera ut frequenter, et deficiunt in
minori parte; et aliter in disciplinis, idest in mathematiris, quae sunt semper vera' (Comm. Post. an.
I, lect. 42, n. 3).
39. ' . . . but the end is prior to the efficient cause insofar as it is bringing something about in a
substance and its complement [i.e. its accidents or the new substance it is becoming] since the
action of the efficient cause is completed only on account of the end.'
' . . . sed finis est prior efficiente... inquantum est efficiens in substantia et complemento, cum
actio efficientis non compleatur nisi per finem' (De principiis naturae IV, 25).
notion of final causality in various ways, including applying it to the

problem of providing a necessary causal explanation for statistical pheno
mena by describing the effect as afinalcause. If, for example, the flourish
ing of a tree is the final cause of its seed, then one can argue that whenever
the final cause occurs, its correlative efficient cause must have occurred;
thus if there is aflourishingtree, there was a seed.40 As Aquinas' successors
hastened to point out, however, such arguments cannot yield plausible
scientific explanations.41 Even if such explanations 'on the supposition that
the effect has occurred' were plausible, they still would not satisfy all the
requirements for genuine demonstration. But his attempt shows that while
in principle Aquinas remains a strict adherent to 'the Aristotelian ideal of
demonstrative science', in practice he welcomes various sorts of scientific
explanations which do not satisfy its requirements.42
Duns Scotus' modifications of the concept of demonstrative science

In the writings of Duns Scotus the concept of demonstrative science
undergoes a substantial change which may have had roots in the Con
demnation of 1277. Among the propositions condemned by bishop
Tempier wefindthe assertion that: 'One should not hold anything unless it
is self-evident or can be manifested from self-evident principles.' 43 The
view that the canons of demonstrative science cannot provide the sole
40. '... we will adduce as proof that if it happens that an ultimate end occurs, that through which the
end was attained must precede.'
*... argumentabimur quod si fieri contingit finem ultimum, oportet praecedere ea per quae
pervenitur ad finem' (Comm. Post. an. 11, lect. 9, n. 11). Aquinas was not the only philosopher
interested in enhancing the status of explanations from final causes. According to Longeway
1977, ΡΡ· 320-55, Simon of Faversham held that a demonstratio potissima comprises a major
premiss which identifies an essential cause of a subject s of type T, a minor which identifies the
final cause/of Τ things, and a conclusion which shows that f necessarily inheres in any token of
type T. The fact that the property/ascribed in the minor and in the conclusion is extensionally the
same property apparently does not detract, in Simon's eyes, from the interest of establishing its
inherence as a reasoned fact.
41. For example, Scotus writes, 'For the end is not the cause of the efficient cause.'
'Nam illud quod est finis non est causa eius quod est efficiens' (Scotus 1966, pp. 28-9).
Ockham writes, 'But to say that a thing outside the mind is mutable according to one considera
tion of mine and that it is immutable according to another is simply false and asinine.'
'Sed quod ilia res quae est extra propter unam considerationem meam sit mutabiiis et propter
aliam considerationem meam sit immutabilis, est simpliciter falsum et asinine dictum' (Ockham
1957b, P· 14)-1" Ockham 1974a, III, II, 17, he explains that a definition taken from afinalcause
cannot serve as a middle term because it is a nominal, not a real definition.
42. For Boethius of Dacia's different interpretation of the relationship between demonstration and
the objects of natural science, see Pinborg 1975a.
43. 'Quod nichil est credendum, nisi per se notuiii, vel ex per se notis possit declari' (Denifle and
Châtelain 1889-97.1, P- 54s). For a discussion of the condemnation, including detailed treatment
of both primary and secondary sources, see Wippel 1977. See esp. Hissette 1977.
criteria for knowledge found redoubtable advocates in Scotus and later in

Ockham. Although Scotus agrees that demonstrative syllogisms yield
knowledge of reasoned facts, other aspects of his thinking undermine the
traditional status of demonstration. These include renewed attention to
God's relation to knowledge and a revived interest in the Augustinian
paradigm of knowledge as vision.
Scotus on truth
For Scotus, truth is a 'likeness through imitation, as a copy is to a pattern'. 44
Arguing against Henry of Ghent, he proposes three interpretations of this
correspondence relation in support of the view that there is genuine
knowledge of truths apart from special instances of divine illumination.45
Truth can be understood as the correspondence between terms in an
indubitable proposition such as: 'The whole is equal to the sum of its parts.'
Or it can be taken as the relationship between a particular idea and the
particular individual or state of affairs which is its object. In a third sense,
truth can be the correspondence between a general idea and the essence of
objects ofthat genus. Scotus' criterion for successful apprehension of truths
suggests that he espouses at least the first two interpretations of the cor
respondence relation. He uses the evidentness of cognition as a criterion for
truth and certainty, even though it is not part of his definition of truth.
'Evident cognition' is a technical term which designates both the possible
objects and the quality of the cognition; its possible objects are the existence
of particular individuals or states of affairs a person is perceiving and
propositions whose truth is known from their terms alone. 46 Because
evident cognitions are clear and indubitable, they serve as an epistemic
ideal and a criterion for truth.
44- '•. - sirnilitudo per imitationem sicut est ideati ad ideam' (Quaest. quodl. Q. 13, n. 12). For a
discussion of this point, see Vier 1951, pp. 31—8.
45. 'From all this, it is clear why a special illumination is not necessary in order to see in the eternal
rules, for Augustine assumes that we sec in them only such "truths" as are necessary by the force of
their terms.'
'Et ex isto apparet qualiter non est necessaria specialis illustratio ad videndum in regulis aeternis,
quia Augustinus non ponit in eis videri nisi "vera" quae sunt necessaria ex vi terminorum' (Scotus
1954, dist. 3, 1, Q. 4; p. 164.13-15).
46. Scotus' ideal of certain and evident cognition echoes the Augustinian paradigm of knowledge as
direct vision of its objects: 'Perfect knowledge is certain and evident cognition Furthermore,
vision of the extremes of a contingent truth and of their union necessarily causes evident certitude
concerning such evident truth.'
' . . . in scientia illud perfecrionis est, quod sit cognitio certa et evidens... Visio autem ex-
tremorum veritatis contingentis et unionis eorum necessario causât evidentem certitudinem de
tali veritate evidente'(Scotus 1950-, 1,4, Q. 1-2; p. 144.10-11 and p. 145.10-12). The same ideal
of certain and evident cognition is satisfied in grasping a self-evident proposition, i.e., one which
Scotus on evidentness and necessity

Although Scotus' view of truth does not decisively distinguish his theory of
science from that of earlier authors, his interest in evidentness does. Like
Aristotle and Aquinas, he believes that first principles can be evident on the
basis of experience,47 but he insists that the apprehension of the correct
premisses of a scientific syllogism does not suffice for scientific knowledge.
We also need evident cognition of the phenomenon being explained by
the demonstration.48 To follow the difficulties this apparently innocuous
point creates in Scotus' treatment of demonstrative science, it is helpful to
attend to his innovative view of necessity. The most significant divergence
from the Aristotelian theory of science in Scotus is his denial of necessity in
the natural world. He agrees with the traditional requirement that dem-
onstrations of reasoned facts have some sort of necessity, but for theological
reasons denies that there are any unconditionally necessary truths about the
natural world. In the first place, the laws of nature cannot be absolutely
derives its evident truth from knowledge of its terms and contains within itself the sole source of
its certitude: '.. · ilia propositio est per se nota . . . quae ex terminis cognitis habet veritatem
evidentem et quae non habet certitudincm nisi ex aliquo in se' (Scotus 1960a, I, dist. 2, Q. 2;
p. 117.17—18). Scotus insists that if a proposition can be demonstrated to anyone, it is not
self-evident: ' . . . i l i a non est per se nota ex notitia terminorum quae potest esse conclusio
démonstration«' (Ibid., p. 115.5-8).
47. The senses merely serve as an occasion for grasping the evidence of such a relationship: Ί reply
that with respect to knowledge [of principles], the intellect has the senses not as a cause but as an
occasion, because it can have knowledge of simples [i.e., terms] only from the senses.'
'Respondeo quantum ad istam notitiam [principiorum], quod intellectus non habet sensus pro
causa, sed tantum pro occasione, quia intellectus non potest habere notitiam simplicium nisi
acceptant a sensibus' (Scotus 1954, dist. 3,1, Q 4; p. 140.18-20). Scotus' example of a self-evident
truth here is this: ' . . . Every whole is greater than its part - the intellect by its o w n power and by
the force of those terms should assent without doubt to this complex.' ' . . . omne totum est maius
sua partee, intellectus virtute sui et istorum terminorum assentiet indubitanter isti complexioni'
(Ibid., p. 141.4-6).
48. ' . . . unless we first have a simple idea of something in itself, nothing about it is demonstrated
scientifically.'
' . . . quod quia nihil scientifice concluditur de aliquo, nisi in se simpliciter praeconcepto' (Scotus
1966, IV, p. 111). In the Quaest. quodl., Q. 7,2, Scotus says that the objects of scientific knowledge
are concepts and definitions which do not always require intuitive cognition of their extra-mental
referents. Here he is defending the view that the scientific status of theology is not contingent on
the theologian's intuitive cognition of God; it requires only that he have evident cognition of the
concept of God. But the best understanding of natural phenomena requires evident cognition of
an actual effect together with demonstrative knowledge of its causes: 'When you say that
cognition through a cause is more perfect, I say that simple cognition o f the effect, which it causes
itself, must be included here. Cognition of a complex is caused by cognition of the cause and the
effect together; and it is true that [a cognition caused] by the first cause and the second together is
more perfect than that [caused] by the second alone.'
'Cum dicis: cognitio per causam est perfectior, dico quod ibi includitur cognitio effectus simplex
causata ab ipso. Cognitio complexi causatur simul a cognitionibus causae et causati; et verum est
quod a causa prima et secunda simul est aliquid perfectius quam a secunda sola' (foc. (it.),
necessary because the creator chose them by a free act of his will. 49
Secondly, God could at any moment will to abrogate the laws he originally
ordained. Since demonstrative premisses were to be necessary in and of
themselves, this theologically based scepticism appears to undermine the
possibility of a demonstrative science of nature as Aristotle conceived it.
Scotus on the purpose of demonstrative science

But rather than rejecting demonstrative science, Scotus reinterprets its
purpose; its new object is to discover not what is necessary in nature but
what is possible or compossible. 50 Thus science primarily concerns not
phenomena as they actually stand, but propositions about what can occur;
its truths are now de dicto, not de re assertions. In this new view, science deals
primarily with relations among ideas, but these ideas are grounded in
reality by the assumption that indemonstrable premisses are found in
experience and by the requirement that scientific knowledge of natural
phenomena include evident cognition of the effect explained.
Unfortunately, this new philosophy of science creates some confusion
about the explanatory power of demonstrations in natural science. If
evidentness is the criterion for truth, and scientific knowledge requires
evident cognition of an effect, what additional insight does a demon-
stration contribute? It seems that an evidently cognised particular fact must
always be more evident than its premisses. Aristotelian science was de-
signed to explain a particular phenomenon by showing it to be a token of a
type whose essential nature always necessitates that exact kind of pheno-
menon under the specified conditions. Scotist science is designed to elicit
answers to two different questions: what laws of nature make this type of
phenomenon possible? What particular causes brought about this par-
ticular instance of its occurrence? There is some reason to believe that
Scotus himself did not always fully appreciate the distinction between these
two questions, as the following examples suggest.
49. 'Proof: Whatever God causes immediately he causes contingently ... therefore, [he causes]
everything [in this way], because the contingent does not naturally precede the necessary, nor
does the necessary depend on the contingent.'
'Probatur, quid quod immediate causât, contingenter causât... igitur et quidlibet, quia con-
tingens non praecedit naturaliter necessarium, nec necessarium dependet a contingente' (Scotus
1966, IV, p. 91).
50. While this conclusion appears more clearly in Ockham, Scotus recognises in the Quaest. quodl.,
Q. 7,2, that there are necessary truths about the possibilities of contingent things; see Scotus 1895,
p. 292.
Scotus on the regularity of nature

Scotus believes that nature functions in a regular manner, deeming it
evident that: 'Whatever occurs in many instances by means of a cause that
is not free is the natural effect ofthat cause.' 51 Experience enables us to
identify 'unfree' causes, i.e. causes which do not directly express an agent's
will, and to associate these causes with their frequent effects.52 It is natural
to expect that having introduced this principle Scotus will apply it to the
problem of explaining statistical phenomena; but he encounters some
difficulty in doing so when he takes up the traditional example of the lunar
eclipse. The problem is that he assumes the explanandum is a particular
eclipse which is now being observed; since what is intermittent is not a
given instance of an eclipse but the occurrence of that type of celestial
event, there is nothing intermittent left to explain by means of the prin
ciple of the regularity of nature. 53
The second case Scotus considers involves the proposition that a certain
herb or particular sprig of an herb is hot to the taste. 54 The standard
interpretation of Scotus' treatment of the example runs as follows. On the
basis of the principle of the regularity of nature, together with our induc
tive evidence that sprigs of, say, coriander are hot, we can conclude with
the lowest grade of scientific certainty that the next sprig tasted will be hot,
or at least that it is apt to be hot. According to this interpretation, Scotus is
pointing out that even the principle of the regularity of nature cannot fully
51. 'Quidquid evenit ut in pluribus ab aliqua causa non libera, est effcctus naturalis illius causae'
(Scotus 1954, dist. 3, 1, Q. 4, p. 142.1-2).
52. Ά natural cause brings about its effect to the best of its power whenever it is not impeded.'
'Causa naturalis agit ad cffectum suum secundum ultimum potentiae suae quando non est
impedita' (Ibid., dist. 3.1, QQ. 1-2, p. 52.1 -3). It is interesting that Leonardo Da Vinci makes a
similar point: Ά principle being given, it is necessary that its consequences flow from it, if it has
not been hindered ...' (Frammenti lelterari efilosofici, ed. Salmi 1925, quoted in Blake. Ducassc.
and Madden 1966, p. 18).
53. 'And if someone will have discoveted by way of division that the earth is such a body interposed
between sun and moon, our conclusion will no longer be known merely by experience as was the
case before we discovered this principle. It will now be known most certainly by a demonstration
of the reasoned fact, for it is known through its cause.'
'Et si inventum fuerit per divisionem quod terra tale est corpus, intcrpositum inter solem et
lunam, scietur certissime demonstratione propter quid (quia per causam) et non tantum per
cxperientiam, sicut sciebatur ista conclusio ante inventionem principii' (Scotus 1954, dist. 3, 1,
Q. 4, P· 143 12-16).
54. Vier 1951 (p. 147) and Wolter 1962 (p. 111) suppose Scotus is discussing a generalisation about a
type of herb; but a literal translation of the relevanr passage suggests that he is interested in a
particular instance, e.g., a particular sprig of coriander: 'suppose that this herb of such a species is
hot', 'puta quod haec herba talis speciei est calida' (Scotus 1954, p. 143.20-144.1).
overcome the fallibility of induction, since the next sprig may not be hot. 5 5
But another reading of the example yields a different interpretation which
parallels the treatment of the lunar eclipse. If the assumption of the example
is that Scotus has tasted a particular sprig of coriander, and so has an evident
cognition of its hotness, reference to the principle of the regularity of
nature can add nothing to the evidentness of his cognition that it is h o t . 5 6
The failure of these examples to provide a clear illustration of the utility
of the principle of the regularity of nature which Scotus introduced with
some emphasis suggests that he may not have fully understood that his
theory entails a radical revision of Aristotelian science. Addressing the
challenge he raises of explaining a particular event in its particularity would
involve abandoning the demonstrative format as it was traditionally con
ceived. Although the laws of nature have no absolute necessity, Scotus
holds that what occurs at a given moment is uniquely determined by a
concatenation of causes, including what God wills t h e n . 5 7 In principle a
particular event, for example that this seed germinated at this moment,
could be explained if one could specify in sufficient detail the set of relevant
conditions. But if, per impossibile, we could give such an explanation, it
would involve so many additional causes, conditions, and principles that it
would hardly resemble the traditionally simple demonstrative syllogism. It
is no wonder that Scotus seems uncertain about what demonstrative
55· Vier 1951 supports this interpretation in maintaining that: 'The only infallible knowledge the
inductive process conveys is that the herb is capable of producing warmth. It is for this reason that
the knowledge resulting from the application o f the inductive process is termed by Scotus the
lowest degree of scientific knowledge' (p. 147)- This reading supposes that the principle of the
regularity of nature, together with appropriate experiences o f coriander, tells us only that a
particular sprig of coriander has 'an aptitude for warmth'; but it cannot tell us with certainty that a
particular sprig will be hot. In this view, the fallibility and uncertainty of the supposition that an
untastcd sprig of coriander will be hot is removed to some extent by adverting to the regularity of
nature. Vier's point requires reading 'licet tunc certitudo et infallibilitas removeantur per istam
propositioncm' rather than 'licet tunc incertitudo et fallibilitas...'. The manuscripts provide no
decisive witness for either reading; a number o f them omit at least patt o f this vexed point; see
Scotus 1954, p. 144η.
56. O n this interpretation, the text reads: 'although the certitude and infallibility [of a present evident
cognition] may be put aside through that proposition [that coriander is only usually h o t ] . . . ' .
Thus the evidentness of a particular present experience of a sprig's hotness may be ignored in
favour of a less evident and certain explanation of the cause of the general phenomenon that
coriander is hot. This reading squares with Scotus* comment that, 'individuals are only im
perfectly understood in a universal reason, since they are not understood according to the full
positive entity in them, as I have shown in the question concerning individuation'.
'Individua in ratione universalis imperfecte intelliguntur, quia non secundum quidlibet entitatis
positivae in eis, sicut in quaestione de individuatione oscendi' (Scotus 1966, IV, p. 107).
57. He writes in Scotus 1966, III: ' . . . for whatever can be produced has some cause [to which it is]
essentially ordered'.
' . . . nam omne effectibile habet aliquam causam essentialiter ordinatam' (p. 61).
Demonstrative science SU
syllogisms can contribute to the understanding of particular events in their

particularity.58
Ockham on propositions as the objects of science

William Ockham endorsed many of the tenets of Aristotle's theory of
demonstrative science while using the notions of evidentness and divine
power to erode its epistemic importance. He holds that: Ά demonstrative
syllogism is one in which primary cognition of a conclusion can be
obtained from propositions that are necessary and evidently known.' 5 9
But he maintains that the resulting sciences are collections of true pro
positions, and not necessarily a mirror of the inner constitution of nature. 6 0
The premisses and objects of demonstrative science must be propositions
rather than natural phenomena because God is the only necessary being. 61
These necessary premisses may be derived from matters of fact by a two-
step process; first we learn contingent truths about what is the case by
experience, then we transform them into correlative necessary truths about
what must be possible.62 Even though Ockham insists that properly
speaking demonstrative science concerns ideas and propositions, 'mental
contents which are common', he confidently assumes that it reflects the
causes of these ideas in the physical world. 63 His confidence stems from the
58. Thus Scotus' view should not be assimilated to Humean scepticism about induction. He is not
worried about induction per se, but about how to reconcile his belief in the explanatory power of
demonstration with his commitment to using the evidence of cognition as a criterion for truth
and certainty.
59. 'Syllogismus demonstrativus est ille, in quo ex propositionibus necessariis evidenter notis potest
adquiri prima notitia conclusionis' (Ockham 1974a, III, I, I; p. 359.14—16).
60. 'For the object of science is the whole known proposition.'
'Nam obiectum scientiae est tota propositio nota' (Ockham 1957, p. 9).
61. 'In one way something is called necessary... because it can begin or cease being through no
power; and thus God alone is... necessary In another sense, a proposition which cannot be
false is called necessary And in this sense demonstration is of necessities... that is, of
propositions which cannot be false, but always true.'
'Uno modo dicitur aliquid necessarium . . . quia per nullam potentiam potest inciperc vel desinere
esse; et sic solus Deus est... necessarius. . . . Aliter dicitur necessarium ... propositio quae non
potest esse falsa Et isto modo demonstratio est necessariorum . . . hoc est propositionum, quae
non possunt esse falsae sed tantum verae' (Ockham 1974a, III, II, 5; p. 512.27-32).
62. '... propositions in a demonstration are necessary which are negative or hypothetical or which
concern the possible ..."
'... propositiones... in demonstratione sunt necessariae quae... sunt negativae vel hypotheticae
vel dc possibili...' (Ibid., p. 513.65—514.2).
63. 'But, properly speaking, natural science is about mental intentions common to such [corrup
tible and mutable] things and suppositing precisely for such things in many propositions...
Nevertheless, metaphorically and improperly speaking, natural science is said to be about
corruptible and mutable things, because it is about those terms which supposit for such things'.
'Sed, proprie loquendo, scientia naturalis est de intentionibus animae communibus talibus rebus
[rebus corruptibilibus et mobilibus] et supponentibus praecise pro talibus rebus in multis pro-
belief that intuitive, evident cognition of particulars, which provides the

basis for knowledge, is by definition true. 6 4
Ockham's devaluation of demonstrative science
But his conception of evidentness dramatically devalues demonstrative

science. If the conclusions of demonstrative syllogisms are known only
indirectly rather than directly and evidently, they must be dubitable in
some sense. 6 5 Hence, he contends, scientific knowledge is not epistemically
decisive; a demonstrative syllogism is in principle no more certain than a
dialectical one which rests on probable premisses - those which are con
sidered true by everyone, or by most people, or by the wisest - since such a
syllogism may 'produce firm faith without any doubt'. 6 6 Further challeng
ing the claims of demonstration to epistemic priority, Ockham argues that
cognition acquired by demonstration and that acquired from experience
do not differ in kind. 6 7 But even while denying demonstrative science its
traditional pre-eminence, Ockham presents nominalist interpretations of
traditional features of the demonstrative syllogism, such as the necessity of
the premisses and their priority with respect to the conclusions derived. 6 8
He raises and discusses various questions (stimulated by Scotus' work)
concerning the relationship between experience and demonstration e.g.:
'Can there be demonstration propter quid of a currently observed pheno-
positionibus... Tamen, metaphorice et improprie loquendo, dicitur scientia naturalis esse de

corruptibilibus et de mobilibus, quia est de illis terminis qui pro talibus supponunt' (Ockham
1957, p. 11).
64. For exposition and criticism of this tenet, see Adams 1970, Scott 1971, and Boler 1976.
65. ' . . . every conclusion of a [full-fledged] demonstration is dubitable, thus... it is not self-evident',
' . . . omnis conclusio demonstrationis [potissimae] est dubitabilis, ita . . . non est per se nota'.
(Ockham 1974a, III, II, 9; p. 521.7-8). For a discussion of this point, see Webering 1953, p. 76; for
a related point in Scotus, see n. 46 above.
66. Ά topical [or dialectical] syllogism is not deficient in matter or in form but often even produces
firm belief without any doubt.'
' . . . Syllogismus topicus nec peccat in materia nec in forma ... sed etiam frequenter facit firmam
fidem, sine omni dubitatione' (Ockham 1974a, III, I, I; p. 360.35—6 and 40—1).
67. 'And if it should be asked whether cognition of some conclusion acquired through experience
and that acquired through demonstration are of the same species: and similarly whether cognition
acquired through diverse premisses is of the same species, it can probably be said that if such a
cognition should be precisely cognition of the conclusion and nothing else, it is not inappropriate
to posit that such a cognition is of the exact same species.'
'Et si quaeratur an notitia accepta per experientiam alicuius conclusionis et notitia eiusdem accepta
per demonstrationem sint eiusdem speciei; et similiter, notitia accepta per diversas praemissas sit
eiusdem speciei, potest probabiliter dici quod si talis notitia praecise sit notitia conclusionis et nihil
aliud, non est inconveniens ponere quod talis notitia sit eiusdem speciei specialissimae." (Ibid., Ill,
II, it; p. 524.13-18).
68. See esp. Ockham 1974a, III, II, chapters 3, 5, 11, and 15.
menon?' 6 9 'Can current experience serve in lieu of a definition, principle,

or term?' 7 0
Nicholas of Autrecourt
Some philosophers who extended the nominalist programme thought that
Ockham had not gone far enough in applying his razor to demonstrative
science.71 For example, Nicholas of Autrecourt thought that Ockham had
not interpreted the concept of evidentness strictly enough, since only
immediate experience and the law of non-contradiction are truly evident
and certain. 72 The vast majority of true propositions are only probable
because theoretically they could be falsified by God's exercise of his
absolute power. 73 Since non-mathematical demonstrative premisses fall
into this class, it is impossible for them to produce absolutely certain
knowledge of natural phenomena.
föhn Buridan
In response to Autrecourt's strict interpretation of evidentness, John
Buridan took up the defence of natural science. His strategy was to concede
that its principles and conclusions are not absolutely evident in Autre-
court's sense, but nevertheless to insist that they are sound if they have been
confirmed in many instances and have not been falsified.74 It has been
suggested that Buridan 's methodological stricture that a single exception in
nature falsifies an explanatory principle enables him to maintain - without
entering into complex metaphysical controversies — that natural phenom-
ena as well as propositions are the immediate objects of science. 75
69. Ibid., Ill, II. 23.

70. Ibid., Ill, II, 29.
71. For a detailed exposition and discussion of these doctrines in Ockham, see Adams 1977.
72. '... I am evidently certain concerning the objects of the five senses and concerning my actions.
... Every certitude had by us is reduced to this principle [that contradictories cannot be simul-
taneously true].' '... sum certus evidenter de obiectis quinquc sensuum et de actibus mcis.
... Omnis certitudo a nobis habita resolvitur in istud principium [contradictoria non possunt
simul vera]' (Nicholas of Autrecourt 1908, p. 6.15-6; p. 7.5-6, and p. 6.33).
73. Nicholas rejects the strategy of defending the evidentness of scientific inferences by adding the
proviso that God is not engaging in miraculous intervention, on the grounds that we cannot be
certain if or when the proviso is satisfied (Ibid., pp. 2-3 and p. 13.26-30).
74. 'It is therefore concluded ... that certain people argue very badly wanting to undermine the
natural sciences,... because... in possible cases they can be falsified supernaturally, because strict
evidentness is not required in such sciences...'
'Ideo conclusum est ... quod aliqui valde male dicunt, volentes interimere scientias naturales
... eo quod ... possunt falsificari per casus supernaturaliter possibiles, quia non requiritur ad tales
scientias evidentia simpliciter ...' (Buridan 1588 (actually 151S), Book II, Q. 1, f. 9rb).
75. Moody 1970, p. 605.
Buridan does make the traditional assumption that scientists can pursue
their investigation without giving an account of the metaphysical con-
stituents of what they study; 76 and he evidently believes, correctly, that
scientific explanations which lack the absolute certitude or necessity of
demonstration can be well enough established to be interesting and en-
lightening. But these assumptions need not rest on an innovative falsifi-
cationist methodology; indeed, it would be misleading to consider Buri-
dan a strict falsificationist. He admits, for example, that miracles do not
negate scientific truths, which deal only with the common course of
nature. 77 He adds a more significant qualification in response to the
observation that even in the common course of nature, expected effects are
often impeded by other natural phenomena: explanations in natural
science implicitly assume the proviso that no impediments occur. 78 Thus at
least some sorts of exceptions will not falsify explanatory principles. In the
light of these qualifications, it is not surprising that Buridan considers
demonstrative science only a small subset of science. Only propositions or
phenomena which happen all or almost all the time can be objects of
demonstrative, as opposed to natural, science; most natural phenomena are
only objects of natural science. Thus for Buridan the phenomenon of the
lunar eclipse cannot be an object of demonstrative science, nor can the
dispositional property of the heavens to have eclipses from time to time. 7 9
76. ' . . . for no science besides metaphysics has to consider the quiddity of a thing simpliciter. . . . the
natural scientist does not have to know simpliciter what a man is or what an ass is; he should
describe such things by some change or by some operations.'
'... nulla enim scientia praeter metaphysicam habet considerare de quiditate rei simpliciter . . .
physicus non habet scire simpliciter quid est homo vel quid est asinus: licet possit talia describere
per aliquos motus vel per aliquas operationes' (Buridan 1588, Book I, Q. 1, f. 3vb).
77. ' . . . there is firmness of truth [e.g. that "fire is hot") on the supposition of the common course of
nature ... even though God could thus make it cold, and so that [principle] "every fire is hot" is
falsified'.
'... est firmitas veritatis [quod "ignis est calidus"] ex suppositione communis cursus naturae . . .
non obstante quod deus posset sic facere ignem frigidum et sic falsificatur ista "omnis ignis est
calidus" ' (Ibid., Book I, Q. 1, f. 8vb).
78. 'Although we naturally should assent to principles ofthis sort ["every fire is hot"] nevertheless this
[proviso] must be understood, "if an impediment does not occur".'
'Unde quam vis naturaliter assentiamus huiusmodi principiis ["omnis ignis est calidus"] hoc tamen
est intelligendum "si non occurat impedimentum" ' (Ibid., Book II, Q. 2, f. 9vb).
79. *... only in the first way is science about propositions and not about all things, but only about true
and necessary things or at least those which are true in most cases... In a second way, science is
very much about all things of the world, but not as though (it was only) about demonstrable
conclusions."
'... primo modo solum est scientia de propositionibus et non de omnibus: sed solum de veris et
necessariis vel saltern de Ulis que (sic) ut in pluribus sunt vere . . . Secundo modo est bene scientia
de omnibus rebus mundi . . . sed non tanquam de conclusionis demonstrabilibus' (Ibid., Book VI,
Q. 3. f. 34"-b)
Buridan seems willing in practice to relegate demonstrative science to a

peripheral status, and he devotes himself to defending the reliability of
natural science in a number of commonsensical anti-sceptical arguments.80
Conclusion
The revival of explicitly realist interpretations of demonstrative science
and the elaboration of its methodology by Aristotelians of the Renaissance
lie beyond the scope of this survey.81 But it is clear that even in the
medieval portion of its history, the so-called 'Aristotelian ideal of dem-
onstrative science', like most long-lived ideals, underwent numerous
transformations, many of them theologically motivated. What accounts
for the persistent interest and influence of this theory of demonstrative
science? On the basis of our survey, it is tempting to begin a response to this
question by speculating that the malleability of the notion of demonstra-
tive science in the hands of diverse philosophers is part of the answer.
80. Ibid., Book I, Q. 1, f. S1*. See also Scott 1965.

81. For a view of the extensive Renaissance discussion of methodological issues, see Gilbert i960.
Traditional versions of the theory of demonstrative science survived well past the fourteenth
century; indeed, in many texts even Galileo seems to espouse such a strict version of the theory
that interpreters have been hard-pressed to identify the novelties one might expect. For recent
discussions, see Butts and Pitt 1978.
VII
NATURAL PHILOSOPHY
25
THE I N T E R P R E T A T I O N OF
ARISTOTLE'S PHYSICS AND THE
SCIENCE OF M O T I O N
Natural philosophy,firstphilosophy, and moral philosophy

When the 'new' Aristotelian books of philosophy were incorporated into
the curriculum of the medieval Faculty of Arts by 1252, they were simply
added as 'the three philosophies' to an existing curriculum of the seven
liberal arts, a course requiring up to eight years before one became a
Regent Master. The 'new logic' (logica nova), namely the two Analytics,
Topics, and Elenchi, had merely expanded the old study of logic, which had
even assimilated the logica modernorum without substantially changing the
curriculum. But the addition of the hitherto proscribed (1210-01. 1237)
libri naturales and Metaphysics, together with the Nicomachean Ethics trans-
lated in full (1245-7) by Robert Grosseteste, expanded the curriculum
substantially to include three new 'sciences': natural philosophy, first
philosophy (or metaphysics), and moral philosophy. Thus during the
second half of the thirteenth century was inaugurated what might be called
the Faculty of Arts and Sciences.
The assimilation of the new learning

It was a period that saw an unprecedented assimilation of 'the new learn-
ing' not only in the Faculty of Arts, but more especially in the Faculty of
Theology. However, the most notable assimilation and syntheses of the
new learning both in philosophy and in theology were accomplished by
theologians who had already passed through the university system and
embarked on their own re-thinking of Christian truths 'new and old'. This
was particularly true of such leading scholars as Robert Grosseteste, Albert
the Great, Thomas Aquinas, Roger Bacon, and later Thomas Brad-
wardine. In most cases their writings (philosophical and theological) are
extra-curricular in the sense that they were not delivered in the lecture halls
or intended for the classroom. On the other hand, by far the most
numerous writings of most thirteenth-century Masters do stem from the
classroom, either as lectures by Masters in Arts or Bachelors in Theology
522 Natural philosophy
commenting on the Sentences, or Masters disputing questions or comment-

ing on Scripture. In this latter group, some of the more outstanding
scholars were Siger of Brabant, Bonaventure, Henry of Ghent, John Duns
Scotus, and in the early fourteenth century, William Ockham.
While the tendency of mid-thirteenth-century Masters was to effect a
harmony of faith and reason, using the once-proscribed Aristotle as an aid
to reason and faith, the tendency of early fourteenth-century Masters was
to separate the demands of 'demonstration' from the simplicity of the
Christian faith, seeing Aristotle as a pagan and his followers as hostile to the
faith. This was particularly true among the Franciscans and the Austin
Friars. (It may well be that this change in the intellectual climate is in part a
consequence of the Condemnation of 1270 at Paris and the Condemna-
tions of 1277 at Paris and Oxford.) During the second half of the thirteenth
century all three of the new sciences were avidly studied and commented
upon. Indeed, commentaries on the entire Aristotelian Corpus (as it was
then conceived of) abound during this period, as they do not in the
fourteenth century.
The special status of natural philosophy

But obviously all the Aristotelian books together with the seven liberal arts
could not be covered in the eight years one spent in the Faculty of Arts. The
books of logic were naturally given special emphasis, since they were the
indispensable tool (organon) of all scientific knowledge. Inevitably logic
remained one of the main pillars of medieval education and of scholastic
thought, even in the later Middle Ages. Among the 'real' sciences, how-
ever, natural philosophy was acknowledged as fundamentally important
by all the scholastics, whatever their views of the status of moral philosophy
or metaphysics. All acknowledged that the libri naturales, the Aristotelian
and pseudo-Aristotelian books of natural science, provided the foundation
of ethics, metaphysics and - to a certain extent - theology, in the sense in
which Anselm had described it as 'faith seeking understanding'. While
interest in moral philosophy and metaphysics declined in the later Middle
Ages, as is evident from both university requirements and extant com-
mentaries, interest in natural science never flagged, although different
emphases developed with succeeding generations and different centres of
learning. Amid all vicissitudes, two Aristotelian books of natural philos-
ophy stand out as of primary importance: the Physics and the De anima.
Around them centred many of the problems that methodologically would
have arisen in the context ofother books, had they been treated. The eight
Aristotle's Physics and the science of motion 523
books of Aristotle's Physics became the focal point of all basic problems of
natural science, while the three books of Aristotle's De anima became the
focal point of all basic problems of human psychology, especially that of
the separated soul. Few bothered to deal with De caelo. De generatione et
corruptione, Meteora or with the details of De animalibus, while the so-called
parva naturalia were covered in class in a matter of a few weeks.
Aristotelian natural philosophy extended from very general considera-
tions in the Physics to particular subjects discussed in the other books
dealing with the heavens, metals, minerals, flora, fauna, environment and
human behaviour. The most remarkable example of the medieval version
of this kind of Aristotelian science is found in the writings of Albert the
Great. There was also a non-Aristotelian medieval approach to natural
science, one that sought explanations of natural species and phenomena in
perspective and the mathematics of light rays; perhaps the most remarkable
examples of this 'Pythagorean' approach - the term is Albert's - are found
in the writings of Robert Grosseteste and later in Roger Bacon. The
Aristotelian approach remained clearly dominant, however. Even while
Walter Burley was concerned with a wide range of problems for the
'realist' dialectician in natural philosophy, and John Buridan was con-
cerned with new problems of 'impetus' and the via moderna in its widest
application, the Physics of Aristotle remained the cornerstone of medieval
natural philosophy. William Ockham wrote no commentaries on moral
philosophy or metaphysics, but he composed both an Expositio and
Quaestiones on Aristotle's Physics, and made it the focal point of his logical
analysis of motion and quantity. The influential De proportionibus veloci-
tatum in motibus of Thomas Bradwardine centred on the problem of ratios
in Aristotle's Physics VII, 4 - 5 , while his De continuo (against Grosseteste,
Walter Chatton, and Henry of Harclay) developed the basic principle in
Aristotle's Physics VI, 1-2. In the later Middle Ages, proforma requirements
of the curriculum notwithstanding, logic and natural philosophy were the
two pillars of philosophical education. Aristotle's Physics clearly ranked as the
base of the second pillar, with the whole of psychology (including 'phdos-
ophy of mind') considered as an integral part of one single science, variously
called scientia naturalis, philosophia naturalis, scientia de naturalibus or the like.
The medieval understanding of the structure of Aristotle's Physics

Medieval schoolmen took the eight Books of Aristotle's Physics as they
received them from the Greek (divided into chapters) and from the Arabic
(divided into texts), without raising questions about authenticity, possible
autonomy of books, recensions, or possible posthumous compilation.

They were, however, very much concerned with the arrangement and
content of its Books. As to their arrangement, all the scholastics saw the
first two Books as dealing with the general subject of natural philosophy,
ens mobile, which had 'motion' (Bks III-VIII) as its proper attribute (propria
passio).
Interpretations of Physics I
Thomas Aquinas saw more clearly than Albert that Book I discussed the
possibility of all change (natural or unnatural) in the face of the pre-
Socratic dilemma concerning genesis (fieri): 'no thing can come from
nothing' and 'what is does not come to be'. Aristotle's resolution of this
dilemma in terms of potentiality and actuality was not uniformly ap-
preciated by all scholastics. Many, clinging to imagination, conceived 'first
matter' as a substratum having some minimal actuality of its own, even
prior to first actuality, or 'form'. Aquinas, however, and many others
insisted that unless this 'first matter' were pure potentiality (pura potentia),
having no actuality whatever of its own (not even esse), the dilemma faced
by Aristotle could not be resolved. It would still be impossible for the
ultimate substance and reality of anything truly to change (fieri, mutari).
For this reason, the whole of Aquinas' natural philosophy is based on the
absolute unicity of substantial form in every material composite. For him,
the one and only esse a substance has comes entirely from the actualising
form (forma dat esse); whatever is posterior to the first, immediate actualis-
ing form must consequently be accidental to the composite. The many
opponents of this position conceded not only a minimal esse and actuality
to 'first matter', but insisted that a plurality of substantial forms, hierarchi-
cally arranged, was necessary to safeguard certain truths of the Christian
faith. This 'Christian' position was the one defended by Robert
Kilwardby, Dominican Archbishop of Canterbury, when he condemned
the unicity doctrine at Oxford on 18 March 1277, a condemnation that was
reiterated by his successor, the Franciscan John Peckham, seven years later.
Interpretations of Physics II
While Book I discussed the possibility of any change taking place in the
world, Book II was seen to limit 'the subject of this science' to nature, which
was identified as 'a principle of motion and rest in those things to which it
belongs per se and not accidentally' (iQ2 b 2i-3). Thus the natural philoso-
pher is not properly concerned with those things that come about in the
world by human ingenuity (art) or by pure chance. But, as is evident from

Book I, the word 'nature' can be used in two senses: in. the active sense of
'form' as a formal principle (ut principium formale seu activum) or in the
passive sense of 'matter' as a material principle (ut principium materiale seu
passivum). Therefore the natural philosopher must study both the active
principle (form) and the passive principle (matter) of all natural things. But
since no passive principle can be actualised without some extrinsic, efficient
cause acting for a definite purpose, the natural philosopher must also study
the efficient and final causes of all natural things. In this regard he differs
from the mathematician who applies his mathematical principles to the
same natural phenomena, since the mathematician 'abstracts from sensible
matter and from motion as such' * and is concerned only with the measur-
able aspects of those phenomena. For Albert, Thomas, and even Grosse-
teste, those measurable aspects constitute the objects of special sciences
really distinct from natural science. Thomas called them scientiae mediae,2 -
e.g., astronomy and optics - intermediate between natural philosophy and
pure mathematical sciences, but formally mathematical, not natural. Thus,
for Thomas, the astronomer and the naturalist may demonstrate the same
conclusion, e.g., the sphericity of the earth, but they do so through
formally different principles of demonstration. 3 This vast area of inter-
mediate science developed rapidly in succeeding centuries; its concern was
the measurable aspects of natural phenomena and what could be de-
monstrated about them, disregarding the 'natural causes' that brought
them about, whether they be strictly material, formal, efficient or final.
Medieval attitudes towards the role of chance in

Aristotelian natural philosophy
Surprisingly, not all the scholastics appreciated the important role 'chance'
(tuchë, casus) played in Aristotelian natural philosophy. While 'nature'
necessarily and per se tends toward one predetermined end as to a final
cause, by definition there can be no per se cause of a 'chance event'.
Precisely because a 'chance event' is indeterminable, it cannot have a per se
cause or be a per se cause (ia6 b 22-3). Thus, for Aristotle 'chance is a per
accidens cause, happening in a minority of cases and in the sphere of those
1. Boethius i860, i, 2; PL 64; col. 1250; cf. Aquinas, In Boeth. De Irin., q. 5, a. 3 (per totum); In V
Metaph., lect. 16, n. 989. See analysis in Laso 1952, pp. 1-29.
2. In Boeth. De trin., q. 5, a. 3 <ni6;also/n II Phys., lect. 3,n. 8; ST, Ilallae, q.9, a. 2 ad 3. See Pseudo-
Albertus 1977, c. 3, p. 14.11-2.
3. Aquinas, ST, Ilallae, q. 1, a. 1.
actions that are done for the sake of something involving purpose'
(ΐ97*5 _ 6)· Aristotle's world was not fatalistic or mechanistic. Nevertheless,
many Christians in the Middle Ages sought to explain apparently chance
events on earth by astrology (astronomia iudicialis) or some other form of
divination. That is, they sought to explain future events by Fate, the stars,
or some created superpower, short of divine providence itself 4 The prob
lem with this fatalism is that it denies man's free will, which is of course
important for Christianity. Medieval scholastics were nevertheless imbued
with the idea of the influences of celestial bodies and intelligences on
terrestrial phenomena. Many of the scholastics, notably Albert the Great
and Roger Bacon, were highly credulous about these influences on the
atmosphere, bodily humours and human emotions. But even apart from
the influences of celestial bodies, man's free will (ad utrumlibet), and nature's
predetermination to an end actually attained in most cases (ut in pluribus),
medieval natural philosophy recognised a whole range of events that do
genuinely happen by chance (ut inpaucioribus), which are outside the range
of true 'scientific' knowledge. Regarding those chance events we can have
only 'probability' or 'likelihood', not certainty.
Interpretations of Physics III-VIII

Physics III—VIII was seen as a systematic examination of all the basic aspects
of motion in general: its essential nature (Book III), its natural conditions of
place and time (Book IV), types of true motion (Book V), continuity and
quantitative divisibility (Book VI), and efficient causes of various motions
ordered per se ultimately to a First Unmoved Mover, itself devoid of
motion, matter, quantity, and limitation of power (Books VII-VIII).
The study of Aristotelian natural philosophy

The entire ensemble of the eight Books presented only the basic theory of
natural philosophy, presupposed by subsequent investigation, but needing
to be filled out with increasingly detailed studies of celestial and terrestrial
motions down to the most specific differences of metals, minerals, flora,
fauna, and mankind in its complex nature, races, and sexuality. Although
medieval students had numerous summaries and manuals of Aristotelian
natural philosophy and studied some of the actual texts in class together
with various commentaries, notably that of Averroes, there was not
enough time assigned in the university calendar to study the whole of
4. The principal source of medieval astrology was not Stoic determinism but Muslim fatalism,
particularly that of Albumbazar (Abu Ma'shar),
Aristotle's natural science. The most that could be hoped for was a general
idea of some of the main principles and a serious study of at least the Physics
and the De anima, a study that always presupposed logic, grammar, and
some of the mathematical sciences.
Aristotle's account of motion

Aristotle clearly stated that since 'nature' is 'a principle of motion and
change' and is the subject of natural philosophy, it is obviously necessary
'that we understand the meaning of "motion"; for if it were unknown, the
meaning of "nature" too would be unknown' (200 b i2-i4). The facts that
some things really move and that moving is objectively different from not
moving Aristotle took as primary data of sense (185*12-14), within which
he included genesis (fieri). For him there is an obvious difference between
the state of a body before it moves (potency) and its state after it has begun
to move (actuality). The reality of movement itself must somehow be a
kind of actualisation ofthat potentiality, precisely as it is potential, toward
completed actuality, just as the process of'building' is the actualisation of
the 'buildable' only as long as the building is 'being built' (cf. 20i b o-i5).
Albert the Great's exposition of Aristotle on motion

In this exposition the scholastics, notably Albert the Great,5 recognised
three kinds of 'definition' of motion: (i) 'the actualising of what exists
potentially, insofar as it exists potentially' (20i a io-ii), which they called
formal; (ii) 'the actualising of the movable qua movable' (20i a 28-o), which
they called material, because it is better known to us and demonstrable from
the prior definition; (iii) the simultaneous actuality of the mover as cause
and of the moved as that in which the change takes place, which Albert
called a total definition, because it includes the efficient (and implicitly the
final) cause whereby motion takes place. Hence, for Albert, the category of
agere (acting) belongs to the agent as coming from him (a quo), while the
category of pati (undergoing) belongs to the recipient as existing in him (in
quo).
Medieval misunderstandings of Albert's exposition

Some misunderstanding arose among later scholastics concerning Albert's
exposition of Aristotle's doctrine.6 The background was Avicenna's in-
sistence, based on his conception of all nature as predominantly passive,
5. Physica III, tr. 1, cc. 1-8; Albert the Great 1890-9, III, pp. 177-202.
6. This misunderstanding has persisted in some modern historians. See, e.g., Maier 1944; 1949,
pp. 9-25; 1958, pp. 61—76; Dijksterhuis 1961, pp. 174-6; Pedersen 1953.
that all motion, of whatever kind, belongs to the category of passio as a

fluxusformae, caused by intelligences. But Averroes, in a famous text (Phys.
Ill, comm. 4; V, comm. 9), insisted that motion cannot belong to any one
category; it must be ζ forma reducible to one of the three terms or ends of
motion: quantity, quality, or place. The alternative conceptions of motion
as fluxus formae (a succession of form) orforma fluens (a successive form) are
nowhere presented in Avicenna, Averroes, or Albert; further, Albert in no
way aligns himself with either Avicenna or Averroes, nor does he present
Averroes simply as a defender of any forma-fluens theory; finally, no
fourteenth-century misreading of Albert could have prepared the way for
Ockham's denial of motion as a reality distinct from form. 7
Albert and Thomas on 'motion' as an analogical term

Albert's main objection to Avicenna is that he places motion in a single
category of passio, essentially different from the term of motion, as though
it were a univocal concept. His main objection to Averroes is that he fails to
distinguish sufficiently between motion as a via (or process) and the term
(or end) of motion. For Albert, motion is essentially a fluxus alicuius entis (a
succession of a certain being) existing in the moved, but belonging to the
agent as its cause. He even defines motion as a continuus exitus formae (a
continuous flow of a form), which belongs to the same category as its term,
differing from it not as a distinct category (as Avicenna had held), but as a
process differs from its term. The main point for Albert (and for Thomas) is
that 'motion' is an analogical term, more easily applied by us to loco
motion, but also true of growth and of the intension and remission of
qualities. The analogical term becomes even more diverse when applied to
'substantial change' (mutatio substantialis), and most tenuous when applied
to 'creation' (creatio), where there is no pre-existent matter.
Albert and Thomas on sources of motion

Unlike Averroes, neither Albert nor Thomas ever spoke of the substantial
form of an inanimate body as a conjoined mover (motor coniunctus); this
term was reserved exclusively for the substantial form of higher living
things which by definition 'move themselves'. Although Albert himself
denied the view, he explained that Aristotle and most of the Peripatetics
thought the heavens to be 'animated' with a motor coniunctus that moved
with the celestial body under the influence of a separated intelligence. 8 O n
7. Anneliese Maier to the contrary notwithstanding.

8. Albert, L/lier de causis, II, tr. i, c. 2; II, tr. 2, c. I, etc.
the other hand, for Averroes the Aristotelian principle 'everything that is
moved is moved by another' required the constant presence of a conjoined
mover to account for natural as well as violent motions. Thus, although
Averroes recognised the difference between living and non-living things,
he created a separation between the 'form' as mover of the medium and
'matter' as moved by the form through the medium (i.e., per accidens),
which meant that there could be no motion whatever in a void.9 For
Albert and Thomas, however, the 'form' of any inanimate body is simply a
formal, active principle from which natural motion spontaneously and
immediately flows toward an end unless impeded.10 Such a 'form' is in no
way the efficient cause of its natural motion. But a 'living form', an anima,
is not only a formal principle of certain elemental natural motions, but also
a vital principle of other natural motions such as growth that cannot be
classified as 'self-motion'. In most animals, the 'form' is not only the active
principle of elemental motions, spontaneous growth and digestion, but it is
also the real efficient cause of some of its own motions, such as walking,
swimming, crawling, or flying. ' ' Only in this last case can the anima or
forma be called a conjoined mover in any proper sense of the term, for it is
truly an efficient cause of some of its movements.
The basic problem, however, is that each of these bodies must 'be moved'
in some way in the first place, and therefore 'moved by another'. It is
obvious that every terrestrial body must be generated in the first place in
order to have the nature it has, whereby it acts naturally. Celestial bodies
'are moved* perpetually without coming to any final term, or rest; more-
over their 'souls' (if such there be, according to Aristotle) are themselves
'moved' by reason of the bodies they animate, i.e., per accidens. But the
whole point of Aristotle's Books VII and VIII is that all such motions have
no 'explanation' unless there is some First Mover, itself entirely unmoved
per se and per accidens, immaterial (separated from matter), having infinite
power to move the entire universe as a whole by means of the first heaven
for all eternity. Albert and Thomas thought that even if the universe
moved and existed from all eternity, the First Uncaused Mover would still,
as First Being, have 'to create' (producere, or movere) the entire universe ex
nihilo - even on Aristotle's own principles.12
9. See Weisheipl 1974b and Moody 1951.

10. Weisheipl 1955, pp. 1-32.
11. Albert the Great 195 ia. Lift«· deprincipiis motusprocessivi, pp. 47-75; cf. Liber de causis II, tr. 3, c. 11,
and c. 17; Albert the Great 1890-9C, pp. 560-ia, 568-93.
12. Albert the Great 1890-9C, II, tr. 1, c 17; tr. 2, c. 17; Aquinas, In II Sent., dist. 1, q. 1, a. 2; De
potentia, q. 3, a. 5.
Ockham and the 'common view' of motion

William Ockham, lecturing on the Sentences at Oxford around 1317-18,
was not troubled about any controversy over motion as forma fluens or
fluxus formae. ' 3 Rather, he was concerned with 'a common view' of
motion, defended by Walter Burley and others, that motion is (i) a forma
diminuta distinct from the body in motion, 1 4 and (ii) increased and de-
creased by a succession of distinct forms, more or less intense, of indivisible
duration, 15 a view Burley claimed to have taken from Godfrey of
Fontaines, if not from Aristotle.
Theories of motion and Ockham's razor

The problem of motion for Ockham was only one of many that seemed to
multiply entities not at all required by sound philosophy or the Christian
faith. For him, the problem of what has real existence in itself reduces to the
basic question of distinguishing a res absoluta as distinct from what is only
relative or connotative - i.e., from what is not a thing but only a way of
knowing things or of talking about them. But for Ockham there are only
two kinds of res absolutae, substances and qualities: 'Besides absolute things,
namely substance and qualities, there is nothing imaginable either in act or
in potency.' l 6 An 'absolute thing' (res absoluta), for Ockham, is an objec-
tively existing reality (a res permanens), distinct in place and subject from
every other permanent thing, capable of existing, at least by God's absolute
power, with nothing prior or posterior to it. O n this basis, only individual
substances (either matter or form) and individual qualities (such as colour,
heat, shape, and weight) are res absolutae (the latter because of their occur-
rence in the Eucharist). All the other Aristotelian 'categories' are therefore
simply ways of understanding things or talking about them.
Motion not a distinct reality

Concerning the first point on motion, the Aristotelian text to which
Ockham constantly appeals is: 'There is no such thing as motion over and
above things' (200b34). Aristotle merely meant that there is no special
13. Ockham was under the impression that Averroes defended the view of motion asfluxusformae
(Ockham 1494-6, Sent. II, q. 9, obj. 7 H), and he seems to have misunderstood the presumed
point of the distinction as presented by A. Maier (ibid., ad 7 S); cf. Ockham 1944, pp. 49~52.
14. Super octo libros Physicorum, III, text. 4; Burley 1501. f. 6ovb and f. 62".
15. De intensione et remissioneformarum, Burley 1496, ff. 2'-i 5". This view is summarised and rejected
by Gregory of Rimini, Sent. I, dist. 17, q. 4, a. 2; Gregory of Rimini 1522, ff. I04 vb -i07 v '. Cf.
Godfrey of Fontaines 1914, Quodl. VII, q. 7; discussed in Duhem 1906—13, III, pp. 327-8.
16. Summa logicae I, c. 49; Ockham 1974a, OP I, p. 154.23-4.
category of'motion' over and above the ten discussed in the Categories, but
for Ockham this meant that motion is not a distinct reality over and above
the body in motion. In fact, 'to move' means simply to be in one place after
another 'without interruption'. Consequently, 'motion' is not something
over and above the body in motion, but is the body itself acquiring part
after part successively, i.e., without interruption. Since being 'without
interruption' signifies an absence, a negation, an ens rationis, it cannot be
something real and positive added to the body in reality. Therefore,
motion, as such, cannot be a reality over and above the body in motion, but
only a way of speaking about individual bodies. n
One of Ockham's major concerns was about 'the fiction of abstract
nouns', since many philosophers erroneously imagine that distinct nouns
correspond to distinct realities outside the mind. 18 Abstract nouns, such as
'motion,' 'change,' 'action,' or 'passion', which are derived from verbs, are
used in human speech, according to Ockham, only 'for the sake of brevity
of speech or ornamentation of language'.19 Thus the noun 'locomotion'
briefly expresses the cumbersome phrase: 'a body that is in one place and
later in another place in such a way that at no time does it rest in any place'.
But this should not be taken to mean that locomotion (or any other
motion) is a reality over and above the body in a place.20 Hence, according
to Ockham the term 'motion' has a double signification: one positive,
signifying the body (res absoluta) itself, the other negative (an ens rationis),
connoting uninterrupted succession of part after part. Since negation is an
ens rationis, it cannot be a reality over and above the body in motion.
Motion and the absence of forms yet to come

Concerning the second point on motion noted above, the essence of
motion consists in the acquisition of part after part, not in the loss of pre-
vious parts or degrees, as Burley proposed. In the obvious case of natural
augmentation (invariably exemplified by rarefaction in the fourteenth cen-
tury) the preceding part remains, but a new extension is added, quantity
itself not being distinct from the quantified body according to Ockham. In
condensation a body naturally decreases its dimensions while retaining all
of its previous parts. According to the doctrine of the Eucharist, God can so
decrease the extension of matter by his absolute power as to make a body
17. Ockham 1495-6, Sent. II, q. 9 (per totum); Cf. Ockham 1944, pp. 32-69.
18. Ockham 1944, p. 46; Ockham 1930,1, q. 1; pp. 52-66.
19. Ockham 1944, p. 37.
20. Ibid., p. 45.
exist without any extension whatever. That is to say, by his absolute power
God can make all the parts of the material universe to exist simul, as at a
point, in such a way that the distinct parts would be without any extension
in space, although not without existing in some place. Therefore, Ockham
argued, 'quantity' or extension in space cannot be a res absoluta distinct
from substance or qualities.
According to Ockham, although it is true that in local motion, in
decrease (in condensation), and in remission (in qualities), the preceding
'form' is lost or left behind, this is by no means essential to the concept of
motion:
For in the motion of augmentation and alteration [i.e., intension of qualities] it is
obvious that prior parts remain with those to be acquired, since increase is achieved
by the addition of one part to another . . . , likewise in local motion, although in
fact one place or location is continuously lost just as another is continuously
acquired, nevertheless by God's power all the lost parts could remain simul-
taneously, since God could make a body move locally in such a way that when it
acquired a new place, it retained the previous one, and so on, because God can
make the same body exist [simultaneously] in different places.21
The point is that although motion sometimes involves a 'loss' of some kind,
it is essentially the absence of forms yet-to-come that is connoted by the
fact of motion. Thus 'motion' signifies principally the existing res per-
manens, the individual substance or quality, and connotes the non-existence
of forms yet to come: 'And through such negations and affirmations
"succession" is explained in motion'. 2 2 For Ockham the 'negation' implied
in motion is not the non-existence of a previous form, but of a future form;
all successive motion is essentially the acquisition of additional parts or
designations. Just as in augmentation and increase of qualities the previous
form of quantity or quality is not lost, but remains with the additional
form, so it can be in local motion.
From this it follows that intension and remission of forms presented
Ockham with no real problems. Intension, for him, simply meant more by
21. 'Nam in motu augmentationis et alterationis manifestum est quod partes priores manent cum
posterioribus, quia augmentatio fit per additionem unius partis ad partem ..., similiter in motu
locali, licet de facto unus locus vel ubi continue deperdatur sicut aliud continue acquiritur, tamen
de potentia dei possunt omnia acquisita simul manere, quia deus potest facere quod corpus
moveatur localiter, et quod semper quando acquirit novum locum retineat primum locum et
secundum, et sic deinceps, quia deus potest facere ut idem corpus sit in diversis locis.' Ockham
1495-6, Sent. II, q. 9 H. The same conception of motion is found in Peter John Olivi 1922-6,
Quaestiones in secundum librum Sententiarum, q. 27; I, p. 470.
22. Ockham 1495-6, Sent. II, q. 9 H.
addition; remission simply meant less by subtraction.23 Ockham never

clearly distinguished between the extension and the intension of any
quality, much less between mass and velocity.
Bradwardine and the commensurability of motions

In 1328 Thomas Bradwardine, Fellow of Merton College and theology
student at Oxford, published an influential treatise entitled De proportione
velocitatum in motibus, which dealt directly with the commensurability of
motions, discussed by Aristotle in Physics VII, 4-5. Since all successive
motions are proportionate to one another in velocity (with direction
intended), natural philosophy, which studies motion, must not ignore the
proportions of motions and velocities. But Bradwardine recognised that
this essential and extremely difficult part of natural philosophy requires a
sound knowledge of mathematics, 'for, as Boethius says, whoever dis-
misses mathematical studies has destroyed the whole of philosophical
knowledge'.24
For Aristotle not all motions are comparable, neither according to term
nor according to velocity, but only those which are of the same species.
Hence alteration cannot be compared with local motion, and even rec-
tilinear motion cannot be compared with rotational motion since a straight
line and a circle are specifically different (248 a i8- b 7; 248 b io-i2). But even
within a given species of motion, there are limits beyond which motions
cannot be compared, as when a moving body is divided in such a way that
its power is less than the resistance offered by the body to be moved, and
conversely, when the body to be moved is doubled beyond the capacity of
the moving power.
Moving power and velocity

The two notions essential to this doctrine are that of 'moving power'
(potentia motiva) over resistance and that of velocity (velocitas in motibus) in
the direction of motion. While motion itself, for Aristotle, is not 'speed' or
'velocity', all natural motions (unlike rest) have a velocity, in each case
involving both a time and kind of'distance', which accelerates toward its
natural end. Hence the basic motion in the whole terrestrial world is
accelerated motion toward a goal, even if the motion be irregular as in the
case of growth and alteration. For this reason Aristotle had no difficulty in
23. See Seni. I, dist. 17, q. 4; Ockham 1977a, OT III, pp. 479—519.
24. Thomas Bradwardine 1955, p. 65. Cf. Boethius 1847, De arithmetica I, c. 1; PL 63, col. 1081.
explaining 'gravitational motion' or 'the acceleration of freely falling

bodies'; it was the natural kind of motion for a heavy body to have.
Velocity considered in respect of its cause or of its effect

To appreciate the significance of Bradwardine's treatise, two essential
distinctions must be noted. Although Bradwardine discussed them only in
passing and took them for granted, they are fundamental to all subsequent
discussions of motion at Oxford and elsewhere. The first is between
velocity considered penes causam (in respect of its cause) and velocity
considered penes effectum (in respect of its effect). The first had to do with
what Gaetano di Thiene called the a priori conditions of the proportion of
force over resistance productive of different speeds in a given direction, 25 a
consideration that later came to be called 'dynamics'. The important point
is that 'velocity considered in respect of its cause' was a variable ratio
(proportio) of force to resistance that was the cause (an efficient cause, so to
speak) of variable velocities. The second consideration had to do with what
Gaetano called the a posteriori conditions of speeds effected, regardless of
direction. This consideration of motion later came to be called 'kine-
matics'. Regarding this distinction, it should be noted first that these
questions were asked not only about local motion, as with Bradwardine,
but also about augmentation and alteration, as with John Dumbleton,
William Heytesbury, and others. Second, Bradwardine was concerned not
only with the variable proportion of moving power and resistance (dy-
namics), but also with the variable proportion of distance to time, that is,
with velocity in the original direction of motion.
Total velocity and instantaneous velocity

The second basic distinction is between the whole motion (velocitas totalis)
from beginning to end and motion at any given instant (velocitas in—
stantanea).26 N o Aristotelian maintained that true motion could take place
in an instant, i.e., without time. 'Velocitas instantanea' was the expression
used to designate the 'intensity' or 'degree' (gradus) of a motion at any
particular instant. It was measured by the velocitas totalis a body would have
if it were moving at a constant speed at the rate it had at that instant, for
example, 50 km. per hour. This distinction corresponded to the older
distinction between the 'quantity of motion' (the time it takes) and the
'quality of motion' (its intensity).
25. Gaetano di Thiene 1494, f. 37™; see Duhem 1906-13, III, pp. 302-9; Clagett 1959, pp. 207-9.
26. See esp. Maier 1951, pp. n i - j i ; 1958, pp. 147-86; Clagett 1959, pp. 199-219.
The initial problem for Bradwardine and the Mertonians (the philoso-
phers associated with him at Oxford) 2 7 was to give some definite meaning
to the expression 'instantaneous velocity' when talking of a uniformly
accelerated body moving from zero to maximum degree compared to the
more intelligible expression velocitas totalis. Their solution, known in the
history of science as 'the Mertonian mean speed theorem', was to say that
the 'total velocity' of a body moving with constant speed 'corresponds' to
the speed at the middle instant of the time taken by a body moving with
uniform acceleration from zero to maximum degree. 28 Obviously, no
general rule can be established for nonuniformly accelerated or decelerated
motions, such as growing and decaying, but a rough mid-point can be
established, although it is only widely analogous to velocitas instantanea. It is
also clear that in questions of pure kinematics such as the Mertonians were
interested in, there was concern only over the spatial-temporal conditions
of speed, not over the direction of motion.
Bradwardine's laws
The aim of Bradwardine's treatise was to determine a universal rule that
would govern proportions between moving power and resistance, on the
one hand, and between distance and time on the other. For him there had
to be a proportion of velocities between motions of the same kind. The
difficulty Bradwardine found with the traditional Aristotelian formula was
twofold: (i) 'it is insufficient, because it does not determine the proportion
of the velocity of motion except in cases where either the mover is the same
or equal, or when the mobile body is the same or equal'. 2 9 (ii) 'It produces a
fallacy of the consequent', because it would then follow that 'any mobile
body could be moved by any mover', whereas for motion to occur, 'the
proportion of the mover to the moved must always be one of greater
inequality'. 30 Bradwardine devised a universal formula in which both
forces and resistances could vary and yet maintain a proportion of greater
inequality and always produce a proportionate velocity in moving bodies.
The formula he suggested was a function for any given ratio of greater
inequality. In this way any exponential variation of a given ratio of greater
inequality would always produce a definite, proportional variation of
speeds in moving bodies. Although Bradwardine's laws primarily concern
27. See Sylla's contribution to this volume.

28. Texts and comment in Clagett 1959, pp. 255-329; Grant 1974, pp. 237-43.
29. Bradwardine 1955, pp. 96-9.
30. Ibid., pp. 98-105.
kinematic representation, i.e., speed (penes effectum), they are firmly ground-
ed in a dynamical cause of motion (penes causam).31
The influence of Bradwardine's work

Bradwardine's treatise became immediately popular both in England and
on the Continent. But undergraduates were introduced to Bradwardine's
doctrine of 'geometrical proportionality' more often through shorter and
simpler summaries, often called Proportiones breves or De proportionibus.32
The new conception of motion at least suggested by Bradwardine can be
seen in the author of an anonymous Tractatus de motu locali diffbrmi, possibly
Richard Swineshead, despite its expression in terms of the traditional
Aristotelian causes:
The material cause of motion is whatever is acquired through motion; the formal
cause is a certain transmutation conjoined with time; the efficient cause is a
proportion of greater inequality of the moving power over resistance; and the final
cause is the goal intended.33
One significant point is that 'velocity' or 'speed' was conceived of as an
intensive quality, a 'form', that could be intensified and remitted as any other
'form'. In the later Middle Ages velocitas, whether considered with a vector
quantity or not, was always discussed under the general heading of the
'intension and remission of forms'. But after Bradwardine every attempt
was made to reduce the kinematics of all intensive qualities to some kind of
measurement that would be amenable to Bradwardine's formula. While
this attempt was doomed to failure, the attempt to mathematise all
motions, including celestial motions (as Bradwardine himself attempted in
Chapter 4 of his treatise) was destined for considerable success.
31. On Bradwardine's mathematics see Molland 1967; 1968a; 1968b; and Mahoney 1978.
32. Besides Albert of Saxony's well-known treatise produced at Paris, two in particular were in
common use among undergraduates. One has been edited and translated in Clagett 1959,
pp. 481-94; the other exists in many MSS and begins, 'Omnis proporcio aut est communiter dicta,
proprie dicta, vel magis proprie dicta ...'
33. 'Unde causa materialis motus seu materia motus est ipsum acquisitum per motum; causa formalis
est ipsa transmutatio quedam coniuncta cum tempore; causa efficiens est proportio maioris
inequalitatis potentie motive super potentiam resistivam; causa finalis est terminus internus per
motum.'Anon., Tractatus de motu locali difformi, MS Cambridge, Gon ville & Caius 499/268, f. 212™.
This treatise occurs between two other short treatises ascribed to Swineshead and with the
Calculationes.
26
THE EFFECT O F T H E C O N D E M N A T I O N
OF 1277
Potentia Dei absoluta

The Condemnation of 219 articles in theology and natural philosophy by
the bishop of Paris in 1277 points to a significant development in the
history of medieval philosophy generally, but especially natural philoso-
phy.1 Whatever may have induced bishop Stephen Tempier and his
advisers to promulgate the condemnation, the most significant out-
come was an emphasis on the reality and importance of God's absolute
power (potentia Dei absoluta) to do whatever He pleases short of bringing
about a logical contradiction. Although the doctrine of God's absolute
power was hardly new in the thirteenth century, the introduction into the
Latin West of Greco-Arabic physics and natural philosophy, with their in-
dependent, and often deterministic, philosophical and scientific explana-
tory principles, conferred on that doctrine a new and more significant
status. After 1277, appeals to God's absolute power were frequently intro-
duced into discussions of Aristotelian physics and cosmology.
The range of the Condemnation

The wide range of topics covered by the Condemnation indicates its
potential impact on natural philosophy. Among the themes at which
several articles were directed are God's knowability, nature, will, and
power; the causation and eternality of the world; the nature and function
of intelligences; the nature and operation of the heavens and the generation
of terrestrial things; the necessity and contingency of events; the principles
of material objects; man and the active intellect. Whether implicitly or
explicitly, many of the articles asserted God's infinite and absolute creative
1. The Latin text appears in CUP I, pp. 543-55, and, reorganised by subject matter, in Mandonnet
1908, pp. 175-91. For an edition of the text with a discussion of its possible sources see Hissette
1977. For a translation of all the articles see Fortin and O'Neill 1963 (reprinted in Hyman and
Walsh 1967, pp. 540-9). A selection of the articles relevant to medieval science appears in Grant
1974. pp. 45-50.
and causative power against those who thought to circumscribe it by the

principles of natural philosophy. Nowhere is the spirit of the Condemna-
tion better revealed than in Article 147, which condemned the opinion
'That the absolutely impossible cannot be done by God or another agent',
if '"impossible" is understood according to nature'. 2
The effect of the Condemnation

The Condemnation was in effect at Paris throughout the fourteenth
century. The seriousness with which it was taken is shown by numerous
direct and indirect references to its articles by such eminent scholastics as
Richard of Middleton, John Duns Scotus, William Ockham, Walter
Burley, Peter Aureoli, John of Ripa, John Buridan, and Nicole Oresme. As
the most characteristic feature of the Condemnation, God's absolute power
was invoked in a variety of hypothetical physical situations. The novel
supernatural alternatives considered in the aftermath of the Condemnation
conditioned scholastics to contemplate physical possibilities outside the ken
of Aristotelian natural philosophy, and frequently in direct conflict with it;
indeed, a concern with such possibilities became a characteristic of late
medieval scholastic thought.
The Condemnation and the concept of a vacuum

N o area of physical thought was more affected by the Condemnation and
its emphasis on God's absolute power than the concept of a vacuum. Here
two articles played a paramount role, the thirty-fourth, which condemned
the claim that the First Cause, or God, could not produce more than one
world 3 and the forty-ninth, which condemned the claim that God could
not move the world with a rectilinear motion because a vacuum would be
left behind. 4 In exploring the consequences of these possibilities, concepts
contrary to Aristotelian physics and cosmology were found plausible
rather than impossible. Not only could God create other worlds, but each
would be a closed system like ours with its own proper centre and
circumference. With the simultaneous existence of a plurality of centres
and circumferences rendered hypothetically intelligible, Aristotle's
argument for the necessary existence of a single centre and circumference,
2. 'Quod impossibile simpliciter non potest fieri a Deo, vel ab agente alio. - Error, si de impossibili
secundum naturam intelligatur.' CUP\, p. 552.
3. 'Quod prima causa non posset plures mundos facere.' CUP I, p. 545.
4. 'Quod Deus non possit movere celum motu recto. Et ratio est, quia tunc relinqueret vacuum.'
CUP I, p. 54°·
The effect of the condemnation of 1277 539
on which he had founded his belief in a unique world, was plainly

subverted.
Articles 34 and 49 also made it appear plausible to suppose that an infinite
empty space existed beyond our world. For if God did make other worlds,
empty space would intervene between them; and if God moved the world
rectilinearly not only would an empty space be left behind but also the
world would move into and out ofother empty spaces that lay beyond. To
John of Ripa, for example, the mere possibility that God could move the
world suggested the actual existence of an extra-cosmic empty space; for
otherwise there could be no places or spaces capable of receiving the world
or any part of it, and so God would be unable to move the world, which
would restrict His power (and violate the intent of article 49).5
Although no articles of the Condemnation concerned vacua within the
cosmos itself, it seemed obvious that if God could create or allow a vacuum
beyond the world, He surely could do the same within the world; and so
God was frequently imagined to annihilate all or part of the matter within
the material plenum of our world. Potential problems that would arise
from such divine action were often discussed. Thus it was asked: Was the
empty interval a proper space? Would the concave surface surrounding
that space be a proper place? Would a stone placed in such a void be capable
of rectilinear motion? Could distances be measured within such emptiness?
Would vision and hearing be possible there? Analysis of these and similar
'thought experiments' in the late Middle Ages were often made in terms of
Aristotelian principles even though the conditions imagined were
'contrary to fact' and impossible within Aristotelian natural philosophy.
Conclusion
As a consequence of the Condemnation of 1277, God's absolute power
became a convenient vehicle for the introduction of subtle, imaginative
questions which generated novel replies. Although the speculative
responses did not replace, or cause the overthrow of, the Aristotelian world
view, they did challenge some of its fundamental principles and
assumptions. For some four centuries many were made aware that things
might be quite otherwise than had been dreamt of in Aristotle's
philosophy.
5. Combes and Ruello 1967, pp. 232.66-8 and 234.6-9.

27
THE O X F O R D CALCULATORS
The identity and the writings of the Calculators or 'Mertonians'

In the second quarter of the fourteenth century a collection of works was
produced at Oxford whose joint impact on European natural philosophy
lasted well into the sixteenth century. The works at the core of this
collection are Thomas Bradwardine's De proportionibus velocitatum (1328),
William Heytesbury's Regulae solvendi sophismata (1335), and Richard
Swineshead's Liber calculationum (usually dated ca. 1350, but probably
earlier). Other treatises were linked with these three through common
interests and approaches in logic, in mathematics, and in physics or natural
philosophy. Among the most closely linked works were Richard Kilving
ton's Sophismata, Walter Burley's De primo et ultimo instanti and Tractatus
primus et secundus de formis accidentalihus (the Tractatus secundus is known
better as De intensione et remissione formarum), Richard Billingham's Con-
clusiones, Heytesbury's Sophismata and the Probationes Conclusionum of his
Regulae, Roger Swineshead's De motibus naturalibus, John Bode's treatise
on the sophisma Ά est unum calidum' and others, the anonymous Sex
inconvenientia, and John Dumbleton's Summa logicae et philosophiae naturalis,
not to mention many treatises on the usual subjects of fourteenth-century
logic — supposition, consequences, obligationes, insolubilia, etc. — or com
mentaries on Aristotle's physical works and other set books of the medieval
curriculum.
Since many of the most famous authors of these works, including
Bradwardine, Heytesbury, Richard Swineshead, Burley, Dumbleton, and
possibly also Bode, had been fellows of Merton College, Oxford, some
recent historians of science call this group of authors the Merton School,
although there is little contemporary evidence that they were called
Mertonians. 1 Contemporary and slightly later Continental philosophers
1. See, e.g., Günther 1923, which has a chapter on the 'Merton School of Astronomy'. Also Maier
1952,p. 265; Clagett 1959, pp. 199-329; Molland 1968a; Weisheipl 1968, i9ô9;SyIla 1971; Bottin
1973, P· 127. The appropriateness of the term is questioned by Weisheipl 1959, pp. 439-40. One
instance of a fourteenth-century identification with Merton is Thomas Bradwardine's addressing
The Oxford Calculators 541
tended to call the members of the group simply 'Anglici' or 'Britannia',

doubdess associating them with the larger group of British logicians whose
contribution to logic was considered noteworthy. 2 By the late fifteenth
and early sixteenth centuries the members of the narrower group were
being called Calculators, with Richard Swineshead known as 'the
Calculator'.3 In what follows, I will treat these authors insofar as the label
'Oxford Calculators' fits them.*
The Calculators considered as natural philosophers

The fifteenth- and sixteenth-century Continental authors who quoted the
arguments of the Oxford Calculators did so most often in the context of
natural philosophy.* Consequently, it was twentieth-century historians of
science such as Pierre Duhem, Anneliese Maier, and Marshall Clagett,
looking for the scientific background of Leonardo da Vinci and Galileo,
who rekindled an interest in the work of the Calculators.6 Seen from this
perspective, the Oxford Calculators have been credited with distinguish
ing between kinematics and dynamics, with developing a concept of
instantaneous velocity, and with the proof of the so-called mean speed
theorem for uniformly accelerated motion. 7 Bradwardine's De propor-
his De causa Dei contra Pelagios 'ad suos Mertonenses', but such a term of address is, of course, no
evidence of the conscious association of a particular school of thought with Merton.
2. For example, in many of the manuscripts of the relevant works the author is identified as
'anglkus''. Richard de Bury in his Philobibton, in referring to the University of Paris, says;
'Involvunt sententias sermonibus imperitis, et omnis logicae proprietates privantur; nisi quod
Angiicanas subtilitates, quibus palam detrahunt, vigiliis furtivis addiscunt', p. 89 in Thomas 1888,
quoted by Gilbert 1976, p. 232. Gaetano di Thiene, in hi» De reactione (1491), says, 'Sed haec
oppositio est britannica. Et melius est tenere opinionem realium.' Thus he seems to associate the
British with nominalism.
3. Swineshead is called the Calculator by, e.g., Angelo of Fossambruno in the early fifteenth
century, as well as by Marliani, Scaliger, Leibniz, and others later. Cf. Clagett 1959, pp. 649,659;
Duhem 1906-13,111, pp. 497-8.
4. This label seems preferable to 'Merton School' because not all scholars whose work seems to be
associated with that of the group can be shown to ha ve been members of Merton College, though
the core certainly was there. I am here retaining 'Oxford Calculators' rather than 'English
Calculators', which would have much more early support, not in order to exclude the possibility
of Calculators at Cambridge, but rather because I want to exclude members of the English nation
at Paris who were never at Oxford.
5. I am thinking of such authors as Blaise of Parma, Messinus, Angelo of Fossambruno, Jacopo of
Forli, Paul of Venice, Gaetano di Thiene, Giovanni Marliani, John Dullaert, Luiz Coronel, Juan
Celaya, Alvarus Thomas, Domingo de Soto, Hieronymus Picus, et al. See Clagett 1941 ; also 1959,
pp. 645-59; Wallace 1969,1971; Lewis 1975,1976.
6. See Pierre Duhem 1906-13, Anneliese Maier 1952 et seq., and Marshall Clagett 1950, 1969.
Others who have studied the Calculators from the perspective of natural philosophy or mathe
matics arc Michalski 1922 etc.; Molland 1967 etc.; Murdoch 1957etc.; Sylla 1970etc.; Thorndike
1934; Weisheipl 1956 etc.; and Wilson 1956.
7. Clagett 1959, p. 205. Cf. Murdoch 1974a, p. 55η.
tionibus (On the Ratios of Velocities in Motions), Roger Swineshead's De

motibus (On Natural Motions), and Dumbleton's Summa (Compendium of
Logic and Natural Philosophy) certainly are primarily works of natural
philosophy. To the modern ear, the name 'Calculators' tends to link the
bearers of the name with mathematics and with science. But if later scholars
have noticed and valued the work of the Oxford Calculators for its physical
and mathematical content, nevertheless, within the fourteenth-century
Oxford academic context, the work of the Calculators probably arose not
in the guise of recognised mathematics or natural philosophy, but within
the standard practice of logical disputations. The specific characteristics of
the work of the Oxford Calculators are much easier to understand if the
work is seen in this disputational context.
The disputational context for the work of the Calculators

The three pedagogical techniques most commonly used in medieval uni-
versities were the lecture, the question, and the disputation. 8 The student
was, first of all, required to hear certain textbooks read, both in cursory
lectures, most often given by bachelors of arts, and in ordinary lectures
given by masters of arts, the latter lectures including the raising and
answering of pertinent questions about the text. The main subjects covered
in the Faculty of Arts at Oxford were the seven liberal arts and the three
philosophies: natural, moral, and metaphysical. The fourteenth-century
Oxford undergraduate heard lectures on grammar, logic, and natural
philosophy or the quadrivium (arithmetic, geometry, astronomy, and
music). The bachelor of arts heard lectures mainly on philosophical topics
and lectured cursorily on logic and natural philosophy. Ideally, the master
of arts was expected to be able to teach all ten disciplines.9
8. I have based the following survey of the Oxford curriculum mainly on Gibson 1931 and
Weisheipl 1964a. Lectures and disputations are clearly distinct - in the former a single lecturer
reads and expounds the text, and in the latter several people debate with one another. Questions
are closely related to both lectures and disputations. They may be integral parts of lectures, merely
mimicking the form of disputation, they may be subjects of independent disputations, or, thirdly,
they may arise from lectures, become the subjects of disputations, and then receive definitive
answers in later lectures (Wallerand 1913, p. 20).
9. Logic dominated the undergraduate curriculum, as is clear from the statutes of 1268 (Gibson 1931,
p. 26) and 1409 (Gibson 1931, p. 200). The 1268 statutes, but not those of 1409, also require
attendance at lectures on the Physics, De anima, and De generatione et corruptione. Bachelors studied
mainly the three philosophies, as is shown by the requirements for inception in the statutes prior
to 1350 (Gibson 1931, pp. 32—3) and in 1431 (Gibson 1931, pp. 234-5). Both sets of statutes
require attendance at lectures on the libri naturales. Inter alia the 1431 statutes specify three terms to
be spent on the Physics or the De caelo, the De anima, or other books of natural philosophy.
It seems clear that in 1431 there was greater effort to cover all the seven liberal arts and three
philosophies than there had been earlier. The statutes for determiners and inceptors before 1350
But besides hearing lectures with or without questions, the student was
required to assist at disputations. After two years of hearing lectures, the
undergraduate might first serve as 'opponent' and then as 'respondent' in
disputations de sophismatibus (1268) or in parviso (before 1350,1409). When
he had done this for a year, he might be admitted to respond de quaestione.
And when he had responded de quaestione for at least part of a year, he
might be given permission to 'determine' in the school of his or some other
master during Lent.10 At this point he became a bachelor of arts.
do not, for instance, mention any books of music or metaphysics. In 1268 natural philosophy was
required for determination, but nothing from the quadrivium, it apparently being possible to
postpone hearing the books of the quadrivium until the bachelor years. (See Gibson 1931, p. 33.
14-20, quoted above.) In 1409, on the other hand, at least the most elementary parts of the
quadrivium were required for determination, but nothing was required of natural philosophy
until inception. According to a statute of 1340, the bachelor was supposed to lecture on a book of
natural philosophy, such as De caelo. De anima, Meteora, De generatione et corruptione, or the Parva
naturalia; so he would have had to have heard at least one of those books earlier (see Gibson 1931,
P- 32).
(It would appear that at least in certain decades of the fourteenth and fifteenth centuries the
Physics could be omitted altogether, replaced by De caelo. Cf. Gibson 1931, p. 32.8-13; p. 200;
PP· 234·32-235·3·>
As far as the masters are concerned, it seems that at some periods they could choose the subjects
of their own lectures. (Cf. Weisheipl 1964a, pp. 149, 160.) But in later periods a certain
organisation was imposed to insure that there were ordinary lectures on all the books required pro
forma (Gibson 1931, pp. 235-6, for 1431).
In this more regulated situation a master might well be expected to change the subject of his
lectures from term to term. Similarly, a late statute for responses of bachelors, citing ancient
custom, supposes that they might respond in disputations concerning each of the ten disciplines in
turn (Gibson 1931, p. 247 for 1432). This statute is perhaps evidence that the ordinary solemn
disputations of the masters were linked to the disciplines or ordinary lectures. This is argued
further in the next two notes.
10. The statutes of 1268 oblige candidates for the determination 'publice de sophismatibus per an-
num integre ... rcspondisse' and in the summer before the Lent term of their degree 'de
questione rcspondisse' (Gibson 1931, p. 26). In the 1409 statutes they must have responded 'de
questione' before the Hilary term of their degree, and they must previously have been 'arciste
générales, parvisum interim fréquentantes, et se ibidem disputando, arguendo, et respondendo
doctrinaliter exercentcs' (Gibson 1931, p. 200). Disputations 'in parviso'(in a place out of doors)
as contrasted with 'in scolis' are mentioned also in statutes of ca. 1350 and 1607 (Gibson 1931,
P· 358).
A comparison of the 1268 and the 1409 statutes leads to the conclusion that the disputations in
parviso were either identical with the disputations de sophismatibus or else took their place.
Weisheipl (1964a, p. 154) says that disputations de sophismatibus were disputations on logic
whereas disputations de quaestione were on natural philosophy. I am inclined to believe that the
responses de quaestione are to be identified not by their subject matter as much as by their location;
in particular, I think that theresponsesde quaestione at issue may be the ones that occurred in the
schools. It seems likely that the regent masters giving ordinary morning lectures on required
books also held ordinary disputations on questions or problems arising from them, and that both
undergraduates and bachelors acted as the respondents in these disputations held in the schools,
usually in the afternoon. (Cf. Gibson 1931, pp. 192 and 194 for 1407 or before.) The advanced
undergraduate would respond de quaestione at the same ordinary or solemn disputations of the
masters at which bachelors were supposed to oppose or respond (Gibson 1931, pp. 235-6,247).
As a bachelor of arts the student would continue to oppose and respond

in disputations. At the end of three years he could be licensed to incept as
master of arts, participating in two disputations as part of inception, and,
after that, disputing on every 'disputable day' for the next forty days and
giving ordinary lectures with accompanying ordinary disputations as a
'necessary regent' for the next two years.11
The master might give a formal reply during a subsequent morning lecture (Gibson 1931, p. 56,
for before 1350).
Undergraduates who were admitted to determine were supposed to dispute only in logic
except on Fridays and special days (Gibson 1931, pp. 27,201—2). From the emphasis on logic in
determinations we may infer that in the disputation in parviso the emphasis was similarly on logic,
as indeed one statute indicates (Gibson 1931, p. 27). This would fit with the fact that in hearing
lectures undergraduates concentrated on logic and with the fact that even if the disputations in
parviso were not, as has been assumed, identical with the disputations de sophismatibus required in
1208, they nevertheless held the same place in the curriculum.
The main sources concerning disputations of bachelors of arts are from 1340 and 1431. The
statutes for 1340 state (Gibson 1931, p. 32): 'Quociens tenenlur artiste arguere et respondere ht
disputacionibus bachilariorum. Item ordinatum est quod arguât quilibet incepturus quater ad minus
puplice in disputacionibus magistrorum, et quod semel disputet vel respondeat quilibet in
disputacione generali bachilariorum facultatis predicte, et hoc pertinencia argumenta adducendo
tantummodo ad questionem vel problema quam vel quod eum contigerit disputare. (Quociens
artiste tenentur respondere magistris antequam incipiant.) Item, ordinatum est quod quilibet in-
cepturus, ante licenciant suam in artibus, respondeat bis ad minus magistris regentibus in
disputacionibus solempnibus, que non fuerint de quolibet, et hoc de questionibus vel semel de
questione et de problemate alias.'
There are at least two sorts of disputations mentioned here. First, there are the solemn
disputations of the masters, which are probably the disputations of single masters in their schools
with responding sophistae and bachelors, as argued in the previous note. At these ordinary or
solemn disputations the bachelor was, in 1340, expected to act as opponent at least four times and
as respondent at least twice, either both times de quaestione or once de quaestione and once de
problemate.
The second sort of disputation mentioned is that apud Augustinienses, or the general disputation
of the bachelors. It is supposed that questions or problems will be disputed there and these are
supposed to be announced at least three days in advance (Gibson 1931, p. 287). Apparently the
role of respondent was more popular than that of opponent. If there was disagreement over roles,
the senior bachelor was to be respondent (Gibson 1931, pp. 286-7, for before 1477). In 134611 was
ordained that each bachelor, when called upon, should take part in these general disputations of
the bachelors twice a year, once as opponent and once as respondent (Gibson 1931, p. 147).
The disputations connected with inception in arts were 'vesperies' and investiture; there were
formal differences between the disputations associated with the two occasions. See Gibson 1931,
pp. 36-9; Weisheipl 1964a, pp. i64-5;Little and Pelster 1934, pp. 44-52. These disputations seem
to have concerned mainly the quadrivium or one of the philosophies. Such disputations can be
found in MS London, Lambeth Palace 221, ff. 262—309, and in MS Oxford Magd. College 38,
ff. i6 v -48 v . See Gibson 1930 and 1931, pp. 643-7. The inceptor took the role of opponent while
those who had most recently incepted responded. (Gibson 1931, p. 38, for before 1350.)
For the disputations for forty days after inception there is a statute for before 1350 (Gibson
I93I.P-39).
In theology faculties there were, besides the disputations so far described, quodlibetal dispu-
tations. Most recent commentators believe that there were also quodlibets in the arts faculty (e.g.
Glorieux 1925-35, II, p. 19; Gibson 1931, p. xcv; p. 32.26; pp. 404, 406; Weisheipl 1964a,
pp. 182—5). ' think this has yet to be shown. Weisheipl 1964a, p. 182, reports that the editors of
Rashdall 1936 similarly doubt the existence of arts quodlibets at Paris before a late date; I, p. 460,
n. 2. It should, of course, not be forgotten that in the extant quodlibets of Oxford theologians arts
As I interpret the statutes, the ordinary disputations of the masters

which, along with the ordinary lectures, formed the backbone of uni-
versity instruction were linked to the lectures and open to participation by
men from all levels of the university. The disputation was held in a school
by a master of arts who, at the end or on a subsequent day as part of his
lecture, 'determined' or gave a final answer concerning the subject of the
dispute. Beyond these ordinary disputations held by masters in their
schools there were other sorts of disputations, typically designed for stu-
dents at a particular level.
Disputations on sophismata
Although the extant university statutes and other evidence for fourteenth-
century Oxford are frustratingly full of gaps, it appears (from information
given in more detail in the notes) that at Oxford there were disputations de
sophismatibus, which were understood mainly as aids to learning logic,
which were held (or called) in parviso, and which were primarily intended
for advanced undergraduates, who were called sophistae because of their
participation in these disputations. 12 Masters or perhaps bachelors of arts
attended and supervised these disputations, but the real work of the
disputation was done by the undergraduate respondent. If the master made
a determination of the sophisma at all, the point of the exercise remained
the activity of the undergraduate. It appears likely that the most typical
works of the Oxford Calculators, especially Heytesbury's Regulae and
Sophismata, were connected with these undergraduate disputations in
parviso.
Sophismata at Paris and at Oxford

Quite a few sophismata determined by named masters of arts survive from
the University of Paris. ' 3 This fits the evidence that at Paris sophismata
were disputed in scolis. Many of these sophismata, even in their written
questions are treated as well as more obviously theological ones. But surely theological quodlibets
were not an arena in which sophismata were determined. A consideration of all these various
types of disputation leads to the conclusion that disputation of sophismata at Oxford took place
mainly in parviso or also in ordinary disputations of the masters linked to the ordinary lectures on
logic, but not usually elsewhere. In the present state ofour knowledge it seems most likely that the
works of the Oxford Calculators are to be linked with disputations in parviso.
12. Cf. Weisheipl 1964a, pp. 177-81. One cannot claim an absolute identity between disputations de
sophismatibus and disputations in parviso. There were certainly disputations de sophismatibus not in
parviso - for instance in the 'determinations' of new bachelors in Lent and probably also in
disputations connected with the ordinary lectures on logic. Later there may have been dispu-
tations in parviso not on logic. I am arguing that in the period of the Oxford Calculators most
disputations in parviso would have been de sophismatibus and vice versa.
13. See Grabmann 1940a.
form, show clear evidence of the live debate in which they originated, and
some are not really sophismata but merely disputed questions. 14 The
sophismata of the Oxford Calculators, on the other hand, do not contain so
much evidence of live debate and are almost always real sophismata and
not disputed questions. 15 I assume that the works of the Oxford Calcula-
tors were composed by men who were masters of arts and not students, but
I doubt that the sophismata of Kilvington and Heytesbury were publicly
determined by them in live debate, although Kilvington's sophismata are
usually determined by him in their literary format. Rather than being
records of live debate, the works of Kilvington, Heytesbury, and Richard
Swineshead seem to have originated as written works intended primarily
to provide ammunition for future undergraduate disputes with little em-
phasis on a particular magisterial determination. In its literary format,
Heytesbury's Regulae solvendi sophismata may be seen as developing a set of
theories which are then extended and applied by means of sophismata. 16 In
its historical genesis, however, the order was most likely the other way
around, as the title of the work would indicate: posing and solving
sophismata was the live activity in the service of which the rules were
compiled. Perhaps with the help of such works, the undergraduate dispu-
tations in fourteenth-century Oxford were carried on with a level of
energy and expertise high enough to merit outside attention. 17
The alleged distinction between 'physical' and 'logical' sophismata

This, then, is the disputational context in which the core works of the
Oxford Calculators should be viewed. ' 8 But, if so, why, we might ask, are
14. See note 10 above. Also Grabmann 1940a and Pinborg 1975b.
15. Pinborg 1979, pp. 28-9.
16. Cf. Wallerand 1913, pp. (29)-(32). Wallerand distinguishes between the use of sophismata as
exercises applying a theory and their use to extend theory. Murdoch 1974, pp. 63-70; Murdoch
1975e, pp. 304-7·
17. Itis striking that theforma for determination in 1409 is introduced by a statement that Oxford has
received many honours for the determinations of its bachelors and that the logical subtlety for
which Oxford is world-famous receives its greatest increase from exercise in determinations
(Gibson 1931, p. 199): 'Quia per sollennes determinaciones bachillariorum in facultate arcium
nostra mater Oxonie universitas, et precipue ipsa arcium facultas, multipliciter honoratur, ac mira
seiende logicalis subtilitas, qua prefata mater nostra supra cetera mundi studia dinoscitur actenus
claruisse, per fructuosum exercicium in eisdem potissimum suscipit incrementum, utile et
expediens visum est magistris ut certa forma provideretur, sub qua bachitlarii sufficientes et
ydonei, exclusis indignis, ad determinacionis actum forent admittendi, modumque et con-
diciones cxpnmere, quos in suo introitu, processu et exitu debeant observare . . . ' For the fame of
Oxford logic at Paris, see Gilbert 1976; Coleman 1975; Murdoch 1978b.
18. Cf. Duhem 1906-13, III, pp. 441-51; Pedersen 1953, p. 141: 'The result... was a new kind of
literature, the so-called sophismata, viz. tracts containing huge collections of calculations in the
most various fields with Suiseth's own book as the most famous example.'
physical as well as logical subjects involved in these rules and sophismata?

What are we to make of the so-called 'physical sophismata'?19 For all the
physics used in these works, it appears that they were not intended to be
part of the student's education in natural philosophy, though they might
use philosophical knowledge already gained. As will be clear from the
more detailed description of the works to follow, in Kilvington's or
Heytesbury's work one cannot separate 'physical sophismata' from logical
or grammatical sophismata; physical concepts are Used in traditional logical
sophismata where they might not have been expected.20 And wherever
natural philosophy or calculations are used, they are used for the sake of
the disputation, either to set up the puzzling results typical of sophismata
or else to unravel such results once they have been set up. Just as one
could derive and explain counterintuitive results using the techniques of
syncategoremata and supposition theory, so one could do so using
calculationes.21
This at least would be the rationale for introducing calculations. We are
not forced to conclude that there is no natural philosophy or mathematics
in the works of the Oxford Calculators. It may have been the case that
because the general disputations of the sophistae in parviso were somewhat
separated from the ordinary lectures there was a greater chance there to
introduce material from non-logical disciplines than would have been the
case in an ordinary disputation covering questions arising from the reading
of a particular book of logic. Perhaps sophistae attracted by the more
advanced topics of natural philosophy and mathematics were tempted to
smuggle these topics into disputations in parviso originally meant as logical
exercises. But even if this is the correct explanation of the appearance of
natural philosophy and mathematics in sophismata, the work would never-
theless have been done in the context and under the influence of logical
disputations.
19. For discussions of physical sophismata, see Duhem 1906-13, III, pp. 441-54; Michalski 1926,
pp. 59-61; Maier 1952, pp. 264-70; Weisheipl 1964a, p. 178, n. 96; Bottin 1973, p. 126; Murdoch
I975>.P· i°°.
20. Wilson 1956, pp. 21 -5. Physical subject matter could seem to fall under scientiae sermocinales when
the emphasis was put upon an analysis of the semantics of the propositions comprising science.
John Murdoch has drawn attention to the 'second intentional' or 'metalinguistic' approach of
fourteenth-century natural philosophy in a recent series of important articles which give ex-
amples of how this occurred in detail and explores various ways in which it may have come about.
See Murdoch 1974a, pp. 60-2, 68-70, 73-4, 100, 105, 111-12; Murdoch 1975a, pp. 287-8,
303-7; Murdoch 1979 and his other recent articles, some still in press at the time of this writing.
21. Alvarus Thomas 1509 provides some vivid descriptions of how calculations might be involved in
disputations. See Duhem 1906-13, III, pp. 537, 541—3.
Richard Kilvington's Sophismata

Kilvington's Sophismata exemplifies as perfectly as one might desire how
close disputations de sophismatibus might come to questions of mathematics
and natural philosophy without ever leaving the confines of logic and
semantics proper and without bringing in calculations to any extent. 22 It
seems likely that this work was composed before Bradwardine's De pro
portionibus in 1328.23 Although not bringing in mathematics or natural
philosophy directly, Kilvington's work was so close to it, so ready for it,
that it was only a very small step for Heytesbury to bring in mathematics
and physics explicitly in his sophismata. Whether this happened primarily
because of a felt need within the sophismata tradition or because of the
impressiveness of Bradwardine's achievement (after Kilvington's?) is not
clear, but both factors must have had an important influence.
Kilvington's sophismata are individually short and topically unified in
comparison with Heytesbury's, but they are organised and connected in a
way that allows a sustained examination of certain problem areas. The first
eleven sophismata all involve combinations of beginning, ceasing, and
comparisons of degrees of whiteness. In the twelfth sophisma the problems
of traversing a space are introduced and these are treated, sometimes in
combination with problems of beginning and ceasing, of motive power,
22. I have based my study of Kilvington on the forthcoming edition of his Sophismata prepared by
Norman and Barbara Kretzmann. I would like to thank them for generously providing me with
the typescript of their edition. Wilson 1956, pp. 163-8, gives a list of the sophisma sentences with
some indications of parts of the arguments. In Wilson's numbering, Sophisma 30 should be
omitted and the subsequent numbers decreased by one. In general I here follow Kretzmann's
terminology, according to which the typical sophisma has the following parts: (1) the 'sophisma
sentence' which is frequently called the sophism in recent literature; (2) the 'hypothesis' or
hypothetical case in light of which the sophisma sentence is to be interpreted; (3) the proof of the
sophisma sentence; (4) the disproof of the sophisma sentence; (5) the solution of the sophisma; (6)
the reply to arguments for the opposing side. All these parts together are called the 'sophisma',
retaining the medieval term to guard against the too easy assumption that medieval sophismata
are all considered sophisms or sophistical in the modern sense. While in many cases the sophisma
sentences are strange or bizarre, often they are not - for instance Ά begins to be true' (Kilvington,
Sophisma 16) or 'Socrates does not move faster than Plato' (Kilvington, Sophisma 32). What does
characterise the sophismata of the Oxford Calculators more universally is that there are plausible
proofs and disproofs of the same sophisma sentence, a plausible proof of something that appears
obviously false, or a plausible disproof of something that seems obviously true.
23. The Sophismata shows no familiarity with Bradwardine's work - for instance it does not draw a
distinction between total (or average) and instantaneous velocity where such a distinction would
have been most useful. Kilvington was a bachelor of theology by 1335. If we assume that seven
years of study of theology were required for that degree, Kilvington's 'necessary regency' in the
arts would have occurred ca. 1326-8 or before. The Sophismata was cited by Adam Wodeham in
1330-1. See Courtenay 1978, pp. 86-9. Courtenay suggests that Kilvington became a bachelor of
theology between 1332 and 1335. It should be noted, however, that Kilvington's Questions on the
Physics are said to take up Bradwardine's dynamics (Maier 1964, p. 253). For a more detailed
consideration of the historical questions surrounding Kilvington's work, see the Introduction to
the forthcoming edition of his Sophismata by Norman and Barbara Kretzmann.
or of comparisons of velocities, in Sophismata 12-19, 27-35, and 42-4.

The intermediate sophismata deal with generation and corruption (20,
39-41), whitening involving non-uniformity of whiteness (21-3, 26),
division of a body (24- 5), and strength (36-8). The last four sophismata are
somewhat separate and deal with problems associated with the verb 'to
know'.
What is remarkable about these sophismata is that all but the last four
(and Sophisma 17 which treats an issue subsidiary to 16) deal with problems
of motion (or change) and yet physical issues concerning motion are very
rarely raised. Thus the first sophisma sentence is 'Socrates is whiter than
Plato begins to be white', for the interpretation of which it is supposed that
Socrates is white in the highest degree and that Plato now is not white and
begins to be white.24 In the whole discussion of the sophisma it is tacitly
assumed that the whiteness Plato takes on increases continuously from zero
degree as an extrinsic limit, but nothing is said (a) about how whiteness is
measured or determined when it may be non-uniformly distributed within
a body, or (b) about what theory of the intension and remission of forms
is assumed - among the many theories hotly debated in the fourteenth
century.25
To put the matter in modern terms, the problem at issue in the first
sophisma arises from the fact that the set of degrees that Plato takes on has
no least member and no intrinsic bound on the smaller side, but rather
contains degrees lower than any given degree. If it is admitted that Socrates
is whiter than Plato begins to be white, then it would seem to follow that he
is infinitely whiter proportionally, since there is no intrinsic lower bound
to Plato's whitenesses: given any ratio between Socrates' whiteness and
some degree of whiteness Plato takes on, there is a greater ratio between
Socrates' whiteness and a smaller whiteness that Plato takes on. But Plato
begins to have not zero whiteness but some whiteness, and so it seems to
follow that if Socrates is infinitely whiter than Plato then Socrates must be
infinitely white, which is false.26
24. This sophisma has been discussed several times in the recent literature. See Bottin 1973 (Latin text
of the initial sophismata); Kretzmann 1977 (English translation of thefirstsophisma); Knuuttila
and Lehtinen 1979.
25. For these theories see Sylla 1973.
26. As Kilvington recognises in his reply to the sophisma, this is a good argument except for the last
sentence. 'Socrates is infinitely whiter than Plato begins to be white' is a true proposition if
'infinitely' is taken in the syncategorematic sense. Then the sense of the proposition is simply that
given any ratio between Socrates' whiteness and some degree of whiteness Plato takes on, there is
a greater ratio between Socrates' whiteness and a degree of whiteness that Plato takes on earlier.
This does not entitle one to conclude that there is an infinite ratio between Socrates' whiteness and
some given degree of whiteness taken on by Plato, and so it does not entitle one to conclude that
Socrates is infinitely white.
Kilvington responds that the sophisma sentence is true, and concedes

that Socrates is infinitely whiter than Plato begins to be white; but he
denies that Socrates is infinitely white on the grounds that there is no given
first and most remiss degree of whiteness that Plato will have. 2 7
In reply to the second sophisma - 'Socrates is infinitely whiter than Plato
begins to be white' - Kilvington again accepts the sophisma sentence, but
denies that it implies that Plato begins to be infinitely less white than
Socrates is now, again on the grounds that there is no first degree of
whiteness with which Plato begins to be white. 2 8 The ostensible basis for
the distinction between the two accepted and rejected propositions is word
order. Expressions standing first in a proposition, before any syncategore-
matic word, refer to determinate supposita, such as given degrees of
whiteness in Socrates or Plato. But expressions occurring later in the
proposition or after some syncategorematic word may refer indetermi-
nately to a whole set of supposita, such as all the degrees of whiteness Plato
takes on. 2 9 Thus Kilvington has solved these sophismata not by using
physical theory or mathematics, but rather by using the standard tech-
niques of terminist logic.
A second example of the logical orientation of Kilvington's sophismata
can be found in the sophismata about traversing a distance. The main issue
raised concerning traversing a distance A is whether one should say that
Socrates traverses it (a) at any time he is in the process of traversing A, or (b)
only when he has completely traversed A. Against the first interpretation is
the consequence that Socrates traverses A infinitely many times, since
infinitely many times he will be in the process of traversing A; but Socrates
traverses A only once. Against the second interpretation is the consider-
ation that when Socrates has traversed A completely, then he is no longer
traversing it; so it hardly makes sense to say that he traverses A only when
he has traversed A completely.
In Sophisma 13 - 'Socrates will traverse distance A' - Kilvington
concludes only by distinguishing the two senses of 'Socrates will traverse
distance A' - either it is equivalent to 'Socrates will be in the process of
traversing distance A', in which case the sophisma sentence is true, and it is
true that Socrates will traverse A infinitely many times, or it is equivalent
to 'Distance A will have been traversed by Socrates', in which case the
27. Bottin 1973, pp. 139-40; Kretzmann 1977, p. 15.

28. Bottin 1973, pp. 140-1.
29. For this standard use of supposition theory see the discussions of supposition and syncategoremata
elsewhere in this volume.
sophisma sentence will not be true before the last instant of the time period
during which Socrates will move over distance A. 3 0
The replies to Sophisma 12 - 'Socrates has traversed distance A' - are
much more complicated. The main question is whether or not the instant
at which Socrates first reaches the end of distance A (call it instant C) is the
first instant at which he has traversed distance A. Kilvington's preferred
answer is that at C it is not true that Socrates has traversed A, because in that
case he would have ceased moving over A, which is false (he has not ceased
moving over A because at C he is still touching the end of A). 3 1 From the
fact that distance A has been traversed by Socrates it does not follow that
Socrates has traversed distance A, but only that he begins to have traversed
A (in the sense that at C he has not traversed A but immediately after C he
will have traversed A). 3 2 An alternative reply, which Kilvington finds less
satisfactory, interprets 'Socrates has traversed A' as equivalent to 'Socrates
was in the process of traversing A'. 3 3
Thus in his preferred reply Kilvington interprets 'Socrates has traversed
A' as equivalent to Ά was previously traversed by Socrates'. A third
alternative which Kilvington mentions but does not develop would be to
interpret 'Socrates has traversed A' as equivalent to Ά has been traversed
by Socrates', which would leave open the possibility that A has just now
been traversed and was not previously traversed by Socrates.3* Given
30. 'Socrates pertransibit A spatium. . . . Ad sophisma dicitur distinguendo de isto termino "per-
transibit". Uno modo sic exponitur: "Socrates pertransibit A spatium" id est, "Socrates erit in
pertranseundo A spatium." Et sic est sophisma verum. Et ulterius conceditur conclusio ultima,
quod infinities isto modo pertransibit Socrates A spatium, quia infinities erit Socrates in per
transeundo A spatium. Alio modo potest sophisma exponi sic: "Socrates pertransibit A spatium"
- id est, "A spatium erit pertransitum a Socrate." Et ita loquendo ante C non pertransibit Socrates
A spatium' (ed. Kretzmann, forthcoming).
31. 'Socrates pertransivit A spatium... Ad sophisma dicitur quod est falsum, quia si Socrates
pcrtransivit A spatium et nullam aliam partem Β sparii Socrates pertransivit, ut suppono - sit Β
quoddam totum spatium cuius medietas est A spatium - igitur Socrates desinebat moveri super A
spatium, quod est falsum' (ed. Kretzmann, forthcoming).
32. 'Ad probationem dicitur quod non valet ilia consequentia; "A spatium est pertransitum a Socrate;
igitur Socrates pertransivit A spatium." Sed bene sequitur "A spatium est pertransitum a Socrate;
igitur Socrates pertransivit A spatium vel incipit pcrtransivissc A spatium"' (ed. Kretzmann,
forthcoming).
33. 'Aliter tamen dicunt quidam concedendo quod Socrates pertransivit A spatium. Et ulterius,
quando arguitur "Socrates pertransivit A spatium; igitur Socrates prius pertransivit A spatium",
concedunt conclusionem. Et ulterius concedunt quod Socrates incepit pcrtransirc A spatium
quando Socrates incepit moveri super A spatium, et quod per totum tempus quo Socrates
movebatur super A spatium fuit haec propositio vera: "Socrates pertransivit A spatium"' (ed.
Kretzmann, forthcoming).
34. 'Alio modo exponitur iste terminus "pertransivit" per verbum passivum sic: "Socrates per
transivit A spatium" - id est, "A spatium est pertransitum a Socrate." Et sic exponendo istum
terminum "pertransivit" non sequitur "Socrates pertransivit A spatium; igitur Socrates prius
Kilvington's preferred reply, Socrates will not begin to traverse A any

sooner than he will begin to have traversed A previously (neither will be
true until immediately after instant C). O n the third alternative reply
Socrates will traverse A when he will not have traversed it previously - at
instant C Ά has been traversed by Socrates' will be true, but Ά was
previously traversed by Socrates' will not be true. Nevertheless, even on
this alternative interpretation both propositions will begin to be true at C;
the first will be true at C and at no time before (one of the two standard
expositions of'begins'), and the second will not be true at C but at any time
thereafter (the other standard exposition of'begins'). All of this is said in
reply to Sophisma 14. 3 S
Having explored in Sophismata 12-14 the results of assuming that
Socrates traverses a distance either (a) when he is in the process of traversing
it or (b) when he has completely traversed it, Kilvington, in Sophisma 15,
considers the results of assuming (c) that a distance is traversed when more
than half of it is traversed. These results need not detain us here, but the
overall circumstances are worth noting. The issues Kilvington considers in
Sophismata 12-15 are very close to the Oxford Calculators'later common
concern with the proper measures of local motion with respect to effect. 36
There is little mathematical or calculatory about Kilvington's procedure
except in Sophisma 15 where he argues that if something is traversed when
more than half of it is traversed, it follows that if distance Β is traversed,
then C, equal to three-halves B, will be traversed. But if C is traversed, then
pertransivit A spatium," quia non sequitur "A spatium est pertransitum a Socrate; igitur Λ
spatium est prius pertransitum a Socrate" ' (ed. Kretzmann, forthcoming).
35. 'Socrates incipiet pertransire A spatium, et Socrates incipiet pertransivisse A spatium, et non prius
incipiet pertransire A spatium quam incipiet pertransivisse A spatium. Et hoc probatur per
expositionem istorum terminorum "pertransire" et "pertransivisse". Nam non prius incipiet
Socrates esse in pertranseundo A spatium quam incipiet fuisse in pertranseundo A spatium, quia
sine tempore medio postquam Socrates incipiet moveri super A spatium erit in pertranseundo A
spatium, et sine medio postquam Socrates incipiet moveri erit verum quod Socrates fuit in
pertranseundo A spatium. Ideo in hoc sensu et secundum hanc expositionem concedendum est
sophisma. Sed alio modo exponendo eosdem terminos sophisma est falsum. Quia prius incipiet A
spatium esse pertransitum quam debet fuisse pertransitum, quia prius erit verum A spatium esse
pertransitum quam A spatium fuisse pertransitum. Quod patet quia in primo instanti in quo
motor deveniet ad Β terminum A spatii erit haec propositio vera: "A est pertransitum." Et tunc
non erit haec propositio vera: "A fuit pertransitum", posito quod A non erit pertransitum nisi
semel ante C instans. Unde licet concedenda sit haec propositio "Prius incipiet A spatium esse
pertransitum quam debet fuisse pertransitum", haec tamen propositio est neganda: "Prius incipiet
A spatium esse pertransitum quam incipiet fuisse pertransitum", quia in C incipiet A spatium esse
pertransitum, exponendo li "incipit" per positionem de praesenti et remotionem de praeterito, et
in C incipiet A spatium fuisse pertransitum, alio tamen modo exponendo li "incipit" - scilicet, per
remotionem de praesenti et positionem de futuro' (ed. Kretzmann, forthcoming).
36. Cf. Clagett 1959, pp. 208-9.
D, equal to 2B, will be traversed since C is more than half D, etc. 3 7 But
with or without this small foray into calculations on Kilvington's part it is
obvious how, beginning from Kilvington's logico-semantical approach,
one could be drawn by progressive stages into calculations.
Bradwardine's De proportionibus
Although this work of Bradwardine's does not represent the Oxford
calculatory tradition at its most typical, it was highly influential in that
tradition.38 Bradwardine devoted thefirstmain section of his treatise to an
exposition of a mathematical theory of ratios. This theory was based on
Book V of Euclid's Elements, but it took advantage of a theoretical tradition
in mathematics that treated ratios as sui generis and entirely different from
fractions in order to develop a system of operations on ratios different from
the normal system of operations on fractions.39
In the second and third chapters of the work, Bradwardine used this
system of ratios to defend a new theory of how forces, resistances, and
velocities are to be correlated in local motions.*0 Bradwardine's new
theory became the dominant one in the succeeding period, being rejected
primarily by those philosophers who rejected his distinction between ratios
and fractions.*1
In the fourth chapter Bradwardine argued that velocities of rotation
should be measured by the speed of the fastest moved point. 4 2 This attempt
to establish a measure for rotations soon blossomed in the work of
Heytesbury and others into attempts to find measures of all sorts of local
37. Ά spatium incipit esse pertransitum. Posito quod A spatium dicatur pertransitum quando maior
pars eius fuerit pertransita, et sit A spatium non pertransitum quando maior pars eius fuerit non
pertransita. Et ponatur quod aliquid incipiat moveri super A spatium. Tunc probatur sophisma
sic. A spatium non est pertransitum ct erit pertransitum, et nullum tempus erit antequam A
spatium erit pertransitum; igitur A spatium incipit esse pertransitum. Consequentia patet et maior
similiter. Et minorem probo, quia si aliquod tempus erit antequam A spatium erit pertransitum,
sit igitur, gratia exempli, quod hora erit antequam A spatium erit pertransitum. Sed probo quod
non, quia lapsa medietate illius horae, verum erit quod A est pertransitum. Quod probo, quia
lapsa medietate illius horae, verum erit quod aliqua pars A est pertransita. Sit igitur quod Β pars
tunc sit pertransita. Tunc sic. Β est pertransitum, et Β est plus quam medietas C; igitur per casum
C tunc erit pertransitum - posito quod C sit unum compositum ex B et alia parte A aequali
medietati B. Et per consimile argumentum D erit tunc pertransitum - posito quod D sit duplum
ad B. Et sic arguendo, sequitur quod lapsa medietate illius horae praedictae, totum A erit
pertransitum' (ed. Kretzmann, forthcoming).
38. The full text of the work is in Crosby 1955. See also Murdoch 1969, pp.- 225-33.
39. See Molland 1968a and 1978, pp. 150-60. 1 have written a paper which tries to trace these
traditions for the Festschrift for I. B. Cohen (forthcoming).
40. See Weisheipl's contribution to this volume.
41. See Clagett 1959, p. 443.
42. Crosby 1955, pp. 128-33, and Clagett 1959, pp. 215-16, 220-2.
motions, augmentations, and alterations.* 3 Thus if Bradwardine's con-

temporaries were tempted by sophismata like Kilvington's to want to
introduce mathematical physics into their sophismata, Bradwardine pro-
vided them with excellent starting materials.
Burley's De primo et ultimo instanti and De intensione et remissione formarum

Although Burley's works were used by Italian philosophers of the fifteenth
and sixteenth centuries as if they belonged to the same universe of discourse
as the works of Heytesbury and Swineshead, it can be seen that, from a
fourteenth-century point of view, they represent stages leading into the
calculatory tradition rather than full-fledged parts of it. Burley, like
Kilvington, very rarely calculates. Nevertheless, for at least two reasons,
Burley's work became very important to the calculatory tradition.
First of all, Burley's question (delivered at Toulouse) De primo et ultimo
instanti (On the First and Last Instant) took the existing logical tradition
of discussing the syncategorematic words 'begins' and 'ceases', a tradi-
tion which had already incorporated some Aristotelian physics in the
thirteenth-century works of Peter of Spain and William of Sherwood,**
and connected it even more inextricably with physics. Henceforth, if one
wanted to know, for instance, whether in a given case of beginning there
would be a last instant of non-existence or a first instant of existence, it was
clear that one ought to look at the underlying physics of the situation to
find an answer. Whereas the earlier tradition had distinguished between
permanent entities for which there would be a first and no last instant of
existence and successive entities (like motion) for which there would be
only a last instant of non-existence in beginning and only a first instant of
non-existence in ceasing, Burley broke down the category of permanent
entities into many sub-categories, e.g. permanent things having an in-
divisible degree of perfection (for which there would be a first and last
instant of existence) versus permanent things existing within a range or
latitude of degrees (for which there would be a first and no last instant of
existence) .*5
Secondly, Burley's treatises on the intension and remission of forms
were important to the calculatory tradition because they offered a well-
43. Cf. Clagett 1959, Ch. 4-6; Wilson i960, Ch. 4.

44. Cf. Wilson i960, Ch. 2; Kretzmann 1976; Murdoch 1979; and Murdoch forthcoming a. A
preliminary edition of the Burley question has been published (Shapiro 1965).
45. Shapiro 1965, pp. 164-6.
developed theory of alteration in terms of the succession of forms, opposed

to the more commonly accepted addition theory of alteration.*6 These
treatises also gave clear expositions of a third theory, the admixture theory,
which was sometimes combined with the addition theory. The com-
petition among these three theories left open a wide range of possible
assumptions concerning qualities which could be exploited by the
Calculators. On Burley's theory a body that was hot would at each point
have some degree of heat and no degrees of cold. On the admixture theory
the sum of hot and cold at any given point of a body always equalled some
maximum degree (usually taken as eight) with a hot body having more hot
than cold and a cold body more cold than hot. The Calculators, who usually
hold the addition theory, vacillated inconsistently between assuming that
there was one quality of a given type at one place and assuming that both
qualities could be present at a given place. This vacillation gave them far
more opportunities for calculation than would otherwise have been the
case.
Kilvington and logic, Bradwardine and mathematics,
Burley and natural philosophy
Thus, very generally speaking, while Kilvington contributed to the origins
of the calculatory movement from the logical side and Bradwardine from
the mathematical side, Burley's contribution came mainly from natural
philosophy and from bringing natural philosophy into juxtaposition with
what had once been simply logical problems. Of course William Ockham
and the Ockhamists also contributed very importantly to the habit of
bringing physical considerations into logic - for instance considerations of
what sorts of entities actually exist in the outside world.47
Heytesbury's Regulae
I take the work of Heytesbury and Swineshead as representing the peak of
the Oxford calculatory tradition. In the prohemium to his Regulae solvendi
sophismata (Rulesfor Solving Sophismata) Heytesbury says that it is meant for
the young men engaged in the first year of the study of logic and that it will
deal not with sophismata presenting enormous difficulties, but with those
which come up in common and everyday exercises and which any re-
46. The Second Treatise was published at Venice in 1496. See Shapiro 1959; Maier 1968, pp. 315-52;
Sylla 1973, pp. 233-8.
47. See Murdoch forthcoming b.
sponder ought to come to know how to unravel. 4 8 In saying this, although

modern historians have doubted it, 4 9 Heytesbury clearly does intend to
write for beginners and not for the older and more advanced sophistae; but
in his reference to responders, there is an indication that the book is not for
the most junior students who would simply listen to lectures on logic, but
rather for those who had advanced to the stage of responding in dispu
tations de sophismatibus or in parviso.
In the prohemium Heytesbury goes on to say that he has organised the
subject matter to make it possible to find a given subject more quickly. s o
Although the Regulae consists mainly of fully resolved sophismata, these
are meant as examples which, along with the explicitly stated rules, will
make it easy for a respondent to solve any sophisma put to him. Almost
every section of the work contains the remark that many other sophismata
can be produced concerning the same subject-matter, all of which can be
solved using the rules stated.* 1
The major sections of the Regulae are on (i) self-reference (insolubilia);
(2) 'to know' and 'to doubt'; (3) relative terms; (4) 'begins' and 'ceases';
(5) maxima and minima; and (6) the three categories of motion, - viz., local
motion, augmentation, and alteration. 5 2 O n e might suppose that the
earlier sections concern logic while the later sections pertain to natural
philosophy, but in fact problems associated with traversing spaces and
alteration are introduced into earlier sections as well as later ones. Each
section provides a separate vocabulary and rules, or what might be called
an 'analytical language', to use John Murdoch's terminology, which will
48. William Heytesbury 1494, f. 4 , a : 'Regulas solvendi sophismata non ea quidem quae apparent!
contradictione undique vallavit iuventorum subtilitas aut quae latere soient quempiam logi-
corum, sed quae adeo existunt communia ut communis quotidianaqueexercitatio ea doceat atque
responsalem quemlibet oportet noscere evolvere: vestrae sollicitudini iuvenes studio logicalium
agentes annum primum prout facultatis meae administraret sterilitas: traderem brevi summa: si
non verbosus tumor sophistarum veterum provectorumque indignatio altiora quaerentium huic
operi obviarcnt. Nescio, enim, nec video inter tot et tantas inventiones novas opinionesque tarn
varias sicut iam de die in diem pullulant et frondescunt qualiter declinarent murmura dum quae
ccrta sunt omnibus ulterius attentarem. Verum quoniam est iste labor facilis ipsumque utilitatem
quamdam spero posse amplecti practensam hanc causam non causabor ut causam sed opusculum
istud ut praemissis studentibus offerre proposui moderata brevitate veluti quodam introductionis
modo aggredi temptabo. Er in sex capitula dividens summulam ne dispendiosae et incompositae
narrationis ob prolixitatem lectorum fastidiat oculos. Viso primum quid in singulis inferius agetur
capitulis ut inveniat quisque promptius quod voluerit.'
49. Weisheipl 1968, pp. 196-7.
50. See text in note 48.
51. Heytesbury 1494, ff. 16*, 2 i A , 2 1 " , 26v», 27", 4 i v b . 4*", 44", 44 Λ ·
52. For the later sections, see Wilson 1956. Paul Spade has translated the section on insolubilia (Spade
1979b)·
The Oxford Calculators 5S7
be useful in solving certain sophismata - often with two or more of the

'analytical languages' applied seriatim to the same sophisma.53
In its structure and purpose Heytesbury's Regulae is very nearly the
prototype of the late fifteenth- and early sixteenth-century printed Libelli
54
Sophistarum ad usum Oxoniensium and Cantabrigiensem. These later
printed books, also designed for the use of sophistae or advanced under
graduates, generally contain a collection of short works on basic logic
along with treatises on obligationes and insolubilia, a short work on fallacies,
a vocabulary of the basic terms of natural philosophy, and a treatise on
proportions or ratios like Bradwardine's. The booklet for Cambridge
sophistae has a section devoted specifically to sophismata which deals with
many of the sophismata found in Heytesbury's collection, but treated more
briefly.55 Within its treatise on insolubilia there is a long section on
maxima and minima. 56 The preface to the Libellus . . . Oxoniensium (Booklet
for the Use of Sophistae at Oxford) indicates that it will help the student
impress others with his solutions of sophismata.57
One final indication that all of Heytesbury's Regulae has an essentially
logical rather than natural philosophical intent is that in several places it is
remarked that the cases treated are impossible in nature, although they are
given consideration as imaginable. Thus, for instance, Heytesbury con
siders acceleration to infinity as imaginable and so available for considera
tion, but he thought it physically impossible.58 In dealing with a case of
diminution to zero quantity, Heytesbury says, 'assume for the sake of
argument that numerically the same magnitude can be diminished part by
53. Murdoch 1974a, pp. 58-60; Murdoch 1975a, pp. 282-7. Among these conceptual languages are
the languages of intension and remission of forms, proportions, beginning and ceasing, first and
last instants, maxima and minima, continuity and infinity, and supposition theory. Such an
'analytical language' would have, not only a set of technical terms, but also a set of standard
moves used to analyse any problem.
54. See Ashworth 1979. For fifteenth-century manuscripts of similar works see De Rijk 1975.
55. It is the seventh tract in the book and appears also in MS Cambridge, Gonville and Caius 182/215,
PP- 73-91. Sec De Rijk 1975, pp. 302-3. I have used the edition of London, Wynandus de
Worde, 1524.
56. This is the sixteenth tract in the book. The main treatise on insolubilia is that of Roger
Swineshead, but other material has been added.
57. London, Richard Pynson, 1499—1500, f. Ai verso: ' . . . quoniam tarn inter se congruentes sunt
logica et sophisma ut qui logicam laudat sophistriam bonum logice seminarium laudet necesse sit.
Haec enim tractat subtilissima sophismata quae si quis bene doctique intellexerit videbitur apud
omnes mirabiles sapientiae et disciplinae.'
58. Heytesbury 1494b, 43"**: 'Sed contra illud forte arguitur posito quod A et Β motus sic Intendantur
saltern difformiter A continue in duplo velocius Β quousque uterque illorum habuerit omnem
gradum velocitatis imaginabilem, videlicet usque ad gradum velocitatis infinitum secundum
imaginationem ...' Cf. Heytesbury 1494c, f. i6i v b .
part until it has no quantity. Although, indeed, this is not possible literally
speaking, nevertheless since the case is not self-contradictory it can satisfac-
torily be admitted for the sake of disputation.' 5 9 Gaetano di Thiene
comments that 'calculators should not flee from a case'. 60 Thus the point is
not to discuss what really happens physically, but rather to dispute imagi-
nary cases in the usual fourteenth-century manner. 6 ' Here it is clear that it is
not God's absolute power that leads to the introduction of imaginary cases,
but rather the disputational rules under which a Calculator is supposed to
operate. Heytesbury in one case gives alternate responses to a single
argument and comments: 'The first response is more fitting for the sophista
and the second pertains more to the facts (ad rem).'62
Heytesbury's Sophismata
Heytesbury's Sophismata provides an excellent example of the use of
physical theory and calculations in sophismata. At the same time, I believe,
this treatise shows that the whole investigation was still considered a part of
logic and not partly logic, partly physics. Of Heytesbury's thirty-two
sophismata, Constantin Michalski and others have characterised the last
two as physical and the others as logical because of their sophisma sentences
(and on the basis of the colophon of a Parisian manuscript). 63 Many of
Heytesbury's first thirty sophismata are familiar from the sophisma-
literature of the thirteenth century. 6 4 By contrast, the last two sophismata
are (31) 'It is necessary that something be condensed if something should be
rarefied' and (32) 'It is impossible that something be heated unless some-
59. Heytesbury 1494b, f. 48va: 'Posito gratia argument! quod eadem magnitudo numéro poterit
diminui per partem ante partem usque ad non quantum. Quamvis enim hoc non sit possibile de
virtute sermonis tamen ex quo casus non claudit contradictionem satis poterit admitti gratia
dispu tationis.'
60. Ibid.: 'Cum quantitas sit aeterna mere est impossibile quod ilia magnitudo condensetur usque ad
non quantum. Sed dicit ille magister bene scis hoc, sed quia non implicat contradictionem et est
satis imaginable, ideo calculatores non debent fugere casum quia est fuga baranorum.' The
significance of the last four words is not clear to me.
61. Cf. Wilson 1956, pp. 24-5; Murdoch 1974a, pp. 64—70; Murdoch 1975a, pp. 281, 292; Murdoch
and Sylla 1978, pp. 246-7.
62. Heytesbury 1494b, f. 2l va : 'Prima responsio plus convenir sophistae et secunda ad rem magis
pertinet.'
63. Michalski 1926, pp. 59-60. The MS is Paris BN lat. 16134. Weisheipl 1964a, pp. 178-80.
64. Heytesbury, Sophismata. Cf. Wilson 1956, pp. 153-63 for a list of the sophisma sentences and a
sketch of the argumentation. The first ten sophisma sentences are: (1) 'Omnis homo est omnis
homo'; (2) 'Omne coloratum est'; (3) 'Omnis homo est totum in quantitate'; (4) 'Omnis homo est
unus solus homo'; (5) 'Omnis homo qui est albus currit'; (6) 'Anima Antichristi neccssario erit';
(7) 'Omnis propositio vel eius contradictoria est vera'; (8) 'Isti fêrunt lapidem': (9) 'Neutrum
oculum habendo tu potes videre'; (10) 'Quilibet homo morietur quando unus solus homo
morietur.'
thing be cooled.'65 But a prominent feature of the last two sophismata is

that they are modal propositions involving 'necessary' and 'impossible';
despite their presentation of physical considerations, they might well be
disputed in connection with logical problems of modality. Conversely, the
earlier sophismata might well be proved or disproved on the basis of
physical considerations. Thus the fifth sophisma — 'Every man who is white
is running' - is used to consider the way in which the distribution of a
quality in a body affects what is said about the whole body: how white does
a man have to be to be called white? This might raise the whole issue of the
proper measures of qualities.
Thus one must distinguish between the 'sophisma sentence' and the
'sophisma', the whole development consisting of the sophisma sentence,
hypothesis, proof, disproof, and resolution. The appearance of physics in
sophisma sentences and the appearance of physics in the full sophismata are
sufficiently independent of each other to argue against assuming that there
were 'physical sophismata', which were considered parts of natural
philosophy.
Looking only at Heytesbury's first ten sophismata, it is clear that physi-
cal and calculatory elements are present throughout and are not reserved
for the last two sophismata. And yet, again, these elements are applied not
to natural cases, but rather only to imaginary cases.66 Again Gaetano
comments: 'For although these cases now posited are not possible de facto,
nevertheless they are imaginable without contradiction, and so they should
be admitted by the logician'.67
If we turn to the last two so-called physical sophismata, we find indeed
that mainly physical arguments are used there, but that logical arguments
are also used, and that the physical cases proposed are not naturally possible,
but only imaginable. In discussing whether it is true that, in the words of
the last sophisma 'It is impossible for something to be heated unless
something should be cooled', Heytesbury must first of all, for the sake of
disputation, agree to use words loosely, because he believes that, strictly
speaking, nothing is heated or cooled, since whenever something is acted
on by a heating or cooling agent it undergoes a substantial as well as an
accidental change and hence does not remain the same thing. Water is
6
5· (31) 'Necesse est aliquid condensari si aliquid raréfiât'; (32) 'Impossibile est aliquid calcfieri nisi
aliquid frigefiat.'
66. Cf. Heytesbury 1494c, 133", and esp. ι6Γ*> and 162". Cf. Wilson 1956, p. 25, n. 65.
67. Heytesbury 1494c, f. 89™: 'Nam licet casus nunc positi de facto non sint possibiles, sunt tamen
imaginabiles absque contradictionc, quare a logico admittendi.'
56ο Natural philosophy
never heated because as soon as a heating agent is applied it begins to

convert the water to air. 6 8
Even beyond this, however, much of Heytesbury's argumentation is
based on cases that are naturally impossible. Supposing that bodies often
simultaneously contain degrees of hot and cold, Heytesbury concedes that
if, per impossibile, there were a body with no hot or cold, then it could
simultaneously be heated and cooled. 6 9 Indeed, he thought that any body
in which the sum of the degrees or latitudes of hot and cold was less than the
maximum degree or latitude could be simultaneously heated and cooled
until the sum of its degrees reached the maximum degree, at which time
any further heating would require a reduction of the cold present and vice
versa. 7 0 But since it would never occur naturally that a body was qualified
with degrees totalling less than the maximum degree, Heytesbury's interest
in the last two sophismata is no more truly physical than in the earlier ones;
he is interested in performing thought experiments, but he is unconcerned
with even their theoretical realisability.
Swineshead's Liber calculationum

If Swineshead's Liber calculationum (Book of Calculations) is considered in
the light of Heytesbury's works, it becomes clear that, for all its exclusive
concern with physical topics, the Liber calculationum, too, is a work concern
ing not real but imaginable cases and designed to be used in disputations.
The treatises of the Liber calculationum consider (i) intension and remission:
(2) non-uniformly qualified bodies; (3) the intensity of an element having
two unequally intense qualities: (4) the intensity of mixed bodies; (5) rarity
and density; (6) augmentation; (7) reaction; (8) powers of things; (9)
difficulty of action; (10) maxima and minima; (11) the place of an element;
(12) light sources; (13) the action of light sources; (14) local motion; (15)
non-resisting media; (16) the induction of the maximum degree. 7 1
The typical treatise first lists the various positions usually taken concern-
68. Ibid., ff. 164«, I65 , b .

69. Ibid., f. 17ο"1: ' . . . dato per impossibile quod esset aliquod corpus alterabile quod nullam cal-
iditatem nec frigiditatem haberet et esset sibi approximatum aliquod aliud corpus quod esset
uniforme per totum aequaliter omnino habens de caliditate sicut de frigiditate, conceditur tunc
quod illud agens agcret in illud alterabile simul caliditatem et frigiditatem quousque totum esset
assimilatum Uli agenti'.
70. Ibid., f. 170•*: 'verumtamen quamvis aliqui duo gradus caliditatis et frigiditatis possunt simul
intendi, non tamen sie possunt omnes gradus, quia nulli duo gradus ultra gradum medium sue
latitudinis possunt intendi. Unde si sint duo gradus caliditatis et frigiditatis medii coextensi simul,
et unus illorurn incipit intendi, alius remitri, ideo etc.'
71. Richard Swineshead 1520. See Murdoch and Sylla 1976 and 1978, pp. 236-7.
ing the subject matter of the section, next recites various conclusions or
sophismata that might be raised against each given position, and finally
shows how the sophismata apparently opposed to the preferred position
might either be accepted as unobjectionable or resolved. Sometimes more
than one position is considered sustainable.72 Several sections end with the
statement that many other sophismata might be raised and solved using the
preceding material.73
This, at least, is the pattern of the earlier treatises of the book. Some of
the later treatises consist only of sets of conclusions following from a given
position, and these conclusions are generally very prominently of a calcu-
latory nature - as in the fourteenth treatise, which gives rules for local
motion following from Bradwardine's law concerning the relations of
forces, resistances, and velocities. Nevertheless, from the juxtaposition of
these treatises, the purpose of the calculations seems clear: they are used to
derive surprising or counter-intuitive results and to determine whether or
not these must be accepted.
Bradwardine and Heytesbury were often content to inspect mathema-
tical relationships in order to give a general classification of the results to be
expected. Thus Bradwardine asserts that if one starts with a force less than
twice a resistance, then by doubling that force one will produce a velocity
more than double the original velocity.7* He does not indicate how to
determine exactly the ratio by which the velocity will increase. Heytesbury
likewise is content to say that given any uniformly accelerated motion not
starting from zero velocity, the ratio of the distance traversed in the second
half of the time to the distance traversed in the first half of the time will be as
the mean velocity of the second half of the motion to the mean velocity of
the first half of the motion. He even seems to suggest that to calculate the
actual numerical values would be more bothersome than helpful.75
But both Bradwardine and Heytesbury made enough progress toward
deriving mathematical results that their successors were stimulated to go
further, and Swineshead's Liber calculationum is the result of many such
further steps toward actual calculation.
Still, it should be pointed out that even in the Liber calculationum most of
the mathematics involved is not very numerical and does not give rise to
72. See Murdoch and Sylla 1976, pp. 190-1.
73. Swineshead IS20, ff. 9", 9 A , I5 A , i6v*, 22*.
74. Crosby 1955, pp. 112-13. Conclusion 6.
75. Heytesbury 1494b, f. 41*. The printed version actually says less difficulty ('sed huiusmodi
calculatio minorem sollicitudinem agcrct quam profectum'), but compare Gaetano's comment
on the passage. Cf. Swineshead 1520, f. 52".
many computations or calculations in the usual modern senses of those

terms. Instead, one has to do with verbal reasoning in which many
variables and their relations are kept in mind and equivalences derived with
very little use of numbers. 7 6 It is in this broad sense of'calculation' that
Heytesbury can say concerning insolubilia; 'Many very prolix and useless
cases arise in this matter in which it is necessary to calculate diligently and
to run from one proposition to another until it appears which of them is
insoluble.' 7 7 In this broad sense, too, any of the familiar fourteenth-
century arguments in terminis, in which the letters A, B, C, etc., were used
to represent, for example, velocities, or degrees of quality, or propositions,
or distances, or instants, may have been considered calculations. 7 8
If Swineshead's book covers only physical topics, it is nevertheless no
more genuine physics than are Heytesbury's works. Having treated many
different distributions of hot and cold, Swineshead remarks, 'All of these
must be conceded imaginarily, and they are to be denied defacto; for in all
these cases it is posited that unequal colds are extended with equal h e a t . ' 7 9
In another place Swineshead supposes that there are in half a body infinite
heats of infinite intensity and infinite colds of infinite intensity, something
obviously not possible in nature. 8 0
The Oxford disputations in the light of the Calculators' writings

If, then, the core works of the Calculators were meant to be used as aids in
preparing for disputations, does this prove anything about the disputations
themselves? Unfortunately, there need not have been a simple connection
between, say, the contents of Heytesbury's Regulae and some particular
series or type of disputations. 8 1 Nevertheless, putting the works of the
Calculators together with the Oxford statutes discussed earlier, the series of
disputations described in the statutes providing the best match with
Heytesbury's and Swineshead's work seems to be the disputations in
76. Cf. Murdoch 1974a, p. 67 and n. 39.

77. Heytesbury 1494b, f. 7 Λ : 'Multi etiam fiunt casus in hac parte prolixi nimis et inutiles in quibus
oportet diligenter calculare et discurrere ab una propositione ad aliam quousque appareat quae
illarum sit insolubilis. '
78. Maier 1952, pp. 257-88, esp. 258-60.
79. Swineshead 1520, f. 15™: 'Omnes istae concedendac sunt imaginarie et negentur de facto. Nam in
omnibus casibus istis ponitur cum aequali caliditate inaequales frigiditates extendi.' Compare
ff. 8", I3 v b , is"'h, 16", 16".
80. Ibid., f. 8".
81. It would be useful to look for more direct records of such disputations. Some of the works already
known but not carefully examined with this issue in mind may prove to be of help. Perhaps one
should not expect undergraduate disputations to be recorded, but more evidence would be
desirable.
parviso. Calculatory techniques could be introduced into disputations on

natural philosophy, but then the results of those disputations were much
more recognisably physical than are the core works of the Oxfore
Calculators. The result would be something like Roger Swineshead's De
motibus or Dumbleton's Summa, but not Swineshead's Liber calculationum.
When the techniques of the Calculators were brought into the areas of the
three philosophies, their application was perhaps questionable because so
many of the cases they considered were thought to be possible only in
imagination.
Conclusion
Although there is excellent precedent in fourteenth-century theories of
scientific method for considering logical analysis as a method of scientific
discovery,82 nevertheless the Calculators carried their analyses and calcu-
lations a bit too far for it to be plausible that their main goal was discoveries
in natural philosophy. If some later natural philosophers found the calcu-
lations subtle and 'beautiful speculations',83 it is not surprising that others
considered them unduly complicated and irrelevant. When later humanists
complained that the English disputational subtleties made too much of
what should have been schoolboy exercises, they may have been right, at
least concerning the context in which the subtleties developed.84 The
historian inclined to admire the work of the Oxford Calculators may wish
to assert that it was indeed the work of masters and not of schoolboys.
While this is no doubt true, it nevertheless seems to be the case that the
work was to be used in disputations de sophismatibus or in parviso, which
were meant primarily for advanced undergraduates. Even while thinking
of the work as that of masters, we ought also to wonder at the level of
logical sophistication that advanced undergraduates in fourteenth-century
Oxford must have attained.
82. See Sylla 1979, esp. pp. 176-7.

83. There is a marginal note in Swineshead's Lifcf r calculationum to this effect (f. 64A): 'Nota pulchram
speculationem.'
84. Cf. Garin 1969a; Guerlac 1979.
28
INFINITY AND C O N T I N U I T Y
The prevalence of issues involving infinity and continuity

Natural philosophy in the fourteenth century is, when compared to that of
the preceding century, more extensive, less repetitious, and more varied in
the problems it treats, the solutions it sets forth, and the approaches and
methods it employs in reaching those solutions. However, if one examines
in some depth not merely the expositions and questions dealing with the
relevant works of Aristotle but also the numerous non-commentatorial
works constituting this literature, one cannot but be impressed by the
unusual amount of time and effort spent in dealing with problems involv-
ing in one manner or another the infinite and the continuous.
Often these problems concern infinity or continuity from the outset; but
equally often the problems are extended or developed by the fourteenth-
century scholar to take into account some aspect of the infinite or the
continuous in a manner that was not apparent in the problem as initially
stated. A discussion of the way in which one should measure a quantity that
varies in intensity throughout its subject might, for example, be carried so
far as to accommodate 'infinite values'. Alternatively, a discussion of
angelic motion might involve one in a rather full investigation of the
composition of all continuous quantities. Indeed, the prevalence of issues
involving the infinite or the continuous in later medieval natural philos-
ophy is such that an exhaustive history of these two notions in the later
Middle Ages would constitute a very large part of the history of natural
philosophy during this period.
The Aristotelian background

Aristotle had himself devoted considerable attention to the infinite and the
continuous: the major portion of Book III and all of Book VI of the Physics,
as well as three substantial chapters in the first book of De caelo, deal
expressly with these notions. And many other, less extensive discussions,
especially in the later books of the Physics, also take up aspects of these
Infinity and continuity 565
notions. Aristotle's evident concern in this regard was not lost on his later
medieval expositors. When one considers that the Physics was the most
commented upon of Aristotle's natural philosophical works through the
first half of the fourteenth century and, further, that it received more
attention at Oxford — where preoccupation with problems involving
infinity and continuity was proportionally greater than on the Continent -
it seems obvious that Aristotle provided a considerable part of the
background and impetus for the later medieval concern with these two
notions. Just how Aristotle served in this regard will be made clearer when
we turn to the specific issues of the infinite and the continuous that were
treated by medieval scholars and cite the Aristotelian source for the prob-
lem at hand.
Aristotle stimulated and influenced the medieval preoccupation with
infinity and continuity not only because of his explicit discussions of those
notions but also because of the central importance of continuous, infinitely
divisible quantities and processes throughout Aristotelian natural philos-
ophy.' Indeed, even if the processes are not themselves continuous - like
locomotion or qualitative change - they are considered as taking place
within an absolutely continuous time, so that the concept of the continuous
is always relevant if only in respect of analysing just how some discontin-
uous change - some substance passing into or out of existence, for example
- could be precisely described against the background of continuous time.
The medieval development of the Aristotelian material

For all the importance of the Aristotelian background to medieval discus-
sions of infinity and continuity, there are few areas in later medieval
natural philosophy in which one can more clearly observe the medievals
going beyond their inherited Aristotelian material. This development of
what could be gleaned from Aristotle with respect to the infinite and the
continuous is apparent even in the very kinds of medieval writings that
provide the sources for the history of these two notions. For we find an
immense amount of relevant material not only in commentaries on Aristo-
telian works and in one or another Summa of natural philosophy, but also a
surprising number of special treatises and quaestiones dealing exclusively
with some aspect of the infinite or continuous.2
1. For an account of the importance of cominuity for Aristotle, see W. Wieland 1970, pp. 278-316.
2. This will be evident from some of the works to be cited in what follows, but there are many other
separate works on infinity and continuity, especially in the fourteenth century.
The influence of theological considerations

Writings produced in the Faculties of Theology were as important as those
produced in the Faculties of Arts; thus theological considerations were
influential, especially in furthering the analysis of the infinite. For example,
the extensive thirteenth- and fourteenth-century debates over the possible
eternality of the world provided a stimulating setting in which to puzzle
over the difficulties involved in the notion of an infinite past time. And
discussions of God's infinity or of the possibility of His knowing an infinite
had similar effects.3
Fourteenth-century commentaries on the Sentences often contained
extensive deliberations concerned with the infinite and the continuous in
less obviously appropriate contexts, such as discussions of whether the will
acts instantaneously or continuously over time, of how one might best
'measure' its meritorious and demeritorious acts, or of how one might
compare the perfections of various creatures with one another and with
God.*
The most influential theological element was the newly precise concept
of God's omnipotence. Writers began to base their analyses of the infinite
and the continuous upon what could obtain in respect of the absolute
power of God. But since God's absolute power extended to everything that
did not include a contradiction, to invoke this power in examining infinity
or continuity was to transfer one's analysis from the realm of the physically
possible (within the confines of Aristotelian natural philosophy) to the
broader realm of the logically possible.
Although this appeal to divine omnipotence arose within a theological
context, it appears as well in works that are clearly devoted to natural
philosophy. 5 Yet whether such an appeal occurred within the one context
or the other, it had the effect of extending the discussion at hand so that in
principle it would take into account all logical possibilities. Invoking the
absolute power of God thus supplied one with a warrant to reason secundum
imaginationem, a manner of reasoning which, when applied to the infinite
and the continuous (or to almost any other area of late medieval philos-
ophy), permitted the exercise of a number of new methods and tools of
3. On God's infinity, see Sweeney & Ermatinger 1958 for a single instance of the kinds of things at
issue. For the problem of God's knowledge of the infinite - a problem that derives from Augus-
tine, De civ. Dei, XII, 18 - see William of Alnwick 1937, pp. 488-551.
4. References to such less obviously appropriate contexts can be found in Murdoch 1975a, pp. 289-
303.
5. For example, John Buridan 1509, Quaestiones Physkorum, III, QQ. 14, 19.
analysis in resolving (in some cases first creating and then resolving)
problems.
Actual and potential infinity

Of all the points made by Aristotle in his treatment of the infinite in Book
III of the Physics, undoubtedly the one most often repeated by the medieval
philosopher was his denying the possibility of an 'actual infinite' of any sort
and admitting only the 'potential infinite' that was associated with the
infinite divisibility of continuous magnitudes. As Aristotle put it, any
permissible infinite is not that beyond which there is nothing (for that
would be a completed infinite in actu), but rather that beyond which there
is always something (the infinite in potentia).6
Almost all the scholastics followed Aristotle's distinction of permissible
from non-permissible infinites, formulating a variety of alternative ways
of expressing this distinction. The most popular of these alternative expres-
sions was the claim that the rejected actual infinite was a quantity so great
that it could not be greater (tantum quod non maius), while the permissible
potential infinite was a quantity that was not so great but that it could be
greater (non tantum quin maius).1 The scholastics themselves often pointed
out that the latter was really only an indefinite finite, as was made explicit
in any number of 'expositions' of propositions involving this type of
infinite.8
The logic of the infinite

The exposition of propositions involving infinity is at the heart of what
might be called the medieval logic of the infinite. The fundamental
distinction in this logic, or analysis, is between the 'categorematic' and
'syncategorematic' uses of the term 'infinite* (or 'infinitely', or 'infinitely
many'). Taken categorematically, the term functioned collectively, very
6. Physics, III, 6,207*1,7-8. The medieval Latin is: 'Non enim cuius nihil est extra, sed cuius semper
aliquid est extra, hoc infinitum est. ... Infinitum quidem igitur hoc est, cuius secundum quanti-
tatem accipientibus, semper est aliquid accipere extra.'
7. The same distinction for multitudes was expressed, respectively, as 'tot quod non plura' and 'non
tot quin plura'. Another frequently used manner of expressing the distinction between Aristotle's
two infinites was to speak of an infinitum in facto esse (=actual) vs. an infinitum infieri(=potential).
On these distinctions, sec Maicr 1964, pp. 41-4.
8. Hence in his Tractatus de continuo, Thomas Bradwardine characterises the potential infinite as
'infinitum privative secundum quid est quantum fini turn, et finitum maius isto, etfinitummaius
isto maiori, et sic sine fine ultimo terminante; et hoc est quantum, et non tantum quin maius'
(MssTorun R, 40, 2, p. 153; Erfurt, Amplon. Q° 385. 17e; Paris BN n.a.i. 625, 71"). Note that
because of this 'always finite' character of the Aristotelian (and hence also of the scholastic)
infinite, Georg Cantor was later to term it an Uneigentlich-unendliche (Cantor 1932, pp. 165,180).
much like an ordinary numerical adjective or modifier; but taken syn-

categorematically it was held to function as a distributive sign. Thus,
'homines infiniti currunt' means that an actually infinite number or collection
of men are running, the key to the proper exposition of this proposition
being that the term 'infiniti' follows the subject term 'homines' and hence
functions categorematically. On the other hand, if the term 'infiniti' pre-
cedes the subject term it modifies, then it occurs in its syncategorematic
sense, so that the proper exposition of 'infiniti homines currunt' is that some
men are running but not so many that no more are running. But this is to
say only that a potentially infinite number of men are running. 9 In this
latter syncategorematic sense, 'infinitely many' ('infiniti') was taken as a
distributive sign, that is, as 'distributing' (or 'multiplying') the common
subject term relative to the predicate.10 For example, Peter of Spain
explains the syncategorematic use of'infinitely many' by claiming that in
that use it functions as equivalent to the phrase 'more than any [number]
you choose' (quolibet plura).' '
The re-examination of the possibility of actual infinity

In Book III of the Physics and Book I of De caelo Aristotle had argued
against the existence of actual, completed infinites largely on the basis of
the inconsistency of an actual infinite with the doctrines of natural place
and the contrary opposition and transformation of the elements; con-
9. See Maier 1964, p. 44. Using the same doctrine of the position of'infinite' relative to what it
modifies, Albert of Saxony distinguishes "in infinitum continuum est divisibile' from 'continuum
est divisibile in infinitum' (Quaestiones Physicorum, III, Q.9; Albert of Saxony Iji8a, f. 37"). The
former takes 'infinitum' syncategorematically and is a true statement about the potentially
infinite divisibility of continua, while the latter takes 'infinitum' categorematically and is a false
statement about the divisibility of continua, since it would imply an actually infinite, completed
division. John Buridan, however, at times does not agree that it makes a difference if 'infinite'
precedes or follows what it qualifies: 'non refert dicere infinita magnitudo et magnitudo infinita,
sicut non refert dicere homo albus et albus homo' (Quaestiones Phys. Ill, 18; (Buridan 1509, f. 61 *).
10. Thus in "infiniti homines currunt,' 'infiniti' distributes 'homines' relative to 'currunt.'
11. PeterofSpain 1972, p. 231: 'Solet autem poni quod "infinitum" quandoquesumitur pro termino
communi, et tunc hec propositio "infinita suntfinita"equipollet huic: "aliqua infinita sunt finita";
quandoque autem sumitui pro signo distributive et tunc ista "infinita suntfinita"equipollet huic
quoad distributionem, scilicet " quolibet plura suntfinita".Et probatur sic: uno plura sunt finita,
doubus plura sunt finita, tribus plura sunt finita, et sic de aliis; ergo quolibet plura sunt finita. Et
tunc dicitur facere intcrscalarcm distributionem, sive interruptam vel discontinuant, quia hec
dictio "plura" in prima propositione supponit pro duobus et deinceps, et in secunda pro tribus et
deinceps, et sic semper gradatim sive scalariter ascendendo. Et ideo hec oratio: "quolibet plura"
facit interscalarem distributionem, quia pro aliis supponit hoc quod dico "quolibet" et pro aliis hoc
quod dico "plura", secundum numerum ascendendo, ut dictum est.' For other analyses of die
categorematic and syncategorematic uses of 'infiniti' see William of Sherwood 1941, pp. S4-$;
1968, pp. 41 —3; Peter of Spain 1945, pp. 118—22 (which contains the probably spurious Tractatus
exponibilium); Albert of Saxony 1502, Sophisma 53; Thomas Bradwardine, Tractatus de continua,
MssTorun R. 40. 2, pp. 153-6; Erfurt Amplon. Q" 385, I7r—18*.
sequently, the arguments elucidating this inconsistency were for the most
part physical in character. But it was not physical possibilities that were in
question when the medievals went beyond Aristotle and asked whether
God in his absolute power could create an actually infinite stone or could
12
complete the infinite division of a continuous magnitude. If the absolute
power of God could not accomplish such 'tasks', it was certainly not mere
defacto principles of natural philosophy that ruled them out. Some stronger
inconsistency would have to be involved were such a restriction validly to
be placed upon God's omnipotence, and the one which seems to have been
most compelling to the medievals is the paradox occasioned by the implied
existence of unequal infinites.
The paradox of unequal infinites used against actual infinity

The structure of the paradox was uncommonly simple and straight
forward: if one allows the existence of actual infinites, then it appears that
some infinites will clearly be greater than other infinites which are equally
clearly parts of the former; but it is axiomatic that all actual infinites are
equal; therefore in this instance a part is not less than, but equal to, its whole
- which is absurd.
The discussion of this paradox was not something the medieval philo
sopher could cull from Aristotle, and although the paradox was treated in
other ancient sources,13 they were either not available in the Latin Middle
Ages or were never cited; the same must be said for Islamic discussions of
the paradox.14 On the basis of presently available evidence, then, philoso
phers and theologians in the Latin West appear to have realised the impor
tance of the paradox on their own. Discussions of it can be found in any
number of thirteenth- and fourteenth-century texts,.but it was probably
Bonaventure's inclusion of it in his examination of the possibility of an
12. These examples are quite common, but sec, e.g., John Buridan 1509, III, Q. i9:f.64'~ v : 'Hec est
impossibilis: in qualibet medietate proportiunali huius dici Deus creat unum lapidem pcdalcm ct
etiam ista: in qualibet medietate proportional! huius dici creabit unum lapidem pcdalcm.
... Omnes partes linee Β Deus potest separarc ab invicem et separatim conservare In qualibet
medietate proportional! huius dici potest Dens facere unum lapidem pcdalcm conservando ipsum
semper post.' (Note the presence of'potest' in the last two propositions and its absence in the first.)
13. Sec, for example, Plutarch, De comm. not. adv. Sloicos, 1079a; John Philoponus, Dr. if ternitate numdi
contra Procluin, I, 3, and apud Simplicium, Phys., VIII, 1, ed. Dicls, p. 1179; Alexander of
Aphrodisias, Quaest. naturales, III, p. 12; Proclus, Comm. in Eiiclidcm, dcf. 17; Proclus, Elem. llieoi,
prop. 1 ; Lucretius, 1.615-26. Only Lucretius and Proclus' Elements of Theology were available to
the Latin West, but neither was cited relative to the paradox in question.
14. Al-Shahrastânî, Kitâb Nihâyatu Ί-Iqdâm ft 'Ilnii'l-Kalhn, ed. A. Guillaume, Ch. 1; Averroes,
Tahafut al-Tahafut, tr. Van den Bcrgh, pp. 9-10, 14, 162-3: Maimonides. Guide for the Perplexed,
tr. Fricdlander, p. 138. Maimonides was translated into Latin, but was not cited in this context.
The translation of Averroes' Tahafut occurred too late ( 1328) to enter effectively into the debate in
the Latin West.
eternal world that gave it the status of an essential ingredient in almost all
subsequent discussions ofthat possibility. ' 5 The use to which Bonaventure
put the paradox represents one of three basic attitudes exhibited by the
medievals towards unequal infinites, the one a part of the other - i.e.,
regarding them as strictly impossible, and maintaining, therefore, that one
had to reject as equally impossible the situation that appeared to generate
such infinites in the first place (in Bonaventure's case, the possibility of an
eternal world). 1 6
Resolutions of the paradox of unequal infinites

However, not all the medievals used the paradox in this way; some attem-
pted to resolve it. O n e group simply concluded that the fact that the same
infinites could be shown to be both equal and unequal meant that one
could not apply 'equal to', 'greater than', and 'less than' to actual infinites at
all. 17 But a second group of those who attempted a resolution of the
15. Bonaventure, 1882-19023, Comm. Sent., II, dist. 1, pars. 1, art. 1, Q. 2 (the first of his arguments
against an eternal world): 'Prima est: Impossibile est infinito addi. Haec est manifesta per se, quia
omne illud quod recipit additionem fit maius, infinito autem nihil maius. Sed si mundus est sine
principio, duravit in infinitum; ergo durationi eius non potest addi. Sed constat, hoc esse falsum,
quia revolutio additur revolutioni omni die; ergo, etc. Si dicas quod infinitum est quantum ad
praeterita, tamen quantum ad praesens quod nunc est est finitum actu, et ideo ex ea parte qua
finitum est actu est reperire maius; contra, ostenditur quod in praeterito est reperirc maius. Haec
est Veritas infallibilis: quod, si mundus est aeternus, revolutiones solis in orbe suo sunt infinitae;
rursus, pro una revolutione solis necesse est fuisse duodccim ipsius lunae; ergo plus revoluta est
luna quam sol; et sol infinities, ergo infinitorum ex ea parte qua infinita sunt est reperire excessum.
Hoc autem est impossibile; ergo etc'.
16. The impossibility of unequal infinites became one of the most popular, and most telling, ways of
disposing of the eternity of the world, especially among Franciscans who, steadfastly refusing to
consider the possibility of an eternal world as a problema neutrum as had Thomas Aquinas (e.g., ST,
I, Q. 46, art. 1—2), did their utmost to establish its impossibility. However, it was not only a
Franciscan trait (which, admittedly, becomes less universal amongfratres minores in the fourteenth
century) to wield the paradox of unequal infinites against the eternity of the world; one finds
a highly developed version of the paradox put to the same task, for example, in Thomas
Bradwardine's De causa Dei, Lib. I, cap. 1, coroll., pars 40; Bradwardine 1618, pp. 121-6; Cf.
Murdoch 1962, pp. 18-20. Mention should also be made of the fact that, like Lucretius (I,
615-26), Walter Chatton used the paradox to refute the possibility of the infinite divisibility of
continuous magnitudes (see his Comm. Sent., Lib. II, dist. 2, Q. 3: MS Paris BN 15887, 94*).
Chatton (who does not mention Lucretius in this connection) was a 'finitist', holding the
extremely strange view that a continuum was composed of a finite number of non-extended
indivisibles. On Chatton and continua, see Murdoch and Synan 1966.
17. For example, Nicole Orcsme, Quest, phys., Ill, Q. 12 (MS Sevilla Colomb. 7-6-30, 37v-39v),
where the major conclusion is: 'Nullum infinitum alteri comparatum per ymaginationcm est ipso
minus vel equate vel maius, sed omne omni est incomparable.' Albert of Saxony's treatment of
the issue parallels that of Orcsme; see Albert of Saxony 1518b, I, Q. 10. Although one cannot
maintain that they knew either Oresme or Albert, both Galileo and Newton come to a similar
conclusion about the inapplicability of'equal to', 'less than,' and 'greater than' to infinites. (See
Galileo, Discorsi (ed. naz.), p. 79; Newton, Letter to Richard Bentley, 1693, in I. B. Cohen (ed.),
Newton's Papers and Letters on Natural Philosophy (Cambridge, Mass., 1958), pp. 293-9.)
paradox asked instead whether different 'axioms' of equality and part-

whole might not be applicable to infinite magnitudes and multitudes. One
might consider, we are told, an infinite line AB beginning at some
arbitrary point A and extended ad infinitum toward B; now if one specifies a
determinate point C on the line at some finite distance from A, then our
paradox arises with respect to the whole infinite line AB and its equally
infinite part CB. Yet one can resolve the paradox by noting that, while AB
is indeed equal to CB relative to the terminus B, they are unequal relative
to the termini A and C.18 Infinites can have, as It were, 'finite ends', and can
hence simultaneously suffer equality and inequality depending upon
'direction'.19
Henry of Harclay
A more fruitful attempt to uncover a way to tailor part-whole and
equality relations for infinites began, it seems, with Henry of Harclay,
Chancellor of Oxford in 1312. A proponent of the existence of actual
infinites and the composition of continua out of indivisibles, Harclay
believed that there can be, and are, unequal infinites and realised that the
crucial issue one must resolve if one is to hold such a belief is to explain just
what kind of part-whole axiom governs such infinites. In anticipation of
his adversaries, he asserts that the traditional Euclidean axiom 'every whole
is greater than its part' is naturally applicable only to finite quantities.
Nevertheless, one must realise that it is subordinate to a more general
axiom which does apply to infinites and in terms of which the inequality of
infinites must be understood: 'that which (e.g., an infinite set) contains
another thing (e.g., an infinite proper subset) and something else beyond
(ultra) it, or in addition to (praeter) it, is a whole with respect to that other
thing'.20
William of Alnwick and Gregory of Rimini

Harclay's first critic, the Franciscan William of Alnwick, claims that there
is a significant difference between the terms 'beyond' and 'in addition to'
which Harclay employed in a seemingly innocuous disjunction in his
18. The argument is that of an unnamed opponent in Peter John Olivi 1922-6,1, pp. 38-9.
19. The 'finite end' of infinites is frequently considered in attempting to resolve the paradox. So it is,
for example, in Gerard of Odo's Comm. Sent., Lib. II, dist. 2, Q. 2 (MS Valencia Catcd. 200,
17'—17*); and in Michael of Massa 's Comm. Sent., addiliones libri secundi (MS Vat. lat. 1087,
151*-154")·
20. Henry of Harclay, Questio de infinito et continuo (MSS Tortosa Cated. 88,83"; Firenze Naz. II.II.281,
951): 'Illud quod continct aliud et aliquid ultra illud vel praeter illud est totum respectu illius."
generalised axiom: 'beyond' entails having more absolutely, while 'in

addition to' entails having more only when it is a question offinitethings of
a certain quantity. Among infinite things, Alnwick maintains, 'in addition
to' entails diversity but not a greater plurality.21
Although in drawing this latter distinction Alnwick may have begun to
move in the direction of what we today consider set-subset relations on the
one hand and cardinality relations on the other, he had still not faced the
question of precisely what it is that is involved in the inequality which can,
or cannot, be inferred on the basis of the distinction he had drawn between
mere diversity and greater plurality. A reasonably satisfactory treatment of
that question is provided in the lectures on the Sentences given at Paris in
1342 by Gregory of Rimini. 22 If one is to apply 'part' and 'whole', 'greater
than', and 'less than' to infinites, one must realise, Gregory claims, that
there are two senses for each term. In one way, everything is 'a whole
which includes something that is something and something else in addition
to (praeter) that something*. But in a second, more restricted sense that is 'a
whole which includes something in the first way and also includes as many
things as that included does not include'. An infinite multitude can,
Gregory continues, very well function as a whole with respect to another
infinite multitude in the first sense, but not in the second sense. It seems
clear, then, that what Gregory intends is, in our terms, distinctions be-
tween whole and part in the sense of set and subset (his first sense) and
between whole and part in the sense of unequal cardinality of the sets
involved (his second sense).23 By such means Gregory, who believed in the
i i . William of Alnwick, Determinatio 2 (MS Vat. Pal. lat. 1805, 10"): 'Dicendum quod refert dicere
"habere aliquid in se et aliquid ultra illud", et dicere "habere in se aliquid praeter illud". Quod
enim habet in se aliquid et aliquid aliud ultra illud est maius illo, quia habere in se ultra aliquam
quantitatem, est habere plus. Sed habere aliquid et praeter illud habere aliud, non includit habere
plus nisi in rebus finitis; sed includit diversitatem sed non maiorem pluralitatem in infinitis.
Tempus igitur habens in se infinitos menses februarios et ianuarios non habet in se aliquod tempus
ultra hos menses aut illos; habet tamen in se aliquid praeter istos menses aut praeter illos.'
22. Although one can find similar attempts to resolve the paradox between Alnwick and Rimini
(e.g., in Ockham 1495-6, Comm. Sent., Lib. II, Q. 8; and Ockham 1491, Quodl. II, Q. 6), they are
not as successful as that of Rimini.
23. Gregory of Rimini 1522, Comm. Sent., Lib. I, dist. 42-44, Q. 4; f. 173": 'Primo modo omne quod
includit aliquid quod est aliquid et aliud praeter illud aliquid et quodlibet illius dicitur totum ad
illud; ct omne sic inclusum dicitur pars indudentis. Secundo modo dicitur totum illud quod
includit aliquid primo modo et includit tanta tot quot non includit inclusum; et econverso tale
inclusum non includens tot tanta quot includens dicitur pars eius. ... Secundo distinguo hos
terminos maius et minus... uno modo sumuntur proprie et sic multitudo . . . dicitur maior quae
pluries continet unum vel plures unitates; ilia vero minor quae paucies seu pauciores. Alio modo
sumitur improprie, et sic omnis multitudo quae includit unitates omnes alterius multitudinis et
quasdam alias unitates ab illis dicitur maior ilia, esto quod non includat plures unitates quam ilia.
Et hoc modo esse maiorem multitudinem alia non est aliud quam includere illam et esse totum
respectu illius primo modo.'
existence of actual infinites,2* removed what was perhaps the most telling
objection to such a belief.
Infinity in continua
Late medieval discussions of the infinite as involved in the divisibility of
continuous magnitudes were perhaps richer and more varied in the new
considerations they brought to the fore than were discussions of actual
infinity. It was all but universally agreed that continua in some way or
other contained, at least potentially, an infinite number of indivisibles.25
They contained them 'in some way or other' because it was realised that the
very existence of indivisibles was itself exceedingly problematic. Indeed,
although Aristotle had devoted considerable attention to issues involving
the presence of indivisibles within continua, he had not spent much effort
in examining their existence as such. The medievals, however, did so, led
by William Ockham.
Ockham on indivisibles
Put succinctly, Ockham's position was that indivisibles do not exist at all. A
point, for example, is not a thing in any proper sense. Points are, of course,
traditionally regarded as the termini of lines. But in Ockham's view we do
not need separately existing points to account for the termination of lines;
all we need are finite lines of this or that length. Indeed, he is willing to say
that the term 'point' is equivalent to the expression (complexum) 'a line of
such and such a length' or something ofthat sort.26 But if we can use such
an equivalence to rid ourselves of points, it follows that we do not need
separately existing points to account for the termination of lines; the lines
24. See Maier 1964, pp. 82-4.
25. The fact that continua contained indivisibles is not to say, of course, that they were composed of
indivisibles. (The latter was a quite different issue and will be treated briefly below.) These
'contained' indivisibles would be points (for one-dimensional geometrical magnitudes), lines (for
two-dimensional magnitudes), or surfaces (for three-dimensional magnitudes), instants for time,
and degrees for intensible and remissible qualities or forms.
26. William Ockham 1930, pp. 36-8 (MSS Basel F.II.24, 23'; Rome, Angelica 1017, 660: 'Et si
queratur quid est punetus, aut est res divisibilis aut indivisibilis, dicendum est quod, si sic dicendo
"punetus est aliquid" vel "punetus est res" vel huiusmodi li "punetus" supponat pro aliquo ita
quod habeat precise vim nominis et non includat equivalcntcr unum complexum ex nomine ct
verbo vel aliquid consimile quod secundum proprietatem vocis potest reddere suppositum verbo,
debet conccdi quod punetus est aliquid ct quod punetus est res; et hoc quia debet concedi quod
punetus est linea et punetus est quantitas, quia tunc hoc nomen "punetus" equivalet toti isti:
"linea tante vel tante longitudinis" sive "linea non ulterius protensa vel extensa" vel alicui tori
composito ex adiectivo et substantivo vel alicui toti composito ex nomine et verbo mediante
coniunctione vel adverbio vel hoc pronomine "qui", secundum quod placet dare diversas
diffinitiones exprimentes quid nominis illius nominis "punetus". Et ideo sicut hoc predicatum
"res divisibilis" prcdicatur de tali substantivo et per consequens de composito ex adiectivo et
substantivo, ita predicatur de puncto.'
are terminated per je ipsas.21 O f course, the term 'point' occurs everywhere
in the propositions of natural philosophy, and there is nothing wrong with
using the term 'point' in these propositions; we should merely realise that
in so doing we are but employing the term in place of more complex
expressions or propositions in which the term 'point' does not appear and
which thus reveal that we need not be committed to the existence of any
such thing as a point. 28
Ockham's attempted removal of such fictive elements as points and
instants parallels his removal of such connotative terms as 'motion' or
'time' from natural philosophy. For just as one must realise that there is no
separately existing thing called motion above and beyond the moving body
itself and the places it successively occupies or the forms it successively
possesses, so one must realise that there are no separately existing things
called points above and beyond terminated lines. 29 Further, just as one
accounts for the fact that there is no thing directly corresponding to the
term 'motion' by reducing all propositions in which this term occurs to
another proposition or other propositions in which only terms standing for
moving bodies and other real things occur, so one can provide for the fact
that there is no thing corresponding to the term 'point' by reducing all
propositions in which this term occurs to other propositions or expressions
in which only terms standing for lines of one sort or another occur. 30 The
technique of propositional analysis employed by Ockham in such cases
preserved a particularist ontology that allowed the existence of only
individual permanent things. 31
Ockham's influence
Although the sparseness of Ockham's ontology for such conceptions as
motion was opposed by some of his successors,32 his denial of the reality of
27. Ockham 1930, p. 32: 'Linea sufficienter est continua et finita per naturam propriam sine omni alia
re addita sibi.' Note that since lines are themselves indivisible in two dimensions and surfaces are
indivisible in one, Ockham must perform the same 'reduction' of lines to surfaces and surfaces to
bodies as he docs for points to lines. Then, with three-dimensional bodies, he will have reached
the level ofindividual res permanentes, of really existing things. Of course, parallel reductions must
be carried out for instants relative to time intervals.
28. Ockham 1930, p. 44.
29. Although lines themselves are no more things than are points (see note 27), once we have
performed the same reduction operations for lines and rhen for surfaces, we are at the level of real
things.
30. On Ockham's analysis of the term 'motion' see Shapiro 1957.
31. On such 'propositional analysis' in general in fourteenth-cenrury philosophy, see Murdoch
forthcoming a and b.
32. Ockham's influence is most evident from the fact that after him the question about the 'reality' of
Infinity and continuity SIS
such entities as points and instants was more warmly received. At Paris,
John Buridan and Albert of Saxony left no doubt of this in following and
developing Ockham's denial of the reality of indivisibles, while closer to
home the Mertonians Thomas Bradwardine and William Heytesbury
assert an equally firm Ockhamist position in this matter.33 Thus, at the end
of his Tractatus de continuo, Bradwardine draws things together concerning
the indivisibles he has been dealing with throughout his treatise in the
explicit claim that 'there are no surfaces, lines, or points at all', adding the
equally Ockhamist corollary that 'a continuum is neither continued nor
rendered finite by such (indivisibles), but by its very self'.3* And
Heytesbury in his Regulae solvendi sophismata allows himself the rare onto-
logical remark that there are no such things in nature as instants, or even
time or motion, even though he alludes to such things in the Regulae at
every turn in his examples and arguments.35
The rise of indivisibilism

In Book VI of the Physics, Aristotle had set forth several arguments
against the possibility that continua are composed of indivisibles. Almost
all medieval philosophers accepted Aristotle's arguments, but a minority
opposition arose at the beginning of the fourteenth century. Indivisi-
bilists or atomists such as Henry of Harclay, Walter Chatton, Gerard of
Odo, and Nicholas Bonet - to mention only those most frequently cited
and criticised - began to maintain that continua were indeed composed of
motion was most often raised in the Ockhamist form of whether motion was some entity above
and beyond permanent things, but his conclusion that there were only such permanent things
involved in motion or change was far from universally accepted. To cite only the most notable
instance of such disagreement, John Buridan accepted Ockham's conclusions with respect to the
motions of alteration and augmentation or diminution, but disagreed with him concerning local
motion (see Murdoch and Sylla 1978, pp. 217-18).
33. For the relevant references to Buridan and Albert, see Zoubov 1961 (which contains an edition of
Buridan's De puncto). Note that Buridan's and Albert's treatments of points occur in works of
natural philosophy, while Ockham's is most fully developed in a theological work because of
Ockham's convicrion that an analysis of quantity was required in order to give an appropriate
account of the Eucharist. In this instance theology appears to have provided an important
starting-point for philosophy.
34. Thomas Bradwardine, Tractatus de continuo (MsTorun R.4 0 ,2, p. 192): 'Superficicm, lineam sive
punctum omnino non esse. Unde manifeste: Continuum non continuari nec fmitari per talia, sed
seipso.'
35. Heytesbury 1494b, f. 26r: 'Quid autem instans sit in rerum natura et qualiter continue sit aliud
instans et aliud, longe est alterius perscrutationis; et multafigmentafalsa admittit modus loquendi
hominis dc instanti, tempore, et motu propter breviloquium et mentis conceptum facilius
exprimendum, quia in rerum natura non est aliquid quod est instans ut instans nec tempus ut
tempus aut motus ut motus, sicut nihil est Sortes prout ipse est homo albus, nec aliquid est Plato
prout ipse est disputaturus eras aut prout ipse debet hodie respondere.'
indivisibles.36 In most instances the component indivisibles were taken to

be extensionless, whether an infinite or only a finite number of them was
thought to compose this or that continuum. Unfortunately, almost all this
indivisibilist literature is devoted to arguing against the Aristotelian posi-
tion and to establishing that continua can be composed in this or that
fashion of indivisibles; very little is said that helps to explain precisely why
this current of indivisibilism arose in the first third of the fourteenth
century or what function it was held to serve. There seems to be no sign of a
resurgence of ancient physical atomism among these late medieval indivi-
sibilists, nor anything resembling a consciously atomistic interpretation of
mathematics. Yet perhaps two motives can be gleaned from the texts
themselves. The first is that indivisibilism may have resulted from attempts
to account for the motion of angels. In Book VI of the Physics (Ch. 10;
240b8-24ia26) Aristotle had proved that the motion of an indivisible
implies the composition of spatial and temporal continua out of indivisibles
and so had rejected the possibility of the motion of an indivisible. But
angels are indivisibles that move. 37 An even more specific motive is found
36. On this fourteenth-century development of indivisibilism, see Murdoch & Synan 1966; Murdoch
1974b. Chatton seems to be alone in holding that continua are composed of finite numbers of
indivisibles. Further, although almost all medieval indivisibilists held their indivisibles to be
extensionless, Nicholas Bonet was an exception in that, like Democritus, he opted for indivisibles
having magnitude. On Bonet, Chatton, and Odo, see Zoubov 1959. Thomas Bradwardine
includes both Harclay and Chatton in the classification he gives in his Tractatus de continuo of the
various kinds of indivisibilism: 'Pro intellectu huius conclusionis est sciendum, quod circa
compositionem continui sunt 5 opiniones famose inter veteres philosophos et modernos. Ponunt
enim quidam, ut Aristoteles et Averroys et plurimi modernorum, continuum non componi ex
athomis, sed ex partibus divisibilibus sine fine. Alii autem dicunt ipsum componi ex indivisibi-
libus dupliciter variantes, quoniam Democritus ponit continuum componi ex corporibus in-
divisibilibus. Alii autem ex punctis, et hü dupliciter, quia Pythagoras, pater huius secte, et Plato et
Waltherus modernus, ponunt ipsum componi ex finitis indivisibilibus. Alii autem ex infinitis, et
sunt biparti«, quia quidam eorum, ut Henricus modernus, dicit ipsum componi ex infinitis
indivisibilibus inmediate coniunctis; alii autem, ut Lyncul<Jniensis^, ex infinitis ad invicem
mediatis...' (MSS Torun, R. 4 0 2, p. 165; Erfurt, Amplon. Q° 385, ff. 25 v -260· There were, in
addition to the four mentioned above, many other supporters of indivisibilism in the fourteenth
century, some of whom are little known or anonymous. Still, at least two further proponents
should be mentioned: Nicholas Autrecourt (see the relevant text in Nicholas of Autrecourt 1939,
pp. 206-17, and the analysis in Weinberg 1948, Ch. 9) and John Wyclif (see the text in Wyclif
1899, Ch. 9), who had several indivisibilist followers at Oxford in the later fourteenth century.
Indeed, Wyclif's condemnation at the Council of Constance in 1414 included (guilt by associa-
tion!) what seems to have been the only medieval condemnation of indivisibilism (Hermann von
der Hardt (ed.). Corpus actorum et decretorum magni Constantiensis concilii, vol. 4, p. 406).
37. Thus in his Comm. Sent. Walter Chatton introduces his discussion of the composition of continua
as follows: 'Et quia non potest sciri de motu angeli utrum sit continuus vel discretus nisi sciatur
utrum motus et alia continua componantur ex indivisibilibus, ideo quero propter motum angeli:
Utrum quantum componatur ex indivisibilibus...' (MSS BN 15887,93'; Firenze Naz. conv. sopp.
C5.357, I871). Gerard of Odo's discussion of continua occurs in the context of a discussion of
God's ubiquity (Dist. 37 of Book I of his Comm. Sent.). Four questions are relevant to this context,
in the earliest of the indivisibilists, Henry of Harclay; for it was his view
that the composition of continua out of indivisibles was required by the
possible inequality of infinites.38 Nevertheless, these 'motives' do not seem
to tell the whole story; except for the two or three sentences in which they
are expressed, they receive no attention in the texts themselves. Indeed, it
appears equally attractive to suppose that these fourteenth-century in-
divisibilists were drawn to their views largely because their analysis of
Aristotle's arguments against indivisibilism uncovered loopholes in them.
Harclay's attempted refutation of Aristotle's 'touching' argument

Whatever may have been the occasion for the development of this in-
divisibilism, the first task each of its proponents had to face was the
refutation of the standard arguments against such a position, those of
Aristotle being the first in line. Foremost among Aristotle's arguments was
the one that proved that no continuum could be composed of indivisibles
not merely because indivisibles cannot be continuous, but because there is
no way in which they can even be in contact with one another. This is so,
Aristotle argued, because indivisibles can have no extremities or parts,
which they must have if they are to be in contact with one another; for if
one whole indivisible were somehow to be in contact with another
indivisible as a whole, the two could not constitute a continuum of a size
greater than a single indivisible (Physics VI, i; 23i a 2i- b 6). Thus, the
fourteenth-century indivisibilists reasoned, if continua are to be composed
of indivisibles, there must be some way in which they can touch or be
'connected to' one another. Harclay's solution is that they do touch whole
to whole but 'in respect of distinct locations' (secundum distinctos situs), since
in that way, he thinks, they could account for the increase in size (faciunt
maius) required for the continuum which they composed, something they
could not do if they were in contact as wholes 'in one and the same
location' (in eodem situ).39
one of which is 'utrum motus angeli habent partem aliquam simpliciter primam". He then
discusses the composition of continua as the central difficulty to be resolved 'pro solutione istarum
[quatuor)qucstionum'(MSS Napoli, Bib. naz. VII.B.25,234"-235r; Valencia Catcd. 139,120v). It
is noteworthy that Duns Scotus' influential (non-indivisibilist) treatment of the composition of
continua (see note 44 below) also occurs in the context of a discussioii of angelic motion.
38. Harclay, Questio de infinito et continuo (MS Tortosa Cated. 88, 86 v -87 r ): 'Contra hoc [viz., the
inequality of infinites] sunt omnia argumenta que probant continuum non posse componi ex
indivisibilibus; probant enim etiam quod in uno continuo non sint plura puncta quam in alio.'
39. Harclay, Questio de infinito et continuo (MSS Tortosa Catcd. 88,89'; Fitenze Naz. II.II.281, g S ' ^ » ) :
'Dico quod indivisibile tangit indivisibile secundum totum, sed potest hoc esse dupliciter: vel
totum tangit totum in eodem situ, et tunc est superpositio sicut dicit Commentator, et non faciunt
Bradwardine against Harclay

Harclay's attempted refutation of Aristotle's 'touching' argument against
indivisibilism did not go unchallenged. Thomas Bradwardine, for exam-
ple, overturned Harclay's contentions more impressively than did
Harclay's other critics. Appealing to geometry, Bradwardine grants
Harclay the Euclidean notion of superposition as a way of interpreting or
accounting for his notion of indivisibles touching 'in respect of distinct
locations' or being immediately next to one another.*0 But he proceeds to
show that all occurrences of superposition in geometry are systematically
disassociated from continuity; that is, no superposed geometrical magni-
tude can form a continuum or be continuous with the magnitude on which
it is superposed.41 And so even the most favorable interpretation of
Harclay's immediate indivisibles will not allow them to serve as com-
ponents of continua.*2
infinita indivisibilia plus quam u n u m . . . . Eodem modo dico ego superficies vel corpora, in quibus
sunt huiusmodi puncta, si essent applicata secundum eundem simm, non facerent aliquid maius. Et
ideo dico quod non propter indivisibilitatem quod unum indivisibile sic additum indivisibili non
facit maius extensive, sed quia additur ei secundum eundem situm et ( n o n ) secundum distinctum situm.
Si tamen indivisibile applicetur immediate ad indivisibile secundum distinctum situm, potest magis
facere secundum situm.' Gerard of O d o has a similar notion of touching indivisibles; see Murdoch
1964, pp. 4 3 1 - 5 .
40. Euclid appeals to superposition in proving the congruence o f figures in several basic theorems (I,
4,8; III, 24). The technique is to apply or superpose one of the figures upon the other. Although
Bradwardine uses the notion of superposition in a sense that is faithful to Euclid, the term
'superpositio' also had other meanings (deriving basically from Averroes) in the medieval
tradition (see, for example, the use o f the term in the preceding note). O n all of this, as well as the
history of superposition in medieval mathematics, see Murdoch 1964.
41. Bradwardine, Tractatus de continuo, Conclusiones 9—13 (MSsTorun R. 4 0 , 2, pp. 158—60; Erfurt
Amplon.Q°385,19'—2i r ): '9 — Lineam rectam secundum totum vel partem magnam rectealteri
superponi et habere aliquod punctum intrinsecum commune cum ista non contingit. 1 0 - L i n e e
rectc unam partem magnam alii recte imponi et aliam partem magnam superponi eidem vel ad
latus distare ab ilia impossibile comprobatur. 11 - Unius recte duo puncta in alia continuari et per
partem eius magnam superponi eidem vel ad latus distare ab ilia non posse. 12 — Linee rectc unam
partem magnam recte alteri superponi et [ad!] aliam ad latus distare ab ista est impossibile
manifestum. 13 — Unius recte duo puncta alteri superponi vel unum imponi, aliud vero
superponi, et magnam eius partem ad latus distare ab ista non posse contingere.'
42. Bradwardine intended his arguments to be valid not only against Harclay, but against any brand
of immediate indivisibilism. In addition, he gives many effective arguments against any view
maintaining that continua are composed of a finite number o f indivisibles (and hence against
another English indivisibilist, Chatton). He also believed that he had overturned the position
maintaining that continua are composed of an infinite number of indivisibles between any two of
which there is always another (something Bradwardine himself refers to - see note 36 above - as
composition ex infinitis [indivisibilibus] ad invicem mediatis), but he was not successful in this
regard. Incidentally, Bradwardine's ascription of this last brand of indivisibilism to Grosseteste ('ut
Lyncul<[niensisy) is not accurate. Grosseteste held that an infinity (indeed, differing infinities) of
indivisibles were contained in differing continua, but he did not hold them to be composed of
indivisibles (see Robert Grosseteste 1963, pp. 91-5).
Mathematical arguments against indivisibilism

Some medieval arguments against indivisibilism were not Aristotelian,
and the most interesting and formidable of them were essentially mathe-
matical. Such arguments can be found in the Metaphysics of Algazel*3 but
also, and more importantly, in John Duns Scotus.*4 These arguments as
well as all the variants which they generated in later philosophers, might be
characterised basically as attempts to reveal the absolute incompatibility of
indivisibilism and geometry. For example, parallel lines drawn from each
indivisible in one side of a square to each indivisible in the opposite side
will destroy the incommensurability of the diagonal with the side, since
these parallels will meet the diagonal in as many, and only as many,
indivisibles as they meet the sides. Similarly, the construction of all the radii
of two concentric circles will entail the absurdity that they have equal
circumferences if both circumferences are composed of indivisibles.45
These arguments, too, had to be answered by the medieval indivisibilist,
but in attempting to answer them he was often forced to introduce
inappropriate material notions into the mathematics of the arguments or
even to reveal himself quite incompetent to deal with the mathematics at
all.46
Euclidean considerations against indivisibilism

These arguments against the indivisibilist composition of geometrical
continua were not the only mathematical factors in the medieval analysis
of continuity. A few scholars were also aware of the assumptions made
about continuity (and about infinity too) within Euclidean geometry itself.
Thomas Bradwardine, for instance, had in effect rested his whole case
against indivisibilism in his Tractatus de continuo on a series of geometrical
propositions he had established at the beginning of his treatise. Yet at the
end of his refutation he astutely asked just which 'axioms' concerning the
composition of continua were assumed in the first place by the very
43. Algazel 1933, pp. 10-13.

44. Duns Scotus, Opus Oxoniense, Lib. II, dist. 2, quest. 9.
45. Examples of these arguments can be found in Murdoch 1062, pp. 24-30; also Murdoch and
Synan 1966, pp. 254-6.
46. Thus, an indivisibilist might account for the intersection of lines in terms appropriate only to the
crossing or overlapping of something having physical magnitude like sticks, since one could then
explain how the parallel 'lines' cutting the diagonal of a square consumed more of the diagonal
(which they intersected 'obliquely') than they did when they cut the sides (which they intersected
at right angles). See, e.g., Murdoch and Synan 1966, pp. 259—62, where Chatton also shows his
lack of understanding of incommensurability when he tries to relate it to odd vs. even numbers.
58ο Natural philosophy
geometry he was using; by basing his refutation so completely on

geometry, had he perhaps begged the whole question? No, he replies, for
one can allow the composition of geometrical magnitudes out of in
divisibles of a certain sort and still prove all the standard theorems of
47
geometry.
Campanus ofNovara on curvilinear angles

A quite different and less unusual awareness of the continuity assumptions
in geometry can be found in the medieval comments on Euclid. The first
proposition of Book X of the Elements states that, given any magnitude A
greater than some magnitude B, we can, in continuing to subtract parts
from A, at some point reach a magnitude less than B. 4 8 The effect of this
proposition was to exclude infinitely large, infinitely small, and minimal
magnitudes. The thirteenth-century mathematician Campanus ofNovara
revealed himself to be quite aware of this effect, when, in preparing his
version of the Elements, he commented that certain magnitudes, namely
certain angles formed by straight lines and the circumference of a circle,
appeared to contradict this proposition. For if we consider (Figure i) the
rectilinear angle ABC greater than the curvilinear angle DBC, it is evident
that in decreasing or subtracting parts from ABC, we shall never reach an
angle less than DBC. 'Therefore it is clear', Campanus adds, 'that any
47. Bradwardine, Tractatus de continuo, (MS Torun. R. 4 0 2, p. 188): 'Posset autem circa predicta fieri
una falsigraphia: A vroys in commento suo super Physicorum (III, c. 31), ubi dicit, quod naturalis
demonsttat continuum esse divisibile in infinitum et geometer hoc non probat, sed supponit
tamquam dcmonstratum in scientia natural!, potest igitur impugnare demonstrationes gco-
metricas prius factas dicendo: Geometriam ubique supponere continuum ex indivisibilibus non
componi ct illud demonstrari non posse. Sed illud non valet, quia suppositum falsum. Non enim
ponitur inter demonstrationes geometricas continuum non componi ex indivisibilibus nec
dyalecticer indigetûr) ubique, quoniam ( n o n ) in 5*° Elementorum Euclidis. Et similiter, nec
geometer in aliqua demonstratione supponit continuum non componi ex infinitis indivisibilibus
medians, quia, dato eius opposito, quelibet demonstratio non minus procedit, ut patet inductive
seiend conclusiones geometricas demonstrate.' Bradwardine also claims that maintaining that
continua were composed of indivisibles immediately next to one another would cause difficulties
in proving the fourth and eighth propositions of Book I of Euclid, since these propositions appeal
to superposition in their proof and superposition is the geometrical notion used by Bradwardine
to interpret the immediacy of indivisibles (see Murdoch 1964, p. 440). What Bradwardine was
doing in this place and in the passage quoted above amounts to probing the logic of assumptions
made in mathematics and (in modern terms) to asking just which axioms were independent of the
system of geometry as he knew it and which were not.
48. This is a simplification of X, I. Euclid speaks not of subtracting parts in general from the greater
magnitude, but of subtracting more than half (or, in a corollary, just half) the difference between
the greater and lesser magnitudes. On the difficulties caused by this restriction in Euclid (as well as
by the fact that most medieval Latin versions of the Elements lacked another form [V, def. 4) of the
continuity assumption asserted by X, I ), see Murdoch 1963, pp. 240-51. These difficulties are not,
however, germane to the point being made here.
Β
Figure ι
rectilinear angle is greater than an infinite number of angles of contin

gence' (that is, curvilinear angles like DBC). Campanus had correctly
identified what in fart was an infinitesimal magnitude which does not obey
the continuity principle set down by Euclid's X, 1, and he appropriately
specifies that the fact that such curvilinear angles are not of the same genus
as rectilinear angles is the reason why the former do not obey this prin
ciple.49 What is more, in his comment on the proposition in Euclid in
which such curvilinear angles are mentioned, Campanus notes that they
violate yet another continuity principle. For the right angle EBC (Figure 1)
is greater than the angle of the semicircle EBD, while the acute angle EBA
is less than EBD. Yet one can decrease the right angle EBC by rotating the
tangent BC about Β toward BA and in so doing reach an acute angle less
than the angle of the semicircle EBD. But this means one has violated the
continuity principle that claims that if one moves from a magnitude (EBC)
greater than some second magnitude (EBD) to a third magnitude (EBA)
49. Euclid, Elementorum geometricorum libri XV cum expositione ... Campant in omnes (ed. Basel, 1558),
p. 244: 'Attendere autem oportet, quod huic propositioni videtur decimaquinta tenii con-
tradicere, proponens angulum contingen tiae minorem fore quolibet angulo a duabus lineis rectis
comento. Posito enim angulo quolibet rectilineo, si ab ipso maius dimidio demarur, itemque de
residuo maius dimidio, necesse videtur hoc toties posse fieri, quousque angulus rectilineus, minor
angulo contingentiae relinquatur, cuius oppositum 15 tertii syllogisat. Sed hi non sunt univoce
anguli, non enim eiusdem sunt generis simpliciter curvum et rectum.... Planum ergo est etiam,
quemlibet angulum rectilineum, infinitis angulis contingentiae esse maiorem.' (In modern terms,
because it does not obey the continuity principle in X, 1, an angle of contingence is a non-
Archimedean magnitude.)
less than the second magnitude, then one moves through a magnitude
equal to this second magnitude. 5 0 In a word, angles of contingence and
angles of semicircles need not obey this continuity principle.
Peter Ceffons and the measure of species

The relevance of this particular portion of medieval mathematics to
medieval philosophy derived not only from the fact that any number of
fourteenth-century philosophers and theologians were aware of the pe-
culiar properties of these curvilinear angles (and often aware of what
Campanus had said about them), 5 1 but also from the discovery that such
angles could be put to use in providing a scale of measure that would prove
effective when applied to the different perfections of radically distinct
species. The paradigm of this application of curvilinear angles was un-
doubtedly the one produced by Peter Ceffons. Lecturing on the Sentences at
Paris in 1348-9, Ceffons spent what seems to be an inordinate amount of
time explaining how his 'calculus of angles' could measure the 'distances'
between species. There is, to begin with, an infinite distance between
individuals of distinct species, between (say) a man and a donkey, a distance
that can be represented by the infinite excess of a rectilinear angle rep-
resenting man over an angle of contingence representing the donkey. For,
like distinct species, these angles are finite things allowing of mutually
infinite excess. Moreover, these angles of contingence can themselves be
increased or decreased (by drawing smaller or larger circles to the relevant
point of tangency) and they are still infinitely exceeded by rectilinear
angles (thus angle ABC (Figure 1) infinitely exceeds angle FBC as well as
angle DBC). This phenomenon corresponds to an allowable increase or
decrease of individuals within a given species while preserving the infinite
distance between individuals of distinct species. 52 Ceffons' fancies in this
regard appear to have drawn fire from at least one of his Parisian successors,
John of Ripa, who did not hold with the value of such 'exempla mathe-
matica'.53
50. Euclid, op. cit., p. 67, additio ad prop. Ill, 15 ( = 111,16 of the Greek): 'Ex hoc notandum, quod non
valet ista argumentatio: Hoc transit a minori ad maius, et per omnia media: ergo per aequale. Nec
ista: Contingit reperire maius hoc, et minus eodem: ergo contingit reperire aequale'. Campanus
goes on to specify that angles of contingence and angles of semicircles provide instances of
magnitudes which, relarive to rectilinear angles, violate these 'continuity principles'.
51. Thomas Bradwardine, Geometria speculativa (ed. Paris, 1503), Tract. 2, Ch. 3, concl. 6.
52. Something more of Ceffons' machinations with curvilinear angles and the perfections of species
can be found in Murdoch 1978a, pp. 61—3 and Murdoch 1969, pp. 242-6, the latter of which
contains other material (pp. 238—41) relating to the 'measure' of species. Ceffons' Comm. Sent, is
unedited. The basic article on him is Trapp 1957.
53. See the text in Murdoch 1969, n. 107.
Harclay, Alnwick, and terminist logic in the analysis of continua

Mathematics was not the only discipline to which the late medieval scholar
turned in analysing continua; logic provided an even more fruitful source
in this regard, especially when it was a question of analysing the limits one
ascribed to continuous magnitudes and processes.
An instance of this application of logic can be found in one of Henry of
Harclay's positive arguments in support of his indivisibilism. Although
men cannot perceive or know all the points of a finite line, Harclay argued,
God surely can. But if God does perceive all the points of a line, then either
he perceives that there is some line segment between the first point of the
line and every other point of the line, or he does not. If he does, then, since
points can obviously be assigned in any such segment, there are points in
the original line which he did not perceive, contrary to the hypothesis. If he
does not, then some point must be immediate to the first point of the line,
which proves that at least one continuum is composed of points in contact
- which, as we have seen, is Harclay's position.54 In criticising this argu-
ment of Harclay's, William of Alnwick insists that we must distinguish
two different propositions: (1) 'between the first point of a line and every
other point known by God of the same line, there is a mean point'; (2)
'there is a mean point between the first point and every other point
perceived by God of the same line'. Alnwick claims that the first prop-
osition is true while the second is false, because, in the first proposition, the
term 'mean point' has a type of 'merely confused supposition' (or in-
determinate reference) (as a result of the fact that 'every' precedes it in the
proposition), while 'mean point' in the second proposition has 'deter-
minate supposition'. On the principles of terminist logic employed by
Alnwick here, in the second proposition we can 'descend' from the term
'mean point' to the specification of either 'this (particular) mean point' or
'that (particular) mean point' or 'that other (particular) mean point', etc.
But this means that the second proposition is false, because there is no one
particular mean point that falls between the initial point of the line and all
other points in the line. On the other hand, the fact that in the first
54. Since in what follows I discuss William of Alnwick's refutation of Harclay's argument, I present
Alnwick's version of the argument here: it is in essence, if not in language, the same as Harclay's
original. 'Deus actualiter videt sive cognoscit primum punctum inchoativum linee et quodlibet
aliud punctum possibile signari in eadem linea. Aut igitur Deus videt quod inter hoc punctum
inchoativum linee et quodlibet aliud punctum in eadem linea potest linea intercipi aut non. Si
non, igitur videt punctum puncto immediatum, quod est propositum. Si sic. cum in linea media
possint assign ar j puncta, ilia puncta media non essent visa a Deo, quod falsum est.' (MS Vat. pal. lat.
1805, f. II')·
5»4 Natural philosophy
proposition the term 'mean point' has merely confused supposition means
that one cannot descend to such a disjunction of singular propositions -
which represents what as a matter of fact is the case, namely, that there is no
one mean point between the initial point of the line and all others. Hence
the first proposition is true. 55
Alnwick's criticism of Harclay is but a single instance of the application
of the doctrine of supposition to the analysis of continua, in particular to
the analysis of the order and denseness of elements within a continuum. If
we look to Ockham, for example, we find him employing the same
technique in dealing with continua and the infinite.56
55. I have given a resume o f William of Alnwick's reply to Harclay's argument (which he cites,
naming Harclay as its author): Alnwick, Determinatio 2 (MS Vat. pal. lat. 1805, 14'""):'Dico autem
bieviter quod ista est vera: "inter primum punctum linee et omnem alium punctum eiusdem linee
cognitum a Deo est linea media". Quelibet enim singularis est vera et eius contradictoria est falsa.
Et hoc ideo est, quia "linea media" in predicato sequens mediate signum universale stat confuse
tantum. Hec tamen est falsa: "est linea media inter primum punctum et omnem alium punctum
eiusdem linee visum a Deo", quia nulla est linea media inter primum punctum et omnem alium
punctum visum a Deo. N o n enim contingit dare aliquam talem lineam mediam, sic enim
mediarct inter primum punctum et seipsam; nec ilia linea esset visa a Deo. Et ideo, cum infertur: si
sic, igitur cum in linea possent puncta signari, et cetera, ibi "linea" stat particulariter; et ideo
arguitur a superior! ad inferius affirmative et sic facit fallaciam consequentis. Similiter arguitur a
termino stante confuse tantum ad eundem terminum stantem determinate sive particulariter, et
commutatur quale quid in hoc aliquid, et fit fallacia figure dictionis.' (In summarising Alnwick's
reply, I have spoken of'mean points' in place of the 'mean line' in which such points can be
designated, but the structure of the argument is precisely the same.) Cf. William of Alnwick 1937,
pp. 501-3.
56. Ockham, Exp. phys.. Ill, ad text. 60 (2o6 b i6), MS M e n . 293,71»: 'Est autem istis adiciendum quod
quamvis hec sit vera: "omni magnitudinc est minor magnitudo", hec tamen est impossibilis:
"aliqua magnitudo est minor omni magnitudinc". Ista enim est vera: "omni magnitudine est
minor magnitudo", quia est una universalis cuius quelibet singularis est vera. Hec tamen est falsa:
"aliqua magnitudo est minor omni magnitudine", quia est una particularis cuius quelibet
singularis est falsa. Et est simile sicut de istis duobus: hec est vera: "omnis homo est animal", et hec
falsa: "aliquod animal est omnis homo". Et ratio diversitatis est quia in ista: "omni magnitudine
est minor magnitudo" ly "minor magnitudo" supponit confuse tantum propter signum uni-
versale precedens a parte subiecti, et ideo ad veritatem sufficit quod ista magnitudine sit una
magnitudo minor et ilia magnitudine sit una alia magnitudo minor et sic de aliis. Sed in ista:
"aliqua magnitudo est minor omni magnitudine" ly "magnitudo" supponit determinate, et ideo
oportet quod aliqua una magnitudo numéro esset minor omni magnitudine, et per consequens
esset minor seipsa.' (Note that Ockham dispenses with the unnecessary 'aliqua' in the first
proposition; it was included in other treatments.) Ockham, Exp. phys,, VI, ad text. 52 (236b3), MS
Mert. 293, 133w: 'Tertio sciendum est pro sophisticis quod ista propositio de virtute sermonis est
concedenda: "ante omne mutari est mutatum esse", quia quelibet singularis est vera, scilicet ista:
"ante hoc mutari est mutatum esse" et "ante illud mutari" et sic de singulis. Sed ista falsa est:
"aliquod mutatum esse est ante omne mutari", quia est una indiffinita cuius quelibet singularis est
falsa. Nec sequitur: "ante { o m n e ) murari est mutatum esse, igitur aliquod mutatum esse est ante
omne mutari", quia in prima "mutatum esse" stat confuse tantum propter hoc quod sequitur
signum universale affirmativum mediate; in conséquente autem stat determinate, quia precedit
signum, et ideo est fallacia figure dictionis ei etiam fallacia consequentis.'
Contradictory opposites and the instant of change in Aristotle

Logic also provided the conceptual toolkit for another problem concern-
ing continua, one which received so much attention by late scholastics that
whole treatises were devoted to it. In this case, however, the problem
underwent a development within natural philosophy proper before it was
submitted to logical analysis. Two passages in Aristotle's Physics furnished
the raw material. In the fifth chapter of Book VI of the Physics
(235b32-236a27), Aristotle had raised the question whether there is a first
instant, a 'primary when', at the beginning of a continuous change or
motion and, secondly, whether there is such a primary element for the
completion of such a change, an instant when one can first truly say 'this
change is now over'. Aristotle's reply was that one should deny that there is
afirstinstant of a continuous change, but affirm that there is an instant at the
end of it - that is, a 'primary when' signifying that the change has been
concluded. The second Aristotelian ingredient also came from the Physics:
Book VIII, Chapter 8 (203b9-26). There Aristotle had dealt with the
'contradictory change' involved in (say) changing from being not-white
to being white.51 Here his decision was that the relevant first instant should
be assigned to the later segment of the time interval in question; that is,
there would be afirstinstant at which the changing subject is white. Unlike
his reply in Chapter 5 of Book VI maintaining that there is no first element
to a continuous change, this passage maintains that there is such a first
element.
De primo et ultimo instanti

These Aristotelian problems were expanded and systematised in the four-
teenth century in questions and treatises usually entitled De primo et ultimo
instanti, the most popular of them by far being Walter Burley's.58 In sum,
these treatises stipulated that since one must, with Aristotle, deny a first
instant at the beginning of a continuous change, one could, indeed one
must, designate a last instant before the beginning of the change, a last instant
of the period of the change's not yet occurring. Similarly, if one turns to
the ending of such a change, since one must, again with Aristotle, affirm a
first instant of the change's no longer occurring, one must also deny a last
57. Aristotle's example is of a change from being white to being not-white, but since the scholastics
reversed the example, I have done so here.
58. The text of Burley's treatise is printed in Shapiro 1965, but it is not a critical edition.
instant of the change itself. O n e can thus say that any continuous change or
motion is limited at its beginning by a last instant of its not-being and at its
end by a first instant of its not-being. But the restriction to not-being
means that the change or motion is extrinsically limited at both its ends,
since if it were intrinsically limited, that would entail the existence of first
or last instants belonging to the change itself, which is categorically denied.
The rules were different, however, when one turned from continuous
changes, res successivae, to res permanentes, a case of which is provided by the
example of something changing from not-white to white. 5 9 Here, follow-
ing Aristotle's decision, there was a first instant of being white, and thus a
(non-continuous) change that was intrinsically limited at its beginning. O n
the other hand, when that being white came to an end, it would do so
without there being a last instant of its existence; there would merely be
another first instant of the existence of whatever the subject changed into
(or even a first instant of the non-existence of the subject itself). Thus a res
permanens like being white was intrinsically limited at its beginning but
extrinsically limited at its end.
IncipitI desinit
Such an interpretation of Aristotle's views about the ascription of limits to
both successive and permanent things was, however, only one cornerstone
of the medieval literature dealing with the topic. And of all this literature,
unquestionably the most important was that which transferred the whole
problem to the context of logic. There was already existent in that context
a new and altogether suitable home for this problem of limits: namely, the
logical tradition of treating as problematic propositions in which the verbs
'begins' ('incipit') or 'ceases' ('desinit') occurred. 60
The new context made a difference. For in the literature of first and last
instants (de primo et ultimo instanti) one spoke directly about the things to
which limits were ascribed and about the instants or limits themselves; but
once the problem of limits was placed within the logical context of discus-
sions of 'incipit' and 'desinit' the elements of analysis became the terms
'incipit' and 'desinit' and the propositions embodying those terms. Instead of
59. Being white wasconsidered a respermanens because it had all of its 'parts' at once, and not, like a res
successiva (any continuous motion, for example), one after the other.
60. This earlier tradition is sketched in Kretzmann 1976. Note, however, that the earliest of the
literature constituting that tradition did not contain treatments of the problem of limits presently
in question. The introduction ofthat problem had occurred by the time of the discussions of
'incipit' and 'desinit' by William of Sherwood and Peter of Spain (for which see Kretzmann,
op. cit.).
asking, for example, whether Socrates' being white was limited at its
beginning by afirstinstant of his being white, one carried out an analysis of
the proposition 'Socrates begins to be white' ('Sortes incipit esse albus').
Accordingly, the considerations regarding first and last instants which
differentiated intrinsic from extrinsic limits were represented by different
kinds of logical 'expositions' that could be given of propositions involving
the terms 'incipit' and 'desinit.'61 The difference between the treatment of
these problems of limits in the literature of first and last instants and their
treatment in the literature of 'incipit' and 'desinit' might be described as a
difference between an object-language treatment and a metalinguistic
treatment of the same problems. This sort of difference can be seen in the
thirteenth century in the sections on 'incipit' and 'desinit' in the works of
William of Sherwood and Peter of Spain, and in the fourteenth century in
Thomas Bradwardine's treatise on the subject.62 But something addi-
tional happens when we follow this incipit/desinit literature further into the
fourteenth century, especially in the works of Richard Kilvington,63
William Heytesbury,6* and, at the close of the century, Paul of Venice.65
The keynote of this later literature was the construction of limit-decision
problems as sophismata, the motive behind this new turn apparently being
the formulation of problems complicated and ingenious enough to allow,
perhaps even compel, the hearer or reader of them to develop real expertise
in the application of the techniques and rules relevant to the resolution of
limit-problems, especially as cast in their incipitI desinit form.
De maximo et minimo
In the literature of first and last instants or of 'incipit' and 'desinit', the
fundamental continuum in question was that of time. For it is against the
'background' of an absolutely continuous time that the changes whose
limits were at stake occur, whether they are continuous or instantaneous.
But fourteenth-century scholastics also struggled with the problem of
ascribing limits in cases in which the continuum of time was not relevant.
Such cases occur when one attempts to set limits to powers or capacities,
61. See, for instance, the expositions provided in the Tractatus syncategorematum ascribed to Peter of
Spain. The whole section of this work dealing with 'incipit' and 'desinit' is conveniently available
in translation in Kretzmann 1976, pp. 122-8.
62. For Sherwood and Peter of Spain, see Kretzmann 1976. For Thomas Bradwardine's Tractatus de
incipit (unedited), see Murdoch 1979, pp. 125-6.
63. Richard Kilvington, Sophismata, currently being edited and translated by Norman and Barbara
Kretzmann.
64. William Heytesbury 1494b, cap. 4: 'De incipit et desinit'; ff. 23"—27r.
65. Paul of Venice I499, Prima pars, tractatus 18; ff. 65 v -70\
maxima that a power (of lifting, for example) can accomplish vs. minima it
cannot. Although the treatment of this kind of limit-problems was more
thoroughly physical than that of setting limits to the existence of things and
events in time, it too had its decidedly logical elements. But this story
66
cannot be told here.
Richard Swineshead's Liber calculationum
Instead, for the final example let us return from continua to the infinite.
We shall at the same time be turning from logic to something more
distinctively mathematical even though the treatment of the problems is
carried out in the format of sophismata. The work in question is Richard
Swineshead's Liber calculationum. The context is the measurement of the
overall intensity of a given quality (heat, for example) in a given subject
when varying intensities ofthat quality are distributed in certain specifiable
ways over the subject.67 One two-part example of the distribution of these
varying intensities brings Swineshead face to face with the infinite.
Imagine, as a first hypothesis, that a given subject is (say) hot in degree ι
over its first half, in degree 2 over its next quarter, in degree 3 over its next
eighth, in degree 4 over its next sixteenth, and so on in infinitum. As a
66. The problem had its origins in Aristotle's contention (De caelo, I, Ch. 11,281 * 1 -27) that a capacity
should be defined in terms of the maximum it can accomplish, a contention to which he added the
correlative information that, if some capacity can accomplish so much, it certainly can accom
plish less, while if it cannot accomplish so much, it surely cannot accomplish more. Aristotle did
not raise the question of a possible boundary between what a capacity can and cannot accomplish,
but Averroes moved in the direction of formulating such a question when he commented that a
capacity should be defined in terms of the maximum it can do, while its correlative incapacity
should be defined in terms of the minimum it cannot do. These notions of maxima and minima
became standard when, in later medieval analyses, the problem was transformed into one of
ascribing limits. Should a capacity such as Socrates' ability to lift things be limited by a maximum
weight he can lift or by a minimum weight he cannot lift? What criteria can be used to decidesuch
a question one way or another? Although some solutions to this problem were reasonably
physical (such as that of Richard Swineshead in Tract. 10 of his Liber calculationum), others', and
especially Heytesbury's (in Ch. 5 of his Regulae), bring logical considerations to bear. Thus,
Heytesbury's approach to the problem amounts to asserting that the kind of (ermi occurring in the
very formulation of the question will enable one to decide whether a maximum quod sic ox a
minimum quod non is appropriate. Thus, the prime consideration is what kind of quantifier
modifies the common noun (for example, 'weight', 'lifting power*, etc.) by means of which one
determines a maximum or minimum, and how many things (supposita) the quantified noun
stands for. Heytesbury even appeals to the verifiability of propositions in order to express a
number of the conditions that must obtain in order that a capacity have a range susceptible to
maxima and minima. On the whole problem, Heytesbury and Swineshead included, see Wilson
1956, Ch. 3.
67. On the context in question, see Murdoch & Sylla 1976, pp. 190-3. The example in question is
drawn from Tractatus II (De difformibus), of Swineshead 1520, ff. 6"-γ.
Degree 1
4 5
Subject A
Degr
Subject Β
Figure 2
whole, the subject is hot in degree 2. 6 8 That is, it is finitely hot as a whole
even though the heat throughout it increases infinitely. Secondly,
Swineshead takes the very same subject having varying intensities of a
quality distributed in infinitum over its decreasing proportional parts in the
manner just described and submits this subject to rarefaction in a specified
fashion. Rarefy the second proportional part (that is, its third quarter) by
any amount howsoever small, rarefy the fourth proportional part (that is,
its fifteenth sixteenth) by an amount half as much (or, as Swineshead says,
rarefy the fourth proportional part half as slowly as the second), and so on,
taking only every 2 n m proportional part and rarefying each twice as slowly
as the preceding one. Given this hypothesis, how hot is the whole? Answer:
nth
infinitely hot. For in selecting only the 2 proportional parts of the
subject, Swineshead had chosen parts whose intensities were, as he has
68. If we allow ourselves a geometric representation (Figure 2) which is not present in Swineshead,
we can see more clearly how he went about establishing his conclusion. Take subject A with
increasing intensities (beginning from an intensity of degree 1 over its first half) 'stacked up' over
its succeeding proportional parts; A clearly has just as much quality' as does subject B. For all we
have to do is redistribute the 'top layer' of Β over the succeeding proportional parts of A as
uniformly intense in degree 1 to obtain A with the increasing intensities ad infinitum. But Β as a
whole is clearly hot (or of whatever quality) in degree 2; therefore, so is A with increasing
intensities in infinitum. In modern terms, what we have is the convergent series J + 4 + i +
Yg + ... + -"y = 2, but it is anachronistic to speak of infinite series in Swineshead. See the text
of this segment of Swineshead's example in Clagett 1968, pp. 58-61.
specified in his first hypothesis, successively double one another. Therefore,

in deliberately stipulating that the rarefaction over these parts should be
successively 'twice as slow' (which means that the amount added to each
part is always half that added to its predecessor), it immediately followed
that, considering both the extension of each part added by rarefaction and
the intensity in that added part, the resulting contribution (no matter how
small) to the measure of the whole would be the same in each instance.
Since there were an infinite number of such added parts, the measure of the
whole immediately becomes infinite. 6 9
Within fourteenth-century natural philosophy, Swineshead's accom
plishment is impressive. But something more of the structure of what he
was doing can be seen if we compare both parts of his 'infinite measure'
example. Recall that in his first example Swineshead had shown that,
astonishing as it might seem, a subject whose quality increases in infinitum
can be only finitely intense overall. But then in his second example he
shows that one can take this very same finitely intense subject, change it by
a finite amount as small as one wishes, and it immediately becomes
infinitely intense. The switch from infinite to finite and then back to
infinite again was more than incidental; it was a feature of Swineshead's
indulgence in the (by then very fashionable) tradition of creating and
solving sophismata. In fact, he even refers to the problem he was dealing
with as a sophisma; the fact that it involved mathematics instead of logic
made no difference.
69. Swineshead 1520, f. η': 'Sit A tale cuius prima pars proportionalis sit aliqualiter intensa, et secunda
in duplo intensior et tripla in triplo intensior, et sic in infinitum (this much gives the initial
"convergent series" mentioned in n. 68 above). Tunc A est solum finite internum, ut predicitur.
Ponatur igitur quod secunda pars proportionalis A aliqualiter velociter rarefiet acquirendo
quantitatem, et quarta pars proportionalis in duplo intensior secunda parte proportionali in duplo
tardius acquirat de quantitate, et pars in quadruplo intensior ilia secunda in quadruplo tardius
acquiret quantitatem quam ilia secunda, et sic in infinitum; et nunc incipiat huiusmodi rarefactio
in illis partibus, stantibus omnibus aliis partibus non rarefactis nec condensatis. Et sequitur quod A
solum finite velociter rarefiet seu maiorabitur et secunda in duplo tardius et sic in infinitum. Et
quod A subito in infinitum intendatur patet, quia ante quodcumque instans habebit A infinitas
partes quarum quelibet tantum faciei ad intensioncm totius, sicut est hoc certum datum,
demonstrando illud quod faciet quantitas acquisita secunde parti proportionali; nam sicut aliqua
pars acquisita alicui parti proportionali erit minor quam quantitas acquisita secunde parti pro
portionali, ita eadem pars erit intensior quam secunda pars proportionalis et quam quantitas sibi
acquisita, ut ponit casus, addendo quod quelibet pars proportionalis in principio et continue sit
uniformis. Sed hoc est generaliter verum: quod si proportionaliter sicut una quantitas seu pars
alicuius est alia parte minor, ita eadem sit alia parte majori intensior, equaliter facient omnes
huiusmodi partes ad totius intensionem; ergo omnes iste quantitates acquisite infinite facient
equaliter ad totius denominationem, id est, ante quodcumque instans, et per consequens cum
erunt huiusmodi partes infinite, sequitur quod in infinitum facient ante quodcumque instans
datum.' We would, anachronistically, interpret Swineshead at this point by saying chat the series
he had consrructed was divergenr, for its general rerm is 2"/22 — 2"/2n~ ' K, where Κ represents
the rarefaction undergone by the first rarefied part.
Logic and mathematics

The absence of a recognised difference between logic and mathematics is
symptomatic of what transpired methodologically throughout fourteenth-
century natural philosophy of the infinite and the continuous, especially in
its more creative phases. Mathematical considerations on the one hand and
logical considerations on the other consistently provided concepts, doc-
trines, and techniques for the analysis at hand. In some instances, the
mathematics was drawn from, or at least rooted in, what had been inherited
from the Greeks. But in other cases, the medievals were forced to go well
beyond their mathematical heritage, perhaps even go against it, since a
'mathematics of the infinite' was certainly not a Greek predilection. A fair
share of the mathematical considerations that were employed in the
medieval treatment of infinity and continuity was, then, 'home grown'.
Moreover, almost all the logic that was employed was similarly a
medieval creation. This is surely the case with respect to the doctrine of
supposition, but it is perhaps less obviously true regarding the propo-
sitional analysis of problems of the infinite and the continuous. O f course,
the invocation of logic and logical considerations was far more frequent in
fourteenth-century philosophy and theology than an appeal to anything
mathematical. Yet judgements of this sort are probably guilty of separating
logical and mathematical techniques in a way in which they were not
separated in the minds of the fourteenth-century thinkers who applied
these techniques. Not only were the techniques often mixed in their
application of them, but it is frequently difficult to determine, as we have
seen, whether in a given instance it might be the mathematics that was
more determinative of the problem or the sophisma-mould deriving from
logic into which the mathematics was poured. In any event, the application
of both mathematical and logical notions was a hallmark of the analytical
bent of the fourteenth century, throughout its philosophy and its theology.
VIII
PHILOSOPHY OF
MIND A N D ACTION
20
THE POTENTIAL AND THE AGENT
INTELLECT
The Aristotelian origins of the doctrine

The conception of potential and agent intellect came to Western medieval
philosophy with the assimilation of Aristotle's theory of soul in his De
anima. In this text, intellective cognition was understood as the reception of
abstract concepts; therefore Aristotle conceived an intellective power ca-
pable of receiving which, in order to accomplish this function, had a purely
potential nature. In several passages of the De anima, this power is called
nous pathètikos (Lat. intellectus possibilis). The process of cognition starts,
however, with the data of sensitive cognition, which are particular and not
universal. Therefore the reception of abstract concepts must be preceded
by the abstraction of the universal content from sensible images. In order to
explain this action, Aristotle conceived of an active power which his Greek
commentators named nouspoietikos (Lat. intellectus agens). Neither the exact
functions of the two powers nor the relation between them was very clear
in the De anima. In some portions of the text, the intellect was described as a
part of the soul, which was defined by Aristotle as a substantial form of the
body, but other sections considered the intellect as having a nature different
from the soul-form of the body. This difference was especially stressed in
the case of the active power, which was at various points described as being
separate from the body and surviving death, or as inseparably joined to the
body. These and other inconsistencies in Aristotle's text opened the way
to different interpretations beginning with such Greek commentators as
Themistius and Alexander of Aphrodisias, and carrying on through
medieval Arabic 'Aristotelian' theories of the soul.
Avicenna and Averroes

Western medieval philosophers became acquainted with Aristotle's theory
along with its earlier interpretations, the most important and influential
among them being Averroes' and Avicenna's. Averroes and Avicenna had
accepted some fundamental theses common to different currents of
596 Philosophy of mind and action
'Aristotelianism': there exist both a potential and an agent intellect, each

of which differs in nature (i.e. potentiality/act) and in function (i.e.
passivity/activity). Avicenna endowed man with a potential intellect
which was the highest part of the soul, spiritual substance accidentally
joined to the body as its form. The agent intellect, however, was according
to Avicenna a separate, spiritual substance, the one and only intelligence
for all men - 'a treasury of concepts' - whose function was to transmit
intelligible forms - concepts - into the individual human mind (possible
intellect). According to Averroes' conception, on the other hand, both
intellects were separate substances, common to all humanity and con-
nected to each man by his individual sensitive cognition. When the
sensitive data came into contact with the potential intellect, they were
illuminated by the agent intellect, their universal content was separated
from their individual conditions, and the abstract concepts thus produced
were then received in the potential intellect. The whole process of intellec-
tive cognition proceeded, therefore, inside the separate psychic substances
and was only communicated to individual men.
The Christian Platonist tradition and the new Aristotelianism
The new texts presenting various interpretations of Aristotle's theory of

the potential and agent intellect came to the West in the twelfth and early
thirteenth centuries. At the time this process began, Western philosophy
conceived of the soul and the process of intellective cognition in a Christian
perspective against the philosophical background of a Platonist tradition
with Boethius and Augustine as its main sources. The human soul was
treated as a spiritual substance, having innate knowledge, illuminated by
God, composed of a ratio superior directed toward the spiritual world and a
ratio inferior tending toward lower beings, not dependent on the body, and
accomplishing its cognitive function all by itself. The encounter of the new
Aristotelian conception with the old Christian tradition produced a con-
flict which was never resolved in the Middle Ages and which provided an
important stimulus for the evolution and development of a new theory of
the soul, incorporating the conception of the potential and agent intellect.
Because of the strong position of this old Christian tradition, the first
attempts to assimilate the Aristotelian theory of the potential and agent
intellect favored those interpretations which stressed 'Platonic' residues in
Aristotle's conception, especially the transcendent nature of the intellect.
The potential and agent intellect 597
Dominicus Gundissalinus
The first Western assimilation of the theory of the intellect occurred in
Spain, in Toledo, where Avicenna's translator, Dominicus Gundissalinus,
wrote his De anima between 1142 and 1152. According to Gundissalinus,
the human soul is a spiritual substance, mover of the body and its perfection
or form. His conception of the intellect is basically Avicennian: the human
soul contains the potential intellect only, and the agent intellect is a spiritual
substance distinct from man. The function of both intellects is Avicennian
too: the potential intellect gives its attention to phantasmata and this
attention prepares it to receive enlightenment from the agent intellect.
Thus, the agent intellect illuminates phantasmata and creates abstract
concepts in the potential intellect. '
fohn Blund
No sequel to this Spanish twelfth-century reception of the Avicennian or
Aristotelian theory of the intellect is known until the beginning of the
thirteenth century when John Blund, student and professor at the Faculty
of Arts in Paris commented on Aristotle's De anima before 12 ίο. 2 The main
sources of his commentary were Augustine, Boethius, and Avicenna, and
these philosophers lay behind his conception of the soul as a spiritual
substance, simple, like a disembodied intelligence, but linked somehow
with the body, directed by its higher part toward contemplation of
spiritual beings and by its lower part toward sensible things. This lower
activity proceeds by way of abstraction, vaguely conceived in an
Avicennian sense. John Blund's sparse remarks on the intellect and its
cognition are very imprecise. He accepts the potential intellect as receiver
of abstract concepts and the active intellect as a distinct substance, a dator
formarum or angelic intelligence, which is necessary for the impression of
universal concepts in the possible intellect because sensible data initiating
3
the process of intellective cognition are 'too far' from the intellect.
William of Auvergne, Philip the Chancellor, and John of La Rochelle

The prohibition of 1210 against teaching Aristotle's texts dealing with
natural philosophy stopped the process of assimilation of the Aristotelian
1. Gilson 1929-30, pp. 82-3, 95, 87.

2. Edited by D. A. Callus and R. M. Hunt, 1970. See also Callus 1955, p. 432.
3. MS Vat. lat. 833, f. 99", 99 y b , 100", 100*, ioo v b .
59° Philosophy of mind and action
theory of the intellect in the Paris Arts Faculty for thirty years. However,
the process continued slowly in the Paris Faculty of Theology, where no
Aristotle texts of any kind had been taught and consequently no pro-
hibition was enforced. William of Auvergne, Philip the Chancellor, and
John of La Rochelle were the main links in this process. It started in 1230
with the De anima of William of Auvergne. William was aware of the
Avicennian conception of the intellect and rejected it in order to maintain
the traditional Christian theory of man. The process continued with Philip
the Chancellor (d. 1236), who incorporated Aristotelian concepts of the
potential and agent intellect in the traditional Christian theory of soul, and
it came to its full development with John of La Rochelle (d. 1245), who
also attempted to incorporate the new conception of the soul and the role
of the possible and active intellect into the tradition. In his Summa de anima
he created a new conception of the soul as a spiritual substance, composed
of quo est and quod est, which was at the same time a perfection of the body.
He also accepted the Avicennian agent intellect, identified, however, with
an angelic intelligence who illuminates human souls giving them knowl-
edge of supra-sensible beings. Nevertheless, he regarded not only the
potential but also the active intellect as internal to the human soul; the latter
has the nature of an actuality and is identical with divine light in the soul,
while the former is described in Aristotelian fashion as a tabula non scripta.
The function of both intellects in the soul is basically Aristotelian: the agent
intellect purifies sensible images of their individual characteristics and
produces universals for reception in the potential intellect. 4
The elements of the early assimilation

Thus was the first step accomplished in the Christian interpretation of
Aristotle: each man was endowed with an individual active intellect. The
agent intellect, which in the De anima was treated sometimes as separate
from the soul and sometimes as belonging to it, and which in Avicenna's
conception was considered a single distinct substance, was now divided
into two different beings: the first was part of the human soul, and the
second identical to an angelic intelligence and thus separate.
Roger Bacon
After thirty years of prohibition, Aristotle texts again appeared in the Paris
Faculty of Arts in 1240, and the first philosopher to comment upon
4. John of la Rochelle 1881, pp. 118-21, 161-3, 290-5.

Aristotle's works on philosophia naturalis was Roger Bacon, master there

between 1240 and 1247. His opinion on the soul and the potential and
agent intellect evolved between his Quaestiones and Quaestiones alterae on
Aristotle's Physics and Metaphysics: in the first phase he considered the agent
intellect as part of the human soul, in the second as a separate Avicennian
substance.
Bacon began by accepting the general Avicennian conception of the
soul as a spiritual substance which can be conceived in union with the body,
but he rejected his concept of the separate agent intellect. In this way he
took the second step on the road first taken by John of La Rochelle. Con-
sidered in its union with the body, the soul contains the potential and agent
intellect, the former being a lower, the latter a higher part of the soul. The
potential intellect is connected with sensitive powers, it is directed toward
spiritual beings, and its object is the material thing. The agent intellect does
not use sensitive powers, it is directed toward spiritual beings and knows
them by its own essence. Conforming to Platonic and Augustinian tradi-
tion, Bacon also accepts some kind of 'confused innate knowledge', ex-
empla in the soul. However, he weakens the meaning of this expression and
understands innate knowledge only as a disposition which inclines the
human intellect to cognition of truth, in other words, there is innate
knowledge only of the first principles of understanding.5
Roger Bacon's conception of the cognition and function of both intel-
lects is dependent on his interpretation of their objects. Individual beings
have a common nature, but this nature can be known by the intellect only
through universals, which are unclear, indictinct images. The agent intel-
lect illuminates phantasmata, dematerialised to some extent by sensitive
powers, liberates them definitively from material conditions and impresses
them upon the potential intellect as concepts, intentions, or universals. The
process of intellective cognition is, however, not limited to this abstraction,
because the agent intellect performs some kind of illumination through its
'innate knowledge' also, but this illumination has a merely regulative role:
the illumination by exempta is not a necessary condition of the cognition
but only of the veracity of knowledge.6
In the second phase of the evolution of his position, Roger Bacon
separates the agent intellect from the intellective soul which becomes
identical with the sole potential intellect. Thus the potential intellect is now
5. Roger Bacon 1909-40, VII, pp. 15-16, n o , 111-12; VIII, pp. 2-3,15-16; XI, p. 9.
6. Bacon 1909-40, VII, p. 31; VIII, pp. 4, 97.
6oo Philosophy of mind and action
considered from a twofold point of view - as spiritual substance, or as a

form of the body. Considered as substance, it has its innate confused
knowledge and is able to know spiritual substances. The potential intellect
as a form of the body is not able to know spiritual substances, although it
was created with innate knowledge, it has lost it as a result of separation
from God and become a tabula rasa with no knowledge at all. The agent
intellect, a distinct spiritual substance, gives its confused knowledge to the
potential intellect by means of impressed intelligible species. Having thus
obtained its confused knowledge, the potential intellect is enabled to
proceed to further cognition.7 The concept of the cognition and function
of the potential intellect is, however, not elaborated in this second phase of
Bacon's developing position. The conception of the intellect explained in
Opus maius and Communia naturalium between 1265 and 1268 belongs to
another period of his scientific activity, and represents a type common in
Augustinian theory.
Other theories in thefirsthalf of the thirteenth century

An anonymous Commentary on the Ethics written in Paris before 1250
presents an opinion similar to the first phase of Bacon's evolution, but it
seems to eliminate any kind of collaboration between the potential and
active intellect.8 Peter of Spain's Scientia de anima, written probably
around 1250 in Salerno or Portugal, conceives of the soul as containing the
potential and agent intellect, but accepts a second agent intellect distinct
from the soul in accordance with Avicenna's model.
The Augustinian - Aristotelian synthesis

The new Aristotelian material became well known by the second half of
the thirteenth century. Elements taken from Aristode and Avicenna, as
assimilated by theologians in Paris, were welded with the Platonic and
Augustinian Christian tradition into the synthesis of Bonaventure. A deep
analysis of the texts of Aristotle and his commentators undertaken with the
aim of modernising Christian thought gave rise to the works of Albert the
Great and Thomas Aquinas. Scholarly interest in original Aristotelian
ideas, clarified by his commentators, gave rise to the work of Siger of
Brabant.9 The work of Albert, Thomas, and Siger will be the subject
7. Bacon 1909-40, X, p. 11; XII, pp. 73, 74; XIII, pp. 7, 8, 10, 11.
8. Lottin 1957, Vol. 1, pp. 512—15.
9. Kuksewicz 1968, pp. 63-6.
of the next chapter: meanwhile something should here be said about

Bonaventure.
Bonaventure
Bonaventure did not approve of the infiltration of Aristotle into Christian
tradition, and his theory of the potential and agent intellect explained in II
Sententiarum and Breviloquium between 1250 and 1257 is only a minor part
of his Augustinian conception of the soul, almost free of Avicenna's
influence and including some elements of Aristotle. There is no separate
agent intellect, because the sole illuminative function belongs to God, who
illuminates souls giving them the possibility of true knowledge and re-
gulating their intellective activity. (See the contribution to this volume by
J. Owens.) The human soul possess its 'higher' and 'lower face'. By means of
the former it achieves the cognition of itself and supra-sensible beings, and
in this field of activity there is no need for Aristotelian concepts of the
potential and agent intellect. However, when describing the cognition of
material objects, Bonaventure introduces Aristotelian terminology. The
'lower face' of the soul, which is connected with the material world,
operates through the potential and agent intellect which are understood
not as different faculties or parts but as two kinds of activity of the same
substance- the soul. The conception of the potential and the agent intellect
is, however, far from Aristotelian. The potential intellect does not have the
nature of pure potentiality, because, according to Bonaventure, it not only
receives but also produces knowledge: it proceeds to abstraction based
upon particular images produced by the senses, consequently, it is active.
Even the process ofabstraction is not conceived of in the Aristotelian sense:
it is merely an act of attaching the human mind to a particular image of a
given object and not a process of purification from individual charac-
teristics. Therefore, there is no need to accept the active intellect as the
agent of abstraction; but Bonaventure endows the soul with this traditional
Aristotelian faculty and gives it the function of illuminating the possible
intellect as a necessary condition for the operation of this intellect.
30
SENSE, INTELLECT, AND
IMAGINATION IN ALBERT, THOMAS,
AND SIGER
Albert the Great

Albert the Great reveals the influence of Avicenna and Averroes in his
psychology, though he certainly does not agree with them on all points. 1
Although he maintains that sense as such is a material and passive power,
Albert admits that after it has been actualised by the sensible form, it can
make judgements. However, it appears to do so only through the common
sense. Albert rejects the argument that just as there is in the soul an agent
intellect which abstracts and thus actualises intelligible species so there must
be an active sense which abstracts and thus actualises the sensible. He replies
that it is something in the nature of the intellect which renders the
potentially intelligible actually intelligible, but it is something in external
reality, not anything in the powers of the soul, which renders material
things actually sensible. However, what is in the sense is certainly not the
form united in existence with matter as found in the external thing, but
rather an intention (intentio) or species of the material thing which enables
us to have sense knowledge of that thing. Since sense apprehends the
sensible object through such an intention, the first grade of abstraction is
I. On Albert's life, see Meersseman 1931; Weisheipl 1980, pp. 13-51. The most ambitious study of
his psychology is Schneider 1903-6. Other general monographs on the subject are Reilly 1934,
Pegis 1934, and Michaud-Quantin 1966. The impact of Albert's own psychological doctrines on
later medieval and Renaissance philosophy has received attention from Park 1980, pp. 501-35
and Mahoney 1980a, pp. 537—63. For Albert's relationship to Averroes and Averroism see Nardi
i960, pp. 108-50; Van Steenberghen 1966, pp. 289-306. Nardi argues that for Albert philosophy
has to do with interpreting Aristotle and the Aristotelian tradition, setting aside miracles and
theological principles so that we do not confuse philosophy and theology. He cites Albert, De
generatione et corruptione, I, tr. 1, c. 22, ad t. c. 14: 'dico quod nihil ad me dc Dei miraculis cum ego
de naturalibus disseram'. Nardi finds Albert frankly Averroist on the possibility of the human
intellect's union with a separate intellect. While admitting the influence of Averroes on Albert's
interpretation of Aristotle, Van Steenberghen argues that Albert's Aristotelianism is orthodox or
Christian, since Albert does not hesitate to denounce Aristotle's errors and those ofother pagan
philosophers. He also notes that Albert himself insists that Aristotle was a human being, not a
god, and he could therefore err just as we can (Albert, Physica, VIII, tr. I, c. 14). For further
discussion on the thorny question of whether Albert's philosophical commentaries contain his
own philosophical position, see Kennedy 1959—60, pp. 121-3 and Kaiser 1962, pp. 53-62.
Sense, intellect and imagination 603
found in sensation, namely, separation from the matter of the external

thing. However, in so far as sensation is of actually existing sensible things,
which are individuated by matter, there must remain in sense apprehension
a reference to matter as present and to the individuating conditions of
matter. The other three grades of abstraction and apprehension occur in the
internal senses and the intellect. 2
The internal senses provided Albert with a topic of discussion in several
of his works, and scholars have sometimes seen inconsistencies in these
accounts. It has been argued, however, that two principles unite these
discussions, namely, the principle of the 7 grades of abstraction and the
theory of animal spirit and its role in the brain. 3 In his De homine, Albert
classifies the common sense, along with the external senses, as a power of
the sensible soul which apprehends what occurs externally (apprehensiva
deforis), while elsewhere he follows Avicenna and lists it as one of the
internal senses (virtutes interiores). The common sense enables us to be aware
of the operations of our external sense, such as seeing and hearing, and also
to make comparisons between the objects sensed by the different external
senses.4 With imagination we advance to the second grade of abstraction,
for imagination retains the images of sensible things when the latter are no
longer present, though it does not abstract from reference to conditions of
matter, that is, uniquely individuating characteristics. It is also imagination
which enables us to form an image before our interior eyes (prae oculis
interioribus facere imaginem) and to prepare images for the use of phantasy
and intellect, which represent the third and fourth grade of abstraction
respectively. Only on the level of the intellect is there apprehension of
2. Albert, De anima, II, tr. 3, c. 1, 3, 4 and 6, pp. 96a-98a, loob-l02b; De homine, q. 34, a. 1, pp.
294a-296a; a. 3, p. 303a. Steneck 1980, pp. 263-90. carefully examines the physical basis of the
grades of abstraction in the sense organs and the brain. For a detailed analysis of the four grades of
abstraction according to Albert, see Dähnert 1934, pp. 27-90. Albert considers the sensible species
which is generated in the sense organ to be a similitudo caused by the sensible object, an imago of
that object. By bringing about an intentio in the sense organ, the sensible object provides us with
knowledge of itself. See De homine, q. 45, a. 2, p. 4i4ab.
3. See Steneck 1974, pp. 193-211. Steneck argues (p. 209) that these two principles set up a graded
hierarchy among the internal senses. The principle of greater abstraction of the form from the
object is connected to the doctrine of spirit in that the higher internal sense is located in a more
subtle spirit in the brain. Albert shows far greater interest than Aquinas in establishing the
relationship between the internal senses and the brain. See also Noble 1905, pp. 91—101; Steneck
1980, pp. 275-8; and Dewan 1980, pp. 291-320. Both Schneider 1903-5, pp. 154-7, and
Wolfson 1935, pp. 116-20 and 1973, pp. 197-301, have discerned discrepancies in Albert's
various accounts of the internal senses. But see Klubertanz 1952, pp. 134-44.
4. Albert, De homine, q. 19, p. 164; q. 37, p. 323; De anima, II, tr. 4, c. 7, p. 157a. Cf. Steneck 1974,
p. 198; Schneider 1903-6, pp. 157-8. See also Ryan 1951, pp. 56-73.
6od\ Philosophy of mind and action
essences which have been separated from all material conditions so as to be

common to many and thus universal.5
Despite his interest in Avicenna and Averroes, Albert insists that both
the potential and the agent intellect are parts of the human soul and are
therefore multiplied according to the number of individual human
beings.6 While Albert had denied the need of an agent sense, since the
sensed objects themselves serve as agents, he argues that an intellect acting
in a universal manner (intellectus agens universaliter) is required in each
human soul. Phantasms provide only knowledge of individuals, and they
therefore cannot move the potential intellect to the universal knowledge
proper to it.7 Consequently, the agent intellect must abstract both from all
material characteristics which individuate forms and also from the par-
ticular itself8 There are two tasks which the agent intellect must perform,
namely, to abstract intelligible forms from phantasms and to illuminate the
potential intellect directly. This means that when the universal intelligible
species are received by the potential intellect, which then achieves intel-
lectual knowledge, they are received in the light of the agent intellect.9
Although Albert rejects the view that the agent intellect is a transcendent
being, he does not exclude God as a causal factor in human cognition.10
5. Albert, De anima, II, tr. 3, c. 4. pp. i o i b - i 0 2 a ; tr. 4, c. 7, p. I57ab; De homine, q. 37, a. 4,

pp. 328a-329b;q. 38, a. I, p.331a. O n phantasy and estimation, see Schneider 1903-6, pp. 161-5;
Dähnert 1934, pp. 6 6 - 9 ; Michaud-Quantin 1966, pp. 6 3 - 7 6 . Presumably phantasy is the sense
power which enables us to imagine things which in no way exist outside the mind, as when we
invent (fingimus) chimeras, goat-stags and other monsters simply by imagining them. See De
anima. III, tr. I, c. 4, p. 169b; De homine, q. 38, a. I, p. 332a. Albert shows keen interest in
explaining sleep and dreams in terms of the internal senses and the 'spirit' in the brain. See De
homine, q. 38, a. 5, p. 335b; q. 43, a. 1, pp. 366a and 368a; q. 44, a. 1, p. 4043b, a. 3, p. 406b; a. 4,
p. 4073; q. 45, a. 1, p. 412a and a. 3, p. 417a.
6. Albert, D e homine, q. 55, a. 4, part 1, p. 470a; a. 6, p. 476a; De anima, III, tr. 2, c. 19, pp. 203b-ao5b.
In his early De homine, q. 55, a. 3, p. 466, Albert believed that Averroes held that the agent intellect
was not a separate substance. Albert later realised that Averroes in fact taught that it was. For an
explanation of why Albert may perhaps have been misled and also an account of how Albert
transposed key concepts o f Averroes' psychology into his o w n , see Miller 1954, pp. 5 7 - 7 i -
7. De homine, q. 55, a. 1, pp. 455b-456a. See also De natura et origine animae, tr. i . e . 6, pp. i s b - i 6 b
for reference to the intellectus universaliter agens within the human soul. When Albert refers to God
by this same term he sometimes adds that God is such simpliciter, since he alone illuminates all else
which exists and is not himself illuminated by anything else before him. See De causis etprocessu
universitatis, I, tr. 2, c. 1, p. 388a.
8. Albert, De intellectu et intelligibili, I, tr. 2, c. 1, pp. 49ob~49ia; Summa Theol., II, q. 14, p. 193a. On
abstraction or separation from the 'appendicies* of matter, see De anima, II, tr. 3, c. 4, pp. i o i b -
102a; III, tr. 2, c. 13, pp. I95b-iy6a; De homine, q. 57, a. 2, p. 491a; q. 61, a. 1, pp. 5i9b-aob.
9. Albert, De anima, III, tr. 2, c. 19, p. 205b; III, tr. 3,c. 11, p. 221 ab; Dénatura et origine animae, I, c. 7,
p. loa; De intellectu et intelligibili, II, c. 3, pp. 5o6a-5o8a; I Sem., d. 3, a. 29, p. I30ab.
ίο. Albert, I Sent.,d. 2, a. 5, pp. 59—60; Summa Theol., I, tr. 3, q. 15, a. 3, pp. 1 lob—11 ia. Albert rejects
making God an agent intellect in the Avicennian sense. See De homine, q. 5 5, ad 18 and 20, p. 468b.
The agent intellect in the human soul and its abstractive action are not
sufficient to account for our knowledge of natural objects — the light of the
uncreated intellect must also be operative. However, illumination from
God seems to be required in a very special way in the case of'divine' things,
that is, beings such as God and the angels which are wholly separate from
matter. These beings, which are studied by metaphysics, are known
through the intellect's direct union with the divine light and in complete
independence of phantasms and abstraction.11 Albert shows particular
interest in earlier discussions of such philosophers as Alexander,
Themistius, Avicenna, and Averroes regarding the union of the human
intellect with a separate intellect.12 He explains in some detail how the
potential intellect isfirstjoined to the light of the agent intellect operating
universally within the soul itself, then to the lights of the different intelli-
gences, andfinallyto the light of the divine intellect itself, which is the goal
of the human intellect's strivings.13
Thomas Aquinas
Albert's student, Thomas Aquinas, was also much interested in Avicenna,
Averroes, and the Greek commentators, though he makes a somewhat
more cautious use of them.14 Thomas maintains that sense is primarily a
passive power, one which is naturally suited to be modified or changed by
an external sensible object. In sensation there is both a material modifi-
cation of the sense organ and also a 'spiritual' or intentional modification
brought about by the production of a species (species) or intention
For discussion regarding Albert's doctrine of divine illumination, see Schneider 1903-6, pp. 342-
8; Michaud-Quantin 1955, pp. 73-5; Nardi i960, pp. 111-17; Johnston i960, pp. 210-11;
Kennedy 1962-3, pp. 23-37.
11. Albert, Summa Theol., II, tr. 4, q. 14, a. 2, part 4, ad 1, p. 196a.
12. Albert, De anima. III, tr. 3, c. 6-11, pp. 2l4b-223b. See also II, tr. 2, c. 5, pp. I83b-i84a.
13. Albert, De intellectu et intelligibili, II, c. 9-12, pp. 5l6a-52ib. See also De anima, III, tr. 3,c. n ,
pp. 22ib-222b. On this striking doctrine, see Schneider 1903-6; Nardi i960, pp. 147-50;
Kennedy 1962-3, pp. 28-37. The resemblance of Albert's doctrine to Averroes here is brought
out by Kennedy 1959-60, pp. 131-7 and Nardi 1958, pp. 127-37. Albert himself indicates how
close he is to Averroes on the topic of conjunction: 'Nos autem in paucis dissentimus ab Averroe,
qui inducit istam quaestionem in commento super librum de anima.' De anima. III, tr. 3, c. 11,
p. 221a, Un. 9-11.
14. On the life and works of Aquinas, see Walz 1962; Chenu 1064; and Weisheipl 1974. For this
philosophical sources, sec Callus 1957, pp. 93—174, and for his use of Themistius' paraphrases 011
the De anima, see Verbeke 1957, pp. ix-lxii. Information on the relation of his thought to
Avicenna and Averroes is provided by Goichon 1951, pp. 114-30; Vanstecnkistc 1953, pp. 457-
507;idem 1957, pp. 585-623;idem 1060, pp. 336-401; deContenson 1958, pp. 3-31; idem 1959.
pp. 31—4, 53, 63-4, 68, 82-3 and 93—4; Gardet 1974, pp. 419—48; idem 1976, pp. 139—49;
Anawati 1974, pp. 449-6S; Gomez Nogales 1976, pp. 161-77.
6o6 Philosophy of mind and action
(intentio), that is, a likeness (similitudo) of the sensible object. Besides this
passive aspect of sensation, there is also an active side: the individual sense
power is able to distinguish or judge among its own proper sensible
objects.15 However, like Albert, Thomas rejects the need for an agent sense
to render the material object fit to be sensed, arguing that sensible things
outside the soul are already in act, that is, actually sensible. On the other
hand, since things outside the soul are not already actually intelligible, an
active element as well as a passive element is needed in the intellect.16
In his analysis of the internal senses in the Summa theologiae, Thomas
refers to Avicenna's list of five internal senses, but his own discussion
reveals the influence of Averroes and also some original developments.
Thomas takes the internal senses to be four in number, namely, the
common sense, the imagination, the estimative or cogitative power, and
memory, all of which are located in the brain. Although he does allow that
the individual senses can make a judgement regarding their own proper
sensibles, for example, sight can discern white from green, he insists that a
separate sense power is needed to make a judgement of discernment in
regard to the objects of all five senses. This separate power, which is called
the common sense, also enables the animal or the man to be aware of the
activity of his external senses.17 The imagination or phantasy serves as the
storehouse of forms (thesaurus formarum), that is, phantasms, of sensible
things when the latter are no longer present to the external senses.18 When
15. Aquinas,Summatheol., I,q. 12, a.2;q.78,a.3-4;q.84,a.3;q. 85,a. i,ady,\-H,q.22,a.z,ady,ln

II Sent., d. 19, q. 1, a. 3. ad 1 ; In de anima, II, lect. 24; Quodl. VIII, q. 2, a. 1 ; De mala, q. 16, a. 8, ai 10.
For discussion regarding the intentionality of sensation, see Boyer 1925, pp. 97-116; Picard 1926;
Rohmer 1951, pp. 5-39; Van Riet 1953, pp. 374-408. On the judgements made by the senses and
by the common sense, see Garceau 1968, pp. 242-8 and Owens 1970, pp. 138-47. For Thomas all
knowledge involves the assimilation of the knower to the known, but this assimilation is brought
about only by means of a likeness (similitudo) of the thing known existing in the knower. See
Summa theol., I, q. 85. a. 2; S.C.G., II, c. 65 and 77; De ver., q. 2, a. 3. ad y, q. 8, a. 5; q. 10, a. 4 and 8.
The likeness (similitudo) serves as the principle of knowing a thing not by reason of its existing in a
knowing power but by reason of the relationship it has to the thing known, that is, as it is
representative of the thing (repraesentativa rei). See De ver., q. 2, a. 5, ed 7. For studies on Aquinas'
doctrine of intentio, see Gome2 Izquierdo 1924, pp. 169—88; Simonin 1930a, pp. 445—63; Rabeau
1938, pp. 13-22; Hayen 1954; Regis 1959.
16. Aquinas, Summa theol., I, q. 79, a. 3, ad 1; Q.D. de anima, a. 4, ad 5; Quodl. VIII, q. 2, a. 1; De ver.,
q. 26, a. 3, ad 4. See the appropriate remarks of Van Riet 1953, pp. 400-7. Thomas' target appears
to be William of Auvergne, who argued in his De anima, c. 7, part 3, p. 207 that just as no agent
sense is required between the sensibles and the senses, so no agent intellect is required between the
intelligibles and the material intellect.
17. Aquinas, Summa theol., I, q. 78, a. 4, c. and ad2; q. 87, a. 3, ad 3; Q.D. de anima, a. 13; In de anima, II,
lect. 13. For discussion see Garceau 1968, pp. 242—8.
18. Aquinas, Summa theol., I, q. 12, a. 9, ad2; q.78,a.4;q. 84, a. 6, ad2; Quodl. VIII, q. 2, a. 1; De ver.,
imagination operates in conjunction with the human intellect, it enables

human beings to form new combinations of the stored phantasms, as for
example the image of a golden mountain. It is the task of the common sense
to judge between the images found in imagination and external reality.
However, since the normal functioning of common sense is suspended in
sleep, we frequently are led to take our dreams for true likenesses of the
external world. On the other hand, the role of the estimative power is to
enable animals to perceive intentions not known directly to the external
senses, such as danger, and to form an instinctive judgement, such as that
they should flee. However, since man makes some rational comparisons
among these intentions and acts by a free judgement, this sense in man is
given the special name of the cogitative power.19 Thomas sometimes
identifies the corruptible 'passive intellect' (intellectus passivus) with this
cogitative power, at other times with the imagination, and on occasion
with the complex of the cogitative power, the imagination, and memory,
all of which prepare the phantasm for the intellect.20 Indeed, Thomas
explains the variation in understanding among human beings in terms of
the difference in development of these three internal senses.21
For Thomas both the potential intellect and the agent intellect are
powers of the human soul, and they are therefore multiplied according to
the number of human beings.22 Thomas denies, in opposition to Avicenna
q. 8, a. 5. Thomas rejects Avicenna's distinction of two imaginative powers, arguing that since
creative imagination appears only in man a distinct power is not required. For two strikingly
different approaches to Thomas' doctrine on imagination, see Brennan 1941 pp. 149—61 and
Kenny 1969, pp. 273-96. Thomas' terminology regarding sensible species and phantasms needs
to be carefully understood. He reserves the word 'phantasm' for the species found in the
imagination, the cogitative power and memory. See Summa theol., I, q. 89. a. 5; S.C.G., II, c. 73. It
is not used of the sensible species as found in the external senses or the common sense. In Summa
theol., I, q. 85, a. 1, ad 3, phantasms are called likenesses just after it is said that colours can impress
(imprimere) their likeness (similitudo) on sight. It would be a misreading, however, to take Aquinas
to mean that the species produced in the power of sight are phantasms.
19. Aquinas, Summa theol., 1, q. 78, a. 4; q. 83, a. 1; Q.D. de anima, a. 13. On Thomas's originality
regarding the cogitative power in comparison with Averroes, see Klubertanz 1952, pp. 276—82.
The central role of the cogitative power is also brought out by Peghaire 1942-3 and Naus 1959,
pp. 123 -6 and 192-8. Thomas allows for the possibility that some people will retain some use of
their common sense during sleep and so be able to judge that what they are seeing arc dreams.
However, he adds that they will be deceived in taking some similitudes for the real thing. See
Summa theol, I, q. 84, a. 8, ad 2; De ver., q. 12, a. 3, ad 2; q. 28, a. 3, ad 6. See n. 5 above.
20. For the relevant texts, sec Klubertanz 1952, pp. 161,183-90, 194-6 and 244-5. Hamlyn 1961,
p. 48 confuses matters somewhat by referring to the potential intellect as the 'passive intellect'.
The distinction between them goes back to Themistius and Averroes.
21. Aquinas, Summa theol., I, q. 76, a. 5; q. 85, a. 7; In de anima, II, lect. 19.
22. Aquinas, De unitate intellectus, c. 1 and 4; Summa theol., I, q. 76, a. 2; q. 79, a. 1 and 4; Despiritualibus
creaturis, a. 10, ad 4; Q.D. de anima, a. 3 and 5; S.C.G., II, c. 59,73,75-8; Comp, theol., I, c. 85-9.
6o8 Philosophy of mind and action
and Averroes, that Aristode ever taught that either intellect was a separate
substance.23 Moreover, he rejects those accounts of intellectual cognition
which involve representations that are innate to the soul or that are sent
into the soul from a higher being, and he likewise denies that the human
24
intellect knows material things in the exemplars of the divine mind. The
potential intellect stands wholly in potency to the intelligible, but the
natures of the sensible things in the world about us are not of themselves
intelligible. It is the task of the agent intellect to render intelligible the
phantasm or image of the external thing. This it does by abstracting
the nature or form from matter and the individuating conditions of the
particular individual which is represented in the phantasm. 25 Since a
phantasm in the senses and in imagination is material and individual, it is
incapable of producing a universal likeness of a nature or essence. The
agent intellect must therefore abstract the bare essence or quiddity of the
sensible thing from all the accidental material conditions presented in the
phantasm. 26 The resulting universal likeness, which is called the intel
ligible species, actualises the potential intellect, and the latter forms in turn
its own likeness of the abstracted nature, namely, the 'concept' or mental
word. 2 7 Although Thomas is clear that the intelligible species and the
concept, just like the phantasm, are not themselves the objects of cognition
but serve rather as the means by which one knows, later critics would argue
23. Aquinas, Summa theol., I, q. 76, a. 2; q. 79, a. 4-5; De unitate intellectus, c.i;De spititualibus creaturis,
a. 9— io;S.C.G., II, c. 74 and 78. However, in the early In IISent., d. 17, q. 2, a. ι, Thomas appears
to accept that Aristotle held the agent intellect to be a separate substance, namely, the lowest of the
intelligences. This passage is analysed by Gilson 1926—7, pp. 111-13. See n. 33 below.
24. Aquinas. Summa theol., I, q. 84, a. 3-5; De ver., q. 10, a. 6; Q.D. de anima, a. 15.
25. Aquinas, Summa the al., I, q. 54, a. 4; q. 79, a. 3—4; q. 84, a. 2 and 6; q. 85, a. I; q. 86, a. I; Q.D. de
anima, a. 4. The phantasm is the similitudo rei particulars (Summa theol., I, q. 84, a. 7, ad 2). In Summa
theol., I, q. 85, a. 1, ad 4, Thomas emphasises that the agent intellect illuminates the phantasms in
order that they will be rendered fit to have intelligible species abstracted from them. For
discussion regarding the nature of this illumination and the interpretations offered in the Thomist
tradition, see Garin 1931, pp. 395-485. On the agent intellect as a lumen naturale, see n. 34 below.
26. Aquinas, Summa theol., 1, q. 85, a. I; S.C.C., II, c. 77; De spititualibus creaturis, a. 10, ad4 and 17; De
ver., q. 10, a. 6, ad 2 and 7; In de anima. III, lect. 8 and 10. On the various meanings of abstraction in
Aquinas, see Blanche 1934, pp. 237-51 • Santeler 1939, pp. 4-66 presents a detailed examination
of Thomas' doctrine on the agent intellect as found in his various works.
27. Aquinas, Summa theol., I, q. 85, a. 2, ady, I—II, q. 93, a. 1, ad2; De ver.,q. 4, a. 2; q. 10, a. 6; S.C.G.,
I.e. 53 and IV, c. 11; De pot. Dei, q. 8, a. i;q.g,a. 5; Comp, theol.. I.e. 38-9; QuoaV. V, q. 5, a. 2. On
the doctrine of the mental word in Thomas'thought, see Garin 1931, pp. 621-828; Paissac 1951,
pp. 101-236; Peifer 1952, pp. 132-79: Chenevert 1961, pp. 192-223 and 370-92; Lonergan 1967,
pp. 1-10,124-33 and 191-6. The presentation of Thomas' theory of knowledge and doctrine of
the mental word by Colish 1968, pp. 161-83 is not wholly accurate. See Bloomfield 1970,
pp. 119-22; Mahoney 1973b, pp. 258-62.
that it was dangerous to postulate such intermediaries in the process of

cognition since they appeared to veil reality from the mind. 2 8
A related doctrine of Thomas also became the object of later criticism,
namely, his insistence that the intellect as such knows universals directly
and individuals only indirectly. Since the intellect abstracts from the
individual as presented in the phantasm, it has to reflect on that phantasm if
it is to know the individual.29 Thomas himself points out that the nature of
a material thing can in fact be known completely only when it is known as
existing in an individual. And since we grasp the individual only through
the phantasms of sense and imagination, we must turn back to the phan
tasm. Indeed, Thomas insists that in this life we cannot actually understand
anything unless we turn in some way to phantasms. This can be seen in the
fact that when we want to understand something, we form phantasms to
help us do so, just as we offer others concrete examples in order to help
them to understand. One consequence of Thomas' position is that humans
can understand purely immaterial things such as God and the angels only
by comparison to things of which they do have phantasms. 30 Another
28. Aquinas, Summa theol-, I, q. 85, a. 2; q. 87, a. 3; 5.C.G., II, c. 75; De spiritualibiiscreaturis, a. 9, ad6;
De vet., q. 10, a. 8, ad 2; In de anima. III, lect. 8. The intelligible species can of course become the
object of knowledge but only secondarily, that is, by a reflection of the intellect back on itself and
its acts of thinking. Historians who have written about Thomas' theory of knowledge have
tended to emphasise that the intelligible species and the concept are merely the means in or by
which we know and are not the object of knowledge itself. See for example Gomez Izquierdo
1924, pp. 169-88;Garin 1931, pp. 692-828; Peifer 1952, pp. 165-79; Regis 1959, pp. 210-21 and
248-52: Carlo 1966, pp. 47-66; Owens 1974, pp. 189-205. However, some of Thomas' own
contemporaries opposed intelligible species precisely because they would be the object of
knowledge and block the intellect's grasp of extramental reality. See Henry of Ghent, QuodUbeta,
III, q. I ; IV, q. 7; V, q. 14; Peter John Olivi, Quaestiones in secundum librum sententiarum, q. 58 and
q. 74. Olivi's remark that anything between the gaze of the intellect and the external object, such
as an intelligible species, would 'veil' (velaret) the object may seem familiar to the contemporary
reader. Similar language is found in Russell 1910-11, p. 119 and Bennett 1965, pp. 2-3.
29. Aquinas, Summa theol., I, q. 86, a. 1; De ver., q. 2, a. 6; q. 10, a. 5; 5.C.G., I, c. 65; Q.D. de anima,
a. 20. ad 1; In de anima. III, lect. 8; Quodl. VII, q. I, a. 3. In these texts Thomas himself connects
abstraction and the need to reflect back on phantasms with his view that individuation is through
matter. Not surprisingly, there has been disagreement as to what Aquinas meant by the reflection
back on the phantasm. Among the more helpful studies arc Allers 1941, pp. 95-163; Isaac 1948,
pp. 338-40; Klubcrtanz 1952b. pp. 135-66; Lonergan 1967, pp. 25-33 and 1S9-71; Kenny 1969,
pp. 288-94. Thomas' doctrine that the intellect only knows the universal directly and can know
sensible individuals only indirectly by a reflection on the phantasm (Summa theol., I, q. 86, a. 1)
was already the subject of attack by thirteenth-century Franciscans such as William of la Mare and
Matthew of Aquasparta. See Simonin 1930b, pp. 289—303.
30. Aquinas, Summa theol., I, q. 12, a. 2,11 and 12; q. 32, a. 1 ; q. 84, a. 7-8; q. 88, a. ι -2; q. 117, a. 1 ; Dc
ver.,q. 10, a. 2, ad7, and a. 11; S.C.G., II, c. 73; III, c. 51-2; IV, c. 1. Actual sensing and thinking
are necessary conditions foi self-consciousness. See Summa theol., I, q. 14, a. 2, ad3; q. 87, a. 1 and
3;/n J//S«il.,d. 23, q. I, a. 2, ad y, Dever.,q. 10, a. 8; Q.D. de an.,a. 16,ad 8. For a helpful analysis
of how the soul knows itself see Gardeil 1934, pp. 219-36.
6ιο Philosophy of mind and action
consequence is that in order to explain how the disembodied soul after

death can know anything Thomas is forced to postulate that God sends
into the soul likenesses (similitudines) or species which are not phantasms
but which can provide the soul with knowledge of individuals.31
One of the most striking aspects of Thomas' psychology of knowledge
is the rejection of such key Augustinian doctrines as the active theory of
sensation and the need for divine illumination to achieve necessary and
certain knowledge. 32 Thomas also opposes some of the theses put forth by
those who pursued the path of Avicennised Augustinianism, to borrow
Gilson's useful term. 3 3 For example, it is clear that for Thomas God in fact
has no special role to play in man's achieving knowledge of the natural
order, even that knowledge which is necessary and certain. The 'natural
light' of the human intellect does not entail for Thomas either God as an
agent intellect or divine illumination.3* Unhke Albert, Thomas sees no
need to postulate any divine illumination to account for man's natural
knowledge, and unhke Bonaventure, he believes that the human intellect's
own abstractive abilities are sufficient to account for its necessary and
31. For an appreciation of the nuances of his thought on this thorny topic, see Aquinas, Summit theol,
I,q. 89, a. 1-8; Q.D. de an., a. 15 and 17-21; De ver., q. 19, a. 1-2, S.C.G., U.c. 80-1; Quodl. Ill,
q. 9, a. 1. For general discussion see Pegis 1974, pp. 131-58.
32. Aquinas rejects Augustine's active theory of sensation in Summa theol, I, q. 75, a. 4; q. 84, a. 6. For
a contrast of Augustine and Thomas on the nature of sensation, see Hessen i960, pp. 251-4 and
258-60. Moreover, Aquinas rejects the need for any divine illumination and also the possibility
that God himself is the first object of human cognition. See Expositio super De trinitate, q. 1, a. I and
3; Summa theol., I, q. 12, a. 11, ad y, q. 84, a. 5; q. 88, a. 3, ad 1-2; I-II, q. 109, a. 1; De ver.,q. 10,
a. 11 ; Quodl. X, q. 4, a. 1 ; De spiritualibus creaturis, a. 10, ad 8. For comment, see Grabmann 1931,
pp. 57-63; idem 1948, pp. 66-86.
33. For references to God as agent intellect, see Summa theol, I, q. 79, a. 4; In II Sent., d. 17, q. 2, a. 1;
d. 28, q. 1, a. 5; S.C.G.,Il,c. 85; Q.D. de anima, a. y, De spiritualibus creaturis, a. 10, especially obj. 1;
De unitate intellectus, c. 5. For clear rejection of any direct and special role for God or an
intelligence in man's natural knowledge and for an emphasis on the relative autonomy of the
agent and possible intellects in man, see Summa theol, I, q. 79, a. 4; q. 84, a. 4, ad 1 and 3; I—II,q. 93,
a. 2; Q.D. de anima, a. 4, "«/7; a. 5, c. and ad 6; De spiritualibus creaturis, a. 10, c. and ad 8-9; S.C.C.,
II, c. 76-8. For discussion see Grabmann 1931, pp. 53-68. By 'Avicennised Augustinianism',
Gilson meant a psychological theory which attributes to God the functions of Avicenna's agent
intellect. That is to say, the intelligible is received into the human soul from the God who
illuminates. Gilson listed as adherents of this theory William of Auvergne, Roger Bacon, and
John Peckham. See Gilson 1926-7, pp. 5-127; idem 1929, pp. 5-107. Fora careful and helpful re
examination of the appropriateness of Gilson's term, see Bcitola 1967, pp. 318-34, and 1971,
pp. 278-320.
34. Sometimes Thomas speaks simply of the lumen naturale of the intellect, of the soul, or of reason
(Summa theol, I, q. 12, a. 12, ad3 and a. 13; q. 88, a. 3, ad 1; I-II, q. 68, a. 2; q. 109, a. I,a<i2; II-II,
q. 8, a. i;/n III Sent.,ά. 39, q. 3, a. 1, ad}), whereas at other times he speaks of the lumen naturale of
the agent intellect (In II Sent., d. 28, q. i,a. 5; In IVSent., d. 49, q. 2,8.4). Thomas occasionally
connects the concept of the lumen naturale with our ability to know selfevident first principles.
See for example Summa theol, I—II, q. 63, a. 3; S.C.G., III, c. 154. The classic paper on Thomas'
doctrine on the lumen naturale is by Guillet 1927, pp. 79-88.
certain knowledge.35 Finally, Thomas is highly critical of any suggestion

that during this life the human intellect can achieve an intuitive cognition
of separated substances, whether they be intelligences, angels, or God. 3 6 In
this opposition, he is making one of his sharpest breaks with Albert and also
setting himself against Avicenna and Averroes. His opposition would be
duly noted by the Latin Averroist tradition.
Siger of Brabant
Siger of Brabant is without doubt the most celebrated representative of so-
called Latin Averroism.37 A member of the Arts Faculty at the University
of Paris in the 1260s and 1270s, he played a key role in controversies with
other members ofthat faculty, with theologians such as Thomas, and with
Church authorities such as Etienne Tempier. While it would be an exagger
ation to say that he eventually became a loyal follower of Thomas, there is
certainly evidence that his views on the soul and intellect underwent a
dramatic evolution and that his final position was cjose to that of Thomas
on some key points. That evolution, long a topic of modern scholarly
debate, has been definitively established with the discovery and publication
of his questions on the Liber de causis.38
The major works of Siger which are dedicated to philosophical psy
chology and which are also generally accepted as genuine are the
Quaestiones in librum tertium de anima, the Tractatus de anima intellectiva, and
the De intellectu.39 In the first of these works, which was probably com-
35. Aquinas, Summa theol, I, q. 84, a. 1 ; De spiritualibus creaturis, a. 10, e<f 8; Expositio super De ttinitate,
q. 5, a. 2, c. and ad 4. Thomas relates the possibility of knowledge with certitude to our grasping
first principles or resolving things back into such principles. See Summa theol, I, q. 85, a. 6; I-II,
q. 90, a. 2,ady,q. 112, a. 5; S.C.G.,III, c.47; IV, c. 54; De ver., q. ι, a. 4, ad$;q. 10, a. 6, c. and ad 6;
q. 11, a. 1. ad 13. For comparison ofThomas with Bonaventure and the Augustinian approach, see
Grabmann 1931, pp. 68-71; idem 1948, pp. 45-51; Quinn 1973, pp. 443-663; idem 1974, pp.
105-40. On Aquinas'theory of judgement and assent, see Wilpert 1931,pp. s6-75",Keeler 1934,
pp. 83—in; Tyrrell 1948; Hocnen 1952; Garceau 1968; Owens 1970, pp. 138-58.
36. Aquinas, Summa theol, I, q. 12, a. 1-2,4-5 and 11; q. 88, a. 1 -2; S.C. C, life. 42-8 and 51-3; De
ver., q. 10, a. 11; Expositio super De trinitate, q. 1, a. 2; q. 6, a. 3. Man's ultimate beatitude for
Thomas was a direct vision of God in the hereafter which involved no intelligible species or
concept and which was given as a free gift by God. See Summa theol, I, q. 12, a. 2-5,9 and 11 ; I-II,
q. 3, a. 8.
37. For a brief sketch of Siger's life, see Glorieux 1941. There is a detailed account of his life and career
in Van Steenberghen 1977, pp. 9-176. For earlier reviews of Siger scholarship, see Van Steen
berghen 1956, pp. 130-47; Maurer 1956, pp. 49—56; Zimmermann 1967—8, pp. 206-17.
38. Various scholars have affirmed that evolution on the basis of the questions on the De causis. See
Dondaine and Bataillon 1966, pp. 206—10; Marlasca 1971, pp. 3—27 and 1974, pp. 431—9;
Fioravanti 1972, pp. 407—64; Mahoney 1974a, pp. 531-53: Van Steenberghen 1977, pp. 338-403.
39. In 1923, Martin Grabmann discovered Quaestiones in libros très de anima in Clm 95 59, a manuscript
of the Staatsbibliothek at Munich. The find was announced in Grabmann 1924. The work
6l2 Philosophy of mind and action
posed during the academic year 1269-70,40 Siger begins by accepting

Averroes' doctrine that there is one intellect for all men. It comes from
without and unites with the vegetative and sensitive parts within the
human being to form a composite soul (anima composita).*1 However, the
intellect cannot be united in its substance to human bodies, since it would
then be inseparable from the body. The intellect exists in the body by
operating in it, namely, by moving the body and also by thinking in the
body. 42 The latter occurs as the intellect works with the actually operative
imaginations of particular human beings. Indeed, Siger argues both that
the single intellect is numbered or diversified among humans through the
different intentions present in their imaginations and also that men have
different shares in the universal intelligibles of the one intellect according to
the number of imagined intentions they possess.43
Our senses know a sensible object by means of a sensible species or a
likeness (similitudo) which that particular sensible object causes in the
sense organ. 44 The senses and imagination apprehend things only under
contained a psychology resembling Thomas'. Van Steenberghen accepted Grabmann's attri-

bution of the work to Siger and edited and published the Munich questions along with similar
questions from Codex 275 belonging to Merton College, Oxford. See Van Steenberghen 1931,
pp. 21 -156. However, Grabmann's ascription of the Munich questions to Siger was subsequently
attacked by Nardi 1936, pp. 26-35 and 1937, pp. 160-4. Van Steenberghen replied in a brief
monograph, 1938, pp. 24-45, D U t Nardi 1939, pp. 453-71 and 1940, pp. 149-56 continued his
attack. He was supported by Gilson 1939, pp. 316-23. Van Steenberghen's case was seriously
weakened by Sajo 19S8, pp. 21-58, who proved that two of the other works in the Munich
manuscript were by Boethius of Dacia. Van Steenberghen 1971, pp. 131-3 and 1978, pp. 66-8
now accepts the ascription of these works to Boethius and even considers him the possible author
of the questions on the De anima.
40. Bazân 1972b, pp. 70*-74* has argued for this dating on the grounds that Siger's work must have
been written after Thomas' disputed question on the De anima, whose final redaction he dates as
1269.
41. Siger of Brabant, In tertium de anima, q. ι, pp. 2-3, lin. 42-68; q. 2, p. 5, lin. 32-41. Nardi i960,
pp. 160-1 argued that Siger's conception of a composite soul reflected the influence of Albert the
Great's De natura et origine animae, but this thesis has been severely criticized by Van Steenberghen
1977, pp. 341-2 and Bazân 1975, pp. 32-4. See also Vennebusch 1966, p. 59, n. 50. Siger's use of
the term anima composita for the union of the intellect with the vegetative and sensitive parts must
be distinguished from his use of the terms 'intellective soul' (anima intellectiva), 'rational soul'
(anime rationalis) and 'our intellect' (nosier intellectus) for the unitary and separate intellect. For
discussion, see Da Palma 1955a, pp. 30-3.
42. Ibid., q. 7, pp. 22-4; q. 8, p. 25. Siger uses (p. 24, lin. 54) the celebrated sailor-ship analogy (De
anima, II, c. 1,413 *7~9) for his own purposes. For a summary of the various ways in which it was
adopted by different ancient and medieval philosophers, see Mansion 1953, pp. 457-65.
43. Ibid., q. 9, p. 28, lines 74—96. Siger points out that this is what Averroes meant when he spoke of
the speculative intellect. Bazân 1975, ΡΡ· 27-30, considers Siger's conception of the speculative
intellect to be a weakness in his theory and not to do justice to its importance in Averroes'
thought.
44. Ibid., q. 18, pp. 67-8. Siger's example is that of sight, which will have a likeness of this particular
coloured thing.
their material conditions, that is, as particulars. Sense and imagination

differ, however, in that a sense can receive the sensible form only when the
sensible object is actually present, while imagination can also receive that
form in the absence of the sensible object. While Siger admits that the
estimative power (existimativa) receives something beside sensible forms,
for example the enmity which the sheep apprehends in a wolf, he argues
that it can apprehend such an insensible form only when it simultaneously
apprehends a sensible form such as colour or size.4* On the other hand,
Siger claims that we are aware of the reception of immaterial and universal
intelligible forms or essences within us which cannot be explained by any
sensible species. That is to say, we experience in ourselves two immaterial
activities and the two powers of the intellect causing those activities in us,
namely, the abstracting by the agent intellect of the universal intelligibles
which previously were intentions of our imagination (intentiones imag-
inatae) and the reception by the material or possible intellect (intellectus
possibilis) of those universal intelligibles.46 Siger seems very doubtful,
however, that the intellect itself can know individuals as individual, even
by a reflection on the phantasm, since he takes it as axiomatic that the
intellect of itself knows only that which is universal.47
Aquinas against the Averroists

Thomas' Tractatus de unitate intellectus contra Averroistas, completed early in
1270, was probably not directed specifically against Siger's In tertium librum
de anima but rather against the oral teaching of Siger and his circle in .the
Arts Faculty.48 The critique which it presents eventually undermined
45. Ibid., q. 4, p. 14, lin. 12-15 a"d p. 16, lin. 67-82. The imagination receives the sensible form
(sensitivaforma) only by means of a sensible image (imago). Siger is of course following Averroes
when he claims that Aristotle called the imagination the 'passive intellect' (intellectus passivus).
46. Ibid., q. 4, pp. 12-13,lin. 70-5 and p. 14, lin. 23-9; q. 13, p. 45.lin. 50-5;q. 14, p. 47,lin. 28-32.
The three principles of intellectual knowledge are therefore the possible intellect, the agent
intellect, and the phantasms or intentions of the imagination which are rendered intelligible by
abstraction. Seeq. 12, p. 37, lin.63-9and p. 40, lin. 21-4; q. 14, p. 47, lin. 30-2, 38-50 and64-82,
and p. 50, lin. 1 -11. Siger does not consider the agent and the possible intellects to be two
individual substances. On the contrary, they are powers of the same single substance, namely.
Our intellect', the unitary intellect of all human beings. See q. 15, p. 58, lin. 42-8.
47. Ibid., q. 18, pp. 65-7. Da Palma 1958, p. 71 sees Aquinas as Siger's target.
48. Van Steenberghen 1942, p. 558 and 1966, p. 435 sûtes that it would be difficult to determine with
certitude whether Thomas had Siger's In tertium de anima specifically in view when he wrote the
De unitate intellectus. He admits (1942, p. 557) that Thomas' final paragraphs arc directed to some
writing or lesson of Siger. More recently Van Steenberghen 1977, pp. 59 and 347 has suggested
that Thomas seems to have made use of course notes reflecting the oral teaching of Siger and his
party. Webet 1970, pp. 29-33 and 41-5 has attempted to reconstruct Siger's teaching from
apparent references to it in Thomas's opusculum. Bazin 1974, pp. 57-66 has justifiably ques-
6i4 Philosophy of mind and action
Siger's confidence in the Averroist psychology. 49 The two major errors of

Averroes regarding Aristotle which Thomas singles out for attack are the
denial that the possible or potential intellect is the substantial form of the
human body and the doctrine that it is one for all men and therefore not
multiplied according to the number of existing human beings. 50 He
marshals passages from Aristotle himself as well as from Alexander,
Theophrastus, Themistius, Avicenna, and Algazel to prove that not just
Latin philosophers but also Greeks and Muslims take Aristotle to maintain
that the potential intellect is part of the individual human soul, which is
itself the form of the body. 5 1 It is particularly noteworthy that in two
different passages Thomas quotes at length from William of Moerbeke's
translation of Themistius' paraphrases of the De anima to prove that
Themistius had held that while there may be a single agent intellect
illuminating all other intellects, there is in each human soul an agent
intellect which, having been illuminated by that transcendent intellect,
illuminates in turn the individual potential intellect that is also found in the
soul. 52 These citations embolden Thomas to denounce Averroes as 'the
depraver of Peripatetic philosophy' for having falsely presented Themistius
as a proponent of the unity of the intellect.
Thomas attacks the Averroists not for having said that the agent intellect
is one for all men, but for having held to the unity of the potential
intellect. 53 Like Siger, Thomas appeals to our own psychological ex-
perience. The difference is that Thomas emphasises that it reveals to us that
in thinking it is this individual man who thinks and not some separate
intellect whose cognition we merely share. Pursuing this theme of the
tioned how Weber can distinguish Siger's teaching from that of his colleagues, and he has gone on
to challenge the specific doctrines which Wéber attributes to Siger. Moreover, Bazân has
effectively discredited Weber's fundamental thesis, namely, that Thomas dramatically changed
his own doctrine on the relation between the human soul and its intellect after studying Siger's
position. See the replies by Wéber 1974, pp. 15-16 and 435 and 1976, pp. 294 and 309. Van
Steenberghen 1977, pp. 357-9 and 412-15 has entered the fray to attack anew Weber's 1970 book
and his scholarship.
49. Much of Thomas' critique can actually be found in his earlier works, where it is directed simply
against Averroes. See Nardi 1947, pp. 56—67. More recent studies on the topic include Verbeke
i960, pp. 220-49; Mazzarella 1974, pp. 246-83; Mahoney 1974a, pp. 535-8; Ncdoncelle 1974,
pp. 284-92.
50. Aquinas, De unitate intellectus (1957), c. 1, par. 1, pp. 1-2.
51. Ibid.,c. 1, par. 3-43, pp. 3-29; c. 2, par. 51-9, pp. 33-8; c. 5, par. 119-21, pp. 76-8.
52. Ibid., c. 2, par. 51-4, pp. 33-5; c. 5, par. 120-1, pp. 77-8. For discussion of the impact of
Moerbeke's translation on Thomas and other late-thirteenth and early-fourtcenth-century philo-
sophers, see Verbeke 19S7, pp. xxxix-lxii; Mahoney 1973a, pp. 422-67.
53. Ibid., c. 4, par. 86, pp. 54-5; c. 5, par. 120, p. 77. Thomas well knew that some of his
contemporaries taught that there is only one agent intellect, namely, God himself. See n. 33
above. On the Franciscans' discussions regarding the agent intellect and whether it is to be
identified with God, see Bowman 1972-3, pp. 251-79.
individuation of cognition in the individual, he analyses the Averroist

theory of the interrelationship of the phantasm, the intelligible species, and
the potential and agent intellects. Thomas takes Averroes to postulate two
subjects for the intelligible species, namely, the separate potential intellect
and the phantasm in the individual human being. Pointing out that when
the intelligible species is in the phantasm it is there only potentially, that is,
before abstraction by the agent intellect has rendered it actually intelligible,
he argues that the potential or possible intellect cannot therefore be joined
to the phantasms of individual men through the intelligible species. The
obvious consequence is that Averroes has no way to account for the in
dividual person actually thinking. 5 * Thomas himself explains the individu
ation of universal cognition by means of his own doctrine of intelligible
species.55 While the intelligible object (intellectum) is certainly one and the
same for all men, namely, the universal nature or essence abstracted from
phantasms, it must be distinguished from the intelligible species (species
intelligibilis). It is the different intelligible species found in the intellects of
individual men which allow universal intellectual cognition of the same
intelligible object to be individuated in these different individuals. That is
to say, while each of them will know the same intelligible object, they will
do so by means of different intelligible species in their individual intellects.
The consequence will be that we can truly say of each of them that it is this
individual human being who is thinking.
Siger's treatise De intellectu

Siger's initial reply to Thomas' De unitate intellectus was his now lost De
intellectu, probably written in 1270 before Tempier's condemnation. 56 It
54. Ibid., c. 3, par. 62-6, pp. 39-42. See also c. 4, par. 91, p. 58. Thomas considers it absurd to believe
that the separate substances could receive anything from the phantasms of humans. See c. 4, par.
95. Ρ·6ι.
55. Ibid.,c. 5, par. 106—13, pp. 68—73. Both Wéber and Bazân have rightly emphasised that Thomas'
doctrine on intelligible species represents a radical departure from Aristotle. Weber 1970 explains
that Thomas introduced intelligible species in his account of intellectual cognition in order to
break with Aristotle's pure and simple identity between the intellect and the intelligible (De
anima. III, 4, 430*2-4), presumably a fundamental principle leading to Averroes' doctrine of the
unity of the intellect (pp. 221,226-9,237-8 and 291). Bâzan 1974, pp. 98-9 rightly praises Weber
for underscoring the importance of the intelligible species in Thomas' critique of and
reply to Averroes.
56. Bâzan 1972b, p. 75*; Van Steenberghen 1966, p. 447 and 1977, p. 63. The proposal of Kuksewicz
1968, p. 44, 72-3 and 76 that the De intellectu was written after Siger's De anima intellectiva in 1272
or 1273 and was thus sent to Thomas after he had left Paris is implausible. Such an ordering of the
two works would make the evolution of Siger's thought erratic in the light of his final position in
the questions on the De causis. Nardi 1945, pp. 20-1 was the first to establish that the De intellectu
and the De anima intellectiva were distinct works and that the former, not the latter, was Siger's
initial reply to Thomas' De unitate intellectus. Mandonnet 1911a, p. n o claimed that Thomas'
opusculum was a reply to Siger's De anima intellectiva. Doncoeur 1910, p. 501 rejected
6ι6 Philosophy of mind and action
has been reconstructed by Bruno Nardi from citations to be found in the

writings of Agostino Nifo. 5 7 Siger reveals himself to be on the defensive
in regard to Thomas and to have made some readjustments in his own
Averroist psychology in order to save the unity of the intellect and yet be
able to evade some of Thomas' arguments. To Thomas' objection that
the intellect must be united to man as his form if he is to be put in the species
of the rational, Siger replies that the intellect, that is, the potential intellect,
is united to man according to its operation and also according to ex
istence. Consequently, it is the form of the human body, providing man
with his specific difference. 58 The potential intellect is united to our cogita
tive power from our birth as a natural form which gives us existence (dans
esse) and puts us in the human species. However, the intellect is united to
our phantasms in a different way when it serves as a principle of knowing. 5 9
The agent intellect is a distinct separate substance, for it is God, who
appears to play here a dual role for Siger. He illuminates and abstracts the
phantasms in men which the potential intellect then knows, and he also can
be the object of an intuitive knowledge providing the potential intellect
and the human being joined to it with complete beatitude. 6 0
Siger's Tractatus de anima intellectiva

These modifications, which were meant to shore up the Averroist psy
chology against Thomas' attack, did not long satisfy Siger. 6 1 About three
Mandonnet's thesis, which had been published in the first edition of his book, and proposed that
either Thomas was attacking another work or he was attacking other Averroists. Chossat 1914,
pp. 2S-52 showed that Siger drew arguments against the unity of the intellect from Thomas
himself, and he therefore argued that Siger no longer holds the doctrine himself in the De anima
intellectiva. Van Steenberghen 1938, pp. 64-75 a n d 1942, pp. 551-8 has summarised in admirably
clear fashion the results of these earlier investigations.
57. Nardi 1945, pp. 11-90. For a summary of the major doctrines of Siger's De intellectu and his De
felicitate, see pp. 46-7.
58. Agostino Nifo, De intellectu, I, tr. 2, c. 8-9, cited by Nardi 1945, pp. 14-15. See Aquinas, Df
unitate intellectus, c. 3, par. 80, p. 50.
59. Agostino Nifo, In libros de anima. III, comm. 5, cited by Nardi 1945, pp. 15-17. See also his De
intellectu, I, tr. 3, c. 18, cited by Nardi 1945, pp. 17-19. By the potential intellect Siger means the
lowest of the separate substances, that is, an intellect that is one for all men.
60. Nifo, De intellectu, I, tr. 4, c. 10; II, tr. 2, c. 17; De beatitudine animae, II, comm. 21, cited by Nardi
1945, PP- 2 I a nd 25-7. Nifo himself incorporated Siger's suggestion that God is the agent
intellect into his own more complicated interpretation of Averroes. For details see Mahoney
1970, pp. 387-409-
61. Bazân 1972, p. 77*; Van Steenberghen 1977, pp. 99,218 and 220-1. Siger's caution regarding the
conflict of faith and reason no doubt reflects some concern on his part resulting from the
Condemnation of 1270. For the text of the latter, see Mandonnet 1911 a, p. 111, η. 1 ; Denifle and
Châtelain 1889, pp. 486-7. For thorough discussion of the condemnations of 1270 and 1277 see
Wippel 1977, pp. 169-201; Van Steenberghen 1977, pp. 74-80 and 149-59; Hissette 1977.
or four years later, that is, in 1273 or more likely in 1274, he completed his
Tractatus de anima intellectiva. The focus of attention is now Aristotle
himself, and the commentator used most frequently to determine his mind
is Themistius, the very authority whom Thomas had used to discredit
Averroes' reliability.62 Averroes himself, the 'Commentator', is men-
tioned by name only twice.63
Siger again accepts that for Aristotle the intellect, which operates in an
immaterial fashion, comes from without and unites with the vegetative and
sensitive parts, which have only bodily activities and are educed from matter.
The result of this union is a composite form (forma composita).6* Siger
attacks both Albert and Thomas, 'distinguished men in philosophy*, for
maintaining that the substance of the intellective soul is so united to the
body that it gives existence to it, while its power, the intellect, is wholly
separate from matter since it needs no bodily organ to operate. They have
thereby contradicted Aristotle. Nonetheless, while Siger emphasises that
the thinking of the intellective soul, which is a form subsisting apart from
matter, is somehow separate from matter, he also admits that that thinking
must somehow be united to matter if we are truly to say this man is
thinking. Siger points out that the intellect by its nature requires both the
phantasm and the imagination if it is to carry out its proper activity, but he
is quick to add that it depends on the phantasm provided by the body only
as an object (obiectum) of cognition and not as an underlying subject
(subiectum) required for its own existence. Thinking can be truly attri-
buted to the individual human being insofar as he operates by one of his
'parts', namely, the single intellect. That is to say, when the intellect thinks,
by its very nature it operates within the human body. 65 Since that which is
62. Forreferenceto Themistius by name, see Siger of Brabant, De anima intellectiva, c. 3, p. 83, lin. 43,
p. 85, lin. 88 and p. 88, lin. 42. In his edition. Bazin has identified many other pasages in which
Siger is relying on Themistius. For further discussion, set Bruckmüller 1908, pp. 13,16-18, 50,
55-6,9J; Nardi 1947, pp. 75-8; Verbeke 1957, pp. xlvii-xlviii; Mahoney 1966, pp. 163-4 and
170-2; idem 1973a, pp. 438-41; idem 1974a, pp. 540 and 544, n. 25. Although Van Steenberghen
1942, p. 653 and 1977, pp. 100,207 and 400-1 does indicate that Themistius' paraphrases on the
De anima helped to undercut Siger's allegiance to Averroes, he does not bring out how much
Themistius influenced the psychology which Siger himself develops in the De anima intellectiva.
On the other hand, Kuksewicz 1968, pp. 32-44 never mentions Themistius at all when analysing
Siger's De anima intellectiva.
63. Siger of Brabant, De anima intellectiva, Prol., p. 70, lin. 9 and c. 6, p. 97, lin. 54. See Bruckmüller
1908, p. 13.
64. Ibid., c. 8, pp. 109-10.
65. Ibid., c. 3, pp. 80-7. Siger's implausible claim (p. 86, lin. 11-12) that the intellect is not man's
'mover' when it funcdons within him in the process of thinking probably represents an attempt
to circumvent Thomas' attack on Averroes for making the intellect a mover and the human
being something moved. See Aquinas, De unitate intellectus, c. 3, par. 57—63 and 79, pp. 42-6 and
6ι8 Philosophy of mind and action
an intrinsic agent (operans intrinsecum) in regard to matter can be called its

'form', at least in an extended sense of the word, the intellect proves to be
the form or perfection of the human body. Though it subsists in itself, the
intellect is like a true form both because that which operates intrinsic to a
body is not spatially separate (non loco separatum) from that body, and also
because the operation of such an intrinsic agent denominates the whole
composite.66
Despite all these tortured distinctions, Siger admits in the seventh chap
ter of the De anima intellectiva that there are philosophical arguments and
authorities for the view that the intellective soul is multiplied according to
the number of human bodies: he cites Algazel, Avicenna, and Themistius,
authorities used by Thomas in his De unitate intellectus.61 He also adopts two
of Thomas' arguments for the multiplicity of intellects, namely, that on the
supposition of the unity of the intellect all men would have the same
knowledge and there would be no need for the agent intellect, since the
intellect would always be filled with all intelligibles. Siger recounts that
because of the difficulty of these and other arguments he has long been in
doubt as to what natural reason should hold regarding the problem and
what Aristotle in fact thought about it. He concludes that in such a state of
doubt he must adhere to faith, which surpasses all human reason.68
50, and also c. 4, par. 87-8, pp. 56-7. It is ironic that while Thomas insists that the soul is united to
the body as its form and not as a sailor to his ship (ibid,, c. 1, par. 10, pp. 8-9; see also c. 1, par, 5,
p. 5; and c. 3, par. 69, p. 44), Siger deliberately cites (De anima intellectiva, c. 3, p. 79, lin. 38-42 and
p. 8s, lin. 85-9) Themistius as an authority to argue that for Aristotle the intellect is related in its
operations in regard to the body precisely as a sailor to a ship. He docs so when presenting his
theory of the operans intrinsecum as the form or perfection ofthat in which it operates. Chossat
1914, pp. 573-4 characterises Siger's theory as a miserable subterfuge. For a more sympathetic
judgement, see MacClintock 1954-S, pp. 187—94 who shows the influence of Siger's doctrine on
John of Jandun. For the relevant text see Jandun, Super libros de anima. III, q. 5, f. 58,1\ who
mentions Siger by name as the source of the operans intrinsecum theory (f. 60™).
66. 'Ad ultimum dicendum quod anima intellectiva perfectio corporis est, secundum quod in
trinsecum operans ad corpus perfectio et forma corporis habet dici. Convertit enim cum forma in
hoc quod intrinsecum corpori non loco separatum, et quia etiam operatio sic intrinseci operands
totum denominat.' Siger of Brabant, De anima intellectiva, c. 3, p. 87, lin. 33-7. For a full
discussion, see Ermatinger 1963, especially pp. 28-40.
67. Ibid., c. 7, p. 107, lin. 42—7. See Aquinas, De unitate intellectus, c. 5, par. 119—20, pp. 76—7. The
strong influence of Thomas' doctrine of individuation on Siger here was first delineated by
Bruckmüller 1908, pp. 140-53. See n. 76 below.
68. Ibid.,c. 7, p. 108, lin. 76-87. See Aquinas, De unitate intellectus, c. 4, par. 90—1 and 95, pp. 57—8; Q.
D. de anima, a. ι; De spiritualibus creaturis, a. 9. See n. 74 below. Since Siger admits his uncertainty
as to Aristotle's position on the intellective soul, it seems out of the question to consider him to be
a convinced 'Averroist' in this work. Furthermore, since he gives Themistius precedence over
Averroes as an interpreter of Aristotle and takes Themistius to maintain the multiplicity of both
agent and possible intellects in human beings, it would be erratic for him to present Aristotle as a
proponent of the unity of the intellect. Siger also raises the problem of the relation between faith
Several points remain to be made regarding the psychology assumed by

Siger in the De anima intellectiva. First of all, it must be emphasised that
Siger breaks with Averroes and his own In librum tertium de anima in that he
denies that it is through the phantasms or intentions of the imagination that
the separate intellect is united to man and made his form and holds instead
that the intellective soul is directly united with the human body itself.69
Secondly, although he agrees with Thomas that the intellective soul knows
by receiving intelligible species of material things, he holds against Thomas
that according to Aristotle it receives those species into its very substance
and not merely in its power.70 Thirdly, since the intellective soul is
described only as understanding or receiving forms, which is the proper task
of the possible or potential intellect, and is never said to be the source of
abstraction, it appears that the intellective soul is identical with the possible
intellect, whereas the agent intellect is some other, unspecified separate
substance.71 And finally, we are told nothing in the work about how
the intellect can know individuals or whether man, when united to the
possible intellect, can have an intuitive knowledge of God and the other
separate substances.
Siger's questions on the Liber de causis

Siger's questions on the Liber de causis were composed sometime after the
De anima intellectiva, but presumedly before his flight from Paris in 1276.72
Siger now rejects the notion of a composite soul, arguing that the substan-
tial form of each individual thing must be simple and provide the basis of all
and reason in De anima intellectiva, c. 7, p. 83, tin. 44 to p. 84, lin. 48, where he ends by quoting
almost to the word Albert the Great, In de gen. et cerr., I, tr. 1, c. 22, ad t. c. 14, cited in n. I
above. Siger did not maintain the so-called 'double truth theory', but neither did any of his
contemporaries. For discussion see Van Steenberghen 1974, pp. 555-70; idem 1976, pp. 351-60;
and Pine 1973, pp. 31-7·
69. Ibid., c, 3, p. 85, lin. 76-89. See the discussion in Bruckmüller 1908, pp. 22-3, 30 and 114-15. See
also Van Steenberghen 1942, pp. 676-7; 1966, pp. 450-1; and 1977, pp. 369-70.
70. Ibid., c. 3, p. 83, lin. 22-38. Siger will not surrender this view in his questions on the De causis.
71. Ibid., c. 3, p. 83, lin. 23-5 and c. 4, p. 90, lin. 36-42. While it is true that Siger mentions the agent
intellect in two passages of the treatise, in neither case does he state that the agent intellect is part of
the intellective soul. Seec. 5, p. 93, lin. 62-5 and c. 7, p. 108, lin. 79-82. It is therefore difficult to
accept the confident assertion of Van Steenberghen 1977, p. 376, n. 82 to the contrary. For further
discussion, see Mahoney I966, pp. 177-8 and Kuksewicz 1968, pp. 37-8. It should be noted that
in the Lisbon Quaestiones naturales Siger maintains that the agent intellect and the possible intellect
belong to different separate substances. It should also be recalled that in his earlier De intellectu
Siger had taken the possible or potential intellect and the agent intellect to be two separate
substances. Consequently, we would perhaps do best to consider Siger as non-committal in the
De anima intellectiva.
72. Dondaine and Bataillon 1966, pp. 206—7 and 210—11; Marlasca 1972, pp. 25—9. On Siger's flight,
see Dondaine 1947, pp. 186-92.
that is predicated of that individual, rationality included.73 Averroes'

doctrine of the unity of the intellect is denounced as heretical and also as
irrational, and whether or not Aristotle agreed with Averroes on this
question, he too was human and could err. 7 4 Although Siger adopts some
of Thomas' arguments to show that the intellective soul is truly the form
and actuality of the human body, he still rejects Thomas' conception of the
soul as united by its essence to matter as its form while separated from it by
its power, the intellect. However, he carefully distinguishes the soul from
ordinary material forms by characterising it is subsistent in itself even
though it perfects matter and is thereby individuated.75 Siger has here
abandoned the standard Averroist dichotomy of forms wholly separate
from matter and forms wholly material. That is to say, he is claiming that
between the ordinary material forms, which are completely bound to
matter and multiplied in the same species, and the wholly separate forms,
of which there can be only one in each species, there is a middle kind of
form, namely, the intellective soul or human intellect. Unlike the forms
which are totally separated from matter and cannot be multiplied, human
intellects are multiplied since they do communicate with matter and have
relationships to individual human bodies as their respective forms. On
the other hand, unlike ordinary material forms they maintain an individual
and subsistent existence in themselves which they even maintain in a
disembodied state after death. 7 6
The intellective soul does not depend on the body as on an underlying
subject (subiectum) for its existence or its act of intellectual cognition, but it
73. Siger of Brabant, Super librum de causis, q. 4. pp. 47~9-
74. Ibid., q. 27, p. m , lin. 114-15; p. 112, lin. 147-52; and p. 115, lin. 248-52. While Siger might still
concede that there can be some debate as to what Aristotle himself taught regarding the unity or
multiplicity of the intellect, he now has no doubts that natural reason and philosophy can prove
its multiplicity. This is surely a dramatic change from the lingering doubt which he had expressed
in the De anima intellectiva. See n. 68 above and also Zimmermann 1973, pp. 426-7. Siger now
appears to demarcate more clearly natural reason and philosophy from the text and teachings
of Aristotle. Siger doubtless borrowed his remark (p. 115) that Aristotle was not a god but
rather a human being who could err from Albert, Physica, VIII, tr. 1, c. 14, cited in η. 1 above.
On possible personal contacts between Siger and Albert, see Vennebusch 1966, pp. 168-9 and
Van Steenberghen 1977, pp. 395-7.
75. Siger of Brabant, Super librum de causis, q. 26, pp. 104-7.
76. Ibid.,q.obis, p. 59,lin. 19-31 ;q. 18, p. 81,lin. 29-36; q. 27, pp. 108-17;q. 53, p. 182,lin. 165-70.
Siger's account of the individuation of the soul through its relationship to matter closely
resembles the position of Thomas. See q. 27. pp. 112-16; q. 32, p. 126; q. 33. p. 127; q. 53,
pp. 183—4. Moreover, Siger's explanation of how the disembodied human soul remains in
dividuated despite the loss of its body (q. 27. pp. 114-15. lin. 205-37) is much like that of Thomas
in De spiritualibus creaturis, a. 9, ad 3 and 4; De unitate intellectus, c. 5, par. 104, p. 67, and other
works.
does depend on the body to provide it with an object (obiectum), namely,

the phantasm. 7 7 However, while the phantasm or sensible species is a
principle of knowledge, it brings with it all the individuating charac
teristics of a particular sensible thing. Only by abstracting the intelligible
species from the phantasm will the intellect have a principle of universal
cognition. That is to say, just as Plato and Socrates have their own
individual sensible species by which they see the same individual sensible
nature, for example, this colour, so by their o w n distinct intelligible
species, which are likenesses (similitudines) of one same thing, they know
the same intelligible object (intellectum) or universal nature. 7 8 The intellec
tive soul extends itself in a universal fashion to all material forms, both
abstracting and receiving those forms: the agent intellect abstracts from the
phantasm the universal intelligible species which is then received by the
potential or possible intellect. 7 9 It is thus the agent intellect and the
phantasm as rendered intelligible which together reduce the possible intel
lect from potency to act. 8 0
While Siger does still maintain some doctrines of his o w n in these
questions, it is obvious that he has moved very close to Thomas on the
nature of the soul and intellect. This rapprochement may be one of the
reasons why Dante placed them together in the fourth heaven of the Divine
Comedy.91
Despite Siger's abandonment of Averroes as the true interpreter of
Aristotle and the heavy blow of Tempier's condemnation of 1277,
Averroism did not disappear from the University of Paris. 8 2 John of
Jandun (d. 1328), w h o taught there in the Arts Faculty in the early
fourteenth century, followed Averroes as the best guide to Aristotle, and he
77. Ibid., q. 26, p. 182, lin. 156-61 : q. 26, p. 105, lin. 88-91. This distinction was already used by Siger
in the De anima intellectiva, c. 3, p. 85, lin. 70-1.
78. Ibid., q. 28, pp. 116-17. Siger has made his own the critical distinction which Thomas used
against Averroes. See Marlasca 1971, pp. 20-1 and 25-7. However, it would be a mistake to
believe that Siger capitulated to Thomas on all major philosophical topics. On the remaining
differences between them see Zimmermann 1967-8, pp. 212-21 and 1973, pp. 438-45.
79. Ibid., q. 52. p. 181.lin. 134-6. See also q. 37, p. 146.1 do not believe, as does Marlasca I97i,p. 14,
cither that the intellective soul is identical with the potential intellect or that the nature of the
agent intellect is left somewhat unclear.
80. Ibid., q. 41, pp. 151-2 and q. 52, p. 178, lin. 30-7.
81. See Paradiso, Χ , lin. 133—8, where Thomas indicates Siger to Dante. For discussion of this passage
and of Dante's own philosophical views, see Nardi 1944, pp. 207-45; idem 1947, pp. 81-9; Gilson
1949, pp. 257—81; Zimmermann 1967-8, pp. 206 and 214; Van Steenberghen 1977, pp. 165—76;
idem 1978, pp. 64-8.
82. On Averroism in the late thirteenth and early fourteenth centuries, see De Wulf 1937, pp. 125 and
128; Gauthier 1947-8, especially pp. 187-9 and 331-6; Kuksewicz 1968, pp. 97-9 and 118-20.
also defended him against both Albert and Thomas. 8 3 Jandun's writings
would have influence on fourteenth- and fifteenth-century philosophy in
Italy, especially at Bologna, and they would continue to be read and cited
during the Renaissance.84 Toward the end of the fifteenth century, both
Jandun's questions on the De anima and also Siger's De intellectu were
closely studied at the University of Padua. 85 This represented no clear
victory for Averroism, however, since there was also close interest in
Albert and Thomas. Indeed, Albert's psychological ideas had such great
impact on philosophers like Nicoletto Vernia (d. 1499) and Agostino Nifo
(ca. 1470-1538) that their thought can to some extent be described as
Albertistic.86 This interest in Siger, Thomas, and Albert testifies to the
continuing vitality of thirteenth-century philosophical discussions during
the Renaissance and beyond. 87
83. O n Jandun's life and works, see Valois 1906, pp. 528-602; MacClintock 1956, pp. 4 - 7 and
103—30; Terrero i960, pp. 331-43; Schmugge 1966, pp. 1-26 and 121-32.
84. O n Jandun and Bolognese Averroism, see Maier 1944, pp. 150 and 157; eadem 1949, pp. 251-78;
eadem 1964, pp. 1—40; eadem 1967, pp. 335—66; Krisreller 1952, pp. 59-65; Ermatinger 1954,
pp. 35-6; Kuksewicz 1968, pp. 315-52; Vanni-Rovighi 1971, pp. 161 - 8 3 . His role in Renaissance
Aristotelianism remains to be studied more carefully.
85. Nardi 1945 is the classic study o f Siger's De intellectu and its diffusion in Italy. Jandun's
psychological doctrines and interpretations of Aristotle, Alexander, and Averroes played a central
role in philosophical discussions at Padua. See Mahoney 1968, pp. 281 and 294-5; idem 1976b,
pp. 153—5; idem 1976c, pp. 291—2 and 298—300; Poppi 1970b, passim.
86. O n Thomas' influence at Padua, see Kristeller 1974. ΡΡ· 46, 49~53 and 62; Mahoney 1974b,
pp. 277—85; idem 1976a, pp. 195-211. The Albertistic tendencies o f Vernia and Nifo have been
underlined by Mahoney 1980, pp. 546-55.
87. This essay was completed during the tenure of a fellowship from the John Simon Guggenheim
Foundation. I am indebted to Professors George W. Roberts and John F. Wippel for criticisms of
earlier drafts.
31
CRITICISMS OF ARISTOTELIAN
PSYCHOLOGY AND THE
AUGUSTINIAN-ARISTOTELIAN
SYNTHESIS
Thirteenth-century criticism of Aquinas' Aristotelianism

Strong reaction among traditional theologians in Paris and Oxford against
the massive introduction of new Aristotelian ideas was still growing at the
time Albert, Thomas, and Siger taught and wrote. It culminated in the
formal act of condemnation by Bishop Tempier in 1277. For a while the
Averroistic trend was halted, and the main target of criticism was Thomas
Aquinas. A conception of the soul too closely connected to the body, too
near to matter was an offense against the entire Christian tradition, which
derived so much from Platonism and Augustine. The criticism also at-
tacked the concept of the potential and agent intellect. It rejected the
potential nature of the intellect which received and did not produce
cognition, it rejected the effect of sensible species on the intellect, and
the independence attributed to human cognition, unassisted by divine
illumination.
Reactions to the criticism

Even Thomas' pupils and defenders stepped back in the face of this
overwhelming pressure. Giles of Rome, although he basically agreed with
Thomas' conception of the potential and agent intellect, described the
agent intellect also as a quasi-Avicennian storehouse of pre-empirical
knowledge and rules of understanding, conceived of as complete potential
knowledge.' Godfrey of Fontaines defended Thomas in his QuodUbeta,
written between 1285 and 1297, and gave his conception of the soul an
even firmer Aristotelian character, but he denied the possibility that phan-
tasmata can be turned into intelligible species by the agent intellect, and
indeed eliminated intelligible species altogether as an element of intellec-
tive cognition.
1. Giles of Rome 1476, In Aristotelis De anima commentum, f. 69*b-,·. Cf. Kuksewicz 1973.
Early attempts at an Augustinian-Aristotelian synthesis: Roger Marston

Most theologians of that time preferred the old Augustinian tradition
systematised by Bonaventure, and tried to create doctrines which would
absorb some generally admitted Aristotelian ideas, but include them as
subordinate parts of a basically Platonic and Augustinian conception of a
spiritual and active soul. Roger Marston, for instance, the author of
Quaestiones disputatae dated between 1280 and 1294, accepted not only the
potential and agent intellect in the human soul, but also a second agent
intellect, separate from men and identified with God who illuminates
human souls.
Matthew of Aquasparta
Less simple, and more interesting and original, was the conception of
Matthew of Aquasparta, author of the Quaestiones de cognitione written
before 1280.2 Matthew accepts the agent and potential intellects as cogni-
tive faculties of the soul, assisted by divine illumination. This Augustinian
correction of the Aristotelian conception was based on the Augustinian
idea of the soul and the Avicennian interpretation of the object of cog-
nition. The perfect spiritual nature of the soul makes it impossible for it to
be affected by sensible objects. Furthermore, the object of human intellec-
tive cognition is not the quiddity or essence of material objects as it was
in Thomas' theory, but simply quiditas which exists in particular objects
and is neutral between particularity and universality. The role of both
intellects founded on these principles could not be conceived of in terms of
the abstraction of universals from their individuation. It is not possible for
sensible species to act on the potential intellect, nor is the action by the
agent intellect on these species necessary; since the quiddity is not par-
ticularised it does not need to be freed. The agent intellect, therefore,
assisted by divine illumination, creates intelligible species by itself and
impresses them on the potential intellect. The role of sensible species is
therefore only that of a necessary condition of intellection.
The conception of neutral quiddity, Avicenna's contribution to the
development of Aristotelianism, became an important element of other
solutions. (See J. Wippel's contribution to this volume.) The deep gulf
between spiritual intellect and material sensory data was admitted too, but
Matthew of Aquasparta's limitation of the role of these sensory data to a
2. Matthew of Aquasparta 1903, Quaestiones disputatae selectae vol. 1, Quaestiones defide et cognitione,
pp. 226-7, 231. «54. 42*·
Criticisms of Aristotle's psychology 625
mere necessary condition did not convince other philosophers. Henry of

Ghent and Duns Scotus looked for a better foundation of the content of
human knowledge and its link with objects to be known.
Henry of Ghent
Henry of Ghent, who in his Quodlibet of 1270 represented a conception of
the potential and agent intellect quite close to the theory of Aquinas,
influenced by the generally critical mood after 1277 and by the turbulent
disputes concerning the concept of the intellect, changed his conception
some years later and elaborated one of the most interesting theories of the
late thirteenth century.3
The potential and agent intellect are, according to Henry, faculties of the
soul (which is a spiritual substance and form of the body), and they are
characterised in medieval Aristotelian language: the potential intellect
receives the concept, the agent intellect illuminates sense images and
activates the process of abstraction. However, Henry rejects some specific
theses which were basic to medieval Aristotelianism as represented by
Thomas Aquinas. First, he denies that an accident could be impressed on a
spiritual subject, and this rule applies to the potential intellect. Second, he
does not accept a transmutation performed by the agent intellect upon
phantasmata transforming them from particular to universal, nor does he
allow an impression by sensory faculties on the potential intellect. Third,
the representation of universals by intelligible species is not possible be-
cause only a universal being can give a universal representation of itself;
universal concepts cannot be impressed by individual objects of the ma-
terial world, nor can they be impressed by spiritual beings, because such
concepts would not represent material objects.4
These principles completely changed the Aristotelian conception of the
role of both intellects. The agent intellect acts upon particular phantasmata
and upon the potential intellect. By the latter operation, it prepares the
potential intellect for intellection; by the former, it proceeds to abstraction.
However, this abstraction has little similarity to the Aristotelian or
Thomistic concept of abstraction. The active intellect does not transform
sensible species into universal concepts: its action upon particular phantas-
mata contained in the imagination relieves them only of their rich, par-
ticular character of clear images and transforms them into shapes, indistinct
3. Bettoni 1954, p. 18.

4. Henry of Ghent 1502, Summa quarumdam ordinarium theologiae, f. 14"; idem, 1613, QuodUbeta aurea,
ff. 78», 206», 262»
images called 'universal phantasmata'. These phantasmata, in no way

identical with quiddity, can initiate the act of understanding, which arises
as produced by and impressed in the potential intellect. 5
The role of the potential and agent intellect does not end there. The
resulting knowledge is not a cognition of essence, nor can it be called
'science'. Scientific knowledge is 'necessary', and cannot, therefore, be
caused by changeable material objects and their nebulous images - the
universal phantasmata - in the intellect. Full perfect knowledge must
concern the nature of the given object, and that is possible only when this
object is conceived under the most general categories. The second stage of
cognition begins in the following way. The moment the potential intellect
is stimulated by the universal phantasma, not only the act of understanding
but also the most general notion of'being' appear in it, and then, with the
illumination of the initial knowledge by the agent intellect, the potential
intellect starts the process of cognition whose result is full knowledge of the
essence of the given object. This last process is assisted by divine illumi-
nation directed toward the concept and giving it full validity. 6
More drastic criticisms of Aristotelianism

A positive synthesis of the Aristotelian and Augustinian traditions was not
the only aim of the late-thirteenth-century philosophers. More drastic
criticisms of the Aristotelian notions of the potential and agent intellect
were also made at that time. Peter John Olivi, who founded his conception
on the basic Platonic and Augustinian opposition between the spiritual soul
and the material world, rejected the potential and agent intellect along
with the process of abstraction, phantasmata, and species.
DHHî Scotus
The turbulent disputes about Aristotelian, Platonic, and Augustinian con-
cepts of the intellect were the occasion of a new and final synthesis by Duns
Scotus, who in his lectures in Paris andOxford in 1297-1302 tried once
more to provide an unshakable theory of human cognition aimed at
satisfying growing doubts.
Duns Scotus' conception of the intellect and its activity, was based upon
the Avicennian concept of the nature of a thing. The nature of each being,
natura communis, is neither individual nor universal and this nature is
contained in each individual being in its pure state, without any individ-
uation. The human soul is conceived of in a twofold way: pro statu naturae
5. Henry of Ghent 1613, p. 2O0r; Γ502, p. I29 v -I3i r .

6. Henry of Ghent 1502, ff. 14", ijcf-tyi", Ι53Γ·
Criticisms of Aristotle's psychology 627
(according to its nature) it is a spiritual substance, and pro statu isto (in its
earthly life) it is connected to the body. The proper objects of cognition are
real individual beings and their natures. However, in this life, the human
soul joined to the material body cannot achieve this kind of cognition.
Therefore, the soul pro statu isto perceives the common nature of each
object in a twofold way: through the senses as it exists in individual beings
and through the intellect as a universal. The soul and its cognition pro statu
isto fit well with the Aristotelian conception, and the potential and agent
intellect seemed to Scotus adequate for explaining the process of intellec
tive cognition, starting with the senses and ending with universals.7
The active intellect is characterised, in line with medieval Aristotelian
tradition, as an active power which produces a real effect in the potential
intellect. The potential intellect, however, loses its Aristotelian character: it
becomes actuality and not potentiality; since it is treated as spiritual in
nature, and so knowable in itself, it must be in actu. Nevertheless, Scotus
saves its description as tabula rasa in the sense that it is able to know all
beings and is, therefore, indeterminate. The potential intellect is thus an
active potency which acts through its own power and needs only a determi
nation of the content of its act.8
The function of both intellects depends mainly on the conception of
their object. Since the quiddity (neutral nature) existing in particular things
is not particularised, there is no need to free it from individual conditions,
and no abstraction effected by the active intellect is necessary. The presence
of the nature in the phantasma joined to the illumination of the potential
intellect by the agent effects the intelligible species. The agent intellect is
therefore the real cause of the species, but the phantasma collaborates in this
act, determining its content. Therefore, there are two causes of species,
which are named causae partiales, where the active intellect is the principal
cause. However, the intelligible species is not identical with the act of
cognition. This act is produced jointly by the intelligible species and the
potential intellect, where the potential intellect - an active power - is the
9
cause of the act and the species is the cause of this act being about what it is.
Thus a new Augustinian conception was created at the turn of the four
teenth century founded partly on Aristotelian principles. But these prin
ciples have profoundly changed their original meaning. The role of the
agent intellect has become limited to action on the potential intellect and to
7. John Duns Scotus 1639, Reportata Parisiensia, 1. 3, q. 3, n6. Opus Oxoniense, 1. 1, d. 3, q. 6, n8.
8. Idem, Reportata Parisiensia 1. 1, d. 3, q. 4, Π4. Opus Oxoniense, 1. 2, d. 3, q. 8, n n .
9. Idem, Opus Oxoniense, 1. 1, d. 3, 1.6, ni5, n8.
the production of intelligible species; the potential intellect became an

active power of the act of understanding, whose potentiality was limited to
the indeterminacy of the content of its act. And the intellect as a whole
became the sole cause of the act of cognition. The separation of the material
from the spiritual was once more deep and unbridged. 10
Criticism and stagnation in the fourteenth and fifteenth centuries

Fourteenth-century philosophy presents a great variety of solutions to the
problem we have been dealing with. In that period, human cognition
interested philosophers and theologians much more than the nature of the
soul and its faculties. Thus the analyses of William Ockham treat kinds of
knowledge, intuition, abstraction, and experience much more than the
faculties of the soul. Peter Auriol, Durand de Saint-Pourçain and Henry of
Harclay tend to simplify the process of intellective cognition, eliminating
steps in the process. Durand acknowledges only the possible intellect, and
eliminates the active intellect and intelligible species. John of Mirecourt
rejects the process of abstraction; Nicolas of Autrecourt eliminates the
cognitive faculties of the soul.
However, the Aristotelian tradition is by no means non-existent in the
fourteenth century. Aristotle's De anima was included in the programme of
the Faculty of Arts, and commented on by all masters, who were thus
obliged to explain the conception of the possible and active intellects.
However, the main interest of these commentaries was centred not on the
nature of the possible and agent intellects, but on their function and the
process of abstraction. Problems concerning the necessity of the active
intellect, the production of intelligible species, the relation of the species to
the act of intellection and the possibility of simultaneous intellection of
several objects became the topics of chief interest.
After the intellectual revolution of Ockham, Auriol, Durand, John of
Mirecourt, and Nicolas of Autrecourt which overturned the traditional
conception of abstraction, the second half of the fourteenth century re-
turned to medieval Aristotelianism and its concepts, with an admixture, in
varying degrees, of neoplatonic ideas. By the end of the century, different
solutions coming from different schools blended together. Various solu-
tions discussed earlier were repeated and revived, but there were hardly any
new ideas of lasting importance.
io. For Ockham's theory of the intellect's role in knowledge, see J. Boler's contribution to this
volume.
32
FREE WILL A N D F R E E C H O I C E
Aristotelian and Christian backgrounds for the medieval discussion

Medieval teaching on free choice was inspired on the one hand by
Christian thinking and on the other by the moral philosophy of Aristotle as
expounded in the Nicomachean Ethics. According to Christian thinkers
man is given by God the possibility of choosing between good and evil, so
that his conscious decisions affect his ultimate fate, although salvation is
impossible without the intervention of divine grace. For Aristotle the
consequences of free choice are limited to life on this earth; but in other
respects there are similarities between his view and the Christian teaching.
For both, man can choose between good and evil as a consequence of his
capacity for rational judgement, which makes him significantly more
independent of his environment than other beings. According to Aristotle
man has a will or desire (boulësis) for what is good for him: when this is
combined with a judgement about what, in concrete circumstances, is con-
ducive to his good there results a choice (prohairesis). The criteria by which
such choices are made may differ from individual to individual and are by
no means uniform for all men. '
Medieval thinkers derived their Christian view of free will principally
from the Book of Genesis and from the Epistles of St Paul, but they were
also acquainted with Augustine and John of Damascus. The former, in
speaking of free choice, emphasised on the one hand the freedom and
spontaneity of human aspirations, and on the other the moral and theore-
tical problems associated with the liberation of human beings from the
fetters of doubt, suffering, and sin.2
In addition to the problems concerning human freedom of choice,
medieval Christian thought was concerned with the degree of freedom
i. Aristotle,Ethics, u 12Ί3-18.
2. Genesis 3.1-8; 2 Cor. 3.17; Rom. 20-3; Augustine, De gratia et libero arbitrio, PL 44,1,1, col. 881;
11,2, col. 882; x, 22, col. 894; Augustine, De libero arbitrio, PL 32,1,12,26, col. 1235; II, 1, i.col.
1240; II, 1, 3, col. 1241; III, 3, 7; Augustine, Retractationes, PL 32, I, 9, 4, col. 596. John of
Dramascus, Defideorthodoxa, PG 94, 928 B-C, 941 C, 944 B-C, 945 A-C.
accorded to angels, and with the independence and freedom which be-
longed to God in virtue of his omnipotence, limited to some degree as this
was through the laws established by him. This last problem was linked on
the one hand with the question of predestination, and on the other with the
miraculous intervention of the Creator in a natural world governed by
laws of his own making. 3
Liberum arbitrium
In presenting in English the thought of medieval philosophers on freedom
of choice, we meet with a linguistic difficulty. The cluster of problems
concerning human freedom and action which are discussed by modern and
contemporary English-speaking philosophers under the title 'freedom of
the will' were discussed in the Middle Ages under the heading 'liberum
arbitrium'. But the Latin expression cannot simply be translated by the
English one, because it does not contain the Latin word for will (voluntas),
and it was a matter of debate, among those who believed in the existence of
liberum arbitrium, whether it was the will, or some other faculty, which was
the bearer of the freedom involved in liberum arbitrium, and indeed whether
the will is free at all. The will itself was defined as the rational appetite, or
the desire for the good apprehended by reason, and not in terms of a
capacity for choosing between alternatives; hence its relation to liberum
arbitrium was something which required investigation. 'Freedom of choice'
is probably a less misleading translation of 'liberum arbitrium'; but here too
there is the difficulty that the Latin expression does not contain the
technical word for choice (electio). In what follows, therefore, we shall use
'freedom of decision' or else leave 'liberum arbitrium' untranslated.
Independence of spirit and determinism of matter

It was a matter of general agreement from the beginning of the Middle
Ages that freedom and the capacity of independent action were a preroga-
tive of beings who possessed intellectual understanding. In accordance
with this belief freedom was attributed in an absolute degree to God, in a
lesser degree to angels, and in a still lesser degree to human beings. From
the beginning of the thirteenth century, under the influence of Aristotle,
we find the notion of freedom associated with that of immateriality: the
less a being is conditioned by matter and material forces, the freer and less
3. Gilson 1932, II, pp. 100-18.

Free will and free choice 631
determined is its activity.4 Thus, Philip the Chancellor writes:

Aristotle says that the potential intellect can understand pairs of contraries, because
it is separate, or separable, with respect to matter, from either of them. Similarly,
the practical intellect is capable of operating in opposite directions, because it is free
in general from the bonds of matter.5
The contrast between the independence of spirit and the determinism of
matter became a characteristic feature of Franciscan thought. Merely
corporeal beings are enslaved by the necessity of nature and compelled by
its laws; irrational animals act by instinct and have no power to judge their
own actions and approve or disapprove them. Driven on by natural forces,
they have no freedom of decision because they lack the power of abstrac-
tion which is necessary in order to compare actions with standards of right
and wrong.6
Various aspects of human freedom

Among corporeal beings, then, only human beings enjoy freedom: but
human freedom has more than one aspect. Freedom from evil is not the
same thing as freedom from coercion. Only in the eternal happiness of
heaven can a man hope to be free from all evil and the many kinds of
misery on earth; the way to heaven is by freedom from the moral evil
which is sin. Freedom from sin is something different from the freedom
from coercion without which neither sin nor right action would be
possible in the first place. These contrasts were frequently drawn early in
the Middle Ages.
Thus Bernard of Clairvaux distinguishes between freedom from neces-
sity, freedom from sin, and freedom from misery. Every rational creature,
whether good or bad, enjoys freedom from necessity; freedom from sin
may be called the freedom of grace (libertas gratiae) since only grace can
confer it; freedom from misery may be called the freedom of glory (libertas
gloriae) since it is the prerogative of the glorious saints and angels in
heaven.7 Many medieval thinkers adopt this hierarchy of freedoms, with
the freedom of the blessed in heaven at its apex.8 Thus William of Auxerre
4. Cf. Lottin 1957, pp. 12-13; Gilson 1932, pp. 104-8.

5. Lottin 1957, p. 73.
6. Thus John of La Rochelle (Lottin 1957, p. 133); Alexander of Hales, Summa II, 403; Odo
Regaldus (Lottin 1957, pp. 160-1).
7. Bernard of Clairvaux, De gratia et libero arbitrio tractatus, PL 182, col. 1006.
8. Thus John of La Rochelle (Lottin 1957, p. 132); Philip the Chancellor (ibid., pp. 80-1) and Odo
Regaldus (ibid., p. 162).
rejects the definition of liberum arbitrium as the freedom to do what you

want, on the grounds that such a definition would make the damned more
free than the blessed, whose wills are fixed on God. 9 Other thinkers used
similar arguments.
Freedom to sin
If freedom is, as this view suggests, essentially freedom from evil, does this
mean that sinful action is not really free? This conclusion was drawn by
Anselm of Canterbury, who defined liberum arbitrium as the power to
maintain a righteous will. 1 0 By contrast, Anselm of Laon defined freedom
of decision as 'the power of acting well or badly'. According to him angels
and Christ possessed the power to do evil, but it would be absurd to think
of Christ making use of that power; it was used only by the angels who
proved unfaithful to God. Albert the Great agreed with Anselm of Laon
that when a human being wills evil he acts freely; but he did not agree that
God and the saints have the power to do wrong. Their inability to will evil
does not take away their freedom, however; it is the result not of coercion
but of the irrevocable fixing of their wills on goodness."
Superior and inferior reason

Just as it became common to deny that God and the saints could do wrong,
similarly there is a suggestion in some authors that even in this life reason,
which is the highest part of man, is incapable of sinning. Thus the author of
the Summa Porretana (between 1195 and 1210) offers a definition of liberum
arbitrium as the power of consenting to the reason or to the will. The will
may be a rational will or a sensual will; if the latter is consented to, an evil
choice results; if the former, the free decision is made for the better part.
According to this definition, then, if a choice is in accordance with reason,
the rightness of the choice is guaranteed. Reason can be said to be re-
sponsible for sin only in the sense in which the king can be said to be
responsible for the crimes committed in Paris in virtue of not repressing
them. But the author adds that this all applies only to the superior part of
the reason, or synderesis. Inferior reason possesses the ability to sin. 12
Like the author of the Summa Porretana, Roland of Cremona (a Domi-
9. William of Auxerre 1500, Summa aarea in quattuor libros Sententiarum a subtilissimo doctore magistro
Guillelmo Altissiodorensi édita, f. 64™.
10. Anselm, Dialogus de libero arbitrio, PL 158, col. 494. On Anselm of Laon, cf. Bliemetzrieder 1919,
pp. 27-8.
11. Albert the Great 1899, Summa de homine, pp. 585—6, 589—90.
12. Lottin 1957, p. 49.
Free will andfree choice 633
nican writing ca. 1230) made use of the distinction between superior and
inferior reason; but though only inferior reason can sin, in his view, both
superior and inferior reasons are free. Just as there are two reasons, there are
two types of liberum arbitrium. Superior reason is the reason which is the
bearer of sapientia (philosophical wisdom and learning), inferior reason is
the locus of prudentia (practical wisdom or prudence). Inferior reason can
turn itself in the direction either of good or of evil: it has freedom in respect
of the lower goods that are the subject matter of prudence. Superior reason
or intellect has freedom with respect to the higher goods: it cannot choose
evil, but it can will one good rather than another. Human beings, there-
fore, make two kinds of free decisions; the really free ones are those which
are made by superior reason, which govern its own actions and the actions
of inferior reason as well. ' 3
Theoretical and practical reason

Instead of, or as well as, the distinction between superior and inferior
reason, other authors in their treatment of freedom of decision made use of
a distinction between practical and theoretical reason. Thus, William of
Auxerre observed that reason makes judgements of two kinds. In a judge-
ment of theoretical reason we simply judge what ought to be done and
what ought not to be done; the other kind of judgement involves not only
deliberation but also a consequent imperative to do one thing rather than
another. It is this imperative judgement of practical reason which is the act
of the power of choice and which constitutes liberum arbitrium.1*
Reason, will, and liberum arbitrium

For William of Auxerre, therefore, free decision is essentially an act of the
reason; Philip the Chancellor, by contrast, maintained that freedom was
principally a matter of the will. But his position was not as distant from
William's as might appear, because he regarded reason and will as being in
substance the same faculty, given two different names because of two
different activities. Yet a third view appears in an anonymous treatise of the
first part of the thirteenth century: liberum arbitrium is to be identified
neither with reason nor with will, but is a third power distinct from both.
Whatever will and reason may tell us to do, it is still in our power to do it if
13. Lottin 1957, pp. 106-7.

14. Lottin 1957, p. 67.
we choose or not do it if we choose not. It is this power which is freedom of

decision, and which is the most powerful element in the soul. ' 5
The relationship between reason, will, and freedom of decision was a
major topic of debate throughout the thirteenth century. Eventually it
came to be characteristic of Dominican writers to link freedom very
closely with reason, and of Franciscan writers to locate liberum arbitrium
rather in the will. But in the earlier part of the thirteenth century there was
considerable variety of opinion among Dominican masters.
Thus Hugh of St Cher, writing ca. 1230-5, adopted a position resem-
bling that of Philip the Chancellor. Reason and will are a single faculty
which can perform different actions: telling the difference between good
and bad is called an act of the reason, choosing to enact one of a number of
proposed courses of action is called an act of the will. Liberum arbitrium,
considered as an act, is a combination of the reason's act ofjudgement and
the will's act of choice; if we consider them as powers or faculties, however,
liberum arbitrium, reason, and will are one and the same. 16
We find, however, a clear distinction between the will and liberum
arbitrium in the writings of the early Oxford Dominican, Richard Fishacre.
Richard distinguished between two types of volition, complete and
incomplete. Incomplete volition is the spontaneous motion of the will
which precedes rational reflection. Liberum arbitrium consists precisely in
the power of rational reflection upon spontaneous volition: it is thus
essentially an intellectual power, a manifestation of the power of self-
reflection peculiar to intellectual beings. Complete volition leads to the
carrying out of action after this reflection upon its moral value. Liberum
arbitrium thus operates to turn incomplete volition into complete volition
and is itself neither one nor the other. ] 7
Albert the Great regarded liberum arbitrium as a power distinct from
reason and will which arbitrates (hence its name) between the dictates of
reason and the aims of will in cases where there is a conflict between the two.
Reason judges a proposed action as good or bad, and will endorses or
rejects an object of desire; free decision chooses either the object approved
by the reason or the object pursued by the will. 18
Three stages may be distinguished in the evolution of Albert's thought.
In his Summa de homine he emphasised the role of free decision as mediator
[5. Lottin 1957, pp. 8 8 - 9 .

16. Lottin 1957, pp. 100—1.
17. Lottin 1957, pp. 110-11.
18. Albert the Great 1899, Summa de homine, p. 575.
Free will andfree choice 635
or arbitrator between the two distinct faculties of reason and will. In the
commentary on the Sentences the function of liberum arbitrium is to link the
desires of will with thejudgements of reason, and the notion of arbitration
is not stressed. Finally, in the commentary on Aristotle's Ethics free deci-
sion, while remaining distinct from both reason and will, is described as
involving the activity of both faculties. In being a judgement, free decision
involves the activity of reason; in being free, it involves the activity of the
will.
Thomas Aquinas likewise adopted different approaches to the relation-
ship between reason, will, and freedom at different periods of his life, from
the early commentary on the Sentences through the De veritate and the
commentary on the Ethics to the Summa theotogiae in the last years of his life.
Having said in the commentary on the Sentences that choice is an act of the
will influenced by preceding rational deliberation, he explained in the De
veritate that liberum arbitrium is not a faculty distinct from the will, but the
will itself considered as the power of choosing, choice being just one of the
several kinds of act it can perform. It is the power by which we judge
freely. Judgement itself is an act of reason, but what gives freedom to
judgement is the will. In the commentary on the Ethics he describes the
function of the will as aiming at the ultimate goal, while choice selects the
means conducive to this goal. Finally, in the Summa theologiae he drew a
comparison between cognitive and volitional capacities. Understanding
self-evident truths and drawing conclusions from them are two different
actions, but both are acts of the same faculty, the intellect; similarly, aiming
at an ultimate goal and freely choosing means to that goal are two different
actions, but both are acts of the same faculty, the will. Hence liberum
arbitrium is not a faculty distinct from the will. But for Thomas it is not the
will but the intellect which has the major role in the moral activity of
human beings. For the intellect is the final or teleological cause of the will's
action: it is the intellect which presents to the will its ultimate aims. This last
opinion gained wide popularity through its acceptance by the Thomist
school, and also had an influence on the outlook of many thinkers frdm
other schools such as Buridan and his followers. 19
The Franciscan masters of the thirteenth century were much more
unified in their approach than the Dominicans. However, three different
positions can be identified. John of la Rochelle followed Philip the
19. Thomas Aquinas, In II Sententiarum, dist. 24, quaest. 1, art. 2; De veritate, quaest. 24, art. 6;
Thomas Aquinas 1969, Sententia libri Ethicorum., pp. 133-4; ST, I, q. 82, 1,2; cf. Siewerth 1954.
Chancellor in identifying reason with will and regarding liberum arbitrium

as being itself identical with this single faculty of rational appetition.
Alexander of Hales thought that freedom of decision was a power distinct
from reason and will which was supreme in the soul and could command
both the other faculties. Liberum arbitrium, reason, and will, he maintained,
formed a trinity in the appetitive part of the soul comparable to the divine
Trinity of Father, Son, and Holy Ghost. Bonaventure, on the other hand,
thought that liberum arbitrium was not a faculty at all, but a habitus or
disposition. Neither reason nor will alone suffices to exercise control
over human action: freedom of decision is a disposition created by the
co-operation of the two faculties.20 Bonaventure's view that freedom
of decision is not a faculty but a habitus was followed at Oxford by
the Franciscan Richard Rufus of Cornwall and by the Dominican Robert
Kilwardby.21
Radical voluntarism
In the second half of the thirteenth century the writers of the Franciscan
school began to espouse a radical form of voluntarism. Thomas Aquinas
had taught that the will acts according to the directives it receives from the
reason. Walter of Bruges attacked this doctrine. He distinguished between
the natural will, a passive appetite which tends towards good, and the real
will which has the power to judge and decide, an active power which can
freely accept or reject what is presented to it by reason. The will is
independent of the intellect in the sense that whatever object the intellect
presents to it may be freely chosen or rejected. The judgement of practical
reason is also free, but not to the same degree as the will, which has absolute
power over its own actions.22 William de la Mare, likewise, thought that
the will was both a passive and an active power: passive in that it must
receive its object from the intellect, active in that it moves itself and
activates all other parts of the soul.23 Similarly Matthew of Aquasparta: the
will while needing enlightenment by the intellect itself activates the intel-
lect, giving it commands and stimulating it to action. The will is superior
to the intellect because its virtue, charity, is superior to the intellect's virtue,
20. John of la Rochelle (Lottin 1959, pp. 129-34); Alexander of Hales, Di libero arbitrio, MS Todi Bibl.
Commun. 121, ff. 45v*-46v"; Odo Regaldus (Lottin 1959, pp. 152 ff.); Bonaventure, In II
Sententiarum, 25,1, 2 and 6.
21. Lottin 1959, pp. 190-2.
22. Lottin 1959, p. 243; similar views were held by Gerard of Abbeville, ibid., pp. 249-50.
23. Lottin 1959, pp. 272-3.
faith; and its object, goodness, is superior to the intellect's object, truth. 24
One who was influenced by the voluntarism of the Franciscan school
was the bishop of Paris, Stephen Tempier, who in 1277 condemned a
number of propositions concerning freedom of choice. He declared un-
orthodox the view that the will is a merely passive power and that human
action can be necessitated by external forces. He also condemned the view
taught by Thomas in the De veritate that the will must follow the dictates of
reason.25
Tempier's condemnation of determinism led the majority of thinkers at
the end of the thirteenth and beginning of the fourteenth centuries to
maintain that the will, and not the reason, was the seat of freedom and
autonomy. This, at least, is the impression we get from the writing of Peter
de Falco, who maintained that the will was the root of human freedom.
Reason is necessary to enable us to conceive the possibilities of choice open
to us, and to deliberate about them; but it is the will which produces free
human action since it depends on the will whether a man accepts or rejects
what the reason proposes.26
Attempts to re-establish the supremacy of reason

The Augustinian Giles of Rome attempted to reconcile Aquinas' doctrine
of the superiority of reason to will with the Franciscan thesis that the will
was self-determining. According to Giles, when reason presents the will
with absolute good the will cannot help but desire it; but the will is free to
control its own actions where a choice between relative degrees of good-
ness is concerned: it can attend to the good or evil aspects of the object
presented by reason.27
In Paris the Dominican John Quidort defended against the Franciscan
thinkers the doctrine of Thomas in the De veritate that the will cannot but
desire what reason presents to it as the better. Similarly, Godfrey of
Fontaines maintained that the Franciscan theory of the self-determination
of the will was unintelligible.
In Oxford as in Paris Aquinas found defenders. Thomas Sutton and
Nicholas Trivet took his side in the debate about the superiority of the
intellect over the will. Sutton maintained that the will is passive with
24. Mathew of Aquasparta 1903,8, ad 4, ad 22; 9, ad 4, ad 10. For a similar position see Henry of Ghent
i6i3,Quod7. I, qq. 14-16.
25. CUP, 543-58, propositiones 151, 152, 157,158, 160,163-6. See Hissette 1977, pp. 230-63.
26. Lottin 1959, pp. 283-5; see also Richard Middleton (Lottin 1959) and Roger Marston 1932,
Quaestiones disputatae, pp. 438-9, 441-2, 447-50.
27. Giles of Rome 1646, pp. 177-9.
63 8 Philosophy of mind and action
respect to the object of its desire, but free in its choice of means to achieve its
object. But even in this choice of means it is dependent on the judgements
of reason. Trivet, likewise, maintains that both will and reason are free,
though each in a different way. The dependence of will on reason does not
conflict with its freedom, which is freedom of choice of means in the
endeavour to achieve the aim presented by reason. Unlike Jean Quidort,
these Oxford masters do not defend Thomas outright: they make smaller
or larger concessions to Tempier's condemnations. 28
Voluntarism in Scotus and Ockham

But again, in Oxford as in Paris, it was the Franciscan supporters of
voluntarism who were dominant at the end of the thirteenth century. The
most important of these was John Duns Scotus. Scotus saw the problem in
terms of a contrast between an order or system of nature and an order or
system of liberty. A single human act with a single object cannot belong to
both of these orders at the same time: it cannot be both the result of natural
necessity and of the freedom of the will. In God, indeed, willing and
necessity can coincide: God necessarily loves (his own) infinite goodness,
but it is not thus with created wills. God's love and God's will are identical
with God's essence, but a human being's love and will are not identical
with his essence. A human will need not aspire to a goal presented to it by
the human intellect; it can act or not act, and it can act in any way it pleases.
It is the will which controls the exercise ofother human abilities: thus it is
superior to the intellect since it controls it too. In line with the Franciscan
tradition Scotus argues that the will is superior to the intellect because
charity is superior to knowledge and goodness is superior to truth. 2 9
William Ockham's theory of the will is similar to Scotus'. Ockham has a
twofold concept of freedom: as independence of constraint, and as spon-
taneity of action. Will is a free and active force, which can desire or not
desire whatever is presented to it or dictated to it by reason. The will is free
to choose its own goal as well as being free in the choice of means to a goal.
Nature is determined by the laws which govern it; in virtue of possessing
free will man belongs not to the order of nature but the order of liberty. 30
28. John of Paris, Quaestiones, MS Basel Bibl. Univ. B. Ill, 13, f. 31™ rb, ioov*-loi™; Sutton (Lottin
1959, PP· 365-71, 349-54); Trivet (Lottin 1959, pp. 378-82).
29. Quaestiones quodlibetales XIV-XXI; John Duns Scotus 1895, pp. 180-1, 184, 188-9, 199, 241;
Quaestiones in quartum librum Sententiarum (Opus Oxoniense); John Duns Scotus 1891-5, XXI,
pp. 97,123, 151, 155; Cf. Gilson 1952b. pp. 586-7, 597.
30. William Ockham 1483, In IVSententiarum, d. 16, q. 1 ; Ockham 1481, QuodUbeta septem, f. b 3 , b , b
4'; Ockham 1491, Expositio super Physicam Aristotelis, f. 117'.
Buridan on two types of human freedom

One of the fullest and most interesting medieval accounts of human
freedom was that of John Buridan, presented between 1342 and 1354.
Buridan distinguished between two types of freedom possessed by human
beings: freedom of choice (libertas oppositions) and freedom to aspire to a
goal (libertasfinalisordinationis). Freedom of choice is not absolute, because
a man must necessarily aim at what is good; but he is free in choosing
different paths to its attainment. In the act of choice both the intellect and
the will are involved, and each of the two faculties is both active and
passive. The intellect derives the content of its knowledge from the
external world, and to that extent is passive; but in informing the will what
to seek and what to avoid it is active. Similarly, the will is passive in taking
information from the intellect, but active in making a free choice among
various possible courses. When it first learns of a good or of an evil, the will
spontaneously, but necessarily, feels pleasure or displeasure; but this reac-
tion is not its specific activity, which is the acceptance or rejection of the
object in question. In this act it is free and independent of the dictates of the
intellect.
Besides this freedom of choice human beings possess another freedom:
the freedom to aspire to a goal selected by the intellect. It is this libertasfinalis
ordinationis which is the more perfect freedom. So Buridan, though he
might be called a voluntarist, emphasises the role of the cognitive faculty in
fixing the goals of action. Since it is the intellect which selects the objects of
human aspiration, the intellect is superior to the will.31
Buridan's followers, especially in central Europe (e.g. in Prague, Vienna,
and Cracow) concentrated on the topic of freedom of choice, and ignored
Buridan's second type of freedom. In doing so they made Buridan appear a
more thoroughgoing voluntarist than he really was.
Divine power andfreedom

In the fourteenth century, alongside the debates on human freedom there
was much discussion of the omnipotence and freedom of God. Duns
Scotus emphasised that God was independent of the laws of nature and
could suspend these laws which he had himself established. His omnipo-
tence was twofold: absolute omnipotence (potentia absoluta) and omnipo-
tence within the limits of the laws which he has himself established (potentia
31. John Buridan 1513, Quaestiones super decern libros Ethicorum III, quaest, 1-6; X , quaest. 1-3. Cf.
Monahan 1954, pp. 7 2 - 8 ; Walsh 1964, pp. 50-61; 1966,1-13; Korolec 1974a, 1974b.
640' Philosophy of mind and action
ordinata). If we consider the first kind of freedom we can say that God has
the power to order men to hate him, thus suspending the law of love which
he has given. The only limits of this omnipotence are laws of logic such
as the principle of non-contradiction. Ockham defended a similar view,
saying that in respect of his absolute will (voluntas beneplaciti, correspond-
ing to potentia absoluta) he can do whatever he wants, whereas in respect of
his manifested will (voluntas signi) he wills only in accordance with the
commands he has given. 32
The divine will and the human will

The relationship between the human will and the divine will is most fully
discussed in Thomas Bradwardine's treatise De causa Dei, written before
1325 against the Pelagian heresy. Bradwardine emphasises the absolute
dependence of man's will on God's: human beings can free themselves
from the influence of psychological forces and from the influence of the
stars, but they cannot become independent of God. But in spite of the
absolute divine influence on human behaviour, it is man and not God who
is the cause of sin. The action itself is caused by God, but its sinfulness is
caused by man himself. 33
Robert Holkot, by contrast, is prepared to admit a sense in which God is
the author of sin. Using Ockham's distinction between voluntas beneplaciti
and voluntas signi, he says that God is the author of sin in so far as he can do
what he desires, but cannot be the author of sin because of the will which
manifests itself in the commandments. 3 4 The same problem was also
discussed by Thomas Buckingham (1333 or 1338) who maintained that
every human action was performed in direct dependence on God. By his
prescriptive will (voluntas approbationis) God desires actions which are
morally good; by his absolute will he permits sin and the corruption of
human action. What counts as good action and what counts as evil action
depends only on the decision of God. 3 5
Conclusion
The most striking feature of the medieval discussions of the problem of the
freedom of choice is the number of scholars who supported a moderate
32. Gilson 1952, p. 366; Ockham 1483, In IV Sententiarum, d. 16, q. I.

33. Leff 1957, pp. 57-64, 91-7; Michalski 1937, 320-2, 326, 344-5, 350-1.
34· Michalski 1937, 303-97. 345~6.
35. Ibid. 307—10,347—8. A similar set of distinctions was drawn up in Paris by John of Mirecourt; see
ibid. 275-9, 326-31, 346-7, 352-8.
voluntarism, in contrast to the rationalism of Thomas Aquinas and his

followers. It was voluntarism which had far the greater influence on the
European mind in the second half of the thirteenth century and in the two
succeeding centuries.
33
THOMAS AQUINAS ON
H U M A N ACTION
Philosophical and theological motivations for Aquinas' work

At least two distinct purposes may be discerned in Aquinas' various
writings on human action. One is to complete and correct Aristotle's
treatment of it in the Nicomachean Ethics, to which he of course pays close
and respectful attention. A second springs from his primary commitment
to theology. Reflecting on what is said in the Scriptures and the writings of
the Fathers about such topics as the fall of Adam, sin, conversion, and the
operation of grace, theologians produced a body of doctrine about various
aspects of human acts. To Aquinas' mind, this teaching settles certain
questions authoritatively: as when it declares that voluntary human acts are
commanded by their agents freely, and not by necessity. In addition it
introduces certain concepts into the theory of action, for example, those of
enjoyment and consent. Aquinas undertakes to incorporate these contri-
butions of theology, where sound, into a revised Aristotelian theory.
Aristotelian causal theories

Aristotelian theories of action are causal, and causal in a distinctive way. To
do something, to perform an act, is to cause something. And causing
something is always to be investigated in terms of a pair of fundamental
concepts, dynamis and energeia, which appear in Thomas' Latin as potentia
(potency) and actus (act). The power or capacity of an object to cause
something — whether a change of state, or a persistence in a state — largely
determines what that object is. Brute animals are distinguished by their
possession of powers of sensation and bodily movement. Human beings
are differentiated from brute animals by their further possession of intel-
lectual powers, making it possible for them to cause changes or persistences
in the state of things in rationally pursuing ends they have rationally set
themselves. So to cause a change or persistence is to perform a human act.
Fundamental distinctions regarding actions

Two distinctions pervade Aquinas' treatise on the nature of human acts (de
conditione humanorum actuum), his final and most elaborate discussion of
Thomas Aquinas on human action 643
human action, which occupies Summa theologiae lallae, 6-17. The first,
drawn in Summa theologiae lallae, 1, 1, is that between those doings of
human beings that are human properly speaking and those that are not. A
human act, properly speaking, is an act in which the human agent exercises
the distinctively human capacities of reason and will. Accordingly, acts in
which those capacities are not exercised, like absent-minded twitches of
hand or foot, or strokings of the beard, although acts of a human being
(actus hominis) are not human acts (actus humant). The second is Aquinas'
equally sharp distinction between the complete human acts in which the
human capacities of will and intellect are exercised, and the exercises of
those powers which go to make up complete human acts. A complete
human act is a unity in which, by virtue of certain intellectual acts, some
possible act is commanded, and by virtue of certain acts of will, the
commanded act is performed. ' Neither an intellectual act, such as thinking
that it would be good to have a higher salary, nor an act of will, such as
willing to do what needs to be done to get more money, are by themselves
considered complete human acts by Aquinas. Yet both may be compo
nents in a given person's complete act of working overtime.
Aristotle on voluntariness and choice

According to Aristotle, an act is voluntary if its moving principle is in the
agent himself, he being aware of the circumstances of his act and of the
objects with which it is concerned (Nie. Eth. m o b 3 2 - 3 ; 1111*22-3).
Hence brute animals, acting from bodily appetites such as hunger, act
voluntarily (1111 b 8). Human action is distinguished from that of brutes as
exemplifying not only voluntariness, but also choice. Choice involves
not only will for an end, but also deliberation about how to attain it
(1139*31-2). Both call for intellect; for deliberation is an intellectual
operation, and will presupposes an intellectual grasp of what is willed.
However, since he held that human beings by nature will to live a good life,
and that their characters reveal what they take a good life to be, Aristotle
was reluctant to acknowledge that a man acting out of character or on
impulse genuinely chooses at all (1139*31-4).
Aquinas' modifications of Aristotle

Aquinas rejected this. He retained the structure of Aristotle's analysis of
chosen acts; but he insisted that if human acts as such involve choice, then
1. '... Imperium et actus imperatus sunt unus actus hominis, sicut quoddam totum est unum, sed est
secundum partes multa' (5 Γ, lallae, 17,4c). '... nihil prohibet, in his quae sunt multa partibus et
unum toto, unum est prius alio; sicut anima quodammodo est prius corpore' (ibid, ad 3).
choice cannot be subject to restrictions according to which acts that are out
of character cannot be chosen. Aristotle's restricted concept of choice
follows from his restricted concept of will. Aquinas therefore proposed a
new and less restricted concept of willing an end, which in turn made it
possible for him to introduce, as the fundamental explanatory concept in
the theory of action, the derivative concept of intention.
Will as rational appetite

In Aquinas' theory, voluntary human acts are identical with human acts
involving choice. All such acts are exercises of the same power, the power
of will, in defining which Aquinas simply appropriates Aristotle's de-
finition of the power of choice: ' "Will" ', he writes, "means rational
appetite" (lallae, 6, 2; cf. Nie. Eth. Ii39 b 4). As rational, will is moved to
activity by the activity of the related power of intellect; as appetitive, its
activities are for the attainment of ends. The power of will, as rational
appetite, is directed to the rationally appetible, that is, to the character of
good (ratio boni), which is intellectually grasped as being common to all
things (bonum in communi) (lallae, 8,1 and 2; 10,1 and 2). A particular act in
which the power of will is exercised can be directed to an object only
inasmuch as it is taken to be good (sub ratione boni) (lallae, 8, ic).
Exercises of an appetitive power, which by its nature has to do with the
attainment of an end, necessarily involve two kinds of subordinate acts:
acts directed to the end to be attained, and acts directed to the means by
which the end is to be attained.
Simple acts of the will

The power of will, according to Aquinas, is exercised in no fewer than
three kinds of acts directed towards ends. One, however, is fundamental;
and, with unfortunate ambiguity, Aquinas calls it by the same name as the
power of will itself, namely, 'voluntas'. Fortunately, when using this
ambiguous word would mislead, he usually has recourse to the less
ambiguous phrases 'actus voluntatis' (act of will), sometimes preceded by the
epithet 'simplex' (simple), and 'motus voluntatis' (motion of the will) (e.g.,
lallae 8,2 and 3; 10,2 and 3). A simple act of will is a 'motion' of an agent
inasmuch as he is intellectual (an interior motion, therefore, and not an
exterior bodily one) towards the attainment of a particular thing taken to
be good as an end.
It is of the first importance to follow Aquinas' example (in lallae, 8,2c.)
and to distinguish between what he says about the power of will and what
he says of the acts in which that power is exercised. Only of the power of
will is the object said to be the character of good (ratio boni) (la Ilae, 8, 2c).
And the sense of saying that is clear: namely, that nothing can be an object
of an act of that power except as having that character. Acts of will,
accordingly, have as their objects particular ends sub ratione boni. By
contrast, the traditional interpretation is obscure, according to which
Aquinas holds that the first act of will in a complete human act has good in
communi for its object.2 What would it be to will something incompatible
with good in communi? Aquinas insists that the simple acts of the power of
will have only ends as their objects, while at the same time arguing that the
power of will has as objects means as well as ends 'because the character of
good, which is the object of the power of will, is found not only in the end,
but also in those things which are for the end' (lallae, 8, 2c). Hence, to foist
on him the doctrine that there is a simple act of will having the character of
good as its object, with the implication that there are simple acts of will
extending to means as well as to ends, would gratuitously make him
contradict himself.
Any intellectual act in which something is affirmed to be an attainable
good can give rise to a simple act of will directed to attaining it. Aquinas
insists that an end willed need not be good as a matter of fact (in rei veritate),
finding for his departure Aristotelian credentials of a sort in Physics II,
195*26 (lallae, 8, ic). From this, taken together with the theological
doctrine that the only good attainable by man that is complete and lacks
nothing, namely, the beatific vision, cannot be intellectually grasped by
natural means, he further concludes that in this life the will is not moved to
any end necessarily (lallae, 10, 2). No end any human being can think of
by his natural powers can move his will necessarily, because, since no such
end is completely good, it is open to him to incline instead to some good he
recognises that end to lack (lallae 10, 2c).
Enjoyment and intention

Besides what he called simple acts of will, Aquinas recognises two other
kinds of acts of will directed towards ends: enjoyment (fruitio) and inten-
tion (intentio). By the former, he enriches the Aristotelian theory from
theological sources, above all from the writings of Augustine, whence are
2. By the 'traditional' interpretation. I mean the interpretation that was to be found in most manuals
of Thomist philosophy until very n-ccntly. The source of many of these Treatments appears to be
the article 'Acte' in Dictionnaire de Théologie Catholique, by A. Gardeil. Compare the scheme
presented by Thomas Gilby in Thomas Aquinas 1964-76, vol. 17, p. 211.
drawn the proof texts in all four articles of the question defruitione (lallae,
11). By the latter, he further develops his revision of Aristotle.
Enjoyment
The Aristotelian characterisation of the will as rational appetite says both
less than can be said about why human beings persist in ends despite
accumulating evidence of their unsatisfactoriness, and less than theologians
have taught about the rewards of persisting in ends that are satisfactory.
Augustine had distinguished enjoyment from passive pleasure by main-
taining that it is through the will that we enjoy. 3 And he had defined
enjoying as 'cleaving with love to something for its own sake'. 4 Since
anything simply willed as an end is willed for its own sake, although it may
be willed for the sake of some further end as well, it is natural to ask
whether every simple act of will is accompanied by an act of enjoyment.
Aquinas' answer is implicit. 'Anyone', he declares, 'has love or delight
from what he ultimately looks for (de ultimo exspectato), which is [his] end'
(lallae, 11, 1 c.). This implies that every simple act of will must have an act
of enjoyment as its counterpart.
W h y did Aquinas not simply add 'involving cleaving with love' to his
account of a simple act of will as an interior motion to something as an end
sub ratione boni? T w o reasons can be found in his text. First, although either
one simply wills something as an end or one does not, one enjoys an end
more the better it seems, and only the beatific vision completely (lallae, 11,
3a); moreover, one enjoys an end perfectly only when it is possessed, and
imperfectly when it is merely intended (lallae, 11, 4 c ) . Enjoyment, in
short, has degrees, but simple willing has not. Secondly, with the actual
possession of the beatific vision, 'the delighted will comes to rest' (lallae,
11, 3c); but that is when enjoyment is at its height.
Intention
The fundamental theme of an Aristotelian theory of action is that willing
an end generates action by way of the agent's deliberation about means
(literally, 'those things that are for the end'). Following Nemesius, Aquinas
held that means are possible acts by which the end can be attained (lallae,
14,4c). Unless an appropriate act of deliberation ensues, simple acts of will
remain barren. But once an appropriate act of deliberation has disclosed
means by which an end willed can be attained, the simple act of willing that
3. Augustine, De trinitate X, 10 (PL 42, col. 981); quoted in ST, lallae, 11,3 obj. 3.
4. Augustine, De doctrina Christiana, I, 4 (PL 34, col. 30); quoted in ST, lallae, n , ic.
end generates a further act directed towards it: namely, an act of willing it
through the means. Aquinas offers a simple but effective illustration. When
we simply will to be healthy, our act is a simple act of will, and no more;
but when we will to be healthy as an end to which some possible act is
ordered, we perform a further act, an intention (lallae, 12, 1 ad 4; cf. 4c).
The deliberation by which an intention is generated can be more or less
complete. Aquinas remarks that 'there can be an intention of an end even
though the means have not yet been determined' (lallae, 12, 4 ad 3). As
soon as it is judged that there is a possible act by which an end can be
attained, the end may be willed through that possible act. Intentions, in
short, may be more or less determinate.
Choice
For reasons which will become plain, Aquinas' treatment of choice largely
recapitulates his treatment of intention. He adopted not only Aristotle's
view that willing an end gives rise to choice by way of deliberation, but
also his conception of acts of deliberation as consisting of analytical ques-
tioning (quaerere et resolvere), terminating in a judgement, in which the end
willed as the effect of the contemplated action is resolved into its simple
causes - the means by which it may be attained (lallae, 14, 1 and 5; cf. Nie.
Eth. i i i 2 b i 2 - 3 i ) .
Complete acts and component acts

A further complication can no longer be put aside. Deliberation is itself
often a complete human act; for example, doing a sum in my head, as part
of the act of balancing my monthly accounts. Hence many human acts
must be complex, in the sense of having other complete human acts as
components. Familiar examples given today are repetitious acts, such as
driving home a nail by repeated blows with a hammer.5 But what may be
called organised acts, consisting in a number of different complete human
acts, the performance of which in a certain temporal order (some or all may
be simultaneous) is designed to bring about a certain result, are more
common and more important. Making a cake and tying a complicated
knot are homely examples.
Deliberation as a component of any complete act

The fact that the act of deliberation which, according to any Aristotelian
theory, is a component of any complete human act, may itself be a
5. Cf. Goldman 1970, pp. 35-7.

complete human act, raises the spectre that deliberation is a process that
must go on in infinitum. This would necessarily be the case if every act of
deliberation were a complete human act. However, Aquinas argues that
some are not. It is true that any question may be made an occasion for a
complete human act of deliberation. But since human action does occur,
there must be simple human acts in which the act of deliberation is not itself
a complete human act. This will be so when the agent's end is resolved into
means unproblematically, because the resolution is a matter either of the
agent's scientific knowledge (disciplina) or technical skill (ars), or has to do
with a trifle in which nothing that might come to mind would seem
possibly to bein error (lallae, 14,4 and 6). Thus in making a cake according
to a recipe I know by heart, my technical skill enables me to judge how
much of a given ingredient it is good to add as soon as the question comes
up, without any process of calculation.
Choosing the means

Aquinas points out that the judgement or verdict in which a 'practical
syllogism' of deliberation terminates (lallae, 13, 1 ad 2) is not apodictic. 6
An agent resolving the attainment of an end into its simple causes will
produce a premiss to the effect that it will be attained on a certain
condition: say 'If M is adopted, Ε will be attained.' Quite obviously, from
such a premiss, taken together with the judgement, Έ is good to attain',
which led to E's attainment being willed, it does not follow that 'M is the
suitable means to adopt.' For the conditional premiss does not exclude
others: for example, 'If Ν is adopted instead of Μ, Ε will be attained'
(lallae, 13, 6 ad 2).
As soon as his deliberation has resulted in a judgement that it would be
well to adopt some determinate means for the sake of the end willed, it is
open to the agent to exercise his will and choose that means - or not to,
either abandoning the end willed, or choosing to deliberate further. If he
chooses the means indicated by his deliberation, however, is his choice not
identical with his intention?
Aquinas leaves no doubt that in certain circumstances it is. In the body of
lallae, 12, 4 he lays it down that 'to tend towards an end and to tend
towards that which is for the end is one and the same motion of the will in
subject (subiecto)'; and he goes on to say that that same motion of the will,
which is one in subject, differs according as we think about it either
6. Anscombc 1965, pp. 152-3.

primarily in terms of the means (choice), or primarily in terms of the end

(intention) (lallae, 12, 4 ad 3). The reason why Aquinas' treatment of
choice largely recapitulates his treatment of intention is now plain: when
its object is determinate, an intention is the same act as the corresponding
choice.
Just as an agent does not merely tend to his ends, but cleaves to them
with love, so having in deliberation arrived at a judgement as to what
means to his ends are good, he does not simply tend to those means,
but approves and loves them (lallae, 15, 3). This Aquinas calls consent
(consensus), identifying it with what the translators of John Damascene
referred to as 'sententia'. However, unlike enjoyment, which is full only
when the end of all human action has been attained, consent is logically
prior to choice, and may precede it in time. For consent, unlike choice, is
accorded to suitable means to one's ends as soon as they are judged to be
such. In most simple human acts, only one means to his end ever comes
into the agent's mind, and he consents to it and chooses it together; but, as
Aquinas remarks, it may happen that deliberation discloses several suitable
means, one of which is judged most suitable, and in such cases, the agent
approves and loves all the means judged suitable, while going on to choose
the one judged most suitable (lallae, 15, 3 ad 3).
Commanded acts and acts of command

Aquinas has now developed his revision of Aristotle's theory of action
sufficiendy to elucidate his own conception of a complete human act as a
unity, among the components of which are a commanded act and an act of
command (lallae, 17,4c). Commanded acts are commanded by the will,
but only as a subordinate power (lallae, 1, 1 ad 2): ultimately they are
commanded by the intellect (lallae, 17, ic). This follows from his de-
finition of commanding as 'ordering (ordinäre) somebody to something
that is to be done, with a certain intimating motion (intimativa motione)'
(lallae, 17,2c). He explains that intimating is primarily expressed by indica-
tive sentences containing gerundives, of the form 'Hoc est tibi faciendum'
(This is to be done by you), and only secondarily by imperatives (lallae, 17,
ic). In a complete human act, a commanded act will be that component
which the agent judges is to be made to happen, and which does happen as a
result of an act of will which he performs in view of that judgement.
Taking Aquinas' theory as so far developed, it is natural to identify the act
of command (imperium) with the judgement that terminates deliberation;
and to identify the mediating act of will, as a result of which the com-
manded act happens, with the act of choice. Aquinas, as we shall see, adds a
further act of will (namely, usus), and provides it with an appropriate
generating intellectual act; but that, I shall argue, is muddled.
Commanded acts, elicited acts, and executive powers

Yet it is essential to Aquinas' theory, as nobody seriously disputes, that in
no complete human act can the component commanded act be identical
with any of the acts of intellect and will that giveriseto it. For every one of
those latter acts -judging an end good, willing to attain it, enjoying it,
deliberating about what would be the suitable means of attaining it,
consenting to such means as there are, and choosing the most suitable - is an
immediate exercise of the power of intellect or of will, which excludes any
exercise of any subordinate power (lallae, 1, 1 ad 2; 6,4c). Each of them is
elicited, not commanded.
What acts, then, are commanded? The most obvious complete human
acts are bodily, for example, raising one's arm.7 When an act is a bodily act,
the commanded act will be a motion of one's bodily members, for
example, one's arm's going up. The occurrence of such a commanded act is
always an exercise of what Aquinas calls the agent's executive power with
regard to his body, an exercise that is immediately caused by some act of
will (lallae, 16,2c; 17, 9c). Each human being finds out by experience the
range of his executive power with regard to his body (lallae, 17,9 ad 2 and
3). Aquinas expressly declares that both acts of intellect and acts of will can
also be commanded, and even, up to a point, acts of sensitive appetite
(lallae, 17,5-7)·
Why he does so appears most clearly from what he says about com-
manded acts of intellect. It follows from the nature of elicited acts, as we
have seen, that no elicited act of intellect can be commanded: one cannot
command dissent from what immediately strikes one as true. But not all
acts of intellect are elicited. When human beings ask themselves whether a
certain answer to a question is true or false, it often happens that no
judgement is immediately elicited. Should their intellectual powers fail
them in this way, it is open to them to deliberate whether their ends will be
served better by assenting to a given answer, or by dissenting from it. And
so, in the absence of an elicited act of assent or dissent, assent and dissent
themselves may be commanded (lallae, 17, 6).
7· Wittgenstein, Philosophical Investigations I §621, has made this example the standard one.
The article (lallae, 17, 5) in which Aquinas treats of commanded acts of

will is perilously succinct. It will be misunderstood if it is forgotten that,
when a certain possible act is commanded in an act of intellect and then
chosen in an act of will, what is commanded is the possible act that is
chosen, and not the act of choosing or willing it. The crucial premiss in
Aquinas' demonstration that acts of will are commanded is that a man 'can
judge that it is good that [he] will a certain thing' (lallae, 17, 5c). He has
not forgotten that what is judged in deliberation is that a certain means is
most suitable: his point is that sometimes the means judged most suitable
for attaining an end may itself be an act of will. And so the act of choice that
ensues, if one does, will be an act of willing that act of will. Although
Aquinas gives no examples, they are not far to seek. Resolving to act in a
certain way in the future, one of the means by which moral virtues are
acquired, is one kind of act of will that can be commanded.
The carrying out or execution of a commanded act, as Aquinas describes
it, is a matter of the operation of certain human interior principles, namely,
the powers of the soul such as intellect, will, and sensible appetite, their
trained dispositions (habitus), and the organs or bodily members, each
conceived as the seat of powers such as to see, to feel, to move in such and
such a way, and the like (lallae, 16,1 c ) . These principles are also referred to
as executive powers (lallae, 17,8c). They are set in operation by acts of will
(lallae, 16, 1). Aquinas therefore holds that, in a complete human act,
the actual execution of the commanded act is related to a certain elicited act
of will - necessarily the last one - as effect to cause. And the causal relation
will not be that of an agent to his complete human acts, or to those of their
component acts that are elicited, but that of any event that is an effect to any
event that is its cause.
The last act of will in a complete act

It remains to determine what is the last act of will in a complete human act:
the act of will which causes the operation of the executive powers that
constitutes the commanded act. There is an obvious reason for identifying
it with the act of choice, and so for identifying the intellectual act of
command with the act ofjudgement that terminates deliberation: namely,
that Aquinas himself lays it down that, 'inasmuch as [the power of will] is in
the person willing as having a certain proportion or order to that which is
willed', the last act of will with respect to the means is choice (lallae,
16, 4c).
Usus
Aquinas' argument for interpolating a further act of will between choice
and the commanded act begins with the premiss that, besides the relation
which the power of will can have to the willed inasmuch as it is in the
person willing as having a certain proportion or order to the willed, there is
a second relation which it can have inasmuch as the willed is in the person
willing as an end which he really possesses (lallae, 16, 4c). The means, as
well as the end, is then declared to be included in this second relation. And
finally, it is argued that, since choice is the last act of will in the first relation:
' Usus belongs to the second relation of the will to the thing willed by which
it tends to its bringing about. From which it is manifest that usus follows
choice, provided that the word "usus" is adopted inasmuch as the will uses
the executive power, by moving it' (lallae, 16,4c). This seems to be one of
those sheer blunders into which even the greatest philosophers fall. For if
the second relation of the will to the willed in fact obtains - namely, the
relation it has when the willed is something really possessed - then the
executive power by which the willed has been brought about must already
have been exercised. Hence if usus is an act which belongs to the second
relation of the will to the willed, it cannot be directed to the operation of an
executive power which ex hypothesi has already operated. 'Using' the
executive powers can only be an act belonging to theîwf relation of the
will to the willed. And Aquinas himself concedes that the last act of will in
that first relation is choice. Usus, understood as the 'using' of the executive
powers, should therefore be identified with choice. And imperium, as the act
of judgement which gives rise to usus, need not be identified with an
elicited intellectual act alleged to follow choice (cf. lallae, 17, 3 ad 1): it
should simply be identified with the judgement with which deliberation
terminates.
Shorn of excrescences that contradict his own principles, Aquinas'
analysis of the structure of a simple complete human act can be set out in
the schematic table below (see Figure 1).
Standard philosophical objections

Aquinas' analysis of human action is rightly held by his admirers to be a
major contribution to philosophy. To his immediate successors, the most
controversial point in Aquinas' theory of action was the priority it affirms
of intellect over will. Freedom of will, according to it, is wholly a matter of
the non-necessity of any judgement a man can arrive at by his natural
FIGURE I
Aquinas' Analysis of a Human Act,

Corrected according to the Present Interpretation
A SIMPLE COMPLETE HUMAN ACT

(e.g. Socrates' raising of his arm)
consists of
I. Acts of intellect [I. Acts of will

A. with regard to the end
Judging that an attainable end is 2. Willing to attain that end (e.g.
good (e.g. Socrates' judging that Socrates' willing to attract Plato's
it would be good to attract Plato's attention).
attention and that he can). 3. Enjoying that end.
4.. Intending that end through
suitable means (cf. 7.).
Β. with regard to the means
Deliberating how to attain that
end, that is:
Elicited (a) asking what within one's
acts power would most suitably cause
the attainment of the end willed,
and
(b) judging that a certain act 6. Consenting to all the means
( = the commanded act) in one's judged suitable.
power would be most suitable 7. Choosing the means judged most
(e.g. Socrates' judging that the suitable (e.g. Socrates' choosing
most suitable means would be his that his arm go up.) [This is iden
arm's going up.). [Aquinas should tical with intending, given that
have identified this judgement one's intention is fully deter
with imperium, or command.] minate.]
III. Act of the executive power caused

by the last elicited act of will
Executed 8. The commanded act takes place, the relevant executive powers having
act been set in operation by the act of choice (e.g. Socrates' arm's going up.).
powers as to the goodness of an end or the suitability of a means. Even

when will seems to fly in the face of intellect, there is always a (foolish,
perhaps vicious) judgement which directs it. Duns Scotus was to develop
the doctrine that the source of the will's freedom is internal to it, and not
external, as Aquinas holds.8 To many twentieth-century philosophers, the
feature of Aquinas' theory that is most objectionable is his retention of
8. See Bonansea 1965, esp. pp. 97-113.

Aristotle's conception of causation as the exercise of a power or capacity,

which allows him to think of human beings themselves, and not only of
events occurring within them, as genuine causes of their actions — 'agent
causes'.
Familiar uninformed objections

Little has been said of the vulgar objections, or rather slogans, by which the
philosophical work of Aquinas and other scholastics is now and then
ignorantly dismissed. The commonest, probably, is that Aquinas' theory
of action is a 'faculty-psychology' in which human beings are resolved into
a collection of faculties or powers, each of which is then treated as a quasi-
agent. For convenience, Aquinas often speaks of what a power such as the
intellect does or can do. But such statements, if they are indispensable to his
theory of action, can readily, if sometimes cumbrously, be reformulated as
statements of about what human beings, as possessing that power, do or
can do. He himself annihilates the objection that acts of will cannot be
commanded because 'the will cannot understand a command' (lallae, 17, 5
obj. 2) by pointing out that 'a man enjoins an act of will on himselfinasmuch
as he understands and wills' (lallae, 17, 5 ad 2).
As for the scarcely less common objection that the elicited acts which
according to Aquinas are components of even the simplest complete
human acts are too numerous to be credible, and correspond to nothing in
our experience of our own acts, the reply — made also by action theorists
today - must be that the components of simple human acts are ascertained,
not by introspecting what happens when we perform them, but by
examining various cases in which an act is begun but not completed. We
recognise cases in which we or others will to attain an end, and then do not
bother to consider means at all; cases in which we or others think about
suitable means but fail to settle on any as most suitable; cases in which we or
others settle upon some means as most suitable, and then do not choose to
adopt it; and finally (as when we or others suffer some unanticipated
impairment of our executive powers) cases in which we choose to bring
about some act which we think possible, and find that it is beyond our
power. It is by reflection on such cases, and not by introspection, that
Aquinas constructs his theory (e.g., lallae, 12,2 ad 4; 12, ad 3; 15,3 ad 3; 16,
4 ad 3; 17, 5 ad 2).
IX
ETHICS
34
THE R E C E P T I O N AND
I N T E R P R E T A T I O N OF
ARISTOTLE'S ETHICS
Thefirsttranslations
During the Middle Ages' the Nicomachean Ethics received less attention
than Aristotle's writings on natural philosophy and metaphysics, and still
less than his logical writings. Although the Ethics was never condemned in
any form, it was apparently not until the second half of the fourteenth
century that it was adopted as a regular textbook in the Arts faculties;2 and
it was only in the fifteenth century, as the number of commentaries shows,
that it began to be studied really intensively.
We are not so well informed about this period as we are about the
beginnings of philosophical ethics in the Middle Ages. According to the
latest historical research,3 the first translation of the Nicomachean Ethics
appeared in the twelfth century; but it covered only the second and third
books (ethica vetus). A second translation, of which only the first book
(ethica nova) and a few fragments remain, came at the start of the thirteenth
century.
The new conception of philosophical ethics

The texts thus made available to Latin readers make two claims for
philosophy: (a) happiness and human perfection are a legitimate object of
philosophical concern; (b) virtue, or good human character, can be ration-
ally discussed without recourse to theology.
These philosophical claims were not easy to accept. The problem does
not lie in the conception of a natural virtue. Such an idea was already
present in twelfth-century theology,4 and so the reception of the Aris-
totelian concept of virtue was not a revolutionary step. The difficulty
lay rather in the assertion that the goal of human life, no less than virtue,
was a topic for philosophy. Earlier thinkers had allowed that even pagan
1. The best and most informative survey is in Gauthier 1970, pp. 111-46.
2. Hcidingsfeldcr 1921, pp. 55fr.
3. Gauthier 1974, pp. XVI-CLI.
4. Wieland (forthcoming), Ch. 6, 1.
658 Ethics
philosophers might deal with the path to the goal, but not with the goal
itself, because that was something reserved exclusively for Christian theo-
logy. 5 The consequence of their view was plain: in so far as ethics is merely
the theory of virtue it is subordinate to theology because the end de-
termines the means to the end. Whether philosophical ethics could hold its
own in medieval Christendom against theology would depend on the
treatment of happiness.
Determining a role for philosophical ethics

In commentaries written before the whole Nicomachean Ethics was trans-
lated (ca. 1246-7) it is on the whole the theological view which prevails. 6
What is probably the oldest commentary on the ethica vetus (Avranches, MS
Bibl. m u n c 232), sees the epitome of happiness in God; so too do the Paris
Commentary on the ethica nova1 and vetus (MSS Bibl. nat. lat. 3804 A and
3572) and the commentaries on the ethica nova and vetus wrongly ascribed
to John Peckham, 8 which are the most important of these early texts.
This identification of happiness and God conflicts with the Aristotelian
text, which, in criticising Plato's idea of the Good, rejects the notion of
a self-subsistent entity constituting happiness. Moreover, happiness for
Aristotle is a human achievement rather than a divine gift. So the commen-
tators were at pains to make distinctions to square the traditional Christian
view with the new philosophical view.
The basic question is whether there is happiness in this life in addition to
the true bliss of everlasting life. According to the commentary on the ethica
nova and vetus perhaps written by Robert Kilwardby, the answer is yes. 9 If
so, philosophical ethics has its own field marked out, at least in principle,
and is thus guaranteed a certain independence from theology.
Theological misinterpretations
The first commentators on Aristotle's Ethics misunderstood important
passages of their text in a theological sense. This is strikingly shown by their
treatment of prudence. ' ° Latin writers at the time of the early com-
mentaries knew only the short passage at the end of the first book where
5. Peter Abelard 1970b, Dialogus, p. 137.

6. Wieland (forthcoming), Ch. 5, 3.
7. Gauthier 1975, pp. 71-141.
8. Mss Oxford, Bodl. misc. lar. 71; Florence, Bibl. naz. conv. soppr. G. 4.853. See Spettmann 1923,
pp. 221-2.
9. MS Cambridge, Peterhouse 206. See Wieland (forthcoming), Ch. 5, 3e.
10. Gauthier 1963, pp. 129-74.
Aristotle's Ethics: reception and interpretation 659
Aristotle introduces his fundamental distinction between moral and intel-

lectual virtues, mentioning wisdom, insight, and prudence only briefly and
without further explication. Thus the Pseudo-Peckham's commentary de-
fines intellectual virtue as knowledge and love of the highest good for its
own sake, and sees in prudence the highest of these virtues because pru-
dence originates in intelligent creatures in w h o m the purest image of the
creator appears. This position makes the intellectual virtues into Christian
virtues and interprets them as Christian contemplation, which differs from
merely theoretical speculation by involving the love of the object of
contemplation.
This misinterpretation of the intellectual virtues is easily explained by
the fact that these Latin writers knew only the first three books of the
Nicomachean Ethics. Their lack of the tenth book, in which Aristotle further
develops the concept of happiness in its theoretical aspects, was especially
important. Even after the translation of the whole öf the Nicomachean Ethics
it was difficult for the Latins to reproduce faithfully the Aristotelian
distinction between theoria and praxis. ' ' This is shown, for example, in the
work of Roger Bacon, who subordinates metaphysics to ethics and ethics
to theology because — like the commentators on the ethica nova and vetus —
he does not see theoria as an independent value in Aristotle's sense.
The contribution of Robert Grosseteste
The Latin translation of the Nicomachean Ethics which became the standard
one throughout the Middle Ages was made about 1246-7 by Robert Gros-
seteste,12 who simultaneously published a collection of Greek commen-
taries in Latin translation. 13 These commentaries were of varied
provenance: Eustratius' commentaries on Books I and VI; an anonymous
third-century commentary on Books II—V; one by Michael of Ephesus on
Books V, IX, and X; an anonymous twelfth-century one on Book VI; and
one by Aspasius on Book VIII.
Now the Latins possessed the resources for a better understanding of the
Nicomachean Ethics, but this did not lead to any substantial improvement of
the position of ethics in the context of the philosophical disciplines. In its
statute of 1245, l4 the Paris Faculty of Arts devotes the same amount of time
11. Wieland 1974, pp. 147-73.

12. Gauthier 1974, p. CCI.
13. Mercken 1973.
14. Chartularium Universitatis Parisiensis ( = CUP) I, n. 246.
66ο Ethics
to the Nicomachean Ethics as to the Liber de sex principiis or Aristotle's De

sensu et sensato or De somno et vigilia.
Albert the Great's commentaries

The first complete Latin commentary on the Nicomachean Ethics was
compiled by Albert the Great, w h o made use of a great deal of the material
prepared by Grosseteste. While the earlier commentaries originated in the
Faculty of Arts, Albert commented on the Aristotelian text as a theologian
during his time in Cologne during the years 1248-52. He adopted the same
method as his predecessors from the Faculty of Arts: first came the ex
position of the text (expositio litterae), which followed Aristotle's train of
thought step by step, then quaestiones, in which the difficulties arising from
the text were treated individually.
Some years later (ca. 1263 -67), ' s Albert wrote a second commentary in
the form of a paraphrase of the Nicomachean Ethics, as part of his massive
project to comment on all the works of Aristotle. Later medieval commen
tators quoted Albert's two commentaries more than any others, and he
thus became the greatest Latin authority in the field of philosophical
ethics. 1 6 O n e reason for this status is that he gives a clear and distinct
exposition of the philosophical point of view. O n the question of happiness
he stresses human agency: ' O u r actions constitute the cause of the happiness
of which the Philosopher is speaking h e r e . ' 1 7 O n the question of the virtues
Albert firmly rejects any reference to theological categories: 'Here we are
discussing not theological but natural virtues.' 1 8 He regards human beings
as the immediate and direct cause of happiness in this life, while believing,
as a Christian, that God is the first and all-embracing cause here too. He is
convinced of the fundamental harmony of the Aristotelian and the
Christian positions; and he reads and interprets the Nicomachean Ethics in
the light ofthat conviction. When his work is considered as a whole, he is
seen to have departed remarkably little from Aristotle.
There is, however, a significant difference between Aristotle and Albert
concerning the relation of theoria and praxis. The Aristotelian conception
of ethics as a practical science can easily be misunderstood if ethics is
regarded either as strictly theoretical or as immediately practical. The
15. Gauthier 1970, p. 123; n. 123; Dunbabin 1963, pp. 232-50.

16. Grabmann 1936a, pp. 324-412; Gauthier 1947-8, pp. 269fr.
17. Albert Super Ethica I, 10, 55: 'operationes nostrae sunt causa felicitatis, de qua hie loquitur
Philosophus'.
18. Albert In Eth. I, tr. 7, c. 5: 'iam non de theologicis, sed de physicis (scl. virtutibus) disputamus'.
commentators on the ethica nova and vetus, who as theologians had a low
opinion of mere speculation, tended to interpret ethics as an immediately
applicable guide to virtuous living;19 but this is to mistake the philosophi-
cal (scientific) character of the discipline. Albert gives a simple-sounding
answer to the question: what is the goal of ethics?20 In its theoretical aspect
- as ethica docens - ethics is for the sake of knowledge; in its practical aspect -
as ethica utens - its purpose is 'to make us good'. Albert can thus see ethics as
a unified whole; but he fails to distinguish between the rationality of moral
action and the philosophical rationality of ethics.
Thomas Aquinas on ethics

Thomas Aquinas is the first to see the point clearly. 'Moral science is indeed
for the sake of action, but this action is not an act of science but rather of
virtue.' 21 Accordingly philosophical ethics is bound to stand at a distance
from the concrete individual action; it is therefore not the immediate cause
of moral behaviour. For the rationality of moral action is guaranteed not
by science, but by prudence, the virtue which recognises, judges, and
prescribes what is actually to be done. Prudence mediates between the
general rules and concrete individual actions. This is possible because it is
both the perfection of reason and immediately linked to desire.22 Thus
according to Thomas prudence is no less a moral than an intellectual
virtue.23 Prudence does indeed depend on the moral virtues, but the
dependence is mutual, not one-sided. The moral virtues cannot exist
without prudence, and prudence cannot exist without the moral virtues.
As is well known, it is in the second part of the Summa theologiae that
Thomas develops his moral theory. The doctrine is unambiguously theo-
logical but philosophy is not discarded; on the contrary, in the field of
ethics philosophical thought is quite clearly determinative. Thomas had
already become acquainted with the Nicomachean Ethics during his studies
in Cologne (1248—52); he heard his teacher Albert lecture learnedly on the
text, and it was he who edited the text of Albert's lectures. No doubt this
fostered his obvious natural interest in questions of ethics.
19. Wieland (forthcoming) Ch. 4, 5b.
20. Albert Super Ethica, Prologus, 5: 'Dupliciter potest considerari scientia ista: secundum quod est
docens, et sic finis est scire; vel secundum quod est utens, et sic finis est, ut boni fiamus.'
21. Aquinas 1948c, In librum Boethii De trinitate 5,1 ad y. 'Scientia vero moralis, quamvis sit propter
operationem, tamen ilia operatio non est actus scientiae, sed magis virtutis.'
22. Kluxen 1964, pp. 30-40.
23. Aquinas ST, Ilallae, 47, 4: 'prudentia non solum habet rationem virtutis quam habent aliae
virtutes intellectuales, sed etiam habet rationem virtutis quam habent virtutes morales, quibus
etiam connumcratur'.
662 Ethics
In 1271-2 he himself wrote a commentary on the Nicomachean Ethics. He

based it on a revised but still imperfect text of Grosseteste's translation.
(These is no evidence for a revision of the translation by William of
Moerbeke, comissioned by Thomas himself) 2 4 Thomas never, or hardly
ever, makes explicit use of Albert's Cologne lectures, but many passages
show that the influence of his teacher is still strong. The number of mistakes
Thomas makes in quoting or paraphrasing Albert show that he was not
consulting Albert's lectures as he w r o t e . 2 5 Methodologically, Thomas'
commentary belongs with the earlier commentaries of the Faculty of Arts;
the style of textual explication (expositio litterae) is substantially the same,
but the usual quaestiones are missing. The commentary is based not on
actual lectures, but is the result of Thomas' private study of the Ethics; it
seems to have served as a preparation for the substantive moral theory of his
Summa theologiae.
Thomas' theological interest is shown in the fact that he draws the
boundaries of philosophical ethics more sharply than his predecessors and
his teacher. Thus, for example, he thinks that Aristotle had consciously
restricted himself to imperfect happiness, because perfect happiness is
incompatible with 'mortal human life'. 2 6 This restriction of the com
petence of philosophers does not imply a low estimation of philosophy or
the ethics. O n the contrary, it seems that Thomas was the first of the Latin
masters to develop a clear awareness of the distinctiveness of a practical
science, because of his correspondingly clear grasp of the Aristotelian
concept of theoria. From this point of view one can say that medieval
philosophical ethics came to be an independent discipline only when
metaphysics was firmly established as the basic theoretical science.
The commentaries of the radical Aristotelians of the thirteenth century

As we have seen, the first commentaries on the Nicomachean Ethics after a
complete translation became available were written by theologians an
xious to avoid any flat contradiction between Aristotelian philosophy and
Christian doctrine. But at Paris ethical questions continued to be treated in
the Arts Faculty also, and there sometimes without consideration of the
Christian tradition. The amount of literary evidence before 1277 is indeed
not very great — it includes the Quaestiones morales and fragments of the
24. Gauthier 1970, pp. 125-31.

25. Gauthier 1969, pp. 254*~57*.
26. Aquinas 1969, Sententia libri Ethicorum Χ 13: 'in hac vita non ponit perfectam lelicitatem, sed
talem qualis potest competere humanae et mortali vitae'.
Liber defelicitate of Siger of Brabant27 and the De summo bono of Boethius of

Dacia - but it is enough to show, on the question of happiness if on no
others, a strikingly self-confident philosophy which sets itself on a par with
theology or even above it.
The philosophical life is not just one way of life amongst others, but is
the absolute standard of human living. 'Whoever does not lead this life
does not lead the right life.'28 Thus, the happiness of political life, which
had been given an independent significance by Aristotle, is eliminated, or
made merely relative. This doctrine involves an exaggerated evaluation of
the theoretical life - something as far removed from Aristotle as was the
interpretation of happiness as Christian contemplation in the commen-
taries on the ethica nova.
The self-affirmation of the philosophers led to a strong theological
reaction. The question of the ordo vivendi is one of the topics of Bona-
venture's conflict with contemporary Aristotelians.29
On 7 March 1277 Stephen Tempier, the bishop of Paris, condemned 219
theses.30 These included propositions which corresponded to the man-
ifestoes of contemporary Aristotelians: 'No station in life is to be preferred
to the study of philosophy';31 'Philosophers alone are the wise ones of the
world.' 32 Some propositions are concerned with happiness, for example:
'God cannot infuse happiness directly';33 'Happiness is to be had in this life
and not in another'.34 Other propositions concern the nature of virtue:
'The only good which can be achieved by men consists in the intellectual
virtues";35 'No virtues are possible other than those which are acquired or
innate';36 continence and humility are not to be classified as virtues.37
Wherever these theses may have come from, the commentaries of the
thirteenth century so far known can hardly be the immediate source. For
both the commentators on the ethica nova and vetus and the theologians
Albert and Thomas adhere so closely to Christian doctrine that they cannot
27. These fragments survive in Agostino Nifo. See Nardi 1945. PP· 24-9 and 36-8.
28. Boethius of Dacia 1936, De summo bono, p. 377: 'haec est vita philosophi, quam quicumque non
habuerit non habet rectam vitam'.
29. Bonaventure, Coll. in Hexaemeron, VIII, 16.
30. CUP I, n. 473.
31. CUP 1, n. 473, Sent. 40: 'quod non est excellentior status quam vacare philosophiae'.
32. Sent. 154: 'quod sapientes mundi sunt philosophi tantum'.
33. Sent. 22: 'quod félicitas non potest a Deo immitti immediate".
34. Sent. 176: 'quod félicitas habetur in ista vita et non in alia'.
35. Sent. 144: 'quod omne bonum, quod homini possibile est, consistit in virtutibus intellectualibus'.
36. Sent. 177: 'quod non sunt possibiles aliae virtutes, nisi acquisitae vel innatae'.
37. Sent. 168 and 171.
664 Ethics
be the authors of these obviously anti-Christian propositions. The other

surviving commentaries38 must probably be dated after 1277; they are in
any case far from unambiguous on these topics.
In the Paris MS Bibl. nat. lat. 14698, there is an incomplete commentary
on the Nicomachean Ethics; it includes quaestiones on Books I—V. Its author
was perhaps James of Douai, a master in the Paris Faculty of Arts who
appears in a list of the Faculty in 1275. The commentary may have been
written shordy after 1277; that would explain the commentator's reticence
on the question of happiness. He conspicuously avoids positions which
conflict with Christian doctrine, as for example the thesis: 'God cannot
infuse happiness directly.' On this point as on others he follows the
doctrine developed by Albert the Great. 3 ' Nonetheless, the commentator,
following Aristotle but in conflict with Christian doctrine, sees poverty as a
hindrance for a virtuous life; moreover, he regards philosophical errors as
basically useful because they promote rational discussion. The author of the
commentary, like Boethius of Dacia, also believes that philosophers spon-
taneously five a life of virtue, because the pleasure of philosophical con-
templation far outweighs sensual pleasures;40 but here he misinterprets
Aristotle, who does not derive the moral virtues from the theoretical life.41
The commentary ascribed to Peter of Auvergne (Leipzig MS Univ. lat.
1386) comes from the same time and the same milieu; there is an un-
mistakable affinity between the two texts. Only the quaestiones on the first
two books have survived. As in his other commentaries, Peter interprets
the Aristotelian doctrine here in such a way as to avoid any contradiction
between philosophy and theology. In doing so he frequently follows the
teaching of Thomas Aquinas, even if he does not always understand it
correctly.42
The Erlangen commentary (MS Univ. 213) is rather different. It does not
hesitate to make clear the contradiction between philosophy and theology
and to take up the philosophical side of the argument.43 This is shown
above all in the treatment of happiness. Here questions are discussed which
have hardly anything to do with Aristotle, for example the problem
whether and how God gives happiness. For Aristotle this is not a topic for
ethics; but in this commentary the question is given serious philosophical
38. Grabmann 1931, pp. 30-60; Gauthier 1947-8, pp. 197-336; Hissette 1976, pp. 79-83.
39. Gauthier 1947-8, pp. 269fr
40. F. I30" _b ; Gauthier 1947-8, pp. 226ff.
41. X , 8;li78 b 5ff.
42. Gauthier 1964, pp. 233-60.
43. Grabmann 1931, p. 54.
discussion, despite its being of theological origin. In the opinion of the

commentator, God cannot infuse happiness directly, because as the ab-
solutely unmoved principle he has a single and eternal effect - namely, the
first Intelligence - from which other effects proceed in a hierarchy of
emanations producing multiplicity and variety in the world. Happiness, as
something newly brought about in individual cases, belongs to the world
of multiplicity and is derivable from God only by means of this series of
intermediate steps. This neoplatonic thought, stemming from the Liber de
causis and Avicenna, abolishes God's freedom, and the commentator is
aware that such a view contradicts the faith. Still, like most of these
commentators, he proceeds on the methodological principle that every
practitioner of a scientific discipline must stay within the boundaries set by
his principles.44 Accordingly, theology cannot assume the function of a
regulative, much less a constitutive, principle in philosophy. In spite of this
clear distinction between philosophy and theology, the author often al-
ludes to Thomas Aquinas, and indeed not just to his Ethics commentary,
but also to the Summa theologiae.
Related to the Erlangen commentary is that of Giles of Orleans (Paris MS
Bibl. nat. lat. 16089), another master in the Faculty of Arts at Paris. Both
commentaries are probably to be dated after 1286. Giles is an exponent of
the philosophical viewpoint and does not take Christian doctrine into
consideration. He comes to the same conclusions as the Erlangen commen-
tator. (The anonymous Ethics commentary from Erfurt (MS Amplon. F. 13)
is a member of this same group.)
The Vatican commentary (MSS Vat. lat. 832 and Vat. lat. 2172), possibly
by Radulphus Brito, follows substantially the same lines of thought as the
commentaries so far discussed.45 One argument of this commentator
deserves special mention because it rejects Aquinas' theological answer to
the question whether earthly happiness is possible. Thomas distinguishes
perfect happiness, which consists in the immediate contemplation of the
nature of God, from the imperfect happiness of this fife represented by the
Aristotelian conception. It is obvious that this distinction is made from a
theological point of view. Against this the commentator defends the
philosophical standpoint: the happiness of this life deserves to be called
perfect for it corresponds perfectly to what is possible for human nature
44. The paradigmatic formulation of this principle is by Boethius of Dacia, e.g., De aetemitate mundi,
p. 347: 'Nullus artifex potest aliquid causare, concedere vel negare nisi ex principiis sui scientiae.'
45. Gauthier I947~53. PP· 75~85·
666 Ethics
under the conditions of this life, even though theologically, earthly happi
ness can rightly be called imperfect. 4 6
In the Ethics commentaries of the late thirteenth century, the quaestio-
form becomes predominant. This shows that the original Aristotelian
problems, while not actually abandoned, were increasingly overshadowed
by topical issues of the time. The reception of Aristotle is now complete, and
internal scholastic discussion more and more takes its place.
The commentaries of the fourteenth and fifteenth centuries

We are far less well-informed about the development of philosophical
ethics and the corresponding commentaries in the fourteenth and fifteenth
centuries than in the preceding period. Nonetheless, ethics seems to have
aroused increasing interest. Thus in 133s the study of the Nicomachean
Ethics became a requirement for the Dominicans of Provence. 4 7 In 1366 the
Faculty of Arts in Paris made it an absolute requirement for the M.A.
degree that every candidate must have attended lectures on the Ethics.*6
Towards the end of the century (1392) a lector Ethicorum*9 is mentioned in
Paris. But the two most significant and influential thinkers of the time,
John Duns Scotus and William Ockham, did not compose commentaries
on the Nicomachean Ethics.
All the same, the number of theologians commenting on the Ethics
increased. These include 5 0 the Augustinian Henry of Friemar, whose
commentary comes from the first decade of the fourteenth century; the
Carmelite Guy Terrena who wrote about 1313; the Franciscans Gerard of
O d o (commentary before 1329) and Peter Coruheda; the Dominicans
Conrad of Ascoli and Guy of Rimini, w h o discuss Thomas' commen
tary rather than the Nicomachean Ethics itself. (Richard Kilvington's
Quaestiones is an independent elementary treatment of individual problems
of ethics rather than a real commentary.) Despite this theological interest in
philosophical ethics, there remain conflicts in the fourteenth century
though they are no longer reflected in the form of binding condemnations.
Thus the Augustinian Ugolino of Orvieto, like Bonaventure in the thir
teenth century, regarded ethics as superfluous and largely misleading. 5 1
46. MS Vat. lat. 832, f. 9 * (2172, f. 12"): 'sed dico, quod in hac vita aliquis homo potest esse felix
felicitate humana, perfecta etiam, secundum quod est possibile in hac vita'. Gauthier 1947-8,
p. 279.
47. Gauthier 1970, p. 135.
48. CUP III, η. 1319.
49. Aucluarium Chartularii Universitatis Parisiensis I, 667.
50. Gauthier 1970, pp. 134-6; cf. Gerard of Odo 1482.
51. Sent. Prologus, qu. 4, art. 1.
Walter Burley's commentary

More influential than the theological commentaries mentioned is the
commentary on the Nicomachean Ethics composed between 1333 and 1345
by Walter Burley 52 who according to his own statements had already
written a more extensive but now lost commentary on the first six books.
Burley very often quotes the comments (notulae) made by Robert Grosse
teste on his Latin translation. He also often cites Eustratius, Averroes, and
Albert the Great; despite clear dependence on Aquinas he mentions his
name remarkably rarely.
Albert of Saxony's commentary and Ockham's influence

The Ethics commentary of Albert of Saxony 53 is evidence of Burley's
influence, sometimes following him word for word. It is remarkable that
Albert, a follower of Ockham, should in ethics follow the lead of a declared
opponent of Ockham. This suggests the question whether any influence of
Ockham's thought is discernible in the Ethics commentaries. Such an
influence would be expected in the case of John Buridan, whose com
mentary was more popular than any other among the adherents of nomi
nalism.54 Ockham believed that human action is good if it follows God's
commands: for which there is no other ground than the will of God, itself
limited only by the principle of non-contradiction. If this is taken as the
foundation of a nominalist ethics, the result would be completely incom
patible with the ethics of Aristotle. On the other hand, Ockham regarded
what we would call meta-ethics as one of the most secure sciences because it
contains many self-evident principles, as, for example, that the will must
agree with right reason.55
John Buridan's commentary

There is no trace in Buridan's commentary of a voluntaristic conception
which stresses the absolute freedom and omnipotence of God and restricts
morality to obligations in the face of merely arbitrary commands. In his
opinion Ockham's view of God's potentia absoluta is no part of philo
sophy.56 One might perhaps think that Buridan's ethics operates on the
plane of the order prescribed by God, the order of the potentia ordinata; but
52. Walter Burley 1481,1500, 1521.

53. Heidingsfclder 1921
54. John Buridan ι s 13 ; see also Walsh 1966, p. 4.
55. William Ockham 1491, II, 14.
56. John Buridan 1513, X, qu. 5, f. 213'''.
668 Ethics
in this commentary there is no room for a concept of the type of obligation

which would follow from such an ordering. And this conception of earthly
good as being similar to and part of the supreme good 5 7 cannot be
interpreted as the result of a voluntary decision of God. Altogether
Buridan's commentary on the Nicomachean Ethics, at least so far as concerns
praxis, seems to be anything but nominalistic, even though on other issues
it takes a thoroughly Ockhamist stand - e.g., the rejection of realist
doctrines of universals and of the Scotist distinctio formalis.
The question of the nominalist character of this commentary needs
further examination, but one thing is clear: for Buridan philosophical
ethics is a practical science. It is practical in a sense stronger than that upheld
by Aristotle and Thomas Aquinas, who regarded ethics as a science which
does not immediately determine human action. For them, that is rather the
task of prudence, which is clearly distinguished from science. In contrast to
this Buridan clearly emphasises the immediate practical character of ethics
as productive of goodness; he goes so far as to identify prudence with
knowledge of ethics. 58 Because of the immediately practical task of ethics
Buridan believed that a special moral logic was required, a logic operative
in poetry and rhetoric and appealing not to abstract reason but to hu-
man beings determined by their passions. 59 This may explain, too, why
Buridan quotes Seneca so strikingly often; perhaps his Epistolae adLucilium
gave support to this interpretation of the task of philosophical ethics.
Scholastic commentaries after Buridan

The influence of Buridan extended well beyond Paris. Prague, especially,
where the Quaestiones were presented in shortened form, 60 became a new
centre of Buridanism. There ethics apparently took its place amongst the
regular lectures right from the beginning; it was accorded as much time in
the curriculum as physics. 61
One commentary which may derive from Buridan's is that of Henry
Totting of Oyta. 6 2 Apart from this there is a series of mostly anonymous
57. John Buridan 1513, I, qu. 5, f. 6": ' . . . quodlibet aliquid sibi adipiscitur boni proprii; per quod
bonum, inquantum est quaedam participata similitudo primi boni, ipsum refertur ad primum et
finale bonum'.
58. John Buridan 1513, VI, qu. 17, f. 131"*: 'videtur mihi, quod habitus acquisitus ex doctrina
librorum legum, decretorum et universaliter librorum moralium pertinet ad prudentiam'.
Korolec 1975, pp. 56fr.
59. John Buridan 1513, Prooemium, f. 2tb.
60. Korolec 1974a, p. 198.
61. Heidingsfelder 1921, p. 56.
62. Lang 1937, p. 133.
editions and abridgements of the Buridan commentary dating from the

fourteenth and fifteenth centuries.63 Textbooks in the form of abridge-
ments are characteristic of the treatment of ethics at the universities of
central Europe, including Cracow, where Buridan's influence extended
beyond the middle of the fifteenth century.64 Probably the most signi-
ficant Cracow commentary is that of Paul of Worczyn. The Cracow com-
mentaries are noticeable for restricting themselves to the first four or five
books of the Nicomachean Ethics thus omitting consideration of the intel-
lectual virtues and the theoretical life.
One indication of the ever-increasing interest in ethics is the translation
of the Nicomachean Ethics into French done by Nicole Oresme in 1370 at the
command of King Charles V. Oresme used the Latin text of Robert Grosse-
teste and added short comments mainly derivative from the commentary
of Thomas Aquinas.65
In the fifteenth century it seems to be mainly the universities of central
Europe where ethics is treated in a comprehensive way. Many com-
mentaries come from masters of the University of Vienna,66 such as
Thomas Eberdorfer of Haselbach, Urban of Melk, Thomas of Wulders-
dorf, Andreas of Schärding, who also expounds Buridan's commentary,
and Andreas Wall of Walzheim, who refers to the Viennese custom of
using only the first five books of the Nicomachean Ethics as the basis of a
disputation, which explains why many of these commentaries cover only
five books, as in Cracow.
The commentary ofJohn Versor of Paris was reprinted several times; it
deals only with the first six books.67 Another commentary on the Ethics
was written by Paul of Venice; he does not claim originality, but follows
the received commentators.
Alongside these works composed by philosophers, the fifteenth century
also offered theological commentaries on the Nicomachean Ethics. It is
especially interesting that we now find commentaries which clearly appeal
to Duns Scotus as their authority. Nicholas of Orbelles wrote a three-part
commentary on the whole philosophy of Aristotle, the third part of which,
concerning practical philosophy, is essentially a commentary on the
Nicomachean Ethics.68 The Scotist Peter Tartaretus wrote quaestiones on the
63. Korolec 1974a, pp· 198-202.
64. Korolec 1974a, pp. 202-8.
65. Gauthier 1970, p. 138.
66. Gauthier 1970, pp. 139fr
67. First edition Cologne 1491; Cologne 1495 (reprint, Frankfurt 1966).
68. Basel 1494, 2 vols; cf. Gauthier 1970, p. 142, η. 169.
670 Ethics
first six books of the Ethics which were reprinted several times. 69 The
Franciscan Peter of Castrovol is one of the first to base his commentary on
Aretino's new translation of the Nicomachean Ethics; in his comments he
remains true to the scholastic method.
In the course of the fourteenth and fifteenth centuries philosophical and
theological schools gained in strength and significance, and disputes broke
out between the via antiqua and the via moderna, but ethics remained
comparatively free from controversy. The reason for this is still unclear.
Humanist translations, editions, and interpretations

The fourteenth century in Italy saw the development of studia humanitatis,
governed not by the theoretical interests of scholasticism, but by primarily
educational considerations. Grammar, rhetoric, poetics, history, and ethics
are the topics of humanistic study. 70 Thus ethics is given a dominant place
amongst the disciplines and treated as primarily practical, not theoretical.
Despite the change, Aristotle remains among the preferred authors -
alongside Cicero and Seneca. He is no longer read as a witness to the truth,
however, but as an ancient author who writes in a particular historical
context.
This explains the fact that the main contribution of the humanists to the
study of Aristotle consists of new translations. In 1416—17 Leonardo Bruni,
called Aretino, completed his translation of the Nicomachean Ethics; it is
hardly more than a revision of the Grosseteste translation. 71 Aretino's
project gave rise to a controversy in which defenders of the old translation
espoused the ideal of philosophical truth against elegance of style; 72 but the
controversy did not prevent the translation from being a great success.
Another widely-read translation was that of John Argyropoulos (1457).
Towards the end of the fifteenth century the first editions of the Greek text
appeared.
O n the basis of the new translations there soon emerged new interpre-
tations. While Aretino's Isagogicon moralis philosophiae gave only a short
introduction to the Ethics, Niccolo of Foligno wrote a commentary on the
basis of Aretino's translation. Donato Acciaiuoli used the Argyropoulos
version for his Expositio super librum Ethicorum Aristotelis. The commen-
taries of Ermelao Barbaro and of Agostino Nifo are well known. The last
69. First edition 1497; further editions Venice 1503, 1513, 1571, 1621.
70. Cf. Kristeller 1961b, pp. 289-335.
71. Gauthier 1970, pp. I47ff.
72. Grabmann 1926, pp. 440-8.
of this line of humanist introductions and commentaries on the Nico-

machean Ethics is that of Faber Stapulensis.
Philip Melanchthon
Martin Luther's attack on scholasticism and its philosopher Aristotle was
much sharper than that of the humanists. It is due to the work of Philip
Melanchton that the study of Aristotle and above all of the Ethics was
revived for two further centuries in the schools and universities of
Germany. He was guided less by his humanist interests than by the need to
delimit the new faith in the face of fanaticism, and the realisation that the
gospel contains no teaching about political life.
Melanchthon wrote on the Nicomachean Ethics on several occasions. In
1529 he published a commentary on the first and second books, in 1532 he
extended it to Books III and V. In 1538 he composed the Philosophiae
moralis epitome and in 1550 the Ethicae doctrinae elementa, written completely
in the spirit of Aristotelian ethics. Unlike Aristotle, however, he made a
sharp separation between the political - social and the individual aspects of
human action.
According to Melanchthon, ethics is 'that part of divine law which
prescribes external actions'.73 Thus it is confined to the habits and customs
of civil life. Consequently, the Aristotelian distinction between ethics and
politics becomes meaningless: ethics is in reality politics.74 Accordingly it
primarily concerns the lawyers and the theologians who deal with political
institutions. Ethics does not deal with the inner relation of man to God,
which eludes reason and philosophy. Consistently with this interpretation
of ethics, Melanchthon commented only on Books I—III and V of the
Nicomachean Ethics, that is, on the texts which treat of civil happiness, the
basic principles of the ethical virtues, and justice. It was no accident that
politics was studied in the Protestant universities to a greater extent than
elsewhere.73
Even independently of Melanchthon's efforts the study of the Ethics was
very much on the increase in the middle of the sixteenth century. New
Latin and vernacular translations appeared and the number of commen-
73. Philosophiae moralis epitome 1850, p. 21: 'Philosophia moralis est pars ilia legis divinae, quae de
externis actionibus praecipit.'
74. In primum librum Ethicorum Aristotelis enarratio, cap. I, 1850, p. 285: 'Aristoteles hoc loco admonet
hanc ipsam ethicen vere esse earn politicam seu practicam, quae principaliter privatos mores et
publica officia regit.'
75. Maier 1966, pp. 59-116.
672 Ethics
taries further increased. The influence of Aristotelian ethics in the schools

and universities of the sixteenth and seventeenth centuries may well have
been greater than it had ever been in the Middle Ages; but philosophical
ethics itself underwent its new development outside the universities, in an
atmosphere of indifference or even open hostility to Aristotle.
35
HAPPINESS: THE P E R F E C T I O N
OF MAN
The Aristotelian concept of happiness and the Christian tradition

The medieval discussion of happiness both before and after the reception of
Aristotle is governed by two basic thoughts: there is no happiness in this
world because 'all men, so long as they are mortal, are also necessarily
wretched';1 true happiness is to be found only in the enjoyment of the
contemplation of God (frui Deo) in the world to come. Thus the concept of
happiness involves an element that transcends human capacities. But the
Aristotelian and Christian conceptions also understand happiness as the
perfection of human nature, the actualisation of the possibilities inherent in
man. This rules out all definitions of happiness in terms of something like
worldly wealth, which is external to human nature, or like the satisfaction
of sensual desire, which is not specific to human nature. For Aristotle as for
others the essence of happiness is to be found in perfection. It is in that spirit,
for instance, that Anselm regards the contemplation of God as the perfec-
tion of human rationality, because without this form of happiness man
would be rational to no purpose.2
The difference between the Christian and the philosophical concep-
tion is particularly clear in the Dialogus inter philosophum, Iudaeum, et
Christianum of Peter Abelard. While the philosophers speak only of ethics,
aiming primarily for the way to the highest good, and accordingly deal
mainly with the virtues, Christians take the goal itself as their starting
point. The best the philosophers can offer as the highest human perfection
is a state in which sin and suffering are defacto absent, not one in which they
are necessarily eliminated. But what is wanted is 'a life quite free of sin',
where 'not only is there no sin, but there can be no sin'.3 Such a life cannot
1. Augustine, De tivitate Dei, IX, 15: 'omnes homines, quamdiu morrales sunt, etiam miseri sint
necesse est'.
2. Anselm, Cur Deus homo, II, 1.
3. Abelard 1970b, p. 103: 'ut earn . . . meliorem esse vitam intelligas, quam et ab istis maus omnino
constat esse immunem et in tantum a peccato prorsus remotam, ut non solum ibi non peccetur,
sed nec pcccari possit'.
674 Ethics
be grasped by merely philosophical means. The adequate description of

happiness as the contemplation of God can be given only by a Christian
who realises that there can be no happiness in this life.
Given such a background, there were bound to be difficulties in as-
similating the Aristotelian concept of happiness. Moreover, the Boethian
concept of happiness, which was generally accepted, is not compatible
with Aristotle's; for Boethius, too, regards the perfecting of man in this life
as impossible. The multiplicity and fragility of earthly things admits of no
perfect condition; there is happiness only in another world, in which the
multiplicity is made one and the fragility is exchanged for permanence. 4
Compared with this Christian Platonist interpretation of happiness, the
Aristotelian conception makes a far more modest claim. It is not measured
in terms of an absolute ideal of perfection, but restricts itself to the humanly
possible. Happiness is accordingly a human good: the highest of human
goals to be sure, and one which is sought for its own sake, but one that man
can attain by his own actions.
Medieval thinkers had no difficulty in accepting the Aristotelian analysis
of human action and its results; like Aristotle they assume that the ultimate
goal of action is happiness. The problem is whether human action suffices
to attain this ultimate goal; whether the highest human perfection is to be
understood as a gift of God or as an achievement of man. Aristotle's
position is clear: happiness is an achievement of man in which an absolute
supreme Good plays no part, because such a Good 'is not a good humanly
realisable or attainable; but that is the kind of good we are looking for'. 5
According to the Christian view, God alone is the source of true perfection;
human achievement cannot be compared on the same scale.
A further difficulty is presented by the distinction Aristotle makes
between kinds of happiness. On the one hand there is social or political
happiness, which consists in the exercise of the moral virtues and character-
ises the good life of the free citizen. Contrasted with this there is the
theoretical life, which consists in pure contemplation of ultimate grounds
and causes. This way of life is superior to social happiness because it is more
stable, pleasant, and self-sufficient; it presupposes leisure and stands out as
the higher form of happiness. Aristotle does not reduce the two forms of
happiness to each other: a philosopher leading a theoretical life has no
4. Boethius, Consolatio, III, pr. 9 and 10.

5. I, 6; I096 b 34·
Happiness: the perfection of man 675
obligation to lead a political one, even though he 'is a man and lives in a
community'. 6
How would Aristotle's distinction be justified in a medieval context?
Aristotle himself recognises, simply as a practical matter, that there is a
plurality of forms of life; he does not need to justify this plurality.
However, as soon as one regards God as the one final goal, it is not easy to
see why there should be various forms of human perfection.
The Aristotelian conception of happiness and the relation of man to God

The first commentators on the Nicomachean Ethics1 found Aristotle's con-
ception of happiness particularly difficult: not only were they influenced by
the theological doctrine that God is the quintessence of happiness, but until
about 1250 they had access to only the first three books of the Ethics, and so
remained unaware of the doctrine of the theoretical life, which Aristotle
develops in the tenth book. Hence a number of important questions arose:
(a) What is the relation between the absolute perfection of the one God in
whom true happiness consists and the imperfections of the many individual
men whose happiness it is? (b) How is man to be united to God and thus
made happy? (c) What sort of happiness is Aristotle talking about in the
Nicomachean Ethics? (I will consider each of these questions in turn.)
God and human happiness

Theologians had long distinguished between uncreated happiness, which is
identical with God, and created happiness, which is the individual perfec-
tion of man. 8 The first philosophers who attempted to deal with the Ethics
made use of this distinction. A collection ofquaestiones stemming from the
Paris Faculty of Arts, for example, asks whether there is one single happi-
ness in which all (or many) participate, and gives the answer: 'Happiness is
a single common good, which is caused in order that many may share in it
as one.' 9 How is this proposition to be understood? God cannot be this 'one
single common good', because it is created (caused); equally it cannot be
6. X, 8;li78 b 5-8.
7. Wieland (forthcoming), chap. 5, 3.
8. William of Auxerre 1500, Summa aurea, III, tr. 20, f. 195": 'Tamen dicimus quod, cum dicitur:
Deus diligitur quia bonus, haec dictio "bonus" praeter di vinam esscmiam, quam signât, connotat
aliquid creatum, scilicet suavitatem vel dilectionem vel beatitudinem creatam, quae est finis quo
quiescitur in Deo; Deus autem est finis in quo quiescitur.'
9. MS Ripoll 109 (Archivo de la Corona de Aragon), f. Ι36 Α : 'félicitas est enim unum commune
bonum causatum, ut a pluribus participetur una'.
676 Ethics
the individual actuality ofhappiness, because there are as many actualities as

there are persons who are happy as happy subjects.
The anonymous commentary of Naples gives an explanation: happiness
is numerically one but at the same time general (communis), 'because
everything tends towards it'. 10 This thought has nothing to do with
Aristotelian ethics, which regards happiness not as a numerical but as a
specific unity: men are happy through virtuous activity and not through
participation in an hypostasised form of happiness. The commentator
apparendy sees the problem in a neoplatonic context: something is re-
quired to mediate between the absolute One and the multiplicity of men,
and this is created happiness.
The Pseudo-Peckham's commentary tries to solve the problem without
the neoplatonic metaphysics. Two aspects of created happiness can be
distinguished: in relation to individual subjects happiness is manifold, in
relation to the cause there isjust one happiness.11 Given such an answer, it is
not clear whether the category of created happiness is necessary at all.
Human happiness as the union of man with God

While the concept of created happiness as a hypostasis mediating between
God and man is completely foreign to Aristotle, there is nothing primafacie
unaristotelian about the general conviction of the early commentators that
contemplation is the activity in which individual happiness consists: 'the
means by which happiness ( = God) is united with us'. ' 2 Because Aristotle
himself regards theoria as the highest form of human perfection, the
medieval authors seem to be in general agreement with him here; but the
appearance is deceptive.
The concept of contemplation cannot be properly understood in terms
of knowledge alone; for 'mere knowledge' (simplex notitia) has no essential
emotional component, while contemplation is incomplete without loving
(notitia amantis).13 Thus it is to practical reason that the Paris commentary
ascribes the activity in which contemplation essentially consists: affective
10. MS Naples, Bibl. naz. VIII G 8, f. 9"*: 'necque dico, quod sit communis [sel. félicitas] sicut
universale, sed est unum et idem numéro sicut commune, quia omnia tendunt ad ipsum'.
11. MS Oxford, Bodl. mise, lat. 71, f. i8v*: 'similiter et de bcatitudine causata; ipsa enim considerata in
sua causa una est; ipsa vero considerata in recipientibus multae est*.
12. Cf. the anonymous Commentary on the ethica nova, Gauthier (ed.) 197s, p. 115: 'virtus secundum
quam attenditur vita contemplativa est medium quo nobis unitur félicitas'.
13. Roben Kilwardby 1935, De natura theologiae, p. 31: 'est scientia notitiae simplicis tantum, quae
radicatur in consensu aspectus nudi ad ratiocinationem aliquam vel visionem; et est scientia
notitiae amantis, quae radicatur in consensu affectus per amorem rectum".
knowledge.14 Knowledge finds its perfection and specific form in love.

Aristotle indeed recognises pleasure as part of the happiness associated with
contemplation; but this is an accompaniment of theoretical activity, not
joy in the object of contemplation, not an element of the essential nature of
happiness.
Early medieval views of the Aristotelian notion of happiness

What of the social happiness which Aristotle describes in the first book?
How can this concept be reconciled with traditional Christian categories?
There is nothing to suggest that Aristotle is speaking of anything other
than created happiness;13 from a Christian viewpoint that much goes
without saying. What is far more suprising is the suggestion in the quaes-
tiones from the Paris Faculty of Arts and in the Pseudo-Peckham's com-
mentary that Aristotle is dealing with happiness after death. According to
these texts, the boundary between the philosophical and theological kinds
of happiness is marked not by death, but by the kind of entity that is to be
considered happy. According to philosophy, it is the soul alone that can be
happy after death; according to theology it is the soul in union with the
body.16
This shows two things: first, the Nicomachean Ethics was not initially
interpreted as hostile to the Christian tradition; second, even philosophical
commentators regard theological categories as the decisive guides to inter-
pretation. On the point of happiness after death it did not take long to reach
clarity. An early commentary, probably by Robert Kilwardby, explains
that in the Ethics Aristotle is talking only about the happiness which
belongs to the realm of politics; the question of happiness after death and of
the true happiness of contemplation lies beyond that realm.17 The author
of this commentary was not in a position to know that for Aristotle too it is
theoretical life which constitutes real happiness, for he composed his
commentary just before Robert Grosseteste's translation of the whole
Nicomachean Ethics.
14. Gauthiet (ed.) 1975, p. 102.

15. MS Ripoll 109, f. 136".
16. Ibid.: 'Anima enim maxime vivit in se post mortem, cum sit a corpore separata, et ideo innuit hie
[= Nie. Eth. IIOO'IO] Aristoteles felicitatcm esse post mortem.' MS Oxford, Bodl. misc. lat. 71,
f. 21*.
17. MS Cambridge, Peterhouse 206, f. 293": 'unde forte intendit non nisi de ilia felicitate quae dicitur
vita secundum ius doctrinae civilis, nec debuit forte doctrina civilis de alia felicitate perserutari.
Utrum enim post mortem fclicitatur animal vel totus homo, forte non pertinet ad ipsam, nec hoc
déterminât Aristoteles.'
678 Ethics
The relation between philosophical and theological

concepts of happiness in Thomas Aquinas
The reception and adaptation of the Aristotelian concept of happiness was
completed by Thomas Aquinas. 18 Thomas does not merely reproduce the
Aristotelian concept; he is a theologian, guided by theological interests,
and he fully accepts the Christian tradition. For him it is clear 'that the
happiness of the life to come lies beyond any inquiry of the reason' and that
'perfect human happiness is reserved for the life to come'. 1 9 Since happiness
means the complete satisfaction of human aspirations, and since human
aspirations are infinite because of human spirituality, only an infinite
object, namely God, can perfectly satisfy man. 2 0 Human beings cannot
achieve this by themselves, because insofar as they are potential, they
cannot actualise their own potentialities.
If God is taken to be the object of the desire for happiness in this way,
something important follows: since God does not belong to the sphere of
human action, to the extent to which happiness is identified with him it
cannot become an object of practical knowledge. Thus, if a theologian
takes God, the ultimate goal, as his starting point, his treatment of happi-
ness must be theoretical. But the essence of human happiness does not
consist in the object which makes men happy but in the activity or
relationship of human beings vis-à-vis this object. To describe this distinc-
tion between the object and the activity of happiness, Thomas uses the
traditional concepts of uncreated and created happiness. 21 He appeals to
Aristotle, who defines happiness as strictly an activity (operatio), since
activity is the final and complete realisation of every active being.
What is the precise activity in which happiness consists? Here Thomas'
answer goes against the tradition. His analysis of the essence of happiness
shows that ultimately it can consist only in a cognitive mental activity. It is
knowledge which constitutes happiness and the possession of God, know-
ledge accompanied in the will by the pleasure which arises from this
possession.22 But the cognitive activity which constitutes happiness is not
an exercise of practical reason concerned with human emotions and ac-
tions; this would be the case only if man were his own ultimate goal. 23
18. Kluxen 1978, pp. 77-91.

19. Aquinas 1964b, I, 9, p. 32: 'félicitas alterius vitae omnem investigationem rationis excedit'; I, 16,
p. 60: 'reservatur homini perfecta beatitudo post hanc vitam'.
20. ST, lallae, 2, 8.
21. ST, lallae, 3, 2.
22. ST", lallae, 3, 4.
23. ST, lallae, 3, 5 ad 3.
Since the ultimate goal of infinite human aspiration is God, happiness must
consist in an activity of theoretical reason.
Thus far Thomas agrees with Aristotle; he differs from him only in that
he takes the concept of perfection in the very strict sense familiar from
theology. According to this, human happiness is nothing less than the
vision of God's essence - an activity of theoretical reason - without
interruption, without end, and unaccompanied by other activities. In other
words, perfect human happiness is a single, continuous, eternal activity.
Obviously such an activity is not possible in this life.24
Given such a strict concept of happiness, one can look on the present life
in various ways. It can be regarded as a condition of misery and unhappi-
ness, impossibly far removed from true happiness; or one can emphasise
those elements of the present life which bear a certain relationship to
perfect happiness. Thomas chooses the second alternative. As a theologian
he judges the happiness of this life to be imperfect; but it is happiness in no
merely equivocal sense. The theological distinction between perfect and
imperfect happiness goes back to William of Auxerre,25 but it was Thomas
who was the first to make it fruitful by treating the concept of happiness in
the Nicomachean Ethics as a paradigm of imperfect happiness.
In accordance with the strict sense of perfection, happiness is realised in a
single eternal activity; what is possible in this life, by contrast, is a double
happiness (duplex félicitas), the happiness of the theoretical life and the
happiness of social life. In the state of perfect happiness there is no room for
such a distinction because the union of man with God, brought about by
contemplation, automatically brings about the perfection of the whole
man. It is different with imperfect happiness; the social and the theoretical
forms of life do not coincide. The perfection of theoretical reason as such
does not mean eo ipso the perfection of the whole man; the ordering of
human action and emotions is the task of practical reason. In general in the
area of imperfect happiness the perfection of the lower faculties is a
necessary condition for the perfection of the higher faculties.26 If the
philosopher is a good man, it is not simply because he is a philosopher. This
is a genuinely Aristotelian idea, but, as we shall see, it is interpreted
differently by some of his most enthusiastic followers.
Unquestionably, Thomas himself understands certain doctrines of the
24. ST, lallae, 3, 2 ad4.
25. William of Auxerre 1500, Summa aurea III, tr. 20, f. 222*: 'distinguenda est beatitudo perfecta et
beatitudo imperfecta; beatitudinem perfectam habebunt sancti in futuro, in praesenti vero habent
beatitudinem imperfectam',
26. ST, lallae, 3, 3 ad 3-
68ο Ethics
Nicomachean Ethics in a manner unintended by Aristotle. This is especially

true of the doctrine of the two types of happiness. Thomas agrees with
Aristotle that the active and contemplative lives cannot be made to coin
cide and that theorising is to be regarded more highly than the activity of
the moral life. But, unlike Aristotle, Thomas regards the life of political
action as directed towards theoretical happiness. For him the duality of
happiness is thus only relative.27 The main reason for this difference is that
because of the unity of human nature Thomas can allow only a single goal
for human life.28
There can be no doubt that the Thomistic conception of happiness is
basically intellectualist. This intellectualism is the result of his analysis of the
nature of happiness. From the point of view of praxis, in the conditions of
the present life, Thomas values the will more highly than reason, because
the will can reach God directly while reason remains subjected to the
senses.29 Happiness in its perfect, intellectual form is realised only in the
other life.
The theoretical life as the paradigm of human happiness:

Boethius of Dacia and Siger of Brabant
Some philosophers who were acquainted with the whole Nicomachean
Ethics and therefore familiar with the Aristotelian concept of theoria were
unwilling to adapt these philosophical doctrines to a theological context.
One such philosopher was Boethius of Dacia, who in his work De summo
bono dealt with happiness, outside the context of theology. For him, the
highest good of man consists in the perfect actualisation of the highest
human capacity, reason. Reason involves both a theoretical and a practical
faculty, the theoretical capacity orientated to knowing truth and the
practical one to doing good. The highest good of man, human happiness,
consists in the exercise of both capacities and in the pleasure derived from
each. All action should be directed to this goal. A man who acts for the sake
of this goal acts correcdy and naturally because this is the goal to which he is
orientated by nature. Moral failure occurs when a man sets up as his goal
something other than his highest capacity - for instance, if all his action is
directed to sensuality. For Boethius this amounts to a disorder which
clashes with nature.
27. Aquinas 1964b, X, II, p. 587.

28. Aquinas 1964b, 1, 9, p. 31: 'necesse est enim unum esse ultimum finem hominis inquantum est
homo propter unitatem humanae naturae'.
29. ST, I, 82, 3; lallae, 27, 2 ad 2.
The distinction between theoretical and practical reason is to a certain

extent reminiscent of the Aristotelian distinction according to which social
happiness consists in the complete exercise of the moral virtues and con-
templative happiness in the exercise of wisdom. Aristode is talking of two
goals which are independent of each other; he emphasises that theoria is
neither the cause of nor a precondition for a good social life. Thomas
Aquinas too takes this view, although he regards active happiness as
belonging to practical reason. Does Boethius of Dacia also maintain the
doctrine of the double happiness? Does he think a good life is possible
without theoretical reason operating as its standard? Boethius is aware that
almost all men follow their disordered desires rather than their natural
appetite for knowledge. This shows that almost everyone leads an un-
natural and blameworthy life, but it does not answer our question. Matters
become clearer when he says that philosophers alone follow the innate
appetite for knowledge and lead a life which corresponds to the natural
ordering; they alone achieve the best condition which is possible for
man.30
In fact according to Boethius, philosophy is not just one form of life
amongst others; it is not even simply the highest life, it is the one test of
human life in general: 'whoever does not lead this (life), does not lead the
right life'.31 Boethius does not recognise social happiness as an independent
form of human perfection; only the philosopher can really live the life of
praxis, because his theoretical activity gives him an intuition of virtue and
vice, which enables him to reach correct decisions. Again, it is the philo-
sopher who prefers the greater intellectual pleasures to sensual enjoyment;
in his purely intellectual activity there is no room for wrongdoing.
Boethius is quite consistent in drawing the conclusion: 'it is easier for a
philosopher to be virtuous than for anyone else'.32 Thus Boethius abolishes
the difference between 'man' and 'philosopher', which was so important in
Aristotelian ethics: the philosopher is the epitome of man; all others simply
'do not lead the right life'.
Is there any limit to philosophical happiness? For Boethius, philosophical
knowledge reaches its term with knowledge of the principle of all things,
namely, 'uncreated Being'. 33 This end is not its limit but its perfection.
Boethius, however, retains the Aristotelian assumption that philosophical
30. Boethius of Dacia 1936, De summo bono, pp. 373a".

31. Ibid., p. 377: 'Haec est vita philosophi, quam quicumque non habuerit non habet rectam vitam'.
32. Ibid., p. 375: 'Ideo philosophus est facilius virtuosus quam alius'.
33. Ibid., p. 375: 'numquam enim satiatur appetitus sciendi. donee sciatur ens increatum'.
682 Ethics
knowledge of the divine starts with created beings and remains bound up
with them. The philosopher can demonstrate the necessity of a first cause,
and he can also recognise the dependence of the world and of man on this
cause. But Boethius does not speak of a knowledge of the essence of
'uncreated being', and this may be looked on as a limitation of philo
sophical happiness. By contrast, the perfect happiness of theology, as
explained by Thomas, consists in the knowledge of the essence of God.
According to the testimony of Agostino Nifo (1472-1538), Siger of
Brabant (a contemporary of Boethius of Dacia and of Thomas Aquinas) in
his lost Liber de felicitate rejected this difference between philosophical and
theological conceptions of happiness. 3 4 According to him, human happi
ness consists in the immediate contemplation of the nature of God: 'we
shall finally be made blessed [happy] by the essence of God itself and not by
any intermediary'. 3 3 This is of course not a usual state or one often
encountered even in the philosophical life; it is the maximum which is
finally possible for the phdosopher, for one who 'is very skilled in philo
sophy'. 3 6 It is obvious that such a notion, even more than that of Boethius,
neglects the practical aspect of happiness.
Such an interpretation of the nature of happiness places philosophy on a
par with theology. This explains the reaction of the theologians which took
shape in the Paris condemnation of 1277. The 219 condemned theses
include some concerning happiness and the self-aggrandisement of philo
sophy. 'There is no better calling than to dedicate oneself to philosophy.' 3 7
This proposition fits well with Boethius' views. The thesis that 'we can
understand God in his essence in this mortal life' 3 8 is propounded in
Boethius' Liber de felicitate. Ά man whose understanding and appetite is
39
properly o r d e r e d . . . is sufficiently disposed for eternal bliss'. Such a
proposition appears to abolish the necessity for divine grace and turns the
'perfect' happiness of the theologians into an object of human activity.
The condemnation of 1277 does not seem to have intimidated the
philosophers. The Vatican commentary on the Ethics (MSS Vat. lat. 832 and
2172), probably written around the end of the thirteenth century, took a
34. Nardi 1945, pp. 24-9 and 34-8.

35. Ibid., p. 27: 'nos quandoque beabimur per essentiam Dei, ut ille, et per nihil medians'.
36. Siger of Brabant 1948, III, 1, p. 84: 'homo multum expertus in philosophia a causatis a primo
posset pervenire ad intellectum essentiae Primi'.
37. CUP I, n. 473, sent. 40: 'quod non est excellentior status quam vacarc philosophiae'.
38. Sent. 36: 'quod Deum in hac vita mortali possumus intelligere per essentiam'. Cf. note 35.
39. Sent. 157: 'quod homo ordinatus quantum ad intellectum et affectum ... est sufficienter dispositus
ad felicitatem aeternam'.
position similar to that of the Liber de felicitate: Talking of the happiness of

the present life, the anonymous author says 'happiness consists in the
knowledge of the divine essence'.40 The author draws the conclusion that if
happiness consists in the knowledge of the nature of God and such knowl
edge is possible in this life, then the will must restrict itself to loving what is
presented to it by knowledge. This contradicts the thought of Aquinas,
who gives the will priority over reason in this life: as long as the essence of
God is not yet seen, the will can get closer to the final goal than reason can.
These thirteenth-century philosophers never entirely lost sight of the
practical aspect of the Nicomachean Ethics. For Aristotle happiness is some
thing possible for most or many men. Happiness as the highest good of
man is, for him, something possible even without the philosophical life;
such practical happiness is a 'second-best' life,41 but nonetheless in its way a
highest good. Most philosophers of the thirteenth century make this
second-best life a mere appendix of the philosophical life - an elitist
position that sacrifices the proper claims of the practical life on the altar of
theoria.
John Buridan and the concept of the contemplative life

Not all theologians accepted the philosophers' decision that theoria is the
epitome of happiness. Bonaventure opposed the thesis of Aquinas that
the reason and not the will is the real bearer of happiness.42 Duns Scotus
saw happiness basically as an act of the will and valued the will more highly
than the reason. Scotus rejected the claim of the philosophers to recognise
the final goal and to reach it by their own powers. Philosophy can
recognise the goal only 'in general', in a way which is not sufficient to
guide conduct. The science which directs human action to its eternal goal is
theology, which through revelation grasps this goal 'in detad'. 43
John Buridan's commentary on the Ethics gives us some idea of the
philosophical response to these criticisms in the fourteenth century. His
description of ethics as an immediately practical science, and his identifi
cation of ethics with prudence can be seen as a reaction against the one-sided
theoretical conception of the philosophers at the end of the thirteenth
century. The topic of freedom plays an important part in his commen-
40. MS Vat. lat. 892, f. io v * (2172, f. 14Λ); 'quando dicitur: félicitas consistit in cognitione divinae
essentiae, verum est'; Gauthier 1947-8, p. 290, n. 1.
41. X, 8; 1178*0·
42. Coll. in Hexaemeron, II, 22-9.
43. Honnefelder 1975, pp. 239 and 274.
684 Ethics
tary; 44 a notable example of this is his assertion that it is the free man who
is the specific object of ethics.43 His commentary treats extensively of the
problem of the relation of reason to will and the relation of both to
happiness.46
In the same context, Buridan mentions a question that had often been
asked since John of Paris: whether happiness consists in the direct contemp
lation of God or in some consequence thereof- e.g., in reflexive knowl
edge of it. Buridan took the side of those who identify happiness with
contemplation. 47 Others, he was aware, drawing a distinction between the
potentia absoluta and the potentia ordinata of God, had argued that in his
absolute power God could turn the joy and love arising from the contemp
lation of God into sorrow and hate, or, conversely, he could detach the love
from the act of knowledge. But Buridan's own mind is made up: Ί am
certain that I do not want such a happiness'.48
In other respects Buridan separates philosophical topics quite clearly
from theological ones: his aim is 'to proceed throughout the book purely
philosophically'.49 But this does not exclude the consideration of theolog
ical positions. He distinguishes between the true happiness of the life to
come (in patria), about which he will not speak, and the happiness of this
life, with which the philosophers deal. 50 The happiness of this fife is
achieved by human exertion; God comes into consideration only as ge
neral cause, being the origin of all good things. Here Buridan is repeating
the reading of Albert the Great. 51
A man can be called happy in the present life, but not in the strict sense
(secundum imaginationem mathematicam) that he has reached the highest point
of all his possibilities. That was true only of Christ. The average happy man
must enjoy certain basic elements of happiness: physical health or at least
the lack of serious physical defect, the moral virtues, prudence and wisdom,
wealth and good fortune. These individual elements admit of degrees, so
that human happiness does not entail a condition of absolute perfection.52
This description of happiness corresponds to the well-known definition
44. Korolec 1974b, pp. 190-252.

va
45. John Buridan 1513, I, qu. 3, f. 4 .
46. Ibid., Χ , qu. 1-3.
47. Ibid., X, qu. s, f. 213"; cf. Gauthier 1947-8, p. 197, n. 2.
48. Ibid., X, qu. 4, f. 209rb: 'constat mihi, quod ego non vellem talem felicitatem'.
49. Ibid., I, qu. 18, vol. 18": '... redeuntes ad nostram intentionem, quae est in hoc toto libro nihil nisi
pure philosophice tractare'.
50. Ibid., I, qu. 10, f. io' b .
51. Ibid., I, qu. 17, f. 17™; Albert, Super Ethica I, 10, 55.
52. Ibid.,X, qu. 4, f. 2ii v '- k .
of Boethius, according to which happiness is a perfect condition consisting

in the union of all goods. 53 Buridan mentions a second definition of
happiness, as the most excellent activity of the noblest capacity upon the
noblest object. This definition he ascribes to Aristotle. The two definitions
do not contradict each other, says Buridan. If one starts synthetically (via
compositionis) from the individual elements, one arrives at 'the unification
of all goods'; if one asks analytically (via resolutionis) about the relation of
the elements to one another, one arrives at the most excellent element to
which all the others are subordinated.54
Neither of the two methods makes quite clear how Buridan understands
the Aristotelian distinction between active and contemplative happiness.
He describes active happiness as the exercise of prudence and contempla
tive happiness as the activity of theoretical reason.55 How does this distinc
tion fit the definitions just mentioned? In the 'Boethian' definition pru
dence and wisdom (contemplation) are both equally elements of happiness;
the 'Aristotelian' definition admits only one act and one object of happiness,
namely the act of contemplating the divine nature (actus speculationis circa
divinam essentiam).56 We may raise the questions whether Buridan regards
active happiness as an independent item in a philosophical ethics or thinks
of it as merely preparatory to true human happiness.
An answer becomes clear when we inquire how Buridan regards the
status of people who do not achieve theoria. He mentions workmen
(mechanici) who look after others and thus without themselves achieving
happiness bring others closer to it. 5 7 Thus on his view there is only one
happiness to which the activity of all men has to be orientated; active
happiness can only be regarded as preparatory to contemplative happiness.
This interpretation is confirmed by two further remarks of Buridan.
'The true metaphysician', he says, 'is a good and perfect man without
qualification', while the prudent man, who is also a good man, does not
achieve the same completion and perfection as the metaphysician. Pru-
53- Ibid., X, qu. 4, f. 210"; Boethius, Consolatio III, pr. 2: 'liquet igitur esse beatitudinem statum
bonorum omnium congregatione perfectum'.
54. Ibid., X, qu. 4, f. 210": 'via resolutionis optimum in homine diceretur ilia dispositio vel ille actus
aut quocumque nomine nominetur, quae vel qui in praedicta congregatione resoluta in suas
partes esset optimum et finalissimum aliorum ... et certum est, quod istum modum loquendi
tenuit Aristoteles'.
55. Ibid., X, qu. 4, f. 209 vb .
56. Ibid., Χ, qu. 4, f· 2ΐοΓ1.
57. Ibid., I, qu. $, f. 6": 'sciendum etiam non omnes illos homines esse frustrates suo fine, qui ipsum
formaliter non adipiscuntur ... sicut sunt mechanici, qui procurant aliis vitae necessaria, quibus
habitis alii possunt tendere iam ad maiora et propinquiora fini1.
686 Ethics
dence is the rule and the measure of the moral quality of a man in the area of
politics or of praxis, but it is not an absolute rule and standard except as
preparatory for perfect happiness. 5 8 Conversely, it is true that prudence
receives its rule from wisdom: 'It is wisdom which enables prudence to
59
prescribe.' This means that politics and ethics are subordinated to
metaphysics.
Buridan ends his commentary with the question whether human happi
ness consists of an act of reason or of the will. 6 0 His answer is in full accord
with the philosophical tradition of the thirteenth century: happiness is an
act of knowledge by which man is united to God directly, while desire and
love reach their object only through the mediation of knowledge. Accord
ing to Buridan this is true not only of the perfect happiness of the life to
come, but in this life too, even when this knowledge is founded not on self-
evident principles, but on faith. Ά simple woman' may have a greater love
for God based on greater faith and accordingly a greater knowledge than a
scholar who may have evidence and proofs at his disposal. 61
After Buridan
From the second half of the fourteenth century commentators frequently
restrict their consideration to the first five books of the Nicomachean Ethics.
This suggests that interest in theoretical happiness had begun to recede. But
it is uncertain whether this is related to a particular conception of ethics as
primarily practical or whether it is simply the result of teaching schedules in
the universities. Melanchthon, for example, gives the impression of being
unwilling to take the Aristotelian conception of theoretical happiness into
account. Given his theological background, he naturally judges sceptically
any philosophical claim to provide a true ultimate goal for man. But his
attitude is perhaps also a sign that people are beginning to doubt funda
mentally the possibility of human perfection by theoria.
58. Ibid.,V, qu. 22, f. 139"'': 'verus metaphysicus est simpliciter bonus homo et pcrfectus; prudcns
autem non metaphysicus est simpliciter, idest absoluto sermonc loquendo, bonus homo, sed non
perfecte et consummate ... (f. 140"] prudentia est régula et mensura totius bonitatis humanae
politicae, non totius bonitatis humanae simpliciter nisi praeparative sicut ministra'.
59. Ibid., I, qu. 6, f. 7": 'prudentia praecipit gratia sapientiae'.
60. Ibid., X.qu. 5, f. 2 i l , b .
61. Ibid., X,qu. 5, f. 2l4"- b .
36
CONSCIENCE
Origins of the medieval discussion: Peter Lombard

Medieval treatises on conscience were divided into two parts, one headed
'synderesis' and the other 'conscientia'. 'Synderesis' is just a corrupted trans-
literation of 'suneidësis' the Greek word for 'conscience', so the medieval
distinction between synderesis and conscientia requires explanation. In the
first instance, the explanation is historical. Conscience was not directly
treated either by Plato or by Aristotle; the way in which it became a
standard topic of later medieval philosophy was curious, almost an
accident. Like many other topics regularly discussed by medieval philos-
ophers, it came to their attention through a passage in Peter Lombard's
Sentences and most of the medieval treatises on conscience are to be found
in commentaries on that work.
Yet Peter Lombard does not actually discuss conscience at all: his
question is how the will can be bad (2.39). As usual, he reports several
answers, though, exceptionally, without pronouncing judgement upon
them at the end. He notes, first, that some people distinguish two senses of
'voluntas', in one of which it is a power, in the other the exercise ofthat
power (1.3). This distinction was probably inspired by a parallel Aristo-
telian distinction, between two senses of'know', the first dispositional, but
the second involving actually thinking about what one knows, as is some-
times necessary when using one's knowledge.1 Similarly, we each have a
host of desires, but it is only at certain times that any one of them makes
itself felt or that we pay attention to it, so that it is then actualised in the
sense of being called to mind.
The problem which Peter sets out requires this interpretation, for, he
continues, the will is part of man's natural endowment, and he rejects the
solution that qua potentiality it is always good but qua actualisation some-
times bad, on the ground that there is nothing wrong with calling to mind
what one knows, so why should there be anything wrong with calling to
1. De anima III, 4; 429*29 ff.
688 Ethics
mind what one wants? He admits, though, that there may be some
occasions when it is bad to call to mind what we know: 'for now and again
a person remembers something bad in order to do it, and seeks to under-
stand the truth in order to attack it'.2
Yet these are exceptional, rather than typical cases, whereas evil desires
are commonplace, and this leads him on to the famous passage in Romans
7 where St Paul describes his own internal conflicts: 'For I do not do what I
want, but do what I do not want' (7.15). Are there, then, asks Peter, two
wills in man? Those who say 'Yes' fall into two camps. The first group
holds that the will by which a man wants to do good in such a conflict is the
will with which he is naturally endowed; it is the spark of conscience
which, as Jerome said, was not extinguished even in Cain, whereas the
other will is a result of the fall of Adam and Eve. The second group takes
the opposite view: the will by which a man wants to do what is bad is
embraced by free choice and is in the ascendant unless and until God's grace
gives greater strength to the will that wants to do what is good. Finally,
there are those who maintain that there is only one will in man, by which
he 'naturally wants what is good and through a defect in it wants and takes
pleasure in what is evil; so that, to the extent that he wants what is good he
is naturally good, but to the extent that he wants what is bad he is evil'.3
Origins of the medieval discussion: Jerome

Peter concludes that the question whether there are two wills in man is a
deep one, leaving it to his successors to decide among the three solutions.
Thus the topic by which he was exercised here is the classical one of
weakness of will, which had been discussed in some detail by Plato,
Aristotle, and Augustine, as well as by St Paul. Conscience is no more than
mentioned, and then only with a reference to Jerome. But his commen-
tators followed up that reference, which is to an allegorical interpretation
of Ezekiel's vision. Ezekiel saw four living creatures coming out of a fiery
cloud; each of them had the form of a man, but with four faces: the front
face was human, the right face that of a hon, the left that of an ox and the
back face that of an eagle (Ezekiel 1.4-14). Jerome interpreted the four
faces as representing the structure of the human soul, correlating the first
three with Plato's tripartite division in the Republic (IV, 436B-441B). The
eagle, however, he identified with conscience:
2. 'Memorat enim interdum quis malum ut faciat; et quaerit intelligere verum ut sciat impugnare'
(2.1).
3. '... quo naturaliter vult homo bonum et ex vitio vult malum eoque delectatur; et in quantum vult
bonum, naturaliter bonus est; in quantum malum vult, malus est' (3.4).
Conscience 689
Most people interpret the man, the lion and the ox as the rational, emotional and
appetitive parts of the soul... And they posit a fourth part which is above and
beyond these three, and which the Greeks call synteresis: that spark of conscience
which was not even extinguished in the breast of Cain after he was turned out of
paradise, and by which we discern that we sin, when we are overcome by pleasures
or frenzy and meanwhile are misled by an imitation of reason. They reckon that
this is, strictly speaking, the eagle, which is not mixed up with the other three, but
corrects them when they go wrong ... However, we also see that this conscience is
cast down among some people, who have neither shame nor insight regarding
their offences, and loses its place.. . 4
Peter's citation ofJerome was thus very apposite, since Plato invoked the
tripartite division of the soul precisely to explain weakness of will. But
Jerome's remarks gave his medieval readers many headaches, although
they took no interest in them as exegesis; the text of Ezekiel and its meaning
play no further part in the discussion.
Jerome proposes, in this passage, a quadripartite soul. By the middle
ages, Aristotle's interpretation of the 'parts' of the soul as basic psycho
logical potentialities was generally accepted. On this reading, Jerome is
saying that conscience is a fourth potentiality, irreducible to any of the
other three. But he then goes on to ask whether a person can cease to have a
conscience and, primafacie, his answer is inconsistent. First, he seems to say
'No', because he tells us that even Cain did not cease to have a conscience, a
rather surprising remark in view of the story of Cain and Abel, for at no
point in the story does Cain show the slightest sign of being sorry for
having murdered his brother. When the Lord asks him: 'Where is Abel
your brother?' he tries to disown any responsibility: Ί do not know, Am I
my brother's keeper?' (Genesis 4.9). Subsequently, after being sentenced to
a nomadic life, he merely complains: 'My punishment is greater than I can
bear* (Genesis 4.13). However, Jerome then goes on to say that very
wicked people do cease to have any conscience, quoting other passages of
Scripture in support (Proverbs 18.3, Jeremiah 3.3).
Medieval philosophers thought that they could resolve this apparent
inconsistency. Jerome introduces the example of Cain in apposition to the
word 'synteresis', 'that spark of conscience which was not extinguished
4. 'Plerique ... rationale animae, et irascitivum, et concupiscitivum ... ad hominem et leonem ac

vitulum referunt. ... Quartamque ponunt quae super haec et extra haec tria est, quam Graeci
vocant ουντήρησιν, quae scintilla conscientiae in Cain quoque pectore, postquam ejectus est de
paradiso, non extinguitur, et qua victi voluptatibus vel furore, ipsaque interdum rationis decepti
similitudine, nos peccare sentimus. Quam proprie aquilae députant, non se miscentem tribus, sed
tria errantia corrigentem . . . Et tamen hanc quoque conscientiam . . . cernimus praecipitari apud
quosdam et suum locum amittere, qui ne pudorem quidem et verecundiam habent in delictis ...'
(Lottin 1948, pp. 103-4).
690 Ethics
even in the breast of Cain . . . and by which we discern that we sin'. At the
end of the passage, however, he uses the Latin 'conscientia' when he says 'this
conscience is cast down among some people . . . and loses its place'. Few
western medieval philosophers knew any Greek, so this suggested a dis
tinction to them between synderesis and conscientia, synderesis being the
'spark of conscientia'. Cain at least regretted the consequences of his action;
though he did not regret the action as such, this residue of regret witnessed
to a spark of conscience remaining within him.
To medieval philosophers, this would have seemed an honest attempt to
make sense of a puzzling passage in Jerome, although, as exegesis, it will
hardly convince a modern reader. For when Jerome says that this con
science loses its place in some people, he must be referring to its place in the
quadripartite soul. Moreover, he would have known that 'conscientia' was
the exact Latin equivalent of 'syneidesis': the latter is a nominalisation from
the compound verb 'sun-oida', meaning originally Ί know in common
with', so that 'con-scio' is exactly parallel to it. 5 Disagreement with the
medieval interpretation ofjerome does not necessarily force us, though, to
write off any distinction between synderesis and conscientia as an unfortunate
mistake. There could be independent reasons for drawing a distinction
within what we simply call 'conscience' — never mind the labels for it - and
the right question to ask is whether the medieval distinction, in spite of its
muddled origin, turned out to be productive. Do the two terms mark a
distinction which is essential for understanding and speaking clearly about
the notion of conscience? If so, then the original motivation for its intro
duction need not trouble us further.
Philip the Chancellor
The first treatise on conscience, which set the pattern for subsequent ones,
was written by Philip the Chancellor about 1235. Philip poses four ques
tions; he distinguishes between synderesis and conscientia in the third, whe
ther synderesis can lead us to do wrong. He begins with four arguments to the
effect that it can, of which the first runs as follows: 'Conscientia is sometimes
mistaken, sometimes right. But in whatever power there is any mistake
6
over what is to be done, in that power there is sin.' Now (citing Jerome),
synderesis is the same as conscientia, so synderesis can lead us to do wrong. In
his subsequent discussion, however, Philip denies that they are the same,
5. See Davies 1962, p. 672.

6. ' . . . conscientia aliquando est erronea, aliquando recta. Sed in qua viesterrorinoperandis, inea vi
est peccatum' (Lottin 1948, p. 150.3—5).
Conscience 691
holding that conscientia comes from a conjunction of synderesis with free

choice, and illustrating his position with an example:
. . . suppose that it is written in synderesis that everyone who makes himself out to
be the Son of God and is not, should die the death; and that this man (indicating
Christ) makes himself out to be the Son of God, and yet is not (so it is thought);
therefore he should die the death. What was contributed by synderesis was un-
changeable and dictated only good, but this conjoined with what was contributed
by reason dictated sin. So, therefore, synderesis plus the reason for a free choice
makes conscientia right or mistaken, and conscientia sticks more to the side of reason;
synderesis itself, however, which is the spark of conscientia, . . . is not mistaken.7
In his fourth question, whether a person can lose synderesis, Philip

discusses a closely related example, that of heretics whose conscientia urged
them to die for their faith. Primafacie, synderesis did not murmur in them in
answer to sin, but Philip replies:
, . . the effect of synderesis, considered as such, is paralysed in them because of the
lack of faith, which is the basis of everything good. But the exercise of conscientia
thrives in them, the evidence of which is that the man is ready to undergo
martyrdom, because he supposes that what he believes is the faith. It is not,
however, synderesis which does this, but what belongs to free choice or reason.
Moreover, synderesis is not extinguished in such a person because, although he may
be mistaken about the particular matter, evil in general still displeases him.8
This contrast between the general and the particular occurs again in an
addendum to the second question, devoted to the sense in which synderesis
is said to murmur back in answer to sin:
. . . synderesis affects free choice by telling it to do good and restraining it from evil,
and moves us to the general good which is found in this or that good. Hence it is
not in itself directed to particular good deeds, but to the general [good] which is
present in them.9
7. '... si in syndcresi sit scriptum quod omnis qui sc fecerit filium Dei et non sit, morte moriatur; sed
iste sc faciat filium Dei, demonstrate Christo, et non est; ita opinantur: morte ergo moriatur.
Quod erat syndereseos erat immutabile et non dictabat nisi bonum; sed ilium coniunctum cum eo
quod erat rationis dictabat peccatum. Sic ergo synderesis cum ratione liberi arbitrii facit con-
scientiam rectam vel erroneam; et conscientia magis se tenet ex parte rationis; ipsa tamen
synderesis non est erronea, que est scintilla conscientie ...' (Lottin 1948, pp. 151-2. 61—9).
8. '... in hiis debilitatur effectus syndereseos per se sumpte propter privationem fidei que est
fundamentum omnium bonorum; sed in hiis conscientie actus viget, ratione cuius paratus est ipse
subire martyrium; supponit enim hanc quam credit esse fidem. Hoc autem non facit synderesis,
sed ea que sunt liberi arbitrii aut rationis. Tamen non est in tali extincta synderesis; quia licet erret
in particulari, tamen in generali displicet ei malum ...' (Lottin 1948, pp. 154—5.53—9).
9. '... synderesis movet liberum arbitrium dictando bonum et cohibendo a malo, et movet in
bonum commune quod invenitur in isto bono aut in illo. Non ergo est in bonum particulare
secundum se, sed in commune inventum in eo' (Lottin 1948, p. 148.128-31).
692 Ethics
Philip's argument is thus that we need to distinguish between general and

particular propositions which are of the form: Ά ought to φ', where the
proper name of any person may be substituted for Ά' and any verb-phrase
containing a verb of action for 'φ' (I shall henceforth call these 'deontic
propositions'). His reason is that a person, in spite of holding a correct
general principle, can always misapply it to particular circumstances and so
mistakenly believe himself obliged to a certain action. If both are subsumed
under a single, undifferentiated notion of conscience, then, faced with
examples of the type Philip cites, we cannot answer the question: 'Is the
man's conscience mistaken?' If, however, we distinguish between synderesis
and conscientia, then we can say that his synderesis is not mistaken but his
conscientia is. Yet we now need a more precise characterisation of the
difference between a general and a particular deontic proposition, which,
apart from his examples, Philip does not give us.
Philip's motivation for distinguishing between synderesis and conscientia is
that, although he allows that conscientia can be mistaken, he holds that
synderesis cannot. This should not be confused with his ground for dis
tinguishing between them, namely, that general deontic propositions can
be misapplied to particular circumstances. He gives us a good reason for
distinguishing between two senses of 'conscience', but we are not com
pelled, in accepting it, to agree that in one of these senses conscience cannot
be mistaken. This remains as a further question, which was controversial
even at the time, for he was taking issue with William of Auxerre. 10 The
distinction could be made out just as well by an example in which the
general proposition was false, yet misapplied to yield a true particular
proposition. But then we should have to regard synderesis as consisting in
the set of our deontic beliefs, i.e. those which could be reported in the form:
Ά believes that he ought to φ', whereas the etymology of'conscience', as
well as many of the things which we ordinarily say about it, suggest that it
is a form of knowledge.
Philip also holds that conscientia involves free choice, whereas synderesis is
non-deliberative. The same motivation is at work here: only where choice
comes into play is wrong-doing possible. But we can easily concoct
examples which, like Philip's, support a distinction between two senses of
'conscience' but which afford no ground to posit choice as a factor in assent
to the minor premiss. Suppose that a man believes that he ought to give
alms to the poor, but mistakenly believes of a certain very rich widow that
10. See Lottin 1948, p. 150.

Conscience «93
she is destitute, and concludes that he ought to give alms to this widow. He
may simply have been misinformed about her and have made a purely
factual mistake totally unconnected with his desires. It seems, then, that
Philip is again asking his distinction between general and particular deontic
propositions to do too much work and that he has been misled by his choice
of examples.
In his first and second questions, Philip is trying to determine how
synderesis fits into the structure of the soul. The first asks whether it is a
potentiality or a disposition, the second how it is related to reason. He finds
a clue to these problems in the impulse to sin, of which he sees conscience as
the counterpart:
Since the soul is not abandoned by its creator so that it has no help in doing what is
good, just as it contains an impulse to sin inclining free choice towards sin or evil,
so, therefore, there will be some aid which, to the extent that it works of itself,
always directs it to what is good and makes it shun what is bad, in the same way as
the impulse to sin works in the contrary direction. But what else can this be except
synderesis?1 x
This counterpart theory of conscience seems to derive from the Rabbinic

doctrine of the two impulses in man, according to which every man is born
with an evil inclination, manifested primarily in bodily and especially in
sexual desires. At the age of puberty it begins to be opposed by an impulse
to good, the struggle between the two impulses continuing for the rest of a
man's life.12 St Paul was writing within this tradition in his description of
chronic conflicts of desires, in which the evil impulse appears as the 'law of
sin', and in Augustine (Confessions, VIII) it is neatly meshed into the
Platonic and Aristotelian analyses of weakness of will. But it was also
modified in Christian hands, the impulse to sin now being seen, not as part
of God's design for men, but as a consequence of the Fall. Thus Philip is
able to say that it is 'outside the substance of the soul' ' 3 and to conclude
from this that the same applies to synderesis, i.e. that neither is a basic
psychological potentiality.
If synderesis is the counterpart of the impulse to sin, then it should be
11. 'Cum non sit destituta anima a creatore suo ut non habeat adiucorium in bonum, sicut habet
fomitem peccati inclinantem liberum arbitrium in peccatum aut in malum, erit ergo aliquod
adiutorium semper dirigens, quantum est de se, in bonum et retrahens a malo, qucmadmodum
fomes peccati opposito modo se habet. Sed quid aliud erit a synderesi?' (Lottin 1948,
p. 141.66-71).
12. Davics 1955, pp. 21—5.
13. Lottin 1948, p. 141.
694 Ethics
related to rational desire as the latter is related to bodily appetites. So Philip

concludes that it is an innate tendency of rational desires, what remains of
. . . the original righteousness of man's powers, which Adam had in the state of
innocence, which remained as a little light leading him to God, lest his reason
should be totally turned or bent to temporal things. ... For it is established that
Adam was naturally righteous by virtue of his judgement, will, and emotions . . .
And each of these looks to the highest good, to which it primarily relates. It will
not, accordingly, be a potentiality separated from these powers to the extent that
they are pliable, but will exist in them inflexibly, the same as each one of them.14
By 'pliable', he means that they can be directed towards good or bad,
whereas synderesis is inflexible, being always directed towards good. His
conclusion, then, is that synderesis is innate and not acquired: that it is not a
grace to compensate for the impulse to sin (a view which Peter Lombard
reports, Sentences 2.39.3.3), but what remains of the full control of bodily
appetites which man possessed in the state of innocence: and that, pace
Jerome, it is not a potentiality distinct from reason.
Is it, then, a potentiality or a disposition? His final answer is that it is
neither, but a dispositional potentiality, a term which was his own inven-
tion but which may have been suggested by William of Auxerre: 15
Synderesis, although the morphology of its name makes it sound more like a
disposition than a potentiality, is nevertheless the name of a dispositional poten-
tiality: I do not say of an acquired disposition, but of an innate one. And thus, qua
disposition it can be applied to what occurs as a disposition, qua potentiality to
what occurs as a potentiality. From this it follows that it has a certain difference
from . . . the impulse [to sin] and sensuality. . . . qua potentiality, it is disparate
from . . . sensuality . . . So, if anyone asks whether it is a potentiality or a dis-
position, the right answer lies in taking something in between: a dispositional
potentiality.16
14. ' . . . rectitudinis prime virium, quam habebat Adam in statu innocentie, que remansit tamquam
modicum lumen in Deum ductivum, ne non esset ex toto ratio ad temporalia inclinata vel
incurvata Constat enim quod Adam habuit rectitudinem a principio iudicii et voluntatis et
irascentie naturalem... Et horum omnium est inspectrix relatione ad summum bonum ad quod
principaliter se habet. Et secundum hoc non erit seiuncta potentia ab illis viribus in quantum
flexibiles sunt, sed in illis existens inflexibilis, eadem cum unaquaque illarum' (Lottin 1948,
p. 147.78-90).
15. See Lottin 1948, p. 139 n.
16. "... synderesis, licet secundum formam nominis magis sonare videatur habitum quam potentiam,
tamen est nomen habitualis potentie, non dico de habitu acquisito, sed innato. Et ita ratione
habitus potest apponi ei quod per modum habitus se habet, ratione potentie ei quod per modum
potentie se habet. Unde habet quamdam disparationem . . . 3 fomite et sensualitate . . . : secundum
rationem potentie, disparationem h a b e t . . . a sensualitate;... Si ergo quaeratur utrum sit potentia
aut habitus, respondendum est accipiendo medium: potentia habitualis' (Lottin 1948, pp. 1 4 1 -
2.79-90).
Conscience 695
The point of the reference to the etymology ofsynderesis is, of course, that it
is a compound of oida and that knowledge is a disposition. With this
solution, Philip appears to be trying to have the best of both worlds; yet
rather than worrying too much about his new category of dispositional
potentialities, we might construe his main contention as being that syn-
deresis is akin to a potentiality in being innate, but akin to a disposition in
embodying a tendency, namely, to what is good.
Bonaventure
The history of the treatise on conscience until the end of the thirteenth
century has been ably documented in Lottin 1948. Here I must be highly
selective, and so have chosen Bonaventure and Aquinas as able representa-
tives of views which differ almost as much as the constraints of the medieval
philosophical tradition allowed, while yet exhibiting a progression from
and continuity with Philip's treatment of conscience. Of the three,
Bonaventure is the odd man out. To begin with, he virtually reverses the
synderesisjconscientia distinction and deals with conscientiafirst;though that
could be misleading, because he also draws the distinction in a somewhat
different way from Philip. His first question on conscientia is whether it
belongs to the thinking or to the desiring part of the soul (pars intellectus,
affectus), his first question on synderesis whether it is to be classified with
apprehension or with desire (cognitio, affectus). These questions relate to the
Aristotelian distinction between theoretical and practical: apprehension,
which may be perceptory or intellectual, aims at truth, the world as it is;
desire, which may be instinctual/emotional (the 'passions' of the soul) or
rational (will), aims at the good, the world as one would like it to be and,
thus, at changing it.
Nor does Bonaventure make any use of Philip's distinction between
general and particular deontic propositions, though he reports Philip's
view, without committing himself to it.17 Instead, he takes his cue from
John of Damascus' remark that conscience is the law of our thought 18 and
avers that 'a law is what we recognise by means of conscientia',19 But this
does not distinguish conscientia from synderesis, since both are related to
natural law. The latter, however, can be understood both as a set of
injunctions and as a psychological disposition. As a set of injunctions, it is
the object both of conscientia and of synderesis, the first telling us them and
17. Bonaventure i882-loo2a, II, dist. 39, 2. I.
18. John Damascene 1955, 95.
19. 'lex enim est illud quod per conscientiam novimus' (1.1, corpus).
696 Ethics
the second inclining us to observe them; moreover, conscientia is a dis-

position of practical reason, synderesis a potentiality of desire, because it
tends towards what is good. 20 In other words, the set ofinjunctions has first
to be known to us (the job of conscientia) but, once we know them, we shall
in some sense want to follow them (the job of synderesis).
Bonaventure develops this line of thought in his discussion of the
question whether conscientia pertains to thought or to desire, after dis-
tinguishing three senses of conscientia: its object (that of which we are
conscious, the natural law), the potentiality of being conscious (ability to
become conscious of the natural law) and the disposition in virtue of which
we are conscious (knowing the natural law but not continuously exercising
our knowledge). Etymological influence is marked here: conscience is not
clearly distinguished from consciousness (a very recent development) and
it is taken for granted that conscience is a form of knowledge rather than of
belief. Bonaventure then comments upon the third sense, which he holds
to be the most usual one:
If, then, it be asked ofwhat potentiality it is a disposition, it should be said that it is a
disposition of a cognitive potentiality, but not in the same way in which theoret-
ical knowledge [is a disposition of a cognitive faculty), because theoretical knowl-
edge perfects our understanding insofar as it is theoretical, whereas conscientia is a
disposition perfecting our understanding insofar as it is practical, or insofar as it
directs us towards deeds. And thus understanding has in a way a motivating
character, not because it produces motion, but because it directs and inclines us
towards motion. Such a disposition is, accordingly, not called just 'knowledge'
(scientia) but 'conscience' (conscientia), in order to signify that this disposition does
not perfect the potentiality for theoretical knowledge itself, but as [that poten-
tiality] is joined in some way to desire and deed. Because of this, we do not say that
conscience dictates principles like 'every whole is greater than any of its parts', but
rightly say that it tells us 'God is to be honoured' and similar principles, which are
like rules for what is to be done.2 '
20. 2.1, ad finem.

21. 'Si ergo quaeratur, cuius potentiae sit habitus; dicendum, quod est habitus potentiae cognitivae,
aliter tamen, quam sit ipsa speculativa scientia: quia scientia speculativa est perfectio intellectus
nostri, in quantum est speculativus; conscientia vero est habitus perficiens intellectum nostrum, in
quantum est practicus, sive in quantum dirigit in opere. Et sic intellectus habet quodam modo
rationem motivi, non quia efficiat motum, sed quia dictât et inclinât ad motum. - Et propterea talis
habitus non simpliciter nominatur scientia, sed conscientia, ut in hoc significetur, quod habitus iste
non perficit ipsam potentiam speculativam in se, sed prout est quodam modo iuncta affectioni et
operation!. Propter quod nos non dicimus, quod dictamen conscientiae sit ad hoc principium:
omne totum est maius sua parte, et ad consimilia; sed bene dicimus, quod conscientia dictât,
Deum esse honorandum, et consimilia principia, quae sunt sicut regulae agendorum' (ι .ι, corpus).
Conscience 697
It is, then, concerned with the knowledge of ends and, in the last sentence,
the comparison with a stock example of a first principle of theoretical
reasoning already hints at Bonaventure's account of how we come to
know these ends.
He pursues this in discussing his second question, whether conscientia is an
innate or an acquired disposition:
. . . since it is necessary to apprehension that two things should be present concur-
rently, namely, what can be apprehended and light by means of which we judge
the former, as we see in the case of sight..., apprehensory dispositions are partly
innate because of a light imparted to the soul, but also pardy acquired because of
the forms... For everyone agrees that there is a light imparted to the ap-
prehensory potentiality, which is called 'natural judgement', but we acquire forms
and likenesses of things by means of the senses . . . For no one would apprehend
whole or part, or father and mother, unless he received its form through one of the
external senses;... However, the light of natural judgement directs the soul itself
in judging both of what can be apprehended and of what can be d o n e . . . Since
conscientia thus names a disposition which directs our judgement with respect to
what can be done, it follows that in one way it names an innate disposition with
regard to the basic dictates of nature, but an acquired disposition with regard to
what is added by education... For I have a natural light which is enough to
apprehend that one's parents are to be honoured and that one's neighbours are not
to be harmed, but I do not have the form offather or form of neighbour naturally
impressed upon me. 22
Transposing this account into modern terminology, Bonaventure's

contention is that we have to learn a language before we can have any
deontic knowledge, but that, having once done so, we see immediately
that certain deontic propositions are true a priori. But Bonaventure does
not maintain that this applies to all deontic propositions:
22. 'Cum enim ad cognitionem duo concurrant necessario, videlicet praesentia cognoscibilis et
lumen, quo mediante de illo iudicamus, sicut videmus in visu... ; habitus cogniti vi sunt quodam
modo nobis innati ratione luminis animae inditi, sunt etiam quodam modo acquisitt ratione
speciei... Omnes enim in hoc concordant, quod potentiae cognitivae sit lumen inditum, quod
vocatur naturale îudicatorium; species autem et similitudines rerum acquiruntur in nobis
mediante sensu ... Nemo enim unquam cognosceret totum, aut partem, aut patrem, aut malrem, nisi
sensu aliquo exteriori speciem eius acciperet... Illud autem lumen sive naturale iudicatorium
dirigit ipsam animam in iudicando tarn de cognoscibilibus quam de operabilibus... - Quoniam
igitur conscientia nominat habitum directivum nostri iudicii respectu operabilium, hinc est, quod
quodam modo habitum nominat innatum, et quodam modo nominat acquisitum. Habitum,
inquam, innatum nominat respectu eorum quae sunt de primo dictamine naturae; habitum vero
acquisitum respectu eorum quae sunt institutionis superadditae. . . . Naturale enim habeo lumen,
quod sufficit ad cognoscendum, quod parentes sunt honorandi, et quod proximi non sunt
laedendi; non tamen habeo naturaliter mihi impressam speciem patris, vel speciem proximi' (1.2,
corpus).
698 Ethics
... the following point is especially to be noted. Just as certain things which can be
apprehended are exceedingly plain, e.g. axioms and first principles, but some
things less plain, e.g. particular conclusions; so, too, some things which can be
done are maximally plain, e.g. 'Do not do to others what you do not want to be
done to you', that one ought to submit to God, and so on. Apprehension of first
principles is said to be innate in us in virtue of that light, because that light is
enough to apprehend them by, once the forms have been assimilated, without any
further persuasion, on account of their own clarity . . . Moreover, apprehension of
the particular conclusions of the various branches of knowledge is acquired in that
the light which is innate to us is not enough to apprehend them, but demands some
persuasion and a new aptitude. This is also to be understood as applying to
deeds "
As an example, the Jews, arguing from the basic deontic premiss that God is
to be obeyed, conclude that circumcision and the dietary prescriptions of
the Mosaic law are still obligatory. But there is a suppressed non-deontic
premiss of the argument, that God commands certain dietary practices and
the circumcision of male children, which, in Bonaventure's view, is no
longer true. 2 4
Conscientia nevertheless includes derived deontic propositions as well as
basic ones; it is infallible only with respect to the latter, and so can be
mistaken. Bonaventure therefore asks whether we are bound to do every-
thing which it tells us to be necessary to salvation. He replies that we are,
provided that it does not tell us to do something which is against the law of
God; in that case, such a
. . . conscientia does not bind us to act or not to act, but binds us to get rid of
i t . . . since whether a man does what it says or the opposite, he sins mortally. For if
he does what his conscientia tells him, and that is against the law of God, and to act
against the law of God is mortal sin, then without any doubt he sins mortally. But
if he does the opposite of what his conscientia tells him,... he still sins mortally, not
in virtue of the deed which he does but because he does it in an evil way. For he
does it in despite of God, so long as he believes, his conscientia telling him so, that
23. 'hoc attendendum est praecipuc; quia, sicut inter cognoscibilia quaedam sunt valde evidentia, sicut
dignitates et prima principia; quaedam sunt minus evidentia, sicut conclusiones particulates; sic et
in operabilibus quaedam sunt maxime evidentia, utpote illud: "quod tibi non vis fieri, alii ne
fëceris", et quod Deo obtemperandum est, et consimilia. Quemadmodum igitur cognitio
primorum principiorum ratione illius luminis dicitur esse nobis innata, quia lumen illud sufficit ad
ilia cognoscenda, post receptionem specierum, sine aliqua persuasione superaddita, propter sui
evidentiam . . . Rursus, quemadmodum cognitio particularium conclusionum scientiarum ac-
quisita est, pro eo quod nobis lumen innatum non plene sufficit ad illa cognoscenda, sed indiget
aliqua persuasione et habilitatione nova: sic etiam intelligendum est ex parte operabilium ..." (1.2,
corpus).
24. 2.3 ad 4.
Conscience 699
this displeases God ... God does not merely take notice of what a man does, but
with what intention he does it.25
The poor man is thus caught in a double bind, for the moment he
recognises his conscience to be mistaken, he will, ipsofacto, have changed it.
This would be reasonable only if no one could have a mistaken conscience
in good faith but, ex hypothesi, the situation can arise only where a derived
deontic proposition is involved, and Bonaventure allows that man's rea-
soning is fallible.
His account of synderesis leads him to reject the counterpart theory of
conscience espoused by Philip. Synderesis, he argues, is an innate tendency
to want to do what is honourable rather than useful, a natural bias of desire
corresponding to the natural light of conscientia by which basic deontic
propositions are known (2.1). It cannot, therefore, be extinguished by sin;
the worst that can happen is that its exercise is temporarily prevented, e.g.
'sometimes in sins of the flesh a man is so engrossed . . . that a sense of guilt
has no place, because men of thefleshare so far carried away by the impulse
to pleasure that reason has then no place [in them]'. 26 This is at once
reminiscent of the impulse to sin, which also features explicitly in an
argument he considers to the effect that synderesis can be extinguished, on
the ground that the impulse to sin can be extinguished (as in the Virgin
Mary). His reply is that they are not comparable, because the impulse to sin
is not constitutive of human nature, whereas synderesis (and conscientia)
are.27 He is equally adamant that the impulse to sin is no guide to conscientia;
replying to the argument that, since the law of the flesh is opposed to the
law of the mind and the former is related to desire, so must be the latter, he
says that the law of the flesh presupposes that imagination and apprehen-
sion represent bodily things to us in a disordered way and, hence, corrup-
tion of the perceptory potentiality, which pertains to apprehension, not
desire.28 But this very point also casts doubt upon his hypothesis that
25. 'conscientia non ligat ad faciendum, vel non faciendum, sed ligat ad se deponendum, .. quia,
sive homo faciat quod dicit, sive eius oppositum, mortaliter peccat. - Si enim.iaciat quod
conscientia dictât, et illud est contra legem Dei, et facere contra legem Dei sit mortale peccatum;
absque dubio mortaliter peccat. Si vero facit oppositum eius quod conscientia dictât,... adhuc
peccat mortaliter, non ratione operis, quod facit, sed quia malo modo faç>t. Facit enim in con-
temptum Dei, dum credit, dictante sibi conscientia, hoc Deo dispilcere non tantum
attendit Deus, quid homo faciat, sed quo animo faciat' (1.3, corpus)
26. 'aliquando enim in peccatis carnalibus ita absorbetur homo . :. ut remorsus locum non habeat,
quia carnales homines tanto impetu delectationis feruntur, ut ratio tunc non habeat locum' (2.2,
corpus).
27. 2.2 ad 4.
28. 1.1 ads-
700 Ethics
desire is naturally biased to what is honourable, for, of the ways in which

the exercise of synderesis can be hampered, the first two (the darkness of
blindness and the wantonness of pleasure) are failures of apprehension,
while the third (obstinacy) involves the conviction that one is right.
Thomas Aquinas
Aquinas consolidated Philip's way ofdistinguishing between synderesis and
conscientia with Bonaventure's distinction between basic and derived de-
ontic propositions. He made a significant modification to Bonaventure's
treatment of mistaken conscience, but his most original contribution was
an account of the meaning of deontic propositions. He discussed con-
science on three occasions; here I shall concentrate upon the Disputed
Questions on Truth, 16-17, which date from 1257-58. Although he wrote
on the topic again ten years later, he did so much more briefly and made no
significant modifications to his previous exposition. He begins, like Philip,
with synderesis.
Synderesis, according to Aquinas, is a natural disposition of the human
mind by which we apprehend the basic principles of behaviour, parallel to
that by which we apprehend the basic principles of theoretical disciplines,
and in both cases these principles are apprehended without inquiry. Just as
there is a natural disposition of the human mind by which it apprehends the
principles of theoretical disciplines, which we call the understanding of
principles, so too it has a natural disposition concerned with the basic
principles of behaviour, which are the general principles of natural law.
The disposition in question, he concludes, is a disposition of the poten-
tiality of reason, but synderesis can be used either to mean this disposition,
which is comparable to that by which theoretical principles are apprehen-
ded, or to mean the potentiality of reason as endowed with this
disposition.29
Conscientia, by contrast, Aquinas holds to be an actualisation, the appli-
cation of deontic first principles known by synderesis. He distinguishes two
kinds of application: the case in which a person asks himself the question,
before acting, 'What ought I to do?' from that in which, afterwards, he asks
himself'Did I do the right thing?' 30
29. Cf. 2.2.
30. 'uno modo secundum quod consideratur an actus sit vel fuerit: alio modo secundum quod con-
siderarur an actus sit rectus vel non ... Et haec duplex via in operativis distinguitur secundum
duplicem viam quae est in speculativis; scilicet viam quae est inveniendi et iudicandi. Ilia enim via
qua per scientiam inspicimus quid agendum est, quasi consiliantes, est similis inventioni, per
quam ex principiis investigamus conclusiones. Ilia autem via per quam ea quae iam facta sunt,
Conscience 701
Since Aquinas holds that basic deontic propositions are known to us

without inquiry, we should expect him to say that synderesis cannot do
wrong. He does indeed say that, but for a different reason - viz., that the
whole edifice of knowledge, whether theoretical or practical, rests upon
first principles, so that, if we could be wrong about them, nothing would
be certain.31 He is here assuming an Aristotelian view of the 'edifice of
knowledge' as a deductive system, which few would agree with today;
mathematics is probably the only discipline for which such a characterisa-
tion is plausible. However, his reason for thinking that deontic first
principles are known to us without inquiry is that they are necessarily
true.32 It has been very widely assumed throughout the history of philos-
ophy that necessary propositions are self-evident, but this is to confuse
epistemology with logic. To classify propositions as necessary or con-
tingent is to draw a logical distinction, with the implication that a different
method will be appropriate in each case to determining whether they are
true or false. Kant made the first breach in the traditional epistemological
assumption by distinguishing a priori propositions into analytic and syn-
thetic and showing that synthetic a priori propositions, far from being self-
evident, had to be justified by transcendental arguments: arguments to the
effect that our experience could not be what it is were they false. But even
he thought that phüosophy need not concern itself with analytic a priori
propositions because they were self-evident. The development of logic in
the last hundred years has taught us otherwise; even with regard to
tautologies and their negations, only a little complexity need be present for
people's intuitions about their truth or falsity to become very fallible.
Aquinas does allow, of course, that conscientia can be mistaken, and
Ulustrates two ways in which this can happen: first, through invalid
reasoning; second, by combining a deontic first principle with a false pre-
miss, when valid reasoning will not be enough to guarantee a true conclu-
sion. His example of the latter is those who killed the apostles, thinking that
thereby they did God a service.33 He thinks, nevertheless, that conscientia is
infallible when the derived deontic proposition 'falls direcdy' under the
first principle, but his examples, 'God is not to be loved by me' and
'Something bad ought to be done' appear to be negated transformations of
examinamus et discutimus an recta sint, est sicut via iudicii, per quam conclusiones in principia
resolvuntur' (17.1, corpus),
yi. 16.2.
32. 16.1 ad 9.
33. 17.2; cf. 16.2 ad 1,2.
702 Ethics
deontic first principles and, again, people can and do make mistakes in this
type of reasoning, especially when multiple quantification is involved. He
also says nothing about the misapplication of deontic propositions to par-
ticular circumstances, as distinct from the derivation of deontic propositions
by inference.
Despite these criticisms, Bonaventure's replacement of Philip's distinc-
tion between general and particular deontic propositions with one be-
tween necessary and contingent deontic propositions, and Aquinas' de-
velopment of the latter, was an extremely important advance, because it
provides the basis for a justification of change in ethics.
Aquinas' most important and original contributions to the medieval
debate about conscience are made in his answer to the question whether
conscience binds us. He says immediately that it does, but goes on to
explain:
'binding', used of spiritual things, is a metaphor taken from bodily ones, which
implies the imposition of necessity,... There are two kinds of necessity which can
be imposed by another agent. The first is a necessity of force, through which
everything absolutely necessarily has to do what is determined by the action of the
agent; the other should not strictly be called force but, rather, inducement. This is a
conditional necessity, that is, deriving from a goal; e.g. there may be a necessity
imposed upon someone that, if he does not do such-and-such, he will not obtain
his reward.
. . . The second kind of necessity can be imposed upon the will, e.g. it may be
necessary to choose such-and-such, if a certain good is to result, or if a certain evil is
to be avoided .". .3*
This reply distinguishes psychological from physical necessity, but can
also be taken as an exposition of the meaning of 'ought' in deontic
propositions. It is now recognised that there is a close, though not exact,
analogy between 'obligatory' and 'necessary' on the one hand, and 'per-
missible' and 'possible' on the other. Thus interpreted, Aquinas' first
contention is that deontic propositions are to be expounded as implicit
conditionals. He is aligning the meaning of 'ought' in the deontic pro-
positions with which ethics is concerned with its meaning in a much wider
range of contexts, those in which it introduces a necessary means to a goal.
34. 'ligatio metaphorice a corporalibus ad spiritualia sumpta, necessitatis impositionem importât...

Est autem duplex nécessitas quae ab alio agente imponi potest. Una quidem coactionis, per quam
omnis absolute necesse habet facere hoc ad quod determinatur ex actione agentis; alias coactio non
proprie diceretur: sed magis inductio. Alia vero est nécessitas conditionata, scilicet ex finis sup-
position; sicut imponitur alicui nécessitas ut si non fecerir hoc, non consequatur suum praemium
. . . Secunda nécessitas voluntati imponi potest; ut scilicet necessarium sit hoc eligere, si hoc bonum
debeat consequi, vel si hoc malum debeat vitare' (17.3, corpus).
Conscience 703
What, then, is the presupposed goal in ethical contexts? T o Aquinas, this

presents no difficulty: it is obedience to God's commands. Aquinas is most
explicit about this in his answer to the question whether a mistaken
conscience binds:
It does not seem possible for someone to escape sin if his conscientia, however much
mistaken, tells him that something is an injunction of God which is indifferent or
bad per se and, such conscientia remaining, he arranges to do the contrary. For so far
as in him lies, by this itself he has the wish that the law of God be not observed;
hence he sins mortally.35
Aquinas is talking only about conscientia and not about synderesis: on his
view, a man cannot be mistaken about basic deontic propositions. So the
sense in which the man envisaged has a mistaken conscience is either that he
has drawn a false deontic conclusion from his stock of basic deontic
propositions or that he has misapplied a deontic proposition to particular
circumstances. Nevertheless, the man in question thinks that God enjoins
upon him a certain type of action or a particular action. But his desire to
obey the law of God will be a concomitant of his belief that God ought to
be obeyed, and this, being a basic deontic proposition, will belong to
synderesis and not to conscientia. So it is not because he acts against his
conscientia that he does wrong but because, in acting against his conscientia,
he is also acting against his synderesis, which is infallible.
Aquinas argues that the way in which conscientia binds us is a special case
of the way in which we are bound by the commands of a ruler, namely, by
the threat of sanctions for disobedience or promise of a reward for obe-
dience: 'e.g. there may be a necessity imposed upon someone that, if he
does not do such-and-such, he will not obtain his reward'. God being the
ruler par excellence, his rewards and sanctions are then the ultimate ones. 36
We have already seen that Aquinas holds that a mistaken conscientia is, in
general, binding, but not how he deals with the case in which a man
believes that be is obliged to do something but the action in question is evil.
His view about this case is slightly more lenient than Bonaventure's, but
only slightly. If the mistake is not itself sinful, e.g. when it arises from some
factual mistake, then the man who follows his mistaken conscientia is
excused; but if the mistake is one of law, then he is not excused, any more
than a plea of ignorance of the law excuses in the courts, because the
35. 'Non videtur autem possibile quod aliquis peccatum evadat, si conscientia, quantumcumque
errans, dictet aliquod esse praeceptum Dei quod sit indifferens sive per se malum; si contrarium,
tali conscientia manente, agere disponat. Quantum enim in se est, ex hoc ipso habet voluntatcm
legem Dei non observandi; unde mortaliter peccat' (17.4. corpus).
36. 4"d5-
704 Ethics
mistake has arisen from ignorance of something he ought to have known.

Aquinas realises that in this situation whatever a man does will be wrong,
but this is not an impossible state of affairs.
As a parallel example, he cites a man who is bound to give alms to
someone, but is going to do so with an intention of vainglory: if that is his
intention, then he is obliged not to give the alms, but ex hypothesi he is
obliged to give them. The dilemma can be resolved quite simply by giving
up the bad intention.37
Yet Aquinas does allow that a mistaken conscientia excuses except where
we can say that the person ought to have known better, and he does justify
his position that a man who is mistaken about a matter of law, i.e. about a
deontic proposition, can never be mistaken in good faith. Basic deontic
propositions are not in question for Aquinas here, because synderesis,
according to him, cannot be mistaken. Hence the objects of a mistaken
conscientia must always be false derived deontic propositions or else misap-
plications of deontic propositions. But the former, on his own admission,
can arise only from fallacious reasoning or from the use of false non-
deontic premisses. Moreover, the latter would not count as ignorance of
fact in the sense in which he is contrasting it with ignorance of law;
ignorance of fact in this legal sense is concerned with mistaken descriptions
of circumstances to which a law is applied. Deriving conclusions from
premisses often is difficult to do correcdy, whether in theoretical or
practical matters, so what right has Aquinas to impute evil to everyone
who believes false derived deontic propositions?
Unwittingly, perhaps, Aquinas laid the foundations of an argument
which could lead to a much more lenient position than he himself en-
visaged. For if there are relatively few basic deontic propositions, then the
scope for a mistaken conscientia which is in good faith will be correspond-
ingly wide. It then becomes important for us to identify the basic deontic
propositions, and it is one of the most remarkable features of the medieval
treatment of conscience, in view of the central role of the synderesis/
conscientia distinction, that no serious attempt to do this was ever made. If
we abandon the attempt, and say, unlike Aquinas, that synderesis is fallible
no less than conscientia, then a man with a mistaken conscientia may not, after
all, be caught in a double bind: he may be justified in acting against it. This
is the opposite of what people commonly say today, so the study of medi-
eval philosophy can help them, at least, to think more deeply about issues
upon which, perhaps, they conclude too hastily.
37· 4 ad y
37
N A T U R A L MORALITY AND
N A T U R A L LAW
Sources of the medieval concept of natural law

The chief sources on which the scholastics drew for their knowledge of
natural law were Cicero, the Digest, St Paul, the Fathers and, later,
Aristode.
St Paul observed in his Epistle to the Romans, 2.12-16, that even
without knowledge of the Old Testament Law pagans have its substance
written on their hearts. Conscience and reason lead men to do by nature
what the Law commands.1 Natural law thus accords with the Decalogue.
Lactantius recorded Cicero's definition of law: true law is right reason in
agreement with nature, being found among all men, summoning them to
duty and prohibiting wrongdoing. True law may not be abolished by Se-
nate or People; it is not different in Rome or in Athens, now or in the
future. Its originator and promulgator is God; disobedience to it constitutes
a denial of the nature of man.2
The Digest in itsfirstchapter distinguished three types of law: ius civile or
the law of the state, ius gentium or the law of nations, and ius naturale or the
law of nature. The jurists cited defined the natural law variously. Ulpian
described it as the common instinct of animals; the union of male and
female, the procreation of offspring and their education have been taught
to animals by nature. But Gaius defined the natural law as those human
laws practised by all nations and dictated to all men by natural reason, and
Paulus said that the natural law consists of what is equitable and good. The
Digest does not establish natural law as a law able to override the laws of the
courts, but it affirms that good laws are those that are grounded on nature,
reason, equity and justice. Among the Christian Fathers, Isidore knew the
definitions of the Roman lawyers, and in the fifth book of his Etymologies
he too described the natural law as the law that is common to all nations
and that is set up by a natural instinct, not by any positive constitution. It
[. 'Gentes quae legem non habent naturaliter quae legis sunt faciunt.'
2. Lactantius Divinarum institutionum lib. VI, c. 8 (PL. 6, 660-1); Cicero De republica. III, xxii, 33.
Cf. Cicero De legibus, I, xii, 33.
7θ6 Ethics
enjoins the union of male and female, the procreation and education of
children, the common possession of all things, universal liberty and the
acquisition of things taken in the sky, on earth, and in the sea. It also
includes the restitution of goods held in trust and the repulsion of violence
3
by force. Following the revival of the study of the R o m a n law from the
end of the eleventh century, the definitions given in the Digest became
more widely available. Ulpian's declaration that natural law is the com
mon instinct which nature has taught all animals found the most favour
with the civilians of the twelfth and thirteenth centuries.
Natural law in twelfth-century philosophy

Before the wide diffusion of the new Aristotle, and among the philoso
phers of the twelfth century, notably Peter Abelard and William of
Conches, there was a particular insistence on the equivalence of nature and
reason. Nature is a force implanted in things (vis insita in rebus); it operates
according to the reason of God. Plato's Timaeus and Calcidius' com
mentary on it were central texts in the discussion of natura which was
sometimes personified (e.g. by Alan of Lille in De planctu naturae) as a
goddess who presided over the universe and taught the natural law. Peter
Abelard in his Dialogus inter philosophum, iudaeum et christianum distin
guished ius naturale, which is the reason within men which stipulates what
is necessary for all men, and iuspositivum which is instituted by men. 4 The
natural law is the primitive law followed according to human reason
before the giving of the Mosaic Law, which consists chiefly of a writing
down of the content of the natural law. 5 The gentile philosophers, how
ever, not knowing this revelation, pursued the summum bonum through the
practice of virtue according to reason, and their lives are as exemplary as
6
those of the Prophets. The moral precepts of the Gospel constitute a
reformation of the natural law followed by the philosophers. 7
The Aristotelian contribution

With the broadening of knowledge of the Aristotelian corpus, beginning in
the late twelfth century, the scholastics were not only able to give natural
3. Etymologiarum, lib. V, c. 4.
4. Dialogus inter philosophum, iudaeum et christianum, in PL. 178, 1656. On this see Gagner i960,
pp. 210—24.
5. Dialogus, in PL. 178, 619-23.
6. Theologia Christiana, II.
7. 'Si enim diligenter moralia Evangelii praecepta consideremus, nihil ea aliud quam reformationem
legis naturalis inveniemus, quam secutos esse philosophos constat', Theologia Christiana, II (PL.
178, 1179D).
Natural morality and natural law 707
morality and natural law a metaphysical foundation but were also able to
read Aristotle's own statement about natural justice. In his Rhetoric I, XIII,
2 (1373b) Aristotle distinguished between laws which are particular to
individual states and general laws which are in accordance with nature and
which reflect a universal idea of natural justice. Similarly in the
Nicomachean Ethics V, vii (1134b- 1135a) Aristotle affirmed the existence of
natural justice and, in writing about political justice which regulates
relations among citizens, he distinguished between natural rules which
have the same validity everywhere and do not depend upon enactment,
and conventional or legal rules, which, for example, lay down the cost of a
ransom or the detailed requirements of sacrifice. These latter rules differ
from state to state.
Gratian and the decretists

In the introductory distinctions of the Concordia discordantium canonum,
Gratian followed Isidore in defining natural law as the law common to all
nations by virtue of being found everywhere because of natural instinct
and not because of any positive constitution. It came into existence with
the very creation of man as a rational being; it does not vary in time, but
remains unchangeable.8 But Gratian said more than Cicero or the Roman
lawyers: 'Natural law prevails in antiquity and in dignity over all laws'.
'Whatever has been recognised by custom, or laid down in writing, if it
contradicts natural law, must be considered null and void.' 9 He cites
Isidore, Etymologies V, 2: 'All laws are either divine or human. Divine laws
are based on nature, human laws on custom. The reason why the latter
differ from one another is that different nations prefer different laws.' 10
Gratian equated the natural law with the basic moral precept contained in
the divine law regarding neighbourly love: 'Mankind is ruled by two
things: Natural law and custom. Natural law is that which is contained in
the Law and the Gospel where everyone is commanded to do to another as
he would be done by and forbidden to do to another what he does not wish
to have done to himself.'ll
8. D.V.l.
9. 'Naturale ius inter omnia primatum obtinet et tempore et dignitate,' D.V.I. 'Quecumque enim
vel moribus rccepta sunt, vel scriptis comprehensa, si naturali iuri fuerint adversa, vana et irrita
sunt habenda', D.VIII.u.
10. 'Omnes leges aut divinae sunt, aut humanae. Divinae natura, humanae moribus constant,
idcoquehae discrepant, quoniam aliac aliis gentibus placent', D.I.
11. 'Humanum genus duobus rcgitur, naturali videlicet iure et moribus. lus naturae est, quod in lege
et evangelio continetur, quo quisque iubetur alii facere, quod sibi vult fieri, et prohibetur alii
inferre, quod sibi nolit fieri', D.I.
7θ8 Ethics
The Decretists who explained Gratian's text tended to follow him in

defining natural law as that which is contained in the Scriptures and which
teaches man to do to others what he would have others do to himself. They
pushed into the background Isidore's natural law common to all nations.
But like the philosophers they readily invoked the concept of natural
justice with references to the Timaeus or to Stoic morality. Thus Rufinus of
Assisi (d. ca. 1190) drew upon Cicero to describe man's capacity to dis
tinguish between good and evil as a natural power. 12 The Decretists were
aware of the different senses of the natural law. They began to list the
various definitions more or less systematically: natural law is the teaching
of Scripture, or it is what is left undetermined by divine command or
prohibition; it is the human capacity to distinguish right from wrong; it is
natural equity; it is also the natural instinct of all animals and as well a
general law of all creation.
Early Franciscan masters

The task of trying to classify and analyse these descriptions was taken up by
the scholastics of the thirteenth century, and particularly by the early
Franciscan masters. The treatise on law by John of La Rochelle (contained
in the third part of the Summa attributed to Alexander of Hales) dis
tinguished three modes of natural law: nativum, humanum, and divinum. The
first is the natural law which has regard to all animate creatures; the second
is that which refers to the order of rational creatures and is the ius gentium;
the third is the law regulating human actions in respect of divine grace,
e.g. the Mosaic Law. Each of these modes derives from the eternal law of
13
God. Discussion also developed regarding the permanence or imperma
nence of the precepts of the natural law. Problems were raised by the
doctrine of the Fall and by Old Testament examples of such irregular
actions as cases of polygamy among the Prophets and the Mosaic bill of
divorce. The question whether God could dispense from the natural law or
whether man could adjust it was treated by the Summa fratris Alexandri in
the light of the distinction between immutable precepts (e.g. the law of
charity) and variable demonstrations of these precepts (e.g. communal
property, human equality and liberty). The demonstrations can be ad
justed to suit changing needs. 14 The natural law regulates human reason
12. 'Vis quaedam humanae creaturae a natura insita ad faciendum bonum cavendumque con-
trarium,' Summa Magistri Rufini, cd. von Schulte, p. 4.
13. Alexandri de Hales... Summa theologica. Liber III, pars II, Inq. II, q. IV, c. 1.
14. Alexandri de Hales ... Summa theologica. Lib. Ill, pars II, Inq. II, q. III.
through the medium of conscience which is formed by the natural law, and
human will through synderesis, the spark of conscience (scintilla conscientiae).
The law lays down that good should be done and evil avoided; man makes
a judgement of reason in the light of this principle before committing an
action and this forms conscience; having made his judgement, synderesis
prompts the will to do what is good. * !
Albert the Creat

Albert the Great reacted against the practice of the lawyers who enu
merated the different senses of the natural law, and he rejected Ulpian's
definition of the natural law as that which nature has taught all animals.
Natural law exists only for man and is nothing other than the law of
reason.16 Thus, the procreative instinct of the animals is not part of natural
law or of natural justice because it lacks the rationality required for human
marriage. However, human reason has in the past justifiably permitted
dispensations from the law of monogamy. 1 7 In his commentary on
Aristode's Ethics, V, where he treats of Aristotle's distinction between
natural and legal justice, Albert also rejects the cosmic notion of justice
associated with the Timaeus. Natural justice, which Albert assimilates to the
moral nature of man, proceeds from human reason. 18 He interprets
Cicero's definition of the natural law as an innate force (innata vis) to mean
the light of the agent intellect.19
Thomas Aquinas
The foundation of Aquinas' classic formulation of the doctrine of natural
law is the teleological principle that all beings by their nature have within
themselves inclinations which direct them to the end which is proper to
20
them. Good has the nature of an end and evil is its contrary. There is a
difference between animals which are driven by instinct (aestimatio natu
ralis) and men who also have the capacity of knowing their end and the
relationship of the means to the end. However, human actions arise both
from a natural inclination in the appetitive faculty and from a natural
15. Op. cit.. Lib. Ill, pars II, Inq. II, q. II, c. 3.
16. De bono (ca. 1242), Tractatus V, De iustitia, q. 1, a. 1-2(12). Ethica (after 1260), lib. V, tract. Ill,
cap. HI. Cf. Super Ethica commentum el quaestiones, lib. V, lectio XI, 421 (written 1250-2).
17. D« sactamentis, Tractatus ΓΧ, De matrimonii, Q. 3. Cf. Commentary on the Sentences (1244-9), Kb.
IV, dist. 33A, a.1-3.
18. Super Ethica, lib. V, lectio XI, 419. De bono. Tract. V, q. 1, a. 2.
19. De bono. Tract. V, q. I. a. I (12).
20. 'Bonum est faciendum et prosequendum et malum vitandum', Aquinas S T , lallae, q. 94, a. 2.
710 Ethics
concept in the cognitive faculty. In his commentary on Aristotle's Ethics

(lib. V, lectio XII; written 1271-2), Aquinas distinguished between man's
animal nature, which is obliged by natural instinct to mate and to rear
offspring, and his rational nature which is obliged by the ius gentium to keep
agreements, to grant immunity to envoys, and so on. Aquinas did not
reject Ulpian's definition of the natural law. Indeed, his reason for admit-
ting that polygamy was allowed by the natural law was that not all the
animals restrict themselves to having one mate. The Old Testament ex-
amples of polygamy and of divorce are thus not dispensations or breaches
of the natural law specially allowed by God. 2 1
In his Summa theologiae (lallae, q. 94, a. 2) Aquinas provided a ranking of
the precepts of the natural law which corresponds to an order of natural
inclinations. First, man shares with all substances the inclination to self-
preservation, and the natural law enjoins those things that preserve human
life. Secondly, man shares with all animals a more particular inclination
towards the union of male and female and the education of offspring.
Thirdly, man by virtue of his rational nature has an inclination to do good
and to know the truth about God and to live in society. Human, moral
obligation is therefore imposed by reason and the rational is natural,
because it has a basis in human nature, which is rational. 'In human affairs a
thing is said to be just by virtue of its being right according to the rule of
reason. The first rule of reason is the law of nature.' 2 2 Man's natural desire
includes the desire to know God. In the Summa theologiae (lallae, q. 4)
Aquinas went beyond Aristotle's ethics of human conduct in this life. The
human will and the human intellect cannot be fully satisfied unless they
achieve their supernatural end, which is the vision of God. 2 3 Aquinas
distinguished two ends for man: the natural end, which consists in an
imperfect knowledge of God through his creatures and in the practice of
the natural virtues, and the supernatural end, which transcends the power
of nature, consists of life eternal, and requires the infusion of the theological
or supernatural virtues into the soul.
In ST, lallae, q. 91, art. 1-2, Aquinas gave a detailed formulation of the
place of natural law in the eternal and divine order. Since the universe is
governed by divine providence and by divine reason, and since this is
21. In SCG, III, c. 122, Aquinas condemns fornication as contrary to the divine law which supports
the natural law.
22. 'In rebus autem humanis dicitur esse aliquid tustum ex eo quod est rectum secundum regulam
rationis. Rationis autem prima régula est lex naturae', ST, lallae, q. 95, a. 2.
23. Cf. Aquinas SCG, III, 39.
outside the limits of time, there is also an eternal law to which all creatures
are subject. The end or final cause of all natures is to be what they are, and
all things derive from the eternal law certain inclinations to those actions
and aims that are proper to them. Rational creatures participate in the
eternal law intellectually and rationally and this participation is the natural
law.24
The first and fundamental precept of natural law is 'good is to be done
and pursued and evil avoided' and this precept is self-evident since all
creatures act on account of their end, which is the good for them. 25 In his
commentary on the Sentences Aquinas identifies what is 'written in
everyone's practical intellect' with the Law given to Moses, which re-
iterates what is known to human reason. He also distinguishes between
primary and secondary precepts of the natural law in the light of a
distinction between primary and secondary ends of nature.26 The primary
ends of nature include the use of food, bodily health, procreation and the
upbringing of children. Anything which prevents the achievement of
those ends is forbidden by the primary precepts of the natural law. Here
Aquinas treats a primary precept of the natural law as one which embraces
all animal behaviour, as in Ulpian's definition, although man has in
addition a certain inclination to know the truth about God. The secondary
precepts regulate specifically human and rational behaviour; they relate to
whatever makes the achievement of human ends more possible or im-
possible. The primary precepts are self-evident; secondary precepts are
deduced by the process of reasoning and may, as in the case of monogamy,
be modified in rare circumstances.
Aquinas' distinction between the primary and secondary precepts of the
natural law is presented afresh in ST, lallae, q. 94, a. 4. The primary and
secondary precepts are, respectively, common principles that are self-
evident to the theoretical or speculative reason and conclusions deduced by
the practical reason from the common principles and relating to the sphere
of contingent happenings. The practical reason is not immune from
making erroneous judgements and it does not everywhere and at all times
draw the same conclusions regarding particular situations. The further the
practical reason goes into detail regarding moral judgements and human
events, the more it is likely to fall into error and to provide divergent
24. 'Lex naturalis nihil aliud est quam participatio legisaeternae in rationali creatura', ST, lallae, q. 91
a. 2.
25. ST, lallae, q. 94, a. 2.
26. In IV Sent. d. 33. Cf. In IV Sent. d. 26, q. 1. a. 1, ad 1.
712 Ethics
decisions. However, the primary or common principles are equally well

known to all men and are wholly immutable; they have a kind of necessity
(aliqua nécessitas). The secondary or particular principles that follow closely
by way of conclusion from these first principles are the same among most
people, although in a few societies passion or entrenched social custom may
obscure their appreciation, as happened, according to Caesar, among the
Germans who turned robbery into a virtue. Aquinas says little about the
content of the two kinds of precept, and what he does say is provided by
way of illustration. For example, it is a common principle that man should
do good according to reason, and this leads to the conclusion that what is
borrowed should be returned. But this conclusion is not invariable because
it might be unreasonable in certain circumstances to return borrowed
goods, e.g. if their owner wished to use them to harm the state.
Natural law in Thomas' expositions stands between eternal law on the
one hand and positive law - both divine and human - on the other. It is
linked with both. It is neither an ultimate nor a detailed source of law or
morality. In a long discussion of the divine law in SCG, III, 1140°. Thomas
presents the divine law not as the Old Testament Law and the Gospel but as
a development of the requirements of nature and as a help to the natural
law, commanding explicitly what is intrinsically and objectively right.
Divine law is the reason of God in relation to the governance of human
beings. 27 Its end is the love of God and of one's neighbour. The things that
are prescribed by the divine law are right not only because they have been
established by law but also because they are in accordance with nature and
are naturally right (SCG, III, 129). The divine law prescribes that man
observe the order of reason in those things that are given over to his use; it
confirms the rational distinction between good and evil.
Human positive law is also, like all law, an act of reason. 28 Its purpose is
to benefit human life by deriving conclusions from the natural law that suit
particular needs and by adding to the precepts of the natural law, just as the
written law is said to supply what was wanting in the natural law.
Through human positive law private property and slavery have been
instituted to meet the convenience of man (ST, lallae, q. 94, a. 5 and q. 95).
In his Commentary on the Ethics, V (lectio XII) Aquinas uses the terms
'conclusio' in respect of principles derived from natural justice (e.g. thieves
27. 'Lex . . . est quaedam ratio divinae providentiae gubernantis rationali creaturae proposita', SCG,
III, 115; 'lex divina est quaedam ratio divinae providentiae ad homines gubernandos', ibid., 128.
28. '[Lex] nihil est aliud quam quaedam rationis ordinatio ad bonum commune ab eo qui curam habet
communitatis promulgata', ST, lallae, q. 90, a. 4.
should be punished) and 'determinatio' in respect of positive laws abo

derived therefrom (e.g. such or such a penalty should be inflicted). Law is
commanded by the ruler's will, but command (imperare) is an act of reason
that springs from the will (ST, lallae, q. 90, 1 ad 3). If the will of the
legislator is not regulated by reason, and if human law is at variance with
natural law, it is not law but a corruption of it (ST, lallae, q. 95, a. 2).
DMHî Scotus
By contrast with Aquinas' emphasis on reason, Franciscan thinkers tended

to base moral values on the free will of God, limited only by the bounds of
logical possibility.
For Duns Scotus law and morality are primarily the product of the
divine will. The divine will is the first rule or rectitude.29 It is the cause of
good, and so by the fact that God wills something it is good. 30 But Scotus
also states that God cannot will anything that is not good (e.g. hatred of
himself) and that an act is naturally good (naturaliter bonus) when it is in
conformity with right reason (recta ratio). The Ten Commandments are
divine precepts that condemn what is wrong also by the law of nature.31
Scotus distinguished between immutable moral laws that are self-evident
and moral laws that are merely in accordance with the primary principles
but are not absolutely necessary. The first group belongs to the natural law
in the strictest sense and is contained in the first two commandments which
are concerned with the very notion of good and of God. The second,
which are provided by the commandments of the second Table from
Sabbath observance onwards, may be dispensed with by God since they
regulate the conduct of human beings in particular ways that admit of
exception and variation. Thus, the Old Testament incidents relating to
Abraham's duty to slay Isaac, the polygamy of the Patriarchs, the spoliation
of the goods of the Egyptians and the fornication of Hosea illustrate the fact
that the secondary precepts are not necessary but arise from the will of
29. 'Voluntas divina quae est prima régula omnium agibilium et omnium actionum, et actio divinae
voluntatis, ex quo est prima régula, est prima rectitudo', Reportata Parisiensia, lib. IV, dist. XLVI,
qu. IV, n. VIII.
30. 'Voluntas divina est causa boni et ideo eo ipso quod vult aliquod, ipsum est bonum', Reportata
Parisiensia, lib. I, dist. XLVIU, qu. unica.
31. 'Omnia peccata, quae sunt circa decern praecepta, formaliter non tantum sunt mala quia
prohibita, sed quia mala, ideo prohibita, quia ex lege naturae oppositum cujuslibet fuit malum, et
per naturalem rationem potest homo videre, quod quodlibet praeceptum ex illis est tenendum',
Reportata Parisiensia, lib. II, dist. XXII, qu. unica. n. 3. On recta ratio see Opus Oxoniense, II, XL.
qu. unica, n. 3.
7U Ethics
God. 3 2 Aquinas had not allowed that the secondary precepts could be
dispensed with; cases of apparent dispensation were cases in which the
circumstances were so altered that the precept no longer applied. But
Scotus saw that the precepts relating, for example, to the inviolability of
private property and the wrongness of stealing admitted of legitimate
exceptions and are not self-evident. They depend on the divine will and
although they are reasonable, they are neither necessary nor strictly of the
natural law. For God is free to command anything except that which con-
tradicts itself.
William Ockham
Ockham considerably developed the implications of the 'voluntarist ethics'
of the Franciscans. Perhaps he meant to exaggerate when he supported his
view that God could have sanctioned anything that falls short of logical
contradiction by saying that he could have commanded even hatred of
himself.33 But he emphasised that God is not under any obligation to do
anything; whatever he wills is by this fact just. 3 4 All obligation has its
foundation in divine command and in man's dependence on God. 'Evil is
nothing else than to do something when one is under an obligation to do
the opposite.' 3 3 Yet Ockham did not abandon natural morality or the
natural law. In his Commentary on the Sentences where this radical 'volun-
tarism' finds its expression, Ockham also wrote, like Scotus, that 'every
right will is in conformity with right reason' 3 6 and that 'there is no such
thing as moral virtue or virtuous action unless it is in conformity with right
reason, for right reason is included in the definition of virtue in the second
book of the Ethics'.31 And in his political tract entitled Dialogus de imperio ac
pontificia potestate Ockham spoke of the ius naturale and distinguished it into
three modes. The first consists of absolutely immutable norms conforming
to the natural reason, e.g. the prohibition of lying and adultery. The
32. Opus Oxoniense, III, 37, qu. unica. Cf. Reportata Parisiensia, IV, d. XVII, qu. unica, scholium
primum.
33. In Sent. II, qu. 19P.
34. 'Obligatio non cadit in deum quia ille ad nihil faciendum obligatur', In Sent. II, qu. 5H. 'Deus ad
nullum actum causandum obligatur, ideo quemlibet actum absolutum potest sine omni malo
culpe causare et eius oppositum', In Sent. II, qu. 19P. 'Deus autem ad nullum actum potest
obligari; et ideo eo ipso, quod Deus vult, hoc est iustum fieri'. In Sent. IV, qu. 9E.
35. 'Malum nihil aliud est quam facere aliquid ad cuius oppositum faciendum aliquis obligatur', In
Sent. II, qu. 5H.
36. 'Omnis voluntas recta est conformis rationi recte', In Sent. I, dist. 41K.
37. 'Nulla virtus moralis nec actus... est virtuosus nisi conformis recte rationi quia recta ratio ponitur
in diiTinitione virtutis secundo ethicorum', In Sent. Ill, qu. 12NN. Cf. Aristotle, NU. Eth., 1107a.
See also Ockham, In Sent. Ill, qu. 12CCC.
second consists of ius naturale that may be legitimately modified by positive

law, e.g. the common ownership of property and universal liberty; this is
what man should uphold through natural equity if there were no customs
or institutions. The third consists of norms that may be deduced by evident
reason and by supposition from the ius gentium or from human behaviour,
e.g. that violence, being against the law of nature, may be lawfully
repelled by force. 38
The fourteenth and fifteenth centuries

Whether the nominalist movement of the fourteenth century resulted in a
growing preference by thinkers to move natural law closer to divine law
and to divine will is a question that has produced varying answers. Thomas
Bradwardine found no place for the natural law and forged no such link
between God's will and man's reason. O n the other hand Gabriel Biel did
not break radically with Aquinas and accepted natural-law teaching. Some
scholars have seen Luther's emphasis on the corruption of human nature as
the climax of the retreat by many thinkers at the end of the Middle Ages
from the 'classical' scholastic theory that natural law is different from the
law that comes from revelation; others have doubted the extent of such an
abandonment of natural-law principles. The general picture as regards the
fourteenth and fifteenth centuries is less clear than one might wish it to be.
The sixteenth century

However, in the sixteenth century there was a revival of discussion of the
natural law and a marked renewal of interest in Aquinas' teaching.
Substantial treatments of natural law doctrine were provided by Robert
Bellarmine, Gabriel Vâsquez, Dominic de Soto, Richard Hooker and
Francisco Suârez. Richard Hooker restated Aquinas' arguments regarding
the divisions of law in the first book of The Laws of Ecclesiastical Polity
(1594). Francisco Suârez in his De legibus ac Deo legislature (1612) likewise
returned to Thomas but provided as well (in Book H) a notable review and
critique of subsequent scholastic thinking from Ockham down to Vâsquez
in his own time.
Francisco Suârez
At the very beginning of his treatise Suârez seeks to establish that natural
law is truly law. He comments upon Aquinas' definition of law as a certain
38. Dialogus de imperio ac pontificia potestate, III, tr. 2, p. 3, c. 6.

7i6 Ethics
rule and measure in accordance with which one is induced to act or is

restrained from acting.39 This he criticises as too broad and general. It
appears to suit all creatures, not only those that are human and rational, and
it appears to embrace other actions than moral ones; it also includes
counsels which are clearly distinct from precepts. The expression 'natural
law' may broadly and metaphorically refer to the inclinations of all natural
things, as it seems to do in Plato's Timaeus and also here in Aquinas' Summa,
or it may just refer to all animate creatures as in the Roman law. But,
strictly speaking, things that lack reason are not susceptible to law and are
not capable of obedience; law is strictly a rule, not a working out or an
effect of a rule.40 Suârez accordingly attempts a tighter definition of the
natural law as 'a certain measure of moral acts in the sense that such acts are
characterised by moral rectitude through their conformity to law, and by
perversity, if they are out of harmony with law'. 41 The natural law is 'that
form of law which dwells within the human mind in order that right may
be distinguished from wrong'. 42
In Book II, c. V, of De legibus Suârez concentrates his criticism upon
Vâsquez and disagrees that natural law is right reason or the faculty of
rational judgement. While rational nature is the foundation of the natural
law and provides the measure or standard of objective right and wrong,
law is a more specific concept than that of a standard or measure. Con-
science too is a broader term than that of law and, moreover, conscience,
unlike natural law, can err (II, V, 15). Law has to be commanded; it requires
a lawgiver, and law and command are not provided by rational nature. If
natural law is rational nature, it merely indicates what should or should not
be done and is thus lex indicativa or demonstrativa. Suârez traces this view
back to Gregory of Rimini and to Gabriel Biel, who, he says (II, VI, 3),
appear to hold that natural law does not come from a lawgiving God and
does not depend on God's will. Since it consists of the dictates of right
reason and manifests the evil that resides intrinsically in evil acts (e.g.
lying), it would have the character of law (or at least of indicative law) even
if God did not exist. Suârez himself believed that only a preceptive law (lex
praeceptiva) which commands and prohibits is truly a law because law
39. Suârez 1971 -7, I.I, 1. Cf. Aquinas ST, lallae, qu. 90, a. 1 : 'Lex est quaedam régula et mensura
secundum quam inducitur aliquis ad agendum vel ab agendo retrahitur.'
40. Suârez 1971 -7, loc. cit. Cf. I, III, 5.
41. 'Mensura quaedam actuum moralium, ita ut per conformitatem ad illam rectitudinem moralem
recti, et si ab ilia discordent obliqui sint', Suire2 1971-77, I, I, 5.
42. 'Lex ergo naturalis propria quae ad moralem doctrinam et theologiam pertinet est ilia quae
humanae mend insidet ad discernendum honestum a turpi', Suârez 1971-77, I, III, 9.
cannot exist apart from a legislator or praecipiens whose will it imposes. On

the other hand, Suârez does not wish to subscribe to the extreme view,
which hefindsin Ockham, 43 and according to which natural law is simply
and entirely what is commanded or prohibited by God as the author and
ruler of nature. In saying that no act is evil except in so far as it is forbid-
den by God and that God can abolish or change the whole natural law,
Ockham and those who agree with him (Suârez cites John Gerson and
Peter of Ailly in De legibus, II, VI, 4) assimilate the natural law into the
eternal law which issues the divine commands and prohibitions. The role
of the natural reason is to reveal God's will to man.
Suârez thus provides an admirable appreciation of the divisions of
opinion towards the end of the medieval period. He proceeds in De legibus
(II, VI, 5ff.), to expand fully a via media that is based on Aquinas and
comprises three principal theses. The first is that natural law is genuinely
preceptive or prescriptive law. It is more than a knowledge of self-evident
principles because contravention of it is an offence against God's will and
mandate. Secondly, natural law is more than precept and prohibition.
God's will prescribes acts that are good or evil for rational nature. Human
reason demonstrates this natural law to mankind and shows as well that acts
contrary to natural law are also contrary to divine law. Thirdly, natural
law is truly divine law and God is its legislator. Suârez sees (De legibus, II,
VI, I4ff.) that the matter turns upon the admissibility of the hypothesis that
intrinsic goodness has to be sought and intrinsic evil has to be avoided even
if God were not to exist. He considers an argument put by Bartholomew
Medina and urges that, even if a preceptive divine law is put aside, an evil
action is still not only morally wrong but also a sin apart from its relation to
God because it transgresses the law of nature.
The remainder of Suârez' treatment of the natural law is not without
importance both in itself and in view of the light it sheds on the divergence
of teachings up to his own time. In De legibus, II, VII Suârez discussed the
scope of the contents of the natural law. He roundly affirmed that the
natural law embraces all moral precepts which are characterised by the
goodness necessary to right conduct. Natural law includes as well as self-
evident principles-both general (e.g. do good and avoid evil) and specific
(e.g. worship God; observe justice) — deductions from self-evident precepts
which may not be equally clear to all men (e.g. the prohibitions of
adultery, usury, and lying). Far from hesitating over the necessity or
43. Ockham, In Sent. II, qu. 19 ad 3, 4.

718 Ethics
properly natural-law character of these deductions, Suârez holds that the

truth of a principle becomes clearer in the conclusion; natural reason works
more through the specific conclusions that can be applied to individual acts
and virtues than through universal principles.
Suârez also provided a defence of the immutable character of the natural
law by distinguishing between intrinsic and extrinsic change and between
positive and negative precepts (De legibus, II, XIII). Intrinsically the pre-
cepts of the natural law cannot vary, but since they cannot cover all the
particular circumstances which may arise (e.g. killing in self-defence,
breaking a secret to safeguard a third party), the precepts may be adapted
extrinsically to fit altering circumstances. Moreover, many things (e.g.
common ownership of goods) fall under the natural law which are not
commanded by it. C o m m o n ownership would naturally occur, as would
nakedness unless man introduced a different arrangement; it was permitted
to man and private ownership was not prohibited (De legibus, II, XIV). So
the institution of private property is not a sign that man can alter the
natural law, for private property falls negatively under the natural law in
the sense that it is not prohibited by it. Natural law positively prescribes not
that all should own all but that all should be able to use what is necessary for
each; consequently, natural law prohibits such undue taking of property as
prevents reasonable use. Just as man cannot alter the natural law, neither
can God. Against Ockham Suârez argued that the precepts of the natural
law cannot be altered by God because they pertain to intrinsic goodness or
wrongness (e.g. God cannot command man to hate him). Against Scotus,
Suârez maintained that God cannot alter or dispense with any of the Ten
Commandments and he argued, like Aquinas, that the alleged instances of
dispensations in the Old Testament (e.g. the spoliation of the goods of the
Egyptians by the Hebrews) are apparent examples only since dominion
over these goods belonged rightfully to the Hebrews by divine judgement
(De legibus, II, XV).
Finally, and especially because Suârez is considered an authority on
international law, it may be noted that he firmly distinguished the ius
gentium and the natural law. Although, like the natural law, the ius gentium
is common to all men, Suârez sees a distinction between, on the one hand,
the conclusions which are drawn from the primary moral principles of the
natural law and which are themselves natural law, and, on the other hand,
the common unwritten customs of the nations and the detailed regulations
that govern relations between nations. These are part of positive, human
law (De legibus, II, X V I I - X I X ) .
The seventeenth century

Later in the seventeenth century Pufendorf came to denounce the
schoolmen's doctrine of the natural law. Hugo Grotius in his De iure belli ac
pads (1625) paved the way to the future by writing about natural and about
international law not as a theologian but as a lawyer. But it is clear that he
followed the later scholastics when he made his famous pronouncement that
natural law would retain its validity even if God did not exist. He built his
theory of laws independendy of theological assumptions but his definition
of the natural law is not revolutionary: 'The law of nature is a dictate of
right reason which points out that an act, according as it is or is not in
conformity with rational nature, has in it a quality of moral baseness or
moral necessity; and that, in consequence, such an act is either forbidden or
enjoined by the author of nature, God.' 44 With this Suzrez, and many
before him, would not have disagreed.
44. 'Ius naturale est dictatum rectae rationis indicans, actui alicui, ex ejus convenientia aut dis—
convenientia cum ipsa natura rationali inesse moralem turpitudinem aut necessitatem moralem,
ac consequenter ab auctore naturae Deo talem actum aut vetari aut praecipi', De iure belli ac pacts,
1,1,10,1.
χ
POLITICS
38
THE R E C E P T I O N AND
I N T E R P R E T A T I O N OF
ARISTOTLE'S POLITICS
The scope of this study

This study makes no pretensions to being a comprehensive survey of
medieval commentaries on the Politics; it is confined to a few of the known
commentaries on Moerbeke's translation, all dated to the late thirteenth
and the fourteenth centuries. Within this narrow range, it aims to discuss,
firstly, how close the earliest commentaries, those of Albert the Great and
of Thomas Aquinas with the Continuation by Peter of Auvergne, came to
explaining Aristotle's meaning accurately and, secondly, what motives
later commentators had in writing, what arguments they considered valid,
and how far, if at all, their own views can be inferred from what they
wrote. To make some useful comparison between the texts, I shall, after a
few introductory remarks on each commentary, concentrate on what each
scholar made of Aristotle's arguments for the proposition that the multi-
tude of freemen in a state should participate in its political life (Politics III,
1281b and 1282a). Because they recognised its importance, all the com-
mentators expanded, illustrated, or interpreted this passage in a way which
brings out their individual characters and sometimes their political ideas.
William of Moerbeke's translation

The exact date at which Moerbeke finished his translation of the Politics is
still uncertain, but it was probably around 1260. Moerbeke tried teerender
Aristotle's meaning without the slightest interpretation of his own - an aim
which he believed could be best fulfilled by translating word for word,
preserving the Greek word order, some Greek double negatives, and even
an occasional Greek term - e.g., 'epikeiea' - for which there was no exact
Latin equivalent. The last chapter of Book II is full of mistakes, and there
are some errors in Books VI and VII;1 but, in general, it is a very accu-
rate rendering. Unfortunately, accuracy is more than counterbalanced by
unintelligibility. So the first function of the earliest Latin commentators,
1. Newman 1887, vol. 2, p. xiv, note 1.

72 4 Politics
Albert, Thomas, and Peter of Auvergne, was to engage in the task which
Moerbeke had declined, to explain what Aristotle was saying, using
methods of exposition and philosophical terms familiar to late-thirteenth-
century scholars and students. Though not without their mistakes, the
commentaries are in general far more correct than incorrect. 2
Albert the Great's commentary

Albert's commentary has been tentatively dated to around 1265, 3 which
would make it the earliest written in the Latin West. It ends with his often-
quoted diatribe against the anti-Aristotelian party in the schools, whom he
compares with the slayers of Socrates; their criticisms he ascribes to laziness
and a desire to paralyse others into their own inertia. 4 But despite the
violence of its ending, and the vast storehouse of knowledge revealed in his
citations from other authorities, this commentary was not among his most
important or influential works. He adopts the normal method of literal
exposition, dividing each book into chapters (not the ones conventionally
adopted by later authors), stating the theme, breaking each chapter into
sections, stating the main arguments, then sub-dividing till he comes to
each individual phrase. But here, instead of paraphrasing loosely, as was his
normal habit, he keeps very close to the wording of the Moerbeke trans-
lation, changing only what is strictly necessary in order to make the phrase
intelligible. T o this, he adds comments on the grammatical constructions,
additional information on persons or places mentioned, the occasional
contemporary example of a phenomenon discussed by Aristotle or quo-
tations from other authorities (classical, biblical, patristic or Arabic)
making the same point as Aristotle. But there is comparatively little
interpretation of the text, and no chance to find out what he himself thinks
of it.
Given this method, what Albert makes of the passage on the participa-
tion of free men is in general predictable: by stating the main lines of
argument first, he makes Aristotle's aporetic style somewhat more tren-
chant, and certainly easier to understand; when he gets down to the
individual sentences, he renders them succinctly, usually accurately, and
without judgement on the merit of the line being pursued. He gives the
kind of assistance which might be expected of a good crib. But there is one
point at which his comment can only confuse the reader: where Aristotle
2. Martin 1951, p. 36.

3. Aquinas 1971a, p. A8.
4. Albert the Great i890-çd, pp. 803-4.
Aristotle's Politics: reception and interpretation 725
states that there is an element of truth in the view that the multitude ought
to be supreme (1281a), the Moerbeke translation renders this oportet dominos
esse liberos et muttitudinem civium. Albert, in explaining the multitudinem
civium, goes back to the definition given at 1278a: a citizen in the strictest
sense is he who shares in the offices and honours of the state.5 This
definition looks odd in conjunction with Albert's very next phrase, which
states that these men have no special standing. His misunderstanding robs
the passage of its radical import. Again, when Aristotle says that it is
dangerous to exclude the multitude totally from office and concludes that
they should have some deliberative and judicial functions, Albert explains
these words as applying only to citizens understood in an elitist sense,6
although he has just correctly explained Aristotle's opposition to their
holding office on ground of their folly. Despite Albert's very great repu-
tation, no later commentators followed him in this reading.7
The commentary of Thomas Aquinas and Peter of Auvergne

The second, and far more influential, commentary which sought to make
Moerbeke intelligible was that of Thomas Aquinas, continued by his
disciple Peter of Auvergne. Thomas' part, perhaps to be dated to his second
Paris regency, 1269—72,* covers Books I—III 6, and there are some, by no
means indisputable, signs that he knew Albert's work. 9 Peter's part, which
starts at the beginning of Book III,10 probably dates from 1274-90." He
certainly had access to Albert's commentary. Both include the same
amount of literal exposition, but are more willing to paraphrase. The
general level of accuracy in rendering individual phrases of the text is high,
though there are some errors - mostly of detail, and commoner in Books V
and VI, replete as they are with historical allusions unfamiliar to the
thirteenth century. More impressively, the summing up of arguments at
the beginning of each section is remarkably accurate - it is certainly
Aristotle's point of view which is being given.
Thomas' part has recently been edited, with a full introduction; no more
need be said of it here. But Peter's has been less intensively studied; and
5. Ibid., p. 259
6. Ibid., p. 260.
7. It is possible, however, that it is to be found in some of the fifteenth-, sixteenth- and seventeenth-
century commentaries, strongly influenced by Albert, discussed in Czartoryski i960, pp. 3-44.
8. Weisheipl 1974a, p. 381.
9. Aquinas 1971a, p. A i o .
10. Grech 1967.
11. Ibid., p. 55.
726 Politics
because his influence on later commentators was very powerful, it justifies

a more extended examination of his method than will be accorded to
others in this study. Unlike Albert, Peter is prepared to interpret his
text when he gets down to the detail of the argument and even to make
additions to it. The best known of these12 was made at III 1284b, where
Aristotle says that kingship is one of the true forms of government; Peter,
however, says that it is the best constitution. It may be that this reflects his
own opinion, though there can be no certainty on the point, and the fact
that he was a loyal subject of the French king will not prove it. It cer-
tainly does reflect his belief that all Aristotle's writings form a unity; he has
simply inserted here the statement Aristotle makes in Book VIII of the
Nicomachean Ethics (1108b). Another, much less controversial, addition
comes at the end of 1283 b; there he concludes from what Aristotle says that
the question whether the legislator who desires to make the justest laws
ought to legislate with a view to the good of the higher classes or of the
many is answered in favour of the many. 13 This is indeed what the text
implies. But Peter, by stating the implication, gives it greater significance
in the course of the argument than it has in Aristotle's text; and the impact
of this addition on later commentators was enormous. Walter Burley even
argues that this is the most important question in the whole of Book III,1* an
impression he could never have got from the original text alone; and since
it was an idea particularly congenial to medieval thinkers, no doubt they
were glad to find it in Aristotle. A third addition comes at V 1301b, where
Aristotle says that those who excel in virtue have the best right to rebel, but
are of all men least inclined to do so; Peter adds to his explanation of this
remark that, if the virtuous have a just cause and also the power to rebel,
they ought to do so, and they sin if they do not. ] s
Aristotle asks whether the principle that the multitude collectively are
better judges than the few good men applies to all multitudes, then says it is
not clear. 'Or rather, by heaven, in some cases it is impossible of appli-
cation; for the argument would equally hold about brutes; and wherein, it
will be asked, do some men differ from brutes?' (1281b). Peter fits this into a
neat shape; multitudes can be of two kinds: either bestialis, without reason,
individually or corporately, and therefore having no claim to a political
12. Daly 1968, p. 43; Thomas Aquinas 1951, p. 169, section 474.
13. Thomas Aquinas 1951, p. 164, section 461.
14. Oxford Balliol MS 95, f. 183", 'haec quaestio est principalis huius libri'.
15. Thomas Aquinas 1951, p. 247, section 714 - a remark hard to reconcile with the authoritarianism
attributed to Peter in Martin 1951, p. 39.
role; or one in which each man has some inclination to virtue (called by
later commentators multitudo bene ordinata), where corporately the whole
has more prudence than a small group of good men, and therefore a role in
political life is merited. The effect of Peter's tidying operation is un-
doubtedly to give greater prominence to the multitudo bestialis than it has in
Aristotle's argument, and to make quite explicit the fact that it is denied a
political role.16 But whether there is any substantial innovation here,
which can be ascribed to an anti-democratic bias,17 is far more doubtful.
Peter defines this multitude entirely in moral and rational terms, not social
ones. Aristotle cites with apparent approval Solon's system, in which the
people elected and called to account the magistrates. Peter specifies this as
the role of the multitudo bene ordinata, thus introducing a new and potent
element into medieval political thinking.
Peter also seeks to explain why Aristode is so hesitant in concluding that
the people, if not degraded, are probably better judges corporately than a
few good men (1282a). He cannot ascribe this, as a modern critic might, to
a slightly irritating stylistic habit; for him everything the Philosopher says
has significance. So the 'forte' ('perhaps') with which Aristotle introduces
his conclusion is ascribed to the fact that this conclusion does not apply to
all constitutions, and in particular not to monarchies, where the people
have no right to elect or punish but must simply obey their moral
superior.18
When Aristotle raises again the claim of the multitude to rule (1283b) on
the grounds that it is stronger than a few men, Peter uses this opportunity
to sum up the whole argument. He states that a multitude which is made up
of some wise and prudent men and others who, though not wise in them-
selves, can be persuaded to accept reason, ought to rule, in the sense of
electing and punishing rulers, because such a multitude has the two re-
quirements for rule, both knowledge of how it should be done, and (in
virtue of its numbers) the power to repel enemies.19 By putting together
two pieces of argument which are separated in the text, and thus stressing
that power is a necessary part of the popular claim to be involved in the
state, Peter has in effect strengthened Aristotle's case for popular
participation.
These three additions to the passage all seem to be inspired simply by a
16. Thomas Aquinas 1951, pp. 151-2, section 427-31.

17. Grignashi 1966, p. 83.
18. Thomas Aquinas 1951, p. 152m, section 435.
19. Ibid., pp. 161-2, section 549.
728 Politics
desire to bring out the inner logic of the argument, for its own sake and not
for the sake of putting a different political complexion on Aristotle's
words. A disciplined approach is imposed upon a notoriously undiscip-
lined text, sometimes successfully, in that underlying implications are
made plain, sometimes forcedly, in that a phrase in an aporetic argument is
burdened with rather more weight than most modern critics would think
it could stand. But there is no serious distortion of Aristotle's argument.
The effect of the Thomas - Peter commentary

With the appearance of the Thomas-Peter commentary, the job of render-
ing the Moerbeke translation intelligible was done, and it did not have to
be done again. Many, if not most, other medieval commentators simply
plagiarised it wholesale. But the other commentators considered in this
study managed to contribute something distinctive of their own. The first
three, Guy of Rimini, Walter Burley and Nicole Oresme, used the literal
form, though in very different ways; the other three commentaries used the
question form.
Guy of Rimini's commentary

Guy of Rimini's commentary20 written in the early years of the fourteenth
century, is presented in a simpler style than either Albert's or that of
Thomas—Peter; it is generally free from scholastic distinctions, uses familiar
thirteenth-century words, and has some rhetorical flourishes, along with a
few light-hearted digressions. Unlike Thomas-Peter, Guy regularly
points out the differences between what Aristotle is saying and what
orthodox Christians believe — he is, for example, the only commentator to
say that the doctrine of natural slavery looks harsh to Christian eyes21 -
and, like Albert, he uses contemporary examples to explain what Aristotle
means, Lombard tyrants, Italian overseas trade and the organisation of
chanceries being just a few of his illustrations. All this suggests that he is
aiming his work at a fairly unscholarly readership, an impression which is
proved by an incidental remark he makes while discussing Aristotle's
criticisms of Plato.22 One of the manuscripts in which Guy's commentary
survives (Venice Marciana 2492), is a beautiful and rather extravagant
20. For some discussion of this, see Aquinas 1971a, p. A14.

21. MS. Venice Marciana 2492. f. 87". 'Sed ut hoc videtur crudele dictum et contra naturalem amorem
qui debet esse inter homines et contra doctrinam et pietatem Christianae religionis, quae praecepit
quod unusquisque proximum diligat sicut seipsum
22. 'Hoc opus fàcimus propter eos qui non multum vel nihil de profundis cognoscunt.' f. 124'.
production, far removed from the usual cramped, messy volumes which
circulated in the university world, and in itself suggestive that Guy's work
was regarded as of more than academic importance. But his work was used
by scholars, too: it forms the basis for an eccentric commentary on Books I
and II of the Politics by the canonist John of Legnano,23 and extracts from it
were used in the form of scholia in manuscripts of the Moerbeke text.24
In his discussion on popular rights in political life Guy simply follows
Peter, including all three additions to the text.
Walter Burley's commentary

Walter Burley's commentary, written about 1338-9" and based on
Thomas-Peter, is a brief paraphrase. Each book of the Politics is introduced
by a list of the questions Aristotle discusses there; each has an index of
headings at the end; and there are occasional diagrams to clarify the
argument. The mere imposition of such a format on the Politics gives it a
novel character; and the assumption that it is a series of questions with
solutions leads Burley to greater clarity than Aristotle in providing the
answers. A condensed summary of this sort must surely have been a
handbook for students; that they were mainly English students is made
plain by the examples.26
Burley's reliance on Peter is immediately evident in his discussion of
popular rights in political life. According to Aristotle, he says, 'a bestial
multitude should in no wise rule (principari), but a multitude of men who
have virtue, even if imperfectly, and an inclination to virtuous actions,
ought to rule, for the reason given above [that is, that corporately they may
have more virtue than a few virtuous men]. However, it does not seem that
this multitude ought to rule with supreme authority (principari summo
principatu) because of its lack of prudence, which leads to poor judgement,
and its injustice.'27 Burley follows Aristotle in arguing that when the
23. Contained in MS Venice Marciana 2653, ff. 117'—130·. I am grateful to Dr Maurice Keen for
telling me of the existence of this commentary.
24. See for example MSS Phillips 891, contained in British Library Photocopy R..P. 60, which contains
Guy's introduction, then the Moerbeke translation, with marginal notes taken mainly, though
not exclusively, from his commentary.
25. Martin 1964, p. 225.
26. The dedications, first to Richard de Bury and then to the Pope, do not seem to me to give more
than a partial indication of whom he was writing for.
27. MS Balliol 95, f. 182': 'Solvit quaestionem, intendens quod multitudo bestialis nullo modo debet
principari; sed multitudo hominum qui habent virtutem quamvis imperfecte et inclinationem ad
actus virtuosos debet principari propter rationem superius dictum. Talis tamen multitudo non
videtur quod debet principari sumo principatu propter imprudentiam ad non recte iudicandum
et iniustitiam.'
730 Politics
magistrates, counsellors, judges and leaders of the assembly (contionatores)

are incorporated into the multitude, that multitude is sovereign in the
state.28 Then, following a line of thought very likely incited by Peter's
exclusion of monarchy from the constitutions described by Aristotle in this
passage, Burley produces an original, though enigmatic, comment: 'It is to
be understood that in true constitutions, other than monarchies, a multi-
tude rules, and this means many men; and in addition, in a kingdom, a
multitude made up of the king and the nobles and the wise men of the
kingdom in some way rules. ' 2 9 He then cites the English parliament, called
to expedite important business, as an example. It is difficult to know what
to make of this. It may be that Burley read this whole passage in the Politics
as an argument for power-sharing and broadly-based government rather
than for the participation of the man in the street; but it is also possible that
he believed parliament to be in some way representative of the people,
offering them participation, though very indirectly, in the ordo principatus
which is the life of the state. But Burley's terse style leaves his real meaning
a mystery.
Nicole Oresme's commentary
Both Guy of Rimini and Walter Burley had as their main aim in com-
menting the desire to render the Politics easily understood, though by
rather different groups of people. Nicole Oresme, writing around 1371,30
only partially shared this aim, because, for the most part, he had made the
task redundant by translating the text into straightforward French. Con-
sequently he has time to expand on Aristotle and to relate the text to the
world as he knows it. He is the only one of the commentators whose own
political views emerge without equivocations. When he misinterprets
Aristotle, it is not because medieval political preconceptions blur his
understanding, but because it suits his book to do so. One of his more
startling uses or abuses of the text comes when Aristotle talks of the
supremacy of law; Oresme applies this passage to the church in his own
day, arguing that it would be well if the papacy could see itself as bound by
28. Ibid., f. i82r: 'Magis videtur conveniens quod multitudo principatur quam unus vel pauci
virtuosi, et hoc loquendo de multitudine quae comprehendit in se consiliarios, iudices, con-
tionatores et alios prudentes.'
29. Ibid., f. 182': 'Intelligendum quod in rectis principatibus aliis a regno principatur multitudo, et
haec est plures; et adhuc in regno multitudo constituta ex rege et proceribus et sapientibus regni
quodammodo principatur.'
30. Menut 1970, p. 19.
law, instead of priding itself on its plenitudo potestatis.31 In the same way,
Aristotle's teaching on the proper distribution of offices is applied to
ecclesiastical benefices;32 and his comment that the priesthood ought not to
be given to husbandmen or mechanics is turned into a criticism of the
friars.33
On popular participation in political life, Oresme begins in the conven-
tional way by explaining the text; but at the end of the first section, he adds
a gloss of his own: '[Aristotle] means, it seems to me, firstly, that it is better
for a few men to have the sovereign authority, though not all lordship;
then, that another multitude, if it is not bestial, should have rights or
authority to counsel or judge, as is said; and thirdly, that everyone should
have a voice in the election or correction of the princes and their deeds.' 34
The distinction between the multitude which counsels and the whole
people which elects is unique to him. In his next excursus he declares that
what Aristotle says on elections may be sensible for aristocracies, but does
not relate to monarchies; he echoes Peter in interpreting Aristotle's 'per-
haps' as limiting the application of the remark. But he goes far beyond
Peter in aruging that election is in itself undesirable for three reasons: even
where the multitude is well-ordered at the time when the power to elect is
given to it, it may degenerate over the years; even if it does not degenerate,
the process of election can cause discord; and even a rational multitude can
be seduced into false decisions.35 In the past open election of kings may
have been a good thing, now hereditary succession is expedient.36
Aristotle argues that the master of the house is a better judge of it than
the builder. Oresme notes that in a book entitled Defensor Pads this
argument is invoked to demonstrate that human law should be made,
promulgated, corrected, or changed by the authority and consent of the
whole community or of its weightier part.37 Thus, he says that Aristotle
31. Ibid., pp. 159-60.
32. Ibid., p. 139.
33. Ibid., p. 307.
34. Ibid., p. 135: '11 veult dire, ce me semble, premièrement que ce est mieux que un peu de gens
vertueux aient non pas toute la seigneurie, mes la souveraine dominacion. Item, que l'autre
multitude, se elle ne est bestial, ait dominacion ou auctorité quant a conseillier et a jugier comme
dit est. Item, que tous ensemble aient voies en election et en la correction des princes et de
leur fais.'
35. Ibid., p. 136.
36. Ibid., p. 109.
37. Ibid., p. 137: 'En un livre intitulé Defensor Pads ceste raison est alléguée a monstrer que lays
humaines positives doivent estre faictes, promulguées, corrigées ou muées de l'auctorité et
consentement de toute la communité ou de la plus vaillant partie.'
732 Politics
would have regarded the Romans as having behaved bestially in handing

over to their emperor the right to legislate. 38 In a later context, he goes on
to remark that he sees no difference between arbitrary rule without the law
and rule according to laws made only by the governors, without the
consent of the multitude, and not in accord with public good. 3 9 Oresme
has twisted Aristotle on popular participation into a statement on the
legislative supremacy of the people. The only question is, w h o m does he
have in mind when he talks of the consent of the multitude?
Oresme's multitude is clearly aristocratic. In Book VI, when discussing
sovereignty, he says that the body which is sovereign in polities and
aristocracies is the multitude, 'not the populace, but the multitude or
universal congregation of all princes or office-holders and of the principal
citizens'. This he compares with the assembly of the Masters in the
University of Paris. In monarchies, he says, the king and his council are
only a small part of the political multitude who ought to consider and
ordian where the public good lies; by implication, all office-holders and the
principal citizens have rights here too. He goes on to argue for the rights of
a similar multitude in the government of the Church, in other words for
the conciliarist principle. 40 Oresme's multitude is clearly an aristocratic
one of properly qualified men, to w h o m rights of counselling and judging
can entirely safely be given. In a work undertaken at the request of the king
of France, and dedicated to him, written in a language easily understood by
those men who stood most to gain by using his arguments, Oresme is
demanding a radical programme of reform, both in church and in state.
Peter of Auvergne's Questions on the Politics

The use of the question form (in place of the literal commentary) allows the
master w h o employs it a greater freedom to select topics for discussion and
to offer his own conclusions. But, as in the case of the literal commentaries,
plagiarism was the rule; and again, the key commentary was by Peter of
Auvergne. Peter's Quaestiones*1 cover the first seven books of the Politics, a
hundred and twenty-six questions in all, with notably wider coverage on
Books I and III than the others. In arguing, he employs a tight syllogistic
38. Ibid., p. 138.

39. Ibid., p. 178.
40. Ibid., p. 274: 'La multitude non pas la populaire mes la multitude et congregation universele de
tous les princeys ou offices et des principalz citoiens.'
41. I have used MS Paris Nat. Lat. 16089, ff. 274'-3igr. I have not been able to compare it with MS
Bologna Bibl. Univ. 1625, which is a shortened form of the Quaestiones, but with a little new
material.
form, applies learning culled from other Aristotelian works to the elucida-
tion of the Politics, and relentlessly drags out the implications of Aristotle's
tentative statements. He also echoes a number of the themes which occur in
the political writings of Thomas Aquinas, whose disciple he was.
On popular participation, Peter first asks whether it is better for the
whole multitude or for a few virtuous men to rule. In favour of the few
virtuous, he argues that it must be good for the state to be ruled by those
who further the purpose for which it was instituted, the life of virtue. Also
since that state which most closely approximates the best is better, and since
aristocracy approximates to kingship, which is the best, aristocracy must
be better than rule by the multitude. In favour of rule by the multitude,
Peter says that they are corporately more prudent than the few good men
and that they aim at the common good. To rule, however, it is necessary to
have not only both prudence and virtue per se, but also power per accidens,
on Peter's view; and this leads to his solution: multitudes are divided into
the bestial and the mixed, in which there are some wise men and all the rest
can be persuaded to reason. Bestial multitudes are deprived of all right to
rule, but mixed ones ought to do so, in the sense that they should elect and
punish princes. Such a multitude has all three requisites for rule, whereas
the few virtuous men have only two, lacking the power which comes from
numbers.42
Peter then asks if the multitude should rule with supreme authority. 43
He argues against this, on the ground that it has prudence and virtue only
in respect of the wise men among it, who are like the heart to the whole
animal, and this is not enough for the task of holding office. And secondly,
the multitude can rule only if it is unanimous, which means it can rule only
if it is like one man; therefore rule by one man is better. One man of
prudence and virtue elected to office by the multitude, would obtain the
necessary power from the fact of his election, and would thus combine all
that was needful.
Then comes the problem of whether the multitude should have the right
to elect the princes and call them to account. In the course of his discussion,
42. MS Paris Nat. Lat. 16089, f· 295'"· 'Haec est multitudo bestialis et nata subesse principatu despotico.
Alia autem est multitudo bene persuadibilis, mixta ex sapientibus et ex vulgaribus bene per-
suadibilibus... Dicendum quod multitudinem vilem et bestialcm non cxpcdit ad istam [clcc-
tionem] attingere: multitudinem tamen mixtam et ordinatam iustum est attingere ad
istam ... quia clcctio principis duo requirit, scilicet consilium de principe bono investigando et
pocentiam ad cogendum clectum.'
43. Ibid., f. 295*: 'Utrum oportet multitudinem principari principatu maximo in civitate bene
ordinata.'
734 Politics
Peter points out that he is talking only of cities which are wholly in-
dependent, since, if there is a superior authority, it will be up to it to elect
and correct the prince (a point which recalls the argument of Aquinas in De
regno, Ch. 6). Peter easily shows that the multitude ought to elect, in
virtue of its corporate prudence and its power. Because men love what they
themselves have made, they will more easily obey a prince whom they
have elected. Equally, the multitude ought to punish a prince who fails in
his duty: punishment ought always to be inflicted by those whose action
will breed the least resentment in the punished — which is the multitude
(presumably because its great numbers make it an impersonal force).44
The Milan commentator

Whatever may have been his own political preferences, Peter in writing
the Quaestiones was aspiring to the detachment which is so marked a
characteristic of his master's work. Such detachment, however, did not
attract others. An interesting commentary, dating from the late thirteenth
or the early fourteenth centuries, and clearly closely related to Peter's
Quaestiones, is to be found in the MS Milan Ambrosiana A ioo Inf.
(ff. i r ~54 r ). The author asks only ninety questions; but they are almost
identical with those asked by Peter, and many of the arguments are the
same. Because this anonymous author and Peter are usually so similar, their
differences on popular participation in political life are all the more
interesting.
In discussing how the multitude ought to participate, the Milan com-
mentator points out that it must be a very varied body; just as the animal
needs many parts to be effective, so the multitude must be made up of
many different kinds of men. Some of them will be naturally good, others
can be persuaded to goodness, and on others no form of persuasion will be
effective.45 He excludes the last category from political rights, calling them
multitudo bestialis et servilis et non persuadibilis aliquo modo ad virtutes; uniquely
among the commentators, he offers an example of such a group, a crowd of
mechanics, presumably because they are corrupted by their work. 46 But
all others are comprehended within the multitude which deserves political
44. Ibid., f. 296': 'Ergo maxime expedit eligere quod in se habet consilium et potestatem; sed haec est
multitudo; quia per sapientcs partes sui habet prudentiam, per se autem potential«; et iterum
homines diligunt sua opera ut filios... et multitudo principi quem eligit magis obediet. Item ad
corrcctionem dum princeps peccaverit, exigatur discretio penam inveniendi et potentiam earn
infligandi. Haec autem conveniuntin multitudine.'
45. Taken from the Ethics, Book X, 1179b— u8oa.
46. MS Milan Ambrosiana A 100 Inf. f. 28'.
rights. As a consequence, the commentator departs in form from Peter,

allowing such a multitude the right to wield supreme authority (principari
summo principatu), but only in the sense that the few prudent men will
actually hold office, while all the rest participate in some way, either in
counselling or electing.47 The Milan commentator even tells us how the
elections should be organised, and in so doing, he passes from the abstract
sphere of Peter to concrete political facts. Elections are out of place, he
observes, in principalities where hereditary succession is normal; but where
elections do take place, they should not be democratic in the full sense.
Representatives should be chosen from the principal families and crafts
(artes) to choose the prince; then they should present their candidate to the
whole people to be accepted; but the people have the right to reject him if
one of their number makes out a case against him. The process is defended
by citing the maxim of Roman law that what affects everyone ought to be
decided by everyone (quod omnes tangit ah omnibus debet pertractari).48
The blend of a representative system with popular right to acclaim or
veto is compatible with Aristotle's intentions in this passage, but it is far too
concrete a notion to emergejust from the reading of the text. Our author is
using the text to urge a particular policy in circumstances which our
ignorance of his identity and background hide from us. The passage stands
out in the Milan commentary as extraordinary; elsewhere the author exhi-
bits the same attitude towards the text as Peter does, one of detachment and
philosophical, rather than political, reflection.
John Buridan's Questions on the Politics

If the discussion of popular participation is a particularly good one for
bringing out the individual character of the Milan commentary, it is not
suitable for the Quaestiones of John Buridan,49 the teacher of Nicole
Oresme. When Buridan turned to the Politics at some unknown date in the
middle of the fourteenth century, he used it to provide an extremely loose
47. Ibid., f. 28': 'Ad hoc dicendum est quod expedit multitudinem principari in maximo principatu
intelligendo quaestionem illo secundo modo, sicut dicebatur, ita quod aliqui virtuosi principentur
in tali multitudine simpliciter et dirigant alios in finem determinatum talis principatus, et alii
habcant aliquod principatum sicut consilium vel principentur aliquo modo exsequendo vel in
eligendo alios qui debent simpliciter principari.'
48. Ibid., f. 28e: 'Ulterius est intelligendum quod non oportet sic multitudinem eligere principem
quod quilibet existens in civitate habeat vocem in eligendo principem; sed sic quod unus de una
tota progenie et in istis artibus quae principaliores [sunt] habeat vocem in eligendo, et non
omnis... isti sic deputati a tota multitudine ducunt ilium quern elegerunt praesentare multi-
tudini, videndum utrum placeat multitudini, et si non placet, potest electio impediri.'
49. John Buridan 1640.
736 Politics
framework for discussing his own views on moral and legal questions,
showing a particular interest in the proper processes for law courts and
judgements, which he introduced into the text with very little justification
from Aristotle. It is these parts which would best exemplify his com-
mentary. Yet, for our purposes it is worth looking at what he says on the
topic here selected for discussion. 50
Buridan asks whether it is permissible for the multitude to elect a prince
and afterwards to punish him if he errs. The arguments put against the
multitude's doing so are that prudence is necessary for election, but multi-
tudes often lack it; that rules of expediency do not apply to serious matters,
of which this is one; that the right to elect or punish rulers confers on the
multitude a form of sovereignty which vile persons ought not to enjoy
over their superiors; and finally that election encourages division. The only
point made in favour of the multitude at this point is that Aristotle believes
in it. Buridan then offers six notabilia, five of which clearly derive from
Peter's Quaestiones. They are that a true prince needs three qualities, virtue,
prudence, and power; that multitudes can be divided into bestial (constituta
ex hominibus vilibus) and well-ordered (constituta ex potentibus viris, virtuosis
et sapientibus); the bestial can be divided into the totally sensual, who cannot
participate in reason, and the persuadibiles, who can be brought to it; the
body which is to elect a prince needs to have both prudence and power,
since it may need to force a good man to accept office; and a prince elected
by the multitude will be loved by it. The sixth point, his own, is that a good
prince is an interpreter of the law; if he is ignorant of the laws, he is not
good; and if he knowingly despises them, he ought to be punished or
deposed. This leads him to his own conclusions: that the bestial multitude
which, in his argument, though not Peter's, includes the persuadibiles,
should not elect or punish; but the well-ordered both can and should, since
it has prudence, love of the community and strength of numbers; such an
election will serve peace and concord, and these will be preserved by
allowing the multitude to punish the prince if he strays from rule according
to the law. All the arguments originally put against the conclusion apply
only to bestial multitudes. 51
It is striking how Buridan is content to copy almost verbatim from
Peter, a man who ought to have been reckoned among his philosophical
opponents. What is less clear is whether Buridan has added a new dimen-
50. See Grignashi i960, p. 138.

51. John Buridan 1640, pp. 140-2.
sion to the argument in his description of the make-up of the two multi-
tudes. He is certainly more exclusive than Peter in including the per-
suadibiles among the bestial, though he makes nothing of this in his
conclusion. But it has been claimed that by describing the bestial multitude
as made up of vile men, and the well-ordered of the powerful, Buridan was
aiming to limit political rights to the nobility and haute bourgeoisie.52 This is
possible; but Buridan's words are susceptible of another interpretation.
And the word 'viles' he applies to the bestial multitude may be a moral
rather than a social adjective; even if it is social, it is as likely in a medieval
context to mean servile as common, and thus preserve a genuinely
Aristotelian distinction.
Conclusion
In conclusion, it is worth emphasising what has been the chief finding of
this study: though the medieval commentaries on the Politics have a
common source, they still differ in their approach, partly as a result of the
needs of the readership for which they were written, partly because they
follow different academic conventions, and partly because some show a
genuine originality of mind.
52. Grignashi i960, p. 138.

39
RIGHTS, NATURAL RIGHTS,
AND THE PHILOSOPHY OF LAW
Rights derived from above: John ofSalisbury

In the period from John of Salisbury to Richard Hooker and Francisco
Suârez the concept of a right and the theory of natural rights emerged from
a religious view of society with which the subsequent politics of rights has
more or less willingly dispensed. At the outset sacral kingship and, more
convincingly, the authority of the Church, especially the papacy, claimed
divine warrant and support. At the same time those who wished to resist
their superiors could usually allege violation of mutual obligation, or
failure to conform to the requirements of rulership, and Christian impulses
to condemn, flee, or find a radical alternative to ordinary worldly life were
always active.
John of Salisbury's Policraticus, although a highly personal work, shows
important aspects of the original view. John regards rights (iura) as the vital
means which an ideal court (an Areopagus) would give each class or
profession in a community as required to perform its proper functions
(Polier. I.3) - functions in an organic social whole, which has as its soul the
priesthood and whose princely head of government is an earthly image of
the divine majesty. ' John draws his organic model for society from classical
sources but adapts it to medieval Christian needs by making the sacerdotal
soul a distinct class, of which the prince is in some sense a minister. 2 In the
same way, he takes over the R o m a n jurists' impersonal definition of
fairness or equity as a rational equilibration of disparate things with respect
to the same laws (iura), bestowing on each what is his own, but identifies
such aequitas with the justice of God, to which rulers are emphatically
subordinate. 3 Rights, in his view, are bestowed on their possessors from
above, that is, from God or from the exigencies of the organic social whole.
They are not in theory arbitrarily determinable by government, however,
1. John ofSalisbury 1909, IV.i, VI.25.

2. Ibid., IV.3; V, pro!.. 1-2, 6; III.i.
3. Ibid., IV.2-7.
Concept of a right and the theory of natural rights 739
and in John's case corporatist principles are compatible with vigorous

individuality. The potential for a theory of natural rights is clearly present.
Virtue as the determiner of rights: Thomas Aquinas

With some exceptions (for example, Roger Bacon's Opus maius) the
philosophically significant political thought of the thirteenth and four-
teenth centuries was part of a struggle within Christian society to accom-
modate, control, or exploit the naturalistic resources of Aristotle's Politics.
According to Thomas Aquinas, an individual man is a part of the com-
munity in all that he is and has; hence his proper good is subordinate to the
more perfect common good.4 Determination and achievement of both
public and private good take place, however, through the virtues of
charity, prudence, and justice. In his complex analysis of these virtues (ST,
lallae) as well as in his treatment of law (ST, lallae, qq. 90-108), Thomas
provides both inspiration and criticism for the attempt to base politics on a
theory of persons and their rights. A right in the sense of an action in some
sense owed to another person, is the object of the moral virtue justice,5
while the discernment of rules of action by which right or just relationships
among persons and things may be achieved is the work of another cardinal
virtue, the intellectual virtue of prudence.6 An individual's concern for the
good of the community is supported by the theological virtue of charity
(discord, strife, and sedition are vices7 opposed to peace, which is an effect
of charity) and is also a matter of rational prudence,8 but the community in
question is primarily a unity of right (or law) and common utility, not
simply an existing power structure. Hence, the tyrant is seditious, not those
who resist him, for it is he who violates this unity.9 With respect to the
principles of justice we are subject to God, not mediately, through a human
superior, but immediately, for God instructs each man by divine and
natural law.10 Again, both the justification of private property rights and
their limitations turn on considerations of rational communal utility more
basic than positive titles. Appropriation can indeed be justified, but only
with respect to care and responsibility for material things, not their use.! '
4. ST, lallae, q. 96, a. 4; q. 90, a. 2.

5. ST, Ilallae, q. 57, a. i;q. 58, a. I.
6. ST, Ilallae, q. 47, a. 6.
7. ST, Ilallae, qq. 37, 41-2.
8. ST, Ilallae, q. 47, a. 10.
9. ST, Ilallae, q. 42, aa. 2-3.
10. ST, Ilallae, q. 104, a. 5, ad 2.
11. ST, Ilallae, q. 66, a. 2.
740 Politics
Accordingly, theft from necessity does not violate the commandment

against stealing (although open robbery with a threat to life does). 12
Thomas' account of the objective order of justice yields a rather short list
of universal human rights. Because all men are equals in nature, all have
rights at a fundamental natural level, such as the right to contract marriage
or vow virginity. 13 Because human law concerns only outward acts, the
mind, as Seneca says, is free; 14 this freedom must be respected even when it
is a matter of accepting or not accepting Christianity. 1S The right to active
participation in political affairs is not universal. It is appropriate for a
morally responsible people, but a corrupt people is more suitably ruled by
others. 16 Thomas thought the former situation more desirable than the
latter, holding that the best political regime (as exemplified in the Old
Testament) is one 'well mixed' from royal, aristocratic, and democratic
elements. 17 Even on an occasion (De regno I) when he extolled the merits of
pure kingship and deplored tyrannicide, he tied this to an emphasis on
virtuous subjects as the goal of good kings and the chief fear of tyrants. In
every case, then, the powers and virtues of persons are central in Thomas'
political thought, but of persons seen in relation to other persons, including
God, not in isolation.
Powers and rights: Fourteenth-century papalism and its radical opponents

Both the term 'plenitudo potestatis' and the papal Rechtsidee associated with it
have a long history in the earlier Middle Ages. 18 The most philosophically
systematic expositions of this view, however, were those presented in the
early fourteenth century in the face of new theoretical, institutional, and
ethical challenges by such writers as Giles of Rome, James of Viterbo, and
Augustinus Triumphus.
Augustinus Triumphus
Augustinus' work, for example, is a Summa de ecclesiastica potestate, a
consideration of the pope's power, first in itself,19 then in relation to the
acts of temporal and spiritual dominion or lordship (dominium) for which it
12. ST, Ilallae, q. 66, aa. 7-8.

13. ST", Ilallae, q. 104, a. 5.
14. ST, Ilallae, q. 104, a. 5.
15. ST, Ilallae, q. 10, a. 8.
16. ST, lallae, q. 97, a. 1; a. 3, ad 3. Ilallae, q. 47, a. 12; q. 50, aa. 1-2.
17. ST, lallae, q. 95, a. 4; q. 105, a. 1.
18. Ullmann 1966a, 1966b, 1970.
19. qq. 1-34.
is ordained,20 and finally in relation to the graded perfections of insti-

tutional status accruing to men from such power.2 ' Just as God cannot
deny Himself to be lord of all, so His vicar cannot exempt anyone from
papal jurisdiction, cannot deny that he has universal dominium. To do so
would be to fall into Manichaeism.22 Accordingly, all other rights are
derived from God to men through the pope, and it pertains principally to
him to maintain others in these derived rights.23 No one else could judge
the pope, but a heretical pope (a possibility often discussed in earlier
canonistic literature) judged, and deposed, himself.24 Augustinus also
incorporated into his treatise on papal power ideals of natural law involv-
ing no particular reference to the Church. His enumeration, under the
heading of theft, of various forms of injustice perpetrated by secular lords
against their subjects is an example of this.25
Marsilius of Padua
Early-fourteenth-century opponents of papal claims to comprehensive
secular power included such distinguished Aristotelian successon of
Aquinas as John of Paris and Dante, both of whom based the autonomy of
secular governments on the rational and moral competence of human
nature, but by far the most radical, if not the most Aristotelian, anti-
hierocratic theoretician was Marsilius of Padua.26 For Marsilius, more than
for any earlier thinker, the natural process of earthly politics had a self-
sufficiency which not only needed no completion or rectification from
higher sources but could be represented as mortally endangered by such
interference.27 The normal human desire for a peaceful life must itself be
allowed to generate the means necessary for its realisation. That is, the
people as a whole must consent to the power of their rulers and, above all,
popular consultation and consent must determine the laws by which
government is to operate.28 The government and laws thus established will
in turn establish and control such other offices or parts of the state (includ-
ing the priesthood) as are necessary for the healthy existence of the com-
20. qq. 35-75.

21. qq. 76-112.
22. q. 6 l , aa. 2-3.
23. q. 1, aa. 3, 7-8; qq. 44-6; q. 75, a. 1.
24. q. 5, a. 1; q. 22, a. 1, <nf 2; q. 1, a. 2.
25. q. S4, a. 4; cf. q. 26.
26. Gewirth's comparative method makes his penetrating study of Marsilius (Gewirth 1951) an
exceptionally valuable account of major aspects of political philosophy in the period.
27. Defensor pacts, Dictio I, cc. 4, 9.
28. Def pac. I.9, sections 5-9; 1.12.3.
742 Politics
munity. 2 9 Should the government signally or persistently fail in its mission

of implementing the popular will, the people may correct or replace it. 30
If Marsilius so directly asserted the natural right of the people as a whole
to control its own political situation, his contribution to the theory of
individual rights was both less direct and less affirmative. While his de-
finition of the right of ownership, for example, as a power of legal control
over a thing 3 J may mark a stage in the doctrine of'subjective right' central
to much modern jurisprudence and political theory, 32 Marsilius did not
regard such rights either as based on natural law or as the building-blocks of
politics. His interest in them was negative: he held a lack of them to be the
hallmark of a genuinely Christian ministry. 33
William Ockham
Like Marsilius, William Ockham became involved in politics as an op-
ponent of the contemporary papacy, but although he eventually developed
detailed ideas on both papal and secular government, the initial basis for
his opposition was religious, not political. Ockham and most other
Franciscans of the time held that Christ and the apostles had abdicated all
ownership of material things and that a life imitating this poverty was the
most perfect status available to a Christian. An important part of the
Franciscan position expounded in the Opus nonaginta dierum, a detailed
attack on John XXII's constitutions on the poverty questions, was the idea
of a natural right to use material things possessed by the human race in
common. Property rights in positive law were seen as excluding the
exercise of this natural right by persons other than the owner or those he
authorised. The 'licenses of using' granted the Franciscans could then be
viewed, not as conferring a new positive-legal right, but simply as cancel-
ling the exclusion created by the owner's legal right, with the result that the
friars were placed in a state of nature with respect to the things they used. 34
For the sake of clarity in the extremely intricate discussions of rights which
the poverty controversy by this time required, Ockham in the second
29. Ibid., I.7,15.

30. Ibid., 1.18.
31. Ibid., II.12.10.
32. Subjective right may be defined as an individual's legally recognised power or freedom with
respect to some good. However, the definition, historical antecedents, and juristic and philosoph-
ical significance of the idea are all debated (see especially Bücher 1965 and Kasper 1967).
33. Def. pac. II.14.6-16.
34. William Ockham 1963-74, Opus nonaginta dierum, cc. 61—2, 65, 87-92; Miethke 1969,
pp. 458-502.
chapter of the Opus nonaginta dierum defined a series of key terms, including
right of use (ius utendi) and ownership (dominium). In these definitions a
legal right is said to be a judicially enforceable 'licit power' to use some-
thing, and this has been taken to be an important appearance of the concept
of subjective right.35 Certainly, later Ockhamists made telling use of the
notion that a ruler's power (or lordship, dominium) over his people was not
to be identified with any such right of ownership.36 It may be doubted,
however, whether Ockham himself intended to do anything more in these
definitions than spell out the concrete meaning of having a (legal) right in
such a way as to draw clearly the issues between his own party of rebel friars
and John XXII.
Another feature of Ockham's early polemic with at least indirect impli-
cations for later discussion of rights is his assertion of what might be called
the right to a reasonable explanation in religion. The theory, accepted by
Ockham and his group, that a heretical pope was ipso facto deprived of all
spiritual power and was therefore no longer really pope, required for
practical effectiveness a general awareness of the pope's defection from the
faith. But since the pope was normally in charge of examinations of faith,
the legitimacy of even discussing his own orthodoxy was questionable, and
it was difficult to work out the obligations towards one another of
Christians holding different views on the subject. In the labyrinth of the
first part of Ockham's Dialogus, the thesis that an erring inferior is not
obligated to give up his opinions at the bare rebuke of an orthodox ecclesias-
tical superior has important bearing on such issues. Ockham argues that a
reasonable explanation of the error must be given the mistaken party
before he is obligated to change his position. This line of argument, like
Ockham's whole anti-papal project, has sometimes been seen as an attack
on the foundations of medieval Christianity, sometimes as a sign of vitality
in a system too often perceived as monolithic.
Doing without rights: Wyclif

Augustine had said that 'everything belongs to the just', 37 and such an idea
of a better kind of relationship to things and persons than the legal relations
of ownership and lordship actually obtaining in the world was richly
developed in the course of the Middle Ages. In his De civili dominio Wyclif
used this idea as the basis for a political philosophy. A soul infected by
35. Villey 1964; McGrade 1980.
30. Skinner 1978, II, pp. 117-22; 176-7, 320.
37. Epist. 93, c. 12.
744 Politics
mortal sin lacks the justice required for possessing or using things well, he
argued, while a just person through God's grace dominates the whole
sensible world with a lordship that cannot be taken from h i m . 3 8 Wyclif is
scornful of those who base lordship on conquest or the fictions of blind
sinners rather than the creation and re-creation of God, or w h o hold that it
requires coercive power rather than virtue. 3 9 True dominium is an effect of
charity, through which all the just possess the world in common and are
reciprocally one another's lords and servants. 4 0 This is not to say that
charity justifies a redistribution of material goods among the godly. O n the
contrary, the ideal is to follow Christ - the primum metrum et mensura of the
genus Christian as fire is the first measure of hot things - in a life of
evangelical poverty free from the temptations to sin which are inherent in
owning property or exercising coercive power. 4 1 It is a diabolical error to
suppose that any legal right is unconditional or that a person can do what
he pleases with his o w n possessions. All are merely custodians of what is
God's. 4 2
In denying absolute standing to conventional principles of justice,
Wyclif contributes much to the negative, revolutionary side of thought
regarding natural rights; but he equally denies the legitimacy of natural
rights themselves, if these are construed in secular fashion as items of moral
property held without continuing reference to God. It is controversial
whether Wyclif's work is the culmination of an earlier tradition (as he
suggests with his quotations from such figures as Augustine, Grosseteste,
and Bernard) or a serious distortion of it. In any case, his work represents a
final stage of some distinctively medieval lines of thought as well as an
example for later reformers who were not always appreciative of its
traditional premises.
Later developments
In the Italian Renaissance, in the midst of despotism and foreign invasions,

a variety of issues concerning man's nature, powers, dignity, present
condition, and prospects were discussed by humanists and others with
searching eloquence. Scholars give a more positive weight to medieval
contributions to this discussion now than previously, but very disparate
38. John Wyclif 1885-1904, De civ. dorn.. Book I, cc. 7, 9. 15.

39. Ibid., 1.11,20-1.
40. Ibid., I.i ι, 14.
41. Ibid., III.4-6, 10.
42. Ibid., I.19-21, II.15.
assessments are still defensible. Machiavelli, for example, so modern in his

analysis of human affairs, can also be regarded as an Augustine or John of
Salisbury manqué. Much political thought in Italy and elsewhere was
connected in one way or another with the Great Schism in the papacy.
Ideas about the authority of church councils elaborated in response to the
temporary disintegration of papal monarchy were developed further in
unsuccessful attempts to constitutionalise, first, ecclesiastical and then sec-
ular government. Detection of Marsilian and Ockhamist influences on
conciliarism and kindred developments depends heavily on the view taken
of these thinkers and of the later movements.43 Marsilius clearly asserted
the regular superiority of general council to pope, which Ockham denied,
but Ockham discussed constitutional questions in both spiritual and tem-
poral government in a way that emphasised the power of a community
(the congregation of believers or the community of mortals) to take
decisive action if ill served by its rulers. Marsilius, whose attack on the
whole conception of a spiritually governed society has been considered as
radical as Marx's attack on capitalism, was undoubtedly more often read
than cited. Ockham, on the other hand, continued to be a dominant
influence in academic theology and philosophy. A convincing history of
political Ockhamism remains to be written, but it would include chapters
on such important thinkers as Peter of Ailly and John Gerson in the
fifteenth century, James Almain and John Major in the sixteenth, and very
likely a chapter on Locke.
Locke's conception of government as the delegate of a community of
reasonable but imperfect individuals, set up to protect the natural rights of
its principals rather than to confer rights on them, and restrained in its
religious power, draws from (while, of course, modifying), not only
Ockhamism, but also the Thomistic tradition, to which Locke was exposed
through Hooker as well as in his own wider reading. The more corporatist
modern natural-rights tradition culminating in Rousseau has medieval
antecedents in the general tendency to think of human association in terms
of organic analogies or principles of ideal justice and in doctrines of
consent, such as those of Marsilius of Padua and Nicholas of Cusa, calling
for a genuine synthesis or concord of desires as a basis for legitimate social
action. The idea of rights against the state or civil society has roots in
medieval forms of scepticism about secular political and economic insti-
tutions, varying from hierocratic insistence on the corrigibility of all
43. For the influence of the canonists, see Tierney 1954,1955a, 1955b, 1066, 1975.
746 Politics
secular rulers in the light of ecclesiastically administered divine law to

Franciscan and Wyclifite renunciation of legal rights in favor of evange-
lical poverty and charity.
Philosophy of law
Much later medieval philosophy of law can be understood only in relation
to factors usually excluded from modern legal discussions: cosmic religious
values and moral leadership. There is other important material, easier to
absorb, in which law appears as a rational response to the problems of
communal existence; and modern legal positivism also has a medieval
background. Some of the relations among positions held may be clearer in
a thematic than in a chronological survey.
Law as the expression of higher values: Augustinus Triumphus and

Richard Hooker
The idea that human society has its value in relation to something beyond
the individuals composing it has two main expressions in our period, one in
the hierocratic conception of the papacy as an institution mediating the
relation of man to God and regulating on God's behalf all relations of man
to man, the other in the conception of law itself as an expression of the
divine order. Augustinus Triumphus will serve as an example of the first
tradition, Richard Hooker of the second.
Augustinus begins his argument that papal authorisation is required for
the validity of imperial civil law with an exegesis of Augustine's reference
to an eternal law, a highest divine reason, by which it is just that all things
should be done 'most orderedly'. Every just law depends doubly on divine
law, he argues: first, effectively and derivatively, simply in being just, since
what is just and what is unjust is discerned only through divine law; second,
materially and subjectively, because the first imperial laws were materially
composed from divine speeches (for example, the Roman law of the
twelve tables from Solomon's laws). Imperial law, therefore, depends on
the pope's authority by the same right by which it depends on divine law,
of which the pope is vicar and minister, especially since, according to
Dionysius, the law of divinity is such that its influence does not pass to
lower things except by intermediates (media). But the intermediate be-
tween God and the Christian people is the pope (Summa de ecclesiastica
potestate, q. 44, a. i). 4 4 Since Christ by his passion has merited judicial
44. Cf. q. I,aa. 1,3-9·

power over every creature, his vicar's authority extends to pagans as well as
Christians.45 The pope can correct, depose, or institute (without the
request or consent of others) the rulers of any realm (q. 46, aa. 1-3).
Anyone suffering unjusdy may appeal to him from the sentence of any
46
man whatsoever, whether king or emperor.
Augustinus argues at one point that a papal precept binds more than the
law of nature, since it binds not only potentially but also actually, not only
universally but also particularly. Moreover, its 'impression' in Christ's
commission of the Church to Peter (Matt. 16.18) is more unfailing than the
impression of the law of nature (which is only habitual and can in its actual
exercise fall into evil or be reformed to good) (q. 63, a. r ; cf. q. 60, a. 1). This
greater concrete effectiveness of papal authority is not, of course, a license
to violate natural law. Augustinus is intent on claiming all ultimate legal or
governmental authority for the papacy - the whole worldly machine is a
single governmental unit and needs a single ruler 47 - but in exercising this
authority the pope only mediates a higher justice and must act accordingly.
He is not to be obeyed if he commands anything contrary to natural or
divine law, for example (q. 22, a. 1). Similarly, it behooves him to set an
example of obedience to imperial laws, and he cannot justly deprive pagans
of political authority (q. 23, a. 3), free slaves (q. 22, a. 5), or arbitrarily take
one person's property and give it to another (q. 54, a. 4, ad 1). For
Augustinus Triumphus the hierarchical structure of ecclesiastical offices,
however unworthy the men who held them, reflected or represented the
perfection of Christ and the Apostles.48 Similarly, Richard Hooker de
fended the laws of English ecclesiastical polity rather than the faulty men
49
who administered them. He held these laws to be superior to the crown,
so that a royal grant contrary to law could be null, 5 0 and he held them to be
warranted by the systematic survey of 'laws and their several kinds in
general' with which he began his work. But Hooker's defence was itself
necessary because of an appeal to higher law, divine law, made by the
Anglican establishment's Puritan opponents. They held that Scripture
provided a rule for every action of life51 and, especially, that it contained a
45- q. 23,a. 1.
46. q. 45, a. 3.
47· q· « , a- 3; q· 49, a. 2; q. 60, a. 4.
48. q. 1, a. 2; qq. 76-112. On the distinction between office and person in medieval political thought
generally, see Kantorowicz 1957.
49. Hooker 1977-81, Of the Laws of Ecclesiastical Polity, Book II, c. I, sect, ι; I.16.1; VII.18 and
24.2-16.
50. Ibid., VIU.2.13.
Si. AM., II.
748 Politics
complete form of church government, 5 2 including a penitential judicial

system replacing episcopal courts with consistories of local laymen. 53
Hooker must thus be read as working out a substantially medieval problem
the problem of how, not whether, a Christian society should conform to
the divine order. Much of Hooker's Laws is a defense of human reason as a
factor in both the acceptance and the application of Scripture, as well as in
deciding matters not addressed by Scripture. 54 In this rational aspect
Hooker's concept of law reflects his vision of reality as wonderfully
ordered and intelligible, yet he clearly declines to present Anglican law as a
unique and necessary consequence of cosmic reason. Accordingly, he is
much concerned with the authority of human law over a range of'prob-
able' determinations in religious practices and with the autonomy of a
national human community and its comprehensive power over individual
members. 3 5 In this dynamic or voluntarist aspect, too, his concept of
human law also reflects his view of the nature of things. Departing from
tradition, Hooker does not include imposition by a superior as a necessary
element in his definition of law. 5 6 The immediate result is that he can speak
of God Himself as voluntarily adopting a law for His own actions, but in
addition he can present the English church as free to make laws for itself in
matters of religion.
It is sometimes thought that Hooker gave legal omnipotence to a
positivist public reason. 57 O n the contrary, he regarded natural and divine
law as limiting the range of human legislation, 58 his interpretation of the
English royal supremacy was strikingly constitutionalist, 59 and his whole
work expresses a sense that history is a process of communal development
and activity, guided by government but not simply a product of rulers'
decisions. Hooker's contention that a political community's laws should
reflect and promote its religion—that church and commonwealth are one if
every citizen is a Christian (Laws VIII.1.2; 6.6.11) - is a late expression of
medieval belief in the wholeness of the aims of human association.
52. Ibid., III.

53. Hooker's treatment of the spiritual-legal topic of lay-elders in Laws VI is largely lost. On the
significance of the treatment of penance which remains, see McGrade (1978).
54. Hooker 1977-81, II.7-8, III.8.
55. Ibid., Preface, 6; III. t, 9-11; V.8, 10.
56. Ibid., 1.2.1,1.3.1; but cf. I.10.7-8, VIII.6.13.
57. See Munz 1952, McGrade 1963, Cargill Thompson 1972.
58. Hooker 1977-81,1.8.5-8, 10.1; VIII.6.5.
59. /W., VIH.2. 3,7, 12-13, 17; 6.8, 13; 8.9. Munz 1952, pp. I07ff·
Moral leadership, revolution, and reform: John ofSalisbury and John Wyclif
PoUtical philosophers who regard law or government as an expression of
higher values - and nearly every medieval thinker did so in one way or
another-have special problems in providing against human failings. While
a distinction between office and person or laws and men is easy enough to
draw in theory, it is difficult to set up procedural safeguards 'over' insti-
tutions that are themselves endowed with supreme spiritual or moral
power. A natural twofold response to this situation consists of first, seeking
to engender the highest possible ethical qualities in rulers (the aim of
numerous 'mirrors' for princes) and, if that fails, countenancing revolution
or extra-legal reform.
While John of Salisbury regarded a true king as an image of deity he
viewed tyrants as the greatest of political horrors. The sole or greatest
difference between a legitimate ruler and a tyrant is rule in accordance with
law (Polier. IV. 1, VII. 17). Law itselfJohn describes in the highest terms as a
gift of God, interpreter or form of equity, norm of justice, image of the
divine will, safeguard of safety, union and consolidation of peoples, rule of
offices, exclusion and extermination of vices, and punishment of all vio-
lence and injury (Polier. IV.2, VII. 17). To encourage proper regard for law
by those in power he uses both theological and philosophical tactics.
Against 'whitewashers' citing Roman law maxims that the prince is legibus
solutus and what pleases him has the force of law, John deploys a sustained
exegesis of Deuteronomy enforcing the subordination of rulers to justice as
a religious principle,60 and he attempts to make such subordination more
plausible psychologically through a sustained philosophical criticism of the
current courtly ethos.61 This provides the framework as well as the subtitle
(On the Follies of Courtiers and Footsteps of Philosophers) of his work. The
brutalising effect of the successful exercise of power, the cultivation of
bizarre means to foretell the future and exciting ways to waste the present,
and the courtier's self-avoiding narcissism are reviewed with a rare combi-
nation of astuteness, stylistic elegance, and fervour.
Instead of the court life so described, John proposes, not heroic sanctity,
but a philosophically moderated civilitas, and existence including the de-
lights of literature and the truth of self-knowledge as well as love ofjustice,
country, and liberty.62 John's aim is to alter a form of life. Literature, in the
60. John ofSalisbury 1909, IV.7.
61. Ibid., I—HI, VII-VIII.
62. Ibid., VIII.5, 8-9; HI.»; VII.25.
750 Politics
broadest sense, serves the function of a constitutional check, providing

both standards of criticism for courtly life and a viable alternative to it. And
if this form of philosophy fails as a check, it supports the most radical ethical
action, tyrannicide. Like most of the Policraticus, the treatment of tyran
63
nicide combines Biblical and classical inspiration. Tyranny is outrageous
as a violation of both divine law and republican liberty. It creates a climate
of violence and deceit and at the same time legitimates these as tactics
against itself. John presents both aspects of the situation effectively in
recounting Judith's assassination of Holofernes and in epigrams adapted
from Cicero on friendship (it is licit to flatter whom it is licit to kill). John's
commendation of tyrannicide is not standard medieval doctrine, but it
deserves attention as the response of a widely admired writer to a widely
recognised problem.
Whereas John of Salisbury sought to augment virtue in order to bring
about royal government in accordance with law, Wyclif urged the king to
override merely human law in order to bring about the rule of virtue. For
both, obedience to divine law is central. Wyclif, however, uses this idea as a
basis, not for subjecting government to moderation by human law, but for
virtually dispensing with human law in favour of the law of charity.
Charity suffices for governing the soul, therefore it suffices for the govern
ment of whatever is accessory to the soul, such as bodily goods and the
goods of fortune; 64 and this applies not only to religious matters but also to
the commonwealth, for Christ's law teaches completely how every sin
should be destroyed and avoided, but it is impossible for a commonwealth
to deteriorate except by occasion of sin. 65 The sharp contrast in Wyclif's
thought between spiritual and material goods 6 6 and his insistence that such
human law as may be necessary should be directed by a vision of the higher
goods had striking implications for the reform of social attitudes and
practices. Christ came for the gradual abolition of private civil property. 67
It would be better even from the standpoint of earthly polity - and Wyclif
68
assures us Aristotle would agree - if all things were held in common.
Accordingly, social distinctions based on inequalities of wealth and power
are fallacious. Servile status, for example, is nothing to be ashamed of
63. Ibid., III.15; IV.i, 12; VII.20; VIII.17-21.

64. John Wyclif 1885-1904,1.17, III.2-3.
65. Ibid., I.44.
66. Ibid., I.12-13.
67. Ibid., II. 16, ΠΙ.4.
68. Ibid., I.14, III.12.
(hereditary servitude is of doubtful legitimacy), while wealth and political

power are more to be feared than sought. 69
For Wyclif, leadership in a good society would necessarily be moral
leadership and would have to come from the Christian ministry. The ideal
picture of England's priests guiding theirflocksby the light of charity is like
that of the best (aristocratic or natural) Biblical polity, in which judges
ruled by the law of the Lord. Given the sinfulness of the present human
condition, firm coercion by kings using civil punishments will also be
necessary, but this must be directed theologically.70 Unfortunately, the
condition of the Christian ministry at the time struck Wyclif as the heart of
the problem. 71 Accordingly, he proceeded to a typically medieval political
paradox, a call for coercively enforced reform to purify an essentially
spiritual governing authority.72 On one level, the expropriation of excessive
or misused ecclesiastical property Wyclif advocated can be understood as a
simple redistribution of wealth. He held that the poor were defrauded of
their share of the Church's goods under present arrangements and that if
clerics were remiss in administering such goods layman might properly
step in. 7 3 But this is not, of course, the primary level on which he sought to
operate. He sought to make the ideal of a Christian transformation of
human life effective in a world where, as he saw it, human life had largely
transformed Christianity instead.
Law as a product of human reason: Thomas Aquinas and William Ockham

Aquinas' famous definition of law as an ordinance of reason for the
common good, made by one having the care of the community, and
promulgated (ST, lallae, q. 90, a. 4), has both traditional and positivistic
elements, but its essence is practical reason. The definition is intended to
74
apply to law in general, including an eternal law (the divine reason
conceived as providentially governing all creation), natural law (the
rational creature's participation in eternal law), and divine law (super-
naturally revealed certifications of natural law and determinations of it
with a view to man's ultimate end of eternal blessedness), as well as human
law (which to be legitimate must be derived from natural law and not
contrary to divine law). By situating human law in the context of these
69. Ibid., I.22, 32-4.
70. Ibid., I.26-7, H 2; cf. III.13, Ι.ιο.
71. Ibid., I.44, III.2-3.
72. Ibid., I.37, II.5.
73. Ibid., I.42, III.14.
74. ST, lallae, q. 91; q. 99, a. 4; q. 100, a. 1.
752 Politics
other laws, Thomas follows medieval Christian precedent. By including

authoritative enactment and promulgation in the concept of law, he
accords some weight to power. Yet even when rules for human action
seem most clearly to be given from above, for Thomas it is a matter of
instruction rather than command, and the instruction is in first principles
constituted by and self-evident to practical reason, and in reason's elabor-
ation of those principles.75 Conversely, the primary function of legal
penalties is to supplement the family disciplina by which the young,
especially, may be educated to perform good acts voluntarily; only in
extreme cases is law a matter of repressing the wicked in order to leave the
good in peace.76 Similarly, in so far as tyrannical law is contrary to reason,
it is not law but a perversion of law, and to that extent it is not binding in
conscience.77
In reason's direction of human life, it moves from determinate ends,
fixed by our various natural inclinations, to rules for action that will be
effective in achieving these ends in particular, indefinitely varied circum-
stances.78 Thomas' own work, then, is partly concerned with making
evident the necessity of the more general practical rational principles. He
does this, for example, in his treatment of the Decalogue (traditionally
taken as a divine prescription of natural law), which he expounds in a
remarkably systematic fashion, so that the self-evidence of these precepts is
not a matter of isolated intuitions but of seeing the point of certain forms of
action as constituents or preconditions of human well-being.79 The more
specific one attempts to make the indispensable principles of reason, how-
ever, the more difficult it is to formulate rules which actually realise the
principles in all cases.80 Accordingly, Thomas also discusses various sorts of
contingency in human affairs and the functions of law and government in
meeting them.
While all legitimate human law is in some way derived from natural
law, some is essentially only a restatement of it. The most properly human
part of our legislation, however, is derived from natural law by way of
'determinations' fitting it to specific circumstances, rather as an artist
embodies a form in a particular matter.81 The implication seems to be that
75. ST, lallae, q. 90, a. i,adz;q. 94, a. I. On natural law as practical reason in Aquinas, Grisez 1969.
76. ST, lallae, q. 95, a. I.
77. ST, lallae, q. 92, a. 1, ad 4; q. 96, a. 4.
78. ST, lallae, q. 51, a. 1; q. 91, a. 2; Ilallae, q. 47, a. 15; q. 120, a. 1.
79. ST, lallae, q. 99, aa. 1-2; q. 100; Ilallae, q. 122.
80. ST, lallae, q. 94, a. 4.
81. ST, lallae, q. 95, a. 2.
there is a permissible range of choice for human lawmakers here, within

which there may be better or worse laws but where even the less adequate
rules are ethically acceptable in themselves and become obligatory through
legitimate enactment. Human law is rightly subject to change, either
because of change in the circumstances to which the law is addressed or
because of the progressive character of human reason, which may discover
better legislative solutions to continuing problems. Contemplated new
legislation must, however, be a considerable improvement if change is
to be justified.82
While practical reason is in principle universally valid as is any other
kind of reason, non-rational factors of various kinds may impede its
operation in particular men or societies.83 Although human law is essen-
tially concerned with making its subjects good, it properly does not
command all virtuous acts or forbid all evil deeds but issues only such
commands and prohibitions as most people are capable of observing
without intolerable strain and resultant greater evils (Thomas quotes
Augustine for the legality of prostitution).84 This and other imperfections
of human law are made good by divine law. Besides allowing everyone to
have certain knowledge of the most necessary principles of natural law,
divine law prescribes all the acts of all the virtues (including inner acts and
acts bearing on personal excellence rather than the common good) and
punishes all vicious acts;85 and divine law calls men to membership in a
higher community, with God, the principalitas of which is love inspired by
the Holy Spirit.86
Human law and government have positive ethical value, and Christians
are subject to secular government, even though its authority is founded on
nature, not grace.87 On the other hand, Christians should not be politically
subject to non-believers, for even though their authority may be in
principle legitimate, it would in most concrete circumstances pose dangers
to the Christians' faith.88 Heresy, as distinct from simple non-belief, is not
merely a vice but a legally punishable crime, for it is a violation of
commitments that a person has explicitly professed.89
82. ST, lallae, q. 97, aa. 1-2.

83. ST, lallae, q. 94, a. 6.
84. ST, lallae, q. 96, a. 2; Ilallae, q. 10, a. 11.
85. ST, lallae, q. 91, a. 4; q. 100, a. 2.
86. ST, lallae, q. 108, a. 1.
87. ST, Ilallae, q. 104, a. 6.
88. ST, Ilallae, q. 10, a. 10.
89. ST, Ilallae, q. 10, aa. 8, 12.
754 Politics
Notwithstanding Suârez' opinion that the issues involved are more ver-
bal than real (Tractatus de legibus ac Deo legislature, I.5.1), Ockham's 'volun-
tarist' view oflaw is sometimes sharply contrasted with Thomas' definition
of law as reason. Ockham's conception of coercion of wrongdoers as the
principal purpose of secular government 90 and his concern with questions
of power throughout his political writings may seem to indicate a legal
positivism consonant with the theological and philosophical positivism
sometimes found in his academic writings. The sceptical or positivist
interpretation of these earlier writings is questionable, however. Certainly,
Ockham made commitment to recta ratio essential to a good will in ethics.91
Rationality, as against arbitrary power, is also a dominant theme in such
major political works as the third part of the Dialogus and the Octo
quaestiones de potestate papae,92 but it is a rationality which gives unusually
full attention to contingent, often irrational circumstances. For example,
Ockham holds that monarchy is normally the best regime, for the world as
well as for a single realm, since one ruler can discharge the functions for
which governments are instituted more effectively than many. In spite of
his role as an imperial apologist, however, his endorsement of monarchy is
qualified: it is expedient to vary regimes or dominions according to the
variety, quality, and needs of the times. Sometimes it may be expedient to
have one secular or ecclesiastical ruler over all mortals, sometimes many
secular or ecclesiastical rulers governing together; at other times it may be
useful for many independent rulers to preside over the different parts of the
world.93
Ockham allows that the subjects of any government may in some cases
(but not without cause) depose their rulers and set up new regimes,94 and
he explicitly rejects the common axiom of papalist and Marsilian political
theory,95 that avoidance of strife requires a unity ofjurisdiction between
spiritual and secular governments (Octo quaestiones III. 3). The apparent
anarchy of these Ockhamist positions96 is offset by certain principles
Ockham sought to establish concerning what is ordinarily desirable in
90. William Ockham 1974b, Octo quaestiones, Q. III. c. 8.

91. William Ockham 1495-6, III, qq. 12, 13.
92. The less nuanced and less widely circulated Breviloquium and De imperatorum etpontificum potestate
are now more accessible.
93. Part III of the Dialogus, Tract II, Book 1, cc. 1, 5-6; Octo quaestiones III.11.
94. Ill Dialogus I.1.16,2.28: II.1.7,29, 31. Octo Quaestiones II.2,9-10; VIII. 6. Breviloquium IV.12-13,
VI.2.
95. Gewirth 1951, pp. 14-20, 30, 115-25.
96. Lagarde 1956-70, V, 258-9 and passim.
politics. Rejecting the Aristotelian monarchic ideal of a legally absolute

outstandingly virtuous ruler as unsuitable to the present day, he argued that
government should ordinarily operate under legal restraints.97 Again,
while he found no absolute line between secular and spiritual affairs and
hence held that lay and ecclesiastical authorities might in extraordinary
circumstances act in one another's jurisdictions,98 acceptance of a con-
siderable distance between secular and spiritual government as normally
desirable was one of his main aims in writing. 99 Finally, Ockham's strong
emphasis on personal liberty, though unaccompanied by endorsements of
popular participation in politics, favours a situation in which rulers must
show cause for restricting their subjects' freedom and in which subjects
have scope for opposition should occasion require it. 100
Law as command
Every political theory has something to say about effectiveness as a feature
of law, but when the sheer survival of a government or of endurable
political life is the paramount concern, it becomes natural to think of
coercive enforceability as law's essence.101 In our period, Marsilius of
Padua is the pre-eminent representative of this viewpoint. It may mislead
to say that whatever a ruler effectively wills has the force of law for
Marsilius, but the needed qualifications of this positivistic formulation all
rest on deeper considerations of effectiveness or pofitical viability. Marsilius
does not admit the legitimacy ofjust any ruler; though indifferent to the
type of regime as between one, few, or many rulers, he insists that
whatever the government is, it should be subordinate to the popular
will.102 But this is because governments lacking popular control are
ineffective, in his view, not merely in achieving or preserving the common
good but even in preserving themselves.103 Again, he holds that govern-
mental action should be determined by law, rather than law being a
product of government, but this is because only in the universal formu-
lations of law can the healthy will of the people find effective expression in
contrast with the partial or diseased wills of individuals.1 °* Finally, he holds
97. Ill Dialogus I.2.6.

98. Ill Dialogus 1.1.16—17; Octo quaestiones III.12; De imperatorum et pontificum potestate, c. 12.
99. McGrade 1974, pp. 84-5.
100. Octo quaestiones 1.6, III.6; III Dialogus I.1.5-8; Breviloquium II.3-4.
101. Gewirth 1951, pp. 131-6; but see Lewis 1963 and Rubenstein 1965.
102. Defensor pads I.8.4,9.9,17.2.
103. Ibid., I.12.6-7.
104. Ibid., I.11-12.
756 Politics
that what the people would will if impediments to the operation of their
desire were removed would not be capricious or irrational, but this is not
because the people have insight into, or commitment to, abstract principles
of rectitude but rather because determining the contents of proposed
legislation in an objectively adequate way is less important in his theory
than the wholesomeness of the popular desire which gives such recom-
mendations legal force. 105 Within very broad limits, then, whatever the
people as a whole approve as a legal command will have legitimacy, in the
sense of effectively promoting the peace and tranquillity that the people as a
whole necessarily desire.
If Marsilian legal positivism grew from an urgently felt need for govern-
mental unity, a higher positivism can be found as part of the hierocratic
tradition. The ruling idea here is that of an omnicompetent authority
whose status is warranted religiously, 'from above'. The popes were
acutely conscious of their legislative power in the area of canon law and
typically claimed superiority to every other positive law as well. In col-
laboration or competition with the papal model, fortified by the tradition
of theocratic kingship, and often instructed by Roman law, secular mon-
archs could sometimes vindicate claims to a similar supra-legal authority. It
has been beyond the scope of this survey to determine what, if anything,
can make political authority sacred, or whether true religion best supports
positivism or an illuminated legal rationalism. Much of later medieval
political philosophy seems understandable without such determinations,
and in the current philosophical climate it is naturally tempting to take up
that portion and leave the rest. The temptation is yet stronger if one feels
unhappy with the combination of ius divinum and Marsilian positivism (in
versions of the latter worked out by Bodin and Hobbes) used to sustain
absolutism in the seventeenth and eighteenth centuries. Separating the
purely reasonable parts of medieval political thought from the spiritual
concerns that originally animated them is surely bad history, however;
and in the present disspirited condition of Western political theory it is
arguably bad philosophy as well.
105. Ibid., I.13.1-4.

40
THE STATE OF N A T U R E AND THE
O R I G I N OF THE STATE
Lordship and ownership

Before the arrival in the west of Aristotle's Politics, the origin of organised
society was usually discussed in terms of the institution of lordship and
ownership (dominium). Dominium was seen to arise from an act of force, an
act of God, human agreement or an amalgam of these, just as in fact the
assumption of power often proved to be a combination of events such as
usurpation, the test of utility and merit, 'divine right', hereditary claims
and election or confirmation by the community or its clerical part.
The view that lordship arose from the forceful assumption of power and
the subjection ofother men had been handed down by Patristic writers. It
was illustrated by the story of the Fall and of the appearance with Cain and
Nimrod of sinful ambition and dominion, and it reflected too the Stoic
assumption (cf. Seneca, Epistola, XIV.2 (90)) that men had enjoyed
equality, freedom, and self-sufficiency in an original state of innocence
which had been lost through the appearance of human wickedness. The
history of the ancient Roman empire attracted much interest since it had
obviously gained authority from conquest. Government, then, was the
consequence of sin and it arose from the lust for power and domination.
But in so far as coercive authority restrained further abuse of free will, it
was a necessary and legitimate remedy of sin. After the loss of innocence
many men were no longer fit to enjoy freedom and equality or to practice
common ownership.
The divine origin ofrulership

The ultimately divine origin ofrulership was generally accepted. Even on
the assumption that government is rooted in sin, it acquires a moral and
sacred function as a remedy of sin. St Paul had written: 'There is no power
but of God. For the powers that be are ordained of God' (Romans 13.1).
For this reason Hugh of Fleury, writing after 1102, dismissed as foolish the
assertion that royal authority sprang from greed, crime, and pride or
758 Politics
through the agitation of the devil. ' Following the example of the writings
of Denis the pseudo-Areopagite many scholastics, including William of
Auvergne2 and Giles of Rome, 3 explained the existence of ranks and
offices on earth as an aspect of the hierarchic order of the universe and, in
particular, as an extension to earth of the celestial model in which God is
king and under him stand nine orders of angels in descending order.
Although men, like angels, are equal by nature, they are divided into a
hierarchy of unequal orders for the maintenance of harmony and for the
effective exercise of different functions. To bring multiplicity to unity, the
lowest must be subjected to the highest through intermediaries, as much in
the temporal as in the spiritual sphere.
There were many demonstrations of the claims of the priesthood to
mediate God's will in the establishment ofrulership. Giles of Rome 4 and
James of Viterbo5 supported the theory that temporal power is caused by
and is subject to spiritual power even in temporal as distinct from spiritual
matters, just as in the universe inferior, corporeal bodies are moved by
spirits and intelligences. Nature is perfected by grace, and the means to an
end is judged in relation to the final end. Some writers, particularly
imperialist writers, argued that the ruler's authority is derived immediately
from God and not mediately through the Church. One of the most
interesting supporters of such 'divine right' theory was Dante who, like
many others in his time, distinguished a twofold end for man correspond
ing to his double nature, part corruptible and temporal and part incor
ruptible and spiritual. The means to each of these ends differ and man needs
a double direction. The emperor guides man to temporal felicity in
accordance with philosophic teachings regarding the practice of the moral
and intellectual virtues and the pope leads the human race to eternal felicity
in accordance with revealed truth. But although both ends are divinely
6
ordained, and although one is inferior to the other, they are distinct. At
7
the root of many disagreements, as John of Paris saw, lay the problem of
evaluating what was meant by the principle that temporal ends are or
dained to spiritual ends.
i. D i regia potestate et sacerdotali dignitate. Ι, ι.

2. De universo, II, 2.
3. Giles o f R o m e 1929, Df ecclesiastica potestate, II, 13.
4. Ibid., I, 4, 5.
5. James of Viterbo 1926, De regimine christiano, II, 6, 7.
6. Dante 1963, De monorchia. III, 16.
7. John o f Paris 1968, De regia potestate et papali, 5; 13; 17.
The state of nature and the origin of the state 759
Agreement as the basis for government

Roman law provided material for the conception that government sprang
from the voluntary agreement of the people or the corporate community.
By the lex regia the people of Rome had 'conferred upon the prince all their
power and authority'.8 That the relationship between ruler and ruled
rested on a sort of compact or pactum was evident to many writers, among
them Manegold of Lautenbach.9 Marsilius of Padua provided a systematic
statement of the role of the community as the corporate Legislator from
whom the power of the ruler is derived. ' ° By the fifteenth century the idea
that secular lordship had originated in a free grant of authority by the ruled
had come to be applied by conciliarist thinkers to spiritual authority. In his
De concordantia catholica Nicholas of Cusa argued that since men are by
nature free, the regulation of freedom requires agreement; rulers should be
constituted through election.11 Suârez rejected the view that political
authority was possessed by 'divine right'; the decision to create a political
jurisdiction must be made by the community.12
Only from the thirteenth century onwards did the origin of the political
community itself receive sustained examination as a natural necessity. The
'organic metaphor' or analogy of the state with a living organism had
already been deployed to indicate the interdependence of all the members
of a community under the purposeful direction of its higher parts, and the
metaphor continued to find favour in later centuries. ' 3 So too did Cicero's
conception of the virtuousness of public life.14 But the decisive develop-
ment was the discovery of Aristotle's teaching that man, unless he was a
beast or a god, could not exist without the state. He is a political animal by
nature. A state or body of citizens is required to meet man's natural needs -
physical, moral and rational. Although the state was 'made for the sake of
living, it exists for the sake of living well'. 15 Writers as different as
Aquinas,16 Giles of Rome 17 and Marsilius of Padua18 found in the Politics
8. ' . . . cum lege regia . . . populus ei et in eum omne suum impcrium et potestatem conférât', Digest,
1,4.1.
9. Liber ad Gebehardum.
10. Defensor pacts, I. 15.
11. Nicholas of Cusa 1964-8, De concordantia catholica, II, 14; II, 19; II, 34; III, 4.
12. Suârez 1971—7, De legibus ac Deo legislator, III, ii, 3—4.
13. Cf. John ofSalisbury 1909, Policraticus, V, 1; Marsilius, Defensor pads, I, 2; Nicholas of Cusa, De
concordantia catholica. III, 41. Also Aristotle, Politics, I, 55; VI, 4, 2.
14. Df republica, 1,1 - 7 . Cf. Lactantius, Divinae insiitutiones, VI, 10 (man as a social animal).
15. Politics, I, i, 8 and 12.
16. Sentemia libri Politicomm, I, 1.
17. De regimint principum. III, 1.
18. Defensor pads, 1,3 and 4.
76ο Politics
the source of their account of the origin of the state. The state is the
consummation of the development of the family, of villages, and of
townships. It reflects the natural increase of population. It is a natural, not a
conventional, institution. But it is the product of natural human reasoning
and issues from rational agreement.
The naturalness or unnaturalness of the state
In his De regimine principum, Aquinas followed Aristotle to the extent that

he argued that man cannot live alone. If he could live alone, he would be
king over himself under God, but by nature he is both a social and a
political animal. Men would not pursue their common end, which is their
common good, without the direction of a ruler. Aquinas invoked the
analogy of a ship blown about by diverse winds and of its helmsman who
selects the course to be taken and steers it to port. But the ultimate end for
Aquinas is not virtue or living well; it is the vision of God, and to attain the
ultimate end a government of priests has been instituted which is distinct
from the government of kings. To kings belongs the duty to order the
good life in a way that is congruous with the achievement of eternal
felicity.19
Discussion of the origin ofrulership and of the state was closely bound to
an assessment of the state of nature. If men are naturally innocent, equal,
and free, if property is naturally held in common, if natural law is unchang
ing and perpetual, the state through its maintenance of private property,
unequal liberties, and serfdom, is a contradiction of nature or at best it is
based on conventions that qualify nature. Many thought that serfdom,
private property, and coercive power do not belong to the state of nature.
The appearance of sin having altered all, arguments had to be advanced and
distinctions had to be found to justify these institutions.
From the canonist Rufinus in the twelfth century 20 to Suârez in the
early seventeenth,21 there were thinkers who met the problem by dis-
tinguishing what is commanded by the natural law and what is indicated
or demonstrated by it without being commanded. Thus, according to
Rufinus, liberty is one of the demonstrations of the natural law. Natural
law neither commands nor forbids it but shows it to be good. It was
expedient before the Fall but now it is no longer expedient and the civil
19. De regimine principum, I, 1, 14; Sentenlia libri Ethicorum, 1,1; Sententia libri Politicorum, I, 1.
20. Summa decretorum, D . 1, Dictum Gratiani ad c. 1.
21. Suârez 1971-7, II, 5.
laws rightly deprive most men of freedom. This is necessary because crime
must be punished and virtue inculcated through discipline. Even slavery is
a means to virtue; thus, slavery is a part of the natural law although it is not
found in the state of nature. Other institutions also arose subsequent to the
state of nature and modify the natural law, but they do not conflict with it.
Customs have wisely developed regarding the union of male and female
that limit this to certain persons after marriage; otherwise ephemeral and
precipitous unions would occur.22
The recovery of Aristotle's natural philosophy led people to take note of
the naturalness of the political community itself; the actual state was itself a
natural state. Aquinas consequently distinguished two types of lordship.23
Lordship in the sense of rule in the interest of the ruler was, as the Stoics and
the Fathers and the medieval lawyers had maintained, absent from the state
of nature or innocence. But lordship in the sense of the direction of free
men for the sake of their common good is, as Aristotle showed, natural, for
man is by nature a political animal and the common good of a society
requires an organising and directing authority. Moreover, men are by
nature unequal in respect of knowledge and the capacity for justice, and it
would be unsuitable for the superior members of society not to be able to
use their superiority for the benefit of others. Thus subjection is natural,
and although slavery is not found in the state of nature, it is not against
nature, for it is an addition made to nature by human reason for the benefit
of man following the introduction of sin.
The naturalness or unnaturalness of private property

Aquinas also bowed to Aristotle in respect of private property. 24 He
recognised the strength of traditional teaching that by natural law all things
belong to all men in common. But he refused to see private property as a
contradiction of nature. He appealed to the distinction between possession
and use. Natural law does not prohibit the acquisition or division of
possessions; it arises by human and reasonable agreement under positive
law as an addition to the natural law. The tranquillity and order of society
are better assured if men are encouraged to acquire property through their
own effort and to maintain it themselves. But the common use of things
prevails over private possession and in cases of necessity goods should be
22. On the handling by the canonists of Plato's teaching regarding the community of wives see
Kuttner 1976.
23. ST, I, q. 92, a. I ad 2. Also Ilallae, q. 57, a. 3.
24. ST, I, 2, q. 94, i. 5. Also Ilallae, q. 32, a. 7 ad 3; Ilallae, q. 66. Cf. Politics, I, 3.
762 Politics
transferred to those in need. Aquinas forthrightly condoned the reasonable-

ness of a poor man's taking what he needs from a well-endowed person,
because by natural law inferior things exist for the common use of men and
human law cannot abrogate natural law. The ruler, for the sake of the
common utility, may tax or confiscate private property. 2 3
Aquinas' doctrines that lordship and private ownership belonged to the
state of nature rapidly gained a wide influence. John of Paris, for example,
held that 'true lordship' over things belonged to those who acquired them
by their o w n labour. Men may dispose of their property at their pleasure as
long as no injury is done to another. But the prince is the judge of rights; he
must punish usurpations of property when found and redistribute it when
the common 'necessity or utility' requires it. 26 Ockham taught likewise. 27
But the controversies over the question of apostolic poverty brought these
discussions into sharper focus. Many scholastics in the fourteenth century
applied a moral or religious test to ownership and held that the right
enjoyed by an individual possessor was dependent on his relationship to
God. The Spiritual Franciscans argued that by nature property is held in
common and therefore those who seek perfection should renounce per-
sonal possession. When applied to the Church or to the wealthy or to those
outside the church such doctrines were of far more than academic interest.
The question, too, of the ruler's powers over the property of his subjects
was crucial to the discussion of lordship.
Giles of R o m e , basing himself upon Augustine's De civitate Dei, II.22
('true justice does not exist except in that commonwealth whose founder
and ruler is Christ'), argued that valid titles to property could not be
enjoyed by infidels and excommunicates and could only be held under the
general lordship of the Church. 2 8 John of Paris rejected this, except in cases
of supreme necessity. 29 So did Ockham who taught that the Church's
mission was purely spiritual and was not concerned with social utility. 30
But Richard FitzRalph argued that since lordship is found in the state of
nature and since by nature men are just and hold all things in common,
original lordship is exercised by all just men in common. Ownership which
makes a thing the property of one man is not original lordship and it may
exclude the just man from the use or the right to use that thing. Ownership
25. De regimine iudaeorum.
26. De potestate regia et papali, 7.
27. Opus nonaginta dierum, II, 88; Breviloquium de principatu tyrannico, III, 14.
28. De ecclesiasticapotestate, II, 7 and 12.
29. De potestate regia et papali, 7.
30. De imperatorum et pontificum potestate, 9 - 1 0 .
is acquired lordship which is also called civil or political or positive

lordship. Original or pure lordship, on the other hand, is the right of the
just man alone.31 Wyclif, who borrowed many of his premisses from
FitzRalph, also wrote in support of the view that the man who is in grace is
lord of all the world, and if there is a multitude of such men they should
have to hold all in common. Misuse of goods, even by the ecclesiastical
community, creates a ground for their expropriation.32 But Wyclif and
FitzRalph did not urge an immediate or unconditional transfer of property
from the unjust to the just. Viewed spiritually, men who are in grace do
have a full right to all property, but the presence of sin constrains all to
accept the fact of private ownership.
In the sixteenth century some Thomist thinkers vigorously reaffirmed
the equal capacity of all men, including unjust men and non-Christians, to
establish, under natural law, their own political societies with true rulers
and legitimate ownership. The Spanish conquests in the Americas and the
seizure of Indian possessions were attacked on these grounds by, among
others, Vitoria in the 1530s in his lectures on The Recently Discovered Indies.
Rulers as delegates of the people

In the scholastic period populist principles - 'people-sovereignty' or the
'ascending' theme of government - challenged 'ruler-sovereignty' and
hierocratic theory - the 'descending' theme of government - even in the
ecclesiastical sphere. The Roman law teaching that rulership is derived
from the people by the lex regia (Digest, 1,1,4) and Aristotle's view that man
is by nature a political animal, capable of being a citizen as distinct from
being a subject (subditus), underlay much of the development of political
philosophy.
Absolute rulership
Roman law did not uncontrovertibly point to the conclusion that the
people retained any control over the authority it had created. In Roman
law the prince is freed from the laws (legibus solutus); his will has the force of
law. The revival of Roman law from the end of the eleventh century
encouraged a growing appreciation that a legislative authority - e.g. a king
or emperor - is the sole, immediate, active source of law, and that only the
prince can make positive, written law. Hence the bold claims of, for
31. Richard Fitzralph 1890, De pauperie salvatoris, I, 2; IV, 1.

32. De civili dominio, 1,14 and 37.
764 Politics
example, Aeneas Sylvius: the emperor is absolute, he is lord of the laws and
he must be obeyed, however unjust he is.33 From the thirteenth century at
least the conception emerged more clearly that there is in the state a
sovereign power, caesar or pope, which alone can provide law (legem
ponere; lex posit(iv)a) and against which there can be no appeal.
The sovereignty of the people and other restraints on rulership

However, the civilians and others who knew about Roman law also
debated another theory, namely that the people had merely conceded
authority to the emperor (through a concessio) and is entitled to resume it.
The Corpus iuris avilis had also laid down that 'it is a thing greater than
empire that a prince submit his government to the laws'. 34 Many were far
from agreeing that the emperor might issue law arbitrarily. For one thing,
law must conform to justice. As John ofSalisbury wrote, 'the prince is said
to be loosed from the bonds of law, not because unjust deeds are permitted
him, but because he ought to be one who cultivates equity from the love of
justice rather than from fear of punishment'.35 And Bracton, while affirm-
ing that the king has no peer in his kingdom and that no writ runs against
the king, held that the king is under the law; what pleases the prince has the
force of law (Digest, I, 4, 1), but the prince should will only what is right
and just and agreed after taking counsel with his great men. 36 Moreover,
the prince should respect the customs of the people. Bartolus took this
further: if the people can by tacit consent create usages and customs, they
can also by explicit consent create the written law as well; the state was itself
sovereign (civitas sibi princeps).37 Later in the sixteenth century Jacques
Almain argued forthrightly that no community can wholly alienate its
authority to a ruler any more than an individual can renounce his right to
self-preservation; although monarchy is the best form of government, it is
limited by being a rule over free men. 38 At the very beginning of the
revival of Roman law, Irnerius had defined law as an ordinance of the
people promulgated after consultation with wise men, especially the Se-
nate.39 Richard Hooker reflected many thinkers in the centuries that
33. De ortu et auctoritate imperii romani, 16; 20; 21.

34. 'Maius iniperio est legibus submittere principatum', Cod. I, 14, 4.
35. 'Princeps tamen legis nexibus dicitur absolutus. non quia ei iniqua liceant, sed quia is esse debet,
qui non timoré poenae sed amore iustitiae aequitatem c o l a t . . . ' Policraticus, IV, 2.
36. Henry of Bracton 1968, De legibus et consuetudinibus, I, 8, 5; III, 9, 3.
37. Woolfi9i3.
38. Almain 17ου, De dominio naturali civili et ecclesiastico. Cf. Almain 1525, De auctoritate ecclesiae, I.
39. Summa codicis, I, 14, 3.
intervene when he wrote that, while wise men may devise laws, their
coercive character may only be given by the whole community or its
representatives or by the prince if he has received this authority from the
community. 'Laws they are not therefore which public approbation hath
not made so.' 40
Limited monarchy
In the thirteenth century the case for a middle road between ruler-
sovereignty and people-sovereignty was developed and the theory of
limited monarchy or of the mixed constitution emerged. The question was
not basically whether rule by one man was better than rule by many.
Under the influence of Aristotle's Politics, I, I, a distinction came to be
drawn between political government according to law and despotic or
arbitrary government. The value put upon each and the position of the line
between the two varied from thinker to thinker. Aristotle in his Politics had
himself indicated a relativist view of the value of different constitutions.
Aquinas in his Commentary on the Politics*1 and Ptolemy of Lucca in his
De regimine principum, II, 8-9 and III, 11, regarded political government
(regimen politicum) as government in which the ruler - whether one or
many - was limited by the laws of the state and governed according to
established laws. Regimen regale was government in which the ruler has
absolute power (plenaria potestas) and is not bound by the laws of which he
is the living source. The form of government to be preferred depended
largely on the qualities of the people to be governed.
Giles of Rome expressed a clear preference for regimen regale over
regimen politicum in his De regimine principum, II, 1,14 and II, 2, 29: the king
should rule according to his own will and according to the laws he has
himself made, not according to the laws made by the citizens. But he is
obliged by the natural law to maintain justice and to seek the common
good, not to pursue his private interest (1,1,12; I, 3,3). The case for rule by
one man received a universal extension in Dante's De monarchia. In this
treatise Dante sought to maximise the liberty of the human race by placing
the affairs common to all men under the rule of one supreme, universal
monarch. Ockham called a monarchy in which one man rules without
legal limit for the benefit of the common good a pure, regal monarchy; this
is 'the best, when it is at its best'.42 Regal monarchy, in Ockham's terms, is
40. Of the Laws of Ecclesiastical Polity, I, 10, 4 - 8 ; VIII, 2, 9 - 1 1 .

41. Senlentia libit Politicorum, I, 1.
42. Dialogus, pars III, fr. I, lib. II, c. 6.
766 Politics
government over free men that does not use men or their goods despoti-
cally for its own advantage. But this is nowhere to be found: 'in these days
there is perhaps in the whole world no instance of regal monarchy'.43 The
best in practice is kingship according to law in which the ruler is bound by
certain laws or customs introduced by men. Sir John Fortescue reconciled
regal and political lordship in a combined dominium politicum et regale. By
this Fortescue meant that the king should rule by such laws as he makes
himself (dominium regale), but such laws should receive the assent of his
people (dominium politicum).**
Aquinas had earlier expressed an influential preference for mixed, con-
stitutional rule. In De regimine principum, I, 2 and 3, and also in 1,6, Aquinas
held that a monarchy devoted to the common good was the best form of
government because it tended to the unity of the state. But to prevent the
monarch becoming a tyrant his power must be limited (temperatur). In the
Summa theologiae Aquinas describes as the best government that of one man
of virtue when other men of virtue govern under him and when the rulers
are elected from all and by all (politia bene commixta).45 The mixed consti-
tution includes the elements of monarchy, aristocracy, and democracy
(regimen commixtum) and in it laws are made by the nobles with the
participation of the people (majores natu simul cum plebibus).46
In the Summa theologiae Aquinas outlined a case for giving everyone a
share in the ruler's authority: the ruler should be elected by all and from all.
John of Paris argued that the climatic, linguistic and political differences
between communities constituted a reason why separate communities
should choose their own rulers and not, pace Dante, be subject to one
supreme monarch.47 Marsilius used a utilitarian argument in favour of
popular participation in government: the involvement of all the citizens
would make government more efficient.48 Nicholas of Cusa went further:
since by nature all men are free and equal in respect of authority, valid rule
and legal coercion can only arise from the voluntary consent of men. All
power, spiritual as well as corporeal, is latent in the people in potency, and
if it is to be actualised so as to regulate and coerce men, the people must
freely subject themselves. The elective principle is the origin of every
43. 'Forte his diebus non est in universo orbe talis principatus scilicet primus regalis', Dialogus, ibid.
44. De natura legis naturae, I, 16 and 24-5; De laudibus legum angliae, 9.
45. ST, lallae, q. ioj, a. 1.
46. ST, lallae. q. 95, a. 4.
47. De potestate regia etpapali, 3.
48. Defensor pacts, 1,12.
ordered superiority.49 Suârez thought likewise: the authority to make law

resides in the community for men are by nature equal and no man
naturally has jurisdiction over other men. The community creates govern-
ment in such a form as it wishes.50
The ruler as private and public person

It was very hard in some concrete situations to disentangle the private,
proprietary rights and actions of kings from their public capacities and the
property of the state entrusted to them. But Roman law distinguished
public and private law and John of Salisbury echoes it when he describes
the king as 'the minister of the public utility': 'the power of all the subjects
is gathered together in him that he may be strong enough to seek out and
perform what is useful for the welfare of each and all... The prince is, as
some define him, a public power and a certain image of the divine majesty
on earth.' 51 Aquinas spoke of the prince as 'the bearer of the person of the
community'.52 Gerson, in Vivat rex, reminded the king of France that a
king is not a private person but a public power ordained for the welfare of
the whole community. The development of elective and representative
institutions and of ideas of consent and the common good, and the
development, moreover, of the notion of the community itself or of the
communitas regni, was intertwined with reflection on the links that bind
together the ruler and the ruled. Marsilius of Padua brought out in a
decisive manner the implications for political philosophy of the idea of
popular sovereignty and of representative government. Following
Aristotle's Politics, III, 11, Marsilius argued that in every 'perfect com-
munity' the authority to make law and to institute the governing head
(pars principans) must belong ultimately to the people or all the citizens
(populus seu civium universitas) or their weightier part (valentiorpars).53 Law
made by the citizen body as a whole will secure the common benefit, and
will be better observed, than law made by one man or by a few. The people
must be the legislator. The pars principans is not necessarily elected; it might
consist of one individual or a small council and inheritance may enter into
it, but the authority granted to it is delegated to it immediately by the
49. De concordantia catholica, II; III, 4.

50. De legibus ac Deo legislature, III, 2, 3.
51. 'In eum omnium subditorum potestas confertur, ut in militate singulorum et omnium ex-
quirenda et facienda sibi ipse sufficiat... Est ergo, ut eum plerique difhniunt, princeps potestas
publica et in terris quaedam divinae maiestatis imago', Policraticus, IV, i.
52. 'Princeps qui curam populi habet et eius personam gerit', ST, Ilallae, q. 57, a. 2.
53. Defensor pacts, 1,12.
768 Politics
Legislator. Thus the pars principans is the executive part (secundaria quasi
Instrumentalis seu executiva pars) of the community and is bound by the law.
The doctrine of popular sovereignty and of elective, representative govern
ment was also applied to the ecclesiastical community. 5 4 Marsilius applied
the principle that coercive authority is located in the whole community to
the Christian community of believers who, in a Christian republic, are
identical with the community of citizens. They should therefore be gov
erned by a general council organised by the 'faithful, human legislator
which lacks a superior'. 5 5 Other conciliarist writers argued towards similar
conclusions.
The limits set upon a ruler's authority varied according to the nature and
source of his authority. Likewise the arguments that were elaborated to
cope with abuse of authority and with tyranny also varied. There were
writers, particularly earlier in the Middle Ages, who argued that since royal
authority was sacred and of divine institution, it could not lawfully be
resisted even when it acted unjustly. 'All power is from God . . . he who
resists the power resists the ordinance of God' (Romans, 13. ι and 2). Even
Nero, as Augustine remarked, 5 6 owed his power to divine providence
which decided that man's condition deserved such a master. Gregory the
Great provided support for a theory of'divine right' when he put the bad
ruler beyond criticism by men and made him accountable to God alone. 5 7
Imperialist writers in the eleventh century tended to agree.
The legitimacy of resisting authority
But this was not the whole story, for the patristic inheritance overwhelm
ingly proclaimed that the purpose of society and of rulership was the
attainment of justice. By this standard the tyrant enjoyed no rightful
power. There were several ways along which theories allowing control of
the ruler might develop and admit of resistance to him and ultimately
justify assassination. O n e of these starts from the distinction between a king
and a tyrant. Isidore of Seville had drawn a sharp contrast between t h e m . 5 8
59
John of Salisbury, too, in his Policraticus describes tyranny as the con
tradiction of kingship which consists in the government of the people in
the interest of public utility and equity and which upholds the law out of
54. Ibid., I, 15-17.

55. Ibid., II, 16-17; Π, 20-1.
56. De civitate Dei, V, 19.
57. Regula pastoralis. III, 4; Moralia in lob, XXII, 24; XXV, 16.
58. Isidore of Seville 1911, Etymologiae, IX, 3; Sententiae, III, 48.
59. Policraticus, III, 15; IV; VIII, 17-21.
the love of justice. A tyrant who oppresses the people by the sword and
who voids the law has no rights against the people and deserves to die by
the sword. But every ruler is the image and vicar of God, and killing a
tyrant is an act that signifies divine vengeance. The tyrant's assassin does
not really act in the name of the outraged community.
Aquinas also based his argument in favour of resistance to tyranny upon
the tyrant's lack of divine sanction. In his Commentary on the Sentences, II,
d. 44, 2, 2, he stated that authority (praelatio) derives from God or it is
defective. However, Aquinas is also careful to indicate the limitations to be
put upon a theory of resistance. Authority may be defective either through
the manner of its acquisition or through the use to which it is put. Resis
tance to defective authority is not allowable if the ruler is simply
unsuitable, nor is it allowable if the ruler obtains power illegally but sub
sequently gains acceptance by his subjects. But it is allowable if the ruler
uses his authority to command acts of sin or to command what is beyond
his rights. Aquinas, through his support of the idea of a mixed constitution,
hoped that the ruler's authority might be limited (temperare) in its use in
such a way that the opportunity for tyranny would be removed. 60 But he
clearly affirmed that resistance to a tyrant is not sedition, because it is the
tyrant who is seditious.6 ' It may be better to obey the tyrant if a worse evil
is likely to ensue from his deposition, but he cannot be obeyed in acts of sin.
Aquinas also applied Aristotle's test of good government which is good if it
serves the common good. The obedience due to a ruler by virtue of his
divinely ordained authority extends only as far as the order of justice
62
requires.
Another form of resistance theory rested on the idea of contract. The
argument used by Manegold of Lautenbach in his Liber ad Gebehardum
was one that reappeared frequently in later centuries: the king is hired or
contracted by the people to do a job. The elective principle was an aspect of
the making of a king and oaths were taken at royal coronations. If the king
chooses to act the tyrant, he can no longer claim fidelity because he has
broken faith and the compact. Hence the people should be free of his
lordship. Manegold also used the argument that the tyrant lacks the
essential qualities of a king who is meant to excel all men in wisdom, justice
and piety. Loss of honour by corruption justifies loss of the royal title.
Aquinas is one of many who appealed to the notion of contract. In De
60. De regimine principum, I, 6.

61. S Τ, Ilallae, q. 42, a. 2.
62. ST, Ilallae, q. 104, a. 6; also q. 42, 2. Cf. Giles of R o m e , De regimine principum, 1,1,12; I, 3, 3; III,
2.34-
770 Politics
regimine principum, I, 6, he held that public authority — though not private

individuals - may remove a tyrant who breaks the contract with the
people, just as it may establish the king. Giles of Rome, too, although he
favoured absolute and hereditary monarchy, nonetheless maintained in De
renuntiatione papae, XVI, i, that the ruler must be established by the consent
of men and by that same consent he may be deposed. Contract theory was,
however, two-edged. Richard Hooker shrewdly observed that the com
munity may require the consent of the ruler before withdrawing from his
dominion. 63
A further ground for the removal or correction of a ruler lay in the
notion that the community possesses sovereignty. Marsilius' doctrine of
popular sovereignty allows the community, or the representatives whom it
appoints, to correct or depose the ruler because Marsilius transfers the
function of making law from the ruler to the community. The ruler (pars
principans) is simply the executive part of the community which is the
legislator, hence the pars principans is accountable at law. 64 Jacques Almain
hardened the argument when he wrote that the people cannot abdicate;
under natural law any person who menaces the community may be
resisted.65
Among the later scholastics the assumption that authority is a concession
by the community was paramount. Molina treated tyranny as the exercise
of more authority than is granted to the ruler by the community which
determines the extent of the ruler's authority. 66 Soto argued that since
authority is conferred by the community it may be withdrawn by it in the
event of manifest tyranny. Soto, although he had a high view of the
authority of the king, represents the view that a legitimately appointed
ruler who becomes tyrannical may be assassinated, after a public judge
ment, by a duly appointed agent; a tyrant usurper may be killed by
67
anyone. The scholastics in general expressed firm support for the pos
sibility of resistance based on the principles of justice and of contract. Such
principles gained ground in a number of different countries. The Wars of
Religion and the divisions following the Reformation were not a prelude
to the rise of resistance theory. To a certain extent they may have been a
sequel.
63. Of the Laws of Ecclesiastical Polity, VIII, 2, 10.

64. Defensor pads, I, 18.
65. De dominio naturali civili et ecclesiastico.
66. Luis de Molina 1659, De iustitia et iure, I, 2, 23.
67. Domingo de Soto 1567, De iustitia et iure, V, ι, 3. Cf. Mariana, De rege, I, 6.
41
THE JUST W A R
The medieval development of theories of war

In the preface to his Tree of Battles, written in 1387 and dedicated to Charles
VI of France, Honoré Bouvet laments that 'all holy Christendom is so
burdened by wars and hatreds, robberies and dissensions, that it is hard to
name but one little region, be it duchy or county, that enjoys good peace'.
War was the normal condition of society in medieval Europe; and pessi-
mistic doctors argued, on theological or astrological grounds, that 'in this
age it is necessary for there to be wars, and the slaughters and infinite
sufferings of war'.1 Some men were dazzled by the pomp and circum-
stance of glorious war; most doubtless agreed that 'warres & bataylles
shold be acursed thyng, & not due'. 2
About that cursed thing arose a prodigious literature — legal and theolog-
ical, philosophical and practical, historical, strategical, and ecclesiastical.
The centrepiece of the medieval discussions, to which they owe their
abiding philosophical interest, is the theory ofjust war.
That theory is now most familiar from Aquinas' brief essay De hello (ST,
Ilallae, q. 40); but in this instance Aquinas was no innovator: he stands in a
long line of theorists, thefons et origo of whose ruminations is to be found in
the writings of Augustine.3 The scattered observations of Augustine and
his successors were collected and ordered by the canon lawyers of the
twelfth century, whose work is best represented by Gratian's Decretum.
The commentators on the Decretum, of whom Rufinus was the first,
developed a theory of war; and the theory was elaborated in the thirteenth
1. Baldus, Consilia V, cons. 439; see esp. John of Legnano, De bello, Ch. 6; Bouvet, Tree of Battles,
III. 2.
2. Christine de Pisan, Fayttes of Armes 1.2.
3. The main Augustinian texts, in chronological order, are: Contra Faustum ΧΧΙΙ.74-8; Ep. 138; De
civ. Dei XV. 4; XIX. 7, 12-is; Ep. 189; Serm. 302; QQ. in Hept. IV.44; VI.io; Ep. 229. The
influence of Augustine was all-pervasive; he is cited frequently in this chapter only because he was
constantly quoted by the medieval theorists. By contrast, Aristotle's influence was negligible (but
see below, nn. 30, 58); for his views see Politics l256 b 23-6, I333bj8—I3J4*2-
772 Politics
century by Decretalists such as Raymond of Pennaforte and by theologians

such as Alexander of Hales and Thomas Aquinas. The theory was vigorous
in the fourteenth century -John of Legnano and Baldus de Ubaldis may
serve as representative figures - and in the sixteenth it was discussed by men
of the stature of Vitoria and Suârez. The final flowering of the tradition
may properly be seen in Hugo Grotius' De jure belli et pads, a masterpiece
standardly taken to mark the beginning of modern political theory.4
Early Christian thinkers inclined to pacifism;5 for the New Testament
seemed firmly opposed to any form of military enterprise: 'All soldiering is
instituted either to repel injury or to inflict punishment - injury is repelled
either from one's own person or from one's associates - and both things are
prohibited by the law of the gospels.'6 Augustine checked that pacifist
inclination: on his view, Christ's 'precepts of patience' do not outlaw war;
for 'those precepts are addressed rather to the preparation of the heart,
which is internal, than to the deed, which takes place publicly.'7 The evil of
war lies in the mind: the soldier who strikes down his enemy from
benevolence and pity acts in accordance with Christ's teaching.8
Augustine's scriptural exegesis may raise a sceptical eyebrow or an
4. My sketch of the just war theory is based primarily on the following sources: Augustine (texts
listed above, n. 3; and see. e.g., De la Brière 1930); Gratian, Decretum Causa XXIII (see, e.g.,
Hubrecht 1955); Rufinus, Summa decretorum ad Causa XXIII; Raymond of Pennaforte, Summa de
casibus ll.v.i 7—18; Alexander of Hales, Summa theologica III, §§466—70; Aquinas, ST, Ilallae, q. 40;
John ofLegnano, Tractatus de bello; Baldus de Ubaldis, Consilia V, cons. 439; Francisco de Vitoria,
De Indis sive de iure belli Hispanorum in barbaros, relectio posterior; Francisco Suârez, De triplice
uirtute theologica, tract. III. disp. XIII, 'De bello". For the earlier part of the period there is an
exhaustive study, citing numerous supplementary texts, in Russell 1975; for the later Middle Ages
see Keen 1965 and Tooke 1965. Grotius 1853 is a mine of historical information.
5. On early Church attitudes to war see, e.g., Bainton 1946; Russell 1975, Ch. 2.
6. Gratian, C XXIII, q. I, introd.: 'Omnis militia vel ob iniuriam propulsandam vel propter
vindictam inferendam est instituta — iniuria autem vel a propria persona vel a socio repellitur -
quod utrumque evangelica lege prohibetur.'
7. Ep. 138.2: 'Ista praecepta magis ad praeparationem cordis, quae intus est, pertinent quam ad opus,
quod in aperto fit' [quoted by Gratian, C XXIII, q. 1, canon 2; cf. Aquinas, ST, Ilallae, q. 40, art.
1, adz); cf. Augustine, Ep. 189.14 (quoted below, n. 8); Contra Faustum XXII. 76.
8. Contra Faustum XXII.74: 'quid enim culpatur in bello?... nocendi cupiditas, ulciscendi crudelitas,
inpacatus atque implacabilis animus, feritas rebellandi, libido dominandi, et si qua similia, haec
sunt quae in bellis iure culpantur' (quoted by Gratian, C XXIII, q. 1, canon 4; Aquinas, ST,
Ilallae, q. 40, art. 1, resp.; Alexander, Summa, III, §466); cf. Augustine, Ep. 189.14:'Sunt ergo ista
praecepta patientiae semper in cordis praeparatione retinenda, ipsaque benevolentia ne reddatur
malum pro malo semper in voluntate complenda est. Agenda sunt autem multa cum invitis
benigna quadam aspen täte plectendis... Ac per hoc si terrena ista res publica praecepta Christiana
custodiat, et ipsa bella sine benevolentia non gerentur, ut ad pietatis iustitiaeque pacatam
societatem viens facilius consulatur... Misericorditer enim si fieri posset etiam bella gererentur a
bonis, ut licentiosis cupiditatibus domitis haec vitia perderentur quae iusto imperio vel extirpari
vel premi debuerunt.'
The just war 773
outraged hackle;9 but it was gratefully accepted and piously parroted by

the medieval political theorists: by returning a negative answer to the
question 'Is soldiering always a sin?', Augustine made room for a morality
of warfare and a theory ofjust war.
Conditions of a just war

The theory states that X wars justly upon Vif and only if certain conditions
are satisfied. The conditions were arranged under formal headings; but
there was no uniformity, either of enumeration or of nomenclature, to the
arrangement. The most elaborate system is that of Alexander of Hales: 'In
order to determine if a war is just or unjust you must mark the authority
(auctoritas), the state of mind (affectus), the intention (intentio), the condition
(conditio), the desert (meritum), and the cause (causa). The state of mind and
the authority must be considered in the person of him who declares war;
the condition and the intention in the person of him who wages war; the
desert in the person of him who is warred upon; the cause in the person of
him for whom the war is waged.' 10 Raymond 11 and Baldus12 list five
conditions: where Alexander writes conditio they write persona, and his
meritum they express by res; his affectus becomes animus; and they omit
intentio.13 John of Legnano also has a list of five items: he has nothing
answering to affectus, intentio, or conditio, and he makes up his numbers by
somewhat obscure references to 'the enemy' and 'the law which allows the
war'. ' 4 Rufinus specifies four conditions: the declarer of the war must have
authority; the soldiery must fight with the right spirit and be men of the
9. E.g., on Matt 26.52 ('all they that take the sword shall perish with the sword'), he comments: 'Ille
utitur gladio qui nulla superiori ac légitima potestate vel iubente vel concedente in sanguinem
alicuius armatur' (Contra Faustum, ΧΧΙΙ.70, quoted by Gratian, C XXIII, q. 4, canon 36; Aquinas,
ST, Ilallae, q. 40, art. 1, ad 1). For the scriptural texts against war, and the standard way of dealing
with them, see, e.g., John of Legnano, De bello, Ch. 10; Suârez, De bello, 1.1—5; cf. Grotius
1853,1.ii.5-10.
10. Alexander, Summa, III, §466: 'Notandum autem ad hoc ut discernas quod bellum iustum sit vel
iniustum, auctoritatem affectum intentionem conditionem meritum et causam. Affectus et
auctoritas debent considerari in persona indicentis bellum; conditio et intentio in persona
peragentis bellum; meritum in persona sustinentis bellum; causa in persona pro qua agitur
bellum.'
11. Raymond, Summa, II. v. 17: 'Ut autem plane liqueat de bello, nota quod quinque exiguntur ad hoc
ut bellum sit iustum, scilicet persona res causa animus et auctoritas.'
12. Baldus, Consilia V, cons. 439: 'Ad bellum iustum requiruntur quinque: sunt persona res causa
animus et auctoritas.'
13. Unless animus embraces both affectus and intentio (see below, n. 19).
14. John of Legnano, De bello Ch. 76: 'Nam licita dicuntur [sc. bella] ratione indicentis, illius contra
quern, rei, et causae, et iuris permittentis.'
774 Politics
right type; the enemy must deserve — or be justly deemed to deserve - his
fate. Thus Rufinus ignores affectus and causa.15 Aquinas holds that 'for any
war to be just, three things are required. First, authority . . . Secondly, a just
cause. . . . Thirdly, it is required that the intention of the warriors be
correct...' 1 6 Aquinas omits conditio; his causa is Alexander's meritum; his
intentio perhaps embraces intentio, affectus, and causa. Finally, Suârez says that
'for war to occur honourably, several conditions must be observed, which
can be reduced to three heads: first, it must be declared by a legitimate
authority; secondly, there must be a just cause and title; thirdly, the proper
mean and proportion must be preserved in its inception, prosecution, and
victory'.17 Causa again answers to meritum; and Suârez' third condition is,
as it were, the external correlate to affectus and intentio.
Conditio
Of the two conditions Alexander lays on the soldiery, thefirst,conditio,
demands that they be secular, not clerical.18 The question of what part the
clergy might play in warfare looms large in the medieval discussions; but
its interest is purely antiquarian and theological.
Intentio and affectus

The second condition on the soldiery, and the first on the declarer of the
war, are psychological: just intentio demands that the soldiers do not fight
from cupidity or a desire for booty (Alexander has no objection to their
fighting for pay); and a just affectus excludes cruelty. Both conditions derive
15. Rufinus, Summa ad C XXIII, q. 2: 'Iustum bellum dicitur propter indicentem, propter belli-
gerentem, et propter eum qui bello pulsatur. Propter indicentem: ut ille qui vi bellum indicit vel
permictit huius rei indulgendae ordinariam habeat potestatem; propter belligerentem: ut ille qui
bellum gerit et bono zelo hoc fàciat et talis persona sit quam bellare non dedeceat; propter eum qui
bello fatigatur: ut scilicet mereatur bello lacerari vel si non meretur iustis tamen praesumptionibus
mereri putetur.'
16. Aquinas, ST, Ilallae, q.40, art. i.resp.: 'respondeo dicendum quod ad hoc quod aliquod bellum
sit iustum tria requiruntur. Primo quidem, auctoritas... Secundo, requiritur causa iusta. . . .
Tertio, requiritur ut sit intentio bellantium recta.'
17. Suârez, Debello, I.7: 'Ut bellum honestc fiat, nonnullae conditiones sunt observandae, quae ad tria
capita revocantur: primum, ut sit a légitima potestate; secundum, ut causa iusta et titulus; tertium,
ut servetur debitus modus et aequalitas in illius initio prosecutione et victoria.'
18. Alexander, Summa, III, §466: 'ut non sit persona clericalis'; cf., e.g., Raymond, Summa, II.v.17: 'ut
sit saecularis, cui licitum est fundere sanguinem, non autem ecclesiastica, cui est prohibitum'. The
more nuanced view in Gratian (C ΧΧΙΙΙ, q. 8; cf. Rufinus, Summa, ad C ΧΧΙΙΙ, q. 8) is expressly
rejected by Alexander, Summa, III, §470. Those authors who do not make conditio a requirement
onjustice still discuss the question: e.g., Aquinas, ST, Ilallae. q. 40, art. 2; q. 64, art. 4; q. 188, art.
3; John of Legnano, De bello, Ch. 82-7; Suârez, De bello, III.
The just war 775
from Augustine's emphasis on the mental aspects of warfare;19 and it is

tempting to dismiss them. For, as Grotius says, 'these things... may argue
a sin, but they do not make the war itself properly unjust': 20 cruelty and
greed, whether in the prince or in the private, are evil; for they will lead to
breaches of the laws of war and to violations of agreements governing its
conduct. But if a war is fought unjustly, it does not follow that it is unjustly
waged; and the theory of just war is a theory of warring, not of fighting.21
That is largely right. But two reservations should be made. First, if a war
cannot be fought jusdy (if, say, innocent non-combatants are bound to be
kdled), then perhaps it cannot be justly waged: a just war requires at least
the possibility of effectively upright intentions. Secondly, although the
cruelties and hatreds of a prince may be strictly irrelevant to the justice of
his war, his intentions, in a broader sense, are not; for his war aims
will affect the justice of his venture. But that issue perhaps falls under
Alexander's heading of causa.
Auctoritas
Anyone who declares war must have 'just authority' to do so. 22 'War
begins', according to Grotius, 'where the lawcourts end'; 2 3 and Aquinas
explains the condition of authority thus: 'first, the authority of a prince, at
19. On intentio Alexander, Summa, III, §466, quotes a text he ascribes to Augustine: 'Apud veros Dei
cultures etiam ilia bella pacata sunt quae non cupiditate aut crudelitate sed pacts studio geruntur,
ut mali coerceantur et boni subleventur' (Gratian, C XXIII. q. 1, canon 6, and Aquinas, ST,
Ilallae, q. 40, art. 1, resp., also quote the passage; but it is not found in our texts of Augustine); on
affectus Alexander quotes Augustine Contra Faustum, XXII.74 (above, n. 7) and Ep. 189.4 (in
Gratian, C XXIII, q. 1, canon 3). Cf. Raymond, Summa, II.v.17: 'animus, ut non fiat propter
odium vel ultionem vel cupiditatem ...'; Baldus, Consilia V, cons. 439: 'ut non fiat propter odium
vel insatiabiiem crudclitatem, sed propter caritatem iustitiam et oboedentiam, et ut sine per-
turbatione in securitate vivatur'. Raymond and Baldus seem to be thinking primarily of affectus.
20. Grotius 1853, U.xxii.17: 'Sed haec, ubi causa iustifica non deest, peccatum quidem arguunt, ipsum
tamen bellum proprie iniustum non faciunt.'
21. On the distinction between ius ad bellum and ius in bello see Walzer 1977, Ch. 2-3.
22. Augustine, Contra Faustum, XXU.75: 'Interest enim quibus causis quibusque auctoribus homines
gerenda bella suscipiant; ordo tamen ille naturalis mortalium paci accommodatus hoc poscit ut
suscipiendi belli auctoritas atque consilium penes principem sit...' (quoted by Gratian, C ΧΧΙΙ1,
q. I, canon 2; Alexander, Summit, III §466; Aquinas, ST, Ilallae, q. 40, art. 1, resp.). On the
controversies over the condition of authority, and its practical importance, see Keen 1965,
pp. 72-81.
23. Grotius 1853, Il.i.2: 'Ubi indicia deficiunt, incipit bellum.' The view that war is legal action
carried on by other means underlies the whole just war theory, it derives from Rome - see e.g.
Cicero, Off. I.xi.34: 'Cum sint duo genera decertandi, unum per disceptationem, alteram per
vim, cumque illud proprium sit hominis, hoc beluarum, confugiendum est ad posterius si uti non
licet superiore.' Cf., e.g., John of Legnano, Dr bello, Ch. 76: 'Cum ergo ab eo qui obnoxius est
iustitia haberi non potest, tunc licet bellum indicere'; Christine de Pisan, Fayttes of Armes, I.2:
'Wane & bataill whiche is made by iuste quarell is non other thing but right execucion of iustyce,
for to gyve the right there as it apperteyneth.'
77ό Politics
whose command the war is to be waged; for it does not belong to a private
person to start a war, since he can prosecute his rights in the court of a
superior'.24 John of Legnano writes: 'the prince alone is competent to
declare war on his own authority, since he has no superior to whom he may
25
have recourse for obtaining justice'. In short, X may war upon y only if
X is a sovereign state; for otherwise X has legal means by which it may and
must settle its claim against Y.
In medieval theory, all Christendom was one realm, of which the
Emperor was sovereign;26 and a few theorists drew the consequence, that
wars were just only if declared on imperial authority. 27 But the sov
ereignty of the Emperor was a polite fiction; and that doctrine would
render most medieval wars unjust.28 More liberal theorists allowed au
thority to the various princes within the Empire: 'it apperteyneth to none
to empryse warre or bataylle for ony maner cause but yf it be to prynces
souerayn lyke as emperours kynges dukes & other lordes terryens which
ben merely princypall heedes of luredictions temporall'. 29 (Vitoria derived
authority not from sovereignty, but from the Aristotelian principle that 'a
24. Aquinas, ST, Ilallae, q. 40, art. I, resp.: 'primo quidem, auctoritas principis, cuius mandato
bellum est gerendum. N o n enim pertinet ad personam privatam bellum movere quia potest ius
suum in iudicio superioris prosequi.'
25. John of Legnano, Df bello, Ch. 14: 'Soli ergo principi competit sua auctoritate, cum non habeat
superiorem ad quern recurret pro iustitia consequenda'; cf. Suârez, De bello, II.2: 'Princeps et res
publica imperfecta, et quicumque superiorem habet in temporalibus, non potest iuste bellum
indicere sine sui superioris auctoritate. Ratio est: huiusmodi princeps potest petere ius a superiore
suo; ergo non habet ius indicendi bellum, quia quoad hoc se habet ut privata persona.'
26. See, e.g., John of Legnano, De bello, Ch. 13: 'Omnes gentes fere quae oboediunt sacrae matri
ecclesiae sunt de populo Romano' (and therefore subject to the emperor). Cf. the remarkable
comment on the Hundred Years War by Jacob Meyer, Commentaria sive annales rerum Flandicarum
(quoted by Keen 1965, p. 75 n. 5): 'Bellum Anglicum exordium habuit quod omnium lon-
gissimum atque atrocissimum fuit; quodque per intervalla centesimum excessit annum. Seditio
potius domestica quam bellum dicendum. Christiana respublica unum regnum unaque domus
est; quaecumque in ea geruntur bella cum magno dedecore geruntur; nec, si verum fateamur,
bella sunt sed turpissimae seditioncs.'
27. See, e.g. Odofredus, Lectura Codicis in Cod. 11.47: 'Nullus poterit movere guerram et arma
portare sine licentia imperatoris' (quoted by Russell 1975, p. 46 n. 31). Many theorists also grant
auctoritas to the Pope or the Church; e.g. Raymond, Summa, II.v.17: 'auctoritas: ut si auctoritate
ecclesiae, praesertim cum pugnatur pro fide vel auctoritate principis'; Aquinas, ST, Ilallae, q. 188,
art. 3, ad 4: war may be waged 'solum auctoritate principum vel ecclesiae' (see also his Comm. in
Sent. Ill,dist. 30, art. 3, adS, quoted below, n. 60); Baldus, Consitia V, cons. 439: 'Dominus Papa
potest dare liccntias probo religioso ut utatur armis, non tamen ita quod effundat sanguinem sed
pro statu fidei et ecclesiae et pro salute patriae.' Alanus Anglicus (cited by Russell 1975, p. 140,
n. 41) appears to assign auctoritas to the Pope alone. The question raises complex historical issues.
28. See John of Legnano, Df bello, Ch. 13: 'Hodie quia sunt populi non recognoscentes superiorem de
facto, non requiritur superioris auctoritas cum non recognoscent. Immo tota die bella indicuntur
a populo contra populum nullo alio requisito.'
29. Christine de Pisan, Fayttes of Armes, 1.3.
The just war 111
republic must be self-sufficient';30 he also allowed that a republic might go

to war if, through the impotence or indolence of its sovereign, it could
not obtain legal redress.)31 The problem here is partly historical (for in
medieval Europe it was hard to determine with whom sovereignty lay);
but it was partly theoretical (for the concept of sovereignty required more
careful articulation).
The condition of authority was derived ultimately from Cicero; and
according to the Roman theory, on which Augustine drew, 'no war is held
to be just unless it is declared, unless it is announced, unless it concerns
wrongs for which redress has been demanded'.32 The requirement that
war must be formally declared found its way, via Isidore, into Gratian's
Decretum;33 but it was generally ignored by the medieval theorists,34 and
indeed it appears to be a requirement of decency rather than ofjustice - to
make war without giving notice is unsporting; but it does not follow that
the war itself is unjust.33
Meritum
Next comes meritum or, in Aquinas' terminology, causa; and 'whether a
war is just or unjust is judged principally from the cause of warring'. 36
Aquinas comments: 'secondly, a just cause is required, viz. that those who
are to be warred upon should deserve to be warred upon because of some
fault. Hence Augustine says: "Just wars are customarily determined as
30. Vitoria, De Indis, q. 2, §8; for the Aristotelian principle ('respublica debet esse sibi sufficiens') see
Politics i28ob33-s.
31. Vitoria. Df Indis, q. 2,§9; cf. Suârez. De bello, 11.1-2. Note that in Suârez' view (ibid. 11.1) auctoritas
is required only for aggressive wars: 'nam potestas se defendendi ab iniusto invasore penes omnes
datur.' (Raymond, Summa, Il.v.i 8. strangely states that auctoritas is not needed in wars fought pro
rebus repetendis or pro defensicme patriae, i.e., not in just wars.)
32. Cicero, Rep. III.xxiii.35 (' fragment preserved by Isidore, Etym. XVIIU.3): 'nullum bellum
iustum habetur nisi denumiatum, nisi indicium, nisi de rebus repetitis' (see below, n. 51); cf. Rep.
Il.xvii; Off. I.xi.36; for de rebus repetitis and the formal demand for redress see, e.g., Livy vii.9; x. 45
(cf. Philippson 1911, pp. 327-40).
33. Isidore, Etym., XVIIl.i.2: 'iustum bellum est quod ex praedicto geritur de rebus repentis aut
propulsandorum hostium causa' (quoted by Gratian, C XXIII, q. 2, canon 1).
34. But see, e.g., More, Utopia, p. 200 (the Utopians wage war only 'repetitis ac non redditis rebus').
When Gratian quotes Isidore (above, n. 34) he writes 'eduto' for 'praedicto', 'repetendis' for
'repetitis', and 'hominum' tot 'hostium': the second alteration at least may be significant.
35. Grotius 1853,111.111.5, asserts that'ut iustum ... bellum sit,.. .oportet... u t . . . publice decretum
sit'; and he cites the passages from Cicero and Isidore quoted above in nn. 32 and 33. But at ibid.
6-8 he allows that the requirement of public declaration is not absolute, commenting thus:
'verum etiam ubi ius naturae non praecipit talem interpellationem fieri, honeste tamen et
laudabiliter interponitur.'
36. Alanus Anglicus (quoted by Russell 1975, p. 128, n. 2): 'quod bellum sit iustum vel iniustum ex
causa bellandi praecipue iudicatur.'
778 Politics
those which avenge injuries, if a nation or state which is to be warred upon

has neglected to punish crimes committed by its people or to restore what
has been unjustly taken away."' 3 7 In general, X may war on y only if Y
has condoned some injury.38
To whom must the injury have been done? Evidently, X has a case
against y if the people of y have injured the people of X. Gratian adds that
'injury to associates (socii) is to be repelled by arms';39 for, as Ambrose says,
'he who does not defend his associate against injury when he can, is as much
to blame as he who inflicts the injury'.40 Aquinas concurs;41 and Suârez
specifies that injury 'against allies or friends' is a cause for war.42 More
generally, 'the prynce may iustly yf it please hym to ayde & helpe euery
prynce baron or other hys alye or frende or ony contre or londe yf he be
requyred in caas that the quarell be iuste'. 43 Thus if y injures Ζ and Ζ calls
on X, relying on a treaty or a bond of friendship or merely on hope, then X
has just cause to war on Y.
Suppose that y injures its own nationals: may X war upon y to avenge
injuries done to members of y? Are such 'humanitarian' wars sanctioned
by the medieval theory? The letter of the theory appears to countenance
them; and More's Utopians will go to war when, 'pitying a people
oppressed by tyranny, they may, for humanitarian reasons, free them from
the tyrant's yoke and bondage'. 44 But the spirit of the theory is against such
37. Aquinas, ST, Ilallae, q. 40, art. 1, resp.: 'Secundo requiritur causa iusta, ut scilicet illi qui
impugnantur propter aliquam culpam impugnationem mereantur. Unde Augustinus dicit (QQ.
in Hepl., IV. 10]: iusta autem bella ea dctiniri soient q uae ulciscuntur iniurias, si qua gens vel civitas
quae bello petenda est vel vindicare neglexerit quod a suis improbe factum est vel reddere quod
per iniurias ablatum est.' (I have given Augustine's text: Aquinas quotes it in slightly different
form; it is badly garbled by Gratian, C XXIII, q. 2, canon 2. The text continues: 'sed etiam hoc
genus belli sine dubitatione iustum est, quod deus imperat' - cf. Contra Faustum, XXII.74, 75.)
38. Rufinus' view, that the enemy need only be thought to deserve attack (above, η. 15), is unusual;
for discussion see Vitoria, De Indis §20.
39. Gratian,CXXIII, q. 3, p. canon 10: 'Eccequod nonnumquam est obviandum perversis,et iniuria
sociorum armis est propulsanda. ut et malis adempta facultas delinquendi prosit et bonis optata
facultas libère consulcndi ecclesiae ministretur.'
40. Ambrose, Off., I.xxxvi: 'qui enim non repellit a socio iniuriam si potest, tarn est in vitio quam ille
qui facit' (quoted by Gratian, C XXIII, q. 3, canon 7).
41. Aquinas,5T, Ilallae,q. 188, art. $,adi: 'aliquis potest non resistere malo dupliciter . . . Alio modo,
tolerando patienter iniurias aliorum. Et hoc ad imperfectionem pertinet, vel etiam ad Vitium, si
aliquis potest convenienter iniurianti resistere' (he quotes Ambrose, Off., I.xxvii, ciïed by Gratian,
C XXIII, q. 3, canon 5).
42. Suârez, De bello IV.3, says explicitly that iniuria contra foederatos sive amicos provides just cause for
war. See also More, Utopia, p. 200, quoted below, n. 44.
43. Christine de Pisan, Fayttes of Armes, I.4.
44. More, Utopia, p. 200: 'non temere capessunt nisi quo aut suos fines tueantur aut amicorum terris
infusos hostes propulsent aut populum quernpiam Cyrannide pressum miserati (quod humanitatis
gratia fäciunt) suis viribus tyranni iugo et Servitute libèrent." Cf. Grotius 1853, ll.xx.40:
The just war 779
interventions; and Suârez is right when he says that 'what some assert, that
sovereign kings have power to punish injuries over the whole world, is
altogether false, and confounds all order and distinction of jurisdictions'.45
A war can redress wrongs only if wrongs have been done. Vitoria
affirms that 'there is one and only one just cause for waging war, viz. an
injury received'; and he adds, correctly, that 'this is the opinion of all the
doctors'.46 Francis Bacon rejected that view: 'neither is the opinion of some
of the schoolmen to be received, that a war cannot justly be made, but upon
a precedent injury or provocation. For there is no question, but a just fear
of an imminent danger, though no blow be given, is a lawful cause of
war'.47 If y threatens to attack X, may not X indulge in 'anticipatory
defence' and attack Y? (Thus Israel attacked Egypt in 1967.) If Y threatens
to occupy Ζ to the detriment of X, may not X indulge in 'anticipatory
breach of neutrality' and attack Z? (Thus was Iceland invaded in 1940.)
The schoolmen might have replied that if Y puts X in 'imminent danger',
then Y has thereby injured X, so that 'anticipatory' wars can be justified by
Vitoria's cause of'injury received'.48 But there is a better reply: 'imminent
danger' does not warrant war; the appropriate (and prudent) response to a
threat is mobilisation and military vigilance, for that both guards against
the danger and preserves the possibility of peace.49
What injuries give cause for war? An iniuria is the violation of a right;
and the general answer to the question is given by Grotius: 'the sources of
war are as many as those of legal actions'. 50 Augustine's reference to the
'Sciendum quoque est reges et qui par regibus ius obtinent ius habere poenas poscendi non tantum
ob iniurias in se aut subditos suos com missas, sed et ob eas quae ipsos non peculiariter tangunt sed
in quibusvis personis ius naturae aut gentium immaniter violant' (Grotius cites Augustine, De civ.
Dei, V. 1, in support of his view; but he wrenches a sentence out of context). On humanitarian
wars see Walzer 1977, Ch. 6.
45. Suârez, De bello, IV. 3: 'Quod quidam aiunt, supremos reges habere potestatem ad vindicandas
iniurias totius orbis, est omnino falsum et confundit omnem ordinem et distinctionem
iurisdictionum.'
46. Vitoria, De Indis, q. 3, §13: 'Unica est et sola causa iusta inferendi bellum iniuria accepta', citing
Augustine and Aquinas, he remarks that that 'est determinatio omnium doctorum;' cf., e.g.,
Suârez, Dr bello, IV. 1: 'Causa haec iusta ct sufTicicns est gravis iniuria quae alia ratione vindicari
aut reparari nequit;' Grotius 1853, II. i. 1: 'Causa iusta belli suscipiendi nulla esse alia potest nisi
iniuria' (citing Augustine and Livy, i. 32).
47. Bacon, Of Empire, quoted by Walzer 1977, p. 77.
48. Cf. Grotius 1853, II. i. 1: 'dantur autem actiones aut ob iniuriam non factam aut ob factam' -
iniuria non facia is nevertheless iniuria.
49. For considerations pro and contra anticipatory wars see Walzer 1977, pp. 74-85; Brownlie 1963,
pp. 257-61.
50. Grotius 1853, II. i. 2: 'ac plane quot actkmum forensium sunt fontes, totidem sunt belli; nam ubi
iudicia deficiunt, incipit bellum' (see above, n. 23). Claims not actionable at law cannot be
prosecuted in war: 'sicut in foro exigi non potest, ita nec armis deposci' (ibid.. 11. xxii. 16).
78ο Politics
'customary determination' ofjustice in war suggests the Roman formula,

which was adopted by Isidore and frequently repeated: 'a war is just if it is
waged . . . for the restitution of goods or for defending one's country'. 51
Defence is an elastic notion; but here it should be narrowly construed: if
Y invades the territory of X, then Y injures X and X has just cause for war.
Under twentieth-century international law (as defined by the Kellogg-
Briand Pact and the United Nations Charter), self-defence is the only
legitimate excuse for war. 5 2 That excuse, or right, was universally recog
nised by the medieval theorists, who grounded it on the domestic analogy:
as a private person may violently defend himself against an assailant,
so a state may repel an invader by force of arms. 53 The right of self-
defence carries a proviso and implies certain restrictions. The proviso is
that the invader be acting unjustly: there is no right to defence against
legitimate attack. The restrictions are that the defence should be immediate
(incontinenti and not ex intervallo) and that it should observe the rule of
proportionality (moderamen inculpatae tutelae): self-defence cannot justify
war unless it is a response to a present danger; and it cannot justify more
than the minimal force needed to repel the attack. 54 .
51. Isidore, Etym. XVIII.i.2 (quoted above, n. 33). Isidore cites Cicero, Rep., III.xxiii.35 (see above, n.
32): 'ilia iniusta bella sunt quae sine causa suscepta; nam extra <quam iniurias?> ulciscendi aut
propulsandorum hostium causa bellum gcri iustum nullum potest.' Cf., e.g., Raymond, Summa,
II.v. 17: 'res, ut sit pro rebus repetendis et pro defensione patriae'; Baldus, Consilia V, cons. 439: 'ut
fiat pro rebus repetendis et pro defensione patriae'; John of Legnano, De bello Ch. 76: 'septimum
[sc. genus bellorum] dicitur necessarium et licitum, quod fâciunt ftdeles iuris auctoritate se
defendendo contra ipsos invadentes; nam vim vi repellere licet'; More, Utopia, p. 200: 'Auxilium
gratificantur amids non semper quidem quo se defendant sed interdum quoque illatas retalient
atque uldscentur iniurias.' Christine de Pisan, Fayttes of Armes, 1.4, distinguishes threejust causes:
'The first... is for to susteyne right & iustice. The second for to withstonde the euyl that wold
defowle grieve & oppresse the londe the contrée & the people and the thirde for to recoure londes
seignoryes or other thynges by other taken & vsurped by iniuste cause.' Her discussion seems to
imply that the first cause in fact embraces the other two.
52. See Brownlie 1963, pp. 74-92: 'apart from self-defence [and police action taken under the U. N.
Charter] war, and indeed any use of armed force, is prohibited for all purposes, and is placed
outside the legal competence of states' (p. 91)·
53. On the right of self-defence see, e.g., Aquinas, ST, Ilallae, q. 64, art. 7; for self-defence as a
ground for war see also ibid. q. 40, art. 1, resp.: 'gladio bellico ad eos [i.e. principes] pertinet
rempublicam tueri ab exterioribus hostibus'; ibid. q. 188, art. 3, resp.: 'propter defensionem divini
cultus et publicae salutis.' Cf. Ambrose, Off., I.xxvii: 'fortitudo quae . . . in bello tuetur a barbaris
patriam . . . plena iustitia est' (quoted by Gratian, C XXIII, q. 3, canon 5; Aquinas, ST, Ilallae,
q. 188, art. 3, ad 1 [see above, n. 41]); Augustine, Ep. 87: 'non quod hos persequantur, sed quia se
defendant' (quoted by Gratian, C XXIII, q. 3, canon 3).
54. On the limits of self-defence see esp. John of Legnano, De bello, ch. 111-120: cf. e.g. Vitoria, De
Indis q. 2, §3; and the Decretalist Johannes Teutonicus, quoted in Russell 1975, p. 132 η. 11.
The just war 781
The medieval theory did not regard self-defence as a necessary condition

for just warring: aggressive war, in their eyes, might be just; 35 for clearly Y
may do injury - and do injury to X - without invading X's territory.
Suârez remarks that 'various kinds of injuries may be causes of just war, but
they reduce to three heads: first, if a prince appropriates someone else's
goods and will not restore them; secondly, if he denies the common rights
of nations without reasonable cause (e.g. the right to passage,56 common
trade, etc); thirdly, severe harm to reputation or honour'. 57 It would not
be profitable to attempt a subtler classification of actionable injuries; but it
is worth observing that 'self-help' in the redressing of wrongs has always
been the most common reason for going to war. Nor is it an implausible
reason: the modern law, which implicitly prohibits war de rebus repetendis,
is not a principle of justice.
Causa
Alexander's final condition on justice in war is not easy to interpret: 'there
must be a just cause, which is the support of the good, the coercion of the
bad, peace for all'.58 The text Alexander proceeds to cite is used by Aquinas
to explicate his claim that 'the intention of the warriors must be correct,
viz., they should intend to promote good or to avoid evil'. 59 Those
passages have been taken to support a doctrine of'ameliorative' warfare: X
55. So, explicitly, Suârez, Df bello I.5: 'bellum etiam aggressivum non est per se malum, sed potest
esse honestum et necessarium.'
56. Augustine invoked this right tojustify the Israelites' attack on the Amoritcs (Num. 21.21 -5), QQ.
in Hept., IV.44: 'notandum est sane quemadmodum iusta bella gerebantur. Innoxius enim
transitus negabatur, qui iure humanae societatis aequissimo patere debebat' (quoted by Gratian,
C XX111, q. 2, canon 3); the text was frequently repeated - see esp. Bouvet. Tree of Battles IV.61.
On the comparable Corfu Channel case of 1946 see Brownlic 1963, pp. 283-9,
57. Suârez. De hello IV.3: 'Varia esse iniuriarum genera pro iusti belli causa, quae ad tria capita
revocantur: unum, si princeps res alterius occupet ac nolit restituere; alterum, si neget communia
iura gentium sine rationabili causa, ut transitum viarum, commune commercium, etc.; tertium,
gravis laesio in fama vel honore.' Suârez does not mention self-defence, since he is dealing only
with aggressive wars.
58. Alexander, Summa, III, §466: 'in persona illius pro quo pugnatur attendenda est causa iusta, quae
est sublevatio bonorum, coertio malorum, pax omnium;' cf. Raymond, Summa, II. v. 17: 'causa, si
propter necessitatem pugnetur ut per pugnam pax acquiratur;' Baldus, Ccmsilia V, cons. 439: 'ut
propter necessitatem pugnetur ut propter pugnam pax acquiratur'. The slogan that 'war is for the
sake of peace' is associated with Augustine (Ep. 189.5: 'non enim pax quaeritur ut bellum
excitetur, sed bellum geritur ut pax acquiratur' - cited by Gratian, C XXIII, q. 1, canon 2;
Aquinas, ST, Ilallae, q. 40, art. 1, ad 3; cf. Augustine, Ep. 229; Dr civ, Dei, XIX. 12); it derives
from Aristotle, Nie. Eth. 1177l'5. to which Baldus alludes, and ultimately from Plato (Laws 628E,
803D; see also Cicero, Off. I.xi.35; cf. Defourny 1977).
59. Aquinas, 5 T, Ilallae, q. 40, art. 1, resp.: 'ut sit intentio bellantium recta - qua scilicet intenditur vel
ut bonum promoveatur vel ut malum vitetur." The Augustinian text is cited above, n. 19.
782 Politics
wars justly upon Yif the overall consequences of war are better, or less bad,
than the overall consequences of abstaining from war. In such cases, justice
is determined not by the antecedent misdeeds of Y (there may be none) but
by the anticipated results of going to war.60 But neither Alexander nor
Aquinas is concerned with ameliorative war; both texts impose a condition
on just wars additional and not alternative to the condition of meritum or
just cause. Alexander seems to mean, in part, that X may go to war only if
good men have been injured by bad men - a pernicious principle. Aquinas
may mean, in part, that X may go to war only if redressing the injuries by
force of arms will not do more harm than leaving the injuries unredressed.
Trivial injuries will not warrant war; as Vitoria notes, 'one must beware
lest greater evils follow from the war itself than are avoided by the war'. 61
Alexander's reference to peace may be intended to advert to the war
aims of the belligerents, although it determines them only in the vaguest
terms. In fact, the legal analogy implicitly specifies the goals of legitimate
warfare, as Vitoria saw most clearly: if X wars justly on Y, then X may
claim, first, compensation for those injuries which were the casus belli;
secondly, compensation for any losses incurred in the course of the war;
and thirdly, punitive damages against the malefactors of Y who pre-
cipitated the war.62 The Utopians, it is true, 'seek only this in war - that
they may obtain that which, had they had it before, they would not have
gone to war'. 63 But the Utopians were foolish, and they sought less than
justice allows.
Those rights effectively determine the legitimate war aims of X: a prince
who aims at the annihilation of his enemies, or a belligerent who demands
their unconditional surrender, is not engaged in a just war. At this point,
Augustine's psychological requirement, that the belligerent not be moved
by hatred, and the legal condition, that the action be proportionate to the
offence, conspire to set a further limit on the scope of legitimate warfare.
60. A third text, contrasting iustitia and bonum, does seem to allow ameliorative wars; see Aquinas,
Comm. in Sent. Ill, dist. 30, art. 4, ad 8: 'ccclesia hoc modo movet bella ad versus iniquos, vel ut
iustitiam faciat vel ut maius malum evitet aut maius bonum inducat.' But this view is
idiosyncratic.
61. Vitoria, Dc Indis, §37: 'Oportet cavere ne ex ipso bello sequantur maiora mala quam vitentur per
ipsum bellum;" cf. Grotius 1853, H.xxiv, title: 'monita de non temere etiam ex iustis causis
suscipiendo bello' (ibid. 1 : 'contra enim evenit ut plerumque magis pium rectumque sit de iure suo
cedere'). This condition will doubtless rule out wars to avenge libel, of the sort envisaged by
Suârez (above, n. 57).
62. Vitoria, De Indis, q. 4, §§15-19: cf., e.g., Suârez, De bello, IV.5.
63. More, Utopia, p. 202: 'Hoc unum illi in bello spectant, uti id obtineant quod si fuissent ante
consequuti bellum non fuerant illaturi.'
The just war 783
Issues associated with the theory of the just war

Such, in outline, is the theory ofjust war: Xjustly wars upon Y if and only
if each of the conditions described above is satisfied. Some theorists ex-
plicitly add that the war must be 'necessary'; that is to say, that there are no
means short of war by which X can achieve its aims.64 But the requirement
of 'necessity' was implicit in every statement of the theory.
About the theory clustered a variety of related considerations: the issue
of conscientious objection was debated; the duty to obey military orders
was discussed; rules of war, dealing with the rights of non-combatants, the
legitimacy of technologically sophisticated weaponry, the use of deception
and espionage, the propriety of fighting on holidays, were elaborated and
analysed; trials of war criminals were held and argued over; the matter of
the spoils of war was at once a theoretical and a practical issue; the crusades
posed problems of their own. Those questions were raised and answered in
the context of the theory ofjust war,63
The significance of the theory

But how significant was the theory? Did it prevent a single bloody battle?
Did it curb a single ambitious prince? Cynics will observe, with Lactantius,
'how far expediency departs from justice: we are taught that by the
Romans - for they solemnly declared wars and legally inflicted injuries and
eternally desired and appropriated other men's goods, and thereby gained
for themselves possession of the whole world'. 66 Theory is impotent: 'this
present werke by somm enuyous myght be reproched sayeng that it is but
ydlenes & losse of tyme'. 67
In general, cynics underestimate the effect of theory on action; but their
dismissive suggestions can be refuted only by detailed historical research.
One episode in that history is the notorious assault on Constantinople by
the soldiers of the Fourth Crusade. Contemporary chroniclers record the
debates which preceded their attack; and it is plain that, whatever else was in
64. E. g. Raymond and Baldus, quoted above, n. 58; Nicholas I, Responsa ad consulta Bulgarorum 46: 'si
nulla urgeat nécessitas, non solum quadragesimali tempore sed etiam omni tempore a proeliis
abstinendum' (quoted by Gratian, C ΧΧ111, q. 8, canon 15); Suârez, De bello IV. 1 (quoted above,
n. 46); cf. Livy, ix.i: 'iustum bellum quibus necessarium, et pia arma quibus nulla nisi in armis
relinquitur spes."
65. On these questions see Russell 1975; Keen 1965; on the special problems of the Crusades see
Riley-Smith 1977.
66. Lactantius, Inst., VI.xi.4: 'Quantum autem ab iustitia recédât utilitas populus ipse Romanus
doeet, qui per fetiales bella indicendo et legitime iniurias faciendo semperque aliéna cupiendo et
rapiendo possessionem sibi totius orbis comparavit.'
67. Christine de Pisan, Fayttes of Armes 1.2.
784 Politics
their minds, the Crusaders were concerned with, and moved to action by,
the very considerations ofjustice which the theorists were debating. Even
in the heat of a campaign, the just war theory could affect the course of
events.68 But the cynic may remain unimpressed; for, having decided to
take Constantinople, the Crusaders ravaged the city with unbridled
ferocity.
68. On this episode see Schmandt 1975.

XI
THE DEFEAT, NEGLECT AND
R E V I V A L OF S C H O L A S T I C I S M
42
THE ECLIPSE OF MEDIEVAL LOGIC
A revised dating of the eclipse

The view that the insights and developments of medieval logic were
eclipsed during the fifteenth century by a humanist, rhetorically-oriented
logic has long been popular, but it needs considerable revision and modi-
fication. In what follows I shall first give a brief account of what happened
to the writing, teaching, and publication of logical works in the medieval
style, by which I mean those which discuss such topics as consequences, in-
solubles, exponibles, and supposition. I shall then examine in more detail
what was actually said about certain medieval doctrines in the late fifteenth
and sixteenth centuries in order to indicate both where logicians of the
period had something new to contribute, and where there were departures
from medieval doctrines which cannot be attributed to new logical in-
sight.1 My conclusion will be that medieval logic as a living tradition did
largely disappear, but that the eclipse dates from about 1530 (in so far as a
specific date can ever sensibly be offered) rather than the mid fifteenth
century.
Fifteenth-century logicians
After the death of Paul of Venice in 1429, the fifteenth century did not give
rise to much important logical writing. There were various logicians in
Italy who deserve mention for their contributions to logic in the medieval
style, including Domenico Bianchelli (Menghus Blanchellus Faventinus),
who wrote a long commentary on Paul of Venice's Logica parva; Paul of
Pergula, who wrote on Ralph Strode's Consequentiae as well as producing
his own Logica; and Gaetano di Thiene, who wrote on Strode, William
Heytesbury, and Richard Ferrybridge. The latter thinkers all formed part
of the logic curriculum at Padua where both Paul of Pergula and Gaetano
1. For further details about the period as a whole, and for some of the doctrines mentioned below,
see Ashworth 1974a, and Risse 1964. For a bibliography of primary sources, see Risse 1965. For a
bibliography of secondary sources, see Ashworth 1978.
788 The defeat, neglect, and revival of scholasticism
taught. Outside Italy we find a few lesser figures such as John Heynlyn
(Johannes de Lapide), author of commentaries on Aristotle's logical works
and a treatise on exponibles, who between 1446 and 1478 studied and
taught at Heidelberg, Leipzig, Louvain, Basle, Paris, and Tübingen. Per-
haps the most successful commentator was the Thomist John Versor (d.
ca. 1480) whose commentary on Peter of Spain was first printed in 1473
and was reprinted in Cologne as late as 1622. There were many other
commentators on Aristotelian and scholastic logical writings, but their
work seems to have had little effect on subsequent developments.
The end of medieval logic outside France

In most European countries work on logic in the strictly medieval style
came to an end during the sixteenth century, though, as will be noted
below, medieval doctrines survived in textbooks. In Italy medieval texts
continued to be printed and read, but there was no important logician who
genuinely belongs to the medieval tradition. In Poland the tradition
produced its last original thinker with John of Glogow (ca. 1445-1507)
who taught at Cracow and wrote a commentary on Peter of Spain. There
were some good synthesisers in Germany in the early years of the sixteenth
century, notably Jodocus Trutvetter (Isenachensis) (d. 1519) who taught at
both Erfurt and Wittenberg, and Johann von Eck (1486-1543), a noted
humanist and theologian as well as a good logician. England presents a
particularly dreary picture. The 1483 Logici published in Oxford shows
some signs of having been organised by one man, though it is largely
composed of earlier writings. But the popular Libellus sophistarum ad usum
Cantabrigiensium, published four times between 1497 and 1524, and the
Libellus sophistarum ad usum Oxoniensium, published seven times between
1499 and 1530, were both unadorned reprints of early-fifteenth-century
manuscript collections, the actual writing having been done during the
fourteenth century. 2 It is no wonder that humanism triumphed so easily in
the English universities.
2. Pollard and Redgrave 1976 give the title 'Logici' to a work which is in fact untitled. It contains
much of Paul of Venice's Logica parva (without acknowledgement) together with treatises
relating to the separate parts of Aristotle's Organon. It has only two treatises — Swineshead's
Insolubilia and Bradwardine's Proportiones — in common with the Libelli sophistarum. Despite the
judgement of Pollard and Redgrave that the two Libelli have 'essentially the same text', the
Cambridge text contains six treatises which are not in the Oxford text, and the Oxford text
contains five treatises which are not in the Cambridge text. Nor are the ten treatises they have
(more or less) in common fully identical. James McConica has pointed out to me that at least one
of the surviving copies bears ownership marks which indicate use throughout the sixteenth
century. For further discussion and complete references, see Ashworth 1979. It should be noted
The eclipse of medieval logic 789
The resurgence of medieval logic in France and Spain

In France, however, we find a renaissance of medieval work which lasted
from about 1490 to about IS20 and which strongly influenced the Spanish
universities, where medieval logic flourished into the 1530s. A possible
starting point for this logical renaissance is the royal decree of 1474 which
forbade the study of nominalism at the University of Paris. So far from
crushing nominalism, the decree, which was rescinded in 1481, aroused
fresh interest in the writings of William Ockham, John Buridan, Albert of
Saxony, Marsilius of Inghen, Peter of Ailly and John Dorp, and this led in
turn to a great increase in new writings on standard medieval topics. The
authors involved are far too numerous to list here; I will mention just a few
of the more outstanding names. The most distinguished of the earlier
writers were the Scotist Peter Tartaret and the nominalist Thomas Bricot
(d. 1516), both of whom completed their logical work before 1500. In the
sixteenth century the leading figure was the Scotsman John Major
(1469-1550), who attracted a brilliant and diverse group of men to the
college of Montaigu where he taught from 1505 to 1517. (He was in Paris
again from 1525 to 1531, but the earlier period is the important one.) The
group included several Scotsmen, several Frenchmen, and a large number
of Spaniards, among whom were Gaspar Lax and Domingo Soto. Apart
from Pierre Crockaert (Peter of Brussels) (d. 1514), who became a leading
Thomist, and Soto, also a Thomist, though eclectic, they tended to be
nominalist in orientation. Other Spaniards working in Paris during this
period included Jeronimo Pardo (d. 1505), Juan Dolz and Fernando de
Enzinas. Most of the Spaniards returned to Spain, and under their leader-
ship the teaching of logic at Alcalâ and Salamanca reached great heights.
A study of the publication of logic texts in France and Spain provides
evidence of the richness of the logical work done in these countries in the
first decades of the sixteenth century, as well as revealing the sudden decline
which occurred after 1530. Leaving aside the publication of such medieval
authors as Buridan and Ailly, and leaving aside compendia of various sorts
including commentaries on the parva logicalia, I know of the following
texts. There are twenty separate treatises on terms, by well-known men
such as John Major as well as by such lesser figures as Gilbert Crab and
Gerard Columel. The only treatises on terms published after 1530 are by
that all my claims about the number of times logic texts were printed are subject to revision. On
the one hand, we know there were sixteenth-century editions of which no copies seem to have
survived; on the other hand, copies of hitherto unknown editions are frequently discovered.
Ferdinando de Enzinas (which first appeared in 1533) and Hermosilla

(1553)· There are nine treatises on exponibles; apart from Soto's treatise
(1529) the last is by George Lokert (ca. 1522). There are ten treatises on
insolubles; again, apart from Soto's (1529), the latest is by Celaya (ca. 1518),
and the same is true of the eight treatises I know of on obligations. O f the
five treatises on consequences, the latest is by Lax (1532). Various works by
Enzinas and Lax were reprinted in the 1530s, but only Soto was reprinted
after 1540. A much revised edition of his Introductiones dialecticae or Sum-
mulae, a collection of virtually independent treatises, appeared in 1547
and was reprinted several times.
The decline after 1530

The impression one gets of a sudden decline in medieval logic after 1530 is
considerably strengthened by an investigation of the publication of
medieval authors and of commentaries on Peter of Spain between 1530 and
1600. Authors who contributed most to the development of late medieval
logic-such men as Buridan, Strode, Heytesbury, Peter of Ailly, Albert of
Saxony, and Marsilius of Inghen - drop out of the picture altogether.
Walter Burley's Super artem veterem expositio was published once during
this period (Venice 1541) as was Ockham's Summa logicae (Venice 1591).
John Duns Scotus' In universam Aristotelis logicam quaestiones was published
twice (Venice 1586 and 1600). The only popular work was the Logica parva
of Paul of Venice, which was published at least eight times in Venice
between 1535 and 1580. Of the late fifteenth-century commentaries on
Peter of Spain, two appeared in print: Versor's once in Naples (1577) and at
least five times in Venice between 1550 and 1593, Tartaret's three times in
Venice, in 1571, 1591 and 1592. The only new commentaries on Peter
of Spain were those of Agostino Sbarroya (treatises 1 and 4, Seville
1533); Pedro Sanchez Ciruelo (Salamanca, 1537); Alphonso de Veracruz
(Mexico, 1554, 1562, Salamanca 1561, 1562, 1569, 1572, 1573, 1593); and
Thomas de Mercado (Salamanca 1571). In addition we find Peter of Spain's
Tractatus de locis dialecticis cum Versorii Parisiensis annotationibus included in
the works of Chrysostom Javellus (Lyon 1580). Somewhat later the
Gymnasium Speculativum edited by F. Augustinus Gothutius (Paris 1605)
included tracts 1, 2, 4, 5, 8, and 9 of Peter of Spain, together with such
works as Josse Clichtove (1472-1543) on terms, Tartaret on the Posterior
Analytics, and Thomas of Erfurt's Grammatica speculativa, which was attri-
buted to Duns Scotus.
The survival of medieval logic in sixteenth-century textbooks

Despite the enormous changes signalled by the record of publication in the
sixteenth century, medieval doctrines were to some extent retained in the
textbooks which were written during that period, particularly in Spain.
Two important Spanish authors, who appeared on the Jesuit ratio studiorum
of 1595, were Francisco de Toledo whose Introductio in dialecticam Aristotelis
(Rome 1561) was reprinted more than twenty times, and Pedro de Fonseca
whose Institutionum dialecticarum libri octo (Lisbon 1564) was printed at least
fifty-three times up to 1625. Both works included discussions of such
medieval elements as consequences and supposition, as well as the newly
emphasised traditional subject matter of categorical propositions, syllog-
isms, and fallacies. Similar authors are to be found in Italy, including the
Dominican Chrysostomjavellus (d. ca. 1538), and Ludovico Carbo, who
taught at Perugia in the second half of the century and published his
Introductiones in logicam in 1597. They both discussed consequences, sup-
position, and insolubles in their textbooks. Even in England echoes of
medieval doctrines are to be found in the works of John Seton (d. 1567) and
John Sanderson (d. 1602), though the level of discussion is much lower than
that prevailing on the continent then.
Non-scholastic sixteenth-century logic

Most sixteenth-century logical writings bear no trace of specifically
medieval doctrines and developments, however. They fall roughly into
four overlapping categories: humanist logic, Ramist logic, Aristotelian
textbooks, and commentaries on the Greek Aristotle. The figure most
directly responsible for the spread of humanist logic in the sixteenth
century was Rudolph Agricola (1444-85), whose De inventione dialectica
circulated in manuscript for many years before it was printed in 1515.
Agricola influenced Philip Melanchthon, who wrote a large number of
simplified textbooks, including at least three logic texts for use in German
schools; he also influenced Johann Sturm, Johannes Caesarius, and Bartho-
lomaeus Latomus, all of whom taught Agricola's work at Paris after it was
printed there in 1529. Peter Ramus (Pierre de la Ramée) developed this
tradition further. Francis Titelman's Compendium dialecticae (1535) is a good
example of the straightforward Aristotelian textbooks. Among the com-
mentators on Aristotle the three Italians, Agostino Nifo, J. F. Burana, and
Jacopo Zabarella are perhaps most noteworthy.
Late developments: the properties of terms

Against the background of this survey of authors and their works, I want
now to examine the fate of the medieval doctrines themselves during the
sixteenth century. About supposition and the distinctively medieval treat
ment of terms there is little to say. In the first decades of the sixteenth
century these doctrines were discussed in great detail; in the later textbooks
they were merely summarised. Appellation as a property of terms either
ceased to be mentioned, or was mentioned only with scorn, as for instance
by Augustin Huens (i 521-78), 3 but otherwise there was no obvious
novelty in the nature of the theories presented. The one original develop
ment found in the works of Parisian logicians and later satirised by Vives
concerned the use of 'a' and ' b ' as special signs of supposition, to be used
especially in the analysis of such propositions as ' O f every man some
donkey is running' and 'Every man has a head', which posed special
problems for the theory of supposition. 4 A simple example of how the
signs were used is 'Every man is b animal' which, unlike 'Every man is
animal', signifies, by virtue of the special sign 'b', that every man is
identical to one and the same animal.
Exponibilia
The discussion of exponibles exhibits more significant advances. 5 In the
first part of the century we find treatises which in their detail, organisation,
and clarity mark a great improvement over the rather muddled discussions
found in Paul of Venice. If we turn to the accounts presented by such later
authors as Francisco de Toledo and Pedro de Fonseca, however, we find
three great changes. First, among the exponibles only exclusive, exceptive,
and reduplicative propositions are discussed; propositions employing such
exponible terms as 'incipit' and 'desinit' are ignored. Second, the analyses
offered of these propositions are much simpler than those that had been
developed by medieval logicians. For instance, the reduplicative 'Every A.
inasmuch as it is Β is C is no longer analysed as 'Every A is B, and every A is
C, and every Β is C, and if something is Β it is C but simply rewritten as
'Every A because it is Β is C Third, there is none of the characteristic
concern of the scholastic logicians with especially difficult cases for the
suggested analysis, or with such issues as the relationship between spoken
or written exponibles and the corresponding mental propositions.
3. Ashworth 1974a, p. 97·

4. See Ashworth 1978a; Guerlac 1979.
5. See Ashworth 1973.
Consequences
The changes in the discussion of consequences follow a similar pattern. At
the beginning of the century we find not just detailed analyses of pro-
positions and arguments but also a discussion of wider issues such as the
definition of a valid inference. We also find some relatively original work,
particularly that concerned with the distinction between 'illative' con-
ditionals, in which it is impossible for the antecedent to be true when the
consequent is false, and 'promissory' conditionals, in which truth demands
only that the antecedent not be true when the consequent is false. In the
later textbooks, however, there is no discussion of wider issues, there is
nothing original, and there are clear classical influences. For instance, the
textbooks contain the five Stoic indemonstrables, two of which (modus
ponendo ponens and modus tollendo tollens) were found in medieval works,
but three of which (two concerning strong disjunction and one concerning
negated conjunction) were not. The standard list of consequences narrows
to consist mainly of those concerned with truth and modality, and they are
presented without analysis or comment.
Obligationes and insolubilia

The treatises of the period have not yet been studied in sufficient depth to
say much about the theories of obligationes presented in them. In the latter
part of the century only Cardillo de Villalpandeo discussed obligationes in
his Summa Summularum (1557). The companion treatises on insolubilia are
better known, however, and here we can definitely say that at least one
original theory is to be found in the early period. Thomas Bricot took
Roger Swineshead's solution, which was adopted by Major and Soto
among others, and revised it in such a way as to avoid the consequences that
two contradictories can be false and that there can be valid inferences with a
true antecedent and a false consequent.6 His view was that an affirmative
proposition is true if and only if it meets both of two conditions, namely
that it signifies things to be as they are and does not falsify itself, whereas a
negative proposition is true if and only if it meets one of two conditions,
namely that it signifies things not to be as they are not, or that its
contradictory falsifies itself. Thus 'This is false' said of itself is false, but 'This
is not false', said of the first proposition, is true. Other popular solutions
were drawn from Ailly, Ockham, and Paul of Venice's Logica parva. By the
second part of the century most writers on logic had ceased to refer to
6. See Ashworth 1977b.

insolubles at all, and significant chapters are to be found only in Cardillo de

Villalpandeo and in M. Doniensis Ormazius' De instrumente instrumentorum
sive de dialectica (1569). What is most noteworthy is the sudden appearance
of classical influences. The term 'insolubilia' is replaced by 'inexplicabiles',
and the references are to Cicero and Aulus Gellius rather than to medieval
authors. Solutions were rarely offered by these authors, but Doniensis
Ormazius did argue that insolubles were not genuine propositions, but
orationes imperfectae.
The semantics of propositions

Semantic theory underwent some developments at Paris in the first three
decades of the sixteenth century which were not echoed at all in later
writings. The developments are of two sorts. In the first place we find
lengthy discussions of complexe significabilia, in which the views of Juan
Dolz and Fernando de Enzinas are particularly significant. 7 They argued
that indicative sentences signified aliqualiter - in some way - rather than
aliquid - some thing. They explained this by arguing that propositions are
analogous to syncategorematic rather than to categorematic words in their
manner of signifying, performing a function other than that of naming.
To the question, 'What does this sentence signify?' one could reply only by
a paraphrase. In particular, questions about the dictum of a sentence were
answered by replacing the dictum with a that-phrase. For instance, ' " M a n is
an animal" signifies man being an animal (hominem esse animal)' was
rewritten as ' " M a n is an animal" signifies that man is an animal (quod homo
est animal)', thus removing the temptation to think that the dictum 'man
being an animal' functions as a name. However, they recognised that there
were some contexts in which the question. 'What is this sentence about?'
could be answered by naming or pointing to an object. O n such occasions a
sentence was said to signify categorematically.
The semantics of terms

In the second place we find an interesting series of attempts to provide a
unified theory of the reference of general terms by arguing that the word
'horse' refers to actual objects both in intentional contexts, such as Ί
promise you a horse', and in modal contexts, such as 'For riding is required
a horse', and by arguing that the word 'chimera' could be taken to refer to
actual objects in order to save the truth of such propositions as Ά chimera is
7. See Ashworth forthcoming.

imagined.'8 The solution to the second problem proposed by Celaya and

Enzinas involved the claim that the sentence is a disguised conditional,
being more properly written as 'If an act of riding takes place, a horse is
ridden', so that the truth of the antecedent always involves successful
reference to a particular horse. The solution to the third problem proposed
by the same authors involved the postulation of imaginary objects to serve
as referents, but the universe was not thereby populated with extra entities,
for they explained that an imaginary chimera was just an actual object
which could be imagined to be a chimera. In the last analysis, however,
Enzinas and others were forced to admit that there were a few sentences
such as '"Chimera" signifies a donkey' of whose general terms a purely
extensionalist account could not be given if one wished to prevent the
sentence containing them from being construed as true. The sentence in
question, said Enzinas, can be true only if the word 'chimera' picks out
those donkeys which can be imagined to be chimeras by means of the
concept chimera, which in fact it does not.
Conclusion
Why did these interesting and varied treatments of medieval logical
themes cease so abruptly after 1530?9 Humanism alone cannot be the
answer, since it apparently triumphed only by default. Italian universities
continued to teach medieval logic long after the attacks on it by such men
as Lorenzo Valla; and Agricola's logic did not capture Paris until the
production of texts in the medieval style had already ceased.10 Humanism
certainly had a part to play in the process, however. Soto, for instance,
came to believe as a result of humanist influences that doctrines which were
difficult and not clearly expressed by Aristotle should be omitted from
logic, and that too much time was devoted to summulist doctrines in the
teaching of logic. Accordingly, the later editions of his Introductiones dia-
8. See Ashworth 1974b, 1976a, 1977a.
9. Schmitt 1975, p. 512, notes that certain branches of medieval physics also declined. He writes:
'... several fourteenth-century traditions - including nominalism, the logical traditions of
sophismata and insolubilia, and the Merton and Paris schools of philosophy of motion - continued
on into the first few decades of the sixteenth century and after that quickly lost ground to other
approaches and sets of problems. The printing-history of the medieval texts in question as well as
new commentaries being written on Aristotle indicate this. Why this happened is not clear.
Humanism had a strong impact, as did the reintroduction of the writings of the Greek commen-
tators on Aristotle, but neither of these facts explains why the calculators and writers on sophismata
lost out, while the commentaries of Averroes did not. In brief, certain medieval aspects of the
tradition expired in the early sixteenth century, while other equally medieval aspects continued to
play an important role.'
10. For a discussion of Valla, Agricola, and their influence, see Jardine 1977.
lecticae were very much altered and simplified. Another instructive ex-
ample is Agostino Nifo's Dialectica ludicra (1520). Here we have an intro-
ductory text written by a leading Aristotelian who had a good knowledge
of medieval doctrines, yet he distorts them completely by describing only
those parts of the scholastic theory of terms and supposition theory which
are directly applicable to standard categorical propositions. 1 ' N o one who
became acquainted with medieval logic through Nifo would understand
the function of the non-Aristotelian parts at all. A very plausible account of
the indirect effect of humanism on logic teaching is provided by Terrence
Heath, whose study of the teaching of grammar at three German uni-
versities at the end of the fifteenth century and beginning of the sixteenth
century shows that the change to non-medieval logic was preceded by the
change to humanistic grammar. 1 2 The significance of this sequence of
changes is brought out in Heath's claim that medieval grammar prepared
the student for medieval logic, whereas humanist grammar did not. One
may also speculate that social changes were influential in creating a need for
men with a new style of education. The rise of modern physics has been
cited as a possible cause, but this suggestion cannot be accepted, given that
modern physics can hardly be said to have risen before the end of the
sixteenth century. 13 The judgement of a contemporary logician might be
that medieval logic came to an end because no further progress was possible
without the concept of a formal system and without the development of a
logic of relations. This view is borne out by the desperate, complicated
attempts to analyse such propositions as 'Every man has a head' that are to
be found in the writings of the Parisian logicians. They certainly pushed
medieval logic to its limits, but whether they gave up in despair because
they realised that that was what they had done is another matter. For the
moment our question must remain without a fully satisfactory answer.
11. See Ashworth 1976b.

12. Heath 1971.
13. Kneale 1962, p. 307. It should be noted that the Kneales speak as if interest in formal logic declined
only during the seventeenth century, so that their reference to modern physics is not implausible in
its context.
43
HUMANISM AND THE TEACHING OF
LOGIC
The humanists' reassessment of the study of language

The traditional account of the impact of humanism on the logic cur-
riculum blames the supposed 'barbarousness' of the mediaeval logicians'
use of language for the humanists' hostility to the logic of the traditional
curriculum. In their standard history of logic the Kneales wrote:
The first blow to the prestige of logic came from the humanists, or classical
scholars, of the Renaissance, i.e. in the fifteenth century. Their objection to
scholasticism, and to medieval logic in particular, was not that it was false in any
details, but rather that it was barbarous in style and unattractive in content by
contrast with the rediscovered literature of antiquity. Who but a dullard would
devote his life to the proprietates terminorum when he might read the newly found
poem of Lucretius De Rerum Natura or learn Greek and study Plato?1
According to this account, a commitment to eloquence as the basis for all

learning led humanists to turn from logic, the study of the technical
manipulation of a formal language, to rhetoric:
The writing of elegant Latin was now the chief accomplishment to be learnt, and
for this Cicero and Quintilian were the authorities. From them the men of the
Renaissance acquired the Roman attitude to scholarship, with the result that
genuine logic was neglected for rhetoric and books which purported to be on logic
quoted Cicero as often as Aristotle.2
In actual fact, the shift of interest amongst humanist students of language

derived from different motives. And what is masked by the so-called
'rhetoricisation' of logic by humanist teachers is a more far-reaching
reassessment of the study of language as part of a general arts education.
The humanist most closely associated with such a reassessment, and whose
influence is to be found in most subsequent humanist work on dialectic, is
Lorenzo Valla (1407-57).
1. Kneale 1962, p. 300.

2. Ibid.
798 77«? defeat, neglect, and revival of scholasticism
Lorenzo Valla and the revival of dialectic

Valla was a grammarian whose view of language was coloured by his
commitment to consuetudo (the established linguistic usage of the great
Roman authors of antiquity) as the standard for eloquent Latin. The study
of grammar in the curriculum ought, he maintained, to be based on Latin
usage in the period of its fullest currency as a living language, and ought to
be regulatively descriptive, rather than prescriptive.3 In his most influential
work, the Elegantiae linguae Latinae libri sex (written around 1444), Valla
showed how the precise use of Latin can be decided by careful reference to
classical quotations in which the appropriate lexical item figures; usage
legislates for future employment of the Latin language.
Valla's emphasis on consuetudo as the ultimate arbiter in the study of
language led him to a drastic reassessment of the study of dialectic in the
curriculum, which he put forward in his Dialecticae disputationes (written in
1439).4 The major part of actual argumentation and controversy, Valla
maintained, is concerned not with certainty (that area of ratiocination
appropriately systematised in formal syllogistic as traditionally taught), but
with persuasion and probability.5 Here the central issue is not whether the
conclusion of the argument has been validly inferred from the premisses; it
might be, instead, how tellingly the position taken up by an orator is
supported by evidence and testimony prior to the formation of a valid chain
of arguments; or it might be how convincingly the speaker discriminates
by choice of supporting material and citation of authorities between two
3. An eloquent statement of this point of view is to be found in an early work by Juan Luis Vives, an
enthusiastic follower of Valla. In his Adversus pseudodialeclicos (1782-90), I.42, he writes (1519):
' "Homo est albus' ' is a Latin sentence not because grammar makes it so (any more than rhetorical
figures impart splendor and refinement to speaking because rhetoric decrees it), but rather
because the Roman people, who spoke true Latin, judged that sentence to be Latin. Therefore the
grammarian does not decree that this is Latin, but he teaches that it is; and it is because certain
figures of speech seemed beautiful and fine to speakers that rhetoric diligently observed and
handed them on.' The translation is from Guerlac 1979, p. 57. Valla's Elegantiae in its entirety is an
example of this approach to Latin usage.
4. On Valla see in particular Di Napoli 1971, Camporeale 1972, and Adorno 1954. On the various
versions of the Dialecticae disputationes and their dates of completion, see also Zippel 1957. Zippel
gives the first printed edition of the Dialecticae disputationes as 1509, but Risse 1965 lists two
editions for 1409. See also Jardine 1977.
5. A succinct statement of this position on dialectic is given by another disciple of Valla's, Rudolph
Agricola. In his De inventione dialectica he writes (ca. 1480): 'Exigua enim portio eorum [huma-
norum studiorum] quae discimus, certa et immota est, adeoque si Academiae credimus, hoc solum
scimus, quod nihil scimus. Certe pleraque, pro cuiusque ingenio, ut accommodatissime ad
probandum quisque excogitare potuerit, alio atque alio trahuntur* (Agricola 1539, p. 2).
Humanism and the teaching of logic 799
widely held views. And it may be that coercive strategies such as sorites6 or
dilemma7 whose inferential status is dubious, play as great a part as reliable
syllogistic inference in driving an opponent in debate to a desired
conclusion.8
Although such considerations had traditionally found their way into the
curriculum of the schools under the heading of rhetoric or sophistic, Valla
argued that they rightly belonged in dialectic, not simply as subsidiary
instruction, but as the very core of the study. He adduced the oratorical
texts of Cicero and Quintilian (in particular the Institutio oratoria of the
latter, which had recently been recovered in its entirety through the efforts
of humanist scholars)9 to support this view. Concerned as these authors
were with ratiocination in the context of the law-courts, they tackled the
practical problems of instruction in proving one's case where certain proof
was unlikely to be available and putting a convincing case was a serious
option.10
The effect and influence of Valla's revisions in dialectic
This was the thinking which prompted Valla's revisions in dialectic. Their
effect, in terms of traditional teaching, was to shift the focus of the cur-
riculum away from syllogistic and its associated analysis of terms which
facilitates the formal employment of terms in syllogisms. Instead, the
theory of the Topics became the centre of the course. As developed by the
Roman orators and systematised in Boethius' De topicis differentiis,11 the
theory of the Topics provided a system of classification for oratorical
material according to its appropriateness for a range of strategies used in
6. Sorites is the 'heap' or acervus argument, which proceeds by small, unobjectionable steps to a
controversial conclusion. It takes its name from the case in which an interlocutor accepts a mound
of sand as a 'heap', and is then asked to decide at which point it ceases to be a heap if the grains are
withdrawn one by one. See Jardine 1977, pp. 161—2.
7. Dilemma or antestrephon is another technically undecidable form, of which the most celebrated
example is probably the one attributed to Bion in Aulus Gcllius' Nodes Atticae, The man who
argues for celibacy urges, 'You are sure to marry a woman who is beautiful or ugly; if she is
beautiful you will share her; if she is ugly she will be a trial to you.' The uxorious man retaliates: 'If
I marry a beautiful woman she will not be a trial to me; if an ugly one 1 alone shall possess her'
(Noifrs Atticae V.xi.i-14).
8. See Hamblin 1970 for a modern attempt to derive a formal logic of question and answer from
strategies such as these.
9. Quintilian's Institutio oratoria had been available only in mutilated form until Poggio discovered a
complete manuscript at Saint Gall in 1416. See Reynolds and Wilson 1974, p. 121. See also
Sabbadini 1905. On Valla's work on the newly recovered text see Winterbottom 1967, esp.
pp. 356fr. On Valla's textual activities see Gaeta 1955.
10. See Michel i960.
11. On Boethius and the theory of the Topics in antiquity see Stump 1978.
8oo The defeat, neglect, and revival of scholasticism
debating, of which the syllogism was only one. Even a single example will
help to show the potentially greater flexibility of this focus on the Topics.
The extrinsic Topic from the greater,12 under which are classified pro-
positions of the form 'a is greater than b\ provides the orator with
arguments of the form 'a is greater than b; b is greater than c; therefore a is
greater than c', which although entirely reliable is nevertheless unavailable
to those confined to syllogistic argument.
During the fifty years after Valla completed his Dialecticae disputationes
there is evidence of its being given serious consideration by Aristotelian
logicians.13 But it was not until the early decades of the sixteenth century
that Valla's humanist dialectic came into its own as a serious competitor
with traditional Aristotelian logic within the traditional teaching establish-
ments of Europe. Its appearance in university records correlates closely
with the introduction into those universities of programmes in classical
reading and the study of Greek and Roman eloquence which in the early
decades of the sixteenth century transformed the arts courses of the nor-
thern European universities. Having been a technical introduction to the
linguistic tools needed for solving problems in philosophy and theology,
the arts course became a general introduction to Latin and Greek language
and literature for students destined for professional careers.14 The textbook
which we find displacing Peter of Spain and Paul of Venice as the introduc-
tion to dialectic within these programmes is the De inventione dialectica of
Rudolph Agricola (1444-85),l5 which plainly advocates Valla's approach
to dialectic.
Rudolph Agricola's De inventione dialectica

In 1523 Zasius wrote to Amerbach that at his university of Freiburg, 'Peter
of Spain has disappeared, all of logic has disappeared, except that which
some profess from Melanchthon's little works, some from the books of
Rudolph Agricola." 6 In 1531 a visitation at the University of Tübingen
reported that Agricola was being studied by the 'moderni' in place of the set
12. See Boethius, De topicis differentiis, PL 64,1190D, for his discussion of'greater than' and 'less than'.
13. On the importance of Valla's dialectic see Vasoli 1968a, Jardine 1977, Kristcller 1972, p. 145,
and 1964, pp. 33-5. Agostino Nifo's Dialectica ludicra (1520) invokes Valla's dialectic as a matter of
course (as it appears), as representing an opposed, 'grammatical' approach to dialectic. He quotes
him verbatim without acknowledgement throughout this work and its companion volume,
Epitomata rethorica ludicra, published in the same year. See Jardine forthcoming.
14. On the changing function of the arts course in education see Stelling-Michaud 1967, pp. 71-142.
On the changed role of Oxford and Cambridge in this period see Stone 1975.
15. On Agricola see Vasoli 1968a, pp. 147-82; Howell 1961; Gilbert i960; and Ong 1958a.
16. Amerbach Korrespondenz II, Ep. 923, pp. 429-31, cited in Guerlac 1979, p. 24.
text, Peter of Spain's Tractatus.17 By the 1530s revisions to existing statutes

for the northern universities replace the traditional scholastic manual for
the arts course with Agricola's De inventione dialectica or a humanist dia-
lectic manual derivative from it. Henry VIll's 1535 injunctions for Oxford
and Cambridge stipulate that students in arts should learn their dialec-
tic from 'the purest authors', including Rudolph Agricola and Philip
Melanchthon, rather than from scholastic texts. ' 8 Of the five college
lecturers out of a total of nine appointed to the task of dialectic instruction
in the 1560 statutes for Trinity College, Cambridge, one was to lecture on
Agricola - the only 'modern' text specified.19
Between 1515 and 1590 Agricola's De inventione dialectica went through
at least 35 editions, not including epitomes of the work designed explicitly
for the classroom. Latomus' epitome alone went through 13 editions
between 1530 and 1575.20 It was followed by a spate of textbooks closer in
format to traditional texts (and therefore more readily incorporated into
existing teaching programmes), with Agricolan emphasis on the theory of
the Topics, the priority of discovery (inventio), understood as the selection
and organisation of material, over confirmation (iudicium), understood
as the deployment of such material syllogistically or persuasively, and
ratiocinative strategies other than the syllogism. Most prominent amongst
these were the textbooks of Philip Melanchthon, Johannes Caesarius, and
Peter Ramus.
Melanchthon and Caesarius

Melanchthon (1497-1560)21 produced a number of teaching texts in
dialectic, co-ordinated with his programme for the humanist education of
reformed Christians. Between 1520 and the turn of the century these went
through a staggering ninety-one editions.22 His Erotemata dialectices
(1547)23 preserves the key characteristics of Agricola's and Valla's dia-
lectical approach. It provides an attenuated treatment of syllogistic, which
is not allowed an important place in the text as a whole. Treatment of the
predicables and categories is minimised (doing little more than familiarise
17. Teufel 1977, p. 80. See also Heath 1971.

18. Statuta academiae Cantabrigiensis (Cambridge, 1785), pp. 137-8.
19. Mullinger 1884, appendix to vol. II. See also Jardine 1974 and 1976.
20. All these figures are taken from Risse 1965. See also Ong 1958b.
21. For biographical details see Vasoli 1968a, pp. 278—309.
22. Risse 1965, passim.
23. On the chronology of Melanchthon's various texts on dialectic see Vasoli 1968a. The first was the
Compendiosa dialectices ratio (1520).
8θ2 The defeat, neglect, and revival of scholasticism
the student with the terminology), demonstration gets cursory treatment,

and there is a bare reference to the parva logicalia, the treatises presenting the
innovations of terminist logic, with the statement that all the problems the
parva logicalia are supposed to tackle are the province of the grammarian
24
and do not arise at all if the precepts of grammar are carefully applied.
The same pattern is followed by Caesarius (1467-1550) in his Dialectica
(1532)," which matched Melanchthon's in popularity, and confirmed the
position of Agricolan dialectic at the heart of the curriculum. 26 Both
Caesarius' Dialectica and one of Melanchthon's manuals were in use at the
University of Wittenberg in the 1530s.27
Peter Ramus
But the 'best-seller' amongst dialectic texts was undoubtedly that of the
infamous Peter Ramus, 2 8 whose calibre as a traditional logician in the
medieval mould has been queried by historians of logic ever since W. J.
Ong's work on the popularity and influence of his Dialectica brought him
to their attention. 29 Ramus (1515-72) was trained at the Collège de
Navarre of the University of Paris during the late 1520s,30 a crucial period
for the teaching of dialectic in the University. The introduction of
Agricola's De inventione dialectica into the dialectic course at Paris is gener-
ally dated from 1529, when Sturm came to the city. Latomus was also
teaching in Paris in 1533.31 Two editions of Agricola's work appeared
from the Paris presses in 1529.32 Juan Luis Vives (1493-1540),33 who had
24. Erotemata dialectices (Wittenberg 1555), p. 418: 'Addita est Aristotelis Dialecticae doctrina, uerius
Grammatica quam Dialectica, quam nominarunt parua logicalia, in qua dum praecepta im-
modice cumularunt, et labyrinthos inextricablies, sine aliqua utilitate finxerunt...'
25. For biographical details see Vasoli 1968a, p. 260. For the widespread influence of both
Melanchthon's and Caesarius' dialectic texts see Risse 1964, pp. 25-31 and 79-121.
26. Both Caesarius and Melanchthon acknowledge their debt to Agricola. See Caesarius, Dialectica
(1568), f. 8r. See Vasoli 1968a, passim, for additional evidence of Agricola's direct influence on
later dialecticians.
27. See Caesarius 1559, preface (dated 1532): '... intellexeram Dialecticam meam uel qualemcunque
ipsius Philippi [Melanchthonis] cura et subornatione publice nunc praelegi atque doceri in
Academia Vuittenbergensi'.
28. See Vasoli 1968a, pp. 333-589 passim; Ong 1958a and 1958b.
29. See most recently Ashworth 1974.
30. See Ong 1958a.
31. The evidence for Sturm's having actually introduced Agricola to Paris is rather vague. Lefranc
1893 attributes this statement to Schmidt 185 5, pp. 120-1. It may, therefore, reflect the prejudice
of a biographer of Sturm. There is, however, no doubt that Sturm's and Latomus' arrival in Paris
ensured widespread dissemination of Agricolan dialectic at Paris, whether or not it had preceded
them there.
32. Ong 1958a, p. 96.
33. On Vives see Guerlac 1979; Vasoli 1968a, pp. 214-46.
studied scholastic logic intensively under Gaspar Lax at the Collège de

Montaigu, between 1509 and 1512, had produced his Adversus pseudodialec-
ticos in 1519.34 This highly polemical little work outspokenly attacked the
entire tradition of scholastic logic, rejecting it in favour of a humanistic
study of usage and everyday ratiocination as pioneered by Valla and
Agricola.35 In this atmosphere Ramus set about a ruthless reduction and
schématisation of Agricolan dialectic to produce an all-purpose, accessible,
and undemanding introduction to the tools necessary for 'clear thinking'.
If Valla and Agricola had pioneered the reform of dialectic instruction,
Ramus pioneered popular education: his manuals were designed for the
humblest members of the intellectual community.
Ramus was probably the last of the dialectical innovators to have been
trained both in the scholastic tradition and in the 'Ciceronian' tradition of
the new humanist arts schools. Much of the 'ignorance' and 'uncouthness'
of which Ramist dialectic is accused stemmed from the fact that his texts
fell into the hands of students who had received no systematic introduction
to traditional Aristotelian logic. There is some evidence that scholastic
manuals lingered on in the universities in the latter half of the sixteenth
century because teachers found it necessary to provide some genuine
formal logic to support the humanist programme.36
Ramus was involved in a programme of reform of the entire arts pro-
gramme at Paris and produced textbooks which achieved great vogue in
each subject of the curriculum (and also in such new subjects as vernacular
grammar). He contended that the oratorical focus of a vocational arts
course (no longer a technical training for clerics and teachers, but a higher
education for a literate laity) called for simplification in the teaching of
dialectic. Most specifically (and this emphasis is already to be found in the
writings of Vives and Melanchthon),37 the arts training was now directed
towards ethics, civics, and politics (the active sciences of the professional
man), in which areas opinion and persuasive discourse had traditionally
34. Sec Guerlac 1979.

35. 'Aristotle did not define even the smallest rule in his entire dialectic so that it would not conform
to the same meaning of Greek speech that scholars and children and women and all the common
people used. For a dialectician does not fashion and transmit a new meaning of language, but
teaches rules taken from that old and familiar meaning.' Opera, p. 53, translated Guerlac 1979,
P-79-
36. See McConica 1979. McConica records the persistence of scholastic manuals alongside humanist
manuals, and suggests that an eclectic pedagogy allowed both to be used.
37. Melanchthon argues energetically in his Erotemata dialectices that there is room for the study of
dialectic as part of an education directed at the spiritual and moral welfare of the Christian
individual. All his examples of logical argument are taken from ethical and scriptural questions.
8θ4 The defeat, neglect, and revival of scholasticism
held sway. For such studies a systematic logic of plausibility or probability

(what may be argued as reasonable with the majority of reasonable men) is
not merely an indulgence, but a necessity.
Like his predecessors in humanist logic, Ramus took as his fundamental
distinction in logic the division into inventio and iudicium. Cicero,
Quintilian, and Boethius provided the precedent for his move, which by
now was the hallmark of a humanist dialectical approach. Ramus used
grammatical well-formedness as the unique test of the appropriateness of a
proposition for dialectical use (any well-formed, affirmative proposition is
'true' for his purposes), thus stressing the contingent nature of the study as
he envisages it. H e taught only an abbreviated list of Topics as the means of
classifying dialectical material; these Topics (as in Boethius) serve as a
checklist for the strategies of argument appropriate to the material stored
under any one head. He curtailed discussion of the syllogism and entirely
ignored the logica moderna. And he introduced method as the rallying point
for his aggressively simplified programme of instruction, using 'method'
to designate all the accumulated strategies for organising units of discourse
in blocks larger than single syllogisms.
Ramus appears to have imagined that a dichotomous key, arranging the
major propositions relating to any of the arts in descending order of
generality, would provide a grid enabling teacher or student to display
everything relevant to that art (including complex argumentation) in a
way that would guarantee clarity and completeness. 3 8 In practice, in the
Ramist handbooks which multiplied in the later sixteenth century, this
Ramist method became a crude and restrictive device for running over the
subject matter of a syllabus, or a clumsy mnemonic device for rote recall in
extemporaneous speaking. It is probably fair to say that the protracted
discussions of Ramist method as the solution for all the intellectual ills of
the later sixteenth century represent the lowest ebb of the pioneering
reforms of the humanists. But it should be remembered that this pedagogic
aberration does not represent the entire achievement of humanist thinking
on dialectic.
The function of the humanist textbooks

A good deal of the manoeuvring which takes place within humanist texts
on dialectic is a direct consequence of their function - to provide elemen
tary instruction in the art of discourse for students enrolled in an arts course
38. On Ramist method see Ong 1958a, Gilbert i960, and Jardine 1974.
with an increasingly humanistic bias. In all of them philosophical questions

are subordinated to pragmatic concerns. Check-lists to simplify standard
procedures for selecting material pertinent to a theme and arguing con-
sistently to a desired conclusion figure prominently. Latomus' successful
epitome of Agricola's De inventione dialectica (a much more discursive
work) is a striking example of the way in which the dialectic manual of the
153os and thereafter is expected to be a pocket blueprint for successful
performance in set disputations. The object of the text is unashamedly to
enable the student to make the appropriate debating moves; the only
criterion for preferring one strategy to another is its effectiveness in
polemic. From the point of view of the historian of logic, it no longer
makes very good sense when considering these manuals to isolate treat-
ment of a traditional section like the categories in order to assess the
humanist dialectician's intellectual contribution to it: he makes of it only as
much as will carry his student to the next heading in his survey course. In
this respect a recent historian of logic is correct in saying that the humanist
dialecticians 'perverted the purpose of scholastic logic, which was also to
teach, but to teach a developed formal logic rather than the art of thinking
clearly'.39
Summary and conclusion

We may summarise the changes in dialectic teaching brought about by the
impact of humanistic studies on the arts course as follows. In the early
decades of the sixteenth century, across northern Europe, we find the
introductory arts course being adapted to meet the requirements of an
influx of students from the professional classes.40 Within this arts course,
with its humanist predilection for Greek and Latin eloquence, and legal
and ethical instruction, there was an acknowledged need for some rigorous
underpinning of instruction in 'clear thinking'. But the meticulous intro-
duction to formal logic and semantic theory provided by the scholastic
programme came to look increasingly unsuitable for this purpose.
One should, however, add that where the introductory arts course
continued to be the preliminary stage in a programme directed toward the
39. Ashworth 1974a, p. 9.

40. It is becoming increasingly clear that despite a certain inertia in the statutes, humanist texts began
to be studied in arts courses throughout northern Europe in the early decades of the sixteenth
century, as study of book inventories ofstudents, commonplace books, and teachers' lecture notes
proves. See Curtis 1959 for the persisting discrepancy between statute requirements and the books
students actually studied and owned. See also Jardine 1974, 1975, 1976,1977, and forthcoming;
also Kearney 1970.
8o6 The defeat, neglect, and revival of scholasticism
study of philosophy and theology, scholastic logic continued to be taught

with its traditional rigour. This was the case at the universities of southern
Europe such as Padua, with its thriving philosophy faculty, and Bologna,
with its sophisticated medical school.41 Humanist studies in the Italian
universities became assimilated into what had always been a thriving
grammar course.42 Aristotelian logicians teaching in the Italian universities
show themselves aware of the alternative programme in practical dia-
lectic,43 and sometimes wrote elementary manuals suitable for a humanist
course, while concentrating most of their attention on the enduring
scholastic tradition.44 One might even suggest that contact with the
humanists' emphasis on the Topics led some of these logicians to introduce
fruitful and original treatments of Topics into their formal and philo-
sophical treatises.45
What was needed in the way of dialectic for the humanist arts course was
a simple introduction to the analysis and use of ordinary language (elegant
Latin) for formal debating and clear thinking. Such an analysis was pro-
vided in the 'alternative' dialectics deriving from the work of Valla and
Agricola. These were committed to 'plausibility' as the measure of success-
ful argumentation. Their authors impressively argued the case for giving
serious consideration to non-syllogistic forms of argument, and strategies
which support or convince rather than prove, as an intrinsic part of the 'art
of discourse'.
In practice, the pedagogues working from the innovatory treatises of
Valla and Agricola tended to discard those earlier authors' intellectual
justifications for their refocused textbooks. But they stuck faithfully to
the selection of subjects for treatment to which those authors had
given emphasis. Thus Caesarius and Melanchthon both give space to non-
syllogistic sorites in their textbooks, although they give no indication why
41. On the school of Padua see Randall 1940, reprinted in Randall 1961. Fora full bibliography see
Schmitt 1971.
42. See Rashdall 1936,1, 92, for an account of this fundamental difference between the structures of
the arts courses north and south of the Alps.
43. Nifo, Zimara, and Zabarella all allude to the humanist programme, and sometimes appear to
incorporate some of the devices (particularly for presentation) from those courses in their more
traditional treatises. For an account of the close contact between humanism and the enduring
Aristotelianism at Padua sec Poppi 1970 and Vasoli 1968b.
44. See Nifo's Dialectica ludicra, which in many ways deviates from traditional Aristotelianism in its
selection of topics and presentation. See also the Tabulae logicae by Zabarella, printed at the end of
his Opera logica.
45. This may well be the case with Nifo, whose commentary on Aristotle's Topics, widely used in the
sixteenth century, incorporates many non-Aristotelian attitudes towards the use and flexibility of
the teaching of the Topics. See Ashworth 1976b.
this topic is of peculiar interest to them.46 They also tend to be extremely

sketchy on just those traditional areas of logic in which the historian of
logic is particularly interested. This may well be because there already
existed a body of standard textbooks of quality in these areas, to which the
student could be referred. There has not yet been an adequate study of the
intellectual contribution of the humanist dialecticians on the issues with
which they are particularly concerned, such as extended argumentation
over units larger than the syllogism, or coercive forms of argument.47
Until such a study is made, we shall not be in a position to decide whether
the humanist intervention in the history of logic represents a decisive
impoverishment or a possible enrichment of the tradition.
46. Soriles is allowed a heading to itself, although no attempt is made at more than a description of the
argument form.
47. The work of Vasoli, Howell, and Gilbert is in the end too descriptive to make much headway.
44
CHANGES IN THE A P P R O A C H
TO LANGUAGE
The humanists' attitude towards scholastic philosophy

There is a certain irony in the fact that it was the humanists with their
enthusiasm for the literature of the ancient world who were in large part
responsible for the demise of scholastic grammar. For inasmuch as human-
ism was a literary and educational, not a philosophical movement, the
attitude of most humanists towards scholastic philosophy was one of
indifference; they seldom manifested outright hostility, and such oppo-
sition as they professed was not, for the most part, philosophical.'
Humanists outside the university
The early humanists were, in fact, either independent men of letters or,
more typically, members of the legal profession holding high office in
church or state. 2 Petrarch (1304-74) is the best-known representative of
the first category. Famous nowadays for having written some of the most
magnificent sonnets in the Italian language, he was known to his contem-
poraries primarily as a promoter of classical studies and the author of a
number of highly regarded original works in Latin as well as a voluminous
correspondence, likewise in Latin. An outstanding example of the second
type of humanist is Petrarch's younger friend Coluccio Salutati (1331 —
1406), who was trained as a notary at Bologna University and for the last
twenty-six years of his life held the office of chancellor in Florence.
1. On the relation between humanism and scholasticism, see Kristeller 1944-5; 1956, pp. 553—83;
Garin 1965, pp. 1-4. The word 'humanist', a fifteenth-century creation, was originally applied to
teachers of classical literature, later to anybody, whether teacher or not, who cultivated the so-
called jludia humanitatis; see Campana 1946, esp. p. 66; Avesani 1970. The abstract noun
'humanism*, on the other hand, is of very recent date. Note that while humanism was not a
philosophical movement, many humanists had profound philosophical commitments, witness
the platonism of Marsilio Ficino; see Garin 1965, pp. 78—135; Kristeller 1961a, pp. 48—69.
2. Onearlyhumanism, see Weiss 1947. A classic account ofltalian and German humanism is Geiger
1882. Voigt 1893 is also still valuable. On the concept of the Renaissance a great deal has been
written, see especially Ferguson 1948; Chabod 1958, pp. 149-200; Garin 1969b, pp. 1-20.
Changes in the approach to language 809
The humanists' involvement in education

Then, from the early fifteenth century onwards, many humanists became
involved in education and left their stamp on generations of students, at
first drawn predominantly from northern Italy, but in course of time from
as far afield as England and Hungary. The two foremost educators,
Guarino Veronese (1374-1406) at the d'Esté court in Ferrara, and Vit-
torino da Feltre (1373-1446) at Mantua, developed a comprehensive
course of studies for the young men in their charge, embracing a wide
variety of physical as well as intellectual pursuits.3 The literary curriculum
was largely based on the principles laid down by Quintilian and comprised
a thorough initiation into both Latin and Greek, together with the inten-
sive study of classical authors, including the then much-esteemed works on
moral philosophy by Cicero, Plato, and Aristotle. One might perhaps
characterise the ideal products of the new educational programme as all-
round individuals capable of turning their talents to any form of public
service. It was this conception of the liberally educated man which dis-
tinguished the humanistic schools from the scholastically dominated uni-
versities of the time.
The humanists' attitude towards the study ofgrammar

As regards the more narrowly linguistic aspects of the movement, the
foremost aim of the humanists was to model their Latin style on that of the
ancients whom they so admired: one of the primary targets of their dis-
approval was the level of contemporary Latinity. At first this ideal was
achieved without the aid of special textbooks. Petrarch spent his life
saturating himself in the works of Cicero and other Latin authors, which
did not fail to affect his own Latin prose style in a profound way, but he
expressed no dissatisfaction with the grammatical literature in general use
at that time. Indeed, he included two medieval dictionaries in his list of
favourite books, namely the anonymous Papias (mid-eleventh century)
and the Catholicon ofJohn of Genoa (late thirteenth century).4
Early humanist grammars

Even when the humanists turned their attention to the composition of new
textbooks, they did so without any explicit criticism of the medieval
3. On Guarino Veronese, see Sabbadini 1891; also 1896, esp. pp. 26-38; on Viitorino da Feltre, see
Woodward 1921.
4. On Petrarch's favourite books, see Ullman 1955, pp. 117-37.
8io The defeat, neglect, and revival of scholasticism
grammarians. The earliest humanist grammar of Latin was the Regulae

grammaticales of Guarino Veronese, composed before 1418.* The two
salient characteristics of this remarkable work are its externe brevity and its
omission of the logical and metaphysical underpinnings characteristic of
the grammatical products of the late Middle Ages. Of genuine originality
there is little, if any: Guarino drew his material from the Latin grammars of
his immediate predecessors, not directly from the Roman grammarians of
antiquity. It is noteworthy too that he made no attempt to support his
rules by means of quotations from the classical authors; indeed many of his
examples are concocted ones of the kind familiar from earlier grammatical
works.6
In spite of the unoriginality of Guarino's grammar it was a considerable
success, but copyists in transcribing it often put back some of the medieval
features which Guarino had been at such pains to remove.7 At least in that
negative respect Guarino was clearly in advance of his age.
The Latin grammars written in the wake of Guarino's Regulae are
characterised by a gradual accentuation of humanistic features and a pro
gressive rejection of medieval traits, but the development is by no means
linear. Thus the grammar of Gasparo Veronese, composed in Rome some
time during the pontificate of Nicholas V (1447-55), shows if anything
more medieval features than Guarino's.8
Fully developed humanist grammars

The work which most profoundly influenced all humanists in the area of
Latin prose style was the voluminous Elegantiarum linguae Latinae libri sex
by Lorenzo Valla (1407-57), which circulated beginning in the early
1440S. The revolutionary feature of Valla's manual was the extensive,
though not exclusive, use of direct quotations from the authors of antiquity
to back up the stylistic precepts. Valla was not concerned with syntax or
5. On Guarino's Latin grammatical works, see Sabbadini 1896, pp. 38-47; 1902; 1906; 1919, pp. 79-
80; 1922, pp. 3-15; Percival 1972; 1975; 1976a.
6. The following are typical of the examples to be found in Guarino's grammar: 'Me taedet
vitiorum', 'Ego emo equum decern ducatorum', 'Mulier trunca manum', 'Animal homo currit.'
The quotations from the classics are almost all stock examples carried over from earlier grammars,
such as 'Urbem quam statuo vestra est' (Vergil, Aen., i. 573), quoted by Priscian and also by the
author of the Graecismus (Wrobel 1887, p. 5.40).
7. For instance, some manuscripts of the Regulae contain the modistic definitions of the parts of
speech — e.g., 'Nomen est pars orationis declinabilis, significans per modum habitus et quietis et
determinatae apprehensionis' (MS Naples, Biblioteca Governativa dei Gerolamini, A 60, f. 1 •). Sec
Roos 1961, p. 10, lines Ι9-2Γ.
8. On the grammar of Gasparo Veronese, see Sabbadini 1896, pp. 45-6.
composition, however; the bulk of the Elegantiae is devoted to the proper

meaning and choice of single words and the avoidance of barbarisms,
i.e., non-classical words. What is truly novel in Valla's work is a pervasive
critical attitude, which extends also to the grammatical authorities of
antiquity: Valla indeed went so far as to criticise Priscian on occasion, for
which he appears to have been censured by his contemporaries.9
After Valla's death, the next major contribution to Latin grammatical
writing was the Rudimenta grammatices by Niccolo Perotti (1429-80),
which was completed in Viterbo in 1468. This neatly organised manual,
which soon appeared in many printed editions and was widely used in Italy
and northern Europe until well into the sixteenth century, was the first
humanistic grammar of Latin to include elementary morphology (the
noun and verb paradigms, for instance) in addition to more advanced
topics, and the last third of it is devoted to epistolary style. It could,
therefore, be used as the basis of a complete course of studies in Latin, and it
may be that it owed its popularity to this fact as well as to its perspicuous
organisation.10
An even more comprehensive grammar was shortly afterwards pro-
duced by the Salamanca humanist Antonio de Nebrija (ca. 1441—1522).
The first printed edition of his Introductiones Latinae appeared in 1481; the
work assumed final shape in 1495 and was reprinted many times, in Spain
and elsewhere.11 Like Perotti's grammar it is lucid and well-organised, and
it also treats topics such as orthography, prosody, and metrics, ending with
a short dictionary, based predominantly on Valla's Elegantiae and the
monumental Orthographia of Giovanni Tortelli (ca. 1400-66), a work
which had appeared in the early 1450s. Nebrija was self-consciously in-
novative, and all the editions from 1495 on are provided with a copious
commentary in which he meticulously explains and justifies his intentions.
His most drastic innovation was perhaps his addition of the gerund and the
supine to the eight canonical parts of speech, which he justified on the
grounds that there had been no consensus among the grammarians of
antiquity as to the number of the parts of speech.12
9. On Valla's Elegantiae, see Barozzi and Sabbadini 1891, pp. 161—73; Casacci 1926; Stevens 1975.
10. A thorough analysis of Perotti's Rudimenta is an urgent desideratum. See, however, Mercati 1925,
pp. 59-61, 131-2; Baebler 1885, pp. 135-9.
11. Nebrija is chiefly known for his grammar of Castilian, which appeared in 1492. On Nebrija's life
and works, see Gonzâlez-Llubera 1926, pp. xvii-xl. His Latin grammar has so far not received the
attention it deserves.
12. After arguing that the gerund is a separate part of speech, he says: 'Neque id mirum videri debet,
cum multi plures, multi paudores quam octo partes orationis posuerint' (Nebrija 1481, sig. c6").
The influence of the rediscovered classics

It is important to realise that Nebrija's sort of statement could not have
been made by grammarians who lived before those spectacular discoveries
of previously unknown, or little known, Latin authors which we associate
with such men as Poggio Bracciolini (1380-1459). 13 It was, in fact, the
discovery of the works of Probus, Diomedes, Charisius, and other Roman
grammarians which made books such as Tortelli's Orthographia possible. 14
In this general category the most important find was undoubtedly the
fragments of Varro's De lingua Latina (first century B.C.), on which the
humanist Giulio Pomponio Leto (1428—98) lectured in R o m e and which
he subsequently edited for publication (1471). 15 The discovery of Varro
was especially important for grammarians because his much earlier ap-
proach to grammatical questions differed in many important respects from
that of the two canonical authors Donatus (ca. 350 A.D.) and Priscian (ca.
500 A.D.). Thus, the humanists' critical attitude to traditional theory was
reinforced, if not brought on, by this exposure to more ancient, hitherto
unfamiliar doctrine.
Another factor which contributed to the widening of linguistic horizons
was the initiation of Greek studies in Italy, first by Chrysoloras in Florence
in the final decade of the fourteenth century, then by other Greek emigres,
notably Theodore Gaza and Constantine Lascaris, and as time went on by
an ever-increasing number of Italians, among them Guarino Veronese,
who was perhaps the first westerner to study Greek in Constantinople
(1403-8). 16 Chrysoloras, Guarino's teacher, wrote a grammatical text-
book, the Erotemata, in the traditional catechetical style, which Guarino
abridged. This epitome was subsequently translated into Latin, and the
resulting bilingual version was the first Greek grammar widely dissemi-
nated in the West. ' 7 The amount of Greek which the general run of
13. On the discoveries of manuscripts made by the humanists, see Sabbadini 1905; 1914.
14. The episde dedicatory in Tortelli's Orthographia makes this abundantly clear; see Garin 1958,
pp. 495-7. On the Orthographia, see Rinaldi 1973.
15. On Pomponio Leto's grammatical works, see Ruysschaert 1954; 1961. The standard treatment of
this humanist is Zabughin 1909-12.
16. On the Greek grammars of the Renaissance, sec Pertusi 1962. The impact of Greek émigré
scholars on the Italian Renaissance has at times been somewhat exaggerated in the popular
imagination, especially in northern Europe. Garin comments as follows: 'The actual contribution
made by Byzantium to humanism consisted mainly in new tools and in precious materials which
enriched occidental civilisation. The East was able to provide useful formulas to a western system
of thought which had already reached an autonomous maturity' (196s, p. 81). In this connection
see also Geanakoplos 1966, esp. p. 112.
17. The curious methodology of Greek instruction in the West is described in Sabbadini 1922,
pp. 19-22. On Chrysoloras' grammar, see Pertusi 1962; Sabbadini 1919, pp. 76-7, 128.
fifteenth-century humanists commanded should not be over-estimated.

But learning a language in many ways so different from Latin posed a
considerable intellectual challenge to the student of those days and may
even have stimulated many teachers to scrutinise traditional grammatical
doctrine.
A similar broadening of perspectives must have resulted from the begin-
nings of Hebrew studies. It should perhaps not surprise us that Nebrija,
who taught Greek and Hebrew as well as Latin, was the first humanist
grammarian to state explicitly that the rules of syntactic agreement ob-
served in Latin are common to all languages, but it is noteworthy that he
appealed to the authority of Aristotle (De int. i6a8) in so doing. 18
The humanists' treatment of logical syntax
This raises the complex issue of logical syntax, an area which had but
recently been developed by the scholastic grammarians with little help
from their classical forebears. An elaborate and theoretically sophisticated
framework for the analysis of sentences had taken shape by the fourteenth
century. It is worthy of remark that while the humanist teachers reduced
the grammatical baggage to the minimum in order to facilitate and hasten
their students' initiation into classical literature, they nonetheless retained
the essential features of the system of logical syntax they had inherited
from the immediate past. Terms such as 'suppositum' ('subject'), 'appositum'
('object', or sometimes 'predicate'), 'agens' ('initiator of the action'), 'pa-
ttens' ('recipient of the action'), and above all 'antecedens' ('antecedent of
a pronoun'), which had constituted the stock-in-trade of the Latin gram-
marians of the fourteenth century, survived well into the sixteenth cen-
tury, and some of them, such as 'agent' and 'antecedent', have remained in
use to this day.
This retention of medieval syntactic concepts is especially evident in the
grammatical textbooks written by the humanists of northern Europe. Thus
the Syntax of Anthonius Haneron (ca. 1400-90), presumably written in the
1430S and perhaps the earliest grammatical work of the humanistic variety
composed north of the Alps, includes a careful discussion of sentential
18. Nebrija's three universal rules are the agreement of adjective with substantive, of relative with
antecedent, and of subject nominative with verb. He comments as follows: 'Ponit tria praecepta
quae non sermonis latini propria, verum etiam omnibus unguis communia, publica, et quodam
modo naturalia sunt, si quidem naturale est, ut inquit Aristoteles, quod apud omnes gentes idem
est' (Introductiones Latinaecum commento, 1495, sig. c8").
8i4 The defeat, neglect, and revival of scholasticism
analysis.19 It is also significant that in the version of Perotti's grammar

edited by Bernard Perger of the University of Vienna with the impressive
title Grammatka Nova (the first printed editions date from the early 1480s),
Perger adds an extra set of syntactic rules after Perotti's.20
The humanists' attitude towards modistic grammar

Regarding the modistic literature, the humanists' attitude was, not surpris-
ingly, one of indifference, if not outright disdain. Their explicit references
to the modi significandi are rather few, but consistently hostile. A passage
from Valla's encomium on Thomas Aquinas perhaps typifies the ambiva-
lence the humanists felt towards certain manifestations of scholasticism.
Valla first reproves the theologians for their superstitious reverence for the
modi significandi - it is, he says, almost as if they regarded them as a newly
discovered sphere or planetary epicycle - then declares that it really makes
no difference to him whether one knows about the modes or not, and
finally hastens to add that it is perhaps preferable to remain ignorant of
them altogether since they are obstacles in the way of better things.21
An organised attack on modistic grammar was launched by the West-
phalian humanist Alexander Hegius (ca. 1433—98), who taught for many
years at the school of the Brethren of the Common Life in Deventer. His
pungent Invectiva in modos significandi, composed around 1480, contains all
the expected humanistic charges, such as the contention that the modistae
were corrupters of Latinity, that the church fathers had written excellent
Latin without knowing anything about the modi significandi, and so forth.
But Hegius offers at least one argument of a more substantive nature,
namely the contention that words which have the same modus significandi
do not necessarily behave the same syntactically. He points out, for in-
stance, that while the genitive of 'Johannes' can be used in the expression
'liber Johannis' ('John's book'), the genitive of the personal pronoun 'ego'
(T), cannot be used in that way: 'liber met' is not permissible. From this the
reader is to conclude that knowing the modi significandi does not entail
knowing Latin syntax.22
19. On Anthonius Haneron, see Ijsewijn 1975, pp. 217-19; Ijsewijn-Jacobs 1976, pp. 29-33.
20. On Bernard Perger and the impact of the new humanistic textbooks in German-speaking
universities, see Heath 1971.
2i. See Ijsewijn 1971, pp. 301-2; Vahlen 1886, p. 394.
22. For the full text of Hegius' invective, see Ijsewijn 1971. On Hegius, see Lindeboom 1913, pp. 70-
81; Hyma, 1951, pp. 151-9; 1965, pp. 125-8. In philosophical penetration Hegius' arguments
against modistic grammar cannot compare with those of the nominalists a hundred and fifty years
earlier; see Pinborg 1967a, pp. 172-85.
The humanists' attitude towards the verse grammars

As for the verse grammars (in particular the Doctrinale of Alexandre de
Villedieu and the Graecismus of Evrard de Bethune), which had been a
staple of the curriculum from the thirteenth century onwards, the attitude
of the humanists towards them seems initially to have been favourable, or
at least tolerant. The Doctrinale was still being used in Guarino's school.23 A
wholesale reaction against the verse grammars did not take place until the
second half of the fifteenth century. There is, for instance, a sharply worded
invective aimed at the Doctrinale in one of the grammatical textbooks
written by the Roman humanist Giovanni Sulpizio Verulano (ca. 1470).24
Significantly, however, his criticisms were motivated primarily by ped-
agogical considerations: he censured the work for its density and obscurity,
not its factual unreliability.
In northern Europe, the attack on the verse grammars generated more
acrimony, and they were all but completely banished from the classroom
by the middle of the sixteenth century.25 Nevertheless, the custom of
casting grammatical rules in verse for mnemonic purposes was by no
means abandoned by the humanist grammarians: parts of Guarino's and
Nebrija's grammars were written in verse, and in general this practice
persisted until well into the nineteenth century.
Universal grammar
The belief that all languages share certain basic grammatical categories
was another medieval legacy, a notion which had been clearly stated by
Gundissalinus in his De divisione philosophiae (late twelfth century)26 and
was presupposed by all the writers of modistic treatises. The early human-
ists do not mention the idea, but it would be rash to assume that they took
exception to it. As we have seen, Nebrija grants universal status to the Latin
rules of concord. After suffering what appears to have been a complete
eclipse, philosophical grammar came back into fashion in the seventeenth
century,27 and some of its new votaries acknowledged that they were not
starting from scratch. Both Campanella, in a treatise published in Paris in
1638, and John Wilkins, author of An Essay Towards a Real Character and a
23. See Sabbadini 1896, pp. 36, 42, 56-7; Woodward 1921, p. 165.
24. For the complete text, see Percival 1976b, pp. 87-8; cf. pp. 82-3.
25. See Reichling 1893, pp. XCHI-CHI; Heath 1971.
26. See Baur 1903, pp. 45-6.
27. On seventeenth-century philosophical grammar, see Padley 1976, pp. 154—209. On the Port-
Royal language textbooks, see Donzé 1967; Padley 1976, pp. 210-59.
8i6 The defeat, neglect, and revival of scholasticism
Philosophical Language (1668) allude explicitly to medieval predecessors. 28

Whether we should posit an unbroken tradition of philosophical gram-
mar linking the modistae and the seventeenth-century universal gram-
marians_has not yet been established. The famous Port-Royal Grammaire
générale et raisonêe (1660) emphasised the difference between substantives
and adjectives in their modes of signifying (Part II, Chap. 2), 2 9 but the same
point had been made by the nominalist Mario Nizzoli in his De veris prin-
cipiis et vera rationephilosophandi a little over a hundred years earlier. 30 We
know that the Port-Royal educators were familiar with such works as
the elder Scaliger's De causis linguae Latinae (1540), a blistering attack on
traditional grammatical terminology and concepts from an avowedly
Aristotelian standpoint, but it is not clear to what extent, if any, they had
first-hand acquaintance with medieval approaches to language. It is per-
haps more likely that they were influenced by scholastic logic than by
survivals of the medieval grammatical tradition. 31
The influence of the humanists' approach to language

Even the more narrowly linguistic aspects of humanism could not fail to
have a profound and lasting influence on philosophy as a whole. Already in
the first half of the fifteenth century Valla had subjected some of the key
terms used by the schoolmen, such as quiditas and ens, to destructive
grammatical analysis. As Valla viewed the matter, a word such as quiditas is
a morphological impossibility in Latin, and the notion of universality
implied in the term ens can have no grammatical justification whatever. 32
28. Campanella 1638, Pars prima, I. i., p. 3; Wilkins 1668, III. i-, p. 297.
29. Grammaire générale et raisonnée, II. ii-, p. 31 (in the first (1660) edition).
30. De veris principiis (Parma 1553), I- iii·, sigs· C2 r ~C3'.
31. Thus the famous analysis o f the sentence 'Dieu invisible a créé le monde visible' into the three
judgements 'Dieu est invisible', 'Ilacréé le monde', and 'Le monde est visible' (Grammaire générale
et raisonnée, (1660) II. ix, pp. 6 8 - 9 ) can be traced back to the discussions centring on 'restrictions'
and 'exponibles' by the medieval logicians, sec Salmon 1969, pp. 180-3; Kretzmann 1975, pp.
189-95.
32. Dialecticae disputationes, I. ii and iv (Opera, sigs. 2S3"-5', 2S6 y -7 r ). Fresh light has recently been
shed on this work by the discovery o f two previously unpublished versions, one earlier and the
other later than the version to be found in the Opera. See Zippel 1957 and subsequent studies, such
as Vasoli 1968a and Camporeale 1972 (esp. pp. 149-71). The pedagogical implementation of the
new humanistic approach to logic can be observed in the manuals o f George o f Trebizond (see
Monfasani 1976, pp. 300-17) and Rudolph Agricola (see Faust 1922; Crescini 1965, pp. 49-67).
The philological method was almost immediately applied to biblical criticism, witness the
Adnotationes in Novum Testamentum, written by Lorenzo Valla in the last years of his life and
published half a century later by Erasmus (Paris 1505). O n the theological aspects of humanism see
Fois 1969, Trinkaus 1970, Camporeale 1972, pp. 211-468; and firisteller 1961a, pp. 79—80; 1974,
p. 65. Procedures which had proved effective in the study of belles-lettres were inevitably turned
to good account by theologians themselves. O n the relation between biblical exegesis and
present-day canons of textual criticism, see Pasquali 1952, pp. 8-12.
Another important service provided by the humanists was to make

available hitherto unknown or imperfectly known philosophical texts, and
concomitantly to bring about a re-evaluation of already familiar texts. In
this area, one may point to the many translations of Greek philosophical
works which began to appear from the early fifteenth century onwards.
The humanists published bilingual Greek-Latin editions of many such
works and provided them with elaborate new commentaries, composed,
naturally, in Latin.33
The greater accessibility of the rich Platonic corpus was a novel
and especially potent factor. As is now widely recognised, a vigorous
Aristotelian tradition, increasingly revitalised by the availabdity of new
texts and fresh translations, maintained itself throughout the Renaissance
and into the early modern period. At the same time, however, Platonism
undoubtedly provided philosophers with valuable new concepts, which
were utilised in a wide variety of contexts. For instance, the recovery of
Plato's Cratylus stimulated and enriched speculation about the nature of
language.34
33. The new humanistic translations of De anima, for example, had to compete with the familiar
version by William of Moerbeke (late thirteenth century), which had provided the schoolmen
with a stock of useful terms for a number of fundamental psychological concepts; see Garin 1951 ;
Cranz 1976. It is clear from such cases that the medieval philosophical tradition, while undergoing
a profound change, survived the humanistic revolution to a considerable extent.
34. See, for instance, Scaliger 1540, III. lxvii, sigs. P3'-4V; Sanctius 1587,1, i, sigs. r\.$'-6".
45
SCHOLASTICISM
IN THE SEVENTEENTH C E N T U R Y
Spain as the bastion of late scholasticism

After the attacks of humanists, Ramists, reformers, and plain haters of
philosophy over much of two centuries, it is amazing that scholasticism
survived at all. Not only did it survive, it experienced a notable revival
throughout much of western Europe towards the end of the sixteenth
century and beginning of the seventeenth. Humanists and reformers were
by no means unanimous in opposing the medieval scholastics. More
important, the Iberian peninsula was comparatively unaffected by the
intellectual and religious ferment of most of the rest of Europe. The schools
of Spain and Portugal had a more or less continuous tradition of scholastic
philosophy, and the leading figures in the general revival of scholastic
thought round the end of the sixteenth century tend to be Spaniards like
Banez, Vâsquez, and Suârez. In northern Europe the scholastic revival
looks more like a self-conscious and deliberate Aristotelian reaction to
Ramists, humanists, and the like, but the northerners of whatever religious
allegiance were happy enough to take guidance and inspiration from Spain.
New trends in late scholasticism

Although in obvious ways continuous with the main medieval tradition,
late scholasticism, whether in its Iberian form or in its northern revival,
shows certain very distinctive characteristics of its own which may be seen
as marking a transition to some of the most prominent themes of early
modern philosophy.
While these philosophers were nearly unanimous in rejecting medieval
nominalism (indeed, in the north this was another of the things they were
reacting against), Scotus, Ockham, and the later nominalistic tradition had
a very powerful influence on them. What is particularly notable is their
tendency to internalise order, to direct our attention away from an order
given in the external world and towards an internal order, the structure of
the mind and patterns in the way in which the mind acts. In what follows I
shall try to trace the result of this internalising of order in logic, in
metaphysics, and in the study of law and morals.
Scholasticism in the seventeenth century 819
Late scholastic logic

It is hard to praise the late scholastics for their contributions to formal
logic.1 For reasons that are complex and a little obscure, progress and
inventiveness in formal logic dried up early in the sixteenth century. With
hindsight we can conjecture that the medieval logicians had gone about as
far as they could go without the mathematical tools of a later age. We find
no new contributions to formal logic in the treatises of the late sixteenth
and early seventeenth centuries, and medieval achievements are mis-
represented or not reported at all. Some of these treatises give little more
formal logic than a rather garbled treatment of syllogistic. Few include any
discussion of supposition theory at all, and those that do either give a very
truncated survey, as Sanderson does,2 or put together doctrines that would
have horrified their medieval predecessors.
In fact, these authors were not much interested in formal logic as such.
They were far more interested in the philosophy of logic and language
than in formal logic. On these matters, while they made few if any original
contributions, they carried on the medieval discussion at a high level of
competence and were the means of transmitting some of the characteristic
doctrines of medieval speculative grammar and the philosophical study of
language to the early modern period. This is a matter of some interest
because we can here trace the history of what Chomsky has called
'Cartesian linguistics' from the medieval logicians and grammarians to the
Cartesians themselves.3 Many of the logicians of the period also show some
Ramist influence in their concern for methodological questions. They are
sometimes accused of psychologising logic; but the genuine scholastics - as
opposed to the out-and-out Ramists or the 'mixts' who tried to combine
Aristotle and Ramus - were concerned about an ideal, inner order of
thought, fit to represent the most general formal characteristics of being. In
this their philosophy of logic was no more psychologistic than that of their
medieval predecessors.
The logic of John of St Thomas

A well-known logic of the period and the only one that has attracted much
attention in our time is that ofJohn of St Thomas. His logical works were
incorporated during his lifetime in his Cursus philosophicus Thomisticus,
although in his first editions he used such titles as Artis logicae pars prima, de
1. Thomas 1964, pp. 297-311.

2. Sanderson 1640.
3. Chomsky 1966. See Trentman 1975; 1976; also Kretzmann 1975.
dialecticis institutionibus, quas summulas vocant and pars secunda, de instruments

logicalibus ex parte materiae.4 The 'first part' is a shorter text intended as an
introduction to logic; the 'second part' is a more extensive work for more
mature students of philosophy. It takes up such Aristotelian subjects as the
predicables, the ten categories, and demonstration, including of course an
account of necessary propositions and per se predication.
In his general logic John, unlike most of his contemporaries, includes a
relatively extensive treatment of supposition theory. He very insistently
claims to be a follower of Thomas Aquinas, and it is interesting to com-
pare what he gives of supposition theory with that of an earlier soi-disant
Thomist, Vincent Ferrer (1350-141 Q).5
John's divisions of supposition are the same as Ferrer's. He has not much
to say about natural supposition, which obtains in necessary propositions
and which Ferrer discussed at very considerable length, but he includes it in
his account, and he also defines simple supposition in the characteristic anti-
Ockhamist way as obtaining when a term stands for what it primarily and
immediately signifies.6 He had no objection whatever to assigning the
property of supposition to predicates as well as subjects. In this he was no
true follower of his master, Aquinas, if Ferrer was right.7 Worse yet, he
makes full use of what the medievals had called 'merely confused sup-
position', which Ferrer had argued does not occur — is, indeed, a merely
confused notion. Furthermore, John uses the notion exactly as Ockham
had done. So at least so far as supposition theory goes John of St Thomas
turns out to be a strange Thomist.8
Noam Chomsky has identified three principal doctrines that characterise
what he calls 'Cartesian linguistics'.9 First, there is a creative aspect of
human language that defies the possibility of mechanistic explanations and
distinguishes our use of language from animal behaviour generally.
Secondly, there are 'grammatical principles' that are universal to all human
languages in the sense that they stand as constraints on possible grammars
of natural languages. Thirdly, one can and must distinguish the deep
structure of language, shared with other rational beings who speak dif-
ferent languages, from the surface structures of these tongues. All these
4. For bibliography see John of St Thomas 1955a.

5. Vincent Ferrer 1977.
6. A good account of these debates can be found in Pinborg 1972.
7. Vincent Ferrer 1977. Cf. Geach 1972; 1961, pp. 76-80. In fact, Aquinas was not consistent on this
point; see Vincent Ferrer 1977, p. 97, n. 1.
8. Cf. Vincent Ferrer 1977, pp. 43 and I4if.
9. Chomsky 1966.
principles are clearly found in John of St Thomas' logic, but in this he was
only following the long tradition of medieval scholastic study of lan-
guage.10 Indeed, John's insistence that entia rationis of second intention are
the formal object of the study of logic is tantamount to the claim that logic
has to do with structures universal to all rational minds. The logician must
give an account of this internal order and relate it, on the one hand, to the
conventional sounds and marks we produce and, on the other, to the things
in the world we want to talk about. Nothing could be more 'Cartesian'.
Scholastic logics of the seventeenth century

We find the same thing wherever we look in scholastic logics of this period.
Franco Burgersdijck's Institutionum logicum libri duo was widely used
throughout northern Europe and in 1646 was translated into Dutch and
went into 'school editions' in the author's homeland.l ' It presents little
formal logic and shows no logical originality. There are signs of Ramist
influence in its concern about questions of method, although its author
records his proper scholastic-revival disapproval of Ramus. On 'Cartesian'
principles, however, Burgersdijck is sound. Much the same can be said of
two other popular logics of the time, the Logica of the Polish Jesuit, Mar-
tin Smiglecki, and the Logicae libri quinque of the Anglican, Richard
Crakanthorpe.12
Much the same can be said of the earlier work of Seton, which was first
printed 'with Peter Carter's annotations' in 1572.13 As a formal logic it is
tedious and uninspired; as philosophy of language it faithfully records the
standard medieval doctrine. Likewise, Edward Brerewood in two logic
texts does little formal logic and none to deserve special notice, but he does
produce a quite interesting discussion of problems in the philosophy of
language that has much in common with Thomas of Erfurt's Grammatica
speculativa.1* A rather more interesting English logic of this period is
Robert Sanderson's Logicae artis compendium.15 Unlike most of his con-
temporaries, Sanderson gives a brief survey of supposition theory; he also
includes as an appendix an interesting survey of the history of logic, in
which he praises the logic of his medieval predecessors but not their Latin
10. Cf. Trentman 1975.

11. See Risse 1964, pp. 515-16, and Guerault 1968.
12. Smiglecki 1634; Crakanthorpe 1622.
13. Seton 1631; cf. Trentman 1976, pp. 180-1 and 184-5.
14. Brerewood 1614 and 1628; cf. Trentman 1976, pp. 187—8.
15. Sanderson 1640.
style. His account of the task of the logician is quite interesting.16 The
logician, using his own appropriate tools, can clarify and make explicit the
structures that may have been hidden in the idiomatic expressions of
everyday speech. This is good medieval doctrine; it also points ahead to
'Cartesian' linguistics in its concern with the inner order of mind and
mental acts.
Late scholastic metaphysics in general

The metaphysics of the late scholastics, like their logic, mixes old and new.
We may note the influence of Ockham and his followers on philosophers
who were by no means Ockhamists and indeed, regarded themselves as
Thomists, as well as the emphasis on the activity of mind and the internal
structure of mental activity. The metaphysics of Francisco Suârez was
particularly prominent and influential.
Suârez on essence and existence

The Suarezian idea that has occasioned the most interest and controversy
through the centuries is his understanding of the distinction between
essence and existence. Suârez inherited the question and its background
from his medieval predecessors. It was the old problem of understanding
the idea of creation and relating the belief in creation to a philosophical
analysis of created being. It is a tenet of orthodox Judaeo-Christian belief
that God exists by his own nature; it is in the very nature of what it means
to be God that God must exist. As it was put, in God essence and existence
are identical. What it means to be a creature is that this is true of no created
thing. For all created things essence must somehow be distinguished from
existence. But how? Suârez distinguishes three possible positions that one
might take in response to this question.17 First, one can say that for finite
beings existence and essence are really distinct. Suârez expresses this option
in a way that requires its defender to hold that existence and essence are
distinct things; existence is a kind of object (rem quandam distinctam). He says
that this 'seems to be St Thomas' opinion',18 but he is sure that it is the
opinion of a wide variety of later Thomists including Cajetan. The second
option holds that existence and essence are not really distinct; they are
formally distinct, i.e., existence can be regarded as a mode of essence.
Mention of the formal distinction suggests Scotus, and Suârez identifies
16. Ibid., p. 75.
17. Suârez 1965, 31, 1; vol. II, pp. 224-8.
18. 'Haec existimatur esse opinio D. Thomae'; cf. pp. 225-31.
this position with him. The third opinion is that existence and essence are
only mentally distinct (tantum ratione). According to Suârez, this opinion
was held by Alexander of Hales, Peter Aureoli, Durandus, Gabriel Biel,
and nominalists generally. Suârez asserts that the first two positions are false
and the third true, but true only if the terms are understood in the proper
sense. By 'existence' we must understand 'actual existence' and by 'essence'
'actually existing essence'.19 Understood in this way the existence and
essence of any created being can be regarded as mentally distinct or
mentally distinguishable. For Suârez, then, the distinction has its origin in
the mind; making such distinctions is one of the things that minds do. But
the mind is not at complete liberty to do what it pleases here; while it is
responsible for the distinction, something about the way things are with
finite beings in the world is the occasion for it to make the distinction. The
distinction itself is something that minds do, but their performing such an
action is somehow grounded in things; this is what he means by his
expression 'distinctio rationis cum fundamento in re'.20
Suârez on individuals and universals

One of the major turning points in the history of western philosophy was
Scotus' formulation of the idea of intuitive cognition. The old problem of
individuation was, crudely, this: given the knowledge of universals, find
the individuals. After Scotus there was a new problem: given the knowl-
edge of individuals, find the universals. Suârez addressed the new problem,
and although he was sharply critical of Scotus, his doctrines of individuality
and universality show their influence throughout Suârez' discussion. All
actually existing things, immediately existing things as distinct from what
can exist only dependently in an individual object, are singular and
individual.21 Furthermore, they can be known in their individuality.
Suârez agreed with Scotus that to say that a thing is individual is to talk
about some thing (aliquid reale) in addition to the nature of the thing, which
is common to many, but he saw no reason for positing a special individ-
uator like the haecceitas to account for this individuality.22 The composite
of form and matter itself contains in its union its own principle of individu-
ation.23 As he puts it, 'the adequate principle of individuation is this matter
19· 31, ι, Π; vol. II, p. 228.

20. 31, 6, 23; vol. II, p. 250. Cf. 7, 1, 4~s; vol. I, p. 251.
21. 5,1,4; vol. I, pp. 146-7.
22. 5,2, 8-9; vol. I, pp. 150-1.
23· 5. 3; vol. I, pp. 161-75.
and this form in union. The form alone is the sufficient and chief principle
so that the composite, as an individual thing of a certain species, can be
considered numerically one and the same thing'. 2 * In many ways this
argumentation in Suârez curiously foreshadows the twentieth-century
debate about so-called bare particulars. Modern logical atomists have
argued that in order to account for the individuality of two things that are
identical in all their nonrelational properties one must posit individuators
that in themselves have no properties (hence their bareness) but ground the
individuality and thereby the distinctness of the things. But the idea of the
bare particular, like that of the Scotist's haecceitas, has a strangeness about it
that does not appeal to many philosophers. 25
Suârez certainly did not regard himself as a nominalist. Nevertheless, it is
interesting to notice how close his position on this issue is to Ockham's. 2 6
Again we see that at the heart of Suârez' doctrine there is a strong emphasis
on the activity of the mind. He insists that the unity of the universal is not a
real, extramental unity but a rational unity, a unity imposed by the mind.
But it would be quite wrong to conclude that there is litde more than
Ockhamism in his doctrine. Suârez thought minds are no more at liberty to
impose whatever they will on external reality than they are to distinguish
existence and essence without some basis or grounding for the distinction.
Universal words are not names that refer to independent entities in the
world, but acts of minds have an ontological grounding in the structure of
external reality; once more in Suârez we meet the concept cum fundamento
in re.21 In his discussion of universals Suârez often appeals to Aquinas,
picking out for special attention Aquinas' insistence on the creativity of
minds to the possible neglect ofother elements in his thought. In the game
whose rules had been drawn up by Scotus, Thomism can be made to look
very much like Ockhamism.
Suarezian individuals, then, are composites of form and matter that
contain in their unity their own principles of individuality. They can be
known directly by the mind, and the mind accounts for what they share
with each other by imposing on them its conceptual scheme, its acts of
universalisation. But its doing so has a ground in reality although not in the
existence of distinct, real things, i.e., extramental universals. Some have
24. ' . . . adaequatum individuationis principium esse hanc materiam et hanc formam inter se unitas,
inter quae praecipuum principium est forma, quae sola sufficit, ut hoc compositum, quatenus est
Individuum talis speciei, idem numéro censeatur'. 5, 6, 15; vol. I, p. 186.
25. See, e.g., Bergmann 1964.
26. His position is worked out in Dist. 6; vol. I, pp. 201—50. See especially sections 5 and 6; pp. 222—8.
27. 'Universalitas est per intellectum cum fundamento in re.' 6, 5, i; vol. I, p. 222.
said that Suârez gives the mind too much to do, so that he runs the danger,
if he does not indeed succumb to it, of letting the mind have such a creative
function that its hold on external reality is weakened if not lost entirely.
This is the burden of the complaint by some Thomists (whose best known
modern representative is Gilson) that Suârez is a philosopher of essence not
existence, that in contrast with Aquinas he is left with an 'essentialist'
metaphysics.28
Suârez on creation
The distinction between essence and existence, it was noted, was the result
of an attempt to provide a philosophical understanding of what is involved
in the theological doctrine of creation. This doctrine requires that, while it
is in God's nature necessarily to exist, created things exist contingently in
dependence on God's creative action. Furthermore, in his act of creation
God must not be thought to work with any kind of raw material and,
therefore, to bring about a kind of change in individual things. Creating
differs from making. Geach has expressed this difference succinctly in the
following way.29 When we mean to say that God (or anybody, A) makes
something, what we say can be expressed in the form
M — (3 χ) (A brought it about that χ is an F)
To say however that something is created is to say
C — (Not M) and (A brought it about that (3 x) (x is an ¥))
Thomists maintain that their doctrine meets the principles expressed in this
analysis completely. Suarezians would deny that Suârez' doctrine of exist-
ing essences causes him the theological embarrassment of having provided
God with uncreated material to work with in creation or the philosophical
embarrassment of a proliferation of odd entities. A Thomistic critic would
maintain that if Suarezians insist that the actualised essence is just as
contingent and dependent on God's creative act as the Thomist's created
individual, they only push the difficulty back.30
What is at issue in this dispute can be usefully illuminated by making
some more modern comparisons. Suârez certainly insists that no created
being exists necessarily. We can think of essences abstracted from their
existence, but in actuality essence and existence cannot exist separately.
Indeed, for Suârez such a supposition makes no sense. Just as it is absurd to
suppose that essence without existence would leave some existing thing as a
28. Gilson 1952, pp. 96-120.
29. Geach 1969, pp. 75-85.
30. Gilson 1952, p. 104.
remainder, so it is absurd to suppose that before God brought it about that

there was this particular actually existing essence there was some part of it;
in denying the real distinction Suârez supposed that he was denying that
such part-whole talk makes sense. Nevertheless, one can think of essence in
abstraction from its existence.
Suârez on possibility and reality
This brings us back to another important Suarezian distinction. Suârez main-

tains that we can talk about being in two different ways, distinguished,
he claims, by the use of a noun (esse) and the use of a participle (ens). The
use of the noun signifies what has a genuine essence whether it actually
exists or not; the use of the participle indicates the act of existence, and by it
we understand that the thing in question actually exists.31 Moreover,
corresponding to this distinction between two ways to talk about being,
Suârez distinguishes two senses of the term 'real'. It can mean 'capable of
existing in the real world', i.e., possibly producible by a cause; something
real in this sense has 'objective potential being', which in turn means that
we can think of its possibly existing. The second sense of 'real', which
Suârez calls its 'proper' sense, is 'actually existing in the real world'. 32 All
this talk of worlds and the real world, actual beings and possible beings
reminds one of Leibniz, who was after all a great admirer of Suârez. It also
reminds one of Leibniz' modern followers who use the language of set
theory to talk about necessity, contingency, and possibility. It is no ac-
cident that some of these tools can easily be made to work with Suârez'
ideas.
We can think of a possible world as an overall state of affairs, the way
things can be or could have been. Obviously, things are and have been a
certain way, but they could have been different. There will be individual
beings that exist in the various possible worlds, some in all, others only in
some possible worlds. Suârez would maintain that God exists in all possible
worlds but Francisco Suârez, for instance, only in some. These beings have
properties in the various worlds, and the properties they have in the various
worlds may of course be described by propositions. In distinguishing his
two senses of'being', Suârez points out that a real essence signified by the
noun 'esse' is something that may not but could exist in the real world. By
31. Suârez 1965, 2, 4, 9; vol. I, p. 90.

32. 31,2,1; vol. II, p. 229. Also 31, 2,10; vol. II, p. 232.
this he means to rule out two possibilities. First, a real essence must be
something that can be coherently thought; 'the round square', 'the thing
that both is F and is not F' cannot designate real essences. So the idea of
possibility utilised in his concept of objective potential being must, first of
all, mean what is in a broad sense logically possible; the round square exists
in no possible world. Secondly, Suârez means to rule out pure flights of
fancy. According to Suârez, fictional characters have no sort of being, not
even possible being,33 and things like the chimera are not possible beings.
Because certain properties are essential to the nature of what we call goats,
no goat can have a lion's head and a serpent's tail; therefore, no such things
can be truly said to exist in any possible world.
Real essences exist in possible worlds, but obviously not all of them exist
in the actual world. We can clearly think of objects that are describable in
logically coherent ways, are not fictional characters, are of such a nature
that they could exist in the actual world; but they do not. Do they yet have
some kind of being or reality? Such things are real in the sense of'having
objective potential being', not in the sense of 'actually existing in the real
world*. But this is too quick, Suârez' critic will say. Surely everything that
exists in a possible world exists, is real, in that world; the so-called actual
world, which we may designate W + , is only one element in the set of
worlds W. Why should we give such prominence to one member of the
set? For any world, W, to say 'This individual χ exists in W is tantamount
to saying 'X actually exists in W ; to say 'x exists in W + ' , therefore, is to say
'x actually exists in W + ' . How does this give any preferred status to W + ,
which we have chosen to call the 'actual world'? It looks as if
(A) This is the actual world
is logically equivalent to
(B) This world is this world.
Plantinga's response to this kind of objection is very much to the point in
discussing Suârez. Plantinga argues that the objector's conclusion is based
+
on a mistake. Β is obviously tautologically true; but if A means ' W is the
actual world', it is contingently true. 3 4 This is exactly Suârez' point; God
has chosen to make this world the actual world, and there's an end on't.
Why he chose to create what he chose to create we shall never know and is
none of our business. As a matter of contingent fact he created the actual
world.
33. 31, 2, 2; vol. II, p. 229. Cf. Plantinga 1974, pp. 153-63.
34. Plantinga 1974, pp. 49-51·
The concept of natural law in late scholasticism

Both Suârez himself and also the other philosophers of the scholastic
revival are probably better known for their contributions to theories of
natural law than for their metaphysics. It has often been thought that their
doctrines of natural law developed in a more and more secularistic direc-
tion, culminating in Grotius' famous comment that the natural law would
still hold and oblige us even if God did not exist. But it is anachronistic to
read this sort of secularisation into Grotius, and those who do so often show
no awareness of the context of Grotius' remark or, indeed, of exactly what
he wrote. He certainly thought God required things and forbade others
because they were in and of themselves right or wrong, and he also thought
it makes no sense to say that God can change the natural law; but he tells us
that the speculation that, even if God did not exist, the natural law would
hold cannot be considered without the most profane wickedness (quod sine
summo scelere dari nequit)35 - hardly the remark of a secular moralist. Later
natural-law theorists may seriously have tried to secularise the theory; in
the period under consideration here the changes in natural-law theory
were different. In fact, they directly parallel the changes we have noted in
the philosophy of language and in metaphysics. Again we see the influence
of Scotus, Ockham, and the Ockhamist tradition on people who were at
some pains to refute Ockhamism, and we see the shift of interest from
order in external nature to an internal order in the minds of rational beings.
And it can also be claimed that few if any of the raw materials of this
philosophy of law and morals were original. Indeed, in the thought of
Thomas Aquinas (and in that of his followers down to the present) there
has always been a certain tension between the extent to which the precepts
of natural law are seen to be read off the order of external nature and the
extent to which minds are illuminated in an Augustinian fashion to know
them. The complex and difficult doctrine of synderesis is adequate testi-
mony to this tension. What happens in the period under consideration is
that the illumination theme tends to assume ever greater prominence at the
expense of confidence in finding the precepts of natural law in an external
order.
Richard Hooker on natural law

Of course, not everyone moved so far so quickly, and within the period
itself we can see this tension working itself out. A good place to start is with
35. Grotius 1853, Prolegomena.

Richard Hooker. Hooker is in many ways a typical product of the north-

ern scholastic revival. He has often been regarded as a transitional figure,
and so he is; but in comparison with others I shall consider shortly, like
Suârez, Holdsworth, and Grotius, he is a man who is constantly looking
back over his shoulder. To put it in his own language, we can say that he
would prefer to inquire 'into the causes of goodness' (by which he means
Aristotelian aitiai), but 'this present age full of tongue and weak of brain'
forces us to settle for second best.36 There is a theological as well as a
philosophical presupposition behind this. Hooker was explicitly reacting
against the doctrines of English Calvinists, who held a low view of the
possibilities for natural knowledge and whom C. S. Lewis aptly charac-
terised as Barthians.37 Luther was little better help; he thought our minds
had capacities sufficient only for recognising that we were sinners and
needed salvation. Melanchthon, who was an Aristotelian and in many
ways a kind of precursor to the later Aristotelian revival, was unable to
improve on this gloomy diagnosis by much. It is the theological problem
of the consequences of the Fall for human knowledge that sets the task for
later protestant defenders of natural law and provides the context of
Hooker's revival of natural law. What can the human mind know of law
and morals on its own without God's grace and revelation? He thinks we
should (in principle) be able to have something like an Aristotelian science
of morals, but it is uncertain how far he actually thought a demonstrative
science of morals a real possibility. He is content to argue that certain
general (mostly formal) principles of natural law are self-evidently known
to us. Otherwise we must go by 'signs and tokens to know good'. 38 As for
these signs and tokens, Hooker lays great stress on what he calls the
'universal consent of men', 39 that these Laws of Reason are generally
known - 'the world hath always been acquainted with them'. 40 In a later
age defenders of kinds of natural-law doctrine commonly appealed to the
supposed fact that all men agreed about certain obligations and pro-
hibitions, while opponents thought the discovery of apparent exceptions
constituted a refutation of such claims. Hooker, however, certainly did not
appeal to any nose-count, nor can one imagine him with a tally-board
recording responses to questionnaires. Indeed, he hedges his appeal to
36. Hooker 1977,1, viii, 2.

37. Lewis 1954. P· +49·
38. Hooker 1977,1, viii, 3.
39. Ibid.
40. I, viii, 9.
universality with a number of crucial qualifications. Children, innocents,

and madmen get no vote; 41 indeed, it is those 'having natural perfection of
wit and ripeness ofjudgement' who are to be listened to. 4 2 It is those who
comprehend the principles of natural law by the light of natural under-
standing, Melancthon's lumen naturale and Aquinas' synderesis.
Hooker, however, has no truck with voluntarism; 'They err therefore
who think that of the will of God to do this or that there is no reason besides
his will.' 4 3 W e may not know the reason for the precepts of natural law,
but we are to be assured that there is one; God in his unsearchable counsel
knows the right and rational order of things, knows it to be good, and
consequently wills accordingly.
Suârez and Vâsquez on natural law

The problem I have introduced was at the heart of the controversy
between Suârez and one of his countrymen, Vâsquez, who like Hooker
had a hankering for an independently knowable external order. Vâsquez
maintained that the natural law is identical with rational nature, by which
he meant that good acts are those that conform to the ideal nature of human
beings and bad acts are those that violate it. This nature is objectively
knowable and constitutes the foundation of all law and morals. 44 Suârez
objected that all this is false to the meaning of the term 'law'. Law in itself is
a matter of will, and Suârez defines 'law' as 'an act of a just and right will by
which a superior wills to oblige his inferiors to do this or that'. 4 5 Will,
therefore, is essential to the meaning of 'law', which definition identifies
Suârez' doctrine with the tradition of medieval voluntarism. But the
divine will is not the sole basis of morality. Will is essential to law, but
God's will attaches to what is intrinsically good and his prohibitions to
what is intrinsically bad in the nature of things. Gregory of Rimini had
held that the natural law is not prescriptive but demonstrative of the order
of things; Ockham (as Suârez interprets him) had held that good is simply
what God wills and God is free to attach his will to anything that can be
consistently described. Suârez aims at avoiding both extremes. Wdl is
essential to the concept of law, but God's will is not arbitrary; He wills
what is in the nature of things good.
41. I, vii, 4.
42. I, viii, 9.
«· I, i". 5-
44. Vâsquez 1620, disp. 150, 3; cf. q. 94, ar. 2 and q. 94, ar. 5.
45. Suârez 1944,1, 5,24, ' . . . actum voluntatis iustae et rectae, quo superior vult inferiorem obligare
ad hoc, vel illud faciendum.'
In support of this via media Suârez offers an important, influential

argument. Aristotle contends in the Ethics (1107*8-27) that some actions or
passions do not admit of an extreme; they are described by names that in
and of themselves imply badness. Aristotle's examples are spite, envy,
shamelessness, adultery, theft, murder. If an action or passion is truly
described in one of these ways, there can be no debate about when, where,
how to perform the action or indulge the passion; it has already been
condemned in the description. Suirez attributes the same doctrine to a
wide variety of authors including the Ockhamist, Gabriel Biel.46 In
fact, Ockham himself gave a certain prominence to this Aristotelian
argument.47
Richard Holdsworth on natural law

An author who gives this argument a central significance in his doctrine of
natural law is Richard Holdsworth, a seventeenth-century English theolo-
gian. Holdsworth was very influential in his own time but has since been
almost entirely forgotten.48 Holdsworth was very much a product of the
scholastic revival, and although he was not only an Anglican but associated
with the Puritans, he readily acknowledged his debt to Suârez and other
scholastics. He went to some pains to try to explicate and use the scholastic
notion of synderesis, but he was unsatisfied either with an appeal to our
supposed awareness of an external order or with any sort of intuitionist
claim that we all really know in our hearts what is right and wrong. Rather
he appeals to Aristotle's argument and adds that anyone who can under-
stand a description expressed in terms of one of Aristotle's 'bad names'
must (assuming him to be in his right mind) be embarrassed and insulted.
Thus to call a person a perjurer, he claims, is to describe his actions in such a
way that the description is as good as an insult. Anyone truly so described is
properly insulted (or corrected, or brought to repentance). Here we see the
basis of natural law sought in a kind of linguistic fact, the fact that certain
concepts not only describe acts but imply a condemnation of anyone
whose doings can be truthfully described by them. As Holdsworth puts it,
this is in effect the voice of nature itself pronouncing judgement on one
who does these things (Natura ipsa clamât quod malum sit, Naturae Lex
46. Suârez 1944, II, vi, 11.

47. William Ockham 1495-6, III, 12, d. 16; cf. McGrade 1964, p. 190, and Trentman 1978, pp. 29-
39-
48. See, e.g., Curtis 1959, and Hill 1965. Holdsworth's lectures were edited and put out by his nephew
after the restoration (Holdsworth 1661).
arguit.)*9 What this means, according to Holdsworth, is that natural law is

founded in the nature of things, but not in any natural order we can readily
observe around us; rather it is found in what he calls 'integral nature', an
ideal order, which has only left us its traces.50 These traces are to be found
in language, in concepts that all rational beings share whatever their
national languages. In this way natural law is an expression of the com-
munity of language users, which of course presupposes that all human
beings as such share certain concepts. This sharing constitutes a community
that in principle unites all human beings and provides them with a rather
tenuous link with integral nature; here, in Holdsworth's terminology, we
find the vestigia of ideal order.
Natural law and the order of nature

What has become of the external order of nature? Hooker, for all his
qualifications, still felt confident enough to assert ' . . . God being the author
of Nature, her voice is but his instrument. By her from Him we receive
whatsoever in such sort we learn.' S1 The new scholasticism as represented
by such thinkers as Suârez and Holdsworth had far more in common with
another contemporary, John Donne, whose well-known fines express his
feeling about the effect of'new philosophy':
Tis all in pieces, all coherence gone;
All just supply, and all Relation.52
In the response of the new scholastics to a picture of the world from which
all coherence was gone we see the themes I have been stressing come
together. As I have noted, Suârez, following the Ockhamist tradition,
insisted that will is essential to the concept of law. Likewise Holdsworth
rejects what he calls the traditional jurists' definition of natural law - that
natural law is what Nature, teaches all living beings. First, he insists that
natural law is essentially about prescriptions for human beings; secondly,
he is sceptical about Nature's directly teaching anything. We need our
reasons, expressed in our linguistic competence and use, to reflect an ideal,
49. Holdsworth 1661, lect. xxviii, pp. 247-8. It might be noted that it would be both anachronistic
and too crude to interpret this in terms of analyticity or, worse, some vague idea of what is 'true
by definition'; it is the nature of the acts in question chat requires necessary connections in our
thought.
50. Holdsworth 1661, p. 246.
51. Hooker 1977,1, viii, 3.
52. 'The First Anniversarie* in John T. Shawcross (ed.), The Complete Poetry offohn Donne, New
York University Press, 1968, p. 278.
divinely ordered order to make sense of any kind of lawfulness. 53

Incidentally, it should be obvious that the natural-law doctrines of these
authors do no violence to the principle of the autonomy of ethics under-
stood as the rule that ethical conclusions cannot be derived from admit-
tedly non-ethical premisses. 54 Indeed, they insist that the right ordering of
rational wills is at the heart of the definition of law.
This account of the scholastic revival has in a way come full circle in that
we find the doctrine of natural law rooted in an internal order to be found
in what is innate in all minds. What is innate is an ordered conceptual
scheme shared by all rational beings, and this idea is explicated in terms of
the 'Cartesian' doctrines in the philosophy of language, inherited from the
logicians and speculative grammarians of the Middle Ages.
Hugo Grotius on natural law

As a final example of the new scholastic doctrine of natural law, we can
turn to Grotius, Much that he says in the Prolegomena to his De jure belli et
pads sounds very like the theories of Suârez and Holdsworth I have just
discussed. The idea of natural law or natural right is based not on general
ideas of order in nature but on an analysis of human nature. Human beings
naturally desire to five in society. This is no news; everybody in the
scholastic tradition would have agreed with Plato and Aristotle about this.
Grotius accounts for it, however, by appealing to the fact that language-use
is a differentia of human beings, and society is based on linguistic
community. Furthermore, Grotius' arguments about the species-specific
characteristics of language use are almost identical to Descartes'. 55 One can
see that the raw materials for the new scholasticism were all very old. It was
the proportion, the emphasis, that was new, but this produced a distinc-
tively different-looking result, a result that looks much more like what we
call early modern philosophy.
The survival of scholasticism in the universities

Obviously the transition from medieval to modern philosophy was neither
so smooth nor so evident as this may suggest. First, at least some Aris-
totelian texts continued to hold a prominent place in the curriculum
for a very long time. How seriously they were taken and how they were
53. Holdsworth 1661, pp. 239-40.

54. Cf. a similat point made about Aquinas in Donagan 1969.
55. Grotius 1660; cf. Descartes, René (1897 & 1913). Discours de la méthode in Charles Adam and Paul
Tannery, Oeuvres de Descartes, vol. vi, Cerf, pp. 56-9.
read and studied we shall soon consider. Secondly, many of the formalities
of medieval scholastic university training - the lecture, the disputation,
other scholastic acts or exercises — continued to be observed, at least after a
fashion, until quite recent times; indeed, their traces are obviously still with
us. Thirdly, the works of medieval philosophers continued to be printed
and came out in new editions. Fourthly, and most important, despite
superficial continuities, the most influential and most read scholastic works
in the seventeenth century and thereafter were not Aristotelian texts, nor
the works of medieval scholastics, but the new textbooks. They were the
main vehicle for transmitting scholasticism to the succeeding centuries.
This may be a cause for lament, since what was criticised as medieval
scholastic doctrine may often have been a demonstrably garbled account
by a seventeenth-century textbook-writer.
Little need here be said about the continuation of Aristotelian texts in
university curricula. The texts prescribed tended often to be Aristotle's
scientific works rather than what we should call the philosophical works,
although the Nicomachean Ethics turns up with some frequency in lists of
prescribed books. Whether the logical works do or do not appear de-
pended perhaps on the degree to which the institution prescribing the
books had been influenced by Ramism. Aristotle's philosophical works
were rarely allowed to speak for themselves, however. In most lists of
prescribed books or directions for students drawn up in the seventeenth
century Aristotelian works are accompanied by some of the new textbooks
to 'explain' them. O n e suspects that while lip service was given to Aristotle
for a long time, it was the explanations by textbook-writers that were read.
The continued reprinting of works by medieval scholastic authors poses
a similar problem. Even leaving out of account reprints in Roman
Catholic countries, where the influence of the commitments of religious
orders was instrumental, the number of new editions and reprints through-
out the seventeenth and into the eighteenth century is astounding. Who
read these books? They generally do not appear in reading-lists for stu-
dents. O f course, they were available in university libraries, and we know
from catalogues of personal libraries that they were evidently bought by
individual scholars. Yet they seem not to have influenced them very much.
One example might help to make this point. Ockham's Summa logicae
appeared in a new printing in Oxford in 1675, 56 but it is very hard to find
any evidence that the book found perceptive readers who made much of its
56. Printed by O. Walker, Oxford.

contents. Formal logic was in a dreary state; Sanderson, perhaps the best of
a bad lot, gives only a very superficial treatment of supposition theory.57
Furthermore, there is little trace in the period of the medieval theory of
consequences; the formal logic is once again syllogistic, not infrequently
garbled. What then did people make of the medieval texts that kept
coming out in new editions? Very little, it seems. They were too busy
reading the new textbooks.
The shift from scholastic commentaries to modern treatises

Where medieval philosophers had commonly worked out their doctrine in
the course of commentaries, either on the works of Aristotle or on the
Sentences of Peter Lombard, the scholastics of the early seventeenth century
took to writing systematic treatises on their own. Suârez' Disputationes
metaphysicae was a prominent example of the new style, a work that aimed
at a systematic treatment of the subject, full of references to other philos-
ophers, to be sure, but organised according to its author's view of a
methodical, systematic presentation of the subject rather than as a com-
mentary on what others had written. The Disputationes metaphysicae hardly
looks like a text for beginners, but the systematic style could easily be used
both for general and purportedly exhaustive treatments of a subject and for
introductions aimed at beginning students. One of the most popular
writers of such textbooks in northern Europe was Franco Burgersdijck,
who produced texts on every branch of philosophy which were very
widely read in the seventeenth century. Another such person was
Bartholomew Keckermann. It was authors like these who were most
widely read and who represented scholasticism to most people in the
seventeenth and eighteenth centuries. Unfortunately, not only were they
not very original, they did not always understand or adequately transmit
the doctrine they had inherited from the Middle Ages.
The best way to get an impression of what in fact was being read in the
universities of the seventeenth and early eighteenth centuries is to examine
a guide for students. An excellent one for our purposes is the Directionsfor a
Student in the Universitie by Richard Holdsworth.58 We know that the
57. Sanderson 1640.

58. Holdsworth, Richard (1648?), Directionsfor a Student in the Universitie, Emmanuel College
(Cambridge) MS 48; reprinted in H. F. Fletcher, The Intellectual Development offohn Milton, 1961,
University of Illinois Press, vol. 11, pp. 623-55. Cf. Bodleian MS Rawlinson D200; in a forth-
coming paper I argue that its ascription of the work to John Merryweather is mistaken.
Trentman, J. A. 'The Authorship of the Seventeenth-century Cambridge Directionsfor a Student
in the Universitie', Transactions of the Cambridge Bibliographical Society.
Directions not only represents university education during its author's

lifetime but also for at least fifty years after his death. When Josiah Barnes
(1654-1712), Regius Professor of Greek in Cambridge University from
1695 until his death, wrote his own directions and advice to students,59 in
the middle of it he copied out the whole of Holdsworth's Directions as
giving the essence ofwhat he had to say. What Holdsworth prescribed was
clearly still being followed or at least urged sixty years later.
It is interesting to note how similar in certain general principles and
presuppositions the Directions is to the Jesuit Ratio Studiorum of 1599. Both
show the influence of humanism in their emphasis on the study of ancient
languages and, further, their emphasis on writing and speaking 'good'
Latin, i.e., the Latin of Cicero and Quintilian. With this a wide range .of
Greek and Latin literatureis prescribed. And, of course, both include much
moral and religious counsel. For both, however, the study of philosophy is
rooted firmly in scholastic Aristotelianism. Holdsworth supposes that the
serious student will read all of Aristotle (in Greek, of course) and all of both
of Aquinas' Summae, but what is much more interesting is how he thinks
these things should be read and where his real emphasis lies. Concerning
philosophical controversies (the successors of the medieval disputations),
which are to be a primary concern of the third year of university study,
Holdsworth writes, 'The reading of Aristotle, will not only conduce much
to your study of Controversy, being read with a Comentator, but all so help
you in Greeke, & indeed crown all your other learning, for he can hardly
deserve the name of a Scholar, that is not in some measure acquainted with
his works.' 60 The important phrase here is 'being read with a Comentator',
and much more space is given in the Directions to the commentators than to
Aristotle and medieval scholastics put together. It is useful here simply to
note who some of these authors were. In logic Burgersdijck is most highly
praised, but we are directed also to Brierwood, Eustachius, Smiglesius,
the Complutenses and the Conimbricenses, Crakanthorpe, Kecker-
mann, Molinaeus, and Sanderson.61 In ethics Burgersdijck again, Golius,
59. Emmanuel College (Cambridge) MS 179.

60. Holdsworth 1648 (?), p. 33.
61. Brerewood (Brierwood) 1614 and 1628. Eustachius (a Sancto Paulo) (1609), Summa philosophiae
quadripartita, Paris. Smiglesius: Smiglecki 1634. Complutenses: the Carmelite professors of the
University of Alcali (formerly Complutum) published a Cursus Artium there in 1624; the author
of the Logic was Diego de Jesus. Conimbricenses: a similar group-venture produced by the Jesuit
professors at the university of Coimbra in Portugal; they wrote a variety of commentaries
and a Cursus philosophicus. Crakanthorpe 1622, Keckermann, Bartholomew (1613), Systema
Systematum, Hanover. Molinaeus: du Moulin, Pierre (1603), Elementa logica, Antwerp, Sanderson
1640.
Eustachius, Morisanus;62 in metaphysics Scheibler, Fonseca, Eustachius,

Suârez, 'or the like'.63 These were the authors that both students and
teachers of philosophy read. And there is abundant evidence that the spirit
and even many of the details of the Ratio studiorum and Holdsworth's
Directions were widely characteristic of university training in western
Europe into the eighteenth century. The pity is that scholasticism was not
represented in the universities either by the best medieval philosophers or
even very often by the more interesting late scholastics. Indeed, Holds-
worth himself wrote more interesting things than many of the authors he
recommends to students.64 Instead scholasticism was transmitted to later
generations by philosophical hacks like Burgersdijck and Keckermann or
ill-tempered clerics like Crakanthorpe. It is little wonder that real scholastic
insights were misunderstood and that Aristotelian scholasticism was often
'refuted' by ignoratio elenchi.
62. Burgersdijck, Franco (1623), Idea Moral! Philosophiae, Leyden. Golius, Theophilus (1631), Epitome
doctrinae moralis, Argentorati (i.e. Strasbourg). Eustachius 1609. Morisanus: Morisan, Bernard
(1625), Commentarii et disputationes in libros logicos, physicos, el ethicos Aristotelis, Frankfort.
63. Scheibler, Christoph (1628), Philosophia compendiosa, Oxford. Fonseca, Petrus da (1577),
Commentariorum ...in libros metaphysicorum Aristotelis, Rome. Eustachius 1609. Suârez 1965.
64. Holdsworth was, however, very reluctant to see his writings published. The only thing to come
out in his lifetime was his sermon 'The Valley of Vision', which was printed as a pamphlet after,
he says, he was three times begged to allow its distribution. It appeared in more permanent form
shortly after his death in Holdsworth, Richard (1651), The Valley of Vision, London.
46
NEOSCHOLASTICISM
Intellectual and religious reaction to the French Revolution

The upheaval of the French Revolution destroyed many academic and
ecclesiastical organisations belonging to the old order, but a reaction was
not slow in coming. As Sainte-Beuve noticed in 1854, the number of Le
Moniteur for Easter Sunday 1802, which published news of the Peace of
Amiens and of the Concordat between Napoleon and the Pope, also
published a review of a recent book: Le Génie du Christianisme, by
Chateaubriand (1768-1848). An appeal to tradition against the excesses
that had followed from rationalism found romantic expression in
Chateaubriand's book, but the concept of tradition was to receive a more
philosophical cast from others - De Maistre (1763-1852), De Bonald
(1754-1840), and Lamennais (1782-1854). All three had idiosyncratic
views on the role of language, and the opposition they provoked in-
fluenced the form taken by the revival of scholasticism. Writing in 1809,
De Maistre claimed that the content of language depends upon the fife and
customs of those who use it; it eludes arbitrary enactments; it was not
invented by men, nor can its diversity be attributed to human means. 1
For De Bonald, the disagreements of philosophers oblige us to seek for
moral science what physical sciences have already: a fixed point, a criterion
of truth, something that will be public, readily accessible, and evident. He
believed this to be 'the primordial and indispensable gift of language,
bestowed upon the human race'. Reason and experience in individuals
need the setting and tradition of society, and of the language it hands on, to
reach truths that go beyond particular facts.2 Lamennais is the best known
of the three, and drew political conclusions of a very different sort, but he
too gives a special place to society and to language. The eccentricities of
individual reason are to be corrected by the beliefs that are held by the
human race as a whole, for these traditions - 'common sense', as he calls
1. De Maistre 1844, §§XLVII-LVII.

2. De Bonald 1830,1, pp. 5?f.
Neoscholasticism 839
them - go back to the truths primordially bestowed by God, bestowed

along with the gift of language.3 His political conclusions were of a
theocratic and 'democratic' character, threatened the relations established
between Church and State after the Napoleonic wars, and were one cause
of his condemnation by Rome in 1832 and 1834.4
The condemnation of Lamennais

In so far as the intervention by Rome had any philosophical motivation, it
was the novelty of what Lamennais wrote that caused offence,5 for there
was little official interest at Rome in philosophical matters. The teaching of
it to ecclesiastics at the Jesuit Collegio Romano was not based upon any
one tradition or authority.6 A good part of the course there then seems to
have been devoted to mathematics and science;7 indeed, a pupil at the
Collegio Romano in 1830 wrote afterwards that the only tenet held by all
was a contempt for Aristotle.8
The condemnation of Lamennais was taken by some at Rome as an
encouragement to return to the scholastics, but a statement9 from one
teaching philosophy there shows that the matter was regarded with tem-
pered enthusiasm. Lamennais and others, it said, had attacked the rationa-
lism of Descartes; but such attacks are rarely mortal, and Cartesianism has
much to offer. And if not Cartesianism, why not the scholastic philosophy
on which Descartes built his own? The Church has endured through the
ages, and does not need novelties in speculation.
The disciplining ofBautain

The opinions of Lamennais were not the only novelty of which the writer
disapproved. He goes on to mention someone else whose opinions were to
be opposed by the neoscholastic revival: Louis Bautain (1796—1867).
Bautain had attacked the scholastic method in philosophy, and had made
3. Lamennais 1819, ce. xiii, xx.
4. DS, 2730t; DB i6i7f; for contemporary comment by ajournai we shall meet again, seeJHL 1;
the political consequences of Lamennais' opinions are reprobated in Boyer 1835. (DS is Enchiri-
dion symbolorum ... quod primum edidit Henricus Denzinger et quodfunditus relractavit... Adolfus
Schbnmetzer S. I. DS gives extracts from ecclesiastical documents, with ample bibliographical
information. First edited by Denzinger in the nineteenth century, it was repeatedly revised; the
revision by Schonmetzer changed the numbering of the paragraphs found in the earlier standard
edition by Bannwart (DB).fHL is Journal historique el littéraire.)
5. DS 2732.
6. Dmowski 1836, p. 511, in a passage about Lamennais.
7. JHL 1: 197, part of a general account of studies at Rome.
8. Cited by Pelzer 1911, p. 251.
9. Anonymous 1834.
both the validity of arguments in favour of religious belief and the accep-
tance of miracles depend upon faith. Denounced by his bishop, he went to
R o m e in 1838 to present his case. 10 While recording the indifference at
R o m e to philosophy as such, he also shows that it was felt there that too
severe a condemnation of his fideistic views would be invoked in favour of
yet another school of thought - theologians in Germany sympathetic to
Georg Hermes (177S-1831), who had been condemned in 1835. n
Hermes, trying to come to terms with Kant's philosophy, would have
theology begin with a sceptical doubt about everything - including re-
ligion - and then recover the tenets of Christianity by an adaptation of
Kantian techniques of argument. The disfavour shown by R o m e towards
what was seen as rationalism mitigated its reaction to Bautain, when he had
to sign a formula in 1844. 12 Disapproval of Bautain, and of the other
philosophical innovations, did at least imply a preference for older ways of
thought, and the disapproval was eventually to strengthen the hands of
those who wished to revive them.
Victor Cousin
Meanwhile, however, interest in the older ways was being shown in purely
secular contexts. Victor Cousin (1792-1867) was the best known if not the
most skilful pioneer of the history of medieval philosophy, 13 and this
discipline was, eventually, to exercise the most important influence of all
on the neoscholastic revival.
Early nineteenth-century Thomism

Among the places that had preserved a teaching of scholastic philosophy
was, not surprisingly, Spain, and the Catalan Cardinal Boxadors had issued
a letter in 1757 demanding that St Thomas Aquinas be studied. One of
those who followed the lead which Boxadors had given was another
Catalan, the Jesuit B. Masdeu. When the Society was expelled from Spain
in 1767, Masdeu and others went to Italy. From 1799 to 1806 he taught at
Piacenza, in the Collegio Alberoni, and among the pupils he converted to
Thomism was V. Buzzetti (1777-1824), who was dissatisfied with the
teaching he had been receiving there — Condillac had lived at Parma, and
10. His Roman Journal, in Poupard 1964, is in my opinion a minor literary masterpiece.
11. DS 273 8f.
12. DS 27$6f.; Poupard 1964 and 1961 give a careful estimate of its force and setting.
13. In works such as Cousin 1836 and 1838.
his philosophy was popular in that part of Italy. 14 At all events, Buzzetti
himself taught Thomism at Piacenza to some who were to carry it to
Rome and (eventually) to favour there. Among these were Domenico and
Serafino Sordi, who both became Jesuits, and played a part in the curious
mixture of intrigue and politics that followed. During his novitiate in the
Society, Serafino handed on the doctrine to Luigi Taparelli d'Azeglio
(1793-1862) who in 1824 became the Rector of the newly-restored
Collegio Romano. We have already seen what variety of philosophical
teaching obtained there. Taparelli was grieved by this, and recruited a
group of students to whom, under pledge of secrecy, he imparted the
teaching he had himself received.15 One thing at least is clear: all those
involved in the affair touched up their memoirs in later years so as to read
back as much Thomism into their old selves as possible. Why, we shall soon
see; for the present, we can notice that one of the pupils so initiated was
Gioacchino Pecci, the future Leo XIII (1878-1903). Among the Jesuits
themselves, there was much opposition to the revival of scholasticism, and
Taparelli and one of the Sordi brothers were removed, first to Naples, then
to Sicily.16 The revolution of 1848 changed the balance of opinion at
Rome: the upheaval seemed a consequence of modern thought, and some
new instrument was needed for the defence of the Church. At the in
sistence of the new Pope (Pius ΓΧ, 1846-1878), the Jesuits founded the
periodical Civiltà Cattolica; its editorial board was dominated by those who
favoured a return to scholasticism,17 and from 1850 onwards the perio-
dical was advocating it as part of a more general policy of militant
ultramontanism.
The neoscholasticism of Joseph Kleutgen

The victory of ultramontanism still lay in the future. There existed a
variety of philosophical opinions in Roman Catholic institutions, and
some were to shape the development of neoscholasticism by the debates
and disagreements they caused. In Germany, for instance, Hermes had
14. In giving this account, I have followed Pelzer 1911, BatUori 1944, and Jacquin 1943. On
Boxadors, see also Tusquets 1923, and for an anthology of early neoscholastic writing in Italy, see
Dezza 1942-4. For other works, not seen by me (especially Masnovo's investigations), see the
exhaustive Rossi 1959, who himself minimises the influence of Masdeu, on grounds that
Scholastic philosophy had never been wholly neglected in Piacenza.
15. I follow here the works cited in n. 14 above, and notice that the footnotes in Jacquin 1943 are
valuable for their lengthy extracts from sources not easily available.
16. For Pelzer 1911, the reasons were philosophical; Jacquin 1943, c. 3, gives evidence that ad-
ministrative difficulties also played a part.
17. Jacquin 194J, c. 5.
been followed by Günther (1783-1863) and Frohschammer (1821-93),

who went to Hegel and Fichte respectively for categories in which tradi-
tional doctrines might be expressed. Their condemnations at Rome 18 are
associated with polemical activities by one of the first Germans to join the
cause of neoscholasticism, Joseph Kleutgen (1811-83). Kleutgen came to
scholastic philosophy only as an afterthought. Strongly opposed to all
forms of idealism, he wrote a defence of traditional theology against it; but
Giinther's attacks on the scholastic method as such led to his defending the
older philosophy as well in a much-translated book.19 Kleutgen had come
to live in Rome in the 1840s as a teacher of rhetoric, and his profession
shows in the crusading style of his book - a precedent that was to be widely
followed. In one sense, he does what Lamennais had done: he appeals to
tradition. But the tradition is one of approved philosophical speculation in
the Church, and to this believers must return;20 tradition in the sense that
Lamennais and others had used - the socially transmitted and primaevally
linguistic embodiment of knowledge - he rejects, just as he rejects the
fideism of Bautain and the rationalism of the Germans. For Kleutgen, what
matters most is a defence of objectivity in knowledge against idealism, and
the first half of his book is entirely concerned with epistemology. His
preoccupations made him see the scholastic revival in a way that many
would see it after him: a rejection of idealism, a rational vindication of
elements in religious belief, a call to return to a neglected but enduring
heritage. These preoccupations made him, just as they were to make others,
insensitive to values in what he was opposing: the creative function of the
understanding, the relation of arguments about religion to the setting of
religious belief, the role of language and of social instruction, the historical
setting of philosophical speculation.
Cartesian scholasticism
Another (French) tradition tended to neglect those same issues. A readily
accessible source, Renan's Souvenirs, shows us something of this tradition
in Renan's own time as a student. At Tréguier, the clergy at the college he
attended in the 1830s distrusted Chateaubriand, just as the Sulpician priests
at the seminary at Issy distrusted Lamennais.21 At Issy, the college for
18. 1857 and 1862; DS 2828f. and 285of.

19. Kleutgen i860; see also Lakner 1933.
20. i860, Einleitung §1.
xi. Renan 1975: III, ί, p. 106; IV, i, p. 142; IV, ii, p. 155. Boyer, cited earlier as opposing Lamennais,
was superior at Issy; Renan writes of his contact with Rome at IV, i, p. 142.
philosophy, various traditions of speculative thought were represented,22

but the textbook was scholastic and in Latin. However, the scholasticism
taught was not 'the barbarous and childish scholasticism of the thirteenth
century, but what might be called cartesian scholasticism, that modified
form of cartesianism which was widely adopted for ecclesiastical instruc-
tion and set down in three volumes known as the Philosophie de Lyon'.23 It
was this modified cartesianism that was to be the greatest obstacle to the
neoscholastic revival, and yet was to have so great an effect on it. In the first
place, it took for granted Descartes' starting-point in philosophy: by
reflection and criticism, the individual was to scrutinise in himself what
purported to be knowledge, and to discern the genuine from the spurious.
The neoscholastics were to oppose the cartesians and others in the name of a
revived medieval philosophy, but their starting-point was to be the same.
A full return to the Middle Ages would have called it in question, but the
neoscholastics did not do this; indeed, their rejection of theories that gave
pride of place to language and to society confirmed them in their
cartesianism.
That they were so confirmed, they showed by sharing the belief of
contemporaries that medieval logic was unworthy of serious notice. For
Renan, the vocabulary of medieval scholasticism was barbarous and child-
ish; an ecclesiastical writer of the time on philosophy could deplore the
Middle Ages, when 'the sacred sciences were held in the harsh bondage of
dialectic';24 and the Jesuit we saw opposing Taparelli could list 'Barbara
Celarent' along with alchemy as among the absurdities that a revival of
scholasticism would bring. 25 That neoscholastics thought much the same
can be seen from the flood of textbooks that started in the 1860s: whatever
else in them constitutes a revival, the section on logic does not. Indeed, the
editor of one of the first (it was a translation of a seventeenth-century book)
defends his printing of obsolete scientific opinions, but concedes that late
medieval logic had become over-subtle,26 and a similar remark about
over-subtlety is about the only qualification to the praises of Aquinas in a
Letter of Leo XIII.27
22. Renan 1975: IV, ii, p. 157.
23. '...non la scolastique du XIIIe siècle, barbare et enfantine, mais ce qu'on peut appeler la
scolastique cartésienne, c'est-à-dire ce cartésianisme mitigé qui fut adopté en général pour
l'enseignement ecclésiastique, au XVIIIe siècle, et fixé dans les trois volumes connus sous le nom
dc Philosophie de Lyon' (Renan 1975: IV, ii, p. 156).
24. Ferrari 1833, p. XVII.
25. Jacquin 1943, p. 295, n. 150.
26. Goudin 1864, p. xvii.
27. DS 3140.
Ontologism
There was a third school of thought which neoscholasticism opposed, and

which left its mark thereby upon the revival. Its defenders were largely in
Belgium, at the University of Louvain. 28 T o understand it, we must realise
that 'cartesian scholasticism' of the kind taught to Renan regarded itself as
the heir to a tradition going back to Plato. Augustine had reformed
Platonism according to Christian belief, and had spoken of God as the
illuminating sun of knowledge, making all else to be understood. 29 In
medieval times, the tradition had been represented by Bonaventure; the
innate ideas of Descartes, and even more Malebranche's vision of all things
in God, were in the same line of thought, making our understanding
depend upon divine enlightenment. 3 0 At Louvain, G. C. Ubaghs and
others combined this 'Augustinian' tradition with that emphasis upon
language and upon the social transmission of knowledge that we saw held
earlier in the century. Under the influence of the Italian philosopher
Gioberti (1802-53), then exiled to Brussels for republicanism, they saw his
'ontologism' as a continuation of Augustine's view. 31 An intuition of God
is the base of all further knowledge, and divine illumination gives knowl-
edge its necessity. 32 In controversy with neoscholastics, Ubaghs used
medieval sources then being made available to show the variety of opinion
permitted in medieval times, and to suggest that a dangerous tendency to
empiricism lay in Aristotelian elements in the opinions of Aquinas about
knowledge. 3 3 It is this point, rather than the quasi-religious terminology of
the debate, that was important for the development of neoscholasticism.
Precisely because the ontologists appealed to the 'Augustinian' strain in
traditional theories of knowledge, neoscholastics like Liberatore stressed
whatever in Aquinas was distinct from it. 3 4 Conceptual thinking was
explained by them in terms of an 'abstraction' that did not do justice to the
'illuminating' or active nature of such thinking. 3 5
28. A good account of this Louvain school is in Henry 1924; apart from the primary sources to be
quoted, the JUL and the Revue Catholique have an abundance of material from the 1840s to the
1860s.
29. A favourite text was Soliloquia, I 8; PL 32: 877.
30. For a statement of this belief, see Branchereau 1855, iii.
31. Labis, 1845.
32. Ubaghs i860.
33. Ubaghs 1861; see also Fabre 1862 and Claessens i860.
34. Liberatore 1855.
35. A comment on the debate from an English admirer of Branchereau is Meynell 1866. For
MeyneH's use of Branchereau, see letters of his in 1869 to Newman, who sought his criticism
when wriring the Grammar of Assent, Newman 1961-, XXIV: 306, 353.
The victory of neoscholasticism

If neoscholasticism was affected by its disagreement with all these other
schools of thought, the disagreement itself was solved by political rather
than by philosophical means. The advance of ultramontane opinions
favoured the adoption of a uniform system of philosophy throughout the
Church. 36 A further pragmatic and extraneous consideration was that
Aquinas was used in theology, and that his technical terms there are
unintelligible without some instruction in his philosophy. 37 In Goudin
1851, the editor's introduction contains personal reminiscence to this effect.
In all this, the ambiguity latent in the revival from the start becomes
palpable: neoscholasticism is rationally preferable; it must be, because the
preference has been part of Catholic tradition. That the revival bore the
marks of recent debates, and shared quite unmedieval presuppositions and
preferences with its opponents, was beyond the grasp or concern of many
who were promoting it. There was something especially paradoxical in
reviving a philosophy conspicuous for its attention to language, while at
the same time discounting the role of language. 38 But the battle with
ontologism was canonically over by the late 1860s, although R o m e was
slower to move than the Civiltà Cattolica. By then, supposedly ontologist
theses had been condemned as tending to pantheism and irrationalism, 39
but even the Vatican Council of 1870 did not commit itself to specific
philosophical positions. 40 Matters changed when, in 1878, Leo XIII suc-
ceeded Pius IX. As a student in R o m e , he had had scholasticism imparted to
him in secret at the Collegio Romano; as Nuncio in Brussels in the 1840s,
he had seen the beginning of the debates over ontologism; as Archbishop of
Perugia from 1846, he had (with his ex-Jesuit brother) enforced its teaching
there; as Pope, one of his first Encyclical letters was a call for the philosophy
of Aquinas to be taught and honoured throughout the R o m a n Church.
Aquinas has gathered together and increased the wisdom of all earlier
teaching: 'he has refuted by himself the errors of preceding times, and has
36. Ramièrc 1861 uses this argument among others (pp. 10-12,40-3); see also Ventura 1861, vol. i,
pp. cxv, 618-19; on the preposterous Padre Ventura, and on much else, see the valuable Fouchcr
1955.
37. Ramièrc 1861, pp. 21-3.
38. See, e.g., Ramicre 1861, p. 84.
39. DS 2841; ASS 3:204-24; further details in Henry 1924, p. 135f. (ASS is the Acta Sanctae Sedis in
compendium opportune redacta, which was succeeded in 1909 by the Ada Apostolicae Sedis as the
official journal for Roman documents and decrees.)
40. DS 30oof.
provided invincible weapons for the refutation of errors that were to be

ever springing up in days to come.' 4 1
Neoscholastic attention to medieval philosophy

It is fair to say that the circumstances of the victory of neoscholasticism
brought out its most conspicuous defect: the lack of a sense of history. The
need to place all human utterances in their cultural and temporal setting
had not been felt in the Middle Ages as it was being felt in the nineteenth
century; yet the patronage accorded to the revival was accorded by a
Church which found it difficult to place its own doctrines in that way, and
for which the elaborate vocabulary and distinctions of scholasticism
seemed to offer a way out. A neglect of the role of history in theology both
confirmed and was confirmed by an analogous neglect in philosophy.
Neoscholasticism took its first step towards maturity only when it began a
dispassionate examination ofjust what medieval philosophy had been, and
an inventory ofjust what had survived of it. One scholar to be mentioned
here links this study of history with an institute that played a special part in
the neoscholastic revival: Maurice De Wulf (1867-1947), who from 1894
to 1939 held the Chair of The History of Medieval Philosophy at the
Institut Supérieur de Philosophie at Louvain. Leo XIII, aware of the popu-
larity that ontologism had enjoyed at that university, wanted a chair of
Thomistic philosophy there. The first holder was D. Mercier (1851-1926),
whose foresight and determination eventually secured the founding in
1890 of the Institute as a place for graduate studies. 42 Among this gener-
ation of historians, M. Grabmann (1875-1959) contributed more than
anyone else to the discovery and editing of previously unknown texts. 43
O f the results achieved by these and by others, none was more important
than the display of the sheer variety in medieval philosophy - a variety not
only in opinions and interests, but in the genres of philosophical and
theological writing then. 4 4 W e can measure the advance made by the
41. The text of the letter - in the drafting of which Kleutgen is said to have played a part —is in ASS
12: 97-115; extracts are in DS 3I35Î, where omissions have given it an air of reasonableness.
Reactions to it in England are given in Holmes 1975; for its eventual effect on philosophical
teaching in a seminary there, see Milburn 1964, pp. 302-6.
42. An entertaining and well-documented account of the politics of the matter is in De Raeymaeker
1951; Tambuyser 1958, gives details of its complicated pre-history. Judicious estimates of
Mercier's own philosophical work are in De Raeymaeker 1952, and in Van Riet 1946, p. I34f. De
Wulfs example and influence were of the highest order; for an affectionate and perceptive
account, see Van Steenberghen 1948.
43. Schmaus 1959 shows just how monumental his achievements were.
44. For examples of how work done here could affect estimates by neoscholastics of their own past,
see Moreau 1951, in a volume dedicated to De Ghellinck.
history of medieval philosophy if we take a theme already mentioned - the

balance in it between the 'Augustinian' and 'Aristotelian' traditions - and
look back to the state of affairs at the turn of the century. At that time
pressure from Leo XIII45 obliged the editors of Bonaventure to claim his
identity in thought with Aquinas; and at that time a history like Picavet's
(1905), defending the interesting thesis that the clue to the scholastics lay in
an understanding of Plotinus, exhibited the exaggerations of a pioneering
book. Since those days, a debate in works by Gilson46 and Van
Steenberghen47 has successively explored and appraised the diversity of
elements in the thought of the thirteenth century; the synthesis in Van
Steenberghen 1966 has the strength it has because of the long tradition on
which it builds. The study of the history of medieval philosophy has
changed the concept of what a scholastic revival might be.
Neoscholasticism in relation to the philosophy and science of its own time

But if historical work showed medieval philosophy to be varied and
complex, and to have been occupied with its own particular range of
philosophical problems, it thereby pointed to the need for neoscholasticism
to consider the philosophical questions of its own day. The need was most
readily met in epistemology, concerning which, as we have seen, the
neoscholastics shared presuppositions of their contemporaries. If epistemo-
logical questions come again to be put where they are now out of fashion,
the sheer quantity and variety of answers offered will prove useful.48
Things were very different for any kind of philosophy of nature, for
medieval questions here seemed bound up with so much that had been
abandoned. The very obviousness of the difficulty has led to a multiplicity
of answers. Some have been as much historical as philosophical. At the
beginning of this century, Duhem had shown continuities between the
'new sciences' and what had gone before, and his example has been
followed. A work like Maier 1949 can be seen as doing for Galileo what
Gilson 1912 did for Descartes: showing, without compromising his place
in a long history, just how much he was part ofthat history. Philosophical
answers have not always been as successful. One has been to separate the
findings of modern science from natural philosophy, and to make the latter
45- ASS 31:264-7·

46. 1936,1955,1961a.
47· 1930. 1955.1966.
48. Van Riet 1946 is a lucid and wide-ranging account, which also throws light on other interests of
neoscholasticism.
alone concerned with the ultimate principles of things. For Maritain,

physical science of our day is concerned only that the final results of its
calculations should coincide with measurements made, and, unhke philos-
ophy, is not concerned with the inner ontological nature of reality. 49 The
view goes back to the early neoscholastic Zigliara, for w h o m metaphysics
decides on the essence of bodies, while physics touches only their 'thres-
hold'. 5 0 For obvious reasons, it has remained popular with theologians.
Other neoscholastics have taken a more serious view of modern science,
and have submitted that medieval developments of Aristotle can do it
justice in a way that mechanistic philosophy, with its reduction of all things
to extension and motion, cannot. An adversary such as 'mechanistic
science' is not only vague, but — modern science being what it is - can be
little more than a straw man. However, this claim too goes back to the
early days of the revival. 51 It is difficult to see how the conceptual distinc-
tions of medieval Aristotelianism can be combined with experimental
evidence expressed in terms of modern science. Thus, a scholastic adversary
of Hoenen holds that chemistry shows him to be wrong in claiming that
elements in a compound remain 'virtually, not formally'; they rather
remain 'formally, but without a new form of the compound'. 5 2 H o w the
matter could conceivably be settled, and what contribution by what
discipline would effect the settlement, is not investigated by either party,
Metaphysics in neoscholasticism
We saw that Kleutgen had argued against Kant as against other idealists,
and later neoscholastics such as Mercier devoted attention to him. 5 3 It was,
however, in the work of J. Maréchal (1878—1944) that neoscholasticism
tried most elaborately to come to terms with the challenge of the critical
philosophy. In a series of 'Cahiers' from 1922, he examined theories of
knowledge, and in particular the Kantian position. 54 His contention was
that, by beginning where Kant did, and by analysing what was involved in
49. Maritain 1959, p. 60.

50. Zigliara 1876, vol. 2. p. 81.
51. Comoldi 1893 is an English version of a work of 1864. The classic exposition of this claim,
supported with a massive background of science and history, is Hoenen 1945, part of which is in
English as Hoenen 1967.
52. Moran 1951, pp. 413—36, who gives references to Hoenen; the passages in Hoenen 1945 are
nn. 308-10, referring to nn. 275, 291.
53. Van Riet 1946, p. 177.
54. Maréchal 1926.
that starting-point, we can regain the metaphysics that Kant had banished
from the speculative intellect. Maréchal was influenced by Blondel
(1861-1949), whose philosophical speculations were based on analyses of
human action and of the implications of its needs.55 This attempt to begin
with a theory of understanding and to end with metaphysics has found
much support among German neoscholastics, of whom the best known is
probably the theologian, Karl Rahner - indeed, for Rahner, the 'tran-
scendental method', as it is called, is the legitimate heir to what was
neoscholasticism.56 Bernard Lonergan writes in this general tradition: his
work of 1957 will be admired for its aphorisms and phenomenological
descriptions even when its remoter purposes fail to win assent, or com-
prehension. Earlier articles assembled in Lonergan 1968 should be con-
sulted (despite their plethora of undiscriminated references to the text of
Aquinas) for an elaborate account and appraisal of what Aquinas says about
conceptual thinking and for comparisons with the views of other
scholastics.
Some neoscholastic contributions worthy of note have dealt with the
notion of esse ('existence', approximately) in Aquinas.57 Gilson and Fabro,
for instance, seek to vindicate for the speculations of Aquinas, and in
particular for what he says about God, a distinctive role for esse, a role that,
they claim, has been obscured by his successors. The philosophical part of
their claim raises questions that have been answered in different senses by
two writers on Aquinas outside the neoscholastic tradition. For Geach the
account of esse in what Aquinas writes about God in his mature works is
coherent, while for Kenny it is not. 58 1 am not concerned with the merits of
their disagreement, but I point out that they are agreed about the kind of
argument to be used; that the arguments they do use - often with close
reference to the text of Aquinas - are not of the sort found in Fabro or
Gilson.59 What causes the gap here - and, of course, not only here -
between two types of attention to Aquinas? Are we faced with a con-
sequence of the lack of interest in logic shown by neoscholastics? Certainly,
55. See, e.g., Blondel 1893.

56. Foreword to Muck 1968. Well-written examples of what this tradition is attempting, and of its
contact with other schools, are the essays in Lotz 1955.
57. Gilson 1952a is a well-known example, while Fabro 1939, Geiger 1942, and Fabro i960, combine
the investigation with an account of those elements in the thought of Aquinas that are concerned
with the relation of participation in creatures towards God.
58. Geach 1961; Kenny 1969.
59. Geach has expressed elsewhere (1972, pp. 263-5) dissent from arguments employed by Gilson.
the lack is worthy of note, and we should also note that the exceptions
themselves are odd.
Logic in neoscholasticism
It is the rambling and at times ridiculous Sanseverino who shows an
acquaintance with the then recent controversy of De Morgan with
Hamilton, and who speaks with respect of Stoic logic, citing Diogenes
Laertius and Sextus Empiricus.60 And, by a strange turn of history, the
chief logical influences of neoscholasticism in a broad sense came through
two men who had broken with the Roman Church: Franz Brentano
(1838-1917) and Anton Marty (1847-1914). Brentano was above all a
student of Aristotle, but he had also studied medieval philosophy under
Clemens, a friend ofKleutgen.61 His philosophical influence in Poland was
to have important results. Twardowski was impressed with 'scholastic'
elements in Brentano 's philosophy, such as an insistence on clear de-
finitions and on logical rigour. He passed on these standards to his pupil
Lukasiewicz, who had attended, but had not valued, lectures at the Institut
Supérieur at Louvain.62 This zeal for exactness of statement was found in
Marty's writings by Lesniewski, another logician from Poland.63 And
indeed Marty sets a similar goal for philosophy, and gives logic and
grammar a place in it that is thoroughly 'scholastic', but which no neoscho-
lastic had given them. 64 The influence of Lukasiewicz on the study of
ancient and medieval logic is notorious: it is worth recalling the devious-
ness of the route by which it all started.
Methodological obscurity
Nonetheless, I see this lack of interest in logic as only part of a more general
weakness in neoscholasticism, which I would describe as a lack of methodo-
logical clarity, already noticed when considering neoscholastic approaches
to natural science. The growth of the new sciences is more than a juxta-
position of new questions with old; it calls for an account of the old
questions that will show them to be still worth asking.65 We can go for
60. Sanseverino 1862,1, vol. 2, cap. 2, art. IX; cap. 3, art. IV.
61. Biographical and critical material is in McAlister 1976; Brentano 1968 gives an interesting
appreciation of Aquinas (written in 1008), with remarks on neoscholasticism.
62. Sobocinski 1956, pp. 3f.
63. Kotarbinski 1967.
64. See Marty 1908, dedicated to Brentano.
65. McMuUin 1963, which contains an exchange between a neoscholastic and an analytic philo-
sopher, shows an awareness of how real the difficulty is.
another example of the difficulty to a topic already met - the scholastic

account of conceptual thinking. It has an elaborate terminology of'active
intellect', 'passive intellect', 'phantasm', 'illumination', 'abstraction', and
the rest. That we have metaphors here need be no demerit in so funda-
mental a matter. But what check has been made upon the growth of the
metaphor? How far does each term have an independent sense? What are
we supposed to be doing in drawing the distinctions? Too often, qualifying
words like 'metaphysical' or significantly untranslatable terms like 'entia
quibus' have been used to block questions that needed putting, here as in the
metaphysical speculations on 'esse'. Methodological obscurity can be said
to go with a lack of attention to logic and language, for a greater attention
might at least have brought out the varied functions of words, and the
impossibility of getting a satisfactory semantics by simply making all
words 'mean things', and then affixing deterrent labels to the things meant.
Philosophy can err in the matter of self-examination by defect as well as by
excess.
Neoscholasticism's loss of distinctiveness

These obscurities have lost the significance they once had, because neo-
scholasticism is no longer the distinctive body of thought it once was. One
reason for this development lies in the Roman Church. From the turn of
the century, it has been trying to face the relationship between doctrine and
history, and neoscholasticism proved of little assistance in the modernist
crisis of the early 1900s.66 Further dissent, especially since 1945, found
expression in the Second Vatican Council, where philosophical pluralism
was acknowledged.67 Another reason has been the absorption of neoscho-
lastic activities into the general stream of philosophy. The extent of the
absorption can be seen by an inspection of periodicals over the years,68 or
by surveying the unrivalled 'Chroniques' in the Revue Philosophique de
Louvain. The Institut Supérieur itself has seen a shift of interest towards
Husserl, whose Nachlass is deposited there.69 And, as its 'Centre De Wulf-
Mansion', research goes on into medieval logic among other things.70 The
entry of the neoscholastic revival into the general course of philosophy can
66. I have argued this in FitzPatrick 1973.
67. Baldanza 1970; Mclnerny 1966 illustrates the problems this has created.
68. On them, and on other activities among neoscholastics, ample information is to be found in De
Raeymaeker 1947.
69. On which, see Van Breda 1971 ; on the division of the University in 1970, see Revue Philosophique
de Louvain 69:164, and later 'Chroniques*.
70. See, e.g., the commemorative articles of 1977 in Revue Philosophique de Louvain 75.
be seen in this History itself. It would hardly have been written if the revival
had not taken place, but it reflects many interests that would not have been
stressed by those who, from a mixture of motives, worked a century and
more ago for a revived scholasticism. That it can reflect those interests is a
tribute to the rich diversity of what was being revived.
BIOGRAPHIES
BIOGRAPHICAL I N F O R M A T I O N ON
MEDIEVAL A U T H O R S
These very concise presentations of biographical information are intended solely as a

convenience for readers of this volume. They cannot replace the biographies in such
standard reference works as The New Catholic Encyclopedia, The Encyclopedia of Philosophy,
or The Dictionary of Scientific Biography. Only the more prominent medieval Latin authors
who belong to the period specifically covered by this History are included here, and only so
much of their careers and writings as is directly pertinent to the purposes of the History is
presented. For most medieval authors very little biographical information of any kind is
available. We have tried in general to provide at least the following information: the
author's name (along with well-known or confusing variants); dates (or the death date, or
the ßoruit); texts of the author's works (editions whenever possible, manuscripts when
necessary); secondary literature (a selection chosen with a view to guiding the reader to
other articles and books on the author and his work).
ADAM W O D E H A M (Wodham, Wodam, Godam)

BIOGRAPHICAL DATA: b . ca. I298; d. I358.
TEXTS: Quaestiones in libros Sententiarum (is 12). Editiomedia, ed. Henry of Oyta, published
by John Major, Paris; Prologue to Ockham's Summa logicae, q. v.; Tractatus de indivisibilibus:
MS Flor. Bibl. Naz. Conv. Soppr. A HI 508 ff. I 3 s n - i 4 7 * ; MS Flor. Bibl. Naz. Conv.
Soppr. Β VHff. I33 r -H3'·
SECONDARY LITERATURE; Courtenay, William J. (1978). Adam Wodeham: An
Introduction to His Life and Writings, E.J. Brill; Murdoch, John and Edward Synan (1966).
'On the Composition of Continua', Franciscan Studies 26:267-88.
AENEAS S Y L V I U S (PopePius II)

BIOGRAPHICAL DATA: b. I4O5; d. I46+.
TEXTS : De ortu et auctoritate imperii romani (1446); De iurisdictione, auctoritate et preaeeminentia
imperiali (1566), ed. S. Schard, Basle.
A L A N O F L I L L E ( Alanus ab Insulis)
BIOGRAPHICAL DATA: d. I203.
TEXTS: Deplanctu naturae, PL 210; The Complaint of Nature (1908), D.M. Moffat (tr.) (Yale
Studies in English, XXXVI), H. Holt.
SECONDARY LITERATURE: D'Alverny, M. T. (1965). Alain de Lille: Textes inédits,]. Vrin.
A L B E R T T H E G R E A T (Albertus Magnus, Albert of Cologne)

BIOGRAPHICAL DATA: b. ca. Ι20Ο; d. 1280. Joined Dominican order 1223 while studying
at Padua. Lector of theology in Germany, 1220s to 1230s. Sent to Paris, 1240s. Master of
Sacred Theology, 1245. Taught at Cologne, 1249, where Thomas Aquinas was one of his
students. Paraphrased or commented on most of Aristotle's works, 1250s to 1260s.
TEXTS: Opera omnia (1651), ed. P. Jammy (21 vols.), Lyon; Opera omnia (1890-9), ed.
A. Borgnet (38 vols.), Vives; Opera omnia (1951-), ed. Bernard Geyer ef ai, Aschendorff.
856 Biographies
SECONDARY LITERATURE: Catania, Francis J. (1959). Ά Bibliography of St. Albert the

Great', The Modern Schoolman 37:11-28; Houde, R. (1961). Ά Bibliography of Albert the
Great: Some Addenda', The Modem Schoolman 39:61-4; Meersemann, G. G. (1931).
Introductio in Opera Omnia B. Alberti Magni, Bruges, Beyaert; Pelster, Franz (1920). Kritische
Studien zum Leben und zu den Schriften Alberts des Crossen, Stimmen der Zeit; Schooyans, M.
(1961). 'Bibliographie philosophique de S. Albert le Grand (1931-1960)', Revista da
Universidade Catolica de Sao Paolo 21:36-88; Schwertner, T. M. (1932). St Albert the Great,
Marquette University; Weisheipl, J. Α., ed. (1980). Albertus Magnus and the Sciences,
Pontifical Institute of Mediaeval Studies.
A L B E R T O F S A X O N Y (Albertus Parvus, Albertutius, Albertus de Saxonia)

BIOGRAPHICAL DATA: d. 1390. Studied in Prague and Paris. Master of Arts at Paris, 1351.
Taught in arts faculty at Paris, 13 51 -62. Helped to found University of Vienna, of which he
became the first rector in 1365.
TEXTS : Commentarius in Posteriora Aristotelis (1497,1522), Venice; Expositio aurea et admodum
utilis super artem veterem édita per venerabilem inceptorem fratrem Gullielmum de Ockham cum
quaestionibus Albertiparvi de Saxonia (1496), Bologna; Perutilis Logica (1518, 1522), Venice
(Reprinted Olms 1974); Quaestiones de caelo et mundo (1520), Venice; Quaestiones de gene-
ratione et corruptione (1568), Paris; Quaestiones totius libri Physicorum (1518), Paris; Sophismata
(1495, 1502), Paris (Reprinted Olms 1975).
SECONDARY LITERATURE: Boehner, Philotheus (1952). Medieval Logic, Manchester
University Press; Clagett, Marshall (i959)· The Science of Mechanics in the Middle Ages,
University of Wisconsin Press; Gonzalez, Atanasio (1958-9). 'The Theory of Assertoric
Consequences in Albert of Saxony', Franciscan Studies 18:290-354 and 19:13-114;
Heidingsfelder, Georg (1927)· Albert von Sachsen: Sein Lebensgang und sein Kommentar zur
nikomachischen Ethik des Aristoteles (BGPM XXII, 3-4), Aschendorff; Moody, Ε. Α. (1970).
'Albert of Saxony', in Dictionary of Scientific Biography.
A L E X A N D E R O F HALES
BIOGRAPHICAL DATA: b. ca. 1185 in Gloucestershire; d. 1245. Studied and taught theology
at Paris. His Summa is the earliest philosophical contribution by a Franciscan and one of the
earliest medieval works to be based on full knowledge of the Aristotelian corpus and the
Arabic commentators.
TEXTS: Summa theologica (1924-48), Collegium S. Bonaventurae; Glossa in quatuor libros
sententiarum (1951-7), Collegium S. Bonaventurae.
SECONDARY LITERATURE: Gössmann, Elisabeth (1964). Metaphysik und Heibgeschichte:
eine theologische Untersuchung der Summa Halensis, Hueber.
A N D R É DE N E U F C H Â T E A U (Andreas de Novo Castro)

BIOGRAPHICAL DATA: fl. ca. 1360.
TEXTS: In primum librum Sententiarum (1514), Paris. Parts translated into French in Hubert
Elie (1937), Le Complexe significabile, J. Vrin; pp. 83-138.
SECONDARY LITERATURE: Elie, Hubert (1937)· 'André de Neufchâteau, dit "Le docteur
très ingénieux", Étude bio-bibliographique', in his Le Complexe significabile, J. Vrin;
pp. 225-52.
ANSELM O F C A N T E R B U R Y (Anseimus Beccensis)

B I O G R A P H I C A L D A T A : b. ca. io33;d. 1109.
TEXTS: Sancti Anselmi Opera Omnia (1938-61), ed. F. S. Schmitt, Nelson (Reprinted
Stuttgart, Frommann 1968); Schmitt, F. S. (1936). 'Ein neues unvollendetes Werk des hl.
Biographies 857
Anselm von Canterbury' (BGPM XXXIII, 3), Aschendorff; Schmitt, F. S. and R. W.

Southern (1969). Memorials of St Anselm, British Academy.
SECONDARY LITERATURE: Henry, D. P. (1964). The De Grammatico of St Anselm
(Publications in Medieval Studies, 18), University of Notre Dame Press; Henry, D. P.
(1967). Tfie Logic of Saint Anselm, Clarendon Press; Henry, D. P. (1974). Commentary on 'De
Grammatico' (Synthese Historical Library, 8), Reidel; Hopkins, Jasper (1972). A Companion
to the Study ofSt Anselm, University of Minnesota Press; Hopkins, Jasper and H. Richaidson
(1974-6). Anselm of Canterbury (4 vols.), Edwin Meilen.
A U G U S T I N U S T R I U M P H U S (Agostino Trionfo; of Ancona)

BIOGRAPHICAL DATA: b. ca. 1270/3; d. 1328. Augustinian Order of Hermits. Studied at
Paris 1297-1300; lectured on Sentences there 1302-4 or 1304-6. Lector in the Augustinian
School at Padua; returned to Paris as Master of Theology 1313 -15. Chaplain to Charles, son
of King Robert of Naples, in 1322.
TEXTS: Summa de potestate ecclesiastica (1582, 1583, 1584, 1585), Rome; Tractatus brevis de
duplici potestate prektorum et laicorum (1903), ed. R. Scholz (Kirchenrechtliche Abhand
lungen, 6-8), Enke.
SECONDARY LITERATURE: McCready, William D. (1974). 'The Problem of the empire
in Augustinus Triumphus and late medieval papal hierocratic theory', Traditio 30:325-49;
McCready, William D. (1977). 'The papal sovereign in the ecclesiology of Augustinus
Triumphus', Medieval Studies 39:117-205; Ministen, Β. (1953). De Vita et Operibus
Augustini de Ancona O.E.S.A. (ΊΊ328); Wilks, Michael (1963). The Problem of Sovereignty in
the Later Middle Ages: The Papal Monarchy with Augustinus Triumphus and the Publicists
(Cambridge Studies in Medieval Life and Thought, New Series 9), Cambridge University
Press.
B A L D U S DE U B A L D I S
BIOGRAPHICAL DATA: b. ca. 1320; d. 1400. Roman lawyer and canonist who taught at
Bologna, Perugia, Pisa, Florence, Padua, and Pavia.
TEXTS: Consilia (1608-9). Venice; L'opera di Baldo nel V centenario delta morte del grande
giutetansulto (sooo-i) ed. I, Tarducci (Annali dell' Université di Perugia, Facoltà di
Giurisprudenza, 10-11), University of Perugia Press.
SECONDARY LITERATURE: Horn, Norbert (1968). Aequitas in den Lehren des Balduis
Böhlau; Lange, Hermann (1973). Die Consilien des Baldus de Ubaldis (f 1400), Verlag der
Akademie der Wissenschaften und der Literatur (Mainz).
BARTHOLOMEW KECKERMAN
BIOGRAPHICAL DATA: b. 1571; d. 1609. Student at Wittenberg, Leipzig, and Heidelberg.
Taught at Heidelberg and Danzig. Wrote and published widely in astronomy, physics, and
mathematics, in addition to theology and philosophy.
TEXTS: Opera Omnia (1614), Geneva.
SECONDARY LITERATURE: Rose, Paul Lawrence (1973). 'Keckerman, Bartholomew' in
Dictionary of Scientific Biography.
BARTOLUS OF SASSOFERRATO
BIOGRAPHICAL DATA: d. 1357.
TEXTS: Opera omnia (1588), Basle.
SECONDARY LITERATURE: C. N. S. Woolf (1913). Bartolus of Sassoferrato, Cambridge
University Press; Calasso, F. (1964)- 'Bartolo da Sassoferrato' in Dizionario biogrqfico degli
italiani, Istituto delta Enciclopedia italiana, Società Graftca Romana; vol. 6, pp. 640-9.
858 Biographies
B O E T H I U S O F D A C I A (Boethius Dacus)
BIOGRAPHICAL DATA: fl. 1275. Master of arts in Paris ca. 1270. Involved in the condem-
nation of 1277.
TEXTS: Opera omnia (1969-), in Corpus Philosophorum Danicorum Medii Aevi, ed. Jan
Pinborg, Gad; Godfrey of Fontaine's Abridgement of Boethius of Dacia's Modi Significandi
(1980), ed. and tr. A. Charlene Senape McDermoct, Benjamins.
SECONDARY LITERATURE: Fioravanti, G. (1969-70). 'Scientia, fides, theologia in Boezio
di Dacia', Atti delta Accademia delle scienze di Torino (Classe di scienze morale), pp. 525-632;
Pinborg, Jan (1974). 'Zur Philosophie des Boethius de Dacia. Ein Ueberblick', Studio
Mediewistyczne 15:165-85.
B O N A V E N T U R E (John of Fidanza)
BIOGRAPHICAL DATA: b. ca. 1217; d. 1274. Studied in Paris beginning 1234/5. Master of
arts in Paris around 1243. Joined Franciscans 1243/4. Master of theology 1254/5. Elected
minister general of Franciscans 1257. Active against radical Aristotelians (the so-called Latin
Averroists) in Paris from 1260s until his death.
TEXTS: Opera Omnia (1882-1902), ed. Collegium S. Bonaventurae.
SECONDARY LITERATURE: Bougerol.J. Guy (1964). Introduction to the Works ofBonaven-
ture, St Anthony Guild Press; Brady, I. (1975). 'The Opera Omnia of Saint Bonaventure
Revisited', Proceedings of the Seventh Centenary Celebration of the Death of Saint Bonaventure,
The Franciscan Institute; Quinn, J. F. (1973). The Historical Constitution of St Bonaventure's
Philosophy, The Pontifical Institute of Mediaeval Studies.
BONSEMBIANTE BEDUARIUS OF PADUA

BIOGRAPHICAL DATA: b. 1332; d. 1389.
TEXTS: Quattuor principia; MSS Codex Latinus Monacensis 26. 711; Codex Vaticanus
Latinus 981.
SECONDARY LITERATURE: Elie, Hubert (1937). Le complexe significabile,]. Vrin.
C A M P A N U S OF N O V A R A
BIOGRAPHICAL DATA: b. ca. 120s; d. 1296. Primarily a mathematician; cited by Roger
Bacon in 1267 as excellent in mathematics. Probably a master at Paris or Bologna. Closely
associated with Pope Urban IV (1261-64).
TEXTS: Euclidis Megarensis mathematics clarissimi Elementorum geometricorum libri XV cum
expositione Theonis in priores XIII a Bartholomaeo Veneto Latinitate donata, Campani in omnes et
Hypsicles Alexandrini in duos postremos (1546), Basle; Theorica planelarum (1971), ed. and tr.
F. S. Benjamin and G. J. Toomer in Campanus ofNovara and Medieval Planetary Theory
(University of Wisconsin Publications in Medieval Science, 16), University of Wisconsin
Press.
SECONDARY LITERATURE: Murdoch, John E. (1968). 'The Medieval Euclid: Salient
Aspects of the Translations of the Elements by Adclard of Bath and Campanus ofNovara',
Reime de Synthese 89:67-94; Toomer, G.J. (1971). 'Campanus ofNovara' in Dictionary of
Scientific Biography.
DANTE ALIGHIERI
TEXTS: De monorchia (1963), ed. E. Moore and P. Toynbee in Le Opere di Dante Alighieri
(5th edn.), Oxford University Press; Dante Alighieri: Monorchia (1965), ed. P. G. Ricci,
Mondadori.
SECONDARY LITERATURE: Gilson, Etienne (1939). Dante et la philosophie,]. Vrin.
Biographies 859
D O M I N G O BAISEZ
BIOGRAPHICAL DATA: b, 1528; d. 1604. Joined Dominicans 1547. Studied under Melchior
Cano at Salamanca. Taught at Salamanca, Avila, Alcalâ, Valladolid. Advisor and confessor
of St Teresa.
TEXTS: Scfioliiiiifo commentaria in primam partem Summae theologiae D. Thomae Aquinatis
(1934), ed. L. Urbuno, F.E.D.A. (Valencia); Commentarium in libros De generatione et
corruptione (1585), Salamanca; Institutiones minores dialecticae (1590), Salamanca.
D O M I N G O DE S O T O
BIOGRAPHICAL DATA: b. 1494/95; d. 1560. Studied at Paris. Returned to Spain, 1519.
Entered Dominican order, 1525. Taught at Salamanca. Attended council of Trent, iS45·
Confessor to Charles V.
TEXTS: De iustitia el iure (1553/54, 1556/57), Salamanca.
SECONDARY LITERATURE: Munoz-Delgado, V. (1964). Logica formal y filosofia en
Domingo de Soto (Publicaciones del Monasterio de Poyo, 16), Estudios de Madrid.
E D W A R D B R E R E W O O D (Brierwood)
BIOGRAPHICAL DATA: b. ca. 1565; d. 1613. Studied at Brasenose College, Oxford.
Professor of Astronomy at Gresham College, London.
TEXTS: Commentarii in Ethica Aristotelis (1640), Oxford; Elementa logica (1614), London;
Enquiries touching the Diversities ofLanguages and Religions Through the ChiefParts ofthe World
(1614), London; Tractatus quidam logici de praedicabilibus et praedicamentis (1628), Oxford.
FRANCISCO SUAREZ
BIOGRAPHICAL DATA: b. 1548; d. 1617. Entered Society ofJesus in 1564. Studied canon
law at Salamanca. Taught at Avila. Segovia, Valladolid, Rome, Alcala, Salamanca, and
Coimbra. His theological writings include a commentary on Aquinas' Summa theologiae.
TEXTS: De bello [= De triplici virtute theologica, tr. Ill, disp. ΧΠΪ] (1954), ed. L. P. Vincente,
Madrid Instituto Francisco de Vitoria; De legibus (1971-), ed. L. Perena ei al. (Corpus
Hispanorum de Pace, 11), Madrid Consejo Superior de Investigaciones Cientificas,
Instituto Francisco de Vitoria; On Formal and Universal Unity (1964), J. F. Ross (tr.),
Marquette University Press; On the Various Kinds of Distinctions (1947), C. Vollert (tr.),
Marquette University Press; Opera omnia (1856-78), Vives; Disputationes metaphysicae
(Reprinted from Opera omnia, Olms 1965); Selections from Three Works (1944), G. W.
Williams and James Brown Scott (ed. and tr.) (Classics of International Law, 20),
Clarendon Press.
SECONDARY LITERATURE: Grabmann, Martin (1926). 'Die Disputationes Metaphysicae
der Franz Suarez in ihrer methodischen Eigenart und Fortwirkung' in his Mittelalterliches
Geistesleben, vol. 1, Hueber; Mahieu, L. (1921). Francois Suarez, sa philosophie et les rapports
qu'elle a avec sa théologie, Desclée; Mugica, Placido (1948). Bibliografia suareciana,
Universidad de Granada; Wilenius, Reijo (1963). The Social and Political Theory ofFrancisco
Suârez (Acta Philosophica Fennica, 15), Societas Philosophica Fennica.
F R A N C I S C O DE V I T O R I A
BIOGRAPHICAL DATA: b. 1483/86; d. 1546. Spanish theologian and jurist. Studied and
taught in Paris around 1506-23. Held chair of philosophy at Salamanca from 1526.
TEXTS: Relecciones leolègicas (i960), ed. T. Urdânoz, Biblioteca de Autores Cristianos;
Francisa de Victoria De Indis et De iure belli relectiones (1917), ed. E. Nys, Carnegie Institution
of Washington.
86ο Biographies
SECONDARY LITERATURE: Hamilton, Bernice (1963). Political Thought in sixteenth-

century Spain: A study ofthe Political Ideas of Vitoria, De Soto, Suarez, and Molina, Clarendon
Press.
F R A N C O B U R G E R S D I J C K (Burgersdicius)
BIOGRAPHICAL DATA: b. 1590; d. 1629. Studied at the University of Leiden. Studied
theology and taught at Samur. Professor of logic and moral philosophy (later also natural
philosophy) at Leiden.
TEXTS: Idea philosophiae moralis (1623), Leiden; Institutionum logkarum libri duo (1626),
Leiden; Institutionum logkarum synopsis sive rudimenta logica (1626), Leiden.
GABRIEL VÂSQUEZ
BIOGRAPHICAL DATA: b. ca. 1551; d. 1604. Entered Society ofJesus in 1569. Student at
Alcalâ. Taught at Alcalâ, Ocafla, Madrid, Rome.
TEXTS: Commentaria ac disputationes in primam partem S. Thomae (1620), Antwerp; (1631),
Lyon.
SECONDARY LITERATURE: Gilson, Etienne (1930). Études sur le rôle de la pensée médiévale
dans la formation du système cartésien,]. Vrin; Maurer, Armand (1962). Medieval Philosophy,
Random House.
G A E T A N O DI T H I E N E
BIOGRAPHICAL DATA: b. 1387; d. 1465. Taught at Padua and held the chair of logic there
between 1422 and 1430.
TEXTS: Complementum expositionis Messini de tribus praedicamentis (1494), Venice; Declaratio
super tractatu Hentisberi Regularum (1494), Venice; De intensioneet remissioneformarum (1491),
Venice; Expositio in libros De coelo et mundo (1484), Venice; Recollectae super Consequentias
Strodi; MS Venice, San Marco, lat. VI. 160, ff. I09 r -n8 r ; Recollectae super octo libros
Physicorum Aristotelis (1496), Venice; Recollectae super Sophismatibus Hentisberi (1494),
Venice.
SECONDARY LITERATURE: Clagett, Marshall (1959). The Science of Mechanics in the
Middle Ages, University of Wisconsin Press; Valsanzibio, P. S. da (1948). Vita e Dottrina di
Gaetano di Thiene, Scuola tipografica Madonna di Castelmonte; Wilson, Curtis (1956).
William Heytesbury: Medieval Logic and the Rise of Mathematical Physics, University of
Wisconsin Press.
GARLANDUS COMPOTISTA
BIOGRAPHICAL DATA: fl. n t h century. Native of Lorraine. Studied at one of theLiégian
schools. Went to England under the reign of Harald I (1036-40). Magister scholarum at
Besançon in 1084.
TEXTS: Garlandus Compotista, Dialectica (1959), ed. L. M. De Rijk, Van Gorcum.
SECONDARY LITERATURE: Henry, D. P. (1975). 'The Singular Syllogisms of Garlandus
Compotista', Revue internationale de philosophic 113:243-70; Stump, Eleonore (1980).
'Dialectic in the Eleventh and Twelfth Centuries: Garlandus Compotista', History and
Philosophy of Logic 1:1-18.
G E O F F R E Y O F H A S P H A L L (Aspall)
BIOGRAPHICAL DATA: fl. 1270. Master and regent of arts in Paris before 1265. Prominent
at Oxford around 1270. Wrote several commentaries on Aristotle's works.
TEXTS: Quaestiones in Metaphyskam, MS Cambridge Gonville and Caius College 509.
SECONDARY LITERATURE: Callus, D. A. (1943). 'The Introduction of Aristotelian
Biographies 86ι
Learning to Oxford', Proceedings of the British Academy 29:229-81; Macrae, E. (1968).

'Geoffrey of Aspall's Commentaries on Aristotle', Medieval and Renaissance Studies
6:94-134; Pinborg, Jan (1979). 'The English Contribution to Logic Before Ockham',
Synthese 40:19-42.
G E R A R D O F O D O (Guiral Ot, Geraldus Odonis)

BIOGRAPHIC AL DATA: fl. ca. 1325, in Paris. Elected Franciscan minister general in 1329. As
a result of criticism by John the Canon, he became known for his unorthodox views on the
existence of a void and the composition of continua.
TEXTS : In quatuor libros sententiarum, MS Paris, BN lat. 3068; De suppositionibus (1975)· ed. S.
Brown, Franciscan Studies 35:5-44; Expositio in Aristotelis Ethicam (1482), Brescia.
SECONDARY LITERATURE: Bartolomé, L. (1928). Fray Gerardo de Odén, Ministro General
de la Orden Franciscana (1329-42), Tipografia San Francisco; Langlois, C. (1927)· 'Guiral
Ot (Geraldus Odonis), Frère mineur', Histoire littéraire de la France 36:203-25; Teetaert
A. (1932). O t , Guiral,' Dictionnaire de théologie Catholique, vol. i l , 1658-63; Walsh, J.
(1975). 'Some Relationships Between Gerald Odo's and John Buridan's Commentaries on
Aristotle's Ethics', Franciscan Studies 251237-75.
GILES O F R O M E (Aegidius Romanus, Egidius Colonna)

BIOGRAPHICAL DATA: b.ca. 1243-7 in Rome; d. 1316. Joined Hermits ofSt Augustine at
age fourteen. Sent to Paris to study in 1260. Completed study ofliberal arts, 1266. Studied
theology, probably under Thomas Aquinas, 1269-72. Involved in the condemnation of
1277 and returned to Italy. Resumed teaching as master of theology at Paris 1285-91 and
was the Augustinians' first regent master in theology. Elected General of his order in 1292.
TEXTS: De ecclesiastica potestate (1929), ed. R. Scholz, H. Boehlaus Nachfolger (Reprinted
Aalen, Scientia Verlag 1961); De medio in demonstration potissima (1976), ed. Jan Pinborg,
Miscellanea Medievalia 10:254-68; De regimine principum (1502), Venice; (1607), Rome; De
renuntiatione papae (1698), ed. J. T. Rocaberti, Bibliotheca maxima pontificia, Rome;
Theoremata de esse et essentia, ed. E. Hocedez, Louvain, Museum Lessianum (1930). Also in
Theorems on Existence and Essence (1952), M. V. Murray (tr.), Marquette University Press.
SECONDARY LITERATURE: Bruni, G. (1936). Le Opere di Egidio Romano, Olschki;
Huebener, W. (1968). Studien zur Theorie der kognitiven Repräsentation in der mittelalterlichen
Philosophie, unpublished Habilitationsschrift, Freie Universität, Berlin; Mandonnet, P.
(1910). 'La carrière scolaire de Giles de Rome', Revue des sciences philosophiques et théologiques
4:480-99; Nash, P. (1967). 'Giles of Rome', New Catholic Encyclopedia.
GODFREY OF F O N T A I N E S
BIOGRAPHICAL DATA: b. ca. 1250; d. 1306-9. Studied arts at Paris in the early 1270s.
Studied theology under Henry of Ghent and Gervais of Mt St Elias at the Sorbonne
beginning at least as early as 1274. Master in the theology faculty from 1285 to 1298-9 and
again ca. 1303-4.
TEXTS: Disputed Questions 1, 7, 8, 13, 15, in Neumann, B. (1958), Der Mensch und die
himmlische Seligkeit nach der Lehre Gottfrieds von Fontaines, Limburg, Lahn-Verlag,
pp. 152-66; Disputed Questions 4 and 5, in Lottin, O. (1954), Psychologie et morale au XIIe et
XIIP siècles, Abbaye du Mont César, vol. 4, pt. 3, pp. 581-8 and 591-7; Disputed Questions
9,10, and 12, in Wippel, J. F. (1973), 'Godfrey of Fontaines: Disputed Questions 9,10, and
12', Franciscan Studies 33 '.351—72; Disputed Question n , in Lottin, Ο. (1949), Psychologie et
morale au XIF et XHF siècles. Abbaye du Mont César, vol. 3, pt. 2, pp. 497-502; Disputed
Question 15, in Koch, J. (1930), Durandi de S. Porciano O. P. Tractatus de habitibus Quaestio
Quarta (Opuscula et Textus, Series scolastica, 8), Münster, Aschendorff, pp. 60—6; Disputed
862 Biographies
Question 19, in Lottin, O. (1954) 'Les vertus morales acquises son-telles de vraies vertues?',
Recherches de Théologie ancienne et médiévale 21:114-22; Les Quodlibets, in Les Philosophes
Belges, l'Institut Supérieur de Philosophie de Louvain.
SECONDARY LITERATURE: Arway, R. J. (1962). Ά Half Century of Research on
Godfrey of Fontaines', The New Scholasticism 36:192-218; Neumann, B. (1958). Der
Mensch und die himmlische Seligkeit nach der Lehre Gottfrieds von Fontaines, Lahn-Verlag;
Tihon, P. (1966). Foi et théologie sehn Godefroid de Fontaines, Desclée de Brouwer; Wippel,
J. F. (1967). 'Godfrey of Fontaines', New Catholic Encyclopedia 6:577-8; (1980). The
Metaphysical Thought of Godfrey of Fontaines; A Study in Late Thirteenth-Century Philosophy,
Catholic University of America Press.
G R A T I A N O F B O L O G N A (Franciscus Gratianus)
BIOGRAPHICAL DATA: fl. 1140. Entered Camaldulian monastery in early life and wrote his
compilation of canon law in the monastery of San Felice in Bologna. Generally considered
the founder of the science of canon law.
TEXTS: Concordia discordantium canonum (1879), ed. A. Friedberg (CorpusJuris Canonici, 1),
Tauchnitz.
G R E G O R Y O F R I M I N I (Gregorius Ariminensis)
BIOGRAPHICAL DATA: b. ca. 1300 at Rimini; d. 1358. Studied in Italy, at Paris (B.A. ca.
1323), and in England. Taught at Paris, Bologna, Padua, Perugia. Returned to Paris
1341-51. Taught at Rimini 1351-6. Elected prior general of the Augustinians 1357.
TEXTS: Super primum et secundum Sententiarum (1522), Venice (Reprinted, The Franciscan
Institute 1955); Lectura super primum et secundum Sententiarum (1979-), ed. D. Trapp et al.
(Spätmittelaltcr und Reformation, Texte und Untersuchungen, 9), De Gruyter.
SECONDARY LITERATURE: Elie, Hubert (1937)- Le Complexe significabile,]. Vrin; Gâl,
Gedeon (1967). 'Gregory of Rimini', in New Catholic Encyclopedia 6:797; Leff, Gordon
(1961). Gregory of Rimini: Tradition and Innovation in Fourteenth Century Thought,
Manchester University Press; Nuchelmans, Gabriel (1973). Theories of the Proposition:
Ancient and Medieval Conceptions ofthe Bearers of Truth and Falsity (North-Holland Linguistic
Series, 8), North-Holland; Trapp, Damasus (1958). 'Gregory of Rimini: Manuscripts,
Editions, and Additions' Augustiniana 8:425-43; Trapp, Damasus (1962). 'New
Approaches to Gregory of Rimini', Augustinianum 2:115-30.
G U A R I N O V E R O N E S E (Guarino da Verona, Guarinus Veronensis)

BIOGRAPHICAL DATA: b. 1373; d. 1460. Studied in Padua under Giovanidi Conversino, in
Constantinople under Manuel Chrysoloras (1403-8). Taught in Florence (1410-14),
Verona (1419-29), and Ferrara (1430-60).
TEXTS: Epistolario di Guarino Veronese (1915-19), ed. R. Sabbadini (Miscellanea di storia
veneta, 3rd ser., vols. 8,11, 14), (Reprinted Bottega d'Erasmo 1967); Regulae grammaticales.
MSS Milan, Biblioteca Trivulziana, Cod. 631; Oxford, Bodleian Library, Lat. misc. e. 123;
Venice, Biblioteca Marciana, Lat. XIII. 113 ( = 4042).
G U Y O F R I M I N I (Guido Vernani de Arimino)

BIOGRAPHICAL DATA: d. after 1344. Twice lector in Studium generale in Bologna
between 1310 and 1324.
TEXTS: De potestate summi pontifias (1746), Bologna; De reprobatione Monarchiae compositae a
Dante Alighiero (1746), Bologna; Expositio super décrétait: Unam Sanctam édita a domino
Bonifacio papa octavo (1934), ed. Martin Grabmann (Sitzungsberichte der Akademie der
Wissenschaften, Munich; Philosophisch-historische Abteilung, 2:144—57).
Biographies 863
S E C O N D A R Y L I T E R A T U R E : Grabmann, Martin (1934). 'Studien über den Einfluss der

aristotelischen Philosophie auf die mittelalterlichen Theorien über das Verhältnis von
Kirche und Staat', Sitzungsberichte der Akademie der Wissenschaften, Munich; Philosophisch-
historische Abteilung, 2:76-100; Kappeli, T. (1937-8). 'Der Dantegegner Guido
Vernani O. P. von Rimini', Quellen und Forschungen aus italienischen Archiven und
Bibliotheken 28:107-46.
HENRYBRACTON
TEXTS: Bracton on the laws and customs of England (1968-). S. E. T h o m e (tr.), Belknap
Press of Harvard University Press; De legibus et consuetudinibus Angliae, (1915-42), ed. G. Ε.
Woodbine, Yale University Press.
SECONDARY LITERATURE: Fesefeldt, W. (1962). Englische Staatstheorie des 13.
Jahrhunderts: Henry de Bracton und sein Werk, Musterschmidt.
H E N R Y O F G H E N T (Henricus Gandavensis, Henricus de Gandavo)

BIOGRAPHICAL D A T A : b. ca. 1217; d. 1293. Taught at the University of Paris, 1276-92.
TEXTS: La 'Lectura ordinaria super sacrant scripturam' attribuée à Henri de Gand (1972), ed. R .
Macken, Editions Universitaires de Louvain; QuodUbeta (1518), Paris. (Reprinted
Bibliothèque S-J. de Louvain 1961) Les 'Quaestiones in librum De causis' attribuées à Henri de
Gand, (1974), ed.J. P. Zwaenepoel, Editions Universitaires de Louvain; Summa quaestionum
ordinariorum (1520), Paris (Reprinted The Franciscan Institute 1943); Henrici de Gandavo
Opera omnia (1979-), ed. Raymond Macken et al. E. J, Brill.
SECONDARY LITERATURE: Bettoni, Ε. (1954). // processo astrattivo nella concezione di Enrico
di Gand, Societi éditrice 'Vita e Pensiero'; Huet, F. (1838). Recherches historiques et critiques sur
la vie, les ouvrages et la doctrine de Henri de Gand, Leroux; Mauter, Armand (1948). 'Henry of
Ghent and the Unity of Man', Mediaeval Studies 10:1-20. Paulus.J. (1938). Henri de Gand:
Essai sur les tendances de sa métaphysique,]. Vrin.
H E N R Y OF H A R C L A Y
BIOGRAPHICAL D A T A : b. 1270; d. 1317. Master of arts by 1296. Ordained priest in 1297.
Studied theology at Paris while Duns Scotus was there. Bachelor of theology by 1308.
Master of theology in 1312. Elected Chancellor of Oxford in 1312.
TEXTS: Gâl, Gedeon (1971). 'Henricus de Harclay: Quaestio de Significato Conceptus
Universalis', Franciscan Studies 31:178-234; Maurer, Armand (1954). 'Henry of Harclay's
Question on the Univocity o f Being', Mediaeval Studies 16:1-18; Maurer, Armand (1957).
'Henry of Harclay's Questions on Immortality', Mediaeval Studies 19:79-107; Maurer,
Armand (1961). 'Henry o f Harclay's Questions on the Divine Ideas', Mediaeval Studies
23:163-93.
S E C O N D A R Y L I T E R A T U R E : Balic, C. (1959). 'Henricus de Harclay et loannes Duns
Scotus', Mélanges offerts à Etienne Gilson, Pontifical Institute of Mediaeval Studies; Kraus, J.
(1932). 'Die Universalienlehre des Oxforder Kanzlers Heinrich von Harclay und ihre
Mittelstellung zwischen skotistischem Realismus und ockhamistischem Nominalismus',
Divus Thomas (Freiburg) 10, II; Pelster, F. (1924). 'Heinrich von Harclay, Kanzler von
Oxford und seine Quästionen', Miscellanea Francesco Ehrle I (Studi e Testi, 37), pp. 307-56.
H E R V A E U S N A T A L I S (Harvey Nedellec, Noel)

B I O G R A P H I C A L D A T A : b. ca. 1250/60; d. 1323.
TEXTS: Quaestiones in quatuor libros sententiarum (1505), Venice; (1647) Paris (Reprinted
Gregg 1966); Quaestiones disputatae (1513), Venice; QuodUbeta (i486, 1513), Venice
(Reprinted Gregg 1966); Tractatus desecundis intentionibus (1489), Paris; (1513), Venice.
864 Biographies
S E C O N D A R Y L I T E R A T U R E : Allen, E. (i960). 'Hervaeus Natalis: An Early Thomist on the

Notion of Being', Mediaeval Studies 22:1—14; Glorieux, P. (1933). Répertoire des Maitresen
Théologie de Paris au XIIIe Siècle, (n. 64) J. Vrin; Stella, P. (1959). 'La prima critiea di
Hervaeus Natalis O.P., alla Noetica di Enrico di Gand: il 'De intellectu et specie' del
cosidetto'Dequattuormateriis', Salesianum zi : 125-70; Vollen, C. Ο. (1947). The Doctrine
of Hervaeus Natalis on Primitive Justice and Original Sin (Analecta Gregoriana, 42), Gregorian
University.
H U G O G R O T I U S (Huigh de Groot)
BIOGRAPHICAL D A T A : b. 1583; d. 1645. Entered the University of Delft in 1594. Doctor of
law at Orleans in 1598. Diplomat and lawyer. Studied with Arminius at Leiden. Involved in
controversy with mainstream Calvinists; imprisoned but fled to France in 1621. Became
Swedish Ambassador to France.
TEXTS: De iure belli ac pacts (1625), Paris. Reprinted with English translation by F. W.
Kelsey et. al. (1925) (Classics of International Law, 3), Oxford, Clarendon Press.
S E C O N D A R Y L I T E R A T U R E : Dumbauld, Edward (1969). The Life and Legal Writings of
Hugo Grotius, University of Oklahoma Press; Eysinga, Willem J. M. van (1945). Huigh de
Groot, een Schels, H. D. Tjeenk Willink; Friedmann, Wolfgang (1967). 'Grotius, Hugo' in
Paul Edwards (ed.), The Encyclopedia of Philosophy, vol. 3; Meulen, Jacob ter and P.J.J.
Diermanse (1950). Bibliographie des écrits imprimés de Hugo Grotius, Martinus Nijhoff.
I R N E R I U S O F B O L O G N A (Guarnerius)
B I O G R A P H I C A L D A T A : fl. 1118.
TEXTS: Summa codkis (1894), ed. H. Fitting, Guttentag; Questiones de iuris subtilitatibus
(1894), ed. H. Fitting, Guttentag.
S E C O N D A R Y L I T E R A T U R E : Besta, E. (1896). L'opéra d'Imerio: Contributo alia storia del
diritto italiano (2 vols.), Loescher; Fitting, H. (1888). Die Anfinge der Rechtsschule zu Bologna,
Guttentag; Kantorowicz, H. and W . W. Buckland (1938). Studies in the Glossators of the
Roman Law, Cambridge University Press.
J A M E S A L M A I N (Jaques Almain)
BIOGRAPHICAL D A T A : b. ca. 1480; d. 1515. Studied under John Major. Asked by the
Sorbonne in 1512 to compose its official response to Louis XII's request for a rebuttal of
papal claims to supremacy over church councils.
T E X T S : Libellus de auctoritate ecclesiae; Quaestio resumptiva; De dominio naturali, civili, et
ealesiastico; and Expositio de suprema potestate ecclesiastica et laica, circa quaestionum decisiones
Magistri Guillermi de Ockham super potestate summi pontifias, in Jean Gerson's Opera Omnia,
ed. L. E. DuPin (1706), Antwerp, Sumptibus Societatis, vol. 2; Opuscula (1518), Paris.
S E C O N D A R Y L I T E R A T U R E : LaBrosse, Olivier (1965). Le pape et le concile. Editions du
Cerf; Oakley, Francis (1965). 'Almain and Major: conciliar theory on the eve of the
Reformation', The American Historical Review 70:673-90; Skinner, Quentin (1978). The
Foundations of modern Political Thought, 2, Cambridge University Press.
JAMES OF D O U A I
BIOGRAPHICAL D A T A : fl. 1275. Master of arts in Paris.
T E X T S : Commentaries on the Prior and Posterior Analytks and De anima; Quaestiones in
Physicam. MSS Paris Bibl. nat. lat. 14721 and 14698.
S E C O N D A R Y L I T E R A T U R E : Grabmann, Martin (1956). 'Jakob von Douai, ein Aristoteles-
kommentator zur Zeit des heiligen Thomas von Aquin und des Siger von Brabant', in his
Mittelalterliches Geistesleben 3:158-79, Munich, Hueber.
Biographies 865
JAMES OF V I T E R B O
BIOGRAPHICAL D A T A : b. ca. 1255; d. 1308. Joined the order of Augustinian Hermits
around 1270 in Viterbo. Studied theology at Paris from 1278/9-1282/3. Became master of
theology at Paris ca. 1293 and succeeded Giles of R o m e in the Augustinian chair, function-
ing as regent master until 1296/7. In charge of Studium generale in Naples beginning 1300.
Bishop of Benevento 1302. Archbishop of Naples 1303.
TEXTS: De regimine chrisliano, in H.-X. Arquillière (1926), Le plus ancien traité de l'église:
Jacques de Viterbe, De Regimine Christiano (1301-2), Gabriel Beauchesne; QuodUbeta, in Jacobi
de Viterbio O.E.S.A. Disputatio Prima de Quolibet (1967),Jacobi de Viterbio O.E.S.A. Disputatio
Secunda de Quolibet (1969), Jacobi de Viterbio, O.E.S.A. Disputatio Tertia de Quolibet (1973),
Jacobi de Viterbio Disputatio Quarta de Quolibet (1975), ed. E. Ypma, Augustinus-Verlag.
SECONDARY LITERATURE: Gutierrez, David (1939). De Β. lacobi Viterbiensis O.E.S.A.
Vita, Operibus, et Doctrina Theologica, Analecta Augustiniana; Ypma, E. (1974). 'Recherches
sur la carrière scolaire et la bibliothèque de Jacques de Viterbe t 1308', Augustiniana 24:
247-82; Ypma, E. (1975)· 'Recherches sur la productivité littéraire de Jacques de Viterbe
jusqu' à 1300', Augustiniana 25:223-82.
JOHN AURIFABER
B I O G R A P H I C A L D A T A : fl. ca. 1330, Erfurt.
TEXTS: Commentary on the Tractatus of Peter of Spain, possible fragment in MS Krakow,
BJ 742, ff. 130-44.
Sophism against the modi significandi, in Jan Pinborg (1967), Die Entwicklung der Sprachtheorie
im Mittelalter (BGPM XL\l, 2), Aschendorff; Sophism on dimensions, MS Leipzig, Ü B , 1444,
ff. I49 v -l52 v ; Tractatus de demonstratio^, MSS Harburg I, 2, 80, 10, ff. 123-127; München
Clm. 331, ff- 45-7.
S E C O N D A R Y L I T E R A T U R E : Pinborgjan (1975). Ά N o t e o n Some Theoretical Concepts
of (Medieval] Logic and Grammar', Revue Internationale de Philosophie 113 :286-96.
J O H N B O D E (John Body, Bodi)

BIOGRAPHICAL D A T A : fl. mid-fourteenth century. Fellow of Merton College, Oxford,
possibly in 1334, 1338. Benedictine monk; D. Th. mentioned in 1357, 1358.
TEXTS: A est unum calidum, MSS Paris B. N . lat. 16134, ff. 7 3 r - 8 o v ; Venice, San Marco, lat.
VI. 30, ff. 113r— 133 r ; Venice, San Marco, lat. VI. 155. ff. 6 5 r - 8 2 v ; Vatican, lat. 4447,
ff. 229™-303 vb .
SECONDARY L I T E R A T U R E : Busard, H. L. L. (1965). 'Unendliche Reihen in A est unum
calidum', Archive for History of Exact Sciences 2:387-97; Duhem, Pierre (1906-13). Etudes sur
Léonard de Vinci, Hermann et Fils; Vol. 3, pp. 474-7; Duhem, Pierre (1913-59). Le Système
du Monde, Hermann et Fils; Vol. 7, pp. 648-50; Maier, Anneliese (1952). An der Grenze von
Scholastik und Naturwissenschaft (2nd ed.), Edizioni di Storia e Letteratura.
J O H N B U R I D A N (Jean Buridan, loannes Buridanus)

BIOGRAPHICAL D A T A : b. ca. 1295/1300; d. after 1358. Studied at Paris and obtained master
of arts there around 1320. Rector of the university in 1328 and again in 1340. Lecturer in the
faculty of arts at Paris. Taught Albert of Saxony, Marsilius of Inghen, Nicole Oresme.
TEXTS: Consequentiae, in Iohannis Buridani Tractatus de consequentiis (1976), ed. H. Hubien,
Publications universitaires de Louvain; De motibus animalium (1967), ed. F. Scott and
H. Shapiro, Isis 58:533-52; In Metaphysicam Aristotelis Quaestiones (1518), Paris (Reprinted
Minerva 1964); Quaestio de punctis (1961), ed. V. Zoubov, Mediaeval and Renaissance Studies
5:63-95; Quaestiones et decisiones physicales insignium virorum (1489), Paris (Reprinted
Minerva 1966); Quaestiones super libris quattuor de caelo et mundo (1942), ed. Ε. A. Moody, The
Mediaeval Academy of America; Quaestiones super decern libros Ethicorum {1513), Paris
866 Biographies
(Reprinted Minerva 1968); Quaestiones super octo Physkorum libros (1509), Paris (Reprinted
Minerva 1963); Sophismata, in Johannes Buridanus: Sophismata (1977), ed. T. K. Scott,
(Grammatica Speculativa, 1) Frommann-Holzboog. Also in Sophisms on Meaning and Truth
(1966), T. K. Scott (tr.) (Century Philosophy Sourcebooks), Appleton-Century-Crofts;
Summulae de dialectica (also called Summulae dialecticae, Summulae logicae, Compendium logicae,
Compendium totius logicae), Tractatus IV (De suppositionibus) (1957) in Maria Elena Reina
(ed.) 'Giovanni Buridano: Tractatus de suppositionibus,' Rivista crilica distoria dellafilosofia
12:175-208 and 323—52. Also Perutile compendium totius logicaeJoannis Buridani cum praeclar-
rissima solertissimi viriJoannis Dorp expositione (1499), Venice (Reprinted Minerva 1965).
SECONDARY LITERATURE: Faral, Edmond (1946). 'Jean Buridan: Notes sur les manu-
scrits, les éditions et le contenu de ses ouvrages', Archives d'histoire doctrinale et littéraire du
moyen âge 21:1-53 ; Faral, Edmond (1949). 'Jean Buridan: Maître es arts de l'Université de
Paris', in Histoire littéraire de la France, Imprimerie nationale; vol. 38; Ghisalberti, A. (1975).
Giovanni Buridano dalla metafisica allafisica (Publicazioni délia Universitâ Catolica del Sacro
Cuore), Vita e pensiero; Moody, E. A. (1975). 'Jean Buridan', in Studies in Medieval
Philosophy, Science, and Logic. Collected Papers IQ33-1Q60 (Publications of the Center for
Medieval and Renaissance Studies at UCLA, 7) University of California Press, pp. 441-53;
Pinborg, Jan, ed. (1976). The Logic ofJohn Buridan (Opuscula graecolatina, 9), Museum
Tusculanum.
JOHN CAPREOLUS
BIOGRAPHICAL DATA: b. I380; d. I444.
TEXTS: Defensiones theologiae divi Thomae Aquinatis (1900-8), ed. C. Paban and T. Pègues
(7 vols.), Cattier (Reprinted Minerva 1967).
SECONDARY LITERATURE: Degl'Innocenti, U. (1967). 'Capreolo, Giovanni' in Enciclo-
pedia Filosofica 1, Sansoni; pp. 1202-3; Grabmann, Martin (1944). 'Joannes Capreolus, der
"Princeps Thomistarum" und seine Stellung in der Geschichte der Thomistenschule', DIVMJ
Thomas (Freiburg) 22:85—109, 145-70.
JOHN DUMBLETON
BIOGRAPHICAL DATA: fl. mid-fourteenth century. Fellow of Merton College, Oxford
from 1338 through 1347/8.
TEXTS: Compendium sex conclusionum (Expositio capituli quarti Bradwardini De proportionibus).
MS Paris, B. N. Nouv. acq. lat. 625, ff. 70 v -7i v ; Summa logicae et philosophiae naturalis. MSS
Cambridge, Peterhouse 272, ff. 1-111; Cambridge, Gonville and Caius 499/268, ff 1-162;
Vatican, lat. 6750, ff. 1-202.
SECONDARY LITERATURE: Clagett, Marshall (1959). The Science of Mechanics m the
Middle Ages, University of Wisconsin Press; Maier, Anneliese (1968). Zwei Grundprobleme
der scholastischen Naturphilosophie (3rd ed.), Edizioni di Storia e Letteratura; Molland,
A. G. (1973). 'John of Dumbleton', Dictionary ofScientific Biography; Sylla, Edith (1970). The
Oxford Calculators and the Mathematics of Motion 1320-1350. Unpublished Ph.D. dissertation.
Harvard University; Sylla, Edith (1973). 'Medieval Concepts of the Latitude of Forms: the
Oxford Calculators', Archives d'histoire doctrinale et littéraire du moyen-âge 40:251-71;
Weisheipl, James A. (1956). Early 14th-century Physics and the Merton 'School' with special
reference to Dumbleton and Heytesbury, unpublished D.Phil, thesis, Oxford University;
Weisheipl, James A. (1959). 'The Place of John Dumbleton in the Merton School'. Lis 50:439-
54; Weisheipl, James A. (1969). 'Repertorium Mertonense', Mediaeval Studies 31:174-224.
JOHN DUNS SCOTUS

BIOGRAPHICAL DATA: b. ca. 1265; d. 1308. Ordained in St. Andrews, Northhampton in
1291. Studied at Oxford between 1288 and 1301. Lectured on Sentences at Oxford during
Biographies 867
that time, again at Paris in 1302-3, and possibly also at Cambridge in 1303-4. Returned to
Paris in 1304. Probably became master in 1305. Lectured as professor of theology at
Franciscan study house in Cologne from 1307 until his death.
TEXTS: loannis Duns Scott Doctoris Subtilis et Mariant Opera Omnia (1950-), ed. Carl Balic et
al., Typis Polyglottis Vaticanis; Joannis Duns Scoti Doctoris Subtilis Ordinis Minorum Opera
Omnia (1639), ed. Luke Wadding, Lyon. Republished by Vives, Paris (iSçi-s);John Duns
Scotus, A Treatise on God as First Principle (1966). A. B. Wolter (ed. and tr.) (Forum Books),
Franciscan Herald Press; Johannes Duns Scotus, Abhandlung über das erste Prinzip (1974).
W. Kluxen (ed. and tr.) (Texte zur Forschung, 20), Wissenschaftliche Buchgesellschaft;
JohannL· Duns Scoti Tractatus de Primo Principio (1941), ed. M. Müller (Bücher
Augustinischer und Franziskanischer Geistigkeit, Erste Reihe A 1), Herder; The De Primo
Principio of John Duns Scotus (1949), E. Roche (ed. and tr.), The Franciscan Institute.
SECONDARY LITERATURE: Balic, Carl (1065). The Life and Works of John Duns
Scotus', in John K. Ryan and Bernardine M. Bonansea (cds.) John Duns Scotus 1263—1963,
Catholic University of America; Gieben, S. (1965). 'Bibliographia Scotistica recentior
(iQSs-i96s)\Lautentianum6;Sdaekr,Od\il((i9SS)-Bibtiographiadevita,operibusetdoctrina
loannis Duns Scoti, Saec. XIX-XX, Orbis-Catholicus-Herder; Schaefer, Odulf (1967).
'Resenha abreviada da bibliografia escotista mais récente (1954-1966)', Revista Portuguesa de
Filosofia 23:338-63.
SIR JOHN FORTESCUE

BIOGRAPHICAL DATA: b. ca. 1394; d. ca. 1476.
TEXTS : De laudibus legum angliae (1942), ed. S. B. Chtimes, Cambridge University Press; De
natura legis naturae (1869), in Lord Claremont (ed.) Works, London.
JOHN GERSON
BIOGRAPHICAL DATA: d. 1429. Chancellor of the University of Paris in 1395, succeeding
Peter of Ailly, with whom he played an important part in the conciliar movement
culminating in the Council of Constance.
TEXTS: Oeuvres Complètes (1960-5), ed. P. Glorieux, Desclée; Opera omnia (1706), ed.
L. DuPin, Antwerp: Sumptibus Societatis, vol. 2.
SECONDARY LITERATURE: Combes, A. (1940). Jean Gerson, commentateur dionysien,
J. Vrin; Figgis, J. N. (1907). Studies of Political ThoughtfromGerson to Grotius, Cambridge
University Press; Morrall.John B. (i960). Gerson and the Great Schism, Manchester Univer-
sity Press; Pascoe, Louis B. (1973). Jean Gerson: Principles of Church Reform (Studies in
Medieval and Reformation Thought, 7), E.J. Brill; Schäfer, Carl (1953). Die Staatslehre des
Johannes Gtrson, Buchdruckerei Beyer und Hausknecht K.-G.
JOHN OFJANDUN
BIOGRAPHICAL DATA: b. ca. 1285/9; d. 1328. Master of arts in Paris in 1310. Fled to the
Emperor Ludwig of Bavaria with Marsilius of Padua in 1326.
TEXTS: Quaestiones in XII libros Metaphysicae et super libros Aristotelis De anima (1553,1583),
Venice (Reprinted Minerva 1966); Quaestiones super VIII libros Physkorum Aristotelis (1551),
Venice (Reprinted Minerva 1969).
SECONDARY LITERATURE: MacClintock, S. (1956). Perversity and Error, Indiana
University Press; Schmugge, L. (i966)._/ofcen«ei von Jandun (Pariser Historische Studien, 5),
Hiersemann.
J O H N OF L E G N A N O
BIOGRAPHICAL DATA: b. ca. 1320; d. 1382/83. A lawyer of Bologna. Wrote on theology,
law, politics, astrology; very active in Italian politics.
868 Biographies
TEXTS: Tractatus de bello (1917), ed. T. E. Holland, Oxford University Press.

SECONDARY LITERATURE: Ermini, G. (1933)· I trattati délia guerra e della pace di Giovanni
da Legnano, Imola.
J O H N MAJ O R (John Mair)

BIOGRAPHICAL DATA: b. 1467/69^. 1550. Studied at Cambridge and then at Paris, where
he received his master's degree in 1495. Taught at the University of Glasgow and at St
Andrews between 1518 and 1526, after which he returned to Paris, 1526-31. An Ockhamist
in theology and politics. Taught Pierre Crockaert, George Buchanan, John Calvin, and
John Knox.
TEX TS : De statu et potestate ecclesiae and De Potestate Papae in Temporalibus, in Gerson's Opera
Omnia (1706), ed. L. DuPin, Antwerp, Sumptibus Societatis, vol. 2; Le traité 'De l'infini' dt
Jean Mair (1938), ed. H. Elie, J. Vrin.
SECONDARY LITERATURE: Burns, J. H. (i9S4). 'New Light on John Major', The Innés
Review 5:83-100; Oakley, Francis (1962). 'On the road from Constance to 1688: the
political thought of John Major and George Buchanan', The Journal of British Studies
2:1-31; Oakley, Francis (1965). 'Almain and Major: conciliar theory on the eve of the
Reformation*, The American Historical Review 70:673-90; Renaudet, Augustin (1953).
Préréforme et humanisme à Paris pendant les premières guerres d'Italie (1494-1317) (2nd edn),
Librairie d'Argences; Skinner, Quentin (1978). The Foundations ofModem PoUtical Thought,
2, Cambridge University Press.
J O H N OF M I R E C O U R T
BIOGRAPHICAL DATA: fl. Ι345.
TEXTS: 'Questioni inédite di Giovanni di Mirecourt sulla conoscenza (Sent. I, q. 2-6)'
(1958), A. Franzinelli (ed.), Rivista critica di storia della filosofia 13:319-40, 415-49;
'Questioni inédite tratte dal I libro del Commento aile Sentenze di Giovanni di Mirecourt
(q. 13-16)* (1978), M. Parodi (ed.), Medioevo 3, 237-84.
SECONDARY LITERATURE: Birkenmaier, Α. (1922). Ein Rechtfertigungsschreiben Johannes
von Mirecourt (BGPM XX, 5), Aschendorff; Parodi, M. (1978). 'Recenti studi su Giovanni
di Mirecourt', Rivista critica di storia della filosofia 33:297-307; Stegmueller, F. (1933). 'Die
zwei Apologien des Jean de Mirecourt', Recherches de théologie ancienne et moderne 5:40-78
and 192-204.
J O H N LE P A G E (Johannes Pagus)
BIOGRAPHICAL DATA: fl. I 2 3 0
TEXTS: Rationes super Praedicamenta Aristotelis (1934), excerpted in E. Franceschini,
'Giovanni Pago: Le sue "Rationes super Predicamenta Aristotelis" e la loro posizione nel
movimento aristotelico del secolo ΧΙΙΓ, Sophia 2:172-82; 329-50; 476-86.
SECONDARY LITERATURE: Gründe!,J. (1958). 'Die Sentenzenglosse des Johannes Pagus
(ca. 1243-45) in PaduaBibl. Ant. 139', Münchener theologische Zeitschrift 9:171-85; Pelster,
F. (1930). 'Literaturgeschichtliches zur Pariser theologischen Schule aus den Jahren 1230 bis
1256', Scholastik 5:46—78; Braakhuis, H. A. G. (1979). De 130e eeuwse traetaten over syncate-
gorematische lernten (Deel I), Krips Repro Meppel.
J O H N O F P A R I S (John Quidort, Johannes Dormiens, Surdus, Soardus)

BIOGRAPHICAL DATA: d. 1306. Dominican. Lectured on the Sentences at Paris, 1293-4.
Master of theology at Paris, 1304. Involved in both political and theological controversy:
supported Philip the Fair's call for a general council against Boniface VIII. Object ol
Biographies 869
theological censure by an episcopal board under Giles of Rome and suspended from
teaching in 1305.
TEXTS: Correctorium corruptorii 'Circa' (1941), ed. J. P. Muller (Studia Anselmiana, 12-13),
Herder; De regia potestate et papali, in F. Bleienstein (1968), Johannes Quidort von Paris über
königliche und päpstliche Gewalt (De Regia Potestate et Papali) (Frankfurter Studien zur
Wissenschaft von der Politik), E. Klett. Translated in On Royal and Papal Power (1971), J. A.
Watt (tr.), Rome Pontifical Institute of Mediaeval Studies; and John of Paris on Royal and
Papal Power (1974), Arthur Monahan (tr.), Columbia University Press.
SECONDARY LITERATURE: Grabmann, Martin (1922).'Studien zu Johannes Quidort von
Paris O. Pr.', Sitzungsberichte der bayerischen Akademie der Wissenschaften, (Philosophisch-
Historische Klasse), 3; Kaeppeli, Thomas (1975). Scriptores Ordinis Praedicatorum medii aevi,
Istituto Storico Domenicano Roma; II, 517-24; Leclercq, Jean (1942). Jean de Paris et
l'ecclésiologie du XIII' siècle, J. Vrin; Tierney, Brian (1955). Foiim/nrions of the Conciliar
Theory, Cambridge University Press.
J O H N P E C K H A M (John Peckam.John Pecham)

BIOGRAPHICAL DATA: b. ca. 1225; d. 1292. Archbishop of Canterbury, 1278-92.
TEXTS: Perspectiva communis (1970), ed. and tr. D. C. Lindberg inJohn Pecham and the Science
of Optics (University of Wisconsin Publications in Medieval Science, 14), University of
Wisconsin Press; Quaestiones tractantes de anima (1918), ed. H. Spettmann (BGPM XIX,
5-6), Münster, Aschendorff; Quodlibet Romanum (1938), ed. F. Delorme (Spicilegium
Pontificii Athenaei Antoniani, 1), Pontifkium Athenaeum Antonianum; Summa de esse et
essentia (1928), ed. F. Delorme, Studi francescani 14:1-18; Tractatus de anima (1949), ed. G.
Melani, (Biblioteca di studi francescani, 1), Edizioni Studi Franciscani; Tractatus de perspectiva
(1972), ed. D. C. Lindberg (Franciscan Institute Text Series, 16), The Franciscan Institute.
SECONDARY LITERATURE: Callebaut, Α. (1925). 'Jean Peckham Ο. F. M. et
l'augustinisme', Archivum Franciscanum Historicum 18:441—72; Douie, D. L. (1952).
Archbishop Pecham, Clarendon Press; Spettmann, H. (1919). Die Psychologie des Johannes
Pecham (BGPM XX, 6), Aschendorff.
JOHN OF READING
BIOGRAPHICAL DATA: b. ca. 1285; d. 1346. A Franciscan and one of the earliest &
staunchest supporters of Duns Scotus. Lectured on the Sentences in Oxford before 1320.
Became regent master in theology, 1320/21. Strongly opposed Ockham's nominalism.
TEXTS: Super Sententias I, 4.3, q. 3, in Gedeon Gâl (1969), 'QuaestioJoannis de Reading de
Necessitate Specierum Intelligibilium Defensio Doctrinae Scoti', Franciscan Studies
7:66-156.
SECONDARY LITERATURE: Longpré, P. (1924). 'Jean dc Reading et le B.Jean Duns Scot',
La France Franciscaine 7:99-109.
J O H N OF RIPA
BIOGRAPHICAL DATA: fl. ca. 1355, in Paris. His work influenced a number of later
theologians, including Peter of Candia.
TEXTS: Conclusiones (1957), ed. A. Combes (Etudes de philosophie médiévale, 44), Vrin;
Determinationes (1957), ed. A. Combes (Textes philosophiques du moyen âge, 4), Vrin;
Lectura super 'Primum Sententiarum' (1961-70), ed. A. Combes and F. Ruello (Textes
philosophiques du moyen âge, 8,16), Vrin; Quaestio degradu supremo (1964), ed. A. Combes
and P. Vignaux (Textes philosophiques du moyen âge, 12), Vrin.
SECONDARY LITERATURE: Combes, André (1956). 'Présentation de Jean de Ripa',
Archives d'histoire doctrinale et littéraire du moyen âge 22:145-242.
870 Biographies
J O H N O F ST T H O M A S (Jean Poinsot)
BIOGRAPHICAL DATA: b. 1589; d. 1644. Entered Dominican order 1612. Studied at
Coimbra and Louvain. Taught at Piacenza, Madrid, and Alcalâ. Advisor and confessor of
Philip IV of Spain.
TEXTS: Cursus Philosophkus Thomisticus (1930—7), ed. B. Reiser, Marietti; Cursus
Théologiens (1930-53), ed. Benedictines of Solesmes, Desclée; Outlines of Formal Logic
(i955). Francis C. Wade (tr.), Marquette University Press; The Material Logic of John of St
Thomas (1955). Y. R. Simon, J. J. Glanville, G. D. Hollenhorst (trs.), University of Chicago
Press.
J O H N OF SALISBURY
BIOGRAPHICAL DATA: b. ca. 1120; d. 1180. Studied at Paris under Abelard and at Chartres
under Gilbert de la Porrée. Recommended by St Bernard, he served for several years as
secretary to Theobald, Archbishop of Canterbury. Close friend to Thomas Becket. Bishop
of Chartres in 1176.
TEXTS : Entheticus de Dogmate Philosophorum (1843), ed. Christian Petersen, Meisser; Historia
Pontificalis (1927), ed. R. Lane Poole, Clarendon Press; Metalogkon (1929), ed. C. C.J.
Webb, Clarendon Press; Policraticus (1909), ed. C. C.J. Webb, Clarendon Press (Reprinted
Minerva 1965.), Portions translated in The Statesman's Book of John ofSalisbury (1927), John
Dickinson (tr.), Knopf, and in Frivolties of Courtiers and Footsteps of Philosophers (1938),
Joseph B. Pike (tr.), Oxford University Press.
SECONDARY LITERATURE: Eberenz, James H. (1969). The Concept of Sovereignly in Four
Medieval Philosophers: John ofSalisbury, St Thomas Aquinas, Egidius Colonna and Marsilius of
Padua, University Microfilms; Liebeschütz, H. (1950). Humanism in the Life and Writings of
John of Salisbury, Warburg Institute; Liebeschütz, H. (1968). 'Chartres and Bologna:
Naturbegriff und Staatsidee bei Johannes von Salisbury*, Archiv für Kulturgeschichte
50:3-32; Rouse, Richard H. and Mary A. (1967). 'John ofSalisbury and the doctrine of
tyrannicide'. Speculum 42:693—709; Smalley, Beryl (1973). The Becket Conflict and the
Schools: A Study of Intellectuals in Politics, Rowman and Littlefield; Turk, Egbert (1977).
Nugae curialium: Le règne d'Henri II Plantagenet (it 43-1189) et l'éthique politique (Publications
du Centre de Recherche d'Histoire et Philologie de la IVe Section de l'École pratique des
Hautes Études, 5: Hautes Médiévales et Modernes, 28), Droz; Webb, C.C.J. (1932).John of
Salisbury, Methuen.
JOHN SETON
BIOGRAPHICAL DATA: b. ca. 1498; d. 1567· Student and Fellow of St John's College,
Cambridge. Chaplain to Bishop Gardiner. Disputed with Cranmer, Ridley, and Latimer at
Oxford. Imprisoned after accession of Elizabeth I. Fled to Rome.
TEXTS: Dialectica: annotationibus Petri Carteri, ut clarissimis, ita brevissimis, explicata (1572),
London.
J O H N WYCLIF
BIOGRAPHICAL DATA: b. ca. 1330; d. 1384. Fellow of Merton College, Oxford, in 1356-
Master of Balliol College in 1360; doctor of divinity in 1372. Involved in theological
controversies; forty-five articles from his works were proscribed at the Council of Con-
stance in 1415.
TEXTS: Select English Works ofJohn Wyclif (1869-71), ed. Thomas Arnold, Clarendon
Press; The English Works of Wyclif Hitherto Unprinted (1ΆΪ0), ed. F. D. Matthew, Clarendon
Press; Works (1882-1924) (Wyclif Society), London, Tübner (ReprintedJohnson Reprint
Corporation 1966).
Biographies 871
SECONDARY LITERATURE: Daly, Lowrie John (1962). The Political Theory of John Wyclif,
Chicago, Loyola University Press; Farr, William (1974). John Wyclif as Legal Reformer, E.J.
Brill; Leff, Gordon (1967). Heresy in the Later Middle Ages, Manchester University Press;
Robson, J. Α. (1961). Wyclifand the Oxford Schools (Cambridge Studies in Medieval Life and
Thought, Second Series, 8), Cambridge University Press; Tatnall, Edith Comfort (1965).
Church and State According to John Wyclyf, London, University Microfilms.
JUAN D E M A R I A N A
BIOGRAPHICAL DATA: b. i536;d. 1623/4.
TEXTS: De rege et regis institutione (1605), Mainz.
SECONDARY LITERATURE: Lewry, G. (i960). Constitutionalism and Statecraft during the
Golden Age of Spain. A Study of the Political Philosophy ofJuan de Mariana, S J. (Travaux
d'humanisme et renaissance, 36), Droz.
LAMBERT O F A U X E R R E
BIOGRAPHICAL DATA: fl. 1250, in Paris. Dominican. See Alessio's 'Introduzione' in
Lambert of Auxerre 1971.
TEXTS: Logica, or Summa Lambert! (De proposittonibus, De praedicabilibus, De praedkamentis.
De postpraedkamentis. De sillogismo. De locis. De fallaciis, De suppositionibus et de signifi-
cationibus) edited in Lambert of Auxerre 1971, La Nuova Italia Editrice.
L O R E N Z O V A L L A (Laurentius Vallensis)
BIOGRAPHICAL DATA: b. 1407; d. 1457. Studied under Giovanni Aurispa and Ranuccio
da Castiglion Fiorentino. Taught in Pavia, 1430-3, and in Milan, 1433-4. Entered the
service of Alfonso of Aragon, King of Naples, in 1435. Worked for Nicholas V in the papal
curia beginning in 1448.
TEXTS: Opera (1543), Basle (Reprinted Bottega d'Erasmo 1962); Opuscula quaedam (1503),
Venice (Reprinted Bottega d'Erasmo 1962).
SECONDARY LITERATURE: Camporeale, Salvatore I. (1972). Lorenzo Valla: umanesimo e
teologia, Istituto Nazionale di Studi sui Rinascimento; Fois, Mario (1969). II pensiero cristiano
di Lorenzo Valla nel quadro storico-culturale del suo ambiente (Analecta Gregoriana, 174),
Libreria Editrice dell' Università Gregoriana; Garin, E. (1952). Prosatori Latini del
Quattrocento (La letteratura italiana: storia e testi, 13), Ricciardi; Gerl, Hanna-Barbara
(1974). Rhetorik als Philosophie: Lorenzo Valla (Humanistische Bibliothek, 13), Fink Verlag.
LUIS D E M O L I N A
BIOGRAPHICAL DATA: b. I535; d. Ι6ΟΟ.
TEXTS: De iustitia et iure (1876), P. Lethielleux.
SECONDARY LITERATURE: Stegmüller, F. (1935). Geschichte des Molinismus (BGPM
XXXII), Aschendorff.
MANEGOLD OF LAUTENBACH
BIOGRAPHICAL DATA: d. II03.
TEXTS: Liber ad Gebehardum (1891) (Monumenta Germaniae historica, I), Impensis
Bibliopolii Hahniani.
M A R S I L I U S O F I N G H E N (Marsilius of Inghem, Marsilius de Novimagio)

BIOGRAPHICAL DATA: b. ca. 1330; d. 1396. Probably studied at Paris under John Buridan.
Became master of arts at Paris in 1362. Served twice as rector of the University of Paris (i 367
and 1371). Went to the University of Heidelberg in 1382, where he became the first rector
(1386-92).
872 Biographies
TEXTS: Abbreviation's Libri Physkorum (1521), Venice; De Generatione (1518), Venice; Parva
Logicalia (forthcoming), ed. E. P. Bos; Quaestiones super libros Priorum Analytkorum (1516),
Venice. (Reprinted Minerva 1968); Quaestiones super quattuor libros Sententiarum (1501),
Strassburg. (Reprinted Minerva 1966).
SECONDARY L I T E R A T U R E : Clagett, Marshall (1959). The Science of Mechanics in the
Middle Ages, University o f Wisconsin Press; Maier, Anneliese (1964, 1967). Ausgehendes
Mittelalter I and II, Edizioni Storia e Letteratura; Möhler, W. (1949). Die Trinitätslehre da
Marsilius von Inghen: ein Beitrag zur Geschichte der Theologie des Spätmittelalters, Limburg;
Nardi, B. (1967). 'Marsilio di Inghen', in Enciclopediafilosofica4, Sansoni; Ritter, G. (1921).
Studien zur Spätscholastik: I. Marsilius von Inghen und die okkamistische Schule in Deutschland,
Sitzungsberichte der Heidelberger Akademie der Wissenschaften, 4. Abhandlung, Winter.
M A R S I L I U S O F P A D U A (Menandrinus of Padua, Marsiglio dei Mainardini di

Padova)
BIOGRAPHICAL D A T A : b. 1275/80; d. 1342/3. Rector o f the University o f Paris in 1313,
where he lectured on natural philosophy and engaged in medical research. Completed the
Defensor Paris in 1324. When his authorship o f the work became known in 1326, he fled
with John of Jandun to the court o f Ludwig of Bavaria. Remained under Ludwig's
protection until his death.
T E X T S : Defensor minor (1922), ed. C. K. Brampton, Cornish Brothers; Defensor pacts (1928),
ed. C. S. Previté-Orton, Cambridge University Press. Also (1932-3), ed. R. Scholz,
Hahnsche Buchhandlung. Translated A. Gewirth (1956), Marsilius ofPadua: The Defender of
Peace (The Records of Civilization, Sources and Studies, 46), Columbia University Press;
De translation imperii and De iurisdictione imperatoris in causis matrimonialibus (i960), in
M. Goldast (ed.) Monarchia Sancti Romani Imperii, Akademische Druck und Verlagsanstalt;
Oeuvres mineures (Defensor Minor, De translatione imperii) (1979), ed. Colette Jeudy and
Jeannine Quillet, Editions du Centre National de la Recherche Scientifique.
S E C O N D A R Y L I T E R A T U R E : D'Entrèves, A. Passerin (1939). The Medieval Contribution to
Political Thought: Thomas Aquinas, Marsilius of Padua, Richard Hooker, Oxford University
Press; Eberenz, James H. (1969). The Concept of Sovereignty in Four Medieval Philosophers:
John ofSalisbury, St. Thomas Aquinas, Egidius Colonna, and Marsilius of Padua, University
Microfilms; Quillet, J. (1970). La Philosophie politique de Marsile de Padoue (L'Église et l'état
au Moyen Age, H ) , J . Vrin; Stout, Harry S. (1974). 'Marsilius of Padua and the Henrician
Reformation', Church History 43:308-18.
M A R T I N O F D A C I A (Martinus Dacus)
BIOGRAPHICAL D A T A : d. 1304. Professor at the arts faculty at Paris in the 1270s. Master of
theology at Paris around 1285. Chancellor of the Danish King Erik VI around 1288-1300.
Afterwards returned to Paris.
TEXTS: Opera (1961), ed. H. Roos (Corpus Philosophorum Danicorum Medii Aevi, 2),
Gad.
S E C O N D A R Y L I T E R A T U R E : Roos, H. (1952). 'Die Modi significandi des Martinus de
Dacia', (BGPM XXXVII, 2), Aschendorff.
M A R T I N S M I G L E C K I (Smiglesius)
BIOGRAPHICAL D A T A : b. 1564; d. 1618. Joined the Society of Jesus in 1581. Professor of
Philosophy and Theology at Vilna. Governor o f Colleges of Pubtusk, Posen, and Kalisch.
TEXTS: Logica (1618), Ingolstad; (1634, 1638) Oxford.
Biographies 873
MATTHEW OF A Q U A S P A R T A
BIOGRAPHICAL D A T A : b. ca. 1237; d. 1302.
TEXTS: Quaestiones De Anima VI (1958), ed. A.-J. Gondras, Archives d'histoire doctrinale et
littéraire du moyen âge 24:203-52; Quaestiones disputatae De Anima XIII (1061), ed. A.-J.
Gondras, Etudes de philosophie médiévale 50; Quaestiones disputatae de gratia, cum introduetione
critica (1935), ed. V. Doucet, Collegium s. Bonaventurae; Qu. disp. de anima separata, de
anima beata, de ieiunio, et de legibus (1959), Collegium s. Bonaventurae; Qu. disp. defideet de
cognitione (1957), Collegium s. Bonaventurae; Qu. disp. de productione rerum et de Providentia
(1956), ed. G. Gâl, Collegium s. Bonaventurae.
SECONDARY LITERATURE: Beha, H. M. (1960-1). 'Matthew of Acquasparta's Theory of
Cognition', Franciscan Studies 20:161-204; 21:1-79; Prezioso, F. (1950). 'L'attività del
soggetto pensante nella gnoseologia di Matteo d'Acquasparta e di Ruggiero Marston',
Antonianum 25:259-326.
MICHAEL OF MASSA
BIOGRAPHICAL DATA: fl. ca. 1325. Augustinian friar, still almost entirely unknown. The
second book of his Commentary on the Sentences contains many lengthy additiones dealing
with topics in natural philosophy.
TEXTS: Commentary on the Sentences, MS Vat. lat. 1087.
SECONDARY LITERATURE: Trapp, Damasus (1956). 'Augustinian Theology of the
Fourteenth Century', Augustiniana 6:146-274, esp. 163-75; Trapp, Damasus (1965).
'Notes on some Manuscripts of the Augustinian Michael of Massa (t 1337)', Augustianum 5 :
S8-133.
N I C H O L A S O F A U T R E C O U R T (Nicolaus de Ultricuria, Autricuria)

BIOGRAPHICAL DATA: b. ca. 1300; d. after 1350.
TEXTS : Exigit ordo executionis ( Tractatus universalis magistri Nicholai de Ultricuria ad videndum
an sermones Peripateticorum fuerint demonstrati) (1939), ed. J. R. O'Donnell, Mediaeval Studies
ι : 179-280; Letters (1908), ed. J. Lappe, 'Nikolaus von Autrecourt' (BGPM VI, 2),
Aschendorff. Translated E. A. Moody, in H. Shapiro, ed. (1964), Medieval Philosophy,
Modern Library; The Universal Treatise ofNicholas ofAutrecourt (1971), tr. L. A. Kennedy et
al. (Medieval Texts in Translation, 20), Marquette University Press.
SECONDARY LITERATURE: DalPra, Mario (i95i).Mcp'<ji/iv4Hrreccwri, Fratelli Bocca; Dal
Pra, Mario (1952). 'La fondazione del empirismo e le sue aporie nel pensiero di Nicola di
AatK-court\ Rivistacriticadistoriadellafilosofias : 389-402; Moody,E. A. (1947). 'Ockham,
Buridan, and Nicholas of Autrecourt', Franciscan Studies 7:113-46; Scott, T. Κ. (1971).
'Nicholas of Autrecourt, Buridan, and Ockhamism', Journal of the History of Philosophy
9:15-41; Weinberg, Julius (1948). Nicholas of Autrecourt, Princeton University Press.
NICHOLAS B O N E T
BIOGRAPHICAL DATA: d. 1343. Taught in Paris. Known to have followed Gerard of Odo
in many of his contentions concerning indivisibles and continua.
TEXTS: Praedkamenta; Philosophia naturalis; Metaphysica; Theologia naturalis (1505), Venice.
SECONDARY LITERATURE: Barcelone, Martin de (1925). 'Nicolas Bonet (t 1343),
Tourangeau, Doctor Proficuus O. M.', Étudesfianchcaines 37:638-57; O'Briain; F. (1937).
'Nicolas Bonet' in Dictionnaire d'histoire et géographie ecclésiastiques 9:849-52.
NICHOLAS OF C U S A
BIOGRAPHICAL D A T A : b. 1401; d. 1464.
TEXTS : Opera omnia ( 193 2-), (Heidelberger Akademie), Meiner; Nicholas ofCusa on Learned
874 Biographies
Ignorance: A Translation and Appraisal of De docta ignorantia (1981), tr. Jasper Hopkins, A.J.
Banning; Nicholas of Cusa"s Debate withJohn Wenck: A Translation and Appraisal ofDe ignota
litteratura and Apologia doctae ignorantiae, tr. Jasper Hopkins, A. J. Banning.
S E C O N D A R Y L I T E R A T U R E : Danzer, R. (1964). 'Cusanus-Bibliographie 1961-64 und
Nachträge', Mitteilungen und Forschungsbeiträge der Cusanus-GeseUschafi 3:223-37;
Gandillac, M. de (1941). La Philosophie de Nicolas de Cues, Éditions Montaigne; Hopkins,
Jasper (1978). A Concise Introduction to the Philosophy of Nicholas of Cusa, University of
Minnesota Press; Kleinen, H. and R. Danzer (1961). 'Cusanus-Bibliographie 1920-61 ',
Mitteilungen und Forschungsbeiträge der Cusanus-Gesellschaft 1:95—126; Schnarr, H. (1973).
Modi essendi. Interpretationen zu den Schriften De docta ignorantia, De coniecturis, und De
venatione sapientiae von Nikolaus von Kues (Buchreihe der Cusanus-Gesellschaft, 5),
Aschendorff; Sigmund, P. E. (1963). Nicholas of Cusa and Medieval Political Thought,
Harvard University Press; Traut, W. (1967). 'Cusanus-Bibliographie 1964-67 und
Nachträge', Mitteilungen und Forschungsbeiträge der Cusanus-Geselbchaft 6:178-202;
Watanabe, M. (1063). The Political Ideas of Nicholas of Cusa (Travaux d'Humanisme et
Renaissance, 58), Droz; Zellinger, H. (i960). Cusanus-Konkordanz, Hueber.
N I C H O L A S OF P A R I S
BIOGRAPHICAL DATA: fl. 1250.
TEXTS: Syncategoremata (1979), ed. H. A. G. Braakhuis in his De I3de eeuwse tractaten ovtr
syncategorematische termen, Krips Repro Meppel; Summe Metenses (1967), in L. M. De Rijk
1962-7, II (2).
SECONDARY L I T E R A T U R E : Grabmann, Martin (1926). 'Die logische Schriften des
Nikolaus von Paris und ihre Stellung in der aristotelischen Bewegung des XIII.
Jahrhunderts', in his Mittelalterliches Geistesleben, Hueber.
NICOLE ORESME
BIOGRAPHICAL D A T A : fl. at Paris 1345-60; d. 1382. Wrote Quaestiones on Aristotle's
Physics, De caelo. De generatione et corruptione, De anima. De sensu, and Meteorologica; a group
of works (in French as well as Latin) opposing astrology; and four important treatises in
natural philosophy. Towards the end of his career he translated Aristotelian and other
treatises into French, often with commentary.
TEXTS: Ad pauca respicientes; De proportionibus proportionum (1966), ed. and tr. Edward
Grant in Nicole Oresme: De proportionibus and Ad pauca respicientes (University of Wisconsin
Publications in Medieval Science, 9), University of Wisconsin Press; Le livre du ciel et du
monde (1968), ed. and tr. A. D . Menut and A. J. Denomy in Nicole Oresme: Le livre du ciel et du
monde (University o f Wisconsin Publications in Medieval Science, 11), University of
Wisconsin Press; Quaestiones in libros De anima (1980), ed. Peter Marshall, with introduction
and commentary, unpublished Ph.D. dissertation, Cornell University; Questiones super
libros De celo (1965), ed. Claudia Kren, unpublished Ph.D. dissertation. University of
Wisconsin; Tractatus de commensurabilitate vel incommensurabilitate motuum celi (1971), ed. and
tr. Edward Grant in Nicole Oresme and the Kinematics of Circular Motion (University of
Wisconsin Publications in Medieval Science, 15), University of Wisconsin Press; Tractatus
de configurationibus qualitatum et motuum (1968), ed. and tr. Marshall Clagett in Nicole Oresme
and the Medieval Geometry of Qualities and Motions (University of Wisconsin Publications in
Medieval Science, 12), University of Wisconsin Press.
SECONDARY L I T E R A T U R E : [See Introductions and Commentaries in volumes listed
above.] Clagett, Marshall (1974). 'Nicole Oresme' in Dictionary of Scientific Biography.
Biographies 875
P A U L O F P E R G U L A (Paulus Pergulensis)
BIOGRAPHICAL DATA: d. 1451/5. Studied under Paul of Venice in Padua. Master of arts
ca. 1420. Doctor of theology by 1430. Taught in Venice from 1421 to his death.
TEXTS: Paul of Pergula: Logica and Tractatus de sensu composito et diviso (1961), ed.
M. A. Brown (Text Series, 13), Franciscan Institute.
SECONDARY LITERATURE: Boh, Ivan (1965). 'Paul of Pergula on suppositions and
consequences', Franciscan Studies 25:30-89.
P A U L O F V E N I C E (Paolo Nicoletti, Paulus Venetus)

BIOGRAPHICAL DATA: b. ca. 1369; d. 1429.
TEXTS: Logica magna (1499), Venice; Paul of Venice: Logica Magna Part I, Fascicule ι (i979)·
Norman Kretzmann (ed. and tr.), The British Academy and Oxford University Press; Paul
of Venice: Logica Magna Part II, Fascicule 6 (1978). Francesco del Punta and Marilyn M.
Adams (ed. and tr.), The British Academy and Oxford University Press; Paul of Venice:
Logica Magna (Tractatus de Suppositionibus) (1971). A. R. Perreiah (ed. and tr.), Franciscan
Institute.
SECONDARY LITERATURE: Lohr, C. H. (1973). Ά Note on the Manuscripts of Paulus
Venetus, Logica', Manuscripta 17:35-6; Perreiah, A. R. (1967). Ά Biographical
Introduction to Paul of Venice', Augustiniana 17:450-61.
P E T E R A B E L A R D (Petrus Abaelardus, Abailardus, Abaielardus, Adbaielardus)

BIOGRAPHICAL DATA: b. 1079; d. 1142. Studied with a number of teachers, notably
Roscelin and William of Champeaux. Disagreement with the latter led him to set up his
own school at Melun in 1104 and later at Corbeil. Because of ill health, he stayed in Brittany
1106-8. Returned to Paris and began teaching first at Melun and then at Mont-Sainte-
Geneviève. Studied theology under Anselm of Laon beginning in 1113. Returned again to
teach in Paris in 1116. There followed the famous love affair with Heloise, their secret
marriage, and the castration ordered by her uncle in 1118. Wrote the Theologia summi boni in
1118-20, which was condemned by the synod of Soissons in 1121. Abbot of St Gildas in
Brittany 1125-32. Teaching again in Paris, 1136-40. St Bernard of Clairvaux succeeded in
getting his theology condemned at the synod of Sens in 1140. Peter the Venerable of Cluny
mediated the dispute and arranged for Abelard to spend his last days in quiet study.
TEXTS: Dialectica (1970), ed. L. M. de Rijk (Wijsgerige teksten en studies, 1) (2nd ed.),
Van Gorcum; Dialogus inter Philosophum, Iudaeum et Christianum (1970), ed. R. Thomas,
Friedrich Frommann; Ethica or Scito te ipsum (1971), in D. E. Luscombe (ed.) Peter Abelard's
Ethics (Oxford Medieval Texts), Clarendon Press; Logica 'Nostrorum Petitioni' and Logica
'Ingredientibus', in B. Geyer, ed. (1919-27), Peter Abaelards Philosophische Schriften (BGPM
XXI, 1-4), Aschendorff; Theologia 'Summi Boni' (1939), ed. H. Ostlender (BGPM XXXV,
2-3), Aschendorff; Pelri Abaelardi Opera theologica I-II (1969), ed. E. M. Buytaert (Corpus
Christianorum, Continuatio Mediaevalis, 11-12), Brepols; Pietro Abeiardo Scrittifilosofici:
Editio super Porphyrium, Glossae in Categorias, Editio super Aristotelem De interpretatione, De
divisionibus, Super Topica glossae (1954), ed. Mario dal Pra, Fratelli Bocca; Theologia
'Scholarium' (or Introductio ad theologiam), in Victor Cousin (ed.) (1859), Petri Abaelardi
Opera, II, Durand; Twelfth Century Logic, Texts and Studies: Abaelardiana inedita (1958).
L. Minio-Paluello (ed.), Edizioni di storia e letteratura.
SECONDARY LITERATURE: FumagaUi, M. T. Beonio-Brocchieri (1970). The Logk of
Abelard (Synthese Historical Library, 1), Reidel; Jolivet, Jean (1969). Arts du Language et
Théologie chez Abelard (Études de Philosophie Médiévale, 57), J. Vrin; Luscombe, D. E.
(1969). The School of Peter Abelard: The Influence of Abelard's Thought in the Early Scholastic
Period, Cambridge University Press; Sikes, J. G. (1932). Peter Abatlard, Cambridge
876 Biographies
University Press; Tweedale, M. M. (1976). Abailard on Universals, North-Holland; Ulivi,

L. U. (1976). La pskologia di Abelardo e il 'Tractatus de intellectibus', Edizioni di Storia
e Letteratura.
P E T E R O F AILLY (Pierre d'Ailly, Petrus de Aliaco)

BIOGRAPHICAL DATA: b. 1350. d. 1420/1. Entered Collège de Navarre at University of
Paris as bursar (recipient of a scholarship) in 1364. Became bachelor of arts in 1367. Entered
faculty of theology in 1368. Became doctor of theology in 1381. Named Cardinal 1411.
Active in Council of Constance (1414-18).
TEXTS: Conceptus et insolubilia (1495), Paris. Translated in Peter of Ailly (1980). Concepts and
Insolubles, tr. Paul Vincent Spade, Reidel; Destructiones modorum significandi (ca. 1490-5),
Lyons; Quaestiones super libros Sententiarum (1474), Brussels; (1490) Strassburg (Reprinted
Minerva 1968); Tractatus de arte obligandi (1489), Paris. (Of doubtful authenticity; perhaps by
Marsilius of Inghen.); Tractatus exponibilium (1494), Paris.
SECONDARY LITERATURE: Bottin, Francesco (1976). Le antinomie semantiche nella logica
medioevale (Publicazioni delPistituto di storia della filosofia e del centro per ricerche di
filosofia medioevale, nuova série, 23), Editrice Antenore; Cardin, Α. (1967). 'Ailly, Pierre
de', in Encklopediafilosofica 1, Sansoni; Elie, Hubert (1937). Le Complexe significabile,]. Vrin;
Gandillac, Maurice Patronnier de ( 193 3). 'De l'usage et de la valeur des arguments probables
dans les questions du cardinal Pierre d'Aily sur le "Livre des Sentences" ', Archives d'histoire
doctrinale et littéraire du moyen âge 8:43-91 ; Glorieux, Palémon (1965). 'L'oeuvre littéraire de
Pierre d'Ailly', Mélanges de science religieuse 22:61-78; Meiler, Bernhard (1954). Studien zur
Erkenntnislehre des Peter von Ailly (Freiburger theologischen Studien, 67), Herder; Oakley,
Francis (1964). The Political Thought of Pierre d'Ailly: The Voluntarist Tradition (Yale
Historical Publications, Miscellany 81), Yale University Press; Salembier, Louis (1931). Le
Cardinal Pierre d'Ailly, Georges Frères; Tschackert, Paul (1877). Peter von Ailly (Petrus de
Alliaco): Zur Geschichte des grossen abendländischen Schisma undder Reformconcilien von Pisa und
Constanz, Friedrich Andreas Perthes.
P E T E R A U R E O L I (Aureolus, Aureol, Oriole)

B I O G R A P H I C A L D A T A : b. ca. 1280; d. 1322.
TEXTS: Compendiosa expositio evangelis Joannis (1951), cd. Friedrich Stegmüller,
Franziskanische Studien 33:207—19; Compendium sensus lateralis sacrae scripturae (1896), ed.
P. Seebaeck, Collegium S. Bonaventurae; Scriptum super primum Sententiarum (1953), ed.
E. M. Buytaert, Franciscan Institute.
SECONDARY LITERATURE: Dreiling, R. (1913). 'Der Konzeptualismus in der Univer-
salienfrage des Franziskaner Erzbischofs Petrus Aureoli' (BGPM II, 6), Aschendorff;
Teetaert, A. (1934)· 'Pierre Auriol', in Dictionnaire de Théologie Catholique, vol. 12.
P E T E R O F A U V E R G N E (Petrus de Alvernia, Peter of Alverny)

BIOGRAPHICAL DATA: d. 1303. Probably rector of the university of Paris around 1275.
Master of theology by 1296. Became Bishop of Claremont in 1302.
TEXTS: Hocedez, E. (1932). 'Les Quaestiones in metaphysicam de Pierre d'Auvergne', Archives
de Philosophie9:515-70; 'Quaestiones in Metaphysicam Petri de Alvernia' (1955), ed. A. P.
Monahan in J. R. O'Donnell (ed.) Nine Mediaeval Thinkers (Studies and Texts, 1),
Pontifical Institute of Mediaeval Studies; The Commentary of Peter of Auvergne on Aristotle's
Politics (1967). G. M. Grech (ed.), Desclée; Ά New Ms. of the Questions on the Post. Anal.
Attributed to Petrus de Alvernia with the Transcription of some Questions Related to
Problems of Meaning' (1973). Jan Pinborg (ed.) CIMAGL 10:48-62.
Biographies 877
SECONDARY LITERATURE: Grech, G. M. (1964), 'Recent Bibliography on Peter of

Auvergne', Angelkum 41:446-9; Hocedez, E. (1930). 'La théologie de Pierre d'Auvergne',
Gregorianum 11:520-2; Hocedez, E. (1933). 'La vie et les oeuvres de Pierre d'Auvergne',
Gregorianum 14:3-36; Monahan, A. P. (1954). 'The Subject of Metaphysics for Peter of
AuveTgne', Mediaeval Studies 16:118-30.
PETER CEFFONS
BIOGRAPHICAL DATA: fl. at Paris, ca. 1350. A Cistercian who lectured on the Sentences at
Paris, 1348-49. His Commentary on the Sentences is an immense work showing the
influence of fourteenth-century English philosophy and theology.
TEXTS: Commentary on the Sentences, MS Troyes 62.
SECONDARY LITERATURE: Murdoch, John E. (1978). 'Subtilitates Anglicanae in
Fourteenth-Century Paris: John of Mirecourt and Peter Ceffons' in Machaut's World:
Science and Art in the Fourteenth Century, Annals of the New York Academy of Sciences
314:51-86; Trapp, Damasus (1957). 'Peter Ceffons of Clairvaux', Recherches de Théologie
ancienne et médiévale 24:101-54.
P E T E R HELIAS (Petrus Heliae)

BIOGRAPHICAL DATA: fl. ca. II40, Paris.
TEXTS: Summa super Priscianum, Books I-IU (1975), in Reilly, Petrus Helias' summa super
Priscianum I-IU, an edition and study, unpublished Ph.D. dissertation, University of
Toronto; Summa super Priscianum, Books XVII-XVIU (1978), i n j . H. Toison (ed.), 'The
Summa of Petrus Helias on Priscianus Minor', CIMAGL 27—28:1-210.
SECONDARY LITERATURE: Fredborg, Κ. M. (1973). 'The Dependence of Petrus Helias'
Summa super Priscianum on William of Conches' Close super Priscianum', CIMAGL
11:1-57; Fredborg, Κ. M. (1975). 'Petrus Helias on Rhetoric', CIMAGL 13:31-41; Hunt,
R. W. (1949-50). 'Studies on Priscian in the Twelfth Century', Mediaeval and Renaissance
Studies 1:194-231; 2:1-55. Rijk, L. M. de (1967), in De Rijk 1962-7, II(i), 227-34.
P E T E R J O H N OLIVI
TEXTS: Quaestiones in secundum librum Sententiarum (1922-6), ed. B. Jansen (3 vols.),
Collegium S. Bonaventurae.
SECONDARY LITERATURE: Bettoni, E. (1959)· Le dottrinefilosofichedi Pier di Giovanni
Olivi, Publicazione dell'Università del S. Cuore; Burr, David (1976). The Persecution ofPeter
Olivi (Transactions of the American Philosophical Society, n.s., 66, pt. 5), American
Philosophical Society; Gieben. S. (1968). 'Bibliographia Oliviana 1885-1967', Collectanea
Franciscana 38:167-95; Partee, C. (i960). 'Peter John Olivi: Historical and Doctrinal
Study', Franciscan Studies 20:215-60.
PETER L O M B A R D
BIOGRAPHICAL DATA: b. ca. 1095; d. 1160. Taught in the cathedral school of Paris
from 1140.
TEXTS: Libri IVSententiarum (1916), ed. A. Heysse, Collegium S. Bonaventurae; Sententiae
in IV libris distinctae (1971-), Collegium S. Bonaventurae.
SECONDARY LITERATURE: Delhaye, P. (1961). Pierre Lombard, sa vie, ses oeuvres, sa
morale, Institut d'études médiévales de Montréal.
PETER OF SPAIN (Petrus Hispanus Portugalensis, Petrus Juliani, Pope John XXI)
BIOGRAPHICAL DATA: Bom in Portugal; d. 1277. Studied arts at the University of Paris,
878 Biographies
then medicine perhaps in Montpellier or Salerno. Professor of medicine at the University of

Siena, 1245—9. Dean of Lisbon and Archdeacon of Braga in 1250. Became court physician
of Pope Gregory X in 1272. Archbishop of Braga in 1273 and then Cardinal Archbishop of
Tusculum. Elected Pope (John XXI) in 1276.
TEXTS: Exposicâo sobre os livros de Beato Dionisio Areopagita (1957), ed. M. Alonso, Instituto
de alta cultura, Universidade deLisboa; Obrasfilosoficas (1941-52), ed. M. Alonso (Consejo
superior de investigaciones cientificas. Madrid, Instituto filosofico Luis Vives);
Syncategoremata (1502), Venice. Also translated inj. P. Mullally, tr. (1964), Peter of Spain,
Tractatus Syncategorematum and selected anonymous treatises (Medieval Philosophical Texts in
Translation, 13), Marquette University Press; Tractatus called afterwards Summule logkales
(1972), ed. L. M. de Rijk (Wijsgerige teksten en studies, 22), Van Gorcum.
SECONDARY LITERATURE: Grabmann, Martin (1936). 'Handschriftliche Forschungen
und Funde zu den philosophischen Schriften des Petrus Hispanus, des späteren Papstes
Johannes ΧΧΓ (Sitzungsberichte der Bayerischen Akademie der Wissenschaften, Phil.-hist.
Abt., 9), Hueber; Rijk, L. M. de (1970). 'On the life of Peter of Spain, the author of the
Tractatus called afterwards Summule logkales'. Vivarium 8:123-54; Rijk, L. M. de (1972).
'Introduction' in De Rijk (1972).
P E T E R O F T A R A N T A S I A (Pope Innocentius V)
BIOGRAPHICAL DATA: b. ca. 1225; d. 1276.
T E X T S : In quattuor libros Sententiarum commentaria (1649-52), ed. T. Turco and G. B. De
Marinis, Toulouse (Reprinted The Gregg Press 1964).
S E C O N D A R Y L I T E R A T U R E : Stella, P. (1967). 'Pietrodi Tarantasia,'inEncidopediafilosofica
4, Sansoni.
PHILIP THE C H A N C E L L O R
BIOGRAPHICAL D A T A : d. 1236. Studied theology and probably Canon Law in Paris.
TEXTS : Quaestiones de anima (1937), in Ex Summa Philippi cancellarii Quaestiones de anima, ed.
L. W. Keeler, Aschendorff.
S E C O N D A R Y L I T E R A T U R E : Lottin, Ο. (1927). 'Le Créateur du traité de la syndérèse',
RevueNéo-Scholastique de Philosophie 29:197-220; Schneyer.J. B. (1963). Die Sittenkritik in
den Predigten Philipp des Kanzlers (BGPM X X X I X , 4), Aschendorff.
PHILIPP M E L A N C H T H O N
B I O G R A P H I C A L D A T A : b. 1497; d. 1560.
TEXTS: Opera quae supersunt omnia (1834—60), ed. C. G. Bretschneider (28 vols.),
Schwetzke.
S E C O N D A R Y L I T E R A T U R ε : Huschke, R. Β. (1968). Melanchthons Lehre vom Ordo politicus,
Gütersloh; Kisch, G. (1967). Melanchthons Reechts- und Soziallehre, De Gruyter.
PSEUDO-PETRUS
BIOGRAPHICAL D A T A : The Pseudo-Petrus is the author or authors of material printed in
the Cologne editions (1489 and 1496) o f the logical works o f Peter of Spain and either
mistakenly attributed to Peter (e.g., the Tractatus exponibilium) or presented in ways that
make the attribution confusing or doubtful (e.g., the Tractatus syncategorematum).
P T O L E M Y O F L U C C A (Tolomeo)
B I O G R A P H I C A L D A T A : d. 1326/7.
TEXTS : De regimine principum, in R . M. Spiazzi, ed. (1954), DiVi Thomae Aquinatis Opuscula
Philosophica, Marietti.
Biographies 879
RADULPHUS BRITO
BIOGRAPHICAL DATA: d. 1320. Professor in the arts faculty at the University of Paris
around 1296-1306. Lectured on the Sentences at Paris, 1308-9. Master of theology around
1311-12. Provisor of the Sorbonne, 1315-20.
TEXTS: 'Der Kommentar des Radulphus Britonis zu Buch III De Anima' (1974). W. Fauser
(ed.) (BGPM, neue Folge, 12), Aschendorff; "Radulphus Brito's Commentary on Boethius'
De differentiis topicis and the sophism "Omnis homo est omnis homo"' (1978), ed. N. J.
Green-Pedersen, CIMAGL 26:1-121; 'Radulphus Brito on Universals' (1980). ed. Jan
Pinborg, CIMAGL 35:56-142 Radulphus Brito's Quaestiones super Priscianum minorem
(1980), ed. H, W. Enders and Jan Pinborg (Grammatica Speculativa, 3), Frommann-
Holzboog; 'Radulphus Brito's Sophism on Second Intentions' (1975), Jan Pinborg (ed.),
Vivarium 13:119-52.
SECONDARY LITERATURE: Glorieux, Palemon (1966). Aux Origines de la Sorbonne,
J. Vrin; Pinborg, Jan (1974). 'Zum Begriff der Secunda Intentio: Radulphus Brito, Hervaeus
Natalis, und Petrus Aureoli in Diskussion', CIMAGL 13:49-59; Pinborg, Jan (1975). 'Die
Logik der Modistae', Studia Mediewistyczne 16:39-97.
RALPH STRODE
BIOGRAPHICAL DATA: Fellow of Merton College, 1359-60. Friend of Chaucer. May be
identical with a London lawyer of the same name who died in 1387, survived by his wife
Emma and his son Ralph. See Blackley 1967; Emden I957~9. 3:1807-8; Workman 1926,
1:125-9, 242-3, 2:412-4.
TEXTS: Logica (De arte logica, De principiis logkalibus. De suppositionibus, Consequentiae,
Obligationes, De insolubilibus). The Consequentiae and Obligationes, which were required texts
at several universities, were printed in several renaissance editions; see Hain 1948,
1S093-15100. MSS: Oxford, Bodleian, MS Canon. Misc. 219, ff. i3 r , -52 v b ; individual
treatises of the Logica are preserved in several other manuscripts. Alfonso Maierù is
preparing a partial edition of the Logica. The Consequentiae have been edited and translated
by W. K. Seaton in an unpublished Ph.D. dissertation, 1973, University of California at
Berkeley.
R A Y M O N D LULL
BIOGRAPHICAL DATA: b.cd. 1232; d. 1315/6. Received informal education as courtier. At
the age of 30, he experienced a religious conversion and joined the Franciscans. Considered
it his mission to convert Muslims. Studied Arabic; vigorously attacked Latin Averroism at
the University of Paris. Active in Mallorca, Montpellier, Paris, and Naples. Made three
missionary journeys to North Africa: to Tunis in 1293, to Algeria in 1307, and again to
Tunis in 1314-1 $. Some 280 works are attributed to him, of which 240 survive.
TEXTS: Ars Generalis Ultima (1645), Palma de Mallorca (Reprinted Minerva 1970); Obres de
Ramon Llull (1901 -3), ed. Rosello, Palma de Mallorca; Obres de Ramon Lull (1906-50), ed.
M. Obrador ef ai, Palma de Mallorca; Opera Latina (1959-), ed. Friedrich Stegmüller et ai,
(Maiorcensis Schola Lullistica), Palma dc Mallorca. (Vol. VII onwards published in Corpus
Christianorum, Continuatio Medievalis, 32, Brepols.); Opera Omnia (1721-42), ed. I.
Salzinger, Mainz (Reprinted Minerva 1965).
SECONDARY LITERATURE: Brummer, R. (1976). Bibliographia Lulliana, Ramon-Lull-
Schrifttum 1S70-1973, Gerstenberg; Colomer, E. (1961). Nikolaus von Kues und Raimund
Llull. Quellen und Studien zur Geschichte, De Gruyter; HillgarthJ. N. (15171). Ramon Lull and
Lullism in Fourteenth Century France (Oxford-Warburg Studies), Clarendon Press; Platzeck,
E. W. (1962-4). Raimund Lull, sein Leben, seine Werke (Biblioteca Franciscana, 5-6), Roma-
Düsseldorf.
88ο Biographies
R A Y M O N D OF P E N N A F O R T E
BIOGRAPHICAL DATA: b. ca. 1180; d. 1275. Spanish canon lawyer. Studied in Bologna
around 1210-20 before returning to teach law at Barcelona. Master General of the Domini
cans. Thomas Aquinas wrote the Summa contra gentiles on his advice.
TEXTS: Opera omnia (1945-), ed. J. Rius et Serra, University of Barcelona.
SECONDARY LITERATURE: Balmé, F. and C. Paban, eds. (1900-). Raymundiana; seu
documenta quae pertinent ad S. Raymondi de Pennaforti vitam et scripta, In domo generalitia.
R I C A R D U S S O P H I S T A (Magister Abstractionum)
BIOGRAPHICAL DATA: 'Ricardus Sophista' is the designation used for the author of the
Abstractiones found in at least five MSS; the designation appears in MS Oxford, Bodleian,
Digby 24, and MS Bruges, Bibliothèque de la Ville, 497. Manuscript evidence and internal
evidence suggest that this author must have written in the first decades of the thirteenth
century. Both Richard Fishacre and Richard Rufus of Cornwall have been recently
suggested in attempts to identify Ricardus Sophista.
R I C H A R D B I L L I N G H A M (Belyngham, Bilegam)
BIOGRAPHICAL DATA: fl. mid-fourteenth century. Fellow of Merton College 1344-61.
TEXTS : Conclusiones, MS Vatican lat. 3065 ff. 21 *-25 v a ; De sensu composito et diviso, MS Paris,
B. N. lat. 14715 ff. 79"—82rb; De significate proposkionis, MS Worcester, Bibl. Cathed. F35
ff. I09 v b -no v a ; Sophisma, MS Worcester, Bibl. Cathed. F 35 ff. I07-I09 vb ; Speculum
puerorum [see Maierù 1969, below].
SECONDARY LITERATURE: Knuuttila, Simo and Anja Inkeri Lehtincn (1979). 'Plato in
infinitum remisse incipit esse albus', in E. Saarinen et al. (eds.), Essays in Honour qfjaako
Hintikka, Reidel; Maierù, Alfonso (1969). 'Lo Speculum puerorum sive Terminus est in quern di
Riccardo Billingham', Studi Medievali, 3" série, 10, A Giuseppe Ermini, vol. 3, pp. 297—397;
Rijk, L. M. de (1976). 'Richard Billingham's Works on Logic',· Vivarium 14:121-38;
Weisheipl, James (1969). 'Repertorium Mertonense', Mediaeval Studies 31:174—224.
R I C H A R D B R I N K L E Y (Brenkell, Brinchil, Brinkel, Brinkil, etc.)

BIOGRAPHICAL DATA: fl. 3rd quarter of fourteenth century; O.F.M.
TEXTS: Several theological writings, including a Lectura on the Sentences and a
Distinctiones, attributed to him; Summa logicae (De termino in genere, De universalibus, De
praedicamentis. De suppositionibus terminorum. De propositionibus ingenere, De insolubilibus. De
obligationibus). MSS: Leipzig, Karl Marx Universität, MS 1360, ff. i n - i 0 5 v b (an inferior
copy); Prague, Statni Knihovan CSR, MS III.A. 11, ff. 3 i rl -i04™ (formerly Univ. Library
396); London, British Library, MS Harley 3243, ff. 49 rl -56 rb (De insolubilibus only). De
insolubilibus edited in Spade, Paul Vincent (1969), An Anonymous Fourteenth-Century Treatise
on Insolubles: Text and Study, L.S.M. Thesis, Pontifical Institute of Mediaeval Studies.
SECONDARY LITERATURE: Gâl, Gedeon, and Wood, Rega (forthcoming), 'Richard
Brinkley and his Summa logicae', Franciscan Studies.
R I C H A R D OF CAMPSALL
BIOGRAPHICAL DATA: d. ca. 1350/60. Cleric of the archdiocese of York. Master of
theology; fellow of Balliol and later of Merton Colleges at Oxford.
TEXTS : Logica Campsale Anglici valde utilis et realis contra Ockham, MS University of Bologna,
2635. Parts transcribed in 'The Universal and Supposition in a Logica attributed to Richard
of CampsaU' (1955), ed. Edward Synan i n j . R. O'Donnell (ed.) Nine Mediaeval Thinkers,
Rome, Pontifical Institute of Mediaeval Studies; Notabilia quedam Magistri Richardi camassale
pro materia de contingencia et presciencia dei, in 'Sixteen Sayings by Richard of Campsall on
Biographies 88ι
Contingency and Foreknowledge' (1962), Edward Synan (ed.) Mediaeval Studies

24:250-62; TTie Works of Richard of Campsall, vol. I: Questiones super librum Priorum
Analeticorum (1968), ed. Edward Synan, Pontifical Institute of Mediaeval Studies.
SECONDARY LITERATURE: Synan, Edward (1952). 'Richard of Campsall, an English
Theologian of the Fourteenth Century', Mediaeval Studies 14:1-8; Synan, Edward (1968).
'Introduction' to his edition of Campsall's Questiones super librum Priorum Analeticorum,
Pontifical Institute of Mediaeval Studies.
R I C H A R D O F C O R N W A L L (Ricardus Rufus, Ricardus de Cornubia, Ricardus

Rufus Cornubiensis)
BIOGRAPHICAL DATA: fl. mid-thirteenth century. Probably joined Franciscan order in
Paris sometime before 1238. Lectured on the Sentences at Oxford, 1250-3. Lectured on the
Sentences in Paris, 1253-5. Became regent master at Oxford around 1256, the fifth
Franciscan master at Oxford.
SECONDARY LITERATURE: Callus, D. A. (1939). 'Two Early Oxford Masters on the
Problem of Plurality of Forms, Adam of Buckfield, Richard Rufus of Cornwall', Revue
Néoscolastique de Philosophie 42:439-45; Ebbesen, Sten and Jan Pinborg (1970). 'Studies in
the Logical Writings Attributed to Boethius de Dacia', CIMAGL 3:1-54; Henquinet,
F. M. (1936). 'Autour des écrits d'Alexandre de Hales et Richard Rufus', ΛηιοηιαπΜΐΜ 2:
197-9 and 201-9; Pelster, F. {1929). 'Roger Bacons Compendium studii theologiae und der
Sentenzkommentar des Richardus Rufus', Scholastik 4:410-6; Pelster, F. (i949)· 'Der
Oxforder Theolog Richardus Rufus O.F.M. über die Frage: Utrum Christus in triduo mortis
fuerit homo', Recherches de Théologie ancienne et médiévale 16:259-80.
RICHARD CRAKANTHORPE
BIOGRAPHICAL DATA: b. 1567; d. 1624. Studied at Queen's College, Oxford, where he
was also a Fellow. Chaplain to diplomats and bishops; involved in theological controversies.
TEXTS: Introductio in Metaphysicam (1619), Oxford; Logicae libri quinque de Praedicabilibus,
Praedicamentis, etc. (1622), London.
RICHARD FISHACRE
BIOGRAPHICAL D A T A : d. 1248.
SECONDARY LITERATURE: Callus, D. A. (1943)- 'The Introduction of Aristotelian
Learning to Oxford', Proceedings of the British Academy 29:229-81; Pelster, F. (1930). 'Das
Leben und die Schriften des Oxforder Dominikanerlehrers Richard Fishacre', Zeitschrift fir
Katholische Theologie 54:518-53; Sharp, D. E. (1933)· 'The Philosophy of Richard
Fishacre', The New Scholasticism 7:281-97.
R I C H A R D F I T Z R A L P H (Armacanus)
BIOGRAPHICAL DATA : b. ca. 1295; d. 1360. A native of Dundalk, Ircland; Fellow of Balliol
College, Oxford; Vice-chancellor 1333; Archbishop of Armagh 1347.
TEXTS: De pauperie salvatoris, in R. Lane Poole, ed. (1890), Wyclif's Latin Works, VIII: De
dominio divino. Trübner; pp. 257-476.
SECONDARY LITERATURE: Leff, Gordon (1964). Richard FitzRalph, Commentator of the
Sentences, A Study in Theological Orthodoxy, Manchester University Press.
RICHARD HOLDSWORTH
BIOGRAPHICAL DATA: b. 1590; d. 1649. Studied at St John's College, Cambridge, where
he was also a Fellow. Professor of Divinity, Gresham College, London. Master, Emmanuel
College, Cambridge. Vice-Chancellor, Cambridge. Deprived of his appointments and
imprisoned by parliament in 1643. Named Dean of Worcester by Charles I in 1647.
882 Biographies
TEXTS: Directionsfor it Student in the Universitie, transcribed in H. F. Hetcher (1961), The

Intellectual Development ofJohn Milton (vol. 2), University of Illinois Press; Praclectioncs
Theologkae, habitae in Collegio Greshamensi apud Londinenses (1661), ed. Richard Pearson,
London; The Valley of Vision (1651), London.
RICHARD HOOKER
BIOGRAPHICAL DATA: b. ca. 1554; d. 1600. Studied at the Latin School at Exeter and at
Corpus Christi College, Oxford, where he was also a Fellow. Lecturer in logic and deputy
professor of Hebrew at Oxford. Master of Temple Church in London, 1585-91. Rector of
Bishopsbourne.
TEXTS : Of the Laws ofEcclesiastical Polity, in W. Speed Hill (ed.) (i977~), The Folger Library
Edition of the Works of Richard Hooker (5 vols.), Cambridge, Mass., The Belknap Press of
Harvard University Press.
SECONDARY LITERATURE: D'Entrèves, Alessandro Passerin (1939). The Medieval
Contribution to Political Thought: Thomas Aquinas, Marsilius of Padua, Richard Hooker,
Oxford University Press; Hill, W. Speed, ed. (1972). Studies in Richard Hooker: Essays
Preliminary to an Edition of His Works, Case Western Reserve University Press; Munz, Peter
(1952). The Place of Hooker in the History of Thought, Routledge and Kegan Paul.
R I C H A R D K I L V I N G T O N (Kilmington, Chillington, Climenton, etc.)

BIOGRAPHICAL DATA: d. 1361. At Oxford, perhaps Oriel College, before 1330. Along
with Thomas Bradwardine and others, a member of Richard de Bury's circle. Dean of St
Paul's, London.
TEXTS: Sophismata (forthcoming), ed. N. and B. Kretzmann.
SECONDARY LITERATURE: Bottin, Francesco (1973). 'Analisi linguistica e fisica
Aristotelica nei "Sophysmata" di Richard Kilmyngton', in C. Giacon (ed.) Filosofia e
Politica e altri saggi, Editrice Ancenore; Bottin, Francesco (1974). 'Un testo fondamentale
neu' ambito della "Nuova Fisica" di Oxford: I Sophismata di Richard Kilmington',
Miscellanea Mediaevalia 9:201-5; Kretzmann, Norman (1977). 'Socrates is Whiter than
Plato begins to be White', Nous 11:3-15.
R I C H A R D LAVENHAM
BIOGRAPHICAL DATA: d. 1399. Born at Lavenham in Suffolk. Became Carmelite friar at
Ipswich. Received doctorate in theology from Oxford. Prior of Bristol convent of
Carmelites. Confessor to Richard II.
TEXTS: At least sixty-three works attributed to him. Compiled a collection of heresies
drawn from works of the Wyclifite John Purvey. Wrote commentaries on Aristotle's
Physics and Ethics, treatises on astronomy, logic, natural philosophy, and other topics. See
Klingsford 1949-50 and Emden 1957-9, 2:1109-10; Treatises on logic and natural
philosophy in MS British Library Sloane 3899, ff. l r -90 v ; Consequentiae, Exceptivae,
Suppositiones, Tractatus excltisivarum, Tractatus qui differt et aliud nuncupatur edited in Spade,
Paul Vincent (1974a); 'Five Logical Tracts by Richard Lavenham' in J. Reginald
O'Donnell (ed.). Essays in Honour of Anton Charles Pegis, Pontifical Institute of
Mediaeval Studies; De propositionibus modalibus edited in Spade, Paul Vincent (1973a), "The
Treatises On Modal Propositions and On Hypothetical Propositions by Richard Lavenham',
Mediaeval Studies 35 :49~59; De propositionibus hypotheticis partially edited in Spade 1973a;
edition completed in Spade, Paul Vincent (1975b), 'Notes on Some Manuscripts of Logical
and Physical Works by Richard Lavenham', Manuscripla 19:139-46; Obligationes edited in
Spade, Paul Vincent (1978b), 'Richard Lavenham's Obligationes: Edition and Comments',
Rivista Critica di Storia della Filosofia 33:225-42, Summulae logkales partially edited in Spade,
Biographies 883
Paul Vincent (forthcoming a), 'Notes on Richard Lavenham's So-called Summulae logkales,
with a Partial Edition of the Text', Franciscan Studies. (Edition completed in Spade 1973a
and Spade 1975b.)
SECONDARY LITERATURE: Spade, Paul Vincent (1980a), 'Richard Lavenham and the
Cambridge Logic', Historiographia Linguistka 7:241—7.
R I C H A R D S W I N E S H E A D (Calculator, Suisseth, Suiset)
BIOGRAPHICAL DATA: fl. 1340—55. Fellow of Merton College, Oxford. Later commonly
called 'The Calculator' because of his major work Liber calculationum (ca. 1350).
TEXTS : De motu. De motu locali, and De caelo, MS Cambridge, Gonville and Caius 499/268,
ff. 2i2 t l -2i3 , b ,2i3' b -2i5 r b , 204 la -2ii vb ; Liter Calculationum (1477). Padua; (1520), Venice.
SECONDARY LITERATURE: Clagett, Marshall (1941). Giovanni Marliani and Late Medieval
Physics, Columbia University Press; Clagett, Marshall (1950). 'Richard Swineshead and
Late Medieval Physics', Of iris 9:131-61; Clagett, Marshall (1959). The Science of Mechanics
in the Middle Ages, University of Wisconsin Press; Murdoch, John and Edith Sylla (1976).
'Swineshead, Richard', Dictionary of Scientific Biography, vol. 13; Sylla, Edith (1971).
'Medieval Quantifications of Qualities: the "Merton School"', Archive for History of Exact
Sciences 8:9-39; Sylla, Edith (1973). 'Medieval Concepts of the Latitudes of Forms: the
Oxford Calculators', Archives d'histoire doctrinale et littéraire du moyen âge 40:223—83;
Weisheipl, J. A. (1964). 'Roger Swyneshed, O.S.B., Logician, Natural Philosopher, and
Theologian', in Oxford Studies Presented to Daniel Callus, Oxford University Press;
Weisheipl, J. A. (1969). 'Repertorium Mertonense', Mediaeval Studies 31:174-224.
ROBERT BACON
TEXTS: Syncategoremata (forthcoming), ed. H. A. G. Braakhuis.
SECONDARY LITERATURE: Callus, D. A. (1943). 'The Introduction of Aristotelian
Learning to Oxford', Proceedings of the British Academy 29:229-81; Smalley, Beryl (1948).
'Robert Bacon and the Early Dominican School at Oxford', Transactions of the Royal
Historical Society (Fourth Series) 30:1-19; Braakhuis, H. A. G. (1979). De I3tk eeuwse
tractaten over syncategorematische termen (Deel I), Krips Repro Meppel.
R O B E R T FLAND
BIOGRAPHICAL DATA: Wrote between 1335 and 1370. Probably associated with the
University of Oxford, perhaps with Merton College. See Spade 1978c.
TEXTS: Consequentiae, Insolubilia, Obligationes, in MS Bruges, Bibliothèque publique de la
Ville, 497, ff. 4i r *-46 n ; Consequentiae edited in Spade, Paul Vincent (1976), 'Robert Fland's
Consequentiae: An edition', Mediaeval Studies 38:54-84; Insolubilia edited in Spade, Paul
Vincent (1978c). 'Robert Fland's Insolubilia: An Edition, with Comments on the Dating of
Fland's Works', Mediaeval Studies 40:56-80; Obligationes edited in Spade, Paul Vincent
(1980), 'Robert Fland's Obligation«; An Edition', Mediaeval Studies 42:41-60.
R O B E R T G R O S S E T E S T E (Lincolniensis)
BIOGRAPHICAL DATA: b. ca. 1168/75; d. 1253. Studied at Oxford, possibly also at Paris,
1209-14. First chancellor of Oxford and first teacher to Friars Minor (1224/32-35). Elected
bishop of Lincoln, 1235. Translated many works from Greek, including Aristotle's Nko-
machean Ethics, but more inclined to Augustinian Neo-Platonism than to Aristotelianism.
TEXTS: Commentatius in Posteriorum analyticorum libros (1981), ed. P. Rossi, Leo S. Olschki
Die philosophischen Werke des Robert Grosseteste (1912), ed. L. Baur (BGPM, IX),
Aschendorff; Roberti Grossetesste Commentarius in VIII libros Physicorum Aristotelis (l 963), ed.
R. C. Dales, University of Colorado Press.
884 Biographies
SECONDARY LITERATURE: Callus, D. Α., ed. (1955). Robert Grosseteste, Scholar and
Bishop, Oxford University Press; Crombie, A. C. (1953). Robert Grosseteste and the Origins of
Experimental Science, Oxford University Press; Thomson, S. H. (1940). The Writings of
Robert Grosseteste, Cambridge University Press.
R O B E R T H O L K O T (Holcot)
BIOGRAPHICAL DATA: h.ca. 1290; d. 1349. Oxford Dominican commentator, theologian,
and philosopher. Studied and taught at Oxford ca. 1326-34. Became master in theology in
1332. His view that God is the cause but not the author of sin was condemned at Paris in
1347·
TEXTS: Ά Quodlibetal Question of Robert Holkot, O. P., on the Problem of the Objects
of Knowledge and Belief* (1964), E. A. Moody (ed.), Speculum 39:53-74; In quattuor libros
Sententiarum quaestiones (1518), Lyon (Reprinted Minerva 1967); QuodUbeta 1,6, in William
J. Courtenay (1971), Ά revised text of Robert Holkot's quodlibetal dispute on whether
God is able to know more than he knows', Archiv fur Geschichte der Philosophie 53 :1-21;
'Utrum theologia sit scientia: A quodlibetal question' (1958), J. T. Muckle (ed.), Medieval
Studies 20:127-53.
SECONDARY LITERATURE: Schepers, H. (1970). 'Holkot contra dicta Crathorn I;
Quellenkritik und biographische Auswertung der Bakkalareatsschriften zweier Oxforder
Dominikaner des XIV. Jahrhunderts', Philosophisches Jahrbuch der Görres-Gesellschaft
77:320-54.
ROBERT KILWARDBY
B I O G R A P H I C A L D A T A : d. 1279.
TEXTS: De natura theologiae (1935), ed. Friedrich Stegmüller (Opuscula et Textus, Series
Scholastica, 17), Aschendorff; De ortu scientiarum (1976), ed. A. G. Judy (Auctores Britannici
Medii Aevi, 4), British Academy; 'The Commentary on "Priscianus Maior" Ascribed to
Robert Kilwardby' (1975), Jan Pinborg et al., CIMAGL 15:1-146.
SECONDARY LITERATURE: Sommer-Seckendorff, E. M. F. (1937). Studies in the Life of
Robert Kilwardby, Ο. P. (Dissertationes Historicae, 7), Institutum Historicum FF. Praedica-
torum Romae ad S. Sabinae.
ROBERT SANDERSON
BIOGRAPHICAL DATA: b. 1587; d. 1663. Student and Fellow at Lincoln College, Oxford.
Regius Professor of Divinity, Oxford. Ousted by parliament in 1648; reinstated in 1660.
Held various ecclesiastical appointments, including chaplain to Charles I and Bishop of
Lincoln after the Restoration. Participated in the revision of the Book of Common Prayer in
1661.
TEXTS: Logicae Artis Compendium (1615, 1618), Oxford.
ROGER BACON
BIOGRAPHICAL DATA: b. ca. 1214; d. 1292/4. Studied at Oxford. Master of arts in Paris by
1237, where he lectured on Aristotle for at least ten years. Relinquished his position and
turned to experimental science and astronomical tables. Joined the Franciscans around
1252-7. At the request of Pope Clement IV (1265-8), he outlined his reform of learning,
the Opus maius, in 1267. The Opus minus and Opus tertium followed in 1268. Condemned for
certain suspect doctrines (probably Joachite positions) by the Minister General of his order
and Pope Nicholas III (1277-80). Imprisoned around 1254-7. Later allowed to return to
England.
Biographies 885
TEXTS: 'An Unedited Part of Roger Bacon's Opus maius: De signis' (1978). K. M.
Fredborg, L. Nielsen, and Jan Pinborg (eds.), Traditio 34:75-136; Compendium studii
theologiae (1911), ed. H. Rashdall, Aberdeen University Press (Reprinted Gregg 1966);
Moralis Philosophia (1953), ed. F. Delorme and E. Massa, Antenore; Opera hactenus inedita
R. Baconis (1905-40), ed. R. Steele, Oxford, Clarendon Press; Opera inedita (1859), ed.
J. S. Brewer (Rerum Britannicarum Medii Aevi Scriptores), London (Reprinted Kraus
1966); Opus Maius (1897-1900), ed. J. H. Bridges, Oxford University Press (Reprinted
Minerva 1964); The Greek Grammar of Roger Bacon (1902), ed. E. Nolan and S. A. Hirsch,
Cambridge University Press.
SECONDARY LITERATURE: Alessio, F. (1959). 'Un secolo di studi su Ruggero Bacone',
Rivista critica di storia dellafilosofia14:81 -102; Crowley, T. ( 1950). Roger Bacon: The Problem
of the Soul in His Philosophical Commentaries, Louvain, Publications Universitaires; Easton,
S. C. (1952). Roger Bacon and His Search for a Universal Science, Blackwell; Little, A. G., ed.
(1914). Roger Bacon Essays, Oxford University Press (Reprinted New York, Russell and
Russell 1972).
R O G E R S W I N E S H E A D (Swyneshed, Sineshead, Schweinshaupt, Suicet, Suiscth,

etc.)
BIOGRAPHICAL DATA: fl. before 1335; d. ca. 1365. Master of theology. Benedictine monk
of Glastonbury.
TEXTS: De motibus naturalibus, MS Erfurt, Amplon. F. 135, ff. 25 va -47 rb ; Insolubilia, in Paul
Vincent Spade, ed. (1979a), 'Roger Swyneshed's Insolubilia: Edition and Comments',
Archives d'histoire doctrinale et littéraire du moyen âge 46:177-220; Obligationes, in Paul
Vincent Spade, ed. (1977), 'Roger Swyneshed's Obligationes: Edition and Comments',
Archives d'histoire doctrinale et littéraire du moyen âge 44:243-85.
SECONDARY LITERATURE: Clagett, Marshall (1941). Giovanni Marliani and Late Medieval
Physics, Columbia University Press; Sylla, Edith (1973). 'Medieval Concepts of the Latitude
of Forms: the Oxford Calculators', Archives d'histoire doctrinale et littéraire du moyen âge
40:223-83; Weisheipl, James (1964). 'Roger Swyneshed, O. S. B., Logician, Natural
Philosopher, and Theologian', in Oxford Studies Presented to Daniel Callus, Clarendon Press.
R U F I N U S O F ASSISI
BIOGRAPHICAL DATA: d. ca. 1190. Bishop of Assisi, Archbishop of Sorrento. Canon
lawyer and theologian. Taught at Bologna from ca. 1150. Wrote a popular, systematic
commentary on Gratian.
TEXTS: Die Summa Decretorum des Magister Rufinus (1902), ed. H. Singer, Schöningh; Die
Summi magistri Rufini zum Dekretum Grattant (1892), ed. J. F. von Schulte, Roth.
SIGER OF B R A B A N T (Sigerus, Segerus, Sogerus, Sugerus)

BIOGRAPHICAL DATA: b. ca. 1240; d. 1284. Cited by the papal legate Simon of Brion in
1266 in connection with political disturbances in the arts faculty at the University of Paris,
where he was a master of arts, and one of the leaders of the radical Aristotelian movement.
Some of his beliefs condemned by St Bonaventure in his Lenten Conferences of 1267 and
1268 and in 1273, and by Stephen Tempier, Bishop of Paris in 1270. Thomas Aquinas'
attack in 1270 on the doctrine of the unicity of the intellect caused Sigerto modify his views
somewhat, but his career continued to be stormy. Called to appear before the tribunal of the
Inquisitor of France, Simon du Val, in 1276, but was already away from Paris. Appealed to
the papal curia and ended his days in Italy, without being convicted of the charge of heresy.
TEXTS: Les Quaestiones super librum de causis de Siger de Brabant (1972), ed. A. Marlasca
(Philosophes Médiévaux, 12). Publications universitaires de Louvain; Quaestiones in tertium
886 Biographies
De Anima, De anima intellectiva, De aetemitate mundi (1972), ed. B. Bazân (Philosophes

Médiévaux, 13), Publications universitaires de Louvain; Questions sur la Métaphysique
(1948), ed. C. A. Graiff (Philosophes Médiévaux, 1), Publications universitaires de Louvain;
'Die Questiones metaphysice très des Siger von Brabant' {1966), ed. J. Vennebusch, Archiv
fir Geschichte der Philosophie 48:163-89; Siger de Brabant. Ecrits de logique, de morale et de
physique (1974), ed. B. Bazân (Philosophes Médiévaux, 14), Béatrice-Nauwelaerts.
SECONDARY LITERATURE: Steenberghen, F. Van (1977). Maifre Siger de Brabant
(Philosophes Médiévaux, 21), Béatrice-Nauwelaerts.
SIGER O F C O U R T R A I
BIOGRAPHICAL DATA: d. 1341. Studied in Paris around 1300. Appointed dean of the
chapter of the Church of Our Lady at Courtrai in Flanders shortly after 1305; remained in
office till shortly before 1330. Again in Paris in 1315. Belonged to the school of the modistae.
TEXTS: Les oeuvres de Siger de Courtrai (1913), ed. G. Wallerand (Les philosophes Belges, 8),
Louvain, Publications Universitaires; Summa modorum significandi, Sophismata (1977),
ed. Jan Pinborg (Amsterdam Studies in the Theory and History of Linguistic Science, 14),
John Benjamins; Zeger van Kortrijk, Commentator van Perihermeneias (1964), ed. C. Verhaak,
Paleis der Academiën.
SECONDARY LITERATURE: Bursill-Hall, G. L. (1971). Speculative Grammars of the Middle
Ages, Mouton; Trentman, John (1976). 'Speculative Grammar and Transformational
Grammar: A Comparison of Philosophical Presuppositions' in Herman Parret (ed.) History
of Linguistic Thought and Contemporary Linguistics, De Gruyter.
S I M O N O F F A V E R S H A M (Symon Anglicus)
BIOGRAPHICAL DATA: b. ca. 1260; d. 1306/7 Master of Arts and Doctor of Theology at
Oxford. Chancellor of Oxford, 1304. Archdeacon of Canterbury, 1305.
TEXTS: Opera omnia. Vol. I, Opera logica, tomus prior: Quaestiones super libro Porphyrii,
Quaestiones super libro Praedicamentorum, Quaestiones super libro Perihermeneias (1957), ed. P.
Mazzarella, Cedam; Qiidestion« in très libros De anima (15)63) inj. Vennebusch, Ein anonymer
Aristoteles Kommentar des XIII. Jahrhunderts: Quaestiones in très libros De anima, Schöningh;
Quaestiones super tertium De anima (1934) in D. Sharp, 'Simon of Faversham's Quaestiones
super tertium De anima", Archives d'histoire doctrinale et littéraire du moyen âge 9:307-68;
Sophisma 'Universale est intentio' (1969) in T. Yokoyama, 'Simon of Faversham's Sophisma
'Universale est intentio', Mediaeval Studies 31:1 — 14; Quaestiones novae et veteres super librum
Elenchorum (forthcoming), ed. Del Punta et al., Pontifical Institute of Mediaeval Studies.
SECONDARY LITERATURE: Grabmann, Martin (1933). Die Aristoteleskommentare des
Simon von Faversham: Handschriftliche Mitteilungen (Sitzungsberichte der Bayerischen
Akademie der Wissenschaften, Philosophisch-historische Abteilung, 3), Verlag der
Bayerischen Akademie; Longeway, John (1977). Simon of Faversham's Questions on the
Posterior Analytics: A Thirteenth-Century View of Science, unpublished Ph.D. dissertation,
Cornell Univenity; Pinborg, Jan (1971). 'Simon of Faversham's Sophisma: Universale est
intentio: A Supplementary Note', Mediaeval Studies 33:360-5; Rijk, L. M. de (1968). 'On
the Genuine Text of Peter of Spain's Summule logkales. II Simon of Faversham (d. 1306) as a
Commentator of the Tracts I-V of the Summule', Vivarium 6:69-101; Vennebusch, J.
(1965). 'Die Quaestiones in très libros De anima des Simon von Faversham', Archiv fir
Geschichte der Philosophie 47:20-39; Wolf, F. J. (1966). Die Intellektslehre des Simon von
Faversham nach seinem De anima Kommentaren, Inaugural dissertation, University of Bonn.
T H O M A S A Q U I N A S (Thomas d'Aquino)
BIOGRAPHICAL DATA: b. ca. 1225; d. 1274. Studied with the Benedictines at Monte
Cassino beginning in 1231 and then with secular masters at the University of Naples,
Biographies 887
1239-44. Joined Dominicans at Naples in 1244. Studied in Paris, 1245-8. Studied and
taught under Albert the Great at Cologne, 1248-52. Master of theology at Paris in 1256,
where he taught until 1259. Spent next ten years in Italy, where he wrote the first part of his
Summa theologiae. Ordered back to Paris in 1269, where he composed the second part of the
Summa and most of his commentaries on Aristotle. During this time, he was active in the
defense of the mendicant Orders against secular clerics and of his own orthodox use of
Aristotle against the Augustinians and the Latin Averroists. Died on his way to the Council
of Lyon, leaving unfinished the third part of his Summa and various commentaries on
Aristotle.
TEXTS: De regimine principum, in R. M. Spiazzi, ed. (i954)· Divi Thomae Aquinatis Opuscula
Philosophica, Turin, Marietti. Translated in On Kingship to the King of Cyprus (1949),
G. Phelan and I. T. Eschmann (trs.), Pontifical Institute of Mediaeval Studies; In decern libros
Ethkorum Aristotelis adNicomachum expositio (1964), ed. R. M. Spiazzi (3rd edn.), Marietti; In
octo libros Politicorum Aristotelis expositio (15)66), ed. R. M. Spiazzi, Marietti; Opera omnia
(1852-73), Parma, Tipis Petri Fiaccadori (Reprinted Musurgia 1948-50); Opera omnia
(1871-80), ed. S. E. Fretté and Paul Mare Vives; Opera omnia (1882-), ed. Leonine
Commission, Vatican Polyglot Press; S. Thomae Aquinatis opuscula omnia (1927), ed.
P. Mandonnet, Lethielleux.
SECONDARY LITERATURE: Eckert, W. P., ed. (1974). Thomas von Aquino. Interpretation
und Rezeption, Grünewald; Gilson, Etienne (1956). The Christian Philosophy of St Thomas
Aquinas, Random House; Kluxen, W., ed. (1975). Thomas von Aquinas im philosophischen
Gespräch, Alber; Oeing-Hanhoff, L., ed. (1974). Thomas von Aquin 1274-1974, Kösel; Walz,
A. (1962). Saint Thomas D'Aquin (Philosophes médiévaux, 5), Béatrice-Nauwelaerts;
Weisheipl, J. Α. (ΐ974)· friar Thomas d'Aquino: His Life, Thought and Works, Blackwell.
T H O M A S B R AD W A R D I N E (Thomas Beer de Hertefeld)

BIOGRAPHICAL DATA: b. ca. 1295; d. 1349. Fellow of Balliol College in 1321 and of
Merton College by 1323. Proctor of University of Oxford in 1325 and again in 1327.
Master of arts before 1326. Left Merton College in 1335 to join household of Richard de
Bury. Master of theology at Oxford and then Chancellor of Saint Paul's, London, 1337-49.
Confessor to King Edward in 1346. Doctor of theology before 1348. Archbishop of
Canterbury in 1349. Vigorous opponent of contemporary Pelagianism.
TEXTS: Insolubilia, in Marie-Louise Roure (1970), 'La problématique des propositions
insolubles au XIIIe siècle et au début du XIVe, suivie de l'édition des traités de W.
Shyreswood, W. Burleigh et Th. Bradwardine', Archives d'histoire doctrinale et littéraire du
moyen âge 37:205-326; Summa de causa Dei contra Pelagium et de virtute causarum ad sues
Mertonenses libri très (1618), ed. H. Savile, London (Reprinted Minerva 1964); Thomas
Bradwardine His Tractatus de Proportionibus (1955), ed. H. Lamar Crosby, Jr., University of
Wisconsin Press.
Middle Ages, University of Wisconsin Press; Leff, Gordon (1957). Bradwardine and the
Pelagians, Cambridge University Press; Murdoch, John (1976). 'Bradwardine, Thomas', in
Dictionary ofScientific Biography.Obermati, H. A. (1957). Archbishop Thomas Bradwardine. A
Fourteenth Century Augustinian, Zemink and Zoon; Weisheipl, James A. (1968). Ockham
and Some Mertonians', Mediaeval Studies 30:163-213; Weisheipl, James A. (1969).
'Repertorium Mertonense', Mediaeval Studies 31: 174-224.
T H O M A S OF E R F U R T
BIOGRAPHICAL DATA: fl. ca. 1300.
TEXTS: Grammatica speculativa (1972). G. L. Bursill-Hall (ed. and tr.) (The Classics of
Linguistics, 1), Longmans.
888 Biographies
S E C O N D A R Y L I T E R A T U R E : Bursill-Hall, G. L. (1971). Speculative Grammars of the Middle

Ages. The Doctrine of the partes orationis of the Modistae, Mouton; Grabmann, Martin (1943).
Thomas von Erfurt und die Sprachlogik des Mittelalterlichen Aristotelismus (Sitzungsberichte deT
Bayerischen Akademie der Wissenschaft, Heft 2), Hueber; Pinborg, Jan (1967). Die
Entwicklung der Sprachtheorie im Mittelalter (BGPM, XLII, 2), Aschendorff.
U G O L I N O O F O R V I E T O (Hugolin)
TEXTS: Prologus to Commentaria in quattuor libros Sententiarum, in A. Zumkeller (1941),
Hugolin von Orvieto und seine theologische Erkenntnislehre (Cassiciaeum, Eine Sammlung
wissenschaftlicher Forschungen über den hl. Augustinus und den Augustinerorden, Band
IX, 2. Reihe, 3. Band) Augustinus-Verlag; Der Physikkommentar Hugolins von Orvieto
OESA (1972), ed. Willigis Eckermann. De Gruyter.
S E C O N D A R Y L I T E R A T U R E : Corvino, F. (1967). 'Ugolino di Orvieto', in Enciclopedia
filosofica 6, Sansoni.
V I T A L D U F O U R (Vitalis de Furno)
TEXTS : De rerum principio, in Johannis Duns Scoti Opera omnia (1639), ed. L. Wadding, Vives;
Quaestiones disputatae, in F. M. Delorme, ed. (1927), 'Le Cardinal Vital du Four, Huit
questions disputées sur le problème de la connaissance', Archives d'histoire doctrinal et
littéraire du moyen âge 1:151—337; QuodUbeta Tria (1947), ed. F. M. Delorme, Spicilegium
Pontificii Athenaei Antoniani; Speculum morale totius sacrae scripturae (1513), Lyon.
SECONDARY L I T E R A T U R E : Lynch, John Ε. (1972). The Theory of Knowledge of Vital du
Four, The Franciscan Institute; Pisvin, A. (1949). 'Die Intuition und ihre metaphysische
Wert nach Vitalis de Furno und Gonsalvus Hispanus', Wissenschaft und Weisheit 12:147-73;
Untervintl, L. von (1955). 'Die Intuitionslehrc bei Vitalis de Furno', Collectanea Franciscan»
25:53-113.225-53·
W A L T E R B U R L E Y (Burleigh, Bayle, Borle, Bourle, Brulleus, Burghle, Burlaeus,

Burle, Burlie)
BIOGRAPHICAL D A T A : b. ca. 1275; d. 1344/5. Master of arts at Oxford and Fellow of
Merton College by 1301. Remained Fellow of Merton until 1305. Ordained in 1309.
Studied theology under Thomas Wilton at Paris beginning before 1310. Master of theology
at Paris around 1320-2. Fellow of the Sorbonne by 1324. Appointed envoy to the papal
court at Avignon by Edward III in 1327. Held quodlibetal disputation in Toulouse in 1327
and in Bologna in 1341. Clerk of the household o f Richard de Bury, Bishop o f Durham in
1333, and o f the King's household in 13 36. Held a number of benefices, including a canonry
at York and one at Salisbury.
TEXTS : An Introduction to the Treatise 'De obligationibus' With Critical Editions of the Treatises
of William of Sherwood (?) and Walter Burley (1963), ed. Romuald Green, unpublished
doctoral dissertation, Institut Supérieur de Philosophic de Louvain; De primo et ultimo
instanti (1965), ed. H. and C. Shapiro, Archiv für Geschichte der Philosophie 47:157—73; De
puritate artis logicae tractatus longior, with a revised edition of the Tractatus brevior (1955), ed.
Philotheus Boehner, Franciscan Institute; De relattvis (1962), ed. H. Shapiro, Franciscan
Studies 22:155-71 ; 'La problématique des propositions insolubles au XIII e siècle et au début
du XIV e , suivie de l'édition des traites de W . Shyreswood, W . Burleigh et Th.
Bradwardine' (1970). M.-L. Roure (ed.), Archives d'histoire doctrinale et littéraire du moyen âge
37:205-326; 'The De potentiis animae of Walter Burley' (1974). J. M. Kitchel (ed.),
Mediaeval Studies 33:85-113; 'Walter Burleigh's Middle Commentary on Aristotle's
Perihermeneias' (1973). S. F. Brown (ed.), Franciscan Studies 33 :42-134; 'Walter Burleigh's
Biographies 889
Quaestiones in librum Perihermeneias' (1974). S. F. Brown (ed.), Franciscan Studies 34:200-95;

'Walter Burleigh's Treatise De Suppositionibus and Its Influence on William of Ockham'
(1972). S. F. Brown, Franciscan Studies 32:15-64; 'Walter Burley's Treatise De formu'
(1970). F.J. D. Scott (ed.), Bayerische Akademie der Wissenschaften (Veröffentlichungen d.
Kommiss, f. d. Hrsg. ungedr. Texte aus d. Mittelalt. Geisteswelt, 4).
SECONDARY LITERATURE: Martin, C. (1964). 'Walter Burley', in Oxford Studies Presented
to Daniel Callus (Oxford Historical Society, n.s., 16), Clarendon Press; Murdoch, John and
Edith Sylla (1970). 'Burley, Walter', in Dictionary of Scientific Biography; Una Juârez, A,
(1978). Lafilosofia del siglo XIV. Contexte cultural de Walter Burley, Biblioteca 'La Ciudad de
Dios', Real Monasterio de El Escorial; Weisheipl, James A. (1968). 'Ockham and Some
Mertonians', Mediaeval Studies 30:163-213; Weisheipl, James A. (1969). 'Repertorium
Mertonense', Mediaeval Studies 31:174-224.
W A L T E R CH A T T O N (Chaton, Catton, Caton, Cepton, Schaton)

BIOGRAPHICAL DATA: b, 1285; d. 1344. Entered the Franciscan order as a boy. Ordained
subdeacon in 1307. Commented twice on the Sentences, once (the Reportatio) in 1322/3 at
Oxford, and again later (Lectura) at some unknown place and date. May have used
Ockham's Sentence commentary in producing his own. Adam Wodeham attended
Chatton's lectures and showed his notes to Ockham. Returned to Oxford as fifty-third
regent in 1330. Was in Assisi with Gerard of Odo, Minister General of the Franciscans, in
1332. At the papal court in Avignon in 1333, where, as one of the Pope's advisors, he was
involved in various theological controversies and proceedings.
TEXTS: Commente alle Senienza Prologo-Questione Terza (1973). Luciano Cova (ed.),
Edizioni deirAteneo; De paupertate evangelka, in D. Douie, ed. (1931-2), 'Three Treatises
on Evangelical Poverty', Archivum Franciscanum Historkum 14:341-69,15:36-58,210—40;
'Ein ockhamkritischer Text zu Signifikation und Supposition und zum Verhältnis von
erster und zweiter Intention' [In Sent. I d. 22-23] ('975)· Christian Knudsen (ed.),
CIMAGL 14:1-26; 'Gauthier de Chatton et son Commentaire des Sentences' (1943).
L. Baudry (ed.). Archives d'histoire doctrinale et littéraire du moyen âge 14:3 37-69; 'Gualteri de
Chatton and Guillelmi de Ockham Controversia de Natura Conceptus Universalis' (1967).
Gedeon Gâl (ed.), Franciscan Studies 27: 191-212; 'La prima questione del Prologo del
Commento aile Sentenze di Walter Catton' (1970). M. E. Reina (ed.), Rivista critica di storia
dellafilsofia25: 48-74, 290-314; 'La quarta questione del Prologo del Commento aile
Sentenze di Walter Catton' (1975). Luciano Cova (ed.), Rivista critica di storia della filosofia
30: 303-16; "The Second Question of the Prologue to Walter Chatton's Commentary on
the Sentences. On Intuitive and Abstractive Knowledge' (1955). J. J- O'Callaghan (ed.), in
R. O'Donnell (ed.) Nine Mediaeval Thinkers, Pontifical Institute of Mediaeval Studies; 'Two
Questions on the Continuum: Walter Chatton (?), O. F. M. and Adam Wodeham, O. F. M.'
(1966). E. Synan and J. Murdoch (eds.), Franciscan Studies 26:212-88; 'Walter Chatton on
the Univocity of Being: A Reaction to Peter Aureoli and William Ockham' (1971). Noel
A. Fitzpatrick (ed.), Franciscan Studies 31:88-177.
SECONDARY LITERATURE: Auer, J. (1953). 'Die "skotistische" Lehre von der Heilsge-
wissheit: Walter von Chatton der erste "Skotist"', Wissenschaft und Weisheit 16:
1-19; Brady, Ignatius (1967). 'Walter of Chatton', in The New Catholic Encyclopedia;
Knudsen, Christian (1975). Walter Chattons Kritik an Wilhelm von Ockhams Wissen-
schaftslehre, unpublished doctoral dissertation, Rheinische Friedrich-Wilhelms Universität,
Bonn.
WILLIAM OF A L N W I C K
BIOGRAPHICAL DATA : f). ca. 1315. Lectured on the Sentences at Paris before 1314. He was
associated with Duns Scotus in the opening years of the fourteenth century and may be the
890 Biographies
author o f the additiones to Books I and II of Scotus' Commentary on the Sentences. His
unedited Determinationes is a collection of questions dealing largely with topics in natural
philosophy.
TEXTS: Quaestiones disputatae de esse intelligibili et de quolibet (1937), ed. A. Ledoux,
Collegium S. Bonaventurae.
S E C O N D A R Y L I T E R A T U R E : Ledoux, A. (i937). 'Introduction' in his edition of Quaestiones
disputatae.
W I L L I A M A R N A U D (Guillielmus Arnaldi)
BIOGRAPHICAL D A T A : fl. mid-thirteenth century. Master of arts at Toulouse in 1235-44.
Wrote one o f the earliest commentaries on Peter of Spain's Tractatus.
T E X T S : 'On the genuine text of Peter of Spain's Summulae logkales, IV: The Lectura
Tractatuum by Guilelmus Arnaldi, Master of arts at Toulouse (1235-1244)' (1969). L. M. de
Rijk (ed.), Vivarium 7:120—62.
W I L L I A M O F A U V E R G N E (William of Paris, Guilelmus Alvernus, Guilelmus

Arvernus)
BIOGRAPHICAL D A T A : b. ca. 1180; d. 1249. Taught theology at Paris. Bishop of Paris from
1228-49.
T E X T S : De universe (1516), Paris. Opera omnia (1674), Paris (Reprinted Minerva 1963);
'Tractatus magistr. Guillelmi Alvernensis De bono et malo' (1946). J. R. O'Donnell (ed.),
Mediaeval Studies 8:245-99; 'Tractatus secundus Guillelmi Alvernensis De bono et malo'
( J 954)· J· R- O'Donnell (ed.), Mediaeval Studies 16:219-71.
S E C O N D A R Y L I T E R A T U R E : Baumgartner, Matthias (1893). 'Die Erkenntnislehre des
Wilhelm von Auvergne' (BGPM II, ι), Aschendorff; Chariand, T. M. (1936). Artes
praedkandi, Vrin, Forest, A. (1952). 'Guillaume d'Auvergne, critique d'Aristote', in Études
médiévales offertes à M. la doyen A. Fliehe, Presses universitaires françaises; Masnovo, A.
(1930-45). Da Guglielmo d'Auvergne a S. Tommaso d'Aquino (2 vols.), 'Vita e pensiero'.
Moody, E. A. (1975). 'William of Auvergne and His Treatise De anima' in his Studies in
Medieval Philosophy, Science, and Logk, University of California Press; Valois, Noël (1880).
Guillaume d'Auvergne, évêque de Paris (1228-1249), sa vie et ses ouvrages, Alphonse Picard.
W I L L I A M O F A U X E R R E (Guilelmus Autissiodorensis)
BIOGRAPHICAL D A T A : b.ca. 1140/50; d. 1231. Master of theology at Paris. Archdeacon of
Beauvais.
T E X T S : Summa aurea in quattuor libros Sententiarum (1500), Paris (Reprinted Minerva 1964);
(1518), Paris; (1591), Venice.
S E C O N D A R Y L I T E R A T U R E : Ottaviano, Carmelo (1931). Guglielmo d'Auxerre, la vita, le
opere, il pensiero, "L'Universale" Tipografia Poliglotta; Principe, Walter H. (1963). William
of Auxerre's Theology of the Hypostatic Union, Pontifical Institute of Mediaeval Studies.
WILLIAM OF C O N C H E S
B I O G R A P H I C A L D A T A : b. ca. 1080; d. ca. 1154.
TEXTS: Dragmatkon (1567), ed. G. Gratarolus under the title Dialogus de substantias physi-
cis . . . · , Argentorati ( = Strasbourg); Glosae super Platonem (1065), ed. E. Jeaneau, Vrin;
Philosophia mundi, among the works of Bede, in PL 90, and of Honorius, in PL 172;
Philosophia mundi. Book I (1974), ed. G. Maurach, E.J. Brill.
S E C O N D A R Y L I T E R A T U R E : Fredborg, Κ. M. (1973). 'The Dependence of Petrus Helias'
Summa super Priscianum on William of Conches' Close super Priscianum', CIMAGL
11:1—57; Rijk, L. M. de (1967), in De Rijk 1962-7, II (1), 221-8.
Biographies 891
W I L L I A M H E Y T E S B U R Y (Hentisberius, Hentisburius, Hesberi, Hesbri, Tisberius,

Tisbery)
BIOGRAPHICAL DATA: b. before 1313; d. 1372/3. Fellow of Merton College by 1330.
Foundation Fellow of Queen's College in 1340, but soon returned to Merton. Doctor of
theology by 1348.
TEXTS: Tractatus Gulielmi Hentisberi de sensu composito el diviso, Regulae ejusdem cum
Sophismatibus... (1494), Venice; William Heytesbury: On 'Insoluble' Sentences: Chapter One
of His Rules for Solving Sophisms (1979). ed. Paul Vincent Spade, Pontifical Institute of
Mediaeval Studies.
Middle Ages, University of Wisconsin Press; Maier, Anneliese (1952). An der Grenze von
Scholastik und Naturwissenschaft (2nd ed.), Edizioni di Storia e Letteratura; Maierù, Alfonso
(1966). Ί1 "Tractatus de sensu composito et diviso" di Guglielmo Heytesbury', Rivista
critica di Storia della Filosofia 21 =243-63; Maierù, Alfonso (1969). *I1 problema della verità
nelle opere di G. Heytesbury1, Studi Medievali 7:40—74; Murdoch, John (1979).
'Propositional analysis in Fourteenth-Century Natural Philosophy', Synthese 40:126-38;
Weisheipl, James A. (1968). 'Ockham and Some Mertonians', Mediaeval Studies
30:163-213; Weisheipl, James A. (1969). 'Repertorium Mertonense', Mediaeval Studies
31:174-224; Wilson, Curtis (1956). William Heytesbury: Medieval Logic and the Rise of
Mathematkal Physics (Publications in Medieval Science, 3), University of Wisconsin Press;
Wilson, Curtis (1972). 'Heytesbury, William', in Dictionary of Scientific Biography.
WILLIAM DE LA M A R E
BIOGRAPHICAL DATA: d. ca. 1290.
TEXTS: Correctorium Corruptorii, in P. Glorieux, ed. (1927), Les premières polémiques thomistes
(Bibliothèque thomiste, 9), Kain.
SECONDARY LITERATURE: Pelzer, F. (1955). 'Einige ergänzende Angaben zum Leben
und zur Schriften des Wilhelm de la Mare Ο. F. M.', Franziskanische Studien 37:75-80.
WILLIAM O C K H A M (Guillelmus de Ockham, Occam, Hockham)

BIOGRAPHICAL DATA: b. ca. 1285; d. 1347/9. Joined Franciscans, probably subdeacon in
1306. Student of theology at Oxford. Lectured on Sentences there, 1317—19. Completed all
requirements as bachelor before going to London friary in 1320 to await his turn to incept as
master. Never became regent master, partly because there were many waiting to occupy the
Franciscan chair and partly because of the opposition of the chancellor of the university
John Lutterell. Taught and wrote at the Studium generale of the London Custody until 1324.
Probably lived in the same house as Walter Chatton; their running debate is reflected in the
writings of both men. Summoned to Avignon in 1324 to answer charges of heresy. While
in Avignon he became involved in the controversy concerning Franciscan poverty. Fled
Avignon with the Minister General of the order Michael of Cesena in 1328. After being
excommunicated, took refuge in Munich under the protection of Ludwig of Bavaria.
Carried on polemic against Pope John XXII and his successors, writing numerous political
treatises.
TEXTS: Breviloquium de potestate papae (1944: reprinted 1952) in R. Scholz (ed.) Wilhelm von
Ockham als politischer Denker und sein Breviloquium (Schriften des Reichsinstituts für ältere
deutsche Geschichtskunde, 8), K. W. Hiersemann. Also L. Baudry, ed. (1937) (Études de
philosophie médiévale, 24),J. Vrin; De imperatorum etpontificum potestate (1944) in R. Scholz
(ed.) Unbekannte Kirchenpolitische Streitschriften aus der Zeit Ludwigs des Bayern 1327—1334,
Loescher. Also C. K. Brampton, ed. (1927), Clarendon Press; Opera philosophica et theologica
(1967-), ed. Gedeon Gâl, Stephen Brown, et al., Franciscan Institute; Opera plurima
892 Biographies
(1494-6), Lyon. (Reprinted Gregg Press 1962); Opera politica (1956-), ed. J. G. Sikes,
R. F. Bennett, and H. S. Orfler, Manchester University Press; The De Sacramento Altaris of
William of Ockham (1930). T. B. Birch (ed. and tr.), Lutheran Literary Board; The Tractatus
de praedestinatione et de praescientia dei et defuturis contingentibus of William Ockham (1945), ed.
Philotheus Boehner, Franciscan Institute. Also in William Ockham: Predestination, God's
Foreknowledge and Future Contingents (1969). M. M. Adams and N. Kretzmann (trs.),
Appleton-Century-Crofts/Hackett.
S E C O N D A R Y L I T E R A T U R E : Baudry, Léon (1949). Guillaume d'Occam: Sa vie, ses oeuvres,
ses idées sociales et politiques,]. Vrin; Baudry, Léon (1958). Lexique philosophique de Guillaume
d'Occam: Etude des notions fondamentales, P. Léthielleux; Boehner, Philotheus (1958).
Collected Articles on Ockham, ed. E. M. Buytaert, Franciscan Institute; Ghisalbeiti,
Alessandro (1972). Guglielmo di Ockham, Vita e Pensiero; Heynck, Valens (1950).
Ockham-Literatur: 1919—1949', Franziskanische Studien 32:164—83; Junghans, Helmar
(1968). Ockham im Lichte der neueren Forschung, Lutherisches Verlaghaus; Koelmel, W.
(1962). Wilhelm Ockham und seine Kirchenpolitischen Schriften, Ludgerus Verlag; Leff,
Gordon (1975). William of Ockham: The Metamorphosis of Scholastic Discourse, Manchester
University Press/Rowman and Littlefield; McGrade, Arthur S. (1974). The Political
Thought of William of Ockham, Cambridge University Press; Miethke, J. (1969). Ockhams
Weg zur Sozialphilosophie, De Gruyter; Reilly, James P. (1968). Ockham Bibliography:
1950-1967', Franciscan Studies 28:197-214; Weisheipl, James A. (1968). 'Ockham and
Some Mertonians', Mediaeval Studies 30:163-213.
W I L L I A M O F S H E R W O O D (Guilelmus de Shyreswood, Shyreswode, Schirewode,

Syrewude)
BIOGRAPHICAL D A T A : b. ca. 1200/1210; d. between 1266 and 1272. Studied and taught at
Oxford; master at Oxford in 1252. Treasurer of Lincoln from 1254/8 onwards. Rector of
Aylesbury.
T E X T S : De obligationibus, in Romuald Green, ed. (1963), The Logical Treatises De
obligationibus: An Introduction with Critical Texts of William ofSherwood (? ) and Walter Burley,
unpublished doctoral dissertation, Louvain; Introductiones in logicam (1937), ed. Martin
Grabmann, (Sitzungsberichte Bayerischen Akademie der Wissenschaften, Phil.-hist. Abteilung,
Heft 10), Hueber. Also in William of Sherwood's Introduction to Logic (1966), Norman
Kretzmann (tr.), University o f Minnesota Press; Syncategoremata (1941), ed. J. R.
O'Donnell, Mediaeval Studies 3:46-93. Also in William of Sherwood's Treatise on
Syncategorematic Words (1968), Norman Kretzmann (tr.), University o f Minnesota Press.
S E C O N D A R Y L I T E R A T U R E : Braakhuis, H. A. G. (1977). 'The views of William of
Sherwood on some semantical topics and their relation to those of Roger Bacon', Vivarium
15:111—42; Kretzmann, Norman (1967). 'William of Sherwood', in The Encyclopedia of
Philosophy; Malcolm, J. (1971). 'On Grabmann's text of William of Sherwood', Vivarium
9:108-18; Rijk, L. M. de (1976). 'Some thirteenth-century tracts on the game of obliga-
tion', Vivarium 14:26-49. Spade, Paul Vincent and Eleonore Stump (forthcoming), 'The
Treatise on Obligations Attributed to William of Sherwood'; Stump, Eleonore (1980).
'William of Sherwood's Treatise on Obligations', Historiographia Linguistica 7:249-61.
BIBLIOGRAPHY
Note: Because of the length of time required for the preparation of this History, works
published after 1976 could not always be taken into account; and many works published
before the publication of this History are listed here as forthcoming.
For additional bibliographical information on individual philosophers see also the
Biographies.
Abelard, Peter. See Peter Abelard
Ackrill, J. L. (1963), Aristotle's Categories and De Interpretatione. Translated, with notes.
Clarendon Press
Adams, Marilyn McCord (1970). 'Intuitive Cognition, Certainty and Scepticism in
William of Ockham*, Traditio 26:389-98
(1973)· 'Did Ockham Know of Material and Strict Implication? A Reconsideration*,
Franciscan Studies 33: 5-37
(1976). 'Ockham on Identity and Distinction', Franciscan Studies 36:5-74
(1977). Ockham's Nominalism and Unreal Entities', Philosophical Review 86:144-76
(1978). 'Ockham's Theory of Natural Signification', Monist 62:444-59
Adams, Robert M. (1977). 'Middle Knowledge and the Problem of Evil', American
Philosophical Quarterly 14:109-17
Adorno, F. (1954). 'Di alcune orazioni e prefazioni di Lorenzo Valla*, Rinascimento
5:191-225
Agricola, Rudolph (1539). Df inventione dialectka. Cologne
Alan of Lille (1855). Regulae de sacra theologia, PL 210
Albert the Great (1890-9). Opera omnia (38 vols., ed. Augustus Borgnet), Vives
(1890-93). Liber I Priorum analytkorum (vol. 1), Vives
(i890-9b). De anima libri très (vol. 5), Vives
(1890-9C). Liber de causis et processu universitatis (vol. 10), Vives
(i890~9d). Politicorum Aristotelis commentant (vol. 8), Vives
(i890-9e). Super Ethica (vol. 7), Vives
(1899). Summa de homine (vol. 38), Vives
(1951-). Opera omnia (50 vols., ed. Bernhard Geyer, et al.), Institutum Alberti Magni
(1951a). Liber deprincipiis motus processivi (vol. 12), Institutum Alberti Magni
Albert of Saxony (1495). Sophismata, Paris
(1496). Quaestiones super artem veterem, in Expositio aurea et admodum utilis super artem
veterem édita per venerabilem inceptorem fratrem Gullielmum de Ockham cum quaestionibus
Alberti parvi de Saxonia, Bologna
(1497a). Commentarius in Posteriora Aristotelis, Venice
(1497b). Quaestiones super Analytica posteriora, Venice
(1502). Sophismata, Paris
(1518). Perutilis logica, Venice
(1518a). Quaestiones Physicorum, Paris
894 Bibliography
(1518b). Quaestiones in De caelo, Paris

(1522). Perutilis logica, Venice
Albinus (1858). Didascalkus (in Platonis dialogi... vol. 6, ed. C. F. Hermann), Teubner
Alcuin (1851). Dialectica, PL 101
(194t). The Rhetoric of Alcuin and Charlemagne, tr. W. S. Howell, Princeton University
Press
Alexander of Aphrodisias (1883). In Aristotelis Analyticorum priorum librum I commentarium,
ed. M. Wallies (Commentaria in Aristotelem Graeca, II. 1), Reimer
(1891). In Aristotelis Topkorum libros octo commentaria, ed. M. Wallies (Commentaria in
Aristotelem Graeca, II, 2), Reimer
(1893). Defato, ed. I. Bruns (Commentaria in Aristotelem Graeca, Suppl. II, 2),
Reimer
Alexander of Hales (1924-48). Summa theologica (4 vols, in 5), Collegium S.
Bonaventurae
Alexander of St Elpidius (1969). Tractatus de Ecclesiastica Potestate i n j . T. Rocaberti (ed.)
Bibliotheca Maxima Pontificia, 2, Akademische Druck- und Verlagsanstalt
Algazel (1933). Metaphysics. A Medieval Translation, ed. J. T. Muckle, C.S.B., Pontifical
Institute of Mediaeval Studies
Allan, D.J. (1950). 'Mediaeval Versions of Aristotle, De Caelo, and the Commentary of
Simplicius', Mediaeval and Renaissance Studies 2:82-120
Allen, J. W. (1957). A History ofPolitkal Thought in the Sixteenth Century, revised ed.,
Methuen
Allers, Rudolf (1941). 'The Intellectual Cognition of Particulars', The Thomist 3 : 9 5 - i 6 3
Allmand, C. T., ed. (1973). Society at War, the Experience of England and France during the
Hundred Years War, Oliver and Boyd
ed. (1976). War, Literature and Politics in the Late Middle Ages, Liverpool University Press
Alluntis, F. (1965). 'Demonstrability and Demonstration of the Existence of God', in
J. K. Ryan and B. M. Bonansea (eds.) John Duns Scotus 1263-1963, Catholic
University o f America
Alluntis, F., and Wolter, A. (1975)· John Duns Scotus. God and Creatures: The Quodlibetal
Questions, Princeton University Press
Almain, Jacques (1611). Expositio de Suprema Potestate Ecclesiastica et Laka, circa
Quaestionum Decisiones Magistri Guillermi de Ockham super Potestate Summi Pontifias
in M. Goldast (ed.) Monorchia Sancti Romani Imperii, 1:588-647, Conrad Biermann.
(Reprinted i960 Akademische Druck- und Verlagsanstalt) (Also in L. Du Pin (ed.)
Jean Gerson, Opera Omnia, 2: cols. 1013—1120, Sumptibus Societatis Antwerp)
(1525). Moralia ...et Libellus de Auctoritate Ecclesiae, Claudius Chevallon. (LibeiÎMj also
in L. Du Pin (ed.) Jean Gerson, Opera Omnia, 2: cols. 976-1021, Sumptibus
Societatis Antwerp)
(1706). Quaestio Resumptiva, De Dominio Naturali, Civili et Ecclesiastico in L. Du Pin
(ed.) Jean Gerson, Opera Omnia, 2: cols. 961-76, Sumptibus Societatis Antwerp
Altschul, Michael (1971). 'Kingship, government, and politics in the middle ages: some
recent studies', Medievalia et Humanistica, N.S. 2:133—52
Alvarus Pelagius (1969). De Planctu Ecclesiae in J. T. Rocaberti (ed.) Bibliotheca Maxima
Pontificia, vol. 3, Akademische Druck- und Verlagsanstalt
Alvarus Thomas (1509). Liber de triplici motu. Paris
Ammonius (1&95). In Aristotelis Categorias commentarius, ed. A. Busse (Commentaria in
Aristotelem Graeca IV.4), Reimer
(1897). In Aristotelis De Interpretatione commentarius, ed. A. Busse (Commentaria in
Aristotelem Graeca IV. 5), Reimer
Bibliography 895
Anawati, Georges C. (1974). Études de philosophie musulmane (Études musulmanes,

No. 15), Vrin
André de Neufchâteau (1514). In primum librum Sententiarum, Paris
Andrés, Teodoro de (1969). El nominalismo de Guillermo de Ockham comofilosofia del
lenguaje (Biblioteca hispânica de filosofia), Editorial Gredos
Angelelli, I. 1970. 'The Techniques of Disputation in the History of Logic'. Journal of
Philosophy 67:800-15
(1972). 'Franciscus Sebastiani's Logica (1791)', Journal of the History of Philosophy
10:76-82
Anonymous (14th c). Tractatus de motu locali difformi, Cambridge, Gonville & Caius MS
499/268, ff. 2I3 r b -2I5 r b
Anonymous (161 la) Disputatio inter Militem et Clericum in M. Goldast (ed.) Monorchia
Sancti Romani Imperii, 1:13-18; Conrad Biermann (Reprinted i960 Akademische
Druck- und Verlagsanstalt)
Anonymous (161 lb). Somnium Viridarii de lurisdictione Regia et Sacerdotali in M. Goldast
(ed.) Monorchia Sancti Romani Imperii, ι : 58-229; Conrad Biermann (Reprinted
i960 Akademische Druck- und Verlagsanstalt)
Anonymous (1834). "Lettre écrite de Rome à un professeur de philosophie'. Journal
historique et littéraire ι : 541-51
Anonymous (1902). Vita beati Joannis Psichaitae, in Ven, P. van den 'La vie grecque de
S. Jean le Psichaïte', Le muséon. Études philologiques et religieuses. N.S. 3:97-125
Anonymous (1929). Anonymi logica et quadrivium cum scholiis antiquis, ed. J. L. Heiberg
(Det Kongelige Danske Videnskabernes Selskab. Historisk-filologiske Meddelelser
XV.i)
Anonymous (1940a). Quoniam ignoratis communibus in Grabmann 1940, pp. 27-9
(excerpts)
Anonymous (1940b). Sophismata logicalia in Grabmann 1940, pp. 50-1 (excerpts)
Anonymous (1940c). Sophismata Parisius determinata in Grabmann 1940, pp. 33—41
(excerpts); supplemented in Braakhuis 1979, p. 8i
Anonymous (1954). Rhetorica ad Herennium, tr. Harry Caplan (Loeb Classical Library),
Harvard University Press
Anonymous (1962-73). Ars Emmerana in De Rijk 1962-7, II (2), 143-74
Anonymous (1962-7^. Ars Meliduna in De Rijk 1962-7, II (1), 292—390 (excerpts)
Anonymous (1ÇXÎ2-7C). Dialectica Monacensis in De Rijk 1962-7, II (2), 453-638
Anonymous (io62-7d). Quaestiones Victorinae in De Rijk 1962-7, II (2), 731-69
Anonymous (i962-7e). Tractatus Anagnini in De Rijk 1962-7, II (2), 215-332
Anonymous (1962-7Ο. Tractatus de univocatione Monacensis in De Rijk 1962-7, II (2),
33-51
Anonymous (1977). Tractatus implicilarum in Giusberti 1977
Anonymous (1979). Sincategoreumata Monacensia in Braakhuis 1979,1. PP· 95-104
Anonymous of York (1897). Tractatus Eboracenses in Monumenta Germaniae Historica
(Libelli de Lite) 3:645-87. Impensis Bibliopolii Hahniani
(1977). Der Codex 415 des Corpus Christi College Cambridge, ed. Karl Pellens and Ruth
Nineham, Steiner
Anscombe, G. E. M. (15)65). 'Thought and Action in Aristotle' in Renford Bambrough
(ed.) New Essays on Plato and Aristotle, Routledge and Kegan Paul
Anscombe, G. E. M., and P. T. Geach (1961). Three Philosophers: Aristotle, Aquinas,
Frege, Cornell University Press
Anselm of Canterbury (1936). Ein neues unvollendetes Werk des hl. Anselm von Canterbury,
ed. F. S. Schmitt (BGPM, XXXIII.3), Aschendorff
896 Bibliography
(1946). Opera omnia, ed. F. S. Schmitt, Nelson

(1946a). De libertate arbitrii in F. S. Schmitt (ed.) Opera omnia, vol. I, Nelson
(1946b). De veritate in F. S. Schmitt (ed.) Opera omnia, vol. 1, Nelson
(1946c). De casu diaboli in F. S. Schmitt (ed.) Opera omnia, vol. 1, Nelson
(r946d). De ccmcordia praescientiae et praedestinationis etgratiae Dei cum libero arbitrio in
F. S. Schmitt (ed.) Opera omnia, vol. 2, Nelson
(1946e). De incamatione verbi in F. S. Schmitt (ed.) Opera omnia, vol. 2, Nelson
Apollonius Dyscolus (1878). Scripta minora, cd. R . Schneider (Grammatici Graeci II.i),
Teubner
(1910a). De syntaxi, ed. G. Uhlig (Grammatici Graeci II.2), Teubner
(1910b). Fragmenta, ed. R . Schneider (Grammatici Graeci II.3), Teubner
Apuleius (1908). De philosophia libri, ed. P. Thomas (Apulei opera quae supersunt, vol.
Ill), Teubner
Aquinas, Thomas, see Thomas Aquinas
Arbus, M . - R . (1957), 'Le droit romain dans l'oeuvre de saint Thomas', Revue Thomiste
57:325-49
[Aristoteles Latinus] (1939). Codices: Pars prior, ed. G. Lacombe ei al., La Libreria dello
Stato
(1955). Codices: Pars posterior, ed. G. Lacombe et al, Cambridge University Press
(1957). VII.2, Physica: Translatio Vaticana, ed. A. Mansion, Desclée de Brouwer
(1961a). Codices: Supplementa altera, ed. L. Minio-Paluello, Desclée de Brouwer
(1961b). 1.1—5, Categoriae vel Praedkamenta, ed. L. Minio-Paluello, Desclée de
Brouwer
(1961c). XXLX.i, Politica: Translatio imperfecta, ed. P. Michaud-Quantin, Desclée de
Brouwer
(1962). III.1-4, Analytica priora, ed. L. Minio-Paluello, Desclée de Brouwer
(1065a). II.1-2, De interpretatione vel Periermenias, ed. L. Minio-Paluello and
G. Verbeke, Desclée de Brouwer
(1965b). X I . 1 - 2 , De mundo, ed. W . L. Lorimer, rev. L. Minio-Paluello, Desclée de
Brouwer
(1966a). I.6—7, Categoriarum supplementa: Porphyrii Isagoge et Liber sex principiorum, ed.
L. Minio-Paluello, Desclée de Brouwer
(1966b). XVII.2, De generatione animalium, ed. H. J. Drossaart-Lulofs, Desclée de
Brouwer
(1968a). IV.1-4, Analytica posteriora, ed. L. Minio-Paluello and B. G. Dod, Desclée de
Brouwer
(1968b). X X X I I I . 1 - 2 , De arte poetica, ed. L. Minio-Paluello, Desclée de Brouwer
(1969). V . 1 - 3 , Topica, ed. L. Minio-Paluello, Desclée de Brouwer
(1970). X X V . i - i a , Metaphysica: Translatio lacobi ('Vetustissima') et Translatio composita
('Vetus'), ed. G. Vuillemin-Diem, Desclée de Brouwer
(1972-4). X X V I . 1 - 3 , Ethka Nicomachea, ed. R . A. Gauthier, Brill/Desclée de
Brouwer
(1975). V I . 1 - 3 , De sophisticis elenchis, ed. B. G. Dod, Brill/Desclée de Brouwer
(1976). XXV.2, Metaphysica: Translatio anonyma ('Media1), ed. G. Vuillemin-Diem,
Brill
Armstrong, A. H . , ed. (1967). The Cambridge History of Later Greek and Early Medieval
Philosophy, Cambridge University Press
Armstrong, R . A. (1966). Primary and Secondary Precepts in Thomistic Natural Law
Teaching, Martinus Nijhoff
Arway, R . J . (1962). Ά Half Century of Research on Godfrey of Fontaines', The New
Scholasticism 36:192-218
Bibliography 897
Ashworth, E.J. (1972). 'The Treatment of Semantic Paradoxes from 1400 to 1700',
Noire Dame Journal of Formal Logic 13:34-52
('973)· 'The Doctrine of Exponibilia in the Fifteenth and Sixteenth Centuries',
Vivarium 11:137-67
(1974a). Language and Logk in the Post-medieval Period (Synthese Historical Library, 12),
Reidel
(1974b). '"For Riding is Required a Horse": A Problem of Meaning and Reference
in Late Fifteenth- and Early Sixteenth-Century Logic', Vivarium 12:94-123
(1976a). " Ί Promise You a Horse": A Second Problem of Meaning and Reference in
Late Fifteenth- and Early Sixteenth-Century Logic', I, Vivarium 14:62-79, II, Ibid.,
14:149-55
(1976b). 'Agostino Nifo's Reinterpretation of Medieval Logic', Rivistct critica di storia
dellafilosofia 31:355-74
(1977a). 'Chimeras and Imaginary Objects: A Study in the Post-medieval Theory of
Signification', Vivarium 15:57-79
(1977b). 'Thomas Bricot (d. 1516) and the Liar Paradox', Journal of the History of
(1978). The Tradition of Medieval Logic and Speculative Grammar from Anselm to the End
of the Seventeenth Century: A Bibliography from 1836 Onwards (Subsidia Mediaevalia,
9), Pontifical Institute of Mediaeval Studies
(1978a). 'Multiple Quantification and the Use of Special Quantifiers in Early Sixteenth
Century Logic', Notre Dame Journal of Formal Logic 19:599-613
(1979). 'The "Libelli Sophistarum" and the Use of Medieval Logic Texts at Oxford and
Cambridge in the Early Sixteenth Century', Vivarium 17:134-58
(forthcoming). "Theories of the Proposition: Some Early Sixteenth-Century
Discussions', Franciscan Studies
Atti del XVII Convcgno di Studi, Todi (1978). Le scuole degli Ordini Mendicanti (secoli
XIII-XIV), Accademia Tudertina
Aubert, Jean Marie (1955). Le droit romain dans l'oeuvre de Saint Thomas, Vrin
Augustine (1845a). InJohannis evangelium tractatus, PL 35
(1845b). De nuptiis et concupiscentia, PL 44
(1877). De magistro, PL 32
(1894). De Genest ad litteram, ed. J. Zycha (CSEL, 28), Tempsky
(1895). Epistulae, ed. A. Goldbacher (CSEL, 44), Tempsky
(1899-1900). De civitate Dei, ed. E. Hoffmann (CSEL, 40), Tempsky
(1922). Contra academkos, ed. P. Knöll (CSEL, 63), Tempsky
(1950). Against the Academics, tr. J. J. O'Meara (Ancient Christian Writers, 12),
Newman
(1955). De civitate Dei, ed. B. Domhart and A. Kalb (Corpus Christianorum. Series
Latina, 47-8), Brepols
(1956). De libero arbitrio (CSEL, 74), Hoelder-Pichler-Tempsky
(1961). De magistro, ed. G. Weigel (CSEL, 77), Hoelder-Pichler-Tempsky
(1963a). De doctrina Christiana (CSEL, 80), Hoelder-Pichler-Tempsky
(1963b). The Trinity, tr. S. McKenna (The Fathers of the Church, 45), Catholic
University of America
(1968). De Trinitate, ed. W.J. Mountain (Corpus Christianorum. Series Latina, 50-
50A), Brepols
(1975). Df dialectica, tr. wth introduction and notes by B. D.Jackson from the text
newly edited by J. Pinborg (Synthese Historical Library, 16), Reidel
Augustinus Triumphus (1479). Summa de ecclesiastica potestate, Rome
(1903). Tractatus Brevis de Duplici Potestate Prelatorum el Laicorum in R. Scholz (ed.) Die
898 Bibliography
Publizistik zur Zeit Philipps des Schonen und Bonifaz' VIII, Enke
Aureoli, Peter, see Peter Aureoli
Averroes (1562-743). Media expositio in libros Priorum resolutoriorum (Aristotelis opera cum
Averrois commentariis, Li), Venice
(i502-74b). Quaesita octo in libros Priorum analyticorum (Aristotelis opera cum Averrois
commentariis, 1.2b and 3), Venice
(1562-74C). Media expositio Topicorum Aristotelis (Aristotelis opera cum Averrois
commentariis, I.3), Venice
(i562-74d). Commentary on the Physics (Aristotelis opera cum Averrois commentariis,
IV), Venice
(1562-740). In libros De coelo commentant (Aristotelis opera cum Averrois com-
mentariis, V), Venice
(1562-741"). Commentary on the Metaphysics (Aristotelis opera cum Averrois com-
mentariis, VIII), Venice
(1953). Commentarium magnum in Aristotelis De anima, ed. F. S. Crawford (Corpus
commentariorum Averrois in Aristotelem), Mediaeval Academy of America
(1954). Tahafut al-Tahafitt, tr. S. Van den Bergh, Luzac
(1974). Averroes on Plato's Republic, tr. with introduction and notes by R . Lerner,
Cornell University Press
Avesani, A. (1970). 'La professione dell' "umanista" nel Cinquecento', Italia medioevale e
umanistica 13:205-32
Avicenna (1508). Avicennaeperhypateticiphilosophi ac medicorumfacileprimi opera...,
Venice (Reprinted Minerva and Bibliothèque S.J. de Louvain 1961)
[Avicenna Latinus] (1968) Liber De anima ÎV-V, ed. S. Van Riet, Editions
Orientalistes/Brill
(1972). Liber De anima I—III, ed. S. Van Riet, Ε. Peeters/Brill
(1977). Liber de philosophia prima, ed. S. Van Riet, E. Peeters
Bacon, Roger, see Roger Bacon

Baebler.J. J. (1885). Beitrage zu einer Geschichte der lateinischen Grammatik im Mittelalter,
Buchhandlung des Waisenhauses
Baeumker, Clemens (1895). Avencebrolis (Ihn Gebirol) Fons vitae ex Arabico in Latinum
translatus ab Johanne Hispano et Dominico Gundissalino (BGPM, I), Aschendorff
(1913). 'Die Stellung des Alfred von Sareshel und seiner Schrift " D e Motu Cordis" in
der Wissenschaft des beginnenden XIII. Jahrhunderts', Sitzungsberichte der königliche
Bayerische Akademie der Wissenschaften, Philos.-philol. u. hist. Klasse 9 : 3 5 - 4 0
(1923). Des Alfred von Sareshel (Alfredus Anglicanus) Schrift 'De Motu Cordis' (BGPM,
XXIII), Aschendorff
Bainton, R. H. (1946). 'The Early Church and War', Harvard Theological Review
39:189-212
Baldanza, G. (1970). Ί1 Problema della filosofia in "Optatam totius"', Seminarium n.s.
10:1-18
Baldus de Ubaldis (1608-9). Consilia (2 vols.), Venice
Balic, C. (1931). 'Une question inédite d e j . Duns Scot sur la volonté', Recherches de
théologie ancienne et médiévale 3:191-208
(1965). "The Life and Works o f John Duns Scotus', i n j . K. Ryan and B. M. Bonansea
(ed.), John Duns Scotus 1263-1963, Catholic University o f America
Bânez, Dominic (1934). Scholastica commentaria in primam partem Summae Theologiae S.
Thomae Aquinatis, ed. L. Urbano, Editorial F.E.D.A. (Reprinted Brown Reprint
Library 1966)
Bibliography 899
Barcia Trelles, Camilo (1933). 'Francisco Suârez (1548-1617)', Académie de droit inter-
national: recueil des cours 43: 385-553
Barnes, Jonathan (1972). The Ontological Argument, Macmillan and St Martin's Press
(i(>75). Aristotle's Posterior Analytics, tr. with notes by Jonathan Barnes, Clarendon Press
Barnes, Jonathan, Malcolm Schofield, and Richard Sorabji (1975—9). Articles on Aristotle
(4 vols.), Duckworth
Barozzi, L. and R. Sabbadini (1891). Studi sui Panormita e sui Valla (R. Istituto di studi
superiori pratici e di perfezionamento in Firenze), Le Monier
Barth, T. (1939). 'De fundamento univocationis apud Joannem Duns Scotum',
Antcmianum 14:181-206, 277-98, 373-92
(1953)· 'De univocationis entis scotisticae intentione principali necnon valore critico',
Antcmianum 28 ; 72-110
(1965). 'Being, Univocity, and Analogy According to Duns Scotus', i n j . Κ. Ryan
and Β. Μ. Bonansea (ed.), John Duns Scotus, 1263-1963, Catholic University of
America
Batllori, M. (1944). Baltasar Masdeu y el neoscolastkismo italiano, Balmesiana
Battaglia, F. (1965). Metafiska religione epolitico net pensiero di Nkolo da Cusa, Patron
Baudry, Léon (1934). 'Les rapports de Guillaume d'Occam et de Walter Burleigh',
Archives d'histoire doctrinale et littéraire du moyen âge 9:155—73
(1943). 'Gauthier de Chatton et son commentaire des Sentences', Archives d'histoire
doctrinale et littéraire du moyen âge 18:337-69
(1949). Guillaume d'Occam: Sa vie, ses oeuvres, ses idées sociales et politiques, vol. 1
(Études de Philosophie Médiévale, 39), Vrin
(1950). La Querelle des futurs contingents {Louvain 1463-1473) (Études de Philosophie
Médiévale, 38), Vrin
(1958). Lexique Philosophique de Guillaume d'Ockham, Lethielleux
Baur, Ludwig (1903). Dominicus Gundissalinus: De divisione philosophiae (BGPM, IV.2—3),
Aschendorff
(1912). DU philosophische Werke des Robert Grosseteste, Bischofs von Lincoln (BGPM,
IX), Aschendorff
(1917). Die Philosophie des Robert Grosseteste (BGPM XVIII), Aschendorff
Bayley, C. C. (1949). 'Pivotai concepts in the political philosophy of William of
Ockham', Journal of the History of Ideas 10:199-218
Bazân, Bernardo Carlos (1972). Siger de Brabant: Quaestiones in tertium de anima. De anima
intellectiva, De aeternitate mundi. Édition critique (Philosophes médiévaux, Tome 13),
Publications Universitaires and Béatrice-Nauwelaerts
(1973). 'La eternidad y la contingencia del intelecto en Sigerio de Brabante',
Philosophia (Mendoza), No. 39:63-84
(1974). 'Le dialogue philosophique entre Siger de Brabant et Thomas d'Aquin. A
Propos d'un ouvrage récent de E. H. Weber O.P.', Revue philosophique de Louvain
72:53-155
(15)75). 'Ea union entre el intelecto separado y los individuos, segûn Sigerio de
Brabante', Patristica et mediaevalia (Buenos Aires), 1:5-35
Becker, A. (1933). Die Aristotelische Theorie der Möglichkeitsschlüsse, Junker und
Dünnhaupt Verlag
Becker, O. (1952). Untersuchungen über den Modalkalkül, Westkulturverlag Anton Hain
Becker-Freyseng, A. (1938). Dif Vorgeschichte des philosophischen Terminus "contingens"
(Quellen und Studien zur Geschichte und Kultur des Mittelalters, Reihe D, Heft 7),
Selbstverlag F. Bilabel
Beha, H. (1960-1). 'Matthew of Aquasparta's Theory of Cognition', Franciscan Studies
900 Bibliography
20:161-204; 2 1 : 1 - 7 9 , 3 8 3 - 4 6 5
Bellarmine, Robert
(1870). De Summo Pontifice in Justin Fèvre (ed.) Politiani Opera omnia, 11449-615 and
2:5-167 (Reprinted Minerva 1965)
(1870-4). De Conciliis in Justin Fèvre (ed.) Politiani Opera omnia, 2:187-407
(Reprinted Minerva 1965)
(1950). TJie Poller of the Pope in Temporal Affairs, against William Barclay, tr. and ed.
George Albert Moore, 2nd ed., Country Dollar Press
(1951). Extracts on Politics and Government from the Supreme Pontiff, tr. and ed. George
Albert Moore, Country Dollar Press
Behnond, S. (1928). 'Le mechanisme de la connaissance après Pierre Olieu, dit Olivi', La
France Franciscaine 12:291-401; 463-87
(1930). 'L'Intellect actif après Jean Duns Scot', Revue de philosophie 30:31-54
Beltran de Heredia, V. (1968). Domingo Banez y las controversias sobre la gracia: textos y
documentos, Consejo superior de Investigaciones Cientificas
Bendiek, J. (1952). 'Die Lehre von den Konsequenzen bei Pseudo-Scotus', Franziska-
nische Studien 34:205-34.
Berges, W . (1938). Dif Fürstenspiegel des hohen und späten Mittelalters, K. W. Hiersemann
Bergh, B. (1978). Paleography and Textual Criticism (Scripta Minora Regiae Societatis
Humaniorum litterarum Lundensis 1979-80:2), CWK Gleerup
Bergmann, Gustav (1964). Logic and Reality, University of Wisconsin Press
Bernard o f Clairvaux (1963). De Consideratione ad Eugenium Papam i n j . Leclercq and
H. Rochais (eds.) S. Bernardi Opera, 3 :381-493, Editiones Cistercienses
(1976). Five Books on Consideration: Advice to a Pope, ed. John D. Anderson and
Elizabeth T. Kennan (Cistercian Fathers 13, Works of Bernard of Clairvaux 13),
Cistercian Publications
Bernard Morisan (1625). Commentarii et disputationes in libros logicos, physicos, et ethicos
Aristotelis, Frankfurt
Bertola, Ermencgildo (1967). Έ esistito un avicennismo latino nel medioevo?', Sophia
3S:3i8-334
(1971). Έ esistito un avicennismo latino nel medieoevo?', Sophia 39:278—320
Berube, C. (1964). La connaissance de l'individuel au moyen âge. Presse Universitaire de
Montréal
Bettoni, Efrem (1954)· Il processo astrattivo nella concezione di Enrico di Gand, Vita e
Pensiero
(1961). Duns Scotus, tr. B. M. Bonansea, Catholic University of America
Bird, Otto (1959). 'The logical interest of the topics as seen in Abelard', The Modem
Schoolman, 3 7 : 5 3 - 7
(i960). 'The Formalizing of Topics in Mediaeval Logic', Notre Dame Journal of Formal
Logic 1:138-49
(1961). 'Topic and Consequence in Ockham's Logic', Notre Dame Journal of Formal
Logic, 2 : 6 5 - 7 8
(1962a). "The Tradition o f the Logical Topics: Aristotle to Ockham', Journal of the
History of Ideas 23:307-23
(1962b). 'What Peirce Means by Leading Principles', Nofrf Dame Journal of Formal
Logic 3 : 1 7 5 - 8
Black, Anthony (1970). Monarchy and Community: Political Ideas in the Later Conciliar
Controversy 1430-1430, Cambridge University Press
Blackley, F. D. (1967). 'Ralph Strode', New Catholic Encyclopedia, 12:71
Blake, Ralph, Curt Ducasse, and Edward Madden (1966). Theories of Scientific Method:
Bibliography 901
The Renaissance through the Nineteenth Century, University of Washington Press

Blanche, F.-A. (1934). 'La théorie de l'abstraction chez Saint Thomas d*Aquin' in
Mélanges Thomûtes (Bibliothèque Thomiste, vol. 3), Vrin
Blanche, Robert (1970). La Logique et son Histoire: d'Aristote à Russell, Armand Colin
Blemmydes, Nicephorus. see Nicephorus Blemmydes
Bliemetzrieder, F. (1919). Anseimus von Laon systematische Sentenzen (BGPM, XVIII,
2-3), Aschendorff
Bloch, M. (1973). The Royal Touch: Sacred Monarchy and Scrofila in England and France,
tr. J. E. Anderson, Routledge and Kegan Paul
Blondel, Maurice (1893). L'Action, Alcan
Bloomfield, Morton W. (1970). Review of Marcia L. Colish, The Mirror of Language: A
Study in the Medieval Theory of Knowledge, Yale University Press, 1968, in Speculum
45:119-22
Bochenski, I. M. (1938). 'De Consequcntiis Scholasticorum Earumque Origine',
Angelicum 15:92-109
(1947a). La Logique de Théophraste (Collectanea Friburgensia, n.s. 32), Librairie de
l'Université Fribourg en Suisse
ed. (1947b). Pf tri Hispani Summulae logkales, Marietti
(1951). Ancient Formal Logk (Studies in Logic and the Foundations of Mathematics),
North-Holland
(1956). Formale Logik (Orbis Academicus III, 2), Karl Alber
(1961). History of Formal Logic, tr. and ed. Ivo Thomas, University of Notre Dame
Press
Bodin, Jean (1641). Df Republica libri sex, Frankfurt
(1962). The Six Bookes of a Commonweale, Richard Knolles' trans, of 1606 edn. with
introd. and notes by Kenneth Douglas McRae, Harvard University Press
(1975). Colloquium of the Seven about Secrets of the Sublime, tr. Marion L. D. Kuntz,
Princeton University Press
Boehner, Philotheus (1943a). Ockham's political ideas', Review ofPolitks 5 :426-87
(1943b). "The Notitia Intuitiva of Non-Existents According to William Ockham',
Traditio 1, 245-75
(1946). 'The Realistic Conceptualism of William Ockham', Traditio 4:307-35
(1951a). 'Bemerkungen zur Geschichte der De Morganschen Gesetze in der
Scholastik', Archiv für Philosophie 4:113-46
(1951b). 'Does Ockham Know of Material Implication?', Franciscan Studies 11: 203—50
(i952). Medieval Logic: An Outline of Its Development from 1230-c. 1400, Manchester
University Press
ed. (1955). Walter Burleigh: De puritate artis logicae Tractatus longior with a revised edition
of the Tractatus brevior (Franciscan Institute Publications, Text Series, 9), The
Franciscan Institute
(1957). Ockham: Philosophical Writings, Nelson
(1958). Collected Articles on Ockham, ed. Ε. M. Buytaert (Franciscan Institute
Publications, Philosophy Series, 12), The Franciscan Institute
(1965). 'History of Logic III - Scholastic Logic', Encyclopedia Britannica 14:226-31
Boethius (1847). De arithmetica, PL 63
(1860a). De differentiis topicu libri quattuor, PL 64
(1860b). De syllogismo categorico, PL 64
(1860c). Df trinitate, PL 64
(i86od). In Topica Ciceronis Commentariorum libri sex, PL 64
(1860e). Introductio ad syllogismos categoricos, PL 64
902 Bibliography
(ι 86of). Lifefr de divisione, PL 64

(1860g). In Categorias Aristotelis libri quaituor, PL 64
(i86oh). Commentaria in Porphyrium, PL 64
(18601). Df unitate Trinitatis cum Gilberti Porretae commentario, PL 64
(1877). Commentarii in librum Aristotelis ΠΕΡΙ ΕΡΜΗΝΕΙΑΣ pars prior versionem
continuam et primam editionem continens, ed. Carl Meiser, Teubner
(1880). Commentarii in librum Aristotelis ΠΕΡΙ ΕΡΜΗΝΕΙΑΣ pars posterior secundum
editionem et indices continens, ed. Carl Meiser, Teubner
(1906). In Isagogen Porphyrii Commenta, ed. S. Brandt (Corpus Scriptorum Eccle-
siasticorum Latinorum, 48), F. Tempsky
(1918). De trinitate in Tractates, De consolatione philosophiae, ed. H. F. Stewart and Ε. K.
Rand (Loeb Classical Library), William Heinemann, Ltd
(1957). Philosophiae consolatio, ed. L. Bieler (Corpus Christianorum, Series Latina, 94),
Brepols
(1969). De hypotheticis syllogismis, ed. L. Obertello, Paideia éditrice
(1978). Boethius's De topicis differentiis, tr. Eleonore Stump, with introduction, com-
mentary, and supplementary essays, Cornell University Press
Boethius of Dacia (1936). De Summo Bono sive De Vita Philosophiae in M. Grabmann,
Mittelalterliches Geistesleben, 2:209-16, M. Hüber
(1969). Quaestiones super Priscianum maiorem ed.J. Pinborg and H. Roos (Corpus
Philosophorum Danicorum Medii Aevi 4), Gad
(1976). De aeternitate mundi, ed. N. J. Green-Pedersen in Boethii Daci Opera (Corpus
Philosophorum Danicorum Medii Aevi, 6), Gad
(1976). Quaestiones sup er librum Topicorum, ed. N . J . Green-Pedersen and J. Pinborg
(Corpus Philosophorum Danicorum Medii Aevi VI, 1), Gad
Boggess, W. F. (1971). 'Hermannus Alemannus' Rhetorical Translations', Viator 2:227-
50
Boh, Ivan (1962). Ά Study in Burleigh: Tractatus de Regulis Generalibus
Consequentiarum', Notre Dame Journal of Formal Logk 2:83-101
(1963a). 'Burleigh: On Conditional Hypothetical Propositions', Franciscan Studies
23:4-67
(1963b). 'Walter Burleigh's Hypothetical Syllogistic', Noire Dame Journal of Formal
Logic 4:241-69
(1964). 'An Examination of Some Proofs in Burleigh's Propositional Logic', The New
(1965). 'Paul of Pergula on Suppositions and Consequences', Franciscan Studies 25:30-
89
(15)66). 'Propositional Connectives, Supposition and Consequence in Paul of Pergula',
Notre Dame Journal of Formal Logic 7:109-28
(1977). "The "Conditionatim"-Chuse: One of the Problems of Existential Import in the
History of Logic', Notre Dame Journal of Formal Logic 18:459-66
Boler, J. F. (1963). 'Abailard and the Problem of Universels', Journal of the History of
Philosophy 1137-51
(j-973)· Ockham on Intuition', Journal of the History of Philosophy 11:95-106
(1976). Ockham on Evident Cognition', Franciscan Studies 36:85-98
Bonald, L. G. A. de (1830). Recherches philosophiques sur les premiers objets des connaissances
morales (2 vols.), Vanryckegem-Hovaere
Bonansea, Bernadino M. (1965) 'Duns Scotus's Voluntarism' inj. K. Ryan and Β. M.
Bonansea (eds.) John Duns Scotus 1265—1963. Catholic University of America
Bonaventure (1882-1902). Opera omnia (10 vols.). Collegium S. Bonaventurae
Bibliography 903
(ι882-1902a). Commentarius in quattuor libros Sententiarum Petri Lombardi (Opera

omnia, I—IV), Collegium S. Bonaventurae
(1911). Tria opuscula: Breviloquium, Itinerarium mentis in Deum, De reductione artium ad
theotogiam, Collegium S. Bonaventurae
Borak, H. (1964). 'Principia doctrinae politicae apud S. Bonaventuram', Laurentianum
5:301-20,487-523
Bos, E. P. (1976). John Buridan and Marsilius of Inghen on Consequences', in Jan
Pinborg (ed.), The Logic of John Buridan (Opuscula Graecolatina, Supplementa
Musei Tusculani, 9) Museum Tusculanum
Bosley, R. (1978). 'In Support of an Interpretation of On Int. 9', Ajatus 37:29-40
Bottin, Francesco (1973). 'Analisi linguistica e fisica Aristotelica nei "Sophismata" di
Richard Kilmyngton' in Carlo Giacon (ed.) Filosofia Politico e altri saggi (Universiri
di Padova, Pubblicazioni dell' Istituto di Storia della Filosofia e del Centro per
Ricerche di Filosofia Medioevale, n.s. 14), Antenore
(1973a). 'L'Opinio de Insolubilibus di Richard Kilmington', Rivista Critica di storia
dellafilosofia28:568-90
(1974). 'Un testo fondamentale nell' ambito della "Nuova Fisica" di Oxford: I
Sophismata di Richard Kilmington', Miscellanea Mediaevalia 9:201—5
(1976). Lf antinomie semantiche nella logica médiévale (Pubblicazioni dell' Istituto di
Storia della Filosofia e del Centro per Ricerche di Filosofia Medioevale, n.s. 23),
Antenore.
Bougerol, J. Guy (1964). Introduction to the Works of Bonaventure, St Anthony Guild Press
Bouillard, H. (1949). 'L' Intention fondamentale de Maurice Blondel et la théologie'.
Recherches de Science Religieuse 36:321-402
Bourgeois, R. (1936). 'La théorie de la connaissance intellectuelle chez Henri de Gand",
Revue de Philosophie 36:238-59
Bourke, V.J. (1964). TTte Essential Augustine, The New American Library
Bowman, Leonard J. (1972-3). 'The Development of the Doctrine of the Agent Intel-
lect in the Franciscan School of the Thirteenth Century', The Modern Schoolman
50:251-79
Boyer, Charles (1925). 'Réflexions sur la connaissance sensible selon Saint Thomas',
Archives de philosophie, vol. 3, Cahier 2, Gabriel Beauchesne
Boyer, M. (1835). Défense de l'ordre social contre le carbonarisme modem avec un jugement sur
M. de La Mennais considéré comme écrivain et une dissertation sur le romantisme, Le Clerc
Braakhuis, H. A. G. (1967). 'The Second Tract on Insolubilia Found in Paris, B.N. Lat.
16.617: An Edition of the Text with an Analysis of Its Contents', Vivarium 5:111—
45
(1977)· 'The Views of William of Sherwood on Some Semantical Topics and Their
Relation to Those of Roger Bacon*, Vivarium 15:111 -42
(1978). /. De Ontwikkeling van de théorie van de syncategoremata tot aan de i3de eeuw; II.
De I3de eeuwse syncategoremata-tractaten; III. Enkele aspecten van de ontwikkeling van de
théorie van de syncategoremata in de I3de eeuw, Typescript; preliminary version
(1978a). Preliminary editions of Robert Bacon (?), Sincategoreumata, and Henry of
Ghent, Sincathegoreumata, Typescript; preliminary version
(1979). De ljdf eeuwse Tractaten over syncategorematische Termen (Deel I: Inleidende
Studie; Deel II: Uitgave van Nicolaas van Parijs' Sincategoreumata), Krips Repro
Meppel
Brady, Ignatius (1950). 'Law in the Summa Fratris Alexandri', Proceedings of the American
Catholic Philosophical Association 24:133-47
(1967). 'Bonaventure, St', New Catholic Encyclopedia 21658-64
904 Bibliography
(!975)· 'The Opera Omnia of S Bonaventure Revisited', Proceedings of the Seventh

Centenary Celebration of the D of Saint Bonaventure, pp. 47-59, The Franciscan
Institute
Brampton, C . K. (1963). 'The Pre' Order of Ockham's Non-Polemical Works',
Traditio 19:469-83
B[ranchereau], L. (1855). Praelectiones philosophiae in majori seminario Claromontensi primum
habitae, Leroux & Jouby
Breda, H. L. van (1971). ' T h e Actual State of the w o r k on Hussert's Inedita: achieve
ments and projects', Analecta Husserliana 2:149-59
Brennan, Robert Ε. (1941). 'The Thomistic Concept of Imagination', The New Scholas
ticism 15:149-61
Brentano, F. (1968). "Thomas von Aquin' in Die vier Phasen der Philosophie ... nebst
Abhandlungen über Plotinus, Thomas von Aquin ... (Philosophische Bibliothek, 195),
Felix Meiner
Brerewood, Edward (1614). Elementa Logicae, London
(1614a). Enquiries touching the Diversities of Languages and Religions through the chief parts
of the World, London
(1628). Tractatus quidam logici de Praedkabilibus et Praedicamentis, Oxford
(1640). Commentarii in Ethica Aristotelis, Oxford
Brosch, H.J. (1931). Der Seinsbegriff bei Boethius, mit besonderer Berücksichtigung der
Beziehung von Sosein und Dasein (Philosophie und Grenzwissenschaften 4, pt. 1 : 1 -
120) Rauch
Brounts, A. (1970). 'Nouvelles précisions sur la pecia*. Scriptorium 2 4 : 3 4 3 - 5 9
Brown, J. V. (1971). 'Sensation in Henry of Ghent', Archiv für Geschichte der Philosophie
53:238-66
Brown, Mary A. (1966). 'The Role of the Tractatus de Obligationibus in Medieval
Logic', Franciscan Studies 26:26—35
Brown, Stephen F. (1965). 'Avicenna and the Unity of the Concept of Being', Franciscan
Studies 25:117-50
(1972). 'Walter Burleigh's Treatise De suppositionibus and Its Influence o n William of
Ockham', Franciscan Studies 32; 15-64
(1973)· 'Walter Burleigh's Middle Commentary on Aristotle's Perihermeneias',
Franciscan Studies 33:42-134
(1974). 'Walter Burley's Quaestiones in librum Perihermeneias', Franciscan Studies 34:
200-95
Browne, I. (1963). International Law and the Use of Force by States, Clarendon Press
Browne, M. (1932). 'Circa intellectum et eius illuminationem apud S. Albertum
M a g n u m ' , Angelicunt 9:187-202
Bruckmüller, Franz (1908). Untersuchungen über Sigers (von Brabanl) Anima intellectiva,
G . J . Manz (Inauguraldissertation, Ludwig-Maximilians-Universität, Munich)
Bruder, Κ. (1928). Die philosophischen Elemente in den Opuscula sacra des Boethius, Meinar
Bruni, G. (1936). Le Opere di Egidio Romano, Olschki
Brunner, F. (1963). Platonisme et Aristotélisme, Publications Universitaires de Louvain
Bucher, Eugen (1965). Das subjektive Recht als Normsetzungsbefugnis,]. C. B. Mohr
Bucher, T. (1948). Theologischen Grundlegung der menschlichen Sozialanlage nach der Lehre
des hl. Bonaventura, Weger
Buchner, H . (1970). Platins Möglichkeitslehre (Epimeleia, Beiträge zur Philosophie, 16),
Anton Pustet
Buisson, L. (1958). Potestas und Caritas: die päpstliche Gewall in Spätmittelalter (Fors-
chungen zur kirchlichen Rechtsgeschichte und zum Kirchenrecht, 2), Böhlau
Bibliography 905
Burgersdijck, Franco (1623). Idea Morali Philosophiae, Leyden

(1626a). Institutionum logkarum synopsis sive rudimenta logica, Leyden
(1626b). Institutionum logkarum libri duo, Leyden
Buridan, John, see John Buridan
Burley, Walter, see Walter Burley
Burns, J. H. (1954). 'New light on John Major', The Innes Review 5:83-100
Bursill-Hall, G. L. (1971). Speculative Grammars of the Middle Ages (Approaches to
Semantics, 11), Mouton
Busard, H. L. L. (1965). 'Unendliche Reihen in A est unum calidum', Archive for History of
Exact Sciences 2:387-97
Butts, Robert and Joseph Pitt (1978). New Perspectives on Galileo, Reidel
Buytaert, E. M. (1964). 'The Tractatus Logicae Minor of Ockham', Franciscan Studies 24:
34-100
ed· (1974)· Peter Abelard (Mediaevalia Lovaniensia series I/studia II), Louvain, Presses
Universitaires
Byrne, Edmund F. (1968). Probability and Opinion, Mouton
Caesarius, Johannes (1559). Dialectica, Cologne

Cairns, H. (1949). Legal Philosophy from Plato to Hegel, Johns Hopkins University Press
Cajetan, Thomas (1938-9). Commentaria in De anima Aristotelis, ed. P. I. Coquelle
(2 vols.), Angelicum
Callus, D. A. (1943). 'The Introduction of Aristotelian Learning at Oxford', Proceedings
of the British Academy 29:229-81.
ed. (1955a). Rooerf Grosseteste, Scholar and Bishop, Clarendon Press
(1955b). "The Treatise ofJohn Blund On the Soul' in Autour (PAristote, Publications
Universitaires de Louvain
(1957). 'Les sources de saint Thomas: État de la question' in Aristote et saint Thomas
d'Aquin (Chaire Cardinal Mercier, 1955), Publications Universitaires de Louvain
and Éditions Béatrice-Nauwelaerts
(i960). 'San Tommaso d'Aquino e Sant'Alberto Magno*, /4Mgfiici«ri 37:133-61
Callus, D.A., and R. W. Hunt, eds. (1970). Johannes Blund, Tractatus de Anima, British
Academy
Camara, Helder (1978). 'What would Saint Thomas Aquinas, the Aristotle commen-
tator, do if faced with Karl Marx?' The Journal of Religion 58 Supplement: S 174-82
Campana, Α. (1946). 'The origin of the word 'humanist', Journal of the Warburg and
Courtauld Institutes 9:60-73
Campanella, Tommaso (1638). Philosophiae rationalis partes quinque, Paris
Camporeale, S. I. (1972). Lorenzo Valla, umanesimo e teologia, Nella Sede dell 'Istituto
Palazzo Strozzi
Canizzo, G. (1961), 'La dottrina del "Verbum mentis" in Pietro d'Auvergne', Rivista di
filosofia neoscolastica 53:152-68
Cantor, Georg (1932). Gesammelte Abhandlungen, ed. Ε. Zermelo, Springer
Capelli, A. (1961). Lexicon Abbreviaturarum (6th ed.) Hoepli
Cargill Thompson, W. D. J. (1972). 'The philosopher of the "politic society": Richard
Hooker as a political thinker' in W. Speed Hill (ed.) Studies in Richard Hooker:
Essays Preliminary to an Edition of his Works, Case Western Reserve University
Carlo, William (1966a). 'Idea and Concept: A Key to Epistemology' in Frederick J.
Adelmann (ed.) The Quest for the Absolute (Boston College Studies in Philosophy,
No. 1), Boston College and Martinus Nijhoff
(1966b). The Ultimate Reducibility of Essence to Existence in Existential Metaphysics,
9θ6 Bibliography
Martinus Nijhoff
Carlyle, R. W. and A. J. (1903-36). A History of Medieval Political Theory in the West
(6 vols.), Blackwood
Casacci, A. (1926). 'Gli "Elegantiarum libri" di Lorenzo Valla', Atene e Roma (ser. 2)
7:187-203
Casado, F, (1951—3). Έ1 pensamiento filosofico del beato Santiago de Viterbo', La
Ciudad de Dios 163:437-54; 164:301-31; 165:103-44, 283-302, 489-500
Cassiodorus (1937). lnstitutiones, ed. R. A. B. Mynors, Clarendon Press
Cassirer, Ernst (1946). The Myth of the State, Yale University Press
Chabod, F. (1958). Machiavelli and the Renaissance, Bowes & Bowes
Charette, Léon (1972). 'Philosophie politique et méthode chez Saint Thomas d'Aquin',
Revue de l'Université d'Ottawa 42:83-96
Chenevert, Jacques (1961). 'Le verbum dans le Commentaire sur les Sentences de Saint
Thomas d'Aquin', Sciences ecclésiastiques 13 :191-223 and 359-390
Chenu, M.-D. (1934). 'Contribution à l'histoire du traité de la foi. Commentaire
historique de IIaIIM q. 1, a. 2' in Mélanges Thomistes, publiés par les Dominicains de la
province de France à l'occasion du VF centenaire de la canonisation de Saint Thomas
d'Aquin (18 juillet 1323), Bibliothèque Thomiste III, Le Saulchoir, 1923, 2e édition,
J. Vrin, 1934
(1936). 'Grammaire et théologie aux Xlle et XIHe siècles', Archives d'histoire doctrinale
et littéraire du moyen âge 10:5—28
(r954)· Introduction à l'étude de Saint Thomas d'Aquin,]. Vrin
(1964). Toward Understanding Saint Thomas, trans. A. M. Landry and D. Hughes,
Henry Regnery
Chevrier, G. (1952). 'Remarques sur l'introduction et les vicissitudes de la distinction du
ius privatum et du ius publicum dans les oeuvres des anciens juristes français', Archives
de philosophie du droit, 1:5—77
Chodorow, Stanley (1972). Christian Political Theory and Church Politics in the Mid-
Twelfth Century: The Ecclesiology qfGratian's Decretum, University of California
Press
Chomsky, Noam (1966). Cartesian Linguistics, Harper and Row
Chossat, M. (1914). 'Saint Thomas d'Aquin et Siger de Brabant', Revue de philosophie
24:.553-575.2S:25-52
(1932). 'L'Averroisme de saint Thomas. Notes sur la distinction d'essence et
d'existence à la fin du XIIIe siècle', Archives de philosophie 9:129(465)-177(513)
(1939)· 'Dieu', Dictionnaire de théologie catholique 4:1180
Chrimes, S. B. (1936). English Constitutional Ideas in the Fifteenth Century, Cambridge
University Press
Christine de Pisan (1937). The Boke ofFayttes of Armes and Chyvalrye trans. William
Caxton, ed. A. T. P. Byles (Early English Text Society 184), Oxford University
(1973). Fayttes of Armes, excerpted in Allmand 1973
Chroust, A. H. (1947). 'The Corporate Idea and the Body Politic in the Middle Ages',
Review of Politics 9:423-52
Chroust, A. H. and Corbett, J. A. (1949). 'The fifteenth century Review of Politics of
Laurentius of Arezzo', Mediaeval Studies 11:62-76
Church, Alonzo (i960). 'Intention' in D. D. Runes (ed.) Dictionary of Philosophy,
Philosophical Library
Claessens, P. (i860). Raison et Révélation. Exposé sommaire de quelques notions et de principes
généraux auxiliaires d'une philosophie catholique, Lethielleux
Clagett, Marshall (1941). Giovanni Marliani and Late Medieval Physics, Columbia
Bibliography 907
University Press
(1950). 'Richard Swineshead and Late Medieval Physics', Osiris 9:131—61
(1959). The Science of Mechanics in the Middle Ages, University of Wisconsin Press
(1968). Nicole Oresme and the Medieval Geometry of Qualities and Motions, University of
Wisconsin Press
Clarembald of Arras (1965). Life and Works ofCkrembald of Arras, ed. Nikolaus Häring,
Pontifical Institute of Mediaeval Studies
Clark, David W. (1971). 'Voluntarism and rationalism in the ethics of Ockham',
(1973). 'William of Ockham on right reason', Speculum 48:13-36
Clarke, W. N. (1952a). 'The Limitation of Act by Potency: Aristotelianism or
Neoplatonism?', The New Scholasticism 26:167—94
(1952b). 'The Meaning of Participation in St Thomas', Proceedings of the American
Catholic Philosophical Association 26:147-57
Classen, P. (i960). Gerhoch von Reichersberg, F. Steiner
Classen, S. (i960). 'Collectanea zum Studien- und Buchwesen des Mittelalters mit
besonderer Berücksichtigung der Kölner Universität und der Mendikantenstudien',
ArchivfirGeschichte der Philosophie 42:159-206 & 247-71
Cobban, Α. Β. (1975). The Medieval Universities: their Development and Organisation,
Methuen
Coing, Helmut (ed.) (1973). Handbuch der Quellen und Literatur der neueren europäischen
Privatrechtsgeschkhte, I; Mittelalter (1100-1300). Die gelehrten Rechte und die
Gesetzgebung, C. H. Beck
Coleman, Janet (1975), 'Jean de Ripa, O. F. Μ. and the Oxford Calculators', Mediaeval
Studies 37:130-89
Coleman, T. (1971). Modistic Grammar. Unpublished dissertation, University of Toronto
Colish, M. L. (1968). The Mirror of Language: A Study in the Medieval Theory of
Knowledge, Yale University Press
Collins, J. (1947). The Thomistic Philosophy of the Angels, The Catholic University of
America
Combes, A. (1940). Jean Gerson, commentateur dionysien,]. Vrin
Combes, A. and Ruello, F. (eds.) (1967). 'Jean de Ripa I Sent. Dist. XXXVII: De modo
inexistendi divine essentie in omnibus creaturis', Traditio 23:191-267
Composta, D. (1954). '11 diritto naturale in Graziano', Studia Gratiana 2:151-210
Contenson, P.-M. (1959a). 'S. Thomas et l'avicennisme latin', Revue des sciences
philosophiques et théologiques 43:3-31
(1959b). 'Avicennisme latin et vision de Dieu au début du XIIIe siècle'. Archives
d'histoire doctrinale et littéraire du moyen âge 34:29-97
Copleston, Frederick (1963). A History of Philosophy, vol. 3, Part II, Doubleday Image
Books
Cornoldi, G. M. (1893). The physical system of St Thomas, tr. Edward Heneage Dering,
London & Leamington Art & Book Co.
Costello, Frank B. (1974). The Political Philosophy of Luis de Molina, S.J. (1535-1600)
(Bibliotheca Instituti Historici S. 1.), Institutum Historicum S. I.
Cotta, S. (1955). // concetto di legge nella Summa theologiae di S. Tommaso d'Aquino
(Universitâ di Torino. Memorie dell'Istituto Giuridico, série 2, mem. 89),
G. Giappichelli
Courtenay, William J. (1971). Ά revised text of Robert Holcot's quodlibetal dispute on
whether God is able to know more than he knows', Archiv fir Geschichte der
Philosophie 53:1-21
9θ8 Bibliography
(1972-3). 'John o f Mirecourt and Gregory of Rimini on Whether God Can Undo the
Past', Recherches de Tliéologie Ancienne et Médiévale 39:224-56; 40:147—74
(1978). Adam Wodeham. An Introduction to His Life and Writings, E.J. Brill
Cousin, Victor (1836). Ouvrages inédites d'Abelard, pour servir à l'histoire de la philosophie
scolastique en France, Imprimerie Royale
(1838). Historie générale de la Philosophie, Didier
(1849). Petri Abaelardi Opera (2 vols.), A. Durand
Crakanthorpe, Richard (1619). Introductio in metaphysicam, Oxford
(1622). Logkae Libri Quinque, London
Cranz, F. E. (1940). Aristotelianism in Medieval Political Theory: a study of the reception of
the Politics, Harvard University Graduate School: Summaries o f theses, pp. 133—6
(1971). A Bibliography of Aristotle Editions, 1501—1600 (Bibliotheca Bibliographica
Aureliana, 37), Valentin Koerner
(1976). 'The Renaissance reading of the De Anima', De Pétrarque à Descartes 32:359-
76
Crawford, F. S., ed. (1953). Averrois Cordubensis Commentarium Magnum in Aristotelis De
Anima (Corpus Commentariorum Averrois in Aristotelem), Mediaeval Academy of
America
Crescini, A. (1965). Le origini del metodo analitko: il Cinquecento, Del Bianco
Crombie, A. C. (1953). Robert Grosseteste and the origins of experimental science: 1100-1700,
Clarendon Press
Crosby, H. Lamar, Jr. (1955). Thomas of Bradwardine: His Tractatus de Proportionibus; Its
Significance for the Development of Mathematical Physics, (Publications in Medieval
Science, 2), University o f Wisconsin Press
Crowe, M. B. (1974). 'St Thomas and Ulpian's natural law' in A. Maurer ft al. (eds.)
St Thomas Aquinas, 1274-1974, 1, Pontifical Institute of Mediaeval Studies
(1977)· The Changing Profile of the Natural law, Martinus Nijhoff
Crowley, T. (1950). Roger Bacon. The problem of the Soul in his Philosophical
Commentaries, Duffy
(1952). 'Roger Bacon and Avicenna', Philosophical Studies (Maynooth) 2 : 8 2 - 8
Cunningham, F. (1962). 'Distinction According to St Thomas', The New Scholasticism
36:279-312
(1964). 'Textos de Santo Tomas sobre el esse y esencia', Pensamiento 20:283—306
(1970). "The "Real Distinction" in John Quidort', Journal of the History of Philosophy
8:9-28
Curtis, Mark H. (1959). Oxford and Cambridge in Transition 1558-1642, Clarendon Press
Czartoryski, P. (i960). 'Gloses et commentaires inconnus sur la Politique d'Aristote
d'après les mss. de la bibliothèque Jagellone de Cracovie', Mediaevalia Philosophica
Polonorum 5 : 3 - 4 4
Dabin, P. (1950). Le sacerdoce royal des fidèles dans la tradition ancienne et moderne, L'Édition
Universelle
Dähnert, Ulrich (1934). Die Erkenntnislehre des Albertus Magnus gemessen an den Stufen der
"Abstractio" (Studien und Bibliographien zur Gegenwartsphilosophie, vol. 4),
S. Hirzel
Dales, R . C , ed. (1963). Roberti Grosseteste Commentarius in VIII Libros Physicorum
Aristotelis, University of Colorado Press
Dal Pra, Mario, ed. (1954a). Pietro Abelardo Scritti Filosofici, Editio Super Porphyrium,
Glossae in Categorias, Editio Super Aristotelem De Interpretatione, De Divisionibus,
Super Topica Glossae, Bocca
Bibliography 909
(1954b). 'Sulla dottrina della impositio prima et secunda', Rivista critica di storia della
filosofia 9:390-9
D'Alverny, M.-T. (1952). 'Notes sur les traductions médiévales des oeuvres philoso-
phiques d'Avicenne', Archives d'hutoire doctrinale et littéraire du moyen âge 19:337-58
(*957)- 'Les traductions d'Avicenne (Moyen âge et Renaissance)' in Avkenne nella
storia della cultura medioevale. Relazioni e discussione, Accademia Nazionale dei Liucei
(1961-72). 'Avicenna Latinus', Archives a"'histoire doctrinale et littéraire du moyen âge
28-39
Daly, Lowrie J. (1962). The Political Theory of John Wyclif, Loyola University Press
(1968). 'Medieval and Renaissance Commentaries on the Politics of Aristotle',
Duquesne Revkw 13:41-55
(1973). 'Wyclif's political theory: a century of study', Medievalia et Humanistica, 4:
177-87
Damascene, John. See John Damascene
D'Angelo, Francesco Saverio (1968). Lafilosofiadella politica in S. Tommaso a"Aquino,
Caltanissetta: A cura del Seminario vescovile
Dante Alighieri (1963). Monorchia in E. Moore and P. Toynbee (eds.) Le Opere di Dante
Alighieri, 5th ed., Oxford University Press
Danto A. C. (1973). Analytical Philosophy of Action, Cambridge University Press
Da Palma Campania, G. (1955). La dottrina sull'unità dell'intelletto in Sigieri di Brabante,
CEDAM
David, M. (1954). La souveraineté et les limites juridiques du pouvoir monarchique du IX' au
XV' siècle, Librairie Dalloz
Davies, William David (1955). Paul and Rabbinic Judaism, S.P.C.K.
(1962). 'Conscience' in Buttrick et al. (eds.) The Interpreter's Dictionary of the Bible I :
671-6, Abingdon Press
Davis, C. T. (1957). Dante and the Idea of Rome, Clarendon Press
Davitt, T. E. (1951). The Nature of Law, Herder
Day, S. (1947). Intuitive Cognition: A Key to the Significance of the Later Scholastics, The
De Andres, Teodoro, see Andres, Teodoro de
De Benedictis, M. M. (1946). The Social Thought of Saint Bonaventure, Catholic
University of America
De Bonald, L. G. Α., see Bonald, L. G. A. de
Decker, B. (1967). Die Gotteslehre des Jakob von Metz (BGPM XLII, l), Aschendorff
De Finance, J., see Finance, J. de
De Fourny, Maurice (1977). 'The Aim of the State: Peace', inj. Barnes, M. Schofield,
R. Sorabji (eds.) Articles on Arhtotle: 2. Ethics and Politics, Duckworth
Deku, H. (1956). 'Possibile logicum', Philosophisches Jahrbuch der Görres-GeselUchafi, 64:
I-21
De la Brière, Y., iff La Brière, Y. de
De Lagarde, G., set Lagarde, G. de
De Lapparent, P., see Lapparent, P. de
De La Vaissière, J., see La Vaissière, J. de
De Libera, Α., jef Libera, A. de
Delhaye, P. (1947). 'L'organisation scolaire au XII e siècle', Traditio 5:211-68
(1964). 'Notes sur l'histoire et le sens actuel de la vertu de justice', Mélanges de sciences
religieuses 21:1-14
Del Prado, N. (1911). De veritate fundamental! philosophiae Christianae, Ex typis con-
sociationis saneti Pauli, Friburgi Helvetiorum
910 Bibliography
De Maistre, J., sie Maistre, J. de

Dempf, A. (1929). Sacrum Imperium: Geschkhts- und Staatsphilosophie des Mittelalters und
der politischen Renaissance, R. Oldenbourg
(1931). Die Ethik des Mittelalters, R. Oldenbourg
Denifle, H., and A. Châtelain (1889-97). Chartularium Universitatis Parisiensis (4 vols.),
Delalain
D'Entrèves, Alessandro Passerin (1932). Riccardo Hooker: Contributo alla teoria e alla storia
del diritto naturale, Presso L'Istituto Giuridico della R. Universitâ
(1939). The Medieval Contribution to Political Thought: Thomas Aquinas, Marsilius of
Padua, Richard Hooker, Oxford University Press
(1951). Natural Law. An introduction to legal philosophy, Hutchinson's University
Library
(1952). Dante as a Political Thinker, Oxford University Press
De Pater, W., see Pater, W. de
De Raeymaeker, L., see Raeymaeker, L. de
De Rijk, L. M., see Rijk. L. M. de
Descartes, René (1897 and 1913). Diicours de la méthode in Charles Adam et Paul
Tannery (eds.), Oeuvres de Descartes, L. Cerf
Destrez, J. (1935). La Pecia dans les manuscrits universitaires du XIII' et du XIV siècle, Paris
De Vaux, R., see Vaux, R. de
Dewan, Lawrence (1980). 'St Albert, the Sensibles, and Spiritual Being' in James A.
Weisheipl (ed.) Albertus Magnus and the Sciences: Commtmorativt Essays 1980,
De Wulf, Maurice, see Wulf, Maurice de
Dexippus (1888). In Aristotelis Categorias commentarium, ed. A. Busse (Commentaria in
Aristotelem Graeca IV. 2), Teubner
Dezza, P., ed. (1942—4). / neotomisti italiani del XIX secolo (2 vols.), Fratelli Bocca.
Diem, G. (1967). 'Les Traductions gréco-latines de la Métaphysique au moyen âge: le
problème de la Metaphysica Vêtus', Archiv für Geschichte der Philosophie 49:7-71.
Dijksterhuis, E. J. (1961). The Mechanization of the World Picture, tr. C. Dikshoorne,
Clarendon Press
Di Napoli, G. (1971). Lorenzo Valla:filosofiae religione nell' umanesimo italiano, Edizioni
di storia e letteratura
Diogenes Laertius (1925). Lif« of Eminent Philosophers I-II. With an English translation
by R. D. Hicks (Loeb Classical Library), William Heinemann and Harvard
University Press
(1964). Vitat philosophorum, ed. H. S. Long (Oxford Classical Texts), Clarendon Press
Dirks, Heinrich (1928). Des hi. Bonavtntura Ideen über Staat und Recht, Görres-Druckerei
(Koblenzer Volkszeitung)
Dmowski, J. (1836). Logica et metaphysica quae traduntur in Collegio Romano Societatis lesu a
R. P. Dmowski in exeunte anno 1836 et proximo 1837, Collegio Romano
Dod, B. G. (1970). 'The Study of Aristotle's Posterior Analytics in the Twelfth and
Thirteenth Centuries', Unpublished B. Litt, thesis, Oxford University
Döring, Κ. (1972). Die Megariker. Kommentierte Sammlung der Testimonien (Studien zur
antiken Philosophie, 2), B. R. Grüner
Doig, J. (1965). 'Science première et science universelle dans le "Commentaire de la
métaphysique" de saint Thomas d'Aquin', Revue philosophique de Louvain 63:41-96
(1972). Aquinas on Metaphysics. A historko-doctrinal study of the Commentary on the
Metaphysics, Martinus Nijhoff
Domanski, J. (1966). 'Stephani de Reate de secundis intentionibus', Mediaevalia
Bibliography 911
Philosophica Polonorum 12:67-106

(15)67). Oaves intentionum', Materialy 7:3-22
Donagan, Alan (1969). 'The Scholastic Theory of Moral Law in the Modern World' in
Anthony Kenny (ed.) Aquinas: A Collection of Critical Essays, Anchor Books
Doncoeur, P. (1910). 'Notes sur les averroistes latins: Boèce le Dace', Revue des sciences
philosophiques 4:500-11
Dondaine, A. (1947). 'Le manuel de l'inquisiteur (1230-1330)', Archivum Fratrum
Praedkalorum 17:85-194
(1956). Les secrétaires de Saint Thomas, Commissio Leonina
(1967). 'Un cas majeur d'utilisation d'un argument paléographique en critique
textuelle', Scriptorium 21:261-76
Dondaine, A. and Bataillon, L. J. (1966). 'Le manuscrit Vindob. lat. 2330 et Siger de
Brabant', Archivum Fratrum Praedicatorum 36: IS3-261
Donzé, R. (1967). La grammaire générale et raisonnée de Port-Royal: contribution à l'histoire
des idées grammaticales en France, Francke
Dowdall, H. C. (1923). 'The word "state"'. Law Quarterly Review 39:98-125
Duhem, Pierre (1906-13). Études sur Léonard de Vinci (3 vols.), Hermann
(1913-59). Le système du monde: histoire des doctrines cosmologiques de Platon à Copernic
(10 vols.), Hermann
Duin, J. J. (1954). La doctrine de la providence dans les écrits de Siger de Brabant, Éditions de
l'Institut Supérieur de Philosophie de Louvain
Dumbauld, Edward (1969). The Life and Legal Writings of Hugo Grotius, University of
Oklahoma Press
Dumitriu, Anton (1977). History of Logic, (4 vols.), vol. II, Abacus Press
Dunbabin, Jean (1963). 'The Two Commentaries of Albertus Magnus on the
Nicomachean Ethics', Recherches de Théologie Ancienne tt Médiévale 30:232-50
(1972). 'Robert Grosseteste as Translator, Transmitter, and Commentator: the
Nicomachean Ethics', Traditio 28:460-72
Duns Scotus, John, iff John Duns Scotus
Durand of Saint-Pourçain (1506). De lurisdictione Ecclesiastica,]. Petit
(1677). De Origine et Usu Iurisdictionum in Maxima Bibliotheca Veterum Palrum
26:127-35, Lyon
(1929). Quaestio de natura cognitionis, ed. J. Koch (BGPM, XXVI), Aschendorff
During, I. (1966). Aristoteles. Darstellung und Interpretation seines Denkens, Winter
Dürr, Karl (1942). 'Alte und neue Logik',Jahrbuch der Schweizerischen Philosophischen
Gesellschaft 2:104-22
(1951). The Propositional Logic of Boethius, North-Holland
Duzy, Erminius Stanislaus (1944). Philosophy of Social Change according to the Principles of
Saint Thomas, Catholic University of America
Dyckmans, W. (1937). Das mittelalterliche Gemeinschaftsdenken unter dem Gesichtspunkt der
Totalität: eine rechtsphilosophische Untersuchung (GSrresgesellschaft
Veröffentlichungen der Sektion für Rechts- und Staatswissenschaften 73),
Schöningh
Dziewicki, Michael, ed. (1893-9). JO/IM Wyclif. Tractatus de Logica (Wyclif's Latin
Works) (3 vols.). Trübner, for the Wyclif Society
Ebbesen, S. (1972). 'Anonymi Bodleiani in Sophisticos Elenchos Aristotelis

Commentarii fragmentum', CIMAGL 8:3-32
(1973a). 'Manlius Boethius on Aristotle's Analytica Posteriora', CIMAGL 9:68-73
(1973b). 'Paris 4720A. A I2th-Century Compendium of Aristotle's Sophistici Elenchi',
912 Bibliography
CIMAGL ίο : 1-20
(i976a). 'Anonymus Aurelianensis II, Aristotle, Alexander, Porphyry and Boethius.
Ancient Scholasticism and 12th century Western Europe', CIMAGL 16:1-128
(1976b). "The Summulae, Tractatus VII De Fallaciis' inj. Pinborg (ed.) Thf Logic of
John Buridan. Acts of the 3rd European Symposium on Medieval Logic and Semantics
(Opuscula graecolatina, 9), Museum Tusculanum
ed. (1977a). Incertorum auctorum Quaestiones super Sophistkos Elenchos (Corpus
Philosophorum Danicorum Medii Aevi, 7), Gad
(1977b). 'Jacobus Veneticus on the Posterior Analytics and some early 13th-century
Oxford Masters on the Elenchi', CIMAGL 21:1-9
(1977c). 'Can Equivocation Be Eliminated?', Studia Mediewistyczne 18:103-24
(!979)· "The Dead Man is Alive', Synthese 40:43—70
(1979b). 'Anonymi Aurelianensis I Commentarium in Sophisticos Elenchos',
CIMAGL 34
(1981a). 'Analyzing Syllogisms or Anonymus Aurelianensis III - the (presumably)
Earliest Extant Latin Commentary on the Prior Analytics, and its Greek Model',
CIMAGL 37:1-20
(1981b). Commentators and Commentaries on Aristotle's Sophistici Elenchi, 3 vols., Brill
Ebbesen, S. and J. Pinborg (1970). 'Studies in the Logical Writings Attributed to
Boethius de Dacia', CIMAGL 3:1-54
Eberenz, James H. (1969). The Concept of Sovereignty in Four Medieval Philosophers: John
ofSalisbury, St Thomas Aquinas, Egidius Colonna and Marsilius of Padua, University
Microfilms
Eckermann, Willigis (1972). Dfr Physikkomentar Hugolins von Orvieto OESA, De Gruyter
Ehrle, Franz (1925). 'Der Sentenzenkommentar Peters von Candia', Franziskanische
Studien Beiheft 9
(1954). Zur Enzyklika "Aeterni Patris", Text und Kommentar. Zum 75 jährigen Jubiläum
der Enzyklika neu herausgegeben von Franz Pelster S.J. (Sussidi Eruditi, 6), Edizioni di
Storia e Letteratura
Elie, Hubert (1937). Le complexe significabile,]. Vrin
Emden, A. B. (1957-9). A Biographical Register of the University of Oxford to a.d. 1500
(3 vols.). Clarendon Press
Enders, Heinz (1975). Sprachlogische Traktate des Mittelalters und der Semantikbegriff
(Veröffentlichungen des Grabmann-Instituts, neue Folge, 20), Schöningh
Engelbert of Admont (1614). De Ortu, Progressu et Fine Regnorum, et Praecipue Regni seu
Imperii Romani in M. Goldast (ed.) Politica Imperialia, pp. 754-73, John Bringer
Engelhardt, P. (1976). 'Intentio' inj. Ritter and K. Gründer (eds.) Historisches Wörterbuch
der Philosophie 4:466-74, Wissenschaftliche Buchgessellschaft
Epictetus (1898). Dissertationes, ed. H. Schenkl, Teubner
Ermatinger, Charles J. (1954). 'Averroism in Early Fourteenth Century Bologna',
Mediaeval Studies 16:35-56
(1963). The Coalescent Soul in Post-Thomistic Debate. Ph.D. Dissertation, Saint Louis
University
(1969). 'John ofJandun in his Relations with Arts Masters and Theologisns' in Arts
libéraux et philosophie au moyen âge, Institut d'Études médiévales and J. Vrin
Eschmann, I. T. (1943). Ά Thomistic Glossary on the Principle of the Preeminence of a
Common Good', Mediaeval Studies 5:123-65
(1944). 'Bonum commune melius est quam bonum unius', Mediaeval Studies
6:62-120
(1946). 'Studies on the notion of society in St Thomas Aquinas', Mediaeval Studies
Bibliography 913
8:1-42
{1947). 'Thomistic Social Philosophy and the Theory of Original Sin', Mediaeval
Studies 9:19-55
(1958). 'St Thomas Aquinas on the two powers', Mediaeval Studies 20:177—205
Eustachius a Sancto Paulo (1609). Summa philosophiae quadripartita, Paris
Evans, G. R. (1977). 'Inopes verborum sunt Latini: Technical Language and Technical
Terms in the Writings of St Anselm and some Commentaries of the Mid-Twelfth
century', Archives d'histoire doctrinale et littéraire du moyen âge 54:115-34
(1978). Anselm and Talking about God, Clarendon Press
Eynde, D. Van den (1957). L'oeuvre littéraire de Geroch de Reichersberg, Pontificium
Athenaeum Antonianum
(1962). 'Les écrits perdus d'Abelard', Antonianum 37:467-80
Eysinga, Willem J. M. Van (1945). Hui^ft de Groot, een Schets, H. D. Tjeenk Willink
Fabre [].] (1862). Défense de I'Ontologisme contre les attaques récentes de quelques écrivains qui
se disent disciples de Saint Thomas, Casterman
Fabro, Cornclio (1939). La nozione metafisca di partecipazione seconda San Tommaso
d'Aquino, Vita e Pensiero
(1950). La nozione metafisica di partecipazione secondo s. Tommaso d'Aquino (2nd ed.),
Società Editrice Internazionale
(i960). Partecipazione e causalità secondo San Tommaso, Società Editrice Internazionale
(1961). Participation et causalité. Publications Universitaires de Louvain
Faral, Edmond (1946). "Jean Buridan: Notes sur les manuscrits, les éditions et le contenu
de ses ouvrages', Archives d'histoire doctrinale et littéraire du moyen âge 21:1—53
(1949)· 'Jean Buridan: Maître es arts de l'Université de Paris' in Histoire littéraire de la
France Vol. 38, pp. 462-60S, Imprimerie nationale
Farr, William (1974). John Wyclif as legal Reformer, Brill
Farrel, W. (1930). The Natural Moral Law according to St Thomas and Suârez, St
Dominic's Press
Fasso, G. (1964). La legge delta ragione. Il Mulino
(1966-70). Storia dellafilosofiadel diritto (3 vols.), II Mulino
Faulkner, Robert K. (forthcoming). Richard Hooker and the Politics of a Chrktian England,
University of California Press
Fauser, W. (1973) têr Kommentar des Radulphus Brito zu Buch III De Anima (BGPM,
η. F., XII), Aschendorff
Faust, A. (1922). 'Die Dialektik Rudolf Agricoles: ein Beitrag zur Charakteristik des
deutschen Humanismus*, Archiv für Geschichte der Philosophie 34 (n. F. 27): 118-35
(1931-2). Der Möglichkeitsgedanke. Systemgeschichtliche Untersuchungen I-II (Synthesis,
6-7), Winter
Ferguson, W. K. (1948). The Renaissance in Historical Thought: Five Centuries of
Interpretation, Houghton Mifflin
Fernândez-Santamaria, J. A. (1977). The State, War and Peace: Spanish Political Thought in
the Renaissance, 1516—1559, Cambridge University Press
Ferrari, A. (1833), Ethices Christianae institutiones ad usum clericorum deductae (4 vols.),
G. Vincenzi
Ferrée, William (1942). The Act of Social Justice: An Analysis of the Thomistic concept of
Legal Justice..., Catholic University of America
Figgis.J. N. (1907). Studies of Political Thought from Gerson to Grotius, Cambridge
University Press
(1914). The Divine Right of Kings, Cambridge University Press
Finance, J. de (i960). Être et agir. Librairie Éditrice de l'Université Grégorienne
914 Bibliography
Fink-Errera, G. (i960). 'De l'édition universitaire', L'homme et son destin. Actes du l"
congrès international de philosophie médiévale, pp. 221-8, Nauwelaerts
(1962). 'Une institution du monde médiéval: la pecia'. Revue Philosophique de Louvain
60 (III. série, 66): 184-243
Fioravanti, G. (1972). 'SulP evoluzione del monopsichismo di Sigieri di Brabante', Atti
della Accademia délie Scienze di Torino 106:407-64
Fisher, Luke Francis (1948). A Philosophy of Social Leadership according to Thomistic
Principles, Catholic University of America
Fitch, Frederic B. (1970). 'Comments and A Suggestion", in Robert L. Martin (ed.), TTie
Paradox of the Liar, pp. 75-7, Yale University Press
FitzPatrick, P.J. (1973). 'Fact and Fiat: one theme in the modernist crisis', Durham
University Journal, n. s. 34:151-80
Fletcher, H. F. (1961). The Intellectual Development of John Milton, University of Illinois
Press
Flückiger, F. (1955). Geschichte des Naturrechts, Evangelischer Verlag
Fobes. F. H., and Kurland, S., eds. (1956). Averrois Cordubensis Commentarium medium in
Aristotelis De Ceneralione et Corruptione Libros (Corpus Commentariorum Averrois
in Aristotelem), Mediaeval Academy of America
Fois, Mario (1969). IIpensiero cristiano di Lorenzo Valla nel quadro storko-culturele del suo
ambiente (Analecta Gregoriana, 174), Libreria Editrice dell'Università Gregoriana
Folz, R. (1969). The Concept of Empire in Western Europe from the Fifth to the Fourteenth
Century, tr. S. A. Ogilvie, Arnold
Fonseca, Pierre da. Sef Petrus da Fonseca
Forest, A. (1956). La structure métaphysique du concret selon saint Thomas d'Aquin (2nd ed.),
J.Vrin
Fortescue, Sir John (1864). De natura legis naturae. Printed from a manuscript in the
Lambeth Library for Lord Clermont
(1942). De Laudibus Legum Anglie - The Praise of the Laws of England, tr. and ed. S. B.
Chrimes, Cambridge University Press
Fortin, Ε. L. and O'Neill, P. D. (tr.) (1963). 'The Condemnation of 1277', in R. Lerner
and M. Mahdi (eds.) Medieval Political Philosophy: A Source Book, The Free Press of
Glencoe
Foucher, L. (1955). La philosophie catholique en France avant la renaissance thomiste, et en
rapport avec elle (1800-1880),]. Vrin
Fraassen, Bas C. Van (1971). Formal Semantics and Logic, Macmillan
Franceschini, E. (1933). Roberto Grossatesta, vescovo di Lincoln, e le sue traduzioni latine
(Atti del Reale Istituto Veneto di Scienze, Lettere ed Arti, 93. 2). C. Ferrari
Francis of Mayron (1940-2). 'L'oeuvre politique de François de Mayronnes', ed. P. de
Lapparent, Archives d'histoire doctrinale et littéraire du moyen âge 15-17:5-151
Franklin, J. H. (1973). Jean Bodin and the Rise of Absolutist Theory, Cambridge University
Press
Fredborg, Κ. M. (1973). 'The Dependence of Petrus Helias' Summa super Priscianum on
William of Conches' Glose super Priscianum', CIMAGL 11:1-57
(1977). 'Tractatus glosarum Prisciani in MS Vat. lat. i486', CIMAGL 21:2i~44
Fredborg, K. M., N.J. Green-Pedersen, L. Nielsen, and Jan Pinborg (1975). 'The
Commentary on "Priscianus Maior" ascribed to Robert Kilwardby', CIMAGL 15:
1-143
Fredborg, K. M., L. Nielsen andj. Pinborg (1978). 'An Unedited Part of Roger Bacon's
Opus Maius: De signis', Traditio 34:75-136
Frede, Michael (1974a). Die stoische Logik (Abhandlungen der Akademie der
Bibliography 915
Wissenschaften in Göttingen, Phil.-hist. Klasse, Dritte Folge, 88), Vandenhoeck

und Ruprecht
(1974b). 'Stoic vs. Aristotelian Syllogistic', Archiv für Geschichte der Philosophie
56:1-32
(1977)· 'The Origins of Traditional Grammar' in Butts and Hintikka (ed.) Historical and
Philosophical Dimensions of Logk, Methodology and Philosophy of Science, Reidel
(1978). 'Principles of Stoic Grammar' inj. Ν. Rist (ed.) The Stoics, University of
California Press
Friedmann, Wolfgang (1967). 'Grotius, Hugo' in Paul Edwards (ed.) The Encyclopedia of
Philosophy (vol. 3), Collier-Macmillan
Friedrich, C.J. (1958). The Philosophy of Law in Historical Perspective, University of
Chicago Press
FumagaUi, M. T. B-B. (1970). The Logk of Abelard, Reidel
Gabriel, A. (1977). The Economic and Material Frame of the Medieval University (Texts and
Studies in the History of Medieval Education 15), Notre Dame University Press
Gaeta, F. (1955). Lorenzo Valla:filologia estoria nell' umanesimo italiano, Istituto italiano
per gli studi storici
Gaetano di Thiene (1491). Df reactione, Venice
(1494). Recollecte super De motu locali, Venice
Gagner, Sten (i960). Studien zur Ideengeschklite der Gesetzgebung (Acta Universitatis
Upsaliensis. Studia Iuridica Upsaliensia, I), Almqvist and Wiksell
Gâl, Gedeon (1967a). 'Gregory of Rimini* in New Catholic Encyclopedia 6:797
(1967b). 'William of Ockham' in New Catholic Encyclopedia 14:932-5
(1967c). 'Gualteri de Chatton et Guillelmi de Ockham Controversia de Natura
Conceptus Universalis', Franciscan Studies 27:191-212
(1969). 'Quaestio loannis de Reading de Necessitate Specierum Intelligibilium
Defensio Doctrinae Scoti', Franciscan Studies 29:66-156
(1971). 'Henricus de Harclay: Quaestio de Significato Conceptus Universalis',
(1977)· 'Adam of Wodeham's Question on the "Complexe Significabile" as the
Immediate Object of Scientific Knowledge', Franciscan Studies 37:66-102
Gâl, Gedeon, and Rega Wood (forthcoming). 'Richard Brinkley and His Summa
logicae', Franciscan Studies
Galen (1874). De placitis Hippocratis et Piatonis libri novem, ed. I. Mueller, Teubner
(1884-93). Scripta minora, ed. Marquardt, Mueller, Helmreich (3 vols.), Teubner
(1896). Institutio logica, ed. C. Kalbfleisch, Teubner
(1977)· Libellus de captionibus quae per dictionemfiunt,ed. C. G. Gabler, in R. B. Edlow
(ed.) Calfη on Language and Ambiguity (Philosophia Antiqua, 31), Brill
Galindo Romeo, P. and L. Ortiz Muhoz (1946). Antonio de Nebrija, Gramatka castellana,
texto establecido sobre la ed. "princeps" de 1492, Edicion de la Junta del Centenario
Gandillac, Maurice Patronnier de (1933). 'De l'usage et de la vsleur des arguments
probables dans les questions du cardinal Pierre d'Ailly sur le "Livre des Sentences" '.
Archives d'histoire doctrinale et littéraire du moyen âge 8:43-91
Garceau, Benoit (196%). Judicium: Vocabulaire, sources, doctrine de Saint Thomas d'Aquin
(Université de Montréal, L'Institut d'Études Médiévales, No. 20), Institut d'Études
Médiévales and Vrin
Garcia, F. M. (1910). Lexicon scholasticum philosophico-theologkum, Ad Claras Aquas
Gardeil, A. (1934). 'La perception expérimentale de l'âme par elle-même d'après Saint
Thomas' in Mélanges Thomistes (Bibliothèque Thomiste, vol. 3), Vrin
9i6 Bibliography
Gardet, Louis (i974)- 'Saint Thomas et ses prédécesseurs arabes' in St. Thomas Aquinas
1274-1974: Commemorative Studies, vol. 1, Pontifical Institute of Mediaeval Studies
(1976). 'La connaissance que Thomas d'Aquin put avoir du monde islamque' in
G. Verbeke and D. Verfielst (eds.) Aquinas and the Problems of His Time (Mediaevalia
Lovaniensia, Series I, Studia 5), Louvain, Publications Universitaires
Gardiner, Stephen (1930). The Oration of True Obedience in P. Janelle (ed.) Obedience in
Church and State, Cambridge University Press
Garin, Ε. (1951). 'Le traduzioni umanistiche di Aristotele nel secolo XV, Atti e memorie
dell'Accademia Fiorentina di Scienze Morali, La Colombaria 8 ^7-104
(1952). Prosatori latini del quattrocento (La letteratura italiana: storia e testi, 13), Ricciardi
(1958). IIpensieropedagogico dell' umanesimo, Giuntine e Sansoni
(1965). Italian Humanism: Philosophy and Civic Life in the Renaissance, tr. P. Munz,
Blackwell
(1969a). 'La cultura fiorentina nella secunda meta dell Trecento e i "barbari
Britanni"', in his L'età nuova. Richerche di storia della cultura dal XII al XVI secolo,
Morano
(1969b). Science and Civic Life in the Italian Renaissance, Doubleday
Garin, Pierre (1931). La théorie de l'idée suivant l'école thomiste, Desclée De Brouwer
Garlandus Compotista (1959). Garlandus Compotista. Dialectica, ed. L. M. de Rijk, Van
Gorcum
Gauthier, R.-A, (1947-8). 'Trois commentaires "averroistes" sur 1* Éthique â
Nicomaque', Archives d'histoire doctrinale et littéraire du moyen âge 22/23:187-336
(1947-53). Review of Lottin 1942-60, II—III» Bulletin thomiste 8, no. 1:6o-86
(1963). 'Arnoul de Provence et la doctrine de lafronesis, vertu mystique suprême',
Revue du Moyen Age latin 19:129—70
(1964). 'Les Questiones supra Librum Ethicorum de Pierre d'Auvergne', Revue du Moyen
Age latin 20:233-60
(1969). 'Praefatio' in Thomas Aquinas (1969), pt. 1
(1970). Aristote. L'Ethique à Nicomaque (Tome I, Première Partie: Introduction),
Nauwelaerts
(1974). Ethica Nicomachea. Praefatio (Aristoteles Latinus XXVI, 1-3, fasciculus primus).
Brill
(1975)- 'Le cours sur VEthica nova d'un maître es arts de Paris (123 5-1240)', AHDLMA
42:71-141
Geach, P. T. (1957). Mental Acts: Their Content and Their Objects, Routledge 6c Kegan
Paul
(1961). 'Aquinas' in G. E. M. Anscombe and P. T. Geach, Three Philosophers,
Blackwell
(1962). Reference and Generality, Cornell University Press
(1969)· 'Causality and Creation' in P. T. Geach, God and the Soul, Routledge & Kegan
Paul
(1972). Logic Matters, Blackwell
Geanakopolos, D.J. (1966). Byzantine East and Latin West, Two Worlds of Christendom in
Middle Ages and Renaissance: Studies in Ecclesiastical and Cultural History, Harper &
Row
Geiger, L. (1882). Renaissance und Humanismus in Italien und Deutschland, Grote
(1942). La participation dans la philosophie de saint Thomas d'Aquin,]. Vrin
(1947). 'Abstraction et séparation d'après s. Thomas. In de Trinitate, q. 5, a. 3', Revue
des sciences philosophiques et théologiques 31:3-40
Gelber, Hester Goodenough (1974). Logk and the Trinity: A Clash of Values in Scholastic
Bibliography 917
Thought, 1300-1335 (Unpublished Ph.D. Dissertation), University of Wisconsin

Genicot, L. (1976). 'Le De Regno: speculation ou réalisme?* in G. Verbeke and D.
Verhelst (eds.) Aquinas and the Problems of His Time (Mediaevalia Lovaniensia, Series
1 .Study 5), Nijhoff
Gerard of Odo (1482). Expositio in Aristotelis Ethicam, Brescia
Gerl, Hanna-Barbara (1974). Rhetorik als Philosophie: Lorenzo Valla (Humanistische
Bibliothek, Reihe I: Abhandlungen, Band 13), Fink Verlag
Gerson, John, set John Gerson
Gewirth, Alan (1951). Marsilius ofPadua and Medieval Political Philosophy, Vol. ι of
Marsilius of Padua: The Defender of Peace (Records of Civilization, Sources and
Studies, 46), Columbia University Press
(1961). 'Philosophy and political thought in the fourteenth century' in F. L. Utley
(ed.) The Forward Movement of the Fourteenth Century, Ohio State University
Geyer, Β. (1913). 'Die Stellung Abaelards in der Universalienfrage nach neuen hand
schriftlichen Texten' (BGPM, Supplementband 1), Aschendorff
(1919-27). Peter Abaelards Philosophische Schriften (BGPM, XXI, 1-4), Aschendorff
(1963). 'Albertus Magnus und die Entwicklung der Scholastischen Metaphysik', in
P. Wilpert (ed.) Die Metaphysik im Mittelalter (Miscellanea Mediaevalia, 2), De
Gruyter
Ghisalberti, Alessandro (1972). Guglielmo di Ockham (Pubblicazioni della Università
Cattolica, Milano, scienze filosofiche, 3), Vita e Pensiero
Gibson, Strickland (1930). 'The Order of Disputations', Bodleian Quarterly Record
6:107-12
ed. (1931). Statuta Antiqua Universitatis Oxoniensis, Clarendon Press
Gierke, O. (içjoo). Political Theories of the Middle Ages, tr. with an introduction by F. W.
Maitland. Cambridge University Press
(1934). Natural Law and the Theory of Society, 1500-1800, tr. Ernest Barker (2 vols.),
Cambridge University Press
(1977). Associations and Law: Classkal and early Christian Stages, tr. George Heiman,
University of Toronto Press
Giesey, Ralph E. (1973). 'Medieval jurisprudence in Bodin's concept of sovereignty' in
Horst Denzer (ed.) Jean Bodin: Proceedings of the International Conference on Bodin in
Munich, Beck
Gilbert, Felix (1965). Machiavclli and Guicciardini, Princeton University Press
Gilbert, Neal Ward (i960). Renaissance Concepts of Method, Columbia University Press
(1976). 'Richard de Bury and the "Quires of Yesterday's Sophisms'", in Edward P.
Mahoney (ed.) Philosophy and Humanism. Renaissance Essays in Honor of Paul Oskar
Kristeller, Columbia University Press
Gilbert of Poitiers (1966). The Commentaries on Boethius by Gilbert of Poitiers, ed. Nikolaus
M. Häring, Pontifical Institute of Mediaeval Studies
Gilby, Thomas (1958). Principality and Polity: Aquinas and the Rise of State Theory in the
West, Longmans Green
Giles of Rome (1476). In Aristotelis De anima commentum, Venice
(1481). Df Corpore Christi Theoremata, Bologna
(1498). De Regimine Principum, Venice
(1503). Egidius Romanus de esse et essentia, de mensura angelorum, et de cognitione
angelorum, Venice (Reprinted Minerva 1968)
(1521). In Primum Sententiarum, Venice (Reprinted Minerva 1968)
(1550). Aegidii Romani... opus super authorem de causis, Apud Iacobum Zoppinum
918 Bibliography
(1646). QuodUbeta, Louvain (Reprinted Minerva 1966)

(1929). De ecclesiastica potestate, ed. R. Scholz, H. Bählaus Nachfolger
(1930). Aegidii Romani Theoremata de esse et essentia. Texte précédé d'une introduction
historique et critique, ed. E. Hocedez, Museum Lessianum
(1944). Giles of Rome: Errores philosophorum, ed. with introduction and notes by J.
Koch, tr. J. O. Riedl, Marquette University Press
Gilmore, Myron P. (1941). Argument from Roman Law in Political Thought, 1200—1600,
(Harvard Historical Monographs, 15), Harvard University Press
Gilson, Etienne (1913). Index Scolastko-Cartésien, F. Alcan
(1921). 'La doctrine de la double vérité' in Études de philosophie médiévale (Publications
de la Faculté des Lettres de l'Université de Strasbourg, No. 3), Commission des
Publications de la Faculté des Lettres
(1926-7). 'Pourquoi Saint Thomas a critiqué Saint Augustin', Archives d'histoire
doctrinale et littéraire du moyen âge 1: 5—127
(1927). 'Avicenne et le point de départ de Duns Scot', Archives d'histoire doctrinale et
littéraire du moyen âge 2:89-149
(1929—30). 'Les sources gréco-arabes de l'augustinisme avicennisant', Archives
d'Histoire Doctrinale et Littéraire du Moyen Age 4:5-149
(1930). Études sur le rile de la pensée médiévale dans la formation du système cartésien,
J. Vrin
(1936). The Spirit of Medieval Philosophy, tr. A. H. C. Downes, Sheed and Ward
(1948a). Dante the Philosopher, Sheed and Ward
(1948b). 'L'Objet de la Métaphysique selon Duns Scot', Mediaeval Studies 10:21-92
(1950). 'La preuve du De ente et essentia', Acta III Congressus Thomistici Internationalis:
Doctor Communis 3 :257-60
(1952a). Being and some Philosophers, Pontifical Institute of Mediaeval Studies
(1952b). Jean Duns Scot: introduction à ses positions fondamentales, J. Vrin
(1952c). Les métamorphoses de la Cité de Dieu, Publications universitaires de Louvain
(*-953)· La philosophie de saint Bonaventure,]. Vrin
(1955)· History of Christian Philosophy in the Middle Ages, Random House
(1961a). The Christian Philosophy of St Thomas Aquinas (With a Catalogue of St
Thomas' Works by I. T. Eschmann), tr. L. K. Shook, Victor Gollancz
(1961b). 'Trois leçons sur le problème de l'existence de Dieu', Divinitas 5122-87
(1968). 'Sur la composition fondamentale de l'être fini', in De doctrina loannis Duns
Scoti. Acta Congressus Scotistici Internationalis Oxonii et Edimburgi 11-17 sept. 1966
célébrait. V. II: Problemata Philosophica, Cura Commissionis Scotisticae
Giorgianni, V. (1948). Pensiero morale e politico dt Bonaventura da Bagnorea, Lupa
Giusberti, Franco (1977). Ά Treatise on Implicit Propositions from around the Turn of
the Twelfth Century: An Edition with Some Introductory Notes', CiMv4GL 21:45~
115
Glorieux, P. (1925). 'La littérature quodiibetique de 1260 à 1320', Revue des sciences philo-
sophiques et théologiques 14
(1925-35). La littérature quodiibetique (2 vols.) (Bibliothèque Thomiste, 5, 21),
J. Vrin
(1931). 'Un recueil scolaire de Godefroid de Fontaines', Rtcherches de Théologie
ancienne et médiévale 3:37-53
(1933). Répertoire des maîtres en théologie de Paris au XIII' siècle (2 vols.), Vrin
(1939)· 'Notations brèves sur Godefroid de Fontaines', Recherches de Théologie ancienne
et médiévale 11:168-73
(1965). 'L'oeuvre littéraire de Pierre d'Ailly', Mélanges de sciences religieuses 22161-78
Bibliography 919
(1966). 'L'enseignement au moyen âge. Techniques et méthodes en usage à la Faculté

de Théologie de Paris, au XIIIe siècle', Archives d'histoire doctrinale et littéraire du
moyen âge 35 165-186
(1967). 'Jean de Falisca·. La formation d'un maître en théologie au XIV e siècle',
Godfrey of Fontaines (1904-37). Les Quodlibets de Godefroid de Fontaines, ed. M. de Wulf
and J. Hoffmans (Les Philosophes Belges), Institut supérieur de Philosophie de
l'Université de Louvain
(1914). Lfs Quodlibets V, VI, Vil de Godefroid de Fontaines, ed. M. de Wulf and
J. Hoffmans (Les Philosophes Belges, 3), Institut supérieur de Philosophie de
l'Université de Louvain
(1973)· 'Godfrey of Fontaines: Disputed Questions 9,10 and 12', ed. J. F. Wippel,
Goichon, A.-M. (1937). La distinction de l'essence et de l'existence d'après ibn Smâ, Desclée
de Brouwer
(1951). La philosophie d'Avicenne et son influence en Europe médiévale, 2nd ed., Adrien-
Maisonneuve
Goldman, Alvin I. (1970). A Theory of Human Action, Prentice Hall
Golius, Theophil us (1631). Epitome doctrinal moralis, Argentorati ( = Strasbourg)
Gollancz, Israel (1949-50). 'Strode, Ralph', Dictionary of National Biography (Oxford
University Press), 55 =57-9
Gomez Caffarena, J. (1957). 'Ctonologia de la "Suma" de Enrique de Gante por relacion
a sus "Quodlibetos" ', Gregorianum 38:116-33
(1958). Serpartkipado y ser subsistente en la metafisica de Enrique de Gante, Apud Aedes
Universitatis Gregorianae
Gomez Izquierdo, Alberto (1924). 'Valor cognoscitivo de la "intentio" en Santo Tomas
de Aquino', La Ciencia Tomista 29:169-88
Gomez Nogales, Salvador (1976). 'Saint Thomas, Averroes, et l'averroisme' in
G. Verbeke and D. Verhelst (eds.), Aquinas and tht Probltms of His Time (Mediae-
valia Lovaniensia, Series I, Studia 5), Louvain, Publications Universitaires
Gonsalvus of Spain (1935). Fr. Gonsalvi Hispani O.F.M. Quaestiones disputatae et
de Quodlibet, ed. Leo Amoros, O.F.M. (Bibliotheca Franciscana Scholastica Medii
Aevi, IX), Collegium S. Bonaventurae
Gonzalez, Atanasio (1958-9)- 'The Theory of Assertoric Consequence in Albert of
Saxony', Franciscan Studies 18:290-354 and 19:13-114
Gonzâlez-Llubera, I. (1926). Nebrija: Gramâtica de la lengua castellana, Humphrey Milford
Goudin, Α. (1851). Philosophia juxta inconcussa tutissimaque Divi Thomae dogmata, ed.
Roux Lavergne, Editores Bibliothecae Novae
(1864). Philosophie suivant les principes de saint Thomas, tr. Thomas Bourar,
Poussielgue-Rusand
Grabmann, Martin (1909-11). Geschichte der scholastischen Methode, I-II, Herder
(1924a). Neu aufgefundene Werke des Siger von Brabant und Boetius von Dacien
(Sitzungsberichte der Bayerischen Akademie der Wissenschaften, Philosophisch-
historische Abteilung, 1924, No. 2)
(1924b). 'Doctrina S. Thomae de distinctione reali inter essentiam et esse ex documen
ta ineditis saeculi XIII illustratur', in Acta Hebdomadae Thomisticae Romae Celebratae
19-25 Novembris 1923, Apud Sedem Academiae S. Thomae Aquinatis
(1925). Die Kulturphilosophie des heiligen Thomas von Aquino, Β. Filser
(1926-56). Mittelalterliches Geistesleben I (1926), II (1936), III (1956), Max Hueber
(1926a). 'Das Naturrecht der Scholastik von Gratian bis Thomas von Aquino', in his
920 Bibliography
Mittelalterliches Geistesleben I, M. Hueber

(1926b). 'Die Disputationes Metaphysicae der Franz Suarez in ihrer methodischen
Eigenart und Fortwirkung', in his Mittelalterliches Geistesleben I, M. Hueber
(1926c). 'Eine ungedruckte Verteidigungsschrift der scholastischen Übersetzung der
Nikomachischen Ethik gegenüber dem Humanisten Lionardo Bruni' in his
Mittelalterliches Geistesleben I, M. Hueber
(1931). Der lateinische Averroismus und seine Stellung zur christlichen Weltanschauung
(Sitzungsberichte der Bayerischen Akademie der Wissenschaften, Phil-hist. Abt.,
Heft 2)
(1931a). Der göttliche Grund menschlicher Wahrheitserkenntnis nach Augustinus und Thomas
von Aquin (Veröffentlichungen des Katholischen Instituts fur Philosophie, Albertus-
Magnus-Akademie zu Köln, Band 1, Heft 4), Aschendorff
(1933). Die Aristoteleskommentare des Simon von Faversham: Handschriftliche Mitteilungen
(Sitzungsberichte der Bayerischen Akademie der Wissenschaften, Philosophisch-
historische Abteilung, Jahrgang 1933, Heft 3). Verlag der Bayerischen Akademie
der Wissenschaften
(1934a). 'Eine für Examinazwecke abgefasste Quaestioncnsammlung der Pariser
Artistenfakultät aus der ersten Hälfte des 13. Jahrhunderts', Revue Néoscelastique de
Philosophie 36:211-26
(1934b). 'Studien über den Einfluss der aristotelischen Philosophie auf die mit-
telalterlichen Theorien über das Verhältnis von Kirche und Staat* (Sitzungsberichte
der Bayerischen Akademie der Wissenschaften, Phil.-hist. Abt., Heft 2)
(1936a). 'Der Einfluß Alberts des Großen auf das mittelalterliche Geistesleben. Das
deutsche Element in der mittelalterlichen Scholastik und Mystik' in his Mittelalter-
liches Geistesleben II, Hueber
(1937). Bearbeitungen und Auslegungen der aristotelischen Logik aus der Zeit von Peter
Ahaelard bis Petrus Hispanus. Mitteilungen aus Handschriften deutscher Bibliotheken
(Abhandlungen der Preussischen Akademie der Wissenschaften, Phil.-hist. Klasse, 5)
(1938a). 'Kommentare zur aristotelischen Logik aus dem 12. und 13. Jahrhundert in
Ms. lat. fol. 624 der Preussischen Staatsbibliothek in Berlin', (Sitzungsberichte der
Preussischen Akademie der Wissenschaften, Phil.-hist. Klasse, 18:185-210)
(1938b). 'Ungedrucktc lateinische Kommentare zur aristotelischen Topik aus dem 13.
Jahrhundert', Archiv fir Kulturgeschichte 28:210-32
(1939). Methoden und Hilfsmittel des Aristotelesstudiums im Mittelalter (Sitzungsberichte
der Bayerischen Akademie der Wissenschaften, Philosophisch-Historische
Abteilung, Heft 5).
(1940a). Die Sophismataliteratur des 12. und 13. Jahrhunderts mit Textausgabe eines
Sophisma des Boethius von Dacien (BGPM XXXVI, 1), Aschendorff
(1940b). 'Das Studium der aristotelischen Ethik an der Artistenfakultät der Universität
Paris in der ersten Hälfte des 13. Jahrhunderts', Philosophisches Jahrbuch der Görres-
Gesellschqft 55:339~54
(1941). 'Die mittelalterlichen Kommentare zur Politik des Aristoteles'
(Sitzungsberichte der Bayerischen Akademie der Wissenschaften, Philos.-hist. Abt.,
2, Heft 10)
(1944). 'Die Aristoteleskommentare des Heinrich von Brüssel und der Einfluss Alberts
des Grossen auf die mittelalterliche Aristoteleserklärung'. (Sitzungsberichte der
Bayerischen Akademie der Wissenschaften, Philos.-hist. kl. 1943, Heft 10)
(1946). Guglielmo de Moerbeke O.P., il traduttore delle opere di Aristotele (Miscellanea
Historiae Pontificiae, 11)
(1947). 'Ein Tractatus de Universalibus und andere logische Inedita aus dem 12.
Bibliography 921
Jahrhundert im Cod. lat. 2486 der Nationalbibliothek in Wien', Mediaeval Studies

9:56-70
(1950). 'Aristoteles im 12. Jahrhundert', Mediaeval Studies 12:123-62
Graeser, Α. (1973). Die logischen Fragmente des Theophrast (Kleine Texte für Vorlesungen
und Übungen, 191), DeGruyter
Grajewski, M. (1944). The Formal Distinction of John Duns Scotus, Catholic University of
America
Grant, E. (ed.) (1974). A Source Book in Medieval Science, Harvard University Press
Gray, F.J. (1967). 'Peter of Candia (Alexander V, Antipope)', New Catholic Encyclopedia
11:213
Grech, G. M., ed. (1967). The Commentary of Peter of Auvergne on Aristotle's Politics. The
inedited part: Book III, lessons I—VI, Pontifical University of St Thomas Aquinas
Green, Romuald (forthcoming). The logical Treatise 'De obligationibus': An Introduction
with Critical Texts of William of Sherwood and Walter Burley, The Franciscan Institute
Green-Pedersen, N.J. (1973). 'On the Interpretation of Aristotle's Topics in the 13th
Century', CIMAGL 9:1-46
(1974). 'William of Champcaux on Boethius" Topics according to Orleans Bibl. Mun.
266', CIMAGL 13:13-30
(1976). 'The Summulae of John Buridan, Tractatus VI De locis', inj. Pinborg (ed.)
The Logic of John Buridan, Museum Tusculanum
(1977a). 'Discussions about the Status of the Loci Dialectici in Works from the Middle
of the 13th Century,' CIMAGL 20:38-78
(1977b). 'The Doctrine of "maxima propositio" and "locus differentia" in
Commentaries from the 12th Century on Boethius' "Topics" ', Studio Medie-
wistyczne 18:125-63
Gregory of Rimini (1522). Super primum et secundum Sententiarum, Venice (Reprinted
The Franciscan Institute 1955)
Grignashi, M. (1960a). 'Nicolas Oresme et son commentaire à la Politique d'Aristote' in
Album Helen Maud Cam (Études présentées à la Commission Internationale pour
l'histoire des assemblées d'états, 23), Publications universitaires de Louvain
(1060b). 'Un commentaire nominahste de la Politique d'Aristote: Jean Buridan',
Commission International pour l'histoire des assemblées d'états. Ancien Pays et assemblées
d'états 19:123-42
(1966). 'La définition du "civis" dans la scolastique' Commission International pour
l'histoire des assemblées d'états, Ancien pays et assemblées d'états 35:7l-ioo
(1972). 'Les Traductions latines des ouvrages de la logique arabe et l'abrégé
d'Alfarabi' Archives d'histoire doctrinale et littéraire du moyen âge 39:41-107
Grisez, Germain G. (1965). 'The First Principle of Practical Reason', Natural Law Forum
10:168-96
(1969). 'The first principle of practical reason' (abridgement of Grisez 196s) in
Anthony Kenny (ed.) Aquinas: A Collection of Critical Essays, Doubleday
Grosseteste, Robert, see Robert Grosseteste
Grotius, Hugo (1953). De Jure Belli et Pacts, ed. William Whewell, Parker
(1913-25). De iure belli ac pads libri très. Editio nova with translation by F. W. Kelsey
(2 vols, in 4), Clarendon Press
Guelluy, R. (1947). Philosophie et Théologie chez Guillaume dOckkam, Nauwelaerts
Guerault, Martial (1968). Spinoza, I, Aubier-Montaigne
Guerlac, Rita (19-79). Juan Luis Vives Against the Pseudodialecticians: A Humanist Attack on
Medieval Logic (Synthese Historical Library, 18), Reidel
Guillaume, Alfred (1934)· The Summa Philosophiae of al-Shahrastani, Oxford University
922 Bibliography
Press
Guillet, J. (1927). 'La "lumière intellectuelle" d'après S. Thomas', Archives d'histoire
Günther, Robert (1922—3). Early Science at Oxford (2 vols.) (Publications of the Oxford
Historical Society, 77, 78), Clarendon Press
Gutierrez, G. (1939)- De B. Iacobi ViterbiensL· O.E.S.A. Vita, Operibus, et doclrina
Theologica, Analecta Augustiniana
Guy of Rimini (1937-8). De Reprobatione Monarchiae Compositae a Dante, ed. T. Käppeli,
(Q«e/Ifn und Forschungen aus italienischen Archiven und Bibliotheken herausgegeben vom
Deutschen Institut in Rom, 27), Regenberg
Guy Terrena (1926). Quaestio de Magisttrio Infallibilis Romani Pontificis, ed. B. F. M.
Xiberta, Aschendorff
Gyekye, Kwame (1971). 'The Terms "Prima Intentio" and "Secunda Intentio" in
Arabic Logic', Speculum 46:32-8
Hadot, Pierre (1970). 'Forma Essendi: Interprétation philologique et interprétation

philosophique d'une formule de Boèce', Les Études classiques 38:143-56
(1971). Marius Victorinus. Recherches sur sa vie et ses oeuvres. Études Augustiniennes
Hain, Ludwig (1948). Repertorium Bibliographicum in quo Libri Omnes ab Arte Typographica
Inventa usque ad Annum MD (4 vols.), G. G. Görlich
Halm, Charles, ed. (1863). Rhetores Latini minores, Teubner
Hamblin, C. L. (1970). Fallacies, Methuen
Hamilton, Bernice (1963). Political Thought in Sixteenth-Century Spain. A Study of the
political ideas of Vitoria, De Soto, Suàrtz, and Molina, Clarendon Press
Hamlyn, D. W. (1968). Aristotle's 'De anima', Books II and III (with Certain Passages from
Book I), Translated with Introduction and Notes, The Clarendon Press
Hamm, Marlis (1974). 'Engelbert von Admont als Staatstheoretiker', Studien und
Mitteilungen zur Geschichte des Benediktiner-Ordens und seiner Zweige 85:343-495
Hamman, A. (1950). 'La doctrine de l'église et de l'état d'après le Breviloquium
d'Occam', Franziskanische Studien 32:135-41
Harding, Alan (1976). 'The reflection of thirteenth century legal growth in Saint
Thomas's writings' in G. Verbeke and D. Verhelst (eds.) Aquinas and the Problems of
His Time (Mediaevalia Lovaniensia, Series 1, Study 5), Nijhoff
Haskins, C. H. (1927). Studies in the History of Mediaeval Science (Harvard Historical
Studies, 27), Harvard University Press
Hayen, André (1954). L'Intentionnel selon Saint Thomas, 2nd ed., Desclée de Brouwer
Heath, Terrence (1971). 'Logical Grammar, Grammatical Logic, and Humanism in
Three German Universities', Studies in the Renaissance 1819-64
Heidingsfelder, Georg (1921). Albert von Sachsen: Sein Lebensgang und sein Kommentar zur
nikomachischen Ethik des Aristoteles (BGPMXXII, 3-4), Aschendorff
Hendley, Brian Patrick (1967). Wisdom and Eloquence: A New Interpretation of the
Metalogicon of John ofSalisbury, University Microfilms
Henke, E. L. T. and Lindenkohn, G. S., eds. (1851). Petri Abaelardi Sic et Non, Librariae
Academ. Elwertianae
Henry of Bracton (1968). De Legibus et Consuetudinibus Angliae, ed. and tr. Samuel
E. Thome, Bracton on the Laws and Customs of England, Harvard University Press
Henry, Desmond Paul (1958). 'Why "Grammaticus"?', Archivum Latinitatis Medii Aevi,
28:165-80
(1964). The De grammatico of St Anselm: Theory ofParonymy (Publications in Mediaeval
Studies, 18), University of Notre Dame Press
Bibliography 923
(1965). 'Ockham and the Formal Distinction', Franciscan Studies 25:285-92

(1967). The Logic of Saint Anselm, Clarendon Press
(1972). Medieval Logic and Metaphysics, Hutchinson
(1974). Commentary on De grammatico: The Historical-Logical Dimension of a Dialogue of
St Anselm's (Synthese Historical Library, 8), Reidel
Henry, J. (1924). 'Le traditionalisme et Pontologisme à l'Université de Louvain
1835-1865', Annales de l'Institut Supérieur de Philosophie de Louvain 5
Henry of Ghent (1518). QuodUbeta magistri Henrici Goethals a Gandavo doctoris soiemnis (2
vols.), Paris (Reprinted Bibliothèque S. J. de Louvain 1961)
(1520). Summa quaeslionum ordinariatum (2 vols.), Paris (Reprinted Franciscan Institute
1953)
(1613). QuodUbeta aurea, Venice
(1974). 'The absolute and the ordained powers of the pope', ed. John Marrone,
(1978). Sincathegoreumata, ed. in Braakhuis 1978
Hervaeus Natalis (1513). QuodUbeta Hervei: subtilissimi quodlibeta undecim cum octo ipsius
prqfundissimus tractalibus..., Venice
(1647). In Quatuor Libros Sententiarum Commentaria, Quibus adiectus est eiusdem auctoris
Tractatus de Potestate Papae, Paris (Reprinted Gregg 1967)
(1937-8). Df Paupertate Chrkti et Apostolorum, ed. J. G. Sikes, Archives d'histoire
doctrinale et littéraire du moyen âge 12—13 1209-97
(1959). De lurisdictione, ed. L. Hödl, Hueber
Hessen, Johannes (i960). Augustins Metaphysik der Erkenntnis, 2nd ed., E. J. Brill
Hexter, J. H. (1973). The Vision of Politics on the Eve of the Reformation: More, MachiavelU
and Seyssel, Basic Books
Heynick, Valens (1950). 'Ockham-Literatur: 1919-1949', Franziskanische Studien
32:164-83
Heytesbury, William, see William Heytesbury
Hickmann, L. Α. (1971). Logical second intentions - Late Scholastic theories of higher level
predicates. Unpublished Ph.D. dissertation, University of Texas
Hill, Christopher (1965). Intellectual Origins of the English Revolution, Clarendon Press
Hill, W. Speed, ed. (1972). Studies in Richard Hooker, Case Western Reserve University
Press
Hintikka, Jaakko (1973). Time and Necessity: Studies in Aristotle's Theory of Modality,
Clarendon Press
Hintikka, Jaakko, and U. Remes (1974). Ancient Geometrical Method, Reidel
Hintikka, Jaakko, U. Remes, and S. Knuuttila (1977). Aristotle on Modality and
Determinism (Acta Philosophica Fennica, 29.1), North-Holland
Hissette, R. (1976). 'La date de quelques commentaires à l'Éthique', Bulletin de la
Philosophie Médiévale 18:79-83
(1977). Enquête sur les 219 articles condamnés à Paru le 7 Mars 1277, Publications
Universitaires de Louvain
Hocedez, E. (1925). Richard de Middleton. Sa vie, ses oeuvres, sa doctrine, Spicilegium
Sacrum Lovaniense
(1927). 'Gilles de Rome et Henri de Gand sur la distinction réelle (1276-1287)',
Gregorianum 8:358-84
(1928). 'Le premier Quodlibet d'Henri de Gand (1276)', Gregorianum 9:92-117
(1929). 'Deux questions touchant la distinction réelle entre l'essence et l'existence',
Gregorianum 10:365-86
(1930). Aegidii Romani Theoremata de esse et essentia. Texte précédé d'une introduction
924 Bibliography
historique et critique, Museum Lessianum

(1932). 'La condamnation de Gilles de Rome', Recherches de Théologie ancienne et
médiévale 4:34-58
Hoenen, P. (1945). Cosmologia (3rd ed.), apud aedes Universitatis Gregorianae
(1952). Reality and Judgment According to St Thomas, trans. Henry F. Tiblier,
H. Regnery
(1967). 'Descartes' mechanicism', in W. Doney (ed.), Descartes, a collection of critical
essays, Macmillan
Hoeres, W. (1962). Der Wille als reine Vollkommenheit nach Duns Scotus, Pustet
(1965). 'Wesen und Dasein bei Heinrich von Gent und Duns Scotus', Franziskanische
Studien 47:121-86
Hoffmann, Fritz (1971). 'Der Satz als Zeichen der theologischen Aussage bei Holcot.
Crathorn und Gregor von Rimini', Miscellanea Mediaevaiia, 8:296-313
Holdsworth, Richard (1648?). Directionsfor a student in the universitie, Emmanuel College
(Cambridge) MS 48 [cf. Bodleian MS Rawlinson D 200]
(1651). The Valley of Vision, London
(1661). Praelectiones Theologkae, habitae in Collegio Creshamensi apud Londoninenses, ed.
Richard Pearson, London
Holmes, J. D. (1975). 'Some English reactions to the publication o f Aetemi Patris',
Downside Review 93:269-80
Holscher, Ε. Ε. (1932). Dif ethische Umgestaltung der römischen Individual Justitia durch die
universalistische Naturrechtslehre der mittlelalterlkhen Scholastik (Görres-Gessellschaft
Veröffentlichungen der Sektion fur Rechts- u. Staatswissenschaft 59), Schöningh
Honnefelder, L. (1975). 'Duns Scotus: Der Schritt der Philosophie zur scientia trans-
cenden' in W. Kluxen (ed.) Thomas von Aquin im philosophischen Gespräch, Alber
Honoré Bouvet [Bonet] (1949). L'Arbre des Batailles, ed. and tr. G. W. Coopland,
Liverpool University Press
Hooker, Richard (1977-81). Of the Laws of Ecclesiastical Polity in W. Speed Hill (ed.)
The Folger Library Edition of the Works of Richard Hooker (3 vols.), Harvard
University Press
Hopkins, Jasper (1972). A Companion to the Study of St Anselm, University of Minnesota
Press
(1976). Hermeneutical and textual problems in the complete treatises of St Anselm (vol. 4 in
his St Anselm), Edwin Mellin
Horvâth, A. (1929). Eigentumsrecht nach dem hl. Thomas von Aquin, Moser
Hoskin, Michael, and A. G. Molland (1966). 'Swineshead on Falling Bodies: An
Example of Fourteenth Century Physics', British Journal for the History of Science
3:150-82
Hourani, G. (1972). 'Ibn Sïnâ on Necessary and Possible Existence', The Philosophical
Forum 4:74—86
Howell, Wilbur S. (1961). Logic and Rhetoric in England 1500—1700, Crowell and Crowell
Hubert, M. P. (1949). 'Quelques aspects du latin philosophique aux XII e et XIII e siècles',
Revue des études latines 27:211—33
Hubien, Hubert (1976). loannis Buridani Tractatus de Consequentiis Libri iv (Philosophes
Médiévaux, 16), Publications universitaires de Louvain
Hubrecht, G. (1955). 'La "Juste Guerre" dans le Décret de Gratien', Studio Gratiana
3:161-77
Hugh of Saint Victor (1951). On the Sacraments of the Christian Faith (De Sacramentis), tr.
R o y J. Deferrari, The Medieval Academy of America
Hunt, R . W . (1948). "The Introductions to the Artes in the Twelfth Century', Studia
Bibliography 925
Mtdiaevalia in Honorem R. Martin, Brugge

(1949-50). 'Studies on Priscian in the Twelfth Century', Mediaeval and Renaissance
Studies 1 :i94-23i; 2:1-55
(1975)- 'Absoluta, The Summa of Petrus Hispanus on Priscianus Minor', Historiographia
Linguistica 2:1-23
(1977). 'The Preface to the "Speculum Ecclesiae" of Giraldus Cambrensis', Viator
8:189-213
Hutchins, Robert Maynard (1949). Si Tficmej and the World State, Marquette University
Press
Hyma, A. (1951). Renaissance to Reformation, Eerdmans
(1965). The Christian Renaissanct: a History of the 'Devotio Moderna' (2nd edn.), Archon
Books
Hyman, Α., and Walsh, J. J., eds. (1967). Philosophy in the Middle Ages, Harper and Row
Ijsewijn, J. (1971). 'Alexander Hegius (d. 1498): Invectiva in modos significandi', Forum for
Modem Language Studies 7:299-318
(1975). 'The coming of humanism to the Low Countries' in H. A. Oberman and
T. A. Brady (eds.) Itinerarium Italkum, Brill
IJsewijn-Jacobs, J. (1976). 'Magistri Anthonii Haneron (ca. 1400-1490) opera gram-
matica et rhetorica, IP, Humanistka Lovaniensia 25:1-83
Ijsewijn, J. and Paquet, J. (eds.). (1978). Tfie Universities in the Late Middle Ages.
(Mediaevalia Lovaniensia 6), Leuven University Press
Imle, Fanny (1930). 'Die Gemeinschaftidee in der Theologie des hl. Bonaventura',
Franziskanische Studien 17:325-41
Incerti auctores (1977). Quaestiones super Sophisticos Elenchos, ed. S. Ebbesen (Corpus
Philosophorum Danicorum Medii Aevi, VII), Gad
Isaac, Jean (1948). 'Sur la connaissance de la vérité'. Revue des sciences philosophiques et
théologiques 32:337-50
(1963). Lf Pfri hermeneias en occident de Boèce à Saint Thomas (Bibliothèque Thomiste
29), J. Vrin
Isidore of Seville (1911). Isidori Hùpalensis episcopi Etymologiarum sive Originum libri xx,
ed. W. M. Lindsay, Clarendon Press
lung, N. (i93i). Un franciscain théologien du pouvoir pontifical au XIV siècle: Alvaro Ptlayo,
évique et pénitencier de Jean XXII,]. Vrin
Jacob, E. F. (1963). Essays in the Conciliar Epoch, revised ed.. University of Notre Dame
Press
Jacquin, R. (1943). Taparelli, Lethielleux
Jalbert, Guy (1961). Nécessité et contingence chez saint Thomas d'Aquin et chez ses pré-
décesseurs. Éditions de l'Université d'Ottawa
James of Viterbo (1926). De Regimine Christiano in H.-X. Arquillière (ed.) L· plus ancien
traité de l'église: Jacques de Viterbe, De Regimine Christiano (1301-2), G. Beauchesne
(1968). Jacobi de Viterbio O.E.S.A. Disputatio prima de quolibet, ed. E. Ypma,
Augustinus-Verlag
(1973)· Jacobi de Viterbio O.E.S.A. Disputatio tertia de quolibet, ed. E. Ypma,
Augustinus-Verlag
Jardine, Lisa (1974). 'The Place of Dialectic Teaching in Sixteenth-Century Cambridge',
Studies in the Renaissance 21131-62
(1975a). Francis Bacon: Discovery and the Art of Discourse, Cambridge University Press
(1975b). 'Humanism and the Sixteenth-Century Cambridge Arts Course', History of
926 Bibliography
Education 4:16-31
(1976). 'Humanism and Dialectic in Sixteenth-Century Cambridge: A Preliminary
Investigation' in R. R. Bolgar (ed.) Classical Influences on European Culture, AD
1500-1700, Cambridge University Press
(1977). 'Lorenzo Valla and the Intellectual Origins of Humanist Dialectic', Journal of
the History of Philosophy 15:143-64
(forthcoming). 'Dialectic or dialectical rhetoric? Agostino Nifo's Criticism of Lorenzo
Valla', Proceedings of the First International Conference of Rhetoric (1977)
Jeauneau, E. (1973). Lectio Philosophorum (Recherches sur l'École de Chartres), A. M.
Hakkert
John Blund (1970). Johannes Blund: Tractatus de anima, ed. D. A. Callus, O. P., and
R. W. Hunt (Auctores Britannici medii aevi, 2), British Academy
John Buridan (1499). Perutile compendium totius logicae Joannis Buridani cum praeclarissima
solertissimi viri Joannis Dorp expositione, Venice (Reprinted Minerva 1965)
(after 1500?). Sophismata Buridani, [Paris]
(1509). Quaestiones super octo libros Physkorum Arktotelis, Paris
(1513). Quaestiones super decern libros Ethicorum, Paris (Reprinted Minerva 1968)
(1588). In Metaphysicam Aristotelis Quaestiones, Paris (Reprinted Minerva 1964).
(1640). Quaestiones in octo libros Politicorum Aristotelis, ed. G. Turner, Oxford
(1957). 'Giovanni Buridano: "Tractatus de suppositionibus"', ed. Maria Elena Reina,
Rivista critica di storia dellafilosofia12:175-208; 323-52
(1966). Sophisms on Meaning and Truth, tr. T. K. Scott, Appleton-Century-Crofts
(1967). Questions on Aristotle's Metaphysics (Book II, q. 1), tr. J. Walsh in A. Hyman
and J. Walsh (eds.) Philosophy in the Middle Ages, Harper and Row
(1976). Tractatus de consequentiis, ed. H. Hubien (Philosophes médiévaux, 16),
Publications universitaires de Louvain
(1977). Johannes Buridanus: Sophkmata, ed. T. K. Scott (Grammatica Speculativa, 1),
Frommann-Holzboog
John Capreolus (1900—8). Defensiones theologiae divi Thomae Aquinatis, ed. C. Paban and
T. Pègues, Cattier (Reprinted Minerva 1967)
John of Dacia (1955). Opera, ed. A. Otto (Corpus Philosophorum Danicorum Medii
Aevi, 1), Gad
John Damascene (1953). Dialectica, ed. Owen A. Colligan (Franciscan Institute Text
Series, 6), The Franciscan Institute
(1955). De Fide Orthodoxa, ed. E. M. Buytaert, E. Nauwelaerts / F. Schöningh
John Duns Scotus (1639a). Opera omnia, ed. L. Wadding (12 vols.). Lyons (Reprinted
Georg Olms 1968)
(1639b). Quaestiones miscellaneae de formalitatibus in John Duns Scotus (1639a) vol. 3
(1639c). Quaestiones subtilissimae super libros Metaphyskorum Aristotelis in John Duns
Scotus (1639a) vol. 4
(1630C1). Reportata Parisiensia in John Duns Scotus (1639a) vol. 11
(1891—5). Opera omnia. Editio nova juxta editionem Waddingi XII tomos con-
tinentem a patribus Franciscanis de observantia accurate recognita (26 vols, in 13),
Vives
(1891). In duos libros Perihermenias, operis secundi, quod appellant, Quaestiones octo in
John Duns Scotus (1891-5) vol. I
(1895). Quaestiones quodlibetales in John Duns Scotus (1891-5) vol. XXV
(1950-). Opera omnia, ed. C. Balic fi al., Vatican Scotistic Commission
(1950). Ordinatio, Prologus, in John Duns Scotus (1950-) vol. I
(1950-8). Ordinatio, Prologus - Dist. 48, in John Duns Scotus (1950-) vols. 1-6
(1954). Ordinatio, Dist. 3, in John Duns Scotus (1950-) vol. 3
Bibliography 927
(1960a). Lectura in librum primum Sententiarum, Prologus - Dist. 7, in John Duns Scotus
(1950-) vol. 16
(1960b). Lectura in librum primum Sententiarum, Dist. 8ff, in John Duns Scotus (1950-)
vol. 17
(1966). De Primo Principio in A. Wolter (ed.) A Treatise on Cod as First Principle,
Franciscan Herald
(1968). Questiones Quodlibetales in F. Alluntis (ed. and tr.) Obras del Doctor Sutiljuan
Duns Escoto, Biblioteca de Autores Cristianos
(1975). God and Creatures: The Quodlibetal Questions tr. F. Alluntis and A. Wolter,
John Gerson (1953). On the unity of the Church, tr. and ed. James K. Cameron in
Matthew Spinka (ed.) Advocates of Reform: From Wyclif to Erasmus, Westminster
Press
(1965). De Potestate Ecclesiastica in P. Glorieux (ed.) Oeuvres complètes (6:210-50),
Desclée
(1978). De lurisdictione Spirituali et Temporali in Posthumus Meyjes, Jean Gerson et
l'Assemblée de Vincennes (Studies in Medieval and Reformation Thought, 26), Brill
John of La Rochelle (1882). Summa de anima, ed. T. Domenichelli
John of Legnano (1917). Tractatus de bello, ed. T. E. Holland, Oxford University Press
John le Page (1979). Syncategoremata (excerpted in Braakhuis 1979)
John Major (1706a). Df Potestate Papae in Temporalibus in L. Du Pin (ed.) Jean Gerson,
Opera Omnia, 2: cols. 1145-64, Sumptibus Societatis
(1706b). De Statu et Potestate Ecclesiae in L. Du Pin (ed.) Jean Gerson, Opera Omnia, 2:
cols. 1121-30, Sumptibus Societatis
(1938). Le traité 'De l'infini' de Jean Mair, Nouvelle édition avec traduction et annotations,
ed. and tr. H, Elie, J. Vrin
John of Paris (1941). Le Correctorium Corruptorii 'Circa' de Jean Quidort de Paris, ed. J. P.
Muller (Studia Anselmiana, 12-13), Herder
(1968). De Regia Potestate et Papali in F. Bleienstein (ed.) Johannes Quidort von Park über
königliche und päpstliche Gewalt (De Regia Potestate et Papali) (Frankfurter Studien
zur Wissenschaft von der Politik), E. Klett
(1971). On Royal and Papal Powtr, tr. with an introduction by J. A. Watt, Pontifical
Institute of Mediaeval Studies
(1974). John of Paris on Royal and Papal Power, tr. Arthur Monahan, Columbia
University Press
John Peckham (1948). Tractatus de anima loannis Pecham, ed. G. Melani, Edizioni 'Studi
Francescani's. Francesco
John Philoponus (1905). In Aristotelis Analytica Priora Commentaria, ed. M. Wallies
(Commentaria in Aristotelem Graeca, XIII. 2), G. Reimer
(1909). In Aristotelis Analytica Posteriora commentaria, ed. M. Wallies (Commentaria in
Aristotelem Graeca, XIII. 3), G. Reimer
John Quidort, see John of Paris
John of Reading (1969). 'Quaestio loannis de Reading de Necessitate Specierum
Intelligibilium Defensio Doctrinae Scoti", ed. Gedeon Gâl, Franciscan Studies
29:66-156
John of St Thomas (1930-7). Cursus Philosophicus Thomisticus, ed. B. Reiser, Marietti
(1931-53). Cursus Theologicus, Desclée
(T955>)· Outlines of Formal Logic, tr. with an introduction by Francis C. Wade,
Marquette University Press
(1955b). The Material Logic of John of St Thomas, tr. Y. R. Simon, J. J. Glanville, and
928 Bibliography
G. D . Hollenhorst, University of Chicago Press
John ofSalisbury (1909). Policraticus, ed. C. C . J . W e b b (2 vols.), Clarendon Press
(1927). The Statesman's Book of John ofSalisbury, tr. by John Dickinson (Policraticus 4 - 6
and selections from 7—8), Knopf
(1929). Metalogicon, ed. C. C.J. W e b b , Clarendon Press
(1938). Frivolities of Courtiers and Footprints of Philosophers, tr. Joseph B. Pike
(Policraticus 1—3 and selections from 7—8), Oxford University Press
(1962). The Metalogicon, tr. Daniel D. McGarry, University of California Press
Johnston, Herbert (i960). 'Intellectual Abstraction in St Albert', Philosophical Studies
(Maynooth) 10:204-12
John of Turrecremata (1489). Summa dc ecclesia, R o m e
John Versor (1495). Quaestiones super libros Ethicorum Aristotelis, Cologne
John Wyclif (1871). TTie Church and Her Members in Thomas Arnold (ed.) Select English
Works of John Wyclif, vol. 3, Clarendon Press
(1885—1904). De Civili Dominio (4 vols.), Trübner (Reprinted Johnson Reprint Co.
1966)
(1887). Tractatus de Officio Regis, Trübner
(1899). Tractatus de logica, ed. M. H. Dziewicki (Wyclif Society, 3), Trübner
(1907). Tractatus de Potestate Papae, Trübner (Reprinted Johnson Reprint Co. 1966)
Jolivet, Jean (1969). Arts du langage et théologie chez Abelard (Études de Philosophie
Médiévale, 57), J. Vrin
(1974). 'Comparaison des théories du langage chez Abelard et chez les nominalistes du
XlVe siècle', in E. M. Buytaert (ed.), Peter Abelard. Proceedings of the international
conference Louvain May 10-12, 1971, Leuven University Press and Martinus Nijhoff
(1975) 'Vues médiévales sur les paronymes', Revue Internationale de Philosophie
113:222-42
Joseph, H. W . B . (1916). An Introduction to Logic (2nd ed.), Clarendon Press
Joynson, James Turner (1975). Ideology, Reason, and the Limitation of War: Religious and
Secular Concepts, 1200—1740, Princeton University Press
Juan de Mariana (1599). De Rege et Regis Institutione, P. Rodericus
(1948). The King and the Education of the King, tr. George Albert Moore, Country
Dollar Press
Kaiser, Rudolf (1962). ' Z u r Frage der eigenen Anschauung Alberts d. Gr. in. seinen
philosophischen Kommentaren', Freiburger Zeitschrift für Philosophie und Theologie
9:53-62
Kaluz a, Zenon (1979). 'Le problème du " D e u m non esse" chez Etienne de Chaumont,
Nicolas Aston et Thomas Bradwardine', Mediaevalia Philosophica Polonorum
24:3-19·
Kaminsky, H. (1963). 'Wyclifism as ideology of revolution', Church History 32:57-74
Kantorowicz, E. H. (1955). 'Mysteries of state: an absolutist concept and its late medieval
origins'. Harvard Theological Review 48:65—91
(1957). The King's Two Bodies. A Study in Mediaeval Political Theology, Princeton
University Press
Kasper. Franz (1967). Das subjektive Recht-Begriffsbildung und Bedeutungsmehrheit
(Freiburger Rechts- und Staatswissenschaftliche Abhandlungen, 25). C. F. Muller
Kearney, H . (1970). Scholars and Gentlemen, Cornell University Press
Keckermann, Bartholomew (1613). Systema Systematum, Hanover
Keeler, Leo (1934). The Problem of Error from Plato to Kant, Gregorian University Press
Keen M . A. (1965). The Laws of War in the Late Middle Ages, Routledge and Kegan Paul
Bibliography 929
Reicher, Otto (1913). 'Zur Lehre der ältesten Franziskanertheologen vom "intellectus
agens" ' in Abhandlungen aus dem Gebiete der Philosophie und ihrer Geschichte. Eine
Festgabe zum 70. Geburtstag George Freiherrn von Hertling, Herder
Kelley, F. E. (forthcoming) 'Some Observations on the fictum theory in Ockham and its
Relation to Hervaeus Natalis' Franciscan Studies
Kennan, Elizabeth (1967). 'The Df Consideratione of St Bernard of Clairvaux and the
papacy in the mid-twelfth century: a review of scholarship', Traditio 23:73-115
Kennedy, Leonard A. (1959-60). 'The Nature of the Human Intellect According to St
Albert the Great', The Modern Schoolman 37:121-37
(1962-3). 'St Albert the Great's Doctrine of Divine Illumination', The Modern
Schoolman 40:23-37
Kenny, Anthony (1969a). 'Intellect and Imagination in Aquinas' in Anthony Kenny
(ed.) Aquinas: A Collection of Critical Essays, Doubleday
(1969b). The Five Ways: St Thomas Aquinas' Proofs of God's Existence, Routledge and
Kegan Paul
ed. (1970). Aquinas. A Collection of Critical Essays, Macmillan
Kidd, I. G. (1978). 'Posidonius and Logic', in Les stoïciens et leur logique. Actes du colloque
de Chantilly 18-22 septembre 1976,]. Vrin
Kilvington, Richard, iff Richard Kilvington
Kilwardby, Robert, see Robert Kilwardby
Kirshner, Julius (1973). 'Civitas sibi faciat civem: Bartolus of Sassoferrato's doctrine on the
making of a citizen', Speculum 48:694-713
Kitchel, M. Jean (1974). 'The "De potentiis animae" of Walter Burley', Mediaeval
Studies 33:85-113
Kleineidam, E. (1930). Das Problem der hylomorphen Zusammensetzung der geistigen
Substanzen im 13. Jahrhundert, behandelt bis Thomas von Aquin, Breslau (Dissertation)
( I 0 73)· 'Geschichte der Wissenschaft im mittelalterlichen Erfurt'. Geschichte Thüringens
II 2:150-87, Böhlau
Kleutgen, J. (i860) Dif Philosophie der Vorzeit vertheidigt (2 vols.), Thcissing'sche
Buchhandlung
Klingsford, C. L. (1949-50). 'Lavenham, or Lavynham, Richard', Dictionary of National
Biography (reprint, Oxford University Press), 11:652-3
Klubertanz, George P. (1952). The Discursive Power: Sources and Doctrine of the "Vis
Cogitativa" According to St Thomas Aquinas, The Modern Schoolman
(1952a). 'St Thomas and the Knowledge of the Singular', New Scholasticism
26:135-66
(1953). The Philosophy of Human Nature, Appleton-Century-Crofts
(1954a). 'St Thomas on Learning Metaphysics', Gregorianum 35 :3—17
(1954b). "The Teaching of Thomistic Metaphysics', Gregorianum 35:187-205
(i960). St Thomas Aquinas on Analogy, Loyola University Press
Kluge, E. W. (1973-4). 'William of Ockham's Commentary on Porphyry. Introduction
and English Translation', Franciscan Studies 33 (171-254) and 34 (306-82)
Kluxen, W. (1964). Philosophische Ethik bei Thomas von Aquin, Grünewald
(1978). 'Glück und Glücksteilhabe. Zur Rezeption der aristotelischen Glückslehre
bei Thomas von Aquin' in G. Bien (ed.), Dif Fra^f nach dem Glück, Frommann-
Holzboog
Kneale, William and Martha (1962). The Development of Logic, Clarendon Press
Knecpkcns, C. H. (1977). "The Relatio simplex in the Grammatical Tracts of the Late
Twelfth and Early Thirteenth Century', Vivarium 15:1—30
(1978). 'Master Guido and his View on Government: On Twelfth Century Linguistic
930 Bibliography
Thought', Vivarium 16:108-41

Knudsen, Christian (1975). 'Ein ockhamkritischer Text zu Signifikation und Supposition
und zum Verhältnis von erster und zweiter Intention', CIMAGL 14:1-26
Knuuttila, Simo (1975). Alka Ja Modaliteetti Aristottlistssa Skolastiikassa, Societas
Missiologica et Oecumenica Fennica
(1978). 'The Statistical Interpretation of Modality in Averroes and Thomas Aquinas',
Ajatus 37:79-98
(1981). 'Time and Modality in Scholasticism' in S. Knuuttila, ed., Reforging the Great
Chain of Being: Studies of the History of Modal Theories (Synthese Historical Library
20), Reidel
Knuuttila, Simo, and Anja Inkeri Lehtinen (1979). 'Plato in Infinitum Remisse Incipit
Esse Albus. New Texts on the Late Medieval Discussion on the Concept of Infinity
in Sophismata Literature', in E. Saarinen, R. Hilpinen, I. Nüniluoto and
M. Provence Hintikka (eds.) Essays in Honour of Jaakko Hintikka, Reidel
Koch, J. (1930). Durandi de S. Porciano Ο. P. Tractatus de habitibus Quaestio Quarta,
(Opuscula et Textus, Series scolastica, 8), Aschendorff
(1944). Giles of Rome. Errores Philosophorum, Marquette University Press
Kölmel, W. (1962). Wilhelm Ockham und seine kirchenpolitischen Schriften, Ludgerus-
Verlag Hubert Wingen
(1970). Regimen Christianum: Weg und Ergebnisse des Gewaltenverhältnkses und des
Gewaltenverständnisses (8-14. Jahrhundert), De Gruyter
König, E. (1970). 'Aristoteles' erste Philosophie als universale Wissenschaft von den
ΑΡΧΑΓ, Archiv fir Geschichte der Philosophie 52:225-46
Korolec, J. B. (1974a). 'Le commentaire de Jean Buridan sur l'Éthique à Nicomaque et
l'université de Cracovie dans la première moitié du XVe siècle', Organon
10:187-208
(1974b). 'La philosophie de la liberté de Jean Buridan', Studio Mediewistyczne
15:109-52
(i97S)- 'Les principes de la philosophie morale de Jean Buridan', Mediaevalia
Philosophica Polonorum 21:53-72
Kosegarten, J. G. L. (1857). Geschichte der Universität Greifswald I, Greifswald
Kotarbinski, T. (1967). 'Notes on the Development of Formal Logic in Poland in the
Years 1900-1939' in S. McCall (ed.) Polish Logic 1920—1939, Clarendon Press
Kretzmann, Norman (1966). William of Sherwood's Introduction to Logic, University of
Minnesota Press
(1967). 'History of Semantics' in P. Edwards (ed.) Encyclopedia of Philosophy,
Macmillan
(1968). William of Sherwood's Treatise on Syncategorematic Words, University of
Minnesota Press
(1970). 'Medieval Logicians on the Meaning of the Propositio', Journal of Philosophy
67:767-87
(!975)· 'Transformationalism and the Port-Royal Grammar', i n j . Rieux and Β. Ε.
Rollin (eds.), General and Rational Grammar: the Port-Royal Grammar (Janua
Linguarum, Series Minor, 208), Mouton
(1976). 'Incipit/Desinit* in P. Machamer and R. Turnbull (eds.) Motion and Time,
Space and Matter, Ohio State University Press
(1977). 'Socrates is Whiter than Plato Begins to be White', Nous Ii =3-15
(forthcoming a). 'The Culmination of the Old Logic in Peter Abelard' in R. L.
Benson and G. Constable (eds.) The Renaissance of the Twelfth Century, Harvard
University Press
Bibliography 931
ed., (forthcoming b). Infinity and Continuity in Ancient and Medieval Thought, Cornell
University Press
(forthcoming c). 'Richard Kilvington and the Logic of Instantaneous Speed', in Studi
sui XIV secolo in memoria di Anneliese Maier, ed. A. Maierù and A. Paravicini-
Bagliani, Edizioni di Storia ε Letteratura
(forthcoming d). 'Continuity, Contrariety, Contradiction, and Change', in Kretzmann
(forthcoming b)
Kretzmann, Norman, John Longeway, Eleonore Stump, and John Van Dyk (1975).
[Review of De Rijk 1972], The Philosophical Review 84:560-7
(1978). 'L. M. De Rijk on Peter of Spain', Journal of the History of Philosophy
16:325-33
Kristeller, Paul Oskar (1944-5). 'Humanism and scholasticism in the Italian
Renaissance*, Byzantion 17:346-74
(1952). 'Petrarch's "Averroists": A Note on the History of Aristotelianism in Venice,
Padua, and Bologna', Mélanges Augustin Renaudet, Bibliothèque d'Humanisme et
Renaissance 14:59-65
(1956). Studies in Renaissance Thought and Letters (Storia e letteratura: raccolta di studi e
testi, 54), Edizione di storia e letteratura
(1961a). Renaissance Thought: the Classic, Scholastic, and Humanist Strains, Harper &
Row
(1961b). 'The Moral Thought of Renaissance Humanism' in Chapters in Western
Civilization (3rd ed.), Columbia University Press
(1964). Eight Philosophers of the Italian Renaissance, Stanford University Press
(1972). Renaissance Concepts of Man, Harper Torchbooks
(1974). Medieval Aspects of Renaissance Learning (Duke Monographs in Medieval and
Renaissance Studies, 1), Duke University Press
Kuksewicz, Zdzislaw (1968). De Siger de Brabant à Jacques de Plaisanct - la théorie dt
l'intellect chez les averrotstes latins des XUF et XIV s., Ossolineum
(1973). Albertyzm i tomizm w XV wieku w Krakowie i Kolonii, Wroclaw
Kuttner, S (1976). 'Gratian and Plato' in C. N. L. Brooke et al. (eds.) Church and
Government in the Middle Ages, Cambridge University Press
Labis, F. (1845). 'Théorie des idées selon M. Gioberti', Retw Catholique 3:8-19, 62-73
La Brière, Y. de (1930) 'La conception de la paix et de la guerre chez Saint Augustin'
Revue de Philosophie 30, 557-72
LaBrosse, Olivier de (1965). Le Papf tt le concile, Éditions du Cerf
Lachance, Louis (1933). Le concept de Droit selon Aristote et S. Thomas, Sirey
(1965). L'humanisme politique de saint Thomas d'Aquin, individu et État (rev. ed.), Sirey
Ladner, Gerhart Β. (1967)· 'Homo viator: medieval ideas on alienation and order'.
Speculum 42:233-59
Lagarde, G. de (1934-46). La naissance de l'esprit laique au déclin du moyen âge (6 vols.),
Saint-Paul-Trois-Châteaux
(1943-5)· 'La philosophie sociale d'Henri de Gand et Godefroid de Fontaines',
Archives de l'histoire doctrinale et littéraire du moyen âge 14:73-142
(1956-70). La naissance de l'esprit laïque au déclin du moyen âge (3rd ed.),
E. Nauwelaerts
Lakner, F. (1933). 'Kleutgen und die kirchliche Wissenschaft Deutschlands im 19.
Jahrhundert', Zeitschriftfirkatltolische Theologie 57:161-214
Larabert of Auxerre (1971). Logica (Summa Lamberti), ed. F. Alessio (Pubblicazioni della
facoltà di lettere e filosofia dell'Università di Milano, 59), La Nuova Italia Editrice
932 Bibliography
Lamennais, F. de (1819). Essai sur l'indifférence en matière de religion (5th ed., 4 vols.),
Tourachon-Molin et H. Seguin
Landgraf, A. M. (1935). 'Quelques collections de Quaestiones de seconde moitié du XII e
siècle'. Revue de théologie ancienne et médiévale 7 : 1 2 4 - 7
Lang, A. (1937). Heinrich Totting von Oyta (BGPM, xxx, 4-5), Aschendorff
(1962). Die Entfaltung des apologetischen Problems in der Scholastik des Mittelalters, Herder
(1964). Die theologische Prinzipienlehre der mittelalterlichen Scholastik, Herder
Langhadc, J., and Grignaschi, M., eds. (1971). Alfarabi, deux ouvrages inédits sur L·
Rhétorique, Dar el-Machreq (Beirut)
Lapparent, P. de (1940-2). 'L'oeuvre politique de François de Mayronnes', Archives
d'histoire doctrinale et littéraire du moyen âge 15-17:5-151
Laso, José Alvarez (1952). La Filosofia de las Mathematicas en Santo Tomas, Editorial Jus
Latham, R . Ε. (1965). Revised Medieval Latin Word-List from British and Irish Sources,
Oxford University Press
La Vaissière, J. de (1925). 'Le sens du mot "verbe mental" dans les écrits de Saint
Thomas', Archives de philosophie, vol. 2, Cahier 2, Beauchesne
Lawrence, C. H. (i960). St Edmund of Abingdon, Clarendon Press
Le Bras, Gabriel, ed. (1955-). Histoire du droit et des institutions de l'Église en Occident, Sirey
Leclercq.J. (1931). 'Note sur les théories politiques d'Alvaro Pelayo: à propos d'une
these récente', Recherches de science religieuse 2 1 : 5 8 2 - 9
(1942): Jean de Paris et l'ecclésiologie du XIIFsiècle,]. Vrin
Leff, Gordon (1957). Bradwardine and the Pelagians. A study of His De causa Dei and its
Opponents, Cambridge University Press
(1961). Gregory of Rimini: Tradition and Innovation in Fourteenth Century Thought,
Manchester University Press
(1967a). Heresy in the Later Middle Ages (2 vols.), Manchester University Press
(1967b). 'The apostolic ideal in later medieval ecclesiology'. Journal of Theological
Studies, New Series 18:58-82
(1967c). 'Pierre d'Ailly* in P. Edwards (ed.) Encyclopedia of Philosophy, Macmillan
(1975). William of Ockham: The Metamorphosis of Scholastic Discourse, Manchester
University Press
Lefranc, Abel (1893). Histoire du Collège de France, Hachette
Le Goff, J. (1957). Les Intellectuels au moyen âge, Éditions du Seuil
Lerner, Ralph, and Muhsin Mahdi, eds. (1963). Medkval Political Philosophy, Cornell
University Press
Lemay, R . (1963). 'Dans l'Espagne du XI e siècle. Les Traductions de l'arabe au latin'.
Annales, Économies, Sociétés, Civilisations 18:639-65
Lewis, C. J. T. (1975). The Merton Tradition and Kinematics in late Sixteenth and Early
Seventeenth Century Italy, unpublished Ph.D. thesis, University o f London
(1976). "The Fortunes o f Richard Swineshead in the Time of Galileo', AnnaL· of Science
33:561-84
Lewis, C. S. (1954). English Literature in the Sixteenth Century excluding Drama, Clarendon
Press
Lewis, Ewart (1938). 'Organic tendencies in medieval political thought*, American
Political Science Review 32:849-76
(1939-40). 'Natural Law and expediency', Ethics 1:144-63
(1954). Medieval Political Ideas. 2 vols., Routledge Sc Kegan Paul
(1963). 'The "positivism" of Marsiglio o f Padua', Speculum 38:541-82
Lewry, O. (forthcoming), 'The Oxford Condemnations of 1277 in Grammar and Logic'
in Proceedings of the 4th European Symposium on Medieval Logic and Semantics
Bibliography 933
Libera, A. de (forthcoming). 'Roger Bacon et le problème de l'appellatio univoca',
(Procetdings of the 4th European Symposium on Medieval Logic and Semantks)
Liberatore, M. (1855). Institutiones philosophicae, Typis Propagand3 Fidei
Liebeschütz, Hans (1950). Humanism in the Life and Writings of John ofSalisbury, Warburg
Institute
(1967). 'Western Christian Thought from Boethius to Anselm' in A. H. Armstrong
(ed.) The Cambridge History of Later Greek and Early Mtdieval Philosophy, Cambridge
University Press
(1968). 'Chartres und Bologna: Naturbegriff und Staatsidee bei Johannes von
Salisbury', Archiv fir Kulturgeschichte 50:3-32
Lindeboom, J. (1913). Het bijbelsch humanisme in Nederland, Adriani
Linhardt, R. (1932). Dif Sozialprinzipien des hl. Thomas von Aquino, Herder
Lio, Ermengildo (1950). 'De démentis traditionalibus justitiae in primaeva schola
Franciscana', Franciscan Studies 10:164-85
(1957). Estne obligatio justitiae subvenire mueris?: Quaestionis positio et evolutio a Petro
Lombardo ad S. Thomam ex tribus S. Augustini textibus, Desclée
(1959)· 'De jure ut objecto justitiae apud S. Thomam II-II, q. 57, a. 1', .Ä/>oIIinaris
32:16-71
Little, Α. (1949). The Platonic Heritage of Thomism, Golden Eagle Books Limited,
Standard House
Little, A. G., and F. Pelster (1934). Oxford Theology and Theologians c. A. D. 1282-1302
(publications of the Oxford Historical Society, 96), Clarendon Press
Lloyd, A. C. (1956). 'Neoplatonic logic and Aristotelian Logic', Phronesis 1:58-72;
146-60
(1967). 'Porphyry and lamblichus' in A. H. Armstrong (ed.) The Cambridge History of
Later Greek and Early Medieval Philosophy, Cambridge University
Lohr, Charles H. (1965). 'Logica Algazelis: Introduction and Critical Text', Traditio
211223-90
(1967-74). Medieval Latin Aristotle Commentaries (Authors A - F in Traditio 23:313-
413; Authors G - I, ibid. 24:149-245; Authors Jacobus -Johannes Juff, ibid. 26:
135-216; Authors Johannes de Kanthi - Myngodus, ibid. 27:251-351; Authors
Narcissus - Richardus, ibid. 28:281-396; Authors Robertus - Wilgelmus, ibid.
29:93-197; Supplementary Authors, ibid. 30:119-44.)
(1974-). Renaissance Latin Aristotle Commentaries. (Authors A - Β in Studies in the
Renaissance 21 (1974) :228—89; Authors C in Renaissance Quarterly 28 (1975) = 689-
741; Authors D - F, ibid. 29 (1976) : 714-45; Authors G - Κ, ibid. 30 (1977) :68ι-
741; Authors L - M , ibid. 31 (1978):532-603; Authors Ν - Ph, ibid. 32 (i979):
529-80.)
Lombard, Peter, sfe Peter Lombard
Lonergan, B. J. F. (1957)· Insight. A Study of Human Understanding, Longmans Green
(1967). Verbum: Word and Idea in Aquinas, ed. David Burrell, University of Notre
Dame Press
Longeway, John (1977). Simon ofFaversham's Questions on the Posterior Analytics: A
Thirteenth-Century View of Science, unpublished Ph. D. Dissertation, Cornell
University
Longpré, Ε. (1922) 'Pietro de Trabibus, un discepolo di Pier Giovanni Olivi', Studi
Francescani 19
(1944). The Kingship of Jesus Chrkt according to St Bonaventure and Blessed Duns Scotus,
tr. Daniel J. Barry, St Anthony Guild Press
Loserth, Johann (1918). 'Johann von Wyclif und Robert Grosseteste, Bischof von
934 Bibliography
Lincoln,' Sitzungsberichte der Akademie der Wissenschaften Wien, phil.-hist. Kl. 186
Lottin, O. (1925). 'La définition classique de la loi (comment, de la lallae, 5)0)', Revue
Néoscolastique de Philosophie 26:129-45; 244-73
(1931). Les droit naturel chez S. Thomas d'Aquin et ses prédécesseurs (2nd ed.), E. Dillon
(1932). 'La composition hylémorphique des substances spirituelles', Revue Néosco-
lastique de Philosophie 34:21-41
(1942-60). Psychologie et morale aux XIIe et XIII'siècles (6 vols, in 7), Abbaye du Mont
César
(1948). [in Lottin 1942-60]
(1954). 'Les vertus morales acquises sont-elles de vraies vertus?'. Recherches de Théologie
ancienne et médiévale 21:101 -29
(1957). Psychologie et morale aux XII' et XIII' siècles (2nd ed.), vol I, Problèmes de
Psychologie
(1959)· [in Lottin 1942-60]
Lotz, J. B., ed. (1955). Kant und die Scholastik heute, Verlag Berchmannskolleg
Loux, M. J. (1974). Ockham's Theory of Terms. Part I of the Summa logicae, tr. with an
introduction, University of Notre Dame Press
Lovejoy, A. O. (1936). The Great Chain of Being. A Study of the History of an Idea,
Harvard University Press
Lukasiewicz, Jan (1935). 'Zur Geschichte der Aussagenlogik', Erkenntnis $ : 111-31
(1957). Aristotle's Syllogistic from the Standpoint of Modern Formal Logic (2nd ed.),
Clarendon Press
Lupoid of Bebenberg (1566). De lure Regni et Imperii Romani in S. Schardius (ed.) Df
lurkdictione, Auctoritate et Praeeminentia Imperiali ac Potestate Ecclesiastica, Basel
Luquet, G. H. (1901). 'Hermann l'Allemand', Revue de l'histoire des religions 441407-22
Luscombe, D. E. (1969). The School of Peter Abelard: The Influence of Abelard's Thought in
the Early Scholastic Period (Cambridge Studies in Medieval Life and Thought, 14),
Cambridge University Press
(1971). Peter Abelard's Ethics, Clarendon Press
Lynch, J. E. (1972). The Theory of Knowledge of Vital du Four, The Franciscan Institute
Lynch, John Patrick (1972). Aristotle's School, University of California Press
Lyons, J. (1977). Semantics I-II, Cambridge University Press
McAlister, L., ed. (1976). The Philosophy of Brentano, Duckworth

McCabe, H. (1954). 'Categories', Dominican Studies 7: 147-79
McCall, R. S. (1963). Aristotle's Modal Syllogisms (Studies in Logic and the Foundations
of Mathematics), North-Holland
MacClintock, Stuart (1954-5). 'Heresy and Epithet: An Approach to the Problem of
Latin Averroism', The Review of Metaphysics 8:176-99, 342-56 and 526-45
(1956). Perversity and Error: Studies on the " Averroist" John of Jandun (Indiana
University Publications, Humanities Series, No. 37), Indiana University Press
McConica, J. (1979). 'Humanism and Aristotle in Tudor Oxford', The English Historical
Review 94:291-317
McCready, William D. (1973). 'Papal plentitudo potestatis and the source of temporal
authority in late medieval papal bierocratic theory', Speculum 48:654-74
(1974). 'The problem of the empire in Augustinus Triumphus and late medieval papal
hierocratic theory', Traditio 30:325-49
(1975)· 'Papalists and anti-papalists: aspects of the Church - State controversy in the
later middle ages', Viator 6:241-73
(1977). 'The papal sovereign in the ecclesiology of Augustinus Triumphus', Mediaeval
Bibliography 935
Studies 39:177-205.
McDermott, A. C. S. (1972). 'Notes on the Assertoric and Modal Propositional Logic of
the Pseudo-Scotus', Journal of the History of Philosophy 10:273-306
McDonald, William Joseph (1939). The Social Value of Property according to St Thomas
Aquinas, Catholic University of America
McDonnell, Kevin (1974). 'Does William of Ockham have a theory of natural law?',
McEvoy, James (1977). 'La connaissance intellectuelle selon Robert Grossesteste', Refuf
philosophique du Louvain 75:5-48
McFarlane, K. B. (19S2). John Wycliffe and the beginnings of English Nonconformity, English
Universities Press (Hodder & Stroughton Educational)
McGrade, A. S. (1963). 'The coherence of Hooker's Polity: the books on power', Journal
of the History of Ideas 24:163-82
(1974). The Political Thought of William of Ockham. Personal and Institutional Principles
(Cambridge Studies in Medieval Life and Thought, third series, 7), Cambridge
University Press
(1978). 'Repentance and Spiritual Power: Book VI of Richard Hooker's Of the Laws
of Eccltsiastkal Polity', Journal of Ecclesiastical History 29:163-76
(1980). Ockham and the birth of individual rights' in Brian Tierney and Peter
Linehan (eds.), Authority and Power: Studies on Medieval Law and Government
Prestnted to Walter Ullmann on his Seventieth Birthday, Cambridge University Press
McGrade, A. S. and Brian Vickers, eds. (1975). Richard Hooker: Of the Laws of
Ecclesiastical Polity: An Abridged Edition, Sidgwick and Jackson
Mcllwain, C. H. (1932). The Growth of Political Thought in the West. From the Greeks to
the End of the Middle Ages, MacMillan
(1940). Constitutionalism, Ancient and Modern, Cornell University Press
Mclnerny, R. (1966). Thomism in an Age of Renewal, Doubleday
(1974). 'Boethius and Saint Thomas Aquinas', Rivista di Filosofia neo-scolastica 66:
219-45
McKeon, R. (1929-30). Selections from Medieval Philosophers (2 vols.), Scribners
(1938). "The development of the concept of property in political philosophy: a study
of the background of the constitution'. Ethics 48:297-366
McMullin, Eman (1963). 'Matter as a Principle', with comment by W. Sellars, in E.
McMullin (ed.) The Concept of Matter, University of Notre Dame Press
Maccarrone, Michèle (1954). ' "Potestas directa" e "potestas indirècta" nei teologi del
XII e XIII secolo', Miscellanea Historiat Pontificiae 18:27-47
MachiavelH, Niccolo (1965). Chief Works, and Others, tr. Allan Gilbert, Duke University
Press
(1968). Operf, vol. I, G. Salerno
Macken, R. (1971). 'La temporalité radicale de la créature selon Henri de Gand',
Recherches de Theologie ancienne et médiévale 38:211—72
Macken, R. (1973). 'Les corrections d'Henri de Gand à ses Quodlibets'. Recherches de
théologie ancienne et médiévales 40:5-51
(1976a). 'Le statut de la matière première chez Bonaventure', Franziskanische
Forschungen 28194-103
(1976b). 'La subsistance de la matière première selon Henri de Gand', in San
Bonaventura maestro di vita francescana e di sapienza cristiana (Atti del Congresso
internazionale per il VII centenario di san Bonaventura da Bagnoregio, Roma
19-26 settembre 1974), 3
Madkour, I. (1934). La place d'al Farabi dans l'école philosophique musulmane, Vrin
936 Bibliography
Mahieu, L. (1921). François Suarez, sa philosophie et les rapports qu'elle a avec sa théologie,
Desclée
Mahoney, Edward P. (1970). 'Pier Nicola Castellani and Agostino Nifo on Averroes'
Doctrine of the Agent Intellect', Rivista critica di storia dellafilosofia25:387-409
(1973a). 'Themistius and the Agent Intellect in James of Viterbo and Other Thirteenth
Century Philosophers (Saint Thomas, Siger of Brabant and Henry Bate)',
Augustiniana 23:422—67
(1973b). Review of Marcia L. Colish, Tiie Mirror of Language: A Study in the Medieval
Theory of Knowledge, Yale University Press, 1968, in Journal of the History of
(1974a). 'Saint Thomas and Siger of Brabant Revisited', The Review of Metaphysics
27:531-53
(1974b). 'Saint Thomas and the School of Padua at the End of the Fifteenth Century',
Proceedings of the American Catholic Philosophical Association 48:277-85
(1976a). 'Agostino Nifo and Saint Thomas Aquinas', Memorie Domenicane, n. s.
7:195-226
(1976b). 'Nicoletto Vernia on the Soul and Immortality' in Edward P. Mahoney (ed.)
Philosophy and Humanism: Renaissance Essays in Honor of Paul Oskar Kristeller,
Columbia University Press and E. J. Brill
(1980a). 'Albert the Great and the Studio Patavino in the Late Fifteenth and Early
Sixteenth Centuries' in James A. Weisheipl (ed.) Albertus Magnus and the Sciences:
Commemorative Essays 1980, Pontifical Institute of Mediaeval Studies
(1980b). 'Metaphysical Foundations of the Hierarchy of Being according to Some
Late Medieval and Renaissance Philosophers' in Parviz Morewedge (ed.), Ancient
and Medieval Philosophies of Existence, Fordham University Press
Mahoney, Michael (1978). 'Mathematics' in D . C. Lindberg (ed.) Science in the Middle
Ages, University of Chicago Press
Maier, Anneliese (1944). 'Ein Beitrag zur Geschichte des Italienischen Averroismus im
14. Jahrhundert', Quellen und Forschungen aus den Italienischen Archiven und Biblio-
thekin 33:136-57
(1944a). 'Die scholastische Wesensbestimmung der Bewegung als forma fluens oder
fluxus formae und ihre Beziehung zu Albertus Magnus', Angelicum 2 1 : 9 7 ~ I I I
(1949). Die Vorläufer Galileis im 14. Jahrhundert (Studien zur Naturphilosophie der
Spätscholastik, vol. 1), Edizioni di Storia e Letteratura
(1951). Zwei Grundprobleme der Scholastischen Naturwissenschaft (Studien zur Natur-
philosophie der Spätscholastik, vol. 2) (2nd ed.), Edizioni di Storia e Letteratura
(1952). An der Grenze von Scholastik und Naturwissenschaft (Studien zur Natur-
philosophie der Spätscholastik, vol. 3) (2nd ed.), Edizioni di Storia e Letteratura
(1955). Metaphysische Hintergründe der spätscholastischen Naturphilosophie (Studien zur
Naturphilosophie der Spätscholastik, vol. 4), Edizioni di Storia e Letteratura
(1958). Zwischen Philosophie und Mechanik (Studien zur Naturphilosophie der
Spätscholastik, vol. 5), Edizioni di Storia e Letteratura
(1964). 'Diskussionen über das aktuell Unendliche in der ersten Hälfte des 14. Jahr-
hunderts' in Maier (1964-7), vol. 1
(1964-7). Ausgehendes Mittelalter (2 vols.), Edizioni di Storia e Letteratura
(1966). Die Vorläufer Galileis im 14. Jahrhundert. (Studien zur Naturphilosophie der
Spätscholastik, vol. 1) (2nd. ed.), Edizioni di Storia e Letteratura
(1968). Zwei Grundprobleme der Scholastischen Naturphilosophie (Studien zur Natur-
philosophie der Spätscholastik, vol. 2) (3rd ed.), Edizioni di Storia e Letteratura
Maier, H . (1966). 'Die Lehre der Politik an den deutschen Universitäten vornehmlich
Bibliography 937
vom ιό. bis 18. Jahrhundert' in D. Oberndörfer (ed.) Wksenschafilicht Politik. Eine
Einfuhrung in Grundfragen ihrer Tradition und Theorie (2nd ed), Wissenschaftliche
Buchgesellschaft
Maierù, Α. (1966). 'Il "Tractatus de sensu composito et diviso" di Guglielmo Heytes
bury', Rivùta crilica di Storia della Filosofia 21 ; 243-63
(1969a). 'Il problems della verità nelle opere di G. Heytesbury', Studi Medievali 7:4ο-
74
(1969b). 'Lo Speculum puerorum sive Terminus est in quem di Riccardo Billingham' in A
Giuseppe Ermini, Studi Medievali (Serie terza), 10.3:297—397
(1972). Terminologia logica della tarda scolastica (Lessico intellettuale europeo, 8),
Edizioni dell'Atenco
Maimonides, Moses iff Moses Maimonides
Maistre, J. de (1844). Considerations sur la France, suivies de l'Essai sur le principe générateur
des constitutions politiques . . . . La Société Nationale pour la propagation de bons
livres
Major, John sef John Major
Makdisi, G. (1974). "The Scholastic Method in Medieval Education: An Inquiry into its
Origins in Law and Theology', Speculum 49:640-61
Malcolm, John (1971). 'On Grabmann's text of William of Sherwood', Vivarium
9:108-18
Mandonnet, P. (1908-11). Si^cr de Brabant et l'averroïsme latin au XIIF siècle (2nd ed.) (2
vols.), (Les Philosophes Belges), Institut Supérieur de Philosophie de l'Université de
Louvain
(1910). 'La carrière scolaire de Gilles de Rome', Revue des Sciences philosophiques et
théologiques 4:480-99
(1911). 'Autour de Siger de Brabant', Revue thomiste 19:314-37, 476-502
(1928). 'Chronologie des questions disputées de saint Thomas d'Aquin'. Revue
Thomute 23:267-79
Manegold of Lautenbach (1880). Liber ad Gebehardum, PL 155
Manser, G. (1944). Das Naturrecht in thomistischtr Beleuchtung, Verlag der Paulusdruckerei
Mansion, Augustin (1953). 'L'immortalité de l'âme et de l'intellect d'après Aristote',
Revue philosophique de Louvain 51:457-65
Manzalaoni, Mahmoud (1977). Secretum secretorum, Oxford University Press
Maréchal, J. (1926). Le Point dt Départ dt la Métaphysique. Leçons sur le développement
historique et théorique du problème de la connaissance. Cahier V: Le Thomisme devant la
Philosophie critique, Éditions du Museum Lessianum.
Mariana, Juan de, iff Juan de Mariana
Mariani, U. (1957). Chiesa e stato mi teologi agostiniani del secolo XIV (Uomini e dottrine,
5), Edizioni di storia e letteratura
Maritain, J. (1940). Scholasticism and Politics, tr. M.J. Adler, Macmillan
(1959). Degrees of Knowledge. Translated from the 4th French edition (1946] under
the supervision of G. B. Phelan, Geoffrey Bles
Markowski, Mieczstaw (1968). 'Les questions de Jean Buridan sur les Topiques
d'Aristote', Mediaevalia Philosophica Polonorum 13 :3-7
Marlasca, Antonio (1971). 'La antropologia sigeriana en las "Quaestiones super librum
de causis'", Estudicsfilosàficos20:3-27
(1972). Les Quaestiones suptr librum dt causis de Siger dt Brabant, Publications Universi-
taires de Louvain
(1974). 'De nuevo. Tomâs de Aquino y Siger de Brabante', Estudios filosàficos
938 Bibliography
23:431-9
Marsilius of Inghen (1495). Commentum in primum et quartum traetatum Petri Hispani,
Hagenau (Reprinted Minerva 1967)
(1501). Questiones Marsilii super quattuor libros Sententiarum, Strasburg (Reprinted
Minerva 1966)
(1516). Quaestiones super libros Priorum Analytkorum, Venice (Reprinted Minerva 1968)
Marsilius of Padua (1614a) De Translatione imperii in M. Goldast (ed.)Moriarcftia Sancti
Romani Imperii, 2 :147-53; Conrad Biermann (Reprinted Akademische Druck- und
Verlagsanstalt i960)
(1614b). De lurisdictione imperatoris in causis matrimonialibus in M. Goldast (ed.)
Monorchia Sancti Romani Imperii, 2:1386—91; Conrad Biermann (Reprinted
Akademische Druck- und Verlagsanstalt i960)
(1922). Defensor minor, ed. C. K. Brampton, Comish Brothers
(1928). Defensor pack, ed. C. W. Previte-Orton, Cambridge University Press
(1932-3). Defensor pacts, ed. R. Scholz, Hahnsche Buchhandlung
(1956). The Defensor Paris, ed. and tr. A. Gewirth, Marsilius ofPadua: The Defender of
Peace, 2 (The Records of Civilization, Sources and Studies, 46), Columbia Univer-
sity Press
(1968). Le Défenseur de la paix, tr. with introduction and commentary by Jeannine
Quillet, J. Vrin
Martin, Conor (1949). The commentaries on the Politics of Aristotle in the late thirteenth and
fourteenth centuries, with reference to the thought and political life of the time, unpublished
D. Phil, thesis, Oxford University
(1951). 'Some medieval commentaries on Aristotle's Polities', History 36:29—44
(1964). 'Walter Burleigh' in Oxford Studies Presented to Daniel Callus (Oxford
Historical Society, new series, 16), Clarendon Press
Martin of Dacia (1961). Opera, ed. H. Roos (Corpus Philosophorum Danicorum Medii
Aevi, 2), Gad
Marty, Anton (1908). Untersuchungen zur Grundlegung der allgemeinen Grammatik und
Sprachphilosophie, Max Niemeyer
Masnovo, Amato (1945). Da Guglielmo d'Auvergne a S. Tommaso d'Aquino, vol. 3:
L'uomo (Pubblicazioni delTUniversità Cattolica del Sacro Cuore, n.s. vol. 10),
Società Editrice 'Vita e Pensiero'
Mates, Benson (1965a). Elementary Logic, Oxford University Press
(1965b). 'Pseudo-Scotus on the Soundness of Consequentiae' in A.-T. Tymieniecka
(ed.) Contributions to Logic and Methodology in Honor of I. M. BocheAski, North-
Holland
Matthew of Aquasparta (1903). Quaestiones disputatae selectat, vol. I: Quatstiones defideet
cognitione. Collegium s. Bonaventurae
Matthews, Gareth B. (1973). 'Suppositio and Quantification in Ockham', Nous 7:13-24
(1977)· 'Consciousness and Life', Philosophy 52:13-26
Mattos, Gonçalo de (1940). 'L'intellect agent personnel dans les premiers écrits d'Albert
le Grand et de Thomas d'Aquin', Revue néoscolastique de philosophie 43:145-61
Maurer, Armand (1946). 'Ewf and Essentia in the Metaphysics of Siger of Brabant',
(1955)· 'Boethius of Dacia and the Double Truth', Mediaeval Studies 17:233-9
(1956). 'The State of Historical Research in Siger of Brabant', Speculum 31149—56
(1958). 'Ockham's Conception of the Unity of Science', Mediatval Studits 20:98-112
(1962). Medieval Philosophy, Random House
(1967). Ά Promising new Discovery for Sigerian Studies', Mediaeval Studies 29:364-9
Bibliography 939
(1976). 'Some Aspects of Fourteenth-Century Philosophy', Medievalia et Humanistica,
n.s. 7:175-88
Mazzarella, Pasquale (1974). 'La critica di San Tommaso all' "averroismo gnoseolo-
gico" ', Rivista difilosofianeo-scolastica 66:246—83
Meersseman, Gilles M. (1931) Introductio in opera omnia B. Alberti Magni O. P., C.
Beyaert
(1932). 'Le droit naturel chez S. Thomas d'Aquin et ses prédécesseurs', Angelicum
9:63-76
(1932a). 'Die Einheit der menschlichen Seele nach Albertus Magnus', Ditft« Thomas
10:81-94
(1932b). 'Les origines parisiennes de Talbertisme', Archives d'histoire doctrinale et littéraire
du moyen âge 7:121-42
(1933-5)· Geschichte des Albertismus, 2 vols. (Institutum Historicum Fratrum
Praedicatorum, Dissertationes historica, vols. 3 and 5), Santa Sabina
Melanchthon, Philipp (1520). Compendiaria dialectices ratio, Leipzig
(1555). Erotemata dialectices Wittenberg
(1834-60). Opera quae supersunt omnia, ed. C. G. Bretschneider (28 vols.), Schwetzke
(1850a). Philosophiae moralis epitomes [in Melanchthon 1834—60]
(1850b). In primum librum Ethicorum Aristotelis enarratio [in Melanchthon 1834-60]
Melani, G. (1948). Tractatus de anima loannis Pecham, Edizioni 'Studi Francescani' s.
Francesco
Meiler, Bernhard (1954). Studien zur Erkenntnislehre des Peter von Ailly (Freiburger
theologischen Studien, 67), Herder
Menges, M. C. (1952). The Concept ofUnivocity Regarding the Predication of God and
Creature According to William Ockham, The Franciscan Institute
Menut, Albert D. (1969). Ά Provisional Bibliography of Oresme's Writings: A Supple
mentary Note', Mediaeval Studies 31: 346-7
Mercati, G. (1925). Per la cronologia della vita e degli scritti di Niccolo Perotti, arcivescovo di
Siponto (Studi e testi, 44), Biblioteca apostolica vaticana
Mercken, H. P. F., ed. (1973). The Greek Commentariis on the Nicomachean Ethics in the
Latin Translation of Robert Grosseteste, Bishop of Lincoln (fi253). Vol. I (Corpus
Latinum Commentariorum in Aristotelem Graecorum VI, 1), Brill
Merzbacher, F. (1956). 'Das Somnium viridarii von 1376 als Spiegel des gallikanischen
Kirchenrechts', Zeitschrift der Savigny-Stiftungfür Rechtsgtschichte. Kanonistische
Abteilung 42:55-72
Mesnard, Pierre (1936). L'Essor de la philosophie politiqut au XVIe siècle. Ancienne librairie
Furne, Boivin Sc cie
(1969). L'Essor de la philosophie politique au XVIe siècle (3rd. ed.),J. Vrin
Meulen, Jacob ter and P. J. J. Diermanse (1950). Bibliographie des écrits imprimés de Hugo
Grotius, Martinus Nijhoff
Meynell, C. (1866). Padre Liberatore and the Ontologists: A Review, Burns, Oates, and Co.
Michael of Ephesus (1898). Alexandri quodftrtur in Aristotelis Sophisticos Elenchos Commen-
tarium (ed. M. Wallies) (Commentaria in Aristotelem Graeca II.3), G. Reimer
Michael Pscllus (1926). Chronographie. Tome I, ed. E. Renaud, Les Belles Lettres
Michalski, Κ. (1922). 'Les Courants philosophiques à Oxford et à Paris pendant le XIVe
siècle'. Bulletin internationale de l'Académie polonaise des sciences et des lettres. Classe
d'histoire et de philosophie, et de philologie, les années 1919, 1920, pp. 59-88
(1926). 'Le Criticisme et le scepticisme dans la philosophie du XIV e siècle*, Bulletin
internationale de l'Académie polonaise des sciences et des lettres. Classe d'histoire et de
philosophie, et de philologie, l'année 1925, pp. 41-122
9+0 Bibliography
(1927). 'Les Courants critiques et sceptiques dans la philosophie du XIV e siècle',

Bulletin internationale de l'Académie polonaise des sciences et des lettres. Classe d'histoire
et de philosophie, et de philologie, l'année 1925, Part II, pp. 192-242
(1928). 'La Physique nouvelle et les différents courants philosophiques au XIV e siècle',
Bulletin internationale de l'Académie polonaise des sciences et des lettres. Classe d'histoire
et de philosophie, et de philologie, l'année 1927, pp. 93-164
(i937)· 'Le Problème de la volonté à Oxford et â Paris au XIV e siècle', Studia
Philosophica 2:233-365
(1969). La philosophie au XIV siècle, Minerva
Michaud-Quantin, Pierre (1955). 'Albert le Grand et les puissances de l'âme', Revue du
moyen âge latin 11:59-86
(1966). La psychologie de l'activité chez Albert le Grand, Vrin
Michel, Alain (i960). Les Rapports de la rhétorique et de la philosophie dans l'oeuvre de
Cicéron, Presses universitaires de France
Michel, S. (1932). La notion thomkte du Bien Commun: Quelques-unes de ses applications
juridiques, Vrin
Miczka, G. (1970). Das Bild der Kirche bei Johannes von Salisbury (Bonner historische
Forschungen, 24), L. Röhrscheid
Miethke, J. (1969). Ockhams Weg zur Sozialphilosophie, De Gruyter
Mignucci, M. (1969). 'Albert the Great's Approach to Aristotehan Modal Syllogistic' in
Arts libéraux et philosophie au moyen âge. Actes du IV Congrès international de
philosophie médiévale, Insitut d'études médiévales, Montréal and J. Vrin
Milburn, D. (1964). A History ofUshaw College, Ushaw (Durham)
Milet, A. (1945). 'Les "Cahiers" du Père Maréchal. Sources doctrinales et influences
subies', Revue néoscolastique de philosophie 43 1225-51
Miller, Robert (1954). 'An Aspect of Averroes' Influence on St. Albert', Mediaeval
Studies 16:57—71
Minio-Paluello, L. (1952). 'lacobus Veneticus Grecus: Canonist and Translator of
Aristotle', Traditio 8:265-304
(1954). 'Note sui!' Aristotele Latino Médiévale IX: GH Elenchi Sofistici: redazioni
contaminate coll'ignota versione di Giacomo Veneto (?); frammenti dell* ignoto
commento d'Alessandro d'Afrodisia tradotti in latino', Rivista di Filosofia Nco-
Scolastica 46:223—3 1
(1956). 'Remigio Girolami's De bono communi: Florence at the time of Dante's
banishment and the philosopher's answer to the crisis', Italian Studies 2:56-71
(1956—8). Twelfth Century Logic: Texts and Studies (2 vols.), Edizioni di Storia e
Letteratura
(1962). 'Note suir Aristotele Latino Médiévale XIV: Frammenti del commento
perduto di Alessandro d'Afrodisia ai Secondi Analitici tradotto da Giacomo Veneto
in un codice di Goffredo di Fontaines, Parigi B. N . lat. 16080', Rivista di Filosofia
Neo-Scolastica 54:131-7
(1970a). 'Aristotle: Tradition and Influence' in C. C. Gillispie (ed.), Dictionary of
Scientific Biography, Charles Scribner's Sons
(1970b) 'Boethius' in C. C. Gillispie (ed.), Dictionary of Scientific Biography, Charles
Scribner's sons
(1972). Opuscula, Hakkert
(1974). 'Moerbeke, William o f in C. C. Gillispie (ed.), Dictionary of Scientific Biogra-
phy, Charles Scribner's Sons
Ministeri, B. (1952—3). 'De vita et operibus Augustini de Ancona, O.E.S.A. (*I328)',
Analecta Augustiana 3 1 . 7 - 5 6 , 1 4 8 - 2 6 2
Bibliography 941
Mittelstrass, J. (1962). Die Rettung der Phänomene: Ursprung und Geschichte eines antiken
Forschungsprinzips, De Gruyter
Mochi Onory, S. (1951). Fonti canonistiche dell'idea modema dello stato, Vita e pensiero
Modde, A. (1949). 'Le bien commun dans la philosophie de Saint Thomas', Revue
philosophique de Louvain 47:221-47
Modric, Luca (1978). 'Relazione del P. Luca Modric sui lavori della comissione
Scotistica', Acta Ordinis Fratrum Minorum 97:82-3
Molina, Luis de (1659). De iustitia et iure libri sex (2 vols.), Moguntiae ( = Mainz)
(1876). Concordia liberi arbitrii cumgratiae demis, divina praescientia, Providentia, praedesti-
natione, et reprobatione, Lethielleux
Molinaeus, iff Pierre du Moulin
Molland, A. G. (1967). The Geometria Speculativa of Thomas Bradwardine, unpublished
Ph.D. dissertation, Cambridge University
(1968a). 'The Geometrical Background to the "Merton School'", British Journal for the
History of Science 4:108—25
(1968b). 'Richard Swineshead and Continuously Varying Quantities', XIV Congrès
International d'Histoire des Sciences, Blanchard
(1973). 'John of Dumbleton' in C. C. Gillispie (ed.) Dictionary of Scientific Biography,
Charles Scribner's Sons
(1978). 'An Examination of Bradwardine's Geometry', Archive for History of Exact
Sciences 19:113-75
Monahan, A. P. (1954). 'The Subject of Metaphysics for Peter of Auvergne', Mediaeval
Studies 16:118-30
Monfasani, J. (1976). George of Trebizond: a Biography and a Study of his Rhetoric and Logic
(Columbia Studies in the Classical Tradition, 1), Brill
Monfrin, J., ed. (1962). Abelard: Historia Calamitatum,]. Vrin
Montagnes, Β. (1963). La doctrine de l'analogie de l'être d'après saint Thomas d'Aquin,
Publications Universitaires de Louvain
Moody, Ernest A. (1935). The Logic of William of Ockham, Sheed & Ward
(1942). Iohannis Buridani Quaestiones super libros quattuor de caelo et mundo, The Medieval
Academy of America
(1947). 'Ockham, Buridan, and Nicholas of Autrecourt', Franciscan Studies 7:113—46
(1951). 'Galileo and Avempace: The Dynamics of the Leaning Tower Experiment',
Journalofthe Hktory of Ideas 12:163-93; 375 - 422
(1953). Truth and Consequence in Mediaeval Logic (Studies in Logic and the Foundations
of Mathematics), North-Holland
(1964). Ά quodlibetal question of Robert Holkot, O. P. on the problem of the objects
of knowledge and of belief, Speculum 39: 53-74
(1965). Tfif Logic of William of Ockham (2nd edn.), Russell & Russell
(1966). 'The Medieval Contribution to Logic', Studium Generale 19:443-52
(1967). 'Medieval Logic', under 'History of Logic', in P. Edwards (ed.) The Encyclo
pedia of Philosophy, Macmillan
(1970). 'Buridan, Jean' in C. C. Gillispie (ed.) Dictionary of Scientific Biography, Charles
Scribner's Sons
(1975). Studies in Medieval Philosophy, Science, and Logic. Collected Papers 1933-1969
(Publications of the Center for Medieval and Renaissance Studies, UCLA),
University of California Press
Morân, J. G. (1951). Cursus Philosophicus Collegii Maximi Ysletensis Societatis lesu. Pars iv,
Cosmologia, Buena Prensa
Moraux, Paul (1973). Der Aristotelismus bei den Griechen. Erster Band (Peripatoi, j), De
942 Bibliography
Gruyter
More, Thom3s (1965). Utopia in Edward Surtz and J. H. Hexter (eds.) The Complete
Works of St Thomas More, vol. 4, Yale University Press
Moreau, E. de (1951). Introductory memoir to Mélanges Joseph de Ghellinck, Duculot
Moreno, A. (1966). 'The Nature of Metaphysics', Thomist 30:109-35
Morewedge, P. (1972). 'Philosophical Analysis and Ibn Sînâ's "Essence-Existence"
Distinction', Journal of the American Oriental Society 92:425-35
(1973)· T7K Metaphysica of Avicenna (ibn Smâ). A critical translation-commentary and
analysis of the fundamental arguments in Avicenna's Metaphysica in the Danish Nâma-
i'alâ'î(The Book of Scientific Knowledge) (Persian Heritage Series, 13), Routledge
and Kegan Paul
Morisanus, Bernard, iff Bernard Morisan
Mornet, E. (1978). 'Pauperes scholares. Essai sur la condition matérielle des étudiants
Scandinaves aux XIV e et XV e siècle', Le Moyen-Age 84:53-102
Morrall, John B. (1949). 'Some notes on a recent interpretation of William of Ockham's
political philosophy', Franciscan Studies 9:335-69
(1958). Political Thought in Medieval Times, Hutchinson
(i960). Gerson and the Creat Schism, Manchester University Press
Morris, Christopher (1953). Political Thought in England: Tyndale to Hooker, Oxford
University Press
Morris, Colin (1972). The Discovery of the Individual, 1050-1200, Harper and Row
Moses Maimonides (1951). Guide for the Perplexed, tr. M. Friedlander, Dover
(1974). The Guide of the Perplexed I-II, translated with an introduction and notes by
S. Pines, University of Chicago Press
Muck, O. (1968). The transcendental method. With a foreword by Karl Rahner, tr.
William D. Seidensticker, Herder and Herder
Muckle, J. T. (1958). 'R. Holcot, "Utrum theologia sit scientia": A Quodlibet Ques-
tion', Mediaeval Studies 20:127-53
Müller, I. von (1897). Ueber Galens Werk vom wissenschaftlichen Beweis (Abhandlungen
der philosophisch-philologischen Classe der königlich bayerischen Akademie der
Wissenschaften, XX. Band, II. Abteilung)
Mugica, Placido (1948). Bibliografia suareciana, Universidad de Granada
Mullally, Joseph P. (1964). Pelfr of Spain: Tractatus Syncategorematum and Selected
Anonymous Treatises, Marquette University Press
Muller-Thym, Bernard J. (1940a). 'The Common Sense, Perfection of the Order of Pure
Sensibility', The Thomist 2:315-43
Mullick, M. (1971). 'Does Ockham Accept Material Implication?', Noire Dame Journal of
Formal Logk 12:117-24
Mullinger, J. B. (1884). The University of Cambridge, Cambridge University Press
Mulvaney, Robert J. (1973). 'Political wisdom: an interpretation of Summa Theol. II-II,
50', Mediaeval Studies 35:294-305
Munoz Delgado, V. (1964). La Logica Nominalista en la Universidad dt Salamanca
(1510-1530) (Publicaciones del Monasterio Poyo, 11), Edita Revista 'Estudios'
Munz, Peter (1952). 77ie Place of Hooker in the Hktory of Thought, Routledge & Kegan
Paul
Murdoch, John E. (1957). Geometry and the Continuum in the Fourteenth Century: A
Philosophkai Analysis of Thomas Bradwardine's Tractatus de continuo, unpublished
Ph.D. dissertation, University of Wisconsin
(1962). Rationes Mathematice: Un aspect du rapport des mathématiques et de la philosophie
au Moyen Âge, Université de Paris
Bibliography 943
(1963). 'The Medieval Language of Proportions: Elements of the Interaction with
Greek Foundations and the Development of New Mathematical Techniques', in A,
C. Crombie (ed.), Scientific Change, Heinemann
(1964). 'Superposition, Congruence and Continuity in the Middle Ages' in Mélanges
Alexandre Koyré, Hermann
(1969). 'Mathesis in Philosophiam Scholasticam Introducta. The Rise and Develop-
ment of the Application of Mathematics in Fourteenth Century Philosophy and
Theology', in Arts Libéraux et Philosophie au Moyen Age, Actes du Quatrième
Congrès International de Philosophie Médiévale, Montréal, Institut d'Études
Médiévales, pp. 215-54
(1974a). 'Philosophy and the Enterprise of Science in the Later Middle Ages' in Y.
Elkana (ed.) The Interaction Between Science and Philosophy, Humanities Press
(1974b). 'Naissance et développement de l'atomisme au bas moyen âge latin', in La
science de la naturt: théories et pratiques (Cahiers d'études médiévales, 2), Bellarmin
(1975a). 'From Social into Intellectual Factors: An Aspect of the Unitary Character of
Late Medieval Learning', inj. Ε. Murdoch and E. D. Sylla (eds.) The Cultural Context
of Medieval Learning, Reidel
(1975b). Ά Central Method of Analysis in Fourteenth-Century Science', XlVth
International Congress of the Hktory of Science. Proceedings No. 2 (Tokyo), pp. 68-71
(1978a). "The Development of a Critical Temper: New Approaches and Modes of
Analysis in Fourteenth-Century Philosophy', in S. Wenzel (ed.) Medieval and
Renaissance Studies, No. 7, Proceedings of the Southeastern Institute of Medieval
and Renaissance Studies
(1978b). 'Subtilitates Anglkanae in Fourteenth Century Paris: John of Mirecourt and Peter
Ceffons' in Madeleine Peiner Cosman and Bruce Chandler (eds.) Machaut's World.
Science and Art in the Fourteenth Century (Annals of the New York Academy of
Sciences, 314)
(1979). 'Propositional Analysis in Fourteenth-Century Natural Philosophy', Synthese
40:117-46
(forthcoming a). "The Analytic Character of Late Medieval Learning: Natural
Philosophy without Nature*, in Lawrence Roberts (ed.) Nature in the Middle Ages
(forthcoming b). 'Scientia mediantibus vocibus: Metalinguistic Analysis in Late Medieval
Natural Philosophy', in Sprache und Erktnntnis im Mittelalttr (Acts of the Sixth
International Congress of Medieval Philosophy, Bonn, 29 August-3 September
1977)
(forthcoming c). 'William of Ockham and the logic of infinity and continuity' in
Kretzmann (forthcoming b).
Murdoch, John E. and Edith Sylla, eds. (1975). The Cultural Context of Medieval Learning,
Reidel
(1976). "Swineshead, Richard", in C. C. Gillispie (ed.) Dictionary of Scientific Biography,
(1978). "The Science of Motion' in David C. Lindberg (ed.) Science in the Middle Ages,
University of Chicago Press
Murdoch, John E. and Edward Synan (1966). 'Two Questions on the Continuum; Walter
Chatton (?), O.F.M. and Adam Wodeham, O.F.M.', Franciscan Studies 26:212-88
Murray, Alexander (1978). Reason and Society in the Middle Ages, Clarendon Press
Nagy, A. (1897). Die philosophischen Abhandlungen desja'qûb ben Ishâq Al-Kindî (BGPM
II), Aschendorff
Nardi, B. (1936). 'Il preteso tomismo di Sigieri di Brabante', Giornale critico della
944 Bibliography
filosofia italiana 17:26-35

(I937)- 'Ancora sui preteso tomismo di Sigieri di Brabante', Giomale critico della
filosofia italiana 18:160-4
(1938). 'L'averroismo di Sigieri e Dante', Studi Danteschi 22:83-113
(1939). 'Una nuova monografia su Sigieri di Brabante', Giornale critico della filosofia
italiana 20:453-71
(1940). Review of Fernand van Steenberghen, Les oeuvres et la doctrine de Siger der
Brabant, Académie royale de Belgique, in Studi Danteschi 25:149-56
(1944). Nelmondo di Dante (Storia e letteratura, vol. 5), Edizioni di 'Storia e letteratura'
(1945). Sigieri di Brabante nel penskro del Rinascimento Italiano, Edizioni Italiane
(1947). 'Introduzione' in S. Tommaso cVAquino: Trattato sull'unità dell intelletto ccmtrogli
Averroisti, Sansoni
(1949). 'Note per una storia deH'averroismo latino, V: L'averroismo bolognese nel
secolo XIII e Taddeo Alderotto', Rivista critica di storia dellafilosofia4:11-22
(1950). 'L'anima umana secondo Sigieri', Giomale critico difilosofiaitaliana 29:317-325
(i960) Studi difilosofiamédiévale, Edizioni di Storia e Letteratura
(i960). 'La dottrina d'Alberto Magno sulT Inchoatio formae', in B. Nardi (ed.) Studi
di Filosofia Médiévale, Edizioni di Storia e Letteratura
Nash, P. (1950). 'Giles of Rome on Boethius' "Diversum est esse et id quod est"',
(1950-1). 'Giles of Rome, Auditor and Critic of St. Thomas', The Modem Schoolman
28:1-20
('957)· 'The Accidentality of Esse According to Giles of Rome', Gregorianum
38:103-15
(1967). 'Giles of Rome', New Catholk Encyclopedia 6:484-5
Nasr, S. H. (1964). An Introduction to Islamic Cosmologkal Doctrines, Harvard University
Press
Natalis, Hervaeus, iff Hervaeus Natalis
Naus, John E. (1959). The Nature of the Practical Intellect According to Saint Thomas Aquinas
(Analecta Gregoriana, vol. 108), Gregorian University
Nebrija, Elio Antonio de (1481). Introductiones Latinae, Salamanca
(1495). Introductiones Latinae cum commento, Salamanca
Nédoncelle, Maurice (1974). 'Remarques sur la réfutation des averroïstes par saint
Thomas', Rivista difilosofianeo-scolastica 66:284—92
Neumann, B, (1958). Der Mensch und die himmlische Seligkeit nach der Lehrt Gottfrieds von
Fontaines, Lahn-Verlag
Newman, J. H. (1961-) The Letters and Diaries of John Henry Newman. Ed. C. S. Dessain
et al., Nelson
Newman, William Lambert. (1887). The Politics of Aristotle, Clarendon Press
Nicephorus Blemmydes (1885). Epitomes isagogicce liber primus. Epitome logica (P. G. 142),
Brepols
Nicholas of Autrecourt (1908). Prima epistola ad Bernardum and Epistola Magistri Nicolai de
Autricort ad Bernardum (BGPM VI, 2), Aschendorff
(1939). 'Tractatus universalis', ed. J. R. O'Donnell, C.S.B., Mediaeval Studies ι : 179-
28θ
Nicholas of Cusa (1964-8). De Concordantia Catholica, Felix Meiner
Nicholas of Paris. Sincategoreumata in Braakhuis 1979, II
Summe Metenses in De Rijk 1962-7, II (1). 449-90 (excerpts)
Nicolas, J.-H. (1948). 'Chronique de Philosophie'. Revue Thomiste 48:561-4
Nicole Oresme (1970). Maistre Nicole Oresme: Le Livre de politique d'Aristote, ed. A. D.
Bibliography 945
Menut (Transactions of the American Philosophical Society, New Series 60, Part 6),
American Philosophical Society
Nielsen, L. (1976). 'On the Doctrine of Logic and Language of Gilbert Porreta and his
Folio wen', CIMAGL 17:40-69
Niemeyer, Sister Mary Fredericus (1951). The One and the Many in the Social Order
according to Saint Thomas Aquinas, Catholic University of America
Nifo, Agostino (1520). Dialectica ludicra, Florence
(1521). Epitomata rethorka ludicra, Venice
Noble, Henri-Dominique (1905). 'Note pour l'étude de la psychophysiologie d'Albert le
Grand et de S. Thomas. Le cerveau et les facultés sensibles', Revue Thomiste
13:91-101
Normore, Calvin G. (forthcoming). 'Walter Burley on Continuity' in Kretzmann
(forthcoming b)
Nuchelmans, Gabriel (1973). Theories of the Proposition: Ancient and Medieval Conceptions of
the Bearers of Truth and Falsity (North-Holland Linguistic Series, 8), North-Holland
(1980). Late Scholastic and Humanist Theories of the Proposition, North-Holland
Oakley, Francis (1961). 'Medieval theories of natural law: William of Ockham and the
significance of the voluntarist tradition', Natural Law Forum 6:65-83
(1962). 'On the road from Constance to 1688: the political thought ofJohn Major and
Geoige Buchanan', The Journal of British Studies 2:1-31
(1964). The Political Thought of Pierre d'Ailly: The Voluntarist Tradition (Yale Historical
Publications, Miscellany 81), Yale University Press
(1965). 'Almain and Major: Conciliar theory on the eve of Reformation', The
American Historical Review 70:673-90
(1967). 'Pierre d'Ailly' in B. A. Gerrish (ed.), Reformers in Profile, Fortress Press
(1969). 'Figgis, Constance and the divines of Paris', The American Historical Review
75:368-86
(1974). The Medieval Experience: Foundations of Western Cultural Singularüy, Charles
Scribner's Sons
Oberman, H.A. (1967). Tire Harvest of Medieval Theology: Gabriel Biet and Late Medieval
Nominalism, Harvard University Press
Obertello, Luca (1969). A. M. Severino Boezio: De hypotheticis syllogismis, testo, tradu-
zione, introduzione e commento, Paideia éditrice
(1974). Severino Boezio (2 vols.) (Collana di monografie, 1), Accademia Ligure di
Scienze e Lettere
O'Brien, A.J. (1964). 'Duns Scotus' Teaching on the Distinction Between Essence and
Existence', The New Scholasticism 38:61-77
O'Brien, T. C. (i960). Metaphysics and the Existence of God, The Thomist Press
O'Callaghan.J. J. (1955). 'The Second Question of the Prologue of Walter Carton's
Commentary on the Sentences. On Intuitive and Abstractive Knowledge', in
J. R. O'Donnell (ed.), Nine Mediaeval Thinkers - A collection of Hitherto Unpublished
Texts, Pontifical Institute of Mediaeval Studies
Ockham, William. See William Ockham
O'Connor, Daniel John (1967). Aquinas and Natural Law, Macmillan
O'Donnell.J. Reginald (1939-42). 'Nicholas of Autrecourt', Mediaeval Studies 1:179-
28014:97-125
ed. (1941). 'The Syncategoremata of William of Sherwood', Mediaeval Studies
3:46-93
(1958). 'Themistius' Paraphrase of the Posterior Analytics in Gerard of Cremona's
946 Bibliography
Translation', Mediaeval Studies 20:239—315

Oeing-Hannhoff, L. (1963). 'Die Methoden der Metaphysik im Mittelalter*. Miscellanea
Mediaevalia 2, De Gruyter
Offler, H. S. (1977)- 'The Three Modes of Natural Law in Ockham: A Revision of the
Text', Franciscan Studies 37:207-18
Olgiati, F. (1944). // concetto digiuridkità in San Tommaso a"Aquino (2nd ed.), Vita e
pensiero
O'Malley, John W. (1968). Giles of Viterbo on Church and Reform (Studies in Medieval
and Reformation Thought, 5), Brill
(1972). 'Man's dignity, God's love, and the destiny of Rome: a text of Giles of
Viterbo', Viator 3: 389-416
Onclin, W. (1948). 'Le droit naturel selon les romanistes des XIIe et XIIIe siècles' in
Miscellanea moralia in honorem A.Janssen (Bibliotheca Ephemeridum theol.
Lovaniensium, I, ii), E. Nauwelaerts
Ong, Walter Jackson (1958a). Ramus, Method, and the Decay of Dialogue, Harvard
University Press
(1958b). Ramus and Talon Inventory, Harvard University Press
Opicinus de Canistris (1914). De Praeeminentia Spiritual! Imperii in R. Scholz (ed.)
Unbekannte kirchenpolitische Streitschriften aus der Zeit Ludwigs des Bayern (1327-1354),
vol. 2., Loescher (W. Regenberg)
Ostlender, H. (1939). Peter Abelards Theologia 'Summi Boni' (BGPMXXXV), Aschen-
dorff
Otte, J. K. (1972). 'The Life and Writings of Alfredus Anglicus', Viator 3 1275-91
(1976). 'The Role of Alfred of Sareshel (Alfredus Anglicus) and his Commentary on
the Metheora in the Reacquisition of Aristotle', Viator 7:197-209
Overbeke, P.-M. Van (1955). 'Saint Thomas et le droit', Rftw Thomiste 55:519-64
(1957)- 'La loi naturelle et le droit naturel selon saint Thomas', Revue Thomiste
57:53-78;450-95
(1958). 'Droit et morale, essai de synthèse thomiste', Revue Thomiste 58:285-336;
674-94
Owen, G. E. L., ed. (1968). Aristotle on Dialectic. Proceedings of the Third Symposium
Aristotelicum, Clarendon Press
Owens, Joseph (1948). 'Up to What Point is God Included in the Metaphysics of Duns
Scotus?', Mediaeval Studies 10:163-77
(1953). 'The Conclusion of the Prima Via", The Modem Schoolman 30:109-21
(1965). 'Quiddity and Real Distinction in St. Thomas Aquinas', Mediaeval Studies
27:1-22
(1966). 'Aquinas and the Prooffrom the "Physics" ', Mediaeval Studies 28:119—50
(1970). 'Judgment and Truth in Aquinas', Mediaeval Studies 32:138-58
(1974). 'The Primacy of the External in Thomistic Noetics*. Église et Théologie
5:189-205
(1978). TTif Doctrine of Being in the Aristotelian 'Metaphysics', A Study in the Greek
Background of Mediaeval Thought, 3rd ed., Pontifical Institute of Mediaeval Studies
Ozment, S. (1974). 'Mysticism, nominalism, and dissent' in C. Trinkaus and H. A.
Oberman (eds.) The Pursuit of Holiness in the Late Middle Ages and the Renaissance,
Brill
Padley, G. A. (1976). Grammatical Theory in Western Europe 1500-1700: the Latin

Tradition, Cambridge University Press
Paissac, H. (1951). Théologie du verbe: Saint Augustin et Saint Thomas, Les Éditions du
Bibliography 947
Cerf
Pape, I. (1966). Tradition und Transformation der Modalität I. Möglichkeit-Unmöglichkeit,
Felix Meiner
Park, Katharine (1980). 'Albert's Influence on Late Medieval Psychology' in James A.
Weisheipl (ed.) Albertus Magnus and the Sciences: Commemorative Essays 1980,
Parkes, Μ. Β. (1976). 'The Influence of the Concepts of Ordinatio and Compilatio on
the Development of the Book', Medieval Learning and Literature. Essays presented to
R. W. Hunt. Clarendon Press
Partee, C. (i960). 'Peter John Olivi: Historical and Doctrinal Study', Franciscan Studies
20:215-60
Pascoe, Louis B. (1973). Jean Gerson: Principles of Church Reform (Studies in Medieval and
Reformation Thought, 7), Brill
Pasquali, G. (1952). Storia delta tradizione e critica del testo, Felice Le Monnier
Passerin d'Entrèves, Α., iff D'Entrèves, Alessandro Passerin
Pater, Walter de (1965). Les Topiques d'Aristote et la dialectique platonicienne; La Méthodo-
logie dt la définition, (Études thomistiques, 10), Éditions St Paul
(1968). 'La Fonction du lieu et de l'instrument dans les Topiques', in G. E. L. Owen
(ed.) Aristotle on Dialectic. Proceedings of the Third Symposium Aristotelicum,
Clarendon Press
Pattin, A. (1953). 'Gilles de Rome, O.E.S.A. (ca. 1243-1316) et la distinction réelle de
l'essence et de l'existence', Reviif de l'Université d'Ottawa 23:8o-n6
(1974-5). 'Pour l'histoire du sens agent au moyen âge'. Bulletin de philosophie médiévale
16-17:100-13
Paul of Pergula (1961). Logica and Tractatus dc sensu composito et diviso, ed. Mary Anthony
Brown (Franciscan Institute Text Series, 13) The Franciscan Institute
Paul of Venice (1472). Logica ( = Logica parva), Venice. (Photoreprint Georg Olms, 1970.)
(1493a). Quadratura, Venice
(1493b). Sophismata, Venice
(1499). Logica magna, Venice
Paulus, J. (1933). 'La théorie du Premier Moteur chez Aristote', Revue de philosophie n.s.
4:259-94:394-424
(1938). Henri de Gand. Essai sur les tendances de sa métaphysique,]. Vrin
(1940-2). 'Les disputes d'Henri de Gand et de Gilles de Rome sur la distinction de
l'essence et de l'existence', Archives d'Histoire doctrinale et littéraire du Moyen Age
13:323-58
(1949). Ά propos de la théorie de la connaissance d'Henri de Gand', Revue philo-
sophique du Louvain 47:493—6
Payer, Pierre J. (1979). 'Prudence and the principles of natural law: a medieval develop-
ment', Speculum 54:55-70
Pedersen, Olaf (1953). 'The Development of Natural Philosophy 1250-1350', Classica et
Medice valia 14:86-155
Peghaire, Julien (1936). 'Intellectus' et 'Ratio' selon S. Thomas d'Aquin (Publications de
L'Institut d'Études Médiévales d'Ottawa, No. 6), Vrin and Institut d'Études
Médiévales
(1942-3). Ά Forgotten Sense, The Cogitative according to St Thomas Aquinas',
The Modem Schoolman 20:123-40 and 210-29
Pegis, Anton C. (1934). St Thomas and the Problem of the Soul in the Thirttenth Ctntury, St
Michael's College
(1973). 'St Thomas and the Coherence of the Aristotelian Theology', Mediaeval
948 Bibliography
Studies 35:67-117
(1974). 'The Separated Soul and its Nature in St Thomas' in St Thomas Aquinas
1274-1974: Commemorative Studies, vol. 1, Pontifical Institute of Mediaeval Studies
Peifer, John F. (1952). The Concept in Thomism, Bookman Associates
Pelagius, Alvarus. Sef Alvarus Pelagius
Pelster, F. (1953) 'Die indirekte Gewalt der Kirche über den Staat nach Ockham und
Petrus de Palude: eine Übersicht', Scholastik 2 8 : 7 8 - 8 2
Pelzer, A. (1911). 'Les initiateurs italiens du thomisme contemporain', Revue nèo-
scholastique de philosophie 18:230-54
(1924). ' U n Traducteur inconnu: Pierre Gallego, franciscain et premier évêque de
Carthagène' in Miscellanea Francesco Ehrle (Studi e Testi, 37), pp. 407-56
(1964). Études d'histoire littéraire sur la scholastique médiévale (Philosophes Médiévaux 8),
Louvain Presses universitaires
(1966). Abréviations latines médiévales (supplement to Capelli 1961), Louvain Presses
Universitaires
Perrival, W . K. (1972). 'The historical sources of Guarino's Regulae Grammaticales: a
reconsideration of Sabbadini's evidence' in Civiltà dell' umanesimo: Atti del VI, VII,
VIII convegno del Centro di studi umanistici 'Angelo Poliziano', Leo S. Olschki
(1975)· 'The grammatical tradition and the rise of the vernaculars' in T. A. Sebeok
(ed.) Current Trends in Linguistics (Historiography of Linguistics, 13) Mouton
(1976a). 'Deep and surface structure concepts in renaissance and mediaeval syntactic
theory' in H. Parret (ed.) History of Linguistic Thought and Contemporary Linguistics,
De Gruyter
(1976b). 'Renaissance grammar: rebellion or evolution?' in Interrogativ! dell'umanesimo,
Leo S. Olschki
Pertusi, A. (1962). 'Erotemata: per la storia e le fonti delle prime grammstiche greche a
stampa', Italia medioevale e umanistica 5:321—51
Peter Abelard (1851). Sic et Non, ed. Henke and Lindenkohn, Librarise Acsdem.
Elwertisnae
(18553) Dialogus inter Philosophum, Iudaeum et Christianum, PL 178, Brepols
(1855b). Theologia Christiana PL 178, Brepols
(1859). Introductio ad theologiam, ed. V. Cousin in Petri Abaelardi Opera II, Durand
(1919—27). Logica 'Ingredientibus', ed. Bernhard Geyer in Peter Abaelards Philosophische
Schriften (BGPM, X X I , 1-3), Aschendorff
(1933)- Logica 'Nosirorum petition! sociorum', ed. Bernhard Geyer in Peter Abaelards
Philosophische Schriften (BGPM, XXI, 4), Aschendorff
(1956). Dialectica, ed. L. M . De Rijk (Wijsgerige teksten en studies, ι), Van Gorcum
(1969-). Opera theologica, ed. Ε. M. Buytaert (Corpus Christianorum, Continuatio
mediaevalis, n - i 2 ) , Brepols
(1969a). Editio super Aristotelem De interprétât tone, ed. Mario dal Pra in Pierro Abelardo:
Scritti di logica. La nuova Italia Editrice
(1969b). Super Topica glossae, ed. Mario dal Pra in Piffro Abelardo: Scritti di logica, La
nuova Italia Editrice
(1970a). Dialectica, ed. L. M . de Rijk (Wijsgerige teksten en studies, 1) (2nd ed.), Van
Gorcum
(1970b). Dialogus inter Philosophum, Iudaeum, et Christianum, ed. Rudolph Thomas,
Frommann-Holzboog
Peter of Ailly (1489). Tractatus de arte obligandi, Paris
(1490). Quaestiones super libros Sententiarum cum quibusdam in fine adiunctis, Strasburg
Bibliography 949
(ca. i490-5a). Desiructiones modorum significandi, Lyon
(ca. I490~5b). Conceptus et insolubilia, Lyon
(1494). Tractatus exponibilium, Paris
(ca. 1495). Conceptus et insolubilia, Paris
(1706). De Ecclesiae ... et Cardinalium Auctoritate in L. Du Pin (ed.) Jean Gerson, Opera
Omnia, 2: cols. 925-60, Sumptibus Societatis Antwerp
(1964). Tractatus de Materia Concilii Generalis in Oakley 1964
(1980). Concepts and Insolubles: An Annotated Translation, Paul Vincent Spade (tr.)
(Synthese Historical Library, vol. 19), D. Reidel
Peter Aureoli (1596-1605). Commentarium in primum Sententiarum (2 vols.), Vatican
(1952-6). Pffri Aureoli Scriptum super primum Sententiarum, ed. E. M. Buytaert
(2 vols.) (Franciscan Institute Text Series, 3), The Franciscan Institute
Peter of Auvergne (1951). [Continuation (from 1280*6) of Thomas Aquinas' In octo libros
Politicorum Aristotelis expositio] in Thomas Aquinas 1951
(1967). The commentary of Peter of Auvergne on Aristotle's Politics. The inedited part: Book
III lesson i-vi, ed. Gundisalvus M. Grech, Desclée
Peter Bertrandus (1614). De lurisdictione Ecclesiastica et Politico in M. Goldast (ed.)
Monarchia Sancti Romani Imperii, 2:1361-83, Conrad Biermann (Reprinted
Akademische Druck- und Verlagsanstalt i960)
Peter John Olivi (1922-6). Quaestiones in secundum librum Sententiarum, ed. B.Jansen (3
vols.), Bibliotheca Franciscana Scholastica Medii Aevi, Collegium S. Bonaventurae
Peter Lombard (1916). Libri IV Sententiarum, ed. A. Heysse, Editiones Collegii S.
Bonaventurae ad Claras Aquas, Grottaferrata
(1971-). Sententiae in IV libris distinctae, Editiones Collegii S. Bonaventurae ad Claras
Aquas, Grottaferrata
Peter of Poitiers (1961). Sententiae, ed. P. S. Moore and M. Dulong (2nd ed.)
(Publications in Medieval Studies, 7), University of Notre Dame Press
Peter of Spain (1489). Textus omnium tractatuum Petri Hispani etiam sincathegreumatum et
parvorum logkalium cum copulatis secundum doctrinam divi Thome Aquinatis iuxla
processum magistrorum Colonie in bursa Montis regentium, Cologne
(1944). Commentaria al 'De anima' de Aristoteles, ed. M. Alonso, Bolanos y Aguilar
(1945). Summulae logkales, ed. & tr. J. P. Mullally (Publications in Mediaeval Studies,
8), University of Notre Dame Press
(1964). Tractatus syncategorematum and Sekcted Anonymous Treatises, tr. J. P. Mullally,
with an introduction by J. P. Mullally and R. Houde (Mediaeval Texts in
Translation, 13), Marquette University Press
(1972). Tractatus called afterwards Summule Logkales, ed. L. M. De Rijk (Wijsgerige
teksten en studies, 22), Van Gorcum
(1979). Syncategoremata excerpted in Braakhuis 1979, I
Peter of Tarantasia (1652). In quattuor libros Sententiarum commentaria, Toulouse
(Reprinted Gregg Press 1964)
Peter Tartaretus (1497). Quaestiones super sex libros Ethicorum Aristotelis, Paris
(1514a). Expositio in Summutas Petri Hispani, Basle
(1514b). In Perihermenias, Basle
Peters, F. E. (1968). Aristotle and the Arabs. The Aristotelian tradition in Islam, New York
University Press
Petrarch (1614). Epistolae de Iuribus Imperii Romani in M. Goldast (ed.) Monarchia
Sancti Romani Imperii, 2:1345-1465, Conrad Biermann (Reprinted Akademische
Druck- und Verlagsanstalt i960)
Petrus da Fonseca (1577). Commentariomm ... in libros metaphyskorum Aristotelis, Rome
950 Bibliography
Pfligersdorflèr, G. (1953). 'Zu Boethius, De Interpr. Ed. sec. I, p. 4, 4 sqq. Meiser nebst
Beobachtungen zur Geschichte der Dialektik bei den Römern', Wiener Studien
66:131-54
Philippe, M.-D. (1948). Άφαίρεσις, πρόσθεσις, χωρίζειν dans la philosophie d'Aristote',
Revue thomiste 48:461-79
(i97S)· 'Originalité de "l'ens rationis" dans la philosophie de saint Thomas', Angelicum
52:91-124
Phillipson, C. (1911) Tht International Law and Custom of Ancient Greece and Rome,
Macmillan
Philoponus, John, iff John Philoponus
Picard, Gabriel (1926). Essai sur la connaissance sensible d'après les scolastiquts (Archives de
Philosophie, vol. 4, Cahier 1), Beauchesne
Picavet, F. (1905). Esquisse d'une histoire générale et comparée des philosophies médiévales,
F. Alcan
Piernikarczyk, E. (1930). 'Das Naturgesetz bei J. D. Scotus', Philosophisches Jahrbuch
43:67-91
Pierre de la Palu (1506). De Causa Immediata Ecclesiasticae Potestatis,fols.24-80, J. Petit
Pierre du Moulin (1603). Elementa logka, Antwerp
university of Uppsala (Acta universitatis Upsalensis. Skrifter rörande Uppsala
universitet, C 36), Almquist & Wicksell
Pighius, Albertus (1697-9). Hierarchiae ecclesiasticae assertio inj. T. Rocaberti (ed.)
Bibliotheca maxima pontificia, 2 (21 vols.) (Reprinted Akademische Druck- und
Verlagsanstalt 1969-70)
Piltz, Α. (i977). Studium Upsalense. Specimens of the oldest lecture notes taken in the medieval
Pinborg, Jan (1967a). Dif Entwicklung der Sprachtheorie im Mittelalter (BGPM XLII, 2),
Aschendorff
(1967b). 'Walter Burleigh on the Meaning of Propositions', Classica et Mediaevalia
28:394-404
(1969). 'Topik und Syllogistik im Mittelalter' in F. Hoffman et al. (eds.) Sapienter
ordinäre. FestgabefirErich Kleineidam (Erfurter Theologische Studien, 24), St Benno
Verlag
(1971a). 'Bezeichnung in der Logik des Mittelalters' in Der Begriff der Repraesentatio im
Mittelalter (Miscellanea Mediaevalia, 8), De Gruyter
(1971b). 'Simon of Faversham's Sophisma: Universale est Intentio: A Supplementary
Note', Mediaeval Studies 33 1360-5
(1972). Logik und Semantik im Mittelalter - Ein Überblick (Problemata, 10), Frommann-
Holzboog
(!973)· 'Neues zum Erfurter Schulleben des XIV. Jahrhunderts nach Handschriften der
Jagiellonischen Bibliothek zu Krakow', Bulletin de philosophie médiévale 15:146-51
(1973a). 'Petrus de Alvernia on Porphyry', CIMAGL 9:47-67
(1973b). Ά New Ms. of the Questions on the Post. Anal. Attributed to Petrus de
Alvernia with the Transcription of some Questions Related to Problems of
Meaning', CIMAGL 10:48-62
(1974). 'Zum Begriff der Intentio Secunda - Radulphus Brito, Hervaeus Natalis und
Petrus Aureoli in Diskussion', CIMACL 13149-59
(1975a). 'Die Logik der Modistae', Studia Mediewistyczne 16:39-97
(1975b). 'Radulphus Brito's Sophism on Second Intentions', Vivarium 13:119-52
(1975c). Ά Note on Some Theoretical Concepts of [Medieval] Logic and Grammar',
'Grabmann*, Revue Internationale de Philosophie, 29e année, 113:286-96
(1975α1). 'Classical Antiquity: Greece' in T. A. Sebeok (ed.) Current Trends in
Bibliography 951
Linguistics vol. 13, Historiography of Linguistics, pp. 69-126, Mouton

(1976a). 'Some Problems of Semantic Representations in Medieval Logic' in H.
Parrett (ed.) History of Linguistic Thought and Contemporary Linguistics, De Gruyter
ed. (1976b). The Logk of John Buridan, Acts of the 3rd European Symposium on
Medieval Logic and Semantics (Opuscula Graecolatina, 9), Museum Tusculanum
(1976c). 'Magister Abstractionum', CIMAGL 18:1-4
(19760). 'Nochmals die Erfurter Schulen im XIV. Jahrhundert', CIMAGL 17:76-81
(1976e). 'Diskussionen um die Wissenschaftstheorie an der Artistenfakultät',
Mkcellanea Mediaevalia 10, De Gruyter
(1979)· 'The English Contribution to Logic before Ockham', Synthese 40:19-42
(1980). 'Radulphus Brito on Universals', CIMAGL 35:56-142
Pinborg,Jan, O. Lewry, K. M. Fredborg, N.J. Green-Pedersen, and L. Nielsen (1975).
"The Commentary on "Priscianus maior" ascribed to Robert Kilwardby*,
CIMAGL 15:1-146
Pine, Martin (1973). 'Double Truth' in Dictionary of the Hktory of Ideas, vol. 2, Scribner's
Piper, Paul, ed. (1882). Dif Schriften Notkers und seiner Schule, vol. I, Akademische
Verlagsbuchhandlung vonj. C. B. Mohr
Plantinga, Alvin (1974). The Nature of Ntctssity, Clarendon Press
[Plato Latinus] (1940). I, Meno interprète Henrico Aristippo, ed. V. Kordeuter and
L. Labowsky, Warburg Institute
(1950). II; Phaedo inttrprete Henrko Aristippo, ed. L. Minio-Paluello, Warburg Institute
Plotinus (1951-9). Opfra / - / / , ed. P. Henry and H.-R. Schwyzer (Museum Lessianum,
Series philosophica, 33-4), Desclée de Brouwer
Pocock.J. G. A. (1971). Politics, Language, and Time, Atheneum
(1974). The Machiavellian Moment, Princeton University Press
Pollard, A. W. and G. R. Redgrave (1976). A Short Title Catalogue of Books Printed in
England, Scotland and Ireland and of English books Printed Abroad 1475—1640. Second
Edition, Revised and Enlarged, begun by W. A.Jackson and F. S. Ferguson.
Completed by Katharine F. Pantzer, Volume 2,1-Z. The Bibliographical Society
Ponet, John (1942). A Short Treatise of Politic Power in Winthrop S. Hudson, John Ponet
(I5i6?-i5s6): Advocate of Limited Monarchy, University of Chicago Press
Poppi, Antonino (1970a). Introduzione all' Aristotelismo Padovano, Editrice Antenore
(1970b). Saggi sulpensiero inedito di Pittro Pomponazzi (Centro per la Storia della
Tradizione Aristotelica nel Veneto, Saggi e Testi, No. 8), Editrice Antenore
Porphyry (1886). De abstinentia in A. Nauck (ed.) Porphyrii philosophi platonki opuscula
selecta, Teubner
(1887). Isagoge et in Arktotelis Categorias commentarium, ed. A. Busse (Commentaria in
Aristotelem Graeca, IV. i), Reimer
Post, Gaines (1964). Studies in Medieval Legal Thought, Princeton University Press
Posthumus Meyjes, Guillaume Henri Marie (1963). Jean Gerson, zijn kerkpolitek en
ecclesiologie, Nijhoff
Potts, Timothy C. (1980). Conscienct in Medieval Philosophy, Cambridge University Press
Poupard, P., ed. (1958). 'Lettre de Möhler à Bautain sur les rapports de la raison et de la
foi', Rtvut des sciences philosophiques et théologiques 42:455-82
(1961). L'Abbé Louis Bautain. Un essai de Philosophie Chrétienne au dix-neuvième siècle,
Desclée
ed. (1964).journal Romain de l'abbé Louis Bautain, 1838, Edizioni di Storia e Letteratura
Prantl, Cari (1855-67). Geschichte der Logik im Abendlande (4 vols.), S. Hirzel (Reprinted
Akademische Druck- und Verlagsanstalt 1955)
Prentice, R. (1969). 'Univocity and Analogy according to Scotus's Super libros elen-
952 Bibliography
chorum Aristotelis', Archives d'histoire doctrinale el littéraire du Moyen Age 35:39-64

Preuss, James S. (1972). 'Theological legitimation for innovation in the middle ages',
Viator 3:1-26
Prezioso, F. (1963). La 'species' médiévale e i poredromi delfenomenkmo moderno, instituto
Universitario di Magistero di Catania
Prior, A. N. (1953). 'On Some Consequentiae in Walter Burleigh', The New Scholasti-
cism 27:433-46
(1962). Formal Logic, Clarendon Press
(1962a). 'Some Problems of Self-Reference in John Buridan', Proceedings of the British
Academy 48:281 -96
(1969). 'The Possibly-True and the Possible', Mind 78:481-92
Priscian (1855-9). Institutionesgrammaticae, ed. Martin Hertz (Grammatici Latini, 2-3),
Teubner (Reprinted Georg Olms 1961)
Proclus (1951). 'Prodi Elementatio theologica translata a Guilelmo de Moerbeke', ed.
C. Vansteenkiste, Tijdschrift voor Philosophie, 131263-302; 491-531
Psellus, Michael, see Michael Psellus
Pseudo-Albertus (1977). Speculum astronomiae, ed. S. Caroti, M. Periera, and S. Zamponi
under the direction of Paola Zambelli, Domus Galilacana
Pseudo-Augustinus (1961). Anonymi paraphrasis Themistiana (Pseudo-Augustini Categoriae
decern) in Aristoteles Latinus (1961b)
Pseudo-Petrus (1489). Tractatus exponibilium in Peter of Spain (1489)
Pseudo-Scotus (1639). In librum I Priorum Analyticorum Aristotelis quaestiones in John Duns
Scotus (1639), vol. I
(1891). Super librum I Priorum in John Duns Scotus (1891-5), vol. II
Ptolemy of Lucca (1909a). De Origine ac Translatione et Statu Romani Imperii, ed.
M. Krammer (Fontes Iuris Germanici Antiqui in Usum Scholamm ex Monumentis
Germaniae Historicis separatim editi), Impensis Bibliopolii Hahniani
(1909b). Determinatio Compendiosa de lurisdictione Imperii, ed. M. Krammer (Fontes
Iuris Germanici Antiqui in Usum Scholarum ex Monumentis Germaniae Historicis
separatim editi), Impensis Bibliopolii Hahniani
(1954). Completion of Thomas Aquinas' De Regimine Principum [De Regno] in R. M.
Spiazzi (ed.) Divi Thomae Aquinatis Opuscula Philosophica, Marietti
Quillet, Jeannine (1970). La philosophie politique de Marsile de Padoue, J. Vrin

(1977). La Philosophie politique du 'Songe du Vergier" (1378): Sources Doctrinales, J. Vrin
Quinn, J. F. (1972). 'Chronology of St. Bonaventure (1217-1257)', Franciscan Studies
32:168-86
(1973). The Historical Constitution of St Bonaventure's Philosophy, Pontifical Institute of
Mediaeval Studies
(1974). 'St. Bonaventure's fundamental conception of natural law' in S. Bonaventura
1274-1974, 3, Collegium S. Bonaventurae, Grottaferrata
(1974a). 'Certitude of Reason and Faith in St Bonaventure and St Thomas' in
St Thomas Aquinas 1274-1974: Commemorative Studies, vol. 2, Pontifical Institute
of Mediaeval Studies
Quintilian (1920—2). The Institutio oratoria of Quintilian, with an English translation by
H. E. Butler (4 vols.) (Loeb Classical Library), Heinemann
Rabeau, Gaston (1938). Species-Verbum: L'activité intellectuelle élémentaire selon S. Thomas

d'Aquin (Bibliothèque Thomiste, vol. 22), Vrin
Radetti, G. (1953). Lorenzo Valla: Scrittifilosofiae religiosi; introduzione, traduzionc e
Bibliography 953
note a cura di Giorgio Radetti, Sansoni
Radulphus Brito (1978a). Commentary on Boethius' De differentiis topicis, ed. N.J. Green-
Pedersen, CIMAGL 26:1-92
(1978b). 'The Sophism "Omnis homo est omnis homo" ', ed. Jan Pinborg and N. J.
Green-Pedersen, CIMAGL 26:93-114
(1980). Quaestiones super Priscianum minorem, ed. Heinz W. Enders and Jan Pinborg
(Grammatica Speculativa, 3), Frommann-Holzboog.
Raeymaeker, L. de (1947). Introduction à la Philosophie, Editions de l'Institut Supérieur de
Philosophie
(1951). 'Les origines de l'Institut Supérieur de Philosophie à Louvain', Revue
Philosophique de Louvain 49: 505-633
(1952). 'Dominanten in de philosophische persoonlijkheid van Kardinaal Mercier',
Medelingen van de Koninklijke Vlaamse Acadtmit... Klasse der Letteren, XIV: no. 7
(1954). The Philosophy of Being, Herder
Rahman, F. (1958). 'Essence and Existence in Avicenna', Mediaeval and Renaissance
Studies 4:1-16
(1963). 'Ibn Sina' in M. M. Sharif (ed.) A History of Muslim Philosophy, v. 1, Otto
Harrassowitz
Ralph Strode (1493). Consequentiae et Obligationes, cum commentis, Venice
Ramière, H. (1861). Df l'Unité dans l'Enseignement de la Philosophie au sein des Écoles
Catholiques, Régis Ruffet
Ramirez, S. (1955). El derecho de gentes: examen critico de lafilosofiadel derecko de gentes
desde Aristoteles hasta Francisco Suâres, Ediciones Studium
Ramirez, Santiago Maria (1955). Doctrinapolitica de Sto. Thomas, Publicaciones del
Instituto Social Léon XIII. 1
Rand, Edward Kennard (1946). Cicero in the Courtroom of St Thomas Aquinas, Marquette
University Press
Randall, John Herman (1940). 'The Development of Scientific Method in the School of
Padua', JOwrna/ of the History of Ideas ι : 177-206
(1961). The School of Padua and the Emergence of Modem Science, Editrice Antenore
Rashdall, Hastings (1936). TJie Unii>ersi!ifi of Europe in the Middle Agts (2nd ed.), ed.
F. M. Powicke and Α. Β. Emden (3 vols.), Clarendon Press
Raymond Lull (1744). Logica nova, etc., Palma de Mallorca (Reprinted Minerva 1971,
with introduction by C. H. Lohr)
Reardon, John Joseph (1943). Selfishness and the Social Order: A Study in Thomistic Social
Philosophy, Catholic University of America Press
Régis, Louis Marie (1959). Epistemology, trans. Imelda C. Byrne, The Macmillan
Company
Reichling, D. (1893). Das Doctrinale des Alexander de Villa-Dei (Monuments Germaniae
Paedagogica, 12), A. Hofmann
Reilly, George Cajetan (1934). The Psychology of Saint Albert the Great compared with that
of St Thomas, The Catholic University of America
Reilly, James P. (1968). 'Ockham Bibliography: 1950-1967*, Franciscan Studies 28:197-
214
Reina, Maria Elena (1957). 'Giovanni Buridano, Tractatus de suppositionibus', Rivista
critica di storia dellafilosofia12:17S-208 and 323-52
(i959)· Ή problems del linguaggio in Buridano I: Voci e concetti', Rivista critica di
storia dellafilosofia14:367-417
(1960a). Ί1 problema del linguaggio in Buridano II: Significazione e veritâ', Rivista
critica di storia dellafilosofia15:141-65
954 Bibliography
(1960b). 'Il problema del linguaggio in Buridano III: 11 linguaggio', Rivista critica di
storia della filosofia 15:238-64
(1970). 'La prima questione del prologo del "Commento alle Sentenze" di Walter
Catton', Rivista aitka di storia della filosofia 25:4e-74; 290-314
Remigio de Girolami (1934). Tractatus de bono communi excerpted by R. Egenter,
Scholastik 9 : 7 9 - 9 2 (Excepts translated in Minio-Paluello 1956)
(1959). De bono pads, ed. Charles T. Davis, Studi Danteschi 36:123-36
Renan, Ε. (1975). Souvenirs d'enfance et de jeunesse. Chronologie et introduction par Henriette
Pskhari. Notes et archives de l'oeuvre par Laudice Rétat, Garnier-Flammarion
Renard, H. (1956). 'What is St Thomas' Approach to Metaphysics?' The New
Renard, R . G. (1930). La théorie de l'institution: Essai d'ontologie juridique, Sirey
("•935)· 'De l'institution à la conception analogique du droit', Archives de Philosophie du
Droit 3:81-145
Renaudet, Augustin (1953)· Préréforme et humanisme à Paris pendant les premières guerres
d'Italie, 1494-1517 (2nd edn.). Librairie d'Argences
Renna, Thomas J. (1973). 'Kingship in the Disputatio inter clerkum et militem', Speculum
48:675-93
(1978). 'Aristotle and the French monarchy, 1260-1303', Viator 9:309-24
Rescher, Nicholas (1963). Studies in the History of Arabic Logk, University of Pittsburgh
Press
(1964). 'Aristotle's Theory of Modal Syllogisms and Its Interpretation' in M. Bunge
(ed.) The Critical Approach to Science and Philosophy. In Honor of Karl R. Popper, The
Free Press o f Glencoe
Reynolds, L. D., and Wilson, N . G. (1974). Scribes and Scholars: A Guide to the
Transmission of Greek and Latin Literature (2nd ed.), Oxford University Press
Rhabanus Maurus (1851). De institutione clericorum, PL 107
(1901). De institutione clericorum libri très, ed. Aloisius Knoepfler.J. J. Lentner
Ricardus sophista (forthcoming). Abstractiones, ed. Katherine Tachau et al.
Richard of Campsall (1968). Questiones super librum Priorum Analeticorum, in Edward A.
Synan (ed.), The Works of Richard of Campsall, vol. I; Pontifical Institute of
Mediaeval Studies
Richard Fitzralph (1890). De Pauperie Salvatoris, Books 1-4, ed. R . Lane Poole in
Wyclif s Latin Works, vol. 7: De Dominio Divino, Trübner (Reprinted Johnson
Reprint Co. 1966)
Richard Kilvington (forthcoming). Sophismata, ed. N. Kretzmann and Β. E. Kretzmann,
with introduction, translation, and commentary
Richard Swineshead (1520). Liber calculationum, Venice
Riesenberg, P. N . (1956). Inalienability of Sovtrtignty in Medieval Political Thought, AMS
Press
Riet, G. Van (1946). L'Épistémologie thomiste. Recherches sur le problème de la connaissance
dans l'école thomiste contemporaine, Éditions de l'Institut Supérieur de Philosophie de
Louvain
(1953)· 'Ea théorie thomiste de la sensation externe', R f i w philosophique dt Louvain
51:374-408
Riet, S. Van, ed. (1968). Avicenna Latinus: Liber De Anima IV-V, Éditions Orientalistes/
Brill
ed., (1972). Avicenna Latinus: Liber de anima seu sextus de naturalibus I—II—III, E. Peeters
and E.J. Brill
Rijk, L. M. de, ed. (1956). Petrus Abaelardus: Dialectica, Van Gorcum
Bibliography 955
ed. (1959). Garlandus Compotista: Dialectica, Van Gorcum

(1962-7). Logica modernorum: A contribution to the History of Early Terminist Logic.
I: On the Twelfth Century Theories of Fallacy (1962); II, 1: The Origin and Early
Development of the Theory of Supposition; II, 2: Texts and Indices (1967), Van
Gorcum
(1966a) 'Some new Evidence on Twelfth-Century Logic', Vivarium 4:1-57
(1966b). 'Some Notes on the Mediaeval Tract Df insolubilibus with the Edition of a
Tract Dating from the End of the Twelfth Century', Vivarium 4:83-115
(1968a). 'On the genuine text of Peter of Spain's Summule logicales, V Vivarium 6:1-34
(1968b). 'On the genuine text of Peter of Spain's Summule logicales IV, Vivarium
6:69-101
(1969a). 'On the genuine text of Peter of Spain's Summule logicales, ΙΙΓ, Vivarium
7:8-61
(1969b). 'On the genuine text of Peter of Spain's Summule logicales, IV, Vivarium
7:120-62
(1970). Ptttr Abelard, Dialectica (Wijsgerige teksten en Studies, 1) (2nd edn.), Van Gorcum
(1971-3). "The Development of Suppositio naturalis in Mediaeval Logic', Vivarium 9:
71-107; 11:43-79.
(1972). Pfifr of Spain (Petrus Hispanus Portugalensis) Tractatus called afterwards 'Summule
Logicales', Van Gorcum
(1974-6). 'Some Thirteenth Century Tracts on the Game of Obligation', Vivarium
12:94-123; 13:22-54; 14:26-49
(1975). 'Logica Cantabrigiensis - A Fifteenth-Century Cambridge Manual of Logic*,
Revue internationale de philosophie, Grabmann, 29 e année 1131297-315
(1976a). 'Some thirteenth century tracts on the game of obligation', Vivarium
14:26-49
(1976b). 'Richard Billingham's Works on Logic', Vivarium 14:121-38
(1977a). 'On Ancient and Mediaeval Semantics and Metaphysics', Vivarium 15:81-110
(1977b). 'Logica Oxoniensis', Medioevo 3:121-64
Riley-Smith, J. S. C. (1977). What Wert The Crusades?, Macmillan
Rinaldi, M. D. (1973). 'Fortuns e diffusione del "De orthographia" di Giovanni
Tortelli', Italia medioevale e umanistica 16:227—61
Risse, Wilhelm (1964). Dif Logik der Neuzeit (Band I: 1500-1640), Frommann-
Holzboog
(1965). Bibliographia Logica. Verzeichnk der Druckschriften zur Logik mit Angabe ihrtr
Fundorte, (Band I. 1472-1800), Georg Olms
Rivera Recio.J. F. (1951). 'Personajes hispânicos asistentes en 1215 al IV. Concilio de
Lateran', Hispania Sacra 4:335-55
Rivero, M. L. (1976). 'William of Sherwood on Composition and Division', Historio-
graphia Linguistica 3 :17-36
Rivière, J. (1926). Le problème de l'église et de l'état au temps de Philippe le Bel, E.
Champion
Robert Bacon (?) (1978). Sincategoreumata, ed. H. A. G. Braakhuis in Braakhuis 1978
Robert Grosseteste (1514). In Aristotelis Posteriorum Analyticorum libros, Venice
(1912a). De libtro arbitrio, ed. L. Baur, in Baur 1912
(1912b). De luce seu de inchoatione formarum, ed. L. Baur, in Baur 1912
(1912c). Df veritate, ed. L. Baur, in Baur 1912
(1963). Commentarius in VIII libros Physicorum Aristotelis, ed. Richard C. Dales,
University of Colorado
956 Bibliography
Robert Holkot (1518). In quattuor libros Sententiarum quaestiones, Lyon (Reprinted

Minerva 1967)
(1518a). Sex articuli, in Robert Holkot (1518)
Robert Kilwardby (1935). De natura theologiae, ed. F. Stegmüller (Opuscula et textus,
Series scholastica, 17), Aschendorff
(!975)· 'The Commentary on Priscianus maior Ascribed to Robert Kilwardby', ed. J.
Pinborg et al., CIMAGL 15:1-146
(1976). De ortu scientiarum, ed. A. G. Judy (Auctores Britannici Medii Aevi, 4), The
British Academy and The Pontifical Institute of Mediaeval Studies
Roberts, L. D. (1973). 'Indeterminism in Duns Scotus' Doctrine of Human Freedom',
The Modern Schoolman 51:1-16
Robins, R. H. (1951). Ancient and Mediaeval Grammatical Theory in Europe, G. Bell
Robson, J. Α. (1961). Wyclif and the Oxford Schools (Cambridge Studies in Medieval Life
and Thought, Second Series, 8), Cambridge University Press
Roger Bacon (1897-1900). The 'Opus majus' of Roger Bacon, ed. J. H. Bridges (3 vols.),
Clarendon Press (vols. I and II) and Williams and Norgate (vol. Ill)
(1902). T/ie Greek Grammar of Roger Bacon and a Fragment of his Hebrew Grammar, ed.
E. Nolan and S. A. Hirsch, Cambridge University Press
(1905-40). Opera hactenus inedita Rogeri Baconi, ed. Robert Steele (16 fascicules in 12
vols.), Clarendon Press
(1911). Compendium studii theologiae, ed. Hastings Rashdall (British Society of
Franciscan Studies, 3), Aberdeen University Press
(1928). The Opus Majus of Roger Bacon, tr. Robert Belle Burke (2 vols.), Oxford
University Press
(1940). Sumule dialectices, ed Robert Steele, in Opera hactenus inedita Rogeri Baconi, fasc.
15, Clarendon Press
(1978). De signis, ed. K. M. Fredborg et al., in Fredborg et al. 1978
Roger Marston {1932). Quaestiones disputatae, ed. a pp. Collegii s. Bonaventurae ad
Claras Aquas, I., Quaracchi
Rohmer, Jean (1951). 'L'intentionnalité des sensations de Platon à Ockham', Revue des
sciences religieuses 25: 5-39
Rolland-Gosselin, Bernard (1928). La doctrine politique de saint Thomas d'Aquin, M.
Rivière
(1949). La morale chrétienne, Desclée, De Brouwer
Rommen, H. A. (1947). The Natural Law. A study in legal and social history and philosophy,
tr. T. R. Hanley, B. Herder Book Co
Roos, H. (1961). Martini de Dacia Opera (Corpus Philosophorum Danicorum Medii
Aevi, 2), Gad
(1963). 'Ein unbekanntes Sophisma des Boethius de Dacia', Scholastik 38:378-91
(1974)· 'Zur Begriffsgeschichte des Terminus "apparens" in den logischen Schriften des
ausgehenden 13. Jahrhunderts', in Pïrfi« Politica, Festgabe zum 75. GeoMrfiia^ von
Alfons Hufnagel, Frommann-Holzboog
(1977)· 'Henrici Roos in Memoriam. Nachgelassene Papiere', CIMAGL 24:1-84
Rose, Paul Lawrence (1973). 'Keckermann, Bartholomew' in C. C. Gillispie (ed.)
Dictionary of Scientific Biography, Charles Scribner's Sons
Ross, W. D. (1949). Aristotle's Prior and Posterior Analytics. A revised text with introduc-
tion and commentary, Clarendon Press
Rossi, G. P. (1959). La Filosofia nel Collegio Alberoni e il Neotomismo, Divus Thomas
Rossi, P. (1975). 'Per Tedizione del Commentarium in Posteriorum Analyticorum
Libros di Roberto Grossatesta', Rivista di Filosofia Neo-Scolastica 67:489-515
Bibliography 957
Roure, Marie-Louise (1962). 'Le traité "Des propositions insolubles" de Jean de Celaya',
(1970). 'La problématique des propositions insolubles au xiiie siècle et au début du
xive, suivie de l'édition des traités de W. Shyreswood, W. Burleigh et Th.
Bradwardine', Archives d'histoire doctrinale et littéraire du moyen âge 37:205-326
Rouse, Richard H. and Mary A. .(1967). 'John ofSalisbury and the doctrine of tyran-
nicide', Speculum 42:693-709
Rubenstein, Nicolai (1965). 'Marsilius of Padua and Italian political thought of his time'
inj. R. Hale, J. R. L. Highfield, and B. Smalley (eds.) Europe in the Ute Middle
Ages, Northwestern University Press
Russell, Bertrand (1910-11). 'Knowledge by Acquaintance and Knowledge by
Description', Proceedings of the Arktotelian Society, n.s. vol. 11:108-28
Russell, Frederick H. (1975). The Just War in the Middle Ages, Cambridge University
Press
Russell, J. C. (1933). 'The Preferments and "Adiutores" of Robert Grosseteste', Harvard
Theological Review 26:161-72
Ruysschaert, J. (1954). 'Les manuels de grammaire latine composés par Pomponio
Leto', Scriptorium 8:98-107
(1961). Ά propos des trois premières grammaires latines de Pomponio Leto',
Scriptorium 15:68-75
Ryan, Edmund J. (1951). The Role of the 'Sensus Communis' in the Psychology of St
Thomas Aquinas, Messenger Press
Rzadkiewicz, Arnold Ladislas (1949). The Philosophical Basis of Human Liberty according to
St Thomas Aquinas, Catholic University of America Press
Sabbadini, R. (1891). Vita di Guarino Veronese, Tipografia del R. Istituto Sordo-Muti

(1896). La scuola egli studi di Guarino Guarini Veronese, F. Galati
(1902). 'Dei metodi nell' insegnamento della sintassi latina: considerazioni didattiche e
storiche', Rivista difilologia 30:304-14
(1905). Lf scoperte dei codici latini t greet net secoli XIV t XV (2 vols.; reprinted 1967),
Sansoni
(1906). 'Elementi nazionali nella teoria grammaticale dei Romani', Studi italiani di
filologia classica 14:113-25
(1914). Le scoperte dei codici latini e greci ne' secoli XIV e XV: nuove ricerche col riassunto
filologico dei due volumi, Sansoni
(1916). 'Document! guariniani', Atti dell' Accademia di Verona, 4th Series, 18:211-86
(1919). Epistolario di Guarino Veronese, vol. 3: Commento (Miscellanea di storia veneta,
Ser. 3 vol. 14), A spese della Società
(1922). // metodo degli umanisti. Le Monier
Sabine, George H. (1963). A History of Political Theory (3rd ed.), Holt, Rinehart, and
Winston
Sainte-Beuve, C. A. (1854). 'Chateaubriand: anniversaire du Génif du Christianisme',
Causeries du Lundi (17 April 1854)
Sajo, Géza (1958). 'Boèce de Dacie ct les commentaires anonymes inédits dc Munich sur
la Physique et sur la Génération attribués à Siger de Brabant', Archives d'histoire
doctrinale et littéraire du moyen âge 33:2i-58
Sala, Gervasio (1957)· 'Il valore obligatorio della conscienza nei primi scolastici', Studi
francescani 54:174-98
Salamonius, Albertischus Marius (1955). Df Principatu Libri Septem (Pubblicazioni dell'
Istituto di diritto pubblico e di dottrina dello stato della Facoltà di scienza politiche
958 Bibliography
dell' Università di Roma, Ser. 4, n. 5), Giuffrè

Salamucha, J. (1950). 'Die Aussagenlogik bei Wilhelm von Ockham', Franziskanische
Studien 32:97-134
Salembier, Louis (1886). Petrus de Alliaco,]. Lefort
(1931). Le Cardinal Pierre d'Ailly, Georges Frère
Salman, D. (1939). The Mediaeval Latin Translations of Alfarabi's Work', The New
Salmon, V. (1969). [Review o( Cartesian Linguistics by N. Chomsky], Journal of Linguistics
5:165-87
Salter, H. E., cd. (1923). Registrum Annalium Collegii Mertonensis 1483-1521, Clarendon
Press
Sambursky, S. (1959). Physics of the Stoics, Routledge & Kegan Paul
Sanctius Brocensis, Franciscus (1587). Minerva, Salamanca
Sanderson, Robert (1640). Logicae Artis Compendium, Oxford
Sanseverino, G. ( 1862). Philosophia Christiana cum antiqua et nova comparata, Manfredi
Santeler, Josef (1939). Der Platonismus in der Erkenntnislehre des heiligen Thomas von Aquin,
Rauch
Santonastaso, G. (1957). 'Occam e la plenitudo potestatis', Rassegna di scienze filosofiche
10:213-71
Sarton, G. (1931). Introduction to the History of Science, II. 1-2, Williams and Wilkins
Sarubbi, Antonio (1971). Chiesa t stato comunale nelpensiero di Remigio de' Girolami,
Morano
Sassen, F. (1931). 'Siger de Brabant et la doctrine de la double vérité', Revue néoscolas-
tique de philosophie 33:170-9
Savonarola, Girolamo (1534). 'De politia et regno' in Compendium Totius Philosophiae,
Venice
Scaliger, J. C. (1540). De causis linguae Latinae, Lyon
Schafer, Carl (1935). Die Staatslehre des Johannes Gerson, Buchdruckerei Beyer und
Hansknecht
Scheibler, Christoph (1628). Philosophia compendiosa, Oxford
Schenk, Günther (1973). Zur Geschichte der logische Form. Erster Band. Einige Entwicklungs-
tendenzen von der Antike bis zum Ausgang des Mittelalters. VEB Deutscher Verlag der
Wissenschaften
Schepers. H. (1963). Möglichkeit und Kontingenz. Zur Geschichte der philosophischen
Terminologie vor Leibniz (Studi e richerche di storia della filosofia), Edizioni di
'Filosofia'
(1972). 'Holkot contra dicta Crathorn II: Das "significatum per propositionem".
Aufbau und Kritik einer nominalistischen Theorie über den Gegenstand des
Wissens', Philosophisches Jahrbuch der Görres-Gesellschaft 79:106-36
Schilling, O. (1930). Die Staats- und Soziallehre des hl. Thomas von Aquin (2nd ed.).
Hüber
Schmandt, R. H. (1975). 'The Fourth Crusade and the Just-war Theory', 77ie Catholic
Historical Review 61:191-221
Schmaus, M. (1959). 'Leben und Werk Martin Grabmanns' in Mkcellanea Martin
Grabmann (Mitteilungen des Grabmann-Instituts der Universität München, 3),
University of Munich
Schmidt, C. (1855). La Vie et les Travaux de Jean Sturm, Schmidt
Schmidt, R. W. (1966). The Domain of Logic according to Saint Thomas Aquinas, Nijhoff
Schmitt, Charles B. (1971). A Critical Survey and Bibliography of Studies on Renaissance
Aristotelianism 1958-1969, Editrice Antenore
Bibliography 959
(j-975)· 'Philosophy and Science in Sixteenth-Century Universities: Some Preliminary

Comments' inj. Ε. Murdoch and E. Sylla (1975)
Schmitt, F. S. (1936). Ein neues unvollendetes Werk des hl. Anselm von Canterbury (BGPM
XXXIII, 3), Aschendorff
Schmugge, Ludwig (1966). Johannes von Jandun (1285/89-1328): Untersuchungen zur
Biographie und Sozialtheorie eines lateinischen Avetroisten (Pariser Historische Studien,
No. 5), Anton Hiersemann
Schneider, Arthur (1903-6). Die Psychologie Alberts des Grossen nach den Quellen
dargestellt (BGPM IV, 5-6), Aschendorff
Scholz, H. (1961). A Concise History of Logic, Philosophical Library
Scholz, R. (1902). Aegidius von Rom, Druck der Union deutsche Verlagsgesellschaft
(1911-14). Unbekannte Kirchenpolitische Streitschriften aus der Zeit Ludwigs des Bayern
(1327-1354) (2 vols.), Loescher
(1936). "Marsilius von Padua und die Genesis des modernen Staatsbewusstseins',
Historische Zeitschrift 156:88-103
(1962). Die Publizistik zur Zeit Philipps des Schonen und Bonifaz VIII, P. Schippers [first
published 1903)
Schrimpf, G. (1966). Die Axiomenschrifi dts Boethius (De Hebdomadibus) als Philosophisches
Lehrbuch des Mittelalters, Brill
Schütz, Ludwig (1895). Thomas-Lexikon (2nd edn.), Schöningh
Schulz, F. (1945). 'Bracton on kingship', English Historical Review 60:136-76
Schumacher, Leo Sebastian (1949). The Philosophy of the Equitable Distribution of Wealth,
Catholic University of America Press
Scott, T. K. (1965). 'John Buridan on the Objects of Demonstrative Science', Speculum
40:654-73
(1969). 'Ockham on Evidence, Necessity and Intuition', Journal of the History of
Philosophy 7:27-49
(1971). 'Nicholas of Autrecourt, Buridan, and Ockhamism', Journal of the Hktory of
Philosophy 9:15-41
Serene, Eileen F. (1979) "Robert Grosseteste on Induction and Demonstrative Science',
Synthese 40:97-115
Seton, John (1631). Dialectica annotationibus Petri Carteri, Cambridge
Sextus Empiricus (1958), Opera, ed. H. Mutschmann, vol. I, Teubner
Shapiro, Herman (1957). Motion, Time and Place according to William Ockham, The
(1959)· 'Walter Burley and the Intension and Remission of Forms', Speculum
34:413-27
Shapiro, Herman and Charlotte (1965). 'De Primo ct Ultimo Instanti des Walter
Burley', Archiv fir Geschichte der Philosophie 47:157-73
Sharp, D. (1934). 'Simon of Fâversham's Quaestiones super tertium De anima', Archives
d'histoire doctrinale et littéraire du moyen âge 9:307—68
Sharp, E. D. (1964). Franciscan Philosophy at Oxford in the Thirteenth Century, Oxford
University Press (first published 1930)
Sheppard, Vincent F. (1949). Religion and the Concept of Democracy: A Thomistk Study in
Social Philosophy, Catholic University of America Press
Shields A. L., ed. (1949). Averrois Cordubensis Compendia Librorum qui Parva Naturalia
Vocantur (Corpus Commentariorum Averrois in Aristotelem), Mediaeval Academy
of America
Siematkowska, Z. K. (i960). 'Avant l'exil de Gilles de Rome: Au sujet d'une dispute sur
les "Theoremata de esse et essentia" de Gilles de Rome', Mediaevalia Philosophica
96ο Bibliography
Polonorum 7: 3-67
Siewerth, G. (1954). Die menschliche Willensfreiheit. Texte zur thomistischen
Freiheitslehre, Schwann
[Pseudo-] Siger of Brabant (1941). Questions sur la Physique d'Aristote ed. P. Delhaye,
Louvain, Institut supérieur de philosophie
Siger of Brabant (1948). Sigfr de Brabant. Questions sur la métaphysique (Quaestiones in
Metaphysicam), ed. C. A. Graiff, Éditions de l'Institut Supérieur de Philosophie dc
Louvain
(1966). Die Questiones metaphysice très des Siger von Brabant', ed. J. Vennebusch,
Archiv für Geschichte der Philosophie 48:175-89
(1972a). Quaestiones in tertium De anima, De anima intellectiva, De aetemitate mundi ed.
B. Bazân, Publications Universitaires de Louvain
(1972b). Les Quaestiones super librum de causis de Siger de Brabant, cd. A. Marlasca,
Publications Universitaires de Louvain
(1974). Écrits de logique, de morale et de physique ed. B. Bazân, Publications Universi-
taires de Louvain
Siger of Courtrai (1913). Ars Priorum in G. Wallerand (ed.) Les Oeuvres de Siger de
Courtrai (Les Philosophes Belges. Textes et Études), Louvain Institut Supérieur de
Philosophie
(1978). Summa Modorum Significandi - Sophismata, ed. J. Pinborg (Amsterdam Studies
in the Theory and History of Linguistic Science, III: Studies in the History of
Linguistics, 14), Benjamins
Sigmund, Paul E. (1963). Nicholas of Cusa and Medieval Political Thought, Harvard
University Press
(1971). Natural Law in Political Thought, Winthrop
Sikes, J. G. (1932). Peifr Abailard, Cambridge University Press
(1934)· 'John de Pouilli and Peter de la Palu', English Historical Review 49:219—40
Simon of Faversham (1957). Opera omnia, vol. 1; Opera logica, tomus prior: Quaestiones
super libro Porphyrii, Quaestiones super libro Praedicamentorum, Quaestiones super libro
Perihermeneias, ed. P. Mazzsrella, Cedam
(forthcoming). Quaestiones novae et veteres super librum Elenchorum, ed. F. Del Punta,
S. Ebbesen, T. Izbicki, J. Longeway, E. Serene, and E. Stump, Pontifical Institute of
Mediaeval Studies
Simon of Tournai (1966). Expositio super Simbolum, ed. N. Häring in Archives d'histoire
doctrinale et littéraire du moyen âge 41
Simonin, H. D. (1930). 'La connaissance des singuliers matériels d'après les maîtres
franciscains de la fin du xiüe siècle' in Mélanges Mandonnet. Études d'histoire littéraire
et doctrinale du moyen âge (2 vols.),J. Vrin
(1930a). 'La notion d' "intentio" dans l'oeuvre de S. Thomas d'Aquin*, Revue des
sciences philosophiques et théologiques 19:445-63
(1930b). 'La connaissance humaine des singuliers matériels d'après les maîtres francis-
cains de la fin du XIIIe siècle' in Mélanges Mandonnet, vol. 2 (Bibliothèque
Thomiste, vol. 14), Vrin
(1931). 'Connaissance et similitude', Revue des sciences philosophiques et théologiques
20:293-303
Simplicius (1907). In Aristotelis Catégoriels Commentarium, ed. C. Kalbfleisch
(Commentaria in Aristotelem Graeca VIII), Reimer
(1971-5). Commentaire sur les Catégories d'Aristote, ed. A. Pattin (Corpus latinum
commentariorum in Aristotelem Graecorum, V. 1-2), Publications universitaires de
Louvain (1971); Brill (1975)
Bibliography 961
Siwek (1948). Psychologia Metaphysica. Pontificia Universitas Gregoriana

Skinner, Quentin (1978). The Foundations of Modem Political Thought. Vol. 1: The
Renaissance. Vol. 2: Thf Age of Reformation, Cambridge University Press
Smalley, Beryl, ed. (1965). Trends in Medieval Political Thought, Basil Blackwell
U973)- The Becket Conflict and the Schools: A Study of Intellectuals in Politics, Rowman
and Littlefield
Smet, A. J., ed. (1968). Alexandre äAphrodisias, Commentaire sur les Météores d'Aristote
(Corpus Latinum Commentariorum in Aristotelem Graecorum IV), Publications
Universitaires de Louvain/Éditions Béatrice-Nauwelaerts
Smiglecki, Martin (1634). Logica, Oxford
Smith, V. (1954). 'Prime Mover, Physical and Metaphysical Considerations', Proceedings
of the American Catholic Philosophical Association 28:78-94
(1958). The General Science of Nature, Bruce
Sobocinski, B. (1956). 'In memoriam Jan Lukasiewicz', Philosophical Studies (Maynooth)
6:3-46
Sorabji, Richard (1972). Aristotle on Memory, Duckworth
Soto, Domingo de (1567). Libri decern de iustitia et iure, Antwerp
(1587). In Porphyrii Isagogen, Aristotelis Categorias librosque de demonstratione commentaria,
Venice (Reprinted Minerva 1967)
Southern, R. W. (1963). Saint Anselm and his Biographer, Cambridge University Press
(1970). Medieval Humanism and other Studies. Blackwell
Southern, R. W. and F. S. Schmitt (1969). Memorials of St Anselm, British Academy
Spade, Paul Vincent (1969). An Anonymous Fourteenth-Century Treatise on Insolubles: Text
and Study, L. S. M. thesis, Pontifical Institute of Mediaeval Studies.
(1971). 'An Anonymous Tract on Insolubilia from Ms Vat. lat. 674: An Edition and
Analysis of the Text', Vivarium 9:1-18
(x973)· 'The Origins of the Mediaeval /nioin&iiia-Literature', Franciscan Studies
33:292-309
(1973a). 'The Treatises On Modal Propositions and On Hypothetical Propositions by
Richard Lavenham', Mediaeval Studies 35:49-59
(1974a). 'Five Logical Tracts by Richard Lavenham', inj. Reginald O'Donnell (ed.),
Essays in Honour of Anton Charles Pegis, Pontifical Institute of Mediaeval Studies
(1974b). 'Ockham on Self-Reference', Notre Dame Journal of Formal Logic 15:298-3O0
(1974c). 'Ockham's Rule of Supposition: Two Conflicts in His Theory', Vivarium
12:63-73
(1975a). The Mediaeval Liar: A Catalogue of the Insolubilia-Literature (Subsidia
Mediaevalia, 5), Pontifical Institute of Mediaeval Studies
(1975b). 'Notes on Some Manuscripts of Logical and Physical Works by Richard
Lavenham', Manuscripta 19:139-46
(1975c)· 'Ockham's Distinctions between Absolute and Connotative Terms', Vivarium
13:55-75
(i975d). 'Some Epistemological Implications of the Burley-Ockham Dispute',
(1976). 'Robert Fland's Consequentiae: An Edition', Mediaeval Studies 38:54-84
(i977)- 'Roger Swyneshed's Obligationes: Edition and Comments', Archives d'histoire
(1978a). 'John Buridan on the Liar: A Study and Reconstruction', Notre Dame Journal
of Formal Logk 19:579-90
(1978b). 'Richard Lavenham's Obligationes: Edition and Comments', Rivista critica di
storia dellafilosofia33:225-242
962 Bibliography
(1978c). 'Robert Fland's Insolubilia: An Edition, with Comments on the Dating of

Fland's Works', Mediaeval Studies 4 0 : 5 6 - 8 0
(1979a). 'Roger Swyneshed's Insolubilia: Edition and Comments', Archives d'histoire
(1979b). William Heytesbury: On 'Insoluble' Sentences: Chapter One of His Rules for
Solving Sophisms (Mediaeval Sources in Translation, 21), Pontifical Institute of
Mediaeval Studies
(1979c). 'Recent Research on Medieval Logic", Synthese 4 0 : 3 - 1 8
(1980). 'Robert Fland's Obligationes: An Edition', Mediaeval Studies42:41-60
(1980a). 'Richard Lavenham and the Cambridge Logic', Historiographia Linguislka 7:
241-7
(forthcoming a). 'Notes on Richard Lavenham's So Called Summulae logicales, with a
Partial Edition o f the Text', Franciscan Studies
(forthcoming b). 'Quasi-Aristotelianism' in Kretzmann (forthcoming b)
Spade, Paul Vincent, and Eleonore Stump (forthcoming), 'The Treatise on Obligations
Attributed to William of Sherwood'
Spettmann, H. (1923). 'Der Ethikkommentar des Johannes Peckham' in Abhandlungen
zur Geschichte der Philosophie des Mittelalters (BGPM suppl. Bd. 2), Aschendorff
Stahl, William Harris (1971). Martianus Capella and the Seven Liberal Arts. Volume I, The
Quadrivium of Martianus Capella. Latin Traditions in tht Mathtmatkal Sciences 50 B.c-
A.D. 1250, Columbia University Press
Starkey, Thomas (1948). A Dialogue between Reginald Pole and Thomas Lupset, ed.
Kathleen M. Burton, Chatto and Windus
Steele, R., ed. (1920). Secretum secretorum (Opera hactenus inedita Rogeri Baconi, 5),
Clarendon Press
ed. (1935). Quaestiones supra librum De causis (Opera hactenus inedita Rogeri Baconi,
12), Clarendon Press
Steenberghen, F. Van (1930). 'Siger de Brabant d'après ses oeuvres inédites', Revue
néoscolastique de philosophie 32:403 - 2 3
(1931). Sigfr de Brabant d'après ses oeuvres inédites, vol. 1: Les oeuvres inédites (Les
philosophes belges. No. 12), Éditions de L'Institut Supérieur de Philosophie
(1938). Les oeuvres et la doctrine de Siger de Brabant (Académie royale de Belgique.
Classe des lettres et des sciences morales et politiques. Mémoires, vol. 39, fasc. 3),
Hayez
(1942). Siger de Brabant d'après ses oeuvres inédites, vol. 2: Siger dans l'histoire de
l'aristotélisme (Les philosophes belges, N o . 13), Éditions de L'Institut Supérieur de
Philosophie
(1948). 'Maurice de Wulf, historien de la philosophie médiévale', Revue Philosophique
de Louvain 46:421-47
(1951). 'Siger o f Brabant', The Modem Schoolman 29:11-27
(1955). Aristotle in the West. The Origins of Latin Aristotelianism, tr. L.Johnston,
Nauwelaerts
(1956). 'Nouvelles recherches sur Siger de Brabant et son école', Revue philosophique de
Louvain 54:130-47
(1966). La Philosophie au XIII' siècle (Philosophes médiévaux 9), Publications uni-
versitaires de Louvain Béatrice-Nauwelaerts
(1971). 'Un commentaire semi-averroïste du traité de l'âme (Oxford, Merton College
275, f. 67-84; Munich, Clm 9559, f. 74-82)' in Maurice Giele, Fernand Van
Steenberghen, Bernard Bazân (eds.) Trois commentaires anonymes sur le traité de l'âme
d'Aristote (Philosophes médiévaux, vol. 11), Publications Universitaires and
Bibliography 963
Béatrice-Nauwelaerts
(1974). Introduction à l'étude de la philosophie médiévale (Philosophes médiévaux, vol.
18), Publications Universitaires and Béatrice-Nauwelaerts
(1976). ' "Averro'isme" et "double vérité" au siècle de saint Louis' in Septième centenaire
de la mort de saint Loua, Actes des roiicxju« dt Royaumont et de Paris (21-27 mai 1970)1
Société d'Édition 'Les Belles Lettres'
(1977). Mdilrf Siger de Brabant (Philosophes médiévaux, No. 21), Publications
Universitaires
(1978). 'Siger de Brabant et la condamnation de l'aristotélisme hétérodoxe le 7 mars
1277' Bulletin de l'Académie royale de Belgique, Classes des lettres et des sciences morales et
politiques, 5th scr., Tome 44, Académie Royale de Belgique
(1979)· 'Etienne Gilson, historien de la pensée médiévale'. Revue philosophique de
Louvain 771487-508
(1980). Thomas Aquinas and Radical Arktotelianism, Catholic University of America
Press
Stegmüller, Friedrich (1935). Geschichte der Molinkmus: Neue Molinaschriften (BGPM
XXXII), Aschendorff
Stelling-Michaud, Sven, ed. (1967). Les Universités Européennes du XIV au XVIIF siècle:
Actes du Colloque International à l'occasion du VF centenaire de V Université Jagellonne de
Cracovie, Droz
Steneck, Nicholas H. (1974). 'Albert the Great on the Classification and Localization of
the Internal Senses', Isis 65:193-211
(1980). 'Albert on the Psychology of Sense Perception' in James A. Weisheipl (ed.)
Albertus Magnus and the Sciences: Commemorative Essays 1980, Pontifical Institute of
Mediaeval Studies
Stevens, H.J. (1975). 'Lorenzo Valla and Isidore of Seville', Traditio 31 =343-8
Stiegler, Α. (1958). Der kirchliche Rechtsbegriff·. Elemente und Phasen seiner Erkenntnis
geschichte, Schell Steiner
Stone, Lawrence, ed. (1975). The University in Society (vol. I), Princeton University
Press
Stout, Harry S. (1974). 'Marsilius of Padua and the Henrician Reformation', Church
History 43:308-18
Stratenwerth, G. (1951). Die Naturrechtslehre des Johannes Duns Scotus, Vandenhoeck und
Ruprecht
Strato (1967-9). Fragments, in vol. V of Wehrli (1967-9)
Strauss, L. and J. Cropsy, eds. (1963). A History of Political Philosophy, Rand McNally
Strayer, Joseph R. (1970). On the Medieval Origins of the Modern State, Princeton
University Press
Streckenbach, G. (ed.) (1972). 'Paulus Niavis, Latinum ydeoma pro novellis studentibus',
Mittellateinisches Jahrbuch 7:187-251
Stump, Eleonore (1974). 'Boethius's Works on the Topics', Vivarium 12:77-93
(1978). Boethius's De topicis differentiis. Translated, with notes and essays on the text;
Cornell University
(1980). 'Dialectic in the Eleventh and Twelfth Centuries: Garlandus Compotista',
History and Philosophy of Logic 1:1-18
(1980a). 'William of Sherwood's Treatise on Obligations', Historiograpkia Linguistka
7:249-61
(1981). 'Boethian Theory of Topics and its Place in Early Scholastic Logic' in
Proceedings of the International Boethius Conference in Pavia, ed. L. Obertello,
Brill
964 Bibliography
(forthcoming a). 'Dialectic in the Eleventh to the Thirteenth Centuries', in David

Wagner (ed.) The Seven Liberal Arts, Indiana University Press
(forthcoming b). 'Peter of Spain on the Topics' in Boethius and tht Liberal Arts, ed.
Michael Masi, Verlag Peter Lang
(forthcoming c). 'Theology and Physics in De saaamento altaris: Ockham's Theory of
Indivisibles' in Kretzmann (forthcoming b)
(forthcoming d). Boethius's In Ckeronis Topica
Stump, Eleonore and Paul Vincent Spade (forthcoming). "The Treatise on Obligations
Attributed to William of Sherwood'
Suarez, Francisco (1612). Tractatus de legibus ac Deo legislature, Coimbra
(1856-77). Opera omnia (26 vols.), L. Vives
(1944). SelectionsfromThree Works, ed. James Brown Scott (2 vols.) (The Classics of
International Law, 20), Clarendon Press
(1947). On the various kinds of Distinctions, tr. C. Vollen, Marquette University Press
(1964). On Formal and Universal Unity, tr. with an introduction by J. F. Ross,
Marquette University Press
(1965). Disputationes metaphysicae, Georg Olms
(1971-7). De Legibus, ed. Luciano Perena (6 vols.) (Corpus Hispanorum de Pace,
11-17), Consejo Superior de Investigaciones Cientificas, Instituto Francisco de
Vitoria
Suârez, G. (1948). 'FJ pensamiento de Egidio Romano en torno a la distincion de esencia
y existencia', La Ciencia Tomista 75:66-99; 230-72
Südhoff, Κ. (1914). 'Die kurze "Vita" und das Verzeichnis der Arbeiten Gerhards von
Cremona', Archiv fir Geschichte der Medizin 8:73-82
Sullivan, Mark W. (1967). Apuleian Logic (Studies in Logic and the Foundations of
Mathematics), North-Holland
Susemihl, F., ed. (1872). Aristotelk Politicorum libri octo, cum vetusta translalione Guilelmi de
Moerbeke, Teubner
Sweeney, Leo (1963). 'Existence/Essence in Thomas Aquinas's Early Writings',
Proceedings of the American Catholic Philosophical Association 37:97-131
Sweeney, Leo and C.J. Ermatinger (1958). 'Divine Infinity according to Richard
Fishacre', The Modern Schoolman 35:191-235
Swiezawski, S. (1934). 'Les intentions premiers et les intentions secondes chez Jean Duns
Scot', Archives d'histoire doctrinale et littéraire du moyen âge 9:205-60
Swiniarski, John J. (1970). Ά New Presentation of Ockham's Theory of Supposition
with an Evaluation of Some Contemporary Criticisms', Franciscan Studies
30:181-217
Sylla, Edith (1970). 77ie Oxford Calculators and the Mathematics of Motion, 1320-50, Physics
and Measurement by Latitudes, unpublished Ph.D. dissertation, Harvard University
(1971). 'Medieval Quantifications of Qualities: the "Merton School"', Archive for
History of Exact Sciences 8 =9-39
(1973). 'Medieval Concepts of the Latitude of Forms: The Oxford Calculators',
(1979). 'The A Posteriori Foundations of Natural Science. Some Medieval Commen-
taries on Aristotle's Physics, Book I, Chapters 1 and 2', Synthese 40:147-87
(forthcoming). 'Infinite Indivisibles and Continuity in Fourteenth-Century Theories of
Alteration' in Kretzmann (forthcoming b)
Synan, E. A. (1962). 'Sixteen Sayings by Richard of Campsall on Contingency and
Foreknowledge', Mediaeval Studies 24:250-62
Bibliography 965
Tachau, Katherine, Calvin Normore, Mary Sirridge, and Paul Streveler, eds. (forthcom-
ing). Ricardus sophista: Abstractiones
Tacitus (1949). Cornelii Taciti Dialogus de oratotibus, ed. M. L. de Gubematis (Corpus
Scriptorum Latinorum Paravianum), G. P. Para via
Tambuyser, R. (1958). 'L'érection de la chaire de philosophie thomiste à l'Université de
Louvain (1880-1882)'. Revue philosophique de Louvain 56:478-509
Tatnall, Edith Comfort (1965). Church and State According to John Wyclyf, University
Microfilms
Terrero, José Riesco (i960). 'Juan de Janduno y el Gandavense' Salmanticensis commen-
tarius de sacris disciplinis 7:331 -43
Teufel, Waldemar (1977). Universitas studii Tuwingtnsis: Dit Tübingtr Universitätsvtr-
fassung in vorreformatorischer Ztit (1477-1534), Mohr
Themistius (1900). Analyticorum Posteriorum paraphrask, ed. M. Wallies (Commentaria in
Aristotelem Graeca V. i), G. Reimer
Theophrastus (1973). Die logischen Fragmente des Thtophrast, ed. Andreas Graeser (Kleine
Texte für Vorlesungen und Uebungen, 191), De Gruyter
Théry, G. (1931). 'L'Augustinisme médiéval et le problème de l'unité de la forme
substantielle', in Acta Hebdomadae Augustinianae-Thomisticae, Marietti
Thierry of Chartres (1971). Commentaries on Boethius by Thierry of Chartres and his School,
ed. Nikolaus M. Häring, Pontifical Institute of Mediaeval Studies
Thomas Aquinas (1874-89). Opera omnia, ed. S. E. Fretté and P. Mare (34 vols.),
L. Vives
(1882-). S. Thomae Aquinatk Doctoris Angelici. Opera Omnia. Iussu impensaque Leonis
XIII, P.M. édita (Leonine ed.) Vatican Polyglot Press
(1882). Commentaria in Aristotelis Peri Hermeneias 'et Posteriorum Analyticorum in Thomas
Aquinas 1882-, vol. 1
(1888-1906). Summa theologiae in Thomas Aquinas 1882-, vols. 4-12
(1889). In libros Aristotelis De caelo et mundo expositio in Thomas Aquinas 1882—, vol. 3
(1925). Quaestiones disputatae, ed. P. Mandonnet (3 vols.), P. Léthielleux
(1929-56). Scriptum super Sententiis, ed. P. Mandonnet and M. F. Moos, P. Léthielleux
(1933). In decern libros Ethicorum Aristotelis ad Nicomachum expositio, ed. A. M. Pirotta,
Marietti
(1934a). Summa contra Gentiles, in Thomas Aquinas 1882-, vols. 13-15
(1948a). In Aristotelis librum De anima commentarium, ed. A. M. Pirotta, Marietti
(1948b). Lf 'De ente et essentia' de s. Thomas d'Aquin, ed. M.-D. Roland-Gosselin,
J.Vrin
(1948c). In librum Boethii De Trinitate quaestiones quinta et sexta, Société Philosophique
and E. Nauwelaerts
(1948-50). Summa theologiae, ed. P. Caramello (from the Leonine ed.) (3 vols.),
Marietti
(1949a). Quaestiones dkputatae, ed. R. M. Spiazzi et al. (8th ed.) (2 vols.), Marietti
(1949b). De veritate in Thomas Aquinas 1949a
(1949c). On Kingship to the King of Cyprus, tr. Gerald B. Phelan, rev. with introd. and
notes by I. T. Eschmann, Pontifical Institute of Medieval Studies
(1950a). De principiis naturae. Société Philosophique and Ε. Nauwelaerts
(1950b). In librum beati Dionysii Dt divinis nominibus expositio, ed. C. Pera, Marietti
(1950c). Summa Theologiae: Prima Pars; Prima Secundae, Marietti
(1951). In octo libros Politicorum Aristotelis expositio, ed. R. M. Spiazzi, Marietti
(19533)· Quaestiones disputatae, ed. R. M. Spiazzi ft αϊ. (9th rev. ed.) (2 vols.), Marietti
966 Bibliography
(1953b). Quaestiones disputatae de potentia, in Thomas Aquinas 1953a

(1953c). Quaestio disputata de spiritualibus creaturis in Thomas Aquinas 1953a
(1954a). Opuscula philosophica, ed. R. M. Spiazzi, Marietti
(1954b). Df wife ii essentia in Thomas Aquinas 1954a
(1954c). Df regimine principum ad regem Cypri in Thomas Aquinas 1954a
(i954d). Sancti Thomae de Aquino super Librum de Causis Expositio, ed. H. D. Saffrey,
Société Philosophique
(1954e). In octo libros Physicorum Aristotelis expositio, ed M. Maggiolo, Marietti
(1954t). Opuscula theologica, ed. R. A. Verardo, R. M. Spiazzi, and M. Calcaterra (2
vols.), Marietti
(1954g). Expositio super Boetium De trinitate et De hebdomadibus in Thomas Aquinas
I954f
(1954h). Compendium theologiae in Thomas Aquinas I954f
(1955). Expositio super librum Boethii De Trinitate, ed. B. Decker (Studien und Texte
zur Geistesgeschichte des Mittelalters, 4), Brill
(1956). Quaestiones quodlibetales, ed. R. M. Spiazzi (9th rev. edn.), Marietti
(1957). Tractatus de unitate intellectus contra averroistas, ed. L. W. Keeler (Textus et
Documenta, Series philosophica, 12), Pontificia Universitas Gregoriana
(1959). In Arktotelis librum De anima commentarium, ed. A. M. Pirotta (4th edn.),
Marietti
(1961-7). Liber de veritate Catholkaefidei contra errores infidelium, seu Summa contra
gentiles, ed. C. Pera, P. Marc, and P. Carmello (3 vols.), Marietti
(1963). Treatise on Separate Substances (Tractatus de Substantik Separatis), ed. and tr.
F. J. Lescoe, Saint Joseph College, West Hartford, Conn.
(1964a). In Aristotelis libros Peri hermeneias et Posteriorum analyticorum expositio, ed.
R. M. Spiazzi (2nd edn.), Marietti
(1964b). In decern libros Ethicorum Aristotelis ad Nicomachum expositio, ed. R. M. Spiazzi
(3rd edn.), Marietti
(1964c). Df veritate in Thomas Aquinas 1964-5
(1964—5). Quaestiones disputatae, ed. R. M. Spiazzi ei al. (loth edn.) (2 vols.), Marietti
(1964—76). Summa theologiae, Latin text and English translation, Introductions, Notes,
Appendices, and Glossaries (60 vols.), Blackfriars, in conjunction with Eyre and
Spottiswoode and McGraw Hill
(1969). Sententia libri Ethicorum in Thomas Aquinas 1882-, vol. 47, pts. 1-2
(1971a). Sententia libri Politicorum in Thomas Aquinas 1882-, vol. 48
(1971b). In duodecim libros Metaphyskorum Aristotelis expositio, ed. M. R. Cathala and
R. M. Spiazzi (2nd edn.), Marietti
(1976). Defallaciis in Thomas Aquinas 1882-, vol. 43
Thomas Bradwardine (1503). Geometria speculativa, Paris
(1618). Summa de causa dei contra Pelagium et de virtute causarum ad suos Mertonenses libri
1res, Lyon (Reprinted Minerva 1964)
(1955). Thomas of Bradwardine His Tractatus de Proportionibus, ed. and tr. H. L. Crosby,
Jr., University of Wisconsin Press
Thomas, Ernest C , ed. and tr. (1888). The Philobiblon of Richard de Bury, Kegan Paul,
Trench and Co
Thomas, ivo (1953), 'Kilwardby on Conversion', Dominican Studies 6:56-76
(*954)- 'Maxims in Kilwardby', Dominican Studies 7; 129-46
(1964). 'Medieval Aftermath: Oxford Logic and Logicians of the Seventeenth
Century' in Oxford Studies Presented to Daniel Callus (Oxford Historical Society, 16),
Clarendon Press
Bibliography 967
Thomas, Rudolph, ed. (1970). Pfirtu Abaelardus: Dialogus inter Philosophum, Iudaeum et
Christianum, Frommann-Holzboog
Thomson, S. Harrison (1940). The Writings of Robert Grosseteste, Cambridge University
Press
(1947). 'Walter Burley's commentary on the Politics of Aristotle' in Mélanges Auguste
Pelzer (Receuil des Travaux d'histoire et de philosophie, 3rd series, 26) Université
de Louvain
(1969). Latin Bookhands of the Later Middle Ages 1100-1500, Cambridge University Press
Thorndike, Lynn (1934). 'Calculator and the Rise of Mathematics', in his History of
Magic and Experimental Science, vol. 3, Columbia University Press
(1944). University Records and Life in the Middle Ages, Columbia University Press
(Reprint 1971, Octagon Books)
(i°S9)· 'John of Seville', Speculum 34:20-38
(1965). Michael Scot, Nelson
Thurot, Charles (1868). Extraits dc divers manuscrits latins pour servir à l'histoire des doctrines
grammaticales au moyen-âge, Imprimerie impériale
Tierney, Brian (1953). 'The Canonists and the medieval state', Review of Politics
15:378-88
(1954). Ockham, the conciliar theory and the canonists'. The Journal of the History of
Ideas 15:40-70
(1955a). 'Grosseteste and the theory of papal sovereignty', Journal of Ecclesiastical
History 6:1-17
(1955b). Foundations of the Conciliar Theory, Cambridge University Press
(1966). 'Medieval canon law and western constitutionalism', The Catholic Historical
Review 52:1-17
(1972). Origins of Papal Infallibility, 1150-1350 (Studies in the History of Christian
Thought), Brill
(1975)· '"Divided sovereignty" at Constance: a problem of medieval and early
modern political theory', Annuarium Historiae Conciliorum 7:238-56
Tihon, P. (1966). Foi et théologie selon Godefroid de Fontaines, Desclée de Brouwer
Tooke.J. (1065). The Just War in Aquinas and Grotius, SPCK
Trapé, G. (15)64). Ί1 Platonismo di Egidio Romano', Aquinas 7: 309-44
(1966). Ί1 Neoplatonismo di Egidio Romano nel commento al "De causis"', Aquinas
9:49-86
(1967). 'La dottrina della partecipazione in Egidio Romano', Aquinas 10:170-93
(1968). 'Caratteristiche dell' "esse" partecipato in Egidio Romano', Lateranum
34:351-68
(1069). 'L' "esse" partecipato e distinzione reale in Egidio Romano', Aquinas
12:443-68
Trapp, Damasus (1956). 'Augustinian Theology of the Fourteenth Century',
Augustiniana 6:146-274
(1957). 'Peter Ceffons of Clairvaux', Recherches de théologie ancienne at médiévale
24:101-54
(1958). 'Gregory of Rimini: Manuscripts, Editions, and Additions', Augustiniana
8:425-43
(1962). 'New Approaches to Gregory of Rimini', Augustinianum 2:115-30
Trentman, J. A. (1975). 'Speculative Grammar and Transformational Grammar; A
Comparison of Philosophical Presuppositions', in Herman Parret (ed.) History of
Linguktic Thought and Contemporary Linguistics, De Gruytei
(1976). 'The Study of Logic and Language in England in the Early 17th Century',
968 Bibliography
Historiagraphia Linguistica 3 :179—201

(1978). 'Bad Names: A Linguistic Argument in Late Medieval Natural Law Theories',
Nous 12:29-39
(forthcoming). 'The Authorship of the Seventeenth-Century Cambridge Directionsfor
a Student in the Universitie', Transactions of the Cambridgt Bibliographical Socitty
Trinkaus, C. (1970). In Our Image and Likeness: Humanity and Divinity in Italian Humankt
Thought, Constable
T r o m p , S. (1933). 'Progressus doctrinalis in tractatibus S. Roberti Bellarmini de
Praedestinatione', Gregorianum 14:313-55
Troxler, Ferdinand (1973). Die Lehre vom Eigentum bei Thomas von Aquin und Karl Marx:
Eine Konfrontation, Imba-Verlag
Tschackert, Paul (1877). Peter von Ailly (Petrus de Alliaco): Zur Geschichte des grossen
abendländischen Schisma und der Reformconcilien von Pisa und Constanz, Friedrich
Andreas Perthes
Tuck, Richard (1977). Natural Rights Theories before Locke, unpublished Ph.D. disser-
tation, Cambridge University
Turk, Egbert (1977). Nugae curialium: Le règne d'Henri II Plantegenit (1145-1189) et
l'éthique politique (Publications du Centre de Recherche d'Histoire et de Philologie de
la IV e Section de l'École pratique des Hautes Études, Paris, V: Hautes Études
Médiévales et Modernes, 28), Droz
Tusquets, J. (1923). 'El cardenal J.-T. de Boxadors i la seva influèneia en el renaixement
del tomisme*, Anuari de la Societal Catalana de Filosofia, Institut d'Estudis Catalans
1:243-304
Tweedale, M. M. (1976). Abailard on Universals, North-Holland
Tyrrell, Francis M. (1948). The Role of Assent in Judgment, A Thomistic Study, The
Catholic University of America Press
Ubaghs, G. C. (i860). Essai d'idéologie ontologique, ou considérations sur la nature de nos idées
et sur l'ontologisme en général. De Vanlinthout
(1861). DM problème ontologique des universaux et de la véritable signification du réalisme,
Vanlinthout
UUman, B . L. (1955). Studies in the Italian Renaissance, Edizioni di storia e letteratura
Ullmann, W . (1944). 'The influence ofjohn ofSalisbury on medieval Italian jurists',
English Historical Review 59:384-92
(1949). 'The Development of the Medieval Idea of Sovereignty', English Historical
Review 6 4 : 1 - 3 3
(1961). Principles of Government and Politics in the Middle Ages, Methuen
(1965). A History of Political Thought: the Middle Ages, Penguin Books Ltd
(1966a). The Individual and Society in the Middle Ages, Johns Hopkins University Press
(1966b). Principles of Government and Politics in the Middle Ages (2nd edn.), Methuen
(1970). The Growth of Papal Government in the Middle Ages (4th edn.), Methuen
(1975). Law and Politics in the Middle Ages. An Introduction to the Sources of Medieval
Political Ideas, Cornell University Press
Una Juarez, A. (1978). La filosofia del sigh xiv. Contexte cultural de Walter Burley.
Biblioteca 'La ciudad de Dios', Real Monasterio de el Escorial
Untervintl, L. von (1955). 'Die Intuitionslehre bei Vitalis de Furno', Collectanea
Franciscana 25 :53 —III; 225—58
Urban, Linwood (1973). 'William of Ockham's theological ethics', Franciscan Studies
33:310-50
Ushaw (1839). Sf Cuthbert's College, Ushaw, Durham. The course of studies with the theses
and examination papers, Dolman
Bibliography 969
Vahlen.J. (1886). 'Lorenzo Valla über Thomas von Aquino', Vierteljahrsschrift für Kultur
und Litteratur der Renaissance 1:384-96
Valla, Lorenzo (1540). Opera omnia (2 vols.), Basle (Reprinted 1962)
Valois, Noel (1903). 'Jean de Jandun et Marsile de Padoue, auteurs du Defensor Paris' in
Histoire littéraire de la France 33:528-623
Van Breda, H. L., ste Breda, H. L. Van
Van den Eynde, D., see Eynde, D. Van den
Van Eysinga, Willem J. M. iff Eysinga, Willem J. M. Van
Van Fraassen, Bas C. îef Fraassen, Bas C. Van
Vanni-Rovighi, S. (i960). 'L'intenzionalitâ della conoscenza secondo P. Aureolo', in
•Actes du premier congrès international de philosophit médiévale 673 -80
(1971). 'Gli averroisti bolognesi' in Oritnte e occiàentt ntl medioevo: Filosofia e scienze,
Convegno internazionale 9-15 Aprile 1969 (Accademia Nazionale dei Lincei, Atti dei
Convegni, vol. 13), Accademia Nazionale dei Lincei
Van Overbeke, P. M., see Overbeke, P. M. Van
Van Riet, G., see Riet, G. Van
Van Riet, S., see Riet, S. Van
Van Steenberghen, F., sef Steenberghen, F. Van
Vansteenkiste, C. (19151). 'Prodi Elementatio Theologica Translata a Guilelmo de
Moerbeke', Tijdschrift voor Philosophie 131263-302; 495-509
(1953). 'Avicenna-citaten bij S. Thomas', Tijdschrift voor Philosophie 15:457-507
(1957). 'San Tommaso d'Aquino ed Averroe', Scritti in onore di Gukeppe Furlani:
Rivkta degli studi orientait 32:585-623
(i960). 'Autori arabi e giudei nell' opera di San Tommaso', Angelkum 37:336-401
Vasoli, Cesare (1954). 'Il pensiero politico di Guglielmo d'Occam', Rivista critica di storia
dellafilosofia9:232-S3
(1968a). La dialettica e la retorica dell' umanesimo: "invenziont" e "mttodo" ntlla cultura del
XV e XVI secolo, Feltrinelli
(1968b). Studi sulla cultura del Rinascimento, Lacaita
Vâsquez, Gabriel (1620). Commentaria ac Disputationes in primam partem S. Thomae,
Antwerp
Vaux, R. de (1933). "La Première Entrée d'Averroès chez les Latins', Revue des Sciences
Philosophiques et Théologiques 22:193-245
Vennebusch, J. (1963). Ein anonymer Aristoteles Kommentar des XIII. Jahrhunderts:
Quaestiones in très libros De anima, Schöningh
(1965). 'Die Quaestiones in très libros De anima des Simon von Faversham', Archiv fir
Geschichte der Philosophie 47:20-39
(1966). 'Die Qnfsiion« metaphysice très des Siger von Brabant', Archiv fir Geschichte der
Philosophie 48:163-89
Ventura de Raulica, J. (1861). La philosophie chrétienne (3 vols.), Gaume
Verbeke, Gérard (1949). 'Le développement de la connaissance humaine d'après saint
Thomas', Rfiw philosophique de Louvain 47:437-57
(1957)· 'Themistius et le commentaire de S. Thomas au "De anima" d'Aristote' and
"Themistius et le "De unitate intellectus" de saint Thomas' in Gérard Verbeke (ed.)
Themistius: Commentaire sur le Traité de l'âme d'Aristote, Traduction de Guillaume de
Moerbeke, Édition critique et étude sur l'utilisation du commentaire dans l'oeuvre de saint
Thomas (Corpus latinum commentariorum in Aristotelem graecorum I), Publica-
tions Universitaires de Louvain and Éditions Béatrice-Nauwelaerts
(i960). 'L'unité de l'homme: saint Thomas contre Averroès', Revue philosophique de
Louvain 58:220-49
ed. (1961). Ammonius, Commentaire sur le Péri Hermenias d'Aristote (Corpus Latinum
970 Bibliography
Commentariorum in Aristotelem Graecorum II), Publications Universitaires de

Lou vain/Éditions Béatrice-Nauwelaerts
ed. (1966). Jean Philopon, Commentaire sur le De Anima d'Aristote (Corpus Latinum
Commentariorum in Aristotelem Graecorum III), Publications Universitaires de
Louvain/Éditions Béatrice-Nauwelaerts
(1968). 'Le "De anima" d'Avicenne: Une conception spiritualiste de l'homme* in
S. Van Riet (ed.) Avicenna Latinus: Liber de anima seu sextus de naturalibus IV-V,
Éditions Orientalistes and E. J. Brill
(1977). 'Introduction doctrinale' in y4i>icfnna Latinus: Liber de Philosophia Prima sive
Scientia Divina I-IV, E. Peeters
Vereecke, L. (1965). 'Individu et communauté selon Guillaume d'Ockham', Studio
Moralia 3:150-77
Verger, J. (1973). Les universités au moyen-âge, Presses universitaires françaises
Verstecgh, C. H. M. (1977). Greek Elements in Arabie Linguistic Thinking, Brill
Vescovini, G. F. (1965). Studi sulla prospettiva medioevale (Universitâ di Torino,
Pubblicazioni della Facoltà di lettere e filosofia 16.1), Giapichelli
Veuthey, L. (1951). 'Les diverse courants de la philosophie augustino-franciscaine au
moyen âge', Scholastica: Ratione Hktorico-critica Instauranda, 8:627-52
Victorinus (1863). Explanationum in Rhetorkam M. Tullii Ciccronis libri duo, ed. C. Halm
in Halm 1863
(1971). Liber de definitionibus, ed. Stang], in Hadot 1971
Vier, Peter (1951)· Evidence and its Function according to John Duns Scotus (Philosophy
Series, 7), The Franciscan Institute
Vignaux, Paul (1977). 'La problématique du nominalisme médiéval peut-elle éclairer des
problèmes philosophiques actuels?', Revue philosophique de Louvain 75:293-331
Villey, M. (1954). 'Le droit naturel chez Graden', Studia Gratiana 3 : 8 3 - 9 9
(1962). Leçons d'histoire de la philosophie du Droit, Dalloz
(1964). 'La genèse du droit subjectif chez Guillaume d'Occam', Archives de philosophie
du droit 9:97-127
Vincent Ferrer, (1909). De suppositionibus dialecticis (Oeuvres de Saint Vincent Ferrier,
H. Fages (ed.), Vol. 1, pp. 3-88), A. Picard et fils
(1977)· Tractatus de suppositionibus, ]. A. Trentman (ed.) (Grammatica speculativa, 2),
Frommann-Holzboog
Vital du Four (1927). 'Huit questions disputées sur le problème de la connaissance, ed,
F. Delorme, Archives d'histoire doctrinale et littéraire du moyen âge 2:151-337
Vitoria, Francisco de (1946). The Principles of Political and International Law in the Works of
Francisco de Vitoria, extracts, with introd. and notes by A. T. Serra, Ediciones
Cultura Hispanica
(i960). Obras, Editorial Catolica
Vives, Juan Luis (1782-90). Opera omnia, ed. Majans (8 vols.), Valencia
Voigt, G. (1893). Die Wiederbelebung des classischen Alterthums oder das erste Jahrhundert des
Humanismus (3rd edn.), G. Reimer
Wallace, William A. (1969). 'The "Calculatores" in Early Sixteenth-Century Physics',

The British Journal for the History of Science 4:221 - 3 2
(1971). 'Mechanics from Bradwardine to Galileo', Journal of the History of Ideas
32:15-28
(1972-4). Causality and Scientific Explanation (2 vols.; vol. I, Medieval and early modem
science; vol. II. Classical and contemporary science). University o f Michigan Press
Wallerand, G. (1913). Les Oeuvres de Siger de Courtrai (Les Philosophes Belges, 8), Institut
Bibliography 971
Supérieur de Philosophie de l'Université de Louvain

Wallies, Maximilian (1878). Defontibus Topkorum Ckeronis, A. Haack
Walsh, James J. (1964). Ms Buridan a Sceptic about Free Will?', Vivarium 2:50-61
(1966). 'Nominalism and Ethics: Some Remarks about Buridan's Commentary",
Journal of the History of Philosophy 4:1-13
Walter Burley (1478). In categorias, etc., Venice
(1481, 1500,1521). Expositio super decern libros Ethicorum Aristotelis, Venice
(1491, 1501). Super octo libros Physicorum, Venice
(1496). De intensione et remissioneformarum, Venice
(1497). Expositio super artem veterem Porphyrii et Aristotelis, Venice (Reprinted Minerva
1967)
(1501). In Physicam Aristotelis expositio et quaestiones, Venice (Reprinted Georg Olms
1972)
(1507). Burleus Super Artem Veterem Porphyrii et Aristotelis, n.p.
(1955). De Puritate Artis Logicae Tractatus Longior, ed. P. Boehner (Franciscan Institute
Publications, Text series, No. 9), The Franciscan Institute
Walter Chatton (1955)- Lecturat Chaton Anglici in Sententias, Prol., q. 2, in O'Callaghan
1955
Walther, Helmut G. (1976). Imperiales Königtum, Konziliarismus und Volkssouveränität:
Studien zu den Grenzen des mittelalterlichen Souveränitätsgedankens, Wilhelm Fink
Walton, D. (1976). 'Logical Form and Agency', Philosophical Studies 29:75-89
Walton, Izaak (1927). Tftf Lives of John Donne, Sir Henry Wotton, Richard Hooker, George
Herbert and Robert Sanderson, ed. with an introduction by George Saintsbury,
Oxford University Press
Walz, Angelo (1962). Saint Thomas d'Aquin, trans. Paul Novarina (Philosophes mé-
diévaux, No. 5), Publications universitaires françaises
Walzer, Michael (1977). Just and Unjust Wars, Basic Books
Watanabe, Morimichi (1963). The political Ideas of Nicholas of Cusa: with special reference
to his 'De Concordia Catholica', Droz
Webb, C. C.J. (1932). John ofSalisbury, Methuen
Wéber, Edouard-Henri (1970). La controverse de 1270 à l'Univtrsitè de Paris et son retentisse-
ment sur la pensée de S. Thomas d'Aquin (Bibliothèque thomiste, No. 40), J. Vrin
(1974). Dialogue et dissensions entre Saint Bonaventure it Saint Thomas d'Aquin α Paris
(1252-1273) (Bibliothèque thomiste, No. 41), J. Vrin
(1976). 'Les discussions de 1270 à l'Université de Paris et leur influence sur la pensée
philosophique de S. Thomas d'Aquin' in Albert Zimmermann (ed.) Dif
Auseinandersetzungen an der Parker Universität im XIII. Jahrhundert (Miscellanea
Mediaevalia, Band 10), De Gruyter
Webering, Damascene (1953). Theory of Demonstration according to William Ockham
(Philosophy Series, 10), The Franciscan Institute
Wehrli, F., ed. (1967-9). Dit Schult des Aristoteles; Ttxtt und Kommentare (10 vols.; 2nd
edn.), B. Schwabe
Weigand. R. (1967). Die Naturrechtslehre der Legkten und Dekretisten von Imerius bis
Aecursius und von Gratian bis Johannes Teutonicus (Münchener theologische Studien,
3. Kanonistische Abteilung, XXVI), Hueber
Weinberg, Julius R. (1948). Nicolaus of Autrecourt. A Study in 14th century Thought,
(1964). A Short History of Medieval Philosophy, Princeton University Press
Weisheipl, James A. (1955). Nature and Gravitation, Lyceum
(1956). Early 14th-century Physics and the Merton 'School' with special reference to
972 Bibliography
Dumbleton and Heytesbury, unpublished D. Phil, thesis, Oxford University

(1959)· 'The place ofJohn Dumbleton in the Merton School', Isk 50:439-54
(1963). 'The Concept of Matter in Fourteenth-Century Science' in E. McMullin (ed.),
The Concept of Matter in Greek and Medieval Philosophy, University of Notre Dame
Press
(1964a). 'Curriculum of the Faculty o f Arts at Oxford in the Early Fourteenth
Century', Mediaeval Studies 26:143—85
(1964b). 'Roger Swyneshed, O.S.B., Logician, Natural Philosopher, and Theologian'
in Oxford Studies Presented to Daniel Callus (Oxford Historical Soceity, new series,
16), Clarendon Press
(1965a). 'The Concept of Matter in Fourteenth Century Science' in E. McMullin (ed.)
The Concept of Matter in Greek and Medieval Philosophy, University of Notre Dame
Press
(1965b). 'Classification o f the Sciences in Medieval Thought', Mediaeval Studies
27:54-90
(1966). 'Developments in the Arts Curriculum at Oxford in the Early Fourteenth
Century', Mediaeval Studies 28:151—75
(1968). Ockham and Some Mertonians', Mediaeval Studies 30:163-213
(1969). 'Repertorium Mertonense', Mediaeval Studies 31:174-224
(1974a). Friar Thomas d'Aquino: his life, thought, and work, Blackwell
(1974b). 'Motion in a Void: Aquinas and Averroes', in St Thomas Aquinas 1274-1974:
Commemorative Studies, Vol. I, Pontifical Institute of Mediaeval Studies
(1976). 'The Relationship of Medieval Natural Philosophy to Modern Science: The
Contribution o f Thomas Aquinas to its Understanding', Manuscripta 20:181-96
(1980). 'The Life and Works o f St Albert the Great' in James A. Weisheipl (ed.)
Albertus Magnus and the Sciences: Commemorative Essays 1980, Pontifical Institute of
Mediaeval Studies
Weiss, R. (1947). The Dawn of Humanism in Italy, H. K. Lewis
Wey, J. C. (1949). 'The Sfrmo Finalis of Robert Holcot', Mediaeval Studies 11 =219-24
Wieland, G. (1974). 'Ethik und Metaphysik. Bemerkungen zur Moralphilosophie Roger
Bacons' in A. Möller (ed.) Virtus politica, Frommann-Holzboog
(forthcoming). Ethica - scientia practica. Die Anfange der philosophischen Ethik im 13.
Jahrhundert (BGPM), Aschendorff
Wieland, Wolfgang (1962). Dif aristotelische Physik, Vandenhoeck & Ruprecht
(1970). Dif arktotelische Physik (2nd edn.), Vandenhoeck und Ruprecht
Wilenius, Reijo (1963). The Social and Political Theory of Francisco Suarez (Acta
Philosophica Fennica, XV), Societas Philosophica Fennica
Wilkins, John (1668). An Essay towards a Real Character and a Philosophical Language,
London, reprinted 1974, Clearwater Press, N e w York (Linguistics 1 3 t h - 18th
Centuries series)
Wilks, Michael (1963). The Problem of Sovereignty in the Later Middle Ages: The Papal
Monarchy with Augustinus Triumphus and the Publicists (Cambridge Studies in
Medieval Life and Thought, N e w Series, 9), Cambridge University Press
(1969). 'The early Oxford W y c l i f , Studies in Church History, 5:69-98
(1972). 'Reformatio regni: Wyclif and Hus as leaders of religious protest movements'
in D. Baker (ed.), Schism, Heresy and Religious Protest (Studies in Church History,
9), pp. 109-30. Cambridge University Press
William of Alnwick (1937). Quaestiones disputatae dt esse intelligibili, ed. A. Ledoux,
Collegium s. Bonaventurae
William of Auvergne (1674). Opera omnia (2 vols.), Orleans and Paris (Reprinted
Bibliography 973
Minerva 1963)
(1976). De Trinitate, ed. Bruno Switalski, Pontifical Institute of Mediaeval Studies
William of Auxerre (1500). Summa aurea, Paris (Reprinted Minerva 1964)
William Heytesbury (1494a). Tractatus Gulielmi Hentisberi de sensu composito et divko,
Regulae eiusdem cum Sophismatibus . . . . Venice
(1494b). Regulae solvendi sophismata in William Heytesbury 1494a.
(1494c). Sophismata in William Heytesbury 1494a.
WilUam de la Mare (1927). 'Le Correctorium Corruptorii "Quare" ', ed. P. Glorieux in
Lei premières polémiques thomktes, Kain
William Ockham (1483). Quaestiones et deckiones in primum libarum Sententiarum, Urach
(1491). QuodUbeta Septem, Strasbourg (Reprinted Éditions de la Bibliothèque S.J.
Louvain 1962)
(1494-6). Opera plurima (4 vols.), Lyon (Reprinted Gregg 1962)
(1494). Dialogus in William Ockham 1494-6, vol. I
(1495-6). Super quattuor libros Sententiarum in William Ockham 1494-6, vols. Ill and
IV
(1496). Expositio aurea et admodum utilk super artem veterem édita per venerabilem
inceptorem fratrem Guilielmum de Occham cum quaestionibus Alberti parvi de Saxonia,
Bologna (Reprinted Gregg 1964)
(?) (1914). Df Electione Caroli Quarti, found only in Conrad of Megenberg's Tractatus
contra Wilhelmum Occam in R. Scholz (ed.) Unbekannte kirchenpolitische Streitschriften
aus der Zeit Ludwigs des Bayern (1327—1354), 2:346-63, Loescher
(1914-24). Df Imperatorum et Pontificum Potestate, ed. R. Scholz (1914), Unbekannte
kirchenpolitische Streitschriften aus der Zeit Ludwigs des Bayern (1327-1354), 2:453-80,
Loescher (W. Regenberg); continuation ed. W. Mulder (1923-4), Archivum
Franciscanum Historicum 16:469-92,17:72-97
(1930). The De Sacramento Altaris of William of Ockham, ed. and tr. T. B. Birch, The
Lutheran Literary Board
(1944). Tractatus de successivis, ed. P. Boehner (Franciscan Institute Publications, 1),
The Franciscan Institute
(1945). The Tractatus de praedestinatione et de praescientia dei et defuturk contingentibus of
William Ockham, ed. Philotheus Boehner (Franciscan Institute Publications, 2), The
(1951). Summa logicae (in part) ed. P. Boehner (2 vols.), The Franciscan Institute
(1952). Breviloquium de principatu tyrannko super divina et humana, specialiter autem super
Imperium et subiectos imperio, a quibusdam vocatk summis pontificibus usurpato, ed. R.
Scholz in Wilhelm von Ockham ab politischer Denker und sein Breviloquium de prin-
cipatu tyrannico, Anton Hiersemann [reprint of 1944 edn.]
(1956a). Contra Bentdictum, ed. H. S. Offler, Guillelmi de Ockham Opera Politica, 3,
(1956b). Contra Ioannem, ed. H. S. Offler, Guillelmi de Ockham Opera Politica, 3,
(1957a). Ockham: Philosophical Writings, ed. and tr. P. Boehner, Thomas Nelson
(1957b). Expositio super viii libros Physicorum, Prologus in William Ockham 1957a
(1963-74). Opus nonaginta dierum, chapters 1-6 ed. H. S. Offler (1974), Guillelmi de
Ockham Opera Politica, 1 (rev. edn.), Manchester University Press; chapters 7-124
ed. R. F. Bennett and H. S. OfHer (1963), Guillelmi de Ockham Opera Politica, 2,
(1964). 'The Tractatus logicae minor of Ockham', ed. Eligius M. Buytaert, Franciscan
Studies 24:34-100
974 Bibliography
(1965-66). 'The Eltmentarium logicae of Ockham', ed. Eligius M. Buytaert, Franciscan

Studies 25:151 -276 and 26:66-173
(1967-). Opera philosophica et theologica, ed. Juvenal Lalor, Stephen Brown, Gedeon
Gâl, Angelo Gambatese, and Michael Meilach, The Franciscan Institute
(1967). Opera theologica [OT] I: Saiptum in librum primum Sententiarum, Ordinatio,
Prologus et Distinctio prima, ed. G. Gâl and S. Brown in William Ockham 1967-
(1969). William Ockham: Predestination, God's Foreknowledge, and Future Contingents,
translated with introduction and notes by Marilyn McCord Adams and Norman
Kretzmann (Century Philosophy Sourcebooks), Appleton-Century-Crofts/William
Hackett
(1970). OT II: Scriptum in librum primum Sententiarum, Ordinatio; Dktinctionts II—III,
ed. S. Brown and G. Gâl in William Ockham 1967-
(19743). Opera philosophica [OP] I: Summa logicae, ed. P. Boehner, G. Gâl, and
S. Brown, in William Ockham 1967-
(1974b). Octo quaestiones de potestate papat, ed. H. S. Offler, Guillelmi dt Ockham Opera
politica, 1 (rev. edn.), Manchester University Press
(1974c). Consultatio de causa matrimoniali, ed. H. S. Offler, Guillelmi de Ockham Opera
politica, 1 (rev. edn.), Manchester University Press
(i974d). An princeps pro suo succursu, scilicet guerrae, possit recipere bona ecclesiarum etiam
invito papa, ed. H. S. Offler, Guillelmi de Ockham Opera Politica, 1 (rev. edn.),
(1974e). Ockham's Theory of Terms: Part 1 of the Summa logicae, translated with
introduction and essays by Michael J. Loux, University of Notre Dame Press
(1977a). OT III: Scriptum in librum primum Sententiarum, Ordinatio; Distinctiones
IV-XVII, ed. G. I. Etzkorn in William Ockham 1967-
(1977b). Dialogus II, 3, vi, ed. H. S. Offler in Offler 1977
(1978). OP II: Expositionis in libros artis logicae prooemium et Expositio in librum Porphyrii
de praedicabilibus, ed. E. A. Moody; Expositio in librum Praedicamentorum Aristotelis,
ed. G. Gâl; Expositio in librum Perihermenias Aristotelis, ed. A. Gambatese and
S. Brown; Tractatus de praedestinatione et de praescientia Dei respectufiturorumcon-
tingentium, ed. P. Boehner and S. Brown, in William Ockham 1967—
(1979a). OP III: Expositio super libros Elenchorum, ed. F. del Punta, in William Ockham
1967-
(1979b). OTIV: Scriptum in librum primum Sententiarum, Ordinatio; Distinctiones
XIX-XLVIII, ed. G. I. Etzkorn and F. E. Kelley, in William Ockham 1967-
(1980). ΟΤΓΧ: QuodUbeta Septem, ed. Joseph C. Wey, in William Ockham 1967-
(19803). Ockham's Theory of Propositions: Part 2 of the Summa logicae, tr. A.J. Freddoso
and Henry Schuurman, University of Notre Dame Press
William of Sherwood (1937), Introductiones in logicam, ed. M. Grabmann
(Sitzungsberichte der Bayerischen Akademie der Wissenschaften, Phil.-hist.
Abteilung 1937,10), Verlag der Bayerischen Akademie der Wissenschaften
(1941). Syncategoremata, ed. J. R. O'Donnell, Mediaeval Studies 3146-93
(1966). William of Sherwood's Introduction to Logic, tr. with introduction and notes by
Norman Kretzmann, University of Minnesota Press
(1968). William of Sherwood's Treatke on Syncategorematic Words, tr. with introduction
and notes by Norman Kretzmann, University of Minnesota Press
Williams, G. H. (1951). TTie Norman Anonymous of 1100 A.D. (Harvard Theological
Studies, No. 18), Harvard University Press
Wilpert, Paul (1931). Das Problem der Wahrheitsskherung bei Thomas von Aquin: Ein
Beitrag zur Geschichte des Evidenzproblems BGPM XXX, 3, Aschendorff
Bibliography 975
Wilson, Curtis (1952). 'Pomponazzi's Criticism of Calculator', Isis 44:355-63

(1956). William Heytesbury: Medieval Logk and the Rise of Mathematical Physics
(Publications in Medieval Science, 3), University of Wisconsin Press
(1972). 'Heytesbury, William', in C. C. Gillispie (ed.) Dictionary of Scientific Biography,
Winterbottom, M. (1967). 'Fifteenth-Century Manuscripts of Quintilian', Classical
Quarterly 17:339-69
Wippel, J. F. (1964). 'Godfrey of Fontaines and the Real Distinction Between Essence
and Existence', Traditio 20:385-410
(1967). 'Godfrey of Fontaines', New Catholic Encyclopedia 6:577-8
(1971). 'Godfrey of Fontaines: The Date of Quodlibet 15', Franciscan Studies
31:300-69
(1973). 'Commentary on Boethius' Df triwifafe' Tftf Thomist 37:133-54
(1974a). 'The Dating ofJames of Viterbo's Quodlibet I and Godfrey of Fontaines'
Quodlibet VIII', Augustiniana 24:348-86
(1974b). 'Godfrey of Fontaines and Henry of Ghent's Theory of Intentional
Distinction Between Essence and Existence' in T. W. Köhler (cd.) Sapientiae
Procerum Amort. Mélanges Médiévistes offerts à Dom Jean-Pitrre Müller O.S.B (Studia
Anselmiana, 63), Herder
(1977). 'The Condemnations of 1270 and 1277 at Paris', The Journal of Medieval and
Renaissance Studies 7:169-201
(1978). 'Metaphysics and Separatio According to Thomas Aquinas', The Review of
Metaphysics 31:431 -70
(1979)· 'Aquinas's Route to the Real Distinction: A Note on Df «iff ff «senfia*, The
Thomist 43:279-95
(1980). The Metaphysical Thought of Godfrey of Fontaines: A Study in Late Thirteenth-
Century Philosophy. Catholic University of America Press
(forthcoming). 'The Relationship Between Essence and Existence in Late Thirteenth
Century Thought: Giles of Rome, Henry of Ghent, Godfrey of Fontaines, and
James of Viterbo', in Ancient and Medieval Philosophies of Existence, Fordham
University
Wolf, F.J. (1966). Dif Intellektslehre des Simon von Faversham nach seinem De Anima
Kommentaren, Inaugural Dissertation, University of Bonn
Wolfson, Harry Austryn (1935). 'The Internal Senses in Latin, Arabic, and Hebrew
Philosophic Texts', Harvard Theological Review 28:69-133
(1973). Studies in the Hktory of Philosophy and Religion, vol. 1, ed. I. Twersky and
G. H. Williams, Harvard University Press
Wolin, Sheldon S. (i960). Politks and Vision: Continuity and Innovation in Western Political
Thought, Little, Brown
Wolter, A. B. (1946). The Transcendentals and their Function in the Metaphysics of Duns
Scotus, The Catholic University of America
ed. and tr. (1962). Duns Scotus: Philosophical Writings, Nelson
(1965a). 'The Ockhamist Critique' in E. McMullin (ed.) Tftf Concept of Matter in
Greek and Medieval Philosophy, University of Notre Dame Press
(1965b). 'The Formal Distinction', inj. Κ. Ryan and B. M. Bonansea (eds.) John Duns
Scotus, 1265-1965, The Catholic University of America
(1967a). 'Bacon, Roger' in Paul Edwards (ed.) Tftf Encyclopedia of Philosophy,
MacmiUan
(1967b). 'Duns Scotus, John', in P. Edwards (ed.) Tftf Encyclopedia of Philosophy,
MacmiUan
976 Bibliography
(1968). 'Is Existence for Scotus a Perfection, Predicate, or What?' in De doctrina Ioannk
Duns Scoti. Acta Congressus Scotistki Internationalis Oxonii et Edimburgi 11—17 sept.
1966 celebrati. V. II: Problemata Philosophica, Cura Commissionis Scotisticae
Woodward, W. H. (1921). Vittorino da Feltre and Other Humanist Educators, Cambridge
University Press
Woolf, C. N. S. (1913). Bartolus of Sassoferrato, Cambridge University Press
Workman, Herbert B. (1926). John Wyclif: A Study of the English Medieval Church
(2 vols.), Clarendon Press
(1966). John Wyclif: A Study of the English Medieval Church, Arcon Books (ist edn.
1926)
Wright, N. A. R. (1976). 'The Tree of Battles of Honoré Bouvet and the Laws of War',
in C. T. Allmand (ed.) War, Literature and Politics in the Late Middle Ages, Liverpool
University Press
Wrobel, Ioh. (1887). Eberhardi Bethuniensis Graecismus, G. Koebner
Wüsdörfer, Joseph (1917). Erkennen und Wissen nach Gregor von Rimini. Ein Beitrag zur
Geschichte der Erkenntnistheorie des Nominalismus (BGPM, XX, 1), Aschendorff
Wulf, Maurice de (1904): Un théologien-philosophe du XIII' siècle. Étude sur la vie, les
oeuvres et l'influence de Godefroid de Fontaines, Hayez
(1934-7). Histoire de la philosophie médiévale (6th edn.), I (1934), II (1936), III (1937)
Institut Supérieur de Philosophie/Vrin
Wyclif, John. Sef John Wyclif
Yates, Frances (1966). The Art of Memory, Routledge and Kegan Paul
Yokoyama, Tetsuo (1969). 'Simon of Faversham's Sophisma: Universale est Inttntio',
Mediaeval Studies 31 : ι—14
Ypma, Ε. (1974). 'Recherches sur la carrière scolaire et la bibliothèque de Jacques de
Viterbe t 1308', Augustiniana 24:247-82
(J975)· 'Recherches sur la productivité littéraire de Jacques de Viterbe jusqu'à 1300',
Augustiniana 25:223-82
Zabarella, Jacobus (1608). Opera logica, Frankfurt

Zabughin, V. (1909-12). Giulio Pomponio Leto: saggio critico, La vita letteraria
Zavalloni, R. (1951). Richard de Mediavilla et la controverse sur la pluralité des formes.
Éditions de l'Institut Supérieur de Philosophie de Louvain
Zigliara, T. M. (1876) Summa philosophiae in usum scholarum, Beauchesne
Zimmermann, A. (1965). Ontologie oder Metaphysik? Die Diskussion über den Gegenstand
der Metaphysik im 13. und 14. Jahrhundert, Brill
(1967-8). 'Dante hatte doch Recht: Neue Ergebnisse der Forschung über Siger de
Brabant", Philosophisches Jahrbuch 75:206-17
(i973)· 'Thomas von Aquin und Siger von Brabant im Licht neuer Quellentexte' in
Alf önnerfors, Johannes Rathofer and Fritz Wagner (eds.) Liieraiwr und Sprache im
europäischen Mittelalter: Festschrift für Karl Langosch zum 70. Geburtstag, Wissen-
schaftliche Buchgesellschaft
ed. (1976). Die Auseinandersetzungen an der Pariser Universität im XIII. Jahrhundert
(Miscellanea Mediaevalia, 10), De Gruyter
Zippel, Giuseppe (1957). 'Note sulle redazioni della "Dialectica" di Lorenzo Valla',
Archivio storico per le province parmerisi 9:3 ΟΙ—15
Zoubov, V. P. (1959). 'Walter Catton, Gérard d'Odon et Nicolas Bonet', Pftysis
1:261-78
(i960). 'Le traité "De continuo" de Bradwardin', Istoriko-Matematicheskie Issledovaniya
Bibliography 977
13:385-440
(1961). 'Jean Buridan et les concepts du point au quatorzième siècle', Mediaeval and
Renaissance Studies 5:49-95
Zumkeller, A. (1941). Hugolin von Orvieto und seine theologische Erkenntnklthre,
Cassiciacum. Eine Sammlung wissenschaftlicher Forschungen über den hl.
Augustinus und den Augustinerorden, Band IX, 2. Reihe, 3. Band, Würzburg
(1951). 'De doctrina sociali scholae Augustinianae Aevi Medii', Analtcta Augustiniana
22:57-84

The Cambridge History of Later Medieval Philosophy

Uploaded by

Copyright:

Available Formats

The Cambridge History of Later Medieval Philosophy

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Cambridge History of Later Medieval Philosophy

Uploaded by

Copyright:

Available Formats

The Cambridge History of

Later Medieval Philosophy

© Cambridge University Press 1982

First published 1982

Printed in Great Britain at the University Press, Cambridge

Library of Congress catalogue card number: 81-10086

British Library Cataloguing in Publication Data

ISBN 0 52122605 8 hardback

I Medieval philosophical literature 9

1 ANTHONY KENNY, JAN PINBORG II

4 A n c i e n t scholastic l o g i c as t h e s o u r c e o f m e d i e v a l scholastic l o g i c 101

IV L o g i c in the h i g h m i d d l e ages: s e m a n t i c t h e o r y 159

V L o g i c in the high middle ages: propositions and modalities 271

VI Metaphysics and e p i s t e m o l o g y 383

VII Natural philosophy 519

27 The Oxford calculators 54°

28 Infinity and continuity 564

VIII Philosophy o f m i n d and action 593

30 Sense, intellect, and imagination in Albert, Thomas, and Siger 602

31 Criticisms of Aristotelian psychology and the Augustinian-

37 Natural morality and natural law 705

39 Rights, natural rights, and the philosophy of law 738

40 The state of nature and the origin of the state 757

41 The just war 771

XI The defeat, neglect, and revival of scholasticism 785

PROFESSOR EDITH DUDLEY SYLLA

The idea of this History originated in the Cambridge University Press,

Long delays are evidently inevitable in the production of a work of

attacks. After Christianity and Aristotelianism, the most important in-

remaining chapters in each of these Parts deal with specific ethical or

Schools and universities1

2. See also Southern 1970, pp. 61-85.

the foundation of the university in 1392.3 Moreover, as early as the

disciplines (the Quadrivium). Students in the faculties of Arts were normally

8. Van Steenberghen 1979.

Teaching and books

11. Fink-Errera i960. Cf. Thorndike 1944, pp. 112—18.

Textbooks and university courses

I have just been emphasising the oral aspect of medieval philosophical

tive course-book(s), the littera. Research and teaching was obliged to

Greifswald, statutes of 1456:

De anima + parva naturalia j year

14. See, e.g., Isaac 1953, pp. 61-85; Weisheipl 1964a.

Ethica Nie, Oeconomica, Polilica, Theoria planetarum, Perspectiva,

This standard curriculum was supplemented by readings of special texts

Forms of teaching: Lectures16

17. Peters 1968, pp. 10-17.

Disputations of the theologians

In the thirteenth century the principal duties of a Master in Theology were

Disputations in the Arts faculties

24. Weisheipl 1964a; Gilbert 1976.

The origins of disputation

27. Roos 1963; Pinborg 1975b.

agreed about the presuppositions and antecedents of the institution. A

The later development of disputations

The purpose of disputations

38. A typical example of such a textbook is Stwek 1948.

The literary forms: literal commentaries

interpreted Aristotle. 45 It is also interesting to follow the fourteenth-

49. Murdoch 1975, p. 278.