EC - Bonazzi 2020 - The Sophists - GRE - IgrKsr

Download as pdf or txt
Download as pdf or txt
You are on page 1of 162

New Surveys in the Classics no 45

Greece & Rome


New Surveys in the Classics no 45

The Sophists The Sophists

BY Mauro Bonazzi
From Socrates and Plato onwards, the Sophists were often targeted by the authoritative
philosophical tradition as being mere charlatans and poor teachers. This book, translated
and significantly updated from its most recent Italian version (2nd edition, 2013), challenges

The Sophists
these criticisms by offering an overall interpretation of their thought, and by assessing the
specific contributions of thinkers like Protagoras, Gorgias and Antiphon. A new vision of the
Sophists emerges: they are protagonists and agents of fundamental change in the history of
Ancient Philosophy, who questioned the grounds of morality and politics, as well as the nature
of knowledge and language. By shifting the focus from the cosmos to man, the Sophists
inaugurate an alternative form of philosophy, whose importance is only now becoming clear.

Cambridge Journals Online


For further information about this journal Published for the Classical Association
please go to the journal web site at:
journals.cambridge.org/nsy

https://doi.org/10.1017/S0533245119000129 Published online by Cambridge University Press


New surveys in the classics
Regular subscribers to the journal Greece & Rome receive a volume in
the New Surveys in the Classics series as part of their subscription. The
following volumes are also available to purchase as books.
Volume 44 Sophocles (ISBN 9781108706094)
Volume 43 Early Greek Hexameter Poetry (ISBN 9781316608883)
Volume 42 Horace, (ISBN 9781107444447)
Volume 41 Homer, Second Edition (ISBN 9781107670167) Greece & Rome is a journal which delivers scholarly research to a wider
Volume 40 Greek Art, Second Edition (ISBN 9781107601505) audience. It showcases original and informative articles on ancient history,
Volume 39 Roman Landscape: Culture and Identity (ISBN 9781107400245) literature, art, archaeology, religion, philosophy, and reception of the
Volume 38 Epigram (ISBN 9780521145701)
ancient world. Although its content reflects current research and its
Volume 37 Comedy (ISBN 9780521706094)
Volume 36 Roman Oratory (ISBN 0521687225) contributors include leading figures in the field, undergraduates and general
Volume 35 The Second Sophistic (ISBN 0198568819) readers who wish to be kept informed of current thinking will also find it
Volume 34 Roman Art (ISBN 9780198520818) engaging and accessible, as well as professional scholars in Classics and in
Volume 33 Reception Studies (ISBN 0198528655)
Volume 32 The Invention of Prose (ISBN 0198525234)
other disciplines. With the wider audience in mind all Greek and Latin
Volume 31 Greek Historians (ISBN 019922501X) quotations are translated.
Volume 30 Roman Religion (ISBN 0199224331)
Volume 29 Greek Science (ISBN 0199223955) A subscription to Greece & Rome includes an annual supplement of New
Volume 28 Virgil (ISBN 0199223424) Surveys in the Classics.
Volume 27 Latin Historians (ISBN 0199222932)
Volume 26 Homer (ISBN 0199222096)
Volume 25 Greek Thought (ISBN 0199220743)
Volume 24 Greek Religion (ISBN 0199220735)

To place an order from within the United States, Canada and Mexico For information on subscribing or submitting to the journal, please visit
please contact Cambridge Customer Services toll free via +1 800 872 7423 journals.cambridge.org/gar
or by email via information@cup.org/. For orders outside of North America
please call + 44 (0)1223 325566 or email via ukcustserve@cambridge.org/.

THE CLASSICAL ASSOCIATION


The Classical Association has a worldwide membership and is open to all
who value the study of the languages, literature, and civilizations of ancient
Greece and Rome. It creates opportunities for friendly exchange and
co-operation among classicists, encourages scholarship through its journals
and other publications, and supports classics in schools and universities.
Every year it holds an annual conference, and it sponsors branches all over
the country which put on programmes of lectures and other activities.
The Classical Association publishes three journals, Classical Quarterly,
Classical Review and Greece & Rome. Members may subscribe at substantially
reduced cost. Non-members and institutions can subscribe via Cambridge
University Press (email journals@cambridge.org).
For more information and a membership application form please contact
the Secretary, The Classical Association, Park House, 15–23 Greenhill
Crescent, Watford WD18 8PH (e-mail: office@classicalassociation.org).
The Secretary can also give information about the reduced journal
subscription rates, and about the Association’s other occasional publications.
Cover illustration: Sculpture “Light of the Moon” at terrace near the Beelden
aan Zee Museum. Scheveningen, Hague, Netherlands © Igor Mitoraj.

https://doi.org/10.1017/S0533245119000129 Published online by Cambridge University Press


Greece & Rome
NEW SURVEYS IN THE CLASSICS No. 45

THE SOPHISTS
BY
MAURO BONAZZI

Published for the Classical Association


CAMBRIDGE UNIVERSITY PRESS
2020

https://doi.org/10.1017/S0533245119000129 Published online by Cambridge University Press


CAMBRIDGE UNIVERSITY PRESS
The Edinburgh Building, Cambridge CB2 8BS, United Kingdom
32 Avenue of the Americas, New York, NY 10013-2473, USA
477 Williamstown Road, Port Melbourne, VIC 3207, Australia
Ruiz de Alarcón 13, 28014 Madrid, Spain
Dock house, The Waterfront, Cape Town 8001, South Africa

www.cambridge.org
Information on this title: www.cambridge.org/9781108706216

© The Classical Association 2020


This publication is copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press

Printed in the United Kingdom by Bell and Bain, Glasgow, UK

A catalogue record for this publication is available from the British Library

ISBN 9781108706216

Cambridge University Press has no responsibility for the persistence or accuracy of URLs for
external or third-party websites referred to in this publication, and does not guarantee that any
content on such websites is, or will remain, accurate or appropriate.

https://doi.org/10.1017/S0533245119000129 Published online by Cambridge University Press


To FERNANDA DECLEVA CAIZZI
IL MIGLIOR FABBRO

https://doi.org/10.1017/S0533245119000129 Published online by Cambridge University Press


CONTENTS

Foreword v
I The Sophists: History of a Name and Prejudice 1
II Being and Truth, Humanity and Reality 11
III A World of Words: The Sophists at the Crossroads
Between Grammar, Rhetoric, Poetry, and Philosophy 43
IV Justice and Law 65
V Teaching Virtue: The Sophists on Happiness and Success 96
VI The Gods and Religion 111
Appendix 1: The Protagonists 125
Appendix 2: The Sophists and Specialist Forms of Knowledge
(Technai) 138

Bibliography 141
Index 153

https://doi.org/10.1017/S0533245119000129 Published online by Cambridge University Press


FOREWORD: A NOTE ON THE QUOTATION OF
FRAGMENTS AND TESTIMONIES PERTAINING TO THE
SOPHISTS

Probably as a consequence of the negative judgement which often


weighed against them – yet not unlike the case for many other ancient
writers – almost no complete texts by the sophists survive: we must rely
on quotes and testimonies. Although recently discovered papyri have
broadened our knowledge, to this day the essential edition of the
works of the sophists – and Presocratic philosophers more generally –
is Hermann Diels’s Die Fragmente der Vorsokratiker, later revised and
corrected by Walter Kranz (Diels and Kranz 1951–2, henceforth
D.-K.). This collection is in three volumes: the first two contain all
the sources, arranged in ninety chapters (each chapter usually corre-
sponding to an author), while the third volume features a word index
and a selective index of names and passages. The section on the soph-
ists coincides with Chapters 79–90 of the second volume. For each
author, the chapters are divided into ‘testimonies’ and ‘fragments’,
which are respectively marked with the letters A and B (this subdivision
is not adopted in all cases, but only when the number of passages allows
it – this is not the case, for instance, with Xeniades, Lycophron, and the
Dissoi logoi); in some cases we also find a third section, C, containing
‘imitations’. Thus 80B3 will stand for fragment 3 of Protagoras,
while 87A6 will stand for testimony 6 on Antiphon, and so on (the
same criteria generally apply to all other Presocratic philosophers as
well).
Though yet unsurpassed, Die Vorsokratiker can now be fruitfully
complemented with a new impressive collection of testimonies and
fragments: André Laks and Glenn W. Most, Early Greek Philosophy,
published in 2016 in the Loeb Classical Library (and in French by
Fayard with the title Les débuts de la philosophie. Des premiers penseurs
grecs à Socrate, henceforth L.-M.). The sophists are presented in the
eighth and ninth volumes, sections 31–42. Each chapter normally com-
prises three sections, with each entry being identified by a number pre-
ceded by a letter: P indicates information on the life, character, and
sayings of a given philosopher; D the doctrines; R the reception of
their thought and writings. Unless not otherwise noted, all translations
of the sophists are based on Laks and Most’s translations. Other useful

https://doi.org/10.1017/S0533245119000014 Published online by Cambridge University Press


vi FOREWORD

translations are Freeman 1948, Kent Sprague 1972, and Dillon and
Gergel 2003. With regard to Plato and Aristotle, our two most import-
ant testimonies regarding the sophists, the translations are taken from
Cooper 1997 and Barnes 1984, respectively. For all the other authors,
the translations are from the Loeb Classical Texts (with minor
changes).
This book, a revised and updated version of my previous book, I
sofisti (Rome, 2010), would never have seen the light without the sup-
port and the competence of the editor of the New Surveys series, Phillip
Horky. My warmest thanks are due to him, and also to Sergio Knipe for
his help with the translation.

https://doi.org/10.1017/S0533245119000014 Published online by Cambridge University Press


I THE SOPHISTS: HISTORY OF A NAME AND PREJUDICE

One day around the year 430 BC, before dawn, a young and promising
Athenian, Hippocrates, son of Apollodorus, hastens to Socrates’ house
to rouse him from his sleep. The reason for this bizarre behaviour soon
becomes clear: Protagoras is in Athens! With a knowing smile, Socrates
answers that he is already aware of this, as though he could not really
understand the reason for all this excitement. But Hippocrates shows
no hesitation: Protagoras, the great sophist, he who ‘makes people
wise’, is in Athens – an opportunity not to be missed! The two of
them must leave immediately, and Socrates must help Hippocrates
gain access to Callias’ house, where the sophist is staying. However,
Socrates insists on enquiring about the reason for Hippocrates’ excite-
ment: does he wish to become a sophist?
‘And if somebody asks you what do you expect to become in going to Protagoras?’
He blushed in response – there was just enough daylight now to show him up – and
said, ‘If this is all like the previous cases, then, obviously, to become a sophist.’
‘What? You? Wouldn’t you be ashamed to present yourself to the Greek world as a
sophist?’
‘Yes, I would, Socrates, to be perfectly honest.’ (Plato, Protagoras, 310a–312a)

This famous page from Plato – the prologue of the Protagoras – prob-
ably constitutes the most compelling example of the ambiguous fame
attached to the name of the sophists. That fame has not changed
much over the centuries: the sophists have always been the object of
violent polemics and passionate vindication. After centuries of criti-
cism, in the modern age the sophists received the support of some
most distinguished philosophers: first from Hegel, who regarded
them as the masters of Greece, then from some English liberals (in par-
ticular George Grote), and finally – and even more staunchly – from
Nietzsche, who saw them as the most genuine representatives of the
Greek spirit.1 Later, Popper was to go so far as to speak of a ‘great gen-
eration’, while other scholars have even suggested that ours is the age of
the Third Sophistic.2 At the same time, however, ‘sophist’ and

1
On the importance and limits of the interpretations put forward by Hegel and English liberals,
see Kerferd 1981a: 4–14; on Nietzsche, see Consigny 1994.
2
Popper 1971: 162. One author who describes the contemporary age as a ‘Third Sophistic’ is
Vitanza 1997; see also Fowler 2014. See too Rorty 1979: 157. The expression ‘Second Sophistic’
is used to describe the revival in the Imperial age of certain elements introduced by the sophists in

https://doi.org/10.1017/S0533245119000026 Published online by Cambridge University Press


2 I THE SOPHISTS

‘sophistic’ are still used to describe ‘a verbal philosophy that is neither


sound nor serious’, to quote Lalande’s Vocabulaire technique et critique
de philosophie – almost a paraphrase of the scathing verdict of
Aristotle, who had dismissed sophistry as the wisdom of appearances,
and hence an appearance of wisdom (Sophistical Refutations, 171b34).
In other words, in antiquity, just as in the modern age, the sophists
elicited sympathy at times, but more often deep hostility; and neither
situation has helped to adequately reconstruct their thought. In order
to make an appropriate assessment, it is necessary first to clarify the
topic to be discussed: the primary aim of the present book is to provide
a reliable reconstruction of the sophists’ thought – as far as the sources
allow it – within their historical and philosophical context. It will then
be up to the reader to assess their importance.

Who were the sophists?

In order to adequately analyse the sophists’ thought, it is necessary to


address a highly challenging preliminary problem: to identify which
thinkers can legitimately be called ‘sophists’, and to clarify what distin-
guishes them from other intellectuals of their age. One often speaks of
the ‘sophistic’ as though its definition was taken for granted, as though
it were clear to everyone in antiquity. Yet, when taken literally, even the
term ‘sophistic’ itself is improper and misleading, insofar as in
fifth-century Greece there was never a homogeneous tradition of
thought or institutionalized school to which the label ‘sophistic’
could be applied.3 In the ancient sources, the term from which our
word ‘sophist’ derives, sophistês, expresses a much more fluid situation,
encompassing a more or less extensive range of meanings: sophistês is
used for a poet like Homer, a politician like Solon, and a mythical figure
like Prometheus. According to Plato, a sophist is ‘troublesome and
hard to grasp’ and there is a risk that each person will assign a different
meaning to the word (Sophist, 218b–d). Such a broad use of the word

the fifth century BC. The focus in this case was on rhetorical and literary themes, while the more
strictly philosophical aspect was largely overlooked. For this reason, I will not be discussing the
phenomenon here (for an interesting attempt to link the fifth-century sophistic and the Second
Sophistic, see Cassin 1995).
3
It may be worth noting that, in the surviving testimonies, the only sophist to describe himself as
such is Protagoras in the Platonic dialogue that bears his name (Protagoras, 317b = 80A5 D.-K. =
31P13a L.-M.).

https://doi.org/10.1017/S0533245119000026 Published online by Cambridge University Press


I THE SOPHISTS 3

risks making it useless, as it would come to generically describe anyone


who has some relationship with knowledge or wisdom (sophia in
Greek).
Fortunately, however, a more precise definition can be found. While
the ancients were quite aware that the ‘sophistic’ was not a homoge-
neous school of thought, they acknowledged the sophists as a group
of thinkers who, over the course of the fifth century, had promoted a
radical renewal of the Greek cultural tradition. This makes it possible
to narrow down the scope of the term: while doubts may be raised
about this or that figure, it is possible to identify a core of thinkers
who may be regarded as part of this ‘movement’ (to quote George
Kerferd, the author of a key study on the sophists).4 In this respect,
the most authoritative contribution comes from Hermann Diels, who
devoted the final section of his extensive edition of the Presocratics to
‘ancient sophistic’, by including testimonies and fragments pertaining
to nine authors and two anonymous treatises: Protagoras, Xeniades,
Gorgias, Lycophron, Prodicus, Hippias, Thrasymachus, Antiphon,
and Critias, to which we should add the so-called Anonymus
Iamblichi or the ‘Anonymous of Iamblichus’, the author of a text
preserved by the Neoplatonist philosopher Iamblichus (third–fourth
century CE), and the anonymous author of the Dissoi logoi, a set of dia-
lectical exercises concerning standard topics for argumentation. Diels
hesitated at first about including the sophists in his work; nonetheless,
the sources he selected – with some possible integrations (for example,
Alcidamas and figures such as Polus, Euthydemus, and Dionysodorus,
who are only known to us through the little information that Plato pro-
vides) – make up the ranks of the ‘sophists’, for us as much as for the
ancients.5
This point can be agreed upon. But what criteria allow us to better
define the sophists? One hint comes from the very meaning of the
term sophistês and from the history of its use: by briefly retracing this

4
Kerferd 1981a.
5
Naturally, this list, while perfectly reasonable, is not decisive. To give just one example of how
difficult it is to precisely identify the sophists: in his writings, Aristotle applies this label to four
authors alone, namely Lycophron, Polyeidos, Bryson, and Aristippus. Only the first of these is
reckoned among the sophists today (Polyeidos appears to have been a poet and literary critic,
while Bryson and Aristippus are possibly to be associated with the milieu of the Socratic schools).
All four are called sophists in relation to language and ethico-political issues, which – as we shall
see – constituted the privileged object of reflection for the sophists.

https://doi.org/10.1017/S0533245119000026 Published online by Cambridge University Press


4 I THE SOPHISTS

history, we can identify some key features.6 Other clues may then be
inferred through an engagement with the ancient sources. The word
sophistês is based on the root *soph-, which we also find in sophos
(knowledgeable, wise) and sophia (knowledge, wisdom), two terms
commonly used from early Greece to express the notion of skill in
some field or craft. Starting from this general meaning, sophos gradually
came to describe individuals who were knowledgeable and accom-
plished in intellectual pursuits. These ‘wise men’ (sophoi in the plural),
usually poets or soothsayers, possessed a kind of knowledge that was
unattainable by other men and which could have beneficial effects.
From the fifth century BC onwards, sophistês entered into use as a syno-
nym of sophos: to be more precise, sophistês is formed from the verb
sophizesthai, which is to say ‘to exercise sophia’. A ‘sophist’, therefore,
is someone who acts as a sophos, someone who possesses intellectual
knowledge either in a general sense or in a specific discipline.
Sophistês, originally a synonym of sophos, thus came to describe some-
thing more narrow, something more suited to the needs of an increas-
ingly sophisticated society: the exercise of knowledge in practice is
teaching, as a profession; hence, a sophist is a teacher, and educator.
This is an initial defining feature, from which others follow.
The professionalizing of the sophist’s role helps to explain another
crucial characteristic that further distinguishes his social position,
namely his salary. Insofar as the sophist imparts some teachings, he
expects some payment in return – a request that is only natural for
us, but which was very unusual for the ancient Greeks, who regarded
the idea of teaching wisdom or virtue to anyone for a fee as something
quite shocking.7 Secondly, what also distinguishes the activity of the
sophist – albeit to a lesser degree than the demand for pay – is his itin-
erant character. The sophists are teachers who share their knowledge in
exchange for a fee in the many cities they visit. This is hardly surprising,
given that almost all sophists were born far away from the major cultural

6
On the meaning of sophistês and its history, see the following analyses: Untersteiner 1949–62:
i.xvi–xxiii; Guthrie 1971: 27–35; and Kerferd 1981a: 24–41. An alternative reconstruction is put
forward by Edmunds 2006, according to whom sophistês only entered into use as a technical term
in the fourth century BC, whereas in the fifth it described a gamut of intellectual figures, including
soothsayers, dithyrambic poets, scientists, and orators. This study confirms the importance played
by Plato and Aristotle in the definition of the sophists’ identity; still, it does not rule out that by the
fifth century the term ‘sophist’ had already come to be associated with the practice of teaching,
which is perhaps the most distinguishing feature of the sophists’ activity; on this, see also Tell
2011.
7
Blank 1985.

https://doi.org/10.1017/S0533245119000026 Published online by Cambridge University Press


I THE SOPHISTS 5

and political centres of their day, such as Athens: travelling for them
was practically a requirement.8 The itinerant character of the sophists’
teaching may seem like a trivial fact, but actually has some important
implications – as we shall see – that at least partly account for the degree
of hostility they incurred.
These three criteria (teaching, the charging of a fee, and travel) allow
us to narrow down the field of our enquiry, insofar as they help us dis-
tinguish a ‘sophist’ such as Protagoras from a ‘sophist’ such as
Prometheus. Protagoras was a teacher who visited many cities of the
Greek world, exercising his profession for a fee; by contrast, when
Aeschylus calls Prometheus a sophist, he is only highlighting his knowl-
edge and commitment to help people (Prometheus Bound, 944). What is
even more interesting is the case of Solon, who had also travelled the
known world: Herodotus (1.29) calls him both sophistês and philosophos,
not because he wishes to portray him as a forerunner of the sophists and
philosophers, but to praise his desire for knowledge (the literal meaning
of philo-sophia being ‘love of wisdom’) and the experience he had
acquired through his travels – two virtues which are certainly note-
worthy, but which do not distinguish the sophists’ teaching activity.

Polemicists, bad teachers, and fake philosophers: the charges against


the sophists

Based on the three criteria just outlined, it is possible to identify a group


of ‘sophists’ in the sense in which Hermann Diels and all modern scho-
lars have understood this term. Still, difficulties remain. Thus far, the
sophists appear to resemble one another in terms of their practical
activities, without having any shared status. But if this is the case,
how do the sophists differ from all other teachers and educators?
Once again, the ancient sources are ambiguous on the matter, and
the simplest answer seems to be that there is no difference between
the sophists and other teachers: in a way, all teachers are sophists.

8
To get an idea of the extent of the sophists’ travels, see the profiles in Appendix 1. It is import-
ant to bear in mind that many sophists also travelled for political reasons, acting as ambassadors of
their cities: see esp. Plato, Hippias, 282b–c, on Gorgias, Prodicus, and Hippias. One may recall
Gorgias’ famous embassy to Athens in 427 BC. Similar embassies have been suggested in relation
to Thrasymachus, albeit on less certain grounds: see White 1995 and Yunis 1997. It is also worth
mentioning Antiphon’s famous embassy to Sparta after the coup of 411 BC – assuming that the
hypothesis that identifies the sophist with the orator by that name is correct.

https://doi.org/10.1017/S0533245119000026 Published online by Cambridge University Press


6 I THE SOPHISTS

Curiously enough, this is the position of Plato’s Protagoras in a passage


of the dialogue that bears his name, where he mentions teachers of
gymnastics and music, along with poets and prophets, as the forerun-
ners of the ‘sophist’s art’ (Plato, Protagoras, 316c–317d, partly quoted
by Diels-Kranz as testimony 80A5 D.-K. = Soph. R11 L.-M.).
However, this very passage shows that the situation is rather more com-
plex: Protagoras admits that the profession of the sophist is very danger-
ous, as it ‘stirs considerable hostility, enmities and intrigues’, to the
point that sophists often even risk coming across as swindlers; and it
is precisely in order to avoid this hostility that Protagoras associates
himself with universally esteemed figures such as the poets Homer,
Hesiod, and Simonides. Protagoras’ move is a brilliant and shrewd
one, but clearly it does not reflect his contemporaries’ beliefs:
there are more differences than affinities between Homer and the soph-
ists, as well as between teachers of gymnastics or music and the
sophists. While affinities lie in the common practice of teaching, the
divergences can be identified in the things taught and the teaching
strategies.
At the present state of our knowledge, it is not easy to establish the
exact nature of these divergences – and not merely because few reliable
testimonies are available. One equally important aspect, which cannot
be overlooked, is the fact that the subject taught varied from sophist to
sophist: Protagoras focused on politics, but some sophists – including
Hippias – also dealt with other disciplines (‘arithmetic, astronomy,
geometry, music, and poetry’: Plato, Protagoras, 318d–e), whereas
Gorgias’ only concern was to make people clever at speaking, ‘to
train skilled orators’ (82A21 D.-K. = 32D47 L.-M.). At first sight,
then, it would seem almost impossible to find any guiding thread in
the sophists’ activity. Upon closer inspection, however, this is not really
the case: if not in positive terms, the sophists at least agree from a nega-
tive standpoint, in terms of their polemical aim. And this, when exam-
ined in the light of their opponents’ responses, can help us more
accurately define the sophists’ activities.
One aspect that constantly emerges from the surviving testimonies is
the fact that the sophists’ teaching – whatever its object – entailed a
staunch opposition to traditional forms of teaching (such as music
and gymnastics) and even more so to other types of knowledge: the
sophists not only upheld the worth of their new intellectual teaching
against traditional education, but even boasted that they were capable
of refuting poets and philosophers, physicians and mathematicians,

https://doi.org/10.1017/S0533245119000026 Published online by Cambridge University Press


I THE SOPHISTS 7

and more generally the experts in all other disciplines.9 As has often
been noted, competition plays a central role in ancient Greece, a society
lacking ‘official’ types of authority: each teacher was required to provide
concrete proof of his superiority over other aspiring ‘wise men’.10 The
sophists offer a striking example of this polemical spirit. Their teaching
constitutes a remarkable challenge to the claims to truth made by other
experts in the polis, and betrays the sophists’ desire to establish them-
selves as the new intellectual masters in Greece.11
From the point of view of their opponents, the sophists’ claim to be
capable of taking the place of experts amounted to the promotion of a
strictly verbal kind of knowledge: the sophists do not really know what
they are talking about; they are only brilliant speakers who conceal their
shortcomings behind linguistic tricks; their only concern is how to win
an argument. Accusations of this sort may be found in many texts by
Hippocratic physicians, who label the sophists ‘professional defamers’
(see Appendix 2). No less eloquent are Aristotle’s observations, which
convey a further and more serious criticism. Aristotle is less drastic
than other detractors and grants the sophists a few merits: it is true –
he states – that the sophists have chiefly focused on language and
argumentative techniques; yet their analyses are still worth discussing,
if only to highlight their errors and fallacies. In Aristotle’s perspective,
however, the real problem is this: the sophists’ exclusive interest in lan-
guage implies that they do not examine reality, causes, and principles,
which is to say that they are not genuine philosophers at all. At best they
are dialecticians, grammarians, and orators, even though it is tempting

9
See Fait 2007: xl–xliv.
10
Lloyd 1979.
11
The sophists’ ambition explains why their activity entailed not only private lessons but also
public performances (epideixis): see Guthrie 1971: 41–4. Privately, the teacher was chiefly con-
cerned with presenting certain argumentative schemes that the pupil could then make use of for
his own benefit (see Natali 1986; it is possible that these arguments were subsequently brought
together to form genuine discourses that might serve as a model for students). Public performance
was a privileged avenue for sophists to promote themselves and their wisdom even before an exten-
sive audience (for instance, during solemn celebrations such as the Olympic Games). For a vivid
description of these performances, see Lloyd 1987: 79–102; for analyses of the possible circulation
of these discourses in written form, see O’Sullivan 1996 and Thomas 2003. While bearing this
distinction in mind, we should not overemphasize the break between public and private, since
even the teaching of arguments could take place in open contexts: this is the case, for instance,
in Plato’s Euthydemus, where Euthydemus and Dionysodorus, mocking their interlocutors, present
argumentative schemes that their pupils can apply to new cases. The truly enduring feature of the
sophists’ activity is its agonistic-competitive character. Interesting points on the historical and cul-
tural context may also be found in Soverini 1998.

https://doi.org/10.1017/S0533245119000026 Published online by Cambridge University Press


8 I THE SOPHISTS

to simply dismiss them as word-jugglers who stand out for their


(unfounded) claim to master the art of discourse.12
Insofar as sophistic teaching was simply reduced to a verbal argu-
ment, it is easy to understand why the sophists were also accused of
being bad teachers, or teachers of immorality: exclusively concerned
with winning arguments, the sophists proved incapable of tackling the
great moral questions that lie at the basis of any real education. In
fact, the importance of this accusation also depended on the social pro-
blems associated with the itinerant character of the sophists’ activity.
Particularly in archaic societies, education was a means to hand
down values and beliefs from father to son, a social mechanism
intended to ensure order and continuity.13 With the sophists, this circle
was broken: their social otherness ensured their independence with
respect to the various communities in which they found themselves
teaching. As foreigners and outsiders in the cities hosting them, they
felt free to question everything and to scrutinize what were traditionally
perceived as absolute, unchangeable, and undeniable values. With the
sophists, the idea that cultural traditions are relative gained ground, as
people realized that the values of a society are not absolute, but are
rather the historical product of such a society. If we add the fact that
the sophists often favoured provocation and paradox for ‘promotional’
reasons, as a means to attract potential clients, it is easy to understand
why within a short time they were made the target of the predictable
and banal accusation of subverting all values. Thus in the Clouds, writ-
ten by the great comic poet Aristophanes and first performed in 423 BC,
the sophists are presented as unscrupulous, hypocritical, quibbling
charlatans (see lines 441–51): they are masters of the ‘worse discourse’,

12
Significantly, in the first book of the Metaphysics, the treatise that is usually regarded as the
first history of philosophy, no mention is made of the sophists, whereas their theses are widely dis-
cussed in the treatises of the Organon – in particular, in the Sophistical Refutations. On Aristotle and
the sophists, see the observations made by Classen 1981, who notes that, although Aristotle does
not regard the sophists as genuine philosophers, he does not simply despise them either. Aristotle’s
interpretation was especially influential among the great Latin orators, from Cicero to Quintilian,
who only dealt with the sophists in relation to their rhetorical studies. Among modern scholars, a
similar interpretation has been upheld by Gomperz 1912 and, more recently, in a completely dif-
ferent context, by several American scholars interested in a reassessment of rhetoric (see e.g.
Schiappa 1991 on Protagoras and Consigny 2001 on Gorgias). Along much the same lines,
Michael Gagarin, who is the author of some of the most enlightening contributions on the ancient
sophists, has stressed the central importance of logos (which can mean ‘word’, ‘speech’, or ‘rea-
son’) as the cornerstone of the sophists’ investigations (see, for example, Gagarin 2002 and
2008). While the importance assigned to logos is indisputable, it does not imply a lack of interest
in ethical and political issues, as we shall see.
13
Goldhill 1986: 222–7.

https://doi.org/10.1017/S0533245119000026 Published online by Cambridge University Press


I THE SOPHISTS 9

the discourse which with unjust arguments overturns the better dis-
course and contradicts the laws in such a way as to make injustice tri-
umph (lines 882–4).14 Naturally, their teaching is not an innocuous
abstraction, but has concrete effects, which affect society to its very
core – or, rather, subvert it: Aristophanes’ comedy ends with a son
beating his father, a gesture which constitutes the most eloquent evi-
dence of the overturning of traditional values.15 The sophists are repre-
sented as the bad teachers of Athens (and the whole of Greece), those
responsible for its moral and political crisis.
I will assess the soundness of these charges at different stages in the
course of the present investigation, in order expose their limits and
prejudiced character. For the moment, I only wish to note that both
accusations in a way depend on what may be regarded as the sophists’
most important contribution: their acknowledgement of the fact that
reality is ‘problematic’.16 It is this awareness which leads the sophists
to investigate new issues and to question established values, as was
already observed by Plato himself – the sophists’ fiercest opponent in
a way, but also the philosopher who had most clearly grasped the philo-
sophical and political significance of the challenge they posed.17 This
not only acquits the sophists of the charge of promoting immorality,
but assigns them a prominent place in the history of philosophy – a
point I will be discussing at length in the following chapters.18 For
the time being, it is important to note that these two accusations
allow us to better define the primary aim of the sophists’ research
and teaching: the charge of playing with words may be explained by
considering the sophists’ utmost interest in the issue of language,
while the accusation of promoting immorality is due to their interest
in practical, ethical, and political problems. The sophists’ favourite
object of enquiry was the art of speech (logos), and in particular its prac-
tical and political applications. Language and politics, then, are two of

14
The thinly veiled allusion here is to Protagoras, who promised to ‘make the weaker argument
the stronger’ (80B6b D.-K.): a provocative claim in its ambiguity, given that ‘weaker’ might also
mean less just.
15
Besides, the sophists had touched upon this theme too in a provocative fashion: see Antiphon
87B44B, 5.4-8 D.-K. = 37D38 L.-M.
16
Paci 1957: 126.
17
See now Corey 2015.
18
Significantly enough, aside from a few exceptions, the dominant view today is precisely that
the sophists’ activity entailed an engagement with ontological and political issues. This may be
inferred from the most authoritative studies on the subject, which, while disagreeing on many
points, agree at least on this one – from Untersteiner 1954 (first published in Italian 1949) to
Kerferd 1981a, from Guthrie 1971 (first published 1969) to Cassin 1995.

https://doi.org/10.1017/S0533245119000026 Published online by Cambridge University Press


10 I THE SOPHISTS

their chief interests. Contrary to what Aristotle suggests, research on


language and arguments is not only of theoretical relevance, but also
entails practical repercussions, insofar as the sophists’ teaching pro-
mises to bring their pupils success in their private and public life.
This – it is worth stressing once more – does not justify any ‘strong’
interpretation of the sophists, as members of the same ‘schools’: the
sophists are grouped together, not because they uphold the same doc-
trines, but because they share the same focuses of interest (namely lan-
guage and politics), the same method of investigation, and similar
aims.19 These points will suffice as an initial overview.
The reticence of the ancient sources notwithstanding, it is possible,
then, to reach some general conclusions on the sophists. They were
travelling teachers, who moved from city to city and taught for a fee
(these being the external criteria). Their thought and teaching focused
on man, his nature, and his needs. This led them to concentrate on
issues pertaining to language and politics, which does not mean – as
we shall see – that they ignored the more traditional topics of
Presocratic reflection (such as physis, for instance). A link between all
the various sophists is to be found in this sharing of the same problems
and attitudes, rather than in the upholding of common theses (internal
criteria). This accounts for the ambition of each sophist, who in polem-
ical opposition to other sophists as much as to poets and philosophers
sought to establish himself as the new intellectual master that Greece
needed. Succinctly put, such is the view which a unitary presentation
of the sophists supports.
The richness and complexity of the sophists’ challenge is eloquently
attested to by the Dissoi logoi, which present the sophists’ aim in the fol-
lowing terms:
I think that it belongs <to the same> man and to the same art to be able to discuss
briefly, to know <the> truth of things, to judge a legal case correctly, to be able to
make speeches to the people, to know the art of speeches, and to teach about the nature
of all things, both their present condition and their origins. (90.8 D.-K. = 40.8.1 L.-M.)

No doubt, it was an ambitious aim.

19
Gagarin 2008: 23.

https://doi.org/10.1017/S0533245119000026 Published online by Cambridge University Press


II BEING AND TRUTH, HUMANITY AND REALITY

In order to understand the meaning and scope of the challenge that the
sophists present to us, we must clear the field of all ambiguities and
prejudices. All too often the sophist is only taken into consideration as a
polemical target, as a counterpart to the philosopher: whereas the philoso-
pher argues in order to seek the truth, the sophist only tries to win an argu-
ment; and whereas the philosopher concerns himself with problems in all
of their complexity, the sophist instrumentally focuses on fashionable
topics that might interest his public of potential pupils (in other words,
people willing to pay him). These contrasts are repeated not only in
relation to the future, which is to say to Plato and Aristotle, but also, retro-
spectively, in relation to the so-called ‘Presocratic philosophers’. As a
consequence, the sophists find themselves in a sort of no man’s land,
and their activities appear to mark a break in the history of philosophy,
interrupting its toilsome and earnest progression from myth to reason.
Like all schematic reconstructions, the one just provided gives rise to
some serious misunderstandings. One initial misunderstanding concerns
the very concept of ‘philosophy’: up until the end of the fifth century there
was no discipline of ‘philosophy’, possessing a distinctive epistemological
status and opposed to other literary genres such as history or rhetoric.
Heraclitus – to quote but one figure among those whom we regard as phi-
losophers – uses the term to mock Pythagoras and Hesiod, accusing them
of knowing many things without really understanding any at all (22B35
D.-K. = Her. 40D L.-M.). As a non-polemical term of praise, the historian
Herodotus calls the poet and politician Solon a ‘philosopher’, commend-
ing his thirst for knowledge (1.30.2); along much the same lines, Pericles
goes so far as to claim that all Athenians are philosophers (Thuc. 2.40.1).
Philosophy, as we understand it, which is to say as an independent discip-
line, is a later invention by Plato and Aristotle: what we find in the fifth
century, rather, are many prose writers (including the sophists) and
poets engaging in an all-round debate that touches upon authors ranging
from Homer to Parmenides, from Anaxagoras to Simonides. (It is import-
ant to remind ourselves that the same kind of openness is also displayed by
Plato, Aristotle, and many other ancient philosophers.)
Only by setting out from a correct historical contextualization can we
address the delicate question of the sophists’ contribution to the history
of ancient thought. The sophists were no doubt interested in language
and all the issues related to the problem of language (argumentative

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


12 II BEING AND TRUTH, HUMANITY AND REALITY

techniques, rhetoric, poetry, and literary criticism), since this was the
topic which most interested their public, especially in democratic
Athens. However, this is not to say that they did not also deal with
other ‘philosophical’ problems which the Presocratics had investigated
and which Plato and Aristotle were to take up again – quite the contrary.
The vagueness of the concept of ‘Presocratic’ has often led to the
misguided attempt to draw a clear-cut distinction between natural
philosophers and sophists, spawning the cliché that the former only
discussed nature, and the latter the human world. Like all clichés,
this one contains both an element of truth and an element of falsehood.
It is true that the Presocratics’ investigations chiefly focused on physis
(nature), while the sophists concerned themselves with what more spe-
cifically pertains to humanity. But it is just as true that the Presocratics
made some very important contributions to the study of language and
human psychology, and that the sophists also dealt with the problems of
physis and human’s relationship with the world. In other words, what
changes is the perspective, not the topics discussed: the Presocratics
searched for the elements of continuity between human beings and
reality, while the sophists were interested in the issue of what distin-
guishes humankind. The change of perspective, however, does not
depend on any lack of interest of the latter towards the former: it is
the consequence of a critical engagement.
Ancient sources repeatedly bear witness to the sophists’ interest in
the Presocratics’ physis. The most significant example is provided by
Gorgias’ treatise On Nature or On Not-Being, which from its very title
sets out to overthrow the theses of Melissus and Parmenides
(Gorgias, after all, was a pupil of Empedocles: see 82A3, 10 D.-K. =
32P4–5 L.-M., and Kerferd 1985). Protagoras (80B2 D.-K. = 31D7,
R2 L.-M.), Lycophron (83.2 D.-K. = 38D1 L.-M.), and probably
Xeniades (81.1 = 39R1 L.-M.) also polemicized against Parmenides
and Zeno. Cicero states – unfortunately, in a very succinct way – that
Prodicus and Thrasymachus, too, investigated the nature of things
(De natura rerum, 84B3 D.-K. = 34D2 L.-M.; and 85A9 D.-K. =
35D4 L.-M.); and Aristophanes describes Prodicus (and the sophist
Socrates) as a ‘meteorosophist’ (which is to say, an ‘expert on celestial
phenomena’: 84A5 D.-K. = Dram.T22–4 L.-M.).1 Finally, recent studies

1
Other testimonies about Prodicus’ naturalistic interests – in particular, that of Galen – have
been collected by Mayhew 2011: T 61–9 (see also Mayhew 2011: 171–5, on Aristotle’s testimony
as a source for reconstructing Prodicus’ cosmology).

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


II BEING and TRUTH, HUMANITY AND REALITY 13

have quite rightly emphasized the relationship between Antiphon and


Democritus; and the former’s relationship with Anaxagoras is yet to
be fully investigated.2 This succinct outline is enough to bring out an
evident yet largely overlooked fact: the sophists were not interested in
the thought of the natural philosophers in general (this category
being so broad as to prove generic); rather, as one would expect, they
focused on the more recent natural philosophers, starting from
Parmenides and the Eleatic school. By and large, the investigations
conducted by many sophists constitute a response to the great chal-
lenge of Parmenidean monism and an attempt to defend the multipli-
city of our world – although many details are yet to be clarified. In this
respect, the sophists’ reflections run parallel to and intersect with those
of the so-called pluralists: Empedocles, Anaxagoras, and Democritus.
In the following pages, I will endeavour to assess the importance of
this engagement. But before proceeding any further, it is necessary to
get back to a crucial point, namely the fact that the sophists do not con-
stitute a homogeneous school: just as they argued with their predeces-
sors, they heatedly argued with one another. As we will see, many of
Antiphon’s theses gain significance in opposition to Protagoras, and
much the same probably holds true for Gorgias, Xeniades, and
Lycophron. This also confirms the richness and liveliness of the debate,
which gave rise to some fundamental philosophical positions that Plato
and Aristotle were bound to engage with.

Protagoras’ truth: the epistemological interpretation

One of Protagoras’ most famous treatises was entitled Truth. In itself,


this title is revealing of the ambitions that lay behind it: ‘truth’ was a
key term for the Presocratics (Detienne 2006);3 in choosing it,
Protagoras clearly sought to enter into the debate by challenging a long-
established tradition. The first sentence of the treatise fully confirms its
author’s provocative intentions: ‘Of all things the measure is man: of
those that are, that they are; and of those that are not, that they are

2
Hourcade 2001. Within the same context, we may also mention those testimonies apparently
concerning a ‘sophistic theory of perception’ (Ioli 2010: 56–9, following in the footsteps of
Monique Dixsaut). Indeed, Platonic passages such as Theaetetus, 153e–154a, and Meno, 76d,
may be taken to suggest that Protagoras and Gorgias had somehow adopted the Presocratic theor-
ies envisaging sense-perception in terms of the flow of particles.
3
Detienne 2006.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


14 II BEING AND TRUTH, HUMANITY AND REALITY

not’ (80B1 D.-K. = 32D9 L.-M.; for the sake of convenience, in the
following pages this sentence will be abbreviated as ‘man-measure’).4
‘Man-measure’ is one of the few authentic Protagoras fragments to
have been preserved, and many ancient sources confirm that the state-
ment was originally formulated precisely in these terms. The problem,
however, is its opacity: even in antiquity there was intense debate con-
cerning what Protagoras really meant, and modern readings have only
increased doubts about its content. As a matter of fact, it cannot be
ruled out that a certain degree of ambiguity was intentional. Truth
was not a scientific treatise but, in all likelihood, an epideixis, the text
of a public performance. In opening his speech, Protagoras sought to
capture the attention of his audience (or reader) by making a striking
and allusive claim – at the expense of exactness and precision – and
putting aside all caution and hesitation (which are hardly the most valu-
able qualities for a public performance). There is nothing strange in any
of this, since a desire to capture the audience’s attention was quite
typical of the sophists, as is confirmed by many other sources.5
This is not to say that the ‘man-measure’ thesis had no specific
meaning. Simply, it is a matter of dealing with potentially ambiguous
elements. In particular, it is necessary to clarify the notion of man:
that is, whether ‘man’ refers to each individual or to humanity in gen-
eral; in discussing this issue, the meaning of the other terms that make
up the sentence will also become clearer. In turn, this will help us assess
the underlying meaning of Protagoras’ statement and its field of applic-
ability, so as to determine whether it is an epistemological thesis (a the-
sis concerning knowledge) or whether it also carries practical or
political implications. By way of anticipating my argument, I note
here that the sentence can be read on different levels, in line with the
archaic logic according to which there is no need to clearly distinguish
the different meanings possessed by a term.
There are three main sources for the ‘man-measure’ thesis: Plato,
Aristotle, and Sextus Empiricus (a second-century CE sceptic). All

4
The sceptic Sextus Empiricus, one of our most important sources, attributes this sentence to a
work entitled Kataballontes logoi (The Overthrower Arguments or, better, The Knockdown Arguments;
on this translation, see Appendix 2): one possible solution may be that Truth was one of the dis-
courses in the Antilogies, which were also known with the (sub)title of Kataballontes logoi: see
Decleva Caizzi 1999: 317. Among the most important studies, see Vlastos 1956; Decleva Caizzi
1978; Barnes 1979: ii.541–53; Mansfeld 1981; Farrar 1988.
5
The sentence introducing the treatise on the gods was just as provocative: see Chapter 6,
p. 112.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


II BEING and TRUTH, HUMANITY AND REALITY 15

three understand the phrase – at any rate at a primary level – as though


‘man’ referred to ‘each individual’. The most interesting testimony is in
all probability that of Plato, who repeatedly engages with Protagoras’
thought in his dialogues, especially the Theaetetus and Protagoras. In
the Theaetetus, Socrates and the young mathematician Theaetetus
extensively discuss the ‘man-measure’ thesis, regarding Protagoras as
an early empiricist thinker. The Protagorean phrase is equated with
the claim that ‘knowledge is sensation’, and paraphrased as follows:
‘just as each thing appears to me, so too it is for me, and just as it
appears to you, so too again for you’. Thus, if the wind appears
warm to me, it is warm for me; and if it appears cold to another person,
it is cold for him or her (Theaetetus, 152a = 80B1 D.-K. = 31R5 L.-M.).
In other words, ‘man-measure’ means that each person is the measure
of his or her own sensations, and Protagoras emerges as a staunch
supporter of an empiricist epistemology that grounds knowledge in
sensory data.
Plato’s testimony is crucial and his reconstruction is also taken up by
other authors (see 80A16 D.-K. = 31R28 L.-M. and 70B1 D.-K.). In
and of itself, however, it is a reductive interpretation that risks proving
misleading, as far as the meaning of ‘man’ is concerned. The limits of
Plato’s testimony become quite evident if we consider the underlying
aim of the Theaetetus, which is not to accurately report the theses of
Plato’s opponent, but to show their philosophical limits. According
to Plato, the ‘man-measure’ thesis is a muddled one; the only way to
make it clear is by assuming that Protagoras was only speaking of sen-
sations; and since (according to Plato) it is impossible to base knowl-
edge on sensations alone, it follows that the thesis is fallacious. In
other words, the discussion on Protagoras is intended to be a refutation
of his thesis. The point here is not to evaluate the legitimacy of Plato’s
criticism (who anticipates or inaugurates one of the most important
problems in the history of philosophy, namely the clash between ration-
alism and empiricism). What matters is the fact that a strict empiricist
reading stems from a polemical interpretation, which may be philo-
sophically legitimate but is not historically reliable: clearly, Protagoras
was not speaking of sensations alone. Remarkably, Plato himself sug-
gests that Protagoras’ phrase had a broader meaning between the
lines of the Theaetetus, in certain passages where it emerges that ‘man-
measure’ also concerned all opinions and judgements (see 157d and
170a–171a). Aristotle and Sextus Empiricus further confirm the need
for a broader interpretation: what are at play are not just sensations

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


16 II BEING AND TRUTH, HUMANITY AND REALITY

but more generally all opinions and judgements (and especially value
judgements). ‘Man is the measure of all things’ means that each indi-
vidual is the ultimate arbiter of all of his or her own judgements –
that the wind is warm or cold, but also that performing a given action
is right or wrong.6
The point just made is crucial in order to reconstruct the overall
meaning of Protagoras’ position. The transition to a broader concep-
tion of human activities (not only sense-perceptions but human judge-
ments as a whole) brings out one aspect of the thesis that scholars have
all too often overlooked. Usually it is taken for granted that Protagoras’
thesis concerned problems related to human knowledge and that these
problems were addressed in abstract terms, by examining the general
ways in which the process of knowledge acquisition unfolds – in
other words, by investigating what happens to a subject X who at a
given moment t experiences sensation f (let us think of the example
of the wind). Yet this is not correct. It is certainly true that
Protagoras’ thesis was concerned with human knowledge. However,
it is not true that it focused on it in an ‘abstract’ way, as though humans
were ‘automata’ that can be stimulated from the outside so as to study
their reactions at given moments and in specific circumstances. This
approach might work in the case of the wind, but not in that of judge-
ments concerning what is good or bad. ‘Man’ refers to the individual,
yet not to an abstract subject: Protagoras has concrete, historically well-
defined people in mind, with their ideas and prejudices. What is meant
by ‘man’, therefore, is each person with his or her own personal history,
experiences, and expectations: that is, people for whom some options
are possible while others are not, and whose judgements largely reflect
their individual experience – people who are responsible for the opi-
nions they have and for what they do. When I claim that a certain
thing or action is good (or beautiful or unjust), I do so on the basis
of a series of ideas and opinions that I have acquired over the course
of my life and that I have continued to put to the test through my
engagement with facts – in other words, on the basis of my experience,
which is different from that of other people. The true measure, then,
is not ‘man’ in the abstract sense but rather each individual’s

6
To be more precise, it must be noted that Protagoras, like many thinkers of his day, does not
distinguish between the mind and sense-perceptions as though they concerned two completely dif-
ferent domains (i.e. the intelligible world and the sensible one): cf. 80A1 D.-K = 31R13 L.-M. on
the soul.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


II BEING and TRUTH, HUMANITY AND REALITY 17

experience.7 The ultimate purpose of Protagoras’ statement is to


re-evaluate human experiences; as we shall soon see, it also carries a
polemical undertone with respect to the previous tradition.
Having clarified the meaning of ‘man’, we can now turn to consider
the second part of the phrase, ‘of those that are, that they are; and of
those that are not, that they are not’. Essentially, it is a matter of clari-
fying what ‘things’ means and in what sense man ‘measures’ them.
Based on what has been noted so far, we can reasonably claim that
what Protagoras meant by ‘things’ was not just material objects and
entities (things that can be perceived by the senses, such as wind):
‘things’ are to be understood in their broadest possible sense as
‘facts’ – for example, the wind, but also an event such as performing
a certain action, and all that happens to man. In this respect, the
term chosen by Protagoras is highly revealing. Chrema derives from
the verb chraomai, meaning ‘to treat’, ‘to use’, ‘to entertain a relation
with’: according to its original meaning, the term describes not so
much ‘things’ in themselves (for in this case Protagoras could have
used onta or pragmata), as in their relation to us; the emphasis is on
the way in which we relate to things, use them, judge them, and so
on. ‘Things’, then, means everything which human beings enter into
contact with: material objects but also events and occurrences – in
other words, facts.
As far as the notion of measure is concerned, it is clear that man is
the measure not of the existence of things (such as the wind or the
occurrence of a certain action), but rather of the way in which things
present themselves (for example, as warm/cold or good/bad). The
expression, that is, must be understood not in an existential sense (in
which case it would be difficult to understand what it means for man
to be the measure of the non-existence of the things that are not),
but rather in a modal-predicative sense. Protagoras is not denying the
existence of the external world, but only limiting the possibility of
knowing it to the way in which it presents itself to our experience
(hence, if I judge the wind to be warm, I am the measure of the fact
that it is warm and not of the fact that it exists; if I judge the wind
not to be warm, I am measure of the fact that it is not warm and not
of the fact that it does not exist). Things in themselves do not possess
any predetermined truth or any intrinsic value: they simply exist, they

7
Mansfeld 1981: 44–6.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


18 II BEING AND TRUTH, HUMANITY AND REALITY

are what surrounds a person, what happens to him or her. What matters
is the way in which we relate to them: each individual, according to his
or her experience, is the judge of these facts, insofar as he or she assigns
them a degree of consistency or value.
This initial reconstruction also enables us to evaluate Protagoras’
position with respect to the previous intellectual tradition: ‘man-
measure’ emerges as a polemical response to the enquiries on physis
conducted by many Presocratics, and in particular to Parmenides’
monism.8 Protagoras’ interest in critically engaging with the Eleatic
school is confirmed by other sources: in particular, Porphyry (a
third–fourth-century CE Neoplatonist) informs us that Protagoras
wrote a treatise On Being (regrettably lost) against those who ‘uphold
the uniqueness of being’ – clearly, Parmenides and his disciples
(80B2 D.-K. = 31D7, R2 L.-M.); a polemic with Zeno is also attested.9
The same polemical intentions are suggested by Plato’s Theaetetus,
which presents Protagoras and Parmenides as the leading exponents
of two opposite ways of conceiving reality, namely as becoming and
as being.
Once again, although Plato’s testimony is crucial, it must be exam-
ined with great caution, so as to avoid pushing our interpretation
of Protagoras too far. The Theaetetus ascribes a sort of radical
Heracliteanism to Protagoras that presents reality as the constant flow
of all things, a flow so incessant that we should speak not of ‘being’

8
Farrar 1988: 48–50.
See 29A29 D.-K. = Zen. D12b L.-M.: ‘For he [Zeno] said: “Tell me, Protagoras, does one
9

grain of millet make a sound when it falls or does the thousandth part of the grain of millet?”
When the other answered that it did not, he said, “Does a medimnus of grains of millet make a
sound or not when it falls?” When the other answered that it did make a noise, Zeno said,
“Well then, is there not a proportion between a medimnus of grains of millet and a single grain
and the thousandth part of that one grain?” And when the other answered that there was one,
Zeno said, “Well then, will there not be the same proportions between the sounds with regard
to one another? For just as the things are that make a sound, so too are their sounds; and since
that is so, if a medimnus of millet makes a sound, a single grain of millet will make a sound
too, and so too the thousandth part of that grain”’. While not impossible in chronological
terms, it seems unlikely that the meeting between the two thinkers really occurred. However,
this passage may still be seen to provide further evidence of the polemical exchanges that the soph-
ist entertained with the Eleatic school. The text in question has sometimes been interpreted as
proof of the fact that Protagoras denied infinite divisibility; but this seems like an incorrect conclu-
sion, since Protagoras – at least in theory – does not deny the possibility of a division into increas-
ingly small parts. What he denies is rather that the sound produced by these portions of millet is
audible. Once again, we may note that the sophist examined things from the point of view of sens-
ible experience, against the abstractions of the natural philosophers: just as sight does not perceive
the touching of a sphere and a tangent at a given point (see 80B7 D.-K. =31D33 L.-M.), so hear-
ing can only perceive sounds up to a certain point. Even from this perspective, man is the measure
of all things.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


II BEING and TRUTH, HUMANITY AND REALITY 19

but only of ‘becoming’. On the opposite side of the spectrum, accord-


ing to the dialogue, stands Parmenides, the philosopher of the stability
of being. This thesis, however, is conditioned by the empiricist inter-
pretation, whose limits I have already highlighted: if we only speak of
sensations, it is clear that what we have in mind is a world exclusively
made up of material objects (since sensations are only engendered
starting from material things, such as a table or the wind); and if sensa-
tions vary (in the sense that the wind will appear warm at one time and
cold at another), this occurs because material reality varies, changing
constantly. In other words, the notion that Protagoras supported a flux-
ist interpretation of reality rests on the assumption that he was referring
to sensations alone. But this, as we have seen, is a reductive interpret-
ation. Significantly, Plato referred to this thesis as Protagoras’ ‘secret
doctrine’ (Theaetetus, 152c), thereby revealing that it was not a thesis
openly upheld by the sophist, but rather a position that in his view
could legitimately be inferred from his claims.10
Protagoras’ engagement with Parmenides does not actually take
place on the ontological level. His main interest was not to oppose a dif-
ferent (and truer) world to that of Parmenides (the world of becoming
to the world of being, as Plato would have it) – a cosmology capable of
accounting for the multiplicity of natural entities.11 In this respect,
Protagoras differs from the so-called ‘pluralist’ philosophers
Empedocles, Anaxagoras, and Democritus, who, in response to
Parmenides, had sought to ‘save the phenomena’. Protagoras rather
set out to ‘save man’: his challenge was a more radical one, and it tar-
geted the very assumptions on which Parmenides’ reflection rested. For
Parmenides (like many other natural philosophers), terms such as
‘man’ and ‘opinion’ were synonymous with error, by contrast to the
truth and being: in his view, the purpose of philosophy was to investi-
gate the universal and absolute first principles. Even the multifaceted

10
On this passage, see Zilioli 2013: 239–43, arguing that Plato’s text attributes to Protagoras
not so much a radical version of a Heraclitean theory of flux as a theory of ontological indetermin-
acy; and Corradi 2012: 79–89.
11
This does not rule out the possibility that Protagoras’ conception of material reality may have
been similar to that of other Presocratics, but it is important to stress once again that this was not
the sophist’s chief interest. In brief, and without delving into the details, it cannot be ruled out that
Protagoras viewed physical reality as something that contains all opposites. If this is the case, the
sophist was only taking up a thesis that was common among natural philosophers and typical of the
polarizing thought of the Presocratics, who were used to envisaging nature in terms of contrasting
or harmonizing opposites (let us think here of the Pythagoreans or of Heraclitus). This is Sextus
Empiricus’ interpretation (Outlines of Pyrrhonism, 1.216–19 = 80A14 D.-K. = 31R21 L.-M.), cf.
Woodruff 1999: 303–4.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


20 II BEING AND TRUTH, HUMANITY AND REALITY

world of humankind was to be investigated with the same perspective,


by identifying its points of contact with being and subordinating what is
relative or contingent within it. In other words, Parmenides (a) had
drawn a clear line between the true world of being on the one hand
and the world of appearances and becoming on the other, while (b)
asserting the possibility of a privileged form of knowledge that moves
beyond the illusory world surrounding us in order to reach true reality
(the truth of being), and (c) discrediting ordinary human knowledge
(unfounded opinions).
Protagoras opposed all these assumptions: (b) because nobody
enjoys privileged access to these alleged higher ‘realities’; (a) because
no-one can ignore the reality that surrounds us, and which is our one
and only world; and (c) because it is not true that the world that sur-
rounds us is helplessly muddled and corrupt, since through our judge-
ments we are capable of ‘measuring’ it and giving it an order – which is
to say, of knowing it. There are no wise men and ignorant people, then:
each man can legitimately express his own opinions and judgements.
Protagoras reverses Parmenides’ perspective, reappraising the unique-
ness of human experience, which is always relative and contingent:
this relativeness and contingency is not a limit but rather the condition
on the basis of which we relate to things. Instead of developing hypoth-
eses on the nature of the all, we should shift our focus back to our
everyday life in order to solve the problems affecting it. This was the
programmatic (and polemical) message conveyed by the first sentence
of the treatise on Truth, as discussed by Versenyi:

The things we are concerned with are chremata, i.e. things we are decisively related to;
thus there is no point in speaking in a grand manner about what things may or may not
be in themselves; what we have to take into account and concentrate on is what they are
for us, in the world we live in, in a world in which our relationship to things, our living
in the world, is decisive.12

Protagoras’ position has been described in many different ways. Much


emphasis has been placed on his ‘subjectivism’, stressing the import-
ance of the subject for the sophist; yet interpreters have also spoken
of ‘objectivism’, noting once again that the ‘man-measure’ thesis pre-
supposes a real and concrete relation with reality.13 Other scholars
have spoken of phenomenalism, stressing the fact that reality always

12
Versenyi 1962: 181.
13
More recently, see also the ‘pluralist’ reading developed by Apfel 2011: 45–78.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


II BEING and TRUTH, HUMANITY AND REALITY 21

presents itself as a phenomenon, as that which appears. Like all ‘-isms’,


these classifications are valid as long as we remember that they are
merely tools that may help identify the important aspects of
Protagoras’ thesis, and not formulas that miraculously explain every-
thing. Given this caveat, the category that probably best describes the
particularity and originality of Protagoras’ thought is that of relativ-
ism.14 Knowledge always depends on the existing circumstances and
the knowing subjects, which is why it makes no sense to speak of a sin-
gle and absolute truth: truth is the relationship which each person
establishes with surrounding reality. There is no one absolute truth,
only as many truths as there are perceiving and judging subjects.

Protagoras’ truth: the practical and political consequences

The thesis that man is the measure of his own judgements faces two ser-
ious objections: the charge that it is self-refuting and the charge of sol-
ipsism. Protagoras was at least partly aware of these two difficulties, and
clarifying them will help us bring out the richness and significance of
his thesis.
The charge of being self-refuting (or peritrope, as the ancients called
it: literally, the charge of ‘tripping up’) was apparently formulated for
the first time by Democritus (80A15 D.-K. = 68A114 D.-K. = 31R22
L.-M.) and then further developed by Plato (in the Theaetetus and
Euthydemus) and Aristotle (in the fourth book of the Metaphysics). In
brief, this accusation states that, if all judgements are true (this being
the implicit meaning of the ‘man-measure’ thesis), then the judgement
that ‘man-measure’ is false is also true; hence, ‘man-measure’ is false.
Protagoras had further claimed that ‘it is not possible to contradict’
(80A19 D.-K. = 31R10 L.-M.) and this led Plato and Aristotle to
add that the sophist denied the principle of non-contradiction, suggest-
ing another version of the peritrope charge: Protagoras claims that all
judgements are true (let us call this thesis P) and his opponent that
Protagoras’ claim is false (non-P); but, according to Protagoras, contra-
diction is impossible and therefore non-P is compatible with P. But in
this case, non-P is true; hence, P is false.15

14
See Woodruff 1999: 300–4; Lee 2005: 30–45; Zilioli 2013. On the adoption of this category
with reference to Protagoras and the sophists, consider the cautious note sounded by Bett 1989.
15
Barnes 1979: ii.548.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


22 II BEING AND TRUTH, HUMANITY AND REALITY

The soundness of this criticism has been the object of a heated schol-
arly debate, yet no shared solution has been reached.16 Protagoras could
nonetheless resort to a few arguments to defend his views. The accus-
ation omits the qualifiers ‘for me’ and ‘for you’, which are crucial to the
‘man-measure’ thesis. By introducing these qualifiers, Protagoras could
reformulate the peritrope charge in such a way as to neutralize it: if what
I judge is true and I judge that Protagoras’ thesis is false, it does
not follow from this that Protagoras’ thesis is false in an absolute
sense, but only that it is false ‘for me’. And the fact that it is
false ‘for me’ does not prevent it from being true for other people: it
remains true, for instance, for Protagoras, who can thus avoid the
charge that he is tripping himself up (as well as denying the principle
of non-contradiction).
This defensive strategy, however, is potentially open to an even more
serious charge, that of solipsism. This risk is clearly present in the afore-
mentioned thesis that it is ‘impossible to contradict’. The thesis in
question is fully compatible with ‘man-measure’, for if each individual
is the measure of his or her own world of sensations and judgements,
contradiction can only be apparent: each individual entertains a
relationship with things whose truth-reality cannot be contradicted or
disputed by others. Still, it is just as evident that people differ, since
some maintain what others deny. But how can we deal with these diver-
gences and oppositions, if everyone is right? If this is how things stand,
what follows is a radical (and intolerable) form or solipsism, whereby
the fact that each person is his or her own judge makes it impossible
to decide things with others. The price that Protagoras must pay in
order to defend the consistency of his doctrines seems far too high.17
In this case, too, Protagoras has a number of good arguments on his
side. Up until this point, the discussion has centred on the problem of
‘truth’, a concept the importance of which can hardly be disputed.
However, no matter how central it may be, truth is not the only criter-
ion that governs human thoughts and actions. Alongside truth there are
other, equally important criteria. One is ‘what is useful’: all judgements
are true, but some are more expedient than others. And this is what
matters the most for Protagoras. To exploit a compelling formula,
Marcel Detienne described the Presocratic thinkers as ‘masters of

16
Burnyeat 1976; Lee 2005: 46–76; and especially Castagnoli 2010.
17
Woodruff 1999: 303.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


II BEING and TRUTH, HUMANITY AND REALITY 23

truth’; Protagoras (along with all sophists) might instead be called the
‘master of usefulness’.
Once more, with regard to what is useful, the most important testi-
mony comes from Plato.18 As we have seen, the main object of the
Theaetetus is knowledge, and it is in relation to this problem that
Plato introduces, discusses, and criticizes Protagoras’ thesis.
However, the dialogue also features a lengthy digression that, while it
does not directly touch upon epistemological problems, nonetheless
helps to clarify Protagoras’ perspective in all of its complexity: this is
the so-called ‘Apology of Protagoras’ (166a–168c), a lengthy speech
that Protagoras would have delivered, had he had the opportunity to
take part in the dialogue (the dialogue of the Theaetetus takes place in
399 BC, the year of Socrates’ trial, by which time Protagoras had already
died). In his fictional reply, Protagoras accuses Socrates of playing the
sophist and of intentionally deceiving his interlocutors by playing on
the affinity between the words ‘to know’/‘to have knowledge’ and ‘to
be wise’/‘to have wisdom’. But, the sophist continues, the truth-value
of a type of wisdom is one thing, the value of its contents another: if
we are to reason correctly, we must not confuse these two things.
Unlike the former, which is changeless, the latter varies: from the
point of view of knowledge and truth, every individual is a measure,
and the fact that each person is a measure depends on his or her rela-
tionship with the things that surrounds him or her – this is a fact. But
from the point of view of the value of the contents (of wisdom), it can-
not be ruled out that some people are capable of showing the individual
how to establish a more expedient relationship with reality.
Socrates puts the following words into Protagoras’ mouth:
For I myself say that the truth is just as I wrote it: each of us is the measure of the things
that are and of those that are not, but one person differs enormously from another pre-
cisely inasmuch as things exist and appear to be something for one person, something
else for another. As for wisdom (sophia) and the wise man (sophos), I am very far from
saying that they do not exist; but I also call the man wise who, by transforming things,
makes them appear to be good and be good for someone to whom they appeared to be
bad and were bad.
(Theaetetus, 166d = 80A21 D.-K. = 31D38 L.-M.; transl. after L.-M.)

18
Very interesting considerations regarding the problem of what is useful and its relativity may
be found in Guthrie 1971: 164–75. De facto, the relativity of what is useful may be understood
both in an objective sense (the usefulness of a thing varies depending on the people or circum-
stances) and in a subjective sense (nothing is good or bad in itself: it is we who establish this).
The first sense best expresses Protagoras’ position.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


24 II BEING AND TRUTH, HUMANITY AND REALITY

Thus, to a sick person food will appear bad and be bad: it would be
meaningless to argue that the sick person is wrong to say that food
tastes bitter to him or her. However, there is someone, namely the doc-
tor, who can help the sick person build a more expedient relationship
with reality, by ensuring that food will taste sweet, not bitter, to
them. Protagoras’ aim is not merely to abolish the idea of truth,
which is shattered into an endless number of private truths; his aim is
to replace this criterion with another, more effective criterion – that
of what is useful. It is by reflecting on the issue of what is useful that
the sophist, while respecting everybody’s opinion, shows his wisdom.
For, ultimately, what is useful varies from one situation to the next
(80A22 D.-K.), and the sophist’s task is to identify what is truly expe-
dient in each case. Therefore, just as the doctor helps the patient’s
body, so the sophist can help his patient build a better, more expedient
relationship with reality: relying on his experience, the sophist can help
his pupil evaluate his or her beliefs and whether he or she regards them
as being truly important, or whether they are merely values that he or
she has passively inherited from tradition. The sophist can verify
whether the pupil’s system of values is consistent, or whether it includes
contrasting beliefs. Finally, the sophist can give the pupil advice on how
to set out to achieve his or her goals. In brief, we may conclude that
Protagoras teaches only one, fundamental thing: how to judge well
on the basis of the existing circumstances.19
By focusing on the criterion of what is useful, Protagoras can thus
justify his teaching within a context in which the space for truth has
been substantially limited: the wisdom he claims to possess does not
stem from any privileged access to an imaginary higher realm of ultim-
ate truths from which other people are barred. Rather, his wisdom is
based on his experience and his capacity to master arguments and pro-
blems: it is a human form of knowledge that is never conclusive but
nonetheless has its usefulness, insofar as it can help the pupil under-
stand what is best for him or her and how to achieve it. Protagoras’
teaching takes the form not of a transmission of data and information,
but of a means to establish a more effective relationship with reality.
The significance of Protagoras’ wisdom emerges even more clearly in
relation to political problems, the field in which he most flaunted his
competences. In a famous passage from another Platonic dialogue,

19
Woodruff 1999: 309; Woodruff 2013.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


II BEING and TRUTH, HUMANITY AND REALITY 25

the Protagoras, the sophist claims that what he teaches is essentially how
to make the best decisions with respect not only to private matters but
also (and especially) to the affairs of the city, noting immediately after-
wards that this is the ‘political art’ ( politike techne).20 The object of pol-
itics is the contrast between the potentially conflictual values of
different people: everyone has his or her own ideas, values, and aims.
However, these values and aims often differ or even clash, and this cre-
ates the premises for conflict. Here is where the opportunity arises for
the sophist to intervene. Like everyone else, Protagoras cannot claim to
objectively know what is good or expedient in absolute terms. But,
unlike many other people, he possesses certain practical competences
and knows how to foster a critical reflection among citizens to help
them reach an agreement on what is best for everyone – to help them
identify those values (which are not absolute values, but ones that
can always be reappraised) that make it possible to establish a better
relation among the individuals of the community, as well as with reality.
The sophist can do so not only in those rare cases in which he can dir-
ectly participate in the political life of a city, but also indirectly, by
teaching his pupils to become good politicians, which is to say politi-
cians capable of building a consensus that will defuse the risk of con-
flict. In both cases, Protagoras’ wisdom proves crucial for the life of
the polis, and in particular for that of a democratic polis such as
Athens.21 Protagoras’ wisdom is not divine but human and, most
importantly, political.
To conclude, we can reappraise the overall meaning of the ‘man-
measure’ thesis. The introduction of the criterion of what is useful
entails a significant shift of perspective. The thesis undoubtedly con-
cerns the problem of knowledge, but it is important to stress that it is
not limited to epistemological issues; rather, one might argue that
Protagoras is chiefly concerned with the problem of knowledge in rela-
tion to its practical consequences. The problem of truth is not just an
epistemological problem, but also a practical one. Consequently, the
term ‘man’ acquires a different meaning as well. The object of ‘man-
measure’ is not just individuals but humans in their mutual relations.

20
‘The object of my instruction is good deliberation about household matters, to know how to
manage one’s own household in the best way possible, and about those of the city, so as to be most
capable of acting and speaking in the city’s interests’ (Protagoras, 318e–319a). Reference to the
political craft is made immediately afterwards, at 319a (both passages are included in the
Diels-Kranz edition as testimony 80A5 = 31D37 L.-M.).
21
Guthrie 1971: 174–5.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


26 II BEING AND TRUTH, HUMANITY AND REALITY

Protagoras’ statement therefore implies increasing degrees of complex-


ity, leading from the individual to the polis (understood as a human
community): if I am the measure of my own judgements, then we as
members of our community are judges of our own decisions. In this
sense, it is possible to note that ‘man’ stands not just for every individ-
ual but for human beings in general, who in the absence of absolute
truths jointly establish those values which will enable them to live
together in the best possible way.
A detailed analysis of the available sources thus reveals the signifi-
cance and richness of the ‘man-measure’ thesis, which fully confirms
the programmatic value that Protagoras sought to assign to it by
employing it as the opening sentence of Truth. The statement has
different, mutually compatible, interpretative levels, which reveal the
complexity of the human dimension. The ‘man-measure’ thesis tackles
the problem of human knowledge – of what humans can know and
how; in this respect, it upholds the importance of each individual’s
experiences, in open opposition to the previous philosophical tradition.
The thesis, however, also concerns practical matters: not just what we
can know and how, but how we can act. As we will see later on, political
reflection arguably constitutes the most interesting aspect of
Protagoras’ speculation. What is important to note at this stage is
that the political aspects of his thought, of which he was so proud,
closely depend on his ontological and epistemological theses, in the
light of which he defines the human experience.

Gorgias: the dialectic of not-being

Even more provocative than Protagoras’ Truth was Gorgias’ text On


Not-Being or on Nature (82B3 D.-K. = 32D26 L.-M.). ‘Nature’ was a
key term for Presocratic thinkers, who used it to describe reality as a
whole. Nature, in other words, coincides with total reality, with that
which is: with being, as the Parmenidean Melissus had concluded by
entitling his treatise On Nature or On Being. If, as would seem to be
the case,22 the title was indeed chosen by Gorgias himself, what we

22
The question of titles is always a thorny one for archaic and classical authors. However, as
rightly noted by Palmer 2009: 205 n. 25, the peculiarity of Melissus’ and Gorgias’ titles would
seem to suggest that they are authentic and that the latter was probably drawing upon the former.
On the possible dating of the text to the mid-fifth century BC (or, more precisely, to the years 444–

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


II BEING and TRUTH, HUMANITY AND REALITY 27

have is a paradoxical equating of reality with nothingness.23 Gorgias’


treatise represents the complete ‘overturning’ of the Presocratics’
grand investigation of nature, or rather its negation: in typical fashion,
the sophist’s argument rests on an intellectual tradition in order to
overthrow it.24
Regrettably, the original text is lost – and with it, in all likelihood,
many of the subtleties of Gorgias’ style and arguments. Still, the
sources at our disposal give us a good enough idea of the content of
this treatise and of the genuine tour de force that Gorgias imposed on
his public.25 As far as we can tell, the text consisted of a tight sequence
of arguments in defence of three paradoxical theses: ‘Gorgias says that
nothing is; and if [sc. something] is, it is unknowable; and if [sc. some-
thing] both is and is knowable, it cannot be indicated to other people’
(M.X.G. 979a12–13 = D26 L.-M.). What were Gorgias’ more general
aims? In order to answer this question, we need to examine the argu-
ments used in support of each of the three theses. In favour of the
first thesis, according to which nothing exists, Gorgias put forward
two series of arguments, by developing some personal arguments and
by dialectically exploiting his opponents’ theses (the anonymous author
of M.X.G. explicitly mentions Melissus and Zeno: 979a22–3). In
‘Gorgias’ own argument’, the chief topic of discussion is the crucial
contrast between being and not-being (979a25–33); in the dialectical
demonstration, what Gorgias investigated was instead some of the fun-
damental properties of being, such as the fact of being generated or
ungenerated, one or many (979b20–980a8).
The aim of the first argument is to demonstrate that neither being
nor not-being can exist, which leads to the conclusion that nothing

441 BC, the date of the eighty-fourth Olympic Games: 82B2 D.-K. = 32P4 L.-M.), see the cautious
remarks by Mansfeld 1985: 247 and Ioli 2010: 15–18.
23
Wardy 1996: 15.
24
Cassin 1995: 27.
25
We have two main sources: an anonymous treatise devoted to a critical discussion of the
thought of Melissus, Xenophanes, and Gorgias (On Melissus, Xenophanes, and Gorgias, usually
abbreviated to M.X.G.; on the identity of this author, see Ioli 2010: 23–8); and Sextus
Empiricus, Against the Mathematicians, 7.65–87. Despite the evident similarities, these two texts
arrange Gorgias’ writing in different ways: the more reliable source is probably M.X.G. (which
I will be following); Sextus’ discussion (the only one published in the Diels-Kranz edition) appears
instead to reflect a desire to reshape Gorgias’ arguments by lending them a sceptical twist.
Particularly crucial are the analyses provided by Calogero 1932; Newiger 1973; and Mansfeld
1985. Among other modern editions, I shall refer to Cassin 1980; Buchheim 1989; and Ioli
2010 and 2013.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


28 II BEING AND TRUTH, HUMANITY AND REALITY

is.26 In order to reach this goal, Gorgias develops his argument in two
stages: first he establishes the identity between being and not-being,
demonstrating that even not-being is; from this he derives two conse-
quences that lead to the impossibility of both being and not-being,
and hence to the conclusion that nothing is. The argumentative struc-
ture is always the same: (a) first Gorgias establishes a starting assump-
tion; (b) then he explicates what follows from it; (c) he explains the
reasons for this consequence; and finally (d) he draws his conclusions.
Here is a schematic reconstruction of the whole passage:
I. Identify being with not-being
(a) starting assumption: if not-being is not-being;
(b) consequence: what is not is no less than what is is;
(c) explanation: for what is not is what it is not, and what is is what
it is;
(d) conclusion: hence, not-being is as much as being.27
This conclusion raises two distinct problems (not-being is; not-being is
in a similar way to being), which are analysed as follows:
II.1. First consequence
(a) starting assumption: if not-being is;
(b) consequence: being, its opposite, is not;
(c) explanation: for if not-being is, then being, its opposite, is not;
(d) conclusion: hence, nothing is.
II.2. Second consequence
(a) starting assumption: if being and not-being are the same thing;
(b) consequence: being is not;
(c) explanation: for what is not is not-being, and hence even what is is
not-being;
(d) conclusion: hence, nothing is.

As has rightly been observed, ‘G[orgias]’ own argument’ presents ‘a


novel exploration’ of the notion of being and not-being, thanks to its
polemical reworking of Parmenides’ theses.28 Indeed, the meaning of
the first thesis becomes clearer if we view it in relation to a fundamental

26
The most convincing analyses are found in Calogero 1932: 189–268; Mansfeld 1985; Striker
1996: 11–14; Palmer 1999: 66–74; Curd 2006; and Ioli 2010: 28–40.
27
The text of the anonymous M.X.G. literally states: ‘so that things ( pragmata) are not more
(ouden mallon) than they are not’. However, we should not rule out the possibility that the choice
of the expression ouden mallon and the use of pragma may depend on the vocabulary (and
Pyrrhonian inclinations) of the source rather than on Gorgias himself: see Mansfeld 1988: 258,
followed by Curd 2006: 187 n. 8; contra, see Ioli 2010: 34–36.
28
Curd 2006: 186.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


II BEING and TRUTH, HUMANITY AND REALITY 29

verse from Parmenides’ poem.29 In presenting the two paths – that of


being, which alone can be travelled, and that of not-being, which can-
not even be spoken or thought of – Parmenides had written with regard
to the latter ‘that it is not and that it is necessary for it not to be’ (28B2,
5 D.-K. = Parm. D6 L.-M.; my translation). What is the subject of this
sentence? Scholars have long debated this issue; on his part, Gorgias
seems to understand the ‘is not’ of the first hemistich as the subject
of the second part of the verse, as though the meaning of the sentence
were: it is not and it is necessary for what is not not to be. Gorgias thus
exploits the ambiguity of Parmenides’ use of the verb being in support
of his own paradoxical thesis: his reasoning rests on the observation of
the fact that if we claim that what is not is not, then we are de facto
implying that what is not is something, namely what is not – just as
what is is something, namely what is.30 To be more precise, the ambi-
guity lies in the double meaning which Parmenides apparently attri-
butes to the verb ‘to be’: an existential and a predicative meaning.
The verb ‘to be’ is existential insofar as to claim that being is is tanta-
mount to claiming that being exists; and it is predicative insofar as to
claim that being is indicates the essential property which distinguishes
it – a property that Parmenides clarifies further on in his poem, when
he explains that the fact of being implies the fact of being ungenerated,
imperishable, and so forth (if what is truly is, then it must be ungener-
ated, since if it were generated, it would previously have been other
than what is and hence not-being, and so on: see 28B8 D.-K. =
Parm. D8 L.-M.).
Gorgias’ innovation lies in the fact of having applied this reasoning
no longer to being but to not-being: if the essential property of what
is is to be, the essential property of what is not is not to be; with regard
to what is not, it must be possible to legitimately claim that somehow
it is, namely that it is that which distinguishes not-being (predicative
sense), and hence that it somehow exists (existential sense). Gorgias,
in other words, brings out the tension created by the equivalence (or
lack of distinction) between the predicative and existential meanings
of the verb ‘to be’. This lack of distinction also explains the two follow-
ing passages: whereas in the first consequence the focus is on the

29
Palmer 1999: 72.
30
See Striker 1996: 12. This reasoning is also referred to by Plato in the Parmenides (162a–b:
see Mansfeld 1985: 258–62; Palmer 1999: 109–17) and Aristotle in the Metaphysics (see 1003b10:
‘It is for this reason that we say even of non-being that it is non-being’).

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


30 II BEING AND TRUTH, HUMANITY AND REALITY

existential meaning (if not-being is, being, its opposite, is not), in the
second one the focus is on the predicative meaning (being is not-being:
if being is identical to not-being, then being too is not-being, and hence
is not). Certainly, this undue transition from one meaning of the verb
‘to be’ to the other risks to invalidate the formal correctness of
Gorgias’ refutation. But Gorgias might have objected to this that
what he was doing was merely reproducing an ambiguity already to
be found in Parmenides, simply applying it to not-being rather than
being.31 In this respect at least, insofar as he successfully combined
being and not-being, Gorgias could have effectively defended his thesis
that nothing is: for Parmenides himself had observed that the equating
of being and not-being ‘simply reduces both to the negative level of
non-existence’.32
In the ‘dialectical argument’, Gorgias targets the speculative activ-
ities of all the Presocratics, and not just the theses of the Eleatics. Yet
in this case, too, the Eleatics are Gorgias’ privileged point of reference,
as may be inferred from the discussion on the generated or ungenerated
character of being, which probably represents the most successful
example of Gorgias’ way of arguing.33 Gorgias accepts the arguments
put forward by the Eleatic school regarding the impossibility for
being to be generated, and adopts the thesis that being could neither
be generated from what is (for, in changing, what is would no longer
be what is), nor from what is not (for nothing can be generated from
what is not: nihil ex nihilo). Hence, being is not generated. The prob-
lem, however, is that being is not ungenerated either, contrary to
what the Eleatics liked to claim: Gorgias refutes the Eleatics on
the basis of their very own theses, by leading the arguments that
Melissus and Zeno had produced into contradiction. According to
Melissus, if being were ungenerated, it would also have to be infinite.
This leap from the temporal level to the spatial one might seem

31
Calogero 1932: 197.
32
Ioli 2010: 31; cf. 28B6, 8–9 D.-K. = Parm. D8, 8–9 L.-M. Another objection to Gorgias’ rea-
soning is put forth by the anonymous author of M.X.G., who observes that Gorgias’ thesis also
implies the opposite conclusion: given that being and not-being coincide, one might conclude
not that nothing is, but that everything is. Yet, once again, the anti-Parmenidean polemical back-
drop is enough to show that this is not a real objection at all for Gorgias, given that his primary aim
is to conflate being and not-being: for Parmenides, claiming that everything is – both what is and
what is not – is no more acceptable than the thesis that nothing is.
33
On Gorgias’ argumentative strategy, which often arranges his opponents’ theses into con-
trasting pairs or multiple sets of arguments (e.g. Encomium of Helen, 13), see Mansfeld 1985
and 1986 and the observations made in Chapter 3, pp. 57–62.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


II BEING and TRUTH, HUMANITY AND REALITY 31

questionable to a modern reader, but it was explicitly upheld by


Melissus (30B2 D.-K. = Mel. D3 L.-M.) on the basis of the assumption
that generating something means giving it form, determining it; con-
versely, being ungenerated means being formless, indeterminate –
and hence infinite. Once this has been granted, it is easy for Gorgias
to refute the Eleatics’ thesis by invoking Parmenides’ other great
pupil: Zeno (29A24 D.-K. = Zen. D13a L.-M.) allows us to argue
that if being were infinite it would be nowhere, since to be somewhere
is to be contained by something and this is impossible for what is infin-
ite. Hence, if being were ungenerated, it would be nowhere and there-
fore would not be at all. But if being is neither generated (as many
Presocratics liked to think) nor ungenerated (as the Eleatics argued),
then it is not: nothing is.34
The flow of Gorgias’ texts always produces a baffling effect on the
reader/listener. After having demonstrated that nothing is/exists,
Gorgias admits that something might be/exist after all: but, even if it
did, it still would not be knowable. This is the aim of the second thesis.
Regrettably, the text of the anonymous M.X.G. is very corrupt here and
it is almost impossible to come up with a definitive reconstruction. One
plausible solution features the intertwining of two arguments.35 The
first is a paradoxical argument, which sets out once again from a
Parmenidean premise, namely the close identity between being and
thought (28B3, B8, 38 DK = Parm. D6, D8, 38 LM): if someone thinks
of something (if he or she has something in mind), that thing is; but this
means that if someone thinks of a chariot flying across the sea, then
there is a chariot flying across the sea. As this is patently absurd, the
conclusion is that external reality does not always coincide with what
we have in mind:36 the problem is precisely to distinguish between

34
Briefly put, Gorgias’ argument would run as follows: if P, then X or Y; but neither X nor Y;
hence, not P either. A similar argumentative structure also underpins the second discussion, focus-
ing on the issue of whether being is one or many: if being were one, it would be bodiless; but what
is bodiless has no magnitude, and with no magnitude it would be nothing, as Zeno observed. On
the other hand, if there is no unity, there cannot be any multiplicity either, since this depends on
unity. Therefore, if being is neither one nor many, it is not: nothing is. In M.X.G. we then find a
discussion on movement, but the fragmentary state of the text does not allow us to determine
whether it constituted a further development of the reflection on unity/multiplicity or whether it
instead marked the beginning of the discussion on the motion/rest pair. In support of the first
(and more reasonable) hypothesis, see Ioli 2010: 46–8.
35
Caston 2002.
36
This point might also be raised against Protagoras, who had upheld a sort of infallibilism by
denying the possibility of error: see Di Benedetto 1955 (according to whom Protagoras is actually
the chief polemical target of the whole treatise); Mansfeld 1985: 249–58; Caston 2002: 217–18;
Ioli 2010: 50–60.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


32 II BEING AND TRUTH, HUMANITY AND REALITY

what is true and what is false, which is to say to choose one mental state
over another. This marks the beginning of the second argument, which
introduces sense-perceptions by assimilating them to judgements (the
introduction of sense-perceptions once again shows that Gorgias did
not have the Eleatics alone in mind; rather, this second part of the rea-
soning seems to run against ordinary assumptions about knowledge).
For, ultimately, sense-perceptions also concern thoughts: if I see a
thing, that thing is in my mind. But, as has just been noted, if a
thing is in my mind, that thing is, which leads to absurd results;
hence, not even sense-perceptions are always reliable. However, if nei-
ther judgements nor sense-perceptions correctly represent reality, we
are no longer capable of grasping the truth – we no longer have any
way of understanding which of the things we think truly exist. The
truth might exist and it might even be thinkable, but we will never
know whether we have thought it. And this is tantamount to saying
that, even if being is, it is unknowable.
In the transition from the second to the third thesis, Gorgias con-
cedes what he had first denied: after having established that being is
unknowable, he allows that it might be knowable; but even if it might
be knowable, it cannot be communicated to other people. Besides,
we may grant that reality exists, and that we know reality, but we cannot
grant that we can communicate our knowledge. This, as we shall see, is
the most important point for Gorgias, a point that is never disputed in
the treatise. He appears to adduce three arguments in defence of the
third thesis (in this case, too, the text is corrupt – the most convincing
analysis remains that of Mourelatos 1985). The first argument rests on
the assumption of the heterogeneity of logos (understood as thought and
discourse) with respect to reality. Just as sight does not see sound, and
hearing does not hear colours, so logos does not speak things, but
words: the only possible experience is direct, first-hand experience.
But this is off-limits for logos, which always represents a translation of
such experience, a failed translation insofar as it proves incapable of
taking the place of its object.37 The other two arguments merely reassert
the isolation of logos, showing that there is no way of confirming
whether what A has communicated has been received by B as A

37
An alternative reconstruction that is worth mentioning is the one suggested by Wardy 1996:
14–21, who justifies the impossibility of transmitting logos on the basis of a physicalist interpret-
ation of it: logoi are not symbols but physical objects; as such, they can only occur in one place
at any one moment; clearly, this makes the transmission of the same logos impossible.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


II BEING and TRUTH, HUMANITY AND REALITY 33

understands it; indeed, not even A can be sure of this, given the diver-
gence between his own perceptions and feelings at different times. An
unbridgeable gulf thus separates things from thoughts and words: even
assuming that being is knowable, it cannot be communicated.38
Now that we have reached the end of this analysis, it is time to evalu-
ate Gorgias’ treatise in all of its complexity. What are the aims of this
treatise and what is its underlying meaning? Certainly, the treatise con-
tains a fair dose of irony and provocation, both in the theses it defends
and in the way in which it defends them. The problem, then, is to
understand the meaning of these provocations. Scholars in the past
fluctuated between two radical interpretations of Gorgias: the first dis-
missed his text as a mere divertissement of no real philosophical inter-
est;39 the second interpretation emphasized its ‘nihilism’.40 It is all
too evident, however, that both theses are one-sided, and that they
fail to adequately account for the challenge launched by Gorgias. A cer-
tain playfulness, which frequently manifests itself in the author’s desire
to dazzle his public with bold arguments, no doubt constitutes a defin-
ing feature of Gorgias’ practice (and of the sophists’ more in general).41
Yet this does not justify the conclusion that there is nothing of philo-
sophical relevance in the treatise (and that the treatise is merely an
example of sophistry – of arguments that are only apparently true42).
On the contrary, we only need to think of the discussions on the verb
‘to be’ or of the problem of communicability to realize that Gorgias
is raising questions of the utmost importance, which all great philoso-
phers – from Plato to those of our own day – have felt compelled to
address.43 As for the ‘nihilist’ reading, even leaving aside the problem
of what ‘nihilist’ means, it is clear that this is a doubly selective inter-
pretation. First, it de facto only considers the first thesis – that nothing

38
Kerferd 1984: 218–21; Palmer 2009: 87–8.
39
See Gomperz 1912: 28.
40
One notable example of this interpretation is furnished by Diels 1884, which reconstructs
Gorgias’ spiritual history in three stages: after initially subscribing to Empedocles’ theses,
Gorgias discovered Eleatic dialectic, which marked the beginning of a ‘period of doubt, or rather
of despair’, culminating with the nihilism of the treatise on not-being; Gorgias then emerged from
this nihilism through a new interest in rhetoric. The nihilistic interpretation crops up again and
again in the literature (see the list in Caston 2002: 205 n. 1) and has recently been taken up
again by Hourcade 2006.
41
Consider too fragment 82B12 D.-K. = 32D18 L.-M.: ‘Gorgias said that we should destroy
our opponents’ seriousness by laughter, and their laughter by seriousness.’
42
See J. Robinson 1973.
43
Besides, as rightly noted by Kerferd 1955–6: 3, Gorgias’ treatise is just as playful as Plato’s
Parmenides (see Parm. 137b).

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


34 II BEING AND TRUTH, HUMANITY AND REALITY

is – while the problem which Gorgias is most concerned with is the


third, namely that which concerns the split between language and real-
ity. Secondly, even if we were only to take into consideration the first
part, Gorgias’ polemical target are the views of the Eleatics and the
other Presocratic thinkers: what Gorgias would be denying are meta-
physical entities such as Parmenides’ being or the atoms of
Leucippus and Democritus. But this hardly justifies the thesis that
Gorgias was a ‘nihilist’ (at most, one could argue that the supporters
of the theses criticized by Gorgias – Parmenides and the other
Presocratics – were nihilists).
What are more interesting, albeit not fully satisfactory, are two other
interpretations which have been put forward. The first one, the ‘sub-
jectivist’ or ‘tragic’ interpretation, was endorsed by Hegel (in his
Lectures on the History of Philosophy, 1832–3) and further developed
by scholars of the calibre of Eduard Zeller (fifth edition 1923),
Wilhelm Nestle (1922), and Mario Untersteiner (1954). The second
interpretation, which we might describe as the ‘empiricist’ interpret-
ation, comes from the British liberal milieu and was mostly upheld in
Britain by scholars ranging from George Grote (1864) to William
Guthrie (1971) and Eric Dodds (1973). Unlike the nihilist interpreta-
tions, the ‘tragic’ one quite rightly emphasizes Gorgias’ main aim,
namely the highlighting of the fracture between reality and human
beings: to paraphrase Untersteiner, there is a tragic antithesis between
our impressions and the world that surrounds us, a gap that cannot be
bridged because we can only rely on language and know reality through
language, and language is deceptive. The result is that each human
being is confined within the prison of his or her language and does
not entertain any rational (or rather ‘logical’, bearing in mind what
Gorgias means by logos) relationship with reality. As far as the experi-
ence of logos is concerned, the contradictory nature of reality is unsolv-
able (see also 82B26 D.-K. = 32D34 L.-M.). It is only through an
irrational act of volition that we can escape from this prison, so as to
adapt ourselves to the multiplicity of reality. The ‘empiricist’ interpret-
ation, by contrast, emphasizes another fundamental aspect of Gorgias’
treatise, namely the centrality of its anti-metaphysical polemic. The
result is a completely different picture of Gorgias, whose attack on
the Presocratics’ metaphysical doctrines is interpreted as an initial
defence of the empirical world of common sense. Gorgias thus emerges
as a leading exponent of the fifth-century Enlightenment, engaged in a
struggle to defend human rationality. Reality exists and we can know it

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


II BEING and TRUTH, HUMANITY AND REALITY 35

through our senses and via correct reasoning, with no need to yield to
the abstract truths of philosophy.
No doubt, these interpretations have the merit of identifying some of
the key aspects of Gorgias’ thought, from his interest in the problem of
language to his focus on logical arguments. However, both of them risk
assigning too much to Gorgias: for – and this is a crucial point – what
authorizes us to take everything that Gorgias says as his own personal
doctrine? On the contrary, what clearly emerges from a reading of his
treatise is a dialectical position:44 his arguments appear to reflect a
desire to show to what absurd conclusions the theses or beliefs of his
opponents lead, and not to convey any personal theories; Gorgias is
arguing against, not in defence of. In other words, his treatise constitutes
an apparently playful reflection on a series of far from playful problems
in human life. Yet, we are not justified in turning these reflections and
the arguments developed by Gorgias into a holistic theory, as the afore-
mentioned scholars have sought to do. Gorgias’ aim is to invite his
interlocutors to acknowledge the complexity of these problems, and
to challenge them to come up with the most satisfactory solutions.
We know nothing at all of his own answers, assuming that he had
any to offer: ‘There isn’t any difficulty in appreciating Gorgias’ argu-
ments, once we see them dialectically – once, that is, we stop thinking
that the only way to be serious is to be dogmatic.’45 It is essential to clar-
ify this point in order to appreciate Gorgias’ teaching. His whole reflec-
tion takes the form of an all-out attack against the foundationalist
claims of philosophy and common sense: what Gorgias questions is
not so much reality itself, as the foundationalist claim that reality
possesses some invariant structures and that humans are capable of
discerning and communicating such structures.46 Whether the target
is Parmenides or the ‘two-headed men’, the polemical aim is this
alleged correspondence between being, thought, and language, along
with everything which this thesis implies: (1) the belief that reality is
something ontologically, logically, and temporally independent of us;
(2) the idea that knowledge coincides with the objective apprehension

44
This is suggested, e.g., by Striker 1996: 11–14; Woodruff 1999: 305–6; Caston 2002: 207–8.
See also Wardy 1996: 9–24.
45
Caston 2002: 208.
46
The main champions of this thesis are Mourelatos 1985, Wardy 1996, and Consigny 2001:
60–73 et passim. Cassin 1995 adopts a similar position, with a markedly idiosyncratic reading.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


36 II BEING AND TRUTH, HUMANITY AND REALITY

of this independent structure; and (3) trust in the fact that the purpose
of language is simply to convey this knowledge.
But – Gorgias asks – are things really so? Can we really speak of an iso-
lated reality, removed from the contingencies of human culture and lan-
guage? And even assuming that an independent reality exists, does
knowledge simply coincide with the correspondence of thought to things?
Rather, is it not the case that our opinions, prejudices, expectations, and
desires condition our approach to reality? Ultimately, these are the pro-
blems that Gorgias is most concerned with. They are substantial problems.
Finally, the anti-foundationalist interpretation just outlined has the
merit of explaining what is probably Gorgias’ chief contribution,
namely his acknowledgement of the importance of language in
human experience. As already noted, his whole treatise highlights the
complexity of the linguistic phenomenon, and this awareness – as we
shall see – also finds significant parallels in the other surviving works
by the same author. What makes Gorgias’ reflections so interesting is
the fact that he freed language (and logos more generally) from all onto-
logical implications, all alleged metaphysical correspondences with
reality: logos is not a reflection of things or the natural means by
which to objectively and impartially describe reality.47 On the contrary,
Gorgias asserts the independence of discourse, with its limits and
potential. As we shall see, a more in-depth investigation of this problem
reveals that his theses constitute a genuine challenge to philosophy.

Two testimonies in the wake of Gorgias’ discussion: Xeniades and


Lycophron

Gorgias’ theses may fruitfully be compared with the thought of another


two sophists, Xeniades and Lycophron. Regrettably, little is known
about them. According to the sceptic Sextus Empiricus, Xeniades
claimed that ‘all things are false, that every representation and opinion
is false, that everything that comes to be comes to be out of what-is-not,
and everything that perishes perishes into what is not’ (81.1 D.-K. =
39D1 L.-M.). The emphasis on the theme of not-being and on the lim-
its of human knowledge constitutes a polemical allusion to
Parmenides48 and is reminiscent of Gorgias, if only in a distorted

47
Segal 1962: 110.
48
Brunschwig 2002.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


II BEING and TRUTH, HUMANITY AND REALITY 37

way, since it misinterprets the latter’s underlying intentions: whereas


for Gorgias the analysis of reality is preliminary to that of knowledge
and language, here the movement would appear to be in the opposite
direction and the outcome is a flight into nothingness. However, it is
easy to dismiss this radical position: if nothing is and everything is
false, then even the thesis according to which nothing is and everything
is false is false.49
More interesting is the case of Lycophron, whom Aristotle in some
cases associates with Gorgias and Alcidamas. In a passage of the
Physics discussing the problematic relation between unity and multipli-
city, Aristotle writes:
Among the ancient thinkers, those who were later [sc. than Parmenides and Melissus]
were also troubled by the question of knowing how to avoid that the same thing be at
the same time one and many of them. That is why some of them suppressed the word
‘is’, like Lycophron, whereas others changed the mode of expression and said ‘the man
has been whitened’ instead of ‘is white’. . .in order not to make the one be many by add-
ing the word ‘is’. (83.2 D.-K. = 38D1 L.-M.)

Lycophron’s aim was apparently to offer a solution to the problem of


being, which had been the underlying focus of Gorgias’ polemic against
Parmenides. In fact, Lycophron’s suggestion is less radical than what
the above passage seems to suggest: a comparison with other, later tes-
timonies from the Neoplatonist commentators of Aristotle shows that,
in order to resolve the ambiguities associated with the double use of the
verb ‘to be’, Lycophron suggested that we abolish only the predicative
meaning of the verb, and not the existential.50 ‘To be’ is only used in
the sense of ‘to exist’, while its copulative use is replaced by concrete
verbs. This thesis, however, is not an isolated claim, but may be traced
back to an empiricist theory of knowledge. Elsewhere, Aristotle reports
that, according to Lycophron, knowledge is constituted of the ‘commu-
nion’ of the act of knowing and the soul (83.1 D.-K. = 38D2 L.-M.):
knowledge is not limited to the statement that Socrates is white, but
rather extends to the concrete, first-hand experience of Socrates’ white-
ness.51 Although this thesis, too, does not necessarily correspond with

49
Pradeau 2009b: 327.
50
See Themistius, Paraphrase of Metaphysics, 6.25–7.2; Philoponus, Commentary on Physics,
43.9–13; Simplicius, Commentary on Physics, 93.29–30 (these passages are not included in the
D.-K. edition). It seems that a passage from Plato’s Sophist (251b) can also be traced back to dis-
cussions of this kind, albeit not necessarily to Lycophron.
51
Bonazzi and Pradeau 2009: 335.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


38 II BEING AND TRUTH, HUMANITY AND REALITY

Gorgias’ positions, it may be read as following in the wake of an


anti-Eleatic polemic. It also shows some interesting points of conver-
gence with the ideas of Antisthenes, who, while famously close to
Socrates, had been a pupil of Gorgias’ as well.52 Be that as it may, with-
out going too far in the interpretation of such fragmentary testimonies,
it is important to note that, for Lycophron, the discussion on reality and
being leads to the problem of language: unlike Xeniades, he appears to
have grasped the meaning of Gorgias’ lesson.

Reality according to Antiphon

Antiphon was certainly among the most prominent sophists, and argu-
ably one of the most fascinating. However, two obstacles have ham-
pered the full appreciation of his merits, unjustly removing him from
the centre stage. The first is the problem of his identity, which has
added to the complexity of interpreting his ethical and political frag-
ments. A second and just as insidious hindrance has been the prejudice
according to which the sophists were only interested in the human
world, which is to say rhetorical and political problems. Thus the
many fragments of Antiphon’s writings that deal with cosmological,
biological, meteorological, geometrical, and medical topics have been
neglected, in the belief that they constitute a confused mass of informa-
tion collected from disparate sources. No doubt, Antiphon’s analyses
are largely indebted to the research conducted by natural philosophers
such as Anaxagoras and Democritus or by the Hippocratic physicians.
However, the conclusions drawn by Antiphon are far from obvious or
banal: on the contrary, the picture of physis, of reality, that emerges
from his writing is highly original and, as we shall see, constitutes the
foundation for even more interesting political considerations. The
case of Antiphon also shows that there is no radical break between
the sophists and the Presocratic natural philosophers: in their analyses,
the sophists did not overlook their predecessors’ theses.
Antiphon’s main work was entitled Truth. This title was not ran-
domly chosen, since it perfectly reflects the context in which its author
sought to operate. The title is also significant in relation to Protagoras,
since Antiphon’s theses, particularly in the political sphere, are very
much indebted to and react against the work of the sophist of

52
See Diogenes Laertius, 6.1; Guthrie 1971: 216–18.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


II BEING and TRUTH, HUMANITY AND REALITY 39

Abdera. One focus of the treatise – perhaps the chief focus – was physis,
which was investigated from a wide range of perspectives. The most
stimulating, and best-known, section of the work concerns human
nature; but the author’s analysis of humankind was based on a more
general analysis of reality, fragments of which also survive.
The following testimony from Aristotle helps clarify the possible
meaning of Antiphon’s position:
Some people think that the nature and substance of the things that are by nature is what
is present first of all in each thing, without possessing configuration (arrhuthmistos) in
itself, as the nature of a bed is the wood, and that of a statue the bronze. Antiphon
says that evidence of this is the fact that if one were to bury a bed and the rotting
could acquire the power of sending up a shoot, it would not become a bed, but
wood, which shows that the arrangement in accordance with the rules of the art is
merely an accidental attribute, whereas the substance is the other, which, further, per-
sists continuously through the process.
(87B15 D.-K. = partially reproduced in 37D8 L.-M.)

The above claim is far less banal than it seems. The passage establishes
an opposition between what – using the Aristotelian terminology – we
might call accidental characteristics, and what truly counts, namely the
essence, that which subsists continuously. This opposition between
convention and reality, couched in terms of the opposition between
nomos and physis, was to play a decisive role in Antiphon’s thought.
Here it is interesting to note the way in which physis, understood as
the ultimate material constituent, is presented: according to
Aristotle’s testimony, the term which Antiphon used to describe nature
is arrythmiston, ‘without configuration’, a rare adjective that may be
traced back to the Democritean tradition. Physis, then, is the unconfi-
gured material substrate that resists all external attempts to lend it
order.53 Other fragments too – unfortunately, very short ones – confirm
this picture (87B10, 12, 22–5 D.-K. = 37D9, 37, 18–21 L.-M.), sug-
gesting that Antiphon championed an organic, dynamic, and material-
ist view of reality. Physis is something living and ever-changing that
obeys its own rhythm, not one imposed from the outside: the natural
constituents of the universe (such as the sun and planets) are the prod-
uct of a chain of causal relations which do not stem from any intelligent
or finalistic plan, but only from the accidental combination of the ele-
ments (87B28–32 D.-K. = 37D24–7 L.-M.).

53
See Romeyer-Dherbey 1985: 96–103.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


40 II BEING AND TRUTH, HUMANITY AND REALITY

The implications of these views are far from negligible, when exam-
ined against the background of their cultural context: Antiphon’s
defence of the idea of a universe governed by necessity but not provi-
dence reflects an attempt on his part to clearly distance himself from
all mythical or religious explanations of reality. Moreover, he distances
himself not just from religious traditions but also from most of the
Presocratics, who – while emancipating themselves from the world of
myth – had assigned the ultimate principle or principles a series of posi-
tive attributes usually assigned to the gods: for Antiphon, the nature of
the all is not divine. Finally, from the opposite perspective, we cannot
rule out a criticism of Protagoras and Gorgias, who, by placing humans
at the centre of everything, had denied the priority of physis. Antiphon
succeeds in the delicate task of striking a middle ground, emphasizing
the centrality of nature while at the same time denying any axiological
superiority to this priority: his nature, true reality, is something neutral,
devoid of any meaning or value that might direct human choices.
Reality is indifferent: the importance of these theses will become
clear once the discussion focuses on humankind.54
Someone who had fully realized the significance of the kind of theses
put forward by Antiphon was Plato, who in the Laws – which is to say in
the last years of his life, at a fair distance from the debates and cultural
climate of the fifth century – felt the need to newly address the same
problems in order to decry their dangers. In Book 10, the Athenian
Stranger sets out to inform his two friends, the Spartan Megillus and
the Cretan Clinias, about those theses which were then circulating in
Athens concerning nature and the gods. Although the thinkers referred
to by Plato are unnamed, it is more than likely that, among others, he
was targeting Anaxagoras (who had denied the divine nature of the sun,
arguing that it was only a fiery rock – a thesis at the root of all problems:
886d–e), Democritus, and even Antiphon. The last, as has been sug-
gested, may in fact be Plato’s chief polemical target.55
The thesis put forward by the anonymous thinkers is that the universe
is the necessary outcome of the accidental combination of its material
constituents, which ‘moved at random, each impelled by virtue of its
own inherent properties’ (889b). While the combination of necessity
and chance, two apparently contrasting concepts, may seem odd at

54
See Gagarin 2002: 69–71.
55
Untersteiner 1949–62: iv.178–95; Decleva Caizzi 1986a; Neschke Hentske 1995: 140–9;
Bonazzi 2012. See also Sedley 2013.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


II BEING and TRUTH, HUMANITY AND REALITY 41

first, it reveals an interesting idea, which confirms the Antiphontic tone of


the passage. What is meant by chance (tyche) is not a blind or irrational
force, but an order of things which does not depend on transcendent
forces or intelligent designs, but on the intrinsic features of its constitu-
ents. Once the theory has been reconstructed in these terms (it is not
that far from modern theories, one may note), the reasons for Plato’s
opposition become clear: what we have is a first attempt to explain reality
in merely materialistic and mechanistic terms, an explanation that –
should it prove convincing – would entail the negation of the existence
of the divine (‘the gods exist. . .not by nature, but by certain legal conven-
tions’, 889e) and hence of all providentialism and finalism. This, in turn,
would amount to the negation of values such as justice and virtue, which
were traditionally associated with the divine order of reality. The result, in
one word, is atheism; and atheism, according to Plato, is the greatest ill of
all, on account of its political consequences: with the disappearance of
the gods, the values of justice and morality safeguarded by the gods
also vanish (‘there is no natural standard of justice at all. On the contrary,
men are always wrangling about their moral standards and altering
them’, 889e). The only remaining law, then, is that of violence: ‘every-
thing one can get away with by force is absolutely justified’; ‘the true
[note the importance of this term] natural life is essentially nothing but
a life of conquest over others, not one of service to your neighbour as
the law enjoins’; and this justifies those acts of impiety and subversion
which have driven Athens and the Greek world to the verge of the
abyss (890a–b; cf. 715a–716c).
As we shall see in the following chapters, it is most unlikely that
Antiphon advocated violence: the link between an independent con-
ception of nature, atheism, and political violence is something inferred
by Plato, who finds in this connection further confirmation of his view
that the sophists’ teaching is a social threat. Still, the fact remains that
Antiphon’s theses posed a radical challenge to intellectual traditions
that were well rooted in the Greek world. And, in this respect,
Antiphon proves himself to be a fine sophist.

From physis to logos: Presocratics and sophists

In the light of the analyses conducted thus far, it is possible to redefine


the sophists’ position vis-à-vis the Presocratic philosophers. No less
simplistic than the claim that the Presocratics focused only on the

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


42 II BEING AND TRUTH, HUMANITY AND REALITY

natural world and not on the human world is the thesis that the sophists
had no interest in physis and only focused on humanity. Rather, what
the sophists appear to have polemically engaged with is the fascination
with what is obscure, with the hidden nature of things.56 However, phy-
sis as such remains the starting point of all enquiries. To be more pre-
cise, reality remains the central problem that cannot be overlooked: the
common denominator in all these disparate theories, often developed
in polemical contrast to one another, is the acknowledgement of the
problematic nature of reality.57
And yet, the sophists are responsible for a fundamental turn in the
history of philosophy, a turn the importance of which can hardly be
ignored: a shift of focus from nature to the human world. Of course,
as has been noted, this does not mean that the Presocratics had not
focused on the human world; but they had done so starting from
their study of physis, by considering human beings as part of the
whole. The sophists adopted a completely new perspective, by setting
out from the belief that things must be turned round, that it is reductive
to consider humans simply as a part of the whole. On the contrary, in
their view it is necessary to set out from human beings and their
uniqueness with respect to other entities. In other words, the sophists
follow in the trail of the Presocratic philosophical tradition and con-
tinue to investigate problems related to nature, the city, and human-
kind, but following an opposite hierarchy, based on a decisive
philosophical intuition that no subsequent thinkers could afford to
ignore: the idea that the relation between reality and human beings is
problematic and cannot be taken for granted, but must rather be con-
structed. And it is on the basis of this acknowledgement that the soph-
ists discussed human beings and their problems: for if nature in itself is
ambiguous, neutral, or devoid of meaning, one must turn elsewhere in
order to discover the meaning of humanity’s experience and life. The
attention shifts from cosmology to epistemology, from the study of real-
ity in itself to that of how we can know reality and relate to it – from
physis to logos.

56
Woodruff 1999: 309.
57
Paci 1957: 126.

https://doi.org/10.1017/S0533245119000038 Published online by Cambridge University Press


III A WORLD OF WORDS: THE SOPHISTS AT THE
CROSSROADS BETWEEN GRAMMAR, RHETORIC, POETRY,
AND PHILOSOPHY

An interest in speech and an admiration for those capable of speaking well


was a recurrent feature of the Greek world from its most archaic period.
Contrary to a certain stereotyped image, the Homeric hero is not cele-
brated only for his strength and beauty; his ability to express himself is
also fundamental: Achilles is the most famous hero, and Phoenix’s duty
was to teach him how to be ‘both a speaker of words and a doer of
deeds’ (Iliad, 9.442–3). In times of war and peace, authority largely
depended on one’s skill in rhetoric, which allowed one to settle disputes
and provide suitable advice (see, for example, Hesiod, Theogony, 83–7).1
Later on, the development of the institution of the polis further increased
the importance of this skill, which became even more crucial in Athenian
democracy, in an age in which courtrooms and assemblies shaped men’s
lives and careers. Hence the sophists’ success, which reflects the fame they
enjoyed as masters of the art of speech: it is because he ‘makes you a clever
speaker’ that the young Hippocrates wishes to rush off to visit Protagoras
in Plato’s dialogue of the same name (Protagoras, 312d–e).
The sophists, in other words, are specialists in logos, in all the senses
encompassed by this term: logos may refer to the whole field of language
and words as much as that of arguments, thoughts, and mental pro-
cesses.2 The sophists asserted their competence in all of these domains,
and not just from a practical perspective; their reflection provided the
foundation for their investigation even from a theoretical point of
view: with the sophists, language and arguments became a privileged
matter of debate. Our sources inform us that they investigated these
problems from all angles, with a truly remarkable breadth of perspective
and competence. To be more precise, we can demarcate three specific
areas that bear witness to their interest in logos: grammar and the issue
of the correctness of names; the criticism of and engagement with
poetry; and rhetoric and the effectiveness of argumentative techniques.
In all these fields, the sophists earned indisputable prestige, as Plato
was to acknowledge in the Cratylus: to learn about these matters,

1
Pernot 2006: 15–22.
2
Gagarin 2008.

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


44 III A WORLD OF WORDS

Socrates observes, ‘the most correct way is to investigate together with


people who already know, but you must pay them well and show grati-
tude besides – those are the sophists’ (Cratylus, 391b).

Grammar and the correctness of names

Many sources bear witness to the sophists’ interest in grammar. This


applies especially to Protagoras, who focused on morphological, syn-
tactic, and stylistic questions: apparently, he was the first to distinguish
the gender of nouns (male, female, and neuter), while also suggesting
many corrections for names in use in his day. Thus he suggested
that the female nouns menis (‘wrath’, ‘frenzy’) and pelex (‘helmet’),
two terms familiar to Homer’s audience, should be regarded as mas-
culine – either on the basis of morphological criteria (because names
ending in sigma or xi are usually masculine) or because of their mean-
ing (insofar as war is an eminently masculine pursuit: see 80A27–8
D.-K. = 31D23–4 L.-M.). Protagoras also distinguished four verbal
modes (indicative, conjunctive, optative, and imperative), which he
linked to four types of speech (prayer, question, reply, and command),
once again taking the occasion to criticize Homer, who had addressed
the goddess with a command (‘Sing, Goddess, the wrath’) rather than
a prayer (80A29 D.-K. = 31D25 L.-M.). Ultimately, these grammat-
ical reflections presuppose a sophisticated conception of language,
which is not limited to names but is rather investigated from the
point of view of its underlying structure and connections, in relation
to the different functions of terms.3 As further confirmation of the
importance of this research, it should be added that Protagoras was
not the only sophist to deal with these problems: Hippias investigated
the rhythms, harmonies, and correctness of letters (86A12 D.-K. =
36D15 L.-M), and Alcidamas proposed an alternative division of
the types of speech from Protagoras’ (affirmation, negation, question,
and address: fr. 24 Patillon with Mazzara 2005). The coining of neo-
logisms can probably be traced back to the same context as well
(Antiphon and Critias appear to have been particularly keen on
inventing new terms, a remarkable number of which survive).
Naturally, it would be a serious mistake to assume that this focus on
language depended on exclusively erudite or grammatical interests, as

3
Brancacci 1996: 116–17.

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


III A WORLD OF WORDS 45

though the sophists simply sought to codify or standardize the linguistic


usages of the Greek world. On the contrary, practical and philosophical
interests were in play. On the one hand, it was a matter of increasing the
potential of words in view of the practical aim of teaching pupils to
exploit language to further their goals: it is by mastering a language
that one can use it more effectively.4 As Aristophanes was to note,
these are the tools that help ‘to make the weaker argument the stronger’
(80B6b D.-K.): in order to learn the ‘unjust argument’ (Aristophanes’
version of Protagoras’ ‘weaker argument’), one must first learn the
rules of grammar (Clouds, 658–91 = 80C3 D.-K. = Dram. T19c L.-M.).
On the other hand, it should be noted that the issues discussed by
the sophists raised some important theoretical questions. One recurrent
focus of their research was the issue of the correctness of names
(orthotes onomaton, orthoepeia5), which is to say the capacity of language
to faithfully portray reality: this kind of enquiry is what contributed to
early philosophical reflection on the nature of language.6 The correct-
ness of names no doubt represents one of the problems most widely
debated in the Greek world, and not just by sophists, either: let us
think here of figures such as Democritus and Antisthenes, not to men-
tion Plato’s Cratylus, whose subtitle was ‘On the Correctness of
Names’.
Among the sophists, the most interesting reflections were made by
Protagoras and Prodicus. The latter’s investigation of the correctness
of names took the form of a detailed analysis of synonyms, which
earned him great repute among the men of his day. By first grouping
synonyms together and then distinguishing them, Prodicus sought to
connect each name to its concrete reality.7 His theory presupposes a
one-to-one relation between words and their referents, such that the
phenomenon of synonymity seems only apparent. Prodicus thus
emerges as a forerunner of Socrates’ theories (significantly, the sources

4
Classen 1976: 223–5.
5
The distinction between these two expressions is unclear: see Guthrie 1971: 205.
6
Guthrie 1971: 220.
7
Momigliano 1930. It is difficult to determine on what basis Prodicus drew his distinctions: in
some cases he would appear to rely on the traditional use of terms (e.g. 84A18 D.-K., partially
reproduced in 34D24 L.-M.), while elsewhere he seems to suggest radical innovations based on
their etymology (84B4 D.-K. = 34D9 L.-M.; see Pfeiffer 1968: 40–1). The small number of testi-
monies makes it difficult to come up with a definite answer to these questions: see Classen 1976:
232–7. As rightly noted by Dorion 2009b: 531 n. 22 in relation to 84A16 D.-K., Prodicus also
investigated the problem of homonymy, which is to say the phenomenon of the semantic ambiguity
of a term (the term in this particular case being manthano, which in Greek means both ‘to under-
stand’ and ‘to learn’).

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


46 III A WORLD OF WORDS

often associate the two8), the main difference being that, whereas
Prodicus appear to proceed by asking ‘how does x differ from y’,
Socrates focuses more directly on individual entities, asking ‘what is
x?’9 In Prodicus’ case, this interest in names and synonyms has
favoured a particular interpretation of his thought, which has empha-
sized his specificity compared to other sophists. It has been noted
that his distinctions usually (but not always: see 84B4 D.-K. = 34D9
L.-M.) refer to terms and concepts pertaining to the field of ethics or
moral psychology.10 This has led some scholars to set Prodicus in con-
trast ‘to people the likes of Callicles and Thrasymachus’, as an oppon-
ent of the relativism typical of these authors and the upholder of a
certain foundation for the moral principles that are to govern people’s
lives.11 Provided that we bear in mind the rhetorical purposes of this
kind of research in Prodicus’ case as well, the reconstruction in ques-
tion is largely valid (we will be returning to it in Chapter 5).12
Prodicus’ positive ethical orientation, however, does not entail such a
radical contrast with the other sophists, as may be inferred from a com-
parison with Protagoras.
Plato’s Phaedrus suggests that one of Protagoras’ works was entitled
Orthoepeia (The Correctness of Language; Phdr. 267c = 80A26 D.-K. =
31D22a L.-M.). Regrettably, the content of this book is completely
unknown; but different sources confirm that Protagoras had an interest
in the problem of the correctness of language at different levels.13
Mention has already been made of Protagoras’ grammatical interests
and criticism of Homer, which implied the thesis of linguistic correct-
ness. On another level, the notion of orthos is used as a criterion of ana-
lysis for poetry: correctness in this case concerns not points of
grammar, but the consistency of a composition, its statements, and
its moral teaching (80A25 D.-K. = 31D31 and 42 L.-M.; we will be
returning to this testimony in the following section). Finally, the
same notion could also be applied in an even more general sense:

8
See Guthrie 1971: 275.
9
Classen 1976: 232.
10
Dumont 1986; Wolfsdorf 2008b.
11
In this respect, it is interesting to note that, in the Euthydemus, Plato mentions Prodicus twice
as a potential opponent of sophists and eristic debaters such as Euthydemus and Dionysodorus:
see Euth. 277e and 305c.
12
Cole 1991: 100.
13
Gagarin 2008: 28–30. More in general, see Rademaker 2013.

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


III A WORLD OF WORDS 47

When a competitor in the pentathlon unintentionally struck Epitimus of Pharsalus with


a javelin and killed him, he [i.e. Pericles] spent a whole day with Protagoras examining
the difficulty whether, according to the most correct reasoning, it was the javelin, or the
man who threw it, or the umpires, that should be considered responsible for this unfor-
tunate event.
(80A10 D.-K. = 31D30 L.-M.)

This testimony is a fine example of Protagoras’ way of reasoning. The


facts are indisputable: a man has unintentionally killed another man.14
However, much remains to be said in regard to the issues of moral
responsibility, legal guilt, and how to judge the whole incident. One
and the same indisputable fact may be viewed from many different per-
spectives: for the physician, the javelin will be responsible for the man’s
death; for the judge, it will be the javelin-thrower; for the person who
has organized the competition, it will be the judge. This contrast
gives the sophist some room for action: he will attempt to create
some agreement, lending meaning and order to the event.15 The notion
of correctness is the criterion which enables him to overcome the above
difficulties: given that there is some truth and validity to all the above
points of view, the problem will be to find the one most suited to the
situation, while foregoing any claim to come up with a single, abstract
answer.16
The notion of ‘correctness’ finds a prominent place in Protagoras’
thought, and allows us to grasp some of its strands of continuity.17 At
the basis of this notion lies the awareness of the problematic relation-
ship between reality and the human world which we discussed in con-
nection with the ‘man-measure’ thesis, and which finds another
expression in the statement that ‘concerning every subject, there are

14
Significantly, Antiphon’s second Tetralogy discusses the same problem. See also Antiphon’s
fr. 87B44, 4.10 (= 37D38 L.-M.), where the criterion of ‘correct reasoning’ is used to establish
what causes pain and what pleasure. Another interesting occurrence of the criterion of correctness
is to be found in the Encomium of Helen, where Gorgias sets out ‘to say correctly what is necessary’
in order to preserve Helen’s honour (82B11, 2 D.-K. = 32D24 L.-M.).
15
Untersteiner 1954: 30–2 and 66.
16
Gagarin 2008: 30.
17
This doctrine also shows that Protagoras was not endorsing a conventionalist theory of
names, as one might well expect (see Pl. Crat. 391b–d and the observations rightly made by
Corradi 2006: 54–5). A conventionalist position is possibly to be found in Euthydemus and
Dionysodorus’ thesis, as presented in Plato’s Euthydemus, and has also been attributed to
Antiphon by Guthrie 1971: 204 on the basis of the pseudo-Hippocratic treatise On Art that
Diels published as an appendix to 87B1 D.-K. (= Med. T2 L.-M.). In this passage, however,
the author’s silence as regards his polemical aims makes it difficult to prove Guthrie’s thesis. In
the same period, in close proximity to the sophists, Democritus appears to have upheld a similarly
conventionalist thesis: see Barnes 1979: i.164–9.

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


48 III A WORLD OF WORDS

two arguments opposed to one another’ (80B6a D.-K. = 31D26


L.-M.). With regard to each thing there are two opposite logoi
(speeches, arguments, points of view) – A and not-A – precisely
because each experience is open to contrasting readings and interpreta-
tions. Logos, which expresses our capacity to think and speak, thus can-
not exhaust the complexity of reality.18 However, as already noted with
regard to the ‘man-measure’ thesis, contradictory points of view do not
imply confusion or the failure of human possibilities. Rather, the chal-
lenge that Protagoras takes up and the promise he makes is to be able to
(and teach others to) unravel this complexity by identifying which of
any two opposite logoi is the better, or rather the more correct, which
is to say the one better suited to the circumstance at hand: this is the
meaning of his well-known claim to ‘to make the weaker argument
the stronger’ (80B6b D.-K.). It is easy to discern the structural analysis
that links this notion of correctness to that of the type of usefulness we
have discussed in relation to the ‘man-measure’ fragment: in both cases
the problem is to find the best, most expedient relationship with things
in each circumstance. We here get a better idea of Protagoras’ teaching,
which centres on the concept of logos, understood both as the ability to
reason and the ability to express oneself (that is, as both thought and
language19). ‘Correctness’, in other words, is to be understood on
two levels, one conceptual and the other linguistic: correct reasoning,
which expresses the best possible solution, must find a counterpart in
formal correctness, which makes one’s speech persuasive and hence
allows one to gain the upper hand in each particular situation.20 The
best speech, therefore, is not the speech that is true but the one that
is correct, the speech best suited to the situation at hand and most cap-
able of outdoing others from a formal and logical perspective: an irre-
futable logos.21
The notion of ‘correct’ thus encapsulates the two underlying motiva-
tions of Protagoras’ thought: on the one hand, the concrete need to win
discussions and debates; on the other, a more profound reflection on

18
The same conception also explains the statement that ‘it is not possible to contradict’ (80A19
D.-K. = 31R 10 L.-M.): to the extent that each individual entertains a relationship with things
whose reality and truth cannot be disputed or contradicted, the opposition here is only apparent.
It is interesting to note that in a recently republished papyrus, the same thesis is also attributed to
Prodicus, only on the basis of different arguments: see PToura III 16, 9–18, reprinted in Bonazzi
2007: 261 (= 34R14 L.-M.).
19
See Corradi 2007b.
20
Classen 1976: 222–5.
21
Brancacci 2002b: 183–90.

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


III A WORLD OF WORDS 49

the human beings and the importance of logos, understood as the cap-
acity to reason and to express oneself – as thought and speech. An
increased awareness of the chasm between human beings and reality
goes hand in hand with a belief that, through logos, humans can become
the measure of all things. This is what Protagoras’ thesis ultimately
amounts to, and it is easy to see that Prodicus and the other sophists
had reached the same conclusion, by different routes.

Literary criticism and the critique of the poetic tradition

In the Platonic dialogue of the same name, Protagoras, when illustrat-


ing the aims of his teaching, assigns a particularly central place to
poetry:
I think. . .that for a man the most important part of education consists in being expert
concerning poems; and this means to be able to understand what is said correctly by the
poets and what is not. (Plato, Protagoras, 338e–339a = 80A25 D.-K. = 31D31 L.-M.)

An interest in language did not merely inspire grammatical investiga-


tions and pave the way for rhetoric: it also constituted a privileged
means to engage with the poetic tradition. Literary criticism first
emerged in this period, also thanks to the sophists’ contributions. Yet
this is not all. Poets had traditionally been regarded as educators and
as the custodians of the most genuine Greek tradition: poetry was a
treasure trove of useful knowledge, an encyclopaedia of ethics, politics,
and history that every good citizen was expected to assimilate as the
core of his education. The poet’s task was to preserve and transmit
the system of values on which the life of his community was founded.
To engage with poetry, therefore, was to engage with the tradition;
and this engagement was a fundamental part of the sophists’ teaching.
Generally speaking, the rhapsodes were the social figures entrusted with
the transmission of the poets’ knowledge. By declaiming the Homeric
poems, they contributed to spreading a whole cultural heritage, consist-
ing of norms and values. The rhapsodes did not merely recite verses:
they also commented on them, and in doing so, they imparted teach-
ings and instructions (see Plato, Ion, 530c–d). By developing different
reading strategies, which often went against the established ones, the
sophists presented themselves as the heirs to the rhapsodes, in some
cases even donning their garb (e.g. Gorgias and Hippias: see 82A9

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


50 III A WORLD OF WORDS

D.-K. = 32P18 L.-M.). In such a way, the sophists reinforced their


claim to be new teachers, educators capable of imparting teachings sui-
ted to the needs of the new world of the polis.22 The sophists’ strategy is
one of appropriation, where an engagement with traditional knowledge
represents the starting point of their attempt to acquire a dominant pos-
ition in the Athenian cultural scene.23
The aforementioned evidence from Protagoras is very clear in this
respect. In the dialogue, the above claim is followed by the reading of
a poem by Simonides, one of the great lyric poets of the Greek
world, with the declared aim of highlighting its incongruities and con-
tradictions with respect to questions such as virtue and the good. Once
again, the criterion used for this analysis is that of ‘correctness’.
Protagoras follows a specific method, that of literal interpretation,
which unfolds in three successive stages (understanding, analysing,
and giving account). This method brings together all the various
aspects of the research on language conducted by Protagoras:
[1] the study and examination of the poetic text governed by the criterion of orthotes; [2]
the analysis of the poetic text from a grammatical, verbal, and logical perspective; [3]
explanation and discussion of the text itself, to the point of turning it into the subject
matter of the dialectical conversation.24

Clearly, the purpose is not to provide a mere exegesis. Rather, it is to


critically discuss the text and fulfil an educational goal: literary criticism
is a useful intellectual exercise which enables the individual to

22
See Pfeiffer 1968: 16–17; Soverini 1998: 6–12. Very interesting reflections are also to be
found in Most 1986, who stresses the importance of the interpretation of literary texts as a distinct-
ive feature of the sophists. Indeed, the sophists’ penchant for the written word constitutes a distin-
guishing element with respect to the oral culture in which poets found themselves operating: see
again Pfeiffer 1968: 24–30. It is worth recalling the fact that several sophists were also the authors
of poetical works: this is the case with Hippias (86A12 and B1 D.-K. = 36D2 and D4 L.-M.),
Critias, and possibly Antiphon (see 87A6 a 9 D.-K. = 37P8 L.-M.).
23
Goldhill 1986: 222–43; Morgan 2000: 89–94.
24
Brancacci 1996: 111. The likely polemical target of this method of literal exegesis is the alle-
gorical exegesis developed by Theagenes of Rhegium in the sixth century BC and later taken up in
Athens by another great intellectual of the period, Anaxagoras of Clazomenae, and by his pupil
Metrodorus of Lampsacus (on these authors, see Rocca-Serra 1990 and Morgan 2000: 98–
101). Evidence of a polemic between Protagoras and the champions of the allegorical method is
possibly to be found in testimony 80A30 D.-K. (= 31D32 L.-M.), in which Protagoras focuses
on a theomachy, a theme dear to Homeric interpreters of the allegorical tradition (who were
wont to interpret theomachies as symbolizing the oppositions between natural elements, such as
hot and cold, or dry and moist): see Brancacci 1996: 118. On Protagoras and Homer see, more
recently, Capra 2005 and Corradi 2006: 56–63.

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


III A WORLD OF WORDS 51

grow familiar with the works of the poets and hence with traditional
values – values that the individual will then be free to engage with, by
either approving or rejecting them.25 Similar conclusions are suggested
by other sources I have mentioned, namely the (regrettably) very suc-
cinct texts in which Protagoras analyses and criticizes some Homeric
verses (80A28–30 D.-K. = 31D24–5, 30 L.-M.): by showing his ability
to discuss the great Homer, while at the same time taking the liberty of
criticizing him, Protagoras justifies his claim to be the new teacher that
the city needs.
The importance of an engagement with the poetic tradition finds
confirmation that is only apparently unexpected in the ‘man-measure’
fragment. In the previous chapter we saw how one of Protagoras’
polemical targets was Parmenides and the philosophical tradition of
the ‘teachers of the truth’. Yet these were not the only ‘teachers
of the truth’ attacked by the sophists, since much the same holds true
of the poets, who had often drawn upon the idea of ‘measure’ to assert
their importance: a poet – to quote some famous verses – is someone
who, by grace of the Muses, knows the ‘measure’ of loving wisdom
(Solon, fr. 1, 51–2 Gentili-Prato) and possesses the ‘measure’ of wis-
dom (Theognis, 873–6).26 A poet, in other words, is someone who,
by virtue of the divine protection he enjoys, is capable of speaking
the truth and distinguishing it from falsehood; he is the custodian of
the order of reality and this justifies his prominent role in society.27
Everything changes with Protagoras: the ‘man-measure’ thesis under-
mines the poets’ claim to truth, just as it does with the truth referred
to by philosophers such as Parmenides (who ultimately were operating
within the same context as the tradition of poetic lore: he too wrote in
verse). The truth is no longer guaranteed by gods and inspired poets,
since humans are now the measure of all things, each according to
their own perspective. And Protagoras is the new teacher, and the
only one who can help others find their bearings in the ambiguous
world that surrounds them, in which contrasting opinions take the
place of absolute truth and falsehood. We can thus appreciate the pro-
grammatic importance of Protagoras’ introductory speech at the begin-
ning of Plato’s Protagoras, where he proclaims himself to be the heir to
an authoritative, centuries-old tradition:

25
Morgan 2000: 94.
26
See Corradi 2007a.
27
Arrighetti 1998: xv–xxi; Detienne 2006: 113–24.

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


52 III A WORLD OF WORDS

I say that the sophistic art is ancient, but that those ancient men who practiced it,
because they feared the annoyance it caused, employed a screen and disguised it,
some using poetry, like Homer, Hesiod, and Simonides, and others initiatory rites
and oracles, the followers of Orpheus and Musaeus; and certain ones, I have heard,
under gymnastic too, like Iccus of Tarentum and another one, still alive, as much a
sophist as anyone: Herodicus of Selymbria, originally a Megarian colony. And music
was the screen employed by your fellow citizen Agathocles, a great sophist,
Pythoclides of Ceos, and many others.28
(Pl. Prot. 316d–317c)

This claim is not merely designed to place Protagoras under the aegis of
a well-rooted tradition; rather, it contributes to a more complex strategy
of appropriation, which, through an apparently slavish adherence,
brings about a radical reversal.
While Protagoras seems to focus entirely on the rational and ration-
ally analysable aspects of poetic language, Gorgias also shows an inter-
est in its psychagogic and creative aspects (without overlooking the
importance of rational arguments, as we will see in the following sec-
tion). The assumptions and aims, however, are the same: the centrality
of logos and the ambition to establish oneself as Greece’s true teacher.
In particular, Gorgias’ reflection is marked by an acknowledgement of
the power of words and by the notion of deception. In this respect, the
most interesting text is his declamation Encomium of Helen, which was
apparently composed to defend the memory of Homer’s famous hero-
ine, guilty of having fled with Paris, bringing about the Trojan War.
Among the various reasons that may have led Helen to flee to Troy,
Gorgias considers the arguments by which Paris might have persuaded
her, and this allows him to embark on a famous digression on the power
of logos and what constitutes it, namely words:
Speech (logos) is a great potentate that by means of an extremely tiny and entirely invis-
ible body performs the most divine deeds. For it is able to stop fear, to remove grief, to
instil joy, and to increase pity. (82B11, 8 D.-K. = 32D24, 8 L.-M.)

The acknowledgement of this power of words, a power that is also


magical and divine, leads Gorgias to introduce the concept of decep-
tion (apate), which lies at the basis of his aesthetics.29

28
Only part of this passage is included in the Diels-Kranz edition, as 80A5 D.-K. (the whole
text appears in Bonazzi 2009b as T6 and L.-M. as Soph. R11). On its importance, see
Brancacci 2002a.
29
See Rosenmeyer 1955; Verdenius 1981; Horky 2006.

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


III A WORLD OF WORDS 53

The notion of deceit can probably be traced back to Parmenides, and


certainly to the poetic tradition of earlier centuries.30 In Gorgias, how-
ever, it lacks the negative nuance that it possesses in Parmenides and
the poets. When Parmenides describes his cosmology as deceptive,
he is not saying that it is false or fallacious; rather, he is warning his
audience that what they are dealing with is still the world of appear-
ances and not that of true reality. Much the same holds true for the
poets. In Gorgias, by contrast, there is no longer any room for a ‘true
divine reality’ beyond the changing world of appearances: all that
remains is phenomena and the uncertain opinions of men. The decep-
tion stems from this precarious situation: every opinion – which is to say
every attempt to make sense of reality – is intrinsically deceitful, since it
is incapable of faithfully representing a reality that cannot be faithfully
represented. But it is evident that in this context deception loses all
negative connotations: for such is the human condition. And it is
from this situation that poetry and all other forms of art can set out
to achieve their goals.
Gorgias provides the first ‘aesthetic’ reflection on art and its power:
the aim of the ‘deception’ embodied by art is to charm the soul by rous-
ing feelings of pleasure, joy, or pain.31 Yet, as we have just seen, the trig-
gering of an emotional response is not an end in itself: it is a way to
know oneself better and build a relationship with reality, which, accord-
ing to Gorgias, is always ‘other’ with respect to us – a way of making
sense of ourselves and the things around us. Deception, in other
words, is to be fostered because it is what allows us to build a relation-
ship with the reality of things and the reality of our very own being: the
problem is not to attain a universal truth but to ‘enrich our universe,
expanding our knowledge and culture, which is to say our
self-understanding, our capacity to explain our own being’.32 With
Gorgias, what we have for the first time is an awareness of the value
of art, ‘a deception, in which the one who deceives is more just than
the one who does not deceive, and the one who is deceived is more
intelligent that the one who is not deceived’ (82B23 D-K. = 32D35

30
As regards Parmenides, see fr. 28B, 56 D.-K. (= Parm. 19D8.57 L.-M.) and Verdenius
1981: 124. On the poetic tradition, see de Romilly 1975: 1–22 and de Romilly 1973, who also
notes that this conception of poetry as something magical and illusionary (see e.g. 82B11,
9 D.-K. = 32D24, 9 L.-M.) might reflect an influence from Empedocles (whose disciple
Gorgias may have been: 82A3, 10 D.-K. = 32P4–5 L.-M., and Kerferd 1985).
31
Segal 1962: 124.
32
Casertano 2004: 83

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


54 III A WORLD OF WORDS

L.-M.).33 Not unlike Protagoras, however, Gorgias implements a subtle


strategy by appropriating traditional poetic lore, as poetry is nothing but
‘a speech (logos) that possess meter’ (82B11, 9 D.-K. = 32D24, 9
L.-M.): what matters, then, is logos and the ability to make suitable
use of it. The real object of the encomium is not Helen but logos.34
This justifies the subsuming of poetry within the broader genre of rhet-
oric, the art of logos which is the object of Gorgias’ teaching: like
Protagoras, Gorgias plays with tradition in order to appropriate it.35
The lore safeguarded by the poet has now been integrated in the wis-
dom of rhetors and sophists.
Among the other sophists, Hippias stands out for his interest in poetry
and traditional lore. We know that he dealt with the division and length
of syllables, probably in relation to metrical and rhythmic issues (see
86A2 and 12 D.-K. = 36D14b and 15 L.-M.). Moreover, he was well
known in antiquity for his ‘antiquarian’ interests: that is, for having gath-
ered and catalogued quotes from the great masters of past centuries –
poets such as Orpheus, Musaeus, Homer, and Hesiod, among others
(86B6 D.-K. = 36D22 L.-M.; Hippias’ interest in Homer is also docu-
mented elsewhere: see 86A10, 9, 18 D.-K. = 36D25, 26, 24 L.-M.).
Hippias’ immediate aim, of course, was to provide a series of quotes
that could be used in speeches and discussions. However, this work of
selection may also be seen to promote a more detached approach to
the tradition, which is viewed from a historical perspective and no longer
as the depository of unquestionable truths.
This work of critical revision of poetic lore finds further confirmation
in the method of memorization (mnemotechnique) that made Hippias
famous (86A2, 5a, 11–12, 16 D.-K. = 36D12–13 L.-M.). Up until
then, memory had served as the poet’s key tool: it had exercised a reli-
gious function that enabled him to get in touch with divine reality and
acquire knowledge of present, past, and future. With Hippias – and
indeed Simonides before him – memory becomes a ‘“secular” tech-
nique, a psychological faculty that each person exercises according to
well-defined rules, rules that are available to everyone’.36 This

33
This idea is also taken up in the Dissoi logoi, 89, 3, 10–12 D.-K. (= 40, 3, 10–12 L.-M.),
which quotes verses by the poets Cleoboulina and Aeschylus.
34
Segal 1962: 102; Poulakos 1983.
35
In Gorgias’ case, appropriation also entails an attempt to adapt the poetic style to the kind of
prose declamations typical of his oeuvre: see 82A29 D.-K. (= 32D21b L.-M.) and de Romilly
1975: 8–11.
36
Detienne 2006: 191.

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


III A WORLD OF WORDS 55

engenders a new attitude to time, regarded not as the ‘power of obliv-


ion’, but as the context in which human endeavours take place.37
Hippias thus contributes to the development of historical awareness,
and the only surviving authentic passage by him, the aforementioned
fragment B6, bears unquestionable witness to the pride that he took
in affirming the innovative nature of his teaching:
Of these [sc. probably: ancient opinions] some have doubtless been expressed by
Orpheus, others by Musaeus, to put it briefly, by each one in a different place, others
by Hesiod, others by Homer, others by the other poets; others in treatises; some by
Greeks, others by non-Greeks. But I myself have put together from out of all these
the ones that are most important and akin to one another, and on their basis I shall
compose the following new and variegated discourse. (86B6 D.-K. = 36D22 L.-M.)38

The case of Critias, in contrast, is a more problematic one. Probably


following Philostratus’ lead, Diels reckoned him among the sophists.
Certainly, Critias’ interest in antiquarian traditions and poetry finds a
parallel in the work of other sophists.39 However, his ideology seems
to follow a radically different direction, insofar as he apparently upholds
a return to tradition and poetry (a genre he practised extensively)
against the threats posed by the new rhetorical education (see Brisson
2009: 395).40 The differences here seem far more significant than the
points of convergence.41

Rhetoric, the sophists, and philosophy

As should be clear by now, the interest in logos is central to the sophists’


conceptualizations. In the previous section, we examined how the

37
Detienne 2006: 192.
38
Besides, it is worth noting that Hippias’ ‘antiquarian’ interests were not limited to poetic
quotes, since he also made lists of the winners at the Olympics, so as to establish a reliable chron-
ology of Greek history (86B3 D.-K. = 36D7 L.-M.), of the founding of cities, and of human
genealogies (86A2 and B2 D.-K. = 36D14b and D30 L.-M.), and of many other topics pertaining
to mythological, ethnographic, geographical, and philosophical traditions (86B6–9, 12 D.-K. =
36D22–3, 26–8 L.-M.). On Hippias’ pursuits as a polymath, see Brunschwig 1984; A. Patzer
1986; Pfeiffer 1968: 51–4; Mansfeld 1986; and Balaudé 2006.
39
Pfeiffer 1968: 54–5.
40
See Brisson 2009: 395.
41
Another text that might help us further clarify the nature of the sophists’ interest in literary
criticism and poetry is an anonymous papyrus (POxy. III 414) that Giuliano 1998 has hypothet-
ically assigned to Antiphon. In his collection of fragments of the sophists, moreover, Untersteiner
had published an anonymous treatise On Music, which deals with similar issues (Untersteiner
1949–62: iii.208–11).

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


56 III A WORLD OF WORDS

sophists used their new form of knowledge in relation to – and in com-


petition with – the traditional knowledge embodied by poetry. We can
now move on to analyse how this interest in logos relates to rhetoric,
which developed as an independent form of knowledge in the late
fifth and early fourth centuries BC. In turn, this will allow us to return
to the thorny question of the relation between the sophists and philoso-
phy, an issue which we began to address in the previous chapter.
The extent of the sophists’ contribution to rhetoric has been at the cen-
tre of a lively debate in recent scholarship. Despite the fact that several
sophists were credited with the authorship of textbooks (the so-called
logon technai), and that Gorgias, Antiphon, and Thrasymachus have trad-
itionally been regarded as the founding fathers of this discipline,42 some
modern scholars have noted that the surviving testimonies seem to sug-
gest that a genuine reflection on this new literary genre only emerged
with Plato and Aristotle.43 The problem is certainly an interesting one,
and it is difficult to take a side in the debate, given that the sources we
have do not allow us to clearly determine the extent to which the sophists
may have contributed to the development of theoretical problems (for
example, the classification of different rhetorical genres, such as the delib-
erative, epideictic, and judicial) or stylistics (for instance, the distinction
between high and low style). What is certain is that, although the sophists
were not the ‘official’ founders of rhetoric, they showed an interest in
logos and what related to it, bringing to the fore a series of questions
that later became the focus of the discipline of rhetoric. A more detailed
analysis of problems of this sort should probably be undertaken by scho-
lars of rhetoric. On this occasion, what I wish to emphasize is that we can
only shed light on certain aspects if we focus on the importance of the
theme of logos for the sophists. Only by doing so will we be able to
reach a more adequate understanding of them.
Let us start from an initial misunderstanding. Usually, the attempt to
confine the sophists to the field of rhetoric implies a contrast with phil-
osophy (as in the case of Plato and Aristotle). In turn, the distinction

42
The traditional reconstruction identified a first Sicilian stage, represented by two almost
unknown figures, Tisias and Corax. From Sicily, rhetoric would then have reached Athens thanks
to Gorgias (who famously visited Athens as an ambassador in 427 BC); in turn, Gorgias would have
influenced other sophists such as Antiphon (assuming, of course, that the rhetor and the sophist of
this name are one and the same person: see p. 128) and Thrasymachus. Among the modern cham-
pions of this view, see Kennedy 1963 and, more recently, Pernot 2006.
43
See esp. Cole 1991: 71–112, and Gagarin 2007. Besides, the very adjective rhetorike, which
has given us the term ‘rhetoric’, may have been coined by Plato: see Schiappa 1991: 40–9, along
with the reservations voiced by Pernot 2006: 34–5.

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


III A WORLD OF WORDS 57

between philosophy and rhetoric entails a contrast between different


argumentative strategies: whereas philosophy is seen to uphold the
ideal of rigorous reasoning, rhetoric is considered susceptible to the
charm of a deceptive persuasion which appeals to people’s feelings.
But if we consider Gorgias’ declamations or Antiphon’s writings (par-
ticularly his Tetralogies, clusters of speeches each containing two
speeches for the defence and two for the prosecution of a given
case), we will soon realize that this is a false opposition. For, while it
is true that the appeal to people’s feelings plays an important role in
the sophists’ strategies, it is equally true that their efforts are largely
focused on developing cogent and rigorous arguments.44
If we consider Gorgias and Antiphon, the two sophists about whom
we have most information, several types of argument may be found in
their texts: most notably, arguments from probability (or likelihood:
eikos),45 antinomy, induction from exemplary cases, reductio ad absur-
dum, and so-called apagoge (where the speaker explains all possibilities
in order to then criticize each of them: this appears to be Gorgias’
favourite strategy).46 We may note that one of the major consequences
of Gorgias’ and Antiphon’s interest in rhetoric is their considerable
effort to identify and develop many different types of argument – the
concrete object of the sophists’ teaching.47 The appeal to feelings is cer-
tainly important, but what is equally important is rational analysis.
Ultimately, the reason for this interest is the need to carry out investi-
gations in contexts where truth is not self-evident; and it would be

44
See, for example, Lloyd 1979: 79–86.
45
One variation of this argument is what we might call the ‘counter-probability’ argument: see
e.g. Antiphon, Tetral. 1.2.2.3 and 2.2.6. A classic example is the case of a fight between a weak
man and a strong one: in order to defend himself, the former argues that, being weak, it is unlikely
that he wished to pick a fight with someone stronger. In turn, the latter replies by turning this rea-
soning on its head: it is unlikely that he was the one to start the fight because, being the stronger, he
would immediately have been blamed for it. In other words, something is claimed to be unlikely
precisely because it is likely: see Aristotle, Rhetoric, 2.24.12 (this argument was apparently
‘invented’ by Corax). The sophists’ interest in the notion of ‘probability’ or ‘likelihood’, however,
does not justify the criticism levelled by Plato, who in Phaedrus, 267a, accuses the sophists of
choosing what is probable over what is true: for one only speaks of probability when the truth is
unclear, which is often the case (unfortunately), but not always. See Gagarin 1994.
46
The most complete analysis is provided by Spatharas 2001; see too Mazzara 1999 and Long
1984. On Antiphon, see the analysis by Gagarin 2007 (who quite rightly reacts to Solmsen 1931,
according to whom all of Antiphon’s orations were marked by the adoption of irrational argumen-
tative schemes, such as the use of oaths and ordeals, which were typical of the archaic age).
Thrasymachus was by contrast famous for his ability to play with the audience’s feelings; see
Macé 2008. For an overview, see also Tinsdale 2010.
47
Natali 1986; see also above, Chapter 1, n. 11.

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


58 III A WORLD OF WORDS

difficult to deny that this effort is irrelevant from a philosophical


perspective.
This focus on argumentative strategies, moreover, makes it possible
to rectify a second assumption, which seeks to reduce the sophists’
teaching to a simple transmission of practical advice designed to ensure
victory in an argument – as though achieving successful persuasion and
winning contests were the only things that really mattered. No doubt,
persuasion was crucial. But if we consider the surviving texts by
Gorgias and the other sophists, we soon realize that it was not only a
matter of persuading the listener: ultimately, it would be simplistic to
think that the aim of declamations such as the Encomium of Helen or
the Defence of Palamedes was simply to convince the audience of the
innocence of two mythological figures.
Let us take, for instance, the aforementioned case of the Encomium of
Helen, which Gorgias allegedly composed to defend the memory of
Homer’s celebrated heroine, guilty of fleeing with Paris and causing
the Trojan War. To absolve Helen, Gorgias lists the four possible rea-
sons that might have led her into Paris’ arms, and shows that in all four
cases she is not responsible: the responsibility would lie with the gods,
or with Paris’ force, or with the power of words, or with desire. Now,
the attempt to cover all possibilities – this text is obviously based on
the method of apagoge – clearly goes well beyond the need to persuade
someone of Helen’s innocence: to reach this objective, Gorgias only
needed to stress traditional topics such as the will of the gods or
Paris’ physical force. Instead, he rather quickly moves beyond these
topics, so as to focus on an analysis of the power of logos and eros.
Indeed, while in this text Gorgias states his intention of persuading
the public, his intention of eliciting pleasure is just as important.
This is the pleasure derived from his display of intelligence, from his
capacity to investigate the potential of language and human thought,
and from his bold attempt to revisit – and at times criticize – traditional
knowledge:

If Gorgias were really trying to persuade his audience of Helen’s innocence, he could
have chosen a different strategy with a much greater chance of success.48 But if his

48
For example, he could have exploited an alternative version of the myth, according to which
Helen never went to Troy (this is the version followed by the poet Stesichorus, among others: see
Plato, Phaedrus, 243a–b; see too Herodotus 2.113–20 and Euripides’ Helen). The argument that
Helen was innocent, despite the fact that she went to Troy, instead betrays a desire to provoke
the audience with a thesis that at first sight seems utterly implausible.

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


III A WORLD OF WORDS 59

purpose was rather to enhance his reputation and demonstrate his intellectual virtuos-
ity, then he may have felt that the more implausible the case appears initially, the more
chance he has to show his skill in arguing for it as well as to develop novel perspective
on the issues involved. . . .It does not matter whether anyone is persuaded of Helen’s
innocence; the important thing is that Gorgias’ arguments open up new ways in
which to think about language, emotion, causation, and responsibility. His case may
be shocking, even perverse; it may be completely unconvincing; but his logos remains
one of the most interesting and intellectually stimulating works of the sophistic
period.49

Besides, all of this does not apply to Gorgias alone. For instance, a
cursory reading of the Dissoi logoi is enough to show that the primary
aim of its anonymous author is not simply to persuade his audience,
for example when he notes that, in a race or war, victory is good for
the winner but bad for the loser: the purpose of a claim of this sort is
certainly not to persuade anyone that good and bad are the same
thing. From Protagoras onwards, antilogy became the typical scheme
of sophistic discussions, with opposing arguments being used as a
means to examine a question in all of its complexity and ambiguity.
When properly employed, this method could be of use for winning
arguments; yet it was just as useful as a means to discuss problematic
cases, investigate or develop different types of arguments, entertain
the public or capture its interest, and showcase one’s skills.50 We
here get to the very heart of the sophists’ conceptualization: the analysis
of logos served not just to develop rigorous and cogent arguments, but
also to investigate various other aspects of human experience.51
This helps to clear up a third misunderstanding. The sophists have
often been accused of being somewhat incoherent and unsystematic.
Now, if no philosophy can exist without a system – that is, without a sys-
tematic and organized exposition of all the problems under scrutiny – it
is legitimate to claim that there is no room for the sophists in this field.
But such a rigid conception hardly does any justice to the richness
of philosophy, which also includes enquiry and critical analysis as
constitutive elements. According to this second, and more reasonable,
definition, the sophists, too, have a place in philosophy, since, in order
for it to truly work, the analysis of logos cannot be reduced to an
investigation of argumentative techniques, but must also take into

49
Gagarin 2001: 285–6.
50
Gagarin 2001: 289.
51
See Solmsen 1975.

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


60 III A WORLD OF WORDS

consideration human reality and things in all their complexity. So it is


easy to realize that the development of rhetoric promotes a deeper
familiarity with human psychology and its underlying mechanisms.
Let us again consider the example of the Encomium of Helen. To bet-
ter understand the phenomenon of communication, Gorgias investi-
gates human physiology, emotional dynamics, and the power of
mechanisms of persuasion: his analyses de facto imply a ‘quasi-
physical’ conception of the soul, on which discourses exert a clear
impact, eliciting concrete reactions.52 In turn, this conception of the
soul has very provocative consequences when it comes to human
responsibility, or rather the lack thereof. For, if this is how the
human soul works, then, no matter from what perspective we approach
the problem, the conclusion will always be that Helen is innocent. But,
clearly, Helen is not the problem here, since what is true in her case also
applies to all human actions. So if Helen is innocent, no person is guilty
of his or her actions.53
This is obviously a provocative thesis, and it may well be that not
even Gorgias himself accepted it. Yet, through his arguments,
Gorgias has the merit of having raised thorny problems that call for a
more in-depth reflection on the concept (and existence) of responsibil-
ity and the idea of the subject that it assumes. Can an action performed
under the impulse of a desire be considered a truly responsible action?
And what about an action performed because of some external con-
straint? Thus far, the answer seems quite straightforward. But to
what extent is Helen responsible with respect to the war that followed
her flight? In a way, it was she who triggered it; but can she really be
held accountable for it? And if so, is she also guilty? The problem
becomes more complicated, and it is no coincidence that Protagoras
and Antiphon had dealt with similar questions, by asking themselves
who is really responsible for the death of a boy struck by a javelin
(see p. 47 above). On the other hand, we might note that the question
is not only interesting in itself, but is also so more generally in relation

52
Segal 1962; Long 2015: 97–103. Along with logos and eros, another phenomenon to which
Gorgias pays particular attention both in the Encomium of Helen and elsewhere (cf. 82B4
D.-K. =32D45a L.-M.) is sight. This has an intermediate function, so to speak, insofar as it transmits
purely physical stimuli to the soul, engendering emotional states such as fear and joy, which in turn elicit
certain behaviours according to the sequence: physical stimulus – emotional response – physical stimu-
lus. In addition to the crucial study of Segal 1962: 105–7, see Casertano 1986 and Ioli 2010: 56–60,
exploring a possible ‘sophistic theory of perception’.
53
See Barnes 1979: ii.524–30; Tordesillas 2008.

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


III A WORLD OF WORDS 61

to human freedom, which now seems to have been obliterated by the


force of human passions and the power of chance. Generally speaking,
the sophists are not concerned with developing an exhaustive philo-
sophical system: it is difficult to determine what the position of some-
one like Protagoras may have been with regard to the problem of
determinism; and it is probably impossible to reconstruct a theory of
moral responsibility for Gorgias. Yet this does not mean that the pro-
blems which their reflections raise in each case are unimportant. And
it is precisely in this capacity to make crucial problems the centre
focus of the debate, bringing out many previously undetected tensions,
that the philosophical interest of the sophists lies.
Now that we have cleared up all these misunderstandings, having
confirmed the importance of logos and the rationalist assumptions
underlying the sophists’ commitment to rhetoric, we can return to
the problem of the philosophical significance of sophistic thought.
The link between rhetoric and philosophy might seem paradoxical at
first, given that traditionally – from Plato onwards – philosophy has
developed in opposition to rhetoric: the claim that it is necessary to
focus on rhetoric in order to understand the philosophical significance
of sophistic thought might seem like utter nonsense. Yet it is actually
true: the sophists’ interest in logos and rhetoric does not lead them out-
side philosophy, but towards a different conception of philosophy.
Their unsystematic approach, intellectual curiosity, and ambiguousness
towards the problem of truth are all based on the insight we discussed
in the previous chapter: the acknowledgement of the problematic
nature of the reality that surrounds us and the awareness of the fact
that logos is our only means to develop a relationship with reality.
Let us take the exemplary case of Gorgias once more. The celebra-
tion of logos that we find in the Encomium complements the thesis
regarding the rift between thought and language put forward in the
treatise On Not-Being. At first sight, this might seem problematic,
and one might object that the treatise On Not-Being and the
Encomium of Helen present mutually incompatible theses: whereas the
former ends with an acknowledgement of the failure of words, the latter
assigns the word a sort of divine omnipotence. Upon closer scrutiny,
however, it is possible to reconcile these two positions, if we accept,
on the one hand, that On Not-Being criticizes a certain conception of
logos, and not logos as such; and, on the other, that the two declamations
pick up from where the treatise had left off, so as to explore alternative
possibilities. In this respect, the two declamations may be seen to

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


62 III A WORLD OF WORDS

constitute the constructive side, so to speak, of a two-part work which


also comprised the deconstructionist treatise On Not-Being.54
The discussion on logos is therefore essential for assessing the dis-
tance which separates Gorgias and the other sophists from the trad-
itional conception of philosophy, the Parmenidean conception, which
was partly taken up by Plato and Aristotle. The sophists’ polemical tar-
get is the claim to establish logos as a foundation, based on the assump-
tion that being, thought, and reality coincide, whereby logos is thought
to express truth in the sense of a true and objective representation of
reality. But this is impossible, according to the sophists: reality is intrin-
sically multifarious and resists any unitary reconstruction; and because
human logos is always subjective or relative, it always represents a given
point of view or opinion, and not absolute truth. This is where the
sophists’ alternative conception comes into play: they assign logos the
function not of stating the truth, but of giving meaning to reality, of
turning multiplicity into some kind of order – a provisional order, yet
one still capable of orienting human actions. To paraphrase
Nietzsche (who appreciated the sophists precisely because of such the-
ses), it is a matter of interpreting things, since there are no facts, but
only interpretations. Alternatively, and more radically, it is possible to
draw a distinction between ontology and logology:

Ontology: discourse celebrates being, its function is to speak it. Logology: discourse
makes being, being is an effect of speaking. In the former case, the exterior imposes
itself, and requires that we speak it; in the latter case, discourse produces the exterior.55

Logos can prove successful because – as we have seen – despite its


apparent inconsistency, it is actually very powerful. Paradoxically,
Gorgias’ emphasis on the limits of human experience ultimately leads
to a celebration of the capacity of logos to give meaning to everything.

From logos to nomos, from language to politics

As repeatedly noted over the course of this chapter, the sophists can
truly make a claim to being specialists of logos. Clearly, this is such
an important term that it can hardly be made the exclusive preserve

54
Ioli 2010: 90. On the function of language in Gorgias, see also Calogero 1932: 262 and esp.
Mourelatos 1985: 627–30.
55
Cassin 1995: 73.

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


III A WORLD OF WORDS 63

of the sophists: the concept of logos, with all that it entails, was just as
important for the Presocratics, as well as for Plato, Aristotle, and all
other Greek philosophers. Still, the sophists’ contribution here remains
crucial. It was thanks to them that an awareness of the fact that logos is
the distinctive feature of humankind emerged. Whereas for many pre-
vious thinkers, logos is what enables us to get in touch with an objective
and well-ordered reality, since logos is also the ordering principle of
reality, for the sophists it is a tool that a human can – and must – use
to give meaning to things, a meaning that things do not necessarily pos-
sess in themselves. The problem, in other words, is to build a relation-
ship with an ever-elusive reality and to adapt to changing situations,
finding the right measure in each case. Thus the problem of logos is
the problem of kairos, the right moment: the problem of finding the
best possible solution in each circumstance and of knowing how to
express it.56
In themselves, when viewed in abstract terms, however, logos and
kairos are not enough to describe the sophists’ position. The attempt
to give meaning to reality cannot be confined to the theoretical level
of words, but must concretely manifest itself in the shaping of a
human world: in order to possess any real value, the sophists’ logos
must engage with the ambiguous world of politics, where nothing is
fixed and duplicity is the norm. It is hardly a coincidence that the
first people to be referred to as sophists were politicians such as
Mnesiphilus, who had helped Themistocles:
Who was neither an orator nor one of the so-called ‘natural philosophers’, but who,
practicing something that at that time was called ‘wisdom’ (sophia) but was really clev-
erness in political matters and pragmatic shrewdness, made this his profession and pre-
served it intact like a sect transmitted to him in succession from Solon. But those
people who followed him mixed this with forensic arts and transferred its sphere of

56
Scholars have long been debating the importance of the notion of kairos for the sophists – a
notion that was widespread in the archaic Greek world. The problem is twofold: first, to determine
whether the sophists developed a rhetorical theory of kairos (thereby acknowledging the import-
ance of improvisation); and second, to determine whether they also assigned kairos a more general
value, which might make it the notion that best expresses the specific nature of sophistic knowl-
edge. The few surviving sources allow us to give a positive answer to the first problem (in
82B13 D.K. = 32D12 L.-M., Gorgias is even credited with a kairou techne; cf. 82A1a, A24 and
B13 D.-K. [= 32D11–12 L.-M.]; 80A1 D.-K. [= 31D20DL.-M.] on Protagoras and
Alcidamas’ speech On Those Who Write Written Speeches, or On the Sophists). As regards the second
problem, the existence of a general doctrine of kairos (a ‘Kairos-Lehre’, to quote Guthrie 1971: 272
n. 4) seems like a projection on the part of modern scholars – a fascinating, yet at least partly ana-
chronistic, view; see Tortora 1985 and Tordesillas 1986.

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


64 III A WORLD OF WORDS

application from action to speeches, and they were called ‘sophists’.


(Plutarch, Life of Themistocles, 2 = Soph. R4 L.-M)57

As is well known – and as one would expect in the historical and cul-
tural context of the fifth century – the sophists extensively dealt with
practical and political matters. It is important to note that this was
not a self-contained interest, but the direct consequence of their enqui-
ries into physis and logos:
to put it in even more provocative terms, the blueprint for the sophists’ politics is pro-
vided by the treatise On Not-Being. . . .Ultimately, being is but a consequence of saying.
Given this, it is clear that the presence of Being, the immediacy of Nature, and the
evidence of a speech designed to adequately express these things, all vanish together:
the natural philosopher who discovers speech gives way to the politician who creates
discourse.58

The emphasis on the creative power of speech enables the sophist to


take the place of the philosopher who expresses being and of the poet
who claims to be the spokesman for divine truth; and this is precisely
how he comes to affirm his leading position in the world of the polis.
Along with physis and logos, a third concept is of crucial importance
in order to understand the sophists, specifically nomos, and it is this
concept that we must now focus on.

57
See also Detienne 2006: 205.
58
Cassin 1995: 152.

https://doi.org/10.1017/S053324511900004X Published online by Cambridge University Press


IV JUSTICE AND LAW

The fifth century is the century of nomos, a concept difficult to translate,


as it covers several different semantic fields. Usually nomos is rendered
as ‘law’, but this is a reductive translation, since the term describes not
just positive laws but also traditions, customs, and established practices.
Besides, as Aristotle was to observe, the two meanings ‘law’ and
‘tradition’ are closely related: ‘For the law has no power to command
obedience except that of habit, which can only be given by time’
(Politics, 1269a20). In short, nomos is anything which is assigned
some value, any norm accepted by a group. From the late sixth century
onwards, this term acquired a central place in the Greek world, becom-
ing a privileged object of reflection for the sophists. The most signifi-
cant element is its spread as a technical term for written law,
particularly in Athenian democracy.1 Prior to this, it chiefly referred
to customs and traditions. Yet its growing importance is not devoid
of problematic aspects. The greater economic and political stability
attained through the victorious wars against the Persians led to an
increase in travel, cultural exchanges, and trade, which contributed to
broadening the perspectives and knowledge of the Greeks. One of the
most striking observations they made concerned the wide range of
laws and traditions in force in the various cities and states: nomos is
indeed a crucial value, as the law and tradition governing the life of a
community; but it is in any case relative, as it varies from city to city.
The acknowledgement of the relativity of nomos is a central theme in
fifth-century literature, as may be inferred from some famous passages
by Pindar, Herodotus, and Aeschylus.2 At first sight, the conclusions
reached by these authors would appear to emphasize the capacity of
nomos to impose itself: as a verse by Pindar states, nomos ‘is the king
of all things’ (fr. 169a Snell-Mähler). Yet the ascertainment of the
range of laws and traditions in existence could also lead to moral rela-
tivism and to the conclusion that there are no absolute, universal or
objective values (Dissoi logoi, 90, 2.18 D.-K. = 40, 2.18 L.-M.).
Confronted with the problem of nomos, the sophists were divided:
some upheld its importance, while others sharply criticized its weak-
ness. In order to better articulate their polemic, the latter could also

1
Ostwald 1969.
2
De Romilly 2005: 49–66.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


66 IV JUSTICE AND LAW

rely on another key concept in Greek philosophical reflection, namely


physis, a term which is usually translated as ‘nature’, although a more
fitting translation for it would probably be ‘reality’, at least with refer-
ence to these debates. As already noted in Chapter 2, physis was an
important term for the reflection of naturalist thinkers, from Thales
to Anaxagoras, from Empedocles to Democritus: it stands for the nature
of things. In the fifth century, chiefly through the contribution of medical
knowledge, physis mostly came to be used to refer to human nature. In all
cases, whether it was used to refer to humans or to all things, the term
described the stable, invariable character of things or human beings.
The relation between nomos and physis is probably the topic most fiercely
debated by fifth-century authors:3 it offers a conceptual scheme that can
fruitfully be applied to a wide range of subjects, such as the legitimacy of
slavery, the equality or inequality of human beings, and the existence of
the gods (is the inequality of humans or the existence of the gods a nat-
ural fact or a cultural product?). Yet it is especially in relation to the
theme of justice and the law that the sophists exploited this pair of
terms in the most original way.4

Justice according to Protagoras

The first author to have extensively discussed justice was probably


Protagoras: indeed, according to the Peripatetic Aristoxenus, ‘the
Republic. . .was almost entirely written in Protagoras’ Antilogies’ (80B5
D.-K. = 31R1a L.-M.). Aristoxenus is often malicious towards Plato,
and his accusation of plagiarism is no doubt excessive. The fact remains
that Plato was very interested in Protagoras’ reflections on political pro-
blems, acknowledging their relation to ontological and epistemological
theses: the aforementioned evidence from the Theaetetus and Protagoras
suggests that the discussion of justice and nomos stemmed directly from
the thesis of relativism.5 Therefore, in order to grasp the meaning of
Protagoras’ analyses of justice and law, it is necessary to briefly return
to the points made in Chapter 2. In the Theaetetus, one of the most sig-
nificant consequences of the ‘man-measure’ thesis is the validity and
legitimacy of all opinions and judgements. In the Protagoras, this

3
Heinimann 1945.
4
Hoffmann 1997.
5
See now Gavray 2017.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


IV JUSTICE AND LAW 67

coincides with relativity and plurality of interests and use (80A22


D.-K., not in L.-M.). There are no absolute and objective values,
such as truth and goodness, to guide the judgements and choices of
humans (and similar considerations also apply to the issue of the
gods, regarding which Protagoras stood out for his agnostic position:
see Chapter 6). All these claims present some interesting theoretical
perspectives, but also some serious practical problems: in the absence
of any objective values or norms, the risk is that of a potential conflict
of everyone against everyone else, whereby each person will pursue
their own interest on the basis of what they consider to be good for
them. But while this is certainly a possibility, it is not the only one.
Protagoras believes that, unlike animals, human beings can reach an
agreement, thereby neutralizing these destructive drives. To promote
this agreement is the duty of the sophist, of the true politician, and of
any good citizen: it is a matter of laying down some communal rules
and values, so as to safeguard everyone’s interests by promoting mutual
collaboration.
This is what can be inferred from the relativism of the ‘man-measure’
thesis, which has significant practical consequences and lays the founda-
tions for the Protagorean conception of justice. In the so-called ‘Apology
of Protagoras’ in the Theaetetus, the sophist – who is explaining the correct
meaning of the ‘man-measure’ thesis to Socrates – states that ‘whatever
seems [or: is decreed to be, dokein] just and fine to each city also is that
for it, so long as it thinks [or: adopts this law, nomizein]’ (80A21a
D.-K. = 31D38 L.-M.). Consistently with what has been noted so far,
justice, not unlike truth or goodness, is not an objective, absolute, or div-
ine value, but the result of human decisions – a set of rules that humans
establish together. In practice, justice consists in nomos, understood both
as a body of written laws (the law code) and as a range of values and tra-
ditions (the moral code) on which a community is founded.
From a more strictly juridical perspective, with reference to written
law, we can argue that Protagoras offers the earliest attestation of
what was later to be described as ‘legal positivism’, which is to say
the thesis that law and justice coincide: the just is what is established
by law.6 In order to provide a complete overview of Protagoras’ theory,
however, we must take account of a third element, which is to say what
is useful. In the absence of any absolute good, what provides the

6
Neschke-Hentske 1995: 56–9.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


68 IV JUSTICE AND LAW

foundations for – and lends legitimacy to – the law (and hence justice)
is precisely the concept of what is advantageous and useful for the par-
ties who are striking an agreement. Naturally, in this case too we cannot
think of what is useful as an objective good: as people’s interests vary
depending on the situation, it is not something which one must con-
form to, but rather something to be constructed out of different points
of view. It is precisely this capacity to promote an agreement based on
what is useful that reveals the importance of the sophist or his pupil, the
politician.7 Their duty is to help the city establish rules and values,
which will then take the concrete form of a body of laws (nomoi) allow-
ing the city to prosper: ‘the clever and competent orators [i.e. those
who publicly discuss political matters, hence politicians and sophists]
make good things seem to be just to cities instead of bad ones’ (A21a
D.-K. = 31D38 L.-M.).8
Similar conclusions are suggested by an analysis of the famous myth
that Plato puts into Protagoras’ mouth in the other dialogue devoted to
the sophist, the Protagoras (80C1 D.-K. = 31D40 L.-M.). Some scho-
lars have questioned the reliability of this testimony, arguing that it is
due more to Plato’s genius than Protagoras’ pen. No doubt, this is a
reasonable observation, because it would be too much to assume that
Plato simply incorporated a long extract from an opponent’s work
into his own text. Yet the reasonableness of this observation does not
justify the opposite – and equally radical – hypothesis that the myth is
simply Plato’s own invention and has nothing to do with Protagoras.
This assumption, too, conflicts with Plato’s way of working, since –
as we have seen in relation to the ‘man-measure’ thesis, for instance
– Plato always sets out from his opponents’ theses in order to criticize
them. The myth (like the speech that follows it) certainly offers some
information on Protagoras’ thought; to what extent, it is difficult to
determine.9
The myth retraces the key stages in the history of humankind, start-
ing from the well-known story of Prometheus, the beneficent daemon.
When the time had come to generate mortal animals (including
humans), the gods entrusted Prometheus with assigning each species

7
See also Denyer 2013: 167.
It is worth noting once more that, obviously, the laws will apply ‘so long as the city thinks [or:
8

adopts this law, nomizein]’: as a human product, no law can claim to have absolute and everlasting
validity.
9
For an analysis of these problems, see Bonazzi 2011 and Manuwald 2013.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


IV JUSTICE AND LAW 69

qualities that would allow it to survive and prosper. Prometheus (liter-


ally, ‘he who understands first, who foresees’) left the task up to his
brother Epimetheus (‘he who understands afterwards’), but the latter
forgot humans: ‘while he was in perplexity, Prometheus came to
check the distribution, and he saw that the other animals were suitably
provided with everything, but that man was naked, unshod, uncovered,
and unarmed’ (Prot. 321c4–7). To make up for his brother’s mistake,
Prometheus stole fire and technical expertise from the gods, allowing
humans to approach the world of the gods, learn how to speak, and
master the technologies required to solve practical life problems – the
provision of food, clothing, and housing. However, despite this pro-
gress, humanity risked becoming extinct, as it lacked political wisdom
(‘for that was in Zeus’ possession’ [Prot. 321d5–6], in a realm to
which Prometheus – who was being punished for the theft – no longer
had access): only this wisdom would allow humans to organize them-
selves into social groups and live together, so as to protect themselves
from wild animals and natural dangers. Humans were trying to save
themselves by coming together in cities, but ‘when they gathered
together, they committed injustice against one another. . .so that they
scattered once again and were destroyed’ (Prot. 322b7–8). Finally, fear-
ing that the human race would meet extinction, Zeus despatched
Hermes to distribute justice (dike) and shame (aidos), not in the same
way as with the other forms of arts (whereby, for instance, one physician
is enough for many patients), but to everyone indiscriminately: ‘for cit-
ies would not exist if only a few men have a share in these things [shame
and justice], as in other arts’ (Prot. 323c8–d2). Such is Zeus’s law
(nomos).
The myth (and the long speech that follows it) helps clarify the
anthropological ideas that lie at the basis of the Protagorean conception
of nomos and justice. Even though it is articulated in three phases, the
myth must not be understood in a chronological sense, as though it
retraced the various stages in human civilization; rather, Protagoras’
account is meant to identify and circumscribe some essential features
of humankind itself: what we have is not a ‘genetic’ myth but a ‘struc-
tural’ one.10 The state of nature in which humans find themselves living
after their creation reflects an impossible situation, and may be under-
stood as a counter-factual example to demonstrate e contrario that

10
For an in-depth investigation of these topics and further bibliographical references, see Farrar
1988: 81–99; Bonazzi 2004; Balla 2018: 91–101.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


70 IV JUSTICE AND LAW

human beings are political and rational animals: they are political in the
sense that they cannot live in isolation, but need one another and are
forced to live together (‘political’ is to be understood etymologically
starting from the polis – the city, community, and state); and they are
rational because they have the means to engage with one another in
order to find possible solutions.
As we have seen, these solutions are only possible through the shar-
ing of collective values: in the myth, justice and shame are not a per-
manent possession of humankind, but rather predispositions that
must be actualized, which is to say that they must take the concrete
form of nomos, to ensure the development of a harmonious commu-
nity.11 In order to clarify Protagoras’ position, it is worth referring
again to the conceptual pair nomos/physis: for Protagoras, human physis,
unlike that of other animals, is not a given or something invariable;
rather, it can only be fully realized through nomos, which actualizes
the specific potential of humanity – a political potential.12 Whereas ani-
mals will always behave in the same way, humans can decide how to
comport themselves: whether to plunge back into the violence of the
animal world or to fashion a world of values for themselves. Nomos –
which is to say, the development of a political society – does not
stand in contrast to physis, but rather constitutes its ‘fulfilment’13 and
ensures the existence of individuals. Moreover, nomos is not only
what ensures survival, but also – and most importantly – the precondi-
tion for humans to fully express their potential. The individual – every
individual human being – is the product of society and attains fulfil-
ment in society: hence, his or her ultimate fulfilment depends on the
existence of a well-ordered society.
Again, this explains the social function of the sophist, a master of pol-
itical virtue: the sophist, just like the politician and any good citizen,
exercises a fundamental function for the community, insofar as he pro-
motes this collective awareness, thereby contributing to perfecting his
community, which can and must be improved.14 In other words, it is

11
Moreover, we should not overlook the fact that, while Zeus is the one who bestows these two
virtues on humanity, he does not concretely define what comprises them by making their content
explicit: in other words, justice is not a divine revelation, but something that humans must achieve
together. In this sense, the mythological framework does not conflict with the agnosticism of frag-
ment 80B4 D.-K. (= 31D10 L.-M.).
12
Later in the dialogue, Protagoras speaks of political virtue as human virtue (andros arete) tout
court (Prot. 325a); on this issue see also Beresford 2013.
13
Casertano 1971: 124.
14
Decleva Caizzi 1999: 319.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


IV JUSTICE AND LAW 71

necessary to promote a critical reflection on the community’s choices.


In its ideal representation, the city emerges as a permanent educational
apparatus,15 while the law, nomos, ultimately coincides with what is use-
ful for the individual and the community: respecting and favouring the
nomoi benefits everyone because it safeguards the community. The the-
ory developed by Protagoras thus seems capable of answering one of
the main problems in political philosophy, that of obligation, of why
it is necessary to obey the law.
Protagoras’ theses concerning justice and humankind certainly
represent some of the greatest achievements of the sophists, the import-
ance of which emerges even more clearly when examined within the
broader context of the Greek tradition. The belief that nomos is what
distinguishes humans from animals, insofar as it allows them to tran-
scend the world of brute force and violence by creating an order
based on shared values, is not a new idea that Protagoras introduced.
The very same thesis is also found in Hesiod, who – along with
Homer – was the great educator-poet of the Greek world:16

Perses, lay these things in your heart


And give heed to Justice, and put violence entirely out of your mind.
This is the law (nomos) that Cronus’ son has established for human beings:
That fish and beasts and winged birds
eat one another, since Justice is not among them;
but to human beings he has given Justice, which is the best by far.
(Hesiod, Works and Days, 274–80)

The thesis found in Hesiod is essentially the same as the one later
adopted by Protagoras, and the reference to this text is intended pre-
cisely to emphasize such a convergence: what is typical of human
beings, and what distinguishes them from animals, is their possession
of justice, which is to say their political capacity. Plato’s Protagoras
makes a brilliant attempt to reclaim the tradition, by highlighting the
continuity between his own theses and its precepts.17 However, this
attempt does not amount to an uncritical borrowing, because
Protagoras’ appropriation is also a deformation, and the differences
with respect to Hesiod are no less significant than the similarities.
According to Hesiod, Justice is divine: it is a deity, the daughter and

15
Pl. Prot. 325c5–326e5. See also Vegetti 1989: 51.
16
See Bonazzi 2011.
17
Sihvola 1989: 39–48.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


72 IV JUSTICE AND LAW

protégée of Zeus, who intervenes when he sees that men fail to respect
her (Works and Days, 257, 265; Theogony, 901–3). All mythological
imagery aside, this means that justice exists regardless of human beings;
human justice is not independent of this order of divine values, but
must rather conform to it. The situation radically changes with
Protagoras: the innovation consists in his emphasis on the human
rather than the divine, and the relative, rather than absolute, character
of justice.18 There is no place for divinity: justice is something human –
it is not what brings us close to the gods but what fulfils our natural
potential. Protagoras’, then, is a radical humanism; and it is in the
light of this humanism that the way in which he modifies Hesiod’s pos-
ition becomes clear. Whereas in Hesiod the present age (the Age of
Iron) is characterized precisely by the flight of Justice and Shame
(Works and Days, 190–201), in Protagoras the association of the
human community, polis, and justice also suggests that some form of
justice and shame is always to be found – as long as there are human
beings. In turn, this strengthens the reasonable belief that acting in
view of justice is in everyone’s interests. Whereas Hesiod defines
human history in terms of decadence (since it is difficult to imitate
the gods), Protagoras presents it as the history of a possible progress.
The engagement with Hesiod, which constitutes the real subtext of
the myth, confirms the underlying aim of Plato’s presentation: to high-
light the great ambition of Protagoras, who had not so much sought to
engage in erudite debate with other wise men, as to present himself as
an heir to the Greek paideia, as one of the great masters or, rather, as the
great master of Greek culture, capable of imparting a teaching that drew
upon tradition but that could also meet the needs of the new times
(consider the observations already made on pp. 51–52).
That this was Protagoras’ aim becomes even clearer when we con-
sider another of the great topics discussed in fifth-century Athens: dem-
ocracy. Generally speaking, Protagorean humanism was well suited to
Athenian democracy, so much so that it has been argued that his theses

18
In all likelihood, another implicit polemical aim of Protagoras’ thesis is to be found in those
‘unwritten and unfailing ordinances (nomima) of the gods, which have life, not simply today and
yesterday, but for ever’ (Sophocles, Antigone, 453–5) – on which see e.g. De Romilly 2005: 27–48.
Nor can it be ruled out that, in emphasizing the importance of these unwritten laws, fifth-century
authors such as Sophocles were precisely targeting positivists such as Protagoras and Pericles: see
Bonazzi 2017: 112–24, in the footsteps of Ehrenberg 1957.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


IV JUSTICE AND LAW 73

provide ‘a theoretical basis for participatory democracy’.19 With its


rejection of alleged absolute truths in favour of the vindication of the
legitimacy of opinions, the ‘man-measure’ thesis – when applied to
the political field – clearly serves to defend one of the cornerstones of
Athenian democracy, isegoria, namely every citizen’s right to express
his ideas in the assembly. Without having to posit that Protagoras’ doc-
trines were developed precisely for this purpose, it is possible to note
that they offered some insights that were far from irrelevant to the
new political phase that Athens was experiencing.
Even more significant are the implications that can be drawn from
the myth of Prometheus and Epimetheus. The defence of the political
dimension of human nature helps bring out the distinctiveness of the
democratic system compared to other political systems: if the human
being is a political animal who only achieves fulfilment in a political
context, it is clear that democracy, as the government in which all citi-
zens are allowed to take part, is the system that allows all humans to ful-
fil their nature (and to concretely bring attention to their needs). In
other words, insofar as it fulfils this requirement, democracy represents
not one form of government among others – albeit a better one – but
rather the form of government par excellence, the only truly ‘human’
one: the government of all. This idea of democracy as government of
all, as the government of the people as a whole and not just of a part
of it, which is to say the popular faction (to use the Latin terminology:
as the government of the populus rather than the plebs), was repeatedly
exploited by Athenian democratic propaganda.20 This emphasis on the
community, moreover, is used to justify another cornerstone of democ-
racy, arguably the most important one, namely the very close bond that

19
Kerferd 1981a: 144. A particularly important study on this topic is Farrar 1988. In order to
better evaluate Protagoras’ thesis – and, more generally, those of Athenian democracy – we must
distinguish between the theoretical level and that of concrete praxis. The reason for this is that
Athenian democracy has often been accused of being anything but a ‘democracy’, given that
roughly three-quarters of the population (women, slaves, and foreigners resident in Attica) were
barred from political participation. These are undeniable historical facts. Yet in no way do they
weaken the interest of theories such as those of Protagoras or Pericles, who promote an important
elaboration of the concept of democracy, understood as the government of the community. Ideas
of this sort still preserve all their importance today, regardless of the inconsistent way in which they
were applied in fifth-century Athens.
20
What proves crucial in this respect is the testimony of Thucydides, who, when discussing the
democratic ideology, repeatedly adopts the same perspective as Protagoras. The most clear-cut
stance is expressed in Athenagoras’ speech in Book 6: ‘democracy (demos) is the name for all, oli-
garchy for only a part’ (Thuc. 6.39). No less important is Pericles’ famous funerary speech for the
fallen of the first year of war, in which democracy is described as the government not of ‘the few’
but of ‘the many’ for ‘everyone’ (on this climax, see Musti 1995: 3–13).

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


74 IV JUSTICE AND LAW

must hold together the city’s interest and the individual’s, where the
former is the precondition for the latter.21 Protagoras, in other words,
emerges as a democratic political thinker: he is a political thinker
because he has shown that the social (or political – from polis)
dimension is fundamental; he is democratic insofar as he suggests
that political action is a collective and shared kind of action.
The new democratic political order is not imposed from above, but is
immanent and structured on two levels – those of competency and
excellence. As citizens, all people have a basic political competence,
which some people know how to make use of more than others
(the sophist here marks out a special area in which he can apply his
expertise); but all this invariably takes place through a mutual
engagement.22 Presenting democracy, the form of government for the
new times, as the heir and culmination of the Greek political tradition,
is certainly a brilliant achievement, which legitimizes Protagoras’
aspirations and ambitions. It is hardly a coincidence that the main
ancient reflections on democracy were often set in relation to
Protagoras’ ideas.23
In Protagoras, therefore, we find a first, vigorous defence of nomos,
an idea of which the fifth-century Greeks were particularly proud.
Protagoras was not completely isolated, as other authors and texts asso-
ciated with sophistry appear to have shared his positive evaluation of
law. The text closest to Protagoras’ views is arguably the so-called
Anonymous of Iamblichus, an anonymous treatise quoted at length
in the Neoplatonist Iamblichus’ Protrepticus. No less interesting are
some extracts from an oration, Against Aristogeiton, falsely attributed
to Demosthenes but more likely dating from the late fifth century. Its
author once again emphasizes the thesis that law and justice are what
allow humans to distinguish themselves from animals, by transcending
the world of violence: nature is disorderly and a source of injustice,
whereas the law ensures what is right, good, and beneficial (anon. On
the Laws, 15–20). Finally, Aristotle credits Lycophron with a ‘protec-
tionist’ conception of law (‘law is. . .the guarantor for each other of

21
For some interesting parallels, see Thuc. 2.40.2 and 2.60.2–4 (Pericles’ speech), and Hdt.
5.78.
22
Farrar 1988: 81–7.
23
In particular, in addition to the aforementioned passages by Thucydides, it is worth mention-
ing the famous comparison between the various forms of government found in Herodotus 3.80–2.
Finally, it is worth recalling that the seventh section of the Dissoi logoi also discussed democracy
and how best to safeguard it.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


IV JUSTICE AND LAW 75

what is just’, 83.3 D.-K. = 38D3 L.-M.), which can probably be under-
stood as one of the first attestations of contractualist theories (of which
there is a compelling formulation in Glaucon’s speech at the beginning
of the second book of Plato’s Republic): the idea that laws guarantee
rights evidently implies a ‘disorderly’ conception of human nature.
Laws are designed to protect people, which is why they must be
respected: at least in this sense, the ‘protectionist’ and the contractual-
ist theories converge.24
Unfortunately, the dearth of information about these authors makes
it impossible to reconstruct in any detail the relation between them and
Protagoras, whose theses certainly attracted much attention in Athens
and Greece; but it is not unreasonable to hypothesize that they drew
upon his work in some way. Besides, the fact that Protagoras consti-
tuted an important point of reference in the debates on such matters
may also be inferred from the theses of sophists opposed to nomos,
whose arguments – as we shall now see – appear to have had precisely
the thinker of Abdera as their target.

Against nomos: from Gorgias to Callicles

The other great sophist, Gorgias, does not appear to have devoted any
special attention to justice and other political issues. Here it is possible
to register a substantial difference compared to Protagoras: whereas the
latter attributed the utmost importance to the teaching of political vir-
tue, the former apparently even went so far as to claim that he did not
teach virtue at all:
What I admire most of all in Gorgias, Socrates, is that you would never hear him prom-
ise this [sc. to be a teacher of virtue]. Instead he laughs at other people when he hears
them make this promise. No, he thinks that one should make people clever at speaking.
(82A21 D.-K. = 32D47 L.-M.)

Contrary to what might seem to be the case, however, the opposition


between the two does not lie so much in that between the centrality
of politics on the one hand and the lack of political involvement on
the other: rhetoric, the object of Gorgias’ teaching, is the political art
par excellence, which teaches one how to persuade a crowd or

24
More generally, on the possible circulation of contractualist theories in the fifth century, see
Guthrie 1971: 135–47 and Kahn 1981.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


76 IV JUSTICE AND LAW

manipulate its decisions; it is the art that ensures domination (82A27


D.-K. = 32D7a–b L.-M.). Therefore, as Plato had realized in the
Gorgias, even Gorgias’ teaching is political, in its own way. The true dif-
ference lies in Gorgias’ greater degree of conformism compared to
Protagoras, who had sought to establish a critical relation to traditional
values, rejecting some values while adopting others. Gorgias instead
seems to have no interest in critically engaging with traditional beliefs,
which he readily accepts (see, for example, 82B18–19 D.-K. = 32D53
L.-M.). What matters is not to call a given value system into question,
but rather to find self-affirmation within such a system: this is the aim of
rhetoric, and the success that it ensures is what Gorgias is concerned
with.
In other words, it could be argued that Protagoras attempts to recon-
cile the society’s reasons with those of individuals, whereas Gorgias
focuses on the interests of the individual in contrast to life in society.
The evidence we have does not allow us to claim that, for Gorgias,
the defence of individuals also entails an openly hostile stance towards
society, as was later the case with other sophists. But this contrast is at
least theoretically possible, and explains why Gorgias too can be
included – if only in a nuanced way – among those sophists who
were critical of nomos, understood as the law governing the relations
between citizens within a community.
Significant confirmation of this is provided by the fragment of an epi-
taph written to commemorate the fallen Athenians (82B5a–b and 6
D.-K. = 32D29, 30a–b, 28 L.-M.). This speech, a fine example of
Gorgias’ prose, is entirely played out in oppositions and contrasts,
which are used to celebrate the qualities of the dead soldiers.
Unsurprisingly, these qualities coincide with traditional values, such
as respect for one’s parents and veneration of the gods. Yet this trad-
itionalism heightens the choice to list justice (dikaion) and law
(nomos) among the negative terms: ‘often preferring gentle equity to
implacable justice, and often the correctness of words to the punctili-
ousness of law’. This is certainly a bold expression, which explains
why, shortly afterwards, Gorgias makes an allusion to the ‘most divine
and most universal law’ (‘considering the most divine and most univer-
sal law to be this: both to say and to leave unsaid, and to do < and to
leave undone > what is necessary when it is necessary’). However, it
is evident that this religious veneer is only used to introduce a recurrent
theme in Gorgias’ thought, namely the importance of the opportune
moment (kairos): the ‘most divine law’ is the capacity to say and do

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


IV JUSTICE AND LAW 77

opportune things at the opportune moment. And this is precisely what


justice and law lack, insofar as they purport to control and dominate the
complex and elusive reality of human experience. The rigidity of the
law is thus set in contrast to the flexibility of intelligence, which is cap-
able of adapting its reasoning to the existing circumstances, so as to
favour the best and hence most ‘correct’ situations.25 This contrast
between intelligence and law betrays a degree of impatience with
nomos and was repeatedly taken up by many thinkers, as we shall see
later on in this chapter and in the following one, to the point that it
came to be regarded as a distinguishing feature of the sophists (see
Thucydides 3.37).
A second interesting insight is provided by the Encomium of Helen:
among the various reasons adduced in Helen’s defence, we find
power. If Helen has followed Paris as a result of the will of a superior
being, the goddess Aphrodite, it is not right to blame Helen:
For by nature the stronger is not prevented by the weaker, but the weaker is ruled and
led by the stronger, and the stronger directs, the weaker follows. But a god is stronger
than a human in force, in intelligence, and in all respects. So if the responsibility is to be
ascribed to Fortune and to a god, Helen too is to be freed from ill repute.
(82B11, 6 D.-K. = 32D24, 6 L.-M.)

In this case, too, the mythological scheme (the reference to the super-
iority of gods over humans – but let us not forget that ultimately the
gods are equated with destiny) conceals a far from obvious idea,
which de facto constitutes a first justification of strength as the criterion
guiding human actions:
the form still conforms to traditional patterns, but the principle is now used to justify
the general law of the supremacy of the stronger, which, within the framework of the
mythological example and scheme, is geared towards the acknowledgement of a truth
that de facto operates in the world of human beings.26

Again, what we have here is an idea destined to prove highly influential


in the future.
A radical redevelopment of the insights provided by Gorgias is
offered by Callicles, a mysterious figure who is only known to us
through Plato’s Gorgias, where he is featured – significantly – as a
friend, guest, and in a way pupil of Gorgias’. Given the silence of our

25
Note the importance of this concept: see Chapter 3, p. 48.
26
Isnardi Parente 1969: 172.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


78 IV JUSTICE AND LAW

sources, some scholars have doubted his very existence. However, it is


more likely that Callicles was an actual Athenian nobleman who died
young in the years of the Peloponnesian War, leaving little memory
of himself outside his circle of acquaintances (which included Plato:
many of the characters associated with Callicles as his friends, lovers,
or relatives may be traced back to Plato’s own milieu). For Plato this
young aristocrat embodied a certain way of thinking, an unscrupulous
‘philosophy of life’ deeply influenced by the sophists’ theses (despite
Callicles’ professed contempt for the sophists: see Gorgias, 520a), a
philosophy which must have enjoyed a considerable degree of popular-
ity in Athens in the war years.27
At first sight, Callicles’ thesis presents itself as an all-out attack
against nomos and as a celebration of physis. In particular, it consists
in the distinction between two different laws: law in accordance with
nature and law against nature. Whereas nature shows that strength is
the only criterion of value (‘nature itself reveals that it’s just thing for
the better man and the more capable man to have a greater share
than the worse man and the less capable man’, Gorgias, 483c–d), in
the human world it is the weak who dominate by forcing the strong
and best to respect laws against nature, which proclaim false values
such as equality and moderation (483b–c). Evidently, the polemical ref-
erence here is to democratic nomos, to which Callicles opposes – with an
almost religious emphasis – nature’s law, prophesying the triumph of
the strongest and of true justice, namely justice in accordance with
nature, which stems from strength and the ability to overpower others
(483e–484a).
Callicles’ statements are provocative not just because of their openly
anti-democratic tone, but also – and especially – because of their impli-
cations. In support of his thesis that nature regulates itself through
force, Callicles adduces the examples of what occurs in cities and fam-
ilies (alluding to imperialist wars of conquest and to the practice of slav-
ery), and of the behaviour of animals. This juxtaposition is far from
neutral, insofar as it entails a rejection of human specificity, which
the Greek tradition from Hesiod to Protagoras had stressed: the idea
of the long journey of human beings, who with much difficulty emerge
from the violent animal world through nomos, is wholly rejected as a
false and tendentious account, which distorts reality out of fear of the

27
Dodds 1959: 12–15.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


IV JUSTICE AND LAW 79

truth. The key word in Callicles’ speech, his truth, is pleonexia, literally
the drive ‘to have more’; and it is this unsuppressible instinct to over-
power others, this will and desire to increase one’s own power, that is
the hallmark of all things – of humans, of animals, and indeed of the
whole universe (cf. Gorgias, 508a).
Callicles’ claims are certainly intriguing, as is shown for instance by
the fact that Friedrich Nietzsche drew upon them in some famous
pages of his Genealogy of Morals.28 However, it is easy to show that
they are weak theses. The main difficulty lies in the conflating of the
descriptive and prescriptive aspects of the concept of nomos, or law.
To clarify the problem, we can consider the modern distinction of
the term in the scientific field and in the legal one: scientific laws,
which describe the behaviour of a given object (be it a lion or a planet),
are one thing; the laws issued by legislators, which prescribe a certain
type of action (such as the paying of taxes or the avoidance of murder),
are another. In other words, to state how things are is different from
stating how they should be.
Callicles’ thesis ultimately boils down to an attempt to make these
two aspects coincide, but it falls short of its mark. Like a scientist, he
sets out from the descriptive aspect, yet, unlike the scientist, he also
assigns it axiological value. Thus, it is an objective fact that reality reg-
ulates itself according to power relations, but this is not understood as
something neutral, but rather as something good and just: to use
Callicles’ own words, there is a natural justice, and this is pleonexia,
which is to say the law of the strongest. But herein lies the problem.
The fact that reality axiologically regulates itself according to power rela-
tions means that the one who prevails is the strongest and hence the
most just. But given that Callicles’ world is dominated by those whom
he contemptuously refers to as the mass of weaklings, it follows that
the weak are actually the strong and hence the just: the democratic
nomos of equality is therefore just and beautiful, whereas the ‘strong’
(that is, those regarded as strong and best by Callicles, the aristocrats)
are de facto losers and weaklings.29 In other words, establishing strength
as the criterion of value implies an acceptance of reality, and not its
rejection, which is what Callicles opts for when he calls for the overthrow
of the tyranny of the weak. Significantly, later on in the dialogue,

28
See Dodds 1959: 387–91.
29
A similar reasoning, intended to show the weakness of the strong, is also to be found in the
Anonymus Iamblichi (89.6 D.-K. = 40.6 L.-M.).

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


80 IV JUSTICE AND LAW

Callicles, when pressed by Socrates, will soon relinquish strength as the


distinctive criterion for the good and the just (489c). But, in doing so, he
strips his theses of their ‘realistic’ foundation, showing them for what
they are: a brilliant yet fruitless rhetorical exercise.30 Rather, as we
shall see in the next chapter, it is in relation to the individual (that is,
to moral rather than political problems) that Callicles’ ideas prove
more compelling: his arguments represent more of ‘a life choice, a nos-
talgia for Homeric times, a hope to free oneself from the bonds of egali-
tarian morality and law, than a genuine theoretical argument’.31

Thrasymachus and political realism

Even though they bear witness to an increasing discontent with democ-


racy (something also attested by many other writings from the same
period),32 Callicles’ statements do not pose any real obstacle to the the-
ses in favour of nomos, which find their most prominent spokesman in
Protagoras. More insidious opponents are to be found in the persons of
Thrasymachus and Antiphon, two leading exponents – along with the
historian Thucydides – of what in modern terms we might describe
as ‘political realism’.
Just as for Callicles, Plato’s testimony proves both crucial and prob-
lematic for Thrasymachus. Although (unlike in Callicles’ case) many
other sources talk about Thrasymachus, our only source for his political
thought is the first book of the Republic. Here the sophist is credited
with different theses on justice that are difficult to reconcile: first, just-
ice is defined as the advantage of the stronger, ‘what is advantageous for
the person who is stronger’ (Republic, 338c = 85B6a D.-K. = 35D21a
L.-M.; the thesis is later rephrased as respect for the laws, 340a–b),
and then as ‘what is good for someone else’ (343c = 35D21a L.-M.).
Modern scholars have poured out rivers of ink in the attempt to recon-
cile these theses, without ever reaching a shared solution. Without delv-
ing into this complex debate,33 it may be noted that the ambiguities are

30
Fussi 2006: 203–20.
31
Vegetti 2018: 198.
32
The most significant text, not least because of the numerous affinities it has with the sophists,
is the anonymous Constitution of the Athenians, with regard to which I will refer to Lapini 1997 and
Gray 2007; also very interesting is Xenophon, Memorabilia, 1.2.40–6.
33
For a reconstruction of the different hypotheses that have been put forward (to show either
the coherence or the incoherence between the two theses, and to favour one or the other), see

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


IV JUSTICE AND LAW 81

at least partly resolved if we consider the fact that the first thesis refers
to more specifically political issues, while the second one concerns indi-
vidual choices and hence ‘ethical’ problems (although we should not
forget that, prior to Aristotle, the distinction between ethics and politics
was not explicitly theorized). Consequently, it is worth focusing here on
the first thesis, while the importance of the second thesis – namely,
what its polemical target is – will become clear in the next chapter,
where we will be discussing happiness.
In arguing with Socrates, or rather in countering his defence of just-
ice, Thrasymachus claims of justice that it is ‘what is advantageous for
the person who is stronger’ (Republic, 338c = 85B6a D.-K. = 35D21
L.-M.), noting immediately afterwards that the stronger is to be under-
stood in the sense of ‘the established rule’ (arche): hence, justice ‘is the
advantage of the stronger’, that is of the established rule (Republic,
338d–e). This thesis has often been compared to Callicles’, but this
is a misleading comparison: while it is true that both of them pay
much attention to strength, Thrasymachus clearly understands strength
in the sense of power, and his emphasis on the theme of power allows
him to avoid the conflation between the descriptive and prescriptive
aspects that had proven so fatal to Callicles. Thrasymachus’ thesis is
merely descriptive, insofar as it does not pass any judgement but only
records a fact: namely that strength and power constitute a fundamental
aspect of human relations. Power is where the strength lies, and both
justice and law are subordinate to it: those who are in a position of
strength – that is, those who wield power – establish rules and laws,
the only purpose of which is to ensure that those in a position of
strength preserve their power and hence pursue their own advantage.34
Despite the apparent similarities, the theses of Callicles and
Thrasymachus differ in some crucial respects. First of all,
Thrasymachus’ thesis acquires a universal value that Callicles’ lacks: it
can be used to describe all the political relations that govern human
lives, without the need to draw any arbitrary distinctions between
good and bad governments. Furthermore – and this is the most
significant divergence – Thrasymachus and Callicles have two different
conceptions of justice. For Callicles, justice is something real, some-
thing that objectively exists and is embedded within reality ( physis): its

Macé 2009: 159–62 and, most recently, with further bibliography, Barney 2017, Wedgwood 2017,
and El Murr 2019.
34
Vegetti 1998: 240–7.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


82 IV JUSTICE AND LAW

purpose (a rather confused one, as we have seen) is to bring human


nomos in line with this natural justice. For Thrasymachus, in contrast,
it makes no sense to speak of natural justice: that is, to speak of justice
as though it is something that exists independently of human beings.
Justice is a purely human concept, which is not rooted in physis at all;
it is a convention that regulates human relations (and does so according
to power relations).
These clarifications help us to better understand the relation between
Thrasymachus and Protagoras, who probably represents the polemical
target of the former’s thesis: Thrasymachus’ realism is developed
within the same framework as Protagoras’ conventionalism, so much
so that one could argue that the affinities between the two are greater
than those between Thrasymachus and Callicles. However, the conclu-
sions reached by Thrasymachus overturn the meaning of Protagoras’
thesis, highlighting its limits and the problems that it poses. What the
two sophists have in common is their starting point, positivism – a
rather widespread idea in the Greek world.35 According to this assump-
tion, there is no justice outside law: only what the laws decree to be just
is just. Yet the identification of law with justice does not guarantee the
correctness of Protagoras’ reasoning, since alongside law and justice, or
rather behind them, there lies power; and the acknowledgement of the
priority of power leads to opposite conclusions from those of
Protagoras. According to the latter, as we have seen, it is the commu-
nity as a whole that establishes the rules, laws, and values that promote
peaceful coexistence and the pursuit of everyone’s interests. According
to Thrasymachus, by contrast, the idea of a harmonious and close-knit
community is an illusion, since all human communities are created on
the basis of power relations that govern them: society is nothing but dif-
ferent parts jostling for power. This entails a different evaluation of
nomos, justice, and interests. Laws are not the result of a collective
action: more crudely, they depend on the choices and interests of the
group that finds itself in a dominant position. Therefore, to the extent
that law is identified with the laws, it is a product of force; it represents
not a shared value capable of neutralizing potential conflicts in the
community, but a partisan means to legitimize a given balance of
power between the members of the community. In modern terms, to

35
Vegetti 1998: 240–2.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


IV JUSTICE AND LAW 83

borrow Marxist terminology, we might speak of justice as an ideology


which offers a false view of reality, so as to justify the partisan reality
of power through the screen of a universal principle.
The same considerations also apply to what is useful. For Protagoras
as much as Thrasymachus, advantageousness is the motive behind
all political decisions. However, given that the decisions are always
made by a part of society rather than the whole, it follows that
advantageousness does not concern the community as a whole,
but only those who find themselves in a stronger position, and who
endorse as what is just their own private interests, which is to say
what is advantageous for the preservation of their own power (hence,
in 343c, Thrasymachus will argue that justice is ‘what is good for
someone else’: the good of the one who is in power, not of the one
who is ruled).
On a more concretely political level, it may be noted that
Thrasymachus distances himself from democracy, yet in a different
way from Callicles’ frontal attack. One consequence of Thrasymachus’
realist, objective, and detached attitude is that it would make no sense
to draw a distinction between good and bad regimes, because every
form of government is simply the implementation of a power relation,
which it would be pointless to criticize: democracy is no less legitimate
than aristocracy or monarchy (Republic, 338e–339a). Upon closer
inspection, however, it is easy to see that this apparent legitimation of
the democratic regime actually conflicts with the assumptions of the
champions of democracy. The main postulate behind the democratic
theory – as may be inferred from Protagoras or Pericles – is the belief
that democracy is the government of everyone, and not just of a part;
and it is precisely in this universal character of democracy that its
specificity and superiority lie: as the government of everyone, democracy
is the only regime which can ensure everyone’s good and advantage.
Thrasymachus’ analyses disprove these ideas: insofar as democracy is
the government of a part of society, albeit the majority, it follows that
it only safeguards the interests of that part and not of everyone. When
Thrasymachus associates the democratic regime with the more
traditional ones of aristocracy and monarchy, all he is doing is once
again proving the centrality of power in human relations (since every
form of government expresses a power relation), without making any
concessions to the alleged merits or specificity of democratic
government.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


84 IV JUSTICE AND LAW

Antiphon and the logic of the individual

While Thrasymachus brought to light the inconsistencies of positivism,


even more scathing observations were put forward by Antiphon, who
opposed the logic of the individual to the Protagorean logic of the com-
munity. The juxtaposition between Thrasymachus and Antiphon might
seem surprising, as Antiphon is usually reckoned among the champions
of physis – unlike Thrasymachus, whose analysis is entirely developed
within a positivist and conventionalist context, which entails no refer-
ence to objective or natural values.36 However, as we have seen in
Chapter 2, this alleged opposition is actually a fallacious one: according
to Antiphon, nature is neutral and does not provide any normative
values. This is the truth of his work On Truth. The same considerations
also apply to fragment 44, again from On Truth, which more specifically
concerns human physis.37 In an apparently trivial way, but actually
drawing upon the anatomical research of physicians and other scien-
tists, Antiphon sets out to systematically classify the various limbs
and organs of human bodies, along with the functions associated
with them (87B44A, 2.15–3.12 D.-K. = 37D38b L.-M.). The outcome
is an acknowledgement of the biological equality of human beings. Yet,
contrary to what one might assume, this biological equality does not
have any humanitarian consequences: it is not the source of any rights,
nor does it lay the foundations for political equality, against the conven-
tions that divide people (such as the distinction between free men and
slaves, or nobles and non-nobles). On the contrary, the consequences
drawn from this acknowledgement entail serious problems: equality
means that humans have the same needs, and this gives rise to a poten-
tially conflictual situation, whereby all people desire the same things
and each person pursues his or her own interests against those of
others.
This analysis of human physis goes hand in hand with – and sets the
ground for – that of nomos: the lengthy papyrus fragment that makes up

36
This juxtaposition might seem even more surprising if we consider the fact that most scholars
tend to interpret Antiphon not just as a champion of physis but as the first spokesman for natural
justice. However, this reading is based on an erroneous reconstruction of the papyrus transmitting
87B44 D.-K. (=37D38b L.-M.), as may be inferred from the recent discovery of another part of
the same papyrus: see the edition of Bastianini and Caizzi 1989 (which I follow for the column
order) and the point made by Narcy 1996.
37
In my analysis of this fragment, I am partly drawing upon Bonazzi 2020, which I refer to for
further bibliographical references. Ostwald 1990 remains the most useful introduction.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


IV JUSTICE AND LAW 85

fragment 44 preserves the most in-depth analysis of the problematic


relations between these two concepts. For Antiphon, nomos, in all of
its meanings (as law, tradition, and custom), is something negative: a
‘bond’ that seeks to regulate what nature has left unregulated, thereby
preventing individuals from fulfilling their needs or desires (87B44B,
1.22–4.8 D.-K. = 37D38a, 1.22–4.8 L.-M.). This attempt to control
physis inevitably proves unfavourable, since it prevents us from pursuing
our own advantage, which always depends on our nature, on our
constitution:
If one reasons correctly, what causes pain does not help nature more than what causes
pleasures; and so what causes pain would not be more beneficial either than what
pleases. For what is truly beneficial must not be harmful but helpful.
(87B44B, 4.9–22 D.-K. = 37D38a, 4.9–22 L.-M.)

With Antiphon, the opposition between physis and nomos becomes


explicit and clear-cut.
At this stage, however, one could raise the following objection: given
the neutral character of physis and the potential for conflict among
humans, nomos might be seen to have a positive value insofar as it is
capable of preventing the war of all against all. A solution of this kind
is implicit in the aforementioned thesis put forward by Lycophron
(83.3 D.-K. = 38D3 L.-M.) and finds an eloquent formulation in the
long speech that Plato puts in his brother Glaucon’s mouth in the
second book of the Republic (358e–359c). By nature, humans tend to
commit acts of injustice towards one another; but, realizing that in
this situation the advantages are fewer than the disadvantages, they
established a corpus of laws to regulate their actions, depriving them
of certain possibilities, yet ensuring security. In other words, the
bonds to which we are subjected also restrain brutes and predators
far worse than us.
The following section of the Antiphon passage precisely addresses
objections of this sort: a careful analysis of social dynamics shows that
nomos is often incapable of ensuring this security. Despite its coercive
power, nomos is incapable of preventing its own violation: for example,
it decrees that stealing is forbidden, but does not actually prevent every-
one from stealing. This leads to a paradoxical situation, insofar as nomos
proves doubly disadvantageous precisely for those who respect it:

If the laws provided some help for those who submit to such situations, and loss for
those who do not submit to them but who oppose them, [col. 6] it would not be useless

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


86 IV JUSTICE AND LAW

to obey the laws. But as it is, it is evident that, for those who submit to such situations, a
just outcome deriving from a law is not enough to provide them help. First, it permits
the sufferer to suffer and the doer to do. And just as, at that time [i.e. before the crime]
it did not prevent either the sufferer from suffering or the doer from doing, so too when
reference is made to it for the purpose of punishment, it is not more on the side of the
one who has suffered than on that of the one who has done. For it is necessary that
he persuade those people who will exact punishment that he has suffered, or that he
be capable of obtaining justice by deceit. But the doer too is permitted to deny
the same things [col. 7]. . .exactly as much of a defence is available to the defendant,
as the accusation that is available to the accuser, and persuasiveness is balanced for
the one who has suffered and for the one who has done.
(87B44B, 5.25–7.14 D.-K. = 37D38a, 5.25–7.14 L.-M.)

Let us keep to the example of stealing. Those who accept the injunction
not to steal will not steal, even in a situation in which stealing might ful-
fil a need (as in the case of those who are hungry and steal food). This is
one disadvantage; but, what is more, respecting the law does not pro-
tect one from the possibility of being in turn robbed by those who do
not respect this law. Consequently, one risks suffering an additional
drawback, which adds insult to injury, considering that the law courts
assign equal rights to the victim and to the guilty party, and that the
outcome of a trial depends more on the capacity to persuade than on
the ascertainment of the truth.
No less interesting are the analyses devoted to the practice of bearing
witness in relation to the problem of justice (87B44C D.-K. = 37D38c
L.-M.).38 People usually believe that it is unjust to commit injustices
against those who have not committed any injustice; and that it is
just to testify to the truth in a court case. But, according to
Antiphon, these two beliefs are mutually incompatible when a truthful
witness leads to the sentencing of a criminal who had not done that wit-
ness any wrong. Moreover, Antiphon observes, in doing so the witness
puts themself at risk, since he antagonizes someone – the criminal –
who will sooner or later seek revenge. All this brings out the limits of
nomos, which not only prevents people from fulfilling their primary
needs, but is even incapable of protecting those who yield to its rule.
It is as though physis and nomos are applied at two parallel levels: the
level of truth and that of opinion, the level of concreteness and that
of abstraction, without these levels ever really meeting. While through
his or her action the witness has upheld the nomos of an abstract

38
See Furley 1981.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


IV JUSTICE AND LAW 87

community, he or she has caused, and has incurred, concrete and per-
sonal damages.
Antiphon’s analyses are important in several respects. First of all, it is
worth noting that they constitute an eloquent example of the sophists’
wide range of interests. It is generally assumed that the sophists had no
interest whatsoever in the research conducted by natural philosophers
and scientists, and that they only dealt with politics. Antiphon’s case
instead presents us with a more nuanced situation: the originality and
importance of the sophists rather lies in their ability to bring together
the two spheres, and to put scientific research to the service of political
aims. This is a matter of method as well as content: the adoption of a
detached and scientific, descriptive and non-judgemental attitude is
what lends shape to the ‘realistic’ approach, which constitutes one of
the most significant contributions offered by the sophists’ political
reflection.
It is crucial to take account of this ‘realistic’ and detached attitude if
we are to fully understand Antiphon’s position: the text clearly suggests
that Antiphon did not wish to present his own conception of justice in
opposition to conventional ones, as Callicles does in the Gorgias.
Rather, his primary aim is to closely engage with the theories of justice
that were then circulating in Athens and in the Greek world more
widely. In particular, the opening lines of the fragment show that the
sophist was particularly interested in addressing the positivist thesis,
which had been developed in particularly compelling terms by
Protagoras: the definition of justice as obedience to the laws of the com-
munity of which one is a citizen (87B44B, 1.6–11 D.-K. = 37D38a,
1.6–11 L.-M.) clearly captures the cornerstone of this thesis, namely
the identification of law with justice (what is just is what the laws
decree).
As we have just seen, the critique of positivism is common to many
sophists. However, Antiphon is the one who best grasps the underlying
crux of the matter by raising the vital question of the foundation of the
law (and hence of justice): if justice is a human convention, meaning
that it simply consists of the laws established by people, what authorita-
tiveness or foundation can it have? Protagoras’ answer, as we have seen,
is the idea of what is useful. The authority and legitimacy of nomos
depend on its capacity to ensure everyone’s interests: this is the reason
why nomos must be respected. Nevertheless, a concrete analysis of how
society works shows the flimsiness of Protagoras’ answer, which is an
abstract one, insofar as he refers to a (ideal) community of citizens

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


88 IV JUSTICE AND LAW

rather than to the concrete needs and interests of individuals. Often


(2.26–7) the law entails coercion without offering any advantages or
security. What are we to do in these cases? Antiphon pushes the posi-
tivist thesis towards a paradoxical conclusion: if the legitimacy of
nomos depends on the advantages it provides, if the foundation of just-
ice is its usefulness, and if in certain cases it is more useful to violate
justice than it is to respect it, it follows that injustice is just – or at
any rate, and more pragmatically, that one should publicly abide by
the law but secretly follow nature. Antiphon’s reasoning should not
be interpreted as a call to go against the law; rather, it is a provocation
that pushes to its extreme (and unexpected) consequences a wide-
spread idea of justice, which had been strongly championed by
Protagoras:
Therefore justice is not to transgress against the legal institutions of whatever city one
happens to be a citizen of. So a man would make use of justice in the way that would be
most advantageous for himself if, in the presence of witnesses, he considered that it is
the laws that are great, but, alone and without witnesses, that it is what belong to nature.
For what belongs to the laws is adventitious, but what belongs to nature is necessary.
And what belongs to the laws is the product of an agreement, but what belongs to
nature is the product of nature, not of an agreement. [col. 2] So if someone transgresses
against legal institutions without being noticed by those who agreed upon them, he
escapes shame and punishment; but if they notice, he does not. But if, contravening
what is possible, he does violence to anything produced by nature, the harm is no
less if no man notices him, and it is not greater if all men see him. For it is not because
of opinion that he is harmed, but because of the truth.
(87B44B, 1.5–2.23 D.-K. = 37D38a, 1.5–2.23 L.-M.)

Certainly, Antiphon’s analyses present many similarities to


Thrasymachus’ thesis, which they usefully integrate: while the latter
had criticized the legitimacy of positivism starting from the problem
of power, Antiphon reaches similar conclusions by setting out from
the problem of usefulness. If anything, Antiphon’s analysis is even
more radical: Thrasymachus was operating on a distinctly political
level, by considering the division of the city into social strata (the people
vs nobles for instance). Antiphon instead appears to be interested in a
more ‘philosophical’ analysis, which he extends down to the ultimate
constituent of the city, individuals. It is here that the choice of
Protagoras as a polemical target becomes clear. Whereas in his myth
Protagoras had described human beings as ‘not yet social’ to show
that his interests are eminently social (the human being is a political
animal: there is no human who is not a citizen, a member of a

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


IV JUSTICE AND LAW 89

group), Antiphon analyses the life of humans in society to show that his
interests are individual and hence anti-social – in other words, they
concern the human as an individual and not as a citizen. This explains
the limits of Protagoras’ defence of nomos: the only reality is each indi-
vidual,39 whereas the laws of the polis, directed as they are at an abstract
citizen community, at most express an external and artificial consensus,
incapable of safeguarding the natural and hence real interests of
individuals.40
Such is the ‘truth’ that Antiphon reaches through an impartial ana-
lysis of reality, and it is a truth that affords few illusions as to the pos-
sibility of building a just and close-knit society. The dominant tone of
his analysis suggests an acknowledgement of the difficult condition of
humanity, caught between natural needs and social constrictions.41
However, this is not to say that it is impossible to live in a satisfying
way. After all, the human being has a most valuable means to make
up for the limits of politics and nomos, a means which unfortunately
he or she rarely resorts to: intelligence. As we shall see in the next chap-
ter, it is by using one’s intelligence that it is possible to find a balance
both within oneself and with others.

Other enemies of nomos

Before wrapping up this long discussion, brief mention must be made


of other testimonies which can be traced back to the sophists’ debates
and which record positions akin to those of Thrasymachus and
Antiphon. One interesting parallel is to be found in a tragedy by
Critias, the Pirithous:

39
Hourcade 2001: 115.
40
Farrar 1988: 115–18. In passing, it goes without saying that these conclusions have anti-
democratic implications. Naturally, by this I do not mean to say that it is possible to infer from
the surviving fragments of Antiphon’s On Truth that he openly sided against democracy: on the
contrary, what makes Antiphon’s analyses so intriguing is his claim that he is not taking any
stand himself, but wishes the dynamics of natural and social reality to emerge on their own.
Still, it must be noted that his analyses of humans and society, his systematic emphasis on the
essentially anti-social character of human beings, and hence his privileging of the needs of the indi-
vidual over the collective rights of citizens constitute one of the most dangerous attacks on
Athenian democracy, which was founded on the identification of the common good with the inter-
ests of individuals. The ‘atomization’ of common interest into many private and mutually conflict-
ing interests amounts to the de facto impossibility of finding the kind of common good which, in
the eyes of Protagoras and Pericles, was the distinguishing trait of democracy.
41
Decleva Caizzi 1999: 327.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


90 IV JUSTICE AND LAW

An honest character is surer than the law:


For the former, no orator could ever pervert it;
While the latter, he often ruins it
By agitating it up and down with his speeches.
(88B22 D.-K. = Dram. T67 L.-M.)

Undoubtedly, this combined criticism of the law and rhetoric, which is


typical of Athenian assemblies, offers additional proof of the growing
dissatisfaction that, in the second half of the fifth century, many intel-
lectuals displayed towards nomos, which was increasingly identified with
democratic nomos. But it should also be noted that, in these verses, the
criticism of rhetoric amounts to an attempt to distance oneself from
sophistry: again, Critias emerges not so much as a sophist as a writer
capable of exploiting some sophistic themes for his polemical
purposes.42
Even more interesting is another testimony from Sisyphus, a satyr-
play previously attributed to Critias but more likely to be the work of
Euripides.43 Some significant analogies with Antiphon may be found:
nomos is assigned an instrumental value, as a means to safeguard com-
munities. Yet this value is denied at the very moment in which it is pos-
ited: laws were introduced to bring an end to the bestial violence that
originally marked human life; however, these laws only partially met
their aim, because people continued to secretly commit injustices,
thereby undermining society. Only the invention of the gods as omnis-
cient and omnipotent controllers laid the foundations for a more stable
society. Leaving aside the issue of the invention of the gods, on which
we will focus in Chapter 6, the similarities with Antiphon’s idea of the
conventional character of nomos and its weakness are undoubtedly
significant.44
What is more difficult to assess is Hippias’ position, as it emerges
from two testimonies from Plato’s Protagoras (337c–e = 86C1 D.-K.
= 36D17 L.-M.) and Xenophon’s Memorabilia (4.4.5–25, only partly
included by Diels among the testimonies on Hippias as 86A14
D.-K.; =36D11 L.-M.). In both passages Hippias emerges as a staunch
critic of nomos, which is accused of imposing arbitrary distinctions

42
See H. Patzer 1974. Critias’ aims are quite clear: an unrelenting struggle against democracy,
in an attempt to import into Athens the values of the oligarchic ideology of Sparta. On this, see
Bultrighini 1999 and Iannucci 2002.
43
See Chapter 6, p. 119.
44
Ostwald 1986: 282.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


IV JUSTICE AND LAW 91

against nature (86C1 D.-K. = 36D17 L.-M.: nomos ‘which is a tyrant


over men, commits violence upon many things against nature’) and
of having an unstable character (Mem. 4.4.14: ‘Laws can hardly be
thought of much account, or observance of them, seeing that the very
men who passed them often reject and amend them’). These are essen-
tially similar objections to those raised by Antiphon and Sisyphus.
According to some scholars, moreover, both Xenophon and Plato pre-
sent Hippias as a resolute champion of physis, or even as the theorizer of
the existence of a divine and/or natural justice.45 However, it is more
prudent to suspend judgement on this interpretation: with regard to
Xenophon’s Memorabilia, because the idea of a divine justice is some-
thing that is rather upheld by Socrates;46 and with regard to Plato,
because the reference to physis is probably to be understood in a
more restricted sense, as we shall see in the next section. The testi-
monies we have, then, allows us to identify Hippias too as a critic of
nomos, even though it is impossible to clearly determine what his overall
stance was on the thorny problem of justice (assuming that he indeed
took a stance on the matter).

The problem of equality

Another strongly debated topic was equality. Once again, the political
context – the ups and downs of Athenian democracy – helps us to
understand the interest in this issue: two of the catchwords of the
democratic regime, those which best represented this form of govern-
ment, were isegoria and isonomia, which is to say everyone’s right to
express themselves and everyone’s equality before the law (notwith-
standing the fact that ‘everyone’ excluded a considerable section of
society47). Moreover, the importance of the topic is confirmed by the
fact that discussions about equality extended to broader fields, touching
upon social, racial, and economic questions.48 Naturally, as in the case
of justice, the topic was addressed by many different authors, and not
just sophists, but the latter’s contribution was often highly original

45
Untersteiner 1954: 278–83.
46
Decleva Caizzi 1985: 203–8, has hypothesized that Socrates’ thesis specifically has Antiphon
as its polemical target.
47
See p. 73, n. 19.
48
Guthrie 1971: 148–63.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


92 IV JUSTICE AND LAW

and raised some crucial questions. Usually, the sophists’ position on


this issue has been invoked in support of an ‘Enlightenment’ interpret-
ation of their thought, presenting them as the first thinkers to ever
emphasize the need to truly free oneself from the prejudices oppressing
people’s lives. The surviving testimonies, however, suggest a more
restrained evaluation: certainly, many sophists must be credited with
having brought some real problems up for discussion. Yet here too
many of them would appear to have been driven not so much by the
need to ‘rouse men from dogmatic slumber’, as by a desire to investi-
gate certain problems and topics – and perhaps too by a wish to provoke
the public by going against the grain.
One interesting example of the sophists’ attitude to these problems is
provided by Antiphon, who in On Truth states that ‘we are all fitted
similarly by nature in all regards to be both barbarians and Greeks’
(87B44A, 2.10–15 D.-K. = 37D38b, 2.10–15 L.-M.). Many have
been inspired by this statement to attribute egalitarian or democratic
ideas to Antiphon. But the passage is open to other readings as well.
The reference to barbarians is actually used to deliver a subtle polem-
ical attack on the inconsistencies of the Greeks, who, while respecting
their gods and laws, display the same prejudices of which they accuse
barbarians: in doing so, Antiphon notes, ‘we have become barbarians
toward each other’ (87B44A, 2.7–10 D.-K. = 37D38b, 2.7–10
L.-M.). Apparently, this claim does not put forward any positive view
of barbarians; rather, in order to achieve the desired effect, it must
imply a negative evaluation: to criticize his interlocutors’ behaviour,
Antiphon adopts their perspective and shows that they (be they
Greeks or Athenians) behave exactly like the people they claim to des-
pise. The reference to barbarians serves an instrumental and polemical
purpose: it is used to stigmatize the incoherent conduct of the people
with whom Antiphon is polemically engaging. But this polemic can
hardly be taken as evidence of his role as a champion of the equality
between barbarians and Greeks.49
A no less delicate matter is the evaluation of another two testimonies,
concerning Hippias and Lycophron. In the aforementioned passage of

49
Decleva Caizzi 1986b. Antiphon’s polemic becomes even more significant if, as seems likely,
the sophist and the oligarchical rhetor were one and the same person: in this case, Antiphon would
have criticized slavery not on humanitarian grounds, but in order to once again highlight the limits
‘of the lame egalitarianism of the democratic polis’ (Canfora 2001: 215).

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


IV JUSTICE AND LAW 93

the Protagoras, Hippias addresses the following words to the other


characters of the dialogue:
I consider that you all belong to the same family, household, and city – by nature
( physis), not by law (or convention, nomos): for what is similar belongs by nature to
the same family as what is similar, whereas law (or convention, nomos), which is tyrant
over men, commits violence upon many things against nature.
(86C1 D.-K. = 36D17 L.-M.)

Aristotle informs us that Lycophron criticized the nobility,


arguing that ‘the beauty of nobility is invisible, dignity exists in
speech’ (83.4 D.-K. = 38D4 L.-M.). As in Hippias’ case, this polemic
against nobility, which is depreciated as a mere name with no basis
in reality, finds its starting point in the opposition between nomos
and physis (of which the opposition between reality and words is a
variant). Certainly, upon a first reading one gets a strong impression
that both authors are defending the cause of the natural equality of
human beings; moreover, this is what many authoritative scholars
have argued. But perhaps it should not be ruled out that this ‘bold
outlook on the old principles of the “noble stock”’ is aimed not at a
democratic defence of the rights of all people but rather at ‘replacing
one kind of inequality with an inequality of an even more radical
kind: what we have is the praise of the new superiority, and supremacy,
of intelligence’.50
Gorgias and Thrasymachus are credited instead with two stances in
favour of the equality of the Greeks against the barbarians: ‘trophies
won from the barbarians demand hymns, those from the Greeks dirges’
(Gorgias in the Funeral Oration, 82B5b D.-K. = 32D29 L.-M.); ‘Shall
we, who are Greeks, be slaves to Archelaus, who is a barbarian?’
(Thrasymachus in the oration For the Larissians, 85B2 D.-K. = 35D18
L.-M.). Yet it is almost certain that these are extemporary claims,
instrumental to the political speeches of which they are part:
The Funeral Oration, which he [sc. Gorgias] pronounced in Athens, was delivered on
those men who had fallen in the wars, whom the Athenian buried at public expenses
with eulogies; it is composed with extraordinary skill. For given that he wanted to
rouse the Athenians against the Medes and the Persians and was arguing with the
same intention as in his Olympic Oration, he said nothing at all about concord
among the Greeks, since he was speaking to Athenians (who desired supremacy,
which could only achieved by choosing a very energetic policy), but instead he praised

50
Isnardi Parente 1977: 32–3.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


94 IV JUSTICE AND LAW

at length the trophies won from the Medes, telling them that ‘trophies won from the
barbarians demand hymns, those from the Greeks dirges’.
(82A1 D.-K. = 32D27 and R19 L.-M.)

As far as Thrasymachus is concerned, his pan-Hellenic plea was direc-


ted against the expansionist aims of Archelaus, a Macedonian whom it
would have been an exaggeration to regard as a barbarian.51 Finally,
Alcidamas is credited with a resolute stance against slavery: ‘god has
set all men free; nature has made no man slave’ (fr. 19). But in this
case, too, we are dealing with a sentence taken from an oration
(to be more precise, a school exercise): it is likely to be a catchy line,
designed more to capture the public’s attention than to defend the
abolitionist cause.
In other words, an adequate contextualization of existing testimonies
often suggests that it would be unwarranted to consider the sophists
champions of human rights, by attributing to them a kind of awareness
that would only be fully developed over the course of the centuries. At
the same time, however, we should not go from one extreme to the
other, by concluding that their reflections have no value whatsoever.
Ultimately, driven by their eagerness to critically examine prejudices
and commonplaces, the sophists established themselves as the first
thinkers to expound theses of universal value, destined to have untold
consequences: ‘the struggle was destined to be long, but it had
begun’.52 And this is no small feat.

The enduring relevance of the sophists

The sophists’ discussions of the controversial phenomenon of justice


constitute a striking refutation of the interpretations of those scholars
who were only willing to acknowledge their importance in the field of
rhetoric. On the contrary, the significance of the problem of logos can
only be fully appreciated by framing it within a practical and political
context. Only at this stage can we grasp the sophistic movement in
all of its complexity. Moreover, only at this stage can we measure its
relevance with respect to our own world and interests. The contempor-
ary age is marked by a loss of foundations (or, rather, by the awareness

51
Bonazzi 2008: 66–8.
52
Guthrie 1971: 160.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


IV JUSTICE AND LAW 95

that there are no foundations); and this problem requires a practical as


well as a theoretical response. If this is the case, we are bound to
acknowledge the profoundness of the sophists, who were the first to
investigate such problems.
Even more relevant is the way in which the sophists tackled them:
whatever the thesis upheld, or the sophist involved, what distinguishes
their approach is their underlying realism – the disenchanted gaze of
someone who is not searching for easy justifications. Truly, it is difficult
not to agree with Nietzsche’s criticism of George Grote, who defended
the sophists by portraying them as perfect Victorian gentlemen. The
sophists’ chief merit – or limit, depending on one’s point of view – is
their realism:
The sophists are no more than realists. . .Grote’s tactics in defence of the Sophists are
false: he wants to raise them to the rank of men of honor and ensigns of morality –
but it was their honor not to indulge in any swindle with big words and virtues.53

Similar considerations, as we shall see, apply to another two topics that


are still constantly debated: happiness and the gods, or religion.

53
Nietzsche 1964: 249.

https://doi.org/10.1017/S0533245119000051 Published online by Cambridge University Press


V TEACHING VIRTUE: THE SOPHISTS ON HAPPINESS AND
SUCCESS

One frequently overlooked aspect of the sophists’ reflection concerns


the kind of problems that today might be included in an ethical treatise,
such as happiness and pleasure. Usually it is claimed that the sophists
shifted the focus of enquiry from nature to the human world, but then
the emphasis is almost exclusively placed on the political side of these
investigations (which is no doubt significant, as we have seen in the pre-
vious chapter). This is a misleading perspective, which applies modern
conceptual schemes to the ancient world, creating distortions and mis-
understandings. Before Aristotle, there is no point in distinguishing
between the ethical and the political spheres, since ethics and politics
ultimately constitute two aspects of the same problem: how to reach
fulfilment and thus lead a happy life. Certainly, in order to answer
this question it is necessary to embark on an analysis of political pro-
blems, such as justice or the best form of government: the fact that
human beings live together is something that must be taken into
account. But the ultimate goal is self-realization. This is the end
towards which everyone tends.
In the fifth-century debate, the issue at stake was that of arete, a term
which is usually translated as ‘virtue’: how can one become virtuous? In
our day, given our idea of virtue, a question of this sort may not seem
very pertinent, and one might object that virtue is rather an obstacle to
self-realization. For the Greeks, however, arete primarily indicates a
capacity to perform, which enables individuals to exercise their func-
tion or task in the best possible way: it is the quality (or range of qual-
ities) that helps one to stand out from one’s group and to fully realize
one’s potential, justifying one’s superiority over all others and earning
the individual admiration, along with material goods. An alternative
translation of arete, which better conveys this underlying meaning, is
‘excellence’. At the risk of oversimplifying things, the problem might
be then redefined in terms of success: how can one obtain success,
so as to lead a gratifying and happy life? This is an ever-relevant ques-
tion, which the Greeks never ceased to address.
In Greece, moreover, the debate on this issue was made even more
heated by the new social and cultural conditions which had emerged
in the fifth century. Whereas in the previous ages, dominated by the
aristocratic outlook, the prevalent thesis had been that a person’s

https://doi.org/10.1017/S0533245119000063 Published online by Cambridge University Press


V TEACHING VIRTUE 97

‘value’ derives from their natural endowments, which they need to cul-
tivate by following their ancestors’ example,1 in the fifth century, with
the rise of democracy, the belief took root that anyone could attain vir-
tue. This made the problem of education and teaching an even more
pressing one. Everyone could become ‘virtuous’, which is to say
develop qualities that would propel them to the summit of the commu-
nity; the problem, then, was to learn how to attain virtue. And this is
precisely what the sophists promised to teach, starting from
Protagoras, who claimed that what people could learn from him was
‘good deliberation about household matters, to know how to manage
one’s own household in the best way possible, and about those of the
city, so as to be most capable of acting and speaking in the city’s inter-
ests’ (Protagoras, 318b = 80A5 D.-K. = 31D37 L.-M.). In other words,
Protagoras promised to teach people virtue (Protagoras, 319b; see also
349a and Meno, 91a). Plato reports that Gorgias instead mocked
those who promised to teach others virtue (Meno, 95b = 82A21
D.-K. = 32D47 L.-M., quoted above, p. 75). Yet the purpose of his
teaching was much the same (see e.g. Plato, Gorgias, 452d–e).2 No
less significantly, even one of the Dissoi logoi, the sixth, is entirely
devoted to refuting the thesis that ‘wisdom and virtue can neither be
taught nor learned’ (90.6.1 D.-K. = 40.6.1 L.-M.). Once again, then,
the sophists prove capable of grasping and addressing the most import-
ant problems of the life of the city, eliciting opposite reactions – enthu-
siasm or aversion – through their bold claims.
Generally speaking, a defining feature of the sophists’ reflection is
the affirmation of the priority of the individual over the community.
The sources we have all appear to go in this direction, with one signifi-
cant exception: Protagoras. Unlike the other sophists, Protagoras seems
to be striving to reconcile the interests of the individual with those of
the public sphere, by emphasizing the eminently political dimension
of human nature. In this he once again shows significant affinities
with the democratic ideology of Pericles, and I will therefore refer
back to the previous chapter for a more detailed exposition of his
ideas. In all other cases, what makes the sophists’ analyses interesting

1
See, for example, Theognis, lines 27 ff.; Pindar, Olympian Odes, 2.86, 9.100, 10.20.
2
Also revealing, in this respect, is an epigram from Olympia that was written in his honour: ‘No
mortal ever invented a finer art than Gorgias / To exercise the soul in competition of excellence. /
And it is of him that, in Apollo’s hollows, the statue is dedicated, / A paradigm not of wealth, but of
the piety of his character’ (82A8 D.-K. = 32P34b L.-M.).

https://doi.org/10.1017/S0533245119000063 Published online by Cambridge University Press


98 V TEACHING VIRTUE

is their awareness of the problematic bond between each individual and


his or her community. Usually this leads to the (predictable) charge
that the sophists promote social egoism: their criticism that justice
and the law only possess conventional value, the doubts they repeatedly
raise concerning the existence of the gods – who were traditionally
regarded as the bastions of order and goodness – and their constant
references to personal interest may well be seen to expose them to
the charge of immoralism. No doubt, the sophists tackled some thorny
social questions, adopting an ambiguous stance vis-à-vis traditional
values, in a subtle play of rejection and adoption. But, while the
destructive aspect of their reflections is what most captures our atten-
tion, we should not overlook the many sources that inform us of how
they also put forward some positive suggestions. The picture of the
sophists as ‘immoralists’ is probably a misrepresentation of their posi-
tions, the outcome of a superficial application of certain theses that
have not been adequately understood, and which must be taken into
account without making them the one and only object of the sophists’
reflection.3

The immoralists

The idea that the sophists’ teaching amounted to a sort of immoralism


which justified any action whatsoever was a common one in fifth- and
fourth-century Athens, and finds authoritative confirmation in
Aristophanes and Plato.4 Arguably the most eloquent testimony is pro-
vided by the Clouds, in which Aristophanes unleashes his comic fury on
all the representatives of the new culture – not just the sophists, but
also natural philosophers and even Socrates, who constitutes the real
polemical target of the comedy. The central point made by the Clouds
is that the responsibility for the ruin of Athens ultimately falls on the
new ‘doctrines’ promoted by these self-styled masters, doctrines which
lead to (and justify) the most subversive act of all for the order of the
city: a son beating his father. From a social point of view, this is also a gen-
erational clash, which takes the form of a rejection on the part of the new
generations of those principles that had made the city of Athens great.5

3
See Bett 2002.
4
A detailed overview of these testimonies may be found in De Romilly 1988: 194–231; see also
Chapter 1, pp. 8–9.
5
Ostwald 1986: 229–50.

https://doi.org/10.1017/S0533245119000063 Published online by Cambridge University Press


V TEACHING VIRTUE 99

Compared to Aristophanes, Plato adopts a more nuanced and hence


more interesting position. First of all, it is worth noting that his polemic
seems to be more restricted: while, generally speaking, his dialogues
deliver an all-round attack against the sophists, when it comes to ethical
questions it is almost invariably Gorgias who is made the central target.
This is rather perplexing, considering that Gorgias even denied that he
dealt with virtue and claimed to confine his teaching to the art of speech
(consider the aforementioned testimony 82A21 D.-K. = 32D47
L.-M.). Plato, however, may have been right in a sense: for it is pre-
cisely on account of his alleged ‘neutrality’ that Gorgias might be
accused of being a bad teacher.6 Indeed, according to Plato, what
makes Gorgias dangerous as a teacher is the fact that, on the one
hand, he refuses to address moral issues but, on the other, promises
to teach how to achieve success: the combination of these two elements
may be exploited as a means to justify egoism, whereby the only
thing that matters is the pursuit of one’s goals, regardless of
other people’s point of view, needs, and rights. This is precisely the
criticism formulated by Plato, who accused Gorgias not of any moral
shortcomings (besides, the sources portray Gorgias as a sober and
moderate man), but on account of the noxious consequences of his
teaching.
Particularly revealing, in this respect, is the presentation of two
pupils of Gorgias – Meno in the dialogue of the same name, and
Callicles in the Gorgias – offering two different versions of the same
problem: a subdued version and a radical one. Meno is a fine example
of the first type, that of the shy pupil, who is eager to obtain the success
that Gorgias has promised him, but who is afraid of taking an open
stance against shared values such as justice. This results in a constant
oscillation between the egotistic pursuit of what might promote his self-
interest and a surface acknowledgement of traditional values. Hence,
Meno goes round in circles, without daring to pursue what is implicit
in Gorgias’ promises (the achievement of success) to its ultimate con-
sequences, yet also without understanding what might make traditional
values legitimate.7

6
See also Trabattoni 2000: 19–21.
7
In the Gorgias, a character who is presented in much the same light as Meno is Polus of
Acragas, historically another pupil of Gorgias and the author of a treatise on rhetoric. See, for
instance, Nails 2002: 252.

https://doi.org/10.1017/S0533245119000063 Published online by Cambridge University Press


100 V TEACHING VIRTUE

Callicles’ case is different, since he lacks neither courage nor frank-


ness. He is mostly known for his ‘political’ theories, which emphasize
the rule of force and the superiority of ‘the best’.8 As we have seen, it
is easy to show that these are only rhetorical quips, with no coherent
content. However, as Nietzsche clearly grasped, the same ideas acquire
a very different significance when viewed in relation to the life choices
of individuals. Struggling to reach an agreement with Callicles on the
issue of who ‘the best’ might be on the political level, Socrates tries
to shift the debate to the individual level: granting for the sake of argu-
ment that ‘the best’ have the right to rule ‘the weak’, Socrates asks
Callicles how ‘the best’ ought to behave towards themselves, whether
they ought to control themselves or not (Gorgias, 490e–491e).
Callicles initially does not understand Socrates, but then rejects his
idea. His reply perfectly illustrates what, according to Plato, was the
moral sickness of Athens:
How could a man prove to be happy if he’s enslaved to anyone at all? Rather, this is
what’s admirable and just by nature – and I’ll say it to you now with all frankness –
that the man who’ll live correctly ought to allow his own appetites to get as large as
possible and not restrain them. And when they are as large as possible, he ought to
be competent to devote himself to them by virtue of his bravery and intelligence, and
to fill them with whatever he may have an appetite for at the time. But this isn’t possible
for the many, I believe; hence, they become detractors of people like this because of the
shame they feel, while they conceal their own impotence. . . .The truth of it, Socrates –
the thing you claim to pursue – is like this: wantonness, lack of discipline, and freedom,
if available in good supply, are excellence and happiness; as for these other things, these
fancy phrases, these contracts of men that go against nature, they’re worthless
nonsense! (Plato, Gorgias, 491e–c)

The human being’s innermost nature lies in their desiderative compo-


nent, in the urges and passions that stir within them: it is there, in the
capacity to fulfil the needs and desires of one’s true self, that the quest
for happiness is conducted, against the conventions and limits of trad-
itional morality. While Nietzsche was to appreciate this answer, it was
overturned by Plato, who later on in the dialogue redefined it as a
defence of radical hedonism and refuted it as such – as a potentially
beastly solution and one utterly incapable of explaining what makes a
life truly excellent and worth living.
In its radical nature, however, Callicles’ challenge raises substantial
problems which Plato was well aware of and which he never ceased

8
See Chapter 4, pp. 77–80.

https://doi.org/10.1017/S0533245119000063 Published online by Cambridge University Press


V TEACHING VIRTUE 101

to investigate.9 First of all, it must be noted that Callicles is promoting


an extreme yet not utterly alien view of traditional morality, which ever
since Homer’s Iliad had rested on the logic of competitiveness and on
the friend/enemy dialectic. In particular, it is important to bear in mind
that, whether we share Callicles’ perspective or not, his challenge may
be interpreted as a particularly disturbing confirmation of a problem
that always needs to be addressed, that of obligation: why should I
limit my legitimate pursuit of happiness out of respect for others?10
This is the underlying question which drives the sophists’ reflection,
as well as that of other Platonic characters, such as the aforementioned
Thrasymachus and Glaucon, who newly address the issue in some
memorable pages of the Republic (Rep. 343a–344a and 358b–361d).
If happiness is the ultimate goal of human life, and respect for others
is a cause of unhappiness, then how are we to conduct ourselves?
Clearly, this is a dangerous question, and the frankness with which
the sophists posed it might create the impression that they did so in
order to promote theses such as that of Callicles. Yet Callicles’ ‘immor-
alist’ solution, his invitation to escape ‘moral captivity’, is only one pos-
sible answer. We do not know what solution Gorgias offered: in all
likelihood he offered no solution, in the sense that – as we have
seen – Gorgias avoided addressing problems of this sort. However,
the sources that we have provide information on two other sophists,
Antiphon and Prodicus; and they suggest that they were not immoral-
ists, but at most realists and utilitarians.

Antiphon’s morality of concord

Not unlike Thrasymachus or Glaucon, Antiphon was well aware of the


urgency of the questions just mentioned. As we have seen in Chapter 4,
the problem of how to relate to the law and other people was one of
the underlying themes of the analyses provided in On Truth. Elsewhere,
too, human happiness and that which hinders it constitute a privileged
topic for Antiphon. In a late source we find a curious anecdote, which
makes him a forerunner of psychoanalysis:
he invented an art of eliminating pain, just as there is a kind of treatment provided by
doctors for people who are sick. In Corinth he established an office next to the

9
Hobbs 2000: 137–74.
10
See Kerferd 1981a: 123.

https://doi.org/10.1017/S0533245119000063 Published online by Cambridge University Press


102 V TEACHING VIRTUE

marketplace and advertised that he was able to treat grief-stricken people by means of
his speeches. And he inquired into the causes and thereby consoled those people who
were suffering. But then, considering that this art was beneath him, he turned to rhet-
oric.
(87A6 D.-K. = 37P10 L.-M.)11

Regrettably, no traces of this art survive; however, the extant fragments


of another text, entitled On Concord can help clarify his thoughts on the
matter, when read alongside On Truth.
Certainly, the advice and suggestions that Antiphon used to give his
patients were not based on blithe optimism. On the contrary, the dom-
inant tone of the fragments is that of a disenchanted pessimism, typical
of much of the Greek gnomic tradition:
Living is like a day-long sentry duty, and the whole length of life is like a single day, as it
were, during which no sooner have we raised our eyes toward the light than we pass on
the baton to other people who come after us. (87B50 D.-K. = 37D51 L.-M.)
It is astonishingly easy to find fault with all of life, my friend, since it possesses noth-
ing exceptional nor great and imposing, but everything in it is little, weak, ephemeral,
and mixed with great pains (87B51 D.-K. = 37D50 L.-M.)

At first sight, these bitter descriptions of human life might seem like yet
another variation on the classic theme of human unhappiness, which
Antiphon presents through brilliant and evocative images. Upon closer
inspection, however, some original insights emerge. In particular, it is
worth noting that the undeniable pessimism of the fragments is con-
stantly associated with the theme of intelligence: according to
Antiphon, the true cause of distress, in general, is not the wretchedness
of the human condition, but the poor use to which people put their intel-
ligence, making things even more difficult instead of solving their pro-
blems.12 Life is complicated and the choices we have to make – when
it comes to things such as marriages, children, money, and friendship
(87B49, 53–8 D.-K. = 37D57, 52–3, 59, 47a–b, 54–5 L.-M.) – are diffi-
cult ones. Unfortunately, instead of using their intelligence to deal with
these thorny problems, humans often allow themselves to be driven by
impulses and passions to commit acts they will soon regret, but which

11
Cf. also 87A6 D.-K. (= 37P9 L.-M.): ‘When Antiphon had attained a great degree of per-
suasive power and was nicknamed “Nestor” because he could succeed in persuading people
when he spoke about anything, he announced that he would give lectures capable of eliminating
pain, as he supposed that no one could name to him a grief so terrible that he could not banish
it from that man’s thought.’
12
Gagarin 2002: 95.

https://doi.org/10.1017/S0533245119000063 Published online by Cambridge University Press


V TEACHING VIRTUE 103

cannot be erased, since ‘it is not possible to retract one’s life like a move
in checkers’ (87B52 D.-K. = 37D48 L.-M.); and ‘the time that they are
neglecting is gone’ (87B53a D.-K. = 37D53 L.-M.).
What is most striking about Antiphon’s distress is the lack of any
moralistic overtones and the lucid realism of his observations. Take
the example of wealth, a conventional good whose importance
Antiphon does not wish to dispute, and which constitutes a recurrent
theme in surviving testimonies (87A3, B53–4 D.-K. = 37D83, 52, 59
L.-M.).13 It is difficult to deny the importance of wealth, and this
view had been memorably expressed by Aristodemus, one of the
seven wise men: ‘Surely no witless word was this of the Spartan, I
deem, “Wealth is the worth of a man (chremat’aner); a poverty void of
esteem”’ (Alceus, fr. 49 Lobel-Page = D.L. 1.31). Whether we like it
or not, wealth is not something that can be ignored: to reject it com-
pletely, as though it served no purpose, is to adopt a fruitless position
that is of little use in everyday life. Instead, we should address various
concrete questions: if wealth marks the social value of a person, we
should also be concerned to ensure that it is acquired in an honest
and correct way; in turn, the problem of the acquisition of wealth
requires us to more deeply investigate its actual management, preserva-
tion, increase, circulation, and so on – and these spawned the first ‘eco-
nomic’ reflections. These are only some of the problems that Antiphon
dealt with, favouring the ‘mercantilistic’ view – a minority view at the
time – according to which the circulation of money was a legitimate
way to gain wealth, in opposition both to the traditional practice of saving
money and to those philosophers who merely criticized wealth, without
making any significant contributions to the debate (Xenophon informs
us of a polemic of Antiphon against Socrates on the matter, although it
is difficult to tell just how reliable this testimony is from a historical per-
spective14). Wealth was a concrete problem for which Antiphon suggests
concrete solutions, capable of making life’ more free and pleasant’
(Xenophon, Memorabilia, 1.6 = 87A3.3 D.-K. = 37D83.3 L.-M.). It is
easy to imagine that he may have formulated similar arguments concern-
ing other everyday problems, from marriage to work and friendship.

13
Predictably, given the importance of the problem for an expanding society such as
fifth-century Athens, Antiphon was not the only sophist to deal with the issue of wealth: see
Gorgias 82B20 D.-K. (= 32D40 L.-M.), Prodicus 84B8–9 D.-K. (=34R4 andP6 L.-M.), and
Anon. Iambl. 89.7.1 and 7.8 (40.7.1 and 8 L.-M.); see too Soverini 1998: 45–65 and 81–9;
Demont 1993; Bonazzi 2009a: 28–31 and 2016; Gavray 2016.
14
See Bonazzi 2009a.

https://doi.org/10.1017/S0533245119000063 Published online by Cambridge University Press


104 V TEACHING VIRTUE

With respect to these concrete analyses, what makes Antiphon’s pos-


ition even more interesting is his underlying psychological conception.
According to him, there are essentially two driving forces in human life:
reason and desire. The problem is to strike a balance between the two.15
Instead of generally censuring the predominance of desire, an obvious
or even trivial fact, we are to employ reason to fulfil those desires and
needs that are truly useful, so that they may contribute to our happi-
ness. Within this context we can grasp the importance of two key con-
cepts for Antiphon, sophrosyne (temperance) and homonoia (concord).
Temperance is the intelligence with which in each case we face or fulfil
our desires, so as to obtain what can truly bring us pleasure or profit:
One would attribute temperance (sophrosyne) most correctly only to that man who
blocks himself from the immediate pleasures of his desire and is himself capable of
dominating and defeating himself. But someone who wants to satisfy his desire imme-
diately wants what is worse instead of what is better. (87B58 D.-K. = 37D55 L.-M.)
Someone who has not desired nor touched what is shameful or evil is not temperate
(sophron): for there is nothing in which he shows himself to be well-ordered by domin-
ating it.
(87B59 D.-K. = 37D56 L.-M.)

The outcome of all this is concord, the inner balance that can be cre-
ated between the different driving forces of reason and desire that
guide human actions (see, for example, 87B44a D.-K. = 37R7 L.-M.,
whose attribution to Antiphon, however, is uncertain). Moreover, this
concord is the precondition for the enjoyment of genuine pleasure
(or at any rate for the avoidance of pain): Antiphon’s pessimism does
not rule hedonism out, but rather leads to it.16
Just as in On Truth, in On Concord Antiphon’s privileged object of
reflection would appear to be the single individual, who is different
from the ‘citizen’ of the polis or from the abstract ‘human being’ of
which philosophers all too often speak. Ultimately, the argument
expounded in On Concord seems to offer a positive counterpart to the
analyses of On Truth: Antiphon offers what we might describe as an eth-
ical solution to political problems, which is consistent with his emphasis
on the priority of the individual over the community (without ever for-
getting, however, that the distinction between ethics and politics had
not yet been systematized in the sophists’ day). As we have seen, On

15
Bonazzi 2006b.
16
Capra 1997: 298–303.

https://doi.org/10.1017/S0533245119000063 Published online by Cambridge University Press


V TEACHING VIRTUE 105

Truth had dispelled the illusion that nomos can solve the problems
which physis poses to individuals; for its part, On Concord shows that
only the individual can find a solution to their problems, to the extent
that they succeeds in striking a balance with themselves and with their
own needs and desires. In other words, by contrast to the positive
assessment of the social and collective dimension of human experience,
Antiphon focuses on the limits of coexistence and finds a way out, not
in the existence of shared rules, but in the individual’s capacity to face
life’s concrete problems in such a way as to derive the greatest benefit
and enjoyment. In opposition to the exterior consensus of nomos,
Antiphon thus invokes inner concord, which does not require any
legal approval in order to remain in force and which each person can
establish within him- or herself, by making intelligence prevail over
the destructive impulses of the passions. Through moderation and tem-
perance it is possible to restore the right balance with ourselves and
with the things that surround us: it is not the laws that ensure security
and happiness, but intelligence and the use that people make of it.17
Undoubtedly, Antiphon’s fragments contain none of the sort of revo-
lutionary truths expounded by those figures who sought to overthrow
the whole system of values of their community (let us think of
Callicles); but nor do they simply pander to widespread prejudices of
the sort that guide many people’s actions. Rather, we might speak
here of an intention to provoke, designed to drive people to become

17
Although it chiefly applies to ethical issues, it cannot be ruled out that this ‘morality of con-
cord’ also had more explicitly political consequences (Bonazzi 2006a). ‘Concord’ is one of the
central terms in the heated ideological and political debate that raged in the years of the
Peloponnesian War, so much so that it became one of the key terms that moderates and oligarchs
invoked against democratic nomos. Likewise, for Antiphon the concept of ‘concord’ does not
amount to a generic call to collective reconciliation, but rather serves to trace the profile of true
wise men, who alone ought to govern, since ‘nothing is worse for human beings than lack of
rules (anarchias)’ (87B61 D.-K. = 37D63 L.-M.). Concord, then, emerges as the capacity to
take care of oneself and others by curbing violent and irrational impulses. Obviously, a thesis of
this sort potentially applies to all people, but it is easy to see how the concept could be exploited
by the oligarchic faction as a powerful way to criticize the alleged freedom of the democrats:
according to oligarchic propaganda – as illustrated by the anonymous author of the Constitution
of the Athenians – democracy is a system which justifies the unruliness of the masses, which allows
the masses to yearn for and achieve more and more, thereby laying the ground for the outbreak of
increasingly violent conflicts. When set in its context, this praise of self-control and privileging
of intelligence over law lose much of their apparently generic quality and prove closely reminiscent
of the typically aristocratic themes that characterize pro-Spartan polemics. Regrettably, the limited
number of surviving fragments of On Concord do not allow us to further investigate this hypothesis.
Be that as it may, the fact that considerations of this sort could be exploited for political and espe-
cially anti-democratic purposes is confirmed by the theses of another opponent of (democratic)
nomos, the oligarch Critias: see e.g. H. Patzer 1974 and Bonazzi 2018: 33–41.

https://doi.org/10.1017/S0533245119000063 Published online by Cambridge University Press


106 V TEACHING VIRTUE

more aware of what they are and of what they do, so as to assign priority
to what truly matters: the exercising of one’s own intelligence, since
‘thought (gnome) leads the body, for all humans, toward health and dis-
ease and toward everything else’ (87B2 D.-K. = 37D1b L.-M.). When
it comes to the issue of happiness, therefore, Antiphon once again
proves himself to be a subtle thinker who exemplifies what is arguably
the most distinctive and interesting aspect of sophistry, namely its cap-
acity to rethink the foundations and assumptions of the Greek tradition,
through a subtle dialectic of correction and transformation.

Prodicus’ (and Hippias’) choice

In an article published in 1930, the great historian of antiquity Arnaldo


Momigliano developed an intriguing interpretation of Prodicus, in
marked opposition to the relativism dominant in the ‘ranks of the
Thrasymachuses and Callicleses’. At least on the ethical level, unlike
in the case of the debate on the origin of the gods (see Chapter 6),
Prodicus appears to have aligned himself with traditional morality,
defending the need for strong moral principles in everyday life. To sup-
port his reconstruction, Momigliano relied on two elements, the pre-
dominantly ethical focus of Prodicus’ research on language, and
especially his famous apologue The Choice of Heracles, reported by
Xenophon.18 One noteworthy aspect of Prodicus’ almost obsessive inter-
est in linguistic distinctions concerned ethical terms, and in particular
the distinction between terms with positive and negative meanings: this
might suggest that one of the primary aims of his research was to curb
the linguistic conventionalism common in his day, and the practical rela-
tivism stemming from it.19 Moreover, the idea that the defence of

18
Xenophon, Memorabilia, 2.1.21–34 (= 84B2 D.-K. = 34D21 L.-M.). Recently, Dorion 2008
has denied that the passage from Xenophon is a reliable testimony on Prodicus, arguing that the
apologue is used to investigate ethical issues addressed by Xenophon’s Socrates. This is an intri-
guing hypothesis, which highlights some interesting affinities between Socrates and Prodicus, but
which fails in its attempt to disprove Prodicus’ authorship. Certainly, this text is not to be taken as
a word-by-word quotation (as suggested instead by Sansone 2004, criticized by Gray 2006);
rather, in the light of the introductory expressions used by Xenophon, it is more reasonable to
assume that what we have is a summary of Prodicus’ argument: see now Mayhew 2011: 201–21.
19
Momigliano 1930: 102–3. The frequency of ethical terms among Prodicus’ synonyms has
also been noted by many other scholars: see esp. De Romilly 1986, Dumont 1986, Tordesillas
2004, and Wolfsdorf 2008a. However, in many cases it is difficult to establish whether we are deal-
ing with reliable testimonies (as in the case of 84A19 D.-K. = 34D6a L.-M.) or mere parodies (as
is probably the case with 84A15 D.-K. = 34D22 and 23–5 L.-M.). A testimony by Galen (84B4

https://doi.org/10.1017/S0533245119000063 Published online by Cambridge University Press


V TEACHING VIRTUE 107

traditional morality was one of the cornerstones of Prodicus’ reflection


would appear to find eloquent confirmation in The Choice of Heracles,
whose edifying tone was widely appreciated even by readers usually hos-
tile to the sophists’ teaching, such as Xenophon himself.
It seems as though Prodicus’ case was not an isolated one.
Broadening our enquiry, we might find an interesting parallel in
Hippias’ Trojan Discourse (86B5 D.-K. = 36D5 L.-M.), which is
another example of what a public declamation by a sophist probably
sounded like, and which seems to share the same aims as The Choice
of Heracles. As far as we can tell, the underlying topic of the two
speeches is the same: to provide moral advice to young people who
are entering adulthood and hence are about to join the true life of
the polis. Prodicus relates that Heracles was considering what conduct
to adopt in his life, when he met two female figures. The first
woman, who is called Happiness by her friends and Vice by her detrac-
tors, wore flashy clothes and make-up, and immediately stepped for-
ward to promise the hero a life of pleasure. The second woman,
Virtue, was instead sober and discrete, and promised the young hero
toil but also fame and glory. Although this is not explicitly stated in
the text, it is easy to infer that Heracles chose the second woman,
who promised to make a model citizen and man of him. Regrettably,
Hippias’ Trojan Discourse has not been preserved, yet in antiquity it
was no less famous than The Choice of Heracles: against the background
of the siege of Troy, the wise Nestor gives young Neoptolemus some
practical advice on how to achieve fame and success. It seems as though
both texts presented the kind of arguments that might have pleased par-
ents who were concerned about their sons’ future. Hippias’ position
could be compared to Prodicus’ and his defence of the need for strong
moral principles in everyday life, in opposition to the more radical the-
ses developed by other sophists. Hence, what emerges is a rather clear-
cut rift between the various sophists on the issue of virtue and
happiness.
This reconstruction, which is not an implausible one, once again
confirms that we cannot regard the sophists as exponents of a compact
and homogeneous movement. However, a closer inspection of the
available sources also suggests that we should avoid drawing sharp con-
trasts and tone down the opposition between subversive and egoistic

D.-K. = 34D9 L.-M.), concerning phlegma (inflammation), shows that Prodicus was not exclu-
sively interested in ethical terms.

https://doi.org/10.1017/S0533245119000063 Published online by Cambridge University Press


108 V TEACHING VIRTUE

sophists on the one hand and sophists open to tradition on the other.
We have already noted how immoralism resulted more from a negative
interpretation of their views by their detractors (Aristophanes, Plato,
and others) than a position which was actually upheld by the sophists.
It is a matter now of reassessing Prodicus’ and Hippias’ theses with
respect to traditional knowledge and values, and of verifying that
some apparent banalities20 conceal interesting insights, which can be
fruitfully compared with what has been noted with regard to other
sophists. The freedom and heterogeneity of the sophists does not
exclude a range of common interests and ideas.
Given the almost complete absence of any information on Hippias’
discourse, we should focus our attention on Prodicus. Undoubtedly,
his discourse complied with the civic ideology that assigned glory to
those who had committed themselves to the public good, resisting
the temptation of immediate gratification (84B2.28 D.-K. =
34D21.28 L.-M.). Still, we must be careful to avoid drawing a clear-cut
contrast between pleasure and virtue, as though these were two incom-
patible concepts. Prodicus’ aim is not to teach contempt for pleasure
and encourage the virtues of toil and hard work, but rather to show
that true pleasure, the kind of enjoyment which can accompany a per-
son throughout their life, often stands in opposition to immediate grati-
fication, which can soon turn into pain and affliction. It is not a matter,
then, of drawing a contrast between pleasure and hard work, but of
knowing what is truly good and hence what really brings pleasure
and happiness:
A life entirely focused on eating before being hungry, on sleeping. . .because one has
nothing to do, on pursuing sexual enjoyment without feeling any sexual urge or need
is certainly not a pleasant life. What Virtue champions, then, is not the negation or con-
demnation of pleasure: it is not a matter of seeking out pleasures in themselves. . .but,
on the contrary, of only indulging in them when this means meeting the vital needs of
man, for only then do they truly bring enjoyment.21

In addition to this, we must bear in mind the utilitarian and individu-


alistic perspective on which the whole discourse is based. The choice of
leading a virtuous and noble life is not an end in itself; rather, it is made
because of the advantages it brings: Prodicus does not invite us to

20
Guthrie 1971: 277–8.
21
Casertano 2004: 78. Some very interesting observations on this hedonistic perspective may
be found in Capra 1997: 289–98.

https://doi.org/10.1017/S0533245119000063 Published online by Cambridge University Press


V TEACHING VIRTUE 109

sacrifice ourselves for the public good, but shows that, ultimately, a life
of commitment will prove more enjoyable and convenient; and it is for
this reason, on account of its usefulness, that it is preferable.22 No
doubt, in his exhortations Prodicus does not oppose traditional morality,
but rather draws upon many of its ideas (cf. Hesiod, Works and Days,
286–97; Simonides, fr. 386 Page). But this does not imply a clear
break with respect to the other sophists: Prodicus’ emphasis on personal
advantage and the centrality of the individual, just like his interest in the
issue of pleasure, is undoubtedly reminiscent of Antiphon.23 In other
words, like Antiphon, Prodicus would appear to adopt a nuanced stance
with respect to traditional morality, which is not rejected (he does not
deny the importance of honour, just as Antiphon does not deny the
importance of wealth) but is rather reassessed according to a different
hierarchy of values (where the individual comes before the community).
To further clarify the complexity of Prodicus’ approach to tradition,
we should not overlook the self-promotional tone of his discourse: the
purpose of a discourse, of a public epideixis, is not simply to convey an
argument but to advertise a teacher.24 Significantly, the distinctive fea-
ture of the discourse, as it has reached us, is not so much the fact that
Heracles chooses the path of virtue as the fact that he does so after
debating two opposing theses.25 In such a way, the discourse implicitly
yet quite clearly opposes the traditional conception of education, which
was often seen as consisting in the passive learning of precepts and
norms of conduct: on the contrary, true education must prepare us
for making distinctions and assessments. This, in turn, highlights the
central importance of the teacher, who alone can teach us how to rea-
son and argue correctly. Heracles was an ambiguous figure, known for
having rid the world of monsters (Pindar, Nemean 3) but also for yield-
ing to anger, desire, and pleasure (consider, for instance, Euripides’
Alcestis, Sophocles’ Women of Trachis, and Aristophanes’ Frogs). This
makes him the ideal pupil for a sophist, and Prodicus can portray

22
In this respect, it has been observed that the criterion of usefulness appears to have played a
part even in the case of linguistic distinctions: in other words, it is reasonable to suggest that the
primary aim of the distinction drawn between positive and negative terms was to identify what
course of action one ought to follow. See Dumont 1986 and Tordesillas 2004: 61–3.
Notwithstanding the reservations expressed in n. 19 above, this is an intriguing hypothesis.
23
Significantly, Prodicus would appear to share Antiphon’s dualist psychology, if De Romilly
1986: 6 is correct in noting that his linguistic distinctions are always based on a contrast between
terms associated with the spheres of irrationality and rationality.
24
See the crucial analysis provided by Morgan 2000: 106–15.
25
Dorion 2009b: 536 n. 34.

https://doi.org/10.1017/S0533245119000063 Published online by Cambridge University Press


110 V TEACHING VIRTUE

himself as the perfect teacher, whose merit lies not in his conformity to
traditional values (something expected in public performance), but in
his capacity to teach one how to put forward reasons for each matter
at issue and adequately assess them. Once again, what distinguishes
the sophists’ discourse is its dynamic relation with tradition, which
combines the borrowing of certain ideas and a degree of conformity
(Heracles ultimately chooses virtue) with certain changes. This is all
the more true given that the tale of Heracles at the crossroads was
invented by Prodicus himself, who thereby adapted myth to suit his
own educational purposes.26
The same kind of tension is also to be found in Hippias’ Trojan
Discourse (86B5 D.-K. = 36D5 L.-M.), where the protagonist’s choice
is even more indicative of the provocative character of the sophists’
message. Whereas Heracles remained an ambiguous figure,
Neoptolemus was an infamous one, insofar as during the siege of
Troy he had killed the elderly Priam, who had taken refuge at the
altar of Zeus. The dramatic context of Hippias’ speech is therefore highly
revealing, because it is precisely after the seizing of Troy, which is to say
just after Neoptolemus’ terrible crime, that Nestor addresses words of
advice to the young man. Neoptolemus’ case is even more problematic
than that of Heracles, but it is also better suited to the world of the
second half of the fifth century, a world that was experiencing a dramatic
war and crisis; and Hippias, like a new Nestor capable not only of edu-
cating the young but of leading them to success, can present himself as
an ideal master for these tumultuous times.27 Like Prodicus, he applies a
subtle strategy of appropriation of tradition and its values: he does not
criticize its assumptions or principles, but exploits them for his own pur-
poses. If we think of the way in which Protagoras and Gorgias had
engaged with poetic lore (see Chapter 3), here too the divergence from
the other sophists turns out to be less marked than what at first appeared
to be the case. All sophists, each in his own way, present themselves as
masters of virtue; and this is why Plato criticized them so harshly, in
the belief that their teaching, far from leading one to the true good,
only exacerbated the defects of traditional morality.

26
Kuntz 1993.
27
In this respect, it is interesting to note that, in the Protagoras, Plato parodies Hippias in the
figure of Nestor: see Brancacci 2004. Nestor is certainly a paradigmatic figure, also apparently
evoked by Gorgias (82B14 D.-K. = 32D52 L.-M.) and Antiphon (87A6 D.-K. = 37P9 L.-M.).

https://doi.org/10.1017/S0533245119000063 Published online by Cambridge University Press


VI THE GODS AND RELIGION

The theme of the gods, of their existence, appearance, and interest in


human affairs, was an object of constant attention for the Greeks
from the most ancient times. As is widely known, Greek religion
presents some highly original features compared with present-day
religions: there are no religious texts establishing a body of orthodox
doctrines, and no figures to whom worship is officially entrusted.
Perhaps precisely because of this fluid situation, the element of the div-
ine marks almost all salient moments in the lives of individuals and cit-
ies.1 This constant presence in itself explains why, from Homer
onwards, all writers were so keen to explore the issue of the gods. In
particular, many Presocratic philosophers staunchly criticized popular
prejudices (often presenting the gods as thieves, adulterers, and sedu-
cers) by suggesting alternative and more rigorous conceptions. In
doing so, the Presocratics changed the content of the divine, without
denying its existence:2 they opposed the theology of the philosophers
to the theology of the poets. The sophists also fit within this broader
movement towards the critical redefinition of traditional religiosity, a
movement which engaged many leading personalities of the age: let
us think of Euripides, of Thucydides’ analysis of the plague, or of the
debate on the ‘sacred disease’ (epilepsy) among Hippocratic doctors.
In particular, the sophists stand out on account of what we might
term their ‘sociological’ perspective: while criticizing the phenomenon
of religion, they acknowledged its importance in human life; searching
for its causes, they embarked on enquiries into the nature of the gods,
the origin of the belief in their existence, their role in people’s lives, and
myths as a means to convey traditional values.
The outcomes of these enquiries are often provocative, to the
point that (in some cases, at least) they border on atheism, a
multifaceted concept:3 some sophists are included in the lists of atheists
that circulated in the Hellenistic and Imperial centuries.4 This partly

1
Muir 1985:193–5.
2
Kahn 1997: 250–3.
3
In the ancient world, the term ‘atheist’ did not only describe someone who denies the exist-
ence of the gods, but also someone who scorns the gods or has been abandoned by them (see
Winiarczyk 1990). In the following pages, the term will mostly be used in its modern sense. In
general on ancient atheism, see now Sedley 2013, Whitmarsh 2015, and Gourinat 2019.
4
Usually these lists included the names of Prodicus, Diagoras, Critias, Theodore of Cyrene,
Euhemerus of Messene, and – with some reservations – Protagoras: see Gigon 1985: 423.

https://doi.org/10.1017/S0533245119000075 Published online by Cambridge University Press


112 VI THE GODS AND RELIGION

explains the violent reactions to their teachings in Athens. Given the free-
dom with which the problem of the divine was investigated from the most
ancient times, one might have expected to find a certain degree of toler-
ance towards debates and criticism even in the fifth century.5 Instead,
the opposite was the case, especially in the second half of the century,
when war raged between Athens and Sparta. While Dodds may have
gone too far in describing this as a period of resurgent irrationalism,
excesses occurred in both directions.6 Probably in 437 BC, Diopeithes’
decree was issued to denounce those who did not believe in the existence
of the gods or spoke on celestial matters (ta meteora).7 On the opposite side
of the spectrum, we know that, a few years later, a club of kakodaimonists
(‘companions of the evil daemon’) was founded, whose members would
meet to enjoy a banquet on what were regarded as unlucky days. Finally,
we have the case of the Hermocopids (those who smashed the Hermes sta-
tues in Athenian streets) and the profanation of the Mysteries in 415 BC.
Within this context, it appears that many representatives of the new cul-
ture were brought to trial.8 Then, of course, we have the most famous
trial of all: we should not forget that one of the two charges in the trial
which led to Socrates’ death was that the philosopher did not respect
the traditional gods and had introduced new deities. The sophists appear
to have been both a symptom and a cause of this climate of unrest.9 Much
of their fame, and of the hatred they attracted, depends precisely on their
provocations in this field.

Protagoras’ agnosticism

Protagoras’ penchant for sensational statements, capable of seizing the


public’s attention, is further confirmed by the opening of another text
by him, probably entitled On the Gods:10

5
Betegh 2006: 625.
6
Dodds 1951: 188–95.
7
The authenticity of this decree has been repeatedly disputed; a recent defence of its historicity
is in Whitmarsh 2015: 117–19.
8
Trials have been recorded for Anaxagoras, Diagoras, Protagoras, Prodicus, Euripides,
Phidias, and Aspasia. However, in particular with regard to Protagoras and Prodicus, it seems
that the sources are not always reliable: see the overview provided by Ostwald 1986: 528–36,
and Bonazzi 2018: 43–7.
9
Burkert 1985a: 311–17.
10
At present, it is still difficult to establish the relation between this text and the Truth. What is
more interesting to note is that the sources also credit Protagoras with a treatise On What Is in
Hades (80A1 D.-K. = 31D1 L.-M.), which would appear to confirm his interest in the topic.

https://doi.org/10.1017/S0533245119000075 Published online by Cambridge University Press


VI THE GODS AND RELIGION 113

About the gods I am able to know neither that they exist nor that they do not exist nor of
what kind they are in form: for many things prevent me for knowing this, its obscurity
and the brevity of man’s life. (80B4 D.-K. = 31D10 L.-M.)11

According to one ancient tradition this statement had a disproportion-


ate effect, far beyond what the sophist could imagine: ‘because of the
beginning of his book [i.e. On the Gods], Protagoras was expelled by
the Athenians, and they burned his books in the marketplace after
they had been collected by a herald from everyone who owned them’
(80A1 D.-K. = 31P19); and ‘while he was travelling from the mainland
to the islands and avoiding Athenian triremes, which were deployed on
all the seas, he suffered shipwreck while sailing a small boat’
(80A2 D.-K. = 31P21 L.-M.). According to Philostratus, the source
of the second quote, Protagoras’ impiety was due to his Persian educa-
tion – as though it were impossible for a Greek to come up with such
ideas.12
Be that as it may, the sophist was made the object of unanimous and
enduring condemnation, to the point of bringing together traditionally
antithetical schools of thought, such as the Christians and the
Epicureans: Protagoras of Abdera, the son of Manander, stated that
neither gods nor God exist at all (Epiphanius, Against Heresies, 3.16),
and
Protagoras of Abdera in effect maintained the same opinion as Diagoras, but he
used different words on the idea that he would thereby avoid its excessive temerity.
For he said that he did not know whether the gods exist. But this is the same
as saying that one knows that they do not exist. (Diogenes of Oenoanda, 80A23
D.-K. = 31R24 L.-M.)

However, the anecdotes we have regarding Protagoras’ life are probably


false,13 and there is no reason to assume a profession of atheism on his
part:
Diogenes crassly conflates a profession of knowledge (‘I know that not-P’) with a con-
fession of ignorance (‘I do not know that P’). To the believer, agnostics may be as bad
as atheists; but to the atheists agnostics are not much better than believers.14

For a recent and very detailed analysis of this fragment, see Corradi 2017 with further
bibliography.
11
On the text, see Di Benedetto 2001.
12
On Protagoras and the Magi, see Gigon 1985: 427–30.
13
See Corradi 2012: 31–43.
14
Barnes 1979: ii.449–50.

https://doi.org/10.1017/S0533245119000075 Published online by Cambridge University Press


114 VI THE GODS AND RELIGION

In other words, this is not an ontological thesis but an epistemological


one:15 Protagoras is the first thinker to take an agnostic stance, which
can hardly be understood as an expression of atheism.
But while it is evident that Protagoras’ statement is not a profession
of atheism, its overall meaning is less clear. Some interpreters have even
gone so far as to deny that he is doubting the existence of the gods in
this passage. This position has been endorsed in particular by
Kerferd, who has established a parallel between the use of hos in this
statement and in the ‘man-measure’ one: in the latter case, as
we have seen, the preposition clearly has a modal meaning (see
pp. 17–18); if we assume that it was used in the same sense in the for-
mer passage, we must conclude that Protagoras was chiefly interested in
the way in which the gods are and present themselves, rather than in
their presumed existence or non-existence. He would thus be claiming
that we do not know ‘how’ the gods are or are not, and the words that
follow (‘nor of what kind they are in form’) would have an almost
epexegetic function (or would more concretely refer to the problem
of their visible appearance16). This would imply, then, that Protagoras
never raised any doubts as to the existence of the gods. Against this
reading, however, it must be noted that all the ancient sources, from
Plato onwards (see Theaetetus, 162d), unanimously present Protagoras’
thesis as though it concerned the existence of the gods. As there are
no reasons to go against such a solid tradition, the most reasonable
hypothesis is that the passage discussed both the existence of the
gods (‘neither that they exist nor that they do not exist’) and their
form (‘nor of what kind they are in form’17). Incidentally, it is worth
noting that the statement possibly presents the first recorded use of
the verb ‘to be’ in an existential sense.18
The reasons which Protagoras reportedly adduced in support of
his thesis – the obscurity of the problem and the brevity of human
life – have been regarded as too banal, or at any rate not particularly
relevant.19 However, they acquire greater significance when viewed in
relation to a concept the importance of which has already been stressed
in relation to the ‘man-measure’, namely personal experience: this is

15
Drozdek 2005: 41.
16
See Kerferd 1981a: 165–9.
17
See Di Benedetto 2001.
18
Kahn 1973: 302.
19
Barnes 1979: ii.450.

https://doi.org/10.1017/S0533245119000075 Published online by Cambridge University Press


VI THE GODS AND RELIGION 115

the only legitimate criterion to verify our knowledge.20 Therefore it


seems that, by highlighting the limits of human experience,
Protagoras was engaging in a polemic with those Presocratic thinkers,
such as Parmenides and Empedocles, who had repeatedly claimed to
be able to transcend the limits of human knowledge, as well as the
poets, who had based their authority on a privileged contact with the
divine world. In this respect, some interesting similarities are to be
found between Protagoras and the sixth-century poet and moralist
Xenophanes of Elea (21B34 D.-K. = 8D49 L.-M.), who also reportedly
criticized human claims to know the truth about the gods (significantly,
for Xenophanes too, obscurity is an obstacle to knowledge).21 The ana-
logies with Xenophanes become even more striking when we consider
the second segment of the first statement, concerning the presumed
appearance of the gods. In raising this problem, Protagoras was in all
likelihood referring to the anthropomorphic conception, which
Xenophanes had already repeatedly censured.22 And in doing so,
Xenophanes had not been alone either, because we can also add
Herodotus’ words:23 ‘what appearance the gods have’ is something
which Homer and Hesiod had established for the Greeks, but it was
only a poetic fiction (Herodotus 2.53.1). The polemic does not concern
only the Presocratics but also, and most importantly, traditional forms
of religious expression.
One intriguing problem for scholars has been to reconstruct the
follow-up to this text. It is difficult to imagine that, after a statement
of this sort, Protagoras continued to speak at length about the gods.
The structure of the sentence (and in particular the presence of the par-
ticle men, which implies a contrasting de) suggests that the author
shifted his attention onto human beings: for the Greeks, the notions
of god and human were mutually defining.24 As Werner Jaeger has sug-
gested, it is likely that, in the continuation of the text, Protagoras

20
Mansfeld 1981: 41–2.
21
It is interesting to note that a statement similar to Protagoras’ has been attributed to
Parmenides’ pupil Melissus, who was a polemical target of the sophists: ‘He said that we ought
not to make any statements about the gods, for it was impossible to have knowledge of them’
(D.L. 9.24).
22
Sassi 2013; see also Corradi 2017: 456–61. Alternatively, Protagoras may have been referring
to the repeated transformations of the gods in Greek myths: this too militated against the possibil-
ity of saying anything definite about their form (Drozdek 2005: 42).
23
On the relation between this passage from Herodotus and Protagoras, see Burkert 1985b,
who has emphasized the dependence of the former on the latter, and the words of caution voiced
by Sassi 2013.
24
Mansfeld 1981: 43.

https://doi.org/10.1017/S0533245119000075 Published online by Cambridge University Press


116 VI THE GODS AND RELIGION

discussed not the problem of the gods in itself, but rather the issue of
what they represent for human beings, thereby inaugurating an
anthropological and sociological approach that considers religion in
terms of its function for human civilization and society.25 This is no
doubt a noteworthy suggestion. What is less convincing is the way in
which this hypothesis has been developed. From Jaeger onwards, scho-
lars have believed that, while leaving the problem of the actual existence
of the gods open, Protagoras celebrated their importance for human
civilization.26 Given the silence of fragment B4, this hypothesis is
based on the myth that Protagoras relates in the dialogue named after
him. Yet, even leaving aside the problem of the Protagorean origin of
this page of Plato’s (a far from trivial problem27), what may be inferred
from the myth is in fact the very opposite. In the myth, Protagoras
acknowledges the importance of belief in the gods as something specif-
ically and universally human; but the very moment that he acknowl-
edges the presence of the religious dimension, he de facto limits its
importance by implying that this belief (or fear) is not enough to pro-
vide a foundation for human society. What ensures that people can
live together in cities is rather politics, which is the object of the soph-
ists’ teaching. Protagoras’ agnosticism is therefore instrumental for his
radical humanism.28

The origin of the belief in the gods

As far as we can tell, Prodicus investigated not so much the problem of


the gods themselves as the origins of the human belief in their existence
(84B5 D.-K. = 34D15–18 L.-M.). In all likelihood, in this respect too it
was Prodicus’ interest in names, a distinguishing feature of his activity
as a sophist, that conditioned his reflections. According to his

25
Jaeger 1947: 176. See also Corradi 2012: 169–70, on the basis of Plato, Cratylus, 400d–401a.
26
In addition to Jaeger, consider for instance Schiappa 1991, pp. 141–148 or, more recently,
Drozdek 2005.
27
See Bonazzi 2011.
28
As further confirmation of this humanism, one might recall the ideas on crime and punish-
ment that Plato puts into Protagoras’ mouth in the dialogue named after him (Prot. 323c–324d;
however, it is difficult to establish whether these are ideas that may historically be traced back
to Protagoras: see Saunders 1981). The underlying thesis – that a just punishment must not be
a reprisal, but must rather serve as a means of correction and deterrent – stands out insofar as
it aims to investigate the problem of injustice and evil among human beings without invoking,
or confiding in, any corrective action on the gods’ part.

https://doi.org/10.1017/S0533245119000075 Published online by Cambridge University Press


VI THE GODS AND RELIGION 117

perspective, the gods are a human product, insofar as people have cho-
sen to regard as ‘divine’ that which has proven useful to their lives: at
first, things such as the sun and moon, rivers and fruits; later, ‘as
human wit entered into competition with nature’,29 this was also
extended to those people who had distinguished themselves with inven-
tions that were particularly useful for the well-being of humanity.30
According to Henrichs’ reconstruction, a transition was essentially
made from an impersonal stage, in which beneficial objects were deified
(e.g. bread or wine), to a personal stage, in which the object of deifica-
tion were those people to whom the invention could be attributed, or
who at any rate had taught others how to correctly employ useful things
(as in the case of Demeter and Dionysus31). The gods and religion,
then, are merely a human product resulting from humans’ attempt to
make sense of the environment in which they live.32 Undoubtedly,
the boldness of these ideas, which ‘make the gods come down from
their pedestal’;33 the parallel reassessment of the human world, which
is capable of making sense of reality; and the emphasis on usefulness
as the distinguishing criterion to understand and guide one’s actions
make this theory a typical testimony to the sophists’ way of reasoning.
Not unlike in Protagoras’ case, the limited number of testimonies we
have prevents us from precisely reconstructing the context of Prodicus’
theories on the matter. It cannot be ruled out that somehow – and,
once again, as in Protagoras’ case – this discussion concerning the
gods constituted a prelude to an analysis of human civilization.34 Yet,
as far as we can infer from a passage by Themistius (a rhetor from
the age of Imperial Rome: 84B5 D.-K. = 34D17 L.-M.), Prodicus
would appear to have stressed the importance of agriculture rather
than politics, as Protagoras did in the Platonic dialogue. It might be
hypothesized, therefore, that

29
Henrichs 1975: 111.
30
Guthrie 1971: 238–42 voiced some reservations with regard to the two-phase reconstruction
of Prodicus’ theory (which entails first the deification of inanimate objects and then that of particu-
larly distinguished human beings), by arguing that the available sources allow us to attribute to
Prodicus only the deification of inanimate objects and not of humans as well (the same view is
taken by Gomperz 1912: 113 n. 251). However, an oft-neglected testimony, a papyrus fragment
from a treatise On Piety by the Epicurean Philodemus (PHerc. 1428 = 84B5 D.-K. = 34D15
L.-M.), has made it possible to cast these reservations aside and to attribute the whole theory to
Prodicus: see Henrichs 1975: 112–23; Mayhew 2011: 180–3.
31
See Untersteiner 1947.
32
Henrichs 1975: 112.
33
Dorion 2009a: 348.
34
Nestle 1936.

https://doi.org/10.1017/S0533245119000075 Published online by Cambridge University Press


118 VI THE GODS AND RELIGION

Prodicus reconstructed a Kulturgeschichte in which the progressive improvement of


humankind’s living conditions was first of all connected to the progress made in the
cultivation of the land; and that particular attention was devoted to the relation between
civilization, agriculture, and religion.35

In the context of the Greek world, whose calendar and festivities


revolved around the key moments in the agricultural year, the link
drawn by Prodicus would have been a particularly incisive one.36
With some caution, Prodicus’ view may be regarded as an expression
of atheism.37 An even more explicit atheistic stance has been recorded
in an extensive fragment from a satyr-play entitled Sisyphus.38 This text
had long been attributed to Critias, and actually constituted one of the
main reasons for including this author among the sophists (88B25
D.-K. = Dram. T63 L.-M.). However, the attribution has been dis-
puted, and the name of Euripides has been suggested instead, based
on the fact that in 415 BC the tragedian composed a satyr-play entitled
Sisyphus to accompany a trilogy consisting of Alexander, Palamedes, and
Trojan Women.39 Whether the author is Critias or Euripides, it is
important to avoid the mistake of assuming that these verses
express his personal views: the lines are spoken by a mythological

35
Soverini 1998: 105. Further developing his hypothesis, Nestle 1936 had set these doctrines in
relation to the famous testimony on Heracles’ choice (see Chapter 5, pp. 106–110): indeed, this
argument was featured in a text entitled Horai (84B1 D.-K. = 34D19 L.-M.), and it may reasonably
be assumed that the term referred to the seasons, and therefore that Prodicus celebrated Heracles as
a symbol of the world of farmers as well. According to this view, then, the Horai provided a praise of
agriculture, a theory on the origin of religion connected to agriculture, and an exhortation to virtue
(something necessary for agricultural life), embodied by the figure of Heracles at the crossroads.
This hypothesis – an intriguing one, yet difficult to prove – may find some confirmation in the
testimonies from Aristophanes (Clouds, 360–2, and Birds, 690–2 = 84A5 D.-K. = Dram. T22 L.-M.)
and Timon (fr. 18 Di Marco). These present Prodicus as a meteorosophist, an expert on celestial matters:
the reference here would not be to cosmological phenomena, but to atmospheric ones such as rain and
wind, which Prodicus claimed to know much about (see Soverini 1998: 90–114). However, it is also
worth noting that, while certainly important, the connection with agriculture did not exhaust the rich-
ness of the apologue on Heracles, which could equally well apply to the world of the polis – indeed, it
especially referred to this.
36
Muir 1985: 204. Besides, Antiphon too is credited with a treatise On Agriculture (87B118
D.-K., not in L.-M.), and often the sophists are regarded as experts on celestial matters (meteora;
see preceding footnote): see e.g. Protagoras 80A11 D.-K. = Dram. T18a and Gorgias 82A17 =
32P35 L.-M.
37
Henrichs 1976; Willink 1983; Mayhew 2011.
38
See Kerferd 1981a: 171.
39
Dihle 1977 is the main champion of this new attribution, which has attracted much consen-
sus: see e.g. Scodel 1980: 124; Ostwald 1986: 281; Kahn 1997. However, we also find some cham-
pions of the Critias attribution: see e.g. Centanni 1997: 144–59 (very useful on the other tragedies
possibly written by Critias); Bultrighini 1999: 213–50; Scholten 2003: 238–57 and Alvoni 2017.

https://doi.org/10.1017/S0533245119000075 Published online by Cambridge University Press


VI THE GODS AND RELIGION 119

character – and not a great one either, given the sinister reputation sur-
rounding Sisyphus, the deceiver par excellence.40 In other words, this
long fragment counts as a document that attests to the ideas circulating
at the time rather than to any independent theory.41
The fragment reconstructs the history of humanity by dividing it into
three crucial stages: (1) in its first stage, humankind’s life was dis-
orderly and bestial, dominated as it was by brute force (lines 1–4);
(2) to remedy the situation, human beings established laws ensuring
justice and peace, and yet these laws only partially attained the result
that was sought after: while they regulated public actions, underhanded
acts of violence continued (5–11); (3) someone, ‘a man who was
shrewd and wise in his planning’ (12), then invented the gods: immor-
tal, omniscient, and omnipotent beings capable of reading human
beings’ thoughts and of acting upon this knowledge – and fear of the
gods finally made city life possible (11–42). What we have here is a
first attestation of the theory of religion as an instrumentum regni, a the-
ory destined to enjoy great popularity, first in Rome and then in the
modern world.
Even though it is impossible to establish any direct links, it is
interesting to note that this passage stands in contrast to the theories
of Prodicus and Protagoras that I have just discussed. Unlike
Prodicus, the author of these verses traces the origin of the belief in
the gods back not to usefulness, but to fear.42 Even more marked is
his divergence from Protagoras and democratic ideology. While
Protagoras celebrates the superiority of politics (and the nomos) over
religion precisely because of its capacity to regulate the social life of
humans, the very opposite is the case in the Sisyphus: it is religion,
understood as fear of the gods, which safeguards human life. But the
real point of divergence does not concern so much the analysis of soci-
ety as the anthropological conception which this analysis implies:
whereas Protagoras stresses a ‘collaborative’ anthropology, which
underlines the political and social dimension of human life, the
Sisyphus passage entails a very different kind of anthropology, which
emphasizes the irremovable nature of the passions and of the instinct

40
Sutton 1981.
41
Kahn 1997: 249–50.
42
An interesting parallel here is with Democritus, 68A75 D.-K. = Atom. D207 L.-M.; see too
Euripides, Helen, 743.

https://doi.org/10.1017/S0533245119000075 Published online by Cambridge University Press


120 VI THE GODS AND RELIGION

to overpower others.43 One truly experiences here the ‘bottomless’


depth of human nature.44

Providence and divine justice: matters of theodicy

One issue which had always been debated in relation to the gods was
what in the modern age came to be referred to as the problem of ‘the-
odicy’, which is to say the problem of the justification of divine justice
vis-à-vis the existence of evil. Faith in the gods entailed confidence in a
corrective action on their part, as had constantly been stressed by the
greatest poets, from Hesiod (Works and Days, 267–73) and Solon
(fr. 1, 25–32 D.) onwards. In the second half of the fifth century BC,
the Greeks started voicing more and more doubts, reservations, and
objections with regard to the effectiveness of this divine intervention,
in some case denying its very existence. As a counterpart to
Aeschylus’ censuring of those who dared to claim ‘that the gods did
not deign to concern themselves / with such mortals’ (Agamemnon,
369–72), it is possible to adduce some verses from Euripides that
clearly reflect the anxiety felt by the people of his day: ‘if the gods are
<sensible>, you, just man that you are, will be rewarded. And if not,
why should we toil?’ (Iphigenia in Aulis, 1034–5). Besides, this was
the implicit moral of the Sisyphus: ‘the truth would be that there are
no gods and that no unjust man has anything to fear insofar as he
can escape the human guardians of law and order’.45 Protagoras, too,
had dealt with this issue in an original way, by attempting to solve
the problem of justice without invoking the gods. It is perfectly reason-
able to assume, then, that the sophists contributed to this climate of
anxiety through the views they expressed.
Other testimonies confirm this, in addition to the ones already
discussed. Consider, for instance, the following statement by
Thrasymachus:

43
Evident affinities are instead to be found with Antiphon: see Chapter 4, pp. 84–89. This sort
of polemic becomes even more interesting when we consider the fact that another target might
have been Socrates’ intellectualism. However, it is difficult to agree with Santoro 1997 that the
wise inventor of the gods is to be identified precisely with Socrates (more reasonably, Palumbo
2005 suggests the name of Xenophanes – but, again, it is difficult to draw any definite
conclusions).
44
Nestle 1948. On this, see now Balla 2018.
45
Burkert 1985a: 315.

https://doi.org/10.1017/S0533245119000075 Published online by Cambridge University Press


VI THE GODS AND RELIGION 121

‘The gods do not notice human affairs: for otherwise they would not have failed to take
notice of what is the greatest good for humans, justice. For we see that humans do not
practice this’ (85B8 D.-K. = 35D17 L.-M.).

Generally speaking, this passage has been used to clear Thrasymachus


of the charge of immoralism suggested by the Republic: scholars have
spoken of the ‘cry of sorrowful pessimism’, of ‘an impassioned denun-
ciation’, and of ‘bitter disillusionment’.46 However, these heartfelt ver-
dicts may be making too much of the statement, which lends itself,
rather, to a more disenchanted reading.47 Indeed, the passage neither
delivers any denunciation nor betrays any bitterness for the way in
which people scorn justice. First of all, it acknowledges some facts:
that the gods are indifferent towards human events and that humans,
while affirming the importance of justice, do not actually abide by it.
This observation thus succeeds in its goal of establishing the conflict
between entrenched beliefs and forms of behaviour: for if humans
truly believed in the power of the gods, they would respect the ‘greater
good’ of justice. In other words, Thrasymachus sticks to the facts; yet,
his observation has a startling outcome. When read in such terms,
Thrasymachus’ statement would appear to be not so much a cry of
anguish against his own times as yet another attestation of the rational-
ism of the sophists, who often build their theses in opposition to trad-
itionally accepted ones, as ‘weaker’ arguments prevailing over
‘stronger’ ones.
Whatever the underlying meaning of Thrasymachus’ words, it is evi-
dent that they could not be taken as an expression of atheism. Still,
atheism remained one possible response to the apparent predominance
of injustice in human events. A stance against divine providence has
been attributed to Antiphon (87B12 D.-K. = 37D37 and R17). In add-
ition to this, considering that Plato seems to allude to him in Book 10 of
the Laws, we might infer that, in Antiphon, a mechanistic view of the
universe went hand in hand with the denial of the existence of the div-
ine.48 In antiquity, however, the atheist par excellence was not a sophist
but a poet, who may have been influenced by the sophists: Diagoras of

46
Untersteiner 1954: 325; Romeyer-Dherbey 1985: 77; and Guthrie 1971: 297 respectively.
47
Bonazzi 2008.
48
See also Chapter 2, pp. 40–41. Some difficulty, in this case, might be raised by fragment
87B10 D.-K. = 37D9a L.-M. (‘that is why he lacks nothing and receive nothing from anyone,
but is unlimited and unlacking’), which many scholars believe to be referring to god. Other solu-
tions are possible, however, because the subject might be the cosmos, the intellect, or nature: see
the discussion in Pendrick 2002: 256–9.

https://doi.org/10.1017/S0533245119000075 Published online by Cambridge University Press


122 VI THE GODS AND RELIGION

Melos. Regrettably, not much is known about him, and the little infor-
mation we have consists of trivial anecdotes which do not really help us
to reconstruct his ideas.49 So it is impossible to establish whether – as
hypothesized by Guthrie50 – Diagoras may be the source of theses of
the sort we find in a fragment from Euripides’ Bellerophon:

Does someone then say that there are gods in heaven?


There are not, there are not, if a man will
Not in folly rely on the old argument.
Consider it yourselves; do not build your opinion
On my words. I say that a tyranny
Kills many men and deprives them of their possessions,
And breaking oaths destroys cities;
And doing this they are more happy
Than those who live each day in pious peace.
And I know of small cities that honour the gods
Which obey greater and more impious ones,
Overcome by the greater number of spears.
(fr. 286 Kannicht)51

In this case, the argument is based on the incompatibility between the


predominance of evil on the one hand and the existence of an omnis-
cient, omnipotent, and benevolent deity on the other. Now, given
that apparently there are no reasons to deny the first point, the most
reasonable conclusion would be to deny the latter: hence, the gods
do not exist.52
One interesting element in Euripides’ verses is the allusion to the
destruction of ‘small cities’, something which the Greeks unfortunately
became aware of during the years of war between Athens and Sparta.
One particularly striking episode was the massacre of Melos,
Diagoras’ home city, in 415 BC: that this might have been the origin
of Diagoras’ atheism is an intriguing hypothesis, yet one impossible
to verify.53 Certainly, the Melos episode inspired some memorable

49
The testimonies were brought together in Winiarczyk 1981 and 2016.
50
Guthrie 1971: 236; but see now the more detailed treatment in Whitmarsh 2015: 109–13.
51
Translated Barnes 1979.
52
See Barnes 1979: ii.455–6 for an analysis of the merits and limitations of this argument.
53
From the few testimonies we have about Diagoras’ life, it seems as though he was charged
with impiety and forced to flee Athens precisely in 415/414 BC: see Winiarczyk 2016: 54–59. A
scholium to Aristophanes would appear to establish a correlation between his impiety (and in par-
ticular his profanation of the Eleusinian Mysteries) and the Melos event (Scholia in Aristoph. Aves,
1073); see also Brisson 1994. However, Diagoras’ atheism had long been known: see Ostwald
1986: 275–7. As far as Euripides’ Bellerophon is concerned, it can be dated between 430 and
426 BC.

https://doi.org/10.1017/S0533245119000075 Published online by Cambridge University Press


VI THE GODS AND RELIGION 123

pages by another great author of the fifth century; once again, this is not
a sophist but a writer deeply influenced by the sophists: Thucydides.
When introducing the Melos episode in Book 5 of his History,
Thucydides imagines a dialogue between the Athenian ambassadors
and the representatives of Melos to discuss the surrender of the
‘small city’ to the imperial power of Athens. The structure of what is
often referred to as the ‘Melian Dialogue’, arranged according to
opposite theses, clearly reflects the influence of sophistic antilogies,
as does the content of the arguments: scholars have, for instance,
repeatedly observed that the Athenians defend the same notion of just-
ice as the one that Plato attributes to Thrasymachus in the Republic,
namely the idea that justice comes from might. By contrast, scholars
have often failed to note that the dialogue also includes a most interest-
ing variation on the theme of the divine and of theodicy.
Towards the end of the dispute, when it is clear that any settlement
between the two parties is impossible, the Melians openly accuse the
Athenians of injustice, while voicing their faith in a providential inter-
vention of the gods (who are expected to somehow force the
Spartans – whose colonists the Melians are – to come to their aid):

Melians: We, too, be well assured, think it difficult to contend both against your power
and against fortune, unless she shall be impartial; but nevertheless we trust that, in
point of fortune, we shall through the divine favour be at no disadvantage because
we are god-fearing men standing our ground against men who are unjust. . .
Athenians: Well, as to the kindness of the divine favour, neither do we expect to fall
short of you therein. For in no respect are we departing from men’s observances regard-
ing that which pertains to the divine or from their desires regarding that which pertains
to themselves in aught that we demand or do. For of the gods we hold the belief, and of
men we know, that by a necessity of their nature wherever they have the power they
always rule. And so in our case since we neither enacted this law nor when it was
enacted were the first to use it, but found it in existence and expect to leave it in exist-
ence for all time, so we make use of it, well aware that both you and others, if clothed
with the same power as we are, would do the same thing. And so with regard to the
divine favour, we have good reason not to be so afraid that we shall be at a disadvantage.
(Thucydides, 5.104–5)

While one might disagree with the Athenians’ words, their originality
can hardly be disputed: to defend themselves, they did not simply reject
the belief that any providential divine intervention was possible, as
Thrasymachus might have done, for instance; neither did they deny
the existence of the gods, like the Bellerophon or Diagoras; nor yet
did they endorse an agnostic stance like Protagoras. On the contrary,

https://doi.org/10.1017/S0533245119000075 Published online by Cambridge University Press


124 VI THE GODS AND RELIGION

the Athenians affirmed their ‘piety’ by developing a sort of ‘negative’


theology. The starting point of their reasoning is the opposition
between knowledge and opinion, which draws upon the reflections of
thinkers such as Protagoras and Xenophanes: certainty only concerns
human facts, whereas when it comes to the gods we can only rely on
opinions – the kind of opinions which make up the Greek tradition.
But what can be inferred from these opinions? What is their teaching?
Whether we read Homer or the theogonies, lyric poems or tragedies,
the moral is always the same: the gods rule because they are more
powerful; and this is the conclusion which we (undoubtedly, in all cer-
tainty) reach if we consider human events. Hence, in the light of shared
opinions, as well as of the actual behaviour of human beings, the
Athenians find confirmation of their thesis: the justice of humans and
gods consists in the predominance of those who are stronger; and if
justice consists in the exercising of strength, then it must be concluded
that the Athenians are those who most faithfully follow the gods, pre-
cisely insofar as they impose their own might.54
What are we to make, then, of this negative theology? Clearly, we
cannot attribute it to Thucydides, nor must we necessarily attribute it
to the anonymous Athenians of his text, given that the latter dialectic-
ally inferred it from traditional morality. Be that as it may, what we have
here is one of the most revealing examples of the way in which many
fifth-century authors reasoned, especially the sophists. Its hallmarks
are an inquisitiveness and desire to investigate problems to their utmost
limits, and a taste for provocation. The historian Gregory Crane has
claimed that Thucydides ‘writes to shock’: the same description per-
fectly applies to the sophists as well.55

54
Significant similarities may be found in Book 10 of the Laws: ‘All this, my friends, is the
theme of experts – as our young people regard them – who in their prose and poetry maintain
that anything one can get away with by force is absolutely justified. This is why we experience out-
breaks of impiety among the young, who assume that the kind of gods the laws tell them to believe
in do not exist; this is why we get treasonable efforts to convert people to the “true natural life”,
which is essentially nothing but a life of conquest over others, not one of service to your neighbor
as the law enjoins’ (Pl. Leg. 10.890a–b). On the possible references to Antiphon in this passage, see
above and Chapter 2, pp. 40–41. It is interesting to note that in antiquity Antiphon was regarded as
one of Thucydides’ masters: the hypothesis that these passages from the Melians’ dialogue betray
the sophist’s influence is intriguing, but impossible to demonstrate based on the available
testimonies.
55
Crane 1998: 6.

https://doi.org/10.1017/S0533245119000075 Published online by Cambridge University Press


APPENDIX 1: THE PROTAGONISTS

As already noted in Chapter 1, it is difficult to pinpoint the number and


identity of all the sophists who in the fifth century BC travelled from city
to city educating Greece, as Hegel put it, or corrupting it, as
Aristophanes believed. In what follows, an attempt will be made to pro-
vide an exhaustive overview of the most authoritative sophists, without
overlooking those minor, or not so well-documented, figures who pre-
sent some interesting features.1

Protagoras of Abdera

In antiquity the life and philosophical career of Protagoras – arguably


the most famous sophist – became the object of many, mostly unreli-
able, stories. We know for certain that he was born in Abdera,
Thrace, around 490 BC – the first half of the fifth century. Despite
some evidence to the contrary, it is likely that he came from a wealthy
family, one which even gave hospitality to Xerxes at the time of the
Persian Wars. It is also reported that, in return, Xerxes granted the
family the rare privilege of having their son Protagoras educated by
the Magi (even if one chooses to regard this story as spurious, it still
confirms the contacts between Protagoras and the culture of the
Ionian colonies and those of the Near East). Later he travelled around
Greece as a teacher, earning great fame and amassing considerable
riches. It is equally certain that Protagoras repeatedly sojourned in
Athens in the time of Pericles, with whom he established personal
ties. It is difficult to tell whether these contacts also led to a direct
involvement in the democratic politics of the great statesman.
Evidence in this direction is apparently provided not just by the soph-
ist’s surviving fragments, but also by the fact that, in 444 BC, Pericles
entrusted Protagoras with drafting the constitution of the new
pan-Hellenic colony of Thurii in Magna Graecia (clearly, an

1
In certain cases, these profiles draw and expand upon the introductory notes provided by
Bonazzi 2007 (on Protagoras, see too Bonazzi 2009b). Other detailed presentations may be
found in Kent Sprague 1972, Kerferd 1981a: 42–58, Guthrie 1971: 261–319, and Pradeau
2009a. With regard to the Platonic figures, Nails 2002 is also very useful. On ancient philosophers
more generally, I will refer to Goulet 1989–2018.

https://doi.org/10.1017/S0533245119000087 Published online by Cambridge University Press


126 APPENDIX 1: THE PROTAGONISTS

appointment of this sort implies that the two men shared the same pol-
itical views). This connection with Pericles can also be confirmed by
the upheavals that marked the last years of Protagoras’ life: the crisis
of Periclean politics seems to find a counterpart in the downfall of
Protagoras, who – like other personalities close to the statesman –
was reportedly put on trial.
But the information we have here becomes very hazy. Some sources
report that, after a public reading of his text on the gods, Protagoras was
charged with impiety and fled Athens, while his books were burned. By
the same account, the sophist died in a shipwreck during his flight.
However, this is very questionable; in particular, it differs from what
Plato writes in his dialogues, where he speaks of Protagoras as though
he had died as a universally esteemed figure in old age, with no refer-
ences to any trials or flights. From Plato we know that Protagoras
lived approximately seventy years (other sources state ninety), which
would place his death at around 420 BC.
Diogenes Laertius attributes many works to Protagoras, but a com-
parison with the other sources reveals that his list is incomplete. Worse
still, what are missing from Diogenes’ list are precisely the most import-
ant and controversial works, such as Truth and On the Gods. One pos-
sible explanation is that these two texts constituted individual entries of
the Antilogiae (Opposing Arguments, mentioned by Diogenes), which
were possibly also known as Kataballontes logoi (The Overthrower
Arguments or, better, The Knockdown Arguments: see Appendix 2).
Yet, as in other cases, any attempt at reconstruction is destined to
remain a mere hypothesis, given the dearth of available fragments.
Indeed, apart from the title of Protagoras’ work, all we have are a few
authentic fragments and many interpretations. Ever since antiquity,
his provocative theses have elicited interest and criticism from his read-
ers. As a result, different reconstructions of his thought are available,
which are not always reliable and are often clearly shaped by dispara-
ging intentions.
The first and most important tradition stretches back to Plato, who
devoted two dialogues to the theses of the sophist of Abdera, the
Protagoras and the Theaetetus. Many of the later testimonies, starting
with Aristotle’s, appear to depend on Plato’s discussion, and this
must be taken into account when examining the information from
our sources. Alongside the Platonic tradition, another tradition proved
influential in antiquity, namely the one that highlighted the sceptical
implications of Protagoras’ thought. However, little is known about

https://doi.org/10.1017/S0533245119000087 Published online by Cambridge University Press


APPENDIX 1: THE PROTAGONISTS 127

this interpretation because sceptics such as Sextus Empiricus explicitly


refused to be associated with Protagoras, whom they regarded as a dog-
matist. This kind of reading certainly served polemical purposes.
Another tradition that was hostile to Protagoras, and which was pos-
sibly partly connected to the sceptical interpretation, was the
Epicurean tradition, which is responsible for the emphasis on the link
between Protagoras and Democritus – in all likelihood, an unfounded
idea. All in all, we must note that many – if not all – the testimonies we
have about Protagoras are vitiated by underlying polemical attitudes of
this sort, which in some cases lead to actual distortions of his thought.
The history of this ancient reception must therefore be taken into
account in order to come up with a suitable reconstruction.
Protagoras’ pupils include Antimoerus of Mende, Carmidas and
Euathlus of Athens, and Theodore of Cyrene. His influence has also
been posited in relation to Prodicus, the so-called anonymous of
Iamblichus and the Dissoi logoi, although it is difficult to measure its extent.

Gorgias of Leontini

Gorgias was born in Leontini (present-day Lentini), a Greek colony in


Sicily, around 485 BC. He received his education in Sicily (he was
reportedly a pupil of Empedocles, and some sources associate him
with Thysias and Corax, traditionally regarded as the ‘inventors’ of
rhetoric), although a life of travel led him to visit many cities and
regions in Greece (Olympia, Delphi, Thessaly, Beotia, and Argos).
Here he practised and taught rhetoric, achieving considerable success,
including financially. Gorgias reached Athens in 427 BC, when he was
despatched as an ambassador by his home city to discuss the delicate
question of the war with Syracuse. It is said that, on this occasion, rhet-
oric too made its first entrance into Athens. He died a centenarian,
probably in Larissa in Thessaly, at the end of a frugal life, ‘never
doing anything in view of pleasure’ – as he once told an interlocutor.
Among Gorgias’ most famous works we find the treatise On
Not-Being or on Nature, which only survives thanks to two paraphrases,
one by Sextus Empiricus and the other by the anonymous author of a
text On Melissus, Xenophanes, and Gorgias. Some sources date this work
to the 84th Olympics, which is to say between 444 and 441 BC, but this
is dubious. No less complicated is the attempt to date other texts
known to us: the two declamations Encomium of Helen and Defence of

https://doi.org/10.1017/S0533245119000087 Published online by Cambridge University Press


128 APPENDIX 1: THE PROTAGONISTS

Palamedes, which are among the few sophistic works to have survived; a
long fragment from a Funeral Oration delivered in Athens, or at any rate
written for an Athenian public; and the fragments of other orations
delivered in various places in Greece. Gorgias may also have been the
author of a treatise on The Art of Rhetoric, and of another text, On the
Opportune Moment (kairos).
The names of many of Gorgias’ pupils have been recorded:
Alcidamas of Elea, Callicles of Acharnae, Isocrates of Athens,
Licymnius of Chios, Meno of Larissa, Polus of Acragas, Protarcus of
Athens, Proxenus of Thebes, and probably Lycophron too.
Thucydides and Critias are also said to have been influenced by
Gorgias, as was Aspasia, Pericles’ hetaira.

Antiphon (of Rhamnous?)

Interest in Antiphon dates back to 1915, when a long papyrus fragment


was brought to light (87B44 D.-K. = 37D38b L.-M.): whereas previ-
ously all that was known of the sophist were short and isolated quotes,
the new discovery made available the longest surviving discussion of the
opposition between nomos and physis, one of the most hotly debated
topics in fifth-century Athens. This finding, however, raised the prob-
lem of Antiphon’s identity: of whether the sophist by this name was
to be identified with the rhetor who promoted the oligarchic coup of
411 BC. The first reconstruction of the fragment gave rise to the myth
of Antiphon as a theorizer of equality and an ardent democrat, thereby
ruling out the hypothesis that this author was the oligarch Antiphon.
But then, in 1984, a new discovery proved the unfoundedness of the
‘democratic’ interpretations of the sophist. The demolition of the latter
also removed the main obstacle to the unitary thesis, which currently
seems to be the more reasonable one, even though it is impossible to
prove it conclusively.2 Clearly, if this is the correct hypothesis, what
we have is a truly central figure on the political and intellectual scene
in Athens, a personality whose importance is confirmed by the

2
For an initial overview, I will refer to Bonazzi 2007: 52–6. Significantly, the two recent mono-
graphs by Hourcade 2001 and Gagarin 2002 favour the unitary reading; see also Decleva Caizzi
1969, Narcy 1989, and now Laks and Most 2016, vol. ix. Nevertheless, it is important to note
with Woodruff 2004 that, however widespread this interpretation may be, it is not universally
accepted: Pendrick 2002 once again supports the arguments of the separatists.

https://doi.org/10.1017/S0533245119000087 Published online by Cambridge University Press


APPENDIX 1: THE PROTAGONISTS 129

admiring description provided by Thucydides, who emphasizes the


intelligence, eloquence, and boldness of the rhetor – the very qualities
that represent defining features of the sophists, in both positive and
negative terms.
From the surviving testimonies on the sophist and rhetor we know
that he was born around 480 BC in the Attic deme of Rhamnous and
that he received his early education from his father, Sophilus, whom
the sources refer to as a ‘sophist’. We know practically nothing of
what happened to Antiphon after that, except for the fact that in
Athens he acquired an ambiguous fame as a ‘logographer’ (that is, a
writer of court speeches for private clients) and that for some time he
worked in Corinth, where he promised to ‘treat the suffering with
speeches’ (87A6 D.-K. = 37P10 L.-M.). Whereas Antiphon appears
to have kept a low profile for most of his career, everything changed
in the last, tumultuous years of his life: according to Thucydides’
account, he was among the promoters of the secret coup which in
411 BC sought to introduce a more or less moderate form of oligarchy
in Athens to replace the democratic regime. As is widely known, the
coup failed, not least because of a lack of support from Sparta (which
Antiphon reportedly visited as an ambassador). Many of the people
involved fled, but not Antiphon, who delivered what Thucydides
describes as ‘the best defence ever offered up to my time in a trial for
a capital crime’ (8.68). Noteworthy as it was, the speech was not
enough to save the sophist’s life: Antiphon was sentenced to death,
his property was seized, and his estates were razed.
Considering Antiphon’s work as both a rhetor and a sophist, he is
among the authors about whom we are most informed. From a philo-
sophical perspective, the most relevant text is no doubt On Truth. This
appears to revolve around the problem of physis, which is explored from
various scientific and political points of view. In On Concord, a work
addressed to a broader audience, Antiphon deals with ethical issues,
giving proof of remarkable psychological acumen. Equal acumen is
reflected in the fragments from the Interpretation of Dreams, which he
based on rational methods. Too little survives of the Statesman to get
an idea of the contents of this work, while all that is known of On
Agriculture is the title. Particularly noteworthy among Antiphon’s rhet-
orical works are the three Tetralogies, each of which consists of two
accusatory speeches and two defensive speeches on fictional cases –
probably exercises intended to explore legal (but also philosophical)
problems. Tradition also credits him with a treatise entitled The Art

https://doi.org/10.1017/S0533245119000087 Published online by Cambridge University Press


130 APPENDIX 1: THE PROTAGONISTS

(of Rhetoric) and with some Prologues. The most interesting works
among the surviving orations are On the Murder of Herodes, Against
the Stepmother for Poisoning, and On the Choreutes. We also have a frag-
ment of what was probably the speech he gave to defend himself after
the 411 BC coup. Finally, some sources identify the sophist Antiphon
with another Antiphon, an author of tragedies: but this identification
is made more problematic by the information that the poet was mur-
dered by orders of the tyrant of Syracuse, Dionysius.

Xeniades of Corinth

A practically unknown figure, at different times Xeniades has been


associated with Protagoras, Gorgias, Xenophanes, and the School of
Elea, yet no reliable confirmation of any of these hypotheses has been
found. His thesis that everything springs from nothing is certainly note-
worthy and earned him an original place in the constellation of
post-Parminedean thinkers.

Prodicus of Ceos

We know that Prodicus came from the island of Ceos in the Cyclades.
The sources make no mention of his date of birth but nonetheless pre-
sent him as a contemporary of Democritus, Socrates, Hippias, and
Gorgias, which would suggest sometime around 460 BC. Like other
sophists, Prodicus made repeated visits to Athens (where by 423 BC
he was already well known) and other Greek cities, either as an emissary
or to deliver speeches or lectures. One of the main reasons for his fame
lies in his linguistic interests, and particularly in his ability to correctly
distinguish the various meanings of terms, clarifying their nuances. Not
without some irony, the importance of this research was acknowledged
by Socrates (who at times went so far as to describe himself as a pupil of
Prodicus) and Plato. These linguistic distinctions are based on the
assumption that language can adequately reflect the variety of things:
the correct use of names is therefore the precondition for an adequate
evaluation of reality.
Prodicus also took an interest in natural problems (a record survives
of a treatise On Nature), although he stands out for his humanistic
emphasis, as is shown by his rationalist criticism of religion, according

https://doi.org/10.1017/S0533245119000087 Published online by Cambridge University Press


APPENDIX 1: THE PROTAGONISTS 131

to which humans first deified all the natural forces on which their exist-
ence depended – such as the sun, the rivers, and the sea – and then the
great benefactors who had contributed to the progress of civilization
(for instance, Demeter and Dionysus for bread and wine). Finally,
we know about the contents of a declamation entitled The Choice of
Heracles,3 in which the choice between vice and virtue provided an
avenue to celebrate the freedom of human beings, who forge their
own destiny. In the eyes of several interpreters, from Xenophon to
George Grote, this call to sacrifice and virtue constitutes the most strik-
ing refutation of the charge of immorality directed against the sophists.
According to a late tradition, Prodicus died in Athens when he was
charged with corrupting the young and forced to drink a cup of hem-
lock. Modern scholars have often dismissed this as an unreliable
account; however, other sources report that he was expelled from a
gymnasium for having addressed inappropriate words to the young:
‘it is not impossible that he did have to face the kind of opposition
which Protagoras spoke of as the common lot of all sophists’.4

Thrasymachus of Chalcedon

Thrasymachus was born in Chalcedon, a Bithynian city on the


Bosphorus, around 460 BC. He chiefly devoted himself to rhetoric
and composed several treatises, contributing to the development of
the so-called mixed style and art prose. After arriving in Athens, pos-
sibly for the first time, in 427 BC, he became involved in local politics,
albeit indirectly, given his status as a foreigner. In all likelihood, the
long fragment from his speech On the Constitution was drafted in the
context of the heated polemics that led to the coup of 411 BC. The ref-
erence to the constitution of the Athenian ancestors (starting from
Solon), the so-called Patrios politeia, was one of the themes exploited
by oligarchic propaganda (in particular that of its more moderate
wing) in order to curb the growing power of Athenian demagogues
(according to a different interpretation, however, this is a fragment of
a speech that the sophist delivered in 407 BC in defence of his home
city of Chalcedon, after a failed uprising against Athens). The fragment
from the speech For the People of Larissa instead refers to the troubled

3
Possibly part of the broader work entitled the Seasons (Horai): see p. 118, n. 35.
4
Kerferd 1981a: 46.

https://doi.org/10.1017/S0533245119000087 Published online by Cambridge University Press


132 APPENDIX 1: THE PROTAGONISTS

political situation in Thessaly and puts forward a pan-Hellenic ideal for


anti-Macedonian purposes.
However, Thrasymachus’ fame is due not so much to the few surviv-
ing fragments of his writings as to Plato, who chose him as the protag-
onist of a memorable debate with Socrates in the first book of the
Republic. The thesis defended by Thrasymachus, according to whom
justice is nothing but the advantage of the stronger, constitutes a first
expression of what later came to be described as ‘political realism’:
there is a logic of force that imposes itself and imposes laws to serve
one’s self-interest, identifying the just with what contributes to assert-
ing one’s own power and interest. Justice, therefore, is no longer the
principle capable of checking force, but rather a means to justify the
predominance of those who find themselves in the stronger position.
It remains to be ascertained, however, to what extent Thrasymachus
actually endorsed theses of this sort. The source informing us of his sui-
cide is dubious, and even the date of his death remains uncertain.

Hippias of Elis

Born in Elis, in the western Peloponnese, probably in 443 BC, Hippias


was the first sophist to hail from a Doric city. He was widely admired
(and generously remunerated) as a teacher and public speaker equipped
with an astonishing memory, and served his city in many important
diplomatic missions to Athens, Sparta, and Sicily. His political commit-
ment also sealed his death, if we are to trust the report that he became
involved with the democratic exiles fighting against the oligarchs in
power in Elis (but it may be that our source here, Tertullian, is confus-
ing the sophist with another famous Hippias, the son of the Athenian
tyrant Peisistratos).
Hippias’ career unfolded under the aegis of polymathia: his remark-
able versatility enabled him not just to discuss the most disparate sub-
jects – including astronomy and mathematics, music and poetry,
painting and sculpture, history and mythology – in a large number of
texts, but to distinguish himself in practical matters too. He wrote
speeches (epideixeis such as the Trojan Discourse, along with actual ora-
tions), epic verses, tragedies, and dithyrambs. He also drew up lists of
winners in the Olympics and various other catalogues – including ones
of the names of various peoples, of the priestesses of Hera at Argos, and
of philosophical doctrines. These interests are reflected in what is

https://doi.org/10.1017/S0533245119000087 Published online by Cambridge University Press


APPENDIX 1: THE PROTAGONISTS 133

arguably Hippias’ most important work, which is simply known as


Synagoge, or Collection.
The sophist probably reached the height of his success at the time of
his journey to Olympia, where he showed up wearing garments, foot-
wear, and rings that he had crafted himself and which were no less
beautiful than the most refined Persian jewels. Not only ancient authors
(Plato) but also some modern interpreters have made some ironic
remarks on Hippias’ qualities, since it is difficult to take a man seriously
who claims to be able to do everything on his own. Others, however,
have praised the sophist of Elis for his all-round ideal of man, combin-
ing brightness of mind and manual dexterity. Be that as it may,
Hippias’ pursuits disprove the prejudice according to which the soph-
ists were only interested in ethical and political matters. His many inter-
ests are not an empty show of erudition but proof of his desire to
register the ongoing advances made by humankind in a wide range of
areas. At least in this respect – namely his awareness of the historicity
of the human experience – it is difficult to deny the importance of
Hippias’ contribution.

Critias of Athens

Critias’ tomb was marked by a personification of oligarchy setting fire


to democracy. In death as much as in life, Critias, the scion of one of
Athens’ noblest families and Plato’s uncle, would appear to have
ardently striven after a single goal: to bring down the popular regime
he loathed. This loathing is a recurrent feature of his life, from his
early youthful experiences (his involvement in the Hermes’ mutilation
in 415 BC and probably his joining of the coup of 411 BC) to the leading
role he played in the much-despised government of the Thirty, which
after the Spartan victory of 404 BC sought to turn Attica into a land
of shepherds. He died fighting against the democrat Thrasybulus on
Munichia in 403 BC.
A mark of infamy condemned Critias’ work to an unjust oblivion;
however, he gave proof of remarkable versatility, applying himself to a
wide range of literary genres and carving out a prominent place for him-
self in the cultural life of Athens. Of particular note are his studies on
the various politeiai, which combined an analysis of the type of consti-
tution with an investigation into the most relevant aspects of civilian life
and military organization. Many fragments survive of his political

https://doi.org/10.1017/S0533245119000087 Published online by Cambridge University Press


134 APPENDIX 1: THE PROTAGONISTS

elegies and other poetic compositions, which were well suited to sym-
posia. Critias also composed some tragedies, although in some cases
their authorship is uncertain. He can probably be credited with the frag-
ments from the Tennes, Rhadamanthys, and Pirithous, which appear to
have explored the sharing – out of love and friendship – ‘of the ultimate
dangers, death or exile’ – dangers whose full weight Critias bore in his
own life.5 The case of the satyr-play Sisyphus is more problematic,
because it can also be attributed to Euripides.
Philostratus featured Critias in his Lives of the Sophists: for this rea-
son, perhaps, Diels chose to include him in his edition; and since
then he has always been reckoned among the sophists. However, if
we rule out the fragments from the Sisyphus, there seem to be no cogent
reasons to regard Critias as a sophist: he was not a teacher and appears
to have championed a kind of old-fashioned, pro-Spartan education,
essentially hostile to the teachings of the sophists (although some points
of convergence between the two may be found). The importance that
Critias assigned to poetry also seems to be at odds with the sophists’
positions. Finally, even if one were to attribute the Sisyphus to him,
this work is a theatrical text that cannot be taken to present doctrines
personally endorsed by Critias. At most, like many other young
Athenian nobles, he may be regarded as a pupil of the sophists (and
of Socrates), but he can hardly be considered to have been a sophist
himself.

Lycophron

Aristotle refers to Lycophron as a ‘sophist’: this is the only certain infor-


mation we have about a figure of whom we otherwise know almost
nothing – not where he came from, when he lived, or what his life
was like. If the Lycophron mentioned by the anonymous author of
the second Platonic letter (314d) were the sophist in question, we
would know that he visited the court of Dionysius II in Sicily between
364 and 360 BC. Aristotle compares his style to that of Gorgias and
Alcidamas, which has suggested the plausible conclusion that
Lycophron was a pupil of Gorgias and a rough contemporary of
Alcidamas. Lycophron would thus be a representative of the second

5
Centanni 1997: 178.

https://doi.org/10.1017/S0533245119000087 Published online by Cambridge University Press


APPENDIX 1: THE PROTAGONISTS 135

generation of sophists. Our sources do not even record the titles of his
works, although from the few testimonies we have it is possible to infer
that he had a good command of the problems tackled by the most fam-
ous sophists – from the issue of language to the opposition between
nomos and physis. This makes the lack of reliable information about
him all the more regrettable.

Callicles of Acharnae

All we know about Callicles is what Plato tells us in the Gorgias. Some
scholars have gone so far as to suggest that he is a fictional character.
More reasonably, Dodds defended the opposite thesis, suggesting
that Callicles truly existed but died young, thus leaving little memory
of himself.6 As far as we can judge from Plato’s dialogue, he is remin-
iscent of another Athenian aristocrat, Critias. Like Critias, he is to be
considered not so much a sophist as a pupil of the sophists – and
Plato arguably used the figure of Callicles to illustrate the dangers of
their theses.

Euthydemus and Dionysodorus of Chios

Most of the information that we have about these figures comes from
Plato’s Euthydemus; other testimonies from the Cratylus and Aristotle
confirm that they were genuine historical figures. As far as we can
tell, the distinctive feature of their work lies in their interest in soph-
isms, of which they make bold use in the Platonic dialogue. Ever
since antiquity, this focus on paradoxical arguments has suggested a
comparison with the Megarian School. Yet, there is no need, perhaps,
to hypothesize, as Dorion 2009a does, that the two figures belonged to
this school, thereby severing all links with the sophistic tradition.
Rather – at any rate according to Plato’s testimony – they represent
the worst version of sophistry, the eristic type whose only goal was
to win debates.

6
Dodds 1959: 12–15. The hypothesis of a premature death is based on a prophecy post
eventum, at Gorgias, 519a.

https://doi.org/10.1017/S0533245119000087 Published online by Cambridge University Press


136 APPENDIX 1: THE PROTAGONISTS

Alcidamas of Elaea

The only useful chronological information we have about Alcidamas is


that, like Antisthenes and Lycophron, he was a pupil of Gorgias. He
may indeed be regarded as a representative of the second generation
of sophists, who appear to have focused almost exclusively on the art
of rhetoric. He was probably one of the pupils of Gorgias most inter-
ested in the theme of improvisation, in contrast to those pupils (such
as Isocrates) who instead privileged technical expertise and the drafting
of written speeches. A trace of this contrast survives in the short speech
entitled On the Writers of Written Discourses or On the Sophists. Along
with a memorable sentence against slavery, Alcidamas is significantly
associated with a speech entitled Odysseus against the Treachery of
Palamedes: a work that, if genuine, would set him in competition with
his master. Among the works now lost, the most important one is argu-
ably Museum.

The Dissoi logoi and other anonymous treatises

Dissoi logoi (Twofold Arguments or Pairs of Arguments, as Laks and Most


propose) is a text written in the Doric dialect that has been transmitted
without a title or author as an appendix to the manuscripts of Sextus
Empiricus, a sceptical philosopher lived under the Roman Empire.
The title by which this text is known today comes from the opening
words: ‘Twofold arguments are spoken in Greece. . .’. A reference to
the end of the Peloponnesian War appears to suggest a date between
the late fifth and early fourth century BC, although this is a controversial
hypothesis. No less controversial is the nature of the text: it has been
taken to consist of school exercises, an orator’s notes, an epideixis, or
even notes taken by a listener. What is clearer is its structure, which
consists ‘in setting up opposing arguments about the identity or non-
identity of apparently opposing moral and philosophical terms such
as good and bad, true and false’.7 Generally speaking, this text has
not been very popular with modern scholars, who have often empha-
sized the banality of many of its arguments. However, its awareness
of the relativity of any point of view and its tendency to arrange argu-

7
Kerferd 1981a: 54.

https://doi.org/10.1017/S0533245119000087 Published online by Cambridge University Press


APPENDIX 1: THE PROTAGONISTS 137

ments into contrasting couples bear witness to ‘the remarkable maturity


of Greek thought in the second half of the fifth century’.8 It is important
to keep this point in mind when approaching the Dissoi logoi, which is
one of the texts that best illustrates the Greeks’ capacity to problematize
reality.9
Along with the Dissoi logoi, there are a number of other anonymous
texts that, according to scholars, can be traced back to the world of the
sophists. Arguably the most interesting of these is the so-called
Anonymus Iamblichi. Iamblichus (third–fourth century CE) was a
Neoplatonist philosopher and the author of a work entitled
Protrepticus, which includes long excerpts from previous authors (for
instance, from Aristotle’s Protrepticus). Scholars have identified roughly
ten pages devoted to a defence of nomos that may be traced back to the
fifth/fourth-century BC debate. What is also unique about this text is its
focus on economic issues. Another text for which a sophistic origin has
been suggested is part of an oration, Against Aristogeiton, preserved in
Democritus’ corpus: in this case too the topic discussed is the value
of nomos. In both cases, however, it may be unwarranted to speak of
a sophistic authorship; a more prudent approach would be to speak
of texts influenced by sophistic themes. What may be inferred from
these texts – without having to assign them to any specific sophist –
is that the problems raised by the sophists continued to be debated
for centuries.10

8
Mazzarino 1966: 285–6.
9
See also T. Robinson 1979 and Maso 2018.
10
A succinct overview of the other anonymous treatises attributed to sophists may be found in
Kerferd and Flashar 1998: 97–107.

https://doi.org/10.1017/S0533245119000087 Published online by Cambridge University Press


APPENDIX 2: THE SOPHISTS AND SPECIALIST FORMS
OF KNOWLEDGE (TECHNAI)

As we have repeatedly verified, competition with other ‘intellectuals’


was crucial for the sophists: while poets and philosophers were their
primary targets, they engaged in an equally staunch polemic against
the representatives of other forms of knowledge (the so-called technai),
such as medicine, music, agriculture, and mathematics. The sophists
affirmed the superiority of their teaching against such people as well.
As far as we can tell, this claim was advanced in two different ways.
In the case of Hippias (a famous polymath), and probably of
Prodicus and Antiphon too, the sophists claimed to be as competent
as specialists in a wide range of subjects: this explains the interest of
sophists such as Hippias and Antiphon in geometrical problems
(86B21 = 36D36L.-M.; 87B13 D.-K. = 37D 36a–b and R14–16
L.-M.), and Prodicus’ focus on agriculture.1
In other cases, especially those of Protagoras and Gorgias, the
competition with specialist forms of knowledge takes a far more heated
form. While these sophists do not claim to have mastered specialist
fields, they argue that their knowledge – the new knowledge of the
sophists – allows them to prevail even over such fields. The techne ton
logon – that is, the art of speeches taught by the sophists – allows
them to take on the other technai in their own fields and even to gain
the upper hand. Thus, in the Gorgias, Gorgias relates that, when he
accompanied his brother, a physician, on his visits, he was often more
successful than him – a specialist – in persuading patients to undergo
treatment. From this he concludes that, in the presence of a public audi-
ence, rhetoric is capable of conquering specialist forms of knowledge: in
the case of conflict, a layman will choose to be treated by a rhetor rather
than a doctor (Gorgias, 456a–c). In Protagoras’ case, too, we have
sources attesting to a polemical engagement with the technai:
Not even this is true, that mensuration deals with perceptible and perishable
magnitudes; for then it would have perished, when they perished. And astronomy
also cannot be dealing with perceptible magnitudes nor with this heaven above us.
For neither are perceptible lines such lines as the geometer speaks of (for no
perceptible thing is straight or curved in this way; for a hoop touches a straight edge

1
See Soverini 1998: 90–114; and Chapter 6, n. 34.

https://doi.org/10.1017/S0533245119000099 Published online by Cambridge University Press


APPENDIX 2: THE SOPHISTS AND SPECIALIST FORMS OF KNOWLEDGE 139

not at a point, but as Protagoras said it did, in his refutation of the geometers [that is
along a line]).
(80B7 D.-K. partly reproduced by L.-M. as 31D33)2

Alongside geodesics, astronomy, and geometry, we cannot rule out medi-


cine as a polemical target. One treatise from the Hippocratic Corpus, sig-
nificantly entitled The Art ( peri technes), polemically alludes to some
unspecified ‘professional slanderers’, and it is tempting to see this as a ref-
erence to Protagoras (and possibly Gorgias as well3). Plato, too, reports
this attitude in the Sophist, alluding to treatises by Protagoras on wrestling
and other arts (232d–e = 80B8 D.-K. = 31D2 L.-M.). In all likelihood,
the reference here is not to wrestling but to the capacity that Protagoras
claimed to possess: the ability to ‘contradict each individual expert in a
given subject’ – be it cosmology, theology, politics, or philosophy. As
insightfully noted by Paolo Fait, these testimonies can also help clarify
the meaning of what was probably Protagoras’ most famous work, the
Kataballontes logoi (The Overthrower Arguments or, better, The
Knockdown Arguments), which contains the ‘man-measure’ thesis.4 In
this respect, it is worth quoting the following testimony on Pericles:
There is on record also a certain saying of Thucydides, the son of Melesias, touching
the clever persuasiveness of Pericles, a saying uttered in jest. Thucydides belonged to
the party of the ‘Good and True’, and was for a very long time a political antagonist
of Pericles. When Archidamus, the king of the Lacedaemonians, asked him whether
he or Pericles was the better wrestler, he replied: ‘Whenever I throw him in wrestling
(katabalo palaion), he, disputing the fall (antilegon), carries the point, and persuades
the very men who saw him fall.’
(Plutarch, Life of Pericles, 8.3–4)

The presence of two key terms, the verbs kataballo and antilego, clearly
suggests a Protagorean context. The Kataballontes logoi, therefore, are

2
Let us also consider Philodemus’ testimony: ‘the <things> are not knowable, <the> words are
not acceptable, <as> Protagoras indeed [sc. said] about ma<thematics>’ (PHerc. 1676 = 80B7a
D.-K. = 31D34 L.-M.). As regards the meaning of the polemic against geometry, we might
posit (with Barnes 1979: ii.546) that in this case too Protagoras exploited his method of two con-
trasting logoi: geometry hinges on physical objects; now, if it does not deal with physical objects, all
it amounts to is an insignificant verbal game; on the other hand, if it does deal with physical
objects, it is subject to empirical evaluation; but any empirical evaluation is bound to offer different
results from those provided by a priori analyses; thus even in relation to the apparent certainties of
mathematics we find that for the same object there are two opposite logoi. Arguments of this sort
were probably also used in the treatise On Mathematics (or On the Sciences) mentioned by Diogenes
Laertius (80A1 D.-K. = 31D1 L.-M.). Besides, even in the Protagoras the sophist does not show
himself to be very keen on the mathematical sciences (318d–e, 80A5 D.-K. = 31D37 L.-M.).
3
See Jori 1996: 333–57.
4
Fait 2007: xli.

https://doi.org/10.1017/S0533245119000099 Published online by Cambridge University Press


140 APPENDIX 2: THE SOPHISTS AND SPECIALIST FORMS OF KNOWLEDGE

those speeches or arguments that have the power to defeat one’s oppon-
ent; and putting pupils in the condition to compete against anyone was
the aim of antilogies.
An analysis of testimonies about the sophists’ engagement with spe-
cialist forms of knowledge once again confirms their agonistic approach:
through such polemics, they sought to present their teaching as superior
to that of other specialists, be they poets, scientists, or philosophers.5
This ‘architectural’ vocation – to quote an Aristotelian expression – of
sophistic knowledge vis-à-vis all other forms of knowledge finds an
exemplary expression in a statement attributed to Gorgias: ‘Gorgias
the orator said that those who neglect philosophy6 but dedicate them-
selves to the ordinary disciplines (ta enkuklia mathemata) are similar to
the suitors, who desired Penelope but slept with the maidservants’
(82B29 D.-K. = 32P22 L.-M.) If we consider the ideological value
that technai possessed in fifth-century democratic Athens, we once
again realize just how provocative the sophists’ activity was.
A reaction from the sophists’ opponents, however, was not long in
coming. The Hippocratic treatise The Art has already been mentioned;
see for instance Chapter 1:
Some there are who have made an art of vilifying the arts, though they consider, not that
they are accomplishing the object I mention, but that they are making a display of their
own knowledge. . . .To be eager to bring shame through the art of abuse upon the
discoveries of others, improving nothing but disparaging before those who do not
know the discoveries of those who do, seems to me to be not the ambition and work
of intelligence, but the sign of a nasty nature, or of want of art.

To this one may add other texts from the Corpus hippocraticum (e.g. On
the Nature of Man, 1–8) or certain passages from Plato (from the Sophist
and Euthydemus) and Aristotle (especially from the Sophistical
Refutations). Indeed, the specious nature of these polemics contributed
to fuel the suspicion that the sophists were incompetent and merely
eristic (from the Greek eris, ‘discord’) debaters: they were incompetent
because they did not really grasp the things they were talking about, and
merely eristic because they were only interested in winning debates. In
certain cases, at least, it is difficult to dismiss these charges.

5
In Protagoras’ case, it seems as though a more open position was adopted in relation to music
(or at any rate the musical theories of Damon, who was a representative of the ‘new culture’): see
Brancacci 2008.
6
Evidently, the term ‘philosophy’ here refers to the kind of activity in which rhetors and soph-
ists engaged.

https://doi.org/10.1017/S0533245119000099 Published online by Cambridge University Press


BIBLIOGRAPHY
Journal titles are abbreviated as in L’Année philologique.

1. Editions and translations

Barnes, Jonathan (ed.) 1984. The Complete Works of Aristotle. 2 vols. Princeton, NJ,
Princeton University Press.
Bonazzi, Mauro 2007. I sofisti. Milan, Rizzoli.
Cooper, John (ed.) 1997. Plato. Complete Works. Indianapolis, IN, and Cambridge,
Hackett Publishing Company.
Diels, Hermann, and Kranz, Walter 1951–2. Die Fragmente der Vorsokratiker. Sixth
edition, Berlin, Weidmann.
Dillon, John, and Gergel, Tania 2003. The Greek Sophists. London, Penguin.
Freeman, Kathryn 1948. Ancilla to the Pre-Socratic Philosophers. A Complete Translation
of the Fragments in Diels’ ‘Fragmente der Vorsokratiker’. Oxford, Blackwell.
Kent Sprague, Rosamund (ed.) 1972. The Older Sophists. Columbia, SC, University
of South Carolina Press.
Laks, André, and Most, Glenn W. 2016. Early Greek Philosophy. 9 vols.
Cambridge, MA, Harvard University Press.
———, and ——— 2016. Les débuts de la philosophie. Des premiers penseurs grecs à
Socrate. Paris, Fayard.
Pradeau, Jean-François (ed.) 2009. Les sophistes. With the assistance of M. Bonazzi,
L. Brisson, M.-L. Desclos, L.-A. Dorion, A. Macé, and M. Patillon. 2 vols. Paris,
Flammarion.
Untersteiner, Mario 1949–62. I sofisti. Testimonianze e frammenti. With the assist-
ance of A. M. Battegazzore. 4 vols. Florence, La Nuova Italia.

2. Works cited

Alvoni, Giovanna 2017. ‘Die Rhesis des Sisyphos über den Ursprung der Religion
(Kritias, Fr. 19 Sn.-K.)’, Paideia 72: 467–81.
Apfel, Lauren J. 2011. The Advent of Pluralism. Diversity and Conflict in the Age of
the Sophists. Oxford, Oxford University Press.
Arrighetti, Graziano 1998. Esiodo. Opere. Turin, Utet.
Balaudé, Jean-François 2006. ‘Hippias le passeur’, in Sassi 2006: 287–304.
Balla, Chloe 2018. ‘πέwυκε πλεονεκτεῖν? Plato and the Sophists on Greed and
Savage Humanity’, Polis 35: 83–101.

https://doi.org/10.1017/S0533245119000105 Published online by Cambridge University Press


142 BIBLIOGRAPHY

Barnes, Jonathan 1979. The Presocratic Philosophers. 2 vols. London, Routledge.


Barney, Rachel 2017. ‘Callicles and Thrasymachus’, in E. N. Zalta (ed.), The
Stanford Encyclopedia of Philosophy (Fall 2017 edition), <https://plato.stanford.
edu/archives/fall2017/entries/callicles-thrasymachus/>
Bastianini, Guido, and Decleva Caizzi, Fernanda 1989. ‘Antipho’, in Corpus dei
papiri filosofici greco e romani, vol. I.1. Florence, Olschki: 176–222.
Beresford, Adam 2013. ‘Fangs, Feathers, & Fairness: Protagoras on the Origins of
Right and Wrong’, in Ophuijsen, Van Ralte, and Stork 2013: 139–62.
Betegh, Gábor 2006. ‘Greek Philosophy and Religion’, in M. L. Gill and P.
Pellegrin (eds.), A Companion to Ancient Philosophy. Oxford, Blackwell: 625–39.
Bett, Richard 1989. ‘The Sophists and Relativism’. Phronesis 34: 139–69.
——— 2002. ‘Is There a Sophistic Ethics?’ AncPhil 22: 235–61.
Blank, David 1985. ‘Socratics Versus Sophists on Payment for Teaching’, ClAnt 4:
1–49.
Bonazzi, Mauro 2004. ‘Atene, i sofisti e la democrazia: Protagora e i suoi critici’,
PPol 27: 333–59.
——— 2006a. ‘La realtà, la legge, la concordia secondo Antifonte’, QS 64:
117–39.
——— 2006b. ‘L’uomo, gli dei, le bestie: a proposito dell’antropologia di
Antifonte’, Elenchos 27: 101–15.
——— 2008. ‘Thrasymaque, la polis et le dieux’, PhilosAnt 8: 61–84.
——— 2009a. ‘Antifonte, Socrate e i maestri d’infelicità’, EPlaton 6: 25–39.
——— 2009b. ‘Protagoras’, in Pradeau 2009a: i.43–90, 443–72.
——— 2011. ‘Il mito di Prometeo nel Protagora: una variazione sul tema delle ori-
gini’, in F. Calabi and S. Gastaldi (eds.). Sankt Augustin, Academia Verlag, pp.
41–57.
——— 2012. ‘Antifonte presocratico’, Elenchos 23: 21–40.
——— 2016. ‘Antiphon le “philochrematos”’, in E. Helmer (ed.), Richesse et
pauvreté chez les philosophes de l’antiquité. Paris, Vrin: 47–58.
——— 2017. Atene, la città inquieta. Turin, Einaudi.
——— 2018. Processo a Socrate. Rome, Laterza.
——— 2020. ‘Ethical and Political Thought in Antiphon’s Truth and Concord’, in
D. Wolfsdorf (ed.), Early Greek Thought. Oxford, Oxford University Press,
forthcoming.
——— and Pradeau, Jean-François 2009. ‘Lycophron’, in Pradeau 2009a:
i.333–41, 522–5.
Brancacci, Aldo 1996. ‘Protagora e la critica letteraria’, in M. S. Funghi (ed.),
Odoi dizesios. Le vie della ricerca. Studi in onore di Francesco Adorno. Florence,
Olschki: 109–19.
——— 2002a. ‘Protagora e la techne sophistike: Plat. Prot. 316d–317c’, Elenchos 23:
11–32.
——— 2002b. ‘Protagoras, l’orthoepeia et la justesse de noms’, in A. Brancacci and
M. Dixsaut (eds.), Platon source de présocratiques. Paris, Vrin: 169–90.

https://doi.org/10.1017/S0533245119000105 Published online by Cambridge University Press


BIBLIOGRAPHY 143

——— 2004. ‘Il logos di Ippia: Plat. Prot. 337c–338b’, in G. Casertano (ed.), Il
Protagora di Platone. Struttura e problematiche. Naples, Loffredo: 390–401.
——— 2008. ‘Protagora, Damone e la musica’, in Musica e filosofia da Damone a
Filodemo. Florence, Olschki: 21–33.
Brisson, Luc 1994. ‘Diagoras de Mélos’, in Goulet 1989–2018: ii.750–7.
——— 2009. ‘Critias’, in Pradeau 2009a: i.391–442, 545–58.
Brunschwig, Jacques 1984. ‘Hippias d’Élis, philosophe-ambassadeur’, in Greek
Philosophical Society 1984: 269–76.
——— 2002. ‘Democritus and Xeniades’, in Caston and Graham 2002: 159–67.
Buchheim, Thomas 1989. Gorgias von Leontinoi. Reden, Fragmente, und
Testimonien. Hamburg, Meiner.
Bultrighini, Umberto 1999. ‘Maledetta democrazia’. Studi su Crizia. Alessandria,
Edizioni dell’Orso.
Burkert, Walter 1985a. Greek Religion. London, Blackwell (first published
Stuttgart 1977).
——— 1985b. ‘Herodot über die Namen der Götter: Polytheismus als historisches
Problem’, MH 42: 121–32.
Burnyeat, Myles 1976. ‘Protagoras and Self-Refutation in Later Greek
Philosophy’, PhR 85: 44–69.
Calogero, Guido 1932. Studi sull’Eleatismo. Florence, La Nuova Italia.
Canfora, Luciano 2001. Storia della letteratura greca. Rome and Bari, Laterza.
Capra, Andrea 1997. ‘La tecnica di misurazione del Protagora’, ASNP, 4th series,
2: 273–327.
——— 2005. ‘Protagoras’ Achilles: Homeric Allusions as a Satirical Weapon
(Plat. Prot. 340A)’, CPh 100: 274–7.
Casertano, Gianni 1971. Natura e istituzioni umane nelle dottrine dei sofisti. Florence
and Naples, Il tripode.
——— 1986. ‘L’amour entre “logos” et “pathos”: quelques considérations sur
“l’Hélène” de Gorgias’, in Cassin 1986: 211–20.
——— 2004. Sofista. Naples, Guida.
Cassin, Barbara 1980. Si Parmenide. Le traité anonyme De Melisso Xenophane et
Gorgia. Lille, Presses universitaires de Lille.
——— (ed.) 1986. Positions de la sophistique. Paris, Vrin.
——— 1995. L’effet sophistique. Paris, Gallimard.
Castagnoli, Luca 2010. Ancient Self-Refutation. The Logic and History of the
Self-Refutation Argument from Democritus to Augustine. Cambridge, Cambridge
University Press.
Caston, Victor 2002. ‘Gorgias on Thought and Its Objects’, in Caston and
Graham 2002: 205–32.
——— and Graham Daniel W. (eds.) 2002. Presocratic Philosophy. Essays in Honour
of A.P.D. Mourelatos. Aldershot, Ashgate.
Centanni, Monica 1997. ‘Atene assoluta. Crizia dalla tragedia alla storia’, Padua,
Esedra.

https://doi.org/10.1017/S0533245119000105 Published online by Cambridge University Press


144 BIBLIOGRAPHY

Classen, Carl J. 1976. ‘The Study of Language Amongst Socrates’ Contemporaries’,


in C. J. Classen (ed.), Sophistik. Darmstadt, Wissenschaftliche Buchgesellschaft:
215–47.
Classen, Carl J. 1981. ‘Aristotle’s Picture of the Sophists’, in Kerferd 1981b: 7–24.
Cole, Thomas 1991. The Origins of Rhetoric in Ancient Greece. Baltimore, MD, and
London, The Johns Hopkins University Press.
Collette Bernard, Gavray Marc-Antoine, and Narbonne J.-M. (eds.) 2019. L’esprit
critique de l’antiquité I. Critique et licence dans la Grèce antique. Paris, Les Belles
Lettres.
Consigny, Scott 1994. ‘Nietzsche’s Reading of the Sophists’, RhetR 13: 5–26.
——— 2001. Gorgias, Sophist and Artist, Columbia, SC, University of South
Carolina Press.
Corey, David 2015. The Sophists in Plato’s Dialogues. Albany, NY: State University
of New York Press.
Corradi, Michele 2006. ‘Protagora e l’orthoepeia nel Cratilo di Platone’, in G.
Arrighetti and M. Tulli (eds.), Esegesi letteraria e riflessione sulla lingua nella cul-
tura greca. Pisa, Giardini Stampatori: 47–63.
——— 2007a. ‘Protagoras dans son contexte. L’homme mesure et la tradition
archaïque de l’incipit’, Mètis 5: 185–204.
——— 2007b. ‘Protagora e la forza del discorso: la tradizione sull’epangelma‘.
SemRom 10: 277–91.
——— 2012. Protagora tra filologia e filosofia. La testimonianza di Aristotele. Pisa,
Fabrizio Serra Editore.
——— 2017. ‘Il περὶ θεῶν di Protagora. Un nuovo tentativo di ricostruzione’,
Maia 69: 444–69.
Crane, Gregory 1998. Thucydides and the Ancient Simplicity. The Limits of Political
Realism. Berkeley, CA, University of California Press.
Curd, Patricia 2006. ‘Gorgias and the Eleatics’, in Sassi 2006: 183–200.
Decleva Caizzi, Fernanda 1969. Antiphontis Tetralogiae. Milan, Cisalpino.
——— 1978. ‘Il frammento 1 D.-K. di Protagora: nota critica’, Acme 31: 11–35.
——— 1985. ‘Ricerche su Antifonte: a proposito di POxy 1364 fr. 1’, in M.
Capasso, F. De Martino, and P. Rosati (eds.), Studi di filosofia preplatonica.
Naples, Bibliopolis: 191–208.
——— 1986a. ‘“Hysteron proteron”: la nature et la loi selon Antiphon et Platon’,
RMM 91: 291–310.
——— 1986b. ‘Il nuovo papiro di Antifonte’, in F. Adorno, F. Decleva Caizzi, F.
Lasserre, and F. Vendruscolo (eds.), Protagora, Antifonte, Posidonio Aristotele.
Saggi su frammenti inediti e nuove testimonianze da papyri. Florence, Olschki:
61–9.
——— 1999: ‘Protagoras and Antiphon: Sophistic Debates on Justice’, in A. A.
Long (ed.), The Cambridge Companion to Early Greek Philosophy. Cambridge,
Cambridge University Press: 311–31.
Demont, Paul 1993. ‘La formule de Protagoras “L’homme est la mesure de toutes
choses”’, in Problèmes de la morale antique. Amiens, Faculté des lettres: 39–58.

https://doi.org/10.1017/S0533245119000105 Published online by Cambridge University Press


BIBLIOGRAPHY 145

Denyer, Nicholas 2013. ‘The Political Skill of Protagoras’, in Harte and Lane
2013: 155–67.
De Romilly, Jacqueline 1973. ‘Gorgias et le pouvoir de la poésie’, JHS 93:
155–72.
——— 1975. Magic and Rhetoric in Ancient Greece. Cambridge, MA, Harvard
University Press.
——— 1986. ‘Les manies de Prodicus et la rigeur de la langue grecque’, MH
43: 1–18.
——— 1988. Les grands sophistes dans l’Athènes de Périclès. Paris, Éditions du
Fallois.
——— 2005. La legge nel pensiero greco. Dalle origini ad Aristotele. Milan, Garzanti
(first published as La loi dans la pensée grecque. Des origines à Aristote. Paris, Les
Belles Lettres, 1971).
Detienne, Marcel 2006. Les maîtres de vérité dans la Grèce archaïque. Second edi-
tion, Paris, Librairie Générale Française (first published 1967).
Di Benedetto, Vincenzo 1955. ‘Il Περὶ τοῦ μὴ ὂντος di Gorgia e la polemica contro
Protagora’, RAL 10: 287–307.
——— 2001. ‘Contributo al testo del frammento di Protagora sugli dei’, RCCM
43: 345–6.
Diels, Hermann 1884. ‘Gorgias und Empedokles’, Sitzungsberichte der Königlich
Preußischen Akademie der Wissenschaften zu Berlin: 343–68 (reprinted in C. J.
Classen (ed.), Sophistik. Darmstadt, Wissenschaftliche Buchgesellschaft,
1976: 351–83).
Dihle, Albrecht 1977. ‘Das Satyrspiel Sisyphos’, Hermes 105: 28–42.
Dodds, Eric R. 1951. The Greeks and the Irrational. Berkeley, CA, University of
California Press.
——— 1959. Plato. Gorgias. A Revised Text with Introduction and Commentary.
Oxford, Oxford University Press.
——— 1973. ‘The Sophistic Movement and the Failure of Greek Liberalism’, in
The Ancient Concept of Progress and Other Essays on Greek Literature and Belief.
Oxford, Clarendon Press: 92–105.
Dorion, Louis-André 2008. ‘Héraclès entre Prodicos et Xenophon’, PhilosAnt
8: 85–114.
——— 2009a. ‘Euthydème et Dionysodore’, in Pradeau 2009a: ii.63–80, 179–87.
——— 2009b. ‘Prodicos’, in Pradeau 2009a: i.343–72, 525–39.
Drozdek, Adam 2005. ‘Protagoras and the Instrumentality of Religion’, AC 74:
41–50.
Dumont, Jean-Paul 1986. ‘Prodicos: de la méthode au système’, in Cassin 1986:
221–32.
Edmunds, Lowell 2006. ‘What Was Socrates Called?’, CQ 56: 414–25.
Ehrenberg, Victor 1957. Sophocles and Pericles. London, Blackwell.
El Murr, Dimitri 2019. ‘Platon contre (et avec) Thrasymaque’, in Collette,
Gavray, and Narbonne 2019: 343–64.
Fait, Paolo 2007. Aristotele. Le confutazioni sofistiche. Rome and Bari, Laterza.

https://doi.org/10.1017/S0533245119000105 Published online by Cambridge University Press


146 BIBLIOGRAPHY

Farrar, Cynthia 1988. The Origins of Democratic Thinking. Cambridge, Cambridge


University Press.
Fowler, Ryan C. 2014. Plato in the Third Sophistic. Berlin, Walter de Gruyter.
Furley, David 1981. ‘Antiphon’s Case Against Justice’, in Kerferd 1981b: 81–91.
Fussi, Alessandra 2006. Retorica e potere. Una lettura del Gorgia di Platone. Pisa,
ETS.
Gagarin, Michael 1994. ‘Probability and Persuasion: Plato and Early Greek
Rhetoric’, in I. Worthington (ed.), Persuasion. Greek Rhetoric in Action.
London, Routledge: 46–68.
——— 2001. ‘Did the Sophists Aim to Persuade?’, Rhetorica 19: 275–91.
——— 2002. Antiphon the Athenian. Oratory, Law, and Justice in the Age of the
Sophists. Austin, TX, University of Texas Press.
——— 2007. ‘Rational Argument in Early Athenian Oratory’, in J. Powell (ed.),
Logos. Rational Argument in Classical Rhetoric. London, Institute of Classical
Studies: 9–18.
——— 2008. ‘Protagoras et l’art de la parole’, PhilosAnt 8: 23–32.
Gavray, Marc-Antoine 2016. ‘De l’Anonyme à Jamblique: ou de l’usage politique
de la richesse’, in É. Helmer (ed.), Richesse et pauvreté chez les philosophes de
l’antiquité. Paris, Vrin: 277–303.
——— 2017. Platon, héritier de Protagoras. Un dialogue sur les fondements de la
démocratie. Paris, Vrin.
Gigon, Olof 1985. ‘Il libro “Sugli dei” di Protagora’, RSF 40: 419–48.
Giuliano, Fabio M. 1998. ‘Un dimenticato frammento di poetica: POxy III 414 e
l’”enciclopedia del sapere”’, in Papiri filosofici. Miscellanea di studi II. Florence,
Olschki: 115–65.
Goldhill, Simon 1986. Reading Greek Tragedy. Cambridge, Cambridge University
Press.
Gomperz, Heinrich 1912. Sophistik und Rhetorik. Berlin and Leipzig,
Wissenschaftliche Buchgesellschaft.
Goulet, Richard (ed.) 1989–2018. Dictionnaire des philosophes antiques. 7 vols.
Paris, Éditions du CNRS.
Gray, Vivienne 2006. ‘The Linguistic Philosophies of Prodicus in Xenophon’s
“Choice of Heracles”?’, CQ, 56: 426–35.
——— 2007. Xenophon on Government. Cambridge, Cambridge University
Press.
Greek Philosophical Society 1984. The Sophistic Movement. Athens, Athenian
Library of Philosophy.
Grote, George 1864. A History of Greece. From the Earliest Period to the Close of the
Generation Contemporary with Alexander the Great. London, Routledge.
Gourinat, Jean-Baptiste 2019. ‘L’athéisme antique, entre accusation et réalité’, in
Collette, Gavray, and Narbonne 2019: 167–89.
Guthrie, William K. C. 1971. The Sophists. Cambridge, Cambridge University
Press (first published 1969 as part 1 of A History of Greek Philosophy, vol. 3,
Cambridge, Cambridge University Press).

https://doi.org/10.1017/S0533245119000105 Published online by Cambridge University Press


BIBLIOGRAPHY 147

Harte, Verit, and Lane, Melissa (eds.) 2013. Politeia in Greek and Roman
Philosophy. Cambridge, Cambridge University Press.
Heinimann, Felix 1945. Nomos und Physis. Herkunft und Bedeutung einer Antithese
im griechischen Denken des 5. Jahrhunderts. Basel, Schweizerischen Beiträge zur
Altertumswissenschaft.
Henrichs, Albert 1975. ‘Two Doxographical Notes: Democritus and Prodicus on
Religion’, HSCPh 79: 93–123.
——— 1976. ‘The Atheism of Prodicus’, CErc 6: 15–21.
Hobbs, Angela 2000. Plato and the Hero. Courage, Manliness and the Impersonal
Good. Cambridge, Cambridge University Press.
Hoffmann, Klaus F. 1997. Das Recht im Denken der Sophistik. Stuttgart and
Leipzig, Teubner.
Horky, Phillip S. 2006. ‘The Imprint of the Soul: Psychosomatic Affection in
Plato, Gorgias, and the “Orphic” Gold Tablets’, Mouseion 6: 383–98.
Hourcade, Annie 2001. Antiphon d’Athènes. Une pensée de l’individu. Brussels,
Ousia.
——— 2006. Sophistique et atomisme. Brussels, Ousia.
Iannucci, Alessandro 2002. La parola e l’azione. I frammenti simposiali di Crizia.
Bologna, Edizioni Nautilus.
Ioli, Roberta (ed. and trans.) 2010. Gorgia. Su ciò che non è. Zürich and New York,
Olms.
——— 2013. Gorgia. Testimonianze e frammenti. Rome, Carocci.
Isnardi Parente, Margherita 1969. ‘Il pensiero politico greco dalle origini alla sofis-
tica: Socrate e Platone’, in L. Firpo (ed.), Storia delle idee politiche, economiche e
sociali. Vol. I. Turin, Utet: 127–289.
——— 1977. Sofistica e democrazia antica. Florence, Sansoni.
Jaeger, Werner 1947. The Theology of the Early Greek Philosophers. Oxford,
Clarendon Press.
Jori, Alberto 1996. Medicina e medici nell’antica Grecia. Saggio sul peri technes ippo-
cratico. Bologna, Il Mulino.
Kahn, Charles 1973. The Verb ‘Be’ in Ancient Greek. Dordrecht, Reidel.
——— 1981. ‘The Origins of Social Contract Theory’, in Kerferd 1981b: 92–108.
——— 1997. ‘Greek Religion and Philosophy in the Sisyphus Fragment’, Phronesis
42: 247–62.
Kennedy, George A. 1963. The Art of Persuasion in Greece. Princeton, NJ,
Princeton University Press.
Kerferd, George B. 1955–6. ‘Gorgias on Nature or What is Not’, Phronesis 1:
3–25.
——— 1981a. The Sophistic Movement, Cambridge, Cambridge University Press.
——— (ed.) 1981b. The Sophists and Their Legacy. Wiesbaden, Steiner Verlag.
——— 1984. ‘Meaning and Reference: Gorgias and the Relation Between
Language and Reality’, in Greek Philosophical Society 1984: 215–22.
——— 1985. ‘Gorgias and Empedocles’, Syculorum Gymnasium 38: 595–605.

https://doi.org/10.1017/S0533245119000105 Published online by Cambridge University Press


148 BIBLIOGRAPHY

——— and Flashar, Helmut 1998. ‘Die Sophistik’, in F. Überweg and H. Flashar
(eds.), Grundriss der Geschichte der Philosophie. Die Philosophie der Antike, vol. 2.1:
Sophistik, Sokrates, Sokratik, Mathematik, Medizin. Basel, Schwabe: 1–137.
Kuntz, Mary 1993. ‘The Prodikean “Choice of Herakles”: AReshaping of Myth’,
CJ 89: 163–81.
Lapini, Walter 1997. Commento all’Athenaion Politeia dello pseudo-Senofonte.
Florence, Dipartimento di Scienze dell’Antichità ‘Giorgio Pasquali’.
Lee, Mi-Kyoung 2005. Epistemology After Protagoras. Responses to Relativism in
Plato, Aristotle, and Democritus. Oxford, Clarendon Press.
Lloyd, Geoffrey E. R. 1979. Magic, Reason and Experience. Studies in the Origins and
Development of Greek Science. Cambridge, Cambridge University Press.
——— 1987. The Revolutions of Wisdom. Studies in the Claims and Practices of
Ancient Greek Science. Berkeley, CA, University of California Press.
Long, Anthony A. 1984. ‘Methods of Argument in Gorgias’ Palamedes’, in Greek
Philosophical Society 1984: 233–41.
——— 2015. Greek Models of Mind. Cambridge, MA: Harvard University Press.
Macé, Arnaud 2008. ‘Un monde sans pitiè: Platon à l’échole de Thasymaque de
Chalcédoine’, PhilosAnt 8: 33–60.
——— 2009. ‘Thrasymaque’, in Pradeau 2009a: ii.7–28, 149–66.
Mansfeld, Jaap 1981. ‘Protagoras on Epistemological Obstacles and Persons’, in
Kerferd 1981b: 38–53.
——— 1985. ‘Historical and Philosophical Aspects of Gorgias’ “On What Is
Not”’, SicGymn 38: 243–71.
——— 1986. ‘Aristotle, Plato, and the Preplatonic Doxography and
Chronography’, in G. Cambiano (ed.), Storiografia e dossografia nella filosofia
antica. Turin, Tirrenia Stampatori: 1–59.
——— 1988. ‘De Melisso Xenophane Gorgia: Pyrrhonizing Aristotelianism’, RM
131: 239–76.
Manuwald, Bernd 2013. ‘Protagoras’ Myth in Plato’s Protagoras: Fiction or
Testimony?’, in Ophuijsen, Van Ralte, and Stork 2013: 163–77.
Maso, Stefano 2018. Dissoi logoi. Rome, Edizioni di storia e letteratura.
Mayhew, Robert (ed. and trans.) 2011. Prodicus the Sophist. Texts, translations,
and commentary. Oxford, Oxford University Press.
Mazzara, Giuseppe 1999. Gorgia. La retorica del verosimile. Sankt Augustin,
Akademia Verlag.
——— 2005. ‘La rhétorique éléatico-gorgienne d’Alcidamas chez Diogène Laërce
(IX, 54) et le quatre fonctions fondamentales du λόγος’, AC 74: 51–67.
Mazzarino, Santo 1966. Il pensiero storico classico, vol. 1. Bari, Laterza.
Momigliano, Arnaldo 1930. ‘Prodico di Ceo e le dottrine sul linguaggio da
Democrito ai Cinici’, Atti della Regia Accademia delle Scienze di Torino 65:
95–107.
Morgan, Kathryn 2000. Myth and Philosophy from the Presocratics to Plato.
Cambridge, Cambridge University Press.
Most, Glenn W. 1986. ‘Sophistique et herméneutique’, in Cassin 1986: 233–45.

https://doi.org/10.1017/S0533245119000105 Published online by Cambridge University Press


BIBLIOGRAPHY 149

Mourelatos, Alexander P. D. 1985. ‘Gorgias on the Functions of Language’,


SicGymn 38: 607–30.
Muir, J. V. 1985. ‘Religion and the New Education: The Challenge of the
Sophists’, in P. E. Easterling and J. V. Muir (eds.), Greek Religion and
Society. Cambridge, Cambridge University Press: 191–218.
Musti, Domenico 1995. Demokratía. Origini di un’idea. Rome, Laterza.
Nails, Debra 2002. The People of Plato. Indianapolis, IN, and Cambridge, Hackett.
Narcy, Michel 1989. ‘Antiphon d’Athènes’, in Goulet 1989–2018: i.225–44.
——— 1996. ‘Les interprétations de la pensée politique d’Antiphon au XXe
siècle’, RFHIP 3: 31–45.
Natali, Carlo 1986. ‘Aristote et les méthodes d’enseignement de Gorgias’, in
Cassin 1986: 105–16.
Neschke-Hentschke, Ada 1995. Platonisme politique et théproe du droit naturel.
Contributions à une archéologie de la culture politique européenne. Vol. I. Le plato-
nisme politique dans l’antiquité. Louvain and Paris, Éditions Peeters.
Nestle, Wilhelm 1922. ‘Die Schrift des Gorgias ‘Über die Natur oder Über das
Nichtseiende”’, Hermes 57: 551–62.
——— 1936. ‘Die Horen des Prodikos’, Hermes 71: 151–70.
——— 1948. ‘Critias’, in Griechischen Studien. Stuttgart, Hansmann: 253–320.
Newiger, Hans-Joachim 1973. Untersuchungen zu Gorgias Schrift über das
Nichtseiende. Berlin, De Gruyter.
Nietzsche, Friedrich 1964. The Will to Power, trans. W. A. Kaufmann. New York:
Vintage (first published in German 1901).
Ophuijsen Johan M. van, Van Ralte Marleen, and Stork Peter (eds.) 2013.
Protagoras of Abdera. The Man, His Measure. Leiden, Brill.
Ostwald, Martin 1969. Nomos and the Beginnings of the Athenian Democracy.
Oxford, Oxford University Press.
——— 1986. From Popular Sovereignty to the Sovereignty of Law. Law, Society, and
Politics in Fifth-Century Athens. Berkeley, CA, University of California Press.
——— 1990. ‘Nomos and Physis in Antiphon’s Peri aletheias’, in M. Griffith and D.
J. Mastronarde (eds.), Cabinets of the Muses. Atlanta, GA, Scholars Press:
293–306.
O’Sullivan, Neil 1996. ‘Written and Spoken in the First Sophistic’, in I.
Worthington (ed.), Voice into Text. Orality and Literacy in Ancient Greece,
Leiden, Brill: 115–27.
Paci, Enzo 1957. Storia del pensiero presocratico. Rome, ERI.
Palmer, John 1999. Platos’ Reception of Parmenides. Oxford, Oxford University
Press.
——— 2009. Parmenides and Presocratic Philosophy. Oxford, Oxford University
Press.
Palumbo, Lidia 2005. ‘Sul riferimento alla teologia di Senofane nel Sisifo satiresco’,
in M. Bugno (ed.), Senofane ed Elea tra Ionia e Magna Grecia. Naples, Luciano
editore: 77–86.

https://doi.org/10.1017/S0533245119000105 Published online by Cambridge University Press


150 BIBLIOGRAPHY

Patzer, Andreas 1986. Der Sophist Hippias als Philosophiehistoriker. Freiburg im


Breisgau and Munich, Alber.
Patzer, Harald 1974. ‘Der Tyrann Kritias und die Sophistik’, in K. Döring and W.
Kullmann, Studia Platonica: Festschrift für Hermann Gundert. Amsterdam, B.R.
Grüner: 41–8.
Pendrick, Gerard 2002. Antiphon the Sophist. The Fragments. Cambridge,
Cambridge University Press.
Pfeiffer, Rudolf 1968. History of Classical Scholarship. From the Beginnings to the End
of the Hellenistic Age. Oxford, Clarendon Press.
Pernot, Laurent 2006. La retorica dei Greci e dei Romani. Palermo, Palumbo.
Popper, Karl R. 1971. The Open Society and Its Enemies. Princeton, NJ, Princeton
University Press.
Poulakos, John 1983. ‘Gorgias’ Encomium to Helen and the Defense of Rhetoric’,
Rhetorica 1: 1–16.
Pradeau, Jean-François (ed.) 2009a. Les sophistes. With the assistance of M.
Bonazzi, L. Brisson, M.-L. Desclos, L.-A. Dorion, A. Macé, and M.
Patillon. 2 vols. Paris, Flammarion.
Pradeau, Jean-François (ed.) 2009b. ‘Xéniade’, in Pradeau 2009a: i.327–31,
520–1.
Rademaker, Adriaan 2013. ‘The Most Correct Account: Protagoras on
Language’, in Ophuijsen, van Raalte and Stork 2013: 87–111.
Robinson, John M. 1973. ‘On Gorgias’, in E. N. Lee, A. P. D. Mourelatos, and R.
M. Rorty (eds.), Exegesis and Argument. Assen, Van Gorcum: 49–60.
Robinson, Thomas M. 1979. Contrasting Arguments. An edition of the Dissoi logoi,
New York, Arno Press.
Rocca-Serra, Guillaume, 1990. ‘Naissance de l’exégèse allégorique et naissance de
la raison’, in J.-F. Mattei (ed.), La naissance de la raison en Grèce. Paris, Presses
universitaires de France: 77–82.
Romeyer-Dherbey, Gilbert 1985. Les Sophistes. Paris, Presses universitaires de
France.
Rorty, Richard 1979. Philosophy and the Mirror of Nature. Princeton, NJ, Princeton
University Press.
Rosenmeyer, Thomas G. 1955. ‘Gorgias, Aeschylus and Apate’, AJPh 1955:
225–60.
Sansone, David 2004. ‘Heracles at the Y’, JHS 124: 125–42.
Santoro, Mariacarolina 1997. ‘Il fr. 19 Snell del Sisifo di Crizia come testimo-
nianza della concezione socratica del divino: Crizia “accusatore” di Socrate?’,
Elenchos 18: 257–76.
Sassi Maria Michela (ed.) 2006. La costruzione del discorso filosofico nell’età dei
presocratici. Pisa, Edizioni della Normale.
——— (ed.) 2013. ‘Where Epistemology and Religion Meet: What Do(es) God(s)
Look Like?’, Rhizai 12: 283–307.
Saunders, Trevor J. 1981. ‘Protagoras and Plato on Punishment’, in Kerferd
1981b: 129–41.

https://doi.org/10.1017/S0533245119000105 Published online by Cambridge University Press


BIBLIOGRAPHY 151

Schiappa, Edward 1991. Protagoras and Logos. Columbia, SC, University of South
Carolina Press.
Scholten, Helga 2003. Die Sophistik. Eine Bedrohnung für die Religion und Politik der
Polis? Berlin, Akademie Verlag.
Scodel, Ruth 1980. The Trojan Trilogy of Euripides. Göttingen, Vandenhoek &
Ruprecht.
Sedley, David 2013. ‘The Atheist Underground’, in Harte and Lane 2013: 329–48.
Segal, Charles 1962. ‘Gorgias and the Psychology of Logos’, HSCPh 66: 99–155.
Sihvola, Juha 1989. Decay, Progress, the Good Life? Hesiod and Protagoras on the
Development of Culture. Helsinki, Societas Scientiarum Fennica.
Solmsen, Friedrich 1931. Antiphonstudien. Untersuchungen zur Enstehung der
attischen Gerichtsrede. Berlin, Weidmann.
——— 1975. Intellectual Experiments of the Greek Enlightenment. Princeton, NJ,
Princeton University Press.
Soverini, Luca 1998. Il sofista e l’agora. Sapienti, economia e vita quotidiana nella
Grecia classica. Pisa, Pubblicazioni della Scuola Normale Superiore.
Spatharas, Dimos G. 2001. ‘Patterns of Argumentation in Gorgias’, Mnemosyne
54: 393–408.
Striker, Gisela 1996. ‘Methods of Sophistry’, in Essays in Hellenistic Epistemology
and Ethics. Cambridge, Cambridge University Press: 3–21.
Sutton, Dana 1981. ‘Critias and Atheism’, CQ 31: 33–8.
Tell, Håkan 2011. Plato’s Counterfeit Sophists. Washington, DC: Center for
Hellenic Studies.
Thomas, Rosalind 2003. ‘Prose Performance Texts: Epideixis and Written
Publication in the Late Fifth and Early Fourth Centuries’, in H. Yunis (ed.),
Written Texts and the Rise of Literate Culture in Ancient Greece. Cambridge,
Cambridge University Press: 162–88.
Tinsdale, Christopher W. 2010. Reason’s Dark Champions. Constructive Strategies of
Sophistic Argument. Columbia, SC, University of South Carolina Press.
Tordesillas, Alonso 1986. ‘L’instance temporelle dans l’argumentation de la
première et de la seconde sophistique: la notion de kairos’, in B. Cassin
(ed.), Le plaisir de parler. Paris, Les éditions de minuit: 31–61.
——— 2004. ‘Synonymique, éthique, philosophie (note sur Prodicos de Ceos)’, in
J.-P. Cometti (ed.), L’éthique de la philosophie. Paris, Éditions Kimé: 55–75.
——— 2008. ‘Gorgias et la question de la responsabilité d’Hélène’, in F. Alesse
et al. (eds.), Anthropine sophia. Studi di filologia e storiografia filosofica in memoria
di Gabriele Giannantoni. Naples, Bibliopolis: 45–54.
Tortora, Giuseppe 1985. ‘Il senso del kairos in Gorgia’, SicGymn 38: 537–64.
Trabattoni, Franco (ed. and trans.) 2000. Gorgia – Platone. Parola e ragione. Milan,
Unicopli.
Untersteiner, Mario 1947. ‘La storia dell’umanità e la soluzione dell’antitesi
nomos-physis in Prodico di Ceo’, RSF 2: 117–22.
——— 1954. The Sophists. transl. by K. Freeman. Oxford, Blackwell (first pub-
lished in Italian 1949).

https://doi.org/10.1017/S0533245119000105 Published online by Cambridge University Press


152 BIBLIOGRAPHY

Vegetti, Mario 1989. L’etica degli antichi. Rome and Bari, Laterza.
——— 1998. ‘Trasimaco’, in M. Vegetti (ed.), Platone. La Repubblica. Vol. I. Libro
I. Naples, Bibliopolis: 233–56.
——— 2018. ‘Antropologie della pleonexia: Callicle, Trasimaco e Glaucone in
Platone’, in Il potere della verità. Saggi platonici. Rome, Carocci: 195–210.
Verdenius, Willem J. 1981. ‘Gorgias’ Theory of Deception’, in Kerferd 1981b:
116–28.
Versenyi, Lazslo 1962. ‘Protagoras’ Man-Measure Fragment’, AJPh 83: 178–84.
Vitanza, Victor 1997. Negation, Subjectivity, and the History of Rhetoric. New York,
State University of New York Press.
Vlastos, Gregory 1956. ‘Protagoras’, in M. Ostwald (trans.) and G. Vlastos (ed.),
Plato. Protagoras. New York, Bobbs-Merrill Company: vii–xxiv.
Wardy, Robert 1996. The Birth of Rhetoric. Gorgias, Plato and Their Successors.
London, Routledge.
Wedgwood, Ralph 2017. ‘The Coherence of Thrasymachus’, OSAP 53: 33–63.
White, Stephen A. 1995. ‘Thrasymachus the Diplomat’, CPh 90: 307–27.
Whitmarsh, Tim 2015. Battling the Gods. Atheism in the Ancient World. New York,
Knopf.
Willink, Charles W. 1983. ‘Prodikos, “Meteorosophists” and the “Tantalos”
Paradigm’, CQ 33: 25–33.
Winiarczyk, Marek (ed.) 1981. Diagorae Melii et Theodori Cyrenaei reliquiae.
Leipzig, Teubner.
——— (ed.) 1990. ‘Methodisches zum antiken Atheismus’, RM 133: 1–15.
——— (ed.) 2016. Diagoras of Melos. A Contribution to the History of Ancient
Atheism. Berlin, De Gruyter.
Wolfsdorf, David 2008a. ‘Hesiod, Prodicus, and the Socratics on Work and
Pleasure’, in Brad Inwood (ed.), Oxford Studies in Ancient Philosophy XXXV:
Winter 2008. Oxford, Oxford University Press: 1–18.
——— 2008b. ‘Prodicus on the Correctness of Names: The Case of τέρψις, χαρά
and εὐwροσύνη’, JHS 35: 1–18.
Woodruff, Paul 1999. ‘Rhetoric and Relativism: Protagoras and Gorgias’, in A. A.
Long (ed.), The Cambridge Companion to Early Greek Philosophy. Cambridge,
Cambridge University Press: 290–310.
——— 2004. ‘Antiphons, Sophist and Athenian’. OSAP 26: 323–36.
——— 2013. ‘Euboulia as the Skill Protagoras Taught’, in Ophuijsen, Van Ralte,
and Stork 2013: 179–93.
Yunis, Harvey 1997. ‘Thrasymachus B1: Discord, Not Diplomacy’, CPh 92:
58–66.
Zeller, Eduard 1923. Die Philosophie der Griechen in ihrer geschichtlichen
Entwicklung. Fifth edition, Leipzig, Reisland.
Zilioli, Ugo 2013. ‘Protagoras Through Plato and Aristotle: A Case for the
Philosophical Significance of Ancient Relativism’, in Ophuijsen, Van Ralte
and Stork 2013: 233–57.

https://doi.org/10.1017/S0533245119000105 Published online by Cambridge University Press


New surveys in the classics
Regular subscribers to the journal Greece & Rome receive a volume in
the New Surveys in the Classics series as part of their subscription. The
following volumes are also available to purchase as books.
Volume 44 Sophocles (ISBN 9781108706094)
Volume 43 Early Greek Hexameter Poetry (ISBN 9781316608883)
Volume 42 Horace, (ISBN 9781107444447)
Volume 41 Homer, Second Edition (ISBN 9781107670167) Greece & Rome is a journal which delivers scholarly research to a wider
Volume 40 Greek Art, Second Edition (ISBN 9781107601505) audience. It showcases original and informative articles on ancient history,
Volume 39 Roman Landscape: Culture and Identity (ISBN 9781107400245) literature, art, archaeology, religion, philosophy, and reception of the
Volume 38 Epigram (ISBN 9780521145701)
ancient world. Although its content reflects current research and its
Volume 37 Comedy (ISBN 9780521706094)
Volume 36 Roman Oratory (ISBN 0521687225) contributors include leading figures in the field, undergraduates and general
Volume 35 The Second Sophistic (ISBN 0198568819) readers who wish to be kept informed of current thinking will also find it
Volume 34 Roman Art (ISBN 9780198520818) engaging and accessible, as well as professional scholars in Classics and in
Volume 33 Reception Studies (ISBN 0198528655)
Volume 32 The Invention of Prose (ISBN 0198525234)
other disciplines. With the wider audience in mind all Greek and Latin
Volume 31 Greek Historians (ISBN 019922501X) quotations are translated.
Volume 30 Roman Religion (ISBN 0199224331)
Volume 29 Greek Science (ISBN 0199223955) A subscription to Greece & Rome includes an annual supplement of New
Volume 28 Virgil (ISBN 0199223424) Surveys in the Classics.
Volume 27 Latin Historians (ISBN 0199222932)
Volume 26 Homer (ISBN 0199222096)
Volume 25 Greek Thought (ISBN 0199220743)
Volume 24 Greek Religion (ISBN 0199220735)

To place an order from within the United States, Canada and Mexico For information on subscribing or submitting to the journal, please visit
please contact Cambridge Customer Services toll free via +1 800 872 7423 journals.cambridge.org/gar
or by email via information@cup.org/. For orders outside of North America
please call + 44 (0)1223 325566 or email via ukcustserve@cambridge.org/.

THE CLASSICAL ASSOCIATION


The Classical Association has a worldwide membership and is open to all
who value the study of the languages, literature, and civilizations of ancient
Greece and Rome. It creates opportunities for friendly exchange and
co-operation among classicists, encourages scholarship through its journals
and other publications, and supports classics in schools and universities.
Every year it holds an annual conference, and it sponsors branches all over
the country which put on programmes of lectures and other activities.
The Classical Association publishes three journals, Classical Quarterly,
Classical Review and Greece & Rome. Members may subscribe at substantially
reduced cost. Non-members and institutions can subscribe via Cambridge
University Press (email journals@cambridge.org).
For more information and a membership application form please contact
the Secretary, The Classical Association, Park House, 15–23 Greenhill
Crescent, Watford WD18 8PH (e-mail: office@classicalassociation.org).
The Secretary can also give information about the reduced journal
subscription rates, and about the Association’s other occasional publications.
Cover illustration: Sculpture “Light of the Moon” at terrace near the Beelden
aan Zee Museum. Scheveningen, Hague, Netherlands © Igor Mitoraj.

https://doi.org/10.1017/S0533245119000130 Published online by Cambridge University Press


New Surveys in the Classics no 45
Greece & Rome
New Surveys in the Classics no 45

The Sophists The Sophists

BY Mauro Bonazzi
From Socrates and Plato onwards, the Sophists were often targeted by the authoritative
philosophical tradition as being mere charlatans and poor teachers. This book, translated
and significantly updated from its most recent Italian version (2nd edition, 2013), challenges

The Sophists
these criticisms by offering an overall interpretation of their thought, and by assessing the
specific contributions of thinkers like Protagoras, Gorgias and Antiphon. A new vision of the
Sophists emerges: they are protagonists and agents of fundamental change in the history of
Ancient Philosophy, who questioned the grounds of morality and politics, as well as the nature
of knowledge and language. By shifting the focus from the cosmos to man, the Sophists
inaugurate an alternative form of philosophy, whose importance is only now becoming clear.

Cambridge Journals Online


For further information about this journal Published for the Classical Association
please go to the journal web site at:
journals.cambridge.org/nsy

https://doi.org/10.1017/S0533245119000130 Published online by Cambridge University Press

You might also like