J. L.

AUSTIN

PHILOSOPHICAL
PAPERS

SECOND EDITION

EDITED BY

J. 0. URMSON
AND

G. J. WARNOCK

OXFORD
AT THE CLARENDON PRESS
1970
 O x j r d University Press, Ely House, London W. 1
GLASGOW NEW YORK TORONTO MELBOURNE WELLINGTON
CAPE TOWN SALISBURY [BADAN NAIROBI DAR ES SALAAM LUSAKA ADDIS ABABA
EIOMBAY CALCUTTA MADRAS KARACHI LAHORE DACCA
KUALA LUMPUR SINGAPORE HONG KONG TOKYO

0 Oxfwd University press 1961, 1970

FIRST PUBLISHED 1961

SECOND EDITION 1970

P R I N T E D I N GREAT B R I T A I N
 FOREWORD

SINCE1961, when ust tin's Philosophicul Papers was first pub-

lished, three further papers by him have appeared in print.
Two of these are included in this new edition, and are further
mentioned below. The third is 'Performatif-Constatif', which
Austin. resented at a (predominantly) ~nglo-French con-
ference at Royaumont in March 1958, and the original French
text of which was published after his death in Cahiers de
Royaumont, Philosophie No. IV, La Philosophie Analytique ( ~ e s
Editions de Minuit, 1962). The editors felt in 1961 that this
piece was too closely similar in content to 'Performative
Utterances' for it to be reasonable to include both in a single
u
volume, and of the two preferred 'Performative Utterances'-
~ a r t l y ,of course, because Austin's own English text of that
paper was available. There seems no reason now to change
that view; however, it should be noted that 'Performatif-
Constatif' is now in print, and also that it has been translated
into ~ n g l i s h(by G. J. Warnock, in Philosophy and Ordinary
Language, edited by C. E. Caton, University of ~llinoisPress,

of the two papers now added to this collection, 'C;.Ya~6vand

~ 6 G a ~ ~ oin a Ethics of Aristotle' is much the earlier in date.
v ithe
~t was written before 1939, when ~richard,on whose views it
comments, was still alive; but it was not published then, and
there is no reason to think that in later years Austin intended
that it should be, though he made in it certain changes and
additions. That is largely why it was not included in the original
edition of this book; another factor was that its text was im-
~erfectin some passages. However, it has now appeared very
fittingly in the collection Aristotle, edited by J. M. E. Moravcsik
( ~ o i b l e d a New
~, York, 1967: Macmillan, London, 1968),
and should now be included with Austin's other ~ublished
papers. 'Three Ways of Spilling Ink' is a rather different case,
vi FOREWORD
since, as the introductory note to that paper explains, Austin's
- - -

manuscript was obviously unfinished and incomplete. Pro-

fessor Forguson, however, found it possible to produce a text
which, though certainly rough, is both authentic and intelli-
gible; and since it adds valuably to ust tin's previously available
papers, publication seemed justified. This piece appeared in the
Philosophical Review in October 1966. In each case we are
grateful for permission to reprint.
In the present volume, then, all Austin's -previously pub- -

lished papers are included with the exception of 'Performatif-

Constatif'-perhaps we should mention also his reviews, and
two short notes on nabi is competitions (Analysi~,vol. xii,
No. 6, 1952 and vol. xviii, No. 5, 1958). 'Ifs and Cans' (1956)
was published in the Proceedings of the British Academy. 'A Plea
for Excuses' was his Presidential Address to the Aristotelian
Society in 1956. 'How to Talk-some simple ways' (1953) was
also published in the Proceedings ofthe Aristotelian Society. 'Are
There A Priori Concepts ?' (1939), 'Other ~ i n d s ' (1946),
'Truth' (1950)~and 'pretending' (1958) were all contributions
to symposia of Joint Sessions of the ~ i n Association
d and the
~ristotelianSociety, and were published in the Supplementary
Volumes of the Proceedings ofthe Aristotelian Society. W e thank
the editors for permission to reprint.
Three further papers appeared for the first time in the first
edition of this book. 'The Meaning of a Word' was read to
the Moral Sciences Club in Cambridge and to the Jowett
Society in Oxford in 1940. 'Unfair to Facts' was read to the
Philosophical Society in Oxford in 1954, and is a sequel to
Austin's symposium on Truth, with P. F. Strawson and D. R.
-

Cousin, in 1950. 'Performative Utterances' is a transcript, with

-

minor verbal corrections, of an unscripted talk delivered in

the Third Programme ofthe B.B.C. in 1956, and our thanks are
due to the British Broadcasting Corporation for their permis-
sion to publish it here.
All the papers in this volume are now arranged in chrono-
- -

logical order of composition-on the suppos~tion,where it

 FOREWORD vii

applies, that that was the same as the order of their public
presentation in print or otherwise.
J. 0. U R M S O N
G. J. W A R N O C K
Oxford 19 69
 CONTENTS

I. ciyaOdv and ~ i G a ~ ~ oin a Ethics of Aristotlc

v ithe

2. Are There A Priori Concepts?

3. The Meaning of a Word

4. Other ~ i i l d s

5 . Truth

6. How to Talk-some simple ways

7. Unfair to Facts

8. A Plea for Excuses

9. Ifs and Calls

10. Per formative Utterances

I I. Pretending

12. Three Ways of Spilling Ink

Index
 ATAOON A N D EYAAIMONIA I N THE
E T H I C S OF A R I S T O T L E 1

THISarticle takes its start from an article by Professor H. A.

Prichard (Philosophy, x [1935], 27-39) on i ~ h Meaning e of
oiyaedv in the Ethics of Aristotle'. It will be seen that I disagree -

with him, but I think his article has the great merit of raising
serious questions.
Statement of Pr.fessor Prirhard's ronclusions. Professor Prichard
begins by stating
- -
his 'heretical' conclusions, as follows:
I. Aristotle really meant by ciaddv 'conducive to our hap-
piness'.
2. ~ristotlemaintained that when a man does an action
deliberately, as distinct from impulsively, he does it simply in
order to, i.e. from the desire to, become happy, this being- so
even when he does what is virtuous or speculates.2 -

(2. I ) A corollary: Aristotle, being anxious to persuade men

first and foremost to practice speculation and secondarily to do
what is virtuous, conceived that, to succeed, what he had to
prove was that this was the action necessary to make a man

I Originally published in Aristotle : A Collection of Critical Essays edited

by J. M. E. Moravcsik, published by Doubleday & Co., he.,New York,
1967, and Macmilhn & Co. Ltd., London, 1968. Copyright 0 1967 by
J. M. E. Moravcsik. Reprinted by permission of the editor and publishers.
Footnotes other than Austin's are by Professor Moravcsik, and the translation
used is that of Sir David Ross. Additional editorial work on the text was
done by J. 0. Urmson.
This distinction between 'speculating' and 'doing what is virtuous' is not
strictly Aristotelian: ecwpla is & l p y r ~ aKUT& T ~ VT E ~ Q L O T ~ T&prr$v.
~V
2 A r A 8 0 N A N D EYAAIMONIA
(The corollary Professor Prichard ignores, and, at least for the
iresent, I shall do the same.)
A

H i s reasonfor excluding certain passagesfrom consideration invalid.

W e must first direct our attention to a curious and important
reservation, which Professor Prichard makes in stating his
view. Aristotle, he says, means by &yaOdv conducive to our
happiness 'in the Nicomachean Ethics [abbreviated NE hereafter
-Ed.] except in the two discussions of pleasure-where dya0dv
is opposed to +a~Aovand poX~7pdv'.W e are not here concerned
with the restriction to the NE, but it is necessary to examine
the further restriction, by which we are precluded from using
NE vxr. xi-xiv and X. i-v.
The argument implied in Professor Prichard's words seems
to be as follows:
(a) In these passages dya0o'v means somethmg different from
what it means in the rest of the NE.
(b) This is shown by the fact that it is, in these two passages,
opposed to +a~Aovand p0~0qpdv.
With regard to (a), it is most unfortunate that Professor
Prichard does not tell us what ciya0dv does mean in these two
1 u
As to (b), we clearly need further explanation. I hope that
the following is a correct expansion of Professor Prichard's
argument.
I. Throughout the NE, with the exception of these two
passages, a'yaO6v is never opposed to +aiAov or pox07pdv, but to
something else, presumably ~ a ~ d v .
2. In these two passages alone, dyaOdv is opposed, not to
~ a ~ dbut
v , to +aCAov and poX0qpdv.
3 . Since we know on independent grounds that K U K ~ V has
a different meaning from +aiAov or it follows that
&ya0dv, in these two passages, must have a meaning different
from that which it has throu~houtthe rest of the NE.
U

(To take a parallel case. Suppose that I do not know the mean-
ing of the adjective 'green': and that throughout a certain work
 A r A @ O N AND EYAAIMONIA 3
I find it opposed to the adjective 'experienced', except in two
passages where it is opposed to 'red' and 'yellow'. Then if I
know on other grounds that 'experienced' means something
sufficiently different from 'red' or 'yellow', I can infer that
'green' must, in these two passages, have a meaning different
from that which it has throughout the rest of the work.)
If this is actually the sort of argument on which Professor
Prichard is relying, I think there are considerations which will
lead him to abandon it.
I. & ~ eis dopposed
~ to poXOqP6velsewhere in the NE-e-g. in
IX. viii. 7-where pleasure is not under discussion. (Not to
mention passages in other works, e.g. Met. 1ozo~21.)I have
not found a case of dyaedv being explicitly opposed to # a ~ h o v
elsewhere, but cf: (3) in&.
2. dyaBdv is constantly opposed to ~ a ~ in
. - -
d the
v two discussions
of pleasure: VII. i. 1-2, xiii. I and 7, xiv. 2 and 9, x. ii. 5 . In
vn in particular, the discussion is introduced and terminated by
an opposition between dyaedv and ~ a ~ dand v , in xiv. z we read :

3. I do not know of any clear distinction between the mean-

ing of K ~ K and ~ V the meanings of + a C k ~and po)(eTpdv (or

novqpdv) any more than I can clearly distinguish between dyaOdv,

in&n~rc~Cs,and oxov8aiov. The words seem to be used almost in-
differently, or at least for 'species' of one another which would
be equivalent in certain contexts.
Very many passages in the NE, such as rn. v. 3 , are evidence
of this. But the point seems clear even from the two discussions
of pleasure. mov8aiov seems equivalent to dyaOdv in vn. xiv. 4,
~ L E U C ~toS dYaB6v in x. ii. I, p~XBqpd~ seems equivalent to or
a species of ~ a ~ in d vvn. xiv. 2, #aGhov to ~ a ~ in d vX. i. 2.
Pleasures are called dyaeal, O V O V ~ ~ ~ U: ,~, ~ ~ L . L K E(also
~s ~ d a i 'etc.)
,
in apparently the same sense or senses not suf5ciently distin-
guished: and similar1yo;K dyaeal, poXOrlpal,+ a i ~ a(also t aiqpal,
etc.) in apparently the same sense. And these adjectives seem
opposed to one another indifferently, cf. e.g. x. v. 6. (It must
be admitted that ~ristotledoes not use the expression K U K ~ ~
4 A r A B O N AND EYAAIMONIA
j8oval: if this requires some explanation, I think one could
- -

easily be found.)
~t is, of course, possible that some distinction can be drawn
betweenm~dvon the one hand and +aCAov and p~x07pdv on the
other. But (a) it is clearly incumbent on Professor Prichard to
draw it-which he does~notdo. (b) Even so he would by no
means be out of the wood, for (i) it does not seem to be true
that a'yaOd~is, in these passages on pleasures, more commonly
opposed to +aCAov and po~87p6vthan to ~ a ~ d pxOTpdv v . only
occurs once in each book, and is only used as opposite of &yaedv
in v11. xiv. 2: and there it is only so oppo&d because it is
equated with ~ a ~ d +aGAov v. is opposed to ciya06v only once in
x. i. z, a rather popular passage: and there, as section 5 of
the same chapter shows, it is equivalent to K ~ K ~ V(ii) . Actually
&yaedv is in these same passages much more commonly opposed
to ~ a ~ d and v , so presumably has its 'normal' sense. (But we
shall see that it is vital for Professor Prichard that, when
Aristotle says '480vi is an dyaed~',dyaOdv should never have its
'normal' sense of 'conducive to our happiness.') (iii) In a most
important passage, x. ii. I, exactly the same remarks are made
about &yaOdv as in I. i. r, a passage on which Professor Prichard
relies in arriving at his interpretation of it as 'conducive to our
happiness'. Here then & a B d v must presumably have that mean-
ing: but this is one of the places where $04is said to be an
dyaOdv, which, on Professor Prichard's interpretation of dyaedv,
-

does not make sense. v.i.

It would seem then that Professor Prichard's ostensible argu-
ment for excluding these two discussions of pleasure from con-
sideration will not bear examination. And it is in any case so
very recondite that we may be tempted to think he would
never have chanced upon it, unless he had been searching for
some reason to justify the exclusion of these passages.
W h y does he wish to exclude these passages ? His interpretation of
~ ; G a b ~ o v $ . Why then, we must ask, should it be important for
Professor Prichard to secure the exclusion from consideration
of the two discussions of pleasure? In order to understand this,
 A r A B O N AND E YAAIMONIA 5
we must first understand that the whole argument of Professor
Prichard's paper is based upon a premiss which is never ex-
- - -

pressed, no doubt because it seems to him obvious; namely,

that 'happiness' (his translation of r68aoula) means a state of
feeling pleased. This may be shown as follows:
I. On p. 39, line 16, 'causing happiness' is substituted without
remark for 'causing pleasure', which was the expression used
- -

in the parallel argumentation on, e.g., pp. 35-45.

2. On p. 38, at the bottom we read 'what he means by
&ya06v is productive of a state, or rather a feeling, of happiness,
i.e., as I thnk we may say in this context, a feeling of pleasure.'
This remark is interesting, since it seems to imply that 'in other
contexts' being happy does not mean feeling pleased. However,
we need not worry, I think, about these other contexts, for
it is quite essential to Professor Prichard's argument at a crucial
point that the word rtisa~povla,or in English 'happiness', should
be and should be known by all to be, at least for purposes of
Ethics, entirely clear and unambiguous in meaning: so that,
if being happy means feeling pleased in some contexts in the
NE, it clearly must do so in all thought that concerns the moral
phlosopher.' (The crucial point referred to is found on p. 37
Ht the bottom, and repeated on p. 39 at the top. professor
Prichard there maintains that ~ristotlecannot be really asking
the question he 'ostensibly' asks, viz. 'What is the nature of
happiness?' for there is no uncertainty about that-'the nature
o f . . . happiness is known to us, for we all know the nature
of that for which the word "happiness" stands'.)
3. On p. 39. in summing up his own contentions, Professor
Prichard says that he has in effect represented Aristotle as a
psychological hedonist; i.e., I understand, he claims that, accord-
ing to Aristotle, all deliberate action is done from a desire to
pGduce in ourselves feelings ofpleasure. Now on p. 27, quoted

Prichard's reservation 'in this context', whatever it means, is not important

here; nor is the contrast between state ('disposition') and feeling. Similarly,
on p. 38, a distinction is drawn between 'happiness' and 'some particular state
of happiness': but I do not think that concerns us here.
6 AI'A8ON AND EYAAIMONIA
above, professor Prichard said that, according to Aristotle, all
deliberate action is done from a desire to become happy. Hence
being happy and feeling pleased are evidently, for Professor
Prichard, equivalent expressions. (And this enables us to see
why he considers his view, as he says on p. 27, 'heretical',
although, at first sight and as here stated, it does not appear very
extraordinary: if we realize that 'become happy' means 'feel
pleased' the view certainly is very strange.)
It is, therefore, clear that Professor Prichard does not dis-
tinguish what we call 'being happy' from what we call 'feeling
pleased': and he has in fact been good enough to tell me that
that is so.
It now becomes evident why the two discussions of pleasure
in the N E should have a peculiar hterest for him.
For if &yaO6v means 'conducive to our happiness', and if
-

'happiness' is equivalent to 'pleasure'-then how can we ask,

as Aristotle does in these passages, whether 480~4is an dyaOdv?
For 480vlj must presumably be translated 'pleasure': so that the
question we are asking becomes 'Is (our) pleasure conducive
t i our pleasure?', which is absurd, or at least absurdly limited.
-

Further similar difficulties arise, if we ask, for instance, what

could be meant by saying that some lj8oval are iyaOai, and,
odder still, that some are not &yaOai.
Hence it is essential for Professor Prichard to maintain that
6ya8dv has in these passages a meaning differeht from 'conducive
to our happiness': but, as we have seen, his reason for saying
dyaOL has a new meaning in these two passages is invalid (nor
does he explain what draOdv does mean in them).
But Professor Prichard has another and quite radical dificulty
-

to face in connexion with Aristotle's discussions of pleasure,

which in his paper he appears not to appreciate. For Aristotle
there discusses the relation of 480~4to ~&3a~povla in such a
manner as to make it quite ~ 1 a i nthat these-two reek words do
not mean the same.I Whereas Professor Prichard's wllole
argument depends on translating rt8a~povla'happinessy, and
i a76 C ~ ! ~ C U T O$04
I ~ 6 8 a ~ o vis V, is not 76 ciP~u~ov:
and so on.
 AFAOON AND EYAAIMONIA 7
taking 'happiness' to be equivalent to 'pleasure' which must (we
assume) be the translation of $80~4:so that $804 and ~LSacpovla
ought to mean the same. Hence Professor Prichard must hold
that, in these discussions, not merely Bya06v, but also either
~68arpoviaor $804 changes its meaning from the normal.
Otherwise his view is untenable. As to which alternative he
would choose, I do not know: but both are very difficult. He
can scarcely hold that ~68acpovlachanges its meaning, for, as
we shall see, much of his argumentation depends on his view
that the meaning of ~$8aipoviawas clear and unambiguous.
As to $80~4,he does hold, as we shall see, that the word is
sometimes used in a special restricted sense to include only the
o w p a ~ i ~$Sovai:
ai but it is obviously quite impossible to hold
that it has only this restricted sense throughout the two full-
dress discussions of pleasure. Unfortunately, Professor Prichard
does not notice this additional difficulty.
Even the exclusion of these passages ~uozrldnot su.ce to save
Professor Prichard's view j o m refutation. It would still be open
to Professor Prichard to maintain (and in view of his low
opinion of Aristotle's ~ t h i c sI, think it possible he might do so)
that, even apart from other arguments such as that about 4aGAov
and pXOTpd~,these very facts which I have just mentioned are
themselves sufficient to show that the two discussions of pleasure
are inconsistent with the rest of the NE and may therefore be
neglected. It need scarcely be pointed out how dangerous this
would be: for we are trying to discover the meanings of a ' y a ~ d ~
(and r68acpovia) and it is scarcely permissible to eliminate a
large -part of the evidence, not otherwise known to be in-
-

compatible with the rest, on the ground that it will not square
with our interpretation of those-meanings. At least it would
be necessary to prove that s68acpovla must elsewhere mean
'pleasure': but, it seems to me, Professor Prichard does not
prove this, he assumes it.
However, it would in any case be of no use to exclude from
consideration the 'two discussions of For pleasure is
mentioned in many other parts of the NE, and precisely the
8 A r A 8 0 N AND EYAAIMONIA
same difficulties for Professor Prichard's view are to be found
in them also.
Let us confine ourselves to book I, since it is upon that book
that Professor Prichard principally relies, including chapters
v and xii, which he cites.
In I. v we are told that o i noMoi K U ~ # O ~ T ~ K ~ T U T maintainO&

that ~ ; S a c ~ o v i ais j8ovlj and that Aristotle himself rejects this

view. According to Professor Prichard's interpretation, it
would seem that he ought to accept it, as tautologcal. Pro-
fessor Prichard did reply, when faced with this, that ?j80lojhere,
- .

being the end of the d.mha~rnc~ds pior, has a special

-
restricted
meaning which includes only the ooParecal i8ovai (cf. VII. xiii.
6).' This is scarcely obvious, and we should have expected
Aristotle's rejection
-
to take a rather different form, if he him-
self held that our end is $ovj, although not merely the u w C L a r b ~ a l
Ij8oval. However, we need not insist on this others are
plainer.
In I. xii. 5 , what Eudoxus said about $806 is compared with
what Aristotle himself says about ~68ac~ovovla:clearly, then,
j8omj and ~ 6 8 a e p v l aare distinct. And, whatever may be true of
Sardanapallus, there is no reason whatever to suppose that
Eudoxus meant by ?j80vjmerely the u ~ ~ aj8ovai. ~ ~ ~ a i
Finally, in I. viii the relation of 4806 to r 6 8 a ~ ~ o v is i a dis-
cussed very much as in books vn and X: ~ ; S a ~ p o v l ais PEO'
lj80YCj~4O ~ &ELI K < 8 0 v i ) ~ ,but
~ quite clearly it is distinct from it.
(Even if they are necessarily connected, we must not confuse
one with the other: cf. EE I. ii. 5.)
W e do not, naturally enough, Gnd in book I a discussion
as to whether +owj is an &ya06~.But the views that it is 76
&yaOdv or 76 GP1u+~vare mentioned, and, though rejected,
not rejected as absurdities (I. v and xii). Moreover, as was
pointed out above, Eudoxus' views are mentioned in I. i and xii
Just as, when Aristotle says?j8ov<is L;yaOdv, Professor Prichard says &yu8dv
has a meaning different from the ordinary, so, when Aristotle distinguishes
480~;)from ~ t ; 8 u i ~ o v lProfessor
a, Prichard says $80~4has a meaning different
from the ordinary.
'Accompanied by pleasure or not without pleasurey.-Ed.
 A r A 8 0 N AND EYAAIMONIA 9

in prettv much the same words as in x. ii, so that it would seem

A J

that the meaning o f dya06v ought to be the same in each case.

W e see, then, that Professor Prichard cannot exclude VII.
xi-xiv and x. i-v from consideration, and that they are fatal
to his view. But even if we d o exclude them, other passages,
equally fatal, can be produced even from book I, on which he
relies. So that, if he still maintains the view, it would seem that
he must be prepared to attribute to Aristotle even more and
graver inconsistencies and oversights than those, already so
numerous, which he attributes to him in his article. Myself,
I am not yet prepared to do this: though I am only too well
aware how imperfect the Ethics is in these respects.
Present state if the problem. So far our results are negative.
r66ay~ovladoes not mean a state or feeling o f pleasure and
6'yaOdv does not mean conducive to our -pleasure. (It is, however,
still possible that &yaOdv may mean conducive to our happiness,
if 'happiness' is not equivalent to 'pleasure'.) It certainly is
important to discover, therefore, what these two words do
mean. Of the two, t 6 8 a ~ ~ o vis,
i a for reasons which will appear,
considerably the easier to elucidate, and accordingly I shall
consider it first.
TL &TLV r6Sa~povla;the meaning of the question. Once again 1
shall take m y start from Professor Prichard's article. In a
passage extending from p. 36 at the bottom to p. 38, he argues
that evidence for his view is to be found in NE I. iv. I must
quote, I am afraid, at some length.

At the beginning of Chapter IV he [Aristotle] directs his hearers'

attention to the question . . . 'What is it that is the greatest of all
achievable goods?' and he proceeds to say that while there is general
agreement about the name for it, since both the many and the
educated say that it is happiness, yet they differ about what happiness
is, the many considering it something the nature of which is clear
and obvious, such as pleasure, wealth, or honour, whereas, he im-
plies, the educated consider it something else, of which the nature
is not obvious. Then in the next chapter he proceeds to state what,
to judge from the most prominent types of life, that of enjoyment,
10 ArAOON AND EYAAIMONIA
the political life, and that of contemplation, various men consider
thatthe good or happiness is, viz. enjoyment, honour, and contem-
plation. And later hc gives his own view . . . that happiness is +vX-ijr
iulpyriCt T L S K ~ T ' CtPcr;7vrrhrlnv.

Here it has to be admitted that Aristotle is expressing himself

in a misleading way. His question, 'What is the greatest of goods?'
can be treated& if it had been the question, 'What is man's ultimate
end?' . . . And his answer to this question, if taken as it stands, is
undeniably absurd. For, so understood, it is to the effect that, though
all men, when asked 'What is the ultimate end?' answer by using
the same word, viz. r;6a~povia, yet, as they differ about what
~66aipovlais, i.e. really, about the thing for which they are using
the word c;6arpovla to stand, some using it to designate pleasure,
others wealth, and so on, they are in substance giving different
answers, some meaning by the word s18arpov~apleasure, others
wealth, and so on. But of course this is not what Aristode meant.
He certainly did not think that anyone ever meant by ri8aipovla
either ~ i p < or X ~ O ~ T Oand
~ ; he certainly did not himself mean by it

$ v f i s 2vipy~cdsrs K ~ T ' dprs?jvsrhslau. What he undoubtedly meant

and tholight - others meant by the word ed6aipoula is happiness.
Plainly too, what he thought lien differed about was not th; nature
of happiness but the conditions of its realization, and when he says
that ~68arpoviais #vxijs 2v+yrca' rig ~ a r 'C ~ ~ E ~T ~~ hYr k awhat
v, he
really means is that the latter is what is required for the realization
of happiness . . . his meaning is similar to that of the man who,
when he asks 'What is colour? or 'What is sound?', really means
'What are the conditions necessary for its realization?'
Here is the passage, I. iv. 1-3: .
A i y ~ p ~ .v . TI rd V&V+WY

d ~ ~ 6 ~ a ~r Lojvv?rpa~rCvdyaddv. dvdPan p;v o h U X E ~ ~6 V~ rGv 6

~ A ~ l u r odpoAoyrirab.
v r j v y Q rl8a~povlav~ a oi i VoMoi ~ a or' i
~ a p l c v r ~X~yovcrivJ
s r6 8' rJ <?jv ~ a 71 l ~ i~?~ ~ ~ T TraJT6v
E L V &TO-

hap/3dvovo~T @ ~ i 8 a c ~ o v ~T iEv ~. ; 82 rqs ci16aipovias, T : E'uriv,

&p+icr/3qro6~~ ~ a 06x
l 6pok0~ OE V O M Osois~ cro+olois ci~o8iSdaoiv.

I 'Let us state. . . what is the highest o f all goods achievable by action.

Verbally there is very general agreement; for both the general run o f men and
people o f superior refinement say that it is happiness, and identify living well
 AI'A@ON AND E Y A A r M O N I A XI

Whence does Professor Prichard derive his confidence that

Aristotle is misrepresenting his own If he is, it must
be admitted that the misrepresentaIion is pretty consistent.
Right through to book x Aristotle always purports to be telling
us 'what e68acPovla is'. (He suminarizes the present passage
in almost the same words again in I. vii. 9.)I Moreover, Aristotle
is, of course, aware of th; very kind of misrepresentation of
which Professor Prichard accuses him: compare what he says
about pleasure in vn. xii. 3 and x. iii..b, rebuking those who
maintained that pleasure is a y k v ~when~ ~ ~they
, really meant that
a certain y6v~obgis the condition of the realization of pleasure.
The real reason for Professor Prichard's confidence is to be
found in his unquestioned assumption that ~ 6 S a i ~ o v Lmeansa
pleasure. This assumption is stated in the passage above:
'What Aristotle undoubtedly meant and thought others
meant by r68arpovLa is happiness.' (This is, of course, rather
odd as it stands, since Aristotle did not know English: it would
lose its plausibility if we substituted 480~4for 'happiness'.) For
if ~ 6 8 a c ~ o v ; awere the Greek word for 'pleasure', it might
well be contended that to ask ~i &riv c68acpovla; must be
misleading: for it might very well be held2 that 'pleasure' stands
for something unanalysable and sui gencris, which we either
-

know, and know with entire adequacy, from experiencing it,

or do not know at all. Pleasure might, on these lines, very well
be considered to be in the same case as colour and soui~d,to
which, accordingly, Professor Prichard without hesitation
- .

compares r68acpovLa. In such cases, one who asks 'what is

so-and-so?' will very probably be found to be asking 'what
are the conditions for its realization?' And whether or not
'pleasure' could in any sense be analysed, at least, e.g., the
and doing well with being happy; but with regard to what happiness is they
differ, and the many do not give the same account as the wise. For the former
think it is some plain and obvious thing, like pleasure, wealth, or honour,
etc. '-Ed.
The only reasonable alternative is to hold that in both passages it is really
76 G p ~ c r ~ oi v O P d n y which is being elucidated.
" I do not inquire whether this would be correct.
x2 A I ' A 8 0 N AND E Y A A I M O N I A
person who said pleasure is honour or wealth would obviously be
suspect of only really intending to give conditions of realization.
However, t68aipov;a does not mean pleasure, as we have
seen. And this very passage proves as well as another that it
does not. This is not merely because the theory that c6Gacpovla
means or 'is' tj80vrj is rejected, but also because in a most im-
portant clause omitted in Professor Prichard's paraphrase, r68ai-
p o ~ i ais said to be equivalent to ~d 5(ijv aid 3.rrp&~crv, which
cannot mean 'feeling pleased'. T o this we shall return shortly.
Professor Prichard seems to make out that, apart from the
fact that the Greeks did not disagree about what rd8aipovla
stands for, Aristotle's actual presentation of his question makes
it in general an absurd one. Certainly, Professor Prichard's
ostensible argument for maintaining that Aristotle here mis-
represents his own problem is not very explicit. According to
him, Aristotle says that men agree -
only on the name for the
r l h s , viz. ~ 6 8 a i ~ o v ~
but
a , disagree about what it is used to
stand for, or to designate. This, he says, is undeniably absurd.
N o w why? Has Professor Prichard any other reason for saying
so, except his belief that there could, in fact (owing to the
unanalysable nature of what c68aipovla does stand for), be
no disagreement about what c;Gacpovovla stands for?
He seems to suppose that, according to Aristotle, (I) men
agree only on the name; (2) that is a substantial measure of
agreement; (3) what is being (mistakenly) asked for is some
synonym for ~6Gaipovla,in the simplest sense-some other
word or phrase which stands for precisely the same as r68aipovih
stands for. (Somewhat as though, when asked for the answer
to a mathematical problem, all should agree that the name for
the answer was 'k' while disagreeing as to the number which
'k' stands for.) I do not know whether even this would be
undeniably absurd (always assuming - that in fact all did not
know aJiarpovia ti stand for - but in any case
Aristotle does none of these things.
I. According to Aristotle, men agree, not merely on the
name t66aipovla, but also that r66acpovla is equivalent to +d E:
 A I ' A 8 0 N AND E Y A A I M O N I A I3

C ~ ~V a76i EST ~ ~ T ~ E L(This

Y . statement Professor Prichard omits
in hs paraphrase.) Moreover, it transpires later that they also
agree on a number ofother propositions about the characteristics
of rfi8a~povla,which are listed in chapters viii and ix-xii.
2. As is shown by I. vii. 9, Aristotle does not think that the
agreement on the name alone is very substantial. And indeed
it- is clear even from chapter iv that this agreement could
cover most radical disagreements.
3 . Aristotle is not, I believe, searching for some simple
synonym for ~ 6 8 a ~ o v / but
a , rather for an 'analysis' of its
meaning. While satisfactory as a preliminary statement, this
does n i t make it sufficiently clear what exactly Aristotle is
doing.' All men know more or less vaguely what is meant by
~ 6 8 a ~ vor i a7-6 EG &jv ~ aT&l cG V ~ ~ T T E
and
L Vagree
, on many
propositions about it: but when they attempt to clarify that
meaning, they disagree. Cf. I. vii. 9: dM' Zuws r;lv p i v

T o search thus for an analysis of the meaning of r 3 a ~ ~ o v l a

does not seem to me absurd, except on the false assumption
that its meaning was, and was known to be, simple
- - and un-
analysable. W e might, to take a similar case, agree that the
- -

aim of the statesman is 'liberty' or 'justice', and yet, in a per-

fectly intelligible sense, disagree about 'what liberty is' or 'what
justice is'.
**There is no doubt, however, that this account of what
Aristotle is asking when he asks 71 b r c v cfi8a~~via; is far from
entirely satisfying. For we need to distinguish from the analysis
of the meaning of ~6Ga~~ovia another procedure altogether,
namely the 'discovery' of those things, or that life, as he would
rather ordinarily say, which satisfy the definition of ~68a~povia
when that has been discovered. As Moore has insisted in the
At this point, in view of the state of the manuscript, some editing by
Mr. Urmson was required.-Ed.
'Presumably, however, to say that happiness is the chief good seems a
platitude, and a clearer account of what it is is still desired.'-Ed,
** See below under the same sign.
14 A r A 8 0 N AND EYAAIMONIA
case of 'good' (which will occupy us later), it is important to
distinguish the discovery of what a word means from the dis-
covery of those things in which the characters meant by the
word in fact reside. (Of course this latter procedure is still not
what Prichard means when he speaks -of discovering 'the
- -

conditions for the realization' of something.) This is too simple

a view for modern times, since few will accept that goodness
is a character in this simple sense. But we still would distinguish
the meaning of ~t;GarpovLa-the best life for man, etc.-from
what we may call the specification of the good life: what
the good life allegedly consists in concretely. The whole
problem arises over the connexion between these two.
I

There is uiustice in Prichard's remark that Aristotle 'certainlv J

does not think that anyone ever meant by ~ G G a r ~ o veither la TL~G

or n h o ~ T o s ' .They were capable of having said that they meant
this, but would have more plausibly claimed that np4 or V A O F T O ~
was what satisfied the specification of r;Ga~povLa. But it is not
so clear that 'he certainly did not himself mean by it $VX+S
Ciffc$ TLS K ~ T '&pf+ijv T E A E ~ Y 'at
, least ;v /31v +rhelY. It is hard
to discover, especially in I. vii, where the analysis ends and the
other process begins. It is perhaps impossible to judge how
much is meant to be analysis. Certainly ~6GabpovLais analysed
( ~ a 6 7 6 vhoAaP/3dvovuc) as r 6 6 [ijv ~ a rd l 8 nPTpdrr~~v.Then
L PIy T E A E I ~(ambiguous phrase!) also seems clearly part of
the meaning of c S 8 a ~ p o v i a .I believe that the whole of I. vii. 9-16
is intended to be an analysis of that meaning. And it results,
as it should, in a clear and full definition, referred to as d Adyo*
of r68aLpovla or of 76 lip'pcurov. Moreover EE 11. i. 10 says that
$vx+ 2vipycLa K ~ T ' (;PET<v is ~6 Yivog ~ a Ti ~ V+ov ~ 6 8 a ~ p o v I a g .
If it had been said that Aristotle did not mean by ~ 6 G a ~ ~ o v i a
OcwpLa, that would, 1 think, be certainly true: and it is O ~ w p i a
which, in Aristotle's theory, occupies the place of <8ov+j and
TAOCTO~in rival theories, as chapter v plainly shows.
But when Aristotle discusses what are in fact the special I

virtues, and which is the most -perfect, he cannot be said any

longer to be analysing the meaning of ~6Sacpovovla.He is asking
 A r A 8 0 N A N D EYAAIMONZA 15

in a different sense: what are the virtues

~ U T L V~ S a ~ p o v i ' a ;

that fill the bill. Even in x. viii. 8, where the conclusion is

4
reached: GOT' Iq ZV ~6SaipovlaO~wpLaTts, it is evident that
> 3
~6SarpovLadoes not mean B~opia-~+ ouov S;I S t a r ~ h ~ ~i E W P ~ U
rt
4
~ ar j i~GSatpovla,~ ao lk p a A o v ~ T C E . ~ ~ ErdLl B E W ~ E ~
~ Va~ Li S a i p o v ~ i v
06 (;u&
K ~ T &U V , L L ~ ~ ~ K ~ S~ a ~ h
T ~ V~ E W ~ [ ~ V . * *

So Aristotle's distinction between analysis and specification

is most unclear. But there is some excuse for ~ r i s t o t l eperhaps,
in that ~ 6 S a i p o v i adoes not stand for some character, such as
goodness might be, but for a certain kind of life, or 2vtpy~ia
(Aristotle is unclear as to which): in such a case it is not so easy
clearly to observe Professor p bore's distinction, or even one
such as Hare's or Urmson's between meaning and criteria.
Suppose we were to ask, for instance, 'What is golfing?'. But
there is this finally to be said. If Aristotle had thought that
~GSaipovla,like golfing, resided in fact in ottly one activity of
one kind of creature, there would have been more excuse for
him than is actually the case. For actually he does think that
~6Saipoviais achieved, in different ways, by gods and by men:
. ..
hence ~GSaipovla cannot mean those activities in which human
~6Saipovi'ais found. (Unfortunately, of course, his statements
on divine d S a i p o v i a are rudimentary, and it is very doubtful,
e.g., how ivpicp T E A E can ~ ~ be a part of the meaning of ~ 6 S a i p o v l a
if the gods are also E;SL;I~OVEF!)
Some distinctions, however, though not this requisite one,
Aristotle does draw: in a way he was perhaps on his way to it.
In Rhet. I. v ad init. he says we must ask: ri ;oriv ~ 6 S a i ~ o v l a
~ ar hi pdpra a6~ljs;and this distinction is common, though not
in the NE. In EE I. v. 13-14, he calls the particular virtues p6pia
nis +ae+js ( w ~ s ,which ovvr~lvovor~ p d c;Gaipovlav.
s (This does
** The material included between the double stars required extensiveediting
by Mr. Urmson because the manuscript on these pages was complicated by
notes, corrections, and revisions. Ross's translation of this Greek passage reads:
'Happiness extends, then, just so far as contemplation does, and those to whom
contemplation more fully belongs are more truly happy, not as mere con-
comitant but in virtue of the contemplation; . . . Happiness, therefore, must
be some form of contemplationy.-Ed.
 I6 A r A 8 0 N AND E Y A A I M O N I A
not, of course, mean that they are 'the conditions for the realiza-
tion' of a6Saipovia: in NE I 1 2 9 ~ 1 8where , the pdpia rdSa~povias
are mentioned, they are distinguished from sh ~ O L ~ T ~L Ka ~ i
~ U ~ ~ K e68atPovias
T L K ~ [cf. VI. xii. a distinction insisted on in,
e.g., EE I. ii. 5 . There is a similar distinction in EE I. i of b
~ 1 dl~ EI? 1 [?v from ~3~~ ~ s d compare v ; also MM I. ii. 9-11 :
0 ; yhp ; D ~ L Ya"MoT LX ~ pT O i ~~T W V4 r6Saipovia dMd sacsa.)' NOW
in effect, the discovery of the pdpra rGSarpovias is the dis-
covery of the activities which together make up the life which,
for man, satisfies the definition of ~6Gatpov;a;and we can in
a sense say, as Aristotle does, that this life is r6Sarpovla. Never-
theless, c6SaLpovia does not mean that life, and its discovery is
posterior to the analysis of the meaning of ~;Saq.t.ovla. Two
erroneous presuppositions were, however, encouraged by
Aristotle's failure to be clear what question he is asking when
he writes ~i ~ O T L V~68atpovia; first9 the presupposition that
cGSa~povla- the ideal life-is not a will-o'-the-wisp, and that
there is only one possible ideal life; second, that the question
of what fills the bill is throughout purely factual.
Summing up then, the question 7 ; ~ O T L V c68atpvla; is
sensible but ambiguous. Aristotle means to ask firstly: what
is the analysis or definition of ~;Satpovia?and secondly: what
life, in for man, satisfies that definition or specifica-
tion? (A subsidiary and distinct question is: What are the
conditions JV 0 6 ~GVEU? and the methods necessary for the
realization of such a life?)
What does c J 8 a ~ ~ o v imean?
a Sorpe fenera1 considerations. W e
must now concern ourselves with the first of these questions:
What does ~36a~povLa mean in Greek and incidentally what is
its translation in English? But I cannot enter into details of
Aristotle's own analysis: I shall only attempt to show what,
according to him, was the vague and common notion of
c6Satpovia from which his analysis starts.

It is important to remember this when interpreting such a passage as

I.vii. 5 . Compare EE 11. i. I 2 , a better statement. [Austin's MM is a reference
to Magrza MoraIia.-Ed.]
 A r A 8 0 N AND EYAAIMONIA I7

According to I. iv. 2 76 ~ 6 8 u i p o v ~was

i v admittedly equivalent
to 76 3{?v ~ a76; 4n p p d ~ ~ ~(In
i v .r. vii. 4 there is a weaker state-
ment: it is admitted 'that the ~ G a C w vlives and acts well'-
but, as Aristotle remarks, on his own account ~68arpoviais
living and acting well: and in both the other Ethics the state-
ment is given as in I. iv. 2; cf. MMI. iii. 3; EE 11. i. 10 and I. i. 7.)
~JGaipovLa,then, means living a life of a certain kind-of what
kind can of course only be discovered by analysing the word
~ 5 and
, hence ciYaOdv (so that the full analysis of ~68atp0vla
includes that of ciYa6dv). ~68aipwv,I suppose, means literally or,
what is often the same, etymologically, 'prospered by a deity':
and what the deities prosper is lives or careers or activities or
parts of these. ~ristotleinsists on two further points-~68~1-
p o v h means a complete life of activity of a certain kind. On the
latter point, that the reference is to ; v & y ~ i a not to ;(is, he is
always firm (cf. also I. xii).' On the former point he is not so
happy: not only is /i?lo;or & A E L O ~ hopelessly ambiguous (cf.
MM I. iv. j), but Aristotle often omits to remember this
qualification. So much so that in the end he never explains
how the /i?ios is made up--only the cipr.rai which predominate it.
At any rate, what is important in all this is, that, though of
course we can speak of a man as ~ 6 S a i ~ wthe v , substantive with
whch ~6Sa;cLwvnaturally goes is pior or a similar word: a man
~.

is only called ~68aLpwvbecause his life is so. Hence the discussion

in chapter v of the various /i?l01which lay claim to being
~ 6 S a i ~ o v i And
u . ~ hence the saying 'call no man ~ 6 S a k w vuntil
he is dead' (I. x. i).
On p. 38, Professor Prichard makes what seem to me serious misstate-
ments about I . xii. 'In other words. . . everything else.' The contrast in the
first sentence I, of course, consider mistaken. As for the rest, the quotation
given is actually from section viii where ~ 6 8 a i ~ o v is l anot contrasted with
dPm$ at all. The contrast with 6per$ is in ii. and vi: 6paslj is ~ V U ~ V pT ~a V
76 noiov K ~ vZ d s 71 e?vai-viz. the 2vlpyeiai (mpd[~is, ;pya) which are
r68aipovia: the 2v4pyriai are ripla and the e68aipovia. Professor Prichard
.. > I
talks as though in x11 svepysral ~ a r~ ' ~ E T $ were
V &ru?rarvsrd:they are ripla!
2 And when in I. iv the suggestion is made that e;SaipovLa is $ov< or
V~OGTOS, that is loose language (as v shows) for the life in which most pleasure
or most wealth is gained.
I 8 A r A 8 0 N A N D EYAAIMONIA
Similarly the forms ~I;Saipov&iv and ~I;Saipoviopds,which
seem to mean 'to congratulate', 'congratulationy. That whereon
I congratulate someone is an achievement, an activity and nor-
mally a completed one (though normally also, of course, of
less extent than his whole career). With reference to this point
consider I. v. 6 and viii. 9.
These considerations show conclusively that ~68ai~ovla
could not wean 'pleasure': pleasure is a feeling, not a life of
a certain kind nor an achievement: nor do I congratulate some-
one on his feeling pleased: and it would bc silly to say 'call no
man pleased until he is dead'.
L

There is, however, Aristotle's own remark in r. iv that many

people do maintain that ~G8aipovi'a is -j604', as others ~ A O G T O C
or sip<.This is a loose remark- as Professor Prichard claims,
though not in his way. His explanation, that Aristotle meant that
some maintain that ~68aipov;ais produced by $Sonj, also ~AOGTOS,
etc., is incorrect. Aristotle himself shows what he meant more
fully in chapter v. The view was that ~;Saipovla is the dno-
-

havosi~ds191og, the life in which most pleasure is felt. Likewise

the identification with ~TAOU^TOSshould-be taken as 'the life in
which most wealth is gained' (not, of course, 'is by definition'
but 'in fact resides in').
If we want a translation of ~i8aipovLawhich will not mislead,
as 'happiness' appears to mislead Professor Prichard, we might
use as a prophylactic 'success'. 'Success' is at least a word of
the same type, so to say, as ~68aipovLa.'Success' does mean
living and acting well: a life or a part of it is 'successful': and
with some hesitations I do congratulate on 'success'; it might
well be said 'call no man successful until he is dead'. ~urther-
more, success demands just that fortunate supply of E G W ~ ; ~ ,
L a , Aristotle admits in r. viii. I 5-1 7 is also a neces-
~ i r ) p ~ ~which
sary condition for ~LSaipov[a.
It is true; however, that 'success' is not a moral notion for
us. Perhaps this is no great disadvantage, for it is doubrful
how far a pagan ethic, such as the Greek (or the Chinese),
. .

gains by a translation which imports our own moral notions:

 A r A Q O N AND EYA A I M O N I A I9

~ ; S a i p o v ~ ais certainly quitc an uichristiatl ideal. Still, we do

require a word to import some form of comrrrendation, as
e;Sa;pwv did, and certain non-personal standards.
**That r6Gaipov;a did mean lifc of activity of a certain kind
is almost certainly the corrcct analysis; and that it did further
mean life of cipsrai seems equally correct. So we may say that
the analysis in I. vii. 8-16 is correct and is supported by r. viii.
However we must also say: ( a ) That roundabout &ay of
bringing into the discussion +vX4 and Z ~ ~ Ois Va piece of un-
necessary Aristotelian metaphysics. It is not really made use
of until the very end, in the argument for the supremacy of
~ c o p i a ; the argument proceeds straight on to the cipsral.
(6) The whole discussion here is not purely factual; its nature
-

is disguised by the transference of commendation to the Gpr~al.

(c) Of the three lives suggested none accords with the popular
view which was that of Tellus the Athenian.' His ideal is
omitted altogether except when allowed in by the backdoor
suddenly in book x. (d) It might be argued that he omits to
4
give sufficient consideration to p0' < ~ O V ~ S O ~ K$VEV { ~ O Y ~ S
and it might be argued further that not enough deference
is shown to +vXiKd~-people may have meant by this some-
thing genuinely internal, as did later the Stoics. Quite
possibly c6SarpovLa did have this 'meaning' too.
-

Nevertheless, 'happiness' is probably after all to be pre-

ferred as a translation, partly because it is traditional, and still
more because it is fairly colourless.~It seems to me very rash to
assume that in common English 'happiness' obviously means
feeling pleased: probably it has several more or less vague mean-
ings. Take the lines:
** . . . ** This passage appears to be a later addition by Austin, which we
inserted here in a somewhat expanded form.-Ed.
' Tellus the Athenian is mentioned by Plutarch in his Lfe 4Solon, where
Solon is said to describe him as a happy man on account of his honesty, his
having good children, his having a competent estate, and his having died
bravely in battle for his country. Tellus appears to have been a plain man whose
happiness, nevertheless, Solon holds up to Croesus.-Ed.
Success also does not always or usually refer to life as a whole: and it has
perhaps a nuance of competitiveness.
 A r A 8 O N A N D EYAAIMONIA
This is the happy warrior, this is He
That every Man at arms should wish to be.

I do not think Wordsworth meant by that: 'This is the

warrior who feels pleased.' Indeed, he is

Doomed to go'in company with Pain

And fear and bloodshed, miserable train.

(Though no doubt his life is rr8' j&Mjs i j o 6 ~d v ~ vjsovijs as

~ristotlelikes to assume or feebly to argue his own chosen
one must be.) What every man at arms is being incited to wish,
is not so much to get for himselffeelings comparable to those
of the -paragon,
-
as to imitate his life. I think, then, that if we
are on our guard against misleading nuances, 'happiness' is still
the best translation for ~68ar~ovia.
The question of the relation of ~68ar~ocovia to 4 8 0 ~ 4has not,
it should be noticed, yet been cleared up. It to a great extent
coincides with the equally difficult problem of the relation of
76 6yayaBdv to sdl j ~ d~ 6 t hmust be reserved for a separate
discussion later.
ciya0dv-Does Aristotle tell us what its meaning is ? Pro-
fessor Prichard's contention is, it will be remembered, that
Aristotle really means by dya66v 'conducive to our happi-
ness'. Not concerning ourselves for the moment with the
precise interpretation given, we may notice, firstly, that
he is purporting to give Aristotle's answer to a question
which Aristotle himself expressly declines to answer, viz.
what is the meaning of dya6dv? And secondly, that the answer
given, implying as it does that dya6dv does have a single
meaning, is of a kind which Aristotle himself is at pains to
prove impossible.
Let us take the translation of 6ya6dv as 'good' here for
granted, and let us once more, following Professor Moore,
distinguish between two very different sorts of question, which
are commonly asked in books on ethics. W e are investigating,
we may say, the Good: but we may intend to ask: (I)w~&
 A r A g O N AND E Y A A l M O N I A 2I

does the word 'good' mean?' or (2) What things are good, and in
what degrees? Of course, these two questions may be formu-
lated in a variety of ways. For ( I ) we may substitute: What
is the nature of goodness? or: In saying of anything that it is
good, what am I saying about it? even: What sort of a predi-
cate is 'good'? and so on. For (2) we may substitute: Of what
things may it be truly said that they are good? And which-is
the best of them? and so on.
T o these two different sorts of question, as Moore claimed,
we shall get two corre~~ondin~l.17 different sorts of answer.
To ( I ) the answer might be: Goodness is a simple unanalysable
quality like yellow, or: 'Good' means 'approved by me',
or: To say of anything that it is good, is to say that it is con-
ducive to happiness, or: 'Good' is an evaluative word. Whereas
to (2), the answer might be: Friendship is good, or: Violence
is better than justice, and so on. Note here that it is assumed
that the only sense in which a'ya6dv has a 'meaning' is some
'factual' sense. Aristotle assumes this too.
Now we must ask, on this distinction does Aristotle con-
cern himself with both these questions or with one only?
And if the latter, with which? And I think that we must
answer, that he concerns himself professedly with the second
only.z
For Aristotle himself is aware of and draws this distinction
between the two questions, and says that, in the ~ t h i c s he
, is
concerned only with the second. This he does in the celebrated
chapter r. vi. He there confutes those who had supposed that
the word dya6dv always stands for a single identical yredicate.3

Which is itself probably an ambiguous question, but we nzay let this pass
for the moment. (We also have to distinguish (a) how to translate a word?
( b ) what does someone say about the analysis of the meaning or of the defini-
tion of a word? and both of these from (c) possibly different senses of 'mean'.)
And with that only in the more special form: 'What particular things
are good, and in what degrees, for man?'
Compare H. W. B. Joseph, Some Problems in Ethics (Oxford, 1g31), p. 75;
'That goodness is not a quality is the burden of Aristotle's argument in NE
I. vi.': but this is not quite correct. Aristotle is anxious to say that dyaedv has no
22 AFABON A N D E Y A A l M O N I A
But in proving this, he does not tell us what are the various
meanings of dyaOdv--the furthest he goes in that direction is to
give us a hint as to how the various meanings may be related
to one another, i.e. how the variety yet forms a unity. He then
dismisses the matter, 1096~30:dM' b o g ~ a c r a+v d+erlov ~ r b
vcv. d ( a ~ p i p o ~ y&p
v hipa h & v dMTs Zv E ~ T~ u \ o o o + ~ aO;KEL&
s
T E ~ O V . ~He then turns to his present problem: What is that
- -
good which is rrparrbv ~ a Kl T V T ~ V dv0P;rrY, that is: What
particular things are good for man (and in what degrees)?
Unfortunately, as is well known, he does not in fact discuss 'the
meaning of &ya0dv' elsewhere.
It is clear, then, that Aristotle declines in general to discuss
the meanings of &yaOdv but argues that it has no single meaning.
In both respects Professor Prichard's view conflicts, prima
facie, with Aristotle's statements. Nevertheless, it may be urged
both that Aristotlc is unjustified in declining to explain the
meaning of dyaOdv, and that he must himself attach some mean-
ing to 2 hl using it throughout the Ethics, which we may be
able to discover. And further, that meaning might be identical
in all important cases (this will be explained later).
If asked to justify himself, there is no doubt how Aristotle
would reply.2 The NE is only intended as a guide for politicians,
- ~

and they are only concerned to know what is not what

goodness means. Probably Platonists would have said that
I cannot discover what is good until I have found the definition
of goodness,
- but ~ristotlewould claim that the definition, if
-
possible at all, is only necessary if we wish to demonstrate
scientifically that certain things are good. So much d ~ ~ @ ~ c a
is not called for in the NE, and in any case one can know what
things are good without knowing the analysis of 'good'.3
sitzgle meaning- whether a quality, a relation, or anything else. As a matter of
fact, he says it docs soinetirncs stand for a quality.
'But perhaps these subjects had better be dismissed for the present; for
perfect precision about them would be more appropriate to another branch of
philosophy.'-Ed.
Aristotle does to some extent reply to this objection in r. vi. 14-16.
3 I do not, however, accept Burnet's exaggerated view of the 'dialectical'
method of the NE.
 A r A $ O N AND E Y A A I M O N I A 23

Whether this reply is satisfactory is doubtful. Certainly there

is much to be said on Aristotle's side. ~ i r s t las
~ Moore
, pointed
out, we can know something to be true without knowing its
analysis. Secondly, as Aristotle pointed out in I. vi. 4, if good-
ness is an isolable, definable property we should be able to
study it in isolation from all subject-matters, whereas in fact
there is no such study. Thirdly, even those who, like Moore,
find goodness 'unanalysable' still go on to discuss what is good.
But in any case it must be admitted that his method has its
dangers; and whether they are serious can best be judged by its
results in the body of the NE. (In at least two cases, those of
jdov4 and +tXla and perhaps also in that of ~ 1 ~ the
- -
4 , lack of

a clearer account of the meaning or meanings of byaedv is in

fact most serious.)
The extent to which Aristotle does in fact attach a discover-
able, and even single, meaning to dyaeo'v will, of course, con-
cern us largely. BG before proceeding, it is worth considering
an example of the extreme lengths to which Professor Prichar d
is prepared to go in imputing inconsistency to Aristotle. His
view, we said, implies that Gyaedv does have a single meaning,
i.e. does always stand for an identical common character in
the subjects of which it is predicated. (Note that we need
'character' here, not 'quality'; for, of course, one of Professor
Prichard's main contentions is that Aristotle mistakes 'being u
conducive to our happiness' for a quality of that which is so
conducive, whereas it is not.) And this, we said, conflicts
with Aristotle's own statement that &yaOdvdoes not do this.
Now it may be thought that I have been wasting unnecessary
words over this, since Professor Prichard himself notices this
objection and answers it on pp. 32-3 :
But Aristotle in saying, as he would have said, that in pursuing,
e.g. an honour, we are pursuing it i s hya96v, could only have meant
that we are pursuing it in virtue of thinking that it would possess
a certain character to which he refers by the term &yaOdv,so that by
dya8o'v he must mean to indicate some character which certain
things would have. Further, this being so, in implying as he does
24 A r A 8 O N AND E Y A A I M O N I A
that in pursuing things of certain different kinds K ~ Bads& ' we are
pursuing them i s dya8d, he must be implying that these things of
different kinds have, nevertheless, a common character, viz. that
indicated by the term a'ya8dv. It will, of course, be objected that he
expressly denies that they have a common character. For he says:
ri&s 6; ~ a #Jpov$ocws
i ~ a 4604s
i ; r c p ~ i ~ a 8ia#Jkpovrcs
i oi Adyoi
sadsY fi a'yaOd (Ethics I. vi. I 1 ) . 1 But the answer is simple; viz. that
this is merely an inconsistency into which he is driven by his inability
to find in these things the common character which his theory
requires him to find, and that if he is to succeed in maintaining that
we pursue these things of various kinds i s 6yaOd he has to maintain
that in spite of appearances to the contrary they have a common
character.
W e are not concerned for the moment with this doctrine
about 'pursuing things &S dva9a'.'z But what we must notice
is that Professor Prichard quotes the sentence about Tip<,
$pdqols, and 480mj, as though it were an unwilling admission,
into which Aristotle's honesty drives him: as though Aristotle
is admitting that in certain cases he cannot find the common
character, which he must find. Yet a glance at the context
will show that, on the contrary, the whole passage is designed
to prove that there is no such common character: and the
sentence quoted clinches the argument.3 It is not too much
to say that, if T L J * and
~ the rest are possessed of a common
character denoted by +Oo'v, the whole argument against the
Platonists is undermined: and Professor Prichard must accuse
Aristotle not so much of inconsistency as of bad faith.
In any case, this passage cannot bk thus lightly dismissed.
The objection retains, it seems to me, very considerable weight.
'But of honour, wisdom, and pleasure, just in respect of their goodness,
I

the accounts are distinct and diverse.'-Ed.

&S Aya8d does not in Greek mean 'in virtue of their possessing 4 certain
characteristic, "goodness" '. It means rather 'pursuing them in the w a y in which
we pursue things we say aregood'. Thus the phrase is noncommittal as to whether
there is a common quality or not. Cf. in English 'stating as a fact', 'stating as
a matter of opinion'.
3 Aristotle does not say he 'cannot find' a common character denoted by
dya06v in these instances: he says that he knows dya0dv stands for dflerent
characters.
 A r A 8 0 N AND EYAAIMONIA 25

iya86v-Doe~ it mean 'that which is desired' ? W e may now

turn to the main problem, the meaning of a'ya86. It must,
I think, be agreed, that there is some chance of our being able
to discover Ghat Aristotle does in fact believe about the mean-
ing of the word: but, since Aristotle appears to decline to
assist us, that chance is small.
It is on p. 3 1 that Professor Prichard begins his attempt to
elucidate the meaning of iya86v: and he says that 'Aristotle's
-

nearest approach to an elucidation' is to be found in I. vi. 7-1 I I

and vii. 1-5. Certain statements there made, he says, 'seem in-
tended as an elucidation of the meaning of dYa86v9v':nevertheless
they fail, even formally, to constitute such an elucidation: and
moreover, if they are 'taken seriously as an elucidation', the
result is to render absurd other well-known doctrines in the
Nicomachean Ethics.
I suppose it cannot be denied that Aristotle is capable of
getting himself into pretty. tortuous confusions; but this time,
at least, I think he can be exonerated.
In the first place, it seems to me that the statements in question
are not intended at all as an 'elucidation of the meaning of
dya8L' in Professor Prichard's sense.2 It is, consequently, not
surprising that they fail formally to constitute such an elucida-
tion: what is surprising, I think, is professor ~richard'sreason
for saying that they so fail. In the second place, if they are
'taken seriously as an elucidation', they do not so much
render absurd the doctrines mentioned by Professor ~richard
as others which it is easy to find. In these passages, says
Professor Prichard, Aristotle
speaks of T& K U ~ ' a;+& $ L W K ~ ~ C~ Ya~&i y a ~ c h p ~ vasa called a'ya8d in
one sense . . . and he speaks of T& T O ~ ~ T LT OK~&T W V{ ~ V h a ~ T l KTaW' S as
called dYaOci in another sense, and he implies that these latter are ~ L W K T ~
~ aa i p ~ s & $ Z~T' E ~ O V(EthicsI . vii. 4) . . . Further he appears to consider
that the difference of meaning is elucidated [sic] by referring to the
I For '7-11' we should read '8-11'.
If any passage deserves the description 'Aristotle's nearest approach to an
elucidation' it is probably I. vii. 10 (not mentioned by Professor Prichard); on
this v.i.
26 A r A O O N AND EYAAIMONIA
former as +ad& K ~ Baha' ' and to the latter as dyagci Std racra, i.e.
ciya9ci 8th ciyadh ~ a 9 aJrd.
' But this unfortunately is no elucidation,
since to state a difference of reason for callirzg two things dYa9dv is not
to state a difference of meaning of dyaddv, and indeed is to imply
that the meaning in both cases is the same. Nevertheless, these
statements seem intended as an elucidation of the meaning of
dyaOdv. And the cause for surprise lies in this, that if they are taken
seriously as an elucidation, the conclusion can only be that dya96v
includes 'being desired' in its meaning, and indeed simply means
T&S or end.

It will, I am afraid, necessarily take some considerable space

to clear this matter up.
Let us, as a preliminary, remark that the passage vii. 1-5 is
not concerned with a t all the same matter as vi. 8-1 I. In vii. I
he asserts that T i a'yaOdv of any activity whatsoever is TA TAOS
(incidentally this passage implies that 76 a'yaO6v and T i r;Aor
do not mean the same): he then proceeds to ask, what is the
T ~ A O S TDY T ~ ~ K T~ T~ ~Vw T w vAnd
? the word &yaOd~is not so
much as mentioned again. Moreover, though he does proceed
to distinguish a +CAos K ~ B a h' b S L W K T(alp~rdv)
~V from a rgAos SL'
hcpov S ~ w ~ r d(alpsrdv),
v he does not assert that the word T ~ A O S
has (in any sense) two meanings, only that there are some
76% which are rdrba (complete or final), others which are not.
Still less, of course, does he assert that r;Ae~a has two meanings, -

or endeavour to elucidate its meaning: he says nothing about

it at all; just as, in vi. 8-11, no mention is made of T ~ A O S
at all.
W e may confine ourselves, therefore, to vi. 8-11. Upon
what is Aristotle there engaged? In the earlier part of the
chapter he has been arguing that a'yaO6v cannot have always an
identical meaning. But now he produces an attempted answer
to his own objections, which is to be understood by referring
to his own logic (Categories 1-15). A word (Gvopa) may
be used either ovvwvGpws, dpwvGrws, or rrapwvGpws. If, on each
occasion of its use, its connotation (d ~ a rocvopa
~ d Adyos njs
o;oLap) is identical, then the word is used o w w v ; ~ ~ If,
. on
 A r A O O N A N D EYAAlMONIA 27
different occasions of its use, its connotations are different,
then the word is used Q w v 6 p w r : e.g. KA& may be used to mean
6 6
'key' or collarbone', ~ + O V animal' or 'picture'. But there
is a third possibility: on different occasions of its use, the word
may possess conilotations which are partly identical and partly
different, in which case the word is said to be used r r a p w ~ 6 ~ w r . '
There are evidently many ways in which a word can be used
paronymously: Aristotle names some of them, and gives
examples ( ~ bJ y , r ~ v b in Met. 1003~33, ~d E a ~ ~ c ~and d v ~d
J Y ~ ~ ~ vin6 Met.
v 1060b37. Also notoriously ~b GY, with which
a'ya0dv here is again compared, and ZV. For more details,
the reader may refer to those passages, or to Burnet's notes
on p. 29 of his edition of the Ethics.)"
'6
o n e such type of paronymity is known as the Tp& ZV".
When we speak of a 'healthy exercise' the word 'healthy'
has a connotation which is only partly the same as that which
it has in the phrase 'a healthy body': a healthy exercise is an
exercise which produces or preserves healthiness in bodies.
Hence healthiness", when predicated of an exercise, means
'productive or preservative of healthinessb', i.e. of healthiness
in the sense in which it is predicated of bodies. Thus 'healthi-
A

nessb' and 'healthinessa' have connotations which are partly

identical and partly different.
Now, in our present passage, vi. 8-1 I , Aristotle is producing
as an objection against himself the fact (admitted elsewhere,
cf. Rhet. I 362a27) that alya0dv is paronymous in this way. His
opponents are supposed to claim that, although a ' y a ~ ddoes ~
not always have an identical meaning, that is merely because
it is paronymous in the above manner. Sometimes it means'x',
sometimes etc., of x', etc.; and clearly it is only
the 'nuclear' meaning of 'x', which is common to both, with
Compare the traditional classification of terms as univocal, equivocal, and
analogous.-Joseph, An Introduction to Logic (Oxford, 1916), p. 46: 'analogous'
is unsatisfactory, since K ~ T '&vahoytav is only one form of paronymity.
This supplements and amends Cat. I. a bit; homonymy and paronymy
are defined by very limited examples there. Consider also Rhet. 1362~21fX
with regard to this point.
28 A P A 8 0 N AND EYAAIMONIA
which they are concerned. And it is always identical. To
which ~rGtotlereplies as follows. First, let us distinguish
ciyaO6v as used in the one sense, from dYaOdvas used in the other,
by substituting for it, in the one case i y a O & v K ~ O ahd,
' and in the
other a'ya0b~BL& dyaOdv TL K ~ O 'a6~Ol.IThen let us disregard all
cases where d p O b is used in the latter sense. And finally ob-
serve that, even in cases where it is used in the former sense
only, it still has not always an identical meaning.
Now let us ask, with reference to Professor Prichard's
argument quoted above, does Aristotle intend here to 'eluci-
date the meaning' of dyaOdv? It is clear that there is no simple
answer, Yes or No, to this question. He does intend to point
out that it has at least two meanings, a t least in a reasonably
strict sense, and to explain how those two meanings are related,
and to show how they are in part identical. But he does not
intend to elucidate that identical part. Moreover, he proceeds
to assert that in fact dyaOdv, used in the nuclear way, has at
least three different meanings (so that there may be a t least
-

six meanings of a'yaO6v in general): but he does not mean to

elucidate any one of the three, nor even to explain how they
are related to ,each other. It is not, then, surprising that Pro-
fessor Prichard should feel dissatisfied; if we ask his over-
simplified question, the passage must 'seem intended as an
elucidation' of the meaning of a'yaOd~, and yet fail to be one-for
plainly, Professor ~richardwould only count as a real elucida-
tion, an elucidation of the meaning of a'yaOdv in the nuclear
sense or senses, which Aristotle is not concerned to give h m .
Professor Prichard's remark, that 'to state a difference of
reason for calling two things a'yaOd~is not to state a difference
of meaning of a'yaOdv, and indeed is to imply that the meaning
in both cases is the same', remains to me obscure. But it seems
clear that he has not appreciated the doctrine of vap&vupa,since
he uses the rigid dichotomy 'same meaning-different meaning'
whereas ?ra@vupa are words which have meanings partly the
same and partly different. His 'difference of reason' might be
I I I F ~ might
~ E have been expected rather than Scd: but 8 ~ dis more general.
 ATAOON A N D E Y A A I M O N I A 29

the point about, say, healthy. Then Aristotle would accept the
claim, and say that what he offers is not an elucidation and is not
intended to be one. W e have seen, however, that Professor
Prichard is prepared to believe that the passage may be seriously
intended as an elucidation of a'ya8dv in the 'nuclear' sense. Seeing
that he has already dismissed the distinction between AyaOci K ~ B '
a;+d and a'ya8ci 6 ~ hracra as 'no elucidation', and seeing - that
Aristotle makes no attempt to elucidate the meaning of dra8dv
in the nuclear sense, and further states that it has at-least three
nuclear senses, it is clearly bound to be difficult for Professor
~richardto find anything in the passage which could be de-
scribed as 'an elucidation of the meaning of alya8dv'. He fastens
-

upon the words ~ C W K ~ ~~ Ea dYandp~va,

Vi ~ ~L~KC ~ LT~ ~K C
O ,~ E V

and concludes that, according to Aristotle here, dya8dv 'simply

means T ~ ~ OorS end'.=
Let us proceed to consider first Professor Prichard's argu-
ments to prove that iya8dv does not have that meaning. After
asserting that it 5uite certainly is not used by Aristotle with
that meaning [which in a way is true: but surely, then, it is
unlikely that he intends to ascribe that meaning to it in this
he proceeds as follows: 'Apart from other considera-
tions, if he did [mean by &ya86vsimply T~AOX], then for him
to say, as he in effect does, that we always aim at a'ya8dv 71
would be to say nothing, and for him to speak, as he does, of
the object of po6h7)a~g as ~ d ~ a 8 dwould
v be absurd.'
The first of these arguments is perhaps too concise to be
clear. 'We always aim at &ya8dvTL' may perhaps mean
whenever we act, we aim at a'ya0dv 71, or
I.
2 . whenever we aim, we aim at ciya8dv TL.
If (I) is meant, I very much doubt whether Aristotle would
maintain it; for we sometimes act from b~8vplaor Bvpds,
and then we aim at 486 Tc or at dv+th;~~a~s. But he might sayz
that in all deliberate action we aim at a'ya8dv ++ as he implies in
I It sems to me doubtful whether ~Chosmeans simply 'something desired':
but for present purposes it will do no harm to suppose it does so.
Wrongly.
30 AFAOON AND EYAAIMONlA
the first sentence of the Ethics.' However, this qualification
need not concern us here (though, since he tvortld certainly have
said that every action aims at some T~AOY, it is clear enough
that a'ya8dv does not mean TCAOT). If, then, ( I ) is meant, and
if a'ycz9dv T L means T ~ A O ST b , Aristotle will be saying 'Whenever
we act, we aim at some end.' But to say that is not to say
noth@: indeed, it is to say something which Professor Prichard
declares, on p. 29, to be false.2 At most the sentence is pleonastic:
for we could write srvds for ~ i A o v r ~ i v d s(or dyaOov^ ~ ~ v d s )
without loss of meaning. But that is a very minor matter.
And this would serve, e.g., to define o'ya9dv.
If (2) is meant, the sentence is certainly emptier: 'Whenever
we aim, we aim at some end.' It might be said to be a matter
of definition. But it might well be said, for emphasis, in ex-
plaining the notion of 'aiming', or in otherwise explaining the
use of words. However, for reasons similar to those given in
the preceding paragraph, I do not consider that Aristotle
would ever have made such a remark about iya8dv, though he
might-
well have made it about T ~ A O YIf. we change it to 'when-
ever we aim deliberately, and only then, we aim at Gya8dv +L',
Aristotle might agree: but then, if we substitute T A O T L~
, for

i y a 9 d v TL,the statement would simply be false. Thus proving,

though not by Professor Prichard's argument, that dyaddv does
not mean T ~ A O S .
Now take Prichard's second argument. Here, no such objec-
tion applies as in the case of the first argument, for Aristotle
certainly would say that the object of /?o&Aqois is 7ciya8dv. If
GyaOdv means T ~ A O Xthe , statement becomes: 'The object of
F o & A ~ L is~ L~b~ T ~ A o F . ' NOWthis is not, as a matter of fact, as it
stands, absurd-it is in fact a remark which Aristotle often
makes, with good sense (cf. e.g. I I I I b26, or I I I 3 'I s,+ /i?o&Aqu&Y
706 T ~ A O V 2OT:v).
S However, ~d r i h o s is there being contrasted
with T & ?rp& T ; T ~ A Oand ~ , /?06Aqcrcr with xpoat'prars: whereas
I The importance o f which should not be exaggerated-
2 Because he understands by TE;\OS only what Aristotle call a ~ C h o srapd
T$V .rrp6[rv, I confess that his criticism at this point seems t o me perverse,
 A r A O O N AND EYAAIMONIA 3I

it is clear that Professor Prichard means, and fairly enough, to

refer to a context where /!Io&A~uL~ is being contrasted with
~ ~ ~ O v andp l a BvPds, and rdya0d.v with r6 j86 and 6 u r ~ A d ~ ~ u ~ s .
In such a context, it is clear enough that we cannot substitute
T ~ A O S for ayaOdv. For in the case of the other two types of
+;EELSwe are also aiming at a T~Aos-, and we need 6yaBdv not
T A Oto~ provide the contrast. Professor Prichard's arguments,
then, reduce to one, which I think is sound: GaOdv cannot
mean 'that which is desired', because Aristotle holds that there
are other objects of desire besides T(, dyaOdv.
Apart from this argument, it is absolutely clear from r. vi
itself that draOdv does not, for Aristotle, mean (or at any rate
merely mean?) 'that which is desired'. It is not merely that he
says that a'yaOdv has no single-
meaning, and hence, afortiori, not
that suggested. But further, he says i f stpi, + p 6 q ~ ~ands , $80~4
that they are all Gcw~dpevaKUB'a;rd, so that, presumably,
their A+OL 5 ~ L W K ~ ~ KE ~ VO ~' a;+&' are identical: whereas their
hdvo~m6rq ?)'yaOdare dgerent. So that being dyaOdv cannot
L

mean being desired.

I expect the reader has found the roof of this tedious.
- -

But it is important to establish the point, since the relation

between 'being +aOdv' and 'being desired' is one of the most
baffling puzzles in Aristotle's, or for that matter Plato's, ethical
theory. This puzzle, like that of the relation between ?)'yaBdv
and must be reserved for a separate discussion.
'Their accounts as things pursued for their own sake.' (Ed. transl.).
 ARE THERE A PRIORI CONCEPTS?I

MR.MAC L A G A N seems to have ransacked every avdable

cupboard for skeletons to grace our feast: I hope I shall not be
thought ungrateful for his assiduity, if I decline to pick with
him the particular bones wluch he proffers to me. My excuse
is that he, hke Mr. M a c h o n , has touched only perfunctorily,
if at all, on certain preliminary matters whch deserve discus-
sion. Frankly, I still do not understand what the question before
us means: and since I hold, nevertheless, no strong - views as to
how it should be answered, it seems best to occupy myself
primarily in discussing its meaning. I only wish I could do h s
more helpfully. I shall, therefore,-first write something about
concepts, and then something about their existence and origin:
finally, because it seems an interesting case, I shall choose to
argue with Mr. Maclagan about resemblance.

Neither Mr. Mackinnon nor Mr. Maclagan would claim, I

think, to have told us carefully what they are t a h g about
when they talk about 'concepts'. Both seem, however, to imply
that the word 'concept' could not be explained without using
the word 'universal':2 and h s seems also the common view,
though how the two are related is no doubt obscure and con-
troversial. I propose, therefore, to make some remarks about
Reprinted fiom Proceedings ofthe Aristotelian Sodefy, 1939, by courtesy of
the editor.
Mr.Maclagan perhaps denies this, in order to allow for the possibility of
'innate ideas': but he does not believe in these. Nor does he explain about
them: his brief account of a 'concept' as an 'element in our knowing or think-
ing' will scarcely suffice for that.
 ARE THERE A P R I O R 1 CONCEPTS ? 33
'universals': because I do not understand what they are, so that
it is most unlkely I shall understand what concepts are.
People (phdosophers) speak of 'universals' as though these
were entities which they often stumbled across, in some
familiar way which needs no explanation. But they are not so.
O n the contrary, it is not so very long since these alleged
entities were calculated into existence by a transcendental
argument: and in those days, anyone bold enough to say there
4
were' universals kept the argument always ready, to produce
if challenged. I do not know' if it is upon
-
h s argument
- that
Mr. ~ a c h o and n Mr. Maclagan are r e l y g-. It may be that
- ~

6
they do claim to stumble across universals' in some easy
manner: or it may be that they rely upon some other argument
which is admittedly transcendental. But I propose to consider,
not very fully, that celebrated argumen; which, above all,
seems suited to Drove
A
the existence-of 'universals' in the most
ordinary sense of that word: it runs as follows:
It is assumed that we do 'sense' t h g s , which are many or
diRerent.2 Whether these h g s are 'material objects' or what
are commonly called 'sensedata', is not here relevant: in fact
the argumentcan be made to apply to the objects of any kind
of 'acquaintancey, even non-sensuous-although such applica-
tions were not originally envisaged. It is assumed, further, that
we make a practick o f c i h g mimy cLfferent sensa by the same
single name: we say 'This is greyp,and ' That is grey', when the
sensa denoted by 'this' and by 'that' are not identical. And
finally it is assumed that this practice is 'justifiable' or indis-
pensable. Then we proceed to ask: How is such a practice
possible ? And answer:
(a) Since we use the same single name in each case, there must
surely be some single identical thing 'there' in each case:
For there are in fact several: see below.
There is a constant and hardid ambiguity here: the sensa are commonly
different both 'numerically' and 'qualitatively' (the former, of course, always).
The 'universal' is alleged to be single and identical in bath ways. Hence, fiom
the start, that fatal conhion of the problem of 'genus and species' with the
problem of 'universal and particular'.
34 ARE THERE A P R I O R 1 C O N C E P T S ?
something of which the name is the name: soinethiilg,
therefore, which is 'common' to all sensa called by that
name. Let this entity, whateter it may be, be called a
6
universal'.
(b) Since it was admitted that the things we sense are many
or different, it follows that this 'universal', which is
single and identical, is not sensed.
Let us consider this argument.
I.This is a transcendental1 argument: if there were not in
existence something - other than sensa, we should not be able
to do what we are able to do (viz. name thngs). Let us not
consider here whether, in general, such a form of argument is
permissible or fruitful: butit is important to notice the follow-
ing points:
(i) The 'universal' is emphatically not anything we stumble
across. W e can claim only to know that, not tohat, it is. 'Uni-
versal' means that which will provide the solution to a certain
-

problem: that x which is present, one and identical, in the

different sensa which we call by the same name. Unfortunately,
as so often happens, succeeding generations of philosophers fell
naturally into the habit of supposing that they were perfectly
well acquainted with these entities in their own right: they have
any amount to tell us about them (partly t h s was due to a
confusion of 'universals' in our present sense with 'universals'
in other senses, as we shall see). For instance, we are told that
they are 'objects of thought9:and myths are invented, about our
'contemplation' of ursaneivls: and so on.
I

(ii) On the same grounds, it must be held that to ask a whole

series of questions which have constantly been asked is non-
sensical. For instance: 'How is the universal related to the
6
particulars?', Could there be universals without instances?',
and many others. For a 'universal' is dejined as something w h c h
is related to certain sensa in a certain way. W e might as well
In Kant's sense. But it is .Lo 'transcendental' in another sense, that o
proving the existence of a class of entities different in kind from sensa.
 ARE THERE A P R I O R 1 CONCEPTS? 35
worry about what is the relation between a man and h s aunt,
and as to whether there can be aunts without nephews (or
nieces).
(iii) Here, however, a point to w h c h I have already twice
referred in anticipation, must be made: this will unfortunately
be a digression. There are other transcendental arguments for
'the existence of universals'. I shall mention one: A true state-
ment is one which corresponds with realty: the statements of
the scientist are true: therefore there are realities which corre-
spond to those statements. Sensa do not correspond to the
statements of the scientist (exactly why, is rather too obscure to
discuss here): therefore there must exist other objects, real but
not sensible, w h c h do correspond- to the statements of the
scientist. Let these be called 'universals'.
That this argument begs many questions, is evident. Are all
sciences alike? Is all truth correspondence? Does no science
rmke statements about sensa? Some, for instance, would dis-
tinguish 'a priori' sciences from 'empirical' sciences: and hold
that the 'truth' of the former is not correspondence, while the
statements of the latter are about sensa. Of course, too, the
assumption that the sciences are true is a large one. I But all this
cannot be entered into.
What it is important to notice for our purposes is that here
too the argument
-
I transcendental. The 'universal' is an x, which
is to solve our problem for us: we know only that it is non-
sensible, and in addition must possess certain characters, the
lack of which prohbits sensa from corresponding to the state-
ments of the scientist. But we do not stumble across these
'universals': though, needless to say, philosophers soon take to
tallung as though they did.
Now it must be asked: what conceivable ground have we
for identifying the 'universals' of our original argument with the
'universals' of this second argument? Except that both are
Even Plato once decided that he ought not to make it. It has been sug-
gested to me that the argument should be formulated in terms of 'having
meaning' rather than of 'being true'. I doubt if this would be any irnprovemmt,
36 ARE THERE A P R I 0 R . I CONCEPTS?
non-sensible, notiwg more is known in whch they are allke.
Is it not odd to suppose that m y two distinct transcendental
arguments could po&ibly be known each to prove the existence
of the same kind of h n g ? Hence the oddity of speaking of
'arguments for the existence of universals': in the fiat place,
no two of these arguments are known to be arguments for the
existence of the same thing: and in the second place, the phrase
-

is misleading because it suggests that we know what a 'uni-

versa' is quite apart from the argument for its existence-
whereas, in fact, 'universal' means, in each case, simply 'the
entity which this argument proves to exist'.
As a matter of fact we can, indirectly, show that the objects
'proved' to exist by the two arguments so far mentioned are
not the same. For firstly, the variety of 'universals' proved to
exist in the case of the first argument is strangely than
in the case of the second argument: the former proves a 'uni-
versal' to exist corresponding to every general name, the latter
only does so when the name is that of an object studied by the
scienti~t.~ But it might still be thought that the 'universals'
proved to exist by the second argument do, nevertheless, form
part of the class of 'universals' proved to exist by the first argu-
ment: e.g. 'circularity' or 'straightness' could be proved to
exist by i t h e r argument. Yet, in fact, no clearer cases could be
chosen for demonstrating that the two kinds of 'universals' are
distinct: for if 'circularity' is to be proved to exist by the first
argument, then I must b; able to say truly of certain sensa 'this
-

is circular': whereas, the 'circularity' which is to be proved to

exist by the second argument must be such that it cannot be
truly predicated of any s e n ~ a . ~

It is to be remembered that, if we are to argue that 'science' is not about

1

sensa, very little can be recognized as 'science'.

a It cannot be sense to say that sensible circles are more or less 'like' the
universal 'circularity': a particular can be like nothing but another particular.
Nor can I agree with Mr. Maclagan that, on his account, the 'sensible figure'
could be an approximation to the 'geometrical figure': for what is sensed can
be like nothing but s o m e t h g Llsewhich is sensed. But I must allow that 'non-
sensubus perception', 'intuitive acquaintance' and so on seem to me to be
 ARE THERE A PRIORI C O N C E P T S ? 37
The purpose of h s digression is to point out that, apart
altogether from questions as to whether the 'arguments for the
existence of universals' are good and as to whether they permit
us to talk further about univenals, an immeasurable confusion
arises from the fact that 'universal' may mean at any moment
any one of a number of different h g s . For example, if
'universal' is being used in the sense of the second argument,
it is good enough sense to ask 'How are univenals related to
particulars?' though any answer would be difficult to find.
(The answer that particulars are 'approximations' to universals
not only implies that the two are the same in kind whereas they
were said to- be different, but also again exposes the difference
between this argument and the other; since it would be absurd
to say, of some non-scientific object like a bed, that there was
no sensible bed whch was really a bed, but all sensible beds
-
6
were only more or less remote approximations' to beds.)
Again, to ask 'Are there universals without instances?' is now
absurd for the reason that a 'universal', in the sense of the second
argument, is not the sort of thing which 'has instances' at all
(indeed, someone will certainly be found to apply the first
argument to the objects 'proved' to exist by the second argu-
ment).
2. SO far, we have not investigated the validity of our
argument.
(i) It is to be observed that if the argument holds in its first
part (a), it certainly also holds in its second part (b). If there
'are universals', then they are not sensed: the whole point of
the argument is that there must exist somethmg of a lund quite

contradictions in terms, attempts to have things both ways. I find confirmation

of this when Mr. Maclagan says a sensible circle might be more than an
'approximation' to a geometrical circle: i.e. as I understand him, it would be
a geometrical circle, although we did not know it. Thus he is making the objects
of intuition the same in kind as the objects of sense-indeed interchangeable.
I wonder if Mr. Maclagan's non-sensuous intuition is such that we can say on
occasion, 'This is a (geometrical)circle'? For whatever reason, we do not ever,
I think, speak so. Yet, surely, if we are 'acquainted' with geometrical circles
we ought to be able to do this.
38 ARE THERE A PRIORI C O N C E P T S ?
different from sensa. Nevertheless, a fatal mistake has been
made by many philosophers: they accept thefirst part of the
argument ('there are universals') which, as we shall shortly see,
is wrong, and they reject the second part, which is a necessary
corollary of the first. Of course, the- talk is at first still to the
effect that universals are 'thought': but theories are soon
formed as to how we 'abstract universals from particulars' and
then 'see universals in particulars'.'
-
Undoubtedly, there are
'reasons' of a h d for constructing these theories and rejecting
the 'separation' of universals, of which the following will be
the most pleasing to self-refuting nominalists: if we accept both
. - and (b), it becomes difficult to give any account of how I
(a) -

come to classify together the various things called 'grey'; true,

if and when I am correct in classifying a certain sensum as
6
grey', then the universal must be 'in' it: but it is not sensed 'in'
it: how then am I to decide whether it is or is not there, or even
guess it?2 Hence we depart from the argument in its pristine
form, and embark on Gythologies: and by the time we have
finished, we may well have reached the position of so many
philosophers, and hold that what I do sense are 'universals',
what ~ h not o sense are 'particulars';) which, considering the
meanings of the words, comprises two self-contradictions.
(ii) Finally, it must be pointed out that the first part of the
argument
- (a), is wrong. Indeed, it is so artless that it is difficult
to state it plausibly. clearly it depends on a suppressed premiss
which there is no reason whatever to accept, namely, that words
are essentially 'proper names', unum nomen unum nominatum.
But why, if 'one identical' word is used, must there be 'one
identical' object present which it denotes? Why should it not
be the whole function of a word to denote many t h g s ? 4
Why should not words be by nature 'general' ? However, it is
in any case simply false that we use the same name for different
D o we taste universals in particulars too?
In this sense of 'giving an explanation of naming', the theory gives none.
3 Other theories: that the particular is just a cluster of universals: that the
universal is a particular of a special sort (an image).
4 Many similar things, on a plausible view: but other views might be held.
 ARE THERE A P R I O R 1 C O N C E P T S ? 39
things: 'grey' and 'grey' are not the same, they are two similar
symbols (tokens), just as the t h g s denoted by 'ths' and by
'that' are similar t h g s . In this matter, the 'words' are in a
position precisely analogous to that of the objects denoted by
them.'
But, it may be objected, by the 'same single' word it was
never meant that it is numerically identical. In what sense, then,
was it meant? If it meant 'qualitatively identical', then it is
clear that the sense in which there is an identical '.type7of the
tokens is just llke the sense in w h c h the sensa share in an identi-
cal common character: hence the former cannot be taken as
self-explanatory while the latter is admitted obscure. If it meant
that all these tokens 'have the same meaning', then we cannot
assume that it is the business of similar tokens to 'mean' some-
thing which is numerically self-identical, without begging our
whole question in the manner already pointed out.
But, it will be further said, I do sense s o m e t h g identical in
different sensa. How this could be I do not understand; but if
it is true, it is clear that this identical something is not an entity
different in kind from sensa.
I conclude that this argument does not prove 'the existence
of universals'; and that, if it did, n o t h g more could be said
about then1 than is said in the course of the argument itself,
except that they are certainly quite different from 'universals'
in other senses of that word, i.e. as 'proved' to exist by other
transcendental arguments.
In a certain sense it perhaps is sometimes not harmful to talk
- -

about 'universals' or 'concepts'; just as it is sometimes convenient

to talk about 'propositions', and as it is very often convenient
to use 'material object language7. To say somethtllg about
'concepts' is sometimes a convenient way of saying somethmg
complicated about sensa (or even aboG othe; objects of a;-
quaintance, if there are any), including symbols and images,
There are ways, of course, in which they are not so analogous: for
instance, that one token is of the same type as another, is determined by con-
vention as well as by similarity.
40 ARE THERE A PRIORI C O N C E P T S ?
and about our use of them:' though very different methods of
translation w d have to be employed on different occasions.
But on the whole there is remarkablv little to be said in favour
d

of 'universals', even as an admitted logical construction: the

plain man did not use it until he acquired the habit from philo-
Jophers, and the errors into which that habit leads are very
common and numerous. For example, in addition to those
already noted, the error of tacing a single w o r d or term, instead
of a sentence, as that which 'has meaning9; hence, given some
word lJce 'resemblance', we search for what it denotes (cf. 111).
Or again, we confuse the view that all sentences are about
sensa, with the view that every word or term denotes a sensum.
O r again, and tlus most concerns us, we think of the 'abstracted'
universal as a solid piece of property of ours, and inquire into
its 'origin'.
I should Lke, then, to learn from Mr. Mackinnon and Mr.
Maclagan, 'what a concept is'.

In criticizing Mr. Mackinnon, and elsewhere, Mr. Maclagan

mentions two questions which, he says, it is important to ask
and to distinguish. hey are formulated in different ways, but
-

a typical formulation is the following:

(i) Do w e (actually) possess such-and-such a concept?
(ii) H o w do w e come to possess such-and-such a concept?
I recognize that these two questions have been distinguished
and asked, by philosophers of reputation: Descartes, for
example, asks both questions, severally, about 'the idea of God'.
Nevertheless, it may be doubted whether these really are
distinct and answerable types of question.
As for the first question it would have been helpful to me if
But we must 'be careful'. We must not say, for example, 'a universal is an
image*:Berkeley probably did not make this mistake, but Hume probably
did: hence Hume is led, whereas Berkeley is not, into a theory about 'the
origin' of olu ideas.
 ARE THERE A PRIORI CONCEPTS? 41
Mr. Maclagan had volunteered to argue it, at least in some one
case: for then I might have learned what it means, or how it
may be decided. I ~ifor the second question, it is, I think, quite
clear that Mr. Maclagan, llke Professor Broad, regards the
distinction between 'a priori' and 'empirical' concepts as a
matter of 'the manner in which we come to acquire' them.
Ths I notice as a preliminary, because, in the opening para-
graph of his section 11, he does not define 'empirical' in h s way.
That he really does mean what I claim is shown by a great
many remarks throughout his paper, but more particularly by
the opening paragraph of his concluding section, and by the
nature of the ensuing conclusions.
The principal objection which I have to bring against both
questions is the general one; that they are examples of the
nonsense into which we are led through the facile use of the
word 'concept'. A concept is treated, by Mr. Maclagan as by
Professor Broad, as an article of property, a pretty straight-
forward piece of goods, which comes into my 'possession', if
at all, in some definite enough manner and at some definite
enough moment:"hether I do possess it or not is, apparently,
ascertained simply by making an inventory of the 'furniture'
of my mind.
Let us consider first the first question: when we ask 'Do
-

we possess the concept?' what are we asking? If we are

a s h g about some individual, or about some group of indi-
viduals, whether he or they 'possess the concept of redness',
some meanings can well be attached to this expression. It might
be supposed -to mean, for example, 'Does he, or do they,
understand the word "red" ?' But that again needs further ex-
planation; we shall almost certainly find that it is s t d ambi-
guous, and that, at least on many interpretations, no precise
Perhaps he does intend to argue it-though, if so, he does not clearly
enough distinguish it from the second question-in the case of 'the causal
continuant': but there the issue is further obscured by our 'having' a concept,
the elements of which do not combine into a 'single' concept, and the content
of which is 'dim'.
a Can I 'lose' a concept, as well as acquire it?
42 A R E THERE A P R I O R 1 C O N C E P T S ?
answer can be given as to whether the individual does or does
not 'understand' the 'word'. Does the word 'red' matter?
Would it not do if he used 'rouge' and understood that? Or
even 'green' if he meant by that what most Englishmen mean
by 'red'? And so on. Perhaps we should say he 'possesses the
concept of redness' if he has paid attention to certain features
in that with which he is acquainted, to call attention to which
most Englishmen would use the words 'red', 'redness', &c., and
has adopied some symbolism to call attention to them, and has
-

not 'forgotten' either the features or the symbolism. This is still

only one of the things which might be meant, and is still not
precise enough: and all the dificultics arise with which we are
&en more familiar in the puzzles about material objects-
surely he might do all you say and yet still not possess the
concept- ?
But we do not need to pursue this question. For it appears
that this is anyhow not the question being asked by Mr.
Maclagan as his first question. 111 asking 'Do we possess, &c.?'
the 'we' does not mean us as contrasted with others; it means
any and everybody at once. It means, as he elsewhere puts it,
'Is there such-and-such a concept?' And now what does that
mean, if anything? It looks as though it had a meaning, because
it seems easy to proceed from the question 'Does Socrates
possess, &c.?' to the question 'Does arly rnan possess, &c. ?' But
although verbally llke the first, the secor~dquestion is really
very different. Similarly, although meaning may be attached
to the question, 'Does he, or dorhey, understand this word?',
it is not obvious that the question 'Does anyone understand this
word?' has a meaning at all. For, in the former of these questions,
to 'understand' means, speaking
-
roughly, to use as we, or as
- -

most Englishmen, or as some other assignable persons use: or

again, the features of his experience,' about which it is asked
whether he has or has not paid attention to them, require to
be indicated by referring to certain definite experiences of other
persons. Clearly nothing of the kind is possible in the case of
' ' I use this as equivalent to 'that with which he is acquainted'.
 ARE THERE A PRIOR1 C O N C E P T S ? 43
the second question. Yet it seems to me that its verbal simi-
larity to the first has led people to pose it-together of course,
with a belief in 'concepts' as palpable objects: if they were such,
the second' question would be rather U e the first. Whereas
it is not.
It seems clear, then, that to ask 'whether we posscss a certain
concept?' is the same as to ask whether a certain word-or
rather, sentences in which it occurs-has any meaning. whether
that is a sensible question to ask, and if so how it is to be an-
swered, I do not know: in any case, it is llkely to be ambiguous.
What it is here of interest to observe is, how our question
( I ) is liable to become confused with (2). Since it is going to
be awkward, to say the least, to prove that a certain concept
simply is not, it is tempting to try another way. Instead of
maintaining that it does not exist, we maintain that it cannot
exist. For instance, in certain cases we may hope to show that
an 'idea' is 'self-contradictory', as Leibniz thought he could
show of the 'infinite number' or Berkeley1 of 'matter'. This
Since Mr. Mackinnon makes much of this, a note may be excused.
Berkeley, I think, maintains exactly the same position with regard to 'matter'
as with regard to 'universals': these two are chosen as typical of the two most
popular kinds of entities alleged to differ in kind from sensa. He expresses
himself much more clearly about 'matter' than about 'universals', (though
always suffering from a lack of technical terms). He holds ( I ) that the plain
man's ordinary statements about 'ideas' or about 'material objects' are transla-
table into other statements which are solely about sensa (including symbols),
(2) that the plain philosopher's theories about the 'nature' of matter (inert,
&c.) and of universals (formed by abstraction, &c.) are nonsense: partly his
descriptions of these entities are self-contradictory (e.g. in the way mentioned
by Mr. Maclagan at the end of his section V), partly he can be shown simply
to have misunderstood the nature of a 'logical construction'. In one sense
'there are' both universals and material objects, in another sense there is no
such thing as either: statements about each can usually be analysed, but not
always, nor always without remainder.
Mr. Mackinnon seems t o me to underrate the second line of attack. I do not
thlnk that Berkeley would have been by any means content simply to pra-
pound the view that matter is a 'logical construction' and then to abandon the
plain man still asking for more: he patiently asks 'What more do you want?',
and laboriously shows that either what is asked for is nonsensical or else he has
granted it already. And this, if we will not be content to let plain men work
out their own damnation, is perhaps all that can be done. Nor do I think that
44 ARE THERE A P R I O R 1 CONCEPTS?
particular
-
method, however, is not suitable in some of the most
crucial cases, namely those of simple 'ideas' hke that of 'neces-
sity'. So here another method is tried: we claim to show that
it is causally impossible for .anyone to possess such a concept.
W e construct a theory about the condition or conditions under
w h c h alone we can 'acquire' concepts : and then we claim that,
in the case of certain alleged concepts, these conditions are not
satisfied in the case of any man: therefore no one possesses them,
i.e. they do not exist.
Thus Hume rapidly deploys a theory that we can never
come by any idea unless we have previously experienced an
impression similar to it: and argues that we cannot possess an
idea of power-in-objects such as we commonly thmk ourselves
to possess, because there is no antecedent impression of it.
But h s roundabout method of showing that I do not,
because I cannot, possess certain concepts, will not do. For how
is Hume's conclusion, that we possess no ideas not derived
from antecedent impressions, established? Presumably by in-
duction. But not merely is his survey of the evidence inade-
quate: if he is to make an induction, he must first consider all
cases where I do possess an 'idea', and discover the antecedent
conditions in each case. He must, therefore, consider whether
we do or do not have the idea of power-in-objects before
making his induction: and he must also, and this is more im-
portant, have some means of establishing in particular cases
whether we do or do not 'possess' a certain idea which is quite
distinct from the, as yet unformulated, theory of how we
Berkeley would say 'There are universals' quite so handsomely as Mr. Ma&-
non makes him do: he would not maintain that there are universals in any
sense in which he would deny that there is matter: Berkeley says that 'there
are general ideas' meaning that statements like 'all demonstration is about
general ideas' have a meaning-but also, he says that 'abstract general ideas'
i.e. general ideas as entities of a kind different from sensa, do not exist. (This
does not mean 'general ideas are sensa'.) Mr. Machnnon gives me the im-
pression, perhaps wrongly, that he thinks 'abstract' general ideas to be a
limited class of general ideas, which Berkeley denies to exist: but it is rather a
theory about the nature of general ideas in general that Berkeley means to deny.
(I omit the supplementary theory of 'notions'.)
 ARE THERE A PRIORI CONCEPTS? 45
'come to' possess them. Now what is that means? In Hume's
case, if the 'idea' were really an iniage, then direct introspection
might do, though it would then be surprising that he has to go
such a roundabout way to prove we have noidea of power-&-
objects: but t h s will scarcely do on some other theories of what
concepts are. I should very much llke to know what Mr.
Maclagan's method is.
It is also often very hard to gather how philosophers are
claiming to establish their theories of 'how we come to acquire
concepts'. Professor Broad, for instance, in the passage referred
to by Mr. Maclagan, says: 'It is quite certain that many, if not all,
simple intuitive dispositional ideas are formed in the following
way', and then proceeds to a theory about 'comparing', 'ab-
stracting', and so on. How is this proved? If it is known, then
I must confess myself bad at knowing. O r again, discussing the
question as to whether we do or do not 'have' a certain idea,
in connexion with Hume's problem about the shade of blue
which I have never actually sensed, Professor Broad says: '(a) If
by an "idea of the missingCshade"you mean an imageSwhichis
characterized by the missing shade, the question is purely a
question for empirical psychology. (b) If by "idea" mean
"intuitive idea", the answer is in the negative'. It is satisfactory
to have the answer: but it would be pleasant also to learn how
it is reached, and by whom.
It seems to me that it is seriously questionable whether Mr.
Maclagan's second question can be kept distinct from his first
question. The latter seemed to amount to the question: whether
a certain word has a meaning? But now does the question about
'origins'
- really not amount to this, that we want to know: how
do words mean? Hume's theory about the 'derivation of our
ideas' really amounts to the theory that a word, x, can only
have meaning provided that I can know, on at least one occa-
sion, that 'ths is an x', where 'hs'denotes something sensible.
And most other theories about thls subject, are really theories
of a very similar sort: Mr. Maclagan himself inclines to such a
view, though he would not, of course, add that the ' h s ' must
46 ARE T H E R E A P R I O R I C O N C E P T S ?
be sensible. The 'origin of a concept' is commonly admitted to
be found, when an occasion is found on w h c h I can say, with
knowledge, ' h s is an x'. Mr. Maclagan almost formulates the
problem hmself in h s way in tile opening of h s section 11.
The question of 'innate ideas' seems-very commonly to be
simply the question: whether a word can have a meaning even
though I can never know 'ths is an x'?' But surely it w d be
verydifficult indeed to keep the two questions: 'Has x a mean-
ing?' and 'How do words mean?' apart. It would appear that
to ask the latter is to ask 'What is meant by "having a mean-
ing" ?'2 Now, ifeither of these questions can be treated inde-
pendently of the other, it seems clear that it is h s latter (which
-

would be contrary to Mr. Maclagan's view); unless the

question, whether a certain word has a meaning, is to be taken
as absolutely unanalysable, and to be answered by nieans of
some sort oC direct &spection.
Nevertheless, it is certain that much discussion is devoted
seriously to the questions of 'origin' and 'acquisition' and 'forma-
tion' of concepts. And where t h s is so I suspect that we are
always told either n o h g or nonsense. For instance, Kant
speaks with emphasis on t h s subject, saying that the whole
W e may distinguish several questions: Do I know the meaning of the
word x? Do I know that there are x's? Do I know that there is an x here? Do
I know that this is the x which is here? But the theory that a word only has
meaning if I can know sometimes that 'this is an x' is so engrained in us that
we confuse these questions together: for, if it is true, I cannot answer any of
the questions in the affirmative unless I sometimes answer the last, and therefore
all of them, in the affirmative. Even Kant, it seems to me, who has every
interest in distinguishing these questions with regard to such words as 'duty'
and 'cause', since, as against Hume's verification theory, he is claiming that the
first three questions can be answered in the affirmative though the last must
be answered in the negative, fails to keep them always apart: just as he fails
sufficiently to distinguish the question, 'How arc synthetic a priori propositions
meaningful?' from, 'How are they known to be true?' Perhaps they cannot
be distinguished. I should say that Mr. Maclagan, in putting to me the question
he does about his classes (4) and (s), is posing the really difficultquestion: can
x have a meaning when I can never know that 'this is an xi? I wish I could
answer it.
This, if true, explains why theories about the 'origins' of our ideas are far
from being as inductive as they apparently should be.
 ARE THERE A PRIORI CONCEPTS? 47
point of Hume's theories, and the whole point of his own
transcendental deduction, is about the origin of a priori con-
cepts: it will be remembered how the wretched Beattie is
slated for not understanding this. With trepidation, I confess
I do not understand it either. Beyond that the 'origin' of these
concepts is 'in the nature of the mind itself', I cannot see that
we are given any information about it. Is the Metaphysical
Deduction intended to show that chis 'origin'
- lies in the forms
of judgement? A strange sense of 'origin'.-
And yet even so,
it seems to me that no account is given of the 'origin' of our
concept of 'necessity'.
The fact is that words k e 'origin'
-
or 'source' and -phrases
k e 'how we acquire' are very vague, and may mean many
dungs. Unless we carefully distinguish them we shall fan
into confusions. For example, Descartes' division of 'ideas' into
innate, adventitious, and factitious, seems, if we attempt to
eliminate the metaphor from the ways of acquiring
to rest on no singlefundamentum divisionis. I suppose that really
these expressions are intended to embrace all conditions ante-
I

cedent to our 'possession' of a concept: but plainly these condi-

tions are of very various h d s , for example, we may be given:
(a) Theories about the agents (spiritual or material) respon-
sible for my possession of concepts, e.g. myself, God,
material objects.
(b) Theories about the operations, w h c h lead to the 'forma-
tion' of concepts, e.g. the honest spadework of 'abstrac-
tion'.
(c) Theories about the materials on w h c h those operations
must be conducted, e.g. non-sensuous intuitions, sensa.
(d) Theories about the sources from whch concepts are
somehow drawn, e.g. the mind.
(e) Theories about the times or occasions on which concepts
are acquired, e.g. at birth, before birth, on the occasion
of sensing x.
Probably t h s list could be extended. I do not see that these
48 ARE THERE A PRIORI CONCEPTS?
questions are anythng but confusing, if taken seriously: (c) and
perhaps (d) might be interpreted to make sense. It is strange that
we should continue to ask about 'concepts' much the same
questions about 'origins' as were for so ling and so &uitlessly
asked about sensa.
I should llke, then, to ask Mr. Mackinnon and Mr. Maclagan -
what they are asking when they ask about 'the formation' of
concepts, or about their 'acquisition': and also what is meant
by our 'having a concept'.

The case of 'resemblance', whch Mr. Maclagan introduces as

one where we are obviously not acquainted in sensation with
the object of a concept, is of course a specid and difficult case
in many ways. Yet it seems to me true to say that I do sense
resemblance, though what that means needs explanation.
I. Mr. Maclagan asks 'What is given in sensation?'-an
- -

unfortunate way of putting a very easy question, to which the

answer must and can only be 'Sensa'. What is unfortunate is
the word 'given' (cf. 'sensedata', 'the data of sense'). For this
suggests (a) that something is here 'given' us by somebody;
(b) that sensa are called 'given' in contrast with somethihg
w h c h is rather 'made' or 'taken', namely, my thoughts: but
it is dubious whether there is any sense in which my thoughts
are in my own control but m i sensa not so; (c) that some
proposition is 'given' to us in sensation as incorrigible, as pre-
misses are 'given' in sciences or 'data' to the detective: but sensa
are dumb, and n o t h g is more surely fatal than to confuse
sensing with thuJting.
2. Owing to h s 'concept' language, Mr. Maclagan seems to
try, when presented with any substantive, to fmd some isolable
part of the sensefield which is an instance of it but not pf
anything else. That is, I admit, vague. Given the sentence 'A
resembles B', it seems that Mr. Maclagan says to himself: That
means there should be three t h g s to sense, A and B and a
 ARE T H E R E A P R I O R 1 C O N C E P T S ? 49
resemblance: now we do sense the col~urs,but we do not sense
the resemblance. But surely, speaking carefully, we do not
sense 'red' and 'blue' any more than 'resemblance' (or 'quali-
ties' any more than 'relations') : we sense something.of G h c h
we might say, if we wished to talk about it, that 'thls is red':
and so we sense somehng of whch we might say, if we
wished to talk about it, that ' h s is s i d a r to that' or 'that h s
red is similar to that red'. If we insist on trying to say what
we sense (whch is impossible), we might try saying that 'I sense
A-resembling-By : and see if that helps.
3 . There are, I w i h g l y admit, special daculties about re-
semblance: but it seems- to me that they are not before us-
for Mr. Maclagan would, as is shown by h s commendation
of Locke and others,' say about other 'relations', and apparently
about all of them, precisely what he says about resemblance.
'Relations are not sensed'. This dogma, held by a very great
number and variety of philosopher^,^ seems to me so odd that;
lJte Mr. Maclagan, I find it dif6cult to discover arguments. If
I say 'this dot is to the right of that dot', is it not quaint to say
that 1 am sensing the two dots but not sensing the to the right
of? It is true that I cannot say I do sense the to the right of:
that is not good English-but then nor is it good English to say
that I do not sense it, or that I intuite it. I sense what in English
is described by peans of two demonstrative pronouns and an
adverbial phrase. To look for an isolable entity corresponding
to the latter is a bad habit encouraged by talk about 'concepts'.
What, I wonder, does Mr. Maclagan say about verbs? If 'he
Does Hume really not allow that any 'relations' are sensed: for instance,
'contiguity'?
Even by Berkeley, who jeopardized his whole theory by doing so. A
pretty anthology might be compiled of the phrases found by philosophers to
express their distrust and contempt for relations: 'entia semi-mentalia' and
what not. 1 suppose it goes back to Aristotle, who assumes, with the plain
man, that 'what is real is t h g s ' , and then adds, grudgingly, 'also their
qualities', these being somehow inseparable from the t h g s : but he draws the
line at relations, which are really too flimsy. I doubt if there is much more
behind the prejudice against relations than this: there was not in Leibniz's case,
and few have hammered relations so hard as he.
40 ARE THERE A PRIORI CONCEPTS?
is batting the cat', do I sense him and the cat, and intuite rela-
tions of batting and of being batted? But perhaps t h s is unfair:
for verbs are only used in this way in unanalysed material-
object language.
4. There are those who hold that 'relations' do not have
'instances' at all: but Mr. Maclagan, hke Mr. Joseph, takes the
less extreme view that they do have instances, but not 'sensible'
ones. For this he has, as-he canchdly admits, no arguments
whatever to adduce. But the difficulty is not so much to argue
for the view as to understand it at all. 'The colours' I sense, I
do not intuite them: I intuite 'the resemblance'-or 'their
resen~blance'? Or 'the resemblance between the two colours' ?
Or what exactly? This 'intuition' is a form of acquaintance:
and if we can be acquainted with relations thus, it seems difficult
to see why we should not be acquainted with them in sensation.
And there seems the absurdity that we have now separated off
'the resemblance' from 'the colours': we have a sensing and
simultaneously an intuiting, as we might feel a stab ofjealousy
while tasting porridge. Even if one is never found without the
other, what has the one to do with the other?
5 . Mr. Maclagan, recognizing that there are arguments
which tell against him, advances a 'purely defensive' argument, I
based on considerations about 'If A then B, if B then C'. Now
in the first place there seems to be some muddle about the
symbolism here. In any ordinary sense of 'if. . . then', I do not
think that the premiss given above will help us at all to answer
the question which is asked, viz. 'What is required if we are to
have C?' For the -premiss tells us about certain sufiient condi-
tions for C, whereas the question is about necessary conditions
for C. W h c h does Mr. Maclagan mean to talk about?
Apparently A is to be 'the sensing of the colours', B 'the
- of the resemblance', and C 'the knowing that the
intuiting
colours resemble'. Now if we interpret 'if. . . then' to mean
that A is a sufficientcondition of B, and B of C, it seems im-
possible to suppose that the 'intuiting' (alone) is a sujicicnt
condition of the knowing. Surely the sensing is also necessary?
 ARE THERE A P R I O R 1 C O N C E P T S ? S1
(As a matter of fact, it seems to me quite untrue that even the
sensing and the intuiting would necessitate the knowing: in
order to know, I must also think about the objects of acquain-
tance. I However, Mr. Maclagan is not called upon to point t h s
out, since the plain man, against whom he is arguing, over-
looks it.) On the other llaild, in the case of the sensing and
'
intuiting, Mr. ~ a c l a g a nmust mean that the former does
necessitate, i.e. is a suficient condition of, the latter: for he is
arguing that the plain man is able to say that 'sensing is the sole
necessary condition of knowing', because, though as a matter
of fact the intuiting is also a necessary condition, the sensing
does of itself necessitate the intuiting.-
It seems to me, therefore, that the symbolism here is merely
confusing: 'if . . . then' cannot mean the same in the case of
A and B as in the case of B and C. Moreover, I am very much
in the dark as to what really are the relations between the
sensing and the intuiting, the intuiting and the knowing.
'Condition' and 'necessitation' are words which niay refer to
entailment, or again to natural causation: is it in either of these
senses that the sensing necessitates the intuiting, or that the
intuiting is necessary to the knowing? I shouId doubt whether
in either case 'necessary' is being used in either of thcse familiar
senses; and further, whether it is being used in the salnc sense
in each case.
Clearly, what we arc most concerned with is the sense in
which the sensing necessitates the intuiting. Mr. Maclagan distin-
guishes the case o f 'resemblance' from that of 'straightness', in
that in the former the intuiting is 'logically inseparable' from
the sensing, whereas in the latter the intuiting is 'psychologically
conditioned' by the sensing. These look, from thc words used,
rather like 'entailmcnt' and 'natural causation' respectively: but
whatever may be true in the casc of 'straightnessy, I do not
Mr. Maclagan permits himself to wonder whether the intuiting ( ?sensing
+ intuiting) can really be distinguished from the knowing. But, if it cannot,
not merely does Mr. Maclagan abandon one of his original premisses, as he
recognizes, but also, in our present problem, 'B'cannot be distinguished from
'C'; which is frustrating.
52 A R E THERE A P R I O R 1 CONCEPTS?
see, in the case of resemblance, how one event can entail another,
so that 'logically inseparable' entirely eludes me. If Mr. Mac-
lagan says it is obscure, we may remember how obscure the
'intuiting' itself is, and be content to note one more example
of the h g h differential fertility of obscurities.
Further, Mr. Maclagan holds that other relations are in the
same case as 'resemblance': if so, then other relations between
our two colours will also be intuited, and these intuitings will
also bc necessitated by the sensing of the 'colours'. What I
wonder about, is: do I, as soon as I 'sense the colours', eo ipso
intuite all these relations (and know them)? For example,
suppose I do know that these two colours are similar: then I
must have 'sensed the colours', for I have had the intuition of
resemblance. Now if I havc 'sensed the colours', I have pre-
sumably already had 'intuitions' about which is the lighter, &c.,
-

those being inseparable from the sensing. Yet it seems clear to

me I may very well know the two colours are similar and
yet, if asked which is lighter, have to look back at them again.
Why so, on Mr. Maclagan's theory?
Mr. Maclagan notices: but does not attempt to defend him-
self against, other arguments drawn from the plain man, viz.
that he says he 'sees' or 'hears' resemblances, and says things are
-

'sensibly' alike or different. These facts seem to me important,

-

and I think the plain man would quite rightly persevere in these
assertions. I have heard it said that it is odd to talk of 'smelling
a resemblance': and certainly it is well to consider other senses
than that of sight. But it is not odd to talk of smelling two
similar smells, or two smells whlch are sensibly alike (though,
the plain man might well ask, how could they be alike except
sensibly?). And if I were forced to say either 'I smell the
resemblance' or 'I intuite it7, I know w1;ich I should choose.
The plain dog would, I am sure, say it smelled resemblances:
but no doubt your phlosophca dog would persuade itself
that it 'inhaled' them.
Similar considerations hold also in the case of other 'relations' :
plainbmen say 'ths tastes sweeterlthan that' or ' h s sounds louder
 A R E THERE A P R I O R 1 CONCEPTS? 53
than that'.' In the case of such 'relations' as these, indeed, I can
scarcely conjecture what it is that Mr. Maclagan would try to
persuade me I really mean. Later on, Mr. Maclagan comes to
discuss 'louder'z and says it is a 'comparative word': I wonder
whether 'louder than' is a 'relation' w h c h has to be intuited?
Surely, in such cases as these, it is evident that sentences con-
taining 'relation9words describe what we sense in precisely thc
same way as sentences containing 'quality' words? For it is
difficult to decide which sort of word 'loud' is.
6. I agree with Mr. Maclagan that, on h s view of 'resem-
blance', a non-sensuous acquaintance with the particular in-
stance would occur, to which nothing similarwould occur
in the case of 'redness'. But he seems to hold that, even if I am
to say 'this is red', some non-sensuous acquaintance (or 'aware-
ness') must occur: apparently because, in order to name this
colour red, I must institute a certain 'comparison'. I am not sure
what t h s means. I agree that, in many cases, where I say, for
example, ' h s is puce', I have compared the present sensum
with some 'pattem', perhaps a memory-image of the insect,
but in any case an entity of the same lund as the sensum itself.
Now this seems to me to require no non-sensuous acquaintance
with anything: but I realize that Mr. Maclagan may think it
does, because, when I compare the present image with the
pattem, I must notice that they resemble. Is it, then, that he
t h & an intuition oj'resemblance is needed even if I am to say
'this is red'? At least this rather qualifies h s original statement
that 'I sense the colours'. However, I am not sure he does mean
thls: he does not state with what I 'compare' the sensum, and
perhaps he is Amking that I compare the sensum with some-
thing whch is the object of a non-sensuous awareness, e.g.,
the universal 'redness' (though it would be surprising to find

Very plain men wlll say 'I hear this louder than that'.
Professor W. G. Maclaga~i,whose paper is here referred to, has suggested
in a private letter that 'loud' nus st be meant here, not 'louder', though it is
the latter that appears in the printed text of 1939. It seems clear that 'loud'
is what the sense requires.-Eds.
54 A R E T H E R E A P R I O R I CONCEPTS?
ourselves 'acquainted' with universals, Mr. Maclagan is pre-
pared to allow the possibility for the sake of argument). I
should like, then, to know what non-sensuous acquaintance
Mr. Maclagan has here in mind.
 THE M E A N I N G OF A WORD

SPECIMENS OF SENSE

I. I. What-is-the-meaning-of (the word) 'rat' ?

I. I I. What-is-the-meaning-of (the word) 'word' ?
1.21. Whatisa'rat'?
I. 211. What is a 'word'?
I . 22. What is the 'muzzle' of a rat?
2. I. What-is-the-meaning-of (the ~hrase)'what-is- the-
meaning-of ?
2. I r . what-is-the-meaning-of (the sentence) ha t-is- the-
meaning-of (the word) "x" ?'?

SPECIMENS OF NONSENSE

I. I. what-is-the-meaning-o f a word ?
I. I I. what-is-the-meaning-of any word ?
I. 12. What-is-the-meaning-of a word in general?
I. 2 I . What is the-meaning-of-a-word ?
I. 2 I I . What is the-meaning-of- (the-word)-'ra t' ?
I . 22. What is the 'meaning' of a word?
I . 221.What is the 'meaning' of (the word) 'rat' ?
2. I . Wha t-is-the-meaning-o f (the phrase) 'the-meanin g-
of-a word' ?
2 . 11. what-is-the-meaning-of (the sentence) 'What is the-
meaning-of- (the-word) -"x" ?' ?
2. 12. what-is-the-meaningof (the sentence) 'What is the
6 1
meaning" of "the word" "x" ?' ?
56 T H E M E A N I N G OF A WORD
THISpaper is aboutAthe phrase 'the meaning of a word'. It is
divided into three parts, of w h c h the first is the most trite and
the second the most muddled: all are too long. In the first, I
try to make it clear that the phrase 'the meaning of a word' is,
in general, if not always, a dangerous nonsense-phrase. In the
other two parts I consider in turn two questions, often asked
in phlosophy, w h c h clearly need new and careful scrutiny if
that facile phrase 'the meaning of a word' is no longer to be
permitted to impose upon us.

I begin, then, with some remarks about 'the meaning of a

word'. I think many persons n o w see all or part of what I shall
say: but not all do, and there is a tendency to forget it, or to
&t it slightly wrong. In so far as I am merely flogging the con-
-

verted, I apologize to them.

A preliminary remark. It may justly be urged that, properly
speaking, what alone has meaning is a sentence. Of course, we
-

can speak quite properly of, for example, ' l o o h g up the

meaning of a word' in a dictionary. Nevertheless, it appears
that the sense in w h c h a word or a phrase 'has a meaning' is
derivative from the sense in which a sentence 'has a meaning', -
to say a word or a phrase 'has a meaning' is to say that there are
sentences in w h c h it occurs w h c h 'have meanings': and to
know the meaning which the word or phrase has, is to know
the meanings of sentences in which it occurs. All the dictionary
can do when we 'look up the meaning of a word' is to suggest
aids to the understanmng of sentences in which it occurs.
-

Hence it appears correct to say that what 'has meaning' in the

primary sense is the sentence. And older phlosophers who
discussed the problem of 'the meaning of words' tend to fall
into special errors, avoided by more recent philosophers, who
discuss rather the parallel problem of 'the meaning ofsentences'.
Nevertheless, if we arc on our guard, w e perhaps need not fall
into these special errors, and I propose to overlook them at
present. I
 T H E M E A N I N G OF A WORD 57
There are many sorts of sentence in which the words 'the
meaning of the word so-and-so' are found, e.g. 'He does not
know, or understand, the meaning of the word handsaw': 'I
shall have to explain to her the meaning of the word pikestqf':
and so on. I intend to consider primarily the comnlon question,
'What is the meaning of so-and-so?' or 'What is the meaning
of the word so-and-so ?'
Suppose that in ordinary life I am asked: 'What is the mean-
ing o f t h e word racy?' There are two sorts of thing I may do
in response: I may reply in words, trying to describe what
raciness is and what it is not, to give examples of sentences in
w h c h one might use the word racy, and of others in w h c h
one should not. Let us call this sort of dung 'explaining the
syntactics' of the word 'racy' in the English language. O n
the other hand, I might do what w e may call 'demonstrating the
semantics' of the word, by getting the questioner to imagine,
or even actually to experience, situations which w e should
describe correctly by mdans of sentences containing the words
'racy' 'raciness', &c., and again other situations where w e
should not use these words. T h s is, of course, a simple case: but
perhaps the same two sorts of procedure would be gone through
in the case of at least most ordmary words. And in the same
way, if I wished to find out 'whether he understands the mean-
ing of the word racy', I should test h m at some length in these
two ways (which perhaps could not be entirely divorced from
each other).
Having asked in t h s way, and answered, 'What is the mean-
ing of (the word) "rat"?', 'What is the meaning of (the word)
"cat"?', 'What is the meaning of (the word) "mat"?', and so
on, we then try, being phlosophers, to ask the further general
question, 'What is the meaning of a word?' But there is some-
thing- spurious about this question. W e do not intend to mean
-

by it a certain question G h c h would be perfectly all right,

namely, 'What is the meaning of (the word) "word"?': that
would be no more general than is asking the meaning of the
word 'rat', and would be answered in a precisely similar way.
58 T H E M E A N I N G OF A WORD
No: we want to ask rather, 'What is the meaning- of a-word-
in-general?' or 'of any wordf-not meaning 'any' word you
like to choose, but rather no particular word at all, just 'any word'.
Now if we pause even for a moment to reflect, ths is a per-
fectly absurd question to be trying to ask. I can only answer
a question of the form 'What is the meaning of "x"?' if "x" is
some particular word you are asking about. T h s supposed
general question is really just a spurious question of a type
which commonly arises in philosophy. - -
W e may call it the
fallacy of aslung about '~ohng-in-particular' whch is a
practice decried by the plain man, but by the philosopher
called 'generalizing' and regarded with some complacency.
Many other examples -
of the fallacy can be found: take, for
example, the case of 'realityy-we try to pass from such
questions as 'HOW would you distinguish a real rat from an
imaginary rat?' to 'What is a real thing?', a question which
merely gives rise to nonsense.
. -

W e may expose the error in our present case thus. Instead

of asking 'What is the meaning of (the word) "rat"?' we
might clearly have asked 'What is a "rat"?' and so on. But if
our questions have been put in that form, it becomes very
difficult to formulate any question whch could impose
on us for a moment. Perhaps 'What is anything?'? Few philo-
sophers, if perhaps not none, have been foolhardy enough to
pose such a question. In the same way, we should not perhaps
be tempted to generalize such a question as 'Does he know the
meaning of (the word) "rat"?' 'Does he know the meaning
of a word?' would be slllv. J

Faced with the nonsense question 'What is the meaning of a

A V

word?', and perhaps dimly recognizing it to be nonsense, we

are nevertheless not inclined to give it up. ~nstead,we trans-
form it in a curious and noteworthy manner. Up to now, we
had been asking ' What-is-the-meaning-of (the word) "rat" ?',
&c. ;and ultimately ' What-is-the-mcaninq-o f a word?' But now,
being baffled, we change so to speak, t6e hyphenation, and ask
'What is the-meaning-+a-word?' o t sometimes, 'What is the
 THE MEANING O F A WORD 59

''meaningWof a word?' (I. 22): I shall refer, for brevity's sake,

only to the other (I. 21).It is easy to see how very different this
question is from the other. At once a crowd of traditional and
A

reassuring answers present themselves: 'a concept', 'an idea',

'an imagey, 'a class of similar sensa', &c. All of w h c h are
equally spurious answers to a pseudo-question. Plunging ahead,
however, or rather retracing our steps, we now proceed to ask
such questions as 'What is the-meaning-of-(the-word) "rat" ?'
which is as spurious as 'What-is-the-meaning-of (the word)
"rat"?' was genuine. And again we answer 'the idea of a rat'
and so forth. How quaint this procedure is, may be seen in the
following way. Supposing a plain man puzzled, were to ask
me 'What is the meaning of (the word) "muggy"?', and I
were to answer, 'The idea or concept of "mugginess" ' or 'The
class of sensa of w h c h it is correct to say his is muggy .
9' 7.

the man would stare at me i s at an imbecile. And that is sufi-

ciently unusual for me to conclude that that was not at all the
sort of answer he expected: nor, in plain English, can that
question ever require that sort of answer.
A A

TO show up this pseudo-question, let us take a parallel case,

where perhaps no one has yet been deluded, though they well
might be. suppose that I ask 'What is the point bf doing so-
and-so?' For example, I ask Old Father William 'What is the
point of standing on one's head?' He replies in the way we
know. Then I follow this up with 'What is the point of balanc-
ing an eel on the end of one's nose?' And he explains. Now
suppose I ask as my third question 'What is the point of doing
anything-not anything in particular, but just anything?' Old
Father William would no doubt luck me downstairs without
the option. But lesser men, raising this same question and find-
-

ing no answer, would very &ely commit suicide or join the

Church. (Luckily, in the case of 'What is the meaning of a
word?' the effeits are less serious, amounting- only to the
writing of books.) On the other hand, more adventurous
intellects would n o doubt take to asking 'What is the-point-of-
doing-a-thing?' or 'What is the "point" of doing a thing?':
60 THE M E A N I N G O F A WORD
and then later 'What is the-point-of-eating-suet?' and so on.
Thus we should discover a whole new universe of a kind of
entity called 'points', not previously suspected of existence.
T o make the matter clearer, let us consider another case
which is precisely unlike the case of 'What is the meaning of?'
I can ask not only the question, 'What is the square root of 4?',
o f 8, and so on, but also 'What is the square root of a number?':
w h c h is either nonsense or equivalent to 'What is the "square
root" of a number?' I then give a definition of the 'square root'
of a number, such that, for any given number x, 'the squarc
root of x' is a definite description of another number y. This
differs from our case in that 'the meaning of p' is not a definite
-

description of any entity.

The general questions w h c h we want to ask about 'mean-
ing' are best phrased as, 'What-is-the-meaning-of (the phrase)
' 9, 7
"what-is-the-meaning-of (the word) 'x ? ? The sort of an-
swer w e should get to these quite sensible questions is that
with w h c h I began thls discussion: viz. that when I am asked
'What-is-the-meaning-of (the word) "x"?', I naturally reply
by explaining its syntactics and demonstrating its semantics.
All this must seem very obvious, but I wish to point out that
it is fatally easy to forget it: no doubt I shall do so myself many
times in the course of t h s paper. Even those who see pretty
- -

clearly that 'concepts', 'abstract ideas', and so on are fictitious

entities, which we owe in part to aslung questions about 'the
meaning of a word7,nevertheless themselves thmk that there
is something which is 'the meaning of a word'. Thus Mr. Hamp-
shire1 attacks to some purpose the theory that there is such a
thing as 'the meaning of a word': what he thmks is wrong is
the belief that there is a single thing called the meaning: 'con-
cepts' are nonsense, and no single
- -
particular
-
'image' can be the -

meaning of a general word. So, he goes on to say, the meaning

of a word must really be 'a class of similar particular ideas'. 'If
we are asked "What does this mean?" we point to (!) a class of
I 'Ideas, Propositions and Signs', in the Proceedings ofthe Aristotelian Society,
1939-40. I
 THE MEANING O F A WORD 6I

particular ideas.' But a 'class of particular ideas7is every bit as

fictitious an entity as a 'concept7-or lbstract idea'. In the same
way Mr. C. W. Morris (in the Encyclopaedia of Unified Science)
attacks, to some purpose, those who think of 'a meaning' as
a definite s o m e h g w h c h is 'simply located' somewhere:
what he thinks is wrong is that t h n k of 'a meaning' as
a kind of entity which can be described wholly without refer-
ence to the total activity of 'semiosis'. Well and good. Yet he
himself makes some of the crudest possible remarks about 'the
designatum9 of a word: every sign has a designaturn, which is
not a particular t h n g but a kittd of object or class of object.
N o w t h s is quite as fictitious an entity as any 'Platonic idea':
and is due to precisely the same fallacy of looking for 'the
meaning (or designatum) of a word'.
-

W h y are we tempted to slip back in t h s way? Perhaps there

are two main reasons. First, there is the curious belief that all
words are names, i.e. in effect -proper
-
names, and therefore stand
for something or designate it in the way that a proper name
does. But t h s view that general names- 'have denotation' in
the same way that proper names do, is quite as odd as the view
that proper names 'have connotation' in the same way that
general names do, which is commonly recognized to lead to
- -

error. Secondly, we are afflicted by a more common malady,

which is thls. When we have given an analysis of a certain
sentence, containing a word or phrase ' x ' , we often feel in-
clined to ask, of our analysis, 'What in it, is "x"?' For exanlple,
we give an analysis of 'The State owns this land', in sentences
about individual men, their relations and transactions: and then
at last we feel inclined to ask: well now, what, in all that, is the
State? And we might answer: the State is a collection of indi-
vidual men united in a certain manner. O r again, when we
-

have analysed the statement 'trees can exist unperceived' into

statements about sensing sensa, we still tend to feel uneasy
unless we can say something 'really does' 'exist unperceived':
hence theories about 'sensibilia' and what not. So in our present
case, having given all that is required, viz. an account of
62 THE M E A N I N G OF A WORD
'What-is-the-meaning-of" w h a t is-the-meaning-of (the word)
6 P Y99
x ? we still feel tempted, wrongly supposing our original sen-
'' ,,?
tence to contain a constituent 'the-meaning-of (the-word)- x ,
to ask 'Well now, as it turns out, what is the meaning of the
word "x", after all?' And we answer, 'a class of sirmlar parti-
cular ideas' and what not.
Of course, all my account of our motives in t h s matter may
be only a convenient didactic schema: I do not think it is-but
I recognize that one should not impute motives, least of all
rational motives. Anyhow, what I claim is clear, is that there is
no simple and handy appendage of a word called 'the meaning
of (the word) "x" '.

I now pass on to the first of the two points whch need now a
careful scrutiny if we are no longer
-
to be imposed upon by that
convenient phrase 'the meaning of a word'. wha; I s h d say
here is, I know, not as clear as it should be.
Constantly we ask the question, 'Is y the meaning, or part of
the meaning, or contained in the meaning, of x?--or is it not?'
A favourite way of putting the question is to ask, 'Is the judge-
ment "x is y" analytic or synthetic?' Clearly, we suppose,
y must be either a part of the meaning of x, or not any part of it.
And, if y is a part of the meaning of x, to say 'x is not y' will
be self-contradictory: whle if it is not a part of the meaning
of x, to say 'x is not-y' will present no difficulty-such a state of
affairs will be readily 'concLivable'. T h s seems to be the merest
common sense. And no doubt it t~otridbe the merest common
sense if 'meanings' were things in sonie ordinary sense which
contained parts in some ordinary sense. But they are not.
~ n f o r t u n a c e l ~many
, who know they are not,
still speak as though y must either be or not be 'part of the
meaning' of x. But this is the point: if'ex~lainingthe meaning
of a word' is r e d v the complicated sort of affair that we have
4 A

seen it to be, and ifthere is really nothing to call 'the meaning

of a word'-then phrases llke 'part of the meaning of the word
 THE M E A N I N G O F A WORD 63
x' are coinpletely undefined; it is left hanging in the air, we do
not know what it means at all. W e are using a working-model
which foils to fit the facts that we really wish to talk about. When
we consider what we really do want to talk about, and not the
working-model, what would really be meant at all by a judge-
ment being 'analytic or synthetic'? W e simply - -
do not know.
O f course, we feel inclined to say 'I can easily produce examples
of analytic and synthetic judgements; for instance, I should
confidently say "Being a professor is not part of the meaning
of being a man" and so forth.' 'A is A is analytic.' Yes, but it is
when Ge are required to give a general definition of what we
mean by 'analytic' or 'synthetic', and when we are required
to justify our dogma that every judgement is either analytic or
synthetic, that we find we have, in fact, n o h g to fall back
-

upon except our working-model. From the start, it is clear that our
working-model fails to do justice, for example, to the distinc-
tion between syntactics and semantics: for instance, tallung
about the contradictory of every sentence having to be eithei
self-contradictory or not so, is to talk as though all sentences
w h c h we are prohbited from saying were sentences w h c h
offended against syntactical rules, and could be formally re-
duced to Gerbal self-contradictions. But t h s overlooks all
semantical considerations, w h c h philosophers are sadly prone
to do. Let us consider two cases of some dungs w h c h we
simply canrzot say: although they are not 'self-contradictory' and
although-and this of course is where many will have axes to
grind-we cannot possibly be tempted to say that we have
'synthetic a priori' knowledge of their contradictions.
Let us begin
-
with a case whlch, being about sentences rather
than words, is not quite in point, but which may encourage us.
Take the well-known sentence 'The cat is on the mat, and I do
not believe it'. That seems absurd. On the other hand 'The cat
is on the mat, and I believe it' seems trivial. If we were to adopt
a customary dichotomy, and to say either a proposition p
implies another proposition r, or p is perfectly compatible with
not-r, we should at once in our present case be tempted to say
64 T H E M E A N I N G OF A WORD
that 'The cat is on the mat' implies 'I believe it': hence both the
triviality of adding 'and I believe it' and the absurdity of
adding 'and I do not believe itY.But of course 'the cat is on the
mat' does not imply 'Austin believes the cat is on the mat': nor
even 'the speaker believes the cat is on the mat'-for the speaker
may be lying. The doctrine which is produced in this case is,
that not p indeed, but asserting p implies ' I (who assert p) believe
p'. And here 'implies' must be given a special sense: for of
course it is not that 'I assert p' implies (in the ordinary sense)
'I believe p', for I may be lying. - -
It is the sort of sense in
which by asking a question I 'imply' that I do not know the
answer to it. By asserting p I give it to be understood that I
believe p.
Now the reason why I cannot say 'The cat is on the mat and
I do not believe it' is not that it offends against syntactics in the
sense of being in some way 'self-contradictoryy. What prevents
my saylng it, is rather some semantic convention (implicit, of
course), about the way we use words in situations. What pre- p

cisely is the account to be givcn in thls case we need not ask.

Let us rather notice one significant feature of it. Whereas
'p and I believe it' is somehow trivial, and 'p and I do not
believe it' is somehow nonsense, a t h r d sentence 'p- and I
m k h t not have believed it' makes perfectly good sense. Let US
call these three sentences Q, not Q, and 'might not Q'. Now
what prohibits us from saying 'p' implies 'I believe y' in the
ordinary sense of 'implies', is precisely shown by ths fact: that
although not-Q is (somehotv) absurd, 'might not Q' is not at
-

all absurd. For in ordinary cases of implications, not merely is

not Q absurd, but 'might not Q yis also absurd: e.g. 'triangles are
figures and triangles
- - have no shape'- is no more absurd than
'triangles are figures and triangles might have had no shape'.
Consideration of the sentence 'might not Q' will afford a
rough test as to whether p 'implies' r in the ordinary sense, or
in the special sense, of 'implies'.
Bearing this in mind, let us now consider a sentence whch,
as I claim, cannot possibly be classified as either 'analytic' or
 THE MEANING O F A W O R D 65

'synthetic'. I refer to the sentence, 'Ths x exists', where x is

a sensum, e.g. 'Ths noise exists'. In endeavouring to classify
it, one party would point to the triviality of 'This noise exists',
and to the absurdity of 'Ths noise does not exist'. They would
say, therefore, that existence is 'part of the meaning of' this.
-

~ uanother
t party would point ;ut, that 'Ths noise might not
have existed' makes perfectly good sense. They would say,
therefore, that existence cannot be 'part of the meaning of' this.
Both -parties, as we are now in a position to see, would be
-

correct in their arguments, but incorrect in their conclusions.

What seems to be true is that using the word 'this' (not: the word
4
this') gives it to C understood that the sensum referred to 'exists'.
Perhaps, hstorically, t h s fact about the sentence-trio, 'Thls
noise exists', ' T h s noise does not exist', and 'Ths noise might
not have existed', was pointed out before any phlosopher had
had time to pronounce that ' T h s noise exists' is analytic, or is
synthetic. But such a pronouncement might well have been
made: and to this day, even when the fact has been pointed out,
many phdosophers worry about the case, supposing the sentence
must be one or the other but painfully aware of the difficulties
in choosing either. I wish to point out that consideration of the
analogy bitween t h s case &d the other, should cure us once
and for all of t h s bogy, and of insisting on classifying sentences
as either analytic or synthetic. It may encourage us to consider
again what the facts in their actual complexity really are. (One
thing it suggests is a reconsideration of 'Caesar is bald' and
- -

similar propositions: but I cannot go into that.)

So far, however, we have scarcely begun in earnest: we have
merely felt that initial trepidation, experienced when the firm
ground of prejudice begins to slip away beneath the feet.
Perhaps there are other cases, or other sorts of cases, where it
wdl not be possible to say either that y is a 'part of the meaning'
of x or that it is not, without being misleading.
Suppose we take the case of 'being thought good by me' and
'being approved of by me'. Are we to rush at t h s with the
dichotomy: either 'being approved of by me' i s part of the
66 THE M E A N I N G OF A W O R D
meaning of 'being thought good by me' or it is not? Is it obvious
that 'I thuJc x good but I do not approve of it' is self-contra-
dictory? Of course it is not verbally self-contra&ctory. That it
either is or is not 'really' self-contradictory would seem to be
difficult to establish. Of course, we thnk, it must be one or the
other-only 'it's difficult to decide which': or 'it depends on
how you use the words'. But are those really the difficulties
whlch baffle us ? Of course, ifit were certain that every sentence
nlrist be either analytic or synthetic, those must be the dif5-
culties. But then, it is not certain: no account even of what the
distinction means, is given except by reference to our shabby
working-model. I suggest that 'I thnk x good but I do not
approve of it' may very well be neither self-contradictory nor
yet 'perfectly p o d sense' in the way in whlch 'I think x
exciting but I do not approve of it' is 'perfectly good sense'.
Perhaps t h s example does not stnke you as awkward. It
cannot be expected that all examples will appeal equally to all
hearers. Let us take some others. Is 'What is good -
ought to
exist' analytic or synthetic? According to Moore's theory, t h s
must be 'synthetic': yet he constantly in Principia Ethica takes
its truth for granted. And that dustrates one of the main draw-
backs of insisting on saying - - that a sentence must be either
analytic or synthetic: you are almost certain to have left on
your hands some general sentences whch are certainly not
analytic but whch you find it difficult to conceive being false:
i.e. you are landed with 'synthetic a priori knowledge'. Take
that sentence of ill fame 'Pmk is more k e red than black'. It is
rash to pronounce t h s 'synthetic a priori knowledge' on the
ground that 'being more llke red than black' is not 'part of the
meaning' or 'part of the definition' of 'pmk' and that it is not
'conceivable' that pmk should be more like black than red:
A

I dare say, so far as those -phrases have any clear meaning, that
it is not:-but the question is: is the thmg therefore 'synthetic'
-

a priori knowledge?
Or, again, take some examples from Berkeley: is extended
'part of the meaning' of coloured or of shaped, or shaped 'part of
 THE M E A N I N G OF A WORD 67
the meaning' of extended? Is 'est sed non percipitur' self-
contradictory (when said of a sensum), or is it not? When we
worry thus, is it not worth considering the possibility that w e
are oversinlplifying ?
What we are to say in these cases, what even the possi-
bilities are, I d o not at present clearly see. (I) Evidently, we must
throw away the old worlung-model as soon as we take account
even of the existence of a distinction between syntactics and
semantics. (2) But evidently also, our new worlung-model, the
supposed 'ideal' language, is in many ways a most inadequate
model o f any actual language: its careful separation of syn-
tactics from semantics, its lists of explicitly formulated riles
and conventions, and its careful delimitation of their spheres of
operation-all are misleading. An actual language has few, if
any, explicit conventions, no sharp limits to the spheres of
- - -

operation of rules, n o rigid separation of what is syntactical

and what semantical. (3) Finally, I think I can see that there are
difficulties about our powers of imagination, and about the
curious way in w h c h it is enslaved by words.
T o encourage ourselves in the belief that t h s sort o f con-
sideration may play havoc with the distinction 'analytic or
synthetic', let us consider a similar and more familiar case. It
seems, does it not, perfectly obvious that every proposition
musthave a contradictory? Yet it does not turn out so. Suppose
that I live in harmony and friendshp for four years with a cat:
and then it delivers a philippic. W e ask ourselves, perhaps, 'Is
it a real cat? o r is it not a real cat?' 'Either it is, or it is not, but
we cannot be sure whch.' N o w actually, that is not so: neither
'It is a real cat' nor 'it is not a real cat' fits the facts semantically:
each is designed for other situations than t h s one: vou could
not say the former of sometlung which delivers phlippics, nor
yet the latter o f somethng w h c h has behaved as t h s has for
four years. There are similar difftculties about choosing be-
tween 'This is a hallucination' and 'This is not a hallucination'.
With sound instinct, the plain man turns in such cases to
Watson and says 'Well now, what would you say?' 'How would
 68 THE M E A N I N G OF A WORD
you describe it? The difficulty is just that: there is no short
description which is not misleading: the only t h R -
to do, and
that can easily be done, is to set out the desc;iption of the facts
at length. Ordinary language breaks down in extraordinary
cases. (In such cases, the cause of the breakdown is semantical.)
Now no doubt an ideal language would not break down, what-
-

ever happened. In doing - -

for example, where our
language is tightened up in order precisely to describe compli-
cated and unusual cases concisely, we prepare ling~isticafl~ for
the worst. In ordinary language we do not: words fail us. If we
talk as though an ordinary must be U e an ideal language, we
shall misrepresent the facts.
Consider now 'being extended' and 'being shaped'. In ordi-
nary life we never get into a situation where we learn to say that
anything is extended but not shaped nor conversely. We have
all learned to use, and have used, the words only in cases where
it is correct to use. both. Supposing now someone says ' x is
extended but has no shape'. Somehow we cannot see what t h s
'could mean'-there are no semantic conventions, explicit or
implicit, to cover this case: yet it is not prohibited in any way-
there are no limiting rules about what we might or might not
say in extraordinary cases. It is not merely the difficulty of
imagining or experiencing extraordinary cases, either, w h c h
causes worry. There is this too: we can only describe what it
is we are trying to imagine, by means of words whch pre-
cisely describe and evoke the ordinary case, whch we are trying
to think away. Ordinary language blinkers the already fieblk
imagination. It would be difficult, in this way, if I were to say
'Can I think of a case where a man would be neither at home
nor not at home?' This is inhibiting, because I t h d of the
ordinary- case where I ask 'Is he at home?' and get the answer,
-

'No': when certainly he is not at home. But supposing I happen

first to thnk of the situation when I call on him just after he has
died: then I see at once it would be wrong to say either. So in
our case, the only thng to do is to imagine or experience aIl
kinds of odd situations, and then suddenly round on oneself
 THE M E A N I N G OF A WORD 69
and ask: there, n o w would I say that, being extended it must
be shaped? A new idiom might in odd cases be demanded.
-

I shbuld llke to say, in concluding this section, that in the

course of stressing that we must pay attention to the facts of
actual language, what we can and cannot say, and precisely why,
another and converse point takes shape. Although - it will not
do to force actual language to accordwith some preconceived
model: it equally will not do, having discovered the facts about
6
ordinary usage' to rest content with that, as though there were
nothing- more to be discussed and discovered. There may be plen-
ty that might happen and does happen which would neednew
and better language to describe it in. Very often philosophers are
only engaged on this task, when they seem to be perversely using
words in a way which makes no sense according to 'ordinary
usage'. There may be extraordinary facts, even abbut our every-
-

day experience, which plain if-xn and plain language overlook.

The last, and perhaps least unimportant point I have to

make is the following: it seems to me that far more detailed
attention ought to be given to that cclcbrated qiestion, the
posing- of which has given birth to, and still keeps alive, so
-

many erroneous theories, namely: why do wc call different

thngs by the same name? In reply to this, the philoprogenitive
invent theories of 'universals' and what not: sorne entity or
other to be that of whch the 'name' is the name. And in reply
to them, the more cautious (the 'nominalists') have usually been
content to reply simply that: the reason why we call different
thngs by the same name is simply that the thlngs are similar:
there is nothlng- identical -present in them. This reply is in-
adequate in many respects: it does not, for example, attack the
misleading form in which the question is posed, Gr sufficiently
go into the peculiarities of the word 'similar'. But what I wish to
object to in it tonight is rather this: that it is rlot in the least trrie
that all the things which I 'call by the same (general)name' are in
general 'similar', in any ordinary sense of that much abused word.
70 T H E MEANING OF A WORD
It is a most strange h g that 'nominalrsts' should rest con-
tent with t h s answer. Not merely is it untrue to the facts; but
further, if they had examined the facts, whch are, in them-
selves, interesting enough, they could have produced with
little trouble a far more formidable case against their opponents.
So long as they say the things are similar, it will always be open
to someone to say: 'Ah yes, similar in a certain respect: and that
can only be explained by means of universals' (or whatever the
name may be that they prefer for that well-tried nostrum) : or
again to maintain that similarity is only 'intelLgible' as partial
identity: and so on. And even those who are not persuaded
entirely, may yet go so far as to allow that the 'similarity' and
'identity' languages are alternatives, the choice between which
is indifferent. But surely, if it were made evident that we often
'call different things by& same name', and for perfectly 'good
reasons',' when the t b g s are not even in any ordinary sense
'similar', it will become excessively difficult to maintain that
there is something 'identical' present in each-and after all, it
is in rPJuting that position that the nominalist is really interested.
Not, of course, that we can really rrfute it, or hope to cure those
incurables who have long since reached the tertiary stage of
universals.
Leaving historical d~sputesaside, it is a matter of urgency
that a doctrine should be developed about the various l&ds of
good reasons for which we 'call different thingsz by the same
name'. T h s is an absorbing - question,
- but habitually neglected,
-
so far as I know, by philologists as well as by philosophers.
Lying in the no man's land bekeen them, it falls between two
schools, to develop such a doctrine fully would be very com-
plicated and -perhaps tedious: but also very useful in many
bays. It denlands the study of actual languages,
-
not ideal ones.
That the Polish semanticists have discussed such questions I 1

neither know nor believe. Aristotk did to a quite considerable

1

extent, but scrappily and inexactly.

We are not interested in mere equivocation, of course.
Strictly, sorts of things rather than particular things.
 THE M E A N I N G O F A WORD 7I
I shall proceed forthwith simply to give some of the more
obvious cases where the reasons for 'calhg different sorts of
t b g s by the same name' are not to be dismissed lightly as -

6
similarity'. And show how consideration of these facts may
warn us against errors whlch are constant in philosophy.
r. A very simple case indeed is one often mentioned by
Aristotle: the adjective 'healthy': when I talk of a healthy body
and again of a healthy complexion, of healthy exercise: th-e
word is not just being used equivocally. Aristotle would say it is
being used 'paronyrn~usly'.~ In this case there is what we may
call a primary nuclear sense of 'healthy': the sense in which
'healthy' is used of a healthy body: I call thls nuclear because it
is 'contained as a part' in the other two senses whlch b a y be
set out as 'productive of healthy bodies' and 'resulting from a
healthy body'.
Ths is a simple case, easily understood. Yet constantly it is
-

forgotten when we start disputing as to whether a certain word

-

has-'two senses' or has not &o senses. I remember myself dis-

puting as to whether 'exist' has two senses (as used of material
objects and again of sensa), or only one: actually we were
agreed that 'exist' is used paronymously, only he called that
'having two senses', and I did not. Prichard's paper2 on hyaOdv
-

(in Aristotle) contains a classic instance of misunderstanding

about paronymity, and so worrying about whether a word
really 'has always the same meaning' or 'has several different
-

Now are we to be content to say that the exercise, the com-

plexion, and the body are all called 'healthy' 'because they are
similar'? such a remark cannot fail to be misleading. Why
make it? And why not direct attention to the important and
actual facts?
2. The next case I shall take is what Aristotle calls 'analogous'
terms. When A : B :: X : Y then A and X are often called by
But there are other varieties of paronymity of course.
'The Meaning of ArABON in the Ethics of Aristotle', by H. A. Prichard.
Reprinted in his Moral Obligation, Oxford, 1949.
72 THE M E A N I N G OF A WORD
the same name, e.g. the foot of a mountain and the foot of a
list. Here there is a good reason for calling the things both
'feet' but are we to say they are 'similar'? Not in any ordinary
sense. We may say that thc relations in whch they stand to
B and Y respectively are similar relations. Well and good:
but A and X are not the relations in whch they stand : and any-
one simply told that, in calling A and X both 'feet' I was calling
attention to a 'similarity' in them, would probably be misled.
Anyhow, it is most necessary to remember that 'similarity'
covers such possibilities if it is to do so. (An especially severe
case of 'analogy' arises when a term is used, as ~ristotlesays
'in different categories': e.g. when I talk about 'change' as
qualitative change, change of position, place, &c., how far is
it true to say these 'changes' are 'similar'?)
3 . noth her case is where 1 call B by the same name as A,
because it resembles A, C by the same name because it re-
sembles B, D . . . and so on. But ultimately A and, say, D do
not resemble each other in any recognizable
-
sense at all. This
is a very common case: and the dangers are obvious, when we
search for something 'identical' in all of them!
4. Another case whlch is commonly found is this. Take a
word like 'fascist': this originally connotes, say, a great many
characteristics at once: say x, y, and z. Now we will use
C
fascist' subsequently of things which possess only one of these
strlking characteristics. So that things called 'fascist' in these
senses, whch we may call 'incomplete' senses, need not be
similar at all to each other. Ths often puzzles us most of all
when the original 'complete' sense has been forgotten: com-
pare the various meanings of 'cynicism': we should be puzzled
to find the 'similarity' there! Sometimes the 'incompleteness'
of the resemblance is coupled with a positive lack of resem-
L A

blance, so that we invent a phrase to mark it as a warning,

e.g. 'cupboard love'.
5. Another better-known case is that of a so-called deter-
minable and its determinates: colour and red, green, blue, &c.,
or rather 'absolutely specific' reds, greens, blues, &c. Because
 T H E M E A N I N G OF A WORD 73
t h s is better known, I shall not discuss it, though I am as a
matter of fact rather sceptical about the accounts usually given.
Instead, it should be pointed out how common t h s sort of
relationship is and that it should be suspected in cases where
we are prone to overlook it. A s t r h n g example -
is the case
of 'pleasure': pleasures we may say not merely resemble each
other in being pleasant, but also d@r precisely in the way in
whch they are pleasant.' No greater mistake could be made
than the hedonistic mistake (copied -
by non-hedonists) of
&&ng that pleasure is always a single similar feehg, some-
how isolable from the various activities whch 'give u
rise' to it.
6. Another case which often provides puzzles, is that of
words lke 'youth' and 'love': which sometimes mean the
object loved, or the thng which is youthful, sometimes the
passion 'Love' or the quality (?) 'youth'. These cases are of
course easy (rather like 'healthy'?). But suppose we take the
noun 'truth' : here is a case where the disagreements
-
between
different theorists have largely turned on whether they inter-
preted this as a name of a substance, of a quality, or of a relation.
7. ~astly,I want to take a specially interesting sort of case,
which is perhaps commoner and at the bottom of more
muddles than we are aware of. Take the sense in whch I talk
of a cricket bat and a cricket ball and a cricket umpire. The
reason that all are called by the same name is perhaps that each
has its part-its otvn special part-to play in the activity called
cricketing: it is no good to say that cricket simply means 'used
in cricket': for we cannot explain what we mean by 'cricket'
-

except by explaining the special parts played in cricketing by

the bat, ball, &c. Aristotle's suggestion was that the word
'good' might be used in such a way: in which case it is obvious
how far astray we should go - if we look for a 'definition' of the
word 'good' in any ordinary simple sense: or look for the way
in which 'good' things are 'similar' to each other, in any
ordinary sense. If we tried to find out by such methods what
If we say that they are all called 'pleasures' 'because they are similar', we
shall overlook this fact.
74 T H E M E A N I N G OF A WORD
'cricket' meant, we should very likely conclude that it too was
a simple unanalysable supersensible quality.
Another thing that becomes plain from such examples is
that the apparently common-sense distinction between 'What
is the meaning of the word x' and 'What particular things ore x
and to what degrees?' is not of universal application by any
-

means. The questions cannot be distinguishid in such cases.

O r a similar case would be some word l k e 'golfing': it is not
sense to ask 'What is the meaning of golfing?' 'What t h g s
are golfing?' Though it is sense to ask what component acti-
vities go to constitute golfing, what implements are used in
golfing ('golf' clubs, &I$ and-in what ways. A ristotle suggests
'happiness' is a word of this kind: in which case it is evident
how far astray we shall go ifwe treat it as though it were a word
like 'wbteness'.
These summarily treated examples are enough to show how
essential it is to have a thorough, knowledge of the different
reasons for whch we call different thngs by the same name,
before we can embark confidently on an inquiry. If we rush
up with a demand for a definition in the simple manner of
-

~ I a t or
o many other philosophers, if we use the rigid dichotomy
-

'same meaning, different meaning', or 'What x means', as

distinguished from 'the thngs which are x', we shall simply
make hashes of things. Perhaps some people are now discussing
such questions seriously. All that is to be found in traditional
Logics
- is the mention that there are, besides univocal and
6
equivocal words, also analogous words': which, without
further explanation, is used to lump together all cases where a
word has not always absolutely the same meaning, nor several
absolutely different meanings. All that 'sirmlarity' theorists
manage is to say that all t b i g s called by some one name are
similar to some one -pattern, or are all more similar to each
other than any of them is to anything else; which is obviously
untrue. Anyone who wishes to see the complexity of the
-

problem, has only got to look in a (good) dictionary under

such a word as 'head': the different meanings of the word
 T H E M E A N I N G OF A WORD 75
'head' will be related to each other in all sorts of different ways
at once.
To summarize the contentions of this paper then. Firstly, the
phrase 'the meaning of a word' is a spurious phrase. Secondly
and consequently, a re-examination is needed of phrases like
the two whch I discuss, 'being a part of the meaning of' and
'having the same meaning'. On these matters, dogmatists
require prodding: although history indeed suggests that it may
sometimes be better to let sleeping dogmatists lie.
 OTHER M I N D S 1

I FEEL that I agree with much, and especially with the more
important parts, of what Mr. Wisdom has written, both in his
present paper and in his beneficial series of articles on 'Other
Minds' and other matters. I feel ruefully sure, also, that one
must be at least one sort of fool to rush in over ground so well
trodden by the angels. At best I can hope only to make a contribu-
tion to one part of the problem, where it seems that a little
more industry still might-be of service. I could only wish it was
a more central part. In fact, however, I did find myself unable
to approach the centre whle still bogged down on the peri-
phery. And Mr. Wisdom hlmself may perhaps be sympathetic
towards a policy of splitting hairs to save starting them.
Mr. Wisdom, no doubt correctly, takes the 'Predicament'
to be brought on by such questions as 'HOW do we know that
another man is angry?' He also cites other forms of the
question-'Do we (ever) know?', 'Can we know ?','How can
we know?' the thoughts, feelings, sensations, mind, &c., of
another creature, and so forth. But it seems 1lkely that each of
these further questions is rather different from the first, whlch
alone has been enough to keep me preoccupied, and to which
1 shall stick.
Mr. Wisdom's method is to go on to ask: Is it like the way
in which we know that a kettle is boiling, or that there is a tea-
party next door, or the weight of thistledown? But it seemed
to me that perhaps, as he went on, he was not giving an alto-
gether accurate account (perhaps only because too cursory a
Reprinted from Proceedings UJ the Arimtelian Society, Supplementary
Volume xx (1946), by courtesy o f the editor.
 OTHER M I N D S 77

one) of what we should say if asked 'How do you know?'

these thmgs. For example, in the case of the tea-party, to say
we knew of it 'by analogy' would at best be a very sophlsti-
cated answer (and one to whch some sophsticates might prefer
the phrase 'by induction'), while in addition it seemsincorrect
because we don't, I think, claim to know by analogy, but only
to argue by analogy. Hence I was led on to consider what sort
of h g does actually happen when ordinary people are asked
'How do you know?'
Much depends, obviously, on the sort of item it is about
w h c h we are being asked 'How do you know ?' and there are
bound to be many h d s of case that I shall not cover at all, or
not in detail. The sort of statement w h c h seems simplest, and
at the same time not, on the face of it, u n k e 'He is angry', is
such a statement as 'That is a goldfinch' ('The kettle is bohng')
-a statement of particular, current, empirical fact. Ths is. the
sort of statement on making w h c h we are liable to be asked
'How do you know?' and &e sort that, at least sometimes, we
say we don't know, but only believe. It may serve for a stalking-
horse as well as another.
When we make an assertion such as 'There is a goldfinch
in the garden' or 'He is angry', there is a sense in which we
imply ;hat we are sure of it or know it ('But I took it you
knew', said reproachfully), though what we imply, in a simdar
sense and more strictly, is only that we believe it. On making
such an assertion, therefore, we are directly exposed to the
questions ( I ) 'Do you know there is?' 'Do you know he is?' and
(2) 'How do you know?' If in answer to the first question we
reply 'Yes', we may then be asked the second question, and
even the first question alone is commonly taken as an invitation
A I

to state not merely whether but also how we know. But on the
other hand, we may well reply 'No' in answer to the first
question: we may say 'No, but I t M c there is', 'No, bur I
believe he is'. F& the im~licationthat I know or am sure
I

is not strict: we are not all (terribly or sufficiently) strictly

brought up. If we do & I Sthen
, we are exposed to the question,
78 OTHER M I N D S
w h c h might also have been put to us without preliminaries,
'Why do you believe that?' (or 'What makes you think so?',
'What induces you to suppose so?', 8tc.).
There is a singular difference between the two forms of
challenge: 'How do you know?' and ' Why do you believe?'
W e seem never to ask 'Why do you know?' or 'Horv do you
believe?' And in ths, as well as in other respects
- to be noticed
later, not merely such other words as 'suppose' 'assume', &c.,
but also the expressions 'be sure' and 'be certain', follow the
example of 'believe', not that of 'know'.
Either question, 'HOW do you know?' or 'Why do you
believe?', may well be asked only out of respectful curiosity,
from a genuine desire to learn. But again, they may both be
asked as pinted questions, and, when they are so, a further
difference comes out. 'How do you know ?' suggests that per-
haps you don't know it at all, whereas 'Why do you believe?'
- ~

suggests that perhaps you oughtn't to believe it. There is no

suggestion1 that you ought not to know or that you don't
believe it. If the answer to 'How do you know?' or to 'Why
do you believe?' is considered unsatisfactory by the challenger,
he proceeds rather differently in the two cases. His next riposte
win be, on the one hand, s b m e h g such as 'Then youbon't
know any such thing', or 'But tha; doesn't prove it: in that
case you don't really know it at all', and on the other hand,
something such as 'That's very poor evidence to go on: you
oughtn't to believe it on the strength of that a l ~ n e ' . ~
The 'existence' of your alleged belief is not challenged, but
the 'existence' of your allegedknowledge is challenged. If we
k e to say that 'I believe', and llkewise 'I am sure' and 'I am
certain', are descriptions of subjective mental or cognitive states
But in special senses and cases, there is-for example, if someone has
announced some top secret information, we can ask, 'How do you know?',
nastily.
An interesting variant in the case of knowing would be 'You oughtn't
to say (you've no business to say) you know it at all'. But of course this is only
superficially similar to 'YOUoughtn't to believe it': you ought to say you
believe it, if you do believe it, however poor the evidence.
 OTHER M I N D S 79

or attitudes, o r what not, then 'I know' is not that, or at least

not merely that: it functions differently in talking.
'But of course', it will be said, ' "I know" is obviously more
than that, more than a description of my own state. If I know,
I can't be wrong. You can .always show I don't know by
showing I am wrong, or may be wrong, or that I didn't know
by showing that I might have been wrong. That's the way in
wluch knowing differs even from being as certain as can be.'
T h s must be cbnsidered in due course, but first we should
consider the types of answer that may be given in answer to
the question 'How do you know?'
Suppose I have said 'There's a bittern at the bottom of the
garden', and you ask 'How do you know ?' my reply may take
very different forms :
(a) I was brought up in the fens
(b) I heard it
(c) The keeper reported it
(d) By its booming
(e) From the booming noise
-

(f) Because it is booming.

W e may say, roughly, that the first three are answers to the
questions 'How do you come to know?', 'HOW are you in a
position to know?', or 'How do you know?' understood in
different ways: whle the other three are answers to 'How can
you tell?' understood in different ways. That is, I may take you
to have been askmg:
(I) How do I come to be in a position to know about
bit terns ?
(2) How do I come to be in a position to say there's a bittern
here and now?
(3) How do (can) I tell bitterns?
(4) How do (can) I tell the thing here and now as a bittern?
The implication is that in order to know h s is a bittern, I must
have:
 O T H E R MINDS
(I) been trained in an environment where I could become
famhar with bitterns
(2) had a certain opportunity in the current case
(3) learned to recognize or tell bitterns
-

(4) succeeded in recognizing or t e l h g t h s as a bittern.

(I) and (2) mean that my experiences must have been of

certain kmds, that I must have had certain opportunities: (3)
and (4) mean that I must have exerted a certain kmd and
amount of acumen.'
The questions raised in (I) and (3) concern our past ex-
periences, our opportunities and our activities in learning to
discriminate or discern, and, bound up with both, the correct-
-

ness or otherwise of the linguistic usages we have acquired.

Upon these earlier experiences depends how well we know ~

things, just as, in different but cognate cases of 'knowing', it is

upon earlier experience that it depends how thoroughly or how
intimately we know: we know a person by sight or intimately,
a town inside out, a proof backwards, a job in every detail, a
poem word for word, a Frenchman when we see one. 'He
doesn't know what love (real hunger) - .
is' means he hasn't
had enough experience to be able to recognize it and to
distinguish it from other t h g s slightly hke it. According to
how well I know an item, and accordmg to the h n d of item
it is, I can recognize it, describe it, reproduce it, draw it, recite
it, apply it, and so forth. Statements llke 'I know very well he
- - ~

isn't angry' or 'You know very well that isn't calico', though of
course about the current case, ascribe the excellence of the
knowledge to past experience, as does the general expression
'YOU are old enough to know better'.2

'I know, I know, I've seen it a hundred times, don't keep on t e h g me'
complains of a superabundance of opportunity: 'knowing a hawk from a
handsaw' lays down a minimum of acumen in recognition or classification.
'As well as I know my own name' is said to typify sometlung I must have
experienced and must have learned to discriminate.
The adverbs that can be inserted in 'How . . . do you know?' are few in
number and fall into still fewer classes. There is practically no overlap with
 OTHER M I N D S 8r
By contrast, the questions raised in (2) and (4) concern the
circumstances of the current case. Here we can ask 'How
definitely do you know?' You may know it for certain, quite
positively, officially, on h s own authority, from unimpeach-
able sources, only indirectly, and so forth.
Some of the answers to the question 'How do you know?'
are, oddly enough, described as 'reasons for knowing' or
'reasons to know', or even sometimes as 'reasons why I know',
despite the fact that we do not ask 'Why do you &ow?' But
now surely, according to the Dictionary, 'reasons' should be
given in answer to &e question 'Why?' just as we do in fact
give reasons for believing in answer to the question 'Why do
you believe?' However there is a distinction to be drawn here.
'How do you know that IG Farben worked for war?' 'I have
every reason to know: I served on the investigating comrnis-
sion': here, giving my reasons for knowing is stating how I
come to be in a position to know. In the same way we use the
expressions 'I know because I saw h m do it' or 'I know because
I looked it up only ten minutes ago': these are similar to 'So it
is: it is How did yo;know?' 'I did quite a bit of
physics at school before I took up phdology', or to 'I ought to
know: I was standing only a couple of yards away'. Reasons
for believing on the oiher hand are normally quitea different
affair (a recital of symptoms, arguments in support, and so
forth), though- there are cases where we do give as reasons for
believing our having been in a position in which we could get
good evidence: 'Why do you believe he was lying?' 'I was
w a t c h g him very closely.'
Among- the cases where we give our reasons for knowing
thngs, a special and important class is formed by those where
we cite authorities. If asked 'How do you know the election
is today?', I am apt to reply 'I read it in The Times', and if
asked 'How do vou know the Persians were defeated at Mara-
J

chon?' I am apt to reply 'Herodotus expressly states that they

those that can be inserted in 'How . . . do you believe?' (firmly, sincerely,
genuinely, &c.).
 82 OTHER MINDS
were'. In these cases 'know' is correctly used: we know 'at
second hand' when we can cite an authority who was in a posi-
tion to know (possibly hmself also onl; at second haid).'
The statement of an authority makes me aware of something, -
enables me to know somethng, whch I shouldn't otherwise
have known. It is a source of knowledge. In many cases, we
contrast such reasons for knowing with other reasons for be-
lieving the very same h n g : 'Even if we didn't know it, even
if he hadn't confessed, the evidence against hun would be
enough to hang him'.
It is evident, of course, that this sort of 'knowledge' is 'liable
to be wrong', owing to the unreliability of human testimony
(bias, mistake, lying, exaggeration, 8rc.). Nevertheless, the
occurrence of a piece of human testimony radically alters the
situation. W e say 'We shall never know what caesar's feelings
were on the field of the battle of Philippi', because he did not
pen an account of them: if'he had, then to say 'We shall never
know' won't do in the same way, even though we may still
perhaps find reason to say 'It doesn't read very plausibly: we
shall never really know the truth' and so on. Naturally, we arc
judicious: we don't say we know (at second hand) if there is
any special reason to doubt the testimony: but there has to be
some reason. It is fundamental in tallung (as in other matters)
that we are entitled to trust others, except in so far as there is
some concrete reason to distrust then. Believing persons,
accepting testimony, is the, or one main, point of t a h g .
W e don't play (competitive) games except in the faith that
our opponent is trying to win: if he isn't, it isn't a game,
but something different. So we don't talk with people

Knowing at second hand, or on authority, is not the same as 'knowing

indirectly', whatever precisely that d&cult and perhaps artificial expression
may mean. If a murderer 'confesses', then, whatever our opinion of the worth
of the 'confession', we cannot say that 'we (only) know indirectly that he did
it', nor can we so speak when a witness, reliable or unreliable, has stated that
he saw the man do it. Consequently, it is not correct, either, to say that the
murderer himselfknows 'directly' that he did it, whatever precisely 'knowing
directly' may mean.
 OTHER MINDS 83

(descriptively) except in the taith that they are trying to

convey.information.
It is now time to turn to the question 'How can you tell?',
i.e. to senses (2) and (4) of die question 'How do you know ?'
If you have asked 'How do you know it's a then
I may reply 'From its behaviour', 'By its marlungs', or, in more
detail, 'By its red head', 'From its eating thistles'. That is,
I indicate, o r to some extent set out with some degree of
precision, those features of the situation which enable me to
recognize it as one to be described in the way I did describe it.
Thereupon you may still object in several ways to my saying
it's a goldfinch, without in the least 'disputing my facts',
which is a further stage to be dealt with later. You may object:

(I) But goldfinches don't have red heads

) that's not a goldfinch. From your own description
( ~ a But
I can recognize it as a goldcrest
(2) But that's not enough: plenty of other birds have red
heads. What you say doesn't prove it. For all you know,
it may be a woodpecker.

Objections ( I ) and ( ~ a claim

) that, in one way or another, I
am evidently unable to recognize goldfinches. It may be
(la)--that I have not learned the right (customary, popular,
official) name to apply to the creature ('Who taught you to
use the word "goldfinch"?'):2 or it may be that my powers of
discernment, and consequently of classification, have never
been brought sharply to bear in these matters, so that I remain

Reliance on the authority of others is fundamental, too, in various special

matters, for example, for corroboration and for the correctness of our own use
of words, which we learn from others.
Misnaming is not a trivial or laughing matter: Jf I misname I shall mislead
others, and I shall also misunderstand information given by others to me. 'Of
course I knew all about his condition perfectly, but I never realized that was
diabetes: I thought'it was cancer, and all the books agree that's incurable: if I'd
only known it was diabetes, I should have thought of insulin at once'. Know-
ing what a thing is is, to an important extent, knowing what the name for it,
and the right name for it, is.
84 OTHER MINDS
confused as to how to tell the various species of snlall British
bird. Or, of course, it may be a bitsofboth. In making this sort
of accusation, you would perhaps tend not so much to use the
'
expression You don't know' or 'You oughtn't to say you
know' as, rather, 'But that isn't a goldfinch (go!df;nch)', or
-

'Then you're wrong to call it a goldfinch'. But still, if asked,

you would of course deny the statement that I do know it is
a goldfinch.
It is in the case of objection (2) that you would be more
inclined to say right out 'Then you don't know'. Because it
doesn't prove it, it's not enough to prove it. Several impor-
tant points come out here:
(a) ~f you say 'That's not enough', then you must have in
mind some more or less definite lack. 'To be a goldfinch,
besides having a red head it must also have the characteristic
eye-marhngs': or 'How do you know it isn't a woodpecker?
Woodpeckers have red heads too'. If there is no definite lack,
which you are at least prepared to specify on being pressed, then
it 's silly (outrageous) just to go on saying 'That's not enough'.
(b) Enough is enough: it doesn't mean everythg. Enough
means enough to show that (within reason, and for present
intents and purposes) it 'can't' be anytlung else, there is no
room for an alternative, competing, description of it. It does not
mean, for example, enough to show it isn't a stuted
(c) 'From its red head', given as an answer to 'How do you
know?' requires caretul consideration: in particular it differs
very materially from 'Because it has a red head', which is also
sometimes given as an answer to 'How do you know ?', and
is commonly given as an answer to 'Why do you believe?' It
is much more akin to such obviously 'vague' replies as 'From
its markings' or 'From its behaviour' than at first appears. Our
claim, in saying we know (i.e. that we can tell) is to recognize:
and recognizing, at least in this sort of case, consists in seeing,
or otherwise sensing, a feature or features which we are sure
are similar to something- noted (and usually named) before,
on some earlier occasion in our experience.~ut,h s that we
 OTHER M I N D S 85

see, or otherwise sense, is not necessarily describable in words,

s t d less describable in detail, and in non-committal words, and
by anybody you please. Nearly everybody can recognize a
surly look or the smell of tar, but few can describe them non-
committally, i.e. otherwise than as 'surly' or 'of tar': many can
recognize, and 'with certainty', ports of different vintages,
models by different fashon houses, shades of green, motor-car
makes from behlnd, and so forth, without being able to say
'how they recognize them', i.e. without being- able to 'be more
specific about it'-they can only say they can tell 'by the taste',
'from the cut', and so on. So, when I say I can tell the bird
'from its red head', or that I know a friend 'by h s nose', I imply
- -
that there is something peculiar about the red head or the nose,
something pecuhar to goldfinches or to h m , by which you can
(always) tell them or him. In view of the fewness and crudeness
of the classificatory words in any language compared with the
infinite number of features which are tecog&ed, or which
could be picked out and recognized, in our experience, it is
small wonder that we often and often fall back on the phrases
beginning with 'from' and 'by', and that we are not able to
say, further and precisely, how we can tell. Often we know
h g s quite
-
well, while scarcely able at all to say 'from' what
we know them, let alone what there is so very special about
them. Any answer beginning 'From' or 'By' has, intentionally,
saving 'vagueness'. But on the contrary, an answer begin-
ning 'Because' is dangerously definite. When I say I know it's
a goldfinch '~ecause-ithas a red head', that implies that all I
have noted, or needed to note, about it is that its head is red
(nothmg special or peculiar about the shade, shape, &c. of the
patch): so that I imply that there is no other small British bird
that has any sort of red head except the goldfinch.
(d) Whenever I say I know, I am always liable to be taken
to claim that, in a dertain sense appropriate to the kmd of
'statement (and to present intents and purposes), I am able to
prove it. In the present, very common, type of case, 'proving'
seems to mean stating what are the features of the current case
86 OTHER MINDS
whch are enough - to constitute it one whch i s correctly de-
scribable in the way we have described it, and not in any other
way relevantly variant. Generally speaking, cases where I can
'prove' are cases where we use the 'because' formula: cases
L

where we 'know but can't prove' are cases where we take

refuge in the 'from' or 'by' formula.
I believe that the points so far raised are those most genuinely
and normally raised by the question 'How do you know?'
But there are other, further, questions sometimes raised under
the same rubric, and especially by phlosophers, whch may be
thought more important. These are the worries about 'reality'
and about being - 'sure and certain'.
Up to now, in challenging me with the question 'How do
you know?', you are not taken to have queried my credentials as
stated, though you have asked what they were: nor have you
disputed my facts (the facts on which I am relying to prove it is
a goldfinch), though you have asked me to detail them. It is
this further sort ofchaflenge that may now be made, a challenge
as to the reliability of our alleged 'credentials' and our alleged
'facts'. You may ask:
(I) But do you know it's a real goldftnch? How do you
know you're not dreaming? Or after all, mightn't it be a
stuffed one? And is the head really red? Couldn't it have been
dyed, or isn't there perhaps an odd light reflected on it?
(2) But .are you certain it's the right red for a goldfinch?
Are you quite sure it isn't too orange? Isn't it perhaps rather
too strident a note for a bittern?
These two sorts of worry are distinct, though very pro-
bably they can be combined or confused, or may run into one
another: e.g. 'Are you sure it's really red?' may mean 'Are you
sure it isn't orange?' or again 'Are you sure it isn't just the
pecuhar light?'

I. Reality
If you ask me, 'How do you know it's a real stick?' 'How do
you know it's really bent?' ('Are you sure he's really angry?'),.
 OTHER M I N D S 87

then you are querying my credentials or my facts (it 's often

uncertain whch) in a certain special way. In various special,
recognized ways, depending essentially upon the nature of the
matter w h c h I have announced myself to know, either my
current experiencing or the item currently under consideration
(or uncertain which) may be abnormal, phoney. Either I myself
may be dreaming, or in delirium, or under the Influence of
mescal, &c.: or else the item may be stuffed, painted, dummy,
artificial, trick, freak, toy, assumed, feigned, &c.: or else again
there's an uncertainty (it's left open) whether I am to blame
or it is-mirages, mirror images, odd lighting effects, &c.
These doubts are all to be allayed by means of recognized -

procedures (more or less roughly recognized, of course),

appropriate to the particular type of case. There are recognized
ways of distinguishing between dreaming and waking ( h o w
otherwise should we know how to use and to contrast the
words?), and of deciding whether a thing is stuffed or live, and
so forth. The doubt or question 'But is it a real one?' has always
(must have) a special basis, there must be some 'reason for sug-
gesting' that it isn't real, in the sense of some specific way, or
limited number of specific ways, in w h c h it is suggested that
this experience or item may be phoney. Sometimes (usually)
the context makes it clear what the suggestion is : the goldfinch
might be stuffed but there's no suggestion that it's a mirage, the
oasis might be a mirage but there's no suggestion it might be
stuffed. If the context doesn't make it clear, then I am entitled
to ask 'How do you mean? Do you mean it may be stuffed or
what? What are you suggesting?' The wile of the metaphysician
consists in asking 'Is it a real table?' (a kind of object w h c h has
no obvious way of being phoney) and not specifying or limit-
ing what may be wrong with it, so that I feel at a loss 'how to
prove' it is a real one.' It is the use of the word 'real' in &IS

Conjurers, too, trade on this. 'Will some gentleman kindly satisfy himself
that this is a perfectly ordinary hat?' This leaves us baffled and uneasy: sheep-
ishly we agree that it seems all right, while conscious that we have not the least
idea what to guard against.
88 OTHER M I N D S
manner that leads us on to the supposition that 'real' has a
single meaning ('the real world' 'material objects'), and that
- .

a highly profound and puzzling one. Instead, b e should insist

always on specifying with what 'real' is being contrasted-'not
what' I shall have to show it is, in order to show it is 'real': and
then usually we shall find some specific, less fatal, word, appro-
priate to the particular case, to substitute for 'real'.
Knowing it's a 'real' goldfinch isn't in question in the
ordinary case when I say I know it's a goldfinch: reasonable
-

precautions only are taken. But when it is called in question,

in special cases, then I make sure it's a real goldfinch in ways
essentially similar to those in which I made sure it was a gold-
fmch, though corroboration by other witnesses plays a specially
important part in some cases. Once again the precautions
cannot be more than reasonable, relative to current intents and
purposes. And once again, in the special cases just as in the
ordinary cases, two further conditions hold good:
(a) I don't by any means always know whether it's one or
not. It may fly away before I have a chance of testing it, or of
inspecting it thoroughly enough. This is simple enough: yet
some are prone to argue that because I sometimes don't know
or can't discover, I never can.
(b) 'Being sure it's real' is no more proof against miracles
or outrages bf nature than a n y h g else E or, sub specie humani-
tatis, can be. If we have made sure it's a goldfinch, and a real
goldfinch, and then in the future it does something outrageous
(explodes, quotes Mrs. Woolf, or what not), we don't say
we were wrong to say it was a goldfinch, we don't know what
to ray. Words literally fail us: 'What would you have said?'
'What are we to say now?' 'What would you say?' When I
have made sure it's a real (not stuffed, corroborated
by the disinterested, &c.) then I am not 'predicting' in saying
it's a real goldfinch, and in a very good sense I can't be
proved wrong whatever happens. It seems a serious mistake
- -

to suppose that language (or most language, language about

real t h g s ) is 'predictive' in such a way that the future can -
 OTHER M I N D S 89
always prove it wrong. What the future can always do, is to
make us revise our ideas about goldfinches or real goldfinches
or anydung else.
Perhaps the normal procedure of language could be schema-
tized as follows. First, it is arranged that, on experiencing a
complex of features C, then we are to say 'This is C' or 'This is
a C'. Then subsequently, the occurrence either of the whole of
C or of a sigdicant and characteristic part of it is, on one or
many occasions, accompanied or followed in definite circum-
stances by another special and distinctive feature or complex of
features, which maces it seem desirable to revise our ideas: so
that we draw a distinction between 'This looks hke a C, but in
fact is only a dummy, &c.' and 'Ths is a real C (live, genuine,
&c.)'. Henceforward, we can only ascertain that it's a real C by
ascertaining that the special feature or complex of features is
present in the appropriate circumstances. he old expression
'This is a C' wdl tend as heretofore to fail to draw any distinc-
tion between 'real, live, &c.' and 'dummy, stuffed, &c.' If the
special distinctive feature is one which does not have to manifest
itself in any defmite circumstances (on application of some
specific test, after some limited lapse of time, &c.) then it is
not a suitable feature on whch to base a distinction between
'real' and 'dummy, imaginary, &c.' All we can then do is to
say 'Some Cs are and-some aren't, some do and some don't:
a i d it may be very interesting or important whether they
are or aren't, whether they do or don't, but they're all Cs, real
Cs, just the same'.' Now if the special feature is one which
must appear in (more or less) definite circumstances, then 'This
is a real C' is not necessarily predictive: we can, in favourable
cases, make sure of
The awkwardness about some snarks being boojums.
Sometimes, on the basis of the new special feature, we distinguish, not
between 'Cs' and 'real Cs', but rather between Cs and Ds. There is a reason
for choosing the one procedure rather than the other: all cases where wc use
the 'real' formula exhibit (complicated and serpentine) likenesses, as do all
cases where we use 'proper', a word which behaves in many ways like 'real',
and is no less nor more profound.
90 OTHER M I N D S
2. Sureness and Certainty
The other way of querying my credentials and proofs ('Are
you sure it's the right red?') is quite different. Here we come
up against Mr. Wisdom's views on 'the peculiarity of a man's
Lowledge of his own sensations', for which he refers us to
'Other Minds VII' (Mind, vol. hi, N.s.,no. 207), a passage with
whch I find I disagree.
Mr. Wisdom there says that, excludmg from consideration
cases Lke 'being in love' and other case; which 'involve pre-
diction', and considering statements like 'I am in pain' which,
in the requisite sense, do not involve prediction, then a man
cannot 'be wrong' in making them, in the most favoured sense
of being wrong: that is, though it is of course possible for him
to lie (so that 'I am in pain' may be false), and though it is also
possible for h m to misname, i.e. to use the word 'pawn', say,
instead of 'pain', which would be liable to mislead others but
would not mislead himself, either because he regularly uses
'pawn' for 'pain' or because the.use was a momentary aberra-
tion, as when I call John 'Albert' while knowing him quite well
to be John-though it is possible for him to be 'wrong' in these
two senses, it is not possible for h m to be wrong- in the most
-

favoured sense. He says again that, with this class of statement

(elsewhere called 'sense-statements'), to know directly that one
is in pain is 'to say that one is, and to say it on the basis of being
in paid: and again, that the -~eculiarityof sense-statements lies
in the fact that-'when they are correct and made by X, then X
knows they are correct'.
This seems to me mistaken, though it is a view that, in more
or less subtle forms, has been the basis of a very great deal of
phdosophy. It is ~erhapsthe original sin (~erkeley's apple, the
tree in the quad) by which thepphdosobher casts hmself out
from the garden of the world we live in.
Very clearly detailed, this is the view that, at least and only
in a certain favoured type of case, I can 'say what I see (or
otherwise sense)' almost quite literally. O n t h s view, if I were
to say 'Here is something red', then I might be held to imply or
 OTHER MINDS 9I
to state that it is really a red thing, a thing whch would appear
red in a standard light, or to other people, or tomorrow too,
and perhaps even more besides: all of whch 'involves predic-
tion' (if not also a metaphysical substratum). Even if I were to
say 'Here is something which looks red', I might still be held to
imply or to state that it looks red to others also, and so forth.
If, however, I confine myself to stating 'Here is somedung that
looks red to me now', -then at last I can't be wrong the(GI
most favoured sense).
However, there is an ambiguity in 'sometlung that looks red
to me now'. Perhaps this can be brought out by italics, though
it is not really so much a matter of emphasii as of tone and
expression, of confidence and hesitancy. Contrast 1Here is
something that (definitely) looks to me (anyhow) red' with
'Here is something that looks to me (something ldce) red (I
should say)'. In the former case I am quite confident that, how-
ever it may look to others, whatever it may 'really be', &c., it
certainly does look red to me at the moment. In the other
case I'm not confident at all: it looks reddish, but I've never
seen a n y t h g quite ldce it before, I can't quite describe it-
or, I'm not very good at recognizing colours, I never feel
quite happy about them, I've constktly been caught out
about them. O f course, ths sounds silly in the case of 'red':
red is so very obvious, we all know red when we see it, it',s
unmistakable.' Cases where we should not feel happy about red
are not easy (though not impossible) to find. But take 'magenta':
'It looks rather Lke magenta to me-but then I wouldn't be
too sure about d i s t i n g u h g magenta from mauve or from
heliotrope. Of course I know in a way it's purplish, but I don't
really know whether to say it's magenta or not: I just can't be
sure.' Here, I am not interested in ruling out consideration of
-

how it looks to others (looks to me) or considerations about

what its real colour is (looks): what I am ruling out is my being
sure or certain what it looks to me. Take tastes, or take sounds:
And yet she always thought his shirt was white until she saw it against
Tommy's Persil-washed one.
92 OTHER MINDS
these are so much better as examples than colours, because we
never feel so happy with our other senses as with our eyesight.
Any description of a taste or sound or smell (or colour) or of a
feeling, involves (is) saying that it is hke one or some that we
have experienced before: any descriptive word is classificatory,
involves recognition and in that sense memory, and only when
we use such words (or names or descriptions, which come
down to the same) are we knowing anyhng, or believing
anythmg. But memory and recognition are often uncertain
and unreliable.
Two rather different ways of being hesitant may be distin-
guished.
(a) Let us take the case where we are tasting a certain taste.
We may say 'I simply don't know what it is: I've never tasted
a n y t h g remotely f i e it before. . . . No, it's no use: the more
I tlunk about it the more confused I get: it's perfectly distinct
and perfectly distinctive, quite unique in my experience'. Ths
dustrates the case where I can find n o h g in my past ex-
perience with whch to compare the current case: I'm certain
it's not appreciably hke a n y t h g .I ever tasted before, not
s&ciently hke anything I know to merit the same description.
This case, though distinguishable enough, shades off into the
more common type of case where I'm not quite certain, or
. -

only fairly certain, or practically certain, that G's the taste of,
say, laurel. In all such cases, I am endeavouring to recognize
the current item by s e a r c h g in my past experience for some-
thing like it, some likeness in virtue of which it deserves, more
or less positively, to be described by the same descriptive word:'
and I am meeting with varying degrees of success.
(b) The other case is different, though it very naturally
combines itself with the first. Here, what I try to do is to savour
the current experience, to peer at it, to sense it vividly. I'm not
sure it is the taste of pineapple: isn't there perhaps just some-
Or, of course, related to it in some other way than by 'similarity' (in any
ordinary sense of 'sirmlarity'), which is yet sufficient reason for describing it
by the same word.
 OTHER MINDS 93
thing about it, a tang, a bite, a lack of bite, a cloying sensation,
w h c h isn't quite right for pineapple? Isn't there perhaps just
a peculiar h t of green, whch would rule out mauve and
would hardly do fir heliotrope? Or perhaps it is famtly odd:
I must look more intently, scan it over and over: maybe just
possibly there is a suggestion of an unnatural shmmer, so that
it doesn't look quite like ordinary water. There is a lack of
sharpness in what we actually sense, which is to be cured
not, or not merely, by thmking, but by acuter clscernment,
by sensory discrimination (though it is of course true that
-

t h h g of other, and more pronounced, cases in our

past experience can and does assist our powers of discrimina-
tion).
Cases (a) and (b) a k e , and perhaps usually together, lead
to our being not quite sure or certain what it is, what to say,
how to describe it: what our feehgs really are, whether the
t i c b g is painful exactly, whether I'm really what you'd call
angry with him or only something rather llke it. The hesitation
is of course, in a sense, over misnaming: but I am not so much
or merely worried about possibly misleading others as about
misleading myself (the most favoured sense of being wrong).
I should suggest that the two expressions 'being certain' and
L
being sure', though from the nature of the case they are often
used indiscriminately, have a tendency to refer to cases (a) and
(b) respectively. 'Being certain' tends to indicate confidence in
our memories and our past discernment, 'being sure' to indicate
confidence in the current perception. Perhaps this comes out
in our use of the concessives 'to be sure' and 'certainly', and
in our use of such phrases as 'certainly not 'and 'surely not'.
But it may be unwise to c h w y language beyond the coarser
nuances.
It may be said that, even when I don't know exactly how to
describe it, I nevertheless know that I think (and roughly how
confidently I thmk) it is mauve. So I do know something. But
This appears to cover cases of dull or careless or uninstructed perception,
as opposed to cases of diseased or drugged perception.
 94 OTHER M I N D S
this is irrelevant: I don't know it's mauve, that it definitely
looks to me now mauve. Besides, there are cases where I really
don't know what I t h k : I'm completely baffled by it.
Of course, there are any number of 'sense-statements' about
w h c h I can be, and am, completely sure. In ordmary cases
ordinary men are nearly always-certain when a thing looks red
(or reddish, or anyhow reddish rather than greenish), or when
they're in pain (except when that's rather difficult to say, as
when they're being tickled): in ordinary cases an expert, a-

dyer or a dress designer, will be quite sure when something

looks (to h m in the present light) reseda green or nigger brown,
though those who are not experts will not be so sure. Nearly
always, if not quite always, we can be quite, or pretty, sure if
we take refuge in a sufficiently rough description of the sensa-
tion: roughness and sureness tend to vary inversely. But the
less rough descriptions, just as much as the rough, are all 'sense-
statements'.
It is, I thmk, the problems of sureness and certainty, w h c h
phlosophers tend (if I am not mistaken) to neglect, that have
considerably exercised scientists, while the problem of 'reality.',
which pldosophers have cultivated, does not exercise them.
The whole apparatus of measures and standards seems designed
to combat unsureness and uncertainty, and concomitantly to
increase the possible precision of language, which, in science,
pays. But for the words 'real' and 'unreal' the scientist tends to
substitute, wisely, their cash-value substitutes, of which he
invents and defines an increasing number, to cover an increasing
variety of cases: he doesn't ask 'Is it real?' but rather 'Is it de-
natured?' or 'Is it an allotropic form?' and so on.
It is not clear to me what the class of sense-statements is, nor
what its 'peculiarity' is. Some who talk of sense-statements (or
sense data) appear to draw a distinction between talking about
simple things like red or pain, and t a h n g about complicated
thngs llke love or tables. But apparently Mr. wisdom does
not, because he treats 'This looks to me now llke a man eating
poppies' as in the same case with 'Ths looks to me now red'.
 OTHER MINDS 95

In this he is surely riglit: a man eating poppies may be more

'complex' to recognize, but it is often not appreciably more
difficult than the other. But if, again, we say that non-sense-
statements are those which involve 'prediction', why so?
True, if I say, 'This is a (real) oasis' without first ascertaining
that it's not a mirage, then I do chance my hand: but if I have
ascertained that it';not, and can recognize for sure that it isn't
(as when I am d r h g its waters), then surely I'm not chant-
ing my hand any longer. I believe, of course, that it WLU con-
tinue to perform as (real) oases normally do: but if there's a
lusus naturae, a miracle, and it doesn't, that wouldn't mean I
was wrong, previously, to call it a real oasis.
With regard to Mr. Wisdom's own chosen formulae, we
have seen already that it can't be right to say that the peculi-
arity of sense-statements is that 'when they are correct, and
made by X,then X knows they are correct': for X may think,
without much confidence, that it tastes to him hke Lapsang,
and yet be far from certain, and then subsequently become
certain, or more certain, that it did or didn't. The other ~o
formulae were: 'TOknow that one is in pain is to say that one
is and to say it on the basis of being in pain' and that the only
4 U A 8

mistake possible with sense-statements is typified by the case

where 'knowing him to be Jack I call him "Alfred", thuJung
his name is Alfred or not caring a damn what his name is'. The
snag in both these lies in the phrases 'on the basis of being in
pain' and 'knowing him to be Jack'. 'Knowing him to be Jack'
means that I have recognized
u
him as Tack, a matter over which
d

I may well be hesitant and/or mistaken: it is true that I needn't

recognize him by natne as 'Jack' (and hence I may call him
'~lfred'),but at least I must be recognizing him correctly as,
for instance, the man I last saw in ~erusalem,or else I shal! be
misleadmg myself: Similarly, if 'on the basis of being in pain'
only means 'when I am (what would be correctly described as)
in pain', then something more than merely saying 'I'm in pain'
is necessary for knowing- I'm in pain: and this something more,
-

as it invoives recognition, may be hesitant and/or mistaken,

96 OTHER MINDS
though it b of course unllkely to be so in a case so compara--

tively obvious as that of pain.

Possibly the tendency to overlook the problems of recog-
nition is fostered by the tendency to use a direct object after
the word know. M;. Wisdom, for example, confidently uses
such expressions as 'knowing the feelings of another (his mind,
his sensations, his anger, hs pain) in the way that he knows
them'. But, although we d i correctly use the expressions
'I know your feehgs on the matter' or 'He knows his own
mind' or (archaically) -
~
'May I know your mind?', these are
rather special expressions, whch do not justify any general
usage. 'Feelings' here has the sense it has in 'very strong feelings'
in favour of or against somedung: perhaps it means 'views' or
'opinions' ('very decided opinions'), just as 'mind' in this usage
- -

is given by the Dictionary as equivalent to 'intention' or 'wish'.

To extend the usage uncritically is somewhat as though, on the.
strength of the legitimate phrase 'knowing someone's tastes',
we were to proceed to talk of 'knowing someone's sounds' or
'knowing someone's taste of pineapple'. If, for example, it is
a case of feehgs such as fatigue, we do not use the ex-
pression 'I know your feelings'.
When, therefore, Mr. wisdom speaks generally of 'knowing
his sensations', he presumably means this to be equivalent to
'knowing what he is seeing, s m e h g , &c.', just as 'knowing
the winner of the Derby' means 'knowing what won the Derby'.
But here again, the expression 'know what' seems sometimes
to be taken, unconsciously and erroneously, to lend support
to the practice of putting a direct object after know: for 'what'
is liable to be understood as a relative, = 'that which'. This is a
grammatical mistake: 'what' can of course be a relative, but in
'know what you feel' and 'know what won' it is an interro-
gative (Latin quid, not quod). In this respect, 'I can smell what
he is smehng' differs from 'I can know what he is smelhg'.
'I know what he is feeling' is not 'There is an x which both I
know and he is feehg', but 'I know the answer to the question
"What is he feehg?'" And similarly with 'I know what I am
 OTHER M I N D S 97
feeling': this does not mean that there is something which I am
both knowing and feeling.
- -
Expressions such as 'We don't know another man's anger
in the way he knows it' or 'He knows his pain in a way we
can't' seem barbarous. The man doesn't 'know his pain': he
feels (not knows) what he recognizes as, or what he knows to
be, anger (not his anger), and he knows that he is feelmg angry.
Always assuming that he does recogni;e the feelmg, which in
fact, though feeling it acutely, he may not: 'Now I know what
it was, it was jealousy (or gooseflesh or angina). At the time I
did not know at all what it was, I had never felt a n y t h g quite
like it before: but since then I've got to know it quite well'.'
Uncritical use of the direct obiect after know seems to be
one h g that leads to the view that (or to talking as though)
sensa, that is h g s , colours, noises, and the rest, speak or are
labelled by nature, so that I can literally say what (that which)
I see: it pipes up, or I read it off. It is as if sensa were literally to
'announce themselves' or to 'identify themselves', in the way
we indicate when we say 'It presently identified itself as a -parti-
-

cularly fme white rhoceros'. But surely this is only a manner

of speaking, a reflexive idiom in which the French, for example,
indulge more freely than the ~nglish:sensa are dumb, and only
previous experience enables us to identify them. If we choose
to say that they 'identify themselves' (and certalnly 'recogniz-
ing' is not a highly voluntary activity of ours), then it must be
admitted that they share the birthright of all speakers, that of
speakmg unclear$ and untruly.

If1 know I can't be wrong

One final point about 'How do you know?', the challenge
to the user of the expression 'I know', requires stdl to be brought
There are, of course, legitimate uses of the direct object after know, and of
the possessive pronoun before words for feelings. 'He knows the town well',
'He has known much suffering', 'My old vanity, how well I know it!'--even
the pleonastic 'Where does he feel his (= the) pain?' and the educative
tautology 'He feels his pain'. But none of these really lends support to the
metaphysical 'He knows his pain (in a way we can't)'.
98 OTHER M I N D S
out by consideration of the saying that 'If you know you can't
be wrong'. Surely, if what has so far been said is correct, then
we are often right to say we know even in cases where we turn
out subsequently to have been mistaken-and indeed we seem
always, or practically always, liable to be mistaken.
Now, we are perfectly, and should be candidly, aware of
this liabllty, which does not, however, transpire to be so very
onerous in practice. The human intellect and senses are, indeed,
inherently fallible and delusive, but not by any means inveterately
so. Machines are inherently liable to break down, but good
machines don't (often). It is futile to embark on a 'theory of
knowledge9which denies this liabllty : such theories constantly
end up by admitting the liability after all, and denying the
existence of 'knowledge'.
'When you know you can't be wrong9 is perfectly good
sense. You are prohibited from saying 'I know it is so, but
I may be wrong', just as you are prohibited from saying
'I I willbut I may fail'. If you are aware you may be
mistaken, you ought not to say you know, just as, if you are
aware you may break your word, you have no business to .
pomise. But of course, being aware that you may be mistaken
doesn't mean merely being aware that you are a fallible
human being: it means that you have some concrete reason to
suppose that you may be mistaken in dus case. Just as 'but I may
fail' does not-mean merely 'but I am a weak human being' (in
which case it would be no more exciting than addmg 'D.V.') :
it means that there is some concrete reason for me to suppose * A

that I shall break my word. It is naturally always possible

('humanly' possible) that I may be mistaken or may break my
word, but that by itself is no bar against using the expressions
'I know' and 'I promise' as we do in fact use them.
At the risk -(long since incurred) of being tedious, the
parallel between saying 'I know' and saying 'I promise' may
be elaborated.1
It is the use of the expresr'ms 'I know' and 'I promise' (fust person
singular, present indicative tense) alone that is being considered. 'If I knew,
 OTHER M I N D S 99

When I say 'S is P', I imply at least that I believe it, and, if I
- .

have been strictly brought up, that I am (quite) sure of it: when
I say 'I shall do A', I imply at least that I hope to do it, and, if
I have been strictly brought up that I (fully) intend to. If I only
believe that S is P, I can add 'But of course I may (very well)
be wrong:' if I only hope to do A, I can add 'But of course
I may (very well) not'. When I only believe or only hope, it
is recognized that further evidence or further circumstances
are liable to make me change my mind. If I say 'S is P' when
I don't even believe it, I am lying: if I say it when I believe
it but am not sure of it, I may be misleading but I am not
exactly lying. If I say 'I shall do A' when I have not even any
hope, not the slightest intention, of doing it, then I am de-
liberately deceiving: if I say it when I do not fully intend to, I
am misleading but I am not deliberately deceiving in the same
way.
But now, when I say ' I promise', a new plunge is taken: I
have not merely announced my intention, but, by using h s
formula (performing this ritual), I have bound myself to others,
and stakid my reputation, in a new way. Similarly, saying
'I know' is takmg a new plunge. But it is not saying 'I have
~erformeda specially s t r h g feat of cognition, superior, in
the same scale as believing and being sure, even to being merely
quite sure': for there is nothing in that scale superior to being
quite sure. Just as promising is not s o m e h g superior, in the
same scale as hoping and intending, even to merely fully in-
- -

tending: for there is n o h g in that scale superior to fully in-

tending. When I say 'I know', Igive others my word: I give others
my aulhority for saying that 'S is P'.
I can't have been wrong' or 'Ifshe knows she can't be wrong' are not worrying
in the way that 'If I ("you") know I ("you") cah't be wrong' is worrying.
Or again, 'I promise' is quite different from 'he promises': if I say 'I promise',
I don't say I say I promise, I promise, just as if he says he promises, he doesn't
say he says he promises, he promises: whereas if I say 'he promises', I do (only)
say he says he promises-in the other 'sense' of 'promise', the 'sense' in which
I say I promise, only he can say he promises. I describe his promising, but I do
my own promising and he must do his own.
 100 OTHER MINDS
When I have said only that I am sure, and prove to have
been mistaken, I am not liable to be rounded on by others in
the same way as when I have said 'I know'. I am sure.for my
part, you can take it or leave it: accept it if you thuJc ~ ' man
acute and careful person, that's your responsibility. But I don't
know 'for my part', and when I say 'I know' I don't mean
you can take it or leave it (though of course you can take
it or leave it). In the same way, when I say I fully intend to, I
do so for my part, and, according as you t M c highly or poorly
of my resolution and chances, you will elect to act on it or not
to act on it: but if I say I promise, you are entitled to act on it,
whether or not you choose to do so. If I have said I know or I
promise, you insult me in a special way by refusing to accept it.
W e allfeel the very great difference between saying even 'I'm
-

absolutrly sure' and saying 'I know': it is llke~thd difference

between saying even 'I firmly and irrevocably intend' and
'I ~rornise'.If someone has promised me to do A, then I am
-

entitled to rely on it, and can myself make promises on the

- of it: and so, where someone has said to me 'I know',
strength
I am entitled to say I know too, at second hand. The right to
say 'I know' is transmissible, in the sort of way that other
authority is transmissible. Hence, if I say it lightly, I may be
responsible for getting you into trouble.
If you say you know something, the most immediate chal-
lenge takes the form of asking 'A& you in a position to know?':
that is, you must undertake to show, not merely that you are
sure of it, but that it is within your cognisance. There is a similar
form of challenge in the case of promising: fully intending is
not enough-you must also undertake to show that 'you are
in a position to promise', that is, that it is within your power.
Over these points in the two cases parallel series of doubts are
apt to infect philosophers, on the ground that I cannot foresee
the future. Some begin to hold that I should never, or practi-
cally never, say I know anything-perhaps only what I am
sensing at this moment: others, that I should never, or practi-
-

tally never, say I promise-perhaps only what is actually witlun

 OTHER M I N D S 101

my power at h s moment. In both cases there is an obsession:

if I know I can't be wrong, so I can't have the right to say I
know, and if I promise I can't fail, so I can't have the right
- -
to say I promise. And in both cases rhls obsession fastens on
my inabht-y to make predictions as the root of the matter,
meaning by predictions claims to know the future. But tlus
is doubly mistaken in both cases. As has been seen, we may be
perfectly justified in saying we know or we promise, in spite of
the fact that thmgs 'may' turn out badly, and it's a more or less
serious matter for us if they do. And further, it is overlooked
that the conditions which must be satisfied if I am to show that
a thmg is within my cognisance or w i h my power are condi-
tions, not about the future, but about the present and the past: it
is not demanded that I do more than believe about the future.'
W e feel, however, an objection to saying that 'I know' per-
forms the same sort of function as 'I promise'. It is h s . Sup-
posing that h g s turn out badly, then we say, on the one hand
'You're proved wrong, so you didtr't know', but on the other
hand 'You've failed to perform, although you did promise'. I
believe that this contrast is more apparent than real. The sense
in which you 'did promise' is that you did say you promised
(did say 'I promise'): and you did say you knew. That is the
gravamen of the charge against you when you let us down,
after we have taken your word. But it may well transpire that -

you never fully intended to do it, or that you had concrete

reason to suppose that you wouldn't be able to do it (it might
even be manifestly impossible), and in another 'sense' of pro-
mise you can't then h&e promised to do it, so that you &dn't
I

Consider the use of other phrases analogous to 'I know' and

'I promise'. Suppose, instead of 'I know', I had said 'I swear':
- - -

in that case, upon the opposite appearing, we should say,

exactly as in the promising case, 'You did swear, but you were
IIf 'Figs never grow on thistles' is taken to mean 'None ever have and none
ever will', then it is implied that I know that none ever have, but only that I
believe that none ever will.
102 OTHER M I N D S
wrong'. Suppose again that, instead of 'I promise', I had said
'I guarantee' (e.g. to protect you from attack): in that case,
upon my letting you down, you can say, exactly as in the
knowing case 'You said you guaranteed it, but you didn't
guarantee it'.' Can the situation perhaps be summed up as
follows? In these 'ritual' cases, the approved case is one where
in the appropriate circvmstances, I say a certain formula: e.g. 'I do'
when standing, unmarried or a widower, beside woman, un-
married or a widow and not within the prohibited degrees
-

of relationship, before a clergyman, registrar, &c., or 'I give'

when it is mine to give, &c., or 'I order' when I have the
authority to, &c. But now, if the situation transpires to have
been in some way not orthodox (I was already married: it
wasn't mine to give: I had no authority to order), then we
tend to be rather hesitant about how to put it, as heaven was
when the saint blessed the penguins. W e call the man a bigamist,
- -

but his second marriage was not a marriage, is null and void
(a useful formula in many cases for avoidmg saying either 'he
did' or 'he didn't'): he did 'order' me to do it, but, having no
authority over me, he couldn't 'order' me: he did warn me it
was going to charge, but it wasn't or anyway I knew much
more about it than he did, so in a way he couldn!t warn me,
didn't warn men2W e hesitate between 'He didn't order me',
'He had no right to order me', 'He oughtn't to have said he
ordered me', just as we do between 'YOUdidn't know', 'You
can't have known', 'You had no right to say you knew' (these
perhaps having slightly different nuances, according to what
precisely it is that has gone wrong). But the essential factors

I 'Swear', 'guarantee', 'give my word', 'promise', all these an'dsimilar words

cover cases both of 'knowing' and of 'promising', thus suggesting the two are
analogous. O f course they differ subtly from each other; for example, know
and promise are in a certain sense 'unlimited' expressions, while when I swear
I swear upon something, and when I guarantee I guarantee that, upon some
adverse and more or less to be expected circumstance arising, I will take some
more or less definile action to nullify it.
' 'You can't warn someone of something that isn't going to happen'
parallels 'You can't know what isn't true'.
 OTHER M I N D S 503

are ( a ) You said you knew: you said you promised ( 6 ) You
were mistaken: you didn't perform. The hesitancy concerns
only the precise &ay in whch we are to round on the original
'I know' or 'I promise'.
To suppose that 'I know' is a descriptive phrase, is only one
example of the descriptive fallacy, so common in philosophy.
Even if some language is now purely descriptive, language was
not in origin so, and much of it is still not so. Utterance of
obvious ritual phrases, in the appropriate circumstances, is not
describing the action we are doing,- but doing it ('I do'): in other
cases it functions, l k e tone and expression, or again l k e punc-
tuation and mood, as an intimation that we are employing
language in some special way ('I warn', 'I ask', 'I define'). Such
phrases cannot, strictly, be lies, though they can 'imply7 lies,
as 'I promise' implies that I fully intend, which may be untrue.

If these are the main and multifarious points that arise in

famhar cases where we ask 'How do you know that this is a
case of so-and-so?', they may be expected to arise likewise in
cases where we say 'I know he is angry'. And if there are, as no
doubt there are, special difficulties in this case, at least we can
clear the ground a little of things which are not special
- d&-
culties, and get the matter in better perspective.
As a preliminary, it must be said that I shall only discuss
the question of feelings and emotions, with special reference
to anger. It seems lkely that the cases where we know that
another man thlnks that 2 and 2 make 4, or that he is seeing -
a
rat, and so on, are different in important respects from, though
no doubt also similar to, the case of knowing that he is angry
or hungry.
In he-first place, we certainly do say sometimes that we
know another man is angry, and we also distinguish these
occasions from others on which we say only that we believe he
is angry. For of course, we do not for a moment suppose that
we always know, of all men, whether they are angry or not,
or that we could discover it. There are many occasions when I
I 04 OTHER M I N D S
realize that I can't possibly tell what he's feeling: and there
are many types of people, and many individuals too, with whom
I (they being what they are, and I being what I am) never can
tell. The feelings of royalty, for example, or f&rs or bushmen
or Wykehamists or simple eccentrics-these may be very hard
to divine: unless you have had a prolonged acquaintance with
such persons, and some intimacy with them, you are not in
any sort of position to know what their feelings are, especially
if, for one reason or another, they can't or don't tell you.
O r again, the feelings of some individual whom you have never
met before-they might be almost anything: you don't know
his character at all or his tastes, you have had no experience of
his mannerisms, and so on. His feelings - are elusive and personal:
people differ so much. It is this sort of dung- that leads to the
situation where we say 'You never know' or 'You never can
tell'.
In short, here even more than in the case of the goldfinch,
-
a great deal depends on how familiar we have been in our past
experience with this type of person, and indeed with this Gdi-
vidual, in this type of situation. If we have no great familiar-
ity, t h a ~we hesitate to say we know: indeed, we can't be
expected to say (tell). On the other hand, if we have had
the necessary experience, then we can, in favourable current
circumstances, say we know: we certainly can recognize when
some near relative of ours is angrier than we have ever seen
him.
Further, we must have had experience
- also of the emotion
or feeling concerned, in this case anger.
- In order to know what
you are-feeling, I must also apparently be able to imagine
(guess, understand, appreciate)-what you're feeling. It seems
that more is demanded than that I shall have learned to dis-
criminate displays of anger in others: I must also have been
angry myself.' Or at any rate, if I have never felt a certain
I We say we don't know what it must feel like to be a king, whereas we
do know what one of our frie~ldsmust have felt when mortified. In this
ordinary (imprecise and evidently not whole-hog) sense of 'knowing what it
 OTHER M I N D S 105

emotion, say ambition, then I certainly feel an extra hesitation

in saying that h s motive is ambition. And thls seems to be due
to the very special nature (grammar, logic) of feelings, to the
special way in w h c h they are related to their occasions and
-

manifestations, whch requires further elucidation.

At first sight it may be tempting to follow Mr. Wisdom,
and to draw a hstinction between (I) the physical symptoms
and (2) the feeling. So that when, in the current case, I am
asked 'How can you tell he's angry?' I should answer 'From
the physical symptoms', whde if he is asked how he can tell he's
angry, he should answer 'From the feeling'. - But this seems to
be-a dangerous over-simplification.
In the first place, 'symptoms' (and also 'physical') is being
used in a way different from orchary usage, and one whlch
proves to be misleading.
'Symptoms', a term transferred from medical usage,' tends
to be used only, or primarily, in cases where that of w h c h
there are symptoms is somethmg undesirable (of incipient
hsease rather than of returning health, of despair rather than
of hope, of grief rather than o f joy) : and hence it is more
colourM than 'signs' or 'indications'. Ths, however, is com-
paratively trivial. What is important is the fact that we never
-

talk of 's~mptoms'or 'signs' except by w a y of implied contrast

with inspection of the item itself: N o doubt it would often be
awkward to have to say exactly where the signs or symptoms
end and the item itself begins to appear: but such a division is
always implied to exist. And hen& the words 'symptom' and
I
sign' have no use except in cases where the item, as in the case

would be like' we do often know what it would be like to be our neighbour

drawing his sword, whereas we don't know (can't even guess or imagine),
really, what it would feel like to be a cat or a cockroach. But of course we
don't ever 'know' what in our neighbour accompanies the drawing of his
sword in Mr. Wisdom's peculiar sense of 'know what' as equivalent to
'directly experience that which'.
Doctors nowadays draw a distinction of their own between 'symptoms'
and '(physical) signs': but the distinction is not here relevant, and perhaps not
very clear.
106 OTHER MINDS
of hsease, is liable to be hidden, whether it be in the future, in
the past, under the skin, or in some other more or less notorious
casket: and when the item is itself before us, we no longer talk
of signs and symptoms. When we talk of 'signs of a storm',
we mean signs of an impending storm, or of a past storm, or of
a storm beyond the horizon: we do not mean a storm on top

The words function U e such words as 'traces' or 'clues'.

Once you know the murderer, you don't get -
any more clues,
only Ghat were or would have been clues: nor is-a confession,
or an eye-witness's view of the crime, a particularly good clue-
these are something different altogether. When the cheese is
not to be found or seen, then there may be traces of it: but not
when it's there in front of us (though of course, there aren't,
then, 'no traces' of it either).
For this reason, it seems misleading to lump together, as a
- -

general practice, all the characteristic features ofPany casual

item as 'signs' or 'symptoms' of it: though it is of course some-
times the case that some t h g s which could in appropriate
circumstances be called characteristics or effects or manifesta-
tions or parts or sequelae or what not of certain items may also
be called signs or symptoms of those items in the appropriate
circumstances. It seems to be t h w h c h is really wrong with
Mr. Wisdom's paradox (Other Minds 111) about loolung in
the larder and finding 'all the signs' of bread, when we see the
loaf, touch it, taste it and so on. Doing these things is not find-
ing (some) signs of bread at all: the taste or feel of bread is
not a sign or symptom of bread at all. What I might be taken
to mean if I announced that I had found signs of bread in
the larder seems rather doubtful, since bread is not normally
There are some, more complicated, cases like that of inflation, where the
signs of incipient inflation are of the same nature as inflation itself, but of a less
intensityor at a slower tempo. Here, especially, it is a matter for decision where
the signs or 'tendencies' end and where the state itself sets in: moreover, with
idation, as with some diseases, we can in some contexts go on talking of
signs or symptoms even when the item itself is quite fairly decidedly present,
because it is such as not to be patent to simple observation.
 OTHER MINDS 107
casketed (or if in the bin, leaves no traces), and not being
a transeuAt event (impending bread, &c.), does not have
any normally accepted 'signs': and signs, peculiar to the
item, have to be more or less normally accepted. I might
be taken to mean that I had found traces of bread, such
as crumbs, or signs that bread had at one time been stored
there, or something of the kind: but what I could not be taken
to mean is that I had seen, tasted, or touched (somethmg like)
bread.
The sort of thmg we do actually say, if the look is all right
but we haven't yet tasted it, is 'Here is something that looks
hke bread'. If it turns out not to be bread after all, we might
say 'It tasted like bread, but actually it was only bread-substi-
tute', or 'It exhibited many of the characteristic features of
bread, but differed in important respects: it was only a synthetic
imitation'. That is, we don't use the words 'sign' or 'symptom'
at all.
Now, if 'signs' and 'symptoms' have this restricted usage, -
it
is evident that to say that we only get at the 'signsy or 'symp-
toms' of anythmg is to imply that we never get at it (and this
goes for 'all. the signs' too). So that,if we say that I only get at
the symptoms of his anger, that carries an important implica-
tion. But is thls the way we do talk? Surely we do not consider
that we are never aware of more than symptoms of anger in
another man ?
6
Symptoms' or 'signs' of anger tend to mean signs of rising
or of suppressed anger. Once the man has exploded, we talk of
something different--of an expression
-
or manifestation or
display of anger, of an exhibition of temper, and so forth. A
twitch of the eyebrow, pallor, a tremor in the voice, all these
may be symptoms of anger: but a violent tirade or a blow in
the face are not, they are the acts in which the anger is vented.
'Symptoms' of anger are not, at least normally, contrasted with
the man's own inner personal feeling of anger, but rather with
the actual display of anger. Normally at least, where we have
only symptoms to go upon, we should say only that we believe
108 OTHER M I N D S
that the man is angry or getting angry: whereas when he has
given hvnself away we say that we know.'
The word 'physical7also, as used by Mr. Wisdom in contrast
to 'mental', seems to me abused, though I am not confident
as to whether t h s abuse is misleading in the current case. He
evidently does not wish to call a mall's feelings, which he cites
as a typical example of a 'mental' event, physical. Yet this is
what we ordinarily often do. There are many physical feelings,
such as giddiness, hunger or fatigue: and these are included by
some doctors among the physical signs of various complaints.
Most feelings we do not speak of as either mental or physical,
especially en~otions,such as jealousy or anger itself: we do not
assign them to the mind hut to the heart. Where we do describe
a feeling as mental, it is because we are using a word normally
used to describe a physical feeling in a special transferred sense,
as when we talk about 'mental' discomfort or fatigue.
It is then, clear, that more is involved in being, for example,
angry than simply showing the symptoms and feeling the feel-
ing. For there is also the display or manifestation. And it is to
- .

be noted that the feeling is related in a unique sort of way to

the display. When we are angry, we have an impulse, felt
and/or acted on, to do actions of particular kinds,.and, unless
-

we suppress the anger, we do actually proceed to do them.

There is a peculiar and intimate relationship between the
emotion and the natural manner of venting it, with which,
having been angry ourselves, we are acquainted. The ways in
which anger is normally manifested are natural to anger just

Sometimes, it is said, we use 'I know' where we should be prepared t o

substitute 'I believe', as when we say 'I know he's in, because his hat is in the
hall': thus 'know' is used loosely for 'believe', so why should we suppose there
is a fundamental difference between them? But the question is, what exactly
d o we mean by 'prepared to substitute' and 'loosely'? W e are 'prepared t o
substitute' believe for know not as an equivalent expression but as a weaker and
therefore preferable expression, in view of the seriousness with which, as has
become apparent, the matter is to be treated: the presence of the hat, which
would serve as a proof of it's owner's preseiice in many circumstances, could
only through laxity be adduced as a proof in a court of law.
 OTHER MINDS 109

as there are tones naturally expressive of various emotions

(indignation,
-
&c.). There is not normally taken to be' such a
thing as 'being angry' apart from any impulse, however vague,
to vent the anger in the natural way.
Moreover, besides the natural expressions of anger, there
are also the natural occasions of anger, of which wehave also
had experience, which are similarly connected in an intimate
way with the 'being angry'. It would be as nonsensical to class
these as 'causes' in some supposedly obvious and 'external'
sense, as it would be to class the venting of anger as the 'effect'
of the emotion in a supposedly obvious and 'external' sense.
Equally it would be nonsensical to say that there are three
wholly distinct phenomena, ( I ) cause or occasion; (2) feeLng or
emotion, and (3) effect or manifestation, whch are related
together 'by definition' as all necessary to anger, though t h s
would perhaps
- -
be less misleading than the other.
-

It seems fair to say that 'being angry' is in many respects like

'having mumps'. It is a description of a whole pattern of events,
includmg occasion, symptoms, f e e h g and manifestation, and
possibly other factors besides. It is as silly to ask 'What, really,
is the anger itsey?' as to attempt to fme down 'the disease' to
some one chosen item ('the functional disorder'). That the man
himself feels something which we don't (in the sense that he
feels angry and we don't) is, in the absence of Mr. Wisdom's
variety of telepathy,2 evident enough, and incidentally nothmg
to complain about as a 'predicament': but there is no call to say
that 'that' ('the feehg')3 is the anger. The pattern of events
A new language is naturally necessary if we are to admit unconscious
feelings, and feelings which express themselves in paradoxical manners, such
as the psycho-analysts describe.
There is, it seems to me, something which does actually happen, rather
different from Mr. Wisdom's telepathy, which does sometimes contribute
towards our knowledge of other people's feelings. W e do talk, for example, of
'feelmg another person's displeasure', and say, for example, 'his anger could
be felt', and there seems to be sometling genuine about this. But the feeling
we feel, though a genuine 'feehg', is not, in these cases, displeasure or anger,
but a special counterpart feeling.
3 The 'feelings', i.e. sensations, we can obsewe in ourselves when angry arc
I 10 OTHER MINDS
whatever its precise form, is, fairly clearly, peculiar to the case
of 'feelings' (emotions)-it is not by any means exactly llke
the case of diseases: and it seems to be t h s peculiarity whlch
makes us prone to say that, unless we have had experience of
a feeling ourselves, we cannot know when someone else is
experiencing it. Moreover, it is our confidence in the general
paitern that-makes us apt to say we 'know' another man is
angry when we have only observed parts of the pattern: for
the parts of the pattern are related to each other very much
more intimately than, for example, newspapermen scurrying
in Brighton are related to a fire in Fleet Street.'
The man himself, such is the overriding power of the
pattern, will sometimes accept corrections from outsiders about
his own emotions, i.e. about the correct description of them.
He may be got to agree that he was not really angry so much
as, rather, indignant or jealous, and even that he was not in
pain, but only fancied he was. ~ n this d is not surprising,
especially in view of the fact that he, ldce d of us, bas primarily
learnt to use the expression 'I am angry' of himself by (a)
noting the occasion, symptoms, manifestation, &c., in cases
where other persons say 'I am angry' of thetnselves (b) being
told by others, who have noted all that can be observed about
him on certain occasions, that 'You are angry9, i.e. that he
should say 'I am angry'. On the whole, 'mere' feelings or
emotions, if there are such h g s genuinely detectable, are
certainly very hard to be sure about, even harder than, say,
tastes, which we already choose to describe, normally, only by
their occasions (the taste 'of tar', 'of. pineapple', ac.).
All words for emotions are, besides, on the vague side, in
two ways, leading to further hesitations about whether we
6
know' when he's angry. They tend to cover a rather wide
and illdefined variety of situations: and the patterns they cover

such things as a pounding of the heart or tensing of the muscles, which cannot
in themselves be justifiably called 'the feeling of anger'.
It is therefore misleading to ask 'How do I get from the scowl to the
anger ?'
 OTHER MINDS III

tend to be, each of them, rather complex (though common

and so not dficult to recognize,
-
very often), so that it is easy
for one of the more or less necessary features to be omitted,
and thus to give rise to hesitation about what exactly we should
say in such an unorthodox case. W e reahze, well enough, that
the challenge to w h c h we are exposed if we say we know is
to prove it, and in t h s respect vagueness of terminology is a
crippling handicap.
So far, enough has perhaps been said to show that most of
the daculties w h c h stand in the way of our saying we know
a thing is a goldfinch arise in rather greater strength in the case
where we want to say we know another man is angry. But
there is still a feelng, and I dunk a justified feehg, that there is
a further and quite special difficulty in the latter case.
Ths dficulty seems to be of the sort that Mr. Wisdom
raises at the very outset of h s series of articles on 'Other Minds'.
It is asked, might the man not exhibit all the symptoms (and
&splay and everythng else) of anger, even ad infinitum, and
yet still not (really) be angry? 4It will be remembered that he
there treats it, no doubt provisionally, as a difficulty similar to
that whch can arise concerning the reality of any 'material
object'. But in fact, it has special features of its own.
There seem to be three distinguishable doubts whch may
arise:
I. w h e n to all appearances angry, might he not really be
labouring under sopme other emotion, that, though he
normally feels the same emotion as we should on occasions
when we, in h s position, should feel anger and in m . h g
displays such as we make when angry, in ths particular case
he is acting abnormally?
2. When to all appearances angry, might he not really be
labouring- under some other emotion, in that he normally feels,
on occasions when we in h s position should feel anger and
when acting as we should act if we felt anger, some feeling
whch we, if we experienced it, should distinguish from
anger ?
I 12 OTHER MINDS
3. When to all appearances angry, might he not really be
f e e h g no emotion at all?
In everyday life, all these problems arise in special cases, and
occasion genuine worry. We may worry (1)as to whether
someone is deceiving us, by suppressing hls emotions, or by
feigning emotions w h c h he does not feel: we may worry (2)
as to whether we are misunderstanding someone (or he us), in
wrongly supposing that he does 'feel hke us', that he does share
emotions llke ours: or we may worry (3) as to whether some
action of another person is really deliberate, or perhaps only
involuntarv or inadvertent in some manner or olher. All three
d

varieties of worry may arise, and often do, in connexion with

the actions of persons whom we know very well.' Any or all
of them may be at the bottom of the passage from Mrs. ~ o o l f : ~
all work together in the feeling of loneliness whch affects
everybody at times.
None of these three special difficulties about 'reality' arises
in connexion with goldfinches or bread, any more than the
special difficulties about, for example, the oasis arise in 'con-
nexion with the reality of another person's emotions. The
goldfinch cannot be assumed, nor the bread suppressed: we
-

may be deceived by the appearance of an oasis, or misinterpret

the signs of the weather, but the oasis cannot lie to us and we
cannot misunderstand the storm in the way we misunderstand
the man.
Though the d&culties are special, the ways of deahng with
them are, initially, similar to those employed in the case of the
goldfinch.
-
There are (more or less roughly)
-
established pro-
-

cedures for dealing with suspected cases of deception or of

misunderstandmg or of inadvertence. By these means we do
very often establish (though we do not expect always to
establish) that someone is acting, or that we were misunder-
There is, too, a special way in which we can doubt the 'reality' of our own
emotions, can doubt whether we are not 'acting to ourselves'. Professional
actors may reach a state where they never really know what their genuine
feelings are.
[Quoted by Wisdom in his contribution to this Symposium. Ed.]
 OTHER MINDS 113

standing him, or that he is simply impervious to a certain

emotion, or that he was not acting voluntarily. These special
cases where doubts arise and require resolving, are contrasted
with the normal cases which hold the field1 unless there is some
special suggestion that deceit, &c., is involved, and deceit,
moreover, of an intehgible kind in the circumstances, that
is, of a kind that can be looked into because motive, &c., is
specially suggested. There is no suggestion that I never know
what other pedpIe's emotions are, nor yet that in particular
cases I might be wrong for no special reason or in no special
way.
Extraordinary cases of deceit, misunderstanding, &c. (whlch
are themselves not the normal), do not, e x vi termini, ordinarily
occur: we have a working knowledge of the occasions for,
the temptations to, the practical limits of, and the normal types
of deceit and misunderstanding. Nevertheless, they may occur,
and there mqy be varieties whch are common without our yet
having become aware of the fact. If t h s happens, we are in a
certain sense wrong, because our terminology is inadequate to
the facts, and we shall have thenceforward to be more wary
about saying -
we know, or shall have to revise our ideas and
terminal-ogy. This we are constantly ready to do in a field so
complex and b a h g as that of the emotions.
There remains, however, one further special feature of the
case, which also differentiates it radically-from the
case. The goldfinch, the material object, is, as we insisted above,
~lninscribedand mute: but the man speaks. In the complex of
occurrences which induces us to say we know another man is
angry, the complex of symptoms, occasion, display, and the
rest, a peculiar place is occupied by the man's own statement as
to what h s feelings are. In the usual case, we accept thls state-
ment without question, and we then say that we know (as it
were 'at second-hand') what his feehgs are: though of course
'at second-hand' here could not be used to imply that anybody
but he could know 'at first-hand', and hence perhaps it is not
'YOUcannot fool all of the people all of the time' is 'analytic'.
114 OTHER MINDS
in fact used. In unusual cases, where h s statement confLcts with
the description we should otherwise have been inched to give
of the case, we do not feel bound to accept it, though we
always feel some uneasiness in rejecting it. If the man is an
habitual liar or selfdeceiver, or if there are patent reasons why
he should be lying or deceiving himself on this occasion, then
we feel reasonably happy: but if such a case occurred as the
imagined one where a man, having given throughout life
every appearance of holding a certain pointless belief, leaves
behind a remark in h s private diary to the effect that he
-

never did believe it, then we probably should not know what
to say.
I should like to make in conclusion some further remarks
about thls crucial matter of our believing what the man says
about h s own feehgs. Although I know very well that I do
not see my way clearly in t h , I cannot help feeling sure that
it is fundamental to the whole Predicament, and that it has not
been given the attention it deserves, possibly just because it
is so obvious.
The man's own statement is not (is not treated primarily as)
a sign or symptom, although it can, secondarily and artificially,
be treated as such. A unique -place is reserved for it in the sum-
mary of the facts of th; case. The question then is: 'Why
believe h m ?'
There are answers that we can give to t h s question, w h c h
is here to be taken in the general sense of 'Why believe h m
ever?' not simply as 'Why believe him this time?' W e may say
that the man's statements on matters other than h s own feehgs
have constantly been before us in the past, and have been
regularly
- verified by our own observations of the facts he re-
ported: s o that we-have in fact some basis for an induction
about h s general reliability. Or we may say that h behaviour
is most simply 'explained' on the view that he does feel emo-
tions hke ours, just as psycho-analysts 'explain' erratic be-
haviour by analogy -
-
with normal behaviour when they use the
terminology of 'unconscious desires'.
 OTHER MINDS 115

These answers are, however, dangerous and unhelpful. They

are so obvious that they please nob6dy: whde on the other
hand they encourage the questioner to push h s question to
I
profounder' depths, encouraging us, in turn, to exaggerate
these answers until they become distortions.
The question, pushed further, becomes a challenge to the
very possibhty of 'believing another man', in its ordinarily
accepted sense, at all. What 'justification' is there for supposing
that there is another mind communicating with you at all?
How can you know what it would be k e for another mind to
feel anydung, and so how can you understand it? It is then that
we are tempted to say that we only mean by 'believing hun'
that we take certain vocal noises as signs of certain impending
behaviour, and that 'other minds' are no more really real than
unconscious desires.
This, however, is distortion. It seems, rather, that believing
in other persons, in authority and testimony, is an essential part
of the act of communicating, an act whch we all constantly
-

perform. It is as much an irreducible part of our experience

as, say, giving promises, or playng competitive games, or
even sensing coloured patches. W e can state certain advan-
tages of such performances, and we can elaborate rules of a
kind for their 'rational' conduct (as the Law Courts and
historians and psychologists work out the rules for accepting
testimony). But there is no 'justification' for our doing them
as such.

Final Note
One speaker at Manchester said roundly that the real crux
of the matter remains still that 'I ought not to say that I know
Tom is angry, because I don't introspect his feelings': and this
no doubt is just what many people do boggle at. The gist of
what I have been trying to bring out is simply:
I . Of course I don't introspect Tom's feelings (we sllould be
in a pretty predicament if I did).
2. Ofcotirse I do sometimes know Tom is angry.
116 OTHER M I N D S
Hence
3. to suppose that the question 'How do 1 know that Tom is
angry?' is meant to mean 'How do I introspect Tom's feelings?'
(because, as we know, that's the sort of thng that knowing
is or ought to be), is simply barking our way up the wrong
gum tree.
 TRUTH'

I. 'WHATis truth?' said jesting Pilate, and would not stay for
an answer. Pilate was in advance of h s time. For 'truth' itself
is an abstract noun, a camel, that is, of a logical construction,
-

which cannot get past the eye even of a grammarian. We

approach it cap and categories in hand: we ask ourselves
whether Truth is a substance (the Truth, the ~ o d of y Know-
ledge), or a .quality (something like the colour red, Inhering
in truths), or a relation ('correspondence').2 But philosophers
should take something more nearly their own size to strain at.
What needs cLscussing - rather is the use, or certain uses, of the
word 'true'. In vino, possibly, 'weritar', but in a sober sym-
posium 'werum'.
2. What is it that we say is true or is false? Or, how does the
phrase 'is true' occur in English sentences? The answers appear
at first multifarious. We say (or are said to say) that beliefs are
. ~

true, that descriptions or accounts are true, that propositions

or assertions or statements are true, and that words or sentences
are true: and thls is to mention only a selection of the more
obvious candidates. Again, we say (or are said to say) 'It is
. .

true that the cat is on the mat', or 'It is true to say that the cat
is on the mat', or ' "The cat is on the mat" is true'. We also
remark on occasion, when someone else has said somethmg,
6
Very true' or 'That's true' or 'True enough'.
Most (though not all) of these expressions, and others besides,
I Reprinted from Proceedings 4 the Aristotelian Society, Supplementary
Volume xxiv (1950)~by courtesy of the editor.
It is sufficiently obvious that 'truth' is a substantive. 'true' an adjective and
'of' in 'true of' a preposition.
E
118 TRUTH
certainly do occur naturally enough. But it seems reason-
able to ask whether there is not some use of 'is true' that is
primary, or some generic name for that whch at bottom we
are always saying 'is true'. Which, if any, of these expressions
is to be taken au pied de la leftre? To answer this will not take
us long, nor, perhaps, far: but in phlosophy the foot of the
letter is the foot of the ladder.
I suggest that the following are the primary forms of es-
pression:
It is true (to say) that the cat is on the mat.
That statement (of his, &c.) is true.
The statement that the cat is on the mat is true.
But first for the rival candidates.
(a) Some say that 'truth is primarily a property of beliefs'.
But it may be doubted whether the expression 'a true belief'
is at all common outside philosophy andtheology: and it seems
clear that a man is said to hold a true belief when and in the
sense that he believes (in) something whirl1 is true, or believes
that something which is true is true. Moreover if, as some also
say, a belief is 'of the nature of a picture', then it is of the nature
of what cannot be true, though it may be, for example,
faithful. I
(b) True descriptions and true accounts are simply varieties
. . - .

of true statements or of collections of true statements, as are

true answers and the like. The same applies to propositions too,
in so far as they are genuinely said to be true (and not, as more
commonly, sound, tenable and so on).2 A proposition
- -
in law or
in is something portentous, usually a generalization,
that we are invited to accept and that has to be recommended
I

by argument: it cannot be a direct report on current observa-

tion-if you look and inform me that the cat is on the mat, that

A likeness is true to life, but not true of it. A word picture can be true, just
because it is not a picture.
Predicates applicable also to 'argvments', which we likewise do not say
are true, but, for example, valid.
 TRUTH 119

is not a proposition though it is a statement. In philosophy,

indeed, 'proposition' is sometimes used in a special way for 'the
meaning or sense of a sentence or family of sentences': but
whether we t h d c a lot or little of this usage, a proposition in
t h s sense cannot, at any rate, be what we say is true or false.
For we never say 'The meaning (or sense) of ths sentence (or
of these words) is true': what we do say is what the judge or
jury says, namely that ' The words taken in t h s sense, or if we
assign to them such and such a meaning, or so interpreted or
understood, are true'.
(c) Words and sentences are indeed said to be true, the
former often, the latter rarely. But only in certain senses.
Words as chscussed by phlologists, or by lexicographers,
grammarians, hguists, phoneticians, printers, critics (stylistic
or textual) and so on, are not true or false: they are wrongly
formed, or ambiguous or defective or untranslatable or un-
pronouncable or mis-spelled or archaistic or corrupt or what
n0t.I Sentences in similar contexts are elliptic or involved or
I

ahterative or ungrammatical. W e may, however, genuinely

say 'His closing words were very true' or 'The third sentence
on page 5 of his speech is quite false': but here 'words' and
6
sentence' refer, as is shown by the demonstratives (possessive
pronouns, temporal verbs, definite descriptions, &c.), w h c h
in t h s usage consistently accompany
- -
them, to the words or
sentence as used by a certain person on a certain occasion. That is,
they refer (as does 'Many a true word spoken in jest') to
statements.
A statement is made and its making is an historic event, the
utterance by a certain speaker or writer of certain words (a

Peirce made a beginning by pointing out that there are two (or three)
different senses of the word 'word', and adumbrated a technique ('counting'
words) for deciding what is a 'different sense'. But his two senses are not well
defined, and there are many more-the 'vocable' sense, the philologist's sense
in which 'gramar' is the same word as 'glamour', the textual critic's stnse in
which the 'the' in I. 254 has been written twice, and so on. With all his 66
divisions of signs, Peirce does not, I believe, distinguish between a sentence
and a statement.
120 TRUTH
sentence) to an audience with reference to an hstoric situation,
event or what not.'
A sentence is made up ofwords, a statement is made in words.
A sentence is not English or not good English, a statement is
not in English or not in good English. Statements are made,
words or sentences are used. We talk of my statement, but of
the English sentence (if a sentence is mine, I coined it, but I do
not coin statements). The same sentence is used in making
different statements (I say 'It is mine', you say 'It is mine'): it
may also be used on two occasions or by two persons in making
the same statement, but for this the utterance must be made with
reference to the same situation or event2 We speak of 'the
statement that S,' bur of 'the sentence "S" ', not of 'the sentence
that S'.3
When I say that a statement is what is true, I have no wish
to become wedded to one word. 'Assertion', for example, will
in most contexts do just as well, though perhaps it is slightly
wider. Both words share the weakness of being rather solemn
(much more so than the more general 'what you said' or 'your
words')-though perhaps we are generally being a little solemn
when we cLscuss the truth of anydung. Both have the merit
1 'Historic' does not, of course, mean that we cannot speak of future or
possible statements. A 'certain' speaker need not be any defrnite speaker.
'Utterance' need not be public utterance-the audience may be the speaker
himself.
'The same' does not always mean the same. In fact .it has no meaning in
the way that an 'ordmary' word like 'red' or 'horse' has a meaning: it is a (the
typical) device for establishing and distinguishing the meanings of ordinary
words. Like 'real', it is part of our apparatus in words for f h g and adjusting
the semantics of words.
3 Inverted commas show that the words, though uttered (in writing), are
not to be taken as a statement by the utterer. This covers two possible cases,
(i) where what is to be discussed is the sentence, (ii) where what is to be dis-
cussed is a statement made elsewhen in the words 'quoted'. Only in case (i)
is it correct to say simply that the token is doing duty for the type (and even
here it is quite incorrect to say that 'The cat is on the mat' is the name of an
English sentence-though possibly The C a t is on the Mat might be the title of
a novel, or a bull might be known as Catta est in matta). Only in case (ii) is
there something true or f&e, viz. (not the quotation but) the statement made
in the words quoted.
 TRUTH 121

of clearly referring to the historic use of a sentence by an utterer,

and of being therefore re cis el^ not equivalent to 'sentence'.
For it is a fashionable mistake to take as primary ' (The sentence)
"S" is true (in the English language)'. Here the addition of the
words 'in the EngLsh language' serves to emphasize that
'sentence' is not being used as equivalent to 'statement', so that
it precisely is not what can be true or false (and moreover, 'true
inthe English language' is a solecism, mismodelled presumably,
and with deplorable effect, on expressions llke 'true in geo-
metry').
3. When is a statement true? The temptation is to answer
(at least if we confine ourselves to 'straightforward' statements):
'When it corresponds to the facts'. And as a piece of standard
English this can hardly be wrong. Indeed, I must' confess I do
not really think it is wrong at all: the theory of truth is a series
of truisms. ~ t d it, can at least be misleading.
If there is to be communication of the sort that we acheve
by language at all, there must be a stock of symbols of some
kind w h c h a communicator ('the speaker') can produce 'at
will' and whlch a communicatee ('the audience') can observe:
these may be called the 'words', though, of come, they need
not be anything very like what we should normally call words
-they might be signal flags, &c. There must also be something
other than the words, whch the words are to be used to com-
municate about: this may be called the 'world'. There is no
reason why the world should not include the words, in every
sense except the sense of the actual statement itself w h c h on
any occasion is being made about the world. Further,
the world must exhibit (wemust observe) similarities and dis-
similarities (there could not be the one without the other): if
everything were either absolutely indistinguishable from any-
thing else or completely unlike anything- else, there would be
nothing to say. And finally (for present purposes-f course
there are other conditions to be satisfied too) there must be
two sets of conventions:
Descriptive conventions correlating the words (= sentences)
122 TRUTH
with the types of situation, thing, event, &c., to be found
in the world.
Demonstrative conventions correlating the words (= state-
ments) with the historic situations, &c., to be found in the
world.
A statement is said to be true when the hlstoric state of affairs
to whch it is correlated by the demonstrative conventions (the
one to whch it 'refers') is of a typez with whch the sentence
used in malung it is correlated by the descriptive conventions.3
3a. Troubles arise from the use of the word 'facts' for the
hstoric situations, events, &c., and in general, for the world.
For 'fact' is regularly used in conjunction with 'that' in the
sentences 'The fact is that S' or 'It is a fact that S' and in the
expression 'the fact that S', all of whch imply that it would be
true to say that S.4
1 Both sets of conventions may be included together under 'semantics'.
But they differ greatly.
'Is of a type with which' means 'is sufficiently &e those standard states
of affairs with which'. Thus, for a statement to be true one state of affairs must
f
be like certain others, which is a natural relation, but also su5cient y like-to
merit the same 'description', which is no longer a purely natural relation. To
say 'This is red' is not the same as to say 'This is like those', nor even as to say
'This is like those which were called red'. That things are similar, or even
'exactly' similar, I may literally see, but that they are the same I cannot literally
see-in calling them the same colour a convention is involved additional to the
conventional choice of the name to be given t o the colour which they are
said to be.
3 The trouble is that sentences contain words or verbal devices to serve both
descriptive and demonstrative purposes (not to mention other purposes),
often both at once. In philosophy we mistake the descriptive for the demon-
strative (theory of universals) or the demonstrative for the descriptive (theory
of monads). A sentence as normally distinguished from a mere word or phrase
is characterized by its containing a minimum of verbal demonstrative devices
(Aristotle's 'reference to time'); but many demonstrative conventions are nbn-
verbal (pointing, &c.), and using these we can make a statement in a single
word which is not a 'sentence'. Thus, 'languages' like that of (traffic, &c.) s i p s
use quite distinct media for their descriptive and demonstrative elements (the
sign on the post, the site of the post). And however many verbal demon-
strative devices we use as auxiliaries, there must always be a non-verbal origin
for these con-ordinates, which is the point of utterance of the statement.
(For footnote 4 see p. gr.
 TRUTH 123

T h may lead us to suppose that

(i) 'fact' is only an alternative expression for 'true state-
ment'. We note that when a detective savs 'Let's look d

at the facts' he does not crawl round the carpet, but

proceeds to utter a string of statements: we even talk of
'stating the facts';
-

(ii) for every true statement there exists 'one' and its own
precisely corresponding fact-for every cap the head
it fits.
It is (i) whch leads to some of the mistakes in 'coherence' or
formalist theories; (ii) to some of those in 'correspondence'
theories. Either we suppose that there is nothing there but the
true statement itself, nothing to which it corresponds, or else
I
we populate the world with linguistic Doppelgliitger (and grossly
overpopulate it--every nugget of 'positive' fact overlaid by a
massive concentration of 'negative' facts, every tiny detailed
fact larded with generous general facts, ahd so on).
When a statement is true, there is, of course, a state of affairs
whch makes it true and whch is toto mundo distinct frdm the
true statement about it: but equally of course, we can only
describe that state of affairs in words (either the same or, with
luck, others). I can only describe the situation in which it is
true to say that I am f e e h g sick by saying
-
that it is one in
-

whch I am f e e h g sick (or experiencing sensations of nausea):'

4 I use the following abbreviations:
S for the cat is on the mat.
ST for it is true that the cat is on the mat.
tst for the statement that.
I take tstS as my example throughout and not, say, tst Julius Caesar was bald
or tst all mules are sterile, because these latter are apt in their different ways
to make us overlook the distinction between sentence and statement: we have,
apparently, in the one case a sentence capable of being used to refer to only
one historic situation, in the other a statement without reference to at least
(or to any particular) one.
If space permitted other types of statement (existential, general, hypo-
thetical, &c.) should be dealt with: these raise problems rather of meaning than
of truth, though I feel uneasiness about hypotheticals.
If t h s is what was meant by ' "It is raining" is true if and only if it is
raining', so far so good.
124 TRUTH
yet between stating, however truly, that I am feeling sick and
f e e h g sick there is a great gulf fixed.'
'Fact that' is a phrase designed for use in situations where the
distinction between a true statement and the state of affairs
about which it is a truth is neglected; as it often is with advan-
-

tage in ordinary life, though seldom in phdosophy-above all

in discussing truth, where it is precisely our business to prise
the words off the world and keep them off it. To ask 'Is the fact
that S the true statement that s or that which it is true of?'
may beget absurd answers. To take an analogy: although
we may sensibly ask 'Do we ride the word "elephant" or
the animal?' and equally sensibly 'Do we write the word or the
animal?' it is nonsense-to ask 'Do we define the word or the
animal?' For defining an elephant (supposing we ever do h s )
is a compendious description of an operation involving both
word and animal (do & focus the image or the battleShp?); -

and so speaking about 'the fact that' is a compendious way

of spedung about a situation involvhg both words and
world.2
3 b. 'Corresponds' also gives trouble, because it is commonly
given too restricted or too colourful a meaning, or one which'
in this context it cannot bear. The only essential point is this:
that the correlation between the words (= sentences) and the
type of situation, event, &c., whch is to be such that when a
statement in those words is made with reference to an historic
situation of that type the statement is then true, is absolutely and
purely conventional. W e are absolutely free to appoint any
symbol to describe any type of situation, so far as merely being
true goes. In a small one-spade language tst nuts might be true
in exactly the same circumstances as the statement in English
It takes two to make a truth. Hence (obviously) there can be no criterion
of truth m the sense of some feature detectable in the statement itself whch will
reveal whether it is true or false. Hence, too, a statement cannot 'without
absurdity refer to itself.
'It is true that S' and 'It is a fact that S' are applicable in the same circum-
stances; the cap fits when there is a head it fits. Other words can fd the same
role u 'fact': we say, e.g. 'The situation is that S'.
 TRUTH 125

that the National Liberals are the people's choice.' There is no

need whatsoever for the words used in making a true statement
to t'mirror' in any way, however indirect, any feature what-
soever of the situation or event; a statement no more needs, in
order to be true, to reproduce the 'multiplicity,' say, or the
'structure' or 'form' of the reality, than a word needs to be
echoic or writing pictographc. T o suppose that it does, is to
fall once again into the error of reading- back into the world the
features of language.
The more rudimentary a language, the more, very often, it
will tend to have a 'single' word for a hghly 'complex' type
of situation: t h s has such disadvantages as that the language
becomes elaborate to learn and is incapable of dealing with
situations which are non-standard, unforeseen, for w h c h there
may jist be no word. When we go abroad equipped only with
a phrase-book, we may spend long hours learning by heart-

and so on and so on, yet faced with the situation where we have
the pen of our aunt, find ourselves quite unable to say so. The
characteristics of a more developed language (articulation,
morphology, syntax, abstractions, &c.), do not make state-
ments in it any more capable of being true or capable of being
any more true, they make it more adaptable, more learnable,
more comprehensive, more precise, and so on; and these aims
-

may no doubt be furthered by m h g the langqage (allowance

made for the nature of the medium) 'mirror' in conventional
ways features descried in the world.
Yet even when a language does 'mirror' such features very
closely (and does it ever?) the truth of statements remains still
a matter, as it was with the most rudimentary languages, of
the words used being the ones conventionally appointed for

We could use 'nuts' even now as a code-word: but a code, as a trans-

formation of a language, is distinguished from a language, and a code-word
dispatched is not (called) 'true'.
 126 TRUTH
situations of the type to which that referred to belongs. A
picture, a copy, a replica, a photograph-these are never true in
so far as they are reproductions, produced by natural or mech-
anical means: a reproduction cai be accurate or lifelike (true to
the original), as a gramophone recordmg or a transcription may
be, but not true (of) as a record of proceedings can be. In the
same way a (natural) sign of s o m e t h g can be infallible or
unreliable but only an (artificial) sign far something can be
right or wrong.'
There are many intermediate cases between a true account
and a faithful picture, as here somewhat forcibly contrasted,
and it is from the study of these (a lengthy matter) that we can
get the clearest insight into the contrast. For example, maps:
these may be called pictures, yet they are highly conventiona-
lized pictures. If a map can be clear or accurate or misleading,
like a statement, why can it not be true or exaggerated? HOW' --

do the 'symbols' used in map-making differ from those used in

statement-making? O n the other hand, if an air-mosaic is not
a map, why is it not? And when does a map become a diagram?
These are the really illuminating questions.
4. Some have said that-
T o say that an assertion is true is not to make any further
assertion at all.
In all sentences of the form ' p is true' the phrase 'is true' is
logically superfluous.
T o say that a proposition is true is just to assert it, and to say
that it is false is just to assert its contradictory.
But wrongly. T S ~ S(except in paradoxical cases of forced
and dubious manufacture) refers to the world or any part of
it exclusive of tst S, i.e. of itself? TstST refers to the world or
Berkeley confuses these two. There will not be books in the running
brooks until the dawn of hydro-semantics.
A statement may refer to 'itself' in the sense, for example, of the sentence
used or the utterance uttered in making it ('statement' is not exempt from all
ambiguity). But paradox does result if a statement purports to refer to itself
in a more full-blooded sense, purports, that is, to state that it itself is true, or to
state what it itselfrefers to ('This statement is about Cato').
 v

TRUTH
any part of it inclusive of tstS, though once again exclusive of
itself, i.e. of tstST. That is, tstST refers to somedung to which
tstS cannot refer. TstST does not, certainly, include any state-
ment referring to the world exclusive of tstS whch is not in-
cluded already in tstS-more, it seems doubtful whether it
does include that statement about the world exclusive of tstS
whch is made when we state that S. (If I state that tstS is true,
should we really agree that I have stated that S? Only 'by
implication'.)' But all ths does not go any way to show that
tstST is not a statement different from tstS. If Mr. Q writes on
a notice-board 'Mr. W is a burglar', then a trial is held to decide
whether Mr. Q's published statement that Mr. W is a burglar
is a libel: finding 'Mr. Q's statement was true (in substance and
in fact)'. Thereupon a second trial is held, to decide whether
Mr. W is a burglar, in whlch Mr. Q's statement is no longer
under consideration: verdict 'Mr. W is a burglar'. It is an
arduous business to hold a second trial: why is it done if the
verdict is the same as the previous finding?2
What is felt is that the evidence considered in arriving at the
one verdict is the same as that considered in arriving at thE other.
Ths is not strictly correct. It is more nearly correct that when-
ever tstS is true then tstST is also true and conversely, and that
whenever tstS is false tstST is also false and conversely.3 And
it is argued that the words 'is true' are logically superfluous
because it is believed that if any two statements are
always true together and always false together then they must
mean the same. Now whether t h s is in general a sound view
may be doubted: but even if it is, why shoild it not break down
in the case of so obviously 'peculiar' a phrase as 'is true'?
Mistakes in pMosophy notoriously arise through thinlung that
And 'by implication' tstST asserts something about the making of a state-
ment which tstS certainly does not assert.
Thls is not quite fair: there are many legal and personal reasons for holding
two tri Is-which, however, do not affect the point that the issue being tried
is not t i t e same.
3 Not quite correct, because tstST is only in place at all when tstS is en-
visaged as made and has been verified.
128 TRUTH
what holds of 'ordinary' words like 'red' or 'growls' must also
hold of extraordinary words hke 'real' or 'exists'. But that
'true' is just such ano;her extraordinary word is obvious.'
There is something peculiar about the 'fact' whch is de-
scribed by tstST, something whch may make us hesitate to
call it a 'fact' at all; namely, that the relation between tstS and
the world whch tstST asserts to obtain is a purely conventional
relation (one whch ' t h h g makes so'). For we are aware
that thu; relation is onq whch we could alter at will, whereas
we like to restrict, the word 'fact' to hard facts, facts which are
natural and unalterable, or anyhow not alterable at will. Thus,
to take an analogous case, we may not hke c a h g it a fact that
the word elephant means what it does, though we can be
induced to call it a (soft) fact-and though, of course, we have
no hesitation in calling it a fact that contemporary English
speakers use the word as they do.
An important point about tlus view is that it confuses falsity
with negation: for according to it, it is the same thing to say
'He is not at home' as to say 'It is false that he is at home'. (But
what if no one has said that he is at home? What if he is lying
upstairs d e a ? ) Too many philosophers maintain, when anxious
to explain away negation, that a negation is just a second order
affirmation (to the effect that a certain first order affirmation is
fdse), yet, when anxious to explain away falsity, maintain that
to assert that a statement is false is just to assert its negation
(contra&ctory). It is impossible to deal with so fundamental
a matter here? Let me assert the following merely. Affirmation
Unum, venrm, bonum-the old favourites deserve their celebrity, There
is something odd about each of them. Theoretical theology is a form of
onomatohtry.
The following two sets of logical axioms are, as Aristotle (though not his
successors) makes them, quite distinct:
(a) No statement can be both true and false.
No statement can beneither true nor false.
(b) Of two contradictory statements-
Both cannot be true.
Both cannot be false.
The second set demands a definition of contradictories, and is usuallyjoined
 TRUTH 129

and negation are exactly on a level, in thrs sense, that no lan-

guage can exist whlch does not contain conventions for both
&d that both refer to the world equally &rectly, not to state-
ments about the world: whereas a language can quite well exist
without any device to do the work of 'true' and 'ralse'. Any
satisfactory theory of truth must be able to cope equally with
falsity:' but 'is false' can only be maintained to be logically
superfluous by making thrs fundamental confusion.
5. There is another way of coming to see that the phrase 'is
true' is not logically superfluous, and to appreciate what sort
of a statement it is to say that a certain statement is true. There
are numerous other adjectives w h c h are in the same class as
'true' and 'false', w h c h are concerned, that is, with the relations
between the words (as uttered with reference to an historic
situation) and the world, and which nevertheless no one would
&smiss as logically superfluous. W e say, for example, that a
certain statement is exaggerated or vague or bald, a description
somewhat rough or misleading or not very good, an account
rather general or too concise. In c&s hke these it is pointless
to insist on deciding in simple terms whether the statement is
with an unconscious postulate that for every statement there is one and only
one other statement such that the pair are contradictories. It is doubtful how
fir any language does or must contain contradictories, however defined, such
u to satisfy both this postulate and the set of axioms (b).
Those of the so-called 'logical paradoxes' (hardly a genuine class) which
concern 'true' and 'false' are not to be reduced to cases of self-contradiction,
any more than 'S but I do'not believe it' is. A statement to the effect that it is
itself true is every bit as absurd as one to the effect that it is itself false. There
are other types of sentence which offend against the hdamental conditions
of all communication in ways distinctfrom the way in which 'This is red and
is not red' 0ffends-e.g. 'This does (I do) not exist', or equally absurd 'This
exists (I exist)'. There are more deadly sins than one; nor does the way to
salvation lie through any hierarchy.
T o be false is (not, of course, to correspond to a non-.fact, but) to mis-
correspond with a fact. Some have not seen how, then, since the statement
which is false does not describe the fact with which it mis-corresponds (but
mis-describes it), we know which kct to compare it with: this was because they
thought of all linguistic conventions as descriptive-but it is the demonstrative
conventions which fix which situation it is to which the statement refers. N o
statement can state what it itself refers to.
130 TRUTH
'true or false'. Is it true or false that Belfast is north of London?
That the galaxy is the shape of a fried egg? That Beethoven
was a drunkard? That Wellington won the battle of Waterloo?
There are various degrees and dimensions of success in makmg
statements: the statements fit the facts always more or less
loosely, in. different ways on different occasions for different
intents and purposes. w h a t may score full marks in a general
knowledge test may in other circumstances get a gamma. And
even the most adroit of languages may fail to 'work' in an
abnormal situation or to cope, or cope reasonably simply, with
novel iscoveries: is it true or false that the dog goes -
round
the cow ?I What, moreover, of the large class of cases where a,
statement is not so much false (or true) as out of place, inept
('All the signs of bread' said when the bread is before us) ?
We become obsessed with 'truth' when discussing state-
ments, just as we become obsessed with 'freedom' when dis-
cussing conduct. So long as we t M that what has always and
alone to be decided is whether a certain action was done freely
or was not, we get,nowhere:
-
but so soon as we turn instead
to the numerous other adverbs used in the same comexion
('accidentally', 'unwillingly', 'inadvertently', &c.) , things be-
come easier, and we come to see that no concluding inference
of the form 'Ergo, it was done freely (or not freely)' is required.
Llke freedom, truth is a bare minimum or an illusory ideal (the
truth, the whole truth and n o h g but the truth about, say, the
battle of Waterloo or the Primavera).
6 . Not merely is it jejune to suppose that all a statement
aims to be is 'true', but it may further be questioned whether
Here there is much sense in 'coherence' (and pragmatist) theories of truth,
despite their failure to appreciate the trite but central point that truth is a
matter of the relation between words and world, and despite their wrong-
headed Gleichschaltung ofall varieties of statemental failure under the lone head
of 'partly true' (thereafter wrongly equated with 'part of the truth'). 'Corre-
spondence' theorists too often talk as one would who held that every map is
either accurate or inaccurate; that accuracy is a single and the sole virtue of a
map; that every country can have but one accurate map; that a map on a
larger scale or showing different features must be a map of a different country;
and so on.
 TRUTH 131

every 'statement' does aim to be true at all. The principle of

Logic, that 'Every proposition must be true or false', has too
long operated as the sLnplest, most persuasive and most per-
vasive form of the descriptive fallacy. Phdosophers under its
influence have forcibly interpreted all 'propositions' on the
model of the statement that a certain t h g is red, as made when
the dung concerned is currently under observation.
Recently, it has come to be realized that many utterances
which have been taken to be statements (merely because they
are not, on grounds of grammatical form, to be classed as
commands, questions, &c.) are not in fact descriptive, nor
susceptible of being true or false. When is a statement not a
statement? When it is a formula in a calculus: when it is a
performatory utterance: when it is a value-judgement: when it
is a definition: when it is part of a work of fiction-there are
-

many such suggested answers. It is simply not the business ofsuch

utterances to 'correspond to the facts' (and even genuine state-
ments have other businesses besides that of so corresponding).
It is a matter for decision how far we should continue to call
6
such masqueraders statements' at all, and how widely we
should be prepared to extend the uses of 'true' and 'false' in
C
different senses'. My own feeling is that it is better, when once
a masquerader has been unmasked, not to call it a statement and
-

not to say it is true or false. In ordinary life we should not call

most of them statements at all, though phlosophers and
grammarians may have come to do so (or rather, have lumped
them all together under the term ofart 'proposition'). We make
a difference between 'You said you promised' and 'You stated
that you promised': the former can mean that you said 'I
promise', whereas the latter must mean that you said 'I
*

promised': the latter, which we say you 'stated', is somethng

which is true or false, whereas for the former, which is not true
or false, we use the wider verb to 'say'. Similarly, there is a
difference between 'You say this is (call this) a good picture'
and 'You state that t h s is a good picture'. Moreover, it was only
so long as the real nature of arithmetical formulae, say, or of
132 TRUTH I

geometrical axioms renlained unrecognized, and they were

thought to record information about the world, that it was
-

reasonable to call them 'true' (and perhaps even 'statements'-

though were they ever so called?): but, once their nature has
been recognized, we no longer feel tempted to call them 'true'
or to dispute about their truth or fdsity.
In the cases so far considered the model 'Ths is red' breaks
down because the 'statements' assimilated to it are not of
a nature to correspond to facts at all-the words are not de-
scriptive words, and so on. But there is also another type of
case where the words are descriptive words and the 'proposi-
tion' does in a way have to correspond to facts, but precisely
not in the way that 'Ths is red' and similar statements setting
up to be true have to do.
In the human predicament, for use in which our language is
designed, we may wish to speak about states of affairswhich have
not been observed or are not currently under observation (the
future, for example). And although we can state a n y t h g 'as a
fact' (which statement will then be true or false1) we need not
do so: we need only say 'The cat may be on the mat'. This
utterance is auite different from tstS-it is not a statement at
1

all (it is not true or false; it is compatible with 'The cat may not
be on the mat'). In the same way, the situation in which we
discuss whether and state that tstS is true is different from the
situation in whch we discuss whether it is probable that S. Tst
it is probable that S is out of place, inept, in the situation where
we can make tstST, and, I think, conversely. It is not our business
here to discuss probability: but is worth observing that the
phrases 'It is true that' and 'It is probable that' are in the same
line of b~siness,~ and in so far incompatibles.
7. In a recent article in Analysis Mr. Strawson has pro-
pounded a view of truth whch it wdl be clear I do not accept.

Though it is not yet in place to call it either. For the same reason, one
cannot lie or tell the truth about the future.
Compare the odd behaviours of 'was' and 'will be' when attached to
'true' and to 'probable'.
 TRUTH 133

He rejects the 'semantic' account of truth on the perfectly

correct ground that the phrase 'is true' is not used in t a b g
about sentences, supporting this with an ingenious hypothesis
as to how meaning may have come to be confused with truth:
-

but h s will not suffice to show what he wants-that 'is true'

is not used in talking about (or that 'truth is not a property o f )
anything. For it is used in taking about statements (which in his
article he does not distinguish clearly from sentences). Further,
he supports the 'logical superfluity' view to t h s extent, that he
agrees that to say that ST is not to make any further assertion
at all, beyond the assertion that S: but he disagrees with it in
so far as he t M t s that to say that ST is to do somethmg more
than just to assert that S-it is namely to confirm or to grant (or
sometlung of that kind) the assertion, made or taken as made
already, that S. It will be clear that and why I do not accept
the first part of h s : but what of the second part? I agree that
to say that ST 'is' very ofteri, and according to the all-important
linguistic occasion, to confirm tstS or to grant it or what not;
but this cannot show that to say that ST is not also and at the
same time to make an assertion about tst S. To say that I believe
you 'is' on occasion to accept your statement; but it is also to
make an assertion, whch is not made by the strialy performa-
tory utterance 'I accept your statement'. It is common for quite
ordinary statements t i have a performatory 'aspect': to saythat
you are a cuckold may be to insult you, but it is also and at the
same time to make a statement which is true or fdse. Mr. Straw-
son, moreover, seems to confine himself to the case where I say
'Your statement is true' or s o m e h g similar-but what of the
case where you state that S and I say n o t h g but 'look and see' that
your statement is true ?Idonot see how this criticalcase, to which
nothng analogous occurs with strictly performatory utter-
ances, could be made to respond to Mr. Strawson's treatment.
One final point: if it is admitted (if)that the rather boring yet
satisfactory relation between words and world which has here
been discussed does genuinely occur, why should the phrase 'is
true' not be our way ofdescribing it?And if it is not, what else is?
 HOW TO TALK1

S O M E SIMPLE WAYS

C A Nto describe X as Y really be the same as to call X Y ? Or

again the same as to state that X is Y ? Have we, in using such
a variety of terms for simple speech-acts, any clear and serious
hstinctions in mind? The presumption must surely be that we
have: and what follows is an attempt to isolate and schematize
some of them. But it is not contended that it contains an exact
or full or final account of olir ordinary uses of any one of the
terms for speech-acts discussed. For one thing, this is a mere
essay at one section of what must be a very large theme; for
another, essential though it is as a preliminary to track down
the detail of our ordinary uses of words, it seems that we shall
in the end always be compelled to straighten them-out to some
extent.
W e shall consider a simplified model of a situation in which
we use language for t a h n g about the world. This model we
shall call by the name 'Speech-situation So9.
Possibly we never are actually in a situation exactly k e So:
more probably we sometimes are so, or, more strictly, regard
ourselves for current intents and purposes as being so. But the
purpose of considering the model is to elucidate some of our
-

orcGnary thought andVlanguageabout the uses of speech: and

it seems hardly deniable that such thought and language we
do, for better or worse and whether consciously or uncon-
sciously, make use of such models (not, of course, necessarily
only one such).
Reprinted from Proceedings of the Aristotelian Society, 1952-3, by courtesy
of tbe editor.
 HOW TO TALK I35
The world, then, in So will consist of numerous individual
items, each of one and only one definite type. Each type is totally
and equally different from every other type: each item is totally
and equally distinct from every other item. Numerous items
may be of the same type, but no item is of more than one type.
Item and type are (to speak with necessary roughness) appre-
hended by inspection merely. (Roughly, the world might
consist of an orderless plurality of amorphous colour-patches,
each of either the same pure red, or the same pure blue, or the
L I

same pure yellow. Then would not they be alike in being

coloured, and possibly in other further general features? This
must be ruled out-perhaps by the consideration that in these
other respects every item in our world is identical with every
other, so that nothing can besaidabout them: or perhaps by altera-
tions and refinements--every item is either a colour-~atchof the
same pure red, or a noise of the same definite pitch, intensity,
&c., or a smell, &c. : but anyway, by the ruling that our language
is not going to be equipped to deal with any such further
features.)
The language in So will permit of the utterance only of
sentences of one form, form S:

Besides the expression 'is a', which is used invariably in every

sentence in the position shown above, our language may con-
tain an indefinite number of other vocables to be inserted in the
place of the 'I' or the 'T' in form S. Assuming that the conven-
tions next to be mentioned have been established, each of these
vocables will be either an I-word or a T-word in the language:
and any utterance consisting of an I-word followed by 'is a'
followed by a T-word will be a sentence in the language. - -
N o t h g else w d be a sentence.
In order for t h s language to be used for taking about this
world, two sets of (semantic) conventions will be needed.
I-conventions, or conventions of reference, are nceded in order
to fix whlch item it is that the vocable w h c h is to be an I-word
136 HOW TO TALK
is to refer to on each (and in our simple case, on every) occasion
of the uttering (assertive) of a sentence containing it: we shall
not concern ourselves here with the nature or genesis of these
conventions, but simply take it that each item has had allotted
to it its own I-word by which it is uniquely referred to, and each
I-word similarly its own item. For I-words we shall use in what
follows numerals, for example, '1227'~and we shall speak of
them in use not as '(proper) names', of which they are at most
only a primitive variety, but as 'references'. T-conventions,
or conventions of sense, are needed in order to associate the
vocables whch are to be T-words with the item-types, one
to one: these conventions we may inaugurate by one or other
of two procedures of hguistic legislation, viz. a

I. Name-giving.
2. Sense-giving .

Name-giving ('naming' in one ordinary sense, but not, for

example, in the sense of 'giving the name of' or of 'putting a
name to') consists in allotting -
a certain vocable to a certain
item-type as its 'name'. Sense-giving ('defining' in the sense of
6
ostensive definition', here in a simpMed world) consists in
allotting a certain item-type to a certain vocable as its 'sense'.
These two procedures, at least in our simplified situation, pro-
duce the same upshot: when either has been gone through, the
item-type, attached by nature to certain items, is attached by
convention to a certain vocable, now a T-word and (as we
shall call it) its 'name', as the 'sense' of that word.1

1 The difference between name-giving and sense-giving is important In

some connexions, though not here. If, not happily, we were to use, as we shall
not, sentences of our form S for inaugurating T-conventions, there would
be between '1227 is a rhombus' (name-giving) and '1227 is a rhombus' (sense-
giving) a difference in direction of allotment similar to that difference in direc-
tion of fit which is to be mentioned shortly. 1 say 'not happily' because, for
example, if our interest were in linguistic legislation, we had better go back
behind 'proper names' to 'that' and 'this'. But we are not here concerned with
how these conventions are established, nor even to assert, for example, that the
two types could be established independently of one another. All this is mere
preliminary.
 H O W T O TALK I37

Every word in our language So (except for 'is' and 'a') has
either a reference fixed by I-conventions or a sense fixed
by T-conventions, but not both, and is accorrGngly either an
I-word or a T-word.
W e shall not go into the 'metaphysical status' of types and
senses (nor of items). If we went back to the rudiments of speech
theory, both might appear as 'constructions'. Nevertheless, to
talk of types and senses,l and, as we shall, of n l a t c h g the one
to the other, is not necessarily inexpedient in all contexts: and
in particular it is expedient in our present context, where we
are engaged to elucidate some of our ordmary language about
speech-acts, since such ordinary language does embody a
model l k e So. Conceive of our items here as, say, a number of
samples or specimens of colours, or of (geometrical) shapes, each
with a reference-numeral allotted to it: conceive of our senses
as a number of standards or patterns of colours, or of (geo-
metrical) shapes, each with a name allotted to it: thmk of name-
giving or sense-giving as involving the selection $ a sample or
specimen as a standard pattern. Ths is not so far from the truth.
- Let us now take the stage of linguistic legislation as over.
-

We proceed to use our language in accordance with that legis-

lation to talk about the world. Then a satisfactory utterance
(assertive) on any particular occasion will be one where the
&emreferred to by the I-word in accordance with the conven-
tions of reference is of a (in So, the) type which matches
the sense which is attached by the conventions of sense to the
T-word. For the utterance to be satisfactory, we require the
presence of
both a conventional link beween I-word and item, and
another between T-word and sense,
and a natural link ( m a t ~ hbetween
)~ type and sense.
W e shall call the expression 'is a' as uttered in the uttering
Or ?: 'types' and 'senses'. 'Talk of' gives trouble with inverted commas,
for reasons which can be understood.
a Here, in So, taken to be purely natural-apprehended by inspection without
admixture of convention. This does not obtain in more complicated situations.
138 HOW T O TALK
(assertive, of course: I shall omit t h s necessary quaLfication in
future) of a sentence of form S by the name 'assertive link', and
any utterance of form S an 'assertion'. Then the assertion is
justified not merely by convention, nor merely by nature, but
in a complex or roundabout way. Diagrammatically:
'I 227' 'is a' 'rhombus'
(I-word) (assertive link) (T'lword)

item/type natural link sense

(sample) (match) (panern)

All t h s is, I hope, simple-and, it may be again emphasized,

highly simplified. It is now time to get out of the way, as a
preliminary, two very primitive ways in which I may on any
occasion go wrong in my utterance. I may be guilty of
I. Misnaming.
2. Misreferring.
In both of these cases I match sample to pattern or pattern, to
sample it may be quite unexceptionably, but
I. I quote (give) the name wrongly.
2. I quote (give) the reference wrongly.
In either case alike I mislead, or more strictly tend pro tanto to
mislead (not myself, at least at the time-if indeed that makes
sense, but) my hearers who know the language. Misleading,
at least :ISI use the term here, goes, it should be noted, to the
meaning of the utterance, not to the facts: whether or not I
create additionally in those hearers, or more strictly tend to
create in them, a misapprehension as to the facts depends
 HOW TO TALK
additionally on whether or not I have been correct in match-
ing sample to pattem (or conversely), a quite distinct considera-
tion. I mislead (as to meaning) when, through my use of the
wrong I-word or T-word, my hearers are caused, in assessing
or relying on the justifiability of my assertion, to advert to a
different sample or a different pattern from that whlch I, in
malung the assertion, was adverting to.'
Misnaming and misreferring allke may be either aberrational
or idiosyncratir. Aberrational misnaming or misreferring is a
sin against my accepted linguistic legislation: my sample list
is rightly numbered, my pattern stock is rightly -labellid, but
through aberration I quote or give the number or the name
wrongly. Idiosyncratic misnaming or misreference is due to a
fault in my accepted hguistic legislation: though I quote or
give the number or the name rightly, nly sample list is wrongly
numbered or my pattem stock is wrongly labelled?
Whenever in So I utter an assertion, I am eo ipso refening and
also naming (using these terms, as I shall here only do, in senses
analogous to misnaming and misreferring
- - as explained above).
But in contrast with other varieties of speech-act, to be discussed
next, which in uttering an assertion we may be said to be per-
forming, 'referring' and 'naming' are terms for only parts, and
we may say ancillary parts, of my performance on any occasion.
By contrast, when we say, for example, that in uttering- the
assertion '122~ is a rhombus' I am identifying 1227as a rhombus
or stating that 1227is a rhombus, then the whole issuing of the
utterance is the m h g of the identlfcation or the malung of
the statement, and the whole utterance is (in my use-not, of

I 'Misleading' is a speech-act of a totally distinct class from those speech-acts

with which this paper is concerned. There are, of course, many such distinct
classes.
Clearly, in combination an aberration and an idiosyncrasy may cancel
each other out, may not 'tend to mislead' on that occasion. Just as, similarly,
a combination of misnaming with misreferring, or of either or both with
mismatching, may not on a particular occasion tend to create a misapprehen-
sion. It is characteristic of ordinary language that it should not (bother to)
have simple names for complex faults, such as these are.
140 HOW TO TALK
course, in every use) the iderirification or the ~tatement.~
You
are guilty of misreferring in using, in malung your assertion,
the tvord '1227', or of misnaming in using the word 'rhombus':
but you are guilty of makmg a misidentification or of making
a misstatement in using the sentence '1227 is a rhombus'. The
issuing of the whole utterance cannot be an act of misnaming
or of misreferring, nor, sirmlarly, of naming (in my usage) or
,- P .

Let us henceforward take it, not merely that the stage of

linguistic legislation is over, but also that we are not, in issuing
our utterances, g d t y of sins against, or of sins through faults
in, our accepted linguistic legislation.
There now arise four distinct uses to whch we may put our
sentence '1227 is a rhombus', four distinct speech-acts which,
in uttering it as an assertion, we may be said to be performing
--four species, if you like, ofthe generic speech-act of asserting.
These will be called:
c-identifying, cap-fitting or placing;
b-identifying, bill-filling or casting;
stating;
instancing.
How does thls complexity arise? For let it be repeated that in
none of these performances are we to be taken as in any way
legislating, but only as performing in accordance with the
terms and purposes of an accepted legislation. And furthermore
we are st;~ excluding a coAplicahon of great importance,
1 Sometimes we use 'the identification', like 'the description' and unlike
'the statement', for, we may say, a part of the utterance: but I am using it only
for the whole utterance, so as to assimilate identieing, as it should be assimi-
lated, to stating and not to naming (irl'rny use). Even if 'the identification' can
be used, as is 'the name', for a part of the utterance, still 'identify-ing' is not a
name for a part of my performance in issuing the utterance (as 'naming', in my
use, always is), but for the whole of it. That part of our utterance is a name
or a reference says nothing to prejudge the type of assertive speech-act to
which our whole performance in issuing the utterance belongs: but that part
of it is an identification or a description teUs us precisely that (in, of course,
our. simplified speech-si tua tions).
 HOW TO TALK 141

which will be introduced only later, namely, that our

vocabulary may be 'inadequate' to the variety presented by the
world we are to talk about: we are s t d taking our legislation
as adequate in the sense that every item in the world is of one
type o d y , whch matches peciseiy the sense of one name only. -

~ o m ~ l e k i tarises,
y nevertheless, owing to the complexity,
whch may escape notice, of the notions of 'fitting' and 'match-
ing'.
We have already noticed in passing, in the case of name-
giving and sense-giving, the distinction in point of direction
between allotting an X to a Y and allotting a Y to an X. In a
similar way, when we operate in accordance with such legisla-
tion, thereis a difference-in direction offit between fitting a name
to an item (or an item with the name) and fitting an item to
a name (or a name with the item). These differ as fitting a nut
with a bolt differs from fitting a bolt with a nut. W e may be
'given' a name, and purport to produce an item of a type w h c h
matches (or is matched by) - the sense of that name: this pro-
~

duction we declare by uttering an assertion of form S with the

given name as T-word and the reference of the produced item
as I-word. T o utter an assertion in h s way is to fit an item to a
name. Conversely, we may be 'given' an item, and purport to
produce a name with a sense whch matches (or is matched by)
the type of that item: t h s production we declare by uttering an
assertion of form S with the reference of the given -
item as
I-word and the produced name as T-word. T o utter an asser-
tion in t b way is to fit a name to an item.
But there is also another distinction to be drawn. W e fit the
name to the item or the item to the name on the ground that -

the type of the item and the sense of the name match. But in
matching X and Y, there is a distinction between matching
X to Y and matching- Y to X, whch may be called a distinction
in point of onus of match. We are apt to overlook thls w ~ t hthe
verb 'match" (especially where it is being taken to mean
I If X matches Y, Y matches X: just as, if X fits Y, Y fits X. But if I match
X to Y, I do not match Y to X, any more than, if I fit X to Y, I fit Y to X.
142 HOW TO TALK
'match exncrly'): but if we consider the analogous word
'assimilate', the distinction between assimilating X to Y, where
the onus of assimilability is on X, and assimilating Y to X,
where the. onus of assimilability is on Y, is clear enough. W e
go wrong in assimilating because we are mistaken about or
misrepresent the nature of the member, X in the first case and
Y in the second, on which the onus of assimilability rests.
When we ask whether we should assimilate X to Y, the ques-
tion is loherher X has the qualities Y has: a simile, 'A is like B',
is a bad simile not because B has not the features which A has
or has features whch A has not, but because A has not the
features which B has or has features which B has not.
These two distinctions generate our four different perform-
ances in uttering '1227 is a rhombus', which are, in the form
of a diagram:
c-iden tifying stating
(placing
'Nfits I' 'I fits N'

(casting)
'Nfits I' '1 fits N'
T o explain first the choice of terms. W e use the useful word
I

'identify', understandably enough, in two opposite ways: we

may speak of 'identifying it (as a daplmia)' when you hand it
to me and ask me if I can identify it, and I say that it is a
daphnia: but we also speak of 'identifying a daphnia' (or
'identifying the daphnia') when you hand me a slide and ask
me if I call identify a daphnia (or the daphnia) in it. In the first
case we are finding a cap to fit. a given object: hence the name
 HOW TO TALK I43

'cap-fitting' or 'c-identifying'. W e are trying to 'place' it. But

in the second case we are trying to find an object to fill a given
b d : hence the name 'b-identifying' or 'bill-filling'. W e 'cast'
t h s thing as the daphnia.' The terms 'stating' and 'instancing'
should need no explanation: to instance is to cite I as an
instance of T.
In the diagram the connecting lines, horizontal and vertical,
indicate the wav in whch the members of each pair connected
J

are similar to each other, as follows:

The horizontal line indicates that the direction offit is the same.
In both placing and stating we are fitting names to given
items, in both instancing and casting we are fitting items
to given names. In the verbalizations given in the diagram,
that which is being Gtted to is shown by italics, in contrast
to that which is being fitted with, which is not in italics.
The vertical line indicates that the onus of match is the same.
In both placing and instancing the type of the item is taken
for and the cluestion mightbe whether the sense
of the T-word is such as really to match it: in both stating
and casting - the sense of the T-word is taken for granted,
and the question
-
might
-
be whether the type of ;he item
is really such as to match it. In the verbalizations given in
the diagram, that, name or item, on the sense or type of
which lies the onus of match is put as the subject.
O r again:
T o glace we have to find a pattern to inatch to this sample.
- -

T o state we have to find a pattern to match this sample to.

T o instance we have to find; sample to match this pattern to.
T o cast we have to find a sample to match to this pattern.
Or again, the differences and likenesses between the four
I Contrast the questions: (a) What (part of speech) is the word underlined
in the following sentence: He was going downhill? (b) Which is the adverb
in the following sentencc: He was going downhill? Answers: ( a ) An adverb
(cap-fitting); (6) 'Downhill' (bill-fdling). W e might even christen c-identi-
fying 'what-identifying' and b-identifying 'which-identifying'.
I44 HOW TO TALK
performances may be brought out by considering wherein, if
any one of them is faultily executed, the fault lies:
TO misidentify (= misplace) shows that we have gone wrong
in our matching through failure to appreciate, to keep clearly
before ourselves for the purpose of this matching, the sense of
the name (T-word). 'I see now, I was wrong to identify it as
-

magenta: magenta, of course, on reflection, is not what this is'.

The mistake is due, we might say, using a word more obviously
appropriate in more complicated situations than So, to his-
conception' of the sense.
Misidentifying must be carefully distinguished from what
we have called 'misnaming'. There, the name is 'wrong' even
though, and whether or not, the sense, wrongly allotted to it,
does match the type of the item: whereas here the name is
b
wrong' because thesense, rightly allotted to it, does not match
the type of the item. If I have misnamed, I should not have said
it was a 'rhombus': if I have misidentified, I should not have
said it was a rhombus. (Ambiguity of 'said'.)
-

It may be asked, Does not misidentifying remain, neverthe-

less, itself a (merely) 'hguistic' mistake? Ths is reminiscent
of the argument that misstatement is impossible, which so long -
entangled the Greeks, and is perhaps Lard to answer because
'hguistic' is vague. The basic-point, never to be surrendered,
is that nlistakes in matchmg are possible, do occur, and that
they may be due to faulty grasp of either of the two elements
being matched. Just as we can (do) advert to the same item and
yet match its type to differing patterns (misperception), so we
can advert to the same sense and match it to differing types of
items. If it is indeed hard to imagine our malung such a mistake
in such a simple situation as So,perhaps after all it is equally hard
to imagine, here, misperception and conscquent misstatement.
(Yet in one way our model of patterns and samples may be
helpful; for it suggests that agreement as to the 'sense' of some
name, a term we have admitted to be not ultimate in speech I

theory, is in the last resort established by agreement upon the

items the types of w h c h are to be standards, leaving those types
 HOW TO TALK I45
themselves stdl to be appraised by perception and so liable to
the, admittedly possible, errors of perception.)
- -
T o misinstance likewise reveals n~isconceptionof the sense
of the name. It is to be distinguished from misreferring: there,
the reference is 'wrong' even though the sense does match the
-

type, whereas here the reference is 'wrong' because the sense

does not match the type. I should not have said '1227' was one:
or, I should not have said 1227 was one.
T o misstate shows that we do not correctly appraise the type
of the item: it is due, we may say in our simple situation, to
misperceiving the sample.
T o mitidentijy (= miscast) hkewise reveals misperception of
the type of the item.
This brings out the sirmkrity between placing and instancing
and again between stating and casting. In another way, we
might bring out the s i d r i t y between placing and stating on
the one hand and between instancing and casting on the other:
to a misplacing or a misstatement we respond with 'But 1227
isn't a rhombus', but to a misinstance or a miscasting we respond
with 'But 1227 isn't a rhombus'. (Needless to say, we canno;
so respond in So, where we are not-equipped with negation: to
introduce negation alters the situation-though of course I have
no wish to suggest that negation is 'posterior' to affirmation,
or that assimilating even 'Lakes sensP without contrasting.)
Finally, one further and perhaps less clear way of bringing
out the contrast between placing and stating, and hkewise
between instancing and casting. In both placing and stating we
fit the name to the item, but in placing the interest is in lm.king
the name to the type via the sense, whereas in stating the interest
is in l i h g the sense to the item via the type:
- - it is primarily, we
may say, the type that we identify' but ~ r i r n a r i;he
l ~ item about
whch we state. Similarly, in both instancing and casting we
It should be unnecessary to point out that I am never using 'identify' here
in the sense o f identifying an item as the same item again, which is a feat
demanding a speech-situation far more complex than So. Moreover, in a
situation where, say, a single item may have more than one feature, the sense
of 'identify' wdl suffer a sea-change.
146 H O W TO TALK
fit the iten1 to the name, but it is ~rimarilythe type that we
instance and primarily the item that we cast.
It is now time to inject a first dose of complexity into our
model of the speech-situation. W e shall now suppose that there
occur in the world to be talked about items of types whch do
not exactly match any of the patterns in our stock (the sense
of any of our names), though they may be more or less similar
to one or to more than one of those patterns. This new model
situation will be called by the name 'Speech-situation S,'.
It is to be noted that each item in the world is still being
assumed to be of one type only-or to possess, we may say, one
feature only, or to be assessable in one dimension only. For
example, if our original patterns are colours, they may be a
(certain shade of) red, a (certain shade of) blue and a (certain
shade of) yellow. Then in So,every item whch occurs will be
either a red (of the pattern shade) or a blue (ofthe pattern shade)
or a yellow (of the pattern shade): but in S,, items may occur
of any colour-they may be (what we should ordmarily call)
white, and so resemble none of the patterns, or none more
than another, or (what we should orhnarily call) pink,
and so resemble one of our patterns appreciably and none
of the others at all, or (what we should ordinarily call)
purple, and so resemble two of our patterns equally but
the rest not at all, or (what we should ordinarily call) crim-
son and so resemble one of our patterns most closely but
another appreciably and the rest not at all, and so on. But
in S, none of the items must be conceived as having, for
example, shape or size to be talked about as well as colour,
though they might have, for example, shape but no colour. It
is only in some further model of the speech-situation, to be
called say 'S,' but not here discussed, that we might introduce
the complication that the same item may possess more than one
feature or be of more than one type o r b e assessable in more
than one dimension.
It is obvious that the 'complication' here introduced in S,
is inevitable in most actual speech-situations. The actual world
 HOW T O TALK 147

is, to all human intents and purposes, indefinitely various; but

we cannot handle an indefinitely large vocabulary; nor,
generally s p e a h g , do we wish to insist on the minutest detect-
able differences, but rather on relative similarities; nor, with
our limited experience both as individuals and as a race, can
we anticipate in our vocabulary vagaries of nature which have
yet to be revealed.
Faced, then, with some such item in S,, the type of which
does not exactly match the sense of any one of the names in
our stock, what courses are open to us? W e can of course (as
with the resources in speech-acts available in So we must, to be
correct) say nodung-in default of some fresh legislation : and
this we shall still do if the type of the recalcitrant item resembles
none of our patterns at all, or none more than another. More
generally, however, the type of the item resembling the sense
of some one of our names su$ciently well, and more clos'ely
than it resembles the senses of others of our names, we may say,
using that name, '1228 is a polygon'. When we speak in h s
way, a new set of terms becomes appropriate for the four
different speech-acts which, in saying '1228 is a polygon', we
may be performing. The four performances are distinguished,
as in So, by means of direction of fit and onus of match, but
we now term them;
Calling Describing
Exemplifying Classing
When we call' 1228 a polygon or describe it as a polygon, it
is admitted, by the use of these terms for our speech-acts, that
-

the name does not exactly fit the item-because in the one case
the sense does not exactly match to the type and in the other
case the type does not exactly match to the sense.
If we are accused of wrongly ca&ng 1228 a polygon, or of
miscalling it a polygon, then we are accused of abusing language,
of doing violence to language. In calling 1228 a polygon, we
admit a multifornlity into our pattern, we modify or stretch
Not: call 1228 a 'polygon'. A'mbiguity of 'call'.
148 HOW TO TALK
the sense of our name, and future uses of the name will be
influenced by the precedent here set. If on the other hand we
are accused of wrongly describing, or of misdescribing, 1228 as
a polygon, we are accused of doing violence to the facts. In
describing 1228 as a polygon we impose admittedly a uni-
formity on our specimens, we are simplifying or neglecting
the specificity of the type of the item 1228, and we are com-
mitting ourselves thereby to a certain view of it.
In the same way, briefly, when we give exaniples as opposed
to instances we admit a multiformity in the pattern to w h c h
justice is not done by one specimen, and when we class some
item as a polygon, as opposed to identifying (casting) it as a
polygon, we admit to a neglect of the f d specificity of the item.
Two warnings may be here repeated, concerning the
'ordinary' use of such terms as 'call' and 'describe'. Firstly, these
same terms may be used of speech-acts performed in envisaged
speech-situations other than S,, for example in a speech situa-
tion in which the same item may possess more than one
feature, to draw attention to feature; of such speech-acts other
than (though no doubt connected with) the features just
described above. Secondly, it is hkely enough that our ordinary
use of the terms is fairly loose, that we do not always distinguish
carefully between them, although -
there is a distinction whch
can be marked by their means. Contrast, for example, the
following:
( I ) You call that crimson ? But surely no crimson can have
so much blue in it? That is not what crimson is at aU.
You describe it as crimson? But look, it has a lot of blue
in it. It is not really llke crimson at all.
(2) He calls me a dictator, in spite. of the fact that I have
notoriously always acted only on the advice of Parlia-
ment!
He describes nie as a dictator, whereas in fact, as he must
have known, I have always acted only on the advice
of Parliament.
 HOW TO T A L K I49
If many such examples are studied, the watershed between
-

calling and describing appears to take sl~apc.

Though cases in which we shall have to call, describe, &c.,
instead of, in black-and-white terms, identifying, stating, &c.,
are sure to arise continually, we feel ourselves sometimes bound
to cope with them as they arise by means of fresh linguistic
legislation. In calling there is indeed already implicit an element
of legislation by precedent-thls is case law and will regularly
be necessary: but we may also demand statute law. Naturally,
if the type of an item is highly novel, and does not appreciably
resemble any of our stock of patterns, or none rather than
another, it will be preferable not to call it or describe it by any
of the names in our existent vocabulary, but rather to allot this
type to some altogether new name as its sense: thls is legislation
of the kmd already familiar in So. But suppose the new type
does match fairly well one, or more than one, of our existing -
stock of patterns, suppose 1228is genuinely like a red, i.e. our
hltherto famhar speclfic red. Our fresh legislation will then
take the typical S1 form of classification and differentiation.
That is, we shall not merely allot a new name, a 'speclfic' name,
say 'crimson', to the type,-but we shall also adopt the conven-
of
tion that crimson is a sort red, thus giving the hxplicitness of
statute law to the modification in the sense of the name 'red'
and recognizing
- -
'red' as the name of a multiform pattern, i.e.
-

as a ~enericname. T h s legislation will show itself, in our

restricted language, by the phenomenon of entailment between
sentences of form S, which now appears for the first time:
henceforward, '1228 is a crimson' will entail '1228 is a red'.
The phenomenon merely of incompatibility was already
present even in So, for even there '1227is a red' was incom-
patible, according to our legislation, with '1227is a green'.
In the above account we have not, of course, dealt with by
any means all of the lunds of case that arise in S,. W e have dis-
cussed the case where the novel type is llke up to a point but
not beyond it, the sense of one of our available names, but not
appreciably l k e the sense of any other-the case, we may say,
150 HOW TO TALK
where there is only one name for more than one variety of
type (calling) or more than one variety of type for a single
name (describing). But there are also, say, cases where thk type
resembles up to a point more than one of our available patterns
-where, we maysay, there are two names to call one type by,
or a single type which may be described by two names. Such
different varieties of cases lead appropriately now to the intro-
duction of specific words (differentiation) and now to the '

introduction of generic words (classification). It should be

unnecessary further to point out that in a f~~lly-fledged language
we have, of course, numerous additional devices for coping
with the kind of cases arising typically in S,, devices such as the
useful words 'Ilkel, 'real', &c.: but in S, our language is still
restricted to sentences of the form S.
This sort of investigation of the nature of speech-acts might
go on more or less indcfit~itely.I propose to stop at this point,
where we have barely begun upon the complexities of 'calling',
'describing', &c. ~ b v i o u s l y\;hat is here written is imperfect
and probably it is wrong in many ways: but what I should like
to have succeeded in doing is in calling renewed attention to
the following points:
I. ~ a m efsi r speech-acts are more numerous, more special-
ized, more ambiguous and more significant than is ordinarily
allowed for: none of them can be safely used in philosophy in
a general way (for example, 'statement' or 'description') without
more investigation than they have, I thnk, yet received. Here of
course we have been concerned with only a few speech-acts
of a single family, but naturally there are other whole families
besides.
2. To some extent we probably do, even in ordinary lan-
guage, make use of models of the speech-situation in using the -

terms that we do for speech-acts. A; any rate, the construction

of such models can help towards clarifying the varieties of
speech-act which are possible.
-
Any such model, even the
simplest, seems bound to be fairly complicated-too compli-
cated for the standard subject-predicate or class-membership
 HOW T O T A L K I5x

model. Moreover, we seem bound to use a whole series of

different models, because the dijierence between one named speech-
.. -
act and another o f e n resides principally in a dtyerence between the
speech-situations envisagedfor their respective performances.
I have touched not merely upon very few speech-acts, but
also upon only very few features of these, and in hgMy simpli-
fied situations only. A feature, for example, in w h c h different
speech-acts even of the same family may differ very much is
that conlmonly discussed in an entirely general way under the
name of 'truth': even, say, with speech-acts which are asscia-
tions, we often prefer for one a different term of approbation .

from that which we refer for another, and usually for good
and understandable reasons. This, however, I shall not pursue
here, but instead conclude by giving- a short example of how
a small variation in our model of the speech-situation, this time
on the language side rather than on the side of the world, will
have repercussions on the speech-acts we perform.
Hitherto we have confined ourselves, in our sentence form S,
to a$rmative assertions. But if we now introduce a second
sentence form 'SN', viz.
'I is not a T',
we find that this, unllke form S, is not equally usable for the
performance of all four of our speech-acts in s,.-By introducing
-
this sentence form, we bring out a resemblance not hitherto
pointed out, between c-identifying and b-identifying in con-
trast with stating and instancing, which might be symbolized
in our diagram by lirkng them with a diagonal line, thus:
c-identifying ' stating

instancing x b - i d e n t i f y i n g
A sentence of form SN will be correct on any occasion of its
utterance if the type of the item referred to by 'I' and the sense
of the name 'T' do not match-where 'I is a T' assimilates sense
and type, 'I is not a T7contrasts thcm. We may call thls speech-
situation, which is the same as So except for the introduction
of the negative sentence form SN, 'speech-si t uation SON7.
152 H O W TO T A L K
When in situation So, I utter the sentence '1229is not a T',
then I may be statitzg something about 1229,but I cannot be
identifying it-to say that 1229 is not somethng is not to
identify it. In both stating and identifying our utterance is
intended to fit a name to, to pin a label to, the item: but there
was a difference between the two performances in point of onus
of match. And it now appears that where the interest is in
matclung a sense to the type, nothing is achievcd to the purpose
by the production of a sense whch does not match the type.
T o tell us that 1229is not a T is not to tell us what it is, nor to
identify it. But where, on the other hand, the interest is in
matching the type to a sense, something is achieved to the pur-
pose even by the discovery that the type does not match some
or any one particular sense. W e identify 1229as red as opposed
to blue, &c., but we state that 1229is red as opposed to is not red.
In a similar but opposite manner, when I utter the sentence
6
1229is not a T', I may be giving a negative or counter instance,
but I cannot be identifying (casting): there is no such thing as
a negative or counter identdication. In both instancing and
casting I am fitting an item to the name: but where I am match-
ing the sense to a type something significant is achieved even by
a refusal to match, whereas where I am m a r c h g a type to the
sense, nothing is acheved by a failure to match. W e identify
(cast) 1229 as a square, as opposed to 1228, &c., but we instance
1229as a square as opposed to not a square.
So far, it has been said that the sentence form SN is in order
when we are matchmg the (given) senseltype to a (produced)
typelsense, but not in order when we are matching a (pro-
duced) senseltype to the (given) type/sense. The same distinc
tion can be put in another way and in our old terms as result g
fiom a combination of the two distinctions of direction of fit
IJ
and onus of match, as shown by the following table on p. 200.
In h s table we may say that with both c- and b-identifying
the direction of fit is parallel to the onus of match, whereas with
both stating and instancing the direction and onus are opposite.
In identifying we fit the name to the item because the sense of
 HOW TO T A L K I53
the former matches the type of the latter, or we fit the item
to the name because the type of the former matches the sense of
the latter: but in stating and instancing we fit the namelitem
to the item/name becausc the type/sense of the latter matches the
sense/type of the former. In the verbalizations given in our
original diagram, parallelism is shown by the subject of the
sentence being in italics.
Direction of Fit Onus of Match

c-identifying .
stating .
instancing
b-identifying .
W e cannot, in either sense of 'identifv', identifv I as not a T:
to identify as not is nonsense for not- to identify. Whereas,
therefore, the use of the affirmative sentence form S will not
decide whether we are identifying or stating or instancing, the
use of the negative sentence form SN makes it clear that we
must be either stating or instancing. In similar ways other
variations in the permitted forms of sentence will in general
have effects on the varieties of speech-acts whch, in uttering
them, we may be performing. [ ~ h o u in ~ hgeneral of course
also the use of any one sentence form does not tie us down to
the performance of some one particular variety of speech-act.)
 U N F A I R TO FACTS

T H I s paper goes back to an old controversy between Strawson

and me about truth. Of course comments on comments,
criticisms of criticisms, are subject to the law of diminishmg
fleas, but I think there are here some misconceptions still to be
cleared up, some of which seem to be still prevalent in generally
sensible quarters.
In this old paper I gave (somewhat qualified) support to the
common English expression that a true statement is one which
'corresponds with the facts'. I professed not to hke thls, in its
- -

own way doubtless unexceptionable, terminology, and pre-

ferred some jargon of my own, in which 'facts' and 'corre-
sponds' do not occur at all, as a description of the conditions
Ghch must be satisfied if we are to saybf a statement that it is
true. Strawson in general, though with reservations, accepted,
I believe, that description, but contended that, nevertheless,
to say a certain statement is true is not to assert that those condi-
tions are satisfied but to do something else such as endorsing the
statement said to 'be true'. This was the principal issue, I think,
between us, but we are not concerned with it here.
Now Strawson expended a good deal of ammunition on
three words occurring in my paper-'statement', 'corresponds',
and 'facts', giving at times, I thmk, the impression that I had
given a pretty myth-eaten description even of the conditions
which must obtain if we are to say of a statement that it is true.
And this he claimed was due to my misunderstandings about
the uses of these three words. I propose therefore to consider
one of them, namely 'fact' (which I remember Ryle said at the
time I should give more attention to). I choose 'fact' because,
 UNFAIR T O FACTS I55
although the use of thls word was not at all essential to my
account of truth, (indeed I only mentioned it to point out that
it caused confusion) I do find that I hnk Strawson's own
account of the word incorrect, and that I do not for a minute
believe, with him, that 'facts' are pseudo-entities and the
notion of 'fitting the facts' a useless notion.
By way of preliminary softening up, I shall take Strawson's
two boldest and negative lines, and try to argue that they are
unplausible and based on insufficient consideration of ordinary
usage. Then I shall take h s two more sophisticated and positive
contentions.
I. What I am referring to as his two boldest and negative
lines are these:
(i) There is a (logically fundamental) type-difference
between 'facts' and things-genuinely-in-the-world, such
as things, persons and events.
(ii) 'Corresponds with the facts' is a mere idiom, as it were
a 'fused' idiom, not to be taken at all at its apparent face-
value.
(i) Taking the first of these: how is this alleged typediffer-
ence shown? Strawson points out, truly enough, that 'whle
we certainly say a statement corresponds to (fits, is borne out
by, agrees with) the facts, as a variant on saying that it is true,
we never say that it corresponds to the thing, person, &c., it is
about'.' This is partly said, I hnk, to refute my analysis of
'corresponds with the facts' in terms of (among other things)
the persons, &c., that the statement hfers to: but it will not
suffice for ths, because, of course, that analysis was offered
as a whole as an analysis of the tvhole expression 'corresponds
with the facts': to substitute part of the analysis for part of the
analysandum is a well-known short way with analyses. Still,
here is a case, certainly, in whch 'things', 'events', &c., are used
differently from 'facts'. And certainly there are important

1 Proceeditzgs ofthe Aristotelian Society, Supplementary Volume xxiv.

156 UNFAIR TO FACTS
chfferences between these words. But firstly, there are im-
portant similarities also.
For example, although we perhaps rarely, and perhaps only
in strained senses, say that a 'thing' (e.g. the German Navy) is
a fact, and perhaps never say that a person is a fact, still, h g s
and persons are far from being all that the ordinary man, and
even Strawson, would admit to be genuinely hgs-in-the-
world whatever exactly that may mean. Phenomena, events,
situations, states of affairs are commonly supposed to be
genuinely-in-the-wodd, and even Strawson admits events are
so. Yet surely of all of these we can say that they are facts. The
collapse of the Germans is an event and is a fact-was an event
and was a fact.
Strawson, however, seems to suppose that anything of which
we say '. . . is a fact' is, automatically, not something in the
world. Thus:
What 'makes the statement' that. the cat has mange 'true' is not
the cat, but the condition of the cat, i.e. the fact that the cat has
mange. The only plausible candidate for the position of what (in
the world) makes the statement true is the fact it states; but the fact
it states is not somedung in the world.1
I cannot swallow this because it seems to me quite plain:
( I ) that the condition of the cat is a fact;
(2) that the conchtion afthe cat is somehgin-the-world-
if I understand that expression at all.
How can Strawson have come to say that the condition of the
cat is not s o m e t h q in the world?
It seems to me that the trouble may arise through not observ-
ing certain rather easily overlooked boundaries between what
may and may not be done with the word 'fact'.
There are-two things we might very well, in of course a
pedagogic way, say:
(a) what makes the statement that the cat has mange true
Proceedings oJ the Arisrorelicmn Society, Supplementary Volume xxiv,
pa 135-
 UNFAIR TO FACTS 157

is the condition of the cat (together, of course, with the

meanings of the words);
(6) what makes the statement that the car has mange true
is the fact that the cat has mange.
(The second seems to me a bit unhelpful and off-colour, but
let it pass.)
But now from these two we most certainly cannot infer
The condition [or 'mangy condition'] of the cat is t h e h c t
that the cat has mange,
nor
The condition [or 'mangy condition'] of the cat is thefact
that this statement [or 'the statement that the cat has
mange'] states.
We cannot infer these things, because, for whatever reason,
they don't make sense. We do not, I think (for whatever
reason), ever say anything of the following form:
x is the fact that t h s statement [or 'the statement that S']
states
whatever be substituted for x.
Nor do we in general say anything of the form
x is the-fact that S
whatever be substituted for x (to this there are some exceptions:
e.g. It is the fact that S, Still more surprising is the fact that S,
What I object to is the fact that S, The last straw is the fact
that S).
It appears to me, perhaps wrongly, that Strawson is led, by
his 'i.e.', into supposing that the condition of the cat i s the fact
that the cat has mange, whch according to me is not sense.
T o thls he is helped by h s doctrine, to be discussed shortlv,
that ' "Fact" is wedded to "that"-clauses9. For it would of
course be good sense to say that
The condition of the cat is a fact:
and if we believe in adding 'that'-clauses we shall ask what fact,
158 U N F A I R TO FACTS
and reply 'The fact that the cat has mange'. But the question
what fact, hke its answer, does not make sense.
Having thus got so far as reducing 'the condition of the cat'
to 'the fact that the statement states', Strawson is then ready to
proclaim, on the strength of his other principal doctrine, that
'Facts are what statements state, no4 what they are about', that
the conchtion of the cat is a pseudo-entity.
To sum h s up. According to me
I. The condition of the cat is a fact, and is something in the
world.
2. It is nonsense to say: the condition of the cat-or with only
irrelevant exceptions anytlung else-is the fact that anything
at all.
3. 'Fact' with 'that' and 'fact' without 'that' behave differently.
a. x is afact is all right b. x is the& that is wrong.
Here I should say (or admit?) that there is, I think, one im-
portant and perhaps obvious difference between 'fact' and say
'event' or 'phenomenon'-a difference which for all I know
might qualify as a logically fundamental typedifference. But this
is very far from making 'facts' to be not 'thmgs-in-the-world'.
It seems to me, on the contrary, that to say that something is
a fact isbat least in part precisely to say that it is something
in the world: much more that than-though perhaps also
to a minor extent also-to classify it as being some special
kind of somethmg-in-the-world. To a considerable extent
'being a fact' seems to resemble 'existing' in the sorts of way
that have led it to be held that 'existence is not a predicate'.
One might compare:
Cows are animals with Fevers are diseases
Cows are t h g s Fevers are conditions
Cows exist Fevers are facts
There are such t h g s as cows Thereare such conditions as fevers
Or one might say: it is as true and yet as queer to say that ')?acts
are things-in-the-world' as that 'Entities are hngs-in-the-
world'. Or one might say that 'fact' resembles. 'person':
 UNFAIR T O FACTS I59
4
persons' are not, of course, facts and facts are not persons
but to say so-and-so is a person or a fact is in part at least
to say so-and-so is real. (We do not on the whole llke, there
is oxymoron in, imaginary 'facts' or 'imaginary persons'-we
prefer imaginary characters. In this way 'fact' does differ from
say 'event'.)
I believe-but this is speculative-that it could be shown
that some of the odd phenomena notorious about the word
'fact' could be shown ;o be connected with this 'existential'
side of it (some of which, however, have been much exagger-
ated!), e.g. that we do not say that 'a fact was at a certain time'
but that 'at a certain time it was a fact'.
(ii) Secondly, how is Strawson really to handle the.gobd
English expression 'corresponds with the facts' (and incidentally
a good many others that go along with it) ? For he claims that
'the demand for s o m e h g in the world to whlch the state-
ment corresponds when it is true' is a tuholly mistaken demand
and that 'fact' is a name for a pseudo-entity invented to satisfy
this bogus demand. But this is to treat a wholesome ~ n g l i s h
expression as though it were a philosopher's invented expres-
sion; to treat 'facts' as though they were in the same -

as 'propositions' t&ng these to be pseudo-entities invented to

answer the bogus demand 'what is it that sentences when
meaningful mean?' How come that English has invented so
unhappy an expression? It seems to me that Strawson has two
rather different lines here, neither of w h d will do.
One dung we might try is to say that 'corresponds with the
facts' is a 'fused' idiom precisely
-
equivalent to 'is true'. It is in
no way to be understood as meaning that anything 'corre-
sponds' with anything in the way that that word ordinarily
means, or as implying that there are such things as 'facts', or
that anything is related in any way to anything-(for that
matter, ~I suppose, no assertion is made about the statement
at all when we say it 'corresponds with the facts', any more
than when we say it 'is true'). This seems to me quite un-
plausible-why should not we be meaning by it that there is
I 60 UNFAIR TO FACTS
some sort of relation between something and something (no
doubt not so simple as it sounds)? Strawson actually allows
that when I say a map corresponds with the topography of the
countryside I am taking about a relation between something
and s o m e t h g- , yet
- still contends that I am not when I say a
statement corresponds with the facts. But how is h s plausible
-for we may remind ourselves that it is quite possible to say
of a map that it does not correspond with the facts (e.g. a
situation-map) and of a statement that it does not correspond
with the topography. (Or should we go back, then, on the
admission that 'corresponds' does mean a relation in the case
of maps-say that topography is what maps map, not what
maps are about, &c.!)
A second, and surely rather different, line, is the following.
For the good ~ n g l i s h'corresponds with the facts' or 'fits the
facts' we substitute the quite unEnglish 'corresponds with the
-

fact that so and so' or 'fits the fact that so-and-so' and proceed to
claim that this is somehow tautological. Thls is a method
which seems to be becoming fashonable. Strawson writes,
'What could fit more the fact that it is raining than
the statement that it is raining? Of course statements and facts
fit. They were made ror eaX other.' But in answer to ths:
surely it is not sense either to ask whether the statement that S
fits the fact that S or to state that it either does or does not.
And I may add that it seems to me, pace Mrs. Daitz, equally
nonsense to ask whether the statement that S fits or corresponds
to the fact that F, where 'F' is dflerent from 'S' not identical
with it (though of course it is not nonsense to ask something
that sounds rather similar to ths, viz. whether the statement
that S squares with or 'doesjustice to' the fact that F ('F' # 'S')).
But even, further, if we allowed the expression: the statement
that S fits the fact that S, surely we cannot allow the suggestion
of 'statements and facts were made for each other'. T o begin -
with, obviously, some statements do not fit the facts; we need
not elaborate on ths. But even if we take it as: ' True statements
fit the facts',' True statements and facts are made for each other'
 U N F A I R TO FACTS 161

we cannot agree that there is any more sense in saying they are
'made for each other', or if there is sense, any more m&ce in
saying it, than there would be in saying, on-the ground that
'well-aimed shots hlt their marks', that marks and shots (well-
aimed) are made for each other.
It may be said-before we proceed further-why bother?
Why raise h s cry 'unfair to facts'? There are two reasons
I hnk.
I. It seems in the past to have been, at least often, inexpedient
to say 'Facts are not things-in-the-world'. Ths misled, for ex-
ampie, Bradley who had some by no means dl-deserved cold
water to pour hpon facts, at the crucial point. Still, on the other
hand, no doubt it may for some purposes also be expedient to
give ourselves a jolt by suggesting that 'there are no facts'.
2. More important altogether, to my mind, than h s is the
following. The expression 'fitting the facts' is not by any means
an isolated idiom-in our language. It seems to have a very
intimate connexion with a whole series of adverbs and adjec-
tives used in appraising statements-I mean, 'precise', 'exact'.
'rough', 'accurate', and the ilke, and their cognate adverbs.
All these are connected with the notion of fitting and measuring
in ordinary contexts, and it can scarcely be fortuitous that they,
along with fitting and correspondmg, have been taken over as
a group to the sphere of statements and facts. Now to some
extent the use of this galaxy of words in connexion with state-
ments may be a transferred use; yet no one would surely deny
that these constitute serious and important notions w h c h can
be, and should be elucidated. I should certainly go much
farther and claim-as I have done before-that these are the
important terms to elucidate when we address ourselves to
the -problem of 'truth', just
-
as, not 'freedom' but notions llke
duress and accident, are what require elucidation when we
worry about 'freedom'. Yet all these terms are commonly
dismissed along with the supposed useless 'fitting the facts'.
T o give
- an dustration of h s . Wamock, in h s excellent
book on Berkeley, says that Berkeley's complaint is that
162 U N F A I R T O FACTS
ordinary language does notfit the facts (or, as he says at other
times, is inaccurate or inexact or loose or imprecise): Warnock
says thls is a confusion: for 'fits the facts (exactly?)' is simply
equivalent to 'is true'-and he reduces the notion to helplessness
by the same device as Strawson did: he asks, e.g.
What statement could possibly fit the fact that there is a table in
my study more exactly than the statement 'There is a table in my
study' ?I
and says
It is really tautologous to say that any true statement exactly fits
the facts; it exactly fits those fans, or that fact; which it states.
It would be most unreasonable to object to a statement becausd
it fails to state facts which it does not state.2
Having thus disposed of 'fitting the facts', Wamock considers
all the rest-inaccuracy,J looseness, &c. (but he does not dis-
tinguish between these!) as apparently disposed of, and offers
us as the only reasonable complaint against statements w h c h
could very well, by confusion, be intended by, for example,
Berkeley, this: that the statenlent is 'logically complex' and
can be analysed into more verbose but logically simpler ex-
pressions. Such an analysis does not he says 'fit the facts' any
better but is merely more explicit or more detailed-less con-
centrated. Well, maybe thls does occur (if it is a single phen*
/
menon?); maybe it was what Berkeley meant, though - t h i s
would need proving; but all the same inaccuracy, looseness,
and the rest do also occur, are important, are distinct, and they
cannot be dismissed from consideration by thls ~ervertedlittle
rigmarole about 'fitting the facts'.
II. Now I shall set out what I take to be Strawson's main
positive contentions about the word 'fact'. i t h n k that others
too hold the same beliefs.
4

G. J. Warnock, Berkeley, Pelican Books, 1953, p. 240.

Ibid.
3 The relation between a n,ugh, loose, inaccurate, inexact, imprecise, &c.
statement and the facts is a most important matter for elucidation.
 U N F A I R TO F A C T S 163
4
1. Fact', says Strawson, 'is wedded to that-clauses'.' He does
not explain this overmuch. But I think it is meant to be a state-
ment about ~ n g l i s husage: and one impression he conveys to
h s readers and I think himself is that, wherever we get the
word 'fact' occurring in a sentence without a 'that-clause' we
at least could insert a that-clause after it, even if we do not
already actually 'supply' such a clause in some way mentaUy.We
have seen already two instances of h s acting on this principle.
ii. 'Facts', says Strawson, 'are what statements (when true)
state; they are Lot what statements are about.' Ths seems to
be a p ~ o s o p h c a lcontention, based upon the linguistic con-
tention I above. And it is this phlosophcal contention whch
leads him to say that facts are not anything genuinely-in-the-
world, but pseudo-entities.
Now let us confront i: 'Fact is wedded to that-clauses' with
the evidence of the Dictionary. If it is indeed so wedded, it
leads, I fear, a double life. It was not born in wedlock, its
marriage was a marriage of convenience and it continues to
lead a flourishmg bachelor existence.
The Dictionary (Oxford English Dictionary) does not take
much account of 'fact that'. 'Fact that', along with 'fact of', is
given as a comparatively recent linguistic device for avoiding
gerundial constructions as the subjects of sentences or as the
- -

domains of prepositions: i.e. in order not to say 'I was unaware

of the htchen's being draughtypor 'the h ~ h e n ' being
s draughty
annoyed him', we say 'I was unaware of the fact that the
lutchen was draughty'
- -
or 'the fact that the kitchen was draughty
annoyed him': it is compared with 'circumstance that'. No
use of 'fact of' is given earlier than the eighteenth century:
examples of 'fact that' are not given, but may even be later.
I believe that if we consult even, say, the philosophers of the
eighteenth century we can readily satisfy ourselves that 'fact
that' is not in use; and the Dictionary's account is borne out
by our experience of the general tendency nowadays to adopt
similar novel 'that'-constructions with other words, hke
Proceedings of the Aristotelian Society, Supplementary Volume ladv.
164 UNFAIR TO FACTS
'circumstance', 'event', 'situation'. (A later feature of such
6.
usages of course, is that the 'that' gets dropped out- m case'
for 'in the case that' and 'in the event' for 'in the event that'.)
To complete the hstory of the word accordmg to the Dic-
tionary. ~ 6 the r first zoo years of its use (sixteen& and seven-
teenth centuries) it meant (cf. 'feat') a deed or action, either the
thing done or the doing bf the dung, and more especially a
criminal action; during the eighteend17centurythls use gradu-
ally died out in favour of a more extended meaning w h c h
began to appear already in the seventeenth century: a fact is
now sornethin. that has really occurred (even classical Latin
extendedfact& to mean 'event' or 'occur rince' )1 or something
that is actually the case (a further extension to the meaning of
'facturn' found in scholastic Latin). Hence and thereafter it came
to mean s o m e h g known to have really occurred, and hence
(according to the-~ictionary)a particular truth known by
observation or authentic testimony, by contrast with what is
merely inferred or a conjecture or a fiction.
From this brief history, I take it as obvious that: (i) 'Fact'
was in origin a name for 'something in the world', if we may
take it that a past action or past actual event or occurrence is
c
something in the world', and there is no reason whatever to
doubt thaiit often still is so. (ii) Any connexion between 'fact'
and 'knowledge', and still more between 'fact' and 'truth' (in
particular the use of 'a fact' as equivalent to 'a truth'), is a
derivative and comparatively late connexion. (iii) The expres-
sion 'fact that' is Gter still, and was introduced as a gram-
matical convenience, because of the already existing meaning
of 'fact'. To ttxplain the meaning of 'fact' in terms of the
expression 'fact that' is to invert the real order of things-just
as much as it would be to explain the meaning of 'circumstance'
in terms of 'circumstance that' (or 'situation' in terms of
'situation that' and so on: indeed, one may suspect that we have
here one pretty general method of misunderstandmg or mis-
representing what language is up to). r'

Actual event or occurrence, of course.

 UNFAIR TO FACTS 165
I should further take it as fairly plain that (iv) when 'facts'
or 'a fact' occur in general in modem English, the usage is just
what it was in the eighteenth century. When we say 'The
mangy condition of the cat is a fact' we mean it is an actual
state-of affairs; when we say 'What are the facts?' we mean
'what is the actual state of affiirs ?', 'what has actually occurred ?'
or the llke. Ths is the meaning, too, in such common exprcs-
-

sions as 'an dtcomplished fact' or 'He has had no personal ex-

perience of the facts he reports'. And finally, I take it that (v)
the expression 'The fact that S' means 'a certain fact [or actual
occurrence, &c.], viz. that correctly described [or reported,
&c.] by saying now "S" [or at other times "S" with a change
of tense].' It was to ths that I referred when I said that 'the fact
that' is a way of speaking compendiously about words and
world together. It is a usage grammatically like (not of course
in all ways the same as) the apposition usage with proper names,
as when we say 'The person
-
Caesar' whch we should interpret
as 'a certain person [or actual man, &c.], viz. the one desig-
nated by the name "Caesar" '.
And two further clear points about 'the fact that':
(vi) 'The fact that S' is a totally different occurrence of
'the fact that' from that in say ' Thefact that he stated'.
In other words, even if 'fact' were wedded to that-
clauses, there are cLfferent that-clauses.
(vii) I feel-but cannot I confess sufficiently 'explain', pro-
bably for want of terminology-that somehow h s
- .

explanation of the special form of expression 'the fact

that S' should be what accounts (a) for our inability
to say t h g s of the form 'the mangy condition of the
cat is the fact that the cat has mange' (or 'the cat's
having mange is the fact that the cat has mange')
and likewise possibly (b) for the queerness of the
problem that Ramsey and Moore discussed as to
whether 'the fact that S' is a name or a description or
what on earth?
165 U N F A I R TO FACTS
Compare persons and facts again: is there a parallel between
'Cicero' 'The cat has mange'
ciccro [The cat's having mange
The mangy condition of the cat
The person Cicero The fact that the cat has mange
Thus (a)..
T o say 'the mangy condition of the cat is the fact thai the
cat has mange' is-Wre saying 'Cicero is the person Cicero'.
These two forms of expression are not designed for combina-
tion in thls manner. But why? Nor for that matter is 'Tully is
the person Cicero', only rather ' "~icero"' is the name of the
person Cicero'.
'Tullv' is a name of the same person as 'Cicero': so 'Everest
/ I

is so high' is a statement of the same fact as 'Gaurisankar is so

hgh'. 'Everest is so high' is a statement of the fact that Everest
-

is so high.
Also (b) asking whether 'The fact that S' is a name o; a
description is U e askmg whether 'the person Cicero' is a name.
But we seem to lack a name for thls type of expression and a
description of its role and limitations.
Now let us ask, what was Strawson actually arguing in
saying 'Fact is wedded to that-clauses'. He was t h h g- of its
-

use in connexion with certain verbs.

c
'Facts', he says, are known, stated, learnt, forgotten, over-
looked, commented on, communicated, or noticed. (Each of
these verbs may be followed by a "thatv-clause or a "the fact
that"-clause.)'' This is not, I think, very clear.
I. It is not clear whether it is meant that wherever we have
a sentence in whicli one of these verbs occurs with 'fact' as its
object we could put a that-clause or the-fac t-that-clause in place
of the obiect, or only that we sornetimes can.
2. It is-not clear whether it matters whether 'fact' occurring -
in the original sentence as object of one or another of these
verbs is to be singular or plural, with the definite or indefinite
article or neither.
Proceedings oJ thc Aristotelian Socicty, Supplementary Volume k i v ,
p. 136.
 U N F A I R TO F A C T S 167
3 . It is not clear whether, if we do substitute a that-clause
or a the-fact-that-clause for 'tact' the meaning of the original
sentence is to be considered altered or unaltered.
But I think the impression Strawson conveys both to himself
and to his readers is that, given any of these verbs, the following
three forms of sentence are equally good English and somehow
interchangeable.
(a) Fact-jiorm. He vcrbed a/the factlfacts
(b) That-form. He verbed that S
(c) he fact-that-form. He verbed the fact that S.
But it will be found that the verbs in his list behave differently
with respect to these three forms of sentence. (With some, such
as 'comment on', it is not the case that all three forms make
sense: we cannot 'comment on thatY.)
The best example of what Strawson is after seems to be the
verb 'forget':
He forgot a fact or the facts
He forgot that S
He forgot the fact that S.
These seem all good English, and reasonably interchangeable-
of course the second two specify what it was he forgot whereas
the first does not, but tllis is fair enough and nothing to boggle
at.
And, of course, with the verb 'state' also, all three forms of
sentence make sense (though not all are equally usual) :
He stated a fact
He stated that S
He stated the fact that S.
But here there is a great difference between the second and the
other two. With 'forget', the fact-form entails the that-form,
and conversely; but with 'states' this is not so. 'He stated that
S' does not entail 'he stated a fact'.
Similarly, with 'state' certain forms of sentence are good
sense whch are not good sense in the same way with 'forget'
or 'know':
168 U N F A I R TO F A C T S
What he stated was a fact
In stating that S he stated a fact.
But not
What he forgot or knew was a fact1
In forgetting that S he forgot a fact.
All that this shows is, of course, that 'state' is not, hke 'know' and
'forget', a succcss or achievement word in Ryle's sense. T o know
that or to notice that is to know or notice a fact; but to state
that is not necessarily to state a fact. 'State' may be wedded to
that-clauses; but 'fact' is wedded to neither.
Thls is of course notorious enough. But surely it is also
enough to cast doubt on 'Statements and facts are made for
each other'. Even 'Knowledge and facts are made for each
other' or ' ~ o r-g e t f h e s sand facts were made for each other'
would contain more truth-though even then one would be
inclined to protest that there may very well be facts that
nobody knows or ever will know, and that to say the facts are
made for the knowledge is curious.
Strawson also produces a list of other verbs, presumably
unconnected with 'that7-clauses, which he suggests are con-
nected with things-in-the-world (things, persons, events) but
not with facts, viz. 'witness', 'hear', 'see', 'break', 'overturn',
6
interrupt', 'prolong', &c. But surely we can witness facts?
And observe-them and have personal acquaintance with them.
I cannot see that there is a clear distinction here.
However, we have seen I dunk in general enough from the
dictionary, and from the troubles that arise when we insert
that-clauses indiscriminately, to abandon the doctrine that fact
is wedded to that-clauses. Let us now turn to the second doc-
trine which was that:
'Facts are what statements (when true) state: they are not
what statements are about.'
Take the second clause first, about which I shall not say
much. The main t h n g I feel about it is that 'about' is too
Point of the joke about the Bourbons 'forgetting nothing'.
 UNFAIR TO FACTS 169
vague a word to bear the weight Strawson wants to put on it.
If we take other verbs, for example, facts are both what we
forget and what we forget about, both what we know and
what we know about (as well as 'of); verbally we can say that
'the number of the planets' is both what a statement is about
and what that same statement states. Moreover, Strawson
would admit, I tllink, that a statement can be about a fact, or
about the fact that . . . I do not, however, wish to insist that
Strawson could not make out a case for this clause of his
doctrine.
'
What I do wish to insist is that the first clause- Facts are
what statements, when true, state' is most misleading. Un-
doubtedly we should agree that 'True statements state facts';
but if we convert it in Strawson's manner we certaiilly give
ourselves the impression that this is somehow or other a
'definition' of facts. It gives; and I think is meant to give, the
impression of reducing 'facts' to an accusative so deeply and
hopelessly internal that their status as 'entities' is hopelessly
compromised.
Strawson represents us as asking: 'What do true statements
state?' and giving ourselves the bogus and unhelpful answer
'Facts', much as philosophers are supposed to have asked them-
selves 'What do meaningful sentences mean?' and to have
-

given themselves the bogus answer 'propositions'. Yet there is

nothmg in the form of such questions to make them neces-
sarily, along with their answer, bogus. Strawson himself is
prepared to ask, for example, 'What do statements refer to?'
and to answer 'Things, persons or events'. Nor need we by
any means take our answer (e.g. Things are what statements
refer to) to be a definition of 'things' or in some manner un-
wholesomely analytic.
To shake off this hypnotisin of thc internal accusative, this
vague menace of the analytic, consider a few parallel examples:
Births are what birth-certificates, when accurate, certify.
Persons are what surnames, when borne, are the surnames of.
Women are what men, when they marry, marry.
70
I U N F A I R TO F A C T S
Wives are what men, when they marry, marry.
Animals are what portraits, when faithful, portray.
Sitters are what portraits, when faithful, portray.
Events are what narratives, when true, narrate.
History is what narratives, when true, narrate.
Which, if any of these, is Strawson's sentence to be taken to
resemble? Does it in particular resemble the 'Events are what
narratives, when true, narrate' case? And would it make sense
6
to go on: they are not what narratives are about'? At any rate,
one thing is plain, that, whatever the analyticity or internal
accusatives or what not involved in any of these pronounce-
ments, none of them have the slightest tendency to convince
us that births, persons, or events, women (or even wives),
animals (or even sitters) are not ' t h g s or happenings on the
surface of the globe'.
Why then worry about 'facts'?
There seems, on closer inspection, to be some trouble about
the expression 'What statements state'. How is it to be taken?
Let us agree by all means that 'true statements state facts',
and that somehow or in some sense tbs is, if you lke, 'true by
definition'-though keeping of course an open mind as to
which definition-that, for example, of 'fact' or that of 'true
statement'. But now if we put this in the shape
Facts are what true statements state
it has to be understood that this variety of 'what' ('what state-
ments state' 'What he stated') is a highly 'ambiguous' one--or
rather, that there are different uses for the expression 'what he
stated'.
T o take an illustration: compare:
(a) what he stated was true, [or 'was a truth'].
(b) what he stated was a fact.
In these two sentences 'what he stated' is not grammatically
identical. For in (a) we can substitute for 'what he stated' the
expression 'hls statement'; whereas in (b) we cannot do thls;
 UNFAIR TO FACTS 171
'his statement was a fact', ifit had any meaning, would anyhow
mean something quite different from 'what he stated was a
fictY.'
O r again, contrast both (a) and (b) a U e with yet another
use of 'what he stated'. In reply to the question 'what did he
state?' we might be told '(He stated) that S', i.e. we get
(c) what he stated was that S.
That this differs from the use of 'what he stated' in (a) and (b)
is plain; but the difference may be brought out by considering
how odd it would be, to give to the question 'What &d he
state?' the answer 'A fact' or 'A truth'. These could o n b be
given, on the contrary, as answers to the quite different yes-or-
no-type questions 'Was what he stated a fact?' or 'Was what he
stated true (a truth)'.
W e can in fact distinguish at least five different ways in
which 'what' is used in connexion with 'state', which are
immediately relevant here. 'State', in this respect, is typical of
a large group of words used in taking about communication,
and I shall illustrate the five different uses concerned by means
of the analogous verb to 'signal'.
Let us take a simple model situation in whch all we are
concerned to do is to signal targets as they bob up in an aper-
ture. The targets are of various different recognizable types
and there is an appointed signal (say a flag of a special colour)
for each different type of target. Very well then-the targets
keep on bobbing up, and we keep on wagging flags-the con-
vention of reference being the simple one (non-'verbal' or
rather non 'flag-al') that each signal rerers to the contemporary
occupant of the aperture. Now, in what ways can we use the
expression 'what we signal'
-
?
(I) There are two uses in w h c h we can use the expression
I This is of course in line with the fact that in 'It is true that S' we can insert
'to state' after 'is true'; whereas in 'It is a fact that S' we cannot insert 'to state'.
Strawson says there is no nuance except o f style between 'It is true that' and
'It is a fact that'; yet he himself admits that the former 'glances at' the making
of a statement, whereas the latter does not.
172 U N F A I R TO F A C T S
'What we signal is always . . .'. Perhaps not very helpful or
usual except for paedagogic purposes. viz.
( I ) 'What we signal is always a target'. CE 'What we state
is always a fact'.
(2) 'What we signal is always a signal'. CE 'What we state
is always a statement'.
There is no conflict between these pronouncements; nor is it
either necessary or legitimate,
- in order to 'reconcile' them, to
conclude that 'targets are signals': they are not, nor are facts
statements, by the same token! All that is needed is to realize
that 'what we signal' has two uses-or that 'signal' can take
two varieties of accusative, both, interestingly enough, com-
pletely 'internal'. ( I ) means that whenever we signal we signal
- -

something as a target, signals are as such signals of targets; so

whenever we state we state something as a fact, statements are
as such statements of fact.'
(2) of course is hlghly paedagogic, though harmless. It
would be very paedagogic to say 'we signal signals' or 'we state
statements'--though parallel to saying, for example, 'we plan
plans'. Such .remarks instruct us as follows: whenever we say
'we plan (signal, state) so-and-so' there will be s o m e h g of
which we can say that it was our plan (signal, statement).
(I) and (2) alike are only of use in giving instructions about the
meaning and use of the verb 'signal' and the substantive 'signal'.
But now further (11), when we do not use the 'always' and
the 'we' of paedagogic generalization, we use 'what . . . signal
(state)' in quite different ways, as follows:
(3) 'What we signal is sometimes, but not always, a target'.
'What we state is sometimes, but not always, a fact'.
(4) 'What we signal is sometimes, but not always, (a) cor-
rect (signal).
'Whar we state is sometimes, but not always, true' (a
truth).
Of course we are excluding from treatment throughout the expression
'statement of opinion', interesting enough in itself but not relevant here.
 U N F A I R TO FACTS I73
It is in use (3) that we speak when we say, for example, 'What he
stated then was a fact but what she stated thereafter was not a fact';
and in use (4) that we speak when we say, for example, -
'What
he stated then was true but what she stated thereafter was not'.
The uses of 'what we signal' in (3) and (4) are of course
connected, but they are not identical. If they were identical we
should get the absurd conclusion that
A target is (a) correct (signal)
and the conclusion, equally absurd I thmk, that
A fact is a true statement (a truth).
The actual connexion between (3) and (4) is not so simple,
though simple enough:
If, and only if, and because what we signal (3) is a target,
then what we signal (4) is correct; and, derivatively, if,
and only if, but not because what we signal (4) is correct,
then what we signal (3) is a target. f

There is an obvious connexion between 'what we signal' in

the sense (3) and 'what we signal' in the paedagogic sense ( I ) ;
and Uewise between sense (2) and sense (4). Perhaps, indeed,
we might even say that the hfference between (I) and (3)
and again between (2) and (q), is not so much a difference
between senses or uses of 'what we signal' as rather a difference
between senses or uses of 'what we signal is'. But at any rate
there is a radical distinction between 'what we signal' in ( I )
and (3) together and in (2) and (4) together-between what
-

we signal as a target (actual or-paedagogically-putative) and

on the other hand as a signal (correct or incorrect). And just
as, in the s i g n h g case, there is not the slightest temptation
to say
'Targets are not h n g s in the world'
4
Targets are (correct) signals'
so there should be no temptation to say
'Facts are not things-in-the-world'
'Facts are true statements'.
174 UNFAIR TO FACTS
But finally (111) there is yet another use of 'what we signal',
or of 'what we signal is' where 'what we signal' is neither
always x, as in (I) and (2)' nor always either x or not x, as in
(3) and (4)-
In this use ( 5 ) .
-what we signal is now red, now green, now purple
-what we state is now that S, now that T, now that U.
T o sum up, then,
Facts are what true statements state
is llke
Targets are what correct signals signal
and has no more tendency to prove that facts are pscudo-
entities than that targets are. And if we choose to say that
either is being defined in terms of the other, we should say the
other way round.
O r again: from the two unexceptionable
( I ) He stated a fact
(2) He stated that S

we can indeed get:

'what he stated was a fact'
'what he stated was that S'.
But we cannot possibly make an inference from these such as
a fact is that S or that T, &c.
(whch makes it look llke a 'bogus entity') because 'what he
stated' does not mean the same in the two premisses.
 A PLEA FOR EXCUSES'

THEsubject of this paper, Excuses, is one not to be treated, but

only to be introduced, w i h n such limits. It is, or might be, the
name of a whole branch, even a ramiculated branch, of philo-
sophy, or at least of one fashion of phlosophy. I shall try,
therefore, first to state lohat the subject is, why it is worth study-
ing, and how it may be studied, all this at a regrettably lofty
level: and then I shall illustrate, in more congenial but desultory
detail, some of the methods.to be used, together with their
limitations, and some of the unexpected results to be expected
and lessons to be learned. Much, of course, of the amusement,
and of the instruction, comes in drawing the coverts of the
microglot, in houndmg down the minutiae, and to t h s I can
do no more here than incite you. But I owe it to the subject
to say, that it has long afforded me what philosophy is so often
thought, and made, barren of-the fun of discovery, the
pleasures of co-operation, and the satisfaction of reachmg agree-
ment.
What, then, is the subject?I am here using the word 'excuses'
for a title, but it would be unwise to freeze too fast to t h s one
noun and its -partner verb: indeed for some time I used to use
'extenuation' instead. Still, on the whole 'excuses' is probably
the most central and embracing term in the field, although this
includes others of importance- cplea', 'defence', 'justification',
and so on. When, then, do we 'excuse' conduct, our own or
somebody else's? When are 'excuses' proffered?
In general, the situation is onewhere someone is accused of
Reprinted f r ~ mProceedings of the Aristotelian Society, 1956-7,by courtesy
o f the editor.
176 A PLEA F O R EXCUSES
having done somedung, or (if that w d keep it any cleaner)
where someone is said to have done s o m e h g which is bad,
wrong, inept, unwelcome, or in some other of the numerous
possible ways untoward. Thereupon he, or someone on h s
behalf, will try to defend h s conduct or to get h m out of it.
One way df going about t h s is to admit flatly that he, X,
did do that very dung, A, but to argue that it was a good thing,
or the right or sensible t h g , or a permissible thing to do,
either in general or at least in the special circumstances of the
occasion. To take t h s line is to justify the action, to give
reasons for doing it: not to say, to brazen it out, to glory in it,
or the like.
A different way of going about it is to admit that it wasn't
a good thing to have done, but to argue that it is not quite fair
or correct to say baldly 'X did A'. W e may say it isn't fair just
to say X did it-; perhaps he was under sohebody's influence,
or was nudged. Or, it isn't fair to say baldly he did A; it may
have been partly accidental, or an unintentional slip. Or, it
isn't fair to say he did simply A-he was really doing somethmg
quite different and A was only incidental, or he was looking
at the whole b g quite differently. Naturally these arguments
can be combined or overlap or run into each other.
In the one defence, briefli, we accept responsibility but deny
that it was bad: in the other, we admit that it was bad but
don't accept full, or even any, responsibility.
By and krge, justificationscan be kept distinct from excuses,
and I shall not be so anxious to talk about them because they
have enjoyed more than their fair share of phllosophcal atten-
tion. But the two certainly can be confused, and can seem to go
very near to each other, even if they do not perhaps actually
do so. You dropped the tea-tray: Certainly, but an emotional
storm was about to break out: or, Yes, but there was a wasp..
In each case the defence, very soundly, insists on a fuller de-
scription of the event in its context; but the first is a justifica-
tion, the second an excuse. Again, if the objection is to the use
of such a dyslogistic verb as 'murdered', h s may be on the
 A PLEA F O R E X C U S E S I77
ground that the killing was done in battle (justification) or on
the ground that it was only accidental if reckless (excuse). It
is arguable that we do not use the terms justification and excuse
as carefully as we might; a miscellany of even less clear terms,
such as 'extenuation', 'palliation', 'mitigation', hovers uneasily
bemeen partial justification and partid excuse; and when we
-

plead, say, provocation, there is genuine uncertainty or am-

biguity as to what we mean-is he partly responsible, because
he roused a violent impulse or passion in me, so that it wasn't
truly or merely me acting 'of my own accord' (excuse)? Or
is it rather that, he having done me such injury, I was entitled
to retaliate (justlfication)? Such doubts merely make it the
more urgent to clear up the usage
- -
of these various terms. But
that the defences I have for convenience labelled 'justification'
and 'excuse' are in Iprinciple distinct can scarcelv be doubted.
I 8

This then is the sort of situation we have to consider under

'excuses'. I will only further point out how very wide a field
-

it covers. W e have, of course, to bring in the opposite numbers

of excuses-the expressions that agSravate, such as 'deliberately',
'on purpose', and so on, if only for the reason
often takes the form of a rebuttal of one of
also to bring in a large number of expressions which at first
blush look not so much lrke excuses as hke accusations-
4
clumsiness', 'tactlessness', 'thoughtlessness', and the like.
Because it has always to be remeibered that few excuses get
us out of it completely: the average excuse, in a poor situation,
gets us only out of the fire into the frylng pan-but still, of
course, any frying
-
pan in a fire. If I have broken your dish or
- -
your romance, maybe the best defence I can find will be
clumsiness.
Why, if &s is what 'excuses' are, should we trouble to
investigate them? It might be thought reason enough that
their production has always bulked so large among human
activities. But to moral phlosophy in particular a study of
them will contribute in special ways, both positively towards
the development of a cautious, latter-day version of conduct,
178 A PLEA FOR E X C U S E S
and negatively towards the correction of older and hastier
theories.
In ethics we study, I suppose, the good and the bad, the right
and the wrong, and this must be for the most part in some
connexion with conduct or the doing of actions. Yet before
we consider what actions are good or bad, right or wrong, it
is proper to consider first what is meant by, and what not, and
what is included under, and what not, the expression 'doing
an action' or 'doing sometlung'. These are expressions still too
little examined on their own account and merits, just as the
general notion of 'saying something' is still too lightly passed
over in logic. There is indeed a vague and comforting idea in
the background that, after all, in the last analysis, doing an
action must come down to the making of physical movements
with parts of the body; but this is about as true as that saying -
something must, in the last analysis, come down to making ,

movements of the tongue.

he beginning of sense, not to say wisdom, is to realize that
4
doing an action', as used in philosophy,' is a hghly abstract
expression-it is a stand-in used in the place of any (or alnlost
any?) verb with a personal subject, in the same sort of way that
'thing' is a stand-in for any (or when we remember, almost
any) noun substantive, and 'quality' a stand-in for the adjective.
-

Nobody, to be sure, relies on such dummies quite implicitly

quite indefinitely. Yet notoriously it is possible to arrive at, or
to derive the idea for, an over-simplified metaphysics from the
obsession with 'things' and their 'qualities'. In a similar way,
-

less commonly recognized even in these semi-sophisticated -

times, we fall for the mvth of the verb. We treat the expression
J L

'doing an action' no longer as a stand-in for a verb with a

personal subject, as whichit has no doubt some uses, and might
have more if the range of verbs were not left unspecified, but
as a self-explanatory, ground-level description, one whch
brings adequately into the open the essential features of every-
This use has little to do with the more down-to-earth occurrences of
'action' ul ordinary speech.
 A PLEA FOR EXCUSES 179

thing that comes, by simple inspection, under it. We scarcely

notice even the most patent exceptions or difficulties (is to
think something, or to-say sornethlng, or to try to do some-
dung, to do an action?), iny more than we fret, in the ivresse
desgrandes profondcurs, as to whether flames are things or events.
So we come easily to think of our behaviour over any time, and
of a life as a while, as consisting in doing now action. A, next
action B, then action C, and so on, just as elsewhere we come
to think of the world as consisting of this, that and the other
substance or material thing, each with its properties. All 'actions'
are, as actions (meaning what?), equal, composing a quarrel
with strking a match, winning a war with sneezing: worse stdl,
we assimilate them one and all to the supposedly most obvious
- -

and easy cases, such as posting letters or moving fingers, just as

we assimilate all 'thmgs' to horses or beds.
If we are to continue to use this expression in sober philo-
sophy, we need to ask such questions as: Is to sneeze to d o an
action? Or is to breathe, or to see, or to checkmate, or each one
of countless others? In short, for what range of verbs, as used
on what occasions, is 'doing an action' a stand-in? What have
they in common, and what do those excluded severally lack?
Again we need to ask how we decide what is the correct name
for 'the' action that somebody chd-and what, indeed, are the
rules for the use of 'the' action, 'an' action, 'one' action, a 'part'
or 'phase' of an action and the like. Further, we need to realize
that even the 'simplest7 named actions are not so simple-
certainly are not &e mere makings of physical movements,
and to ask what more, then, comes in (intentions? conven-
tions?) and what does not (motives?), and what is the detail
of the complicated
- internal machinery we use in 'acting'-the
receipt of intelligence,
- -
the appreciation of the situation, the
invocation of principles, the planning, the control of execution
and the rest.
In two main ways the study of excuses can throw light on
these fundamental-matters. First, to examine excuses is to
examine cases where there has been some abnormality or
I 80 A PLEA F O R E X C U S E S
failure: and as so often, the abnormal wdl throw light on the
normal, will help us to penetrate the blin&ng veil of ease and
obviousness thathides the mechanisnls of the natural successful
act. It rapidly becomes plain that the breakdowns signalized
by the various excuses are of radically different hnds, affecting
different parts or stages of the macLinery, which the excuses
consequently pick out and sort out for us. Further, it emerges
that not every slip-up occurs in connexion with everything that
could be called an 'action', that not every excuse is apt with
every verb-far indeed from it: and this provides us with one
means of introducing some classification into the vast miscellany
of 'actions'. If we classify them according to the particular
selection of breakdowns to which each is liable, this should
assign them their places in some family group or groups of
actions, or in some model of the machinery of acting.
In this sort of way, the philosophical study of conduct can
get off to a positive fresh start. But by the way, and more
negatively, a number of traditional cruces or mistakes in this
field can be resolved or removed. First among these comes the
problem
-
of Freedom. Wlile it has been the tradition to present
t h s as the 'positibe' term requiring elucidation, there is little
doubt that to say we acted 'freely' - (in
- the philosopher's use,
whch is only faiktly related to the everyday ise) is 6 say only
that we acted *ot &-freely, in one or another of the many
heterogeneous ways of so acting (under duress, or what not).
Llke 'real', 'free' is only used to rule out the suggestion of some
or all of its recognized antitheses. As 'truth' is not a name for
a characteristic of assertions, so 'freedom' is not a name for a
characteristic of actions, but the name of a dimension in which
actions are assessed. In examining -
all the ways in which each
action may not be 'free', i.e. the cases in which it will not do
to say simply 'X did A', we may hope to dispose of the problem
of Freedom. Aristotle has. often been chidden for talking about
excuses or pleas and overlooking 'the real problem': in my own
case, it was when I began to see the injustice of this charge that
I first became interested in excuses.
 A PLEA F O R E X C U S E S
There is much to be said for the view that, phlosophcal
tradition apart, Responsibility would be a better candidate for
the role here assigned to Freedom. If o r h a r y language
- -
is to
be our guide, it is-to evade responsibility, or full responsibility,
that we most often make excuses, and I have used the word
myself in h s way above. But in fact 'responsibhty' too seems
not really apt in all cases: I do not exactly evade r&ponsibility
when I plead clumsiness or tactlessness, nor, often, when I plead
that I only did it unwibgly or reluctantly, and still less if I
plead that I had in the circumstances no choice: here I was
constrained and have an excuse (or justification), yet may
accept responsibility. ~t may be, then, that at least two key
terms, Freedom and Responsibhty,
- are needed: the relation
between them is not clear, and it may be hoped that the in-
vestigation of excuses will contribute towards its clarification.'
~ o m u c hthen,
, for ways in whch the study of excuses may
throw light on ethics. BL there are also reasons why it is
-

attractive subject methodologically, at least if we are to proceed

from 'ordmary language', that is, by examining what we should
say when, and so why and what we should mean by it. Perhaps
t h s method, at least as one philosophical method, scarcely
requires justification at present-too evidently, there is gold
in them thar lulls: more opportune would be a warning about -

the care and thoroughnes;keeded if it is not to fall into dis-

repute. I will, however, justify it very briefly.
First, words are our tools, and, as a minimum, we should use
clean tools: we should know what we mean and what we do
not, and we must forearm ourselves against the traps that
1 Another well-flogged horse in these same stakes is Blame. At least two

things seem confused together under this term. Sometimes when I blame
X for doing A, say for breaking the vase, it is a question simply or mainly
of my disapproval of A, breaking the vase, which unquestionably X did:
but sonletimes it is, rather, a question simply or mainly of how far I think X
responsible for A, which unquestionably was bad. Hence if somebody says
he blames me for something, I may answer by giving a justification, so that
he will cease to disapprove of what I did, or else by giving an excuse, so
that he will cease to hold me, at least entirely and in every way, responsible
for doing it.
 I 82 A PLEA F O R E X C U S E S
language sets us. Secondly, words are not (except in their own
little corner) facts or thngs: we need therefore to prise them
off the world, to hold them apart f r o n ~and against it, so that
we can realize their inadequacies and arbitrariness, and can re-
look at the world without bhnkers. Thrdly, and more hope-
f d y , our common stock of words embodies all the distinctions
men have found worth drawing, and the connexions they have
found worth marking, in the lifetimes of many generations:
these surely are llkely to be more numerous, more sound, since
they have stood up to the long test of the sunival of the fittest,
and more subtle, i t least in a~brdinaryand reasonably ~ractical
matters, than any that you or I are likely to thuJr up in our
arm-chairs of an afternoon-the most favoured aIternative
method.
In view of the prevalence of the slogan 'ordinary language', ;

and of such names as 'linguistic' or 'analytic' philosophy or

'the analysis of language', one t h g needs specially emphasizing
to counter misunderstandings. When we examine what we
should say when, what word; we should use in what situations,
we are looking again not merely at words (or 'meanings', what-
ever they may be) but also at the realities we use the words to
, talk about: we are using a sharpened awareness of words to
; sharpen our perception of, though not as the final arbiter of, the
I phenomena. For this reason I think it might be better to use,
for dus way of doing philosophy, some less misleading name
than those given above-for instance, 'hguistic phenomeno-
logy', only that is rather a mouthful.
Using, then, such a method, it is plainly preferable to in-
vestigate a field where orchary language is rich and subtle,
as it is in the pressingly practical matter of Excuses, but cer-
talnly is not in the matter, say, of Time. At the same time
we should prefer a field whch is not too much trodden into
bogs or tracks by traditional phlosophy, for in that case even
6
ordinary' language will often have become lnfected with the
jargon of extinct theories, and our own prejudices too, as the
upholders or imbibers of theoretical views, w d be too readily,
 A PLEA F O R EXCUSES 183

and often insensibly, engaged. Here too, Excuses form an

admirable topic; we can discuss at least clumsiness, or absence
of mind, or inconsiderateness, even spontaneousness, without
remembering what Kant thought, and so progress by degrees
even to discussing deliberation without for once remembering
Aristotle or self-control without Plato. Granted that our sub-
ject is, as already claimed for it, neighbouring, analogous or
germane in some way to some notorious centre of philo-
sophical trouble, then, with these two further requirements
satisfied, we should be certain of what we are after: a good site
forjield work in phlosophy. Here at last we should be able to
unfreeze, to loosen up and get going on agreeing about dis-
coveries, however small, and on agreeing about how to reach
agreement.I How much it is to be wished that similar field work
will soon be undertaken in, say, aesthetics; if only we could
forget for a while about the beautiful and get down instead to
thedainty and the dumpy.
There are, I know, or are supposed to be, snags in 'linguistic'
philosophy,
- -
which those not very familiar with it find, some-
kmes not without glee or relief,diunting. But with snags, as
with nettles, the thing to do is to graspthem-and to climb
above them. I wdl mention two in particular, over whch the
study of excuses may help to encourage us. The first is the snag
of Loose (or Divergent or Alternative) Usage; and the second
the crux of the Last Word. Do we all say the same, and only
the same, things in the same situations? Don't usages differ?
And, Why should what we all ordinarily say be the only or the
best or final way of putting it? Why should it even be true?
Well, people's usages do vary, and we do talk loosely, and
we do say different b g s apparently indifferently. But first, not
nearly as much as one would dunk. When we come down to
cases, it transpires in the very great majority that what we had
thought was our wanting to say different things of and in the
same situation was really not so-we had simply imagined the
All of which was seen and claimed by Socrates, when he first betook him-
self to the way of Words.
I 84 A PLEA F O R E X C U S E S
situation slightly differently: which is all too easy to do, because
of course no situation (and wc are dealing with i ~ n t l ~ i t r esitua-
d
tions) is ever 'completely' described. The more we imagine the
a
situation in detail, with background of story-and it is worth
-

employing the most idiosyncratic or, sometimes, boring means

to stimulate and to discipline our wretched imaginations-the
less we find we disagree - about what we should say. Neverthe-
less, sometimes we do ultimately disagree: sometimes we must
allow a usage to be, though appalling, yet actual; sometimes
we should genuinely use either or both of two different de-
scriptions. But why should this daunt us? All that is happening
is entirely explicable. If our usages disagree, then you use 'X'
where I use 'Y', or more probably (and more intriguingly)
your conceptual system is different from mine, though very
llkely it is at least equally consistent and serviceable: in short,
we can find ~ o h ywe disagree-you choose to classify in one
way, I in another. If the usage is loose, we can understand the
temptation that leads to it, and the distinctions that it blurs:
if there are 'alternative' descriptions, then the situation can be
described or can be 'structured' in two ways, or perhaps it is
- -

one where, for current purposes, the two alternatives come

down to the same. A disagreement as to what we should say
is not to be shied off, but to be pounced upon: for the explana-
tion of it can hardly fail to be illuminating. If we light on an
electron that rotates the wrong way, thit is a discovery, a
portent to be followed up, not a reason for chuclung physics:
and by the same token, a genuinely loose or eccentric talker is
a rare specimen to be prized.
- -

As practice hl learning to handle this bogey, in learning the

essential r u b r i ~ we
, could scarcely hope for a more promising
exercise than the study of excuses. Here, surcly, is just the sort
of situation where people will say 'almost anything', because
they are so flurried, or so anxious to get off. 'It was a mistake',
'It was an accident'-how readily these can appear indifferent,
and even be used together. Yet, a story Qr two, and everybody
will not merely agree that they are completely different, but
 A PLEA F O R EXCUSES 185

even discover for himself what the difference is and what each
means.
Then, for the Last Word. Certainly ordinary language has
no claim to be the last word, if there is such a thing. - It em-
bodies, indeed, somedung - better than the metaphysics
.
of the',
.

Stone Age, namely, as was said, the inherited experience and

acumen of many generations of men. But then, that acumen
has been concentrated primarily upon the practical business
of life. If a distinction works well for practical purposes in
ordinary life (no mean feat, for even ordinary life is full of hard
cases), then there is sure to be something in it, it will not mark
nothmg: yet this is likely enough to bc not the best way of
arranging things if our interests are more extensive or.inte1-
lectual than the ordinary. And again, that experience has been
derived only from the sources available to ordinary men
throughout
-
most of civilized history: it has not been fed from
the resources of the microsco~e L
and its successors. And it must
be added too, that superstition
- and error and fantasy of all kinds
do become iilcorporated in ordinary language - and even some-
times stand up to the survival test (only, when they do, why
should we not detect it ?). Certainly, then, ordinary language
is not the last word: in principle it can everywhere be supple-
,4nented and improved upon and superseded. Only remember,
it is the fint
J

For t h s ~ r o b l e mtoo the field of Excuses is a fruitful one.

I

Here is matter both contentious and practically important for

everybody, so that ordinary language is on its toes: yet also,
on its back it has long had a bigger flea to bite it, in the shape
1 You have a donkey, so have I, and they graze in the same field. The day
comes when I conceive a dislike for mine. I go to shoot it-, draw a bead on it,
fire: the brute falls in its tracks. I inspect the victim, and find to my horror that
it is your donkey. I appear on your doorstep with the remains and say-what?
'I say, old sport, I'm awfully sorry, &c., I've shot your donkey by dccident'?
O r 'by mistake'? Then again, 1x0 to shoot my donkey as before, draw a bead
on it, fire-but as I do so, the beasts move, and to my horror yours falls. Again
the scene on the doorstep-what do I say? 'By mistake'? Or 'by accident'?
And forget, for once and for a while, that ot!rer curious question 'Is it
true?' M a y we?
 186 A PLEA FOR E X C U S E S
of the Law, and both again have lately attracted the attentioils
of yet another, and at least a healthly growing, flea, in the
shape
- of psychology.
- - - In the law a constant stream of actual
cases, more novel and more tortuous than the mere imagina-
tion could contrive, are brought up for derision-that is, for-
mulae for docketing them must somehow be found. Hence it
is necessary first to be careful with, but also to be brutal with,
to torture, to fake and to override, ordinary language: we
cannot here evade or forget the whole affair. (In ordinary life
we dismiss the puzzles that crop up about time, but we cannot
do that indefinitely in physics.) Psychology likewise produces
novel cases, but it also produces new methods for bringing
phenomena under observation and study: moreover, unlke
the law, it has an unbiased interest in the totality of them and
is unpressed for decision. Hence its own special and constant
needio supplement, to revise and to supersede the classifications
of both ordinary life and the law. W e have, then, ample
material for practice in learning to handle the bogey of the
Last Word, however it should be handled.
Suppose, then, that we set out to investigate excuses, what
are the methods and resources initially available? Our object
is to imagine the varieties of situation in which we make
Jexcuses, and to examine the expressions used in malung them.
If we have a lively imagination, together perhaps with an
ample experience of dereliction, we shall go far, only we need
system: I do not know how many of you keep a list of the
lunds of fool you make of yourselves. It is advisable to use
systematic aid;, of whch there would appear to be three at
least. I list them here in order of availability to the layman.
First we may use the dictionary-qui te a concise one will do,
but the use must be thorough. Two methods suggest themselves,
both a little tedious, but repaying. One is to read the book
through, listing all the words that seem relevant; this does not
take as long as many suppose. The other is to start with a widish
selection of obviously relevant terms, and to consult the dic-
tionary under each: it will be found that, in the explanations
 A PLEA FOR E X C U S E S 187

of the various meanings of each, a surprising number of other

terms occur, whch are germane though of course not ofien
synonymous. W e then look up each of these, bringing in more
for our bag from the 'definitions' given in each case; and when
we have continued for a little, it will generally be found that
the family circle begins to close, until ultimately it is complete
and we come only upon repetitions. This method has the
advantage of groupingthe terms into convenient clusters-but
of course a good dial will depend upon the comprehensiveness
of our initial selection.
Working the dictionary, it is interesting to find that a h g h
percentage of the terms connected with excuses prove to be
adverbs, type of word which has not enjoyed so large a share
of the philosophcal limelight as the noun, substantive or
adjective,
- and the verb: this is natural because, as was said, the
tenor of so many excuses is that I did it but only in a way, not
just flatly llke that-i.e. the verb needs modifying. - Besides
adverbs, however, there are other words of all kinds, including-
numerous abstract nouns, 'misconception,' 'accident', 'purpose',
and the U e , and a few verbs too, whlch often hold key posi-
tions for the grouping of excuses into classes at a high level
Ccouldn't help', 'didn't mean to', '&dn't realize', or again
'intend', and 'attempt'). In connexion with the nouns another
neglected class of words is prominent, namely, prepositions.
Not merely does it matter considerably which preposition,
ofien of several, is being used with a given substantive, but
further the prepositions deserve study on their own account.
For the question suggests itself, Why are the nouns in one,
group by 'under', in another by 'on', in yet another
by 'by7 or 'through' or 'from' or 'for' or 'with', and so on?
It wdl be disappointing if there prove to be no good reasons
-
for such groupings.
Our second source-book wlll naturallv be the law. his w d
d

provide us with an immense miscellany of untoward cases, and

also with a useful list of recognized pleas, together with a good
deal of acute analysis of both. No one who tries thls resource
r88 A PLEA FOR EXCUSES
wlll long be in doubt, I think, that the common law, and in
particular the law of tort, is the richest storehouse; crime and
contract contribute some special additions of their own, but
tort is altogether more comprehensive and more flexible. But
even here, and still more with so old and hardened a branch
of the law as crime, much caution is needed with the arguments
of counsel and the dicta or decisions of judges: acute though
these are, it has always to be remembered that, in legal cases-
(I) there is the overridmg requirement that a decision be
reached, and a relatively black or w h t e decision-guilty or
not guilty-for the plaintiff or for the defendant;
(25 there is the general requirement that the charge or
action and the pleadings be brought under one or another of
the heads and procedures that have come in the course of
history to be accepted by the Courts. These, though -
fairly
numerous, are stir few and stereotyped in comparison with
the accusations and defences of daily life. Moreover conten-
tions of many kinds are beneath the law, as too trivial, or out-
side it, as too purely
-
moral-for example, inconsiderateness;
(3) there is the general requirement that we argue from
and abide by precedents. The value of t h s in the law is
unquestionable, but it can certainly lead to distortions of
ordinary beliefs and expressions.
For such reasons as these, obviously closely connected and
stemming from the nature and function of Aelaw, practising
lawyers and jurists are by no means so careful as they might be
-

to give to our ordinary expressions their ordinary meanings

and applications. There is special pleading and evasion, stretch-
ing and strait-jacketing, besides the invention of technical
terms, or technical senses for common terms. Nevertheless,
it is a perpetual and salutary surprise to discover how much
is to be learned from the law; and it is to be added that if a
distinction drawn is a sound one, even though not yet recog-
nized in law, a lawyer can be relied upon
-
to take note of it, for
i t may be dangerous not to-if he does not, h s opponent may.
 A PLEA FOR EXCUSES 189

Finally, the third source-book is psychology, with whch

I include such studies as anthropology and animal behaviour.
Here I speak with even more trepidation than about the Law.
But h s a t least is clear, that some varieties of behaviour, some
ways of acting or explanations of the doing of actions, are here
noticed and classified whch have not been observed or named
by ordnary men and hallowed by ordinary language, though
perhaps they often might have been so if they had been of more
practical importance. There is real danger in contempt for the
6.
jargon' of psychology, at least when it sets out to supplement,
and at least sometimes when it sets out to supplant, the language
of ordmary life.
With these sources, and with the aid of the imagination, it
wdl go hard if we cannot arrive at the meanings of large
numbers of expressions and at the understanhe; and classi-
fication of larg; numbers of 'actions'. Then we shall compre-
hend clearly much that, before, we only made use of adhoc.
Definition, I would add, explanatory definition, should stand
! / - ' h g hamong our aims: it is not enough to show how clever
we are by showing how obscure everything is. Clarity, too, 1
know, has been said to be not enough: but perhaps it will be
time to go into that when we are within measurable distance of
achieving clarity on some matter.

So much for the cackle. It remains to make a few remarks,

not, I am afraid, in any very coherent order, about the types of
significant result to be obtained and the more general lessons
to be learned from the study of Excuses.
I. No modijcation without aberration. When it is stated that
X did A, there is a temptation to suppose that given some,
indeed perhaps any, expression modifying the verb we shall
be entitled to insert either it or its opposite or negation in our
statement: that is, we shall be entitled to ask, typically, 'Did X
do A Mly or not Mly?' (e.g. 'Did X murder Y voluntarily or
involuntarily?'), and to answer one or the other. O r as a mini-
mum it is supposed that if X did A there must be at least one
90
I A PLEA F O R E X C U S E S
modifying expression that we could, justifiably and informa-
tively, insert with the verb. In the great majority of cases of the
use of the great majority of verbs ('murder' perhaps is not one
of the majority) such suppositions are quite unjustified. The
natural economy of language dictates that for the standard case
covered by any normal verb-not, perhaps, a verb of omen
such as 'murder', but a verb llke 'eat' or 'kick' or 'croquet'-
no modifying expression is required or even permissible. Only
if we do the action named in some special
- way or circumstances,
different from those in which such an act is naturally done (and
of course both the normal and the abnormal differ according
to what verb in particular is in question) is a modifying ex-
pression called for, or even in order. I sit in my chair, in the
-
usual way-I am not in a daze or influenced by threats or the
U e : here, it w d not do to say either that I sat in it intentionally
or that I did not sit in it intentioilally,' nor yet that I sat in it
automatically or from habit or what you will. It is bedtime, I
am alone, I yawn: but I do not yawn involuntarily (or volun-
tarily!), nor yet deliberately. T o yawn in any such peculiar
way is just not to just yawn.
2. Limitation of application. Expressions modifying verbs,
typically adverbs, have limited ranges of application. That is,
given any adverb of excuse, such as 'unwittingly' or 'spon-
taneobsly' or 'impulsively', it will not be found that it makes
good sense to attach it to any and every verb of 'action' in any
and every context: indeed, it will often apply only to a com-
paratively narrow range of such verbs. S o m e h n g in the lad's
upturned face appealed to h m , he threw a brickat it-'spon-
taneously'? The interest then is to discover why some actions
can be excused in a articular way but not others, ~ a r t i c u l a r l ~
perhaps the latter.2 Ths w d largely elucidate the meaning
of the excuse, and at the same time will illuminate the

Caveat or hedge: of course we can say 'I did not sit in it "intentionally"'
as a way simply of repudiating the suggestion that I sat in it intentionally.
For we are sometimes not so good at observing what we crzn't say as what
we can, yet the first is pretty regularly the more revealing.
 A PLEA FOR E X C U S E S 191

characteristics typical of the group of 'actions' it picks out: very

often too it will throw light
-
on some detail of the machmery
of 'action' in general (see 4, of on our standards of acceptable
conduct (see 5 ) . It is specially important in the case of some
of the terms most favoured by philosophers or jurists to realize
that at least in ordinary speech (disregarding back-seepage of
jargon) they are not used so universally or so dichotomi~ticall~.
For example, take 'voluntarily' and 'involuntarily': we may
join the army or make a gift voluntarily, we may hccough or
make a small gesture involuntarily, and the more we consider
further actions whch we might naturally be said to do in
-

either of these ways, the more circumscribed and unllke each

other do the two dasses become, until we even doubt whether
there is any verb with which both adverbs are equally in place.
Perhaps there are some such; but at least sometimes when we
may &mk we have found one it is an illusion, an apparent
exception that really does prove the rule. 1 can perhaps 'break
a cup' voluntarily, if that is done, say, as an act of self-im-
poverishment: and I can perhaps break another involuntarily,
if, say, I make an involuntary movement which breaks it.
Here, plainly, the two acts described each as 'breaking a cup'
are really very different, and the one is similar to acts typical of
the 'voluntary' class, the other to acts typical
- -
of the 'involun-
tary' class.
3. The importance of Negations artd Opposites. 'Voluntarily'
and 'involuntarily', then, are not opposed in the obvious sort
of way that they are made to be in phdosophy or jurisprudence.
The 'opposite', or rather 'opposites', of 'voluntanly' might be
' - - -

under constraint' of some sort, duress or obligation or in-

fluence:' the opposite of 'involuntarily' might be 'deliberately'
or 'on purpose' or the lrke. Such divergences in opposites
indicate that 'voluntarily' and 'involuntarily', in spite of their
apparent connexion, are fish from very different kettles. In
general, it will pay us to take nothing for granted or as obvious
I But remember, when I sign a cheque in the normal way, I do not do so
cither 'voluntarily' or 'under constraint'.
 192 A PLEA F O R E X C U S E S
about negations and opposites. It does not pay to assume that
a word must have an opposite, or one opposite, whether it
is a 'positive' word hke 'wilfully' or a 'negative9 word U e
'inad;ertently'. Rather, we shodd be a s G g ourselves such
questions as why there is no use for the adverb 'advertently'.
For above all it will not do to assume that the 'positive' word
must be around to wear the trousers; commorJy enough the
'negative' (loolung) word marks the (positive) abnormality
whlle the 'positive' word, ifit exists, merely serves to rule out
the suggestion of that abnormality. It is natural enough, in view
of what was said in (I) above, for the 'positive' word not to be
found at all in some cases. I do an act A, (say, crush a snail)
inadvertently if, in the course of executing by means of move-
ments of my b o d y parts some other act &
(say, in waking
down the public path) I fail to exercise such meticulous super-
vision over the courses of those movements as would have
been needed to ensure that they did not bring about the un-
toward event (here, the impact i n the snail). I By claiming that -

A, was inadvertent we place it, where we imply it belongs, on

ths special level, in a class of incidental happenings
- whch must
occur in the doing of any physical act. ~ o l i fthe
t act out of this
class, we need and possess the expression 'not . . .inadvertently9:
'advertently', if used for this purpose, would suggest that, if
the act was not done inadvertently, then it must have been done
noticing what I was doing, whch is far from necessarily the
case (e.g. if I did it absent-mindedly), or at least that there is
something in common to the ways of doing all acts not done
inadvertently, which is not the case. Again, there is no use for

Or analogously: I do an act Al (say, divulge my age, or imply you are a

liar), inadvertently if, in the course o f executing by the use o f some medium o f
communication some other act A, (say, reminiscing about my war service)
I fail to exercise such meticulous supervision over the choice and arrangement
of the signs as would have been needed to ensure that. . . , It is interesting to
note how such adverbs lead parallel lives, one in connexion with physical
actions ('doing') and the other in connexion with acts o f communication
('saying'), or sometimes also in co~mexionwith acts o f 'thmking' ('inadver-
tently assumed').
 A PLEA FOR E X C U S E S I93

'advertently' at the same level as 'inadvertently': it1 passing the

-

butter I do not knock over the cream-jug, though I do (in-

advertently) knock over the teacup-yet I do not by-pass the
cream-jug advertently: for at this level, below supervision in
detail, anything that we do is, if you llke, inadvertent, though
we only call it so, and indeed oily call it somedung we have
done, if there is something untoward about it.
A further point of interest in studying so-called 'negative'
terms is the manner of their formation. Why are the words
in m e group formed with un- or in-, those in another with
-less ('aimless', 'reckless', 'heedless', gtc.), and those in another
6
with mis- ('mistake', 'misconception', misjudgemefit', &c.) ?
Why carelessly but inattentively? Perhaps care and attention,
so often linked, are rather different. Here are remunerative
exercises.
4. The machinery ofaction. Not merely do adverbial expres-
sions pick out classes of actions, they also pick out the internal
detail of the machnery of doing actions, or the departments
into whch the business of doing actions is organized. There is
for example the stage
- -
at which we have actually to carry out
some action upon whch we embark-perhaps we have to
make certain bodily movements or to make a speech. In the
course of actually doing these things (getting weaving) we have
to pay (some) attention to what we are doing and to take
(some) care to guard against (likely) dangers: we may need
to use judgement or tact: we must exercise sufficient control
over our bodily- -parts: and so on. Inattention, carelessness,
errors ofjudgement, tactlessness, clumsiness, all these and others
are ills (with attendant excuses) whch affect one specific stage
in the macliinery of action, the executive stage, the stage where
we muff it. But thcre are many other departments in tllc busi-
.-

ncss too, each of which is to be traced and mapped through its

cluster of appropriate verbs and adverbs. Obviously thcre are
departments of intelligence and planning, of decision and
resolve, and so on: but I shall mention one in particular, too
often overlooked, where troubles and excuses abound. It
 A PLEA F O R E X C U S E S
happens to us, in military life, to be in receipt of excellent
intelligence, to be also in self-conscious possession of excellent
principles (the five golden rules for winning victories), and yet
to hit upon a plan of action which leads to disaster. One way
in which this can happen is through failure at the stage of
appreciation of the situation, that is at the stage where we are
required to cast our excellent intelligence - into such a form,
under such heads and with such weights attached, that our
-

equally excellent principles can be brought to bear on it pro-

perly, in a way to yield the right answer.' So too in real, or
rather civilian, life, in moral or -practical affairs, we can know
the facts and yet look at them mistakenly or perversely, or not
fully realize or appreciate somethmg, -
or even be under a total
misconception. Many expressions of excuse indicate failure
at this ~ a r t i c u l a r ltricky
~ stage: even thoughtlessness, incon-
siderateness, lack of imagination, are perhaps less matters of
failure in intelligence or planning than might be supposed, and
more matters- of failure to appreciate the situation. A course
of E. M. Forster and we see things differently: yet perhaps we
know no more and are no cleverer.
5 . Standards ofthe unacceptable. It is characteristic of excuses
to be 'unacceptable': given, I suppose, almost any excuse,
there will be cases of such a lund o; of such gravity that 'we
will not acceptyit. It is interesting to detect the stardards and
codes we thus invoke. The extent of the supervision we exer-
cise over the execution of any act can never be quite unlimited,
and usually is expected to fail within fairly defkte limits ('due
care and attention') in the case of acts of some general kind,
though of course we set very different limits in different cases.
W e may plead that we trod on the snail inadvertently: but not
on a baby-you ought to look where you are putting your
great feet. Of course it was (really), - - if you llke, inadvertence:
but that word constitutes a plea, which is. not going to be
W e know all about how to do quadratics: we know all the needful facts
about pipes, cisterns, hours and plumbers: yet we reach the answer '39 men'.
W e have failed to cast our facts correctly into mathematical form.
 A PLEA FOR EXCUSES 195

allowed, because of standards. And if you try it on, you wdl be

subscribing to such dreadful standards that your last state will
be worse than your first. Or again, we set different standards,
and wdl accept different excuses, in the case of acts whch are
rule-governed, Lke spelling, and whch we are expected
absolutely to get right, from those we set and accept for less
stereotyped actions:-a wrong spelling may be a slip, but hardly
- -

an accident, a winged beater may be ai accident, but hardly

a slip.
6. Combination, dissociation and complication. A belief in
-

opposites and dichotomies encourages, among other things,

a blindness to the combinations and dissociations of adveibs
that are possible, even to such obvious facts as that we can act
at once on impulse
- and intentionally, or that we can do an
action intentionally yet for all that not deliberately, still less
on purpose. W e walk along the cliff, and I feel a sudden
impulse to push you over, which I promptly do: I acted on im-
pulse, yet I certainly intended to push you over, and may even
have devised a little ruse to acheve it: yet even then I did not
act deliberately, for I did not (stop to) ask myself whether to
do it or not.
It is worth bearing in mind, too, the general rule that we
must not expect to find simple labels foi complicated cases.
If a mistake results in an accident, it wdl not do to ask whether
'it' was an accident or a mistake, or to demand some briefer
description of 'it'. Here the natural economy of language
operates: if the words already available for simile cases s-&Ce
in combination to describe a complicated case, there wdl be
need for special reasons before a special new word is invented
for the complication.
-
Besides, however well-equipped
- - -
our
language, it can never be forearmed against all possible cases
that may arise and call for description: fact is richer than diction.
7. Regina v. Finney. Often the complexity and difficulty of
a case is considerable. I will quote the case of Regina v. Finney:'
A somewhat distressing kvourite in the class that Hart used to conduct
with me in the years soon after the war. The italics are mine.
196 A PLEA F O R E X C U S E S
Shrewsbury Assizes. I 874. xz Cox 625.
Prisoner was indicted for the manslaughter of Thomas W a h s .
The Prisoner was an attendant at a lunatic asylum. Being in charge
of a lunatic, who was bathing, he turned on hot water into the bath,
and thereby scalded him to death. The facts appeared to be truly set
forth in the statement of the prisoner made before the committing
magistrate, as follows: 'I had bathed W a h s , and had loosed the
bath out. I intended putting in a clean bath, and asked Watkins if he
would get out. At ths time my attention tvas dratvn to the next bath
by the new attendant, who was asking me a question; and m y
attention was lakenfrom the bath where Watkins was. I put my hand
down to turn water on in the bath where Thonlas W a b was.
I did not itttend to turn the hot water, and I made a mistake in the tap. I did
not Know what I had done until I heard Thomas Watkins shout out;
and I did not find m y mistake out till I saw the steam from the water.
You cannot get water in this bath when they are drawing water at
the other bath; but at other times it shoots out llke a water gun when
the other baths are not in use. . . .'
(It was proved that the lunatic had such possession of his faculties
as would enable him to understand what was said to him, and to
get out of the bath.)
A. Yotrng (for Prisoner). The death resulted frorn accident. There
was no such culpable negligence on the part of the prisoner as WIN
support this indictment. A culpable mistake, or some degree of
culpable negligence, causing death, will not support a charge of
manslaughter; unless the negligence be so gross as to be reckless.
(R. v. Noakes.)
Ltrsh,]. To render a person liable for neglect ojduty there must be
such a degree of culpability as to amount to gross negligence on his
part. If you accept the prisoner's own statement, you find no such
amount of negligence as would come w i b this dehition. It is not
every little trip or mistake that will make a man so liable. It was the
duty of the attendant nor to let hot water into the bath while the
patient was therein. According to the prisoner's own account, he
did not believe that he was letting the hot water in while the deceased
remained there. The lunatic was, we have heard, a inan capable of
getting out by himself and of understanding what was said to him.
He was told to get out. A new attendant who had coine on this day,
was at an adjoining bath and he took off the prisoner's attention. Now,
 A PLEA FOR EXCUSES I97
if the prisoner, knowing that the man was in the bath, had turned
on the tap, and turned on the hot instead of the cold water, I should
have said there was gross negligence; for he ought to have looked
to see. But from h s own account he had told the deceased to get out,
and thought he hadgot out. If you thtnk that indicates gross carelessness,
then you should find the prisoner guilty of manslaughter. But if
you thtnk it inadvertence not amounting to culpability-i-e., what is
properly termed an accident-then the prisoner is not liable.
Verdict, Not pilty.
In this case there are two morals that I will point:
(i) Both counsel and judge make very free use of a large
number of terms of excuse, using several as though they
were, and even stating them to be, indifferent or equivalent
when they are not, and presenting as alternatives those that
are not.
(ii) It is constantly difficult to be sure what act it is that
counsel or judge is suggesting might be qualified by what
r
expression ot excuse.
The learned judge's concluding direction is a paradigm of these
faults.= Finney, by contrast, stands out as an evident master of
the Queen's English. He is explicit as to each of his acts and
states, mental and physical: he uses different, and the correct,
adverbs in connexion with each: and he makes no attempt to
boil down,
8. Small distinctions, and bg too. It should go without saying
that terms of excuse are not equivalent,
- and that it matters
w h c h we use: we need to distinguish
-
inadvertence not merely
from (save the mark) such dungs as mistake and accident, but
from such nearer neighbours as, say, aberration and absence
Not but what he probably manages to convey his meaning somehow or
other. Judges seem to acquire a knack of conveying meaning, and even cany-
ing conviction, through the use of a pithy Anglo-Saxon which sometimes has
literaIly no meaning at all. Wishing to distinguish the case of shooting at a
post in the beIief that it was an enemy, as not an 'attempt', from the case of
picking an ernpry pocket in the belief that money was in it, which is an
'attempt', the judge expIains that in shooting at the post 'the man is never on
the thing at all'.
198 A PLEA F O R E X C U S E S
of mind. By imagining cases with vividness and fullness we
should be able to decide in whch precise terms to describe, say,
Miss Plimsoll's action in writing, so carefully, 'DAIRY' on her
fine new book: we should be able to distinguish between sheer,
mere, pure, and simple mistake or inadvertence. Yet un-
fortunately, at least when in the grip of thought, we fail not
merely at these stiffer hurdles. W e equate even-I have seen it
done-'inadvertently' with 'automatically': as though to say
I trod on your toe inadvertently means to say I trod on it
autonlatically. Or we collapse succumbing to temptation into
losing control of ourselves-a bad patch, ths, for telescoping.1
All t h ~ sis not so much a lesson from the study of excuses as
the very object of it.
9. T h e exact phrase and itsplace in the sentence. It is not enough,
either, to attend simply ta the 'key' word: notice must also be
taken of the full and-exact form of the expression used. In
considering mistakes, we have to consider seriatim 'by mistake',
'owing to a n~istakc','mistakenly', 'it was a mistake to', 'to
make a mistake in or over or about', 'to bc mistaken about',
and so on: in considering purpose, we have to consider 'on',
'with the', 'for the', &c., besides 'purposeful', 'purposeless',
and tile Ue. These varying expressions may function quite
differently-and usually do, or why should we burden our-
selves with more than one of them?
Care must be taken too to observe the precise position of an
adverbial expression in the sentence. This should of course
indicate what verb it is being used to modify: but more than
that, the position can also affect the sense of the expression,
Plato, I suppose, and after him Aristotle, fastened this confusion upon us,
as bad in its day and way as the later, grotesque, confusion of inoraI weakness
with weakness of will. I am very partial to ice cream, and a bombe is served
divided into segments corresponding one to one with the persons at High
Table: I am tempted to help myself to two segments and do so, thus succunlb-
ing to temptation and even conceivably (but why necessarily?) going against
my principles. But do I lose controI of myself? D o I raven, do I snatch the
morsels from the dish and wolf them down, impervious to the consternation
of my colleagues? Not a bit of it. W e often succumb to temptation with calm
and c v a ~with finesse.
 A PLEA F O R E X C U S E S I99

i.e. the way in which it modifies that verb. Compare, for

example:
I

a, He clumsily trod on the snail.

a2 Clumsily he trod on the snail.
6, He trod clumsily on the snail.
b, He trod on the snail clumsily.
Here, in ol and n2 we describe his treading on the creature at
all as a piece of clumsiness, incidental, we imply, to his per-
formance of some other action: but with 6, and 6, to tread on
it is, very likely, his aim or policy, what we criticize is his
execution of the feat.' Many adverbs, though far from all (not,
for example, 'purposely') are used in these two typically
different ways.
10. The style ofperforntonce. With some adverbs the distinc-
tion between the two senses referred to in the last paragraph
is carried a stage further. 'He ate his soup deliberately' may
mean, llke 'He deliberately ate his soup', that his eating his
soup was a deliberate act, one perhaps that he thought would
annoy somebody, as it would more commonly if he deliber-
ately ate my soup, and which he decided to do: but it will often
mean that he wint through the performance of eating h s soup
-

in a noteworthy manner or stykc-pause after each mouthful,

careful choice of point of entry for the spoon, suclung of
moustaches, and so on. That is, it will mean that he ate tuith
deliberation rather than 4 e r deliberation. The style of the
performance, slow and unhurried, is understandably called
'deliberate' because each movement has the typical look of a
deliberate act: but it is scarcely being said that the makmg of
each motion is a deliberate act or that he is 'literally' deliberating.
Ths case, then, is more extreme than that of 'clumsily', which
does in both uses describe literally a manner of performing.
As a matter of fact, most of these examples can be uriderstood the other
way, especially if we allow ourselves inflexions of the voice, or commas, or
contexts. a, might be a poetic inversion for 6:, b,, perhaps with commas
round the 'clumsily', might be used for a,: and so on. S t d , the two senses are
clearly enough distinguishable.
200 A PLEA F O R EXCUSES
It is worth watching out for this secondary use when scrutin-
izing any particular adverbial expression : when it definitely
does not exist, the reason is worth inquiring into. Sometimes
it is very hard to be sure whether it does exist or does not: it
does, one would think, with 'carelessly', it does not with
'inadvertently', but docs it or does it not with 'absent-mindedly'
or 'aimlessly'? In sorne cases a word alun to but distinct from
the primary adverb is used for this special role of describing
a style of performance: we use 'purposefully' in this way, but
never 'purposely7.
11. W h a t modifies wlzat? The Judge in Regina v. Finney does
not make clear what event is being excused in what way. 'If
you t M r that indicates gross carelessness, then. . . . But if you
think it inadvertence not amounting to culpability-i.e. what
is properly called an accident-then. . . .' Apparently he means
that Finney may have turncd otl the hot tap inadvertently:' does
he mean also that the tap may have been turned accidentally,
or rather that Watkins ,,ray have GEEn scalded and killed accident-
ally? And was the carelesst~essin turning the tap or in thinking
Watluns had got out? Many disputes as to what excuse we
should properly use arise becausewe will not trouble to state
explicitly what is being excused.
T o do so is all the more vital because it is in principle always
open to us, along various lines, to describe orrefer t o 'what I
did' in so many different ways. This is altogether too large a
theme to elaborate here. Apart from the more general and
obvious problems of the use of 'tendentious' descriptive terms,
-

there are many special problenls in the case of

What Finney says is different: lle says he 'made a nistake in the tap'. This
is the basic use of 'mistake', where we simply, and not necessarily accountably,
take the wrong one. Finney here artenlpts to account for his rnistakc, by
saying that his attelltio~lwas distracted. But supposc the order is 'Right turn'
and I turn left: no doubt the scrgeant will insinuate that my attelltion was
distracted, or that I canrlot distinguish rny right fro111 my left-but it was not
and I can, this was a simple, pure mistake. As often happens. Neither I nor the
sergeant will suggest that there was any accident, or any inadvertence either.
If Finney had turned the hot tap inadvertently, then it would have been
knocked, say, in reaching for the cold tap: a different story.
 A PLEA F O R EXCUSES 201

'actions'. Should we say, are we saying, that he took her money,

or that he robbed her? That he knocked a ball into a hole, or
that he sank a putt? That he said 'Done', or that he accepted an
offer? How far, that is, are motives, intentions and conventions
to be part of the description of actions? And more especially
here, what is an or one or the action? For we can generally split
up what might be named as one action in severaldistinct ways,
into different stretches or or stages. Stages have already
been mentioned: we can dismantle the i n a c h e r y of the act,
and describe (and excuse) separately the intelligence, the appre-
ciation, the planning, the decision, the execution and so forth.
Phases are rather different: we can say that he painted a picture
or fought a campaign, or else weecan say that first he laid on,
h s stroke of paint and then that, first he fought this action
and then that. Stretches are different again: a single term de-
scriptive of what he did may be made to cover either a smaller
or a larger
-
stretch of events, those excluded by the narrower
description being then called 'consequences' or 'results' or
'effects' or the b e of his act. So here we can describe Finney's
act either as turning on the hot tap, which he did by mistake,
with the result that Watkins was scalded, or as scaldmg Wathns,
which he did ?rot do by mistake.
It is verv evident that the problems of excuses and those of
4 I

the different descriptions of actions k e throughout bound up

with each other.
12. Trailitzg clouds ofetymology. It is thcse considerations that
bring us up so forcibly against some of the most difficult words
in the whole story of Excuses, such words as 'result', 'effect', and
'consequence', or again as 'intention', 'purpose', and 'motive'.
I will mention two points of method which are, experience -

has convinced me, indispensable aids at these levels.

One is that a word never-well, hardly ever-shakes off its
etymology and its formation. In spite of all changes in and
extensions of and additions to its meanings, and indeed rather
pervading and governing these, there will still persist the old
idea. Iq an accident sometlung befalls: by mistake you take the
 A PLEA F O R EXCUSES
wrong one: in error you stray: when you act deliberately you
act after weighing it up (not after thinking out ways and means).
It is worth asking ourselves whether we know the etymology
of 'result' or of 'spontaneously', and worth remembering that
'unwillingly' and 'involuntarily' come from very different
sources.
And the second point is connected with h s . Going back
into the hstory of a word, very often into Latin, we come back
pretty commonly to pictures or models of how things happen
or are done. These models may be fairly sophisticated and
recent, as is perhaps the case with 'motive' o r 'impulse', but
one of the commonest and most primitive types of model is
one whch is apt to baffle us through its very naturalness and
simplicity. W e take some very simp k action, like shoving a stone,
usually as done by and viewed by oneself, and use this, with the
features distinguishable in it, as our model in terms of w h c h
to talk about other actions and events: and we continue to do
so, scarcely realizing it, even when these other actions are pretty
remote and perhaps much more interesting to us in their own
right than the acts originally used in constructing the model
ever were, and even when the model is really distorting the -

facts rather than helping us to observe them. In primitive cases

we may get to see clearly the differences between, say, 'results',
'effects', and 'consequences', and yet discover that these differ-
ences are no longer- clear, and the terms themselves no longer -

of real service to us, in the more complicated cases where we

had been bandying them about most freely. A model must be
recognized for what it is. 'Causing', I suppose, was a notion
taken from a man's own experience of doing simple actions,
and by primitive man eveyevent was construed in terms ot
this model: every event has a cause, that is, every event is an
action done by somebody-if not by a man, then by a quasi-
man, a spirit. When, later, events which are not actions are
realized to be such, we still say that they must be 'caused', and
the word snares us: we are struggling to ascribe to it a new,
unanthr~pomor~hic meaning, yet constantly, in searchmg for
 A PLEA FOR EXCUSES 203

its analysis, we unearth and incorporate the lineaments of the

ancient-model. As happened even-to Hume, and consequently
to Rant. Examining such a word historically, we may well
find that it has been extended to cases that have by now too
tenuous a relation to the model case, that it is a source of con-
fusion and superstition.
-

There is too another danger in words that invoke models,

-

half-forgotten or not. It must be remembered that there is no

necessity whatsoever that the various models used in creating
our vocabulary, primitive or recent, should all fit together
neatly as parts into one single,
-
total model or scheme of, for
instance, the doing of actions. It is ~ossible,and indeed highly
likely, that our assortment of models will include some, or
many, that are overlapping, conflicting, or more generally
simply disparate.'
13. In spite of the wide and acute observation of the pheno-
mena of action embodied in ordinary speech, modern scientists
have been able, it seems to me, to reveal its inadequacy at
numerous points, if only because they have had access to more
-

comprehensive data and have studied them with more catholic

and dispassionate interest than the ordinary man, or even the
lawyer, has had occasion to do. I will conclude with two
examples.
Observation of animal behaviour shows that regularly,
when an animal is embarked on some recognizable pattern of
behaviour but meets in the course of it with an insuperable
This is by way of a general warning in philosophy. It seems to be too,
readily assunled that if we can only discover the true meanings of each of a
cluster ofkey terms, usually historic terms, that we use in some particular field
(as, for example, 'right', 'good' and the rest in morals), then it must without
question transpire that each will fit into place in some single, interlocking,
consistent, conceptual scheme. Not only is there no reason to assume this, but
all historical probability is against it, especially in the case of a language de-
rived from such various civilizations as ours is. W e may cheerfully use, and
with weight, terms which are not so much head-on incompatible as simply
disparate, which just do not fit in or even 0x1. Just as we cheerfully subscribe to,
or have the grace to be torn between, simply disparate ideals-why rnust there
be a conceivable amalgam, the Good Life for Man?
204 A PLEA F O R EXCUSES
obstacle, it will betake itself to energetic, but quite unrelated,
activity of some wild kind, such as S t a n h g on its head. T h s
phenomenon is called 'displacement behaviour' and is well
identifiable. If now, in the lightaoft h , we look back at o r h -
ary human life, we see that displacement behaviour bulks quite
-

large in it: yet we have apparently no word, or at least no dear

and simple
- word, for it. If, when thwarted, we stand on our
heads or wiggle our toes, then we are not exactly just standing
on our heads, don't you know, in the ordinary way, yet is
there any convenient adverbial expression we can insert to do
the trick ? 'It1 desperation' ?
Take, again, 'compulsive' behaviour, however exactly
psychologi~ts define it, compulsive washmg for example.
There are of course hnts in ordinary speech that we do t G g s
in this way-'just feel I have to', 'shouldn't feel comfortable
unless I did', and the like: but there is no adverbial expression
satisfactorily pre-en~pted for it, as 'compulsively' is.- his is
understandable enough, since compulsive behaviour, like dis-
placement behaviour, is not in general going to be of great
practical importance.

Here I leave and commend the subject to you.

 I F S AND CANS'

A R E cans constitutionally iffy? Whenever, that is, we say that

we can do something, or could do somethmg, or could have
done something, is there an if in the offing-suppressed, it may
be, but due nevertheless to appear when we set out our sentence
in full or when we give an explanation of its meaning?
Again, if and when there is an ifclause appended to a main
clause which contains a can or could or could have, what sort or
an ifis it? What is the meaning of the if, or what is the effect or
the point of combining this if-clause with the main clause?
These are large questions, to which philosophers, among
them some whom I most respect, have given small answers: and
it is two such answers, given recently by English philosophers,
that I propose to consider. Both, I believe, are mistaken, yet
somedung is to be learned from examining them. In
sophy, there are many mistakes that it is no hsgrace to have
made: to make a first-water, ground-floor mistake, so far from
being easy, takes one (one) form of phtlosophical genius?
Many of you will have read a short but justly admired book
written by Professor G. E. Moore of Cambridge, which is
called simply Ethics. In it, there is a point where Moore, who is
engaged in discussing Right and Wrong, says that if we are to
discuss whether any act that has been done was right or wrong
then we are bound to discuss what the person concerned could
have done instead of what he did in fact do. And this, he tMrs,

Reprinted from the Proceedirrgs of the British Academy, 1956, by courtesy

o f the editor.
Plato, Descartes, and Leibniz all had this form of genius, besides o f course
others.
206 IFS AND CANS
may lead to an entanglement in the problem, so-called, of Free
Will: because, though few would deny, at least expressly, that
a man could have done s o m e h g other than what he did
if
actually do he had chosen, many people would deny that he
could (absolutely) have done any such other dung. Hence Moore
is led to ask whether it is ever true, and if so in what sense, that
a man could have done s o m e h g other than what he did
actually do. And it is with his answer to dus question, not with
its bearings upon the meanings of right and wrong or upon the
problem of Free W d , that we are concerned.
With his usual shrewdness Moore begins by insisting that
there is at least one proper sense in which we can say that a man
can do s o m e h g he does not do or could have done something
he did not d w v e n though there may perhaps be other senses
of can and could have in which we cannot say such things. This
sense he illustrates by the sentence 'I could have walked a mile
in 20 minutes this morning, but I certainly could not have rut1
two miles in 5 minutes': we are to take it that in fact the speaker
did not do either of the two things mentioned, but h J in no
way h d e r s us from drawing the very common and necessary
distinction between undone acts that we could have done and
undone acts that we could not have done. So it is certain that,
at least in some sense, we often could have done t h g s that we
cGd not actually do.
Why then, Moore goes on to ask, should anyone try to deny
this? And he replies that people do so (we may call them
'determinists') because they hold that everything that happens
has a cause which precedes it, which is to say that once the cause
has occurred the dung itself is bound to occur and nothing else
could ever have happened instead.
However, on examining further the 20-minute-mile
example,
-
Moore argues that there is much reason to think
-

that 'could have' in such cases simply means 'could have if1 had
chosen', or, as perhaps we had better say in order to avoid a
possible complication (these are Moore's words), simply means
'should have if I had chosen'. And if this is all it means, then
 IFS A N D C A N S 207

there is after all no conflict between our conviction that we

often could have, in this sense, done things that we did not
actually do and the determinist's theory: fo; he certainly holds
himself that I often, and perhaps even always, should have done
something different from what I cLd do i f I had chosen to do that
different dung, since my choosing differentlywould constitute
a change in the causal antecedenti of my subsequent act, which
would therefore, on his theory, naturally itself be dfferent. If,
therefore, the determinist nevertheless asserts that in some sense
of 'could have' I could not ever havedone anydung different
from what I did actu~llydo, this must simply be a second sense'
of 'could have' different from that which it has in the 20-
minute-mile example.
In the remainder of his chapter, Moore argues that quite
possibly his first sense of 'could have', in which it simply means
'could or should have if I had chosen', is all we need to satisfy
our hankerings after Free Will, or at least is so if conjoined in
some way wlth yet a thlrd sense of 'could have' in which sense
'I could have done something different' means 'I might, for all
anyone could know for certain beforehand, have done some-
thing different'. This third lund of 'could have' might, I thnk,
be held to be a vulgarism, 'could' being used inForrectly for
'might': but in any case we shall not be concerned with it here.
1; the upshot, &en, Moore leaves us with only one important
sense in Ghich it can be said that I could have done somedung
that I cLd not do: he is not convinced that any other sense is
necessary, nor has he any clear idea what such another sense
would be: and he is convinced that, on his interpretation or
'could have', even t l ~ edeterminist can, and indeed must, say
that I could very often have done things I &d not do. To
summarize his suggestions (he does not put them forward with
complete conviction) once again:
r. 'Could have' simply means 'could have if I had chosen'.
2. For 'could have if I had chosen' we may substitute 'should
have if I had chosen'.
1 About which Moore has no more to tell us.
208 IFS AND CANS
3 . The if-clauses in these expressions state the causal con&-
tions upon which it would have followed that I could or
should have done the thing dfferent from what I &d
actually do.
Moore does not state this thud point expressly himself: but it
seems clear, in view of the co&exions he alleges between his
interpretation of 'could have' and the determinist theory, that
he did believe it, presumably taking it as obvious.
There are then three questions to be asked:
1. Does 'could have if I had chosen' mean the same, in
general or ever, as 'should have if I had chosen?'
2. In either of these expressions, is the if the if of causal
condtion ?
3 . In sentences having can or could have as main verb, are we
required or entitle2 always to supply an if-clause, and in
particular the clause 'if I had chosen' ?
It appears to me that the answer in each case is No.
I . Anyone, surely, would admit that in general could is very
different indeed from should or would.1 What a man could do is
not at all the same as what he ~ v o u l ddo: perhaps he could shoot
you if you were within range, but that is not in the least to say
that he would. And it seems clear to me, in our present example,
that 'I could have run a mile if I had chosen' and 'I should have
run a mile if I had chosen' mean quite different things, though
unfortunately it is not so clear exactly what either of them,
especially the latter, does mean. 'I should have run a mile in
20 minutes this morning if I had chosen' seems to me an un-
usual, not to say queer, specimen of English: but if I had to
interpret it, I should take it to mean the same as 'If I had chosen
to run a mile in 20 minutes this morning, I should (jolly well)
have done so', that is, it would be an assertion of my strength
Since Moore has couched his example in the first person, he uses 'should'
in the apodosis: but of course in the third person, everyone would use 'would'.
For brevity, I shall in what follows generally use 'should' to do duty for both
persons.
 I F S AND C A N S 209

of character, in that I put my decisions into execution (an

assertion which is, however, more naturally made, as 1 have
now made it, with the if-clause preceding- the main clause).
I should certainly not myself understand it to mean that if I
had made a certain choice my making that choice would have
caused me to do s o m e h g . But in whichever of these ways we
understand it, it is quite different from 'I could have walked a
mile in 20 minutes this morning if I had chosen', which surely
says something rather about my opportunities or powers.
Moore, unfortunately, does not explain why he dunks we are
entitled to make this all-importan; transition from 'could' to
'should', beyond saying that by doing so we 'avoid a possible
complication'. Later I shall make some suggestions which may
in part explain why he was tempted to make the transition:
but nothing can justify it.
2. Moore, as I pointed out above, did not discuss what sort
of $'it is that we have in 'I can if I choose' or in 'I could have
if 1-hadchosen' or in 'I should have if I had chosen'. Generally,
philosophers, as also grammarians, have a favourite, if some-
what blurred and diffuse, idea of an +clause as a 'conditional'
clause: putting our example schenlatically as 'If y, then q', then
it will be said that q follows from p, typically either in the sense
that p entails q or in the sense that p is a cause of q, though other
important variations are possible. And it seems to be on these
lines that Moore is thnking of the f i n 'I can if I choose'. But
now, it is characteristic of this general sort of Ji that from 'If p
then q' we can draw the inference 'If not q, then not p', whereas
we can not infer either 'Whether or not y, then q' or 'q' sim-
pliciter. For example,
-
from 'If I run, I pant' we cat1 infer 'If I
-

do not pant, I do not run' (or, as we should rather say, 'If I am

not panting, I am not running'), whereas we can not infer
either 'I pant, whether I run or not' or 'I pant' (at least in the
sense of '1 am panting'). If, to avoid these troubles with the
English tenses, which- are unfortunately prcvalent but are not
allowed to matter, we put the example in the past tense, then
from 'If I ran, I panted' it does follow that 'If I did not pant,
210 IFS AND C A N S
I did not run', but it does not follow either that 'I panted whether
or not I ran' or that 'I panted' period. These possibilities and
impossibhties
-
of inference are typical
- -
of the ijof causal condi-
tion: but they are precisely reversed in the case of 'I can if 1
choose' or 'I could have if I had chosen'. For from these we
.should not draw the curious inferences that 'If I cannot, I do
not choose to' or that 'If I could not have, I had not chosen to'
(or 'did not choose to'), whatever these sentences may be sup-
posed to mean. But on the contrary, from 'I can if I choose' we
certainly should infer that 'I can, whether I choose to or not'
and indeed that 'I can' period: and from 'I could have if I had
chosen' we should similarly infer that 'I could have, whether
I chose to or not' and that-anyway 'I could have' period. So
that, whatever this ifmeans, it is evidently not the ifof causal

This becomes even clearer when we observe that it is quite

common elsewhere to find an or&nary causal conditional in if
connexion with a can, and that then there is no doubt about it,
as for example in the sentence 'I can squeeze through if I am
thm enough', which does imply that 'If I cannot squeeze through
I am not d m enough', and of course does not imply that 'I can
squeeze through'. 'I can if I choose' is precisely different from
this.
Nor does can have to be a very special and ~eculiarverb for
ifs which are not causal conditional to be found in connexion
with it: all kinds of ifs are found with all kinds of verbs. Con-
sider for example th&in 'There are biscuits on the sideboard if
you want them', where the verb is the highly ordinary are, but
the ifis more like that in 'I can if I choose' than that in 'I panted
if I ran': for we can certainly infer from it that here are
biscuits on the sideboard whcther you want them or not' and
that anyway 'There are biscuits on the sideboard', whereas it
would be folly to infer that 'If there are no biscuits on the side-
board you do not want them', or to understand the meaning to
be that you have only to want biscuits to cause them to be on
the sideboard.
 IFS A N D C A N S 21I

The ij; then, in 'I can if I choose' is not the causal conditional
if. What of the ifin 'I shall if I choose'? At first glance, we see
that h s is quite hfferent (one more reason for refusing to
substitute shall for can or should have for could have). For from 'I
shall if I choose' w e clearly cannot infer that 'I shall whether I
choose to or not' or simply that 'I shall'. But on the other hand,
can we infer, either, that 'If I shan't I don't choose to'? (Or
should it be rather 'If I don't I don't choose to'?) I t h k not,
as we shall see: but even if some such inference can be drawn,
it would s t d be patently wrong to conclude that the meaning
of 'I shall if I choose' is that my choosing to do the t h g is suffi-
cient to cause me inevitably to do it or has as a consequence that
I shall do it, whlch, unless I am mistaken, is what Moore was
supposing it to mean. This may be seen if we compare 'I shall
ruin him if I choose' with 'I shall ruin him if I am extravagant'.
The latter sentence does indeed obviously state what would
be the consequence of the fulfilment of a condition specified in
the $-clause-but then, the first sentence has clearly cbfferent
characteristics from the second. In the first, it makes good sense
in general to stress the 'shall', but in the second it does not.'
This is a symptom of the fact that in the first sentence 'I shall' is
the present of that mysterious old verb shall, whereas in the
second 'shall' is simply being used as an aux&ary, without any
meaning of its own, to form the future indicative of 'ruin'.
I expect you will be more than ready at h s point to hear
s o m e h g a little more positive about the meanings of these
curious expressions 'I can if I choose' and 'I shall if I choose'.
Let us take the former first, and concentrate upon the ij:
The dictionary tells us that the words from which our ifis des-
cended expressed, or even meant, 'doubt' or 'hesitation' or 'con-
dition' or 'stipulationy. Of these, 'condition' has been given
a procbgious innings by grammarians, lexicographers, and
philosophers allke: it is time for 'doubt' and 'hesitation' to
1 In general, though of course in some contexts it does: e.g. 'I may very
easily ruin him, and 1 shall if I am extravagant', where 'shall' is stressed to
point the contrast with 'may'.
212 IFS A N D C A N S
be remembered, and these do indeed seem to be the notions
present in 'I can if I choose'. W e could give, on different occasions
and in hfferent contexts, many mfferent interpretations of
this sentence, which is of a somewhat primitive and loose-jointed
type. Here are some:
I can, quaere do I choose to?
I can, but do I choose to?
I can, but perhaps I don't choose to
I can, but then I should have to choose to, and what about that?
I can, but would it really be reasonable to choose to?
I can, but whether I choose to is another question
I can, I have only to choose to
I can, in case I (should) choose to,
and so on.
These interpretations are not, of course, all the same: which it
is that we mean will usually be clear from the context (other-
wise we should prefer another expression), but sometimes it can
be brought out by stress, on the 'if'or the 'choose' for example.
What is common to them all is simply that the asserlion, positive
and complete, that 'I can', is linked to the ~aisingofthe question
whether I choose to, which may be relevant in a variety of
ways. I
Ifs of the lund I have been trying- to describe are common
enough, for example the ifin our example 'There are biscuits
on the sideboard if you want them'. I do not know whether you
want biscuits or nit, but in case you do, I point out that there
are some on the sideboard. It is tempting, I know, to 'expand'
our sentence here to this: 'There are biscuits on the sideboard
which you can (or may) take if you want them': but this, legiti-
mate or not, will not make much difference, for we are still left
with 'can (or may) if you want', which is (here) just llke 'can

If there were space, we should consider other germane expressions: e.g.

'I can do it or not as I choose', 'I can do whichever I choose' (quidlibet). In
particular, 'I can whether I choose to or not' meails 'I can, but whether I choose
to or not is an open question': it does not mean 'I can on condition that 1 choose
and likewise on condition that 1 don't', which is absurd.
 IFS AND C A N S 213

if you choose' or 'can if you lke', so that the ifis still the ifof
doubt or hesitation, not the if of condition.1
I will mention two further points, very briefly, about 'I can
if I choose', important but not so relevant to our discussion
here. Sometimes the can will be the can, and the choice the
choice, of legal or other right, at other times these words will
refer to practicability or feasibhty: consequently, we should
sometimes interpret our sentence in some such way as 'I am
entitled to do it (if I choose)', and at other times in some such
way as 'I am capable of doing it (if I choose)'. We, of course,
are concerned with interpretations of t h s second kind. It would
be nice if we always said 'I may if I choose' when we wished to
refer to our rights, as perhaps our nannies once told us to: but
the interlocking histories of can and may are far too chequered
for there to be any such rule in p r a ~ t i c eThe
. ~ second point is
that choose is an important word in its own right, and needs
careful interpretation: 'I call if I lke' is not the same, although
the 'can' and the 'if' may be the same in both, as 'I can if I
choose'. Choice is always between alternatives, that is between
several courses to be weighed in the same scale against each
other, the one to be prrfrred. 'You can vote whichever way you
choose' is different from 'You can vote whchever way you
k e '.
And now for somethmg about 'I shall if I choose'-what sort
of ifhave we here? The point to notice is, that 'I shall' is not an
assertion ofjzct but an expression of intention, verging towards
the giving of some variety of undertaking: and the $ conse-
An account on these lines should probably be given also of an excellent
example given to me by Mr. P. T. Geach: 'I paid you back yesterday, if
you remember.' This is much the same as 'I paid you back yesterday,
don't you remember?' It does not mean that your now remembering that
I did so is a condition, causal or other, of my having paid you back yester-
day.
Formerly I believed that the meaning of 'I can if I choose' was something
hke 'I can, 1 have the choice', a i d that the point of the +clause was to make
clear that the 'can' in the main clause was the 'can' of right. This account,
however, does not do justice to the role of the 'if', and also unduly restricts in
general the meaning of 'choice'.
214 IFS A N D C A N S
if
quently, is the not of condition but of stipulation. In sentences
Ue:
I shall I marry him if I choose
I intend I to marry him if I choose
I promise I to marry him if he will have me
the if-clause is a part of h e object phrase governed by the
initial verb ('shall', 'intend', 'promise'), if this is an allowable
way of putting it: or again, the ifquaMes the content of the
undertaking given, or of the intention announced, it does not
qualify the giving of the undertaking. Why, we may ask, is it
perverse to draw from 'I intend to marry him if I choose' the
inference 'If I do not intend to marry him I do not choose to'?
Because 'I intend to marry hun if I choose' is not U e 'I panted
if I ran' in h s important respect: 'I panted if I ran' does not
assert anythmg 'categorically' about me-it does not assert that
I did pant, and hence it is far from surprising to infer something.
beginning 'If I did not pant': but 'I intend to marry him if I
choose' (and the same goes - for 'I shall marry him if I choose')
is a 'categorical' expression of intention, an2 hence it is para-
doxical to make an inference leading offwith 'If I do not intend'.
3 . Our third question was as to when we are entitled or
required to supply iJ-clauses with can or could have as main
verb.
Here there is one thing to be clear about at the start. There
are two quite hstinct and incompatible views that may be put
forward concerning ifs and cans, which are fatally easy to con-
fuse with each other. One view is that wherever we have can
or could have as our main verb, an if-clause must always be
understood or supplied, if it is not actually present, in order to
complete the sense of the sentence. The other view is that the
meaning of 'can' or 'could have' can be more clearly repro-
duced by some other verb (notably 'shall' or 'should have') with
an if-clause appended to-it. he first view is that an if is re-
quired to complete a can-sentence: the second view is that an iJ
is required in the analysis of a can-sentence. The suggestion of
Moore that 'could have' means 'could have if I had chosen' is
 IFS A N D C A N S 21 S

a suggestion of the first kind: but the suggcstion also made by

Moore that it means 'should have if I had chosen' is a suggestion
--
of the second kind. It may be because it is so easy (apparently) to
confuse these two kinds oftheorv that Moore was temvted
J L
to talk
as though 'should have' could mean the same as 'could have'.
Now we are concerned at this moment solely with thefirst
sort of view, namely that can-sentences are not complete with-
out an if-clause. And if we tlunk, as Moore was for the most
part thi.dmg, about 'could have' (rather than 'canY),it is easy to
see why it may be tempting to allege that it always requires an
if-clause with it. For it is natural to construe 'could have' as a
past subjunctive or 'conditional', whch is practically as much
as to say that it needs a conditional clause with it. And of course
it is quite true that 'could have' m g be, and very often is, a past
conditional: but it is also true that 'could have' may be and
often is the past (definite) indicative of the verb can. Sonletimes
'I could have' is equivalent to the Latin 'Potui' and means
6
I was in a position to': sometimes it is equivalent to the Latin
1
Potuissem' and means 'I should have been in a position to'.
Exactly similar is the double role of 'could', which is sometimes
a conditional meaning 'should be able to', but also sometimes a
past indicative (indefinite) meaning 'was able to': no one can
doubt tlus if he considers such contrasted examples as 'I could
do it 20 years ago' and 'I could do it if I had a h g u m m y . '
It is not so much that 'could' or 'could have' is ambiguous,
as rather that two parts of the verb can take the same shape.
Once it is realized that 'could have' can be a past indicative,
the general temptation to supply ijklauses with it vanishes: at
least there is no more temptation to supply them with 'could
have' than with 'can'. If we ask how a Roman would have said
'I could have ruined you t h s morning (although I didn't)', it
is clear that he would have used 'potui', and that h s sentence
is complete without any conditional clause. But more than tlus,
if he had wished to add 'if I had chosen', and however he had
expressed that in Latin, he would still not have changed hls
'potui' to 'potuissem': but h s is precisely what he would have
216 IFS AND C A N S
done if he had bceil tacking on some other, inore 'normal' kind
of if-clause, such as 'if I had had one more vote'.'
That is to say, the 'could have' in 'could have if I had chosen'
is a past indicative, not a past conditional, despite the fact that
there is what would, I suppose, be called a 'conditional' clause,
that is an $<lause, with G - ~ n thls
d is, of course, why we can
make the inferences that, as we saw, we can make from 'I could
have if I had chosen', notably the inference to 'I could have'
absolutely. Hence we see how mistaken Moore was in con-
trasting 'I could have if I had chosen' with the 'absolute' sense
of 'I could have': we might almost go so far as to say that the
addition of the 'conditional' clause 'if I had chosen' makes it
certain that (in Moore's language) the sense of 'could have' is
the absolute sense, or as I shouldprefer to put it, that the mood
of 'could have' is indicative.
It might at t h s point be worth considering in general whether
it makes sense to suppose that a language could contain any
verb such as can has been argued or implied to be, namely one
that can never occur without an +clause appended to it. At least
if the ifis the normal 'conditional' ifths would seem very diffi-
cult. For let the verb in question be to X: then we shall never
say simply 'I X', but always 'I X if I Y' : but then also, according
to the accepted rules, if it is true that 'I X if I Y', and nlso true
(whch it must surely sometimes be) that 'I do, in fact, Y', it
must surely follow that 'I X', simpliciter, without any ifabout
it any longer. Perhaps t h s was the 'possible complication' that
led Moore to switch from the suggestion that 'I could have' (in
one sense) has always to be expanded to 'I could have if' to the
suggestion that it has always to be nnalysed as 'I should have i f :

ffthe $-clause is 'if I had chosen', then I was able, was actually in a position,
to ruin you: hence 'potui'. But if the +clause expresses a genuule unfufjlled
condition, then plainly I was not actually in a position to ruin you, hence not
'potui' but 'potuissern'. My colleague Mr. R. M. Nisbet has pointed out to
me the interesting discussion o f this point in S. A. Handford, The Latin Sub-
junctive, pp. 1 3 0 ff. It is interesting that although this author well appreciates
the Latin usage, he still takes it for granted that in English the 'could have' is
universally subjunctive or conditional.
 IFS A N D CANS 217

for of course the argument I have just given does not suffice to
show that there could not be some verb which has always to be
annlysrd as somethng containing a conditional if-clause: sug-
gestions that this is in fact the case with some verbs are common
in philosophy, and I do not propose to argue this point, though
I think that doubt might well be felt about it. The only sort of
'verb' 1 can think of that might always demand a conditional
clause with it is an 'auxiliary' verb, if there is one, which is used
solely to form subjunctive or conditional moods (whatever
exactly they may be) of other verbs: but however this may be, it
is quite clear that can, and I should be prepared also to add shall
and will and may, are not in this poyition.-
To summarize, then, what has been here said in reply to
Moore's suggestions in his book:

"
(a) 'I could have if I had chosen' does not mean the same as
should have if I had chosen'.
(b) In neither of these expressions is the iJklause a 'normal
conditional' clause, connecting antecedent to consequent
as cause to effect.
(c) T o argue that can always requires an if-clausewith it to
complete the sense is totally different from arguing that
can-sentences are always to be analysed into sentences
containing if-clauses.
(d) Neither mn nor any other verb always requires a condi-
tional +clause after it: even 'could have', when a past
indicative, does not require such a clause: and in 'I could
have if I had chosen' the verb is in fact a past indicative,
not a past subjunctive or conditional.

Even, however, if all these contentions are true so far, we must

recognize that it may nevertheless still be the case that can,
could, and could have, even when used as indicatives, are to be
analysed as meaning shall, should, and should have, used as
auxiliaries of tense or mood with another verb (i.e. so as to
make that other verb into a future or subjunctive), followed by
21 8 IFS AND C A N S
a conditional if-clause. There is some plausibility,' for example,
in the suggestion that 'I can do X' means 'I shall succeed in
doing X, if I try' and 'I could have done X' means 'I should
have succeeded in doing X, if I had tried'.
It is indeed odd that Moore should have plumped so simply,
in giving h s account whether of the necessary supplementation
or of the analysis of 'could have', for the one particular if-clause
'if I had chosen', which happens to be particularly exposed to
the above objections, without even mentioning the possibility
of invoking other if-clauses,at least in some cases. Perhaps thc
reason was that choose (a word itself much in need of discussion)
presented itself as well fitted to bridge the gulf between deter-
minists and free-willers, whlch try might not so readily do. But
as a matter of fact Moore does hmself at one point give an
analysis of 'I could have done X' whlch is different in an inter-
esting way from his usual version, although confusible with it.
At a crucial point in his argument, he chioses for his example
'The ship could have gone faster', and the suggestion is made
that this is equivalent to 'The s h p would have gone faster ifher
oficers had chosen'. This may well seem plausible, but so far from
being in line, as Moore apparently t h d s , with his general
analysis, it differs from it in two important respects:

* Plausibility, but no more. Consider the case where I miss a very short putt
and kick myself because I could have holed it. It is not that I should have holed
it if I had tried: I did try, and missed. It is not that I should have holed it if '

conditions had been different:. that might of course be so, but I am talking
about conditions as they precisely were, and asserting that I could have holed
it. There is the rub. Nor does 'I can hole it this time' mean that 1shall hole it
this time if I try or if anything else: for I may try and miss, and yet not be
convinced that I could not have done it; indeed, further experiments may
confirm my belief that 1 could have done it that time although I did not.
But if I tried my hardest, say, and missed, surely there must have been some-
thing that caused mc to fail, that made me unable to succeed? So that I could not
have holed it. Well, a modern belief in science, in there being an exp!anation
of everything, may make us assent to this argument. But such a belief is not
in line with the traditional beliefs enshrined in the word can: according t o
them, a human ability or power or capacity is inherently liable not to produce
success, on occasion, and that for no reason (or are bad luck and bad form
sometimes reasons ?).
 IFS AND CANS 219

(a) the subject of the $-clause ('her officers') is different from

the subject of the main clause ('the ship'), the subject
of the original sentence:
(b) the verb in the ifilause following 'chosen' is clfferent
from the verb in the main clause, the verb in the original
sentence. W e do not readily observe this because of the
ellipsis after 'chosen': but plainly the verb must be, not
'to go faster', but 'to make her go faster' or, for example,
'to open the throttle'.

These two features are dictated by the fact that a shtp is inani-
-

mate. W e do not wish seriouslv to ascribe free will to inanimate

1

objects, and the 'could' of the original sentence is perhaps only

justifiable (as opposed to 'might') because it is readily realized
that some Lverson's free will is in auestion.
I

If we follow up the Lnes of this new type of analysis, we

should have to examine the relations between 'I could have
won' and 'I could, or should, have won if I had chosen to lob'
and 'I could, or should, have won if he had chosen to lob'. I
will do no more here than point out that the difference between
'could' and 'should' remains as before, and that the sense of 'I
could have wonY,if it really is one, in which it means something
of the sort 'I should have won if he had chosen to lob' or 'to let
me win' (the parallel to the s h p example), is of little impor-
tance-the 'if' here is of course the conditional if:
It is time now to turn to a second discussion of $ and cans.
Quite recently my colleague Mr. Nowell-Smith, in another
little book called Ethics, also reaches a point in h s argument at
whch he has to examine the sentence 'He could have acted
otherwise', that is, could have done something that he did not
in fact do. His reason for doing so is that, udess we can truly
say t h s of people, we might find ourselves unable to blame
people for things, and this would be generally regretted. This
reason is not unrelated to Moore's reason for embarking on
h s earlier discussion, and Nowell-Smith's views show some
resemblances to Moore's: perhaps this is because Nowell-Smith,
220 fFS A N D C A N S
hke Moore at the time he wrote his book is willing, if not
anxious, to come to terms with determinism.
Nowell-Smith begins his discussion by saying (p. 274) that
' "could have" is a modal phrase, and modal phrases are not
normally used to make categorical statements'. I am not myself
at all sure what exactly a 'modal phrase' is, so I cannot discuss
thls assertion: but I do not thnk &.IS matters, because he pro-
ceeds to give us two other examples of modal phrases, viz.
- have' and 'would have',' and to tell us first what they
'might
are not (which I omit) and then what they are:
'Would have' and 'might have' are clearly suppressed hypo-
thetical~,incomplete without an 'if. . .' or an 'if. ..not . . .'. Nobody
would say 'Jones would have won the championship' unless (a) he
believed that Jones did not win and (b) he was prepared to add 'if he
had entered' or 'ifhe had not sprained his ankle' or some such clause.
Here (t~) is actually incorrect-we can say 'Jones would (still)
have won the championship, (even) if Hagen had enteredy-but
thls does not concern us. (b),however, seems to be fairly correct,
at least as far as concerns 'would have' (in the case of 'might
have' it might well be doubtedz). So we have it that, when
Nowell-Smith says that 'would have' is a 'suppressed hyp+
Also perhaps 'may have', for he discusses 'It might have rained last Thurs-
day' in terms that seem really to apply to 'It may have rained last Thursday'.
1 refrain here from questioning it in the case of 'would have'. Yet 'would'
is agreed to be often a past indicative of the old verb will, requiring no if-
clause: and I think myself that in, say, 'X would have hanged him, but Y was
against it' 'would have' is likewise a past indicative-indeed it is from this sort
of example that we can see how the past tenses of will have come to be used
as auxiliaries of mood for forming the conditionals of other verbs.
To state what seem to be some grammaticd facts (omitting aU reference to
the use of the words concerned in expressing wishes):
Could have is sometimes a past indicative, sometimes a past subjunctive
of the verb can. When it is the main verb and is a subjunctive, it does require
a conditional clause with it. Can and its parts are not used as auxiliaries of
tense or mood to form tenses or moods of other verbs.
Would have, whether or not it is used as a past indicative or subjunctive
of the verb will, is now commonly used (should have in the first person) as
an auxiliary for forming the past subjunctive of other verbs: hence if it is
the main verb it does in general require a conditiona1 clause with it.
 IFS A N D C A N S 22 I

thetical' he means that it requires the addition of an $clause

to complete the sense. And he goes on to say that 'could have'
sentences also (though not so obviously) 'express hypo-
thetical~',if not always at least in important cases, such as
notably those where we say someone could have done some-
thing- he did not actually do: in these cases 'could have' . . . is
equivalent to 'would have . . . if. . .'.
It will be clear at once that ~oweu-Smith,like Moore, is not
distinguishing between the contention that 'could have' requires
supplementation by an ij-clause and the quite different contention
that its analysis contains an if-clause.1 On the whole it seems
plain that it is the second (analysis) view that he wishes to argue
for: but the argument he produces is that 'could have' is (in im-
portant cases) llke 'would have', the point about which is that
it needs an if-clauseto complete it--& though h s , whch is an
argument in favour of thefirst view, told in favour of the
second view. But it cannot possibly do so: and in any event
could have is liable, as we have already seen, to be in important
cases a past indicative, so that the contention that it is like would
have in requiring a conditional $-clause is unfounded.
Nevertheless, it must be allowed that Nowell-Smith may
still be right in urging that 'could have' means 'would have if'
and that, as he eventually adds, 'can' means 'will if'. What has
he to say in support of thts?
He propounds two examples for discussion, which I tMc do
not differ greatly, so I shall quote only the first. Here it is:
He could have read Emma in bed last night, though he actually
read Persuasion; but he could not have read Werther, because he does
not know German.
1 It is true that he uses two different expressions: 'would have' k a (sup
pressed) hypothetical, while 'could have' sentences express hypotheticals. But
it does not look as if any distinction is intended, and if it is, the protracted
initial analogy between 'could have' and 'would have' seems irrelevant and
misleading. Moreover, discussing the (unimportant) case of 'It could have
been a Morris', he writes that 'it would be absurd to ask under what conditions
it could or would have been a Morris' (my italics): this seems to show an
indifference to the distinction that I am insisting on.
 f F S AND CANS
This is evidentlv of the same kmd as Moore's 20-minute-mile
d

example. The first thing that Nowell-Smith urges is that such

a 'could have' statement is not a categorical, or a 'straight-
forward' categorical, statement. And his argument in favour of
this view is derived from the way in w h c h we should establish
its truth or falsity. No inspection of what the man actually did
will, he says, verify directly that he could have done somedung
else (here, read Emma) which he did not do: rather, we should,
to establish this, have to show
(a) that he has performed tasks of similar daculty sufficiently
often to preclude the possibility of a fluke, and (b) that nothing
prevented him on this occasion. For example, we should have to
establish that there was a copy of Emma in the house.

T o refute it, on the other hand, we should have to show either

'that some necessary condition was absent' (there was no copy
of Emma)' or 'that the capacity was absent'. That is, let us say,
we have to show on the one hand that he had both the ability
and the opportunity to read Emma, or on the other hand that
he lacked either the ability or the opportunity.
Nowell-Smith seems, at least at first, to be less interested in
the matter of opportunity: for he says that we can establish
'directly', i.e. by considering what the facts at the time actually
were, at least that he did not have the opportunity, that is, that
somethmg did prevent h m , and he does not seem daunted by
the obviously greater ddEculty of establishg, in order to
establish that he could have done it, the general negative that
there was nothing to prevent him. At any rate, it is at first upon
our manner of establishing that he had (or had not) the ability
to do this thing that he did not do that Nowell-Smith fastens
in order to support his assertion that the 'could have' statement
is not categorical. That the man had the ability to read Emma
can not, he says, be established 'directly', i.e. by observing what
happened on that past occasion, but only by considering what
prowess he has displayed in the face of similar tasks in the past
on other occasions, or displays now when put to the test: the
 IFS AND CANS 223

argument that we have perforce to use is an 'inductive' one

(and, he adds, none the worse for that).
Now let us pass all thls, at least for the sake of argument.=
What interests us is to discover why Nowell-Smith t M c s that
these considerations show that 'He had the ability to read
Emma' is not a categorical statement. I confess I fail to follow
the argument:
The very fact that evidence for or against 'could have' statements
must be drawn from occasions other than that to which they refer is
enough to show that 'He could have acted otherwise' is not a
straighrfonvard categorical statement.
But do we really know what is meant by a 'straightforward -
categorical statement'? certainly it is not the case that state-
ments made on the strength of inductive evidence are in general
not categorical-for example, the statement that the next mule
born will prove sterile: this seems categorical enough. Perhaps
thls example should be ruled out as not in point, on the ground
that here there will some L y be 'direct' evidence relevant to
the assertion, even if it is nit available at the moment. Could
the same, I wonder, be said of the inductive conclusion 'All
mules are sterile'? Or is that not categorical? I know that thls
has been interpreted by some philosophers to mean 'If anything
is a mule then it is sterile', but I see no reason to support that
curious interpretation.
The situation becomes still more puzzling when we remember
that Nowell-Smith is about to generalize his theory, and to
assert, not merely that 'could have' means 'would have . . . if',
but also that 'can' means 'shall or will. .. if'. Suppose then that
I assert 'I can here and now lift my finger,' and translatethis
as 'I shall lift my finger if. . .': then surely h s will be 'directly'
verified if the conditions are satisfied and I do proceed to lift the
1 Yet I think it is not hard to see that we cannot establish 'directly', at least
in many cases, that something 'prevented' him: he was drugged or dazzled,
which prevented him from reading, which establishes that he could not have
read-but how do we know that being drugged or dazzled 'prevents' people
from reading? Surely o n 'inductive' evidence? And, in short, to be prevented
is to be rendered unable.
 224 JFS A N D C A N S
fmger? If thls is correct, and if the theory is indeed a general
one, then there seems to be no point in insisting on the non-
availability of 'direct' evidence, which is only a feature of
certain cases. Incidentally, it is not in fact the case that to say
d
He could have done it' is always used in a way to imply that - .

he did not in fact do it: we make a list of the suspectsin a murder

case, all of whom we t h k could have done it and one ofwhom
we h n k did do it. True, h s is not Nowell-Smith's case: but
unless we are prepared to assert that the 'could have' in his case
- -

differs in meaning from that in the murder case, and so to rule

out the latter as irrelevant, we are in danger of having to admit
that even 'could have' sentences can be 'directly' verified in
favourable cases. For study of the facts of that past occasion
can prove to us that he did it, and hence that our original 'He
could have' was c0rrect.I
However, to proceed. Whether or not we should describe
our conclusion here as 'categorical' it seems that it should still be
a conclusion of the form 'he could have done so and so', and not
in the least a conclusion concerning what he would have done.
W e are interested, remember, in h s abilities: we want to know
whether hc could have read Emma yesterday: we ascertain that
he did read it the day before yesterday, and that he does read
it today: we conclude that he could have read it yesterday.
But it does not appear that t h s says anything about what he
would have done yesterday or in what circumstances: certainly,
we are now convinced, he could have read it yesterday, but
would he have, considering -
that he had read it only the day
before? Moreover, supposing the view is that our conclusion is
not of the 'could have' but of the 'would have if' form, n o h g
has yet been said to establish this, nor to tell us vhat follows
the 'if'. To establish that he would have read it yesterday

There are, I should myself think, good reasons for not speaking of 'I can
lift my finger' as being directly verified when I proceed to lift it, and likewise
for not speaking of 'He could have done it' as being directly verified by the
discovery that he did do it. But on Nowell-Smith's account I think that thae
would count as direct verifications.
 IFS A N D C A N S 225

i f . . ., we shall need evidence not merely as to h s abilities and

opportunities, but also as to h s charactei, motives, and so on.
It may indeed be thought, and it seems that Nowell-Smith
does at least partly think t h s , that what follows the 'if' should
be suppliable from the consideration that to say he could have,
in the full sense, is to say not merely that he had the ability,
which is what we have hitherto concentrated on, but also that
he had the opportunity. For to establish this, do we not have
to establish that certain conditions were satisfied, as for instance
that there was a copy of Emma available? Very well. But here
there is surely a confusion: we allow that, in saying that he
could have, I do assert or imply that certain conditions, those of
opportunity, were satirfed: but t h s is totally dfferent from
allowing that, in saying that he could have, I assert something
conditional. It is, certainly, entirely possible to assert something
conditional such as 'he could have read Emma yesterday if there
had been a copy - - available', could being- then of course a sub-
junctive: but to say t h s sort of dung is precisely not to say the
sort of thing that we say when we say 'He could have acted
otherwise', where 'could have' is an indicative-imp1 ying, as
we now do, that there was no copy available, we imply that - .

pro tanto he could not have acted otherwise. And the same will
be true if we try saying 'He would have read Emma yesterday if
there had been a copy available': this too certainly implies that
he could not in fact have read it, and so cannot by any means
be what we mean by saying that he could have read it.
In the concludmg paragraph of h s discussion, Nowell-Smith
does finally undertake to give us h s analysis not merely of
'could have', but also of 'can' (whch he says means 'will if').
And this last feature is very much to be welcomed, because if
an analysis is being consciously given of 'can' at least we shall
at length be clear of confusions connected with the idea that
6
could haves is necessarily a subjunctive.'
1 It must, however, be pointed out once again that if w e are to discuss the
assertion that somebody can (now) do something, the previous arguments
that our assertions are not categorical because they are based on induction and
226 IFS A N D C A N S
h as follows. It is
The argument of the last ~ a r a ~ r a pruns
'logically odd' to say something of this kind (I am slightly
emending Nowell-Smith's formula, but only in ways that are
favourable to it and demanded by his own argument):
Smith has the ability to run a mile, has the opportunity to run a
mile, has a preponderant motive for running a mile, but does not
in fact do so.
From t h s it follows directly, says ow ell-smith, that 'can'
means 'will if', that is, I suppose, that 'smith can run a mile'
means 'If Smith has the opportunity to run a mile and a pre-
ponderant motive for running it, he will run it'.
It seems, however, plain that nothing of the kind follows.
Ths may be seen first by setting the argument out formally.
Nowell-Smith's premiss is of the form
Not (p and q and r and not -s)

+
Logically odd (ability+ opportunity+ motive non-action) .
Now from this we can indeed infer
p 2 ((q and r) 3 s),
that is that
If he has the ability, then, if he has the opportunity and the
motive, he will do it.
But we cannot infer the converse
((q and r) 3 s) 3 p,
or in other words that
If, when he has the opportunity and the motive, he does it, he has
the ability to do it.
(I do not say t h s last is not s o m e h l g to which we should,
cannot be verified directly, whether they were good or not, must now be
abandoned: because of course it is possible to verify this 'directly' by the
method Nowell-Smith has sptcified in another connexion earlier, viz. by
getting the man to try and seeing him succeed.
 IFS AND CANS 227

when so put into English, assent, only that it does not follow
from Nowell-Smith's premiss: of course it follows merely from
the premiss that he does it, that he has the abhty to do it, ac-
cordmg to ordinary English.) But unless h s second, convene
implication does follow, we cannot, accordmg to the usual
formal principles, infer that p is equivalent to, nor therefore that
it means the same as, (q and r) 2 s, or in words that abhty means
that opportunity plus motive leads to action.
To put the same point non-formally. From the fact that, if
three thmgs are true together a fourth must also be true, we
cannot argue that one of the three thngs simply means that if the
other two are true the fourth will be true. If we could argue
indeed in this way, then we should establish, from Nowell-
Smith's premiss, not merely that
'He h a s the ability to do X' simply means that 'If he has the
opportunity and the motive to do X, he will do X'
but also equally that
'He has the opportunity to do X' simply means that 'If he has the
ability and the motive to do X, he will do X'
and hkewise that
'He has a preponderant motive to do X' simply means that 'If he
h a s the ability and the opportunity to do X, he will do X'.
For clearly we can perform the same operations on q and r as
on p, since the three all occupy positions in the premiss.
But these are fantastic suggestions. Put shortly, ~oweU-smith
is pointing out in h s premiss that if a man both can and wants
to (more than he wants to do anything else), he w d : but from
thls it does not follow that 'he can' simply means that 'if he wants
to he wd'. Nowell-Smith is strcgglhg to effect a transition
from can to will w h c h presents difficulties as great as those of
the transition from coul&o would: he puts up hk show of effect-
ing it by importing the additional, and here irrelevant, concept
of motive, whch needless to say is in general very intimately
connected with the question of what 'he d' do.
228 IFS A N D C A N S
When, in conclusion, Nowell-Smith finally sets out his
analysis of 'Smith could have read Emma last night', it is ths:
He would have read it, if there had been a copy, if he had not been
struck b h d , Gc., Gc., and if he had wanted to read it more than he
had wanted to read (this should be 'doY)anything else.
But so far from this being what we mean by saying he could
have read it, it actually implies that he could not have read it, for
more than adequate reasons: it implies that he was b h d at the
time, and so on. Here we see that Nowell-Smith actually does
make the confusion I referred to above between a statementb
which implies or asserts that certain conchtions were fulfilled and
I

a conchtional statement, i.e. a statement about what would have

happened if those conditions had been fulfilled. Ths is unfortu-
nately a confusion of a general h d that is not uncommon: I
need only mention the classic instance of Keynes, who confused
asserting on evidence h that p is probable with asserting that
on evidence h p is probable, both of which can be ambiguously
expressed by 'asserting that p is probable on evidence-h', b i t
only the former of whch asserts that p is (really) probable.
Here similarly there is a confusion between asserting on the
supposition (or premiss) that he had a copy that he could/would
have read it, and asserting that on the supposition that he had
a copy he could/would have read it, both of which can be
ambiguously expressed by 'asserting that he could/would have
read it on the supposition that he had a copy', but only the
former of whch asserts that he (actually) could have read it.
T o some extent, then, we learn from studying Nowell-
smith's arguments lessons similar to those that we learned in the
case of bore. But some are new, as for instance that many
assertions about what a man would have done or will do depend,
in critical cases, upon premisses about h s motives as well as, or
rather than, about h s abilities or opportunities: hence these
assertions c m o t be what assertions about h s abilities mean.'
Yet here it must be point^ ;1out once more that it has not been shown that
all assertions about what he would have done are so dependent, so that this
 IFS A N D C A N S 229

On one point I may perhaps elaborate a little further. It has

been maintained that sometimes when we sav 'He could haved

done X' ths is a conditional: it requires completion by an if-

clause, typically 'if he had had the opportunity', and so does
not require us, if we are to establish its truth, to establish that he
did infact have the opportunity. Sometimes on the other hand
it is a past indicative, implying that he did have the oppor-
tunity: in which case we do, to establish its truth, have to
establish that certain conditions were satisfied, but the assertion
is not to be described as a conditional assertion.
Now while I have no wish to retract ths account in general
or in all cases, I doubt whether it is the whole story. Consider
the case where what we wish to assert is that somebody had the
opportunity to do s o m e t h g but lacked the ability-'He could
have smashed that lob, if he had been any good at the smash':
here the i$-clause, whch may of course be suppressed and
understood, relates not to opportunity but to ability. Now
although we might describe thewhole sentence as 'conditional',
it nevertheless manages to assert, by means of its main clause,
s o m e h g 'categorical' enough, viz. that he did have a certain
opportunity. ~ n ind the same way Nowell-Smith's 'He could
have read Emma, if he had had a copy', does seem to assert
'categorically7 that he had a certain ability, although he lacked
-

the opport&ity to exercise it. Looking at it in this way, there

is a temptation to say that 'could have' has, besides its 'all-in'
sense several more restricted senses: t h s would be brought out if
we said 'He could have smashed it, only he is no good at the
smash' or 'He could have read Emma but he had no copy',
where, we should say, 'could have' is being used in the re-
stricted senses of opportunity or of ability' only, and is a past
indicative, not a past conditional.
particular argument against the analysis o f 'could have' as 'would have if' is
not conclusive: in particular, it does not dispose of the possible suggestion that
'could have' means 'would have if he had tried', for here considerations of
motive may be irrelevant.
I talk here and throughout of 'ability' and 'opportunity' only: but I
realize that other abstract nouns like 'capacity', 'skill', and even 'right' arc
230 IFS A N D C A N S
Ths view might be reinforced by considering examples with
the simple 'can' itself. W e are tempted to say that 'He can'
sometimes means just that he has the ability, with nothing said
about opportunity, sometimesjust that he has the chance, with
nothing said about ability, sometimes, however, that he really
actuallyfully can here and now, having both ability and oppor-
tunity. Now nobody, I thmk, would be tempted to say that 'can,'
where it means one ofthe two lesser dungs, for example, 'has the
opportunity', i.e. 'can in the full sense if he has the ability', is
grammatically a subjunctive or conditional. Perhaps, then, it
was not correct to describe 'He could have', either, as always
a conditional where it asserts ability or opportunity only, wi&
nothng said about the other, or even where the other is denied
to have existed.
The verb can is a peculiar one. Let us compare it for a moment
with another peculiar verb, know, with G h c h it shares some
grammatical peculiarities, such as lack of a continuous present
tense. When I say that soinebody kr~olvswhat the thng in my
hand is, I may mean merely that he has the ability to identify
it given the opportunity, or that he has the opportunity to
identitjr it if he has the ability, or that he has both. What do we
say about know here? Certainly we are not prone to invoke the
idea of a conditional, but rather that of different senses, or
perhaps the still obscure idea of the dispositional. I must be
content here merely to say that I do not t h d that the old
armoury of terms, such as 'mood' and 'sense', is altogether
adequate for handling such awkward cases. The only point of
w h c h I feel certain is that such verbs as can and know have each
an all-in, paradigm use, around whch cluster and from which
hvagate, little by little and along different paths, a whole
series of other uses, for many of which, though perhaps not for
all, a synonymous expression ('opportunity', 'realize', and so
on) can be found.
It is not unusual for ail audience at a lecture to include some
equally involved. All these terms need listing and elucidating before w e really
get to grips with 'can'.
 IFS AND CANS 231

who prefer t h g s to be important, and to them now, in case

there are any such -present, there is owed a -peroration. Why, in
short, does all t h s matter? First, then, it needs no emphasizing
that both ifand con are highly prevalent and protean words,
perplexing both grammatically and pMosophically: it is not
merely worth wMe, but essential, in these studes to discover
the facts aboutifs and cans, and to remove the confusions they
engender. In pMosophy it is con in particular that we seem so
often to uncover, just when we had thought some problem
settled, grinning residually up at us l ~ k ethe frog at the bottom
of the beer mug. Furthermore and secondly, we have not here
been dissecting these two words in general or completely, but
in a special connexion which perhaps no one will hold trivial.
It has been alleged by very serious phllosophen (not only the
two I have mentioned) that the things we ordinarily say about
what we can do and could have done may actually be con-
sistent with determinism. It is hard to evade all attempt to
decide whether t h s allegation is true-hard even for those who,
llke myself, are inclined to h k that determinism itself is still
a name for n o t h g clear, that has been argued for oilly in-
coherently. At least I should like to claim that the arguments
considered tonight fail to show that it is true, and indeed in
failing go some way to show that it is not. Determinism, what-
ever it may be, may yet be the case, but at least it appears not
consistent with what we ordinarily say and presumably think.
And finally there is a thlrd point. Reflecting on the arguments
in t h s lecture, we may well ask ourselves whether they might
not be as well assigned to grammar as to philosophy: and this,
I think, is a salutary question to end on. There are constant
references in contemporary philosophy, w h c h notoriously is
much concerned with language, to a 'logical grammar' and a
'logical syntax' as though-
these were things-
distinct from ordin-
ary grammarian's grammar and syntax: and certainly they do
seem, whatever exactly they may be, different from traditional
grammar. But grammar today is itself in a state of dux; for fifty
years or more it has been questioned on all hands and counts
 IFS AND C A N S
whether what Dionysius Thrax once thought was thc truth
about Greek is the trith and the whole truthibout all language
and all languages. Do we know, then, that there will prove to
be any ultimate boundary between 'logical gammar' and a
revised and enlarged Gronlnlor ? In the history ofhuman inquiry,
ph~losophyhas the place of the initial central sun, seminal a i d
tunlultuous: from time to time it throws off some portion of
itself to take station as a science, a planet, cool and well regu-
lated, progressing steadily towards a distant final state. This
happened long ago at the birth of mathematics, and again at
the birth of physics: only in the last century we have witnessed
the same process once again, slow and at the time almost inl-
perceptible, in the birth-of the science of mathematical logic,
through the joint labours of pl~losophersand mathematicians.
Is it not possible that the next century may see the birth, through
the joint labours of philosophers, grammarians, and numerous
othkr students of language, b f a true and comprehensive srieme
-

of language? Then we shall have rid ourselves of one more part

of philosophy (there will still be plenty left) in the only way
we ever can get rid of philosophy, by kiclung it upstairs.
 PERFORMATIVE UTTERANCES

Y o u are more than entitled not to know what the word 'per-
formative' means. It is a new word and an ugly word, and
perhaps it does not mean anything very much. But at any rate
there is one thing in its favour, it is not a profound word. I
remember once when I had been talking on this subject that
somebody afterwards said: 'You know, I haven't the least idea
what he means, unless it could be that he simply means what
he says'. Well, that is what I should like to mean.
Let us consider first how this affair arises. We have not got
to go very far back in the history of phlosophy to find philo-
sophers assuming more or less as a matter of course that the
sole business, the sole interesting business, of any utterance-
that is, of anything we say-is to be true or at least false. Of
course they had always known that there are other kinds of
things which we say-thmgs llke imperatives, the expressions
of wishes, and exclamatioAs-some o f which had eGen been
classified by grammarians, though it wasn't perhaps too easy
to tell always which was which. But still -philosophers - have
assumed that the only things that they are interested in are
utterances which report facts or whch describe situations truly
or falsely. In recent times this kind of approach has been
questioned-in two stages, I think. First of all people began to
say: 'Well, if these things are true or false it ought to be possible
to decide which they are, and if we can't decide which they
are they aren't any good but arc, in short, nonsense'. And this
new approach did a great deal of good; a great many dungs
23 4 PERFORMATIVE UTTERANCES
which probably are nonsense were found to be such. It is not
the case, I t W , that all kinds of nonsense have been ade-
quately classified yet, and perhaps some things have been dis-
missed as nonsense which really are not; but still this movement,
the verification movement, was, in its way, excellent.
However, we then come to the second-stage. After all, we
set some limits to the amount of nonsense that we talk, or at
least the amount of nonsense that we are prepared to admit
we talk; and so people began to ask whether after all some
of those things which, treated as statements, were in danger
of being dismissed as nonsense did after all really set out
to be statements at all. ~ i g h t n ' tthey perhaps be- intended
not to report facts but to influence people in this way or that,
or to let off steam in this way or that? O r perhaps at any rate
some elements in these utterances -performed such functions,
or, for example, drew attention in some way (without actually
reporting it) to some important feature of the circumstances
in which the utterance was being made. On these lines people
have now adopted a new slogan, the slogan of the 'different uses
of language'. The old approach, the old statemental approach,
is son~etin~es called even a fallacy, the descriptive fallacy.
Certainly there are a great many uses of language. It's rather
a pity that people are apt to invoke a new use of language
whenever they feel so inclined, to help them out of ths, that,
or the other well-known philosophcal tangle; we need more
of a framework in which to discuss these uses of language; and
also I think we should not despair too easily and talk, as people
are apt to do, about the infinite uses of language. ~ h i l o s o ~ h e r s
will do this when they have listed as many, let us say, as seven-
teen; but even if there were somethng like ten thousand uses
of language, surely we could list them all in time. This, after
all, is no larger than the number of species of beetle that ento-
mologists have taken the -pains to list. But whatever the defects
of either of these movements-the 'verification' movement or
the 'use of language' movement-at any rate they have effected,
nobody could deny, a great revolution in philosophy and,
 PERFORMATIVE U T T E R A N C E S 23 5
many would say, the most salutary in its history. (Not, if you
come to thnk of it, a very immodest claim.)
Now it is one such sort of use of language- -
that I want to
examine here. I want to discuss a kind of utterance which looks
like a statement and grammatically, I suppose, would be classed
as a statement, which is not nonsensica1,and yet is not true or
false. These are not going to be utterances which contain curious
verbs llke 'could' or 'might', or curious words llke 'good',
which many philosophers regard nowadays simply as danger
signals. They will be perfectly straightforward utterances, with
ordinary verbs in the first person singular present indicative
active, and yet we shall see at once that they couldn't possibly
be true or false. Furthermore, if a person makes an utterance
of this sort we should say that he is doing something rather than
merely saying something. This may sound a little odd, but the
examples I shall give will in fact not be odd at all, and may
even seem decidedly dull. Here are three or four. Suppose, f i r
example, that in the course of a marriage ceremony I say, as
people will, 'I do'-(sc. take this woman to be my lawful
wedded-wife). Or again, suppose that I tread on your toe and
say 'I apologize'. Or again, suppose that I have the bottle of
champagne in my hand and say 'I name this ship the Querrz
Elizabeth'. Or suppose I say 'I bet you sixpence it will rain
tomorrow'. In all these cases it would be absurd a, regard the
thing that I say as a report of the performance of the action
which is undoubtedly done-the action of betting, or christen-
ing, or apologizing. We should say rather that, in saying what
I do, I actually perform that action. When I say 'I name this
ship the Qlrrrrl Elizabeth' I do not describe the christening
ceremony, I actually perform the christening; and when I say
'I do' (sc. take this woman to be my lawful wedded wife),
I am not reporting on a marriage, I am indulging in it.
Now these kinds of utterance are the ones that we call per-
formative utterances. his is rather an ugly word, and a new
word, but there seems to be no word already in existence to
do the job. The nearest approach that I can think of is the word
236 PERFORMATIVE U T T E R A N C E S
'operative', as used by lawyers. Lawyers when tallung about
legal instruments will distinguish between the -preamble, whch
- -

recites the circumstances in which a transaction is effected, and

on the other hand the operative part-the part of it whch
1. A A

actually performs the legal act which it is the purpose

- -
of the
instrumeAt to perform. So the word 'operative' is very near to
what we want. 'I give and bequeath my watch to my brother'
would be an operative clause and is a performadve utterance.
However, the word 'operative' has other uses, and it seems
preferable to have a word specially designed for the use we want.
Now at this point one might protest, perhaps even with
some alarm, that I seem to be suggesting that marrying is
simply saying a few words, that just saying a few words is
marrying. Well, that certainly is not the case. The words have
to be said in the appropriate circumstances, and h s is a matter
that will come up again later. But the one dung we must not
suppose is that what is needed in addition to the saying of the
. -

words in such cases is the performance of some internal spiritual

A A

act, of whch the words then are to be the report. It's very easy
to slip into t h s view at least in difficult, portentous cases,
though perhaps not so easy in simple cases like apologizing. In
the case of promising-for example, 'I promise to be there
tomorrow'-it's very easy to thmk that the utterance is simply
the outward and visible (that is, verbal) sign of the performance
of some inward spiritual
-
act of promising,
- -
and h s view has
certainly been expressed in many classic places. There is the
case of Euripides' Hippolytus, who said 'My tongue swore to,
but my heart did not'-perhaps it should be 'mind' or 'spirit'
-

rather than 'heart', but at any rate some kind of backstage

artiste. Now it is clear from this sort of example that, if we slip
into thinking that such utterances are reports, true or false, of
-

the performance of inward and spiritualacn, we open a loop-

hole to perjurers and welshers and bigamists and so-on, so that
there are disadvantages in being excessively solemn in this way.
It is better, perhaps, to stick to the old saying that our word is
our bond.
 PERFORMATIVE U T T E R A N C E S 237
However, although these utterances do not themselves
-

report facts and are not themselves true or false, saying these
h g s does very often imply that certain t h g s are true and
not false, in some sense at least of that rather woolly word
'imply'. For example, when I say 'I do take this woman to be
my lawful wedded wife', or some other formula in the marriage
ceremony, I do imply that I'm not already married, with wife
- .

living, sane, undivorced, and the rest of it. But still it is very
important to realize that to imply that something or other is
true, is not at all the same as saying something which is true
itself.
These performative utterances are not true or false, then.
But they do sufir from certain disabhties of their own. They
can fail to come off in ~pecialways, and that is what I want to
consider next. The various ways in which a performative
utterance may be unsatisfactory we call, for the sake of a name,
the infelicities; and an infelicity arises-that is to say, the utter-
ance is unhappy-if certain rdes, transparently simple rules,
are broken. I will mention some of these rules and then give
examples of some infringements.
-
irk of all, it is obvious that the conventional ~rocedure
which by our utterance we are purporting to use must actually
exist. In the examples
- given here this -procedure will be a verbal
-

one, a verbal procedure for marrying or giving or whatever

it may be; but-it should be borne in mind that there are many
non-verbal procedures by which we can perform exactly the .
same acts as we perform by these verbal means. It's worth
remembering too that a great many of the h g s we do are
at least in part of h s conventional lund. Philosophers at least
are too apt to assume that an action is always in the last resort
the malung of a physical movement, whereas it's usually, at
least in part, a matter of convention.
The first rule is, then, that the convention invoked must
exist and be accepted. And the second rule, also a very obvious
one, is that the circumstances in which we purport to invoke
this procedure must be appropriate for its invocation. If t h s is
23 8 P E R F O R M A T I V E UTTERANCES
not observed, then the act that we purport to perform would
not come off-it will be, one might - say, a misfire. Thls will
also be the case if, for example, we do not carry through the
procedure-whatever it may be-correctly and completely,
without a flaw and without a hitch. If any of these rules are not
observed, we say that the act which we purported to perform
is void, without effect. If, for example, the ~ u r ~ o r t eact
d was
nn act of marrying, then we should say that we 'went through
a form' of marriage, but we did not actually succeed in marrying.
Here are some examples of this kind of misfire. Suppose that,
living in a country like our own, we wish to divorce our wife.
W e may try standing her in front of us squarely in the room
and saying, in a voice loud enough for all to hear, 'I divorce
you'. Now this procedure is not accepted. W e shall not thereby
have succeeded in divorcing our wice, at least in this country
and others llke it. This is a case where the convention, .we
should say, does not exist or is not accepted. Again, suppose
that, picking sides at a children's party, I say 'I pick George'.
But George turns red in the face and says 'Not playingy.In that
case I plainly, for some reason or another, have not picked
George-whether because there is no convention that you can
-

pick people who aren't playing, or because George in the

circumstances is an inappropriate object for the procedure of
picking. Or consider the case in which I say 'I appoint you
Consul', and it turns out that you have been appointed
--
already
-or perhaps it may even transpire that you are a horse; here
again we have the infelicity of inappropriate circumstances,
inappropriate objects, or what not. Examples of flaws and
hitches -are perhaps scarcely necessary-;ne party in the
marriage ceremony says 'I will', the other says 'I won't'; I say
'I bet sixpence', but nobody says 'Done', nobody takes up the
offer. In all these and other such cases, the act whlch we pur-
port to perform, or set out to perform, is not achieved.
But there is another and a rather different way in which this
kind of utterance may go wrong. A good many of these
verbal procedures are designed for use by people who hold
 PERFORMATIVE UTTERANCES 239

certain beliefs or have certain feelings or intentions. ~ n ifdyou

use one of these formulae when you do not have the requisite
thoughts or feelings or intentions then there is an abuse of the
procedure, there is insincerity. Take, for example, the expres-
sion, 'I congratulate Ths is designed for use by people
- - -
who are glad that the person addressed has achieved a certain
feat, believe that he was personally responsible for the success,
and so on. If I say 'I congratulate you' when I'm not pleased
or when I don't believe that the credit was yours, then there is
insincerity. Lkewise if I say I promise to do something, with-
out having the least intention of doing it or without believing
it feasible. In these cases there is s o m e h g wrong certainly,
but it is not llke a misfire. We should not say that I didn't in
fact promise, but rather that I did promise but promised in-
sincerely; I did congratulate you but the congratulations were
hollow. And there may be an infelicity of a somewhat s i d a r
kind when the performative utterance commits the speaker to
future conduct of a certain description
-
and then in the future
he does not in fact behave in the expected way. This is very
obvious, of course, if I promise to do somethmg and then
break my promise, but there are many kinds of commitment
of a rather less tangible form than that in the case of promising.
For instance, I may say 'I welcome you', bidding you welcome
to my home or wherever it may be, but then I proceed to treat
you as though you were exceedingly unwelcome. In t h s case
the procedure of saying 'I welcome you' has been abused in a
-

way rather different from that of simple insincerity.

Now we might ask whether this Gst of infelicities is com-
plete, whether the kinds of infelicity are mutually exclusive,
and so forth. Well, it is not complete,
- and they are not mutually
exclusive; they never are. Suppose
--
that you are just about to
name the ship, you have been appointed to name it, and you
. .

are just about to bang the bottle against the stem; but at that
very moment some low type comes up, snatches the bottle out
of your hand, breaks it on the stem, shouts out 'I name this shlp
th; Generalissimo Stalin', and then for good measure kicL
240 PERFORMATIVE UTTERANCES
away the chocks. Well, we agree of course on several things.
W e agree that the s h p certainly isn't now named the ~mrrafis-
simo Stalin, and we agree
- that it's an infernal shame and so on
and so forth. But we may not agree as to how we should classify
the particular infelicity in this case. W e might say that here is
a case of a perfectly legitimate and agreed procedure whch,
however, has been invoked in the wrong circumstances,
namely by the wrong person, ths low type instead of the
person appointed to do it. But on the other hand we might
look at it differently and say that h s is a case where the pro-
cedure has not as a whole been gone through correctly, because
part of the procedure for naming a s h p is that you should
first of all get yourself appointed as the person to do the
naming and that's what this fellow did not do. Thus the way
we should classify infelicities in different cases will be perhaps
rather a difficult matter, and may even in the last resort be a bit
arbitrary. But of course lawyers, who have to deal very much
with this land of thing, have invented all kinds of technical
terms and have made numerous rules about different lunds of
cases, whch enable them to classify fairly rapidly what in
particular is wrong in any given case.
As for whether this list is complete, it c e r t d v is not. One
A #

further way in which t h g s may go wrong is, for example,

through what in general may be called misunderstanding.
You may not hear what I say, or you may understand me to
refer to something - different from what I intended to refer to,
and so on. And apart from further additions whch we might
make to the list, there is the general over-riding consideration
that, as we are performing a11 act when we issue these per-
formative utterances, we may of course be doing so under
duress or in some other circumstances whch make us not
entirely responsible for doing what we are doing. That would
certainly be an unhappiness of a hd-any lund of non-
responsibility might be called an unhappiness; but of course
- - -

it is a quite different kind of thing from what we have been

t a h g about. And I might mention that, quite differently
 PERFORMATIVE U T T E R A N C E S 241

again, we could be issuing any of thcse utterances, as we can

issue an utterance of any kind whatsoever, in the course, for
example, of acting a play or m k n g a joke or writing a poem-
in which case of course it would not be seriously meant and we
shall not be able to say that we seriously performed the act
concerned. If the poet says 'Go and catch a fahng star' or
whatever it may be, he doesn't seriously iss~lean order Con-
siderations of &s kind apply to any utterance at all, not nlerely
to performatives.
That, then, is enough to be going on with. We have
discussed the performative utterance and its infelicities. That
equips us, we may suppose, with two s h g new tools to
crack the crib of reality maybe. It also equips
- -
us-it always
does-with two s h i n g new skids under our metaphysical feet.
The question is how we use them.

So far we have been going- firmly ahead, f e e h g - the firm

ground of prejudice glide away beneath our feet whlch is
always rather exhilarating, but what next? You will be waiting
for the bit when we bog down, the bit where we take it all
back, and sure enough that's going to come but it will take
time. First of all let us ask a rathcr simple question. How can
we be sure, how can we tell, whctller ally utterance is to be
classed as a performative or not? Surely, we feel, we ought to
be able to do that. And we should obviouslv verv much l ~ k e
/ /

to be able to say that there is a grammatical criterion for this,

some granunatical illcans of deciding whether an utterance is
performative. All the examples I have given hitherto do in
fact have the same granlmatical form; thcy all of them begin
with the verb in the first person singular present indicative
-

active-not just any kind of verb of course: but still they all
are in fact of that form.. Furtherinore, with these vcrbs that I
have used tllere is a typical asynlmetry between the use of thls
person and tense of ilk verband the use of the same verb in
242 PERFORMATIVE U T T E R A N C E S
other persons and other tenses, a l d this asynlmetry is rather an
important clue.
For example, when we say 'I promise that . . .', the case is
very different from when we say 'He promises that . . .', or in
the past tense 'I promised that. . . . For when we say 'I promise
2

that . . .' we do perform an act of promising-we give a

promise. What we-do not do is to report on somebody's per-
forming an act of promising-in particular, we do not report
on somebody's use of the expression 'I promise'. We actually
do use it and do the promising. But if I say 'He pron~ises',or
in the past tense 'I promised', I precisely do report on an act of
promising, that is to say an act of using this formula 'I
-I report on a present act of promising by him, or on a past
act of my own. here is thus a clear difference between our
first person
-
singular
- -present indicative active, and other
persons and tenses. This is brought out by the typical incident
of little Willie whose uncle says he'll give him half-a-crown
if he promises never to sinoke till he's 5 5. Little Willie's
anxious parcnt wdl say 'Of course he promises, don't you,
Willie?' giving him a nudge, and little Willie just doesn't
vouchsafe. The point here is that lle must do the promising
himself by saying 'I promise', and his parent is going too fast
in saying he proinises.
hat, then, is a bit of a test for wlletller an urteraiice is per-
formative or not, but it would not do to suppose that every
performative utterance has to take this standard form. There
is at least one other standard forin, every bit as common as this
one, where the verb is in the passive voice and in the second or
third person, not in the first. The sort of case I mean is that of
a notice inscribed 'Passengers are warned to cross the line by
the bridge only', or of a document reading 'You are hereby
authorized' to do so-and-so. These are ulldoubtedly per-
formative, and in fact a signature is often required in order to
show who it is that is doing the act of warning, or authorizing,
or whatever it may be. Very typical of tllis kind of yerfornia-
tive--especially liable to occur in writtcil documents of course
 PERFORMATIVE U T T E R A N C E S 243

--is that the little word 'hereby' eitller actually occurs or illight
naturally be inserted.
Unfortunately, however, we still can't possibly suggest
that every utterance which is to be classed as a performative
has to take one or another of these two, as we might call them,
standard forms. After all it would be a very typical performative
utterance to say 'I order you to shut the door'. This satisfies all
the criteria. It is performing the act of ordering you to shut the
door, and it is not true or false. But in the appropriate circum-
stances surely we could perform exactly the same act by simply
saying 'Shut the door', in the imperative. Or again, suppose
that somebody sticks up a notice his bull is dangerous', or
simply 'Dangerous bull', or simply 'Bull'. Does this necessarily
differ from sticking up a notice, appropriately signed, saying
'You are hereby warned that this bull is dangerous'? It seems
that the simple'notice 'Bull' can do just the same job as the
more elaborate formula. Of course the difference is that if we
just stick up 'Bull' it would not be quite clear that it is a warn-
ing; it might be there just for interest or information, llke
'Wallaby' on the cage at the zoo, or 'Ancient Monument'.
No doubt we should know from the nature of the case that it
was a warning, but it would not bc explicit.
Well, in view of this break-down of grammatical criteria,
what we should llke to suppose-and there is a good deal in
this--is that any utterance which is performative could be
reduced or expanded or analysed into one of these two standard
forms beginning 'I . . .' so and so or beginning 'You (or he)
hereby . . .' so and so. If there was any justification for this hope,
as to some extent there is, then we might hope to make a list
of all the verbs which can appear in thcse standard forms, and
then we might classify the kinds of acts that can be performed
by performative utterances. We might do t h s with the aid of
a dictionary, using such a test as that already mentioned-
whether there is the characteristic asyinnletry between the first
person singular present indicative active and the other persons
and tenses-in ordcr to decide whether a verb is to go into our
246 PERFORMATIVE U T T E R A N C E S
make it quite plain that the act being performed is the con-
ventional one of doing obeisance rather than some other act.
Now nobody would want to say that lifting your hat was
stating that you were ~ e r f o r m i nan ~ act of obeisance; it
certainly is not, but it does make it quite m lain that you are.
And so in the same way to say 'I warn you that . . .' or 'I order
you to . . .' or 'I promise that . . . is not to state that you are

doing s o m e h g , but makes it plain that you are-it does

constitute your vcrbal pcrformance, a performance of a parti-
cular kind.
So far we have been going along as though there was a quite
clear difference between our performative utterances and what
we have contrasted them with, statements or reports or de-
scriptions. But now we begin to find that this distinction is nor
as clear as it might be. It's now that we begin to sink in a little.
In the first place, of course, we may feel doubts as to how
widely our berformatives extend. ~f we thlnk up some odd
kinds of expression we use in odd cases, we might very well
wonder whether or not they satisfy our rather vague criteria
for being performative utterances. Suppose, for example,
somebody says 'Hurbrah'. Well, not true or false; he is perform-
ing the act of cheering. Does that make it a performative
utterance in our sense or not? Or suppose he says ' D a m ' ; he
is performing the act of swearing, and it is not true or false.
Does that make it performative? We feel that in a way it does
and yet it's rather different. Again, consider cases of 'suiting
the action to the words'; these too may make us wonder
whether perhaps the utterance should be classed as performa-
tive. Or sometimes, if somebody says 'I am sorry', we wonder
whether this is just the same as 'I a~o1ogize'-ii
-
whlch case of
course we have said it's a performative utterance-or whether
I

perhaps it's to be taken as a description, true, or false, of the

state of his feelings. If he had said 'I feel perfectly awful about
it', then wc should think it must be meant to be a description
of the state of his feelings. If he had said 'I apologize', we should
feel t h s was clearly a performative utterance, going through
 PERFORMATIVE U T T E R A N C E S 247

the ritual of apologizing. But if he says 'I am sorry9there is an

unfortunate hovering- between the two. T h s phenomenon is
quite common. W e often find cases in which there is an
obvious pure performative utterance and obvious other utter-
ances connected with it which are not performative but de-
scriptive, but on the other hand a many in between
where we're not quite sure which they are. On some occasions
of course they are obviously used the one way, on some occa-
sions the other way, but on some occasions they seem -positively
to revel in ambiguity.
Again, consider the case of the umpire when he says 'Out'
or 'Over', or the jury's utterance when they say that they find
the prisoner guilty. Of course, we say, these are cases of giving
verdicts, performing the act of appraising and so forth, but still
in a way they have some connexion with the facts. They seem
to have something like the duty to be true or false, and seem
not to be so very remote from statements. If the umpire says
'Over', t h s surely has at least somethmg - to do with six balls
in fact having been delivered rather than seven, and so 011. In
fact in general we may remind ourselves that 'I state that . . .'
does n i t look so very difkrent from 'I warn you that. . .' or
'I promise to . . .'. It makes clear surely that the act that we
are performing is an act of stating, and so functions just like 'I
warn' or 'I order'. So isn't 'I stHte that . . .' a performative
utterance? But then one may feel that utterances beginning
'I state that . . .' do have to be true or false, that they are state-
ments.
Considerations of h s sort, then, may well make us feel
pretty unhappy. If we look back for a moment at our contrast
between statements and performative utterances, we realize
-

that we were taking statements very much on trust from, as we

said, the traditional treatment. Statements, we had it, were to
be true or false; performative
- utterances on the other hand
were to be felicitous or infelicitous. They were the doing of
something, whereas for all we said making statcments was not
doing something. Now this contrast surely, if we look back
248 PERFORMATIVE U T T E R A N C E S
at it, is unsatisfactory. Of course statements are liable to be
assessed in this matter of their correspondence or failure to
correspond with the facts, that is, being true or false. But they
are also liable to infelicity every bit as much as are performative
utterances. In fact some-troubles that have arisen in the study
of statements recently can be shown to be simply troubles df
infelicity. For example, it has been pointed out that there is
something very odd about saying something like ths: 'The cat
is on the-mat .but I don't believe it is'. Now h s is an out-
rageous thmg to say, but it is not self-contradictory. There is
no reason why the cat shouldn't be on the mat without my
believing that it is. So how are we to classify what's wrong
with this peculiar statement? If we remember now the doctrine
of infelicity we shall see that the person
-
who makcs this remark
about the cat is in much the same position as somebody who
says something llke this: 'I promise that I shall be there, but I
haven't the least intention of being thereY.Once again you can
of course perfectly well promise to be there without having
the least intention of being there, but there is somethmg out-
rageous about saying it, about actually avowing the insincerity
of the promise you give. In the same way there is insincerity
in the case of the person who says 'The cat is on the mat but
I don't believe it is', and he is actually avowing that insincerity
-which makes a pecdar kind of nonsense.
A

A second case that has come to light is the one about John's
chldren-thc case where somebody is supposed to say 'All
John's chldren are bald but John hasn't got any children'. Or
perhaps somebody says 'A11 John's chlldren are bald', when as
a matter of fact-he doesn't say so-John has no children.
Now those who study statements have worried about this;
ought they to say that the statement 'A11 John's chldren are
bald' is meaningless in this case? WeU, if it is, it is not a bit like
a great many other more standard hnds of meaninglessness;
and we see, if we look back at our list of infelicities, that what
is going wrong here is much the same as what goes wrong in,
say, the case of a contract for the sale of a piece of land when
 PERFORMATIVE U T T E R A N C E S 249

the piece of land referred to does not exist. Now what we say
in the case of this sale of land, which of course would be effccted
by a performative utterance, is that the sale is void-void for
lack of reference or atnbiguity of reference; and so we can see
that the statement about all ~ d m ' children
s is likewise void for
lack of reference. And if the man actually says that John has
no children in the same breath as saying they're all bald, he is
malung the same kind of outrageous utterance as the man who
says 'The cat is on the mat and I don't believe it is', or the Illan
who says 'I promise to but I don't intend to'.
In this way, then, ills that have been found to aHict state-
ments can be precisely paralleled with ills that are characteristic
of perforrnative
-
utterances. And after all when we state solllo
.thing or describe something or report s o m e h g , we do per-
form an act wllich is every bit as much an act as an act of
ordering or warning. There seems no good reason why stating
should be given a specially unique position. Of course phdo-
sophers have been wont to talk as though you or I or anybody
could just go round stating anytlung about anythmg aid that
would be perfectly in order, only there's just a little question:
is it true or false? But besides the little question,
*
is it true or
false, there is surely the question: ir it in order? Can you go
round just malung statenlents about anythmg? Suppose for
example you say to me 'I'm feeling pretty nlouldy this morn-
ing'. Well, I say to you 'You're not'; and you say 'What the
devil do you mean, I'm not?' I say 'Oh nothitlg-~'m just
stating you're not, is it true or false?' And you say 'Wait a bit
about whether it's true or false, the question is what did you
mean by malung statements about somebody else's feelings?
I told you I'm feeling pretty mouldy. You're just not in a
position to say, to state that I'm not'. Thls brings out that you
can't just make statetnents about other p&ople's feclkgs
(thougll you can make guesses if you llke); and there are very
many things which, having no knowledge of, not being in a
position to pronounce about, you just can't state. What we
need to do for the case of stating, and by the same token
250 PERFORMATIVE U T T E R A N C E S
describing and reporting, is to take then1 a bit off their pedestal,
to realize that they are speecll-acts
- no less than all these other
speech-acts that we have been mentioning and talking about
as performative.
Then let us look for a moment at our original contrast
between the performative and the statement from the other
side. In handling performatives we have been putting it all the
time as tho~lghthe only thing that a perfomlative utterance
had to do was to be felicitous, to cotnc off, not to be a misfire,
not to be an abuse. Yes, but that's not the end of the mattcr.
At least in the case of many utterances which, on what we have
said, we should have to class as performative--cases where we
say 'I warn you to . . .', 'I advise you to . . .' and so on-there
will be other questions besides simply: was it in order, was it
all right, as a piece of advice or a warning, did it come off?
After that surely there willmbe the question: was it good or
sound advice? Was it ajustified
- warning?
- Or in the case, let us
say, of a verdict or an estimate: was it a good estimate, or
a sound verdict? And these arc questions that can only be
decided by considering - how the content of the verdict or
estimate is related in some way to fact, or to evidence available
about the facts. This is to say that we do require to assess at
least a great many performative utterances in a general dimen-
sion of correspondence with fact. It may still be said, of course,
that this does not make them vsrv like statements because still
I

they are not true or false, and that's a little black and white
speciality that distinguishes statements as a class apart. But
a ~ t u a l l ~ - - t hitowould
~ ~ ~ htake too long to go on about this-
the more you thmk about truth and falsity the more you find
that very few statements that we ever utter are just true or just
false. UsualIy there is the question are they fair or are they not
fair, are they adequate or not adequate, are they exaggerated
or not exaggerated? Are they too rough, or are they
precise, accurate, and so on? 'True' and 'false' are just general
labels for a whole dimension of different appraisals which have
somedung or other to do with the relation between what we
 PERFORMATIVE U T T E R A N C E S ZS 1

say and the facts. If, then, we loosen up our ideas of truth and
faisity we shall see that statements, when assessed in relation to
the facts, are not so very different after all from pieces of
advice, warnings, verdicts, and so on.
W e see then that stating something is performing an act just
as much as is giving an order or giving a warning; and we see,
on the other hand, that, when we give an order or a warning or
a~L
i e c eof advice. there is a question about how this is related
A

to fact whch is not perhaps so very different from the kind of

- -

question that arises when we discuss how a statement is related

to fact. Well, ths seems to mean that in its original form our
distinction between the performative and the statement is
considerably weakened, and indeed breaks down. I will just
make a suggestion as to how to handle this matter. We need
to go very much farther back, to consider all the ways and
senses in which saying anything at all is doing this or that-
because of c o u n e it is always doing a good many different
~

thngs. And one thing that emerges when we do do thls is that,

besides the question that has been very much studied in the
Dast
L
as to what a certain utterance means, there is a further
question distinct from thls as to what was theforce, as we may
call it, of the utterance. We may be quite clear what 'Shut the
door' means, but not yet at all clear on the further point as to
whether as uttered at a certain time it was an order, an entreaty
or whatnot. What we need besides the old doctrine about
meanings is a new doctrine about all the possible forces of
utterances, towards the discovery of which our proposed list
of explicit
-
performative verbs would be a very great help; and
-
then, going on from there, an investigation of the various
terms of appraisal that we use in discussing speech-acts of this,
that, or the other precise kind-orders, warnings, and the like.
The notions tha; we have considered then, are the performa-
tive, the infelicity, the explicit performative, and lastly, rather
hurriedly, the notion of the forces of utterances. I dare say that
aU ths seems a little unremunerative, a little complicated.
Well, 1 suppose in some ways it is unremunerative, and I
252 PERFORMATIVE UTTERANCES
suppose it ought to be remunerative. At least, though, I think
that if we pay attention to these matters we can clear up some
mistakes in philosophy; and after all*philosophy is used as a
scapegoat, it parades mistakes whlch are really the mistakes of
everybody. We might even clear up some mistakes in grammar,
which perhaps is a little more respectable.
And is it complicated? Well, it is conlplicated a bit; but life
and truth and dungs do tend to be complicated. It's not things,
it's philosophers that are simple. You will have heard it said,
I expect, that over-simplification is the occupational disease of
philosophers, and in a way one might agree with that. But for
a sneaking suspicion that it's their occupation.
 PRETENDING'

I N a recent paper2 Mr. Errol Bedford argues that 'angery, Lke

other words whlch would be said to be words for emotions, is
not the name of a feeling, despite the existence of such expres-
sions as 'feeling angryY.'~nge~', he argues, is not a name: nor
is anger a feeling: there is no specific feeling that angry men as
such feel, nor do we, to be angry,
- -
have to feel any feeLng
- at all.
With h s thesis I am not concerned, but only with some re-
marks that he makes, quite incidentally, about pretending (and
I realize it is hard on him to pick these out for intensive
criticism). For he thmks that hls view may be countered by
referring to the case of someone pretending to be angry: is this
not parallel to the case of someone pretending to be in pain, who
precisely does not feel a certain feeling (pain) that the man
who is in pain does feel-a feeling of w h c h 'pain* surely is the

'Can we say that being angry is similar to being in pain in this

respect? Let us contrast the cases of a man who is angry and another,
behaving in a s i d a r way, who is only pretending tdbe. Now it may
well be true.that the former feels angry, whereas the h e r does not,
but in any case it is not this that constitutes the difference between
the fact that the one is angry and the fact that the other is only
pretendmg to be. The objection rests on a misconception of what
pretence is. There is necessarily involved in pretence, or shamming,
the notion of a limit which must not be overstepped: pretence is
always insulated, as it were, from reality. Admittedly this limit may
be vague, but'it must exist. It is a not unimportant point that it is
Reprinted from Proceedings of the Aristotelian Society, Supplementary
Volume (1957-8), by courtesy of the editor.
Proceedings of the Aristotelian Society, I 9 56-7.
254 PRETENDING
usually obvious when someone is pretending. If a man who is behav-
ing as if he were angry goes so far as to smash the furniture or
commit an assault, he has passed the limit; he is not pretending, and
it is useless for him to protest afterwards that he did not feel angry.
Far from this statement being proofthat he was not angry, it would
L 6
be discounted even if it were accepted as true. He was angry, but
he did not feel angry" is not self-contradictory, although it is no
doubt normally false. If in a particular case it is difficult-as it may
be-to settle the question "Pretended or real?" that can only be
because the relevant public evidence is inadequate to settle it. What
we want is more evidence of the same k i d , not a special piece of
evidence of a different kind. Our difficulty in resolving the question
"Is he really in pain?" on the other hand, arises from the fact that
the only dccisive evidence is evidence that he alone is in a position
to give.'
Since pain gets a perhaps undue share of attention in philo-
sophy, and since Mr. Bedford is not s h o c h g us about pre-
tending to be in pain, let us here leave pain out of it, only
r e m a r L g that if pretending to be in pain and pretending to
be angry are actually as different as Mr. Bedford supposes
then surely his statements about pretending, designed as they
are to fit the case of anger, should be put in less general terms.
Our man, then, is 'behaving as if h e w ere ang$. He scowls,
let us say, and stamps his foot on the carpet. So far we may (or
perhaps must?) still say 'He is not (really) angry: he is (only)
pretendng to be angry'. But now he goes further, let us say
he bites the carpet: a n d h e will picture the scene with sympathy
-the carpet innocent, the bite untentative and vicious, the
damage grave. Now he has gone too far, overstepped the
limit between pretence and reality, and we cannot any longer
say 'He is pretending to be angry' but must say 'He is really
angry'. Mr. Bedford's language seems to me on the whole to
mean positively that we must say this because and in the sense that
behaviour of t h s extreme sort constittrtes being really angry,=
At least the bite 'constitutes the difference' between being really angry
and pretending to be angry, the common element being presumably such
behaviour as scowling. Some may recall the textbook example, where it is
 PRETENDING 255

or is just what we mean by 'being really angry'. If, however,

he only means, what he also says, that the extreme behaviour
is decisive evidence that the man is really angry, that is not only
a very different and slightly (if only slightly) more
thesis, but also one too weak to serve for h s argument: for now
we are still not told what really being angry, for whch ths is
only the evidence, is, nor therefore shown that it does not in-
volve, or even reside in, the feeling of a feeling-the evidence
might be evidence that he is feeling a certain feeling.
W e have primarily to consider whether Mr. Bedford is right
in what he says we should say, rather than his claims about
what is shown by our so speaking, if we do. If the man takes
the bite, he cantrot 'be pretending'-here surely Mr. Bedford
carries the philosopher's professional addiction to furniture to
a new pitch of positive concern for it. And if he does really
mean that the difference in behaviour 'constitutes the difference
between the fact that the one is angry and the fact that the other
is only pretending to be', then he must be claiming, not only
that once he has taken the bite we cannot (truly) say 'He is only
pretending to be angry', which seems false, but also that if he
merely stamps and goes no further we cannot (truly) say 'He is
really angry', which seems patently false. I think it must on
reflectio; be agreed that in-whichever of the ways the man
behaves it is open to us to say either 'He is angry' or 'He is only
pretending to be angry', and that either statement can be in
fact true, depending on the (other) circumstat~cesof the case
at least in addition to these features of his behaviour. It is
common enough for someone who is really angry to behave
in no way violently or even conspicuously: and if some-
one is to be angry in some emergency where the
success of the pretence matters seriously, more anyway than
the integrity of any adjacent furniture (which may not even
be his own and may in any case be insured), then surely he
only the hair on a gooseberry that stops it from being a grape: by a 'goose-
berry', then, we mean simply a hirsute grape-atrd by a 'grape' Iikewisc simply
a glabrous gooseberry.
256 PRETENDING
may hit upon biting the carpet as the very t h g to clinch the
deception.
Something has gone very wrong. Yet still there are in fact,
as wc should expect, ways in which limits and the overstepping
of limits are relevant to the concept of pretending, as to so
many others. O n a festive occasion you are ordered, for a for-
feit, to pretend to be a hyena: going down on all fours, you
make a few essays at hldeous laughter and finally bite nly calf,
taking,- with a touch of realism possibly exceeding your hopes,
- -

a fair-sized piece right out of it. ~ e y o n dquestion YOU have

-

gone too far.' Try to plead that you were only pretending, and
I shall advert forcibly to the state of my calf-not much pre-
tence about that, is there? There are liniits, old sport. This sort
of t h g in these circumstances will not pass as ' (only)
ing- to be a hyena'. True-but then neither will it pass as really -

bein'q a hyena. The limit overstepped, a limitation upon violence

as in the carpet-biting case, is not a boundary between pre-
tending to be a hyena and really being a hyena, but between
pretending to be a hyena and behaving like an iincivilized
tough, or perhaps between merely ~retendmgto be a hyena
and pretending to be a hyena with a difirence of some kind,
with knobs on or with ulterior motives. So too if you begin to
assault the bric-a-brac when told to pretend to be angry for
a forfeit, we need not say that you must be really angry, but
only that such antics are too bad and quite uncalled-foi when
pretending in such circumstances, or pcrl~apsthat you are
taking advantage of the opportunity to further private aesthetic
aims (in whch case you may not really be pretending, but only
pretending to pretend), or perhaps somethng else again quite
different but still in its way satisfyingly censorious.
The moral is, clearly, that to be not ret tending to be, and
still more to be not only-pretending to be, is not by any means
necessarily, still less eo i p , to be really being. Tllis is so even
when the way in which we fail to be (only-) pretending is by
In these circumstances. But if Nero ordered you, in the arena, to pretend
t o be a hyena, it might be unwisely perfunctory not t o take a piece right out.
 PRETENDING 257

indulging in excessively 'realistic' behaviour : but of course

there are also numerous other kinds of case, some to be men-
tioned later, in whch we might be taken to be ret tend in^ and
so may be said to be not pretending, where the reasons for
whch we are said not to be (only-) pretending are totally
different from h s , and such that the notion that not-pretending
3 really being could scarcely insinuate itself. W e must not allow
ourselves to be too much obsessed by the opposition, in whch
of course there is something, between pretending and really
being: not one of the following formulae is actually correct:
( I ) not really being 3 pretending
(2) p r e t e n h g 3 not really being
(3) not pretending 3 really being
(4) really being 3 not pretenhg.1
So set out these formulae lose, I realize, some of their attractive-
ness: but arguments llke Mr. Bedford's show that they can
attract; he has actually, if I am not mistaken, fallen principally
for (3), w h c h is not by any means the most tempting, though
some of his arguments seem to favour (2), a quite independent
matter.
'Pretend' is a verb used in various constructions, of whch
I have so far only mentioned 'pretend to be' followed by an
adjective or adjectival phrase or by a substantive with the
article: in such cases excessive behaviour will, as we have seen,
commonly not produce the result that the performer 'really is',
for example, angry. (I hesitate to say it, but surely the obvious
reason is that 'being angry' does not consist merely in behaving
publicly in some manner: to say t h s need not commit us to
saying that being angry is the same as feeling angry-it is not,
any more than being tired is the same as feeling tired-stdl less
that 'anger7 is the name of a feehg.) However, we have to
consider also the construction in w h c h 'pretend' is followed by
'to A' or 'to be A-ing', especially in cases where the verb 'A' is
one w h c h describes the doing of some deed (for example, 'bite' as
Actually, 'really' is, like 'actually', really a broken reed in philosophy.
See how they twist and turn in example (3) below-the window-cleaner.
258 PRETENDING
opposed to, for example, 'believe'), and inore particularly when
that deed is of a pretty 'physical' kind (for example, biting as
opposed to, for example, giving). If we now consider such a case
as this: and if we remember one of the conditions that must be
satisfied whenever I am pretending, viz. that there must be
somedung, and somethni public,-that I am actually doing,
some action I actually am performing, in pretending and in
order to pretend: then we may hope to have found one type of
case in which what Mr. Bedford claims to hold of pretending
in general does in fact hold.
Let us take the casc where someone is to 'pretend to take a
bite out of your calf'. Here it would be agreed that one thing
he must not do,' however lifellke the pretence, is anything that
could be correctly described as '(actually) taking a bite out of
your calf': yet plainly too the action he has, in pretending,
actually to perform is one which will be up to a point genuinely
like the action he is pretending to perform (for what he is pre-
tending is hrre to perform a public physical action), and might,
but for precautions, pass over into itOzIf he goes far enough he
will have really done the thing 11c was only to pretend to do:
and if he does not go so far, he cannot have really done that
dung. Here, then, we seem to have a case on Mr. Bedford's
pattern.
It is owing to the special features of cases of this l n d that an
impasse call arise over pretending to do somethmg, say hole a
putt, in circumstances, say in the presence of a surrounding
crowd, where there seems to be nothing one can do at all llke
holing the putt whch will not result in the putt's being actually
holed.' It is easy to pretend to be sitting on a certain chair when

At least intentionally: I neglect complications about the unintentional.

O f course there is too the rarish and quite different case in which a man
pretending to be angry actually hecomes angry-makes himself angry. I do riot
think this is of comparable interest.
Doubtful, though not inexplicable, cases arise here, because of doubts as
to how much is connoted by a putative description of a 'physical' action.
Can I pretend to cough? Shall I, if I produce a coughing noise, have actually
coughed? O r is 'to cough' different from 'to deliberately cough'?
 PRETENDING 259

it is half concealed b e h d a desk, less easy if it is in f d view.

(Ths is different from the less subtle type
- -
of case where one
cannot pretend to do something because one can do neither it
nor, often by the same token, a n y h g even passably l k e it.
Thus you cannot pretend to curl your trunk-though again,
of course, if you help yourself to that curious object 'a pretend
trunk', i.e. something of whch we pretend that it is your trunk,
you can very Mely curl that, and hence also very Mely pretend
to curl it.)
Is it however the case that at least when we are pretending
to do or to be doing i physical action we are universally de-
barred from actually doing that action itself? W e will consider
three examples:
I . Two miscreants are surprised in the act and hastily agree,
the wherewithal being handy, to pretend to be sawing a tree:
in a trice the blade is humming to and fro a bare inch away
from the bark. How good -
a -pretence is that? And wouldn't
they any longer be pretending to be sawing the tree if they
allowed the teeth to bite in? Surely if they want the pretence to
be convincing they should set about actually sawing the tree?
2. Yet surely again magicians pretend to saw girls, we've
all seen one pretending very successfully to saw a girl in half.
Would it really be still a pretence, and a more convincing one,
if the teeth were biting- in? Or wouldn't it rather have been
transformed into grim reality?
3. That chap over there, he's all right I suppose, he's clean-
ing the windows, eh?
Ah, him, he's pretending to be cleaning the windows right
enough, cleaning 'em a treat too: but I seen h m taking note of
the valuables through 'em all the time.'
T o unravel these examples, we shall need a few more
lemmas: we shall need to bring out more of the f d features
Here is another, trick, example, for exercise purposes only:-a man at a
party decides, in an attempt to amuse, to pretend to behave vulgarly: the party,
however, is of a type at which even to pretend to behave vulgarly is, alas, to
behave vulgarly.
260 PRETENDING
of the situation when we are pretending, which is moderately
complicated. And first for that goddess fair and free (fair1y fair,
frailly frce), ctivincst Etyil~ology.Prrr~>-tcnderein Latin never
strays far from the literal meaning of holding or stretching one
t h n g in front of another in order to protect or conceal or
disguise it: even hl such a figurative use as that in Ovid's
'praetendens culpae splendida verba tuae', the words are still
a fasade to hide the crime. In ~ n ~ l i s weh , do not any longer
explicitly refer, in the construction used with 'pretend', to that
wlich the pretender is hding or dissembling, whch in Latin
- -

does appear in the dative case.' Nevertheless it seems clear that

it still is an important feature of pretending, in classic cases if not
in all, that the pretender is concealing or suppressing something.
In a case of pretending, then, there willtypical@ be:
(PB) The pretence-behaviour, the actual public ~erform-
ance gone through in pretendmg, indulged in, as of course
it is, for the sake of dissembling.
(Rd) The reality-dissembled, about whch the audience
is to be hoodwmked. Ths rnny on occasion include in part,
or be wholly identical with
(RBd) Some real-behaviour-dissenlblen, as for instance
when I am really engaged in biting the carpct but disguise
tlus fact by pretending to be kissing it.
Thus when we speak of someone's angry behaviour being only
a pretence, one thing with whch this pretended anger is com-
monly being contrasted at least in our minds is (Kd) h s real
emotion, feeling, attitude, or what you will, which, whatever
it is, is precisely not 'real anger'. In daily life, indeed, this con-
trast may be of more interest than the quite different con-
trast, whch has been more strcsscd by philosophers, bctween
(PBm) Tllc mere-pretence-behaviour, the actual public

Indeed in English even the accusative case after 'prctend', as in, for
example, 'He pretetrded sickness', though a venerable coilstruction is by n o w
archaistic. In the special construction 'pretending not t o be' there is however
a reference to what is being concealed.
 PRETENDING 261
performance gone through in pretenhng, disregarding its
motivation,
and
(GBs) The genuine-behaviour-simulated, whch PBm
is intended to resemble. This may be related to a further
(Gs) 'Genuinity'I-simulated, as genuinely behaving angrily
is related, for example, to genuinely being angry.
When some simple contrast between 'pretence' and 'reality'
comes up in discbssion, it is all too often uncertain whch of
the things here listed is being contrasted with whch.
T o return now to our three examples. (2)-the girl-sawing- -
simply supports the rule suggested by the precedGg discussion,
that in pretending to do A you must not actually do A, or that
PBm must not coincide with GBs. Defending h s rule, we are
tempted to try some special dodge to get out of (I)--the tree-
sawing. The miscreants are 'pretendmg to be sawing the tree'
and also 'they are sawing it' in fact, but perhaps they are pre-
tending to 'be sawing' it in a sense that covers times earlier and
later than the time during which they 'are sawing' it in fact: so
that PBm does differ from GBs, it extends over a shorter stretch
of time. Or perhaps we should not allow that they 'are
(seriously) sawing' it, for example, in the sense that they are
not embarked on an operation designed to terminate in the fall
of the tree: but it is not clcar whit ths means-suppose the
police are suspicious and continue to hang around indefinitely ?
The case will then become llke that of the man who pretends
to be playing golf by playing a few strokes: can he prolong the
- ~

pretence all round the course and yet not be actually playing
golf? It is llkely that by introducing 'seriously' (and of course
I am driven to this horrible word because I wish to use throughout the
second contrast a different term from 'real', which I have kept for the first
colltrast.
The Gs may staid to the GBs as, say, its 'motivation': then such an expres-
sion as 'pretending to be angry' will commonly run the two together. But
where the GBs is something more purely 'physical', such as 'sawing a girl in
half', the Gs, if any, is at a discount.
262 PRETENDING
it is true that their heart is not in sawing
-
the tree, they are only
doing it at all to cover up something) we are really already on
the way to the treatment which we mitst use for example (3)-
the window-cleaner.
Here surely no dodge will help us; we must allow that he is
indeed actually cleaning the windows, from start to finish and
throughout the whole tiltle he is pretending to be cleaning
them. But it is still a pretence, because what he is redly doing
all the time is something different, namely noting the valuables:
he is only cleaning the windows to disguise and promote tlis
other activity-RBd goes on during the course of PB, which
facilitates it and distracts attention from it. (In other cases RBd
may actually be incorporated into PB as a camouflaged part
of it.) It looks, then, as though it does not matter if PB does
coincide with GBs, so long as the contrast between PB and
RBd is -preserved.'
It is worth noting once more that it will seldom be possible
to decide with certainty that PBtn does coincide exactly with
GBs, because in so rnany cases GBs is apt to be described, and
may only be describable, in terms wllich already import the
Gs w h c h underlines it: thus when someone is 'pretending to
be angry', the GBs will be 'angry behaviour' or 'the behaviour
of an angry man', a description which may be held already to
mean that the actions are done 'in anger'. Only when the GBs
is describable in pretty purely 'physical' terms which disregard
I
motivation' and the like, for example, as 'sawing a girl', shall
we be confident of the coincidence.
In the light of example (3)' it c m now be seen that the sup-
posed rule that in certain cases, such as example (2),PBm must
not coincide with GBs, is really only a marginal case of a more
general rule. The essence of the situation in pretending is (not -

so much that my public behaviour must be non-genuine

behaviour, as rather) that my public behaviour is Incant to
Hcrc is one of the similarities between 'pretence' and 'pretcxt'. A pretext
may be not a geiluine reason or not your real reason: a pretence may be some-
thing you are trot genuinely doing or not what you are really doing.
 PRETENDING 263

disguise some reality, often some real behaviour. From thls it

obviously follows, not only that PB must not coincide with
RBd, in which case there would be no disguise, but also that
PB must not coincide with not-RBd, in which case there would
be nothing to be being disguised. Now in a case like that of the
magician,-the RBd pr&isely is, or includes, not actually sawing
the girl in half, so that the GBs, sawing the girl in half, is
equivalent to not-RBd: hence in such a case it follows directly
from the more general rule that PB must stop short of being
identical with GBs, as = not-RBd. This type of case, where Gs
precisely equals or involves not-Kd, or GBs not-RBd, is of
course quite a common one: 'pretending not to be' is a special
variety of it.
At least in many cases there seems to be a clear difference in
meaning between the expressions 'pretending to A' and 'pre-
tending to be A-ing'. The former seems often to be preferred
where it is being pointed out that PBm does not coincide with
GBs, while the latter stresses that PB does not coincide with
RBd. 'He is only pretending to clean the windows', i.e. what he
is doing does not amount to genuinely cleaning the windows:
but 'He is only pretending to be cleaning the windows', i.e.
what he is really up to is something other than cleaning the
windows. Take, again, Potter's gambit, where he makes three
random moves and then resigns. If we say 'He's only pretend-
ing to play (chess)', we mean that that is not playing chess:'
but if we say 'He's only pretending to be playing (chess)', we
allow that in a way and for all we care he is playing chess, but
we mean that he is really up to some deeper game. Children
who are ignorant may typically be 'pretending to play chess':
children, ignorant or not, who are up to mischief may typically
be 'pretending to be playing chess'. The nlagician who is
pretending to saw the girl, i.e. we reassure ourselves, not
actually sawing her, may also be said to be 'pretending to be
sawing her' if, whether he is or not (and naturally we presume
For some reason. For example, to be genuinely playing chess you must be
making your moves with the object of winning, or at least of iiot loslug.
264 PRETENDING
not), he is surreptitiously engaged in something else rather
crucial for the success of the illusion.
I should not, however, llke to claim that t h s is the whole
story about 'pretending to A' and 'pretending to be A-ing'.
For consider two further cases:
4. Sonleone in the next room out of sight keeps up a string
of remarks such as 'Check', 'Your move', &c., and occasionally
taps pieces of wood together. W e should say 'He is (only)
pretending (for the benefit of us in the next room) to be play-
ing chess', but scarcely 'He is (only) pretending to play chess'.
Why is this?
5 . A boy in an arm-chair is making tugging and twisting
movements with h s arms, accompailied by gear-change and
other raucous noises. He is 'pretending to be driving a racing-
car', but scarcely 'pretending to drive a racing-car'. Why?
A possi6li> answer is this. Ill neither case is the behaviour of the
-

pretending party sufficiently like the genuine article (GBs) for

it to be in point to matk the distinction between the two. T o
I

pretend to drive a racing-car, he would need a racing-car: as

it is, there is no serious prospect of deception. And in case (4)
the deception is worked indirectly, nlairlly by words: if his
actual actions were observed, tl~erewould again be no serious
chance of deception. It might be urged, too, that both these
cases of 'preten&ng to' have some affinity with 'pretending
that', of which more later, which generally requires the con-
tinuous present tense after it. On the othcr hand, the difference
between, say, rete tending to sit' and 'pretending to be sitting9
is at least sometimes clear?v iust the familiar diffience between
I J

'he sits' and 'he is sitting',

-
so that it will not do to claim that the
two forms of expression are used to rnark any one single dis-
tinc tion.
So far we have not strayed very far from our starting-point,
a consideration of the limits which must not be overstepped
in the pretence-behaviour. Only in special
- - cases is the limit
between 'pretending to do A' and 'really doing A' of much
interest, and even then it is of minor importance in clarifying
 PRETENDING 265

the whole notion of pretending. When s o m e h g claimed to

be pretending is ruled out by reason of 'going too far', this will
commonly mean something such as 'going beyond what was
socially permissible on that occasion' rather than 'slipping into
doing- the actual thing'. But now further, there are other con-
-

ditions of a quite general kind to which behaviour must con-

form if it i s t o as pretence-behaviour: the following
examples may serve to bring out some of them:
6. Trapped on a branch against the moon, we decide to
pretend to be owls: you give a colourable hoot while I pull up
my legs and hunch my shoulders up to my ears.
7. As I am engaged in filching one of your goats, you return
inopportunely through the dusk: with a baffled snarl I bound
off into the adjacent bush. Was this 'pretending to be a panther' ?
Or what if instead I slink about the kraal with menacing grunts?
8. Told to pretend to be a hyena at a party, you recline and
appear to sleep.
9. In similar circumstances, you proceed to jump around
powerfully on your hind legs, boxing with your fists and
fondling something in your pocket.
These are all somewhat facetious cases of 'pretendmg to be
an' animal. It may be worth pointing out that 'pretending to
be a hyena' in the let's-pretend, make-believe, party-forfeit
way, is a very recent usage, perhaps no older than Lewis
Carroll, and the same indeed seems to apply to at least most
usages in which we pretend to be something other than our-
selves. One of the most conspicuous
-
facts in the hstory of the
word 'pretend' is that of late it has come to be more popular
and to be applied more widely than formerly.
In (6) I do better than you. We both imitate the owl, you
perhaps rather better in voice than I in silhouette: but you stop
short of pretending to be an owl, because you fail to attempt
to disguise the fact that you are not one-mere imitation does
not imply dissembling anything. In (7). while it seems clear
that I am pretending if I slink around, this becon~esmuch more
doubtful if I bound away, rigkt away and promptly: for it to be
266 PRETENDING
a clear case of pretending I, my human person, must remain on
the scene to be hidden under the pretence, but as things are it is
plainly- -preferring- to be hidden under the bush. ~ f to
-
, startle
me, you quack in a passable way from the undergrowth, you
are scarcely pretending to be a duck (for you are not on the
scene nor in need of disguise), as you would, however, be,
very probably, if I trod on in thc dark and you quacked.
Of course in all these cases you might be trying to rtrake me
believe that you were a panther or a duck: but not all suc-11
deceptions are achieved by pretending-I can make you
believe I am angry by many methods without ever pretending
to be angry.
In case (8), at the party, there is of course no question of my
trying to convince you srriously that I am something other than
myself; but still, on the party level, my performance must be
convincing, I must dissentble my humanity under a simulated
hyenity. I contrive to fail on both counts at once, because my
behaviour is as much human as hyenine-how then could it
distract attention from my humanity, to which so many other
things point, or prompt anyone even to t h n k specially of
hyenas? A pretence must be not n~erelyllke but distinctively -

like the genuine article simulated: you will hardly pretend to

be angry by simulating the behaviour of an angry man in per-
fect control of himself (though of course it might help if you
were to say 'I am angry' too).
In (g), you evidently have a wrong- idea of what a hyena is.
The puzzle, such as it is, is exactly parallel to that about the
man who, trying to draw a map of France, draws an outline
which is that of Italy: its solution throws no special light on
pretending, but rather on doing and intending to do in general
-for pretending to be doing something is of course as good
a case as another of doing somethng. You are meaning or
trying to pretend to be a hyena, but actually behaving llke a
kangaroo: h s is the correct and the shortest accurate way of
-

describing the situation. There is no short answer to the ques-

tion 'IS he preten&ng to be a hyena. or isn't he?' nor to 'Is he
 PRETENDING 267

pretending to be a hyena or a kangaroo?' since such simple

expressions are not adequate to cope with such a complicated
case.
It is quite nlisleading to handle pretending in the way it is
so often handled, as identical with being (or being doing)
- .

except that some special feature is left out-and ~ r . - B e d f o r d

is no worse in this respect than those he is attacking who say,
for example, that pretending to be in pain is just the same as
being in pain except that you do not feel pain, or that pretending
to bd angry is behaving like a really angry man only without
feeling l ~ k eone. Even if there were, what there is not, a general
bar against PBnl being the same as GBs, and even if it were
possible, w h c h even h e n it would not be, to give a general
account of the precise way in which PBm must always fall
short of GBs, still such an account would not explain pretend-
ing: for there are many situations in which I behavd like an
angry man without being really angry, which are nevertheless
not cases of pretendmg. For example, I may be a rough
diamond, or have odd manners, or be strangely-insensitive, or
not be attending to what I am doing: or I may be acting or
rehearsing or merely imitating or mimichng. And yet these
are only some of the simplest things from which m re tend in^
has to be distinguished, much less near to it than, say affecting
or shamming or feigning or posing as. To be m re tend in^, in
the basic case,I I must be trying to make others believe, or to
give them the impression, by means of a current ~ersonalper-
formance in their presence, that I am (really, only, &c.) abc,
in order to disguise the fact that I am really xyz. T o neglect
to notice all this is to put in the bathwater without the baby.
Even so, we are far from having a full account of the nuances
of pretending. For example, in a pretence there is for prcference
I neglect here such parasitic cases as let's-pretending and pretending-to-
oneself, besides, for the present, pretending-that. Still less have I space t o take
on 'pretensions', 'the Old Pretender' and the like: but it is not too difficult in
fact to fit all these into their 3ppropriate niches in the concept, and sometimes
they shed light, as, for example, the contrast bctween 'affected' and 'pre-
tentious' may help to point the contrast betwee11 affecting and pretending.
268 PRETENDING
an element of the extempore, and in the situation that prompts
it an element of emergency-there is at least something that
has to be hldden. True, there are 'elaborate' pretences: but if
-

there is too much of this, with m h g - u p and dressing-up

llke an actor rather than a mimic or a diseuse, we begin to
prefer to speak of, say, impersonation or imposture or disguise.
-

To pretend to be a bear i s one thing, to roam the mountain

valleys inside a bearshn rather another. True, there are pro-
<
longed pretences- How long', the cry goes up from the
eternal drawing-room, 'must we two go on pretending to one
another?'-but still we prefer to say that Col. Barker posed
for twenty years as a man rather than that she pretended for
twenty years. Again, if there is no sort of urgency to hide what
we elect to hide, we may prefer to speak of a leg-pull or of
affectation or a pose. Yet these are nuances, for it is probably
legitimate enough, in these days, to extend 'pretending' to-

c&er most of these cases if we do not care for piecision, just as

-

we can use 'pretended he was going to' to cover those cases

where, more specifically, 'he made a feint', i.e. where he made
-

a small movement in one direction to distract h s opponent's

defence, m a s h g h s true intention.
There remains, however, more to be said about one essential
feature of pretending, namely that the pretender must be
present and active in person to effect the deception by dint of
his current behaviour. In the example of the panther above,
the awkwardness is not merely that what is to be dsguised is
not 'on the scene' to be disguised, but also that the pretender
is not on the scene to do the disguising, features both essential
to pretending though of course not essential to many other
forms of deception. I may camouflage a factory as a housing
estate, in order to deceive the enemy in an emergency, but this
is not to pretend that it is a housing estate (still less does it
pretend to be a housing estate). I may pretend to have been
furious by emerging from the conference room breathing hard
and making derogatory remarks about the proceedings: but
not by leaving traces in the conference room-bitten carpets,
 PRETENDING 269

maybe-designed to make you t h ~ n kI was furious at the time.

In pretending, contemporury brbnviour misleads as to contemporary
fact, here the contemporary fact that I am not one recovering
from or still suffering from the after-effects of f u r - , or niulling
over fresh nlemorics of f ~ ~ r y .
This brings me to the last point I shall consider, the con-
struction 'pretending that'. It may be the availability of t h s
handy and flexible construction that has led to the ever in-
creasing popularity of 'pretend', since such neighbowing verbs
in the family as 'affect', 'feign', 'dissemble', and the llke have
never acquired a 'that' construction. It may even seem that,
equipped with a that-clause, pretending achieves emancipation
- - - - - -

from some of the limitations inherent in pretending-to: when

pretending-to I can deceive only as to my oii?ri states or activities,
and contemporary ones at that, but surely when I 'pretend that
it was in thd garage yesterday' I deceive-as to sometlung other
than my own states or activities, and something non-contem-
porary at that.
However, it is not easy to be certain that there is in fact any
systematic difference between ret ten ding-to and re ten ding-
that, let alone that just suggested.' What is the difference
between pretending to be on your way to Antarctica and pre-
tending that you are on your h a y to Antarctica? O r behveen
pretending not to remember her face and pretendmg that you
do not remembcr it? One feels inclincd to say: with pretending-
that the stress is on the suppression or coilccalnlent of know-
ledge or memory or thought or belief or awareness, in short
of some 'cognitive state', and what is simulated is llkewise some
cognitive state? Thus to pretend that you are in love with her
is to dissemble your awaierless that you are not, to pretend to
be in love with her is to dissemble your indifference or aversion
to her. Hence the fact, it might be argued, that in pretending-
It might be relevant, but would take too long, t o consider the other verbs
('hope', &c.) which can take both constructiotls: they are not particularly
numerous.
Yet it seems scarcely right t o say: 'pretcnd that'
(or the like) that'.
- 'pretend t o believe
270 PRETENDING
that the pretence-behaviour is particularly liable to take the
form of verbal behaviour, since that is particularly apt for
creating impressions about our cognitive states. Moreover the
- -

apparent emancipation of pretending-that can be on these lines

both accounted for and discounted: when I m re tend that it was
in the garage yesterday I am still only dissembling my own
-

currenl awareness (memory, knowledge, belief) that it was not:

but of course awareness can be awareness of things other than
my own states or activities, and of non-contemporary things.
oreo over it seems possible in this way to account-for pie-
tending-to-oneself or let's-Pretending, the former of which
strongly, if not exclusively, prefers the 'that' construction.
Here we have a sort of 'make-believe'-we suppress our actual
beliefs and simulate others.
Yet still in all cases of pretending-that, though it may be
only a cognitive state that is simulated and though verbal
devices may be often employed, it remains true that there is
an i~nmediateconnexion with non-verbal behaviour. Pretend-
ing that I am on top of a mountain may seem a less active affair
at first than pretending to be on top of a mountain, yet still it
differs very considerably from merely imagining that I am on
top of a mountain: pretending-that is a preliminary to or even
accompanied by behaviour suchbas inhaling deeply or pointing
downwards ('Let's pretend we're giraffes and eat the leaves'),
whlle imagining-that is a preliminary perhaps only to asking
myself certain questions-How should I feel?, &c., while my
public behaviour will scarcely go beyond a faraway look,
which is certainly no part of the imagining. For this reason I
can 'always' imagine, for exaniple, that my prison walls are
not there, but it may be 'no good' pretending they are not
there, they are solid enough to stop me doing the things that
follow on the pretending.
But how far can all this be pressed? Is pretending to be
playiilg chess always so very different from pretending that
you are playing chess, or again (perhaps still more) from pre-
tending you are playing chess? Perhaps all that sllould be said
 PRETENDING 27 I:

is that the more it is a case of going through the nlotions the

-

more likely we are to prefer 'to be playing' or 'to play': while

the less this is necessary and the more we can put the deception
across by verbal means or by simulating a belief the more we
shall prefer the 'that' construction.

What, finally, is the importance of all this about pretending?

I will answer t h ~ sshortly, although I am not sure importance
is important:' truth is. In the first place, it does seem that
- -

philosophers, who are fond of invoking pretendmg, have

exaggerated its scope and distorted its meaning. In the second
place, in the long-term project of classifying and clarifying all
possible ways and varieties of rlot exactly doing things,
-
which has
to be carried through if we are ever t o understand properly
what doing things is, the clarification of pretending, and the
assignment to it of its proper place within the family of related
concepts, must find some place, if only a humble one.
I dreamt a line that would make a nlotto for a sober philosophy: Neither
a be-all nor an end-all be.
 T H R E E W A Y S OF S P I L L I N G I N K 1

P A R T SI, 11, and 111 of the following paper arc taken, almost
verbatim, from J. L. Austin's longhand draft, found alllong his
papers after his death, of a lecture he gave to the American Society
of Political and Legal Philosophy, which met in Chicago in 1958.
Authenticity of expression can be claimed for these sections, though
no doubt Austin would have changed much had he hiinself com-
pleted his revisions for publication in the yearbook of the Society:
R~s~orrsibility(Nomos 111), edited by Carl Fricdrich (New York,
1960). A11 outline of this lecture, which was. 'distributed at the
meeting and available to the participants', was published as an
-

appendix to that volume.

The introductory section was not fully written out in Austin's
manuscript, but has been pieced together from notes made on several
separatc sheets, and with the help also of Austin's notes for seminars
he conducted at Oxford, over a period of many years, under the
general title 'Excuses.' In thesc seminars he discussed not only
material incorporated in the present paper, but also much on which
'A Plea for Excuses' was based.
The inanuscript bears the title 'Responsibility', since this was
the general topic of the meeting of the Society at which the lecture
was givcn, The published appendix, however, i s entitled 'Three
Ways of Spilling Ink', and this is lllost probably the title Austin
himself would have chosen had he published thc paper.
The editor is to Mrs. Austin for pertl~issionto subnut
tKe papcr for publicatio~iand to Mr. G. J. Warnock, who has read
both the illanuscript and the present paper and has made many
valuable suggestions for its iniprovemcnt.
L. W. FORGUSON
Reprinted from The Philosophical Review, vol. 75, No. 4 (1966), by per-
mission of chc editors.
 T H R E E W A Y S O F SPILLING I N K 273
M O S Tof what I have to say about respo~lsibilityin general
I have said in anothcr p1ace.l But of coursc the point of what
I had to say therc was that thcrc isn't 111uch point in discussing
it in general terms. I shall reycat it hcre only in summary. It is
-

a view which I havc not so 111uch merely lleld, but used in

practice for so~llctwcnty years, and havc found it consistently
to pay off. Briefly, it is thc idca which Aristotlc had in a primi-
tive way, without having to fight free of the toils of sophistica-
tion that now encuinber us: namely, that questions -
of whether
a pcrson was responsible for this or that are prior to questions
of freedom. Whatever Aristotlc's idea may have been it
~ v o r k e dthis way: to discover whethcr soineone acted freely or
not, wc must discover whcther this, that, or thc othcr -plea will
pass-for cxample, duress, or mistakc, or accident, and so forth.
W e may hopc to profit then, in this arca of inquiry, from
thc careful study of what we may call, for the sakc of a word,
exctises-of the different ways, and different words, in which
on occasion we may try to get out of things, to show that wc
didn't act 'freely' or were not 'responsible'. But if we are
going to consider this subject and thcsc expressions, wc ought
to attend as wcll to what might bc called words of aggravatior~-
words, that is, that not only don't get us out of things, but may
actually nlakc things worsc for us, or at any ratc i l ~ a yoften
bring it1 the very things that excuses, if we had any, whuld bc
designed to rule out. I shall concctltratc here on a pretty
narrow topic, since I don't know c~lougll (or even rliirrk I
know enough) about the whole subject: what follows is a
sample onlyof some contributions that might be of use.
I n considering responsibility, few things arc considered lnore
importailt than to establisll whethcr a lllali irlt~trdcdto do A,
or whether he did A i~ltcntionall~. But thcre are at least two
other familiar words iillportarlt in this respect. Lct us distill-
I L

g ~ ~ i between
sh acting irrtc~rrtioriilll~
and acting dolibc)rntely or on
ptrrposc, as far as this can be donc b y attending to what lallguagc
can teach us.
'A Pica for Excuscs' (1956).
274 T H R E E W A Y S OF SPILLING I N K
A schoolteacher may ask a child who has spilled the mk in
class: 'Did you do that intentionally?' or 'Did you do that
deliberately?' or 'Did you do that on purpose (or purposely) ?'
It appears at first sight to matter little which question is asked.
he^ appear to mean the same or at least to come down to the
same in thls case. But do they really ? There are in fact techniques
available for distinguishing between these expressions. I cannot
exploit these by any means fully here, but only indicate the
resources available. W e may consider, for instance, for a start:
(i)
. . imagined or actual cases, and (ii) the 'grammar', 'etymo-
-

logy', and so forth of the words.

First let us consider some cases. Actual cases would of course

be excellent: we might observe what words have actually
been used by commentators on real incidents, or by narrators
of fictitious incidents. Howcver, wc do not have the time or
space to do that here. W e must instead imagine some cases
(imagine them carefully and in detail and comprehensively)
and try to reach agreement upon what we should in fact say
concerning them. If wc can reach this agreement, we shall have
some data ('experimental' data, in fact) which we can then go
. -

on to explain. Here, the explanation will be an account of the

meanings of these expressions, which we shall hope to reach
by using- such methods as those of 'Agreement'
- and 'DiEerenceY:
what is in fact present in the cases where we do use, say, 'de-
liberately', and what is absent when we don't. Of course,
we shall then have arrived at nothing more than an account
of certain ordinary 'concepts' employed by ~ n g l i s hspeakers:
but also at no less a thing. And it is not so little. These concepts
will have evolved over a long time: that is, they will have faced
the test of practical use, o f continual hard cases better than
their vanished rivals.
Here, then, are some cases.
I. Suppose I tie a string across a stairhead. A fragile relative,
from whom I have expectations, trips over it, falls, and perishes.
 T H R E E W A Y S OF SPILLING INK 275
Should wc ask whcthcr I tied thc string thcrc ii~tc~~tionally?
Wcll, but it's hard to see how I could have donc such a thing
~nintcntionall~, or cvcn (what is not thc sainc) not done it
intentionally. You don't do that sort of thing-by accidcnt?
By mistakc! Inadvcrtcntly? O n thc other hand, would I be
bound to admit I did it '011 purposc' or 'purposcly'? That has
an ugly sound. What could the purpose have bccn if not to
trip at lcast soineoilc? Maybc I had bcttcr claiin I was siillply
passing thc timc, playing cat's cradlc, practising tying knots.
2 I necdcd money to play thc ponies, so I dipycd illto the
till. Of course, I irttermhd (all the time) to put it back as soon
as I had collectcd my winnings. That was illy intention: I
took it with the intention of putting it back. But was that my
ptrrpose in taking it? Did I takc it for the purpose of, or on
purposc to, put it back? Plainly not.
3. As I drivc up, I scc that there is broken glass on thc road-
way outsidc illy home; so I throw it on to the sidewalk, and
a pedestrian later stuinbles over it and is injured. Very likely
I did throw thc glass on to the sidewalk ii~tcntionall~; again,
it was not unintentionally-I kilcw it was the sidcwalk all
right, etc. Rut did I do it on purposc, purposely? Did I do it
deliberately ? Conceivably, but in the way wc should naturally
imagine
-
the incident, certainly not either.
4. The noticc says, 'Do not feed the penguins.' I, however,
t
fecd them peanuts. N o w peanuts happento be, and thcsc prove,
fatal to these birds. Did I fced then1 peanuts inteiltioiially ?
Beyond a doubt: I an1 no casual pcanut shcdder. But de-
libcrately? Well, that sccms perhaps to raise the qucstion, 'Had
I read the notice?' Why does it? Or 'on purpose?' That seems
to insinuate that I knew what fatal results would cnsuc. Again,
why?
W e may also consider cases that are stereotypes, the ones
evoked by c1ichi.s. Here arc some.
W e say that A wounded B with thc intelltion of killing him,
or of causing him grave bodily injury; or, inorc formally,
with intent to kill him, and so forth. W c do not say, 'A
276 T H R E E W A Y S OF SPILLING INK
woundcd B for the purpose of killiilg him.' Why not? Because
the killing and the wouilding arc 'not sufficiei~tlyseparate'-
arc 'too iiltirnatcl y connected'; because therc arc not 'trrw
thirgs' that arc done? But what does this rcally mean?
Again, we ask this young man who is paying attentions to
our daughter to declare his intcntions. What arc his intcntions?
Are his intci~tio~ls honourable? Here, would it make any
difference if we asked him what was thc purpose of thcsc
attentions, whether he has some purpose in view, whether he
is doing these things on purposc or for a purpose? This tnakes
his conduct seem more calculated, frames hiin as an adventurer
or seducer. Instead of asking him to clarify the position,
perhaps to himself as well as to us, are we not now asking him
to divulge a guilty secret?
Another clichi.: we find ourselves fairly often speaking of
a 'deliberate intention'; 'with the deliberate intention of forcing
the lock', for example. Just as we may speak of a deliberate
choice or a deliberate decision. But we do not speak of an
intentional deliberation; nor (except 111 special cases which
cannot here be discussed) of an iiltei~tionaldecisioil or an in-
tentional choice.
Perhaps it would help to think of kinds of cases in which
a thing is done ii~tcntioi~allybut not deliberately, and so forth:
cases, that is, where these adverbial exprcssions are cxprcssly
dissociated. The way this happens will c-omrnoi~l~ reveal somi
'opposite' of one of the three expressions which is rrot an
'opposite' of the othcr two.
For exatnple, supposc I do a thing impulsively, and possibly
cvcii on impulse. Then I shall not be doing it dclibcrately-
and indccd to say that I did it impulsively (and perhaps even
on impulsc) would surcly bc to rile out the suggestioh that I
did it deliberately. For exaiuplc, at a certain juncturc in the
coursc of our quarrcl, inovcd perhaps by soille turn of cmotion
or mclnory, I iillpulsivcly strctch out illy hand to illakc things
up, and excrt all my tact to thc salnc end. Now this is intcn-
tioi~alcnough:I intend to put out my hand, to bury t l ~ chatchet.
 T H R E E W A Y S O F SPILLING INK 277
Actually, I did even stretch out my h a i d on purpose, pur-
posely. Yet it was not donc deliberately: within twenty minutes
I may be regrctting it. The inlpulsc is strong: I didn't stop to
think (but about what?). I act precipitately, so probably not
dclibcratcly, but of course I knew what I was doing and meant
to do it, even perhaps used my wits in doing it adroitly. (I
may havc stopped to think about that: the impulse may have
been merely to make friends, holding out my hand something
I thought up to do the trick.) If I acted not-even on impulse,
but quite sporttatzcorrsly (rather tricky, this), and so even more
evidently not deliberately, it is at least plausible to say that
I still acted intentioilally (cf. Sir Walter Raleigh). Again, a
man put into agony of mind and fearful indecision' by solne
crisis may adopt some course such as running back into the
blaze. No doubt he runs back into the blaze intentionally
enough; he even (perhnp~) decides to run back-though of
course this is not necessary for hiin to do so 'intentionally'.
But I think it might well be agreed he did not do so de-
- -

liberately. These examples will suffice to show that what is

done intentionally and purposely need not be done dcliber-
ately. Moreover, they appear to show sonie colnmon charac-
teristics: there is something 'precipitate' about the act in
every case.
On the other hand, it is fully possible to act both deliberately
and intentionally yet 'not on purpose', or at least (ifthis is the
same thing-there are distinctiotls here which wc shall havc
-

to neglect) for no purpose, without any purpose, purposelessly.

So to act may be, typically, to act wantonly. A gang of boys
decapitates, serintim, every one of the line of young trees newly
planted along our street: this is deliberate, wanton damage.
But they have, we may say, no interest in killing the trees;
very likely they haven't give11the matter a thought. Do chil-
dren pull the wings off Aies 'on purpose?' ~ c see t t l l e ~ at
l ~ it,
and it is patent that they do it in tentionally, and also dclibcrately.
Perhaps owing to someone elsc's fault: but for reasoils givcrl below wc
wiil cscl~tdt*this possibility, and supposc it an 'accidental' crisis, such as firc.
278 THREE W A Y S OF SPILLING I N K
So far we have shown that a thing donc intentionally nccd
not be done deliberately or on purpose, but what about con-
versely? Can soinething bc done deliberately or purposely
but not intentionally? Call wc think of a case in which some-
thing is donc deliberately but not intentionally? Certainly this
seems more difficult. However, tllere are cascs.
I am summoned to quell a riot in India. Speed is imperative.
My mind runs on the action to be taken five miles down the
road at the Residcncy. As I set off down the drive, my cookboy's
child's new gocart, the apple of her eye, is right across the
road. I realize I could stop, get out, and move it, but to hell
with that: I must push on. It's too bad, that's all: I drive right
-

over it and am on my way. In this case, a snap decision is

taken on what is essentially an incidentnl matter. I did drive
over the gocart deliberately, but not intentionally-nor, of
course, unintentionally either. It was never part of my inten-
tion to drive over the gocart. At no time did I intend to drive
over it. It was incidental to anything I intended to do, which
was simply to get to the scene of the riot in order to quell it.
However 'odd' it may sound, I feel little doubt that we should
say here that we did run over the gocart deliberately and
that we should not care to say we ran over it intentionally.
W e never intended to run over it,
A similar account should probably be given, too, of some
things that will, it can be foreseen, follow as consequcnces
-

or results of our doing certain actions-namely, that these

things are 'done' by us deliberately but not intentionally. For
example, I realize that by insisting on payment of due debts
I am going to 'ruin' my debtor-that iS, Le will be ruined as
a consequence of being compellcd to pay. I have absolutely no
wish to ruin him, even wish not to: but maybe if I don't get
payment both I and others are going to suffer severely; and
very likely I think hc has been faintly improvident. So I de-
mand payment. He is ruined and, if you like, I ruined him.
If this is said-I might resist and resent the imputation a bit-
I think it must be admitted that I did ruin him deliberately;
 T H R E E W A Y S O F SPILLING INK 279

not, howcver, that I ruined him intentionally. At 110 time did

I intcild to ruin him; it was nevcr any part of my intention.
(This, if it bc admittcd, is an especially intcrcsti~~g case: for
plainly I am trot herc rcsponsiblc for his ruin.)
Finally, can a thing bc done on purpose, but yet not in-
tentionally? This scei~iscvcn more difficult, and rnay actually
be impossiblc. Howcvcr, the expression 'accidentally on pur-
pose' hints, at least ironically, that something of the sort may
bc possible; for, if done accidentally, it is not done intcn tionally.
But how ironical is this expressioil? (Perhaps a case comparable
to the dcbt-collection case could be constructed herc.)

W e now turn to our second general source of inforination:

grammar and philology. Here we find that 'purpose', 'intend',
and 'dclibcrate' exhibit numerous and striking differences.
I. Delibcmte and deliberotion, thc verb and the noun, differ
from both intend/intention and purposc./prrrpose in some ways
in which the latter pair resemble each other. Thus, 'I am de-
liberating' could only be used to describe a process that is
going on: but 'I am intending' and (if it exists) 'I am pur-
posing' could not be used to describe a process. In line with
this is the fact that deliberations may be protracted, but inten-
tions and purposes cannot be so.
IA. The use of 'I intend' (and, so far as it exists, of 'I purpose')
is quite different from 'I deliberate', which if it exists could
only be a habitual present, describing what I ordinarily do, as
in 'I deliberate before I act'. 'I intend to X' is, as it were, a sort
of 'future tense' of the verb 'to X'. It has a vector, committal
effect like 'I promise to X', and, again, like 'I promise to X',
it is one of the possible formulas for making explicit, on occa-
sion, the force of 'I shall X' (namely, that it was a declaration
and not, for example, a forecast or an undertaking). W e
night feel inclined to say: it is almost an 'auxiliary verb'.
But the fact of the matter is that terms like 'future tense' and
280 T H R E E WAYS O F S P I L L I N G INK
'auxiliary verb' were not invented with the need to do justice
to such a word as 'intend' in mind. A complete reclassification
-

of these archaic ternis is needed. That reclassification is needed

is shown, for example, by the fact that there is some oddity
about the combination 'I shall intend'.
2. If we next consider the adjectival terminations found in
'deliberate', 'intentional', and 'purposeful' or 'purposive.' (which
are also of course incorporated in the corresponding adverbs),
it would seem significant that they are different. 'Deli berate'
is of course formed on the Latin past participle: words of this
kind commonly mean that something has happened or been
done. W e should suspect that the process of deliberation,
whatever that may be, has been gone through. So considerate
behaviour is behaviour which shows that there has been con-
sideration (of the feelings of others, as affected by my
proposed activities).
The termination -ful, on the other hand, is conllnonly used
in cases where something may be present or may not be
present: an accretion or extra. 'Thoughtful', 'careful', 'purpose-
- -

ful' alike refer to things we may (but equally may not) be

doing when doing X: we may be taking thought for the
interests and feelings of others, taking care to guard against
possible accidents, or pursuing a purpose.
The termination -al, as in 'intentional', qualifies, or classifies,
as we may say, the act so described much-more 'directly' and
intimately than -ful or -ate. (Incidental note: -ive, as in 'pur-
posive', would pcrform a similar function. ,But of course it is
a term of psychological art, and to my mind requires some
j us tiftcat ion: because all our ordinary terminology, not merely
the adjectival termination, certainly suggests that intention
is related to our action in a more intimate way than its purpose,
and in quite a different way.)
2a. The same lesson is pointed by the negative forms of the
adjectives (and adverbs). There is no accepted negative for111
of 'deliberate'. 'Purposc' takes -less; I may 'have no purpose
(whatsoever)' in doing something, just as I may take no care.
 THREE WAYS OF SPILLING I N K 28 I
But I don't 'have no intention (whatsoever)' in doing some-
thing.
Here something of a general justificatioli may be required.
Why should we supposc it is significant of anything whatsoever
that, for example, these adjectives and their negatives take
different forms? Why shouldn't it just be that 'thought', for
example, is not a Latin word and so can't readily take the
Latin -ate termination, whereas 'consider' can? Why shouldn't
it all be 'euphony', or chance, or meaningless luxuriation?
-

NOW we can admit, and indeed positively welcome, all

these suggestions, and yet adhere to our superstition that the
forms of words and expressions are highly significant for their
meaning. Briefly, let us assume, for the sake of argument and
because we actually have no right to assume anything else,
that 'in origin' speech consisted in any person making any
noise in any situation to signify anything. Let us also assume,
what in a sense is a tautology, that in the very long rrtrl, the forms
of speech
- which survive will be thefittest (most efficient) forms
of speech. Now one general criterion of efficiency1 is simply
this, stated loosely: that any unit of speech U should sound
tonto qrranto like every other unit of speech that 'means' any-
thing like what U means, and ui~likctarlto qrtnnto every other
unit of speech that means anything unlik; what U means: or
that small variations in meaning should be signified by small
concolnitant variations in sound. This principle will account,
on my view, not merely for the phenotnenon of the survival
of words in groups where similar-sounding words tllean
similar things (for cxample, 'fumblc', 'tumble', 'stumble', and
the like) b i t for much of what ought to be included in cty-
mology, and for thc whole general evolution of morphology,
syntax, and grammar.
In this account of the origin of speech forms on evolutionary
lines, it will be seen that allorvnrlicx may be-indeed, to sollic
extent is already fully-made for chancc, for luxuriancc, for
There are others of great importancc: brevity, Icanl-ability, ctc. But of
these, some are closcly collnccted with the abovc in a variety of ways.
282 T H R E E W A Y S OF S P I L L I N G I N K
sound preferences (euphony), and for borrowings. Blir still,
in the long run, thc exprcssioiis which survive will bc such
that their grammatical and morphological characteristics arc
of the highest significance for their meaning.
3 . The prepositions used with 'intention', 'purpose',
- -
and
'deliberation' to form adverbial or other exprcssions likcwise
point to distinctions between the tlirce words, and associate
them with quite distinct families of words. W e say or1 purposc
(to),for the purpose of, but with the intention of: (possibly also
with the purpose of). It seems clear that '011' and 'for' (coinpare
'on the principle', 'on orders', 'for thc sake of') dissocinte or
sever my purposc froin my current action in a way that 'with'
does not do. There are many cxprcssions containing 'purpose'
('for the usr~nlpurposes', 'to good purpose', 'to some purpose',
- - - -

and so forth) which seem to make purpose as it were iwtpcrsonal

in a way that is never done with intention.
with-'deliberation', the only, and unexciting, pre-
position used is 'after'. 'With deliberation' is indeed found,
-

but then the words are used to dcscribc a certain slow style of
performance, which makes an impression on the observer.
'Deliberately' is used in the same way, as when solileonc eats
his soup dcliberatcly. (Compare the case whcrc he deliberately
eats my soup. Here, if he is well advised, he will make haste
over it.) Now this sort of sccoildary smsc is fairly common
with adverbs of this kind; and 'purposefully' is in fact also
used in this way. W e know the kind of performance it de-
scribes: a purposeful air is one of gctting the preliminaries, the
-

first stages, each stage ovcr trfith, in order to proceed to the next
and get the whole business achieved: it is an air of pressing on.
Strikingly ci~ough,however, there is no expression connected
with 'intentional' which call bc used in this nlanncr. The ex-
planation, whatever it is, would seem to lie in the samc direction
as that of the adjectival ternii~~ations referred to above: inten-
tion is too intimately associated with ordinary action in general
for there to be any sppecial style of performance associated
with it.
 T H R E E W A Y S OF S P I L L I N G I N K 283
4. Finally, we might consider the trailing etymologies of
the three words: for no word ever achieves entire forgetfulness
of its origins. The mctaphor in 'deliberate' is one from 'weigh-
ing' or 'weighing up', that in 'intend' (one which keeps break-
ing through in many cognate words) is from bending or
straining toward (compare 'intent on mischief' and 'bent on
mischief'j. In 'purpose', the idea is that of setting something
up before oneself.

Now let us try to understand the three notions of -purpose,

-
intention, and deliberation in the light of our investigations
so far. W e shan't get so far as defining them though, I fear.
The most subtle of our notions is that bf intention. As I
go through life, doing, as we suppose, one thing after another,
I in general always have an idea-someidea, my idea, or picture,
or notion, or conception-of what I'm up to, what i9men-
gaged in, what I'm about, or in general 'what I'm doing'.
I don't 'know what I'm doing' as a result of looking to see or
otherwise conducting ob~ervations:~ only in rare and perturb-
ing cases do I discover what I've done or mrnc to real& what
I am or have been doing in this way. It is not in such fashion
that I know what I'm doing when I strike the match in the
vicinity of the haystack. (This is the sensc in which in general
-
and obviously I know what I'm doing: contrast the sense in
which you sbpose, dubiously, that I know what I'm doing
when I strike the match so close to the gasoline.) I must be
supposed to have as it were a plan, an operation-order or
something of the kind on which I'm acting, which I am seeking
to put into effect, carry out in action: only of course nothing
-

necessarily or, usually, even faintly, so full-blooded as a plan

proper. When we draw attention to this aspect of action, we
use the words connected with intention.=
II profited when I once heard this remarked by Miss G. E. M. Anscombe.
At this point, the manuscript contains the unfinished sentence, 'When
we use the great majority of 'active' verbs, e.g., 'kick' . . .'What Austin
284 T H R E E W A Y S OF S P I L L I N G I N K
Now although I say that the 'intention' words are connected
with this notion of my idea of what I'm doing, it should not
be supposed that it will always make sensc to-stick in 'inten-
tionally' after evcry verb of action in every ordinary sentcnce
with an active personal verb. Only when there is some sug-
gestion
-
that it might have been uointentional docs it make
non-misleading sense to say, for example, 'I ate my dinner
intentionally'. To this extent, it is true that 'intentionally'
serves to rule out 'unintentionallv'. What would be wholly un-
d J

true is to suggest that 'unintentionally' is thc word that 'wears

the trousers'-that is, that until we have grasped certain specific
--
ways of doing things unintentionally, and except as a way of
'-
ruling these out, intentionally' has no positive meaning. There
are words of this description: 'real', for example, is one. But
in the present case, to mention nothing more, there is the verb
'intend' to take into account, and it-must obviously have a
highly 'positive' sense; it cannot just be used to rule out 'don't
(or didn't) intend'.
Although we have this notion of my idea of what I'm doing
-and indeed we have as a general rule such an idea, as it
were a miner's lamp on our forehead which illutninates
always just so far ahead as we go along-it is not to be sup-
posed that there are any precise rules about the extent and
degree of illumination it sheds. The only general rule is that
the illumination is always limited, and thit in several ways.
It will never extend indefinitely far ahead. O f course, all that
is to follow, or to be done thereafter, is not what I am intending
to do, but perhaps consequences or results or effects thereof.
Moreover, it does not illurninate nil of my surroundings.
Whatever I am doing is being done and to be done amidst
probably had in mind was the fact that most 'active' verbs include, as part of
their sense, some notion of a design or plan to be carried out. Thus, it is generalIy
a mistake to consider them as purely 'bchaviouristic'. That I kick soilleone does
not mean merely that my foot moves sharply into contact with his shin.
Perhaps this is why, in normal contexts, adding the adverb 'intcntionally' is
somewhat redundant. This point was suggested to me by G. j.Warnock.
(L. W. F.)
 T H R E E W A Y S O F SPILLING I N K 285

a background of circumstances (including of course activities

by other agents). This is what necessitates care, to ward off
-

impingements, upsets, accidents. Furthennore, the doing of it

will involve itzcid~.rttnllyall kinds of minutiae of, a t the least,
bodily movements, and often many other things besides.
Thcse will be below thc level of any intention, hotvever detailed
(and it need not of course be detailed at all), that I may have
formed.
There is a good deal of freedom in 'structuringythe history
of someone's activities by means of words like 'intention',
just as when we consider a whole war we can divide it into
campaigns, operations, actions, and the like; but this is fairly
arbitrary except in so far as it is based upon the plans of the
contestants. So with human activities; we can assess them in
terms of intentions, purposes, ultimate objcctives, and the like,
but there is much that-is arbitrary about this unless we take
the way the agent himself did actually structure it in his mind
before the event. Now the word 'intention' has from this
point of view a most important bracketing eject: when the till-
dipper claims that he interlded nll nlorzg to put the money back,
what he is claiming is that his action-the action that he was
engaged upon-is to be judged as a rvhole, not just a part of it
carved out of the whole. Nearly always, of course, such a con-
tention as this will carry with it a contention that his action (as a
whole) is not to be described by the term chosen to describe
[only a part of) it: for example, here, it was not 'robbing' the
till, because the action taken as a whole would not result in the
absence of any money from the till. Recrtier portr wietix sariter
is not to retreat.
Quite distinct is the use of the word 'purpose'. Certainly,
tvherz I arn doing something for a purpose, this will be known
to me, like my intcntioi~s,and will guide illy coildi~ct.Indeed,
like an objective, a purpose will infiuc~lcethe forming of intcil-
tions. But my purpose is soinething to bc achieved or effected
as a result of what-I'm doing, like the death of my aunt, or the
sickness of the penguins if I did indeed feed thch peanuts on
286 THREE W A Y S OF SPILLING I N K

purpose. (Very commonly my purpose is to put myself into

a position to be able to go on with the next action, the next
operatxon in the campaign.) I need not, however, have any
-

purpose in acting (cvcn intenti~nally);~ just as I nccd not take

care or thought. I act for or on (a) purpose, I achieve it; I act
with the intention, I carry it out, realize it.
I act deliberately when I have deliberated-which means
when I have stopped to ask myself, 'Shall I or shan't I?' and .

then decided to do X, which I did. That is to say, I weighed up,

in however rudimentary (sometimes almost notional or fic-
. .

tional) a fashion, the pros and cons. And it is understood that

-

there must be some cons, even when what I do deliberately

is something unexceptionable, such as paying my taxes.
The pros and cons are not confined to moral pros and cons,
nor need I decide in favour of what I think best or what
has the most reasons in favour of it. (Nor, of course, when
I have decided to do it, rr~ustI necessarily carry it out.) De-
liberation is not just any kind of thinking prior to action:
to act with forethought or with or to think
about ways and means-all these show that we took thought,
perhaps over a period of time, but none of these shows that
we were acting deliberately, and indeed are quite distinct
matters from deliberation. Ways and means are a matter for
the planning staff; decision is a matter for the commander.
That there should be slowness in moving into action or
conducting it (so much relied on by lawyers) is the rnerest
symptom.
I will close by adding a general word of warning: there are
overriding- considerations, which inay be operative in any
-

situation in which I act, which may put all three words out of
joint, in spite of the other standard conditions for their use
being satisfied. For instance, I may be acting under a threat:
however much I weigh up the pros and cons, if I act under
the influence of a threat I do not do that act deliberately. This

I E.g. feeding starving children: I need have no purpose here.

 THREE W A Y S O F SPILLING I N K 287

sort of overriding consideration must always bc allowed for

in any case.'
Thc manuscript contains a few further remarks, but not enough to re-
construct a coilclusion that could claim to reflect at all adequately Austin's own
ii~tel~tioi~s
for a conclusion. It call be said from these remarks, however,
that one thing Austin most likely had in mind was this: we should not only
colnpark and contrast thcsc three expressions-'intentionally', 'deliberately',
'on purpose'-with each other, but each should be coinpared and contrasted
with other cxprcssioiis as well (e.g. 'motivc' with both 'intention' m d 'pur-
pose', 'premcditation' with 'deliberation', and 'to mean' with 'to intend').
These are Austin's own examples, reproduced also, though inaccurately, in the
Nornos Appendix. (L. W . F.)
 INDEX
Action, 178-80, 237, 271. Geach, P. T., 2 r 3 n.
Aggravation, 177, 273. 'Good', 14, 20-4.
Anscornbe, G. E. M., 283 11. Grammar, 'logical', 23 1-2.
Arguments, transcendental, 34-7.
Aristotle, x passitrr, 49 n., 70-1, I 2"., Hampshire, S. N., 60.
128 n., 180, 183, 198 n., 273. Handford, S. A ., 2 16 n.
Happiness, 4-19.
Beattie, J., 47. Hare, R. M., 15.
Bedford, E., 253-5, 257-8, "7. H u m , 40 n., 44, 45, 46 n., 47, 49 ll.,
Belief; 77-8. 203.
Berkeley, 40 n., 43, 44 n., 4911., 66,
90, 126 n., 161-2. Idcas, innate, 32 n., 16.
Blame, 181 11. Implying, 63-4, 77, 237.
Bradley, F. H., 161. Infelicity, 2 3 7 fE
Broad, C. n., 41, 45. Instancing, I 40.
Burnet, J., 22 n., 27. 'Intentional', 12passin,.
Intuition, 49-53.
Calling, 147 ff.
Casting, l40.
Joseph, H. W. B., 21 n., 27 n., 50.
Certainty, 90-7. Judgements, analytic and synthetic,
Classing, r 47 ff. 62-9.
Clauses, operative, 236.
Justification, 1 7 6 7 , 181 n.
Coherence, in theory of truth, 130 n.
Concepts, 32-48. Kant, 34 n., 46, 183, 203.
Conditionals, 209-14, 2 I 5-19, 22 I , Keynes, f . M., 228.
229. Knowledge, 4 passirti.
Connotation, 61.
Corres~ondcnce?in of truth, Language, ordinary, 67-9.13 4, '82-6.
121, 124-6, 154-5, 159. -uses of, 234-5.
Daitz, Mrs. E., 160. Leibniz, 43, 49 n,, 20 j 11.

Definition, 134. Locke, 49.

'Deliberate', 12y assint. Mackinnon, 1). M., 32, 33, 40, 43 n.,
Denotation, 61.
Descartes, 40, 47, 205 11. 44 n., 48-
Maclagan,W. G., 32, 33, 36n., .37".,
Describing, 147 tT.
Determinism, 2068, 23 I . 40-53.
Meaning, 45, 46, 3 ym~irtr,251.
Dictionaries, use of, 186-7, 243-4. Misidentifying, 144- j-
Emotions, 103-1 5, 253- j. Misnaming, 83 n., I 3 8 @.
Etymology, 201-3, 260, 283. Misreferring, 138 tT
Eudoxus, 8. Mitigation, r 77.
Excuzes, 8 yassitrt, 273. Moore, G. E., 13, 15, 20, 21, 23, 66,
Exeniplifyii~~, 147 ff. 165, 205-9, 211, 214-22, 228.
Extenuation, 175, 177. Moravcsik, J. M. E., I n.
Morris, C.W., 61.
Facts, 122-4, 129 n., 7 passim.
Feelings, I 03-16. Name-giving, r 36-7, 141.
Freedom, 130, 180-1, 207. 1 Naming, r 39-40.
290 INDEX
Negation, 1289, 145, I 5 1-3. Sensations, 90, 94, 96, 109 n.
Nisbet, R, M., 216 n. Sensa-data, 3 3, 48.
No~ninalis~n,69-70. Sense-giving, I 3 6-7, 14I .
Nowell-Smith, P. H., 2x9-29. Sentences, 56, I r 9-21, 133-
Signalling, I 71-4-
Paronv~~lity, 27-9, 71. Similarity, 69-75, 92 n., 122 n.
Pcirce, C. S., 119 11. Socrates, r 8 3 n.
Placing, 140 ff. Speech-acts, 6 pnssir?~,245, 250-1.
Plato, 31, 3 5 n., 74, 183, 198 11-, Statements, I 1 ~ 8234j:,
, 246-8:
205 n. -and facts, 15663, 168-74.
I'lcasure, 5-19. Stating, 140 tf., 245 K
Pretending, XI pnssim. Strawson, f?. F., 132-3, 154-63,
Yrichard, H. A., 1-14, 17n-, 18, I 67-7 I -
20-6, 28-31, 71. Success, 18.
Pron1ising, 98-103, 236, 239, 242, Sureness, 90-7.
248. Symptoins, and signs, 105-I I .
Propositions, I r 8-19, 13 1-2, 169.
Psychology, 186, 189.
Propositions, I 18-19, I 3 1-2, 169. Tellus, 19.
Psychology, 186, 189. Truth, 5 passirti, 159-61, zSo-I.
Purpose, 12 y nssitri.
Universals, 32-40, 53, 65~70.
Ramsey, F. P., 165- Urinson, J. O., I n., I3 n., rs.
Reality, 58, 67, 86-9, 111-14, 261, U tterances, performative, XO passirfi.
284. - explicit performative, 244.
Referring, I 39-40. - primary performative, 244-
Relations, 49, 50, 52, 53. - forces of, 25 I.
Resemblance, 48-53.
liesponsibility, r 8 I, 273, 279.
Ryle, G., 154, 168.
I Warnock, G. J., r6r-2.
Wisdonl, J., 76, 90, 94-6, 105-6,
Sameness, 120 n., 122 n.
Scllsa, 33-9, 43 11-, 48, 53, 97. I 108-9, 111.
Wordsworth, 20.
 PRINTED I N GREAT BRITAIN
AT T H E U N I V E R S I T Y P R E S S , O X F O R D
BY VIVIAN RIDLER
P R I N T E R TO T H E T J N I V E R S I T Y