Also published in this series

Categories a n d De Interpretatione
J. L. AGKRILL

De Anima Books I I a n d I I I
D. W . H A M L Y N

New impression with supplementary material by Christopher Shields

De Generatione et Corruptione
C. J . F . W I L L I A M S

De Partibus Animalium I a n d De Generatione Animalium I

D.M.BALME
New impression with supplementary material by Allan Gotthelf

Eudemian Ethics Books I , I I , a n d V I I I

MICHAEL WOODS
Second edition

Metaphysics Books M a n d N
JULIA ANNAS

Physics Books I a n d I I
WILLIAM CHARLTON
New impression with supplementary material

Physics B o o k s I I I a n d I V
EDWARD HUSSEY
New impression with supplementary material

Posterior Analytics
JONATHAN BARNES
Second edition

Other volumes are inpreparation

 ARISTOTLE
METAPHYSICS
BOOKSr,J,ANDi;

Translated with Motes

by

CHRISTOPHER K I R W A N

SECOND EDITION

CLARENDON PRESS · O X F O R D

1993
 Oxford University Press, Walton Street, Oxford 0 x 2 6 ϋ Ρ
Oxford New Tork Toronto
Delhi Bombay Calcutta Madras Karachi
Kuala Lumpur Singapore Hong Kong Tokyo
Nairobi Dar es Salaam Cape Town
Melbourne Auckland Madrid
and associated companies in
Berlin Ibadan

Oxford is a trade mark of Oxford University Press

Published in the United States

by Oxford University Press Inc., New Tork
First editionpublished igyi; reprinted with corrections ig8o;
second edition published iggj
© Oxford University Press 1971,1gg3

All rights reserved. N0 part of this publication may be reproduced,

stored in a retrieval system, or transmitted, in any form or by any means,
without theprior permission in writing of Oxford University Press.
Within the UK, exceptions are allowed in respect of any fair dealingfor the
purpose of research or private study, or criticism or review, as permitted
under the Copyright, Designs and Patents Act, ig88, or in the case of
reprographic reproduction in accordance with the terms of the licences
issued by the Copyright Licensing Agency. Enquiries concerning
reproduction outside these terms and in other countries should be
sent to the Rights Department, Oxford University Press,
at the address above

This book is sold subject to the condition that it shall not, by way
of trade or otherwise, be lent, re-sold, hired out or otherwise circulated
without thepublisher'sprior consent in any form of binding or cover
other than that in which it is published and without a similar condition
including this condition being imposed on the subsequent purchaser

British Library Cataloguing in Publication Data

Data available

Library of Congress Cataloging in Publication Data

Data available
ISBN o-ig-824086-4
ISBN0-19-824087-2 (pbk.)

I 3 5 7 9 10 864 2

Printed in Hong Kong

 PREFACE

Т н Е text translated is that o f W . J a e g e r i n the Oxford

C l a s s i c a l T e x t s Series (1957). Footnotes i n the translation
m a r k the few places w h e r e I h a v e diverged from it. M a t t e r
enclosed i n square brackets i n the translation is either speci­
fication o f A r i s t o t l e ' s references, e.g. * [ i o i i * 2 o ] ' , or a l t e r n a ­
tive translations, e.g. *origin [ p r i n c i p l e ] ' , or words not
represented i n the G r e e k , e.g. O r i g i n a l [step]' w h e r e the
G r e e k has *origin'. T h e m a r g i n a l lineations i n the translation
are p l a c e d i n s u c h a w a y that, for instance, the E n g l i s h i n the
line m a r k e d 1003*35 translates matter from lines 34 a n d 35
i n the G r e e k text. T h u s the lineations err, w h e n at a l l , on
the side o f being early. C h a p t e r divisions are traditional
(though not due to Aristotle h i m s e l f ) ; p u n c t u a t i o n , p a r e n ­
theses, a n d p a r a g r a p h i n g are m y o w n .
A m o n g several useful translations o f the Metaphysics into
E n g l i s h S i r D a v i d Ross's still stands out as the most accurate
a n d perceptive. S i n c e there w o u l d h a v e been little point i n
attempting a second time w h a t he has done already, I have
a i m e d at a m o r e literal r e n d e r i n g , for w h i c h there is a greater
need n o w that m a n y students of philosophy come to A r i s ­
totle w i t h o u t G r e e k . I n pursuit of this a i m I h a v e tried to
a v o i d using the same E n g l i s h w o r d to translate two G r e e k
w o r d s ; so, for example, the c o m m o n w o r d 'phanaV is given as
'assert', i n order to reserve 'say' ( a m o n g other renderings)
for the e q u a l l y c o m m o n 4egeirC.
I g l a d l y acknowledge m y debt to m a n y scholars a n d c o l ­
leagues, a n d first to S i r D a v i d R o s s , not only for the m o d e l
w h i c h his translation p r o v i d e d but also for the h e l p I h a v e
received from the e n v i a b l y knowledgeable a n d sensible c o m ­
m e n t a r y i n his edition o f the Metaphysics. I have rarely con­
sulted other commentaries, except A l e x a n d e r . Secondly, I
thank P r o f e s s o r J . L . A c k r i l l , editor ofthis series, whose acute
and thorough criticism of m y translation a n d notes has
vi PREFACE

i m p r o v e d t h e m at m a n y places a n d i n m a n y w a y s . I owe a
great debt to his assistance a n d encouragement. T h i r d l y , I
a m g l a d to r e c o r d m y thanks to m a n y others w h o h a v e r e a d
drafts or h e l p e d m e i n correspondence: to R . J . H a w k i n s w h o
c h e c k e d the proofs a n d m a d e a n u m b e r of useful suggestions;
t o J . B a r n e s w h o c o m m e n t e d o n drafts of Г 3 - 4 a n d E; to M .
S c h o l a r w h o a l l o w e d m e to r e a d his detailed essay o n Г 4 - 5 ;
to W . C h a r l t o n whose version of, a n d notes o n , Physics I I 3
I was able to see before their p u b U c a t i o n ; t o J . A . B a k e r a n d
A . J . P . K e n n y w h o r e a d a n d criticized the translations o f
Г a n d Δ respectively; a n d to M i s s W . F . H i c k e n a n d
P r o f e s s o r P . T . G e a c h . L a s t , I owe thanks to m y College a n d
the U n i v e r s i t y of M i c h i g a n , the one for giving m e the s a b ­
b a t i c a l leave d u r i n g w h i c h h a l f this book was w r i t t e n , the
other for ^ffording m e most congenial surroundings for w o r k ­
i n g o n it d u r i n g p a r t of that leave.

CHRISTOPHER KlRWAN
Exeter College^ Oxford
May 1g70

NOTE ON SECOND EDITION

S o m e corrections were a l r e a d y m a d e i n the 1980 r e p r i n t ;
this second edition introduces a b o u t a d o z e n more. F u r t h e r
C o m m e n t s are a d d e d w h i c h t a k e a c c o u n t o f w o r k p u b l i s h e d
since 1971. T h e B i b l i o g r a p h y is enlarged a n d u p d a t e d . A n
I n d e x L o c o r u m is a d d e d . T h e Subject I n d e x is e x p a n d e d . I
h a v e not tried to revise the Notes, b e y o n d a few m i n o r
changes ( I w o u l d alter m a n y things, i f I w e r e n o w starting
afresh).
M a r g i n a l page references to the F u r t h e r C o m m e n t s h a v e
been a d d e d at a score or so o f p l a c e s i n the Notes.
I w i s h to t h a n k Professor J . L . A c k r i l l , Professor M .
F r e d e , a n d D r R . L . J u d s o n for their kindness i n r e v i e w i n g
the F u r t h e r C o m m e n t s , w h i c h i m p r o v e d t h e m .

CHRISTOPHER KlRWAN
Exeter College, Oxford
November 1gg2
 CONTENTS

TRANSLATION
Book Γ I
BookJ 27
Book Е 66

NOTES
General 75
Book Г 75
Book/1 122
Book£ 183

F U R T H E R C O M M E N T S {1992) 201
Metaphysics 201
Contradiction 203
Identity 208
Being 214
Coincidence 217
Determinism 222

BIBLIOGRAPHY 227

GLOSSARY 239

INDEXLOCORUM 241

INDEXOFNAMESANDSUBJECTS 251
 METAPHYSICS BOOK GAMMA

C H A P T E R 1

1003*21. T h e r e is a discipUne w h i c h studies that w h i c h is

qua thing-that-is a n d those things that h o l d good o f t h i s i n its
o w n right. T h i s is not the same as a n y o f w h a t are c a l l e d the
special disciplines. F o r none o f the others examines i m i -
versally that w h i c h is qua thing-that-is, b u t a l l select some
p a r t o f i t a n d study w h a t is coincidental c o n c e r n i n g t h a t ; as 25
for instance the m a t h e m a t i c a l disciplines. B u t since w e are
seeking origins, i.e. the most extreme causes, it is p l a i n that
these are necessarily a p a r t i c u l a r nature's i n its o w n right. I f
therefore these origins w e r e also sought b y those seeking the
elements o f the things-that-are, the elements too are neces- 30
sarily o f that w h i c h is qua thing-that-is, not coincidentaUy.
H e n c e w e also h a v e to find the first causes o f that w h i c h is
qua thing-that4s.

C H A P T E R 2

1003*33. T h a t w h i c h is m a y be so c a l l e d i n several ways, but

w i t h reference to one thing, i.e. one p a r t i c u l a r nature, not
homonymously. J u s t as that w h i c h is healthy a l l has reference 35
to h e a l t h (either from its preserving, or p r o d u c i n g , or being
a sign o f h e a l t h , or because recipient o f i t ) ; a n d that w h i c h is 1003^
m e d i c a l has reference to m e d i c a l [art] (either it is called
m e d i c a l from possessing m e d i c a l [ a r t ] , or from being n a t u r ­
ally suited t o i t , or from being a n exercise o f m e d i c a l [ a r t ] ) —
aftd w e shall find other things c a l l e d [ w h a t they are] i n w a y s
s i m i l a r to these: j u s t so that w h i c h is m a y also be so c a l l e d i n 5
several w a y s , b u t a l l w i t h reference to one origin. F o r some
are c a l l e d things that are because they are substances; some
because they are affections of a substance; some because

I B
 1003b METAPHYSICS Г2

thcy are a route to a substance, or destructions, o r lacks, or

quaUties, o r productive, o r generative o f a substance o r o f
things c a l l e d [ w h a t they a r e ] w i t h reference to substance;
10 or d e n i a b o f one o f these or o f a substance (that is w h y we
assert that even w h a t is not is a t h i n g that is not).

1003^11. T h e r e f o r e , j u s t as everything that is h e a l t h y falls

to one discipUnc, this is e q u a l l y so i n the other cases too.
F o r it f a l k to one discipline to study not only things c a l l e d
[ w h a t they are] b y v i r t u e o f o n e thing, but also things c a l l e d
[ w h a t they a r e ] w i t h reference to one n a t u r e ; indeed i n a
15 c e r t a i n sense the latter too are called [ w h a t they are] b y
v i r t u e of one thing. P l a i n l y , therefore, the things-that-are
also fall to be studied b y one discipline qua things-that-are.

1003^16. I n every case the fundamental c o n c e r n o f a

discipline is w i t h its p r i m a r y [object], i.e. that o n w h i c h the
others d e p e n d a n d to w h i c h they owe their being c a l l e d
[ w h a t they a r e ] . S o i f t h i s thing is substance, the philosopher
w i l l need to h a v e ^the principles a n d causes of substances.

1003^19. E v e r y one genus falls to one perception a n d

20 discipline; as for instance a l l spoken sounds are studied by
g r a m m a r , w h i c h is one discipline. H e n c e it also falls to
generically one discipline to study a l l the forms o f that
w h i c h is qua thing-that-is, a n d the forms o f those forms.

1003^22. Suppose it true, then, that that w h i c h ts a n d

that w h i c h is one are the same t h i n g — i . e . one n a t u r e — i n that
e a c h follows from the other as origin a n d cause do, not as
25 being i n d i c a t e d by one formula (though it makes no differ­
ence even i f w e believe t h e m like t h a t — i n d e e d it helps). F o r
one m a n a n d a m a n that is a n d a m a n are the same t h i n g ;
a n d nothing different is i n d i c a t e d by the r e d u p l i c a t i o n i n
w o r d i n g of Ъе is one m a n ' a n d 'he is one m a n that is' (it is
p l a i n that there is no distinction i n [the processes of] c o m i n g
30 to be or d e s t r u c t i o n ) ; a n d equally i n the case o f that w h i c h
is one. I t follows obviously that the addition indicates the
same t h i n g i n these cases, a n d that w h i c h is one is nothing

2
Г2 TRANSLATION 1003b

different a p a r t from that w h i c h is. A g a i n , e a c h thing's s u b ­

stance is one n o n - c o i n c i d e n t a l l y ; equally, it is also j u s t w h a t
a c e r t a i n k i n d o f thing-that-is [is]. I t follows that there
are as m a n y forms o f thing-that-is as of thing-that-is-one;
a n d w h a t these are ( I m e a n s u c h things as the same a n d 35
similar a n d others of that kind) falls to be studied b y generi­
cally the same discipline. P r a c t i c a l l y a l l contraries derive 1004*
from this o r i g i n ; but w e must take t h e m as h a v i n g been
studied i n the 'Selection o f C o n t r a r i e s ' .

1004*2. A n d there a r e as m a n y parts of philosophy as

there are substances; so that it is necessary that there be
a m o n g t h e m a first a n d a next. F o r that w h i c h w d i v i d e s 5
directly into g e n e r a ; hence the disciplines too w i l l follow
these. F o r the philosopher is like the m a t h e m a t i c i a n , as he is
c a l l e d ; for that also has parts, a p r i m a r y a n d a secondary
discipline a n d others successively w i t h i n mathematics.

1004*9. S i n c e opposites fall to be studied b y one discipUne,

a n d the one is opposed toplurality (the reason w h y the d e n i a l 10
a n d l a c k o f a thing fall to be studied b y one discipline is that
the one thing o f w h i c h they are the d e n i a l o r l a c k is studied i n
both cases. F o r w e either say b a l d l y that that thing does not
hold good, or o f a c e r t a i n genus; i n the latter case, then,
some differentia is a d d e d to the one, a p a r t from w h a t is there
i n the denial.^ F o r the d e n i a l is the thing's absence, b u t i n 15
the case o f the l a c k a c e r t a i n n a t u r e comes i n too as the
subject o f w h i c h the l a c k is s t a t e d ) — s i n c e the one is opposed
to plurality,^ it follows that it also falls to the discipline m e n ­
tioned to m a k e intelligible the opposites o f the things m e n ­
tioned, both that w h i c h is other a n d dissimilar a n d uruqual^ a n d
everything else c a l l e d [ w h a t it is] either b y v i r t u e o f one of
these or b y virtue o f p l u r a l i t y a n d the one. A m o n g these is 20
contrariety, since contrariety is a k i n d ofdifference, a n d differ­
ence is otherness. I t follows that since the one is so c a l l e d i n
several ways, these things also w i l l be c a l l e d [ w h a t they a r e ]

' Kcep the words obelized in the ОСГГ.

» Kcep the words bracketed in the O C T .

3
 ioo4* METAPHYSICS Г2

i n several w a y s . Nevertheless it falls to one discipline to m a k e

a l l o f t h e m intelligible; for it w i l l fall to another not i f [they
are c a l l e d w h a t they a r e ] i n several w a y s but only i f the
25 formulae are connected neither by virtue o f one t h i n g nor b y
a reference to one thing.

1004*25. S i n c e everything is connected to that w h i c h is

p r i m a r y (as for instance, a n y t h i n g c a l l e d one to the p r i m a r y
o n e — a n d the same c a n be asserted to h o l d also o f the same
a n d other a n d contraries), it follows that after d i v i d i n g the
n u m b e r o f w a y s i n w h i c h e a c h t h i n g m a y be c a l l e d [ w h a t it
is], w e h a v e to display, w i t h reference to w h a t is p r i m a r y i n
e a c h p r e d i c a t i o n , i n what w a y it is so c a l l e d w i t h reference
30 to t h a t ; for some things w i l l be so c a l l e d from possessing it,
some from p r o d u c i n g it, others i n other s u c h w a y s .

1004*31. I t is obvious, therefore, that it falls to one disci-

line to discuss these things a n d substance (that w a s one o f
the perplexities w e listed [B 1. 9 9 5 ^ 1 8 - 2 5 ] ) ; a n d it falls to the
1004^ philosopher to be c a p a b l e o f studying a l l o f t h e m . F o r i f not
the philosopher, w h o w i l l it be w h o investigates w h e t h e r
Socrates a n d Socrates sitting d o w n are the same thing, or
w h e t h e r one is c o n t r a r y to one, or w h a t the c o n t r a r y is a n d
i n h o w m a n y w a y s it is so c a l l e d ; a n d e q u a l l y w i t h the other
5 questions o f t h a t k i n d ? Therefore, since these things a r e i n
their o w n right affections o f that w h i c h is one qua one a n d o f
that w h i c h is qua thing-that-is, not qua n u m b e r s or lines or
fire, p l a i n l y it falls to that discipline to m a k e intelligible both
w h a t they a r e a n d the things c o i n c i d e n t a l to t h e m ; a n d those
w h o e x a m i n e these questions a r e at fault not because they
are not philosophizing, b u t because substance is prior, a n d
10 o f substance they h a v e n o comprehension. F o r j u s t as there
are affections distinctive o f n u m b e r qua n u m b e r — a s for
instance oddness, evenness, c o m m e n s u r a b i l i t y , equality,
excess, d e f i c i e n c y — a n d these h o l d good o f n u m b e r s both i n
their o w n right a n d w i t h reference to one another ( a n d
equally there are others distinctive ofthings solid, changeless,
15 changeable, weightless, a n d possessing w e i g h t ) ; so too c e r t a i n

4
Г2 TRANSLATION

things a r e distinctive o f that w h i c h is qua thing-that-is, a n d

these are the things about w h i c h it falb to the philosopher to
investigate the truth·

1004b17. I t is a sign o f t h i s that dialecticians a n d sophists

assume the s a m e guise as the philosopher. Sophistic is o n l y
i m a g i n e d [to be] science. E v e r y t h i n g is discussed b y d i a l e c - 2 0
ticians, a n d that w h i c h is is c o m m o n to everything; a n d
p l a i n l y these things are discussed b y t h e m because these things
are p r o p e r to philosophy. Sophistic a n d dialectic do indeed
range over the same genus as philosophy, but differ from it i n
the one case b y the type o f c a p a c i t y , i n the other by the life 25
chosen; dialectic probes w h e r e philosophy seeks u n d e r s t a n d ­
ing, a n d sophistic is i m a g i n e d [to be science] but is not really.

1004^27. A g a i n , every c o n t r a r y o n one side o f the table is a

lack, a n d a l l o f t h e m r e d u c e to that w h i c h is a n d that w h i c h is
not, a n d to one a n d p l u r a l i t y (as for instance keeping-the-
same is o n the side o f the one, change o n the side o f pluraUty).
P r a c t i c a l l y everyone agrees that the things-that-are, a n d sub­
stance, are composed out o f contraries: at a n y rate, every- 30
one descrfbes the origins o f things as contraries, whether
o d d a n d even or hot a n d cold o r l i m i t a n d limitless or love
a n d strife, a n d it is obvious that a l l the others also reduce to
the one a n d p l u r a l i t y (we must take the reduction for 1005·
granted), a n d the origins proposed by others also find their
place w i t h o u t exception u n d e r these genera. I t is therefore
obvious from this too that it falk to one discipUne to study
that w h i c h is qua thing-that-is. F o r a l l things either are or are
m a d e u p o f contraries, a n d contraries originate i n the one
a n d pluraUty. T h e latter fall to one discipUne, whether or not 5
they are called w h a t they are by virtue o f o n e thing. D o u b t ­
less the t r u t h is that they are n o t ; nevertheless, even i f t h a t
w h i c h is one is so c a l l e d i n several ways, the others w i l l be so
called w i t h reference to the first; a n d equally so w i l l c o n ­
traries. ( T h i s is so, even i f t h a t w h i c h is, or that w h i c h is one,
is not u n i v e r s a l , i.e. the same i n every case, or separable; 10
doubtless they are not, but some o f t h e m are related to one

5
 ioo5* METAPHYSICS Г2

thing, others form a succession.) T h i s also explains w h y it

does not fall to the geometer to study the question w h a t is the
contrary, or complete, or one, or thing-that-is, or the same, or other,
except o n the basis o f a hypothesis.

1005*13. I t is therefore p l a i n : that it falls to one discipline

to study that w h i c h is qua thing-that-is, a n d those things that
h o l d good o f it qua thing-that-is; a n d that the same discipline
15 undertakes the study not o n l y o f substances but o f w h a t e v e r
holds good o f t h e m also, both the things m e n t i o n e d a n d prior
a n d posterior a n d genus a n d form a n d whole a n d part a n d the
others o f that k i n d .

C H A P T E R 3

1005*19. W e h a v e to say whether it falls to one, or a different,

20 discipline to d e a l w i t h the things w h i c h i n m a t h e m a t i c s are
t e r m e d axioms, a n d w i t h substance. I t is i n d e e d obvious that
the investigation o f these too falls to one discipline, a n d that
the philosopher's; for they h o l d good o f e v e r y thing-that-is
a n d not o f a c e r t a i n genus, separate a n d distinct from the
others. E v e r y o n e uses t h e m , it is t r u e ; because they are o f
25 that w h i c h is qua thing-that-is, a n d e a c h genus is a t h i n g -
that-is. B u t everyone uses t h e m j u s t so far as is sufficient for
h i m , that is, so far as the genus extends about w h i c h he is
c a r r y i n g out demonstrations. S i n c e it is p l a i n that they h o l d
good o f a l l things qua things-that-ar^ (for that is w h a t
they h a v e i n c o m m o n ) , it follows that their study too falls
to h i m w h o makes intelligible that w h i c h is qua thing-that-is.
T h i s explains w h y none of those w h o conduct specialized
30 investigations e n d e a v o u r to say a n y t h i n g about t h e m , as to
w h e t h e r or not they are true. Geometers a n d arithmeticians
do not; some students o f n a t u r e do, but that is not surprising,
since they alone h a v e considered that they w e r e investigating
the w h o l e o f nature, i.e. that w h i c h is. B u t since there is
someone still further above the student o f n a t u r e (for nature
35 is one p a r t i c u l a r genus o f t h i n g - t h a t - i s ) , the investigation of

6
гз TRANSLATION 1005»

these things also must fall to h i m w h o studies w h a t is u n i ­

versal, a n d p r i m a r y substance. T h e s t u d y o f n a t u r e is also a
science, but not p r i m a r y . 1005^

1005^2. T h e endeavours of some o f those w h o discuss

[their] truth, as to h o w [they] ought to be accepted, are due
to l a c k o f t r a i n i n g i n analytics. T h e student ought to come
ready e q u i p p e d w i t h knowledge o f these things, not seek it 5
while listening.

1005b5. P l a i n l y , therefore, it falls to the philosopher, i.e.

the student o f w h a t is characteristic o f a l l substance, also to
investigate the principles o f trains o f reasoning.

1005^8. I t is appropriate for h i m w h o has the best u n d e r ­

standing about e a c h genus to be able to state the firmest
principles o f t h a t a c t u a l subject, a n d hence, w h e n his subject 10
is the things-that-are qm things-that-are, to state the firmest
principles o f everything: a n d this m a n is the philosopher.
A principle about w h i c h it is impossible to be i n error is
firmest of a l l . F o r a p r i n c i p l e o f that k i n d is necessarily the
most inteffigible, since everyone makes mistakes o n matters
about w h i c h he d o e s n o t h a v e u n d e r s t a n d i n g ; a n d is n o n -
hypothetical, since w h a t is necessarily p a r t o f t h e equipment 15
of one w h o apprehends a n y of the things-that-are is not a
hypothesis, a n d w h a t one necessarily understands w h o u n d e r ­
stands a n y t h i n g is necessarily p a r t o f the equipment he comes
w i t h . I t is p l a i n , then, that a p r i n c i p l e o f that k i n d is firmest
ofall.

1005^18. W e h a v e next to state w h a t p r i n c i p l e this is. For

the same thing to hold good and not to hold good simultaneously of the 20
same thing and in the same respect is impossible (given a n y further
specifications w h i c h m i g h t be a d d e d against the dialectical
difficulties).

1005^22. T h i s , then, is the firmest o f a l l principles, for it

fits the specification stated. F o r it is impossible for a n y o n e to
beUeve that the same t h i n g is a n d is not, as some consider 25

7
 1005b METAPHYSICS ГЗ

H e r a c l i t u s said—^for it is not necessary that the things one

says one should also believe. B u t i f it is not possible for
contraries to h o l d good o f the same t h i n g simultaneously
(given that the c u s t o m a r y specifications are a d d e d to this
proposition too), a n d the opinion c o n t r a r y to a n o p i n i o n is
30 that o f the contradictory, t h e n obviously it is impossible for
the same person to believe simultaneously that the same
thing is a n d is n o t ; for anyone w h o m a d e that error w o u l d
be h o l d i n g c o n t r a r y opinions simultaneously. T h a t is w h y
a l l those w h o demonstrate go b a c k to this o p i n i o n i n the e n d :
it is, i n the n a t u r e o f things, the p r i n c i p l e of a l l the other
axioms also.

C H A P T E R 4

35 1005^35. T h e r e are those w h o , as w e said, both themselves

1006* assert that it is possible for the same thing to be a n d not to be,
a n d [assert that it is possible] to believe so. M a n y even of
writers o n n a t u r e m a k e use of this statement. B u t w e h a v e j u s t
accepted that it is impossible to be a n d not be simultaneously,
5 a n d w e h a v e s h o w n by m e a n s of this that it is the firmest o f a l l
principles. S o m e , o w i n g to l a c k of t r a i n i n g , a c t u a l l y ask that
it be d e m o n s t r a t e d : for it is l a c k of t r a i n i n g not to recognize
of w h i c h things demonstration ought to be sought, a n d o f
w h i c h not. F o r i n general it is impossible that there should be
demonstration o f everything, since it w o u l d go o n to i n -
10 finity so that not even so w o u l d it be demonstration. B u t i f
there are some things o f w h i c h demonstration ought not to
be sought, they c o u l d not say w h a t they r e g a r d as a p r i n c i p l e
m o r e fully o f that k i n d .

i o o 6 * i i . B u t even this c a n be demonstrated to be i m ­

possible, i n the m a n n e r of a refutation, i f only the disputant
says something. I f h e says nothing, it is ridiculous to look for a
statement i n response to one w h o has a statement o f n o t h i n g ,
15 i n so far as he has n o t ; s u c h a person, i n so far as he is s u c h ,
is s i m i l a r to a vegetable. B y 'demonstrating i n the m a n n e r o f

8
Г4 TRANSLATION ioo6»

a refutation' I m e a n something different from demonstrating,

because i n demonstrating one might be thought to beg the
original [question], but i f someone else is cause o f s u c h a
thing it must be refutation a n d not demonstration. I n
response to every case o f t h a t k i n d the original [step] is not to
ask h i m to state something either to be or not to be (for that 20
might w e l l be believed to beg w h a t was originally at issue),
but at least to signify something both to h i m s e l f a n d to some­
one else; for that is necessary i f h e is to say anything. F o r i f
he does not, there w o u l d be no statement for s u c h a person,
either i n response to h i m s e l f o r to anyone else. B u t i f h e does
offer this, there w i l l be demonstration, for there w i l l already
be something definite. B u t the cause is not he w h o d e m o n - 25
strates b u t h e w h o submits; for e l i m i n a t i n g statement he
submits to statement. A g a i n , anyone w h o agrees to this has
agreed that somethingis true independently ofdemonstration.

1006*28. F i r s t , t h e n , it is p l a i n that this at least is itself

true, that the n a m e signifies to be or not to be this p a r t i c u l a r 30
thing, so that it c o u l d not be that everything was so-and-so
a n d not s c ^ n d - s o .

1006*31. A g a i n , i f ' m a n ' signifies one thing, let that be

two-footed a n i m a l . W h a t I m e a n by 'signifying one thing' is
this: i f that thing is a m a n , then i f a n y t h i n g is a m a n , that
thing w i l l be to be a m a n . B u t it makes no difference even i f
someone were to assert that it signified more t h a n one thing,
provided t h a t t h e s e w e r e definite; for a different n a m e c o u l d i006^
be assigned to e a c h formula. ( I m e a n , for instance, ifsomeone
were to assert that * m a n ' signified not one but several things,
of one o f w h i c h the f o r m u l a was 'two-footed a n i m a l ' , but
there w a s m o r e t h a n one other as w e l l , but a definite
n u m b e r ; for a distinct n a m e c o u l d be assigned i n respect o f 5
e a c h o f t h e f o r m u l a e . ^ B u t if, instead o f s o assigning, he were
to assert that it signified infinitely m a n y things, it is obvious
that there w o u l d be n o statement. F o r not to signify one
thing is to signify nothing, a n d i f names do not signify,
» Keep the words bracketed in the O C T .

9
 /

ioo6b METAPHYSICS Г4

discussion is e l i m i n a t e d w i t h others; a n d , i n t r u t h , even w i t h

10 oneself, since it is not possible e v e n to conceive i f one is not
c o n c e i v i n g one t h i n g a n d , i f i t is possible, one n a m e c o u l d be
assigned to that a c t u a l thing. L e t the n a m e , then, as was said
originally, signify something a n d signify one thing.

1006b13. T h e n it is not possible that Ч0 be a m a n ' should

signifyjust w h a t 'not to be a m a n ' [signifies], i f ' m a n ' s i g n i -
15 fies not o n l y about one t h i n g but also one t h i n g (for we do not
c o u n t as signifying one t h i n g this, v i z . signifying about one
thing, since i n that w a y 'artistic' a n d 'pale' a n d ' m a n ' w o u l d
signify one thing, so that aU w i l l be one, because s y n o n y m o u s ) .
A n d it w i l l not be to be a n d not to be the s a m e t h i n g unless
20 h o m o n y m o u s l y , as i f others w e r e to t e r m n o t - m a n w h a t w e
t e r m m a n . B u t w h a t is found p e r p l e x i n g is not w h e t h e r it is
possible that the same t h i n g s h o u l d simultaneously be a n d
not be a m a n i n n a m e , b u t i n a c t u a l fact. B u t i f ' m a n ' a n d
' n o t - m a n ' do not signify something different, it is p l a i n that
|^neither] does 'not to be a m a n ' from 'to be a m a n ' , so that to
25 be a maÄ^will be to be a n o t - m a n ; for they w i l l be one thing.
( F o r 'to be one thing' signifies this: being like m a n t l e a n d
cloak, i f t h e f o r m u l a is one.) B u t i f t h e y a r e one thing, 'to be
a m a n ' a n d ' n o t - m a n ' signify one thing. B u t it h a d been
s h o w n that they signify something different.

1006^28. I t is a c c o r d i n g l y n e c e s s a r y , i f i t is true o f a n y t h i n g
30 to say that it is a m a n , that it be a two-footed a n i m a l (for
that w a s w h a t ' m a n ' signified); a n d i f that is necessary, it is
not possible that the s a m e t h i n g should not be, at that time,
a two-footed a n i m a l (for 'to be necessary' signifies this: to be
i n c a p a b l e o f n o t being). C o n s e q u e n t l y it is n o t p o s s i b l e that
it should be simultaneously true to say t h a t the s a m e t h i n g
is a m a n a n d is not a m a n .

1007» 1006b34. T h e s a m e a r g u m e n t applies also i n the case o f

not being a m a n . F o r 'to be a m a n ' a n d 'to be a n o t - m a n '
signify something different, i f e v e n being pale a n d being a
m a n a r e different. F o r the former is m u c h m o r e strongly

10
 TRANSLATION 1007·

opposed, so that it signifies something different. B u t i f [the

disputant] asserts that ' p a l e ' signifies one a n d the s a m e t h i n g 5
too, w e s h a l l repeat j u s t w h a t w a s stated before also, that
everything, a n d not o n l y opposites, w i l l be one. I f that is not
possible, w h a t w e h a v e stated follows, i f he w i l l answer the
question asked. B u t if, asked the question b a l d l y , he appends
the denials also, h e is not a n s w e r i n g the question asked. F o r 10
nothing prevents the same t h i n g being both a m a n a n d pale
a n d a t h o u s a n d other things; nevertheless, i f one is asked
whether it is true to say that this t h i n g is a m a n or not, the
answer ought to signify one thing, not a p p e n d that it is also
pale a n d t a l l . F o r it is c e r t a i n l y impossible to go right t h r o u g h 15
the coincidentals o f a thing, w h i c h a r e infinite; so let h i m go
through either aU o r none. S o equally, even i f t h e same t h i n g
is a t h o u s a n d times a m a n a n d not a m a n , one ought not to
append, to one's a n s w e r to the question w h e t h e r it is a m a n ,
that it is simultaneously not a m a n also; unless one is to
append aU the other things too w h i c h coincide i n it, the
things that it is or is not. B u t i f one does that, there is no 20
discussion.
/
1007*20. T h o s e w h o say this entirely eUminate substance
a n d w h a t it is to be. F o r it is necessary for t h e m to m a i n t a i n
that aU things a r e coincidences a n d that t h e r e i s no s u c h
thing a s j u s t w h a t to be a m a n o r to be a n a n i m a l [is]. F o r i f
anything i s j u s t w h a t t o b e a m a n [is], that w i l l not be to be a
n o t - m a n or not to be a m a n : yet those are its denials. F o r 25
w h a t it signified w a s one thing, a n d that was something's
substance, a n d to signify a thing's substance is to signify
that, for it, to be is n o t h i n g else. B u t if, for it, j u s t w h a t to be a
m a n [is] should be either j u s t w h a t to be a n o t - m a n [is] or
just w h a t not to be a m a n [is], it will be something else; so
that it is necessary for t h e m to say that this k i n d o f f o r m u l a 30
appUes to nothing, a n d that aU things are coincidentaUy.
F o r that is w h a t distinguishes substance a n d the c o i n c i d e n ­
tal: the p a l e coincides i n a m a n because he is pale but not
j u s t w h a t p a l e [is].

II
 ioo7* METAPHYSICS Г4

1007*33. B u t i f everything is s a i d coincidentaUy, there

w i U not be a n y t h i n g w h i c h things are initiaUy about, i f
35 'coincidental' a l w a y s signifies a p r e d i c a t i o n about a c e r t a i n
i007b subject. C o n s e q u e n t l y it wül be necessary to go o n to infinity.
B u t that is impossible, for not even m o r e t h a n two c o m b i n e ;
for the c o i n c i d e n t a l is not c o i n c i d e n t a l i n the c o i n c i d e n t a l ,
unless because both coincide i n the s a m e t h i n g — I m e a n for
instance that the pale m a y be artistic a n d the latter pale
5 because both coincide i n a m a n . B u t Socrates is not artistic
i n that w a y — t h a t both coincide i n some other thing. A c ­
cordingly, since some things are c a l l e d c o i n c i d e n t a l i n the
latter w a y , some i n the former, those so c a l l e d i n the latter, as
the pale is i n Socrates, c a n n o t be a n infinite u p w a r d series,
10 e.g. some other t h i n g c o i n c i d e n t a l i n Socrates the p a l e ; for
not everything makes u p some one thing. N o r i n d e e d w i U
there be a n y other t h i n g c o i n c i d e n t a l i n the pale, as for
instance the artistic; for the latter no m o r e coincides i n the
former t h a n the former i n the latter, a n d at the same time
there is a distinction between things that coincide i n this w a y
a n d things that coincide as the artistic i n Socrates. I n n o n e o f
15 the latter cases does the c o i n c i d e n t a l coincide i n something
coincidental, b u t it does i n aU the former cases; so that not
everything w i U be s a i d coincidentaUy. Consequently, there
w i U be something signifying a substance even i n s u c h a case.
A n d i f that is so, it has been s h o w n that it is impossible to
predicate contradictories simultaneously.

1007^18. A g a i n , i f contradictories a r e a U simultaneously

20 true o f t h e same thing, it is p l a i n that everything w i U be one.
F o r the same thing w i U be both a w a r s h i p a n d a w a l l a n d a
m a n , i f i t is possible either to affirm o r to deny something o f
everything, as is necessary for those w h o state the thesis o f
Protagoras. F o r i f a m a n is thought b y someone not to be a
w a r s h i p , it is p l a i n that h e is not a w a r s h i p ; so that he a k o is,
25 i f the c o n t r a d i c t i o n is r e a l l y true. I n d e e d w e also get the
doctrine o f A n a x a g o r a s , that 'every article is m i x e d together';
so that n o t h i n g is t r u l y one. T h e s e people seem, therefore, to

12
Г4 TRANSLATION 1007b

be stating something indefinite; a n d w h i l e they consider

that they are stating that w h i c h is, their statement is actually
concerning that w h i c h is not (for the indefinite is w h a t is
potentially a n d not i n complete r e a l i t y ) . O n the other h a n d
their statements, at least, must affirm or d e n y everything of
everything; for it w o u l d be a b s u r d i f t h e d e n i a l ofitselfheld
good of e a c h thing, but the d e n i a l o f some other thing, w h i c h
does not h o l d good of it, d i d not h o l d good o f it. I m e a n for
instance that i f i t is true to say o f a m a n that he is not a m a n ,
plainly he is also either a w a r s h i p or not a w a r s h i p . S o i f t h e
affirmation holds good o f h i m , necessarily its d e n i a l does too.
But i f the affirmation does not h o l d good, at least its denial
wiU hold good of h i m more r e a d i l y t h a n his o w n . So i f even
the latter does h o l d good, that o f w a r s h i p w i U too; a n d i f i t
does, its affirmation w i U too.

1008*2. T h i s is one consequence, then, for those w h o state

this thesis; another is that it is not necessary either to assert
or deny. F o r i f it is true that h e is a m a n a n d not a m a n , 5
plainly also he w i U be neither a m a n nor not a m a n ; for the
two h a v e tvy^ denials, a n d , i f both m a k e u p the one former,
there must also be the one latter opposed to it.

1008*7. A g a i n , either this is so i n every case, i.e. a thing

is both pale a n d not pale, both a thing-that-is a n d not a
thing-that-is, a n d i n a similar w a y for aU other assertions a n d 10
denials; or it is so i n some cases but not i n others. I f i t is not
so i n aU cases, these w o u l d be agreed. B u t i f i t is so i n aU, then
i n t u r n either a n y t h i n g asserted m a y also be denied a n d
anything d e n i e d also asserted; or the things asserted m a y
also be denied but not everything denied also asserted. B u t i f 15
thelatter, s o m e t h i n g w o u l d be securelynoi a thing-that-is, a n d
that o p i n i o n w o u l d be firm; a n d i f not to be is something
firm a n d c e r t a i n , the opposite assertion w o u l d be still more
certain. B u t i f a n y t h i n g denied m a y equally be asserted too,
necessarily it is either true to state separately, for instance, that
a thing is pale a n d a g a i n that it is not pale, o r not. I f it is not 20
true to state separately, then n o t o n l y does h e not state these

13
 I

ioo8« METAPHYSICS Г4

things but n o t h i n g w h a t e v e r i s — a n d h o w c a n things-that-

are-not w a l k a n d t a l k ? A l s o everything w o u l d be one, as we
s a i d before [1007^20], a n d a m a n a n d a god a n d a w a r s h i p
25 a n d their contradictories w i U be the same t h i n g ; for i f it
applies e q u a l l y to e a c h thing, n o t h i n g w i U differ from a n y ­
t h i n g else, since, i f it d i d differ, that w o u l d be true a n d dis­
tinctive. E q u a l l y , even i f it is possible to h a v e the t r u t h i n
stating things separately, the result w e h a v e stated follows;
a n d i n a d d i t i o n it follows that everyone w o u l d h a v e the t r u t h
a n d everyone w o u l d be i n error, a n d [the disputant] h i m s e l f
30 is i n error by his o w n admission.

1008*30. A t the s a m e time it is obvious that i n response to

this person there is n o t h i n g for a n investigation to d e a l w i t h ;
for he says nothing. F o r h e says neither that it is so-and-so
nor that it is not so-and-so, but that it both is so-and-so a n d
is not so-and-so; a n d a g a i n h e also denies both these, saying
that it is neither so-and-so n o r not so-and-so. F o r i f h e d i d not,
something w o u l d a l r e a d y be definite.

35 1008*34. A g a i n , i f w h e n e v e r a n assertion is true its d e n i a l

is false a n d w h e n the latter is true its affirmation is false,
there c a n be no s u c h t h i n g as simultaneously asserting a n d
1008^ d e n y i n g the same t h i n g t r u l y . H o w e v e r , they w o u l d d o u b t ­
less assert that this is the question originally posed.

ioo8^2. A g a i n , are w e to say that he w h o believes that

things are i n a c e r t a i n state, or are not, is i n error, w h i l e he
w h o believes both has the t r u t h ? F o r i f h e has the t r u t h ,
w h a t c a n be m e a n t b y saying that the n a t u r e ofthings-that-
5 are is of that k i n d ? I f h e does not h a v e the t r u t h , but has
m o r e t r u t h t h a n the one w h o believes the former w a y , t h e n
the things-that-are w o u l d a l r e a d y be i n some state, a n d that
w o u l d be true a n d not simultaneously also not true. B u t i f
everyone e q u a l l y both is i n error a n d states the t r u t h , there
w U l be n o t h i n g for s u c h a person to speak or s a y ; for he
10 simultaneously says this a n d not this. A n d i f a m a n believes

4
Г4 TRANSLATION ioo8b

nothing, but considers it e q u a l l y so a n d not so, h o w w o u l d

his state be diiFerent from a vegetable's?

i o o 8 ^ i 2 . F r o m w h i c h it is also quite obvious that nobody

actually is i n that condition, neither those w h o state this
thesis n o r a n y b o d y else. F o r w h y does a n y o n e w a l k to
M e g a r a r a t h e r t h a n stay w h e r e he i s , w h e n he considers that 15
he should w a l k there? W h y does h e not proceed one m o r n i n g
straight into a w e l l or over a precipice, i f t h e r e is one about:
instead o f e v i d e n t l y t a k i n g c a r e to a v o i d doing so, as one w h o
does not consider that faUing i n is e q u a l l y a good thing a n d
not a good t h i n g ? I t is consequently p l a i n that h e believes
that one t h i n g is better, another not better. A n d ifso, he must
a k o believe that one t h i n g is a m a n a n d another not a m a n ,
one thing sweet a n d another not sweet. F o r h e neither seeks 20
nor believes everything indifferently w h e n , considering that
it is better to d r i n k w a t e r a n d see a m a n , he thereupon
s e e b to do s o ; a n d yet h e ought to, i f the same t h i n g were
equally a m a n a n d not a m a n . B u t j u s t as w e said, there is
nobody w h o does not evidently take c a r e to avoid some 25
things a n d j i o t others; so that it seems that everyone holds
some beliefs b a l d l y , i f n o t about everything t h e n about w h a t
is better a n d worse. A n d i f t h i s is not knowledge but opinion,
one w o u l d h a v e to be aU the m o r e anxious about the truth,
as a sick m a n is m o r e anxious about his h e a l t h t h a n one w h o
is healthy. F o r i n d e e d a m a n w h o holds a n o p i n i o n is i n a n 30
unhealthy condition w i t h r e g a r d to the t r u t h , c o m p a r e d w i t h
one w h o has knowledge.

1008^31. A g a i n , h o w e v e r m u c h everything is so-and-so

and not so-and-so, at least the m o r e a n d the less are present
i n the n a t u r e ofthings-that-are. F o r w e w o u l d not assert that
two a n d three are equally even, or that one w h o considered 35
that four things were five a n d one w h o considered that they
were a t h o u s a n d w e r e equally i n error. S o i f they a r e not
equally, it is p l a i n that one o f t h e m is less, so that he has more
truth. S o i f w h a t is m o r e is nearer, there m u s t be something 1009»
true w h i c h the m o r e true v i e w is nearer. A n d even i f that is

15
 I

ioo9* METAPHYSICS Г4

not SO, at least there is abready something m o r e firm a n d

m o r e truthHke, a n d w e s h o u l d be rid o f the u n a d u l t e r a t e d
5 thesis w h i c h w o u l d prevent us from h a v i n g a n y t h i n g defi­
nite i n o u r t h i n k i n g .

C H A P T E R 5

1009*6. F r o m the s a m e o p i n i o n also derives the thesis o f

Protagoras, a n d it is necessary that both either are or are not
the case equally. F o r i f e v e r y t h i n g that is thought or i m a g i n e d
is true, it is necessary that everything s h o u l d be s i m u l -
10 taneously true a n d false; for m a n y people h a v e m u t u a l l y
c o n t r a r y beliefs, a n d r e g a r d those whose opinions a r e not
the same as their o w n as i n error, so that it is necessary that
the same thing s h o u l d b o t h be a n d not be. A n d i f the latter,
it is necessary that w h a t is thought to be s h o u l d a l l be t r u e ;
for those w h o are i n error a n d those w h o h a v e the t r u t h h o l d
m u t u a l l y opposite opinions, a n d so, i f that is the state o f
15 things-that-are, aU w i U h a v e the t r u t h . I t is p l a i n , then, that
both theses derive from the s a m e t h i n k i n g .

1009*16. B u t the w a y to confront their proponents is not

a l w a y s the s a m e : some r e q u i r e to be c o n v i n c e d , others to be
defeated. I f they h a v e this beUef as a result o f p e r p l e x i t y
their mistake is easy to r e m e d y , for the conflict is not w i t h
20 their statement b u t their t h i n k i n g . B u t i f they state it for
the sake o f stating it, the r e m e d y is to refute the statement
w h i c h is i n their speech a n d i n their words.

1009*22. T h o s e w h o find themselves i n perplexity derive

this o p i n i o n from perceptible things. O n the one h a n d , [they
derive the o p i n i o n ] that contradictories a n d contraries h o l d
good simultaneously from seeing contraries c o m i n g to be
25 out o f the s a m e t h i n g . If, therefore, it is not possible that
w h a t is not s h o u l d c o m e to be, that a c t u a l thing has aU along
been both equally. T h i s is asserted b y A n a x a g o r a s ('every­
thing is m i x e d i n everything'), a n d b y D e m o c r i t u s , w h o says:

16
Г5 TRANSLATION 1009·

the v o i d a n d the full a r e found e q u a l l y i n every p a r t w h a t ­

ever, a n d yet the one o f these is that w h i c h is, the other that
w h i c h is not. I n response to those, then, whose beliefderives 30
from these sources, w e shall say that i n one sense w h a t they
say is correct, but i n another sense they a r e mistaken. F o r
that w h i c h is m a y be so c a l l e d i n two ways, so that there is one
sense i n w h i c h it is possible for something to c o m e to be out of
what is not (though i n another sense this is not possible), a n d
for the same thing to be simultaneously b o t h a thing-that-is
and a thing-that-is-not (only not i n the s a m e respect)—^for 35
it is possible for the same thing simultaneously to be contrary
things potentially, though not i n complete reaUty. A g a i n , we
shall require these people to believe that a m o n g the things-
that-are there is also another k i n d o f substance o f w h i c h
neither change nor destruction n o r c o m i n g to be h o l d good
at aU.

1009*38. S o , too, from perceptible things some derive the i009^

truth o f w h a t is i m a g i n e d . F o r they consider that it is i n ­
appropriate to j u d g e t r u t h b y large or s m a l l n u m b e r s , but
the same tbihg is thought sweet b y some w h o taste it a n d
bitter b y others; so that i f e v e r y o n e w e r e iU or everyone were 5
out o f h i s m i n d a n d o n l y two o r three people were healthy or
sane, the latter w o u l d be thought i U a n d out o f their m i n d s
a n d not the others. A g a i n , [they say] that the s a m e t h i n g s a r e
imagined i n c o n t r a r y w a y s b y m a n y o f t h e other a n i m a l s a n d
by us, a n d even as perceived b y e a c h person they are not
always thought the same. W h i c h kinds o f these, therefore,
are true or false is u n c l e a r ; for these ones a r e no more true 10
t h a n those, b u t equally. T h a t is w h y at least D e m o c r i t u s
asserts that either there is n o t r u t h or at least to us it is
unclear.

1009b12. I n general it is because they believe that p e r ­

ception is w i s d o m , a n d the former is modification, that they
assert that w h a t is i m a g i n e d i n perception is o f necessity
true. F o r it is for these reasons that b o t h E m p e d o c l e s a n d 15
D e m o c r i t u s a n d v i r t u a l l y everyone else h a v e s u c c u m b e d to

17
 1009b METAPHYSICS Г5

opinions o f t h i s k i n d ; for E m p e d o c l e s actually asserts that

a person's w i s d o m alters as h e alters his state:
Men's cunning grows with what is present to them.

A n d elsewhere h e says:
20 A n d as they modified, so did their thoughts
Gome 2dways modified to mind.

A n d P a r m e n i d e s pronounces i n the same sense:

F o r as each moment stand his limbs composed
So is man's thought arrayed; for what has mind
Is nothing but the nature of the limbs
25 O f all of us: we think what we have more of

I t is also r e c o r d e d as a d i c t u m o f A n a x a g o r a s i n response to
some firiends t h a t 4 h i n g s w i U r e a l l y be for t h e m h o w e v e r
they believe'. People even assert that H o m e r evidently h e l d
this opinion, because h e m a d e H e c t o r , w h e n he w a s u n -
30 conscious from a blow, lie ' w i t h other things i n m i n d ' , as i f
even those w h o a r e out o f their m i n d s h a v e something i n
m i n d , t h o u g h not the same things. S o it is p l a i n that i f both
states a r e states o f w i s d o m , things w i U r e a l l y be so-and-so a n d
not so-and-so simultaneously.

1009b33. A n d it is at this point that the most unfortunate

consequence arises; for i f t h o s e w h o h a v e observed the a v a i l -
35 able truths most c l o s e l y — a n d they a r e those w h o h a v e
sought hardest for t h e m a n d c a r e d most about t h e m — i f t h e y
h o l d opinions o f this k i n d a n d m a k e these pronouncements
about t r u t h , WÜ1 not those e n d e a v o u r i n g to philosophize
legitimately lose h e a r t ? T h e quest for t r u t h w o u l d be a w i l d -
goose chase.

1010» i o i o " i . W h a t caused these people to h o l d their o p i n i o n

w a s that, i n s e a r c h i n g for the t r u t h about the things-that-are,
they believed t h a t the things-that-are are m e r e l y p e r c e p t i -
bles; a n d i n these the n a t u r e ofindefiniteness, i.e. o f being i n
the w a y w e h a v e described, is a n i m p o r t a n t constituent. T h i s
5 explains w h y their statements, though plausible, are not true

18
Г5 TRANSLATION ioio·

(an apter w a y to put it t h a n E p i c h a r m u s used about X e n o -

phanes).
i o i o * 7 . A g a i n , observing that a U this [ w o r l d of] n a t u r e is
i n change, a n d that no t r u t h is h a d about t h a t w h i c h is i n
[process of] alteration, it w a s not possible to h a v e the t r u t h at
least c o n c e r n i n g that w h i c h is i n [process of] every style a n d
m a n n e r o f alteration. T h i s belief blossomed into the most 10
extreme o f a l l the opinions w e h a v e mentioned, that ofthose
w h o p r o c l a i m e d themselves H e r a c U t e a n s ; s u c h as G r a t y l u s ,
who i n the e n d considered that he ought to say nothing, a n d
merely m o v e d his finger. H e also c r i t i c i z e d H e r a c U t u s for
saying that it is not possible to step into the same river twice:
for h e h i m s e l f considered that it is not possible even once. 15

i o i o * i 5 . I n response to this a r g u m e n t too we shall say that

while they do h a v e some a r g u m e n t for considering that at the
time ofits altering a t h i n g i n [process o f ] alteration is noty yet
that is after aU disputable. F o r some o f w h a t is being dis-
c a r d e d i s possessed b y the t h i n g d i s c a r d i n g it, a n d some o f
w h a t i s c o m i n g to be necessarily a l r e a d y is. A n d i n general
i f a t h i n g is i n [process o f ] destruction there w i U be something 20
that is, a n d i f a t h i n g is i n [process o f ] c o m i n g to be there is
necessarily something out o f w h i c h it is c o m i n g to be a n d by
the agency o f w h i c h it is being generated, a n d this does not
go o n to infinity. B u t l e a v i n g that aside, w e m a y say this,
that it is not the s a m e t h i n g to alter i n quantity a n d i n
qualification; allowing, then, that a thing's q u a n t i t y is not
constant, still w e a r e a c q u a i n t e d w i t h everything b y its 25
form.

1010*25. A g a i n , those w h o h o l d this beUef c o u l d legiti­

mately be c r i t i c i z e d for p r o n o u n c i n g equaUy true o f the
whole universe w h a t they observe o n l y i n a n u n o r i t y even o f
perceptible things. F o r o n l y the region o f the perceptible
a r o u n d us is p e r m a n e n t l y i n [process of] destruction a n d
c o m i n g to b e ; a n d that is v i r t u a U y n o p o r t i o n o f the whole, 30
so that it w o u l d b e j u s t e r to a c q u i t that portion because o f t h e
other t h a n to c o n d e m n the other because o f i t .

19
 ioio« METAPHYSICS Г5

1010*32. A g a i n , it is p l a i n that w e shall repeat i n response

to these people w h a t w a s said previously [ 1 0 0 9 * 3 6 - 8 ] : for
they must be s h o w n , a n d they m u s t be c o n v i n c e d , that there
35 is a c e r t a i n n a t u r e w h i c h is changeless. A n d yet the conse­
quence o f m a i n t a i n i n g simultaneously that things are a n d
are not is r e a l l y to assert that aU are at rest, r a t h e r t h a n
c h a n g i n g ; for there is n o t h i n g for things to alter into, for
1010^ everything holds good o f everything.

l O i o ^ i . A s for t r u t h , to show that not everything that is

i m a g i n e d is true: first, even i f p e r c e p t i o n , at least o f w h a t is
special, is not false, still i m a g i n a t i o n is not the same t h i n g as
perception. N e x t , one m a y legitimately be surprised that they
5 should find perplexing the question w h e t h e r magnitudes a n d
colours are s u c h as they a r e i m a g i n e d b y those w h o are at a
distance or those w h o a r e n e a r , a n d b y the healthy or the
sick; or whether w h a t is i m a g i n e d by the w e a k or the strong
is h e a v i e r ; or w h e t h e r w h a t is i m a g i n e d by the sleeping or
the w a k i n g is true. F o r it is obvious that they do not really
10 consider it so: at a n y rate ifsomeone i n L i b y a believes h i m s e l f
one night i n A t h e n s , h e does not set offfor the O d e o n . A g a i n ,
as for the future, as P l a t o also says, the opinions o f a doctor
a n d a n ignorant m a n a r e surely not equaUy authoritative,
as for instance o n the question w h e t h e r someone is or is not
going to be healthy.

15 1010b14. A g a i n , i n the case o f o u r perceptions themselves

the perception o f w h a t is alien a n d special, or o f w h a t is
n e i g h b o u r i n g a n d w h a t is its o w n , are not equaUy a u t h o r i ­
tative, b u t i n the case o f c o l o u r it is sight, not taste, a n d i n the
case o f flavour taste, not sight; a n d e a c h o f tiiese never
asserts about the s a m e t h i n g i n the same time that it is
simultaneously so-and-so a n d not so-and-so.

20 1010^19. N o r , e v e n i n another time, was there dispute

about the affection, b u t o n l y about that i n w h i c h the affection
coincides; I m e a n for instance that the same w i n e might be
thought sweet at one time a n d not sweet at another, i f there

20
Г5 TRANSLATION хоюЬ

is a n alteration either i n it or i n the b o d y ; but the sweet s u c h

as it is, w h e n it is, has never yet altered, a n d one a l w a y s has
the t r u t h about it, a n d a n y t h i n g that is going to be sweet is 25
such o f necessity. Y e t this is e l i m i n a t e d b y aU these t h e s e s —
just as nothing has a substance, so too nothing is ofnecessity.
F o r it is not possible that w h a t is necessary should be thus a n d
otherwise, so t h a t i f a n y t h i n g is o f necessity it w i U not be
both thus a n d not thus. 30

1010^30. A n d i n general, i f i n fact only the perceptible

exists, nothing w o u l d exist unless l i v i n g things existed; for
there w o u l d be no perception. N o w it is doubtless true that
neither perceptible things n o r sense-impressions ( w h i c h are
a n affection o f a perceiver) w o u l d exist; but that the s u b ­
jects w h i c h p r o d u c e perception w o u l d not exist, even i n the
absence o f perception, is impossible. F o r perception is not of 35
itself, but there is some other thing too a p a r t from perception,
w h i c h is necessarily p r i o r to perception; for w h a t changes
somethii;^g is prior i n n a t u r e to the thing changed, a n d this is 1011·
so no less even i f t h e y a r e c a l l e d these things w i t h reference to
one another.

C H A P T E R 6

1011*3. S o m e , both a m o n g those w h o are c o n v i n c e d by aU

this a n d a m o n g those w h o merely state these things, are
perplexed because they w a n t to k n o w w h o wiU j u d g e w h o is 5
healthy, a n d i n general o n e a c h subject [who w i l l j u d g e ] w h o
WÜ1 j u d g e it correctly. S u c h perplexities are similar to the
perplexing question ' A r e we n o w asleep or a w a k e ? ' a n d they
aU h a v e the same force. F o r those w h o pose t h e m ask for a n
argument for e v e r y t h i n g ; for they seek a p r i n c i p l e , a n d they
seek to get it t h r o u g h d e m o n s t r a t i o n — s i n c e it is obvious i n Ю
their actions that they are not a c t u a l l y c o n v i n c e d . T h e i r
trouble is j u s t as w e h a v e stated: for they seek a n a r g u m e n t
for something for w h i c h there is no argument, for a p r i n c i p l e

21
 lOIl· METAPHYSICS Г6

o f d e m o n s t r a t i o n is nöt a demonstration. T h e s e people m i g h t

easily be c o n v i n c e d ofthis point, w h i c h is not a h a r d t h i n g to
15 accept. B u t those w h o seek o n l y to be defeated i n the a r g u ­
m e n t seek the impossible; for they ask that contraries be
stated, forthwith stating contraries.

1011*17. I f n o t everything is relative to something, i.e.

some things a r e also themselves i n their o w n right, not
everything i m a g i n e d c a n be true. F o r w h a t is i m a g i n e d is
i m a g i n e d by someone, so that a n y o n e w h o says that every-
20 thing i m a g i n e d is true makes every thing-that-is relative to
something. H e n c e those w h o seek to be defeated i n the
a r g u m e n t a n d at the same time ask to subject themselves to
a r g u m e n t h a v e to be careful to say that it is not w h a t is
i m a g i n e d that is b u t w h a t is i m a g i n e d by h i m w h o imagines
it a n d at the time a n d i n the style a n d m a n n e r that it is
i m a g i n e d . I f they subject themselves to a r g u m e n t but not o n
25 these terms, they w i U soon find themselves m a k i n g c o n t r a r y
statements. F o r it is possible for the same t h i n g to be i m a g i n e d
honey to the sight but not b y the taste; a n d , since w e h a v e
two eyes, not to be i m a g i n e d the s a m e b y the sight o f e a c h ,
i f their sight is dissimilar. F o r i n response to the reasons
mentioned previously [5. 1009*38-^12] for m a i n t a i n i n g that
30 w h a t is i m a g i n e d is true, a n d that o n this a c c o u n t everything
is equaUy false a n d true—^for the s a m e things are not a l w a y s
i m a g i n e d the same b y everyone, or e v e n b y the same m a n ,
but a r e often i m a g i n e d c o n t r a r y i n respect o f the same time,
as t o u c h says there are two things w h e n the fingers are crossed
b u t sight o n e — [ t h e r e p l y is that they are i m a g i n e d c o n t r a r y ]
35 not b y the same perception a n d i n the same aspect o f i t a n d
1011^ i n the same w a y a n d i n the same time, so that this w o u l d be
true. B u t doubtless it is o n this a c c o u n t necessary that anyone
w h o makes the statement for the sake o f m a k i n g it a n d not o n
a c c o u n t o f perplexity should state that this is, not true, b u t
true to this person.

1011 b4. I n d e e d , as was said before [ i o i 1 *20], it is necessary

5 to m a k e everything relative to something, i.e. to o p i n i o n a n d

22
Г6 TRANSLATION юцЬ

perception, so that n o t h i n g either has come to be or w i U be

without someone first h a v i n g that o p i n i o n ; a n d i f things
have come to be or w i U be, it is p l a i n that not everything c a n
be relative to opinion. A g a i n , i f a thing is one,' it is relative
to one thing or to something definite; a n d i f the same thing
is both a h a l f a n d e q u a l , still the e q u a l is not relative to the
double. So if, relative to that w h i c h has a n opinion, the same 10
thing is a m a n a n d the object o f opinion, the object of o p i n ­
ion, not that w h i c h has it, w i U be a m a n . A n d i f e a c h t h i n g
is to be relative to that w h i c h has a n o p i n i o n , that w h i c h has
a n opinion w i U be relative to things infinite i n form.

1011^13. I t has n o w been fully enough stated that the

opinion that opposite assertions are not simultaneously true
is the firmest of a U , a n d w h a t are the consequences for those
who m a k e this statement, a n d w h y they m a k e it. S i n c e it is 15
impossible for a contradiction to be true simultaneously o f
the same thing, it is obviously impossible too that contraries
should $imultaneously h o l d good of the same thing. F o r one
of a paV of contraries is, i n addition, a l a c k — a l a c k of
substance; a n d a l a c k is a d e n i a l about some definite genus. 2 0
So i f it is impossible simultaneously to affirm a n d deny
truly, it is impossible also for contraries to hold good s i m u l ­
taneously, unless either both h o l d good i n a certain w a y , or
one i n a c e r t a i n w a y a n d the other baldly.

C H A P T E R 7

1011^23. Nor, on the other hand, is it possible that there should be

anything in the middle of a contradiction, but it is necessary either to
assert or to deny any one thing of one thing.

1011^25. T h i s w i U be p l a i n i f w e first define w h a t t r u t h 2

a n d falsehood a r e : for to say that that w h i c h is is not or that
w h i c h is not is, is a falsehood; a n d to say that that w h i c h is is
a n d that w h i c h is not is not, is t r u e ; so that, also, h e w h o says
I Keep the words obelized in the O C T .

23
 юцЬ METAPHYSICS П

that a thing is or not will have the truth or be in error. But it

is said that neither that which is nor that which is not either
is not or is.

30 1011^29. Again, [something] wiU be in the middle ofthe

contradiction either i n the way in which what is grey is in the
middle between dark and pale, or in the way in which
what is neither ofthe two is in the middle between a man and
a horse. If, then, i n the latter way, it could not alter, for a
thing alters from not-good into good or from that into not-
good; but in fact it evidently always does, for there is no
35 alteration except into opposites and things in the middle.
1012» But i f it is in the middle, that way too there would be some
sort of [process of] coming to be pale which was not from
not-pale; and that is not in fact observed.

i o i 2 * 2 . Again, thinking either affirms or denies everything

thought ofand conceived—this is plain from the definition—
whenever it has the truth or is in error: whenever it com­
pounds things this way—in an assertion or denial—it has the
5 truth, whenever this way, it is in error.

1012*5. Again, there must be something apart from every

contradiction, ifthe thesis is not stated for the sake ofstating
i t ; so that someone wiU neither have the truth nor not have
the truth and there wiU be something apart from that which
is and that which is not, so that there wiU be a kind of
alteration apart from [the processes of] coming to be and
destruction.

i o i 2 * g . Again, there wiU be [a middle] even in those

10 genera in which the denial implies the contrary, as for i n ­
stance among numbers a number which is neither odd nor
not odd; but that is impossible, as is plain from the definition.

i o i 2 * i 2 . Again, the process leads to infinity, and the

things-that-are wiU not be merely half as many again but
even more. For it wiU be possible in turn to deny this in
24
Г7 TRANSLATION ioia·

relation to the assertion and denial, and this [denial] wiU be

something; for the substance is something else than it. 15
1012*15. Again, when someone, asked whether something
is pale, says that it is not, he has denied nothing else than its
being; and its not being is a denial.
1012*17. Some people have derived this opinion as they
have others of the paradoxes: unable to resolve captious
arguments they give i n to the argument and endorse the
truth ofits conclusion. While some state the thesis for reasons 20
ofthat kind, others do so because ofseeking an argument for
everything. I n response to aU these people the original [step]
is from a definition. Definition arises from the necessity that
they should themselves signify something, for the formula of
[the thing of] which the name is a sign wiU be a definition.
1012*24. I t seems that while Heraclitus' thesis, which says 25
that everything is and is not, makes everything true, that of
Anaxagoras, that there is something in the middle of a con-
tradiction,Viakes everything false; for when things are mixed
the mixture is neither good nor not good, so that there is
nothing true to be said.

CHAPTER 8
1012*29. I n view of these distinctions it is obvious that the
theses which some people state singly and about everything, 30
whether maintaining that nothing is true (for they assert that
there is nothing to prevent everything being like the com­
mensurability of a diagonal) or that everything is true,
cannot hold good. These theses are practically the same as
that of Heraclitus, for anyone who states that everything is 35
true and everything false also states each of these theses ioi2^
separately, so that i f the former are impossible it is i m ­
possible also that the latter should be. Again, there are
obviously contradictions which cannot be simultaneously
true—or aU false either, though that might indeed be
thought the likelier possibility i n view ofwhat has been said.
25
 ioiab METAPHYSICS Г8

5 I n response to aU such theses the right thing, as was also

stated in the arguments above [4. i o o 6 * i 8 ] , is to beg, not
that something is or is not, but that something signifies; so
that we must base discussion on a definition, having taken
for granted what Talsehood' and 4 r u e ' signify. I f what is
true to assert is nothing else than what is a falsehood to deny,'
10 it is impossible for everything to be false; for i t is necessary
that one portion of the contradiction should be true. Again,
i f i t is necessary that everything be either asserted or denied,
it is impossible that both of the two should be false; for one
portion of a contradiction is a falsehood. Indeed aU such
15 theses end up notoriously by eliminating themselves; for
anyone who says that everything is true also makes the
statement contrary to his own true, so that his own is not
true (for the contrary statement asserts that his is not true),
and anyone who says that everything is false says i t of him­
self. But ifeach makes an exception ofthese cases—the former
that the contrary [of his statement] is alone not true, the
20 latter that his own is not false—they wiU end up none the less
begging an infinite number of statements, true and false;
for the statement which states that the true statement is true
is true, and this wiU go on to infinity.

1012^22. I t i s obvious that the statements made by some

people that everything is at rest, and by others that every­
thing is changing, are not true either. For i f everything is at
25 rest, the same things wiU always be true and false, but
evidently this alters; for the speaker did not exist himself
once and wiU not again. And i f everything is changing,
nothing wiU be true, and consequently everything false: but
that has been shown to be impossible. Again, i t is necessary
that that which is alters, since alteration is from something
into something. On the other hand it is not the case that
30 everything is at rest or changing sometimeSy and nothing
always; for there is something which is always changing the
things that change, and the first changer is itself changeless.
» R e a d τό άληθ^ς φάναι η άποφάναι ^cv8oy.

26
 METAPHYSICS BOOK DELTA

CHAPTER 1

1012^34· We call an O R i G i N that point ofan actual thing from

which someone would move first, as for instance of a length 35
or a road there is this origin here, another from the contrary 1013»
direction;

1013^1. and the point from which each thing would come
to be most satisfactorily, as for instance in learning one
should sometimes begin not from what is first, i.e. the origin
of the actual thing, but from the point from which it is
easiest to learn;

1013*4. and that constituent from which a thing first comes

to be, as ipr instance the keel of a ship and the foundation of 5
a house, and in the case of animals some believe it is the
heart, some the brain, others whatever it may be ofthat kind;

1013*7. and that non-constituent from which a thing first

comes to beand from which change and alteration [in it]
characteristically first begin, as for instance a child comes
to be out of its father and mother, and fighting out of 10
swearing;

1013*10. and that at whose decision things changed are

changed and things altered altered, just as the origins
[authorities] in cities, and dynasties and kingships and
tyrannies, are called origins, and as skills are, especially
master-skills.

1013*14. Again, the point from which one first gets 15

acquainted with an actual thing is also called the thing's
origin [principle], as for instance hypotheses of demon­
strations.
27
 ioi3" METAPHYSICS

1013*16. Causes are also so called i n the same number of

ways; for every cause is an origin.
1013*17. I t is therefore common to every origin to be the
first point from which a thing is, or comes to be, or from
which one gets acquainted with i t ; and some of them are
20 constituents, others external. Hence an origin may be a
nature, an element, a thought, a decision, a substance, or
what a thing is for; for with many things the origin both of
acquaintance and of change is what is good and what is
fine.

CHAPTER 2

1013*24 {Physics I I 3. 194b23). We call a C A U S E , in one sense,

25 that constituent out of which something comes to be, as for
instance bronze of a statue and silver of a bowl, and their
genera:

1013*26 {Physics I I 3. 194b26). in another, the form and

pattern, that is, the formula of what it is to be, and its
genera, as for instance two-to-one and in general number of
the octave; also the parts in a formula:
1013*29 {Physics I I 3. 194^29). again, that from which the
30 first origin ofalteration or rest [proceeds], as for instance one
who planned something is its cause and as a father is of his
childand in general what produces ofwhat is produced and
the alterer ofwhat alters:
1013*32 {Physics I I 3. 194b32). again, as a [thing's]
fulfilment, that is, what it is for, as for instance health of
someone's taking walks (why does he take walks? 'in order
35 to be healthy', we assert, and hold that in so saying we have
displayed the cause):

1013^35· {Physics I I 3. 194^35) also, anything that comes

to be as a means to a fulfilment when what effected the change
28
Δ2 TRANSLATION 1013b

was something else, as for instance of health slimming or

purging or medicines or instruments; for aU these are for
the sake ofthe fulfilment, but differ from one another in that
some are instruments, others operations.

1013^3 {Physics I I 3.195^3). These then are pretty well

aU the ways in which causes are so called. From their
being so called in several ways it follows both that the same 5
things may have several causes, not coincidentaUy, as for
instance a statue has both the art of statuemaking and
bronze, not by virtue of some other thing but qua statue, yet
not i n the same sense but the one as matter and the other as
that from which the change [proceeds]: and that things may
be causes of one another, as for instance exercise of fitness 10
and the latter ofexercise, yet not in the same sense but one as
fulfilment and the other as origin ofchange. Again, the same
thing may be [cause] of contraries, for what when present is
cause ofso-and-so we sometimes hold responsiblewhen absent
for the contrary; for a shipwreck, for instance, the absence
of the helmsman whose presence was cause of preservation.
Both things, presence and lack, are causes, as effecting the 15
change.

1013b16 {Physics ΪΙ 3. 195^15). AU the causes here men­

tioned fall under four senses which are the most obvious. For
elements are causes of syUables, and matter of artefacts, and
fire and earth and aU such things of bodies, and parts of 20
wholes, and hypotheses of a conclusion, as being that out of
which [the latter proceed]; and of them some, e.g. parts,
[are cause] as subject, others—the whole and the composition
and the form—as what it isto be. And a seed, a doctor, one who
planned, and in general what produces are aU that from
A v h i c h the origin of alteration or ofkeeping-the-same [proceeds]. 25
And some things [are causes] as fulfilments of the others, i.e.
as their good; for what other things are for is wont to be best,
and to be their fulfilment (let us assume that there is no
difference in calling it good or imagined good).
29
 1013b METAPHYSICS Δ2

1013^28 {Physics I I 3. 195*26). These, then, are the causes

and this is the number of their forms. Their senses, though
30 numerous, also come under quite few headings. For causes
are so called in several ways, even those of the same form i n
ways prior and posterior to one another; as for instance of
health the doctor and the man-of-skill, of the octave the
double and number, and in every case what includes any of
the particular [causes].

35 1013^34 {Physics I I 3. 195^32). Again, as the coinciden­

tal and the genera of these: as for instance of a statue, i n
one way Polyclitus and i n another a statuemaker, because
1014* being Polyclitus coincides in a statuemaker; and what i n ­
cludes the coincidental, as for instance a man or also i n
general an animal is cause of a statue because Polyclitus is
a man and a man is an animal. Among coincidentals also
5 some are more remote and some nearer than others, as for
instance i f someone pale and someone artistic were called
cause o f a statue, 'and notjust Polyclitus or a man.'

1014*7 {Physics I I 3. 195^3). And apart from^ aU those so

called both properly and coincidentaUy, some are so called
as being capable, others as actually functioning; ofthe build­
ing of a house, for instance, the housebuilder or the house-
buUder housebuUding.

10 1014*10 {Physics I I 3. 195^6). What causes are causes of

wiU also be so called i n the ways mentioned, as for instance of
this statue or a statue or i n general a likeness, and of this
bronze or bronze or i n general matter; and i n the same way
in the case of coincidentals.

1014*13 {Physics I I 3. 195b10). Again, both the former and

the latter wiU be so called in combination, as for instance not
15 Polyclitus or statuemaker but Polyclitus statuemaker.
These words are omitted in the M S S . of the Physics and bracketed in the
O C T of the Metaphysics,
* Read παρά.

30
Δ 2 TRANSLATION 1014·

1014*15 {Physics I I 3. 195^12). However, aU of these

amount to six, so caUed in two ways: either as the particular
or as the genus or as the coincidental or as the genus of the
coincidental or these as combined or as stated baldly; and aU
of them either as actually functioning or in respect of cap- 20
acity. They differ to this extent, that what is actually func­
tioning and what is particular is and is not simultaneously
with the things i t causes; as for instance this [man] doctoring
with this [man] being healed and this housebuilder house-
building^ with this thing being built into a house. With
causes i n respect ofcapacity this is not always the case; for
the house and the housebuilder are not destroyed simulta- 25
neously.

CHAPTER 3

1014*26. The primary constituent ofany compound, when

indivisible in form into another form, is caUed an E L E M E N T ;
as for instance the elements of a spoken sound are the things
out of which the spoken sound is composed and into which
it divides ultimately, but which do not themselves divide
further into spoken sounds that are other in form; ifthey do 30
divide, their portions are of the same form, as a portion of
water is water (which is not true of a syllable). EquaUy,
those who speak of the elements of bodies mean the things
into which bodies ultimately divide and which do not divide
further into other things differing in form. Whether there is
one or more than one such thing, that is what they call an 35
element.
1014*35. The elements ofdiagrams, and ofdemonstrations
in general, are so caUed in much the same way. For the
demonstrations that are primary and constituents of more 1014*»
than one other demonstration are caUed elements ofdemon­
strations; of this kind are the primary syllogisms, out of
three [terms] through one middle.
' Read o8c ό οίκοΒόμος οΙκοΒομών.

31
 1014b METAPHYSICS

1014^3. By transference from this, people term an element

anything which, being one and small, is useful for many
5 purposes; that is why what is small and simple and indi­
visible is also caUed an element. From this i t comes that the
most universal things are elements, because each of them,
being one and simple, is a constituent of many things, or of
everything, or nearly everything; also that the one and the
point are thought by some to be origins. Therefore, since the
10 things termed genera are universal and indivisible (for they
have no formula), some people caU genera elements; and
more so than the differentia because the genus is more uni­
versal—^for that of which the differentia holds good implies
the genus, but not everything of which the genus holds good
implies the differentia.
1014^14. I t is common to aU cases that the first constituent
15 ofeach thing is an element of each.

CHAPTER 4

1014^16. We call ΝΑτυκΈ, i n one sense, the coming to be of

things that grow, as i f one were to pronounce the u in ^pkuns'
long;
1014^17. i n another, the first constituent out ofwhich a
growing thing grows;
1014^18. again, what makes the primary change i n any
naturally existing thing a constituent of the thing qua itself
20 AU things are said to grow which gain enlargement through
another thing by contact and assimilation or (as with em­
bryos) adhesion. Assimilation differs from contact, for in the
latter case there is no necessity for any other thing apart
from contact, while in the case of things assimilated there is
some one thing, the same i n both, which makes them assimi-
25 lated instead o f i n contact, and makes them one in respect of
continuity and quantity, though not in respect of qualifi­
cation.
32
 TRANSLATION w14b

1014b26. Again, that out of which a naturally existing

thing first is or comes to be, and which is unstructured and
not subject to loss of its own capacity, is caUed a nature, as
for instance bronze is caUed the nature of a statue and of
bronze artefacts, wood of wooden ones, and equaUy i n the 30
other cases; for each is made out ofthese, the first matter
being conserved. I t is i n this sense that the elements of
naturally existing things are also asserted to be their nature
(some mentioning fire, some earth, some air, some water,
others something else of that kind, and others some or aU of 35
these).
1014b35. I n another sense again, the substance ofnaturally
existing things is caUed their nature, as for instance those
who say that [a thing's] primary composition is its nature, or
as Empedocles says: l0l5·

None has a nature of the things that are;

There is but mixture and exchange of things
Mixed^but nature's name men fix on them.
This explains why, with things that exist or come to be
naturally, although that out of which it is their nature to be
or come to be is already present, we still do not assert that
they possess their nature i f they do not possess their form and 5
shape. What is made up ofboth ofthese, then, exists NATUR¬
A L L Y , as for instance animals and their portions. The nature
is both the first matter (and this in two ways, either first
relative to the thing itself or first in general, as for instance
with works of bronze the bronze is first relative to themselves
but in general it is perhaps water, i f a l l meltables are water), 10
and the form and substance, that is, the fulfilment of their
coming to be.
1015^11. By transference from this case, aU substance i n
general has come to be caUed nature, because nature is also
a kind of substance.
1015*13. O f a l l those mentioned, then, the nature which
is primary and fundamentally so caUed is the substance of
33
 xoi5* METAPHYSICS J 4

15 those things that possess an origin of change i n themselves

qua themselves. For matter is caUed nature from being
recipient of this, and comings to be and growth from being
changes arising from it. And the origin ofchange ofnaturally
existing things is this, being a constituent in a way (either
potentially or i n complete reaUty).

CHAPTER 5

20 1015*20. We call N E C E S S A R Y that without which, as a j o i n t -

cause, it is not possible to live, as for instance breathing and
nourishment are necessary for an animal, because i t is
incapable of existing without them: and anything without
which it is not possible for good to exist or come to be, or for
bad to be discarded or got rid of, as for instance drinking
25 medicine is necessary so as not to be iU, and saiUng to Aegina
so as to get money.
1015 *26. Again, that which is compulsory, and compulsion;
that is, what obstructs and thwarts an inclination or choice.
For what is compulsory is caUed necessary, which is why i t is
also disagreeable, as Evenus asserts,
N0 necessary deed
But has an irksome nature,
30 and compulsion is a kind of necessity, as Sophocles says.
Compulsion does necessitate I do this;
and necessity is, rightly, thought of as not open to persuasion,
for it is contrary to that change which is in accordance with
choice and reasoning.
1015*33. Again, when it is not possible for a thing to be
otherwise, we assert that i t is necessary for i t to be so.
35 Indeed the others are aU i n some way caUed necessary by
virtue of this. For what is compulsory is caUed necessary
i 0 i 5 b either to do or to suffer when i t is not possible to follow
incUnation on account of that which compels, necessity being
34
 TRANSLATION

that on account of which i t is not possible to do otherwise;

and the same is true in the case of the joint-causes of living
and of good, for when i t is not possible that i n the one case
good and i n the other case Hving should exist without certain 5
things, those things are necessary and that cause is a kind of
necessity. Again, demonstration is among the things that are
necessary, because it is not possible for a thing to be otherwise
i f i t has been demonstrated baldly; the cause ofthis is the
initial [premisses], i f the things from which the reasoning
proceeds are incapable of being otherwise.
1015b9. With some things, then, another thing is the cause 10
of their being necessary; with others nothing is, but on
account of them other things are of necessity. I t follows that
the primary, and fundamentally, necessary thing is that
which is simple; for i t is not possible that this should be in
more than one state, nor therefore thus and otherwise—^for
it would thereby be i n more than one state. Consequently,
ifthere are certain invariable and changeless things, there is is
nothing compulsoryx>r unnatural i n them.

CHAPTER 6
1015^16. Things are caUed O N E either coincidentaUy or i n
their own right: coincidentaUy as for instance Coriscus and
the artistic and artistic Coriscus (for i t is the same thing to
say 'Coriscus and the artistic'' and 'artistic Coriscus') and
the artistic and thejust and artistic andjust Coriscus. For aU 20
these are caUed one coincidentaUy, the just and the artistic
because they coincide in one substance, the artistic and
Coriscus because one coincides i n the other. EquaUy the
artistic Coriscus is in a certain sense one with Coriscus
because one of the portions in the formula coincides in the 25
other, I mean the artistic in Coriscus; and the artistic
Coriscus with just Coriscus because a part of each coincides
in the same one thing. The same is true ifthe coincidental is
' Omit <iv>.

35
 1015b METAPHYSICS J6

spoken o f i n the case ofa genus or i n the caseofthe names of

30 something universal, as for instance that a man and an
artistic man are the same thing; for i t is either because the
artistic coincides i n what is one substance, the man^ or
because both coincide i n a certain particular thing, as for
instance Coriscus (except that they do not both hold good
in the same manner, but one doubtless as a genus and in the
substance, the other as a state or affection ofthe substance).
35 Everything caUed one coincidentaUy, then, is so caUed in
this sense.
1015b36. Ofthings caUed one i n their own right, some are
1016* so caUed from being continuous, as for instance a bundle
from its tie and planks ofwood from their glue; and a line,
even i f bent, is caUed one i f it is continuous, as is each part
[ofthe body], as for instance a leg and an arm. But among
these what is naturally continuous is more one than what is
5 artificially so. We caU continuous that whose change in its
own right is one and cannot be otherwise; and a change is
one when indivisible, and indivisible in respect of time.
Whatever is one not by contact is continuous in its own
right; for i f y o u put planks in contact with one another you
wiU not assert that these are one plank or body or anything
else continuous. Continuous things in general, then, are
10 caUed one even ifthey have a bend, but still more those that
have no bend, as for instance the shin or thigh more than the
leg because i t is possible for a change of the leg not to be one
[change]. Also a straight Une is more one than a bent Une;
the line that is bent and has an angle we call both one and
15 not one, because i t is possible for a change to be made in i t
both aU at once and not aU at once, but that in a straight
line is always made aU at once and no portion having magni­
tude is at rest whUe another changes, as happens with the
bent line.
1016*17. I n another sense again, a thing is caUed one
from its subject's being undifferentiated in form, and it is
undifferentiated i f its form is perceptually indivisible. And
36
Δ6 TRANSLATION ioi6»

the subject is either the first or the last relative to the final 20
state; for wine is caUed one and so is water, in that they are
indivisible i n respect of form, and juices (as for instance oil
and wine) and meltables are aU caUed one because the
ultimate subject of aU of them is the same—for aU these
things are water or air.
1016*24. Things are also caUed one whose genus is one,
being differentiated by opposite differentiae; and these are 25
aU caUed one because the genus which is the subject oftheir
differentiae is one (as for instance a horse, a man, and a dog
are one something because aU animals)—in much the same
sense, indeed, as the things whose matter is one. These
things are sometimes caUed one in this way, but sometimes^
the genus above is caUed the same, i f they are the last forms 30
of the genus—that which is further above these;^ as for
instance the isosceles and the equilateral are one and the
same figure because both triangles, but they are not the same
triangles. \

1016*32. Again, things are caUed one when the formula

saying what it is to be is indivisible relative to another
formula which indicates the actual thing (for taken by itself
every formula is divisible). For in this way what has grown 35
and is diminishing is one, because its formula is one, as is that
of the form i n the case of planes. I n general when the con- 1016*
ception which conceives w h a t i t is to be certain things is
indivisible and cannot separate them in time or place or
formula, they are most of aU one, and those that are sub­
stances most of aU among these. (For whenever things are
without division, they are universally caUed one i n that
respect in which they are without it, as for instance ifthey are 5
without division qua man they are one man, i{qua animal one
animal, i{qua magnitude one magnitude).

i o i 6 ^ 6 . While most things, then, are caUed one from

either doing or possessing or being affected by or being
^ Omit <OTi>. ^ Read τό άνωτέρω τούτων.

37
 ioi6b METAPHYSICS J6

related to some other thing that is one, the things caUed one
in the primary way are those whose substance is one, and
one either i n continuity or in form or in formula; for things
10 which either are non-continuous or do not have one form or
do not have one formula we in fact reckon as more than one
thing.

ιοιβ^ιι. Again, although in a way we assert that any­

thing whatever is one which is a quantity and continuous,
in a way we do not i f i t is not some kind ofwhole, that is, i f
it does not possess one form; as for instance ifwe observed the
parts ofa shoe put together anyhow we should not so readily
15 assert that they were one (unless on account of their con­
tinuity), but only ifthey were put together in such a way as
to be a shoe and thereby possess some one form. That is why
a circular line is ofaU lines most one, because it is whole and
complete.

1016^x7. To be one is to be a kind oforigin ofnumber;^

for a first measure is an origin, for what first makes each
20 genus intelligible to us is its first measure. The origin, there­
fore, of our acquaintance with each [kind of] thing is that
which is one. But that which is one is not the same thing i n
every genus; for i t may be here a quarter-tone, there a vowel
or mute, and another thing i n the case of weight and some­
thing else i n the case ofchange. But i n every case that which
is one is indivisible either in quantity or in form. Now what is
25 indivisible i n respect of quantity in aU dimensions is caUed
a unit i f i t has no position, apoint i f i t has position, and what
is divisible in respect ofquantity in one dimension is a line, i n
two a plane, i n aU three a body. I n the reverse order, what
is divisible in two dimensions is a plane, in one dimension a
line, and what is divisible i n no dimensions in respect of
30 quantity is a point and unit, the unit being without position, ,
the point with position.

' R e a d άρ;^^ nvt iariv άριθμοΰ elvai.

38
J6 TRANSLATION ioi6b

1016b31. Again, some things are one in respect ofnumber,

some i n respect of form, some i n respect of genus, some in
respect of analogy: i n number things whose matter is one, i n
form things whose formula is one, in genus things whose
figure of predication is the same, in respect of analogy any
things related as are two further things. I n every case the
earlier imply the later, as for instance what is^>ne in number
is also one i n form but what is one in form is not aU in num­
ber, and whatever is one in form is aU one in genus but what
is in genus is not aU i n form; it is, however, in analogy, but
what is i n analogy is not aU i n genus.

i o i 7 ' 3 . I t is obvious that what is MANY wiU be so caUed in

opposite ways to what is one. For some things wiU be many
from being non-continuous, some from possessing matter
(either the first or the last) which is divisible in respect of
form, some from having more than one formula saying what
it is to be.
\,

CHAPTER 7

1017*7. That which is may be so caUed either coincidentaUy

or in its own right: coincidentaUy, as for instance we assert
someone just to be artistic, and a man artistic, and someone
artistic a man; in much the same way as we say that someone
artistic builds, because being artistic coincides in a house­
builder or a housebuilder in someone artistic (for 'that this is
this' signifi^s'that in this this coincides'). And so it is in the
cases mentione4i for when we say that a man is artistic and
someone artistic a man, or that someone pale is artistic or the
latter pale, i n the one case it is because both coincide i n the
same thing and i n the other because it coincides i n a thing-
that-is; while the artistic is a man because the artistic
coincides in the latter (it is in this way that the not-pale is
said to be, because what it coincides in is). Things said to be

39
 ioi7* METAPHYSICS J7

20 coincidentaUy, then, are so said i n this way: either because

both hold good of the same thing-that-is, or because the
former holds good of a thing-that-is, or because the thing
itsetf, of which that of which it is itself predicated holds
good, is.

1017*22. AU things which signify the figures ofpredication

are said to be in their own right; for Ч0 be' signifies in the
same number of ways as they are said. Since, therefore,
25 among things predicated some signify what a thing is, some
a quaUfication, some a quantity, some a relative, some doing
or being affected, some where, some when, Ч0 be' signifies
the same thing as each of these. For there is no difference
between *a man is one that keeps-healthy' and 'a man
keeps-healthy' or between *a manzj one that walks, or cuts'
30 and 'a man walks, or cuts', and equaUy in the other cases.

1017*31. Again, Ч0 be' and *is' signify that a thing is true,

and 'not to be' that i t is not true but a falsehood, equaUy i n
the case of affirmation and of denial; as for instance that
Socrates is artistic, that this is true, or that Socrates is not-
35 pale, that i t is true; and 'a diagonal is not commensurable'
that i t is a falsehood.

1017* 1017*35. Again, Ч0 be' and 4hat which is' signify those of
the things mentioned which are potentially and those which
are in complete reality; for both that which sees potentially
and that which sees in complete reality we assert to be a
thing-that-sees, and i n the same way both that which is
5 capable of using knowledge and that which is using i t we
assert to know, and both that of which rest already holds
good and that which is capable of being-at-rest we assert
to be-at-rest. EquaUy in the case of substances also; for both
the idol in the stone and the half ofa line and the grain which
is not yet ripe we assert to be. When a thing is or is not yet
capable must be defined elsewhere \Δ i 2 ] .

40
J8 TRANSLATION 1017b

CHAPTER 8

1017b10. We call a s U B S T A N C E both simple bodies, as for 10

instance earth and fire and water and everything of that
kind, and bodies in general and the things constituted out of
them—aiumals and deities and the portions of these; aU
these are caUed substance because they are not said of a
subject but the rest are said ofthem:
1017b14. i n another sense, any constituent of such things 15
(the things not said ofa subject) which is cause oftheir being,
as for instance the soul i n the case ofan animal:
1017b17. again, those constituent portions ofsuch things
which define and signify a this and with whose elimination
the whole thing is eliminated, as for instance the body with
the plane's (as some assert) and the plane with the line's; i n 20
general i t is thought by some that number is of this kind, on
the grounds that when it is eKminated there is nothing, and
it defines everything.
1017^21. Again, what i t is to be, the formula ofwhich is a
definition, is also caUed each thing's substance.
1017^23. I t follows, then, that a substance is so caUed in 25
two senses: both the ultimate subject, which is not further
said of anything else; and whatever, being a this, is also
separable (such is each thing's shape and form).

CHAPTER 9

1017b27. Some things are caUed т н Е SAME coincidentaUy, as

for instance the pale and the artistic are the same thing 30
because they coincide in the same thing, and man and artistic
because one ofthem coincides in the other; and the artistic
[is] a man because i t coincides in the man. And the latter is
the same as each of the other two and each of them as it, for
both the man and the artistic are i n fact caUed the same
41
 1017b METAPHYSIGS J9

thing as the man who is artistic and he as them. This explains

why aU these things are not spoken of universally; for it is not
true to say that every man and the artistic are the same
35 thing, for universals hold good in their own right and things
1018· thatare coincidental are not i n their own right. But in the
case of particulars we do so speak baldly, for Socrates and
artistic Socrates are thought to be the same thing. But
'Socrates' does not apply to a number ofthings, hence we do
not say 'every Socrates' as we say 'every man'.
5 i o i 8 * 4 . Some things, then, are caUed the same in this way;
others in their own right, in as many ways as that which is
one. For things whose matter is one, either in form or
number, are caUed the same, and things whose substance is
one; so that it is obvious that sameness is a kind ofoneness,
either ofthe being ofmore than one thing or when a thing is
treated as more than one (as for instance when someone says
that a thing is the same as itself, which is to treat it as two
things).
10 1018*9. We call things о т н Е К when either their forms or
their matter or the formula of their substance is more than
one; and in general what is other is so caUed in opposite ways
to what is the same.
1018*12. We call D i F F E R i N O any things which are other
while being the same something, not only in number but
either i n form or in genus or in analogy; again, things whose
genus is other; and contraries; and any things which possess
otherness in their substance.
15 1018*15. We call s i M i L A R both things whose affections are
the same in every way; and things more ofwhose affections
are the same than other; and those whose quality is one. And
what has the greater number or the more fundamental of
those contraries in respect ofwhich it is possible for something
to be modified, is similar to that thing.
i o i 8 * i 8 . Things are D i s s i M i L A R i n ways opposite to those
in which things are similar.
42
J10 TRANSLATION ioi8*

C H A P T E R 10

i o i 8 " 2 0 . We c a l l o p p o s r r E S contradiction, a n d contraries, 2 0

and relatives, a n d l a c k a n d possession, a n d the [points] from
w h i c h a n d to w h i c h comings to be a n d destructions ultimately
[lead]. And w h e n e v e r it is not possible for things to be s i m u l ­
taneously present i n that w h i c h is recipient of both of them,
these are said to be opposed, either themselves or the things
they are [ m a d e ] out of; for grey a n d white do not hold good 2 s
simultaneously ofthe same thing, because the things they are
[made] out of are opposed.

i o i 8 * 2 5 . We c a l l coNTRARiEs both those a m o n g the things

that differ i n respect o f genus w h i c h are not capable ofbeing
present simultaneously i n the same t h i n g ; a n d the things that
differ most a m o n g those i n the same genus; a n d the things
that differ most a m o n g those i n the same recipient; a n d the
things that differ most a m o n g those u n d e r the same capacity; 30
a n d the things whose difference i s ^ e a t e s t either baldly or i n
respect of genus or i n respect of form. The r e m a i n i n g c o n ­
traries are so caUed either from possessing such as these, or
from being recipient o f t h e m , or from being productive o f o r
affectible b y t h e m , or [ i n process of] p r o d u c i n g or being
affected b y t h e m ; or are losses or acquisitions or possessions
or lacks of t h e m . 35

1018*35. Since that w h i c h is one a n d that w h i c h is m a y be

so caUed i n several w a y s , they are necessarily followed by
the other things that are spoken o f i n respect ofthese; hence
that w h i c h is the same a n d that w h i c h is other a n d that which is
contrary are other i n respect of e a c h predication.

i o i 8 ' 3 8 . We c a l l oxHER ra FORM both a n y things that are 1018»

o f t h e same genus but not subordinate to one another; a n d
any things that are i n the same genus but possess some
difference; a n d a n y things that possess contrariety i n their
substance. And contraries a r e other i n form f r o m o n e another,
either aU o f t h e m or those so caUed i n the p r i m a r y w a y ; also

43
 ioi8b METAPHYSICS J10

5 any things whose formulae are other in the last form of their
genus, as for instance a man and a horse are inseverable i n
genus but their formulae are other; also any things which are
in the same substance but possess some difference.

1018b7. Things spoken o f i n ways opposite to these are тнЕ

SAME IN FORM.

C H A P T E R 11

1018b9. We call things PRiOR and POSXERiOR as follows.

10 Since i n each genus something is primary and the origin, a
thing may be prior from being nearer a certain origin,
defined either baldly and by itsnature or with reference to
a certain thing or somewhere or by certain people. I n respect
ofplace, for instance, [things are prior] from being nearer to
some place defined either by their nature (as for instance the
middle or the end) or with reference to a chance object; and
15 what is further away is posterior. I n respect oftime [things
are prior] either, i n the case of what has come to be, from
being further from the present (for the Trojan Wars are
prior to the Persian because they are further distant from the
present) or, in the case of what is to come, from being nearer
to the present (for the Nemean Games are prior to the
Pythian because nearer to the present, treated as origin and
20 primary). Inrespect ofchange, what is nearertheprimary
bringer ofchange is prior, as for instance child to man; and
this too is a kind of origin, taken baldly. I n respect ofcapacity,
things which exceed in capacity, i.e. the more capable, are
prior. Such is that whose decision i t is necessary for the other,
25 posterior, thing to follow, so that the latter is changed when
it changes it and when not not; the decision is the origin. I n
respect of arrangement, that is, whenever there is a formula
ranking things with reference to some one defined thing, a
second-ranker, for instance, is prior to a third-ranker and a
next-to-bottom string to a bottom one; the origin being in
44
 TRANSLATION ioi8b

the one case the leader, in the other the middle string. These
things, then, are called prior i n this sense. 30
1018b30. I n another sense, what is prior in acquaintance is
[treated] as also prior baldly.
10x8^31. But among these the cases o f a formula and of
perception are different: i n respect ofthe formula universals
are prior, in respect of perception particulars. Also, i n re­
spect of the formula the coincidental is prior to the whole 35
thing, as for instance the artistic to the artistic man; for
without the part the formula wiU not be whole. And yet it
is not possible to be artistic without being an artistic some­
thing.

1018^37. Again, the affections oi'what is prior are caUed

prior, as for instance straightness to smoothness; for the one
is an affection ofa line i n its own right, the other ofa surface. 1019»
1019*1. Besides things caUed prior and posterior in this
way, a thing is prior in respect of its nature and substance
when i t is possible for i t to be vdthout other things but not
them without it: this division was used by Plato. But since
there are many ways ofbeing, firstly, the subject is prior and S
for that reason the substance is prior; but, secondly, there
is a difference between the cases of capacity and complete
reality, some things being prior in respect ofcapacity, some in
respect ofcomplete reality. Half [of a line], for instance, [is
prior] i n respect of capacity to the whole [line], and the
portion to the whole, and matter to substance; but in respect
of complete reality they are posterior, for when the others 10
are dissolved they wiU be in complete reality.

1019*11. I n a certain sense, everything caUed prior and

posterior is so caUed in respect ofthese last; for i t is possible
for some things to be without the others in respect of coming
to be, as for instance a whole without its parts; and for others
in respect of destruction, as for instance a portion without
the whole. EquaUy i n the other cases also.
45
 ioig· METAPHYSICS Δί2

C H A P T E R 12

15 1019*15. We call a C A P A C i X Y what originates a change or

alteration either in another thing or qua other, as for instance
housebuilding is a capacity which is not a constituent of the
thing being built, but doctoring, which is a capacity, might
be a constituent of the thing being doctored, but not of i t
qua being doctored.

1019*19. We call a capacity what originates an alteration

20 or change not only i n this way in another thing or qua other,
but also by the agency of another thing or qua other. For in any
respect in which a thing affected is at aU affected, we assert it
to be capable of being affected; sometimes whatever the
affection, sometimes not i n respect of every affection but i f it
is for the better.

1019*23. Again, there is the [capacity] to perform thus

satisfactorily or at wiU. For those who have only got about or
25 spoken, but not satisfactorily or not as they chose, we some­
times assert not to be capable of speaking or walking; and
equaUy too in the case of being affected.

1019*26. A g a i n , any state i n respect ofwhich a thing is

wholly unaffectible or unalterable or not easy to change for
the worse is caUed a capacity. For things get broken and
crushed and bent and generally destroyed not from being
30 capable but from being not capable and in some way de­
ficient. A thing is unaffectible by such treatment i f owing
to a capacity—i.e. from being capable and in some state or
other—it does not get affected easily, or seriously. Those are
aU the ways in which capacities are so caUed.
1019*33. Things caUed C A P A B L E in one sense wiU be those
35 which originate change or alteration (for what can keep a
thing the same is i n a way also capable) in other things or
1019^ qua other; in another sense, i f something else possesses such
a capacity over them; i n another sense, i f they possess the
46
J12 TRANSLATION

capacity to alter somehow or other, whether for the worse or

for the better. (For a thing i n [process of] destruction is
thought to be capable ofbeing i n [process of] destruction; or
it would not be destroyed, i f i t were incapable. As things are,
it possesses a certain disposition to, and cause and origin of, 5
this kind of affection. Sometimes, then, i t is thought to be
such from possessing something, sometimes from lacking.
But i f a lack is in a way a possession, aU would be so from
possessing something,^ so that things are capable both from
possessing a sort of possession and origin, and from possessing 10
the lack of this, i f i t is possible to possess a lack; or, ifthat is
not possible, homonymously.^) [Things are caUed capable]
in another sense i f neither other things nor [they] qua other
possess a capacity for, or wiU or%inate, their destruction.
Again, aU these things [are caUed capable] either merely
because they might turn out to come to be or not to come to
be, or because they might do so satisfactorily. For the latter
kind of capacity is inherent even in inanimate things, as for
instance i n instruments; for people assert that one lyre is
capable of sounding, but another not at aU, i f it is out of 15
tune.

1019^15. I N C A P A C I T Y is lack ofcapacity, i.e. ofthe kind of

origin described, either i n general or by something which
characteristically possesses i t or even at a time already
characteristic of its possession. For people would not assert
in the same way that a boy, a grown man, and a eunuch are
incapable of begetting. Again, corresponding to each of the
two capacities (for merely changing things, and for changing 20
them satisfactorily) there is an opposite incapacity.

1019^21. Some things are also caUed m c A P A B L E in respect

of this incapacity.
1019^22. But there is another sense i n which things can be
bothcapable and incapable [possiBLE and i M P 0 s s i B L E ] . That
ofwhich the contrary is true ofnecessity is impossible; as
" Put €t δ€ μη^ ομωννμως at the end of the sentence.

47
 METAPHYSICS J12

for instance that a diagonal be commensurable, because

such a thing is a falsehood ofwhich the contrary is not only
true but also a necessity.^ Consequently, that it is commen­
surable is not only a falsehood but also a falsehood of neces­
sity. The contrary of this, the possible,^ [occurs] whenever i t
is not necessary that the contrary be a falsehood; as for
instance i t is possible that a man be sitting down, for i t is not
of necessity a falsehood that he is not sitting down. I n one
sense, then, the possible, as stated, signifies what is not of
necessity a falsehood; in another what is true; and in another
what admits of being true.
1019^33. A capacity [pov^ER] in geometry is so caUed by
transference.
1019^34. These things, then, are capable not in respect ofa
35 capacity. Those which are so caUed i n respect of a capacity
1020* are aU so caUed with reference to the one which is primary,
that is, the origin of alteration i n something else or qua
something else. For the others are caUed capable either from
something else's possessing a capacity ofthat kind over them,
or from its not possessing it, or from its possessing i t in a par­
ticular way; and equaUy with things which are incapable.
5 I t follows that the fundamental definition of the primary
capacity must be: what originates alteration in something
else or qua something else.

C H A P T E R 13

1020*7. We call a Q U A N T r i Y what is divisible into con­

stituents each ofwhich has the nature ofa one and a this. A
certain quantity is a pluraUty i f i t is countable, a magnitude
10 i f i t is measurable; that which is potentially divisible into
non-continuous parts is caUed a plurality, into continuous a
magnitude. Of magnitudes those which are continuous i n
one dimension are lengths, i n two breadths, in three depths.
Ϊ Omit άσύμμ€τρον €ivai. * Read то hwarov.

48
Δ13 TRANSLATION I020*

Of these, Hmited pluraUty is a number, length a Une,

breadth a surface, depth a body.

1020*14. Again, some things are caUed a certainquantity 15

in their own right, others coincidentaUy, as forinstance a line
is a certain quantity in its own right, the artistic coinciden­
taUy.

1020*17. Ofthose in their ownright some are so in respect

oftheir substance, as for instance a line is a certain quantity;
for 'a certain quantity' is a constituent of the formula which
says what i t is. Others are affections and states of such a 20
substance, as for instance the much and the little, and long
and short, and broad and narrow, and deep and shallow, and
the others of that kind. Both the large and the small, and
larger and smaUer, when so caUed both i n their own right
and with reference to one another, are affections of a quantity 25
in their own right. However, these names are also transferred
to other things.

1020*26. Ofthings caUed quantities coincidentaUy, some

are so caUed i n the way i n which it was said [1020*14-17]
that the artistic and the pale are quantities, from there being a
certain quantity ofwhich they hold good; others as changes
and times, for these are also caUed certain quantities and
continuous, from those things of which they are affections 30
being divisible—I mean not the thing being changed but
that [amount by] which it was changed; for from that being
a quantity the change is a quantity also, and the time from
its being.

CHAPTER 14

1020*33. We caU a QUALmcATiON, i n one sense, the differen­

tia ofa substance, as for instance a man is an animal qualified
in a certain way because he is two-footed, a horse because
it is four-footed; and a circle is a figure qualified i n a certain 35

49
 I020« METAPHYSIGS J14

1020^ way because i t is withoutangles; the difFerentia in respect of

substance being a quaHty. This is one sense in which a quaHty
is so caUed.

i 0 2 0 ^ 2 . Another sense is exempUfied by changeless things

and mathematical objects, as i n the case of numbers being
qualified in certain ways (e.g. composite [numbers] which,
5 being of more than one dimension, are represented by a plane
and a solid, i.e. x timesj and x timesjv times z ) ; and in general
any constituent of the substance apart from quantity. (For
the substance ofeach thing is what [it] is once, as for instance
what six is, not two or three times, but once; for six is once
six.)

i020^8. Again, any affections of changing substances, as

10 for instance hotness and coldness, and paleness and darkness,
heaviness and lightness, and aU of that kind. Bodies are also
said to be modified in respect of alterations of these.

i 0 2 0 ^ i 2 . Again, [there are quaUties] i n respect of excel­

lence and badness, and in general the bad and good.

1020b13. Approximately, a quaUfication could be so caUed

in two senses, and of these one is the most fundamental.
15 For the primary quality is the differentia ofa substance (this
includes the quaUty i n a number, for that is a kind of
differentia of substances, but not of changeables or not of
them qua changing). Others are the affections ofchangeables
qua changing, and the differentiae ofchanges. The affections
include excellence and badness, for these indicate differentiae
20 ofchange and ofactual functioning in respect ofwhich things
in [process of] change act or are 2jfected satisfactorily or not.
For what is capable ofchanging or functioning in this way is
good; what in that—^the contrary—way, evil. The good and
the bad signify qualification i n the case of living things
especially, and among them especially in the case of those
25 which possess [the power of] choice.
50
Δ 15 TRANSLATION ioaob

C H A P T E R 15

1020^26. Some things are called R E L A x r v E as double is

relative to half and triple relative to a third, and i n general
multiple relative to submultiple and exceeding relative to
exceeded; others as the able-to-heat is relative to the heatable
and the able-to-cut relative to the cuttable, and in general 30
the able-to-act relative to the affectible; and others as the
measurable is relative to the measure, and knowable relative
to knowledge, and perceptible relative to perception.
1020^32. The first kind are caUed numerical [relatives],
either baldly or in a definite way, and relative to themselves
or to one. For instance: the double is a definite number
relative to one; the multiple is related to one numerically, 35
but not by any definite number such as this or this; the one- 1021*
and-a-half is related to its reciprocal by a definite number
relative to a number; the one-and-a-bit is related to its
reciprocal by an indefinite number, as the multiple is related
to one. The exceeding is numerically whoUy indefinite
relative to the exceeded; for a number is commensurable 5
and number is not said of the non-commensurable, but the
exceeding is, relative to the exceeded, so much andsomething
again, and this is indefinite—^for i t is as chance has it, either
equal or not equaL These relatives, then, are aU caUed
numerical relatives and affections of number.
i 0 2 i * g . So again, i n another sense, are the equal and
simUar and the same. For they are aU so caUed in respect of 10
the one; for things are the same whose substance is one,
sinülar whose quaUty is one, equal whose quantity is one.
And the one is the origin and measure o f a number, so that
aU these are caUed numerical relatives, but not i n the same
sense.
1021*14. Things able-to-act and afFectible [are relatives] 15
in respect of their capacity to act and be afFected and the
actual functioning of these capacities; as for instance the
51
 I02I» METAPHYSICS Δ 15

able-to-heat is relative to the heatable because it is capable,

and i n turn the heating is relative to the being-heated and the
cutting to the being-cut i n that they are actually functioning.
Numerical relatives do not have actual functionings except
20 i n a sense described elsewhere; actual functionings involving
change do not hold good of them. With some relatives in
respect of capacity a temporal reference is included in the
statement of the relation, as for instance what has produced
relative to what has been produced and what will produce
relative to what wiU be produced. For a father is caUed
father ofhis son in this way; for one ofthem is a thing that
has produced and the other a thing that has been affected
25 in a certain way.

1021*25. Again, some things [are relatives] i n respect o f a

lack of capacity, as the incapable and anything else spoken
of in that way, as for instance the invisible.

1021*26. Things caUed numerical relatives or relatives in

respect of capacity are aU relatives from being caUed just
what they are of something else, not from the other thing
being relative to them. But the measurable and the know-
30 able and the thinkable are caUed relatives from something
else being caUed [what i t is] relative to them. For 'thinkable'
signifies that there exists thought ofit, but the thought is not
relative to that of which i t is a thought (for then we should
have said the same thing twice). And equaUy, sight is the
1021^ sight of something, not of that of which it is the sight (al­
though i t is true to say that) but relative to colour or some
other such thing. The other way we shall be saying the same
thing twice: 'sight is ofthat ofwhichsight is.'

1021^3. Ofthings caUed relatives i n their own right, then,

some are so caUed i n the ways mentioned; others i f their
5 genera are of those kinds, as for instance doctoring is among
the relatives because its genus, knowledge, is thought to be
a relative;
52
 TRANSLATION I02lb

i 0 2 i ^ 6 . again, aU those things i n respect ofwhich their

possessors are caUed relatives; as for instance equaUty is a
relative because the equal is and similarity because the
similar is;
i 0 2 i ^ 8 , other things coincidentaUy; as for instance a man
is a relative because being double coincides i n him, and that l 0
is among the relatives; or the pale, ifbeing double and pale
coincide in the same thing.
i

C H A P T E R 16

i 0 2 i ^ i 2 . We caU c o M P L E T E , in one sense, that outside which

not even one portion is to be found, as for instance the com­
plete time ofeach thing is that outside which there is no time
to be found which is part ofthat time:
1021^14. also, that which i n respect of excellence and 15
goodness cannot be surpassed relative to its genus, as for
instance a doctor is complete and a flautist is complete when
they are without deficiency i n respect of the form of their
own proper excellence. I t is in this way that, transferring it to
the case ofbad things, we speak ofa completescandalmonger
and a complete thief—as indeed we even caU them good:
a good thief and a good scandalmonger. And excellence is 20
a kind of completion, for each thing is complete and every
substance is complete when i n respect ofthe form ofits own
proper excellence no portion of its natural magnitude is
deficient.
1021^23. Again, things which have reached their fulfil­
ment, when i t is worth while, are caUed complete, for they
are complete by virtue of having attained their fulfilment; 25
so that, since a fulfilment is something ultimate, we also say,
transferring it to the case ofworthless things, that a thing has
been completely spoilt and completely destroyed when there
is no deficiency in its destruction and badness but i t has
reached the ultimate. (That is why even life's-end is by
53
 ioaib METAPHYSICS Δ 16

transference caUed a fulfilment, because both are ultimate.

30 A [thing's] fulfilment, i.e. what it is for, is ultimate.)

1021^30. This, then, is the number ofways in which things

caUed complete in their own right are so caUed: some from
their being without deficiency in respect ofgoodness and not
to be surpassed and having nothing to be found outside
them, others in respect ofbeing in general not to be surpassed
1022· in their various genera and having nothing outside them.
The rest are made so by virtue of these, from their either
producing or possessing or fitting something of such a kind,
or from their being somehow or other so caUed with reference
to the things caUed complete in the primary way.

CHAPTER 17

1022*4. caU a L I M I T : the terminus of each thing, i.e. the

5 first thing outside which there is nothing to be found and the
first thing inside which everything is to be found;

1022*5. anything that is the form o f a magnitude or

of [something] possessing magnitude;

1022*6. and the fulfilment of each thing (such is that

towards which change and action [lead], not that from which
—but sometimes both, from which and towards which, i.e.
that for which);'

1022*8. and the substance ofeach thing, i.e. what it is to be

10 each thing, for that is the limit of acquaintance and, i f of
acquaintance, of the actual thing too.

1022*10. I t is obvious from this that a limit is so caUed i n

aU the ways that an origin is, and more: for an origin is a
kind of limit, but not every limit is an origin.
* Close bracket after καΙ то οΰ ерека.

54
 TRANSLATION ioaa·

CHAPTER 18

1022*14. That B Y W H I C H may be so caUed in several ways.

In one sense it is the form and substance ofeach actual thing, 15
as for instance that by which [a man is] good is good itself.
In another sense it is the primary thing in which it is charac­
teristic of something to come to be, as for instance colour in
a surface. WhUe that caUed by which in the primary way is
thus the form, each thing's matter and each thing's primary
subject is so caUed i n a secondary way. I n general, that by
which wiU hold good in the same number of ways as cause; 20
for we say Ъу what did he come?' or ^what did he come
for?' and Ъу what did he miscalculate, or calculate?* or
'what is the cause of his calculation, or miscalculation?'
Again, what involves position is caUed by which: 'by which
he is standing', Ъу which he is walking'—for aU these^
signify position and place.

1022*24. I t follows that that which is B Y i T S E L F [ i N iTS 25

o w N RIGHT] is necessarily also so caUed in several ways. For
in one, [a thing is] i n its own right what it is to be each
thing, as for instance Callias is i n his own right Callias, and
what it is to be CaUias;

1022*27. in another, any constituent ofwhat it is, as for

instance Callias is i n his own right an animal; for animal is a
constituent ofhis formula, for CalUas is a certain animal;

1022*29. again i f i t , or one ofits [parts], is the primary 30

recipient, as for instance a surface is pale in its own right and
a man is alive in his own right (for the soul, the primary
recipient of life, is a certain part of a man);

1022*32. again, what has no other cause; for there are

several causes of a man—animal, the two-footed—but
nevertheless a man is a man in his own right;
Ϊ Read ττάντα yap ταυτα (correcting a misprint in the O C T ) .

55
 ioa2» METAPHYSICS J18

35 1022*35, again, [a thing is] in its own right any things that
hold good of it alone and qua separated on its own account
alone.^

CHAPTER 19

1022^ i 0 2 2 ^ i . We call a D i s p o s m o N the arrangement, either by

place or capacity or form, ofsomething possessing parts. For
it ought to be some kind of position, as the very name
'disposition' indicates.

CHAPTER 20

1022^4. caU a POSSESSION, in one sense [wearing], what is

5 as it were a certain functioning of the wearer and worn, like
a certain action or change. For whenever one thing produces
and another is produced, there is producing between them;
just so there is wearing between one who wears clothes and
the clothes worn. Obviously i t is not possible to possess this
kind of possession, for we shall go on to infinity i f there is
10 such a thing as possessing the possession ofwhat is possessed.
1022^10. I n another sense we call possession [state] a
disposition by virtue of which the thing disposed is either
well or iU disposed, and either in its own right or with
reference to something else; as for instance health is a certain
state, for it is such a disposition. Again, anything that is a
portion of such a dispositionwe call a state; that is why the
excellence of a thing's parts is a certain state.

CHAPTER 21

15 1022^15. We call an A F F E C T i O N , in one sense, a quality in

respect of which i t is possible to be modified, as for instance
the pale and the dark, and sweet and bitter, and heaviness and
lightness, and aU others ofthat kind;
' R e a d καΧ ^ μόνον bi* αύτο Κ€χωρισμ4νον καθ* αύτό.

56
Δ 21 TRANSLATION югаЬ

i 0 2 2 ^ i 8 . i n another sense, the functionings of these, i.e.

the actual modifications;
i 0 2 2 ^ i g . of these again, mainly injurious modifications
and changes, and especially painful injuries. 20
i022^20. Again, severe misfortunes are caUed aiFections.

CHAPTER 22

i022^22. We speak ofa ьлск, in one sense, i f a thing does not

possess something characteristically possessed, even i f i t is
not characteristic of that thing to possess i t ; as for instance
a plant is said to lack eyes;

1022^24. in another sense, i f a thing does not possess what

is characteristically possessed by i t or its genus; as for 2S
instance a blind man and a mole lack sight i n diiFerent ways,
the one i n respect ofits genus, the other i n its own right;
1022^27. again, i f a thing does not possess something
characteristic even at the time when i t is characteristic to
possess it (for blindness is a certain lack, yet one is not blind
at every age but only ifone does not possess [sight] at an age
at which its possession is characteristic); and equally too i f i t 30
does not possess something in that which, and with respect to
that which, and relative to that which, and in the manner
which, is characteristic.

1022b31. Again, the compulsory removal ofeach thing is

called a lack.
1022^32. Lacks are so called i n the same number ofways
as denials containing 'un-' and the like. For a thing is caUed
uKequal from not possessing its characteristic equality, in­
visible from possessing colour either not at aU or i n a meagre 35
degree, and footless from possessing either no feet at aU or
meagre ones.
57
 I022b METAPHYSIGS Δ22

1023* 1022^36· Again, [we speak of a lack] from a thing's

possessing little, as for instance stoneless fruits (this is, in a
way, possessing in a meagre degree);
1023*2. again, from the thing's not being easy or not being
satisfactory, as for instance the uncuttable not just from its
not cutting but also from its not cutting easily or satisfac­
torily;
1023^4. again, from not possessing at aU; for a one-eyed
5 man is not caUed blind, but only a man who possesses no
sight i n either eye. That is why not everyone is good or bad,
orjust or unjust, but there is also the middle [state].

CHAPTER 23

1023*8. POSSESSING is SO callcd in several ways. I n one sense

it is directing a thing according to one's own nature or
10 according to one's own inclination. This is why fever is said
to possess a man and tyrants cities and wearers their clothes;
1023*11. in another sense, that ofwhich, as a recipient,
something is a constituent, as for instance bronze possesses
the form o f a statue and a body possesses disease;
1023*13. in another sense, as a container its contents; for
15 a thing is said to be possessed by the container i n which it is,
as for instance we assert that a bucket possesses liquid and a
city men and a ship seamen. I n this way also a whole possesses
its parts.
1023*17. Again, what prevents something changing or
acting according to its own inclination is said to possess that
thing, as for instance columns their overlying weights and as
20 poets make Atlas possess the sky (as i f i t would coUapse on to
the earth, as even some authorities on nature assert). This is
the sense i n which what fuses things is said to possess the
things it fuses, as i f each would have separated according to
its own inclination.
58
Δ 23 TRANSLATION 10a3»

1023*23. Being ш something is so caUed in simUar, and

corresponding, ways to possessing. 25

C H A P T E R 24

1023*26. We caU being ουτ O F som(ething, in one sense, that

out ofwhich as matter a thing is, and this i n two ways, either
in respect of the first genus or in respect of the last form; as
for instance i n one way everything meltable is out of water
but in another way a statue is out ofbronze;

1023*29. in another sense, as out ofthe first origin that 30

eifected change; as for instance 'What did the fighting
[come] out of?' 'out of the swearing'—because that was the
origin of the fighting;
1023*31. in another sense, as out ofwhat is composed out
ofmatter and shape, as parts are out ofa whole, a line out of
the Iliad, and stones out of a house; for the shape is a fulfil­
ment, and what has attained fulfilment is complete;

1023*35. others as the form is out of the part, as for 35

instance a man is out of the two-footed and a syllable out of
an element (for this is different from the way i n which the 1023ь
statue is out of bronze; for a composite substance is out of
perceptible matter, but a form is also out ofthe matter ofthe
form).

1023^3. Some things, then, are so caUed in these ways;

others i f one of these senses holds good of something in part,
as for instance a child is out of its father and mother, and
plants out of earth, because out ofa certain part ofthem; 5
1023^5. another sense, that after which i n time, as for
instance night is out ofday and a storm out ofgood weather,
because one is after the other. And of these, some are so
caUed from having alteration into one another, like those
just mentioned, others from being merely successive in
59
 METAPHYSICS Δ 24

10 respect of time, as for instance the voyage came to be out of

the equinox because it came to be after the equinox, and the
Thargelia out of the Dionysia because after the Dionysia.

CHAPTER 25

1023^12. We call a P A R T , in one sense, the result ofany kind

of division of a quantity; for what is subtracted from a
quantity qua quantity is always caUed a part of it, as two is
15 caUed a part ofthree in a way;
1023^15. in another sense, only those among such parts
which give the measure ofa thing; that is why two is caUed
a part of three in a way, but in a way not.
1023^17. Again, the results of any non-quantitative
division of a form are also caUed its portions; that is why
people assert that forms are portions oftheir genus;
20 1023^19. again, anything into which a whole, whether a
form or something that possesses a form, is divided, or out of
which it is composed, as for instance both the bronze (that is,
the matter in which the form is) and the angles are parts of
a brunze cube, or a bronze ball.

1023^22. Again, whatever is in the formula indicating

each thing is also a portion ofthe whole; that is why a genus
is also caUed a part of its form, although in a different way
25 the form is a part of its genus.

CHAPTER 26
1023^26. We call a W H O L E both that of which no part is
absent out of those of which we call it a whole naturally;
and what contains its contents in such a manner that they
are one thing, and this in two ways, either as each being one
thing or as making up one thing. For what is universal and
60
J 26 TRANSLATION ioijb

what is said to be as a whole, implying that it is a certain 30

whole, is universal as containing several things, by being
predicated of each of them and by their all—each one—
being one thing; as for instance man, horse, god, because
they are aU animals. But what is continuous and limited [is
a whole] when it is some one thing made up of more than
one thing, especially when these are potential constituents
ofit but, ifnot, when they are actual. Among aU these, what 35
is naturally of such a kind is more [a whole] than what is
artificially, as we said in the case of what is one [ 6. ioi6"4];
wholeness being a kind of oneness.

i 0 2 4 ' i . Again, of quantities possessing a beginning, a i024»

middle, and an end, any i n which position makes no differ­
ence is caUed A L L , any in which it does, whole. Any that
admit both are both wholes and alls; these are the things
whose nature, though not their shape, survives transposition, s
as for instance wax and a cloak; for they are caUed both
whole and aU, for they possess both. But water and other
liquids, and number, are caUed aU; and we do not speak of
the whole number or the whole water, unless by transference.

1024^8. We call E V E R V those things to which 'aU' is applied

as to one thing, 'every' being applied to them as to distinct
things: aU this number, every [one of] these units. 10

CHAPTER 27

i 0 2 4 * i i . Not every chance quantity may be caUed TRUN¬

CATED; it ought to be divisible into parts and also a whole.
For two is not truncated when one of its ones is subtracted,
for what is taken from a thing by truncation and what
remains are never equal; but neither, in general, is any
number, for the substance ought also to remain—if a cup is 15
truncated, it must still be a cup; the number is not the same.
I n addition to this, not even everything that has dissimilar
parts [may be caUed truncated], for in one way a number
6i
 I024» METAPHYSIGS Δ27

also has dissimilar parts, e.g. two and three. But i n general
nothing i n which position makes no difference, as for instance
20 water and fire, can be truncated; to be such, a thing ought to
possess a position by virtue of its substance. Again, [it must
be] continuous; for a musical scale is out ofdissimilar parts
and possesses position, but cannot become truncated. I n
addition to this, not even wholes are truncated as a result of
the lack o{anj portion whatever; for it ought not to be any­
thing fundamental to their substance, but neither can it be
anywhere whatever. For instance, a cup i n which a hole has
25 been drilled is not truncated, but only i f a handle or some
extremity [is removed]; and a man not i f flesh or spleen [is
removed], but i f some extremity is—and not any extremity,
but only one which cannot come to exist once whoUy removed :
for this reason bald men are not truncated.

CHAPTER 28

1024*29. We speak of a G E N U S either i f the coming to be of

30 things possessing the same form is continuous, as for i n ­
stance 'so long as the genus of men is' means 'while their
coming to be is continuous';

1024*31. or i f people [stem] from someone who first

brought them into existence; for in this way some are caUed
Hellenes by genus and others Ionians, from their [stemming]
from Hellen and Ion who first begot them; and more espe-
35 cially from the begetter than from the matter (for people
are caUed by genus from the female too, as some are from
Pyrrha);

i024b 1024*36. again, as the plane is the genus offigures that

are plane, and the solid ofthose that are solid; for eachofthe
figures is either a plane of such and such a kind or a solid of
such and such a kind, that being the subject ofits differentiae;
62
Δ 28 TRANSLATION

1 0 2 4 4 · again, as the first constituent in formulae which is 5

stated in [answer to the question] what a thing is; for this is
the genus (and its quaUties are caUed differentiae).
1024^6. These, then, are aU the ways in which a genus is so
caUed: in respect ofthe continuous coming to be ofthe same
form; in respect of the first thing, having the same form, to
effect change in a thing; and as matter—for what differentiae
and qualities are ofis their subject, which we call the matter.
1024^9. Things are caUed oxHER ш GENUS whose first 10
subjects are other, and wiU not analyse either one into the
other or both into the same thing; as for instance form and
matter are other in genus;
1024^12. also any things spoken of according to different
figures in the predication of that which is (for of the things-
that-are some signify what a thing is, others a certain quali­
fication, others in the other ways previously distinguished);
for these wiU not analyse either into one another or into 15
some one thing.

CHAPTER 29

1024^17. We call a F A L S E H O O D , in one sense, what is a false­

hood as an actual thing: and this sometimes from the thing's
being not compounded, or incapable of being compounded,
aswe say o f a diagonal's being commensurable or ofyour 20
sitting down—for one ofthese is a falsehood always, the other
sometimes (for in this way these things are not things-that-
are) ; sometimes anything which, while being a thing-that-is,
is nevertheless characteristically imagined either not [to be]
such as it is or [to be] something that is not, as for instance a
sketch, and dreams—^for these are something, but not what
they impose on us to imagine [they are]. These, then, are the
ways in which actual things are caUed false, either from their 25
not themselves being or from their giving rise to an imagina­
tion of something that is not.
63
 METAPHYSICS J 29

1024^26. A false formula is, qua f a l s e , o f t h i n g s t h a t are n o t ,

a n d t h a t is why every f o r m u l a i s f a l s e of s o m e t h i n g o t h e r
than t h a t ofwhich it i s t r u e , a s f o r instance that o f a circle is
false of a t r i a n g l e . Each t h i n g h a s , i n one w a y , o n e f o r m u l a ,
30 t h a t o f w h a t i t is t o be; i n another w a y it has many, since
both i t a n d i t a f f e c t e d ( a s f o r instance Socrates a n d a r t i s t i c
Socrates) a r e i n a w a y the s a m e t h i n g . A f a l s e f o r m u l a is,
t a k e n b a l d l y , the f o r m u l a o f n o t h i n g . That is w h y Antisthenes
n a i v e l y c o n s i d e r e d t h a t n o t h i n g c a n l e g i t i m a t e l y be d e s c r i b e d
e x c e p t b y its o w n p r o p e r f o r m u l a , o n e to o n e ; a n o p i n i o n
f r o m w h i c h i t r e s u l t e d t h a t t h e r e is n o s u c h t h i n g a s c o n t r a ­
d i c t i o n , n o r e v e n p r a c t i c a l l y a s f a l s i t y . But t h e r e i s s u c h a
35 t h i n g a s d e s c r i b i n g e a c h t h i n g n o t o n l y b y i t s o w n f o r m u l a
but a l s o b y a n o t h e r ' s ; t h i s m a y be d o n e a l t o g e t h e r falsely,
1025» b u t a l s o i n a w a y t r u l y , a s e i g h t i s d o u b l e , b y t h e f o r m u l a
of two.

1025*1. Apart from these w a y s o f c a l l i n g things false, a

false man i s o n e w h o u s e s s u c h f o r m u l a e r e c k l e s s l y a n d d e l i b ­
erately, not o n a n y other account t h a n their o w n , a n d w h o
5 imposes such formulae o n other people; j u s t as w e assert
t h a t a c t u a l t h i n g s a r e false w h e n t h e y i m p o s e a false i m a g i n a ­
tion. This e x p l a i n s w h a t is m i s l e a d i n g a b o u t t h e a r g u m e n t
i n the Hippias t h a t t h e s a m e m a n i s f a l s e a n d t r u e . For it
t a k e s f o r f a l s e the m a n w h o i s capable o f f a l s i t y ( a n d t h a t i s h e
w h o k n o w s , t h e w i s e m a n ) ; a n d a g a i n it takes for better t h e
10 m a n w h o d o e s w r o n g willingly. The l a t t e r f a l s e h o o d i s g o t b y
i n d u c t i o n : f o r a m a n w h o l i m p s w i l l i n g l y is s u p e r i o r to one
w h o does so u n w i l l i n g l y (meaning by limping pretending,
s i n c e i f h e w e r e w i l l i n g l y lame h e w o u l d d o u b t l e s s be i n f e r i o r ,
as w i t h c h a r a c t e r - t r a i t s ) .

C H A P T E R 30
1025*14. We call coiNCiDENTAL w h a t holds good ofsome-
15 t h i n g a n d is t r u e to s a y , b u t n e i t h e r o f n e c e s s i t y n o r f o r t h e
most p a r t ; as for instance i f someone i n digging a t r e n c h
for a p l a n t f o u n d t r e a s u r e . T h i s — t h e f i n d i n g o f t r e a s u r e — i s

64
 TRANSLATION 1025a

accordingly coincidental for him who is digging the trench;

for the one is not of necessity out of or after the other, nor
does one for the most part find treasure when planting. And
someone artistic might be pale, but since this comes to be 20
neither ofnecessity nor for the most part, we call it coinciden­
tal. Hence, since there are things that hold good and things
they hold good of, and in certain cases a place and a time of
holding good, whatever holds good but not because it is this
or now or here will be coincidental. There is thus no definite
cause of the coincidental, but a chance one, and that is 25
indefinite. I t was a coincidence for someone to visit Aegina
if he went there not in order to visit but having been forced
off course by a storm or captured by pirates. The coincidence
has come to be, or is, but qua another thing, not qua itself; for
the storm was the cause of his not getting to the place he was
sailing for.
1025*30. Things are called coincidental in other ways also, 30
as for instance whatever holds good of each thing in its own
right without being in its substance, as for instance possessing
two right angles [does] of a triangle. These admit of being
invariable, but the former do not. The matter is discussed
elsewhere.

65
 METAPHYSICS BOOK EPSILON

CHAPTER 1

1025^3. We are seeking the origins and the causes of the

things-that-are, and plainly of them qua things-that-are. For
5 there is a particular cause of health and of fitness, and there
are origins and elements and causes ofthe objects ofmathe-
matics, and in general every thinking, or thought-partaking,
discipline deals with causes and origins, more or less precise.
But aU these disciplines delimit a particular thing-that-is—
a particular genus—and treat of i t , not of that which is
10 baldly or qua thing-that-is. Nor do they produce any state­
ment of what it is; but starting from that—having either
indicated i t by the senses or found a hypothesis as to what it
is—they proceed from that to demonstrate, either more or
less rigorously, the things that hold good in its own right of
the genus with which they are dealing. For that reason i t is
obvious that from such an induction there is no demonstration
15 ofsubstance, i.e. ofwhat a thing is, but some other manner of
indicating it. EquaUy, neither is anything said as to whether
the genus of which they treat is or is not, because it falls to
the same thinking to indicate both what a thing is and whether
it is.

1025^18. But since physics is one ofthe disciplines dealing

20 with a particular genus of thing-that-is (for it deals with
the sort of substance in which the origin of change and of
keeping-the-same is in itself), it is plain thatthis discipline
is neither practical nor productive. For in the case ofthe pro­
ducible the origin is in the producer—either intelligence
or art or capacity ofsome kind; and in the case ofthe doable
it is in the doer—choice: for the doable and the choosable
25 are the same. I t follows that, i f a l l thinking is either practical
or productive or theoretical, that concerned with nature
66
El TRANSLATION

must be of a theoretical kind, but a kind which studies such

of the things-that-are as are capable of being changed, and
substance as in a formula for the most part, yet not separable
[substance].

1025^28. We must pay attention to the manner of a thing's

formula, i.e. of what it is to be that thing; since the inquiry 30
wiU get nowhere otherwise. Among things defined, i.e. those
which are what something is, some are like the snub, others
like the concave, and the difference between these is that in
the snub matter is implicit—for the snub is a concave nose—
whereas concavity is independent ofperceptible matter. So i f
every naturally existing thing is caUed [what it is] in the
same way as the snub, as for instance nose, eye, face, flesh, 1026·
bone, and animal as a whole, and leaf, root, bark, and plant
as a whole—^for the formulae ofnone ofthem are independent
of change but always include matter—the manner in which
we need to investigate and define what a thing is in the case of
naturally existing things is plain.

1026*5. [ I t is plain] too that it faUs to the student of nature 5

to study a certain kind of soul, namely any which is not
independent of matter.

1026*6. AU this makes it obvious, then, that the study of

nature is theoretical. But mathematics is also theoretical. On
the other hand, it is not immediately plain whether the
objects of mathematics are changeless and separable, even
though it is plain that some mathematics studies its objects
qua changeless and qua separable. However, i f there is any- 10
thing invariable and changeless and separable, it is obvious
that acquaintance with it faUs to a theoretical discipline, not,
however, to the study of nature (which deals with certain
changeable things) nor indeed to mathematics, but to some­
thing prior to both. For the study ofnature deals with things
that are separable but not changeless, while certain parts of
mathematics deal with things which, though changeless, are 15
67
 1026« METAPHYSICS El

doubtless not separable but as in matter. But the primary

discipline wiU deal also with things separable and changeless.

1026*16. AU causes are necessarily invariable; but these

are especially so, for they are the causes of the divinities
obvious to us.

1026*18. I t follows that there must be three kinds of

theoretical philosophy, mathematical, natural, and theo-
20 logical; for it is not hard to see that the divine is a constituent
of a nature ofsuch a kind, i f of anything. O f these the most
estimable ought to deal with the most estimable genus. The
theoretical are to be preferred, then, among the other disci­
plines, and this among the theoretical.

1026*23. For one might be perplexed as to whether the

primary philosophy really is universal, or deals with a
25 particular genus and one particular nature. For not even
mathematics is aU of a piece in this respect, geometry and
astronomy being concerned with a particular nature, while
universal mathematics is common to aU. I f then there is no
other substance apart from those constituted naturally, the
discipline concerned with nature would be primary. But i f
30 there is some changeless substance, this is prior and is pri­
mary philosophy, and universal in this way, because primary;
and it would fall to it to study that which is qua thing-that-
is, both what it is and the things that hold good of it qua
thing-that-is.

CHAPTER 2

1026*33. But that which is, when baldly so caUed, may be so

caUed in several ways. One of them was that [which is] co-
35 incidentally, another that [which is] as true (and that which
is not, that [which is] as falsehood). Apart from these there
are the figures ofpredication, as for instance what a thing is,
how qualified, of what quantity, where, when, and anything
68
E2 TRANSLATION ioa6b

else that signifies in this sense; again apart fi:om aU these, 1026^
that [which is] potentially and actually.
1026^2. Since that which is may be so caUed, then, in
several ways, it has first to be stated that there is no study
that deals with that [which is] coincidentaUy. A sign of this
is its neglect in every discipline, practical, productive, and 5
theoretical. For one who produces a house does not produce
aU the things which coincide in the house that is coming to
be, for they are infinite. For there is nothing to prevent the
house he has produced being pleasing to some, harmful to
others, beneficial to others, and different from virtuaUy
everything that is; but the discipUne ofhousebuilding is not
productive of any of these things. I n the same manner, a Ю
geometer does not study what is in this way coincidental to
his figures, nor whether a triangle and a triangle possess­
ing two right angles are different. This result is reasonable,
for the coincidental is like a mere name. Hence Plato was
in a way not wrong to classify sophistic as dealing with 15
what is not. For the sophists' arguments are concerned, one
might almost say, more than anything with the coinci­
dental: whether artistic and literate, and artistic Coriscus
and Coriscus, are different or the same thing; and whether
everything that is, but not always,has come to be, so that i f
someone, being artistic, has come to be literate, he has also,
being literate, [come to be] artistic—with aU the other argu- 20
ments of that kind. For what is coincidental is obviously
close to what is not, as is plain also from arguments such as
this: that with things-that-are in another sense there is [a
process of] coming to be and destruction, but with things
[that are] coincidentaUy there is not.
1026^24. We ought nevertheless further to state, as far as
possible, the nature of the coincidental and the cause why it 25
is; for at the same time it wiU doubtless also be plain why
no discipline deals with i t . Since, then, among the things-
that-are some are in the same state always and of necessity
(not necessity i n the sense ofcompulsion but what we call so
69
 югбЬ METAPHYSICS E2

from the impossibility of being otherwise), others not of

necessity or always but for the most part, this is the origin
and this the cause of the existence of the coincidental: for
what is neither always nor for the most part, that we assert
to be coincidental; as for instance i f there is cold stormy
weather i n the dog days, we assert that that is a coincidence,
but not i f there is stiiUng heat, because the one is always or
for the most part, the other not; and i t is a coincidence that
a man is pale, for that is neither always nor for the most part,
but he is not an animal coincidentaUy; and i t is coincidental
that a housebuilder heals somebody, because i t is character­
istic of a doctor, not a housebuilder, to do that, but i t was a
coincidence that the housebuilder was a doctor; and a cook,
aiming to give pleasure, might produce health i n some­
body, but not by virtue of his culinary [art]—hence i t was
a coincidence, we assert, and i n a way he produces i t , but
baldly not.

1027*5. For ofsome ofthem other things are sometimes the

things that produce them;i of others there is no definite art or
capacity. For of things that are or come to be coincidentaUy
the cause is also coincidentaUy.

1027*8. I t follows that since not everything is ofnecessity

and always a thing-that-is or a thing coming to be, most things
10 being so for the most part, it is necessary that there be that
which is coincidentaUy; as for instance someone pale is
neither always nor for the most part artistic, and when this
comes to be he wiU be so coincidentaUy—otherwise every­
thing wiU be of necessity.

1027*13. I t follows that the matter that is capable of

being otherwise than i t is for the most part, is cause of the
coincidental.

15 1027*15. We have to take this as our original [question]:

is there nothing which is neither always nor for the most part?
' R e a d άΛΛαι ivioTC Βυνάμ€ΐς €ujiv ποιητίκαι.

70
E2 TRANSLATION 1027·

or is this impossible? Consequently there are, apart from

these, chance, i.e. coincidental, things.
1027*17. But does it hold good of some things to be for
the most part but ofnone to be always? Or are some things
invariable? These things wiU have to be investigated later.
1027*19. But i t is obvious that no discipUne deals with 20
the coincidental; for every discipUne deals either with that
[which is] always or with that [which is] for the most part.
How else could one learn, or teach another? For a thinghas
to be defined either by that [which is] always or by that
[which is] for the most part: as for instance that for the most
part the fever-patient benefits from honey-water. But the
exception—when he does not, as for instance at new moon— 25
cannot be stated; for that [which is] at new moon is also
either always or for the most part. But the coincidental is an
exception to that.
1027*26. We have stated, then, what the coincidental is
and the cause why it is, and that no discipline deals with i t .

CHAPTER 3

1027*29. I t is obvious that there are origins and causes that

are able to come to be and to be destroyed without [being i n 30
process of] coming to be and being destroyed. For otherwise
everything wiU be of necessity, i f whatever is [in process of]
coming to be and being destroyed necessarily has some cause
non-coincidentally.
1027*32. WiU this be or not? I t wiU i f this comes to be, but
not otherwise; and that i f something else does. And in this
way it is plain that as time is continually subtracted from a 1027^
limited period oftime, we shall come to the present. Thus:
this man wiU die by violence i f he goes out,
and that i f he gets thirsty,
and that i f something else.
71
 io27b METAPHYSIGS E3

I n this way we shall come to what holds good now, or to

something that has come to be. For instance:
. . . i f he gets thirsty,
and that i f he is eating something salty.
5 But this last either holds good or else does not; so ofnecessity
he wiU die or not die. EquaUy, ifonejumps over to what has
come to be, the same argument applies; for t h a t — I mean
what has come to be—^is already a constituent ofsomething.
Consequently, everything that wiU be wiU be of necessity,
e.g. that he who is living dies; for something has already
10 come to be, as for instance opposites i n the same thing. But
whether by disease or violence is not yet [necessary], but
[wiU be] i f this comes to be. I t is consequently plain that i t
runs as far as some origin, but this no further to anything
else; the [origin] ofwhatever may chance wiU therefore be
this, and nothing else is the cause of its coming to be.

1027^14. But what kind oforigin and what kind of cause

15 such a reduction leads to, whether to matter or to what a
thing is for or to what effects a change, needs to be investi­
gated fully.

CHAPTER 4

1027^17. So much for that which is coincidentaUy; it has

been sufficiently distinguished. That which is as true and
that which is not as falsehood are concerned with composition
20 and division and, taken together, with the apportionment
of a contradiction. Fdr truth has the affirmation in the case
ofwhat is compounded and the denial in the case of what is
divided, while a falsehood has the contradictory of this
apportionment. (How we come to conceive things together
or separately is another question—by together and separ­
ately I mean not in succession but so as to make up some
one thing.)
72
E4 TRANSLATION 1027b

1027^25. For falsehood and truth are not in actual things 25

(the good, for example, being true and the bad eo ipso a
falsehood), but in thought; though i n the case ofsimples, i.e.
what things are, not in thought either. What needs study
with regard to that which is and is not in this way wiU have
to be investigated later. But since the combination and the 30
division are in thought, not in actual things, and that which
is in this way is a different thing-that-is from those which are
in the fundamental way (for the thought connects or divides
either what a thing is, or how qualified, or ofwhat quantity
or whatever else it may be), we may leave on one side that
[which is] as coincidental and that which is as true. For the
cause ofthe one is indefinite and of the other is a certain*af-
fection ofthought, and both are concerned with the remain- 1028*
ing genus of thing-that-is and do not indicate the existence
of any extra nature of thing-that-is.
1028*2. So we may leave them aside, and investigate the
causesand origins ofthat which is itself, qua thing-that-is.
1028*4. (In our chapters distinguishing the number ofways 5
in which various things are caUed what they are, it was ob­
vious that that which is may be so caUed in several ways.)

73
 NOTES

G E N E R A L

References to Aristotle's writings are to the Metaphysics unless otherwise

stated. T h e commentary i n S i r D a v i d Ross's edition of the Metaphysics is
referred to as *Ross'.
Scholars designate books of the Metaphysics sometimes, as i n these
notes, by Greek letters, sometimes by numbers. Because the second book
is caUed Little A l p h a , Г is book I V , Δ book V , E book V I .
Г a n d E are epitomized i n chapters 3 - 8 of book K, the authorship of
which is disputed.
I use the formula 'x is essentially F ' to m e a n *if^f exists, it follows that
X is F\

METAPHYSICS BOOK GAMMA

I N T R O D U C T O R Y NOTE

We are told that the fourteen books of Metaphysics were brought into
their present arrangement by editors after Aristotle's death. Book A,
which they set at the beginning, describes the a i m of philosophy as the
removal of surprise a n d perplexity by supplying "knowledge oforiginal
causes", a n d assesses the work of Aristotle's predecessors i n that field.
After the short book designated a , B outlines a set o f " p e r p l e x i t i e s " , most
of w h i c h get examined, more or less directly, i n the rest of the treatise.
T t h u s stands, by the traditional ordering, at the start ofAristotle's m a i n
discussion of metaphysics; it announces its subject-matter i n the first
chapter; a n d its argument is hardly more dependent on what has pre­
ceded than on other parts of Aristotle's works.
T h e book faUs into three parts: chapter 1 briefly states the nature of
metaphysics; chapters 2 a n d 3 to 1005^8 defend its status as a single
subject of inquiry a n d i n d i c a t e i t s scope; the rest of the book from
3. 1005**8 consists of a n examination of what are nowadays known as
the principles of non-contradiction a n d excluded middle.

75
 I003*2I METAPHYSICS Г1

CHAPTER 1

T h i s chapter should be read i n conjunction with E i , w h i c h outlines a

classification of " p h i l o s o p h y " , i.e. systematic knowledge.

A few preliminaries. T h e description w h i c h Aristotle gives of the

" d i s c i p l i n e " whose existence he asserts i n his first sentence applies, a n d
would evidently have been applied by h i m , to everything i n his Meta­
physics; it m a y therefore be taken to define *metaphysics' (literally V h a t
follows physics*), the name coined for the treatise by Aristotle's editors.
H i s o w n label is " p r i m a r y philosophy" (e.g. E i . 1026*30).
'Discipline' translates *episteme% traditionally rendered 'knowledge';
*discipline' has been preferred i n the present translation as possessing a
plural (*science' is now too specialized), but the word a n d its cognates are
translated 'knowledge', 'know', etc., at Г 3. 1005b5, 4. 1008^27, 3^>
Δ 7. 1 0 1 7 ^ 3 , J 15.
' T h a t w h i c h is* (sometimes *what is') translates Ч0 on\ a phrase
consisting of the neuter singular of the definite article followed by the
present participle of the verb 'einai*, w h i c h means both 'to be' a n d 'to
exist'. Parallel expressions with a n adjective, e.g. 'the beautiful', occurred
commonly i n Greek without a following noun. ' T h e beautiful' could
mean either 'the (mentioned) beautiful thing' or 'that w h i c h is beautiful';
and the latter, i n addition to its natural meaning, was used even by n o n -
philosophical writers to designate beauty. Perhaps by analogy with this
last usage, Aristotle's Ίο on* has traditionally been translated 'being'
(which i n the singular must be a verbal noun—das Sein—not a substantive
use ofthe participle—das Seiende); but although Greek adjectives could be
used i n the abstract w a y , it is doubtful whether participles could be. T h e
standard G r e e k for 'being', like the G e r m a n , is definite article plus
infinitive, a n d at Z 1. 1028*20-5 Aristotle himselfindicates a distinction
between 'the to-walk' a n d *the walking (thing)', i.e. that w h i c h walks.
'Thing-that-is' translates the participle *оя' without article.
'QMfl' is the literal L a t i n equivalent ofAristotle's ^hei\ 'as' or *in respect
ofbeing' ('in so far as' a t i ^ 4 . 1006*14, 15). Its meaning is well illustrated
a t J 12. 1019*17-18. A s a n alternative to saying e.g. that a doctor heals
'qua doctor', Aristotle often puts it that he heals ' i n his o w n right',
literally 'by h i m s e l f (see Δ 18); the connection is made explicit at
Posterior Analytics I 4. 73^28-9, " ' i n its o w n right' a n d 'qua i t s e l f are the
same t h i n g " . ' I n its o w n right' is opposed to 'coincidentaUy'. C l e a r l y , the
question whether someone h e a k qua doctor wiU not, while the question
whether he heals i n his o w n right or coincidentaUy wUl, depend on the
description under w h i c h he is identified,
p. 217 " C o i n c i d e n t a l " : Aristotle's verb 'sumbebekenai* is the perfect of 'sum-
bainein\ literally 'come together', w h i c h in its other tenses he uses mostly

76
 NOTES 1003*31

with the sense 'turn out', 'result', 'follow*. Sumbebekota are things w h i c h
might have been apart but have come together. T h e traditional trans­
lation 'accident' loses this etymology, l a c b a corresponding verb, a n d
obscures the Greek word's suggestion (deiued though it is by Aristotle)
that hcmg-a-sumbebekosoi is a synmietrical relation. I have therefore
preferred 'coincidental' a n d 'coincide', even at the price ofinventing the
idiom 'B coincides in A' (e.g. Г 4. 1007b3). F o r other discussions of the
word see the notes on Δ 30, E 2, Δ 6, Δ 7.

Does Aristotle conceive the subject-matter of metaphysics as compris- pp. 201-3

ing everything whatever, or only some among the things-that-are? T h i s
ancient dispute turns on the present chapter a n d £ 1. I t would, ofcourse,
be wrong to take the phrase " t h a t w h i c h i s " as designating a single item:
although grammatically singular, it means the same as ' a l l the things-
that-are'. T h e crucial question is whether "qua thing-that-is" restricts this
phrase,orgoeswith t h e v e r b " s t u d i e s " . (1) (i) laE 1. 1026*23-32 Aristotle
appears to maintain, first, that his inquiry сопсегш oiüy "changeless
substance", a n d secondly that that substance constitutes " a particular
genus and one particular n a t u r e " ; a n d the author oiK 7, w h i c h epito­
mizes E I , describes the discipline as concerned with " t h a t w h i c h is
separable a n d changeless" (1064*33). T h e s e twopassages togetherhave
been taken to show, first, that metaphysics is concerned only with a
restricted ' g e n ^ ' of things-that-are a n d , secondly, that the phrase 'qua
thing-that-is' expresses the restriction, (ii) Some have thought that the
phrase is so used at K 7. 1064*28; it is a k o used to express a restriction,
although a different one, at K 3. 1061*8-10. (iii) Г i's description of
metaphysics as seeking the origins a n d extreme causes w h i c h belong to
" a particular n a t u r e " (or 'some nature') i n its o w n right might sinülarly
imply a restricted subject-matter. ( 2 ) O n the other h a n d , (iv) E 1.
1026*23-32 argues that such changeless substance, i f i t exists, is prior to
the other kinds of substance; hence that its study is " p r i m a r y philosophy,
and universal i n this way, because p r i m a r y " (cf. i ^ 3 . 1005*35). O n this
see notes on E 1. (v) I n the present chapter Aristotle contrasts metaphysics
with the special (literally ' i n part') disciplines, w h i c h "select some part
of" that w h i c h is " a n d study what is coincidental concerning t h a t " , (vi)
I t is, i n any case, usually easier to take 'yaa'-phrases adverbially. T o say,
for example, that a doctor has visited some patient qua patient is not to
say anything about the nature of the patient, but of the visit; and where
no suitable verb is expressed, it is often possible to supply one (e.g.
1003^21, " t o study aU the forms [ w h i c h hold good] ofthat which is qua
thing-that-is"). Probably we should conclude that to say that metaphysics
studies that w h i c h is qua thing-that-is is not to say anything about the
nature ofthe things studied by metaphysics, but about the nature ofthe
study.

77
I003*2X METAPHYSICS Г1

If, according to Г i , metaphysics does study everything-that-is, what

is the special nature of that study conveyed by the phrases ''qua thing-
t h a t - i s " a n d **not c o i n c i d e n t a U y " ? E v i d e n t l y , coincidental knowledge of
everything-that-is w o u l d be exhibited by the universal polymath, who
investigates truths about everything but not truths w h i c h are true about
everything, who studies interdepartmentally but does not study inter­
departmental questions. T h e utterly interdepartmental, or subject-
neutral, questions are the province of metaphysics, w h i c h i n this w a y
manages to be comprehensive i n subject-matter without comprehending
aU other disciplines.

T h e r e are three ways of taking the distinction i n Aristotle's first

sentence between " t h a t w h i c h is qua thing-that-is" a n d "those things
that hold good of this i n its o w n r i g h t " , ( i ) I f the former expression
designates the class of changeless a n d separable substances, Aristotle's
first question might be to ask whether things of such a k i n d exist. T h i s
existential question is posed at B 2. 997*34 i n the form *are there non-
perceptible substances a n d , i f so, o f h o w m a n y kinds are they?' a n d , in
the same form, lengthily examined i n MN. I t is doubtful, however,
whether that question c a n i n practice be separated from inquiry as to
*what holds good of' the class of things so isolated. (2) T h e two expres­
sions are equivalent, the latter being intended to explain the former.
(3) Studying that w h i c h is qua thing-that-is has to be understood as inquir­
ing what truths hold good of everything; " t h e things that hold good of
t h i s " are not truths, but all-pervasive concepts such as those listed at Г 2.
1005*1 i - i 8 . T h u s metaphysics seeks, first, truths that hold good ofevery-
thing-that-is, a n d , secondly, truths that hold good of concepts that hold
good of everything-that-is.
A t A 3. 983*25-6 Aristotle asserts that i f we are properly to know a
thing we must be conversant w i t h its " p r i m a r y c a u s e " (or *reason';
Greek h a d the same word for both). I n tracing, through the remainder
of that book, the course of his predecessors' treatment of causes, he a c ­
knowledges that the word *cause' has more than one sense {A 3. 983*26-32
a n d Δ 2 ) : the earliest *philosophers', for instance, were concerned with
the question *what are things made of?' Aristotle's o w n concerns are
(i) to account for the iÄm^j-that-are i n terms of p r i m a r y or basic things,
w h i c h he identifies as subs'tances (the " p r i m a r y object" of metaphysics,
Г 2. 1003^16) a n d i n particular separable a n d changeless substances
{E I . 1026*16), a n d (ii) to give reasons or arguments i n favour of the
truths w h i c h are true about everything, a n d to lead these back to " p r i m a r y "
or " u l t i m a t e " reasons, w h i c h he also c a l k **elements" a n d " o r i g i n s " (or
principles). T h e notes t 0 P 4 . 1005b35 discuss whether he thinks ofthese
principles as reasons w h i c h do not need, or w h i c h cannot get, further
justification.

78
Г2 NOTES 1003*33

CHAPTER 2

1003*33. Aristotle now considers a n objection to metaphysical inquiry

which m a y be stated as follows. T h e existence o f a designating expression,
e.g. a c o m m o n noun, does not guarantee the existence of truths about
just those things w h i c h the expression designates. I t would be absurd,
for instance, to suppose that there is a body of truths about spilb, each of
which says something true of both splinters a n d tumbles; the study of
spills does not make a single system of knowledge, or discipline. I f the
expression 'thing-that-is' were similarly homonymous, there would be no
study whose subject-matter comprised everything that is, but only a set of
studies e a c h dealing w i t h some genus of thing-that-is. I n the words of
K 3. 1060b33-5, " I f that w h i c h is is so caUed homonymously a n d i n
respect of n o t h i r ^ common, it is not under one discipline, for there is
not one genus of such things".
T h e r e is evidence that this objection to the programme of metaphysics
had been invented, or at least previously used, by Aristotle himself
{Eudemian Ethics I 8. 1217^34). Whatever its provenance, Aristotle came
to believe that it could be met. I n this paragraph he argues that, although
the verb 'be' does indeed have more than one seme (things-that-are are
not " c a U e d what they are by virtue o f o n e t h i n g " , 1003^12-13), never­
theless the difIerent senses are coimected (they are " c a U e d w h a t they are
with reference to one n a t u r e " , 1003^13-14).
T h e first sentence concedes, i n effect, that the verb 'be' has more than
one sense. ( I n *33 it is wrong to translate 'the words " t h i n g - t h a t - i s " . .
because, although the sentence begim w i t h a neuter 'the', w h i c h is
Aristotle's only device for indicating that he is speaking of words rather
than things, such a 'the' does not, of course, always have that force. I n
many occurrences ekewhere of the сопмпоп phrase 'is caUed i n m a n y
ways' the context makes it clear that things, not words, are the subject;
see K 3. 1060b31-4, Δ I . 1012^34 a n d often i n Δ.) I n support of this
thesis Aristotle cites i n the fact that we do not always have the same
reason for describing something as a thing-that-is: it m a y be so caUed
Ъесаше it is a substance, or Ьесаше it is a n affection, a n d so on. H e is not
making the absurd c l a i m that whenever two existential statements are
asserted on different grounds they xaust be understood i n different senses.
T h e kind of ground he has i n m i n d is that w h i c h gives a n account or
explanation o f w h a t it is for something to exist. H e is right to imply that
if one were asked, for instance, w h a t it is for a m a n to exist a n d for the
man's courage to exist, the answers w o u l d be of very different sorts. B u t
is even that enough to justify the thesis that 'exist' has different semes i n
the two cases? I n the absence of a clear criterion for differences of seme
the question is not easy to a m w e r ; a n d even i f it were amwered, w e
should have to allow that Aristotle's o w n expression, here tramlated " i n

79
ioo3*33 METAPHYSIGS Г2

several w a y s " , cannot be held to a n y rigid interpretation (see notes on

Δ 7. 1017*22). Nevertheless, it is fair to ask whether any interpretation
under w h i c h it is true to say that *exist' has m a n y senses or uses or criteria
would be strong enough to support the objector's contention that no
discipline c a n have aU existing things as its subject-matter. O n e such
interpretation is suggested by a n analogy between *thing-that-is' and
•good' w h i c h Aristotle draws at Nicomachean Ethics I 6. 1096*29-34,
where he argues against Platonists that different kinds of knowledge or
expertise are required to discover e.g. w h a t is good i n w a r a n d w h a t is
good i n medicine: there could not be such a thing as a general knowledge
o f w h a t is good w h i c h would short-cut these particular studies. B u t it is a
mistake to suppose that this analogy raises a difficulty i n the w a y of the
inquiry examined i n Metaphysics Г, T h e question about existence w h i c h
parallels the illicit Platonic question *what is good?' is 'what exists?'; a n d ,
while it is true to say that there is no single discipline that includes
answers to such questions as *do electrons exist?' a n d 'does conscience
exist?', metaphysics, according to Aristotle's conception of it, makes no
c l a i m to be such a discipline. I t seems, then, that Aristotle was troubled
by a needless anxiety. N 0 danger lurks i n his concession that the senses
of *be' are multiple, i f that concession is based on the analogy between
*thing-that-is' a n d 'good'.
T h e multiple senses of the w o r d 'healthy' (as i n *healthy exercise',
*healthy climate', 'healthy complexion', 'healthy person') are connected
around the focus (^6 " o r i g i n " ) health. I n the case of the word 'be' the
focal concept is, according to Aristotle, substance. I t is not appropriate to
examine here Aristotle's views about substance, but the following points
m a y be made, (i) H e holds that a statement of the form 'x is a thing-that-
is' (in effect, 'x exists') c a n always be analysed into the form 'x is a n F ' ,
e.g. 'x is a quality*, (ii) H e holds further that the expression replacing ' F '
wiU always imply some reference to a substance, e.g. if x is a quality, x is a
quaUty o f s o m e substance {Z 1. 1028*35-6). (iii) I n several other places
Aristotle telk ш that the senses of 'thing-that-is' are " a s m a n y as the
categories" (see e.g. Δ 7. 1017*22), but of the senses given here only
*substance' a n d *quality o f a substance' introduce items ekewhere treated
as categories. F o r other dkcussions ofconnected senses s e e J 6. i o i 6 ^ 6 - i 1,
Z 4. 1030*32-^3 (*medical'), Θ 1. 1046*4-11, Nicomachean Ethics I 6.
1096*23-9 ('be'), 1096b26-9, EudemianEthics I 8. 1217^26-34 ('be'), V I I
2. 1236*16^32 ('medical'), 1236^17-27, Topics I 15. 106^29-107*2
(*healthy'), Categories 6. 5*38-^10, De Generatione et Corruptione I 6.
322^29-33.
A problem arkes about the inclusion i n ^9-10 of denials among the
things-that-are. Arktotle has i n naind not negative statements (e.g. that
tea is not alcoholic) but negative states of affairs (e.g. tea's not being
alcohoUc). F o r the dktinction between denials a n d lacks (or privations)

80
Г2 NOTES ioo3"33

see Γ 2. i o o 4 * i o - i 6 a n d note. I n the parenthesis, Aristotle asserts that

the existence of denials explains w h y we c a n say that what is not is a
thing that is not. H e r e lurks the false assumption that i n a sentence of the
form 'x is Ф' the function ofthe copulative 'is* must always be to assert the
existence ofsomething. T h i s misconception—tempting because ofthe dual
use of the Greek 'einaV as copulative a n d existential—^had probably
beguiled certain of Aristotle's predecessors into holding that 'x is φ'
entails 'x exists'. Aristotle evidently realizes that the assumption cannot
be correct, e.g. w h e n 'φ* has the value *non-existent'. T h u s the form of
words i n his parenthesis is not intended to cover examples like 4ea is a
thing-that-is-not-alcoholic', i n w h i c h it would still be possible to explain
the 'is' as asserting the existence o f t e a ; it covers examples like 'tea is a
thing-that-does-not-exist' or 'the alcoholic quality of tea is a thing-that-
does-not-exist'. Aristotle's solution to the supposed difficulty is that 'x is
non-existent' entails the existence, not of x, but of a more abstract
entity, the denial of x's existence; i n the first part of the sentence,
therefore, " d e n i a l s o f a substance" (etc.) must be denials ofthe existence
of some substance (etc.), not denials that the substance (etc.) has some
property. So far the solution, though needless, is innocuous. But it
cannot be combined w i t h the thesis that i n a *denial of^:' the status of
the denial as a thing-that-is depends on the status οϊχ as a thing-that-is;
for X is asserted to be a thing-that-is-not. A thing's existence could not be
prior i n that w a y to the existence of the denial of its existence.

1003^11. T h e analogy with health does not seem adequate to A r i s ­

totle's purpose of vindicating the possibility of metaphysics. T h e doctor
can tell w h a t is healthy because he knows about health, and every
healthy thing is somehow connected to health. B u t the metaphysician,
even though he knows about substance, a n d though every existing thing
is somehow connected to substance, cannot pronounce on the question
' W h a t exists?' w h i c h is a j o b for m a n y different specialists. Nor does he
need to, for then his study of that w h i c h is w o u l d be " c o i n c i d e n t a l " .
Aristotle's programme for metaphysics is a coherent one, but this argu­
ment does not show it to be so.

1003^16. Aristotle now intimates for the first time i n Г that substances
are prior to the other things-that-are. I n Z 1 he wiU distinguish three
ways i n w h i c h substance is p r i m a r y — i n formula, knowledge, a n d time.
O n l y the first is mentioned here: i n the words o f Z 1. 1028*35-6, " i n
each thing's formula the formula of a substance is necessarily a con­
stituent". T o " h a v e the principles [or origins] a n d causes of substances"
is perhaps to know w h a t substance is, the question discussed i n ZH\
perhaps to know w h i c h among the substances are basic a n d primary (see
Г 3 . 1005*35).

81
ioo3*i9 METAPHYSICS Г2

1003^19. T h e fact that things-that-are divide into several *Torms"

{eide) the traditional translations are *form' a n d 'species') does not debar
them from being studied by a single discipline, thinks Aristotle, provided
that the forms belong to one genus; but such a discipline is only generi­
cally single, that is, metaphysics has parts. Aristotle here treats things-
that-are as constituting one genus ( c f Г 2. 1004^22); elsewhere he
repudiates that description {B 3 . 9 9 8 ^ 2 2 - 2 7 , PosteriorAnalytics I I 7 . 9 2 ^ 1 4 ) .
A different translation of the second sentence is possible: *Hence it also
faUs to generically one discipline to study aU the forms of that w h i c h is
qua thing-that-is [sc. together], a n d [falls] to the forms [ofthat discipline
to study severally] the forms [ o f t h a t w h i c h i s ] . '

1003^2a. T h i s paragraph, with w h i c h compare / 2. 1054*13-19, inter­

rupts the r u n of argument. T h e preceding division of things-that-are into
forms has evidently suggested to Aristotle or his editor the insertion of a
section designed to show that things-that-are-one divide i n the same way.
T h e first sentence proposes tentatively that " t h a t w h i c h is a n d that w h i c h
is one are the same t h i n g " , a n d after argument the same c l a i m is repeated
in the third sentence. Its meaning is explained i n the first sentence:
*existent' a n d *one' have the same truth-conditions ( " e a c h follows from
the other") even ifthey are not synonymous ("indicated by one f o r m u l a " ) .
I s this so? I t might be objected that although e.g. *Lysistrata is a w o m a n '
and 'Lysistrata is one w o m a n ' are true, *Lysistrata is a n existent w o m a n '
is false. H a d Aristotle noticed the objection, he could plausibly have
replied that i f Lysistrata is one w o m a n she must exist i n some sense or
m a n n e r — i n a p l a y — s i n c e the absolutely non-existent is uncountable
(cf. W . V . Q u i n e , O n W h a t T h e r e I s ' , 4 ) .
T h e second sentence offers support for the first, i n two stages: the
addition of *existent' ( " t h a t i s " ) to a certain expression adds nothing to
what it says; the same is true " i n the case ofthat w h i c h is o n e " . S o m u c h
is clear, but the details of Aristotle's argument are obscure, p a r d y be­
cause of corruption i n the text. I n b g 8 one M S . has *he is one m a n and
he is a m a n ' , others *the m a n is both a m a n a n d one m a n ' ; but the
example must be intended to show the redundancy of 4hat is', w h i c h
Jaeger's emendation, adopted i n the translation, rightly inserts. G i v e n
that emendation, the w o r d 'is' is reduplicated within the second sentence
cited a n d also between the first sentence cited a n d the second; but (i) no
reduplication of the former k i n d need occur if *one' is substituted for
*that is', a n d (ii) the latter reduplication does nothing to show that the
non-reduplicated phrase *that is' is redundant. Ross's emendation differs
from Jaeger's i n omitting the two occurrences of Ъе i s ' ; the two occur­
rences of *one' are also superfluous.
T h e fourth sentence is obscure. " E a c h thing's substance is . . . just
what a certain kind ofthing-that-is i s " : i.e. substances are identical with

82
Г2 NOTES 1003b33

the class ofentities whose substances they are (see f 4 . 1007*20). " O n e
non-coincidentally": the 'parts' ofe.g. paL· Callias are separable, but those
o f C a l l i a s ' substance, viz. two-footed a n d animal, a r e indissolubly one; see
Δ 6. 1015b16. T h e argument seen« to b e : substances are indissolubly
one; substances are to be identified w i t h certain things-that-are; so
certain things-that-are are indissolubly one. I t is not clear how these facts
advance the argument of the paragraph as a whole.
The same a n d similar are not forms of that w h i c h is i n t h e w a y that man
is a form (species) of animal: only some animals are men, but everything
that is, or is one, must be the same (as something) a n d sinülar (to some­
thing)—it is, i n fact, just this feature, universal applicabiHty, w h i c h
makes the study of such concepts a p a r t of metaphysics. Aristotle's
meaning is indicated b y J 15. 1021*9-12: 'equal', 'similar', and ' t h e
same' are defined i n terms of 'one'.
A book on contraries, w h i c h m a y be the 'Selection of Contraries', is
mentioned i n some ancient lists of Aristotle's writings (see introductory
note οτιΔ); according to the commentary o f A l e x a n d e r ofAphrodisias
{floruit A . D . 200) the selection was made from the second book of a lost
work by Plato On the Good.

1004*2. According to the alternative trarwlation of 1003^21-2, each of

the " f o r n M " into w h i c h that-which-is divides is studied by a " f o r m " of
metaphysics. T h e " p a r t s of philosophy" mentioned here mmi be differ­
ent, because the study of a particular k i n d of substance is not, or not
always, metaphysics but m a y be e.g. zoology or geometry or ethics.
Here, therefore, " p h i l o s o p h y " т е а ш 'knowledge'; cf. the "theoretical
philosophies" οΐΕ i . 1026*18-19.
Although branches of mathematics m a y be arranged i n a series,
Aristotle does not explain how the same c a n be done for " p h i l o s o p h y " as
a whole, because he fails to tell us how substances c a n be ordered i n
degrees of priority.

1004*9. Aristotle now uses the thesis argued i n the paragraph begin­
ning 1003^22—that metaphysics studies the concept o f u n i t y — i n order
to show that otherness a n d several related concepts also come within its
scope. T w o arguments are r u n together: (i) plurality is opposed to unity,
and otherness etc. are forms o f p l u r a l i t y ; (ii) othemess etc. are opposed
to sameness etc. T h a t otherness etc. are forms of plurality is not stated,
but c a n be inferred from considerations similar to those m e d i n Δ
15. 1021*9 to show that sameness is a form o f u n i t y : e.g. x,y, z, etc. are
other i f a n d only i f their substances are more than one. "Opposites fall
to be studied by one discipline" because e.g. the study of the conditions
under w h i c h x a n d y are not the same but other is the study of the
conditiom under w h i c h x a n d j i are the s a m e — " t h e one thing of w h i c h
they are the denial or lack is studied i n both cases".

83
ioo4"9 METAPHYSICS Г2

A s Alexander says, the difference between a denial a n d a lack (or

privation) is indicated atPriorAnalßtcs 1 46. 51^25-34: " T o be not-equal
a n d not to be equal are not [the same] either; for one of them, that
w h i c h is not-equal, has a certain subject, a n d this is the unequal, but the
other has none. T h a t is w h y not everything is equal or unequal, but
everything is equal or not equal. . . . I t is therefore obvious that *it is
not-good' is not the denial of *it is good', [but] . . . must be some k i n d of
affirmation." ( T h e distinctions are indicated i n Greek by word-order,
not hyphens.) T h e reason w h y a thing m a y be neither equal nor not-
equal is, as Alexander says, that both of these predicates, unlike *not
equal', imply *capable of being e q u a l ' ; a n d they imply *capable of
being equal' Ьесаше they imply 'quantity'. Q u a n t i t y is thus the " c e r ­
tain genus" («13) a n d the " c e r t a i n n a t u r e " ( * i 6 ) that " c o m e s i n as the
subject of w h i c h the lack is s t a t e d " . SeeZl 22, w h i c h makes it clear that i n
the statement o f a lack this " s u b j e c t " is not always actually mentioned:
hence it might be better to translate *12-13 'either we mean baldly that
that thing does not hold good, or o f a certain genus' ( c f Δ 9. 1018*4).
" I n the latter case, then, some differentia is added to the one, apart
from what is there i n the d e n i a l " : i f the M S . text of this sentence (re­
tained i n the translation, but marked corrupt b y J a e g e r a n d emended by
Ross) m a y stand, its sense wiU be: is not one' expresses a denial, but 'x
is not one F% adding a " d i f f e r e n t i a " , expresses a lack. Against this
interpretation, (i) i t i s h a r d for *differentia' to have the same sense as
*genus'in the preceding sentence, a n d (ii) *not o n e F ' a n d 'not one' fail to
m a r k the distinction (ifit exists) between what merely is not, a n d what is
not capable of being, one. Several other interpretations, none of them
easy, are discussed by Ross, whose o w n text omits " t o the o n e " . T h e
M S S . have 4he differentia', w h i c h perhaps needs emendation (with
Ross) i n order to secure the sense translated.
" T h e one is so caUed i n several w a y s " : s e e J 6.
1004^25. T h e distinction a n d classification o f " w a y s " or senses of 'be',
*one', *the same', a n d other subject-neutral words is undertaken mainly
ΊηΔ a n d 7.
1004"31. Cf.E2. 1 0 2 6 b 2 - 2 4 , E I . 1026*23-32. " A n d ofsubstance they
have no comprehension": they omit that part of metaphysics w h i c h
studies the priority of substance over other things-that-are.
1004^17. ' D i a l e c t i c ' c a n be the general name of the activity of debat­
ing face to face w i t h a n opponent, sometimes, as i n Plato's dialogues, i n
the presence of a n audience a n d sometimes apparently, as i n the second
part of Plato's Parmenides, according to set rules. S u c h debates might
serve a number of different purposes: they might be (i) inquisitorial or
deflationary, as i n Socratic 'refutations', (ii) sporting contests, (iii)
'moots', to practise students i n the art of debate, (iv) aimed at the

84
Г2 NOTES 1004b17

discovery of truth. Aristotle's word for (i) is 'peirastic', w h i c h I have trans­

lated " p r o b e s " , ( i ) , (ii), a n d (iv) are distinguished from dialectic at
De Sophisticis Elenchis 2. 165*38-b^^ but (i) is identified with it i n the
presentpassageandatJ3e5o/)A!ii!cuE'/ei!<:Au 11. 171^9, 172*36. ( i i ) , ( i i i ) ,
and (iv) are aU included under dialectic at Topics I 2. 101*25-8, (i) and
(iv) at Topics V I I I 5. 159^25-37. Sophistic, otherwise eristic {De Sophisticis
Elenchis 11. 171^6-12), seems to cover (ii) and perhaps (iii), but is always
opposed to dialectic by Aristotle as containing " w h a t are imagined to be
reasonings" and " w h a t are imagined to be received opinions"—either i n
the sense that the premisses and arguments of sophistic are always false
and fallacious {E 2. 1026^14-15), or that the sophistical purpose is
indifferent to their truth. Sophists are distinguished by their purpose
("the life c h o s e n " ) , dialecticians by " t h e type of c a p a c i t y " , i.e. their
lack of ability to pursue the higher a i m of truth (cf. Rhetoric I 1.
1355b17-21). Philosophy sticks, ofcourse, to purpose (iv). Since A r i s ­
totle's comparison is designed to indicate the wide scope of metaphysics,
" p h i l o s o p h y " i n this paragraph has to mean 'first philosophy'.

1004^27. Aristotle now offers a new argument to show that " i t faUs to
one discipline to study that w h i c h is qm thing-that-is":
(a) aU things either are or are composed out of contraries;
{b) contraries originate i n the one and plurality;
{c) plurality and the one, being opposites, are studied by the same
discipline;
{d) since the senses of 'one', though many, are connected, what is one
m a y be studied by a single discipline;
(e) therefore aU things m a y be studied by a single discipline.
{b), {c), a n d {d) are d r a w n from the paragraph beginning 1004*9.
Aristotle holds that, among things-that-are, substances and quantities, at
least, are not themselves contraries {Categories 5. 3^24-32 and 6. 5^11-
29); and at Л 10. 1075*28-34 and N 1. 1087*29-^4 he criticizes the
opinion, held by " p r a c t i c a l l y everyone" (^29) among his predecessors,
that the physical world is composed out of contraries. H i s argument here
is therefore apparently ad hominem.
T h e parenthesis at * 8 - i 1 it not easy. (A) Nicomachean Ethics I 6.
1096*23-8 asserts that the good cannot be " a n y t h i n g universally common
and o n e " on the ground that it " m a y be so caUed i n as m a n y ways as that
which is". D e n y i n g universality seems, then, to be equivalent to asserting
multiplicity ofsenses. (B) " S e p a r a b l e " cannot have Aristotle's technical
sense, according to w h i c h substances, but not e.g. affections or matter,
areseparable(PAyjiciI 2. i8^^3i,DeGenerationeetCorruptionel 10. 327^22,
I I I . 329*25): the meaning must be 'having unconnected senses'. T h u s
the parenthesis repeats w h a t was said i n the previous sentence, and is
doubtless a n intrmion. y a n d z are " r e l a t e d to one t h i n g " x w h e n there

85
ioo4b27 METAPHYSIGS Г2

are relations R a n d S such that Ryx a n d &if; they " f o r m a succession"

w h e n Ryz a n d Асдс.
" E x c e p t on the basis of a hypothesis": Ross follows Alexander i n
interpreting *but they proceed on the basis of a n assumed answer (sc.
to the question what is the contrary e t c . ) ' ; cf. Plato, Republic V I 510 c d .

i o o 5 * i 3 . A l l the metaphysical concepts mentioned here a n d earlier in

the chapter are treated i n MetaphysicsA or / .

CHAPTER 3

1005*19. T h i s paragraph answers the question raised a t 5 2. 996^26-33:

" O n the other h a n d , with regard to the demonstrative principles it is
disputable whether they fall to one discipline or more than one. B y
*demonstrative' I m e a n the common opinions on the basis of w h i c h aU
m e n demonstrate; as for instance that it is necessary that everything be
either asserted or denied, a n d that it is impossible simultaneously to be
a n d not to be, a n d aU other propositions ofthat k i n d . Does one discipline
deal with these a n d with substance, or a different one? A n d i f t h e latter,
w h i c h ought to be identified with the discipline we are now s e e k i n g ? "
Aristotle indicates that the word *axiom' (perhaps meaning *requisite') is
taken from mathematics, a n d the parallel passage atuC4. 1061^19 refers
to the " m a t h e m a t i c i a n s ' p r i n c i p l e " , " c o m m o n to aU quantities", that
" w h e n equals are subtracted from equals equals r e m a i n " (cf. Posterior
Armlytics I 10. 76*41, b20, 11. 77*30), w h i c h is the third of E u c l i d ' s
*conmion notions' (see H e a t h , Mathematics in Aristotle, 2 0 1 - 3 ) . I n the
remainder of Г Aristotle wiU discuss only the two absolutely " c o m m o n "
axioms mentioned i n B, the principles of non-contradiction a n d ex­
cluded middle (henceforward P N C a n d P E M ) . T o these two at least he
nught apply the definition of *axiom* given at Posterior Armlytics 1 2.
72*14-18: they are "impossible to s h o w " ( c f . r * 4 . 1006*5-12) a n d " i t is
necessary for anyone who is going to learn anything to h a v e " them (cf.
r * 3 . 1005^16-17).
I n B 2 Aristotle h a d raised two objections to including axioms under
the same discipline as each other and as substance, (i) T h e y are used i n
aU disciplines a n d so cannot be the peculiar study of any one (996^33-
997*2). Aristotle now answers that though "everyone uses t h e m " they
are not studied by everyone ( * 2 9 - 3 0 ) : it is inappropriate for a geometer
to investigate the credentials of P N C because it holds good ofhis subject-
matter, Unes, etc., ''qua things-that-are", a n d so holds good of every
thing-that-is, whether geometrical or not. (ii) T h e axioms cannot be
studied by any discipline; for they do not need to be defiruid a n d , i f they
a r c to be demonstrated by a discipline, everything eke demonstrated by
means o f t h e m (and that k everything whatever) wiU fall under the same
dkcipline, w h i c h k absurd (997*2-11). T h k objection is not answered

86
гз NOTES 1005*19

explicitly i n Γ*. Perhaps Aristotle abandoned its odd assumption that i f C

is demonstrated from B a n d B from Л, A, B, a n d C aU belong to the same
field o f s t u d y ; or perhaps, as Ross suggests, the absurdity is thought not
to arise i f axioms are, strictly speaking, indemonstrable.
O n the errors ofthe students of nature, or *physicists', cf.£^ 1. 1025^18-
21, 1026*27-30, Л 6 - i o , Plato, Sophist 2 4 6 - 7 ; i n one sense *nature* covei^
everything ( c f J 4. 1015*11-13), i n another not ( J 4. i o i 5 " i 3 - i 5 ) .
1005^2. T h e question w h i c h Aristotle dismisses as belonging to
analytics (i.e. logic) m a y concern ( i ) axioms or (2) (Ross) propositions i n
general. I n either case he seems to have i n m i n d especially the fact that
one cannot require every proposition to be demonstrated from another
before it is accepted as true (see notes o n Г 4. 1005b35). Alexander
suggests plausibly that this paragraph ought to follow the next. T h e
argument would then be: metaphysics studies (e.g. tries to overcome
objections to) the principles ofreasoning; analytics shows that principles
are necessary to reasoning. T h e r e is stiU some warrant for this division of
labour.
1005^5. " W h a t is characteristic of aU substance": literally ' a U sub­
stance as it n a t u r a l l y - i s ' ; Aristotle uses a verb cognate w i t h *phusis\
" P r i n c i p l e s of trains of r e a s o n i n g " : literally *syllogistical principles',
but earlier writers, a n d Aristotle himself i n the Topics, used 'sullogismos*
('co-reckoning') a n d its cognates roughly as we use 'reasoning'. E v e n the
definition i n the PriorAnalytics ( I . 1 24^18-20) allows the word to apply
to a n y deductive argument from more than one premiss. So it is best to
avoid the now technical term *syllogistic' (at Δ 18. 1022*22 the trans­
lation is 'calculation').
1005^8. H a v i n g answered the question set at the beginning of the
chapter, Aristotle now leaves the examination of the existence a n d scope
of metaphysics w h i c h has filled the first part of Г. T h e rest of the book
consists of a n exercise i n metaphysics: what are the (chief) axioms, a n d
how c a n their truth be vindicated against challenge? P N C occupies h i m
to the end of chapter 6 ; P E M is m u c h more briefly treated i n Г 7.
H e begins b y asserting that the metaphysician must be able " t o state
the firmest principles of e v e r y t h i n g " , a n d assumes that among these is
a principle "firmest of a U " . I n the second sentence—let us refer to it as
(fl)—the Greek might m e a n either
{ai) ιΐχ is the firmest principle of aU, error with regard to x is i m ­
possible
or
{a2) i f error w i t h regard to x is impossible, x is the firmest principle
ofaU
('x' being universally quantified).Two things support (02). (i) 1005^22-34
wiU argue that P N C is the firmest principle of aU ол the ground that its

87
ioo5^8 METAPHYSICS ГЗ

contradictory is believed by no one, i.e. error w i t h regard to it is i m ­

possible. T h i s p r o o f i s validated by (a2) but not by ( a i ) . (ii) T h e proof
of {a) itselfis by means o f t w o lemmata stated i n the third sentence:
{b) if error with regard to x is impossible, x is most intelligible;
{c) if error with regard to x is impossible, x is non-hypothetical.
( T h e final sentence of the paragraph shows that i n ^13 " o f that k i n d "
means *with regard to w h i c h error is impossible'.) These lemmata w i l l
serve i n a n argument for {a2) but not for {ai).
I n the rest of the third sentence the two further premisses w h i c h
connect {b) a n d {c) with {a) are not stated, but arguments i n support of
{b) and {c) are offered successively, {b) is supported by
{d) anyone who does not understand x is liable to error with regard
to X.
(c) is to be taken as supported by
{e) if error with regard to x is impossible, x is understood by anyone
who understands anything;
(/) if the latter, x is necessarily i n that person's equipment;
{g) if the latter, x is not a hypothesis.
(e) is not stated; it appears to assume that where error is impossible so
are ignorance a n d confusion. I n ( / ) *necessarily' has to govern the
consequent (cf. Г 4. 1006^28-30, 5. 1010^25), it is needed i n the
antecedent of {g), *Intelligible' translates 'gnörimos' ('certain' at Г 4.
1008*17); 'understand' a n d 'have understanding' translate the cognate
verb 'gnorizein% the basic meaning of w h i c h is 'make intelligible' {Г
2. 1004*20, 23, ^26, 3. 1005*28)—to oneselfor others; Aristotle's state­
ment o f ( ^ ) substitutes a synonymous verb 'xunienaV ('apprehend').
F o r 'hypothesis' ^^^PosteriorAnalytics I 10. 76^23-34: a hypothesis is
provable but accepted without proof; a postulate is disbelieved but
assumed for the sake of argument; both are different from a principle
( = axiom) w h i c h " i s necessarily so on its o w n a c c o u n t a n d i s n e c e s s a r i l y
believed".

1005^18. " D i a l e c t i c a l difficulties": Plato remarks that a spinning top

c a n be moving a n d not moving i n different parts (the r i m moving, the
axis at rest. Republic 436 d e ) , a n d accordingly formulates a principle of
contrariety thus: " t h e same*thing will not be willing simultaneously to do
or suffer contraries at least i n the same respect and relative to the same
t h i n g " ( 4 3 6 b 8 - 9 ) . C o m p a r e De SophisticisElenchis 5. 167*23-7, where
Aristotle proposes the guarding formula " o f one a n d the same thing,
actual thing not name. . . . i n the same respect, relative to the same thing,
i n the same w a y a n d i n the same period of t i m e " ; also De Interpretatione
6· 17*33-7· T h e present formulation of P N G covers only unquantified

88
гз NOTES 1005b18

subject-predicate propositions and does not rule out the joint truth of
complex propositions such as Ъе is either upstairs or downstairs or i n m y
lady's chamber' a n d Ъе is neither upstairs nor downstairs nor i n m y
lady's c h a m b e r ' ; for w h i c h we need the more general modern formula
'it is impossible that p a n d not-/>'. Sometimes, indeed (e.g. B 2. 996^30,
Г 4. i o o 6 * i ) , Aristotle employs the phrase "for the same thing to be a n d
not to b e " , i n w h i c h 'to be' might mean 'to be the case' (cf. Г 5. 1009*7);
but even that phrase may be elliptical for the schema 'to be so-and-so',
where 'so-and-so' marks the place for a predicate-expression (cf.
1006^18-20). O t h e r formulations ofthe principle are quite freely used in
some o f t h e arguments w h i c h follow (e.g. 1005^29 w h i c h introduces the
word 'contradiction', 1005^23-4 where the modal operator is omitted,
Г 4 . 1007^18-19, 1008*36).
1005^22. Aristotle now seeks to show that P N C "fits the specification"
of firmest principle by being a principle about which error is impossible.
T h e opinion he cites about Heraclitus (not relied on by modern scholars)
is not
{a) ^x^F (Heraclitus said that it is possible that some people be­
lieve that Fx & ^Fx)
but
{b) 3 x 3 F (Heraclitus said that ^л: & -Fx).
Aristotle points out that we cannot infer from {b) to
{c) ^x^F (Heraclitus believed that Fx & -^л:)
nor therefore to
{d) ^x^F (it is possible that some people believe that Fx & ^Fx)
which is entailed by {c). H e then gives a proofofthe contradictory of {d)y
as follows.
{e) it is impossible for contraries to hold good of the same thing
(b26-8)
(/) \fx^F (belief that Fx is contrary to belief that not-Fx) (b28-9)
{g) therefore V ^ V ^ (it is impossible for anyone to believe that Fx and
believe that not-Fx) (b29-30).
Г б . ι ο ι ι * Ί 5 - 2 2 wiU argue that {e), here introduced with an 'if', is
entailed by P N C itself; but Aristotle nowhere defends the other and more
dubious premiss ( / ) {De Interpretatione 14.23*32-^7 seems irrelevant). T h e
rest of the proofis not explicit: from {g) Aristotle seems to infer, plausibly,
(A) V x V F (it is impossible for anyone to believe that Fx & —Fx)
which contradicts {d); thence, fallaciously ignoring the intensionality of
belief,
(г) it is impossible for anyone to beUeve that n o t - P N C
(where ' P N C ' abbreviates 'VjcVF - [Fx & -FxY)\ and so, validly given
'necessarily P N C ' , to
' 89
ioo5**aa METAPHYSICS ГЗ

(j) it is impossible that ( P N G a n d someone believes that n o t - P N C ) a n d

also impossible that ( n o t - P N C a n d someone believes that P N C ) ,
i.c. error about P N C is impossible. (Aristotle eschews the stronger c l a i m ,
implied by Posterior Analytics I i o . 76^24, that P N C must be believed.)
Г4, 1006*4, " b y means ofthis*', shows that he is aware that the whole
proofdepends on P N C as a premiss. H e does not make clear whether {h)
rules out the possibility of believing even veiled contradictions such as
' B a l a a m rode on a n ass but not on a donkey', *Мепе1аш was king of
S p a r t a but not of Lacedaemon*. O n the paragraph see J . Barnes,
Philosophical Quarterly^ October 1969.
I n b32 " t h i s o p i n i o n " is P N C . Does Aristotle m e a n that every a r g u ­
ment relies on P N C (cf. 1005*23-4), or merely that no argument ques­
tions it? O n l y the latter is supported or explained by the thesis that it
cannot be disbelieved.

CHAPTER 4
T h e notes divide this chapter into a n introduction a n d seven arguments
(Ross runs 1006*28-1007b18 together as the first, a n d separates 1008*2-7
as the third). T h e arguments are variable i n quality a n d sometimes d r a w
on the same material. T h o u g h there are some cross-references, it is
possible that they were collected a n d composed over a long period of
time. Part of Aristotle's purpose m a y have been to encourage his pupils
to compare a n d grade them.

J * 4 : Introduction (1005^35-1006*28)
1005^35· Both ( A ) the conclusion a n d (B) the argument of this p a r a ­
graph are unclear. (A) T h e implication of ioo6*i 1 is that what precedes
has shown that demonstration o f P N C is impossible, unless " i n the manner
of a refutation*'; a n d the author of K 5. 1062*2-4 says explicitly " o n
such matters there is no such thing as demonstration, speaking b a l d l y "
(because no " m o r e trustworthy p r i n c i p l e " c a n be found from w h i c h to
demonstrate i t ) . B u t i n the present paragraph itselfwe are told that " a
demonstration ought not to be sought", w h i c h m a y convey that demon­
stration of P N C is merely needless,
(B) Aristotle's argument is from two premisses: {a) " i t is impossible that
there should be demonstration of e v e r y t h i n g " ; {b) i f anything ought to be
left undemonstrated, P N C ought. T w o criticisms m a y be'brought against
the argument. Firstly, {b) is i n need ofsupport; perhaps Aristotle sees the
rest of the chapter as providing it. Secondly, {a) is ambiguous between
(ai) some things cannot be demonstrated
and
(fl2) it cannot be that everything is demonstrated.
Aristotle m a y have confused these (see e.g. his treatment of *all the air is

90
Г4 NOTES 1005b35

breathable' at Topics V 5 . 1 3 5 * 3 2 - b 1 ) , a n d only {ai) could support thc

stronger version of his conclusion. But even {a2) seems strong enough to
justify
[c) i n any given argument, it ought to be that some things are left
undemonstrated,
w h i c h , together with {b), entails
{d) P N C ought always to be left undemonstrated.
( f l 2 ) — b u t not (fli)—is proved at greater length i n I 3 of the Posterior
Analytics, to w h i c h perhaps those who " l a c k t r a i n i n g " are to be referred;
c f r 3 . 1005b3-4.

i o o 6 * i i . I n this cryptic paragraph Aristotle recommends a strategy pp. 203-4

against the opponent of P N C . W h a t , first, does he mean by " d e m o n ­
strating i n the manner o f a refutation"? ( i ) A n y demonstration of P N C
would be a refutation of its contradictory, as the opponent is at this
stage assumed to admit ( 1 0 0 6 * 5 - 6 ) . H e n c e the special method w h i c h
Aristotle recommends is not merely 'demonstrating by refuting'. ( 2 ) K 5
does not mention refuting, but c a l k for demonstration ad homirum
( " i n response to t h k p e r s o n " , 1 0 6 2 * 3 ) . *^^ hominem' means *from a
premiss accepted only by your opponent', some of the later arguments
in Г follow that procedure, but nothing else i n t h k paragraph suggests
it. ( 3 ) K 5 prescribes, again without the authority of Г 4 , " f i n d i n g
something o f s u c h a kind that it shall be the same . . . but not be thought
to be the s a m e " as P N C ( 1 0 6 2 * 6 - 9 ) . ( 4 ) I n demonstration, says Aristotle
i n * 1 6 - 1 7 , the demonstrator " m i g h t be thought to beg the original
question", but i n refutation "someone else is cause of such a t h i n g " .
According to the definition at Prior Armlytics I I 16. 6 4 ^ 3 4 - 8 it k not
possible to beg self-evident propositions, but Arktotle's opponent,
denying that P N C is seU*-evident, nught accuse h i m of demonstrating
from a premiss w h i c h only P N C itself would give one any reason to
accept ( " o f a nature to be shown by means o f " P N C , 6 4 ^ 4 0 ) . T o avoid
the accusation "someone else", sc. the opponent himself, must choose
the premiss—^which k a feature ofrefutations also. ( I f thk k theprocedure
reconunended by Aristotle, it exploits the fact that every proposition—at
least every one w h i c h the opponent k likely to choose—implies P N C . )
Whether or not the fourth sentence (on w h i c h our M S S · agree,
although Alexander knew some variants) further elucidates " d e m o n ­
strating i n the manner of a refutation", it certainly enjoins more than
merely to give the opponent choice of opening premiss. T h e syntax
leaves it doubtful whether he is required ( i ) to signify something (cf.
K 5 . 1 0 6 2 * 1 1 - 1 4 , make himselfunderstood) or ( 2 ) to state something to
(i.e. that something does) signify. I f ( i ) , we might take it that the require­
ment is to be satisfied either {ia) by uttering a single w o r d , e.g. *man',

91
ioo6*ii METAPHYSICS Г4

o r ( i 6 ) by uttering such a w o r d i n answer to a question such as ' I s Callias

a m a n ? ' T h e difficulty w i t h (10) is that ' m a n ' , not being a proposition,
cannot form the premiss of a n argument (see also *27 " h a s agreed that
something is t r u e " ) ; nor d i d Aristotle believe that ' m a n ' implies 'there is
a m a n ' {De Interpretatione 4. 16^28-30). O n the other h a n d , it is possible
that Aristotle took the utterance of ' m a n ' to imply the proposition
' " m a n " has a significance'. If, by {ib), the opponent is to say ' m a n ' i n
answer to a question, the question must be other than one w h i c h " a s k s
h i m to state something either to be or not to b e " . A t Topics 14. 101^29-33
Aristotle distinguishes the form 'is it the case that/> or not?' (a " p r o b l e m " )
from the form 'is it the case that pV (a " p r e n u s s " ) . Conceivably his
intention here is to prescribe premiss-questions as against problem-
questions, perhaps on the ground that the latter exclude the double
answer '/? and not-/>'. B u t at 1007*7-14, where he does imagine the
opponent to have been asked a question to w h i c h ' m a n ' would be a n
answer, that question is of the problem type a n d the double answer to it
is explicitly excluded. (2) T w o later passages, 1006^11-13 a n d Г 8.
1012^5-8, require the opponent to say that some word has significance,
a n d both refer back, apparently to this sentence.
T h e demand i n the first sentence that the opponent shall " s a y some­
t h i n g " m a y be less innocent than it seems. F o r Aristotle's subsequent
gloss on its contradictory is " h a s a statement o f n o t h i n g " , w h i c h , like the
preceding phrase " l o o k for a statement", employs the noun 4ogos' (see
glossary); a n d while i n the first of these occurrences logos is speech or
statement, i n the second 'has a logos ofsomething' c a n readily m e a n 'has
a definition' ( c f . J 29. 1024^26-1025*1, but c o n t r a s t F 2 . 1004*33 where
the same phrase is translated 'discuss'). Furthermore, at *24-5 it is
concluded that i f the opponent does what is asked of h i m " t h e r e wiU
already be something definite". So perhaps Aristotle, playing on this
double meaning of 4ogos\ requires the opponent not only to state that
e.g. *man' has a significance a n d therefore a definition, but also to state
the definition; cLFy. 1012*21-4, andTS. 1012^5-8 where, however, the
definition is of 'true' a n d 'false'.
" B u t the cause . . . " , * 2 5 - 6 : cause o f w h a t ? IfAristotle means that the
opponent is responsible for the " o r i g i n a l s t e p " , i n that he says something,
nothing is added by the words " e l i m i n a t i n g statement". Probably the
meaning is that the opponent is responsible for his own downfall (cf.
* 1 7 - 1 8 ) ; for he engages i n discourse, but says things w h i c h make dis­
course impossible (cf. Sophist 252 bc).
A t the end of the final sentence some M S S . add, duplicating *30, " s o
that it could not be that everything was so-and-so a n d n o t s o - a n d - s o " ;
others omit the whole sentence, w h i c h пшу therefore be spurious, or a
later addition by Aristotle, or misplaced. I f it belongs here, its force m a y
be: ifsomething is true, something is false; so not everything is true.

92
Г4 NOTES ioo6>28

Γ 4: ist Argument (1006*28-31)

1006*28. T h i s sentence has been given various interpretations. ( i ) * T h e

name chosen signifies e.g. " ( t o be) m a n " or " n o t (to be) m a n " but never
both [ " o r " exclusive]; so it is impossible to be both m a n a n d not m a n . '
T h e premiss is feeble and the argument unconvincing. (2) O n e who says
that X is e.g. a m a n signifies that x is, or is not, something i n particular;
so X is not everything whatever (whatever you like).' T h e r e is a moder­
ately good parallel for this interpretation of 'so-and-so a n d not so-and-so'
at Theaetetus 183 a, but the sense is strained and the conclusion not ob­
viously relevant. (3) *The name chosen signifies being or not being
something; a n d that is the starting-point from w h i c h we shall proceed to
prove P N C T h i s gives a weak sense to " s o t h a t " , and ignores the
implication o f " a g a i n " i n 1006*31 that *28-31 offers a separateargument.
(4) *The name chosen signifies to be or not to be something and does not
also not signify that; so at least one predicate, " s i g n i f y " , does not share
its contradictory w i t h any of its subjects.' T h i s , Alexander's inter­
pretation (cf., for what it is worth, K 5. 1062*17-18), attempts to meet
the difficulty that aU the others require 'anything' rather than 'every­
thing' i n the conclusion. B u t that is aU that commends it, a n d i n any
case a strict reading of the conclusion gives 'at least one subject does not
have any contradictory predicates'. (5) Alexander also construes the
premiss differently: 'the name " t o b e " or " n o t to b e " signifies this par­
ticular thing'. T h i s does not seem to help his own interpretation or to
suggest another; and it is inconsistent with De Interpretatione 3. 16^22-5.

Г 4: 2nd Argument, Part I (1006*31-^34)

Before discussing the text of this extraordinarily mystifying argument, it
wiU be useful to distinguish i n general terms the two types of inter­
pretation it has received. According to the first type. Part I of the argu­
ment (down to 1006^34) seeks to show that no predicate whatever m a y
be predicated jointly with its contradictory; Part I I I (1007*20-^18)
argues independently that the same mtist be true ofnecessary or essential
predicates i n particular. According to the second type, both Parts have
the latter, or a similarly limited, a i m . T h e major attraction of type 1
interpretations is that Part I does not represent itself as a merely partial
defence of P N C a n d , although couched i n terms of the word ' m a n ' , gives
no indication that the choice of a n essential predicate-word is crucial.
O n the other h a n d , type 2 interpretations claim to give a better sense to
1006^13-14 and to explain how ^13-28 connects with Ц 8 - 3 4 ; but there
is some doubt about both these claims. E v e n if type 2 is right, there is no
reason to conclude that Aristotle accepted only the restricted version of
P N C ; indeed he often states it without any such restriction.

93
 ioo6*3X METAPHYSICS Г4

pp. 205-6 1006*31. * I n thc same sense' does not occur among the " c u s t o m a r y
specifications" listed i n the statement of P N C at Г 3. 1005*^19-20
(contrast De Sophisticis Elenchis 5. 167*4, De Interpretatione 6. 17*35).
Y e t Aristotle doubtless knew the child's riddle alluded to at Republic V
479 c : a m a n who was not a m a n (a eunuch) threw a stone that was not
a stone (a punuce) at a bird that was not a bird (a bat) on a twig that
was not a twig (a reed). I n this paragraph he explains a n d defends the
assumption that i n 4he same thing cannot be a m a n a n d not a m a n ' ' m a n '
has just one sense (or explication, see below). Although this seems to
be the force of "signify one t h i n g " (an expression not used elsewhere by
Aristotle), it must be admitted that i n other places *signify' {'semainein')
often means something more like 'denote': e.g. ifthe word 'cloak' is used
as a n abbreviation for 'the horse that is a m a n ' it signifies nothing {De
Interpretatione 8. 18*25, but contrast 1. i 6 * i 6 - i 8 a n d Posterior Analytics
I I 7. 9 2 ^ 5 - 8 ) ; 'every' (De Interpretatione 10, 20*13), a n d perhaps 'is'
(ib. 3. 16^24), do not signify but " c o n s i g n i f y " ; " c o n n e c t i v e s " such as
particles a n d prepositions are " n o n - s i g n i f i c a n t " (Poetics 20. 1456^38-
1457*10). I n choosing to express his assumption of univocity i n this
unusual a n d ambiguous w a y Aristotle is no doubt influenced by the fact
that the word 'signify' has appeared i n the argument already; otherwise
he might have written 'let m a n be said i n one w a y ' or 'let m a n be
synonymous' (Categories 1. 1*6-12).
L a c k i n g inverted commas, Aristotle refers to the name ' m a n ' , a
masculine, by prefixing the neuter form of *the'; cf. Δ 7. 1017*31,
Hterally *the is signifies . . .'. T h e definition 'two-footed a n i m a l ' , though
repeated elsewhere, is not i n the present argument required to be correct
or complete. Aristotle's explanation of *signify one thing' is paraphrased
to the following effect by Alexander a n d Ross: ' i f (a) [what *man' signifies]
is that [viz. two-footed a n i m a l ] , ['man' signifies one thing if a n d o n l y i f ]
(b) i n the case of any m a n , (c) to be a m a n wiU be that [viz. to be a two-
footed a n i m a l ] ' , (a) demands, pace Ross, the textual variant w h i c h omits
the w o r d * m a n ' ; the received text must m e a n either ' i f that [word] is
" m a n " ' (understanding a neuter article before ' m a n ' ) , or ' i f that [one
thing w h i c h ' m a n ' signifies] is a m a n ' , or ' i f that [viz. a two-footed
animal] is a m a n ' . (b) emphasizes that 'two-footed a n i m a l ' must apply to
everything to w h i c h *man' applies.
" P r o v i d e d that theseAvere definite": sc. " a definite n u m b e r " (b4) as
against "infinitely m a n y " (Ь6). Aristotle's argument does not i n fact
require that the significations of a name be finitely many, but only that
there be unit significations, like points on a line, not themselves further
divisible. T h i s is shown by the second sentence after the parenthesis, w h i c h
has to m e a n not that ruimes signify just one thing (a doctrine absurd i n
itself a n d contradicted by *34) but, as indicated by the reference to con­
ceiving, that a speaker must signifyjust one thing, sc. each time he utters

94
Г4 NOTES ioo6*3i

a single name significantly. T h u s the argument is: objects of thought

divide into units w h i c h must be pairable w i t h names ( ^ i o - i i ) ; ther^oere,
if there is speech or thought, names are significant ( ^ 8 - 9 ) ; if names are
significant, each use of a name must have unit signification (^7); if each
use has unit signification, each name must have significations that divide
into units (^5-7). I n claiming that every utterance o f a name signifiesjust
one thing, Aristotle overlooks the phenomenon of double entendre. H e does
not imply that what the name signifies must be a simple entity; cf. De
Interpretatione 8. 18*18-27, ^ 4· 1029^27-8.
" T h e r e would be no statement": or, perhaps, *formula'.

1006^13. T h e opening clause ofthis paragraph is the first major crux

i n the argument. T h r e e interpretations seem possible. T h e first depends
on construing *ouk estaV at the beginning of the second sentence, litei:aUy
(as i n this translation) *will not be', i n the equally common sense *will
not be possible'. T h e effect of this is to make the second sentence antici­
pate the conclusion o f P a r t I , w h i c h wiU be stated at 1006^33-4 i n the
next paragraph. But the second sentence begins not with a 4herefore'
but a n *and', as if explaining the opening clause of the first sentence.
T h i s suggests that that clause also anticipates the conclusion of the
argument of Part I , a n d means

(i) Ч0 be a m a n ' cannot be [sc. truly] predicated of anything of

w h i c h *not to be a m a n ' is [truly] predicated,

i.e. *is a m a n ' a n d *is not a m a n ' have incompatible meanings. However,
(1) gives to *signify x* the sense about to be reserved for *signify about x\
I n face of this it seems impossible to sustain ( i ) , a n d consequently
necessary to construe the second sentence as i n the translation: 'it [sc. to
be a m a n ] wUl not be both to be a n d not to be the same thing . . .'. I n
place of ( i ) , the final sentence of the paragraph suggests a second
interpretation of the opening clause:

(2) Ч0 be a m a n ' cannot m e a n just what 'not to be a m a n ' means,

i.e. 'is a m a n ' a n d 'is not a m a n ' have different meanings. But w h y need
that be " s h o w n " (^28)? I f 'is a m a n ' a n d 'is not a m a n ' h a d ihcsame
meaning, the sentence 'Callias is a m a n a n d not a m a n ' would not state
a contradiction, a n d the opponent i n uttering it would say nothing more
controversial than that Callias is a m a n a n d a m a n . T h i s is the argument
presentedbyAristotle himselfin the third sentence, w h e n h e points out that
i n order to express the c l a i m that the same thing c a n "simultaneously
be a n d not be a m a n . . . i n actual f a c t " the opponent must avoid
speaking " h o m o n y m o u s l y " . I t is at this point that champions of type 2

95
ioo6bi3 METAPHYSICS Г4

interpretations come i n . F a c e d w i t h these objections to ( i ) a n d (2), they

propose that "signify*' has a sense distinct from both *bc predicated of'
a n d *mean', v i z . *be explicated b y ' :

(3) Ч0 be a m a n ' cannot be explicated by just that w h i c h explicates

*not to be a m a n ' .

T h e independence of this sense requires explication to be different from

definition by synonym, a n d G . E . M . Anscombe has suggested i n
illшtration of the difference that *is large' a n d *is not large' might both
(in different contexts) be explicated by *is two feet long'. A s the illustra­
tion shows, sense (3) calls for a corresponding adjustment i n the sense
of *signify one t h i n g ' ; for *large' has perhaps only one sense, but
cannot be substituted for *man' i n ^13. According to (3), the argument
ofthe first three sentences wiU go as follows: *is a m a n ' a n d *is not a m a n '
cannot have the same explication . . . i.e. to be a m a n cannot be
[omitting *possible'] both that a n d not to be a m a n , except i n the u n ­
interesting case i n w h i c h *man' a n d *not-man' are used as synonyms.
T h i s part of the type 2 interpretatbn is, as wiU be seen, independent of
the rest.
Some brief remarks follow about the remainder of the first sentence.
I t is doubtful whether *signify about' occurs elsewhere i n Aristotle,
but similar expressions (e.g. Categories 2. 1^20 f f . , J ' 6, i o i i ^ i 6 ) and the
example i n the parenthesis show that it means Ъе truly predicated of'.
* I f " m a n " signifies not only . . .': i f this clause is to be elucidated by the
following parenthesis, its force must be: it is not enough that *man'
should have two or more senses (explications) aU of w h i c h are, on some
occasions, co-predicable of the same subject; rather aU the senses (ex­
plications) of *man', as understood, must be the same sense (explica­
tion). I n the parenthesis " a U " m a y m e a n *aU these', v i z . pale things a n d
artistic things a n d m e n (as atB^. 999^20 a n d J 26. 1023^32, but not at
1007*21, or 1007*6 w h i c h refers back to this passage); otherwise there is
a suppressed assumption that aU predicates are connected by a c h a i n of
co-predicability. " B e c a u s e s y n o n y m o u s " : because aU wiU have name a n d
formula i n common {Categories 1. 1 * 6 - 7 ) ; they wiU a k o be m a n y - n a m e d ,
i.e. nameable by distinct but (in our sense) synonymous names, but
there is no need to follow Alexander i n importing t h k otherwise u n ­
attested sense of 'sunonumos'.
I t is h a r d to see Arktotle's motive for introducing the distinction be­
tween signifying and signifying about i n these Hnes. I t seems that A l e x a n ­
der took h i m to be replying to the foUowing argument: a m a n is p a l e ; to
be pale is not the same as to be a m a n ; therefore a m a n is not a m a n .
T h e reply goes: the argument is v a l i d only i f its first premiss means *a
m a n i s j u s t what something pale is' ( c f 1007*33); but that is true only i f

96
Г4 NOTES 1006b13

*man' a n d 'pale* signify the same; but that is true only if *signify the
same* is equivalent to *signify about the same*. However, Aristotle is
attempting a proof of P N G , not the demolition of a counterproof.
T h e baffling section from ^22 to ^28 introduces the expression Ч0 bc
a not-man', distinguished from *not to be a man* by word-order i n thc
Greek. T h e r e are two reasons for thinking that nothing is meant to t u m
on the difference of these two, although Aristotle emphasizes it later
(1007*24, 2 8 - 9 ) . (i) b28 (Ч0 be a man* and *not-man' signify something
different) seems to refer back to ^13-14 ('to be a man* cannot signifyjust
what *not to be a man* signifies). (ii) Aristotle maintains elsewhere that *to
be a man* and *not to be a man* are contradictories, while *to be a m a n '
a n d *to be a not-man' are only contraries a n d wiU be false together when
predicated of something non-existent or of something incapable of being
a m a n (PriorAnalytics I 46. 51^36-52*14, De Interpretatione 10. 19^19-30).
T h u s , though it m a y be that *it is not possible that x is F and n o t - F ' does
not demonstrate *it is not possible that x is a n d is not F ' nevertheless one
who conceded the former without the latter would be i n a n intolerably
strange position w h i c h can safely be ignored.
^22-8 operates with five propositions:

(a) *man' and *not-man' do not signify different things;

(b) Ч0 be a m a n ' a n d *not to be a m a n ' do not signify different things;
(c) to be a m a n is to be a n o t - m a n ;
(d) to be a m a n and to be a not-man are one thing;
(e) to be a m a n and to be a not-man have one formula.

A t the beginning Aristotle says that (a) implies (b); at the end he says
that (b) h a d been refuted. T h i s makes it fairly certain that the section
rejects (a) on the basis of the rejection of (b), but it remains unclear,
first, what is the purpose of rejecting (a) a n d , secondly, how the other
propositions fit i n . W e can say at least that the purpose ofrejecting (a)
ought not to be to refute the objection i n ^18-22, against w h i c h the nega­
tion of (b) cannot properly be assumed. A s to the other propositions,
Ross interprets the argument thus: (a) implies (b), w h i c h , since (e) is
equivalent to (d), impHes (c); and (d) implies (b), But this makes (c), (d),
and (e) entirely superfluous. M o r e probably (d) is meant to show how we
are to get from (a) to (b), w h i c h (d) implies, not how we are to get from (b)
to (c); a n d ifso, the " f o r " i n ^25 must explain the conditional i n ^22-4,
not the " s o t h a t " clause w h i c h it inmiediately follows. So we have:
because (e) is equivalent to (d) (^25-7) a n d (d) implies (b) (^27-8),
(a) implies (b) (^22-4). I t is tempting to complete the argument by taking
(c) as the means ofgetting from (a) to (e), but that could be done only
by exchanging the connectives i n ''so that to be a m a n wiU bc to be a
n o t - m a n ; / o r they wiU be one t h i n g " .

97
ioo6b28 METAPHYSICS Г4

юоб^зВ. T h i s notc raises four questions about the final section of

P a r t I : (A) w h a t is the scope of its modal words? (B) c a n it be made a
valid piece ofreasoning? (G) w h y should the opponent accept it? ( D )
h o w does it make use of w h a t has preceded?
(A) Does *neccssary' i n the first sentence govern the whole conditional
or only its consequent? I f t h e first, the argument will go (using * Q ' = *it
is necessary that', ' 0 * = *it is possible that', 'М' = *is a m a n ' , ' T ' = *is
a two-footed animal')
(a) • {Mx ^ Tx)
(b) -O^x{Mx& -Tx)
{c) -O^x{Mx^-Mx)
I f the second, there is a suppressed step:
(fl) V;c {Mx -> • Tx)
{b) yrfx{Mx^-0-Tx)
[{c) y/x{Mx^-0-Mx)]
(d) - 0 3x{Mx8L-Mx)
I n the second version the step from {c) to {d) is valid i n any normal modal
system, provided that *->' is read as a sign of strict implication; but i n
that case propositions having the form of {d) i n the second version wiU be
true only if 'Mx' expresses a n essential predication. I n that version,
therefore, *man' cannot be taken as a stand-in for anypredicatewhatever,
a n d the conclusion does not justify a n unqualified assertion of P N C . F o r
other reasons too the second version is less likely: Aristotle's word order,
though not decisive {cf.PriorAnalytics I i i . 31^7-8) argues against i t ; so do
the words " a t that t i m e " (^31, onutted by one good M S . ) a n d " s i m u l ­
taneously" (^33), w h i c h it renders at best superfluom.
(B) I n neither version does (b) seem to demonstrate {c), for not aU
two-footed a n i m a k are m e n . W e must remember, however (and A r i s ­
totle renunds us), that the role of *two-footed a n i m a l ' is to indicate
the one thing that *man' is being taken to signify; a n d no fallacy is
committed if * T ' means *is a m a n i n the sense "two-footed a n i m a l " '.
( C ) I n defending the step from {a) to {b) Aristotle says that *cannotnot'
is w h a t *must' signifies: could he not have avoided the whole rigmarole
by pointing out that *is not not' is what *is' signifies? W h y should the
opponent not reject the former, as he would the latter?
( D ) {a) refers back to tKe assumption of 1006*31 that *man' signifies
two-footed a n i m a l ; the step from {b) to {c) relies on the assumption of
1006*34-^13 that that is aU *man' signifies; but w h a t is the bearing on
this paragraph of 1006^13-28? T h e rest of this note sinks that question
into a survey ofthe two types ofinterpretation given to the argument as a
whole. Those who favour type 1 must answer the question i n one of two
w a y s : either, the conclusion d r a w n i n ^33-4 was anticipated at 1006^13
or at least ioo6^i8 (this has been discussed); or, 1006^13-28 is not

98
Г4 NOTES ioo6ba8

intended to assist the present argument at aU, but prepares the ground
for Part I I I . T h e comparative feebleness of these arwwers gives the
opening to the champions of type 2, whose case is i n three stages, (i) A t
ioo6^i3 the sense required for 'signify' is such that the proposition there
stated about the word ' m a n ' woiüd not be thought by Aristotle to apply
to every univocal noun a n d adjective (see note a d loc.). (ii) I n ^38 it is
possible, as we have seen, to understand Aristotle to say that thc ex­
plication of ' m a n ' belongs to m e n necessarily, something w h i c h is not
true of the explications ofe.g. 'pale' a n d 'large', since no manisessentially
pale or large. (Alternatively, even i f i n ^28 'necessary' goverm the whole
conditional, Aristotle might stiU think w h a t is said there false of 'pale'
and 'large', on the different ground that the definition of 'pale' is not
true of pale things, but only oftheir pallor—see Categories 5. 2*29-34 but
contrast Г 5. 1010^25-6.) (iu) G i v e n these interpretations of 1006^13
and b28 it is possible to find a connection between 1006^13-28 a n d
the present paragraph, v i z . the suppressed implication: i f ' m a n ' is ex­
plicated by 'two-footed a n i m a l ' a n d 'not-man' cannot be so explicated,
it follows that m e n are necessarily two-footed animals. Unfortunately
it appears that, in whichever w a y we construe the 'necessarily' i n its
consequent, this suppressed implication ought not to be accepted by
Aristotle as true. W h a t is six-foot long is not, i n either way, necessarily a
length ofsix feet; yet 'to be six-foot long' fits the type 2 interpretation of
1006b13, as having no explication w h i c h a k o expUcates 'not to be six-
foot long'. T h e type 2 interpretation thus appears to confuse two different
restricting conditions on predicates—roughly speaking that they be non-
relative and that they be essential. Whether it c a n be purged ofthis con­
fusion, without reducing to the non-controversial assertion that Aristotle
does not apply P N C without qualification to relative predicates, is not
clear, but, even i f n o t , its readings of 1006^13 and ^28 remain possible.

Г 4: sndArgument, Part I I (1006^34-1007*20)

1006^34. " T h e same argument applies" if—Aristotle apparently means
— o n e starts by assuming a signification for 'not-man' (an "indefinite
n a m e " , De Interpretatione 2. 16*32). I t is h a r d to see how the second
sentence explains this contention. I t would fit better with 1006^13-18,
to w h i c h it apparently refers back. B u t the reference is not whoUy apt,
because the previous passage argued to the conclusion, not that ' m a n ' a n d
'not-man' signify different things, but that their signifying different
things would be compatible with their signifying " a b o u t " one thing.
A t b8-9 the opponent is assumed, i n apparent disregard of 1006"18-21,
to have been asked a question such as ' I s Callias a m a n ? ' Aristotle's
objectioш to the answer 'yes and no' are not quite fair. T h e opponent
who appended to his answer ' a n d he is pale' would be adding something
that does not answer the question put, but ' a n d he is not a m a n ' does

99
 1006b34 METAPHYSICS Г4

constitute a n alternative answer. Nevertheless it is reasonable to insist

on dealing with alternative answers one at a time.

Г 4 : 2nd Argument, Part I I I ( 1 0 0 7 * 2 0 - ^ 1 8 )

. 206-8 1007*20. T h i s argument introduces some of Aristotle's more technical
metaphysical notions. Its concluding remark is illuminated by Posterior
Analytics I 2 2 . 8 3 * 2 4 - 3 2 : " A g a i n , things signifying a substance signify,
about that of w h i c h they are predicated, just what so-and-so or just
w h a t a certain so-and-so [is]. A n y things that do not signify a substance,
but are said about another subject w h i c h is not just what so-and-so nor
just what a certain so-and-so [is], are coincidental, as for instance pale
about m a n : for a m a n is n o t j u s t w h a t pale n o r j u s t what a certain pale
[ i s ] ; rather, doubtless, he is a n a n i m a l , for a m a n is just what a n a n i m a l
[is]. A n y things that do not signify a substance must be predicated about
some subject a n d nothing c a n be pale w h i c h is not a certain other thing
that is p a l e . " (Cf. Posterior Analytics I 4 . 7 3 ^ 5 - 8 ; on elision o f t h e verb
after *just w h a t ' cf. 1 0 0 6 ^ 1 3 , where the elided word was taken to be
*signify'.) I n the last sentence of this passage Aristotle means not that
what is pale must have other properties besides, for that is true of what
is a m a n or a n a n i m a l , but that what is pale cannot be identical with the
pale that it is. A particular m a n , therefore, is identical with m a n , a n d
with ' a certain a n i m a l ' (i.e. a kind of a n i m a l ) . Aristotle's theory thus
distinguishes two types of predication: essential predicates, the words
for w h i c h "signify a substance" ('substance' as a t J 8. 1 0 1 7 ^ 2 1 - 3 ) , are
identical w i t h the subjects of w h i c h they are truly predicated; other
predications are true i n virtue of the fact that two distinct items, e.g. a
substance a n d a quality, " c o i n c i d e " . T h i s theory has the consequence
that nothing c a n possess more than one essential predicate ("for it, to
be is nothing e l s e " ) , so that e.g. ' m a n ' a n d ' a certain a n i m a l ' must be
capable of signifying the same thing. Aristotle uses that consequence to
argue that if two contradictory predicates such as 'is a m a n ' a n d 'is not
a m a n ' (or 'is a not-man') are essential, a n d if they are truly predicable
of the same subject, say Callias, then the things they signify must be
identical with that subject a n d so with each other. B u t , as was stated at
1 0 0 6 ^ 1 3 , things signified by contradictory predicates are not identical.
Therefore, iftwo such predicates are truly predicable ofthe same subject,
they are not both essential predicates of that subject; a n d if not both,
then (he assumes) neither. " J u s t w h a t to be a m a n [ i s ] " etc.: Aristotle's
meaning must be 'to be just what a m a n is' but his word-order prevents
that translation by separating 'just w h a t ' from ' a n i m a l ' at * 2 2 - 3 a n d
from ' m a n ' at * 2 8 - 9 . " Y e t those are its d e n i a l s " : a n d so would have to be
identified w i t h it by the opponent.

T h e argument seems vulnerable at two points, (i) I t relies on a dubious

theory of predication; for even if it is possible to make sense of the

100
Г4 NOTES 1007*20

distinction between essential and coincidental predications, the former

are no more statements of identity than the latter are (on the origin of
this confusion not m u c h has been written, but see G . E . L . O w e n , ' T h c
Platonism ofAristotle', 156-63). (ii) Although the premiss carried for­
w a r d , iftaken in sense (2) or sense (3) ofthe note on 1006^13, is extremely
plausible, it does not follow rigorously from the opponent's original a d ­
mission that ' m a n ' is being used with a single signification.

1007"33. T h e argument that not all predications c a n be coincidental pp. 217-19

(i.e. here, non-essential) is i n two stages: (A) ifeverything is a coincidence,
" i t wiU be necessary to go on to infinity" ("33-^1); (B) " b u t that is
impossible" (^1-15).
(A) I t is possible to form a n endless series of true 'is' sentences i n each
of w h i c h the grammatical complement is the expression which was
grammatical subject i n the last: e.g. ' a m a n is pale', 'Callias is a m a n ' ,
'this body is CaUias' . . . According to Aristotle's theory of predication,
the first of these examples predicates one item, pale, of another, m a n ,
whereas both of the other examples predicate an item of itself. F o r that
reason the endless series of sentences does not generate an endless series
of things, but at some point we arrive at a thing not "predicated about a
certain subject", i.e. not predicated ofanything else (in Aristotle's usage,
though not i n ours, x's subject, i f any, is necessarily different from x).
Aristotle's thesis is that without essential predication it is not possible to
terminate thb series of ever different subjects, and " t h e r e wiU not be
anything w h i c h things are initially a b o u t " . I n actual fact, of course, its
termination requires only that there should be true statements of identity,
so that this stage of the argument assumes that aU true statements of
identity predicate essentially, the converse ofthe assumption i n *20-33.
(B) Aristotle next states that, so far from a n endless series of coinci­
dental predications, there cannot be a series with even two members:
given that T coincides in X, it is not possible that any should coincide
in T. T h i s is qualified to allow for the transitive case i n w h i c h ^ a n d T
coincide because both coincide i n X. Setting that case aside, Aristotle
considers two ways i n w h i c h a non-transitive series might be constructed,
(i) T (pale) coincides i n X (Socrates) and ^ coincides i n XT (Socrates thc
pale). T h a t cannot happen, he says, because XT is not a unity (cf. De
Interpretatione 11. 2 i * 7 - i 4 ) . (ii) J^coincides i n X and ^ m T. Aristotle
baldly asserts that that also is impossible, except i n the manner set asidc:
a predicate of X's predicate must be a predicate of X. T h i s argument is
unsatisfactory both because the last assertion is false (e.g. pale is conmion,
but Socrates who is pale is not common, cf. Soph. El. i79*26ff.) and
because, even iftrue, it would not establish ( B ) , viz. the impossibility o f a n
infinite series of coincidental predications. I n fact, Aristofle has perceived
correctly that (B) cannot be established as initiaUy stated, but he hat

IOI
loo7*33 METAPHYSICS Г4

weakened it so m u c h that it no longer combines with (A) to yield his

conclusion. T h e version he should have substituted is: any series o f t r u e
coincidental predications must have a first member (i.e. lead back to a
subject not predicated of a n y other subject). B u t that version, though
defensible, is not defended by Aristotle here.
I f t h e final sentence sums up Part I I I it is, ofcourse, a n exaggeration.
T h e most that could have been shown is that essential predicates are not
co-predicable with their contradictories.

Г 4: 3rd Argument (1007^18-1008*7)

1007^18. Although the contradictory of P N C , as formulated at Г 3.
1005b19-20, is

{a) Здг 3 F O {Fx & -Fx),

Aristotle now, a n d for the rest of the chapter, represents his opponent as
asserting something bolder, the refutation of w h i c h will not establish P N C .
T h e version needed at ^24-5 ( " i f t h e contradiction is really t r u e " ) , a n d
at ^34 ( " i f t h e affirmation holds good o f h i m , necessarily its denialdoes
too") taken w i t h 1008*2 ( " a n d i f i t does, its affirmation wiU t o o " ) , is

{b) y^x^F{Fx<^-Fx),
F r o m {b) the paragraph deduces by two arguments, ^20-5 a n d ^29-^2,
the absurd consequence that " e v e r y t h i n g wiU be o n e " . T h e former
argument appears to be this. T h e "thesis of Protagoras" is
{c) \fp [(someone believes that p) ^ p].

Assuming
{d) y/x y/F [(someone beUeves that Fx) v (someone believes that —Fx)]
it follows that
(e) WxWF{FxV-Fx),
B u t {b) a n d {e) entail

(/) Wxy/F{FxSi -Fx)

w h i c h Aristotle expresses i n two ways: " t h e same thing wiU be both a
warship a n d a w a l l a n d a m a n " sc. a n d so on for every predicate, and
" e v e r y t h i n g wUl be o n e " sc. have the same predicates ("nothing is truly
o n e " i n the diflferent sensfe that nothing is countable or discriminable).
T h e third sentence illmtrates one side of the disjunctive argument from
{d) to ( / ) , taking a premiss of the form *someone believes that —Fx\
(e) construes " s o m e t h i n g " a n d " i t is possible" i n the second sentence as
*anything' a n d *it would be true'. Both readings are required i f the
conclusion is to m e a n that everything wUl be one actually. B u t at ^26
Aristotle is prompted by Anaxagoras' d i c t u m to suggest that his opponent

102
Г4 NOTES t007b18

has without realizing employed Ъе' i n the " i n d e f i n i t e " sense *can bc*
(cf. Г 5. i o o 9 * 3 0 - 6 ) , so that the conclusion ought to go
(Л WxWF{OFx8iO-Fx),
w h i c h woidd follow by {b) from
(0 WxWFO{Fxy -Fx).
T h i s leads to the second argument (^29). T h e opponent c a n bc forced to
accept the stronger proposition {e) rather than (^'), a n d s o ( / ) rathcr
than ( / ' ) . F o r i f the denial of G {AG) holds good of what is G , afortiori
the denial ofF ( Δ ^ ) holds good o f i t i f i t is not F:
(g) Wx[lG{Gx & Δ^^) ^ WF{-Fx ^ АШ
I n the ensuing example Aristotle takes (g) as equivalent to
{h) Wx[3G{Gx & -Gx) ^ WF(Fx v -Fx)].
H c assumes that the closm:e of the antecedent of {h) c a n be asscrtcd
(presumably on the basis of {b)), and (e), the closure ofits cOnsequcnt,
accordingly detached. I n the three final sentences, repeating that i n the
example either Fx or —Fx, he shows that each ofthese combined with {b)
entails that Fx & —Fx, w h i c h generalizes into ( / ) .

1008*2. T h e previous paragraph displayed the opponent as conunitted

to {e), a version of the principle of excluded middle. I t should not be
surprising that Aristotle now represents h i m as committed also to deny­
ing a version of the principle. T h e four propositions are:
(i) he is a m a n ; (iii) he is not a m a n ;
(ii) he is not a m a n ; (iv) he is not not a m a n .
Aristotle argues: i f (i) a n d (ii) " m a k e up the one former", sc. conjunctive
affirmation, (iii) a n d (iv) wiU make up a n opposite conjunctive d e n i a l ;
the former is, b y ( / ) of the previous paragraph, t r u e ; hence, by {b), the
latter is also true. T h e argument overlooks the fact that these two
conjunctions, being o f t h e form 'p & q' a n d & —q\ are not contra­
dictory opposites, as the application of {b) requires.

Г 4: 4th Argument (1008*7-34)

1008*7. T h r e e versions of the opponent's thesis are now distinguished.
T h e boldest is {b)—•or(f), w h i c h the 3rd argument deduced from {b);
alternatively, (A)'s double implication m a y be taken i n one direction
only, giving
{j) Wx WF {Fx ^ -Fx) & - У д : WF{-Fx Fx);
or thirdly the thesis m a y hold " i n some cases but not i n others", w h i c h
would allow it to say as little as
{k) lx3F{Fx&.-Fx).

103
ioo8*7 METAPHYSICS Г4

I t is not clear whether Aristotle would distinguish {k) from (a); i n any
case he dismisses it from the discussion w i t h the inadequate comment
that it narrows the field ofdispute. {j) is, he argues, untenable; forif*^;is
not a m a n ' m a y be " f i r m a n d certain, the opposite assertion would be
stiU more c e r t a i n " . Affirmations are said to be " p r i o r to a n d more cer­
tain t h a n " denials at PosteriorAnalytics I 25. 86^33-4; but that conten­
tion needs to take *certain' ('gnorimos*) i n the sense *intelligible', w h i c h
b not suitable here.
T h e argument of *28-30 came to be known as perilrope or turning of
the tables; cf. Theaetetus iyo-i,rS. 1012^13-22, K^, 1062*36-^7.

1008*30. T h i s s e e m s to b e l o n g w i t h 1008*20-7, a n d to argue against

the opponent who says " i t is not true to state separately". I f *man-and-
n o t - m a n ' is t o b e understood as not predicating two things, it has not
been made intelligible at aU a n d nothing is " d e f i n i t e " ; i f it does pre­
dicate two things, they c a n (and, by 1007*8-20, should) be predicated
separately.

r * 4 : 5 t h A r g u m e n t (1008*34-^2)
1008*34. H e r e , but not at Г 7. 1011b25 a n d Г 8. 1012^7-8, Aristotle
hesitates to appeal to the definitions of *true' a n d *false'.

P 4 : 6th Argument (1008^2-31)

1008^2. T h i s argument continues the discussion of truth initiated at

1008*18. ( T h e two verbs translated *be i n error' are from the same root
as the word meaning *false'.) W e are to consider three beliefs: let A
believe that Callias is pale, B that Callias is not pale, C that Callias is pale
a n d not pale. Aristotle asks his opponent to choose among three alterna-
tivejudgements on C's beUef. T h e third, i n ^7, is that C , like A a n d B , is
i n error; the first therefore, i n ^3, is that C is not i n error, a n d the second,
i n ^5, that C is less i n error than A and B (this follows Ross's restoration
ofthe corrupt text, also adopted by the O C T ; see Ross's notes). Aristotle
proceeds to show that each of the three judgements is untenable. Against
the first he objects that ' C a l l i a s is pale a n d not pale' is unintelligible i f i t
does not entail *Callias is pale' a n d 'Callias is not pale'. I t is not clear
w h y a separate objection should be needed against the second judge­
ment, or how Aristotle's objection works (it would work, as a n antici­
pation of 1008^31-1009*5, i f w e adopted the reading ofsome M S S . 'but
the one who believes the former w a y has more t r u t h ' ; but that reading,
as Ross says, presents the opponent w i t h a n alternative he could not
wish to defend). Against the third alternative Aristotle makes the good
point that speech, a n d also belief, depends on not everything having the
same truth-value.

104
Г4 NOTES ioo8bi2

i o o 8 ^ i 3 . T h e discussion now takes a new turn. Aristotle has already

argbied a priori that P N G cannot be disbeUcved {Г 3. 1005^23-32); now
he appeals to experience as showing that at a n y rate the strong contrary
thesis {b), stated at 1007^18-19, is never believed. H e does not suggest
that thb fact is sufficient to dispose of his opponent. T h e example of
observing a m a n indicates that a person's actions m a y vouch for his
theoretical as well as his practical beliefs; but the cautious conclusion at
b26-7 expresses confidence only about some of the latter. I n thc two
final sentences Aristotle apparently faces the (irrelevant) objection that
from 'everybody believes some simple propositions and not their contra­
dictories' it does not foUow that any simple propositiom are true; for
beliefs, unlike knowledge, m a y be fake (the analogy with health shows
that this is the intended distinction between opinion a n d knowledge).
W h a t k the force ofAristotle's reply? I f h e means 'the believer o f f a k c -
hoods wiU be aiudous to exchange them for truths: so there must be
truths to exchange', that is inadequate i n three w a y s : (i) the prenuss
entails not the exktence, but b e l i e f i n the existence, of truths; (ii) thc
analogy with health is imperfect, since the believer of falsehoods must
suppose, not that he c a n exchange his condition for onc of true belief,
but that his condition is already one of true beUef; (iii) i n Aristotle's
strict usage the attainment of 'knowledge' {episterrw) about what is better
a n d worse, as about contingent matters i n general, is impossible (jVico-
rrmhean Ethics V I 3. 1139^20-1) and so not a subject ofrational anxiety.

Г 4 : ythArgument (1008^31-1009*5)
1008^31. Similar arguments have been brought i n modern times against
the sceptical thesis that everything is uncertain.

CHAPTER 5

T h e r e are echoes i n this chapter and the next ofPlato's Theaetetus 151¬
86, a longer critique of the thesis ofProtagoras that " m a n is the measure
of aU things, ofthose that are that they are a n d ofthose that are not that
they are n o t " {Theaetetus 152 a ) . Plato more than once states the thesis i n
the form " w h a t is thought by [or, seems to] each person, that a k o i s "
(161 c) or " . . . is to h i m w h o thinks i t " (170 a, 177 c ) . Aristotle a k o
employs the verb 'be thought' {'dokein'), but more frequently 'phairusthai',
often rendered 'appear' but i n this translation by the verb 'be imagined',
with whose wide range of uses it corresponds quite closely. T h e noun
'phantasia' {Δ 29. 1024^24, 26, 1025*6) means 'imagination', not 'appear­
ance'.
T h e layout of the chapter is as follows. T h e first two paragraphs
introduce the thesis of Protagoras and distingukh two types of response

105
ioo9*6 METAPHYSICS Г5

to it. L e a v i n g the second type to Г 6 , Aristotle thereafter deploys three

sceptical arguments a n d attempts to expose their weaknesses: at 1009*
2 2 - 3 0 he states, a n d at * 3 0 - 8 responds to, a n argument from changeability
against P N C ; at 1009*38-^12 he states, a n d at ^12-33 a n d more espe­
cially 1010^1-1011*2 responds to, a n argument from contrary appear­
ances to a Protagorean conclusion; a n d at 1010*7-15 he states, a n d at
*15-b1 responds to, a further argument from changeability to the con­
clusion that " o n e ought to say n o t h i n g " (1010*12).

1009*6. Aristotle ignores thc fact that 1008*7-15 distinguished differ­

ent versions ofthe " o p i n i o n " that denies P N C . T h e version he now cites,
" i t is necessary that everything should be simultaneomly true a n d false",
is equivalent to the strong denial ( / ) — f o r the reference-letters, see notes
to 1007b18. H e is clearly right i n saying that ( / ) entails Protagoras' thesis
(c), since i f everything is true, everything that is thought is true. T h e
converse entailment needs a further premiss. B u t w h a t ? Aristotle's—
" m a n y people have mutually contrary beliefs"—is i n effect
(/) 3F (someone believes that Fx a n d someone believes that —Fx),
w h i c h is not strong enough; he needs
{m) Wx ^F (someone believes that Fx a n d someone believes that —Fx).
Y e t who believes, as {m) requires, that e.g. Socrates is a warship? I t wiU
not do to point to the opponent o f P N C himself, for according to Aristotle
he does not believe what he says {P 3. 1005^25-32). Nor is {k), the weak
denial o f P N C , equivalent to Protagoras' thesis {c): given (/), {c) entails
{k), but {k) does not entail {c).

1 0 0 9 * 1 6 . Aristotle now introduces the important distinction between

opponents who are i n genuine perplexity a n d those who relish what De
Sophisticis Elenchis 2. 165^11 calls " c o m b a t i v e a n d eristic" debates, a n d
lookforvictoryordefeatasinamatch;cf.r6. 1011*3-17,7. 1012*17-24,
Topics I 12. 105*16-19. De SophisticisElenchis studies tactics appropriate
against the latter; *philosophy' is interested only i n the former (cf.
Г 2. 1004b25-6). T o gain a sporting t r i u m p h it is necessary to refute
y o u r opponent's conclusion, but if your object is to remove perplexity
you need only tackle the " t h i n k i n g " that led to that conclusion, i.e.
expose the error i n his argument. Y o u need not force h i m to agree with
y o u ; only take away the considerations w h i c h , as he thought, forced h i m
to disagree with you. T h i s therapeutic style ofphilosophy is well exempli­
fied i n the remainder of the chapter. T h e r e is some conflict between
Aristotle's previous c l a i m (Г 3. 1005^25-32) that Heracliteans do not
believe w h a t they say a n d the present admission that a perplexed op­
ponent may " h a v e this belief"; c f Г 6. i o i i ^ i o " t h e y are not actually
c o n v i n c e d " , 1011*3 "those who are c o n v i n c e d " . T h e former description
is shrewder, as H u m e s a w : a m a n w i t h a philosophical problem does not

106
Г5 NOTES ioo9*i6

reaUy believe the paradox of w h i c h he thinks he has a n incontrovertible

proof, but he is still different from the debater who "states it for thc
sake of stating i t " .

1009*22. T h i s first argimient does not deal with Protagoras, but

develops the suggestion made at Г 4. 1007^26-^ that those who say that
e.g. Callias is simultaneously pale a n d not p a l c do so on the grounds,
and i n the sense (*32), that he is simultaneously capable of being pale
and capable of being not-pale. " C o n t r a r i e s come to be out of thc same
t h i n g " : sc. come to hold of the same subject. T h e sceptical concltuion is
extracted by combining this familiar fact w i t h the ambiguous principle
that " i t is not possible that what is not should come to b e " . Aristotle is
clear that his opponent needs to take the principle to say, fakely, that
what is not F cannot come to be F; from w h i c h it is concluded that i f
Callias comes to be now pale, now not pale, he must have been pale a n d
not pale aU along (this must be preferred to the crass argument suppUed
in K 6. 1062^24-30). W h a t is the innocuom sense of the principle i n
which " w h a t they say is c o r r e c t " ? ( i ) Without doubt the dictum was
sometimes taken to m e a n *what is not any G cannot come to b e F ' , w h i c h ,
of course, lends no support to the conclusion that CaUias was pale a n d
not pale aU along. (2) T h e quotation from Anaxagoras suggests another
interpretation: what becomes pale must aU along contain the materiak
of pallor, have, as it were, the pallor inside it waiting to be manifested.
T h i s doctrine need not have the explanatory scientific flavour w h i c h
Anaxagoras seems to have given i t ; at its baldest it amounts to saying
no more than that i f Callias is to become pale he must already be capable
of being pale, be pale " p o t e n t i a l l y " . B u t the scientific hypothesis lends
plausibUity, otherwise lacking, to the description of Callias as being pale
and not pale together (not pale outside, pale somehow *inside'). Most of
Aristotle's expUcation refers plainly to this second account: " t h a t w h i c h
is [F] m a y be so caUed i n two w a y s " — * a c t u a U y F ' a n d *potcntially f ' (cf.
Δ 7. 1017*35-b9). B u t w h e n he says " i t is possible . . . for the samc
thing to be simultaneously both a thing-that-is a n d a thing-that-is-not
(only not i n the same r e s p e c t ) " the objection arises that the respect need
not differ i n the case of potentiality: Callias c a n be at once potentially
pale a n d potentially not paU, Does thc parenthesis confusedly import the
other point, that GaUias c a n be actually F a n d not G together but not
actually F a n d not F together? O r docs Aristotle take the fact that
Callias must be a c t u a l l y p a l e or actually not palc to imply that he cannot
at one time be potentially both? T h i s is the first attempt, at least since
r 4 . 1006*31-1007*20, to attack the bare contradictory o f P N C a n d so to
defend P N C itseU*. I n * 3 6 - 8 Aristotle reverts to his more cautious con­
tention that some contradictions a r c fakc, pointing out that that con­
tention is not to b c overthrown by a n y argument based o n change,

107
ioog*22 METAPHYSICS rs

w h i c h some things do not undergo (cf. 1010*1-7, 2 5 - 3 5 ; i n none ofthese

places does Aristotle suggest that P N C applies only to changeless things).

1009*38. Aristotle now turns to the thesis ofProtagoras. I n accordance

w i t h his prescription to examine not the thesis itself but the " t h i n k i n g "
w h i c h led to it, he proceeds to describe w h a t is to us one of the most
familiar arguments for scepticism, the argument from contrary appear­
ances, especially sense-impressions (cf. Theaetetus 154 a b ) . T h e sceptical
conclusion, d r a w n by m a n y modern philosophers a n d (as Aristotle tells
us) by Democritus, is that it is impossible to know, a n d perhaps meaning­
less to say, that something is e.g. sweet, a n d we must be content to say
e.g. that it tastes sweet to Socrates; Protagoras concluded differently
that 'x is sweet' is not meaningless but elliptical for e.g. 'x tastes sweet
to Socrates'.

1009^12. Aristotle begins his response to the argument from contrary

appearances at i o i o ^ i . M e a n w h i l e , he comments now that the argu­
ment is nourished by confusion between the concepts of wisdom {phro-
nesis, defined at Nicomachean Ethics V I 5. 1140^5-6 as " a true reasoned
state concerned w i t h action regarding w h a t is good a n d bad for a m a n " ,
but here more general), perception (or sensation, aisthesis), a n d modi­
fication {alloi5sis, change o f q u a l i t y , traditionally rendered 'alteration').
I n DeAnima Aristode argues that perceiving ( I I I 7. 431*4-8) a n d under­
standing ( I I 5. 417^2-16) consist not i n being brought into new states
but i n functioning i n accordance w i t h states already possessed; a n d i n
the Theaetetus perception is defined i n such a w a y (152 b 11) that he who
perceives, unlike h i m w h o is wise (*31-3), does not necessarily have the
truth. Aristotle does not indicate the reasoning w h i c h he attributes (not
very convincingly, nor i n every case seriously) to the earlier writers he
mentions. I t m a y be this: if someone is brought into a state i n w h i c h x
appears F to h i m , then he perceives {x to be) a n F t h i n g ; if the latter, then
he knows of {x's being) a n F t h i n g ; i f the latter, x is F; therefore x is as it
appears. I t is not clear that both of the first two premisses i n this argu­
ment c a n be w r o n g ; on the other h a n d it is enough for Aristotle i f o n e of
them is wrong, as is obviously the case. Perception is mentioned i n none
o f t h e sayings he quotes; a n d the theories a d u m b r a t e d i n aU but that of
Anaxagoras are quite un-Protagorean unless combined w i t h the pro­
position that having something i n m i n d is a "state of w i s d o m " , a pro­
position w h i c h is tempting only because the words for *have i n m i n d ' ,
Ъе out of one's m i n d ' , a n d 'with other things i n m i n d ' share the root
' p h r o n - ' w i t h 'phronesis'.

1009^33. Aristotle is not the last philosopher to have expressed dismay

at the legacy of some of his forerunners i n the subject.

108
Г5 NOTES IOIO*I

i o i o " i . W i t h a disconcertingly Platoiüc air this paragraph seems to

concede that Protagoras was right about perceptible things. A s the
reference to indefiniteness shows, however, Aristotle means that he was
right only i n the harmless w a y discussed i n 1009*32-8. T h e allimon to
Epicharmus is lost.

x o i o " 7 . I f things are too changeable to be reidentified (Heraclitus) or

even identified (Cratylus), it is impossible to make true assertions about
them—or false ones, as perhaps Cratylus realized. Cf. Theaetetus 157 bc,
182 d-183 b. Aristotle tells us (A 6. 987*32) that Plato was early i n ­
fluenced by Cratylus. I t is quite possible that Heraclitus' views are
incorrectly reported by both Plato a n d Aristotle.

1010"15. I f t h e sceptic says, for example, that water while getting hot
is neither hot nor cold, Aristotle replies (i) it stiU has some cold and
already has some heat ( c f Physics V I 9. 240*19-26); (ii) its heat, like
anything else i n process ofgeneration, is made out ofsomething else (the
element fire) a n d produced by something else (e.g. a cook), and these
causes must terminate i n something not undergoing change (a 2. 994*1¬
19); (iii) even a changeable thing has one feature, its form, which remains
constant throughout its existence. I t does not foUow from this last point,
as Aristotle oddly implies, that changeable things c a n change only i n
quantity and not i n quality; form is one kind of qualification (cf.
J 14. 1020*33-b2), but that i n respect of w h i c h "bodies are . . . said to be
modified" is another {Δ 14. 1020^8-12).

1010'a5. Aristotle believed that bodies i n the celestial regiora, though

plainly subject to motion, are otherwise changeless.

i o i o " 3 2 . Aristotle does not explain w h y the C r a t y l e a n should be

disturbed that the strong denial of P N C has a consequence inconsistent
with his assumption that everything is changing. Ross's comment that
this a n d the two preceding paragraphs fail to show that " c h a n g e is
reconcUable with the l a w of contradiction" misunderstands their p u r ­
pose, w h i c h is to reconcile change with the possibility of true assertion
(1010*7-10).

i o i o ^ i . R e t u r n i n g to the thesis of Protagoras, Aristotle first looks at

a new argument i n its support (^2-3, discussed below), and then (^3-14)
rebuts the argument already stated at 1009*38-^13. X h e behaviour of
the expatriate who dreams he is i n Athens demonstrates man's abiUty to
grade appearances. Does Aristotle m e a n that we know that the dreamer
is not i n Atherw, or that he knows? I f t h e latter, does he reveal his know­
ledge by not starting for the O d e o n w h e n he wakes, or by not going
there i n his d r e a m (but behaving i n some less consequential m a n n e r ) ,
or by making no actual bodily movements while dreanung? T h e objec-

109
ioiobi METAPHYSICS Г5

tion that two contrary appearances m a y be equally authoritative is

considered i n thc next paragraph a n d at Г 6. i o i ι·25-^ι. " A s Plato also
s a y s " : TheaeUtus 178cd.
Against the argument that what is imagined must be true because
perception is true, Aristotle answers in this paragraph (^2-3) that not aU
that is imagined is perceived, a n d i n the next (1010^14-19) that not aU
perception is true. Both answers a r c , o n a n y reasonable interpretation,
convincing. But both are ambiguous, a n d thc second raises considerable
problems. These notes discuss three interpretative questions: (A) what
kind of imagination is opposed to perception? (B) did Aristotle himself
believe that perception of " s p e c i a l " objects is true? ( C ) what does he
mean by "perception is t r u c " ?
(A) A t least three kinds of imagination (pL·ntasia) differ from percep­
tion. I t is not clear whether Aristotie has i n mind: ( i ) creative imagina­
tion (cf De Anima I I I 3. 427^16-21), whether pictorial as when one
imagines the death o f H e c t o r {De Memoria 1. 450*12, De Insomniis 1.
459*14) or non-pictorial as when one pretends or assumes that ^2 is
rational or giraffes are extinct; (2) the imagination that is allegedly
exercised by somebody when things have a certain appearance to him,
whether or not the appearance does, or is judged by h i m to, corre­
spond to reality {De Anima I I I 3. 428*24-^9, where Aristotle contrasts
such phantasia—a stick looHng bent i n w a t e r — w i t h judgment, a n d see
K . Lycos i n Mind 1964, pp. 496-514); or (3) judgment {De Sophisticis
Elenchis 4. 165^25, Nicomachean Ethics I I I 5. 1114*32, a n d see De Anima
I I I 10. 433^29 on the contrast between ratiocinative a n d perceptive
phantasia).
(B) T h e special objects offour senses are listed at DeAnima I I 6.418*12:
"sight ofcolour, hearing ofsound, taste offlavour, while touch has many
different" objects. DeAnima I I I 3. 427^12 says "perception o f w h a t is
special is always t r u e " a n d I I I 6. 430^29 "seeing what is special [or
" p a l e " — t e x t doubtful] is true, but whether a pale thing is a m a n or not is
not t r u e " . T h i s view is modified at DeAnima I I I 3. 428^18: "perception
of what is special is true or is liable to fabehood to the least possible
extent". Г 6. 1011*34 m a y indicate as Aristotle's opinion (but see note
a d loc.) that perception o f w h a t is special is true w h e n the circumstances
are favourable. T h e evidence of ioio^2 itself is doubtful because of the
text. T h a t translated supplies a n *if* a n d a *not' not i n the M S S . ;
without them the sense would be *not even {oude) perception, at least of
what is special, is false, but . . Editors have noticed that *not even' is
inappropriate i n a n argument designed to prove that imagination is
sometimes false. T r e d e n n i c k conjectured 'ou de% giving the sense *ad-
nuttedly perception, at least of what is special, is not false, but . .
But Alexander's comment suggests the reading translated, w h i c h differs
from Tredennick's i n leaving Aristotle uncommitted.

IIO
Г5 NOTES iotobi

( C ) 'Perception is true' evidently means the same as 'perception is not

mistidcen' {De Anima I I 6. 418*12, De Sensu 442^8). F i v e interpretatiom
are possible, ( i ) Perception is a n achievement, i n the sense that i f some­
one perceives (or perceives something to be) a n F thing, it foUows that
what h e perceives really is F. However, Aristotle's previom distinc­
tion between perception a n d wisdom (1009^12) seemed to t u m o n the
former's not being true i n this w a y ; moreover, 'aisthesis' c a n m e a n 'seiMe¬
impression', a n d sense-impressions do not have to be true i n this w a y ; a n d
finally aU perceptions are true i n this w a y , not only those of special
objects. (2) Sense-impressions are unmistakable, i.e. no one c a n be i n
error as to how things look, soimd, feel, taste, and smeU to h i m . B u t i n ^ g i n -
ation has as strong a c l a i m to be tmmistakable. (3) Sense-impressions
are veridical or incorrigible, i.e. as things look etc. so they reaUy are.
Restriction ofthe doctrine, i n this sense, to special objects would give it a
Berkeleian flavour. T h e objection that Aristotle rejected the Berkeleian
proposition ( К е м у , Mind 1967) tends to show not that he did not use the
word 'true' here i n the sense 'veridical' but that he d i d not accept
without qualification the doctrine that perception is true of its special
objects. (4) K e i m y suggests that b y 'perception of w h a t is special is true'
Aristotle meant 'each perception, i.e. sense, is the court of appeal b y
which w e j u d g e its proper objects': a m a n m a y look pale who is not, b u t
the mistake's detection a n d correction rest finally o n fiu^her looking.
But i n this seme imagination is true also; for if something appears to
a m a n as it is not, h b nüstake is corrected only w h e n the thing appears
to h i m as it is. I n a n y case the Protagorean thesis must go beyond thc
unexceptionable c l a i m that thought a n d imagination a r c arbiters of
truth; a n d its falsity could not therefore be exposed b y contrast w i t h
a doctrine about perception so understood. (5) According to H a m l y n
{Aristoile's 'De Anima', o n I I 6. 418*11), the doctrine of the Metaphysics
passage is that " a sense cannot confuse its object w i t h that of another
sense . . . a person caimot e.g. be nustaken w h e n using hearing as to the
fact that he is hearing s o u n d " . T h i s parades as a version of (3) but, as
H a m l y n notes, depends o n the conceptual point that w h a t is heard must
be a s o u n d ; strictly speaking, Aristotle is not entitled to the c l a i m that
no one who hears c a n mistakenly believe he does not hear a sotmd, but
only to the claim that such a nustakc would be conceptual rathcr than
perceptual. T h e interpretation makes good sensc of Aristotle's contrast,
for i n the case of imagination nothing corresponds to the fact that
sounds cannot be misheard a s colours. A m o n g these interpretations (3)
and (5) seem to be the best, (5) i f ^2-3 gives Aristotle's o w n opinion.

1010'14. T h i s paragraph serves a double purpose, supporting 1010^2-

3 (not aU perception is true) a n d ^3-14 (perceptions differ i n reliabUity).
T h e sense of " a u t h o r i t a t i v e " might be (4), tiiat sight is thc arbiter of

III
 ioiobf4 METAPHYSICS Г5

colour-judgements etc., or (5), that colours cannot be mis-seen as tastes

or sounds, or (3), that one colour cannot be mis-seen as another. T h e last
is a gross exaggeration, but is suggested b y the unrestrictedness of " s o -
and-so a n d not so-and-so".

1010^19. tbe two final paragraphs Aristotle once again, as at

1009"36-^ a n d 1010^25-^1, ends his critique b y insisting that the argu­
ment from contrary appearances would not, even i f cogent, apply to
everything. H e makes two independent claims, (i) T h e premiss of the
argument, viz. that the same thing m a y present contrary appearances, is
true of sweet wine but false, even over a span of time, of " t h e sweet"
w h i c h the wine is. (ii) T h e conclusion ofthe argument, viz. that contrary
properties are jointly predicable of the same thing, is likewise false of
" t h e sweet" w h i c h , being unalterable, cannot possess contrary properties
even at different times ( " i t is not possible that w h a t is necessary should be
thus a n d otherwise"). O f these claims the first is stated by Aristotie
without evidence a n d the second has no force against the Protagorean
argument. Moreover its implicit definition of alteration i n tenns of
contrary properties is one w h i c h opponents o f P N C would have no reason
to accept, since according to them even unaltered things possess contrary
properties. " I n the b o d y " : sc. o f t h e perceiver; Aristotle ignores other
affecting circumstances, such as the m e d i u m of perception. " A n y t h i n g
that is going to be sweet is s u c h ofnecessity": this c a n hardly be intended
to restate ^ i s t o t l e ' s c l a i m that the sweet is itself unalterable, for its
subject must m e a n 'possessors of the quality sweet'; it will support that
c l a i m only i f " o f necessity" governs the whole clause (cf. perhaps Prior
Analytics I 15. 34*7, 17, 21).

i o i o ^ 3 0 . T h e argimient i s : some things do not depend for their

existence o n the existence of perception (are " p r i o r to p e r c e p t i o n " ) ;
perceptibles do so d e p e n d ; therefore some things exist w h i c h are not
perceptible. W h a t Aristotle here distinguishes as " p e r c e p t i b l e s " a n d
"subjects w h i c h produce p e r c e p t i o n " are both called perceptibles at De
Anima I I I 2. 426*23-4, the former i n the sense 'what is being perceived',
the latter i n the sense 'what is capable ofbeing perceived'. T h i s explains
(i) h o w Aristotle c a n defpnd the second premiss above (but are the two
senses of 'perceptible' true oidifferent things?), a n d (ii) that there is no con­
flict between that premiss a n d Categories 7. 7^35-^*12, w h i c h argues that
though the perceptible a n d perception are relatives ( c f " c a U e d w h a t they
are w i t h reference to one a n o t h e r " ) the former is prior (cf. Δ 15.1021 *26-
^3). B u t at the same time it makes Aristotle's conclusion worthless against
the argimient from contrary appearances, w h i c h claims to apply to
everything perceptible i n the Categories sense 'capable ofbeing perceived'.
"Sense-impressions": aisthemata,

II2
Г6 NOTES ioii*3

C H A P T E R 6

Γ 5 has not attempted to refute the thesis of Protagoras, but to under­

mine the arguments i n its favour. Aristotle now explains his objections
to undertaking refutation ( i o i i * i i - i 6 ) but thereafter undertakes it
{ i o i i * i 7 - b i 2 ) ; c f the relation 0 f r 4 . 1005^35-1006*11 to the rest ofthat
chapter.

i o i i * 3 . Comments on the distinction between "those who are c o n ­

v i n c e d " a n d "those w h o merely state these things" wiU be found in
the notes o n Г 5. 1009*16. Aristotle's response to those who desire the
criteria {Theaetetus 178 b) ofcorrectjudgements to be not only stated, as
at Г 5. 1010^1-19, but demonstrated b ambiguous: does he mean that
nothing is both a principle a n d demonstrable, or that nothing c a n be
both a principle and demonstrated? I s either response adequate, without
any indication that alternative principles are inferior to those ordinarily
accepted? F o r the question ' A r e w e Й1еер?' cf. Theaetetus 158 bc.
Aristotle's epigrammatic objection against "those who merely state
these things" has been taken i n v a r i o m ways, ( i ) ' T h e y demand the
right to contradict thenuelves, while at the same time saying something
('refute me') w h i c h is inconsistent w i t h that d e m a n d ' (BulUnger). T h i s
makes the right point, that since the conjunction of a statement of the
form 'p and not-/>' with its contradictory is itseU"ofthe form 'p a n d not.^',
its contradictory is not inconsistent w i t h it. B u t the point comes better
from (2) ' T h e y ask tw to refute them by establishing the contradictory of
what they say, but their o w n assertion is a setf-contradiction (which
implies that contradictories do not refute)'; for the same construction
see Г 4. 1006*5. (3) ' T h e y demand the right to contradict themselves, a
demand w h i c h from the outset contradicts itseU"' (Ross), i.e. t h e y a s k f o r
a licence w h i c h i f granted is not granted.

1011"17. W h a t foUows is oiuy a tentative interpretation of this diffi-

aut paragraph. T h e argument starts w i t h a n (invalid) reductio:
{a) everything imagined i& t r u e ;
{b) everything imagined is imagined by someone;
(i) so everything true is true relative to someone.
I n order to avoid the conclusion, Aristotle says, {a) must be emended to
{ai) everything i n ^ g i n e d by so-and-so etc. is true,
where ' e t c ' represents a specification oftime, sense-organ, m e d i u m , a n d
the like (*21-4). Without explaining w h a t the Protagorean woidd lose
by adopting the emendation, Aristotle next ^ i n t s out that a further

" S
ioii*i7 METAPHYSIGS Г6

consequence of not adopting it ( " i f they subject themselves to argument

but not on these t e m w " ) is to allow a n argument from {a) together with
{d) contraries c a n be imagined together (*25-8)
to
{e) contraries c a n be true together (*24-5).
I t remains to show that the same damaging consequence does not
foUow i f the Protagorean consents to reform his position b y adopting the
emendation of (л), a n d this seems to be the purpose of the possibly
corrupt sentence from •28 to ^ i . I f ( e i ) is taken as universally quantified,
then given {b) it entaib (a) a n d so does have the same consequence. T h u s
Aristotle seems to need, at least from *28, the interpretation
(02) aU the things imagined b y one given person etc. c a n be true
together
(e.g. i f a n y t h i n g looks both red a n d round to one a n d the same eye at one
a n d the same time etc., it c a n be both r e d a n d round together). E v e n
(fl2) wUl lead to {e) unless it c a n be shown that the correspondingly
emended form o({d) is false. *28-^1 argues that it is: the same things are
not ever imagined contrary " b y the same perception a n d i n the same
aspect of it a n d i n the same t i m e " . T h i s assertion has been previoiwly
used, i n Г 5. 1010^14-19, to destroy the argument for {d) given i n
Г 5. 1009*38-^12; a n d some conunentators have thought that it has the
same purpose here a n d that, since it is not for Protagoreans to under­
mine their o w n position, *28-^1 parenthetically repeats Aristotle's ob­
jection to that position. H e would i n that case be arguing, first, that
because truth is not relative {a) has to be emended (* 17-28) a n d , secondly,
that the defence of {a) i n its unemended form is inadequate (*28-^1).
B u t this reversion to Aristotle's earlier argument is needless, whereas
there is need for the reformed Protagorean to show that his emendation
of {a) does the work for w h i c h it was designed a n d w h i c h it wiU not do if,
as " t h e reasons previously m e n t i o n e d " h a d indiscriminately asserted,
contraries c a n be imagined b y the same perception etc. T h e conclusion
" s o that this would be t r u e " , sc. (02), is also more appropriate to the
Protagorean trying to defend the consistency of a new-found position
than to Aristotle for w h o m it would be a parenthesis i n a n already
parenthetical argument.
Some remarks must n o w be made about the textual problem raised
by *28-**i. T h e words " t h e reply is that they are imagined c o n t r a r y " ,
supplied i n the translation, are represented i n our text b y " b u t " (*fl//a'),
w h i c h leaves the sentence without a m a i n verb. O n e solution (Bonitz)
accepts this b a d g r a m m a r as conveying the sense given by the trans­
lation; another (Alexander) supposes that words to the same effect have
dropped out of the text, not indeed from the place where the translation

I14
Г6 NOTES 1011>17

inserts them, w h i c h would not account for the " b u t " , but from
after the first comma (*30), giving " [ t h e reply is that] even o n this
account . . Jaeger also puts the l a c u n a i n *30, but fills it differently:
'[we shall say that their conclusion is that what is imagined b y everyone
is true,] a n d o n this a c c o u n t . . . ' J a e g e r ' s filling imposes the 'parentheti­
cal' interpretation, discussed above. T h e other filling c a n be taken either
that w a y (so Alexander a n d Ross, 'our reply wiU be . . .') or the other
(Bonitz, 'their r e p l y w i U be , . . ' ) . A m o n g c h a m p i o n s o f t h e ' p a r e n t h e t i c a r
interpretation there is disagreement as t o w h e t h e r the words " a n d o n
this account everything is equaUy false a n d t r u e " are (Alexander,Jaeger)
or are not (Ross) part of the reply.
" O n this account necessary" (^1-2): not on account of the emendation
of (a), w h i c h according to *21-4 avoids the necessity of making truth
relative, but (it seems) o n account of the unsatisfactoriness of that
emendation—^viz. 0f(a2)—to anyone detenninedtomaintainProtagoras'
thesis.
T h e r e are difficulties i n the paragraph w h i c h the foregoing interpreta­
tion leaves unresolved. W h y does Aristotle present without separation
two different reductiones of the Protagorean thesis that everything i m a g ­
ined is true, the arguments from {a) a n d {b) to {c) a n d from (a) a n d {d)to
{e)? W h y does he think the former argument valid? W h y does he think
that Protagoreans wiU find the conclmion of the latter argument u n ­
acceptable? C a n b1-3 reaUy m e a n that the emended thesis> (02), is too
weak to satisfy Protagoreans? I f s o , w h y does Aristotle trouble, i n *28-^1,
to absolve it from the consequence attending the unemended {a) ?
T h e experiment with the crossed fingers is described a n d discussed i n
Ross's edition of the Parva Naturalia, o n De Insomniis 2. 460^20-2.

1011^4. Aristotle n o w completes the demolition ofProtagoras' thesis

by refuting, i n three arguments, its alleged consequence that truth is
relative. ( I ) T h e first sentence draws from that consequence the absurd
conclusion that everything true is believed. ( I I ) Compression a n d i n ­
adequate terminology obfuscate the argument i n the second a n d third
sentences, w h i c h m a y perhaps be paraphrased as foUows. {a) W h a t is
e.g. double c a n be relative to more than one thing, evendoublemore than
one thing, but the relation being double has just one converse relation,
being half ("relative to . . . something definite", cf. Categories 7. 6^28-
7b14); it does not also have the converse being equal ( " t h e equal is not
relative to the d o u b l e " , i.e. 'x is equal to У a n d y is double x* are not
equivalent). I n general, then, every relation has just one converse.
(b) Both *is a m a n to' (according to the doctrine that truth is relative)
and *is thought to be a m a n by' express relations whose converse is 4hinks
to be a m a n ' (are " r e l a t i v e to that w h i c h has a n o p i n i o n " ) , {c) Therefore
'is a m a n ' a n d 'is a n object of (such a) thought' express the same relation.

115
1011b4 METAPHYSIGS Г6

{d) Therefore m e n are essentially objects of thought, {e) Therefore men

are not essentially thinkers, w h i c h is absurd. Aristotle does not justify the
step from {d) to {e). ( I I I ) N o t only *is a m a n ' but all other predicates will
express relations w i t h (generically) the same converse *thinks to be (so
a n d so)'. S o the latter relation w i l l have not one but infinitely many
converses.

1011^13. T h e discussion has indicated: that P N C is the firmest opinion

of aU {Г 3. 1005^8-34), the objections to " m a k i n g this statement", sc.
rejecting P N C {Г 4, 6), a n d the arguments adduced b y others for its
rejection {Г 5). A s at Г 5. 1009*6 Aristotle ignores the fact that through
most of Г 4 a n d most of Г 5 he has addressed himself to theses more
vulnerable than the mere contradiction of P N C . O n *contrary', 'lack',
and *denial' s e e J 10 a n d I 4.

CHAPTER 7

T h e arguments i n defence ofthe principle ofexcluded middle ( P E M ) are

stated i n summary form a n d occupy less than a tenth of the space given
to P N C i n Г 3-6. T h i s is not because Aristotle thinks P E M less doubtful,
for nothing c a n be less doubtful than P N C {Г 3. 1005^22-4). T h e reason
is partly that doubt about P E M h a d been expressed by none ofAristotle's
predecessors except—so he thinks—Anaxagoras (1012*24-8); partly
that his diagnosis of the P E M - s c e p t i c ' s state o f m i n d , at 1012*17-24, is
the same as that of the P N C - s c e p t i c ' s , a n d so requires no new discussion.
T h e diagnosis shows that h e does not, either, share the inclination of
some modern logicians to regard P E M as more doubtful than P N C .

1011^23. T h e formulation of P E M is incautious. Although Aristotle

says that, given a predicate F a n d a subject it is necessary either to
assert F of x or to deny F ofx, he cannot reaUy mean to exclude the fur­
ther possibilities of (i) making no judgement as to whether x is F, (ii) per­
forming such non-declarative operations as asking whether x is F or
requesting that x be F, W e must take it that these latter operations are
not " i n the middle o f a c o n t r a d i c t i o n " : that is, do not fall between a pair
offormulae asserting a n d denying F ofx (De Interpretatione 6. 17*33-8).
W h a t are the conditions under w h i c h a n operation is i n the middle of
a contradiction? T h e answer w h i c h suggests itself is: when it involves
rejecting SLs/alse both halves of the contradictory pair. I f s o , Aristotle's
opponent asserts the possibility ofits being false both that x is F a n d that
X is not F , a n d thus, by a part of the definition of falsity w h i c h not even he
wiU wish to give up, the possibility that x is neither F nor not F, Whether,
conversely, Aristotle would accept that his version of P E M entails that it
is necessary that x is either F or not F depends o n his attitude to the
principle that a n assertion or denial that is not false must be true;

ii6
Г7 NOTES 1011b23

according to one interpretation De Interpretatione 9 , while allowing that

two singular contradictories about the future cannot both be false, resists
the conclusion that (at every time) one of them must be true. Nevertheless,
it is likely that Aristotle would have accepted the necessity of *x is either
F or not ; a n d the still more general form *either p or not-^' is implied by
De Interpretatione 9 . 1 9 * 3 2 .

1011**25. T h i s first argument contains {a) a definition of 4rue* a n d

'false', {b) a n inference from the definition ("so that . . . " ) , {c) the
opponent's assertion, {d) a suppressed conclusion. E a c h step саШ for
comment.
{a) " S a y that that w h i c h is is n o t " etc.: i.e. say ofsomething which is
that it is not etc. " I s " could m e a n *exists', *is so-and-so', or *is the case'
(and " i s n o t " s i m i l a r l y ) ; the last alone makes the definitions cover all
truths a n d falsehoods, as Aristotle's argument requires.
(c) " B u t it is s a i d " , sc. by the opponent. T h e sentence is ambiguous
between 'he does not say either that that w h i c h is is not or is, or that that
which is not is not or is' a n d 'he denies both that that w h i c h is is not or is,
and that that w h i c h is not is not or is', but the former is strong enough to
yield Aristotle's conclusion.
(d) T h e conclusion to be supplied is presumably 'what the opponent
says is neither true nor false', w h i c h foUows validly from {a) and (c).
{b), which I assume means '. . . says of a thing that it is or that it is·
n o t . . .', is problematical. I t does not seem to assist the argument, for (d)
foUows from {a) and {c) but not from {b) and {c), Moreover {b) does not
foUow from {a) without the further question-begging assumption that
what goes for that-which-is and that-which-is-not goes for everything. F o r
these reasons some editors have preferred to add a 'this' with Alexander
— " s o that he also who says that this [sc. what is in the middle of a
contradiction] is or not wiU have the truth or be in e r r o r " , which in
addition gives more force to " a l s o " . Against this reading Ross brought
the objection that it does not foUow from {a) that it must be true or false
to say 'is' or 'is not' o f w h a t is in the middle of a contradiction, since what
is in the middle of a contradiction is precisely neither what is nor what is
not. T h i s is an objection w h i c h Ross should have brought against the
M S . reading. Against Alexander's it is not necessarily effective, because
it assumes that the word " t h i s " refers to the subject of the supposed
middle operation, the thing said to be neither F nor not F; whereas the
reference might be to the operation itself, which the opponent does sup­
pose to 'be', i.e. (here) to exist. I f s o , {b) would be, not a step i n the refuta­
tion ofthe opponent, but an aside to the effect that his thesis is refutable.
Although the argument as a whole is valid, it does not damage the
position of the opponent, who has a choice of two replies. (A) Accepting
the implication that, if there is a third operation between asserting a n d

li7
lOIlb25 METAPHYSICS Г7

denying, the products of that operation cannot be either true or fabe, he

nught postulate a third truth-value for them to take. B u t this move is
plausible only if the third truth-value is *designated', i.e. i f predicating
it of a proposition is a w a y of assenting to that proposition. (B) A l t e r n a ­
tively, the opponent might reject premiss {c) on the grounds that the
statement he produces—say 'Socrates neither is sick nor is not s i c k ' —
although different from the denial *Socrates is not sick' is nevertheless
a denial and so does say that that w h i c h is not is not. Moreover, this
second type of denial would meet the condition for being i n the nuddle
of a contradiction, since its truth would involve the falsehood of both
'Socrates is sick' a n d 'Socrates is not sick'.
A t this point it m a y be worth indicating how Aristotle would have
dealt w i t h examples alleged to illustrate the fakity of P E M . (i) 'Socrates
is neither sick nor not s i c k ; he is d e a d ' : Aristotle says that ifSocrates does
n o t e x i s t he is not sick {Categories i o . 13^29-33, but contrast De Inter­
pretations I I . 21*24-8). (ii) ' T h i s stone is neitherjust nor n o t j u s t ; for it is
incapable ofjustice.' Aristotle agrees that stones c a n neither have nor
lack justice, but he holds that the false c l a i m that they lack justice must
be made i n the form 'stones are u n j m t ' or *stones are not-just'. These
claims are false assertions, b u t t h e denial 'this stone is n o t j u s t ' is true {De
Interpretatione 10. 19^5-20*3, Prior Aruilytics I 46. 51^22-5). (iii) ' T h e
present king of F r a n c e is neither wise nor not wise; for the expression
" t h e present king of F r a n c e " is not being, or on the occasion of utterance
cannot correctly be, used to refer to anyone': Aristotle nowhere comments
on this kind of case, but he might have held that the sentence makes no
assertion, or asserts nothing " o f o n e t h i n g " , a n d is therefore no exception
to the nUe that we must assert or deny one thing of one thing (cf. De
Interpretatione 8 ) . (iv) 'Socrates neither wiU be nor wiU not be s i c k ; his
future state of health is not determined.' O n one interpretation, De
Interpretatione 9 makes the lack of inevitability w h i c h characterizes most
future events a reason for denying present truth to e.g. 'Socrates wiU be
sick' a n d 'Socrates wiU not be sick'. B u t we must notice, first, that even
that concession need not be incompatible w i t h the thesis that there is
nothing i n the middle of a contradiction, i f that means that there is
nothing whose truth involves the falsehood of both members of a contra­
dictory p a i r — s i n c e nothing suggests that he held such future contra­
dictories to be already false; a n d , secondly, it appears that, i f h e did make
the concession, he thought it compatible with the truth of the disjunctive
proposition *Socrates either w i U b e or wUl not be sick' {De Interpretatione
9. 19*32), even though these are rwt compatible according to the defini­
tion of truth given i n the present chapter.

1011^29. W h a t alters i n respect ofFness must alter from being F to not

being F, or vice versa. Aristotle distinguishes two ways i n w h i c h the

ii8
Г7 NOTES 1011b39

relation of being F to not being F ndght be pictured so as to admit an

exception to this principle: as the relation between dark (or black), and
pale (or white), w h i c h are contraries having "something intermediate
between t h e m " (Categories 10. 12'9-25), or as the relation between a m a n
and a horse, w h i c h are not contraries [Categories 5. 3^24-7). Aristotle
argues that neither analogy secures the result w h i c h his opponent wants.
Although the intermediates between dark and pale are possible starting-
and stopping-places for change, any intermediate between pale and not
pale would not be; and intermediates between non-contraries are never
possible starting- or stopping-places for change (e.g. nothing could change
into or from a mule). Aristode adds that intermediates always are possible
starting- and stopping-places for change ( " i n fact it evidentiy always
does" sc. alter), so that the man-horse analogy cannot indicate a " w a y
in w h i c h " anything is i n the middle of anything.

i o i 2 * 2 . T h e argument is: thinking something that is true or false is

asserting or denying; [no middle operation is asserting or d e n y i n g ] ;
therefore no middle operation is thinking something that is true or false.
A s i n 1011^25-9, it is assumed (A) that any middle proposition would
have to be true or false a n d (B) that no middle proposition c a n be
a n assertion or denial. F o r the description of truths and falsehoods as
" c o m p o u n d i n g " c f Categories 4. 2"4-10, De Interpretatione 1. i 6 * i 2 - i 8 ,
£ 4 . 1027^17-23.

101a*5. Aristotle does not m e a n that i f a n y middle is true aU must be

true, but that if any makes sense aU must make веше. H e derives two
consequences, (i) I t wiU make sense to say that a proposition is neither
true nor not true, a n d that one who believes it is neither right nor not
right (Aristotle's c l a i m is actually bolder: these things will sometimes
happen), (ii) T h e r e w i l l be a middle between existing and not existing (the
reference to coming to be a n d destruction shows that " t h a t w h i c h i s "
must now be taken existentially). Aristotle does not mean to deny that,
even as things are, there is a " k i n d ofalteration apart from coming to be
a n d destruction", e.g. change of colour; he means that there wUl have
to be a new kind i n addition, standing i n the same relation to coming to
be a n d destruction as change to or from the intermediate between white
and not white stands to ordinary changes of colour.

1 0 1 3 ' 9 . F o r contraries related i n such a w a y that the denial o f o n e

implies the other (i.e. contradictory opposites) see Categories 10. 11^38-
12*9. Aristotle's thesis is that the propositions:
{a) odd is the contrary of even,
{b) 'not odd' implies 'even',
{c) at least one number is neither odd nor not odd,
IOI2*9 METAPHYSIGS П

make a n inconsistent triad. T h e reasoning seems to be: {a) a n d {b) show

that 'even' m a y be defined as *not o d d ' ; by that definition {c) is equivalent
to
{d) at least one number is neither odd nor even,
w h i c h is inconsistent w i t h {b); therefore {c) is inconsistent w i t h the con­
j u n c t i o n of(fl) a n d {b), T h i s argument begs the question, because {a) and
{b) entail that *even' is equivalent to 'not o d d ' only on the assumption
that 'odd' a n d 'not odd' exhaust the field.

i o i 2 * i 2 . T h e interpretation of this argument depends on the queer

phrase " t o deny this [the middle] i n relation to the assertion a n d d e n i a l " ,
( i ) Bonitz understands: to deny both the disjunction o f m i d d l e a n d
assertion a n d the disjunction of middle a n d denial. T h e argument then
is: if between ' F ' a n d 'not F ' is the middle ' G ' , there w i l l be further
middles between ' G ' a n d ' F ' a n d between ' G ' a n d 'not F\ a n d so ad
infinitum, ( 2 ) A simple emendation, adding 'to', would give 'to deny this
<which is> i n relation to the assertion a n d denial', for w h i c h the sense
proposed is: to deny the proposition w h i c h is middle relative to the
original assertion a n d denial. T h e argument then is: i f b e t w e e n ' F ' a n d
'not F ' is the middle ' G ' , there wiU also be the new denial 'not G\ (3) A
less easy emendation expands 'apophesaV ('deny') into 'apophenasthaV
(*declare'), giving the sense: to declare the middle i n both its affirmative
a n d negative forms (cf. PosteriorAnalytics 1 2 . 7 2 * 1 1 - 1 2 ) . T h e argument is
then the same as under ( 2 ) . (4) Alexander understands: to deny both the
middle taken as a n assertion a n d the middle taken as a d e n i a l — t h a t is, the
denial of the middle. T h e argument then is: if between Ψ' a n d 'not F ' is
the middle ' G ' , between that a n d 'not G ' wUl be the further middle ' Я ' ,
a n d so ad infinitum. (4) leads to the same result as ( 2 ) a n d (3). But aU four
of these interpretations face linguistic or textual objections.
" F o r the substance is something else than i t " : the n e w denial (whether
'neither G nor F\ 'not G\ or ' Я ' ) differs i n more than name from what
it denies; for 'substance' meaning 'being' ci.F^. 1007*26.

1012*15. 'Socrates is not pale' (i) denies only that he is pale, not e.g.
that he is medium-dark, a n d (ii) says no more than w h a t it denies ("its
not being is a d e n i a l " ) . So 'Socrates is not pale' is compatible w i t h
everything other than Socrates' being pale. So the aUeged middle
utterance must either be qompatible w i t h 'Socrates is not pale' or else
entail 'Socrates is pale'. I n neither case wiU it fulfil the condition for
being i n the middle of a contradiction.

1012*17. F o r the two " d e r i v a t i o n s " of the opponent's opinion c f

r * 5 . 1 0 0 9 * 1 6 - 2 2 . " T h e original step is a d e f i n i t i o n " : c f i ^ 4 . 1 0 0 6 * 1 8 - 2 6
a n d 8. 1012^5-11.

120
 NOTES x012*24

1012^24. Anaxagoras' alleged thesis is that it is (not only possible but)

necessary neither to assert nor to deny: *good' is of course a representa­
tive for aU other predicates. Against the objection that *neither good nor
not good' is itself something true, Aristotle must refer back to 1011^25^.
" M a k e s everything t r u e " : a n d fake t o o , r * 8 . 1012*29-^2.

CHAPTER 8

1012^29. T h e theses that everything is true a n d that nothing is true here

appear for the first time " s i n g l y " , viz. not conjoined with each other, a n d
" a b o u t everything", e.g. not linuted to things " i n process o f a l t e r a t i o n "
(Г 5. ioio*8). " I f the former are impossible . . . " : i f the unconjoined
theses are impossible, so is their conjunction. T h e argument of 1012^13-
18 could be generalized to prove the converse impUcation, sojustifying
" p r a c t i c a l l y the s a m e " . T h e conjunction is equivalent to 'y/x^F(Fx &
—Fx)\ the strong denial o f P N C (Г4. 1007b18-1008*2); its attribution
to Heraclitus seems once more to confuse it with the weak denial {Г
3. 1005b23-5).
Against these new a n d easy targets Aristotle aims some of the argu­
ments already used i n Jr4-7. " W e must base discussion on a definition",
not, as ίηΓ^, ioo6*3i-^34 ( c f ^ 7 . 1012*21-4), the definition ofsome
specimen " n a m e " chosen by the opponent, but of4rue' a n d *fake'. T h i s
was the procedure employed i n defence of P E M {Г 7. 1011b25-9,
1012*2-5); defence o f P N C it was previously spurned {F 4. 1008*34-
^2). b13-22 states the argument caUed peritrope; cf. Г 4. 1008*28-30,
Theaetetus 170-1. I n ^9 the text is uncertain, but not the sense.

1012^22. " E v i d e n t l y this a l t e r s " : Aristotle's doctrine is not u n ­

reasonable i n holding that truth-values c a n change, but is unreasonable
if it holds that a n assertion changes its truth-value w h e n its asserter dies.
" I f e v e r y t h i n g is c h a n g i n g " , sc. i n aU respects; see /"5. 1010*15-25. " I t is
necessary that that w h i c h is alters", i.e. what alters must be something
that i s ; for, Aristotle argues, if it alters from F to G, there must be some
time at w h i c h it is F a n d some time at w h i c h it is G (cf Physics V I
5. 235^6-30). B u t the argument fails to show that these times are stretches
oftime, and so does not establish that change implies rest (see also Physics
V I I I 3. 253b6-254*1). T h e "first c h a n g e r " is G o d (Л 7); " w h i c h is
always changing the things that c h a n g e " refers, presumably, to the outer­
most sphere of the heavens w h i c h , though Aristotle does not here say so,
is itself constantly i n change (viz. movement, Л 7. 1072*21-3).

I2I
 METAPHYSICS

METAPHYSICS BOOK DELTA

I N T R O D U C T O R Y N O T E

Δ consists of a n examination of the different senses—some i n technical

use b y Aristotle, some not—of about 34 words and their cognates. A t
Г 2. 1003b33-1004*2 the study of the "forms of thing-that-is" a n d " o f
thing that is o n e " , including such concepts as the same a n d similar, was
referred to the "Selection of C o n t r a r i e s " . T h i s is perhaps to be identified
with a lost book. About Contraries, to w h i c h there are references i n the
ancient commentators (see The Works of Aristotle translated into English,
vol. x i i , 109-114), but some o f t h e material may be reused i n J 9-10.
Г 2. ioo4*io-20 mentioned plurality, other, dissimilar, unequal, difference,
a n d contrariety; a n d 1005*12-17 added complete, prior 2Lndposterior, genus,
form, whole and part. T h e discussion of all these, as well as that which is,
one, and substaru:e, falls to " o n e discipline", namely metaphysics (1004*32¬
3); and all but two ofthem are treated i n ten of J ' s thirty chapters. After
the examination of axioms i n Г 3-8, Δ thus comes naturally as a further
exercise i n p r i m a r y philosophy. Nevertheless, many have thought that,
in Ross's words, the book " i s evidently out of place where it i s " . T h e r e
are three reasons for this uneasiness, (i) M a n y of the words discussed in Δ
are not listed i n Г as proper to 'philosophy', although nearly aU seem to
meet its requirement ofubiquity ofapplication. (ii) Aristotle's references
to Δ, in the rest ofthe Metaphysics a n d elsewhere, are to " t h e treatment of
the number of ways i n w h i c h things are caUed what they a r e " (or a
similar formula, e.g. E 4. 1028*4-6); a n d one later writer lists a n A r i s ­
totelian work under the same title. K 1-8 summarizes ВГЕ, i n that
order. A U this suggests a separate origin for the book, (iii) T h e method of
asking e.g. 'how many senses has the word " f a l s e h o o d " ? ' , rather than e.g.
*how is falsehood possible?' may seem more appropriate to the trainer in
philosophical navigation than to the independent explorer. Δ may, then,
be a m a n u a l ofseparate origin, incorporated into Uw. treatise on 'primary
philosophy' by Aristotle himself or a n editor.
Some ofAristotle's distinctions are not so m u c h ofsenses as of criteria,
uses or applications of a word (see e.g. Δ 1. 1013*16-17 a n d note, Δ
7. 1017*24-7 a n d note). H i s own labels for them are o f t w o types, used
interchangeably: (i) adverbs *manyhow', 'twohow', etc., translated ' i n a
number of ways', ' i n two ways', etc., (ii) the noun Uropos', translated
'sense' (but 'manner' at Δ 6. 1015^33, E 1. 1025^15, *type' at Г 2.
1004^24).

122
 NOTES 1012b34

CHAPTER 1

i o i a ^ 3 4 . " W e call a n o r i g i n " : literally 'origin is called', cf. " C i v i l

interests I call life, liberty, health a n d indolency of b o d y " ( L o c k e ,
Letter conc. ToUration, ed. G o u g h , i a 6 ) . " P o i n t " : there is no noun i n
the Greek, here or elsewhere (except 1014^8, 1016^17-31).

1013"1. C f Physics I I , where Aristotle distinguishes between those

origim or elements w h i c h are "clearest and most intelligible to u s " , e.g.
(commonplace facts about) complex macroscopic objects, a n d those
w h i c h are "clearest a n d most intelligible by n a t u r e " , e.g. (descriptions of)
the components of such objects.

1013"4. " S o m e believe it is the h e a r t " : among them Aristotle, see De

Gerwratione Animalium П 4. 740*17-19.

1013"10. T h e word 'arche' h a d two broad meanings i n G r e e k : 'begin­

ning', 'origin', 'starting-point', to w h i c h answers the nuddle voice of the
verb, 'archesthai' (translated 'begin' at *3 and * 8 ) ; and 'rule', 'authority',
'office'—and i n the plural often 'officers', 'authorities'—to w h i c h answers
the active 'archein'. Aristotle's treatment attempts to relate the two
meanings: rulers are authors or initiators o f c h a n g e . A " d y n a s t y " is a
type of oligarchy. Politics I V 5. 1a92^5-10. T h e plurals 'dynasties' etc.
c a n , like 'arche', be used either of a type of government or of the group
w h i c h so governs: c f our use of 'the government'. T h e latter better fits
" t h a t at whose decision" (or ' w i l l ' , 'prohairesu'), but, perhaps influenced
by the former, Aristotle allows himself to say that skilk decide to get
things changed.

1013'14. " T h e point from w h i c h one first gets a c q u a i n t e d " with a thing
is not the same as " t h e point from w h i c h it is easiest to learn " ( * 3 - 4 ) ; for
the former is, but the latter is not, " o r i g i n of the actual t h i n g " . T h e
reference to demoiistratiorw, i.e. things demonstrated, shov*fS that
'getting acquainted' here has the sense 'proving', not 'learning': cf.
A 3. 983*25-6. A t Posterior Armlytics I 10. 76^23-34 a principle is u n ­
provable, a hypothesis provable but accepted without proof; here thc
words are used indifferently of anything accepted without proof

1013*16. " I n the same number of w a y s " : i.e. over the same range of
application, for Aristotie denies that 'origin' and 'cause' are " i n d i c a t e d
by one f o r m u l a " {Г 2. 1003^24-5). E v e n so, from the fact that every
cause is a n origin it does not foUow that the two words apply to the samc
things, and i n fact nothing under 'cause' i n Δ a corresponds with e.g. thc
second serue of 'origin'.

1013*17. " T h o u g h t " belongs under the fifth sense, for men's thoughts
(e.g. that something is attractive) as well as their choices originate

123
ioi3*i7 METAPHYSIGS Δί

action. " W h a t a thing is f o r " : Aristotle m a y m e a n that we prove e.g. that

Callias is i n Thebes by discovering his purpose there (demonstrative
'acquaintance' with a fact), or, more likely, that we understand e.g. what a
lathe is by discovering its purpose (making ourselves acquainted w i t h
what a thing i s ) ; i n either case the purpose m a y originate change as well
as knowledge or understanding—not changes i n the lathe or i n Callias'
situation i n Thebes, but the changes whereby the lathe came to exist (it
was made to serve its purpose) and Callias came to be i n Thebes (he
went there for a purpose).

CHAPTER 2

E x c e p t for a few very minor divergences this chapter is identical w i t h

Physics I I 3. 194b23-195b21; A 3. 983*33-b1 refers for a n adequate
discussion of cause to " t h e books about n a t u r e " , not to J . Physics I I
7. 198*14-16 implies that any true answer to the question ' w h y ? ' gives
a cause {aiiia, aition). H e n c e 'explanation' or 'reason' is often closer to
the sense ofthe Greek words; but a n aitia or aition is generally a reason for
being ('ratio essendi') not a reason for thinking ('ratio cognoscendi', but
see e.g. E 2. 1026^24-31).
T h e traditional names for Aristotle's four causes are material, formal,
motive (or efficient), a n d final. T h i s fourfold classification reappears at
A 3. 983*24-^1. T h e material cause is omitted at DeAnima I I 4.4i5t*8-io,
a n d replaced at Posterior Analytics П 11. 94*20-36 by " t h a t w h i c h being
so, it is necessary for this to b e " , i.e. a necessitating condition (Aristotle's
example is ' w h y is the angle inscribed i n a semicircle a right-angle?
because it is h a l f two right angles'). A 4. 1070^11-30 omits t h e f i n a l
cause a n d splits the formal into form a n d lack (sc. ofform). AtPhysics I I
7. 198*24-33 it is argued that formal, final, a n d motive causes often
" c o m e to the same t h i n g " . W h a t makes Socrates two-footed is (i) his
being a m a n (formal cause), (ii) his father, who must have been a m a n
(motive cause); these are " t h e same i n f o r m " , sc. both m e n although not
the same m a n ( c f A 4. 1070^30-5). F o r m a l a n d final cause m a y be even
numerically the same, for according to Aristotle a natural object has its
nature or substance only when" its development is complete (see Δ
4. 1015*3-5) a n d natural development is not just towards but 'for'
completion or fulfilment (the fullest argument for this is Physics I I 8; c f
also De Anima I I 4. 415^15-21, De Generatione et Corruptione I I 9. 335^6,
Я4. 1044*36-b1).
Aristotle's discussion of cause is no longer influential, chiefly because
he does not raise H u m e ' s problem as to the difterence between 'post hoc'
a n d 'propter hoc'. T h a t problem treats causes as anterior events, but
(i) final causes, if events, occur later than their effects; and (ii) although

124
 NOTES ioi3*34

a final cause may be a n event or outcome (*for the sake of his health*) it
may also be a beneficiary ('for his sake*—Aristotle makes this distinction
at De Anima I I 4. 415^20-1), a n d although a motive cause m a y be a n
event {Posterior Armlytics I I 11. 94*36-^1 " w h y did the Athenians get
involved i n the Persian W a r s ? . . . because they raided Sardis with thc
E r e t r i a n s " ) it m a y a k o , as mostly i n the present chapter, be a n agent.

1013*24 {Physics I I 3. 194b23). "Constituent out o f w h i c h " = material

cause. Bronze is the cause of a statue i n the sense of being cause of the
statue's having such a n d such properties: c f Z 17. 1041*10-11 * " w h y ? "
is always looked for i n this way, " w h y does one thing hold good ofsome
other t h i n g ? " ' . De Anima I I I 5. 430*10-25 shows what Aristotle has in
m i n d : i f a statue is of—i.e. manifests—a brownish colour, that is caused
by {a) its bronze, as matter, and {b) the light shining on it, as external
{A 4. 1070^22-3) origin ofchange. T h e statue's matter has {Z 15. 1039^
2 9 - 3 0 ) , or even is {De Anima I I 1. 412*9), the capacity to acquire and
manifest varying properties; it is therefore a kind of origin of change
(Δ 12. 1019*19-23), but passive, inferior, a n d internal. T h i n g s immaterial
in the modern sense m a y still have matter, i.e. subject-matter or materials,
as vowels and consonants ("elements" 1013^17) are the materials from
w h i c h we make up syllables, and hypotheses the materials (^20) from
w h i c h we get conclusions.

1013"26 {Physics I I 3. 194^26). " F o r m and p a t t e r n " = formal cause.

I t is not uncommon for Aristotle to write as if a thing's form could be
identified with the formula w h i c h expresses it (cf Δ 6. 1015^25). H e h a d
two uses foi the notion of formal cause, w h i c h he did not distinguish.
(i) H e appears to have assumed that being F causes a thing to be G
given only that F s are always or usually Gs, or i n other words that the
conjunction ofFness and Gness is not a coincidence (see notes on Δ 30.
1025^14); thus Physics I I 3. 195^23 tells us that the intermediate cause
of some particular man's building a house is his being a housebuilder.
(ii) Besides citing forms as causes of the properties of individual things,
Aristotle more plausibly casts them as possible answers to general
questions like 'why do housebuilders build houses?' 'why do octaves
span eight notes?' T h e first of these is admittedly more likely to ask for
a final cause but the second, if about something changeless, c a n be
answered only by saying what a n octave i s : " ' w h y ? ' , in the case of
changeless things, reduces i n the end to 'what is i t ? ' " {Physics I I 7. 198*
16-18, c f A 3. 983*28).

1013*29 {Physics I I 3. 194^29). " F i r s t origin ofalteration or r e s t " =

motive cause. " F i r s t " suggests that Aristotle does not regard means as
motive causes: c f our distinction between agent and instrument, a n d *36
" w h a t effected the change was something else", sc. than the means.

125
IOI3*29 METAPHYSICS Δ2

' O r i g i n of alteration or r e s t " , c f 1013^25, E 1. 1025^20-1: w h e n a heavy

body falls, its 'nature' is cause of the downward motion; when it reaches
the ground, the ground causes it to continue thereafter at rest. Both
causes must, according to Aristotle's mechanics, be operative at every
moment of the state they cause.

i o i 3 " 3 2 {Physics I I 3. 194^32). " T h i n g ' s fulfilment" = final cause.

T h e r e is a well-known ambiguity i n this notion, w h i c h Aristotle brushes
aside at ^27-8 with the remark " l e t us assume that there is no difference
i n calling it good or imagined g o o d " . A t Posterior Analytics I I 11. 94^8-26
he offers ' i n order to be healthy' as the answer to two questions: ' w h y
does one take a w a l k ? ' a n d 'why should one take a walk after d i n n e r ? '
But the truth-conditions ofthese answers are different: the answer to the
second question will be true only if walking is good for health, the answer
to the first only i f the walker thinks ( " i m a g i n e s " ) it good. Aristotle
leaves it unclear i n w h i c h ofthese senses it m a y be asserted that " n a t u r e
w o r b for the sake of something" a n d "because it is better s o " {Physics
I I 8. 198b17): on the one h a n d nature does not deliberate (ib. 199b26-8),
but on the other she does make mistakes (ib. 199*33-^1). T h e difference
between 'good' a n d 'imagined good' is discussed i n Nicomachean Ethics
I I I 4.

1013*35 {Physics I I 3. 194^35). A t *35 " a s a means t o " is literally ' i n

the nuddle of' (so translated i n Г 7), showing that Aristode has i n m i n d
a 3-term relation i n w h i c h B is the means b y w h i c h action or thing A
achieves fulfilment C. A, then, is the "something else" w h i c h effected the
change. A doctor (Л), for instance, employs instruments or performs
operations {B) i n order to secure his patient's health ( C ) . I n this situation
it will be true to say both (i) that B, as well as C , is a final cause o f Л —
a fulfilment, although not a " c o m p l e t e " fulfilment {Nicomachean Ethics I
7. 1097*25-8)—and (ii) C is the final cause of B, as of A. Evidently
Aristotle's point is (i).

1013b3 [Physics I I 3. 195*3). " N o t coincidentaUy": see 1013^28ff.

Aristotle's assumption that the same thing cannot have multiple causes
i n the same " s e n s e " m u c h underrates the complexity of the concept of
cause. I n practice he is not so rigid: see 1013^1 i - i 6 a n d the notion o f a
joint-cause or contributory cause, e.g. at DeAnima I I 4. 416*14. " T h i n g s
m a y also be causes of one a n o t h e r " : c f A 3. 983*31-2,
T h e last sentence of the paragraph does not illustrate Aristotle's
contention that the sarru thing m a y be the cause of contraries, for w h i c h
he must rely o n the first half of the penultimate sentence. H e is really
making two concessions: (i) that one cause c a n have different results o n
different occasions, as w h e n the helmsman causes shipwreck by his
absence or non-shipwreck by his presence; a n d (ii) that one result c a n

126
Δ2 NOTES 1013"3

have different causes even on the samc occasion, as when thc motive
cause of sliipwreck is ahernatively given as the helmsman's absence or
simply as the helmsman. " H o l d responsible" translates the vcrb 'ailiasthai',
from the root of 'aitia' but usually meaning 'blame'.

1013^16 {Physics П 3. 195'15). T h i s sums up the first four paragraphs.

i o i 3 ^ a 8 [Physics П 3. 1 9 5 ' 2 6 ) . T h e rest of the chapter deals with a

different classification ofsenses of 'cause', the previous senses [Щ, ^ifi)
being now referred to as forms. Aristotle's procedure is puzzling. .Since
the same individual might be both doctor a n d man-of-skill (the latler
being wider than the former) and since, in spite ofthat, they are said to
be causes ofhealth i n different senses or ways, it must be that the primary
cases o f cause are kinds of thing, not individuals, and the primary cases o(
causal statement are class-relational, not singular. E v e n so, it is odd lo
say that 'doctors cause health' a n d 'men-of-skill cause health' use 'causc'
in different ways, or even mention different kinds of cause. Evidently
Aristotle's meaning is that the predicate 'cause health' (though doubtloss
univocaI) attaches in different wa)-s to the subjects 'doctors' and 'men-of-
skill': for causing health indicates what it is to be a doctor but not what it
is to be a man-of-skill ( s e e ^ i^,E 2. 1 0 2 6 ^ 3 7 - 1 0 2 7 * 2 ) .
" W h a t includes any ol' tbc particular causes", i.e. their genera (in
Δ 2 3 a n d 2 6 the verb is translated 'contains'). Ross's translation construes
differently: " w h a t includes so-and-so [is cause] oftheparticular[effectsj";
and this must be right if the phrase is to have the same meaning as llie
variant Greek at Physics i 9 5 ' ' 3 - ; , liut taking the Metaphysics phrase on its
own, its word-order favours the version adopted.

1013^34 [Pliysics П 3 . 1 9 5 * 3 2 ) . Both doctors a n d men-of-skill are p p . 2 1 8 - 1 g

causes ofhealth i n their own r i L ' l i t . and so not c o i n c i d e n t a l l y ( J 18. 1 0 2 2 "
2 5 - 9 ) . I n showing that Polyclitus is coincidentaUy cause of a statue
Aristotle uses a form ofargument which appears to license the inference
'men a r e animals: being Polyclitus is a coincidence f o r a m a n ; therefore
Polyclitus is coincidentaUy an a n i m a l ' ( c f . PriorAnalytics I 11. ЗI^^I2-2o).
I f , on the other hand, he had argued from the c o n v e i s e minor premiss
'being a statuemaker is a coincidence for PolycIitus', that would have
licensed the equally unwelcome inference 'statuemakrrs cause staturs;
it is a coincidence lor a man-of-skill to be a statucniakcr; therefore
men-of-skiil cause statues coincidentaUy' ( c f PriorAnalytics I 9. 3t)"37-''1).
Thisdifficultyis not to be avoided by replacing 'Polyclitus' with a general
term such as 'the pale'.
I n citing m a n a n d a n i m a l as " i n c l u d i n g " Polyclitus Aristotlc (i) over­
looks the difference between this relationship, of class-menibership, a n d
that between doctor and man-of-skill, w h i c h is ofclass-inchision (cf the
two uses of 'particular' i n 1 0 1 4 * 1 7 a n d 1 0 1 4 * a 1 ) ; aud (ii) ignores the

127
ioi3^34 METAPHYSIGS Δ2

objection that m a n a n d a n i m a l include not only the coincidental cause

Polyclitus but also the non-coincidental cause statuemaker. T h e pale a n d
the artistic are " m o r e r e m o t e " causes t h a n Polyclitus evidently because,
unlike h i m , they cannot directly coincide i n statuemaker; cLF^, 1007*33-
bi8.

1014*7 {Pkysics I I 3. 195b3). T h e classification into proper a n d co­

incidental causes is exhaustive, a n d " a p a r t firom" introduces a cross-
classification; cf. 1014*19-20, E 2. 1026^1. I t is not reaUy necessary to
bring the Metaphysics text, as Jaeger does, into line w i t h the Physics b y
dropping " a p a r t f r o m " . I f w e do, the sense i s : *AU of them, those so
caUed both properly a n d coincidentaUy, are i n some cases so caUed as
being capable, i n others as actually functioning.'

1014*10 {Physics I I 3. 195^6). T h e maker of bronze causes matter

either (Ross) i n the sense that his product is material for someone else, or
i n the sense that he produces a certain k i n d ofmatter (out ofother kinds),

i o i 4 " i 5 {Physics I I 3. 195b12). I t is difficult, but not important, to get

six headings out ofAristotle's list (note that his words for *either' a n d *or'
are the same). T h e distinction between functioning a n d capacity cannot
apply to descriptions such as *Polyclitus' a n d ' m a n ' . Aristotle's point
about simultaneity seems to be: A is doctoring Bjust so long as B is being
healed b y A, but it is not true that A is a doctor just so long as B is a
patient. B u t this contrast has nothing to do w i t h the difference between
capacities a n d the exercise of capacities. Ross's explanation of ' p a r t i c u ­
l a r ' i n *21 is probably correct: w h a t is functioning is particular, i n the
sense that it is always appropriate to ask the question ' w h i c h doctor?'
w h e n someone says 'the doctor is healing' but not always w h e n someone
says 'the doctor is a healer'. T h i s takes 'particular' i n the sense ' i n d i ­
v i d u a l ' , but *17 (cf. 1013b34) used it i n the sense 'specific' to distinguish
e.g. statuemakers from the wider class of men-of-skilL

CHAPTER 3

1014*26. T h e material substances w h i c h Aristotle believed to fit his

first sense of 'element'^are earth, a i r , fire, a n d water (the " b o d i l y
elements" or " s i m p l e bodies" discussed i n De Caelo I I I a n d I V ) . T h e
elements of modern chemistry were originally so caUed i n the same sense,
but a r e n o w k n o w n , of course, to " d i v i d e further into other things
diflTering i n f o r m " : e.g. not aU parts of oxygen are oxygen. Elsewhere
(e.g. Physics I 6. 189b16) Aristotle sometimes names matter, form, and
lack (of a form) as the elements ofsubstances. Elements m a y be divisible,
as v o w e k , or indivisible, as (spoken) consonants.

128
J3 NOTES 1014"35
1014*35· T h e " d i a g r a m s " are solutions to geometrical problems,
hence perhaps geometrical theorems i n general (see B 3. 998*25 a n d
Ackrill's note onCategories 12. 14*26). " S y l l o g i s m s o u t o f t h r e e t e r m s " : as
opposed to sorites {PriorAnalytics I 25. 42^2-3). F o r this use of*element'
seeJ5 3. 998*26 a n d Ross's note on that passage; a k o H e a t h , Mathematics
in Aristotle, 205-6, Proclus, In Euclidem J , ed. Friedlein, 72. E u c l i d ' s book
was, of course, called Elements,

1014^3. Ross refers to Topics I V 1. 121^11-13, " i n aU these cases the

element is that the genus is spoken ofmore widely than the form a n d the
differentia": we m i g h t s a y 4he fundamental thing'. F r o m such general
propositions Aristotle passes to " u n i v e r s a l things", i.e. general coruepts,
w h i c h are elements because widely applicable. H i s mention of points
perhaps adverts to the view, not shared by himsetf {Physics I V 8. 215^18-
19), that lines, planes, a n d solids are composed out of points. " T h e
things termed g e n e r a " are apparently the categories (cf. Δ 6.1016^33-4),
w h i c h are indivisible a n d have no formula Ьесаше they cannot be a n a ­
lysed into genus a n d differentia. Often, of course, Aristotle appUes the
w o r d 'genus' to narrower classes such as animal, but even such genera are
" m o r e " elementary than differentiae because, as Arktotle thinks, a
differentia that k properly given "impUes the g e n u s " — s o that e.g. two-
footed non-animals such as ladders wiU have to be two-footed i n a
derivative sense (cf. Categories 3. 1^16-17, Topics I 15. 107^19-26, V I
6. 144b12-30).

1014^14. A n element k a n origin {A 3. 983^11), but a constituent, not

external, origin {A 4. 1070^22-6). B 3. 998*20-^1^ rakes the question
w h i c h ofthe two kinds of elements of bodies dktingukhed i n t h k c h a p ­
ter—^genera and simple p a r t s — a r e the origins of things.

CHAPTER 4

1014b16. T h e L a t i n ' n a t u r a ' is cognate with *nasci', *to b e b o r n ' , b u t A r k -

totle's Greek word 'phusis' does not usually m e a n 'birth' ("the coming
to be of things that g r o w " ) . E v e n the etymology h e proposes connects
it rather with growth, h k word for 'grow' being 'phuesthaV w h i c h , unlUce
'phusis', has the 'м' long. M o d e r n opinion rejects this etymology and
traces 'phusis' from a root 'phu- meaning 'be', as i n the L a t i n 'fui' (see
R o s s ; also C . S . Lewis, Studies in Words, 34).

1014b18. W h e n a doctor alters h k patient by curing h i m , the change k

not i n the doctor. W h e n a doctor cures himself, the change k i n hinwelf
but not qim himself, since doctors are not restricted to curing themselves.

129
 ioi4^i8 METAPHYSICS J4

G r o w t h , on the other h a n d , is a change w h i c h a living thing works i n

itself qua itself (a self-change). F o r instance, a m a n c a n grow his o w n
fingernails but cannot grow somebody cbc*s; the growth of his finger­
nails is a " c h a n g e i n a naturaUy existing t h i n g " (he is changed) w h i c h is
a "constituent of the t h i n g " (hc effects thc change) ''qua itself" (nothing
else c a n effect the change). Aristotlc confusingly uses thc same w o r d " i n "
first governing the thing changed a n d secondly, i n the phrase *holds
good i n ' (here a n d elsewhere translated " i s a constituent o f " ) , governing
that w h i c h cfFccts the change (cf. E i . 1025^20-1). Nature is " w h a t m a k e s "
it that (literally *whence') changes a r c self-changes: that is, it is because
a m a n has a nature' a n d a diary does not that thc .man grows his
fingernails but the diary docs not grow its daily entries. Not aU changes i n
a m a n are seU*-changes: Aristotlc distinguishes those w h i c h are as " p r i ­
m a r y " . T h e same sense is discussed more fully at Physics I I 1. 192^8-33,
w h i c h concludes " t h e things w h i c h have this kind of origin (sc. of
change) possess a n a t u r e " . I n ^20-^ Aristotle seeks to explain w h y growth
is a self-change even though something outside contributes to it, e.g.
food or the pregnant mother. H i s solution is that the change is due to a
form w h i c h is common to thc growing thing a n d (in the case of food,
after assimilation) the outside agent. H i s words for 'assimilation' {'sum-
phusis') a n d *adhesion' {*prosphusis\ Physics V 3. 227*17) have the appear­
ance ofcompounds oi'phusis*, but m a y be more closely tied to the notion
of growth than that parentage would allow. F o r the requirement that
assimilated things be continuous, i.e. merged or fused, see K 12. 1069*
5-i2.

1014^26. T h i s is thc sense in w h i c h a thing's nature is its matter. I n ^27

onc M S . reads ' a not naturaUy existing thing', w h i c h better fits Aristotle's
example of the statue, but is absurdly restrictive. Ross suggests that the
statue is counted a naturaUy existing thing because it is made of natural
materiab. T h e specification w h i c h Aristotle needs is not *naturally
existing thing' but 'perceptible material object* (see note on 1015*13).
" U n s t r u c t u r e d " : the meaning m a y be ( i ) that bronze is not the sort
o f t h i n g to have or lack a shape, or (2) that a piece ofbronze c a n change
its shape without Umit a n d c a n b c of any shape a n d size, or (3) that a
piece of unworked bronze conmionly has no very regular shape (so
R o s s ; cf. *a shapeless lump*). Aristotle's example is wrong i f h c denies
that wood is " s u b j e c t to loss of its o w n capacity*', since its powers are
permanentiy destroyed by e.g. burning. Perhaps he means that the wood
i n a woodcn artefact does, as a matter offact, retain the powers it possessed
before being w o r k e d ; cf. b31-2 " t h e first (i.e. initial) matter being con­
s e r v e d " . Ifso, his point would be that wood is the nature of e.g. a walking-
stick but not of c.g. a newspaper. B u t i n that case it is not, as he says,
i n the same sense that elements are natures; for e.g. water does lose its

130
J4 NOTES 1014ba6

power of quenching thirst w h e n compounded into sulphuric acid, evcn

though, as Aristotle thinks, the water itself is " c o n s e r v e d " .
O n this sense of'phusis' c f Physics I I 1. 193*9-30.

1014b35. T h e substance of a thing is its essence. Aristotle takes

Empedocles to b e rejecting, i n the modern empiricist manner, the
notion of 'геаГ essence. H a v i n g defined nature ( = matter) a n d nature
( = essence) i n terms of naturally existing things, Aristotle n o w dis­
concertingly defines 'naturally existing' i n tenns of matter a n d form.
T h e latter definition is inadequate, since even i f n a t u r a l l y existing things
are to include artefacts they cannot be meant to include e.g. arguments,
w h i c h are nevertheless " m a d e u p o f " form a n d (subject-) matter (sec
note o n 1015*13). Aristotle should have specified perceptible nuittcr
(cf. Z 10. 1036*9). F o r the two kinds of "first m a t t e r " scc "first g c n u s "
a n d " l a s t f o r m " a t J 24. 1023*27-9.

1015*11. I t is not clear whether Aristotle means ( i ) i n addition to thc

substances ( = essences) of e.g. men, men themselves, as bmng substances,
are natures; or (2) i n addition to substances w h i c h exist naturaUy, arti­
ficial substances are also caUed natures; or (3) *nature' is also used as
a collective name for aU substances taken together.

1015*13. A t 1014b35 nature was thc substance o f " n a t u r a l l y existing

t h i n g s " ; n o w it is " t h e substance of those things that possess a n origin
of change i n themselves qua themselves". T h e n e w locution implies a
definition of *naturally existing things' w h i c h , though plausible i n itseU*
and i n accord w i t h 1014^19, differs from that needed at 1014^27 (viz.
'perceptible material object') a n d from that given at 1015*6-7 (*what is
nmde u p of matter a n d form'). I n the penultimate sentence Aristotic
attempts to relate his fourth a n d first senses to his fifth, a n d i n t h c last
sentence the fifth a n d third senses are identified.

CHAPTER 5

T h i s chapter does not explain t h c important connections between

necessity a n d essence (Posterior Aruilytics I 4. 73^25-74*3) a n d between
necessity a n d regularity {E 2. 1026^27-33). N o r does it accommodate
the apparently Aristotelian sense 'inevitable' {E 3. 1027^10-11, c f De
Interpretatioru 9. 19*23-7). T h e r e is a n interesting discussion of necessity
at De Geruratioru et Corruptioru 11 u .

1015*20. " J o i n t - c a u s e " ; i.e. necessary but not sufficient condition.

Aristotle would not deny that e.g. drinking b a d water пшу b c necessary
(indispensable, needful) for contracting typhoid, or sailing to Paroe for

131
ioi5*ao METAPHYSICS

being captured b y pirates; his point is that neither of these b a d things

w o u l d be called necessary tout court.

1015*26. Perhaps it w o u l d be better to translate 'bia' here by 'duress';

for compvdsion is not contrary to change i n accordance w i t h choice a n d
inclination, since one m a y be compelled to do what, given the choice,
one would anyhow w i s h to do. Compulsion a n d inclination are, however,
different, w h i c h is enough to explain w h y , although a m a n m a y be
dissuaded from doing w h a t he is inclined to do, he m a y not be dissuaded
from doing w h a t h e is compelled to do.
1015*33. T h e definition of *necessary' as *not possible {endechomenon)
that otherwise' m a y be compared w i t h the definition of 'possible'
(dunaton) as *not necessary that not' a t Δ 12. 1019b31-2. Compulsion
requires necessity, i n this sense, plus a contrary inclination. I n the first
sense of *necessary', i.e. 'needful', i f (i) bringing it about that p is neces­
sary i n order that then (ii) it is not possible that {q a n d not-/>), w h i c h is
equivalent to (iii) it is not possible that not (if^, p). Aristotle omits the
last stage i n this analysis, thus failing to exhibit his first sense as a case of
*not possible that otherwise'. E v e n the completed analysis gives the sense
only of expanded statements such as 'it is necessary to breathe i n order
to l i v e ' ; the unexpanded statement ' i t is necessary to breathe' cannot be
analysed i n terms of possibility, i f the 'necessary' i n it means 'needful'.
Demonstrations, i.e. deduced conclusions, are "demonstrated b a l d l y "
w h e n deduced from necessary premisses; i f the premisses are n o n -
necessary or fidse, the conclusion is demonstrated ad hominem (cf. K
5. 1062*2-3) hypothetically (cf. Prior Analytics I 23. 40^23-5, 44.
50*16-28, PosteriorAnalytics 1 3. 72^13), a n d demonstrated to be " n e c e s ­
sary certain things being s o " not "necessary b a l d l y " {Prior Armlytics I
10. 30b31-3, 38-40, Posterior Analytics I I 5. 91b14-19, ц . 94a21-7).
Aristotle's thesis that w h a t foUows from necessary premisses cannot be
otherwise, i.e.
U{P^4)^{UP^-O-q)
is part of every standard system of modal logic; compare the related
theses a t J 12. 1019^22-7, Θ 4. 1047^14-16, PriorAnalytics I 15. 34*5-7.
1015^9. T h a t simples cannot be i n more than one state does not prove
that they are the fundamentaUy necessary things, for according to Θ
10. 1051b9-17 some composites also are " i n c a p a b l e of being otherwise".
Aristotie must be relying o n his contention that only i n the case of simples
is nothing the cause oftheir necessity, i.e. is their necessity indemonstrable.
I n the last sentence he asserts that w h a t is done under compulsion is not
done i n v a r i a b l y : the argtunent seems to be that if x is compelled to do A,
it is possible for x to do A but a k o natural a n d therefore possible for x not
to do A.

132
J6 NOTES 1015b16

CHAPTER 6

Another discussion of the senses of *one' is to be found i n Metaphysics 11.

See also Physics I 2. 185b5.

1015^16. These notes will examine three questions: ( A ) how are the
items listed as examples of coincidental unity meant to be divided u p ?
(B) i n what sense are they one? ( C ) i n what sense is their unity co­
incidental? ( D ) discusses the case of"something u n i v e r s a l " .
(A) b17_18 do not mean that *Coriscus a n d the artistic a n d artistic pp. 221-2
Coriscus' are three examples of coincidental unity (for Coriscus at least
is a unity i n his own r i g h t ) ; nor that they are a single example (for the
subsequent sentences deal only with pairs of expressions). T h i s leaves
three possibilities, ( i ) T h r e e paired examples are to be understood,
Coriscus a n d the artistic, Coriscus a n d artistic Coriscus, the artistic a n d
artistic Coriscus. (2) Since the last of these pairs is not further discussed,
it is more likely that only the first two are intended. (3) I t is also possible
that while *Coriscus a n d the artistic' is the first example, 'artistic
Coriscus' on its own is the second—or rather a repeat of the first. T h i s
seems to be the interpretation ofAlexander, a n d is implied by the modern
punctuation of the O C T . I t also makes good sense of the parenthesis at
b18-19, w h i c h would then be elliptical for 'for it is the same thing to say
" C o r i s c u s a n d the artistic are o n e " a n d "artistic Coriscus is o n e " '.
(Alexander explains this parenthesis with the words 'for it is the same
thing to say " C o r i s c u s a n d artistic [are] o n e " and " t h e artistic coincides
i n C o r i s c u s " a n d " C o r i s c u s is a r t i s t i c " ' ; a n d i n the O C T Jaeger,
presuming that Aristotle's explanation was the same, inserts the word for
'one' after the first 'artistic', giving 'for it is the same thing to say " C o r i s ­
cus a n d the artistic [are] o n e " a n d " C o r i s c u s [is] a r t i s t i c " '—elision of
the verbs would be regular Greek. But it is not necessary to suppose that
Alexander found *one' i n his text of Aristotle; a n d i n a n y case its
presence is less apt i n Aristotle's explanation than i n Alexander's.)
Against this third interpretation stand (i) the oddity of introducing,
among statements of coincidental unity having paired subjects, one
statement having a n unpaired subject, a n d (ii) the fact that ^23 discusses
the unity ofartistic Coriscus with Coriscus. O n the other h a n d , the second
interpretation makes the parenthesis sdll more elliptical, meaning 'for it
is the same thing to say " C o r i s c u s and the artistic are o n e " a n d " C o r i s c u s
a n d artistic Coriscus are o n e " '. I n either case it seems that Aristotle
gives us five examples of coincidental unity:

(fl) Coriscus a n d the artistic (? = artistic Coriscus)

(b) Coriscus a n d artistic Coriscus;
{c) theartisticandthejust;

133
 ιοΐ5^ι6 METAPHYSICS

{d) artistic a n d just Coriscus (thc artistic Coriscus a n d just C o r i s c u s ) ;

(e) a m a n and a n artistic m a n .

pp. 2 0 8 - 1 4 (B) A n analysis of unity m a y include answers to two types ofquestion

(see K . R . Popper, *The Principle of Individuation', Proceedings of the
AristotelianSocietjy,supplemcntaryvolume, 1953, i o o - i ) ; type 1 questions
are of the form 'what are thc conditions under w h i c h x a n d y make u p
one thing, or under w h i c h the combination of^f a n d ^ is singular a n d not
p l u r a l ? ' ; type 2 questions a r c ofthe form 'what are the conditions under
w h i c h X лпау are onc a n d the same thing, a n d not different things?' I n
the paragraph following that about coincidental unity Aristotle addresses
himself to the type 1 question: *is combined out of^: a n d ^ , one thing or
m a n y ? ' I t appears at first as i f t h e discussion ofcoincidental unity deals
with type 2 questions: for e.g. 'Coriscus' a n d *the artistic' c a n be used
as designations of the same thing, a n d Aristotle actually slips into talking
of sameness at ^27. B u t the appearance is nusleading. ^23 says that the
artistic a n d Coriscus are one "because one coincides i n the o t h e r " ; these
items, then, are regarded b y Aristotle as different things, whose relation­
ship of coinciding combines them into a k i n d of unity. I n ^24-5 the
combined items are verbal expressions, "portions i n the f o r m u l a " , but
Aristotle is being careless: it is primarily things, not words, w h i c h are
said to coincide (see e.g. Г 4. 1007*21-2, " i t is necessary for them to
maintain that aU things are coincidences"). T h u s his theory is thatjust
as the expression 'the artistic Coriscus' is a complex of the words 'the
artistic' a n d 'Coriscus', so what it designates is a complex of two n o n -
linguistic items, thc artistic a n d Coriscus. W h e n G o r i s c m is artistic,
thesc items make u p a single complex; otherwise they r e m a i n distinct.
Y c t — a n d here type 1 a n d type 2 questions about unity become con­
fused together—Coriscus a n d thc artistic c a n also be regarded as the
same item, owing to the ambiguity of the expression here literally
translated 'the artistic'. A t Z 6. 1031^22-8 Aristotle tells us that "things
spokcn ofcoincidentally like the artistic a n d pale . . . signify i n two w a y s ;
. . . both that i n w h i c h pale coincides a n d the coincidental . . . affection".
Coriscus a n d the artistic are the same i n the first of these senses—he is a n
artistic (thing). Aristotie holds that this sameness is to be explained by the
fact that the artistic i n its other sense, viz. the affection or property of
being artistic, coincides i n Coriscus. H e n c e his embarrassment over
examples (ό), (d), a n d {e), for e.g. thc p a i r ofexpressions 'Coriscus' and
*artistic Coriscus' cannot be construed as picking out two different items;
only " o n e portion of thc [latter] f o r m i d a " designates a second item—the
property of being artistic—capable of coinciding i n Coriscш.
( C ) B y exploiting this ambiguity Aristotle is able to maintain that the
artistic Coriscus is a unified complex i n w h i c h Coriscus—somehow
detached from his artistry—is a simple part. I n what sense is this unity
Δ6 NOTES 1015b16

coincidental? ( i ) According to the official definition atEs. 1026b31-3

'coincidentaUy F' т с а ш 'F but neither always nor usually F'. I f that is
Aristode's meaning here, his examples are ill-chosen, for (i) the pale
C o r i s c u s — i f n o t artistic Coriscus—might be a unity throughout Coriscus'
life, and (ii) the unity of e.g. a bundle, w h i c h is one i n its own right
( l o i 6 * i ) , might be temporary. {2) T h e r e is no indication that artistic
Coriscus is one i n a derivative sense of 'one' (see i o i 6 ^ 6 - 8 ) . (3) H e is not
non-necessarily one, if 'non-necessarily' means 'non-analytically'. F o r
it is a n analytic truth that every (single) thing is one thing, and Aristotle's
theory offers no reason for denying that truth i n the case ofnon-linguistic
complexes. ( 4 ) S u c h complexes are, however, non-necessarily one, if
'non-necessarily' means 'non-essentially'. F o r although the unity of
artistic Coriscus is a condition of his being artistic Coriscus, it is not a
condition of his existence. I f the elements designated by 'artistic' and
'Coriscus' become dissociated, the artistic Coriscus w i l l not himself
cease to exist, although he will cease to be artistic. T h u s the artistic
Coriscus is non-essentially, though analytically, one just because the
artistic Coriscus is non-essentially, though analytically, artistic. Aristotle
i< right to contrast this case with the unity ofsuch things as a bundle and
a leg, w h i c h are essentially one. Nevertheless his reasons for classifying
the artistic Coriscus and the rest as non-essential unities would have
emerged more sharply had he contrasted them with essential unities also
designated by complex verbal cxpressioiM, such as 'two-footed a n i m a l ' ,
whose parts stand for items that "пшке up one t h i n g " {De Interpretatione
11.20b15-19, c f r 4 . 1007b10,Z12. 1037b10-12).

( D ) T h e case of "something u n i v e r s a l " . T h e truth-conditions for the

unity of thc just a n d the artistic are not that justice be artistic or artistry
just but that some third thing (in ^22 Aristotle assumes, some substance)
be both just and artistic. Does the same apply to («)? Aristotle offers two
analyses: either the artistic coincides i n m a n , or both coincide in Coriscus.
I f t h e latter, two modes ofcoincidence have to be distinguished, for (i) m a n
is not a "state or affection" of Coriscus, but his form, and (ii) m a n co­
incides essentially i n Coriscus. U s u a l l y Aristotle prefers the former alter­
native, even though it implies thc identity of thc form m a n with each
individual пшп (see notcs 0 П Г 4 . 1 0 0 7 * 2 0 ) .

1015^36. A preliminary conMnent. I t is analytic that everything is one

something; a plank is one plank, a bundle of plaiJcs is orte bundle, a
consignment of unbundled planks onc consignment. Most things are
also (made up of) more than one thing: one plank is more than one
molecule, one argument more than one proposition. A t 1 0 1 6 * 1 4 ( " b o t h
one and not o n e " ) Aristotlc seems to acknowledge that the same F can
be o n e F b u t m a n y Gs; a t / 2 . 1 0 5 4 * 1 3 - 1 9 ( " b e i n g o n e does not predicate
anything else i n addition to being each t h i n g " ) he implies also that

135
ioi5^36 METAPHYSIGS J6

everything is one something. O n tht other h a n d , the present paragraph

telb us that a collection of planks is one only i f the planks are tied or
glucd together or otherwise continuous. Aristotle a h o says that some
things are more one t h a n others, e.g. a shin t h a n a leg. T h e s e remarks
appear vubicrable to the criticism of Frege {The Foundations of Anthmetic,
29): " T h e N u m b e r 1, or 100 or a n y other number, cannot be said to
belong to a pile ofplaying cards i n its o w n right, but at most to belong to
it i n view of thc w a y i n w h i c h we have chosen to regard i t ; a n d even then
not i n such a w a y that w e c a n simply assign the number to it as a p r e ­
dicate. W h a t we choose to cali a complete pack is obvioшly a n arbitrary
decision, i n w h i c h the pile of playing cards has no say. B u t it is w h e n we
exanune the pile i n the light of this decision, that w e discover that we
c a n call it two complete packs. Anyone who d i d not know w h a t we call a
complete pack would probably discover i n the pile a n y other N u m b e r
you like before hitting o n t w o . "
I n this paragraph Aristotle mentions, perhaps intentionally, more than
one definition of *continuous'. T h e Greek w o r d , literally 'fused' (see Δ
23. 1023*21-3), is defined at Physics V 3. 227*11-12: " I say that a thing is
continuous w h e n the boundaries at w h i c h each of the two [parts] are i n
contact become one a n d the same a n d , as the name itself signifies, fuse".
T h i s definition is echoed i n the distinction at 1016*7 between *con-
t i n u o m i n its o w n right' a n d *in contact'; but at * i a bundle, whose parts
are merely i n contact, is caUed continuous. *5 interposes yet a third
definition: " t h a t whose change i n its o w n right is one a n d cannot be
otherwise". A thing makes " o n e c h a n g e " (i.e. movement) w h e n aU its
sized parts move simiUtancously (this, not *instantaneom', must be the
meaning o f " i n d i v i s i b l e i n respect o f t i m e " ) , i.e. w h e n it is rigid. A c c o r d ­
ing to *9-10 some non-rigid things (which " h a v e a b e n d " ) are to be
included among the continuous. H e n c e being restricted to one change,
i.e. being rigid, must differ from being restricted to one change i n its o w n
right, w h i c h is the new definition of *continuous'. B u t Aristotle gives no
rules for distinguishing these cases. H e goes o n to pronounce the rigid
" m o r e o n e " t h a n the flexible, a n d the straight than the bent. I t is h a r d to
see w h a t arguments would c o u n t f o r or against these proposals, but we
c a n certainly object to his connection between t h e m ; for straight things
c a n be flexible a n d bent things rigid.
W i t h this p a r a g r a p h c f I 1. 1052*19-21.

i o i 6 * i 7 . Although Aristotle defines, i n this paragraph, a sense i n

w h i c h e.g. a pane ofglass a n d a pond ofwater would be one, his examples
predicate *one' of the m a t e r i a k themselves—^water, wine, j u i c e , etc.
T h e s e a r e divided into two groups: " f i r s t " m a t e r i a k w h i c h are " p e r ­
ceptually i n d i v i s i b l e " , i.e. at the macroscopic level homogeneom (wine
a n d w a t e r ) ; a n d m a t e r i a k whose " l a s t " or " u l t i m a t e " subject is one (the

136
A6 NOTES toi6*i7

repetition of wine shows that at its second mention it, a n d therefore

juices, belong i n the latter group with meltables). A thing's " s u b j e c t " is
its matter (material, substratum), and a " l a s t subject" is a n element
{A 3. 1014*26-35); cf. Л 24. 1023*26-9 where, reversing the roles of
'first' a n d 'last', Aristotle describes the element i n a п о п - с о т р о ш а body
(or perhaps i n any body) as "first genus" of that body's matter. I f w e are
to take the definition i n the opening sentences as applying to both these
kinds of nwtter, first a n d last, Aristotle's contrast wiU be between a body
made of homogeneous material, w h i c h is one because its first subject is
undifferentiated i n form, a n d a non-compound body, w h i c h is one because
it contains only one /oji subject or element, a n d elements must be vm-
difFerentiated i n form. T h e latter of these senses of 'one' implies the
former, but not vice versa. T h e r e are, however, two difficulties with this
account. I n the first place, Meteorologica I V (which m a y , however, not
be by Aristotle) describes some juices a n d meltables as compound; e.g.
some wines are nuxtures of water and earth ( I V 7. 384*3-5, 10. 388*34-
^ i i , but contrast 5. 382^13, 10. 389*10), a n d oil is a nuxture ofwater a n d
air ( I V 7. 383b23, 384*15, 10. 388*32). Secondly, the w o r d 'same' i n »23
suggests that he is saying t h a t j u i c e s a n d meltables have some element t>i
common, a n d so are one collectively, not severally. I f so, the paragraph
shifts from a type 1 to a tyj>e 2 criterion of uiuty: water is one (single)
because h o m o g e n e o v B , oil a n d wine are one (the same) because they share
something homogeneous, viz. a n element.

1016*24. I n the two preceding paragraphs Aristotle has mainly dis­

cussed conditions under w h i c h x a n d j i make up one thing. N o w he t v u : ^
to the type 2 question 'are x &ndy one a n d the same t h i n g ? ' ; a horse and
a m a n , for instance, are one a n d the same living thing but do not
necessarily nuike up a n y single thing, e.g. a n equestrian team. O n the
statement that 'one i n matter' a n d 'one i n genus' are " m u c h the same
sense" c f Δ 28. 1024bg. I n the second sentence the text is corrupt but the
general sense certain. " T h e genus above is caUed the s a m e " : i.e. * andj>
are the same G if both are F a n d G is the genus of F. " I f they are the last
foraos of the g e n m " seems to stipulate (i) that 'x' a n d 'y' m a r k places for
form- (i.e. species-) descriptions rather than proper names a n d (ii) that
the species be the infimae species oSx&nay. I t is not clear w h y either of
these conditions is necessary. " T b a t w h i c h is further above these" m a y
be a gloss, a n d m ^ t i n any case explicate " t h e g e n m a b o v e " , sc. G. I f
" t h e s e " are the last forms, i.e. x andy, "further a b o v e " natut m e a n 'at
one remove above'. T h e Greek word, the comparative of 'above', need
not m e a n so m u c h (it c a n be a synonym for 'above'), but Ross's c l a i m
that it " c a n n o t m e a n " 'higher above' seems rash. I f h e is right, " t h e s e "
mtwt refer, as he says, to F a n d its co-ordinate genera. I n *29 " t h i s w a y "
is 'x лпау are one i n genus', the new w a y 'x xaAy are one G'.

137
ioi6*32 METAPHYSICS J6

1016*32. C f . / I . 1052*29-34. A s indicated by the parenthesis i n

b3-6, formula A is indivisible relative to formula B w h e n A a n d B are the
same formula—that is, w h e n they say the same thing. Aristotle's first
example appears to be of a single individual, e.g. a n a n i m a l , w h i c h c a n
grow larger or smaller a n d yet remain the same individual a n i m a l . T h e
second example m a y be of two plane figures, e.g. squares, w h i c h have one
form or shape a n d therefore, Ьесаше the essence of a figure is its shape,
one formula. T h e third sentence i n eflTect distingvdshes these two kinds of
case: x лпау are " m o s t of aU o n e " w h e n indistinguishabte i n essence and
time a n d place, but they wiU be less one i f both essentially humans but
not the same h u m a n ; c f the distinction between *one i n form' a n d *one
i n number' a t 1016b31, a n d / 1. 1052*29, " o t h e r things are one whose
formula is one, i.e. their conception is one, i.e. indivisible; a n d it is
indivisible w h e n the thing is indivisible i n form or number. T h e p a r ­
ticular, then, is indivisible i n number . . . " F o r the distinction between
having a n essential formula or essence a n d being a substance or essence
see Z 4 a x i d Topics I 9. T h e final parenthesis, i n equating being one w i t h
being without division, glosses over the number/form distinction: i f x
andy are " w i t h o u t division qua m a n " merely i n the seme that both have
the formula o f m a n , it does not foUow that they are one m a n (hence Ross's
surely unjustified gloss *one k i n d of m a n ' ) ; whereas i f they are i n that
seme without division qua a n i m a l or qua magnitude they w i U be one
a n i m a l or one magnitude ( c f Aristotle's usage of *one a n d the same
figure' at 1016*31, a n d 1016b31-5,J 15. 1021*11).

i o i 6 ^ 6 . T h i s is a puzzling paragraph. H a s the disdnction between

being somehow related to what is one a n d h a v i n g one substance already
been d r a w n ? I f so, was this at 1016^3 (being one m a n = having one
substance; being one magnitude = being related to one quantity)? O r
are the secondary unities of the same as the coincidental unities of
the first p a i a g r a p h ? Artistic C o r i s c m could be said to be one from " b e i n g
related to some other thing that is o n e " , but nothing i n the first p a r a ­
graph amwers to " e i t h e r doing or possessing or being affected b y . . . some
other thing that is o n e " . These phrases suggest examples like gas a n d
electricity, w h i c h are heating agents (do one thing), or money a n d
labour, w h i c h c a n be expended or comerved (are affected i n one w a y ) ;
but are such pairs caUed one for these reasom? I n ^8-9 *substance' is used
i n the three senses ofH 1. 1042*3-15: the p r i m a r y unities are things whose
bodies are one i n continuity, whose matter is one i n form or whose what-it-
is-tO'be is one i n formula. These three correspond w i t h the second, third,
a n d fifth paragraphs, a n d doubtless Aristotle comprehends the fourth
paragraph w i t h the third, as treating of " m u c h the samc s e m e " .

i o i 6 ^ i i . C f . / I . 1052*22-8. A m o n g continuom things not only the

straight a n d the rigid (1016*9-17) but, Aristotle now adds, wholes a r e

Ι3δ
 NOTES ioi6bii

more readily asserted to be one. A whole must "possess onc f o r m " i n thc
sense 'have one shape', not as i n 1016"19. But Aristotle does not explain
on w h a t grounds a misassembled shoe could be denied to possess one
form; would it make a difference i f the object so produced h a d a name,
or a use? A straight line need be no less complete than a circle i n any of
the senses of 'complete' given inA 16; but see Physics V I I I 9.

1016^17. C f / I . 1052^15-1053^8. T h e six preceding paragraphs have

distinguished a number of semes of 'one', summed u p (aU but the last)
at ioi6^6-! I under the two headings 'being related to some other thing
that is one' a n d 'having one substance'. These senses are derivative
from that, or those, employed i n their definitions. I t is presumably the
non-derivative senses to w h i c h Aristotle now t u r m his attention. T h e
first sentence is corrupt a n d the tramlation foHows Ross's reconstruction.
J a e g c r i n the O C T has 'being one is the origin ofbeing a certain n u m b e r ' ;
but Ross's version requires a more natural suppressed premiss i n ^17,
viz. 'to be one is to be the first measure o f a n u m b e r ' ; cf. / 6. 1057*3-4.
O n e is the " m e a s u r e " of number because the possibility of counting
depends on the possibility ofcounting i n ones, c f Physics I I I 7. 207^7, " a
number is several ones" (Aristotle did not reckon one itselfas a number:
Physics I V 12. 220^27, M 9 . 1085^10). Aristotle's further stipulation
that to be one is to be the first measure of a genus indicates that, i n the
strict sense now under consideration, what is one i n a genus is the
m i n i m u m part of a member w h i c h is itself a member of the genus. T h i s
explains "indivisible i n f o r m " ( c f / 1. 1052^16-27): Aristotle cannot
m e a n that e.g. one vowel must be indivisible into parts, for the sounding
of a vowel has to occupy a stretch of time; nor that one vowel mшt be
indivisible into other vowels, for most words are indivisible into words,
yet a word is not " t h a t w h i c h is o n e " i n the genus articulate sound; his
point is that one vowel is indivisible into other articulate sounds. Physics
V 4. 228'21 says that " e v e r y change is d i v i s i b l e " , sc. i n quantity; a
unit change must therefore be indivisible i n form. Aristotle gives two
possible criteria for this, at ioi6*6 a n d Physics V 4. 227^20-228*23.
" I n aU d i m e m i o n s " etc.: literally ' i n aU ways' etc., see H e a t h , Mathe­
matics in Aristotle, 206-7.

i o i 6 ^ 3 i . T h i s paragraph secms intrusive. T h e senses it lists are,

unlike m a n y w h i c h have preceded, aU senses i n w h i c h 'one' means 'the
same' (type 2), not 'single' (type 1). T h u s the fourth verb a n d thc last
verb i n the foregoing sentence are one verb i n number, even though that
verb is not without parts, nor even a unit of articulate speech. " O n e i n
g e n u s " repeats thesense of 1016*24, 'having the samc genus', except that
genera are now limited to t h e small number ofcategories (seCid 28. 1024^
12). " O n e i n f o r m " s e e m s t o e c h o 1016*32, ' h a v i n g t h e s a m e f o r m u l a ' , b u t

139
 1016b31 METAPHYSICS J6

this paragraph adds that things numerically different may be the same
i n form, w h i c h was previously glossed over. T h e senses given to *one i n
form' at 1016*17 a n d i o i 6 ^ i i are now inappropriate; the latter was
contrasted with *one i n formula' at 1016^9, as it wiU be again at 1017^»
5-6. * O n e i n analogy" is a sense previously ignored. A n analogy is a
statement that a : b :: c: d\ see for instance J V 6 . 1093^18-20, " a s straight
is i n length so is the level i n b r e a d t h " — t h e straight a n d the level are one
a n d the same i n analogy. I f x лпау are i n one genus G , they are also one
i n analogy i n that x : G wy : G,
i o i 7 * 3 . T h e brief discussion of *many' returns to the classification of
ioi6**6-ii. F o r first and last matter see 1016^17-24. ' M a n y ' is discussed
more fully i n / 3 a n d 6.

CHAPTER 7

pp. 215-16 A s in Г 2 a n d ^ 2 , Aristotle introduces his discussion of the various senses

ofthe verb *to b c ' by means ofthe participle 4 h a t w h i c h i s ' ; but ioi7*22
( " a r e said to b e " ) a n d ioi7*3i (" Ч0 be' a n d *is' " ) show that what he
says is meant to cover aU parts of the verb. H i s four m a i n senses are
examined at length i n later books o f t h e Metaphysics: coincidental being
i n ^ 2 and 3, being i n its o w n right i n Z a n d H, being as truth i n ^ 4 and Θ
10, a n d being as actuality a n d potentiality i n Θ 1-9. A general question
about the first two senses is whether they are to be understood as apply­
ing to the copulative 'be', the existential *be', or one to each. I n a n s w e r i n g
this question it is more convenient to start from the second sense, revers­
ing the order of Aristotle's first two paragraphs.

i o i 7 * 2 a . Aristotle tclk us that the things that a r c i n their own right

are those " w h i c h signify the figures o f p r e d i c a t i o n " , i.e. predicables such
as m a n , the pale, walking (not the words *man', *pale', *walking'). H e
classified aU predicables, together with p r i m a r y substances w h i c h are
" n e i t h e r said o f a subject nor i n a subject" (Categories 5. 2"12-13), under
a small number of types or " f i g u r e s " , eight i n this chapter, ten i n the
longest list at Categories 4. 1^25-2*4. W e call them categories, *kategoria'
being Aristotle's word for *predication'. *22 might alternatively be
construed *all things signified [i.e. indicated] by the figures of predica­
tion', but the translation adopted is more likely ( c f »25 "signify . . . a
qualification"). Aristotle's argument is that since e.g. the pale (and,
doubtless, *pale') signifies a qualification, Ч0 b e ' m a y also signify a
qualification. T h e conclusion of this argument c a n be taken i n any of
three ways, o f w h i c h the first two a r e :
(1) 'something is pale (has pallor)' signifies 'something is qualified i n
a certain w a y ' ,
(2) 'the pale (pallor) exists' signifies 'the pale qualifies something'.

140
ΔΊ NOTES 1017*22

I n general, Aristotle m a y m e a n to distinguish a n d classify ( i ) different

ways of understanding propositions of the form is F ' or (2) different
senses of the w o r d *exist'.
T w o things support ( i ) . First, i n «27-30 Aristotle e x p l a i n s t h a t s e n -
tences employing verbs other than *to b e ' c a n be rephrased i n such a w a y
as to incorporate that verb as a copula (cf. De Interpretatione 12. 21^9-10;
his distinction is literally between e.g. *walks' a n d *is walking', but the
Greek corresponding to the latter is both less idiomatic a n d less special­
ized than our continuous tense). Secondly, PnorArmlytics I 37. 49*6-7
implies category distinctions i n the copulative sense of *be' b y asserting
them ofits technical counterpart *hold good of'. O n the other h a n d , ifthe
sense i n w h i c h a thing is i n its o w n right is to be exemplified b y such
copulative uses of *is' as i n *Goriscus is pale', w h a t is to exemplify the
coincidental sense discussed i n 1017*7-22? F o r at 1017*9 Aristotle tells
us that that w h i c h is is so called coincidentaUy w h e n , for instance, w e
assert a m a n to be artistic. Ross meets this difficulty b y proposing a third
interpretation, according to w h i c h Aristotle's being-in-its-own-right is
restricted to those copulative uses i n w h i c h the predicate is the form or
genus of the subject a n d so necessarily true of it, for example

(3) *The pale (pallor) is a colour' signifies *the pale is a certain

qualification'.

(Another example might be *this colour is pale', w h i c h " s a y s what it is

flWi/signifies a qualification". Topics I 9. 103^31-3.) According to Ross,
the senses of Ъе' implied b y the Pnor Analytics passage are to be taken as
coincidental, mentioned though not subdivided i n 1017*7-22. But Ross
does not explain w h y , w h e n Aristotle divides b y categories the *necessary'
senses of the copulative *be', h e should onfit to do the same for the
*coincidental' senses. F u r t h e r , Ross is forced to explain the intrusion of
non-necessary examples i n *27-30 (e.g. " a m a n is one that w a l k s " ) as
due to Aristotle's wish to illustrate, b y the readiest means at h a n d , the
general point that * " i s " takes its colour from the tenns it connects', sc.
evenifthose terms are participial a n d so r e n d e r t h e verb *is' elidable. B u t
Ross would have to admit that o n his interpretation *27-30 would fit
better i n the p r e v i o m paragraph.
A further consideration favours (2) over both ( i ) a n d (3). T h e copula­
tive *be' reappears, a n d according to *27-30 must implicitly reappear,
i n the analysis of 'something is pale' as 'something is qualified i n a
certain w a y ' a n d of 4 h e pale is a colour' as 'the pale is a certain q u a l i ­
fication'. I t follows that there is no reason for treating these analyses as
exhibiting different senses of the copulative *be'. I n the case of ( i ) , for
instance, the thesis that *something is pale' signifies *something is q u a l i ­
fied i n a certain w a y ' , whereas *something is a thing that waUcs' signifies
*something is acting i n a certain w a y ' , m a y help to show that *pale' a n d

141
1017*22 METAPHYSICS Δ7

*thing that waUss' h a v c genericaUy difFerent meanings, b u t does nothing

to show that *is pale' a n d *is a thing that waDis' employ different senses of
*is'; a n d sinularly i n the case of (3). B u t taken as a n analysis ofexistence,
Aristotle's thesis w o u l d have the merit of providing a n account of the
meaning of *exist' w h i c h is not expressed i n terms of (non-copulative)
existence. I f *the pale exists' signifies 4 h e pale is a qualification of some­
thing', whereas *waUdng exists' signifies *walking is a n action of some­
thing', it is more reasonable to conclude that *exists' has different senses
i n the two cases. T h e passage i n the Prior Armlytics m a y then be explained
as offering a parallel analysis of *holds good of' (as meaning e.g. 'is
a quality of', *is a n action of') w h i c h is not meant to carry over to the
copulative *is'. *27-30 r e m a i n puzzling, but are perhaps to meet the
objection that w e do not commonly say such things as *walking exists'.
A c c o r d i n g to Aristotle *waDdng exists' is implied by *Coriscus walks' (he
holds that *A:'s waUdng exists' is equivalent to 'x is waUdng', not, as w e
might say, to *it makes sense to say that x is waUcing', s e e ^ 3 . 1070*22-3
" f o r [his] health exists just w h e n thc m a n is healthy . . . " ) ; a n d this
implication might be thought to be m a d e more obvious to a Greek by
the consideration that *Goriscus waUcs' already contains a hidden ' i s ' ,
albeit a copulative a n d not existential *is'.
If, i n the second paragraph at least, Aristotle s e e b to distinguish
different senses of *exist', we m a y n o w ask whether he succeeds. F o r
although the availability of different paraphrases for *pale exists' a n d
*walking exists' nught intelligibly be urged as a reason for treating
*exists' as homonymous, it does not follow that the reason is a good one.
W o u l d it not be better to say that pallor a n d walking exist i n the same
sense but under different sorts of conditions, or by the satisfaction of
different criteria? I s it not odd, for instance, to treat a dispute between
a materialist a n d a *realist' over the requirements for the existence of
numbers as a dispute about the sense of a word? Aristotle's parallel
treatment of *good', w h i c h " i s said i n the same number of ways г& that
w h i c h i s " {Nicomachean Ethics I 6. 1096*23-4, Eudemim Ethics 1 8. 1217^
26-7) is open to t h c same objection. O n the other h a n d , the objection
employs a distinction—between identity of sense a n d identity of criteria
— w h i c h is both dubious i n itseU* a n d u n k n o w n to Aristotle; so that it
w o i J d be reasonable to deflect it by interpreting his thesis i n the w a y
w h i c h the objection itsetfconcedes to b e i n n o c e n t , construing Aristotle's
" m a n y w a y s " here a s * m a n y criteria', not *senses'. But another problem
remains: criteria o f existence must be m u c h more numerous than the
eight or ten categories. Aristotle himsetf seems to adntit this i n other
places: for instance, his division of ways of being at Г 2. 1003^6-10
makes no reference to categories other than substance, quality, a n d
action, but adds m a n y other non-categorial headings; De Anima I I
4. 415br3 says that " f o r living things, to be is to be a l i v e " ; a n d the more

142
J7 NOTES ioi7*aa

systematic treatment in Metaphysics H a argues, again without reference

to categories, that since one thing m a y be distinguished from another
either e.g. by the w a y its materials are put together (a bundle of planks
from a wooden box) or by position (threshold from lintel) or by time
(breakfast from dinner) " i t plainly foUows that ' i s ' is also said i n the
same number of w a y s ; for a threshold is because it is positioned i n such
a n d such a w a y . . . " (1042^25-6). T h c last text does, however, suggest
how Aristotle might have thought that the ways of being could be
reduced to the eight or ten 'figures of predication'; for being-in-a-position
is among the ten figures listed i n Categories 4 (2*2). I t is true that threshold
a n d lintel demand different positions: for the threshold, to exist is to be
positioned " i n such a n d such a w a y " . B u t a n unqualified reference to
position, though it cannot fully specify the conditiom of a threshold's
existence, w i U indicate one necessary condition w h i c h the threshold has
in common with the lintel. T h e categories are thus apparently intended
n o t to enumerate but to classify the criteria ofexistence {ci.A 2. 1013^29-
30). Aristotle thought that the n u m b e r of these ultimate kinds cannot
be m u c h reduced below, a n d should not be m u c h increased above, the
eight or ten he usually lists; a n opiruon w h i c h he nowhere attempts to
justify, but w h i c h cannot b c discussed here. N o r does Aristotle explain
how he would deal w i t h things whose existence depends on criteria
faUing under more than one category, as dinner's does on size as well as
time.

1017*7. Aristotle's examples of coincidental being are given i n the

assertions:
(a) 'someone j t B t is artistic';
(b) ' a m a n is artistic';
{c) 'someone artistic is a m a n ' .
T h e subject of each of these assertions begins w i t h a definite article i n
the masculine: literally ' t h e j u s t . . .', 'the m a n . . .', 'the artistic'. T w o
m a i n questior« call for conunent: (A) is the coincidental sense of 'be' a
sense of the copulative or existential ' b e ' ? a n d (B) i n what sense is it
coincidental ?
(A) ( i ) T h e r e a r c two arguments i n favour o f t h e v i e w that Aristotle's
subject is coincidental copulative being, (i) H i s examples are examples of
t h a t (e.g. one w h o is just is not necessarily or always artistic); (ii) the
parenthesis i n *12-13 e x p l a i m the copulative 'is' i n t e m M ofthe relation
coincidence. O n the other h a n d (iii) the second paragraph does not seem
to restrict itself to rucessaty copulative being, as argued above, a n d (iv)
this account does not provide a n y w a y of classifying false predications.
(2) R a t h e r than holding that the copulative sense of 'be' is sometimes to
be understood i n terms of coincidence, Aristotle m a y mean that the

143
ioi7*7 METAPHYSIGS J7
copulative sense of *be' is always a coincidental, i.e. derivative, sense;
a n d that the sense or senses described i n the second p a r a g r a p h , i n w h i c h
*be' means, as w e should say, *exists', are p r i m a r y . T h i s interpretation
w o u l d explain w h y i n *20-2 his analysis of the coincidental things-that-
are reveaied by assertions such as {a)-{c) includes a reference i n each
case to some other thing-that-is: the latter would be a n existent thing e.g.
a m a n , whose existence accounted for the non-existential being of e.g.
the man's being artistic (cf. also De Interpretatione 11. 21*25-8). (3) H o w ­
ever, i n the parallel list of senses of *be' at the beginning of E 2
Aristotle announces that a l l the senses there distinguished, including
the coincidental, belong to " t h a t w h i c h is w h e n baldly so c a l l e d " , a
phrase normally used to pick out the existential Ъе' (see notes a d l o c . ) ;
a n d this favours the view that Aristotle's distinction is between a co­
incidental a n d , i n his second p a r a g r a p h , certain non-coincidental senses
of the existential Ъе'.
(B) W h a t does *coincidental' m e a n i n this chapter? T h e r e are three
possibilities, ( i ) *unusual', (2) *non-essential', (3) 'derivative', ( i ) E 2,
starting from a repetition of A 7's fourfold division of senses of Ъе',
continues w i t h a n examination of coincidental being, i n the course of
w h i c h the coincidental is defined as being " w h a t is neither always nor
for the most p a r t " (1026^31-2). T h i s imphes that *coincidental thing-
that-is' means *(comparatively) unusual thing-that-is', a n d w e m a y
assume that to say that a n artistic m a n is a n unusual thing-that-is is the
same as to say that it is unusual for artistic m e n to exist. B u t there are
difiiculties w i t h this account, (i) i n A 7 coincidental things-that-are are
contrasted w i t h things-that-are i n their o w n right. T h i s seems different
from the contrast between the usual a n d the unusual, (ii) If, as argued
above, things-that-are i n their o w n right include such items as Gallias and
pallor, it is h a r d to see h o w these could be said to exist usually. *An
artistic m a n is unusual' denies that two components usually go together;
*Gallias is u n u s u a l ' cannot be taken i n the same sense—nor therefore
*Gallias is usual'. (2) If, however, *coincidental' means *non-essential',
there is a good contrast w i t h *in its o w n right': for such items as Gallias
a n d pallor are essentially things-that-are, w h i c h means to s a y — t r i v i a l l y
— t h a t their being things-that-are is a condition of their existence. O n
the other h a n d , it seems b y the same token contradictory to assert of
anything that it is яол-essentially a thing-that-is. W e c a n indeed say that
a m a n is non-essentially artistic, for his existence does not depend on his
artistry; but w e cannot say that a n artistic m a n is non-essentially
existent. T h e r e are three reasons w h y Aristotle m a y have overlooked this
difference, (i) I t does not оссшг w h e n 'coincidental' has the sense *unusual':
*it is unusual for a m a n to be artistic' is equivalent to *it is unusual for a n
artistic m a n to exist'. A n d nowhere does Aristotle clearly reveal that he
saw the difference between these two senses of 'coincidental', (ii) W o r d -

144
Δ7 NOTES ioi7«7

order being variable i n Greek, *a m a n is ardstic' could be written, as

Aristotle writes aU his examples i n this paragraph, w i t h *is' at the begin­
ning or end. Idiomatic Greek w o u l d indeed stiU distinguish ' a m a n
artistic is' as copulative from *an artistic m a n is' as non-copulative, but i f
the significance of that idiom escaped Aristotle he h a d no other w a y of
distinguishing *a m a n is non-essentially artistic' from *an artistic m a n
non-essentially is, i.e. exists'. (iii) T h e notes οηΔ 6. 1015^16 suggest that
Aristotle assumed that certain complex referring expressions such as *an
artistic m a n ' refer to complex non-linguistic entities w h i c h we m a y call
states of affairs. H e m a y thus have been tempted to argue that the
existence o f a m a n who is artistic does not depend on that ofthe complex
state of affairs designated *artistic m a n ' : hence that the artistic m a n is
a non-essential existent. B u t whatever w e think of coincidental unities,
this account of coincidental existents must be incoherent. T h e r e is not
only the general objection that it is wrong to regard the state of affairs
designated 'artistic m a n ' as a combination o f t w o items, the m a n a n d the
artistic (either 'the artistic' means 'someone artistic', i n w h i c h case there
is only one item, or it means *artistry', i n w h i c h case we face the impossible
question 'is the man-element i n the complex artistic or not?', cf. Plato,
Parmenides 142 de). I n addition, the theory gives sense to the c l a i m that
a n artistic man's existence does not depend o n a n artistic man's existence
only at the cost of making two distinct items—^man a n d state of affairs—
of the things referred to b y the twooccurrences of 'artistic m a n ' : but of
course a thing is non-essentially F only when its existence does not
imply its—the same thing's—being F .
(3) T h e above explanation h21s the merit of reading Aristotle's dis­
tinction between *coincidentally' a n d *in its o w n right' i n exactly the
same sense as i n J 6 a n d Δ 9. B u t it fails to account for a feature of his
treatment of coincidental being w h i c h has no parallel i n those other
chapters: the presence i n the s u m m a r y at *19-22 ofreferences to some
other thing-that-is. Aristotle holds that i n e.g. the sentence 'someone j u s t
is artistic' w e signify that the just a n d the artistic coincide i n the same
thing-that-is. T h i s suggests that h e takes such a sentence to use ' i s ' i n a
derivative sense, w h i c h has to be explained b y reference to the w a y i n
which something else ' i s ' : specifically that the being (existence) of the
complex item designated 'the j u s t artistic' has to be explained by
reference to the being (existence) of the m a n i n w h o m the elements of
that complex severally coincide (for the use of *coincidentally' to mean
' i n a derivative, or secondary, sense' see e.g. Categories 6. 5*38-^4).
Interpretation (2) asserted that the existence of the simple item desig­
nated by 'the artistic m a n ' does not depend o n the existence of the com­
plex state of affairs designated b y the same expression. Gonversely, (3)
asserts that the existence of the latter does depend o n that of the former,
a n d has to be explained by reference to it. T h e s e two doctrines are not

145
 ioi7*7 METAPHYSICS Δ7

antagonistic but complementary. Indeed it m a y well have been Aristotle's

view that the existence of B is derivative from that of A only i f A c a n
continue to exist w h e n B does not. I f so, (3) implies the substance of (2)
while also giving a better explanation of the meaning of 'coincidental'.
T w o minor points remain. " F o r *that this is this' signifies 'that i n
this this coincides' " : sc. i n the examples given. Aristotle temporarily
overlooks the case allowed i n *15 i n w h i c h two things coincide not one
in the other, but both i n a third thing. 'Someone artistic builds' is really
of this indirect type. " T h e not-pale is said to b e " introduces a n asser­
tion w h i c h , unlike {a)-{c), has not even the appearance ofbeing copula­
tive. Aristotle's point m a y be that just as, i n (c), 'the artistic' designates
something essentially a m a n b u t ' m a n ' designates something n o n -
essentially artistic, so i n 'the not-pale is' 'the not-pale' designates some­
thing essentially existent but ' i s ' designates something non-essentially
not-pale. I f so, the example has nothing to do with coincidental being.

1017*22. S e e p p . 140-3.
1017*31. I t was a common Greek idiom to use 'is' a n d 'is not' i n the
sense 'is the case' a n d 'is not the case' (e.g. Г 5. 1009*7). T h u s it is the
possibility of falsehood w h i c h is at issue i n the Platonic puzzles, e.g. i n
the Euthydemus a n d Sophist, about *saying that w h i c h is not'. B u t the
phrase 'it is (not) that . . .' regularly meant 'it is (not) possible that . . .'.
T h i s m a y explain w h y the examples w h i c h Aristotle gives contain no
subordinate clause; where we should expect 'it is (the case) that Socrates
is artistic' etc. w e have merely *Socrates is artistic' etc. with the 'is' (or 'is
not') emphatically placed at the beginning. Whatever the explanation
of lhis oddity, its effect is to destroy the value of the examples as illus­
trations of a separate sense of 'is'. F o r the fact that 'x is F ' means the same
as *it is true that x is F ' , a n d 'x is not F ' as 'it is false that x is F ' , c a n have
no tendency to show that 'is' c a n m e a n the same as 'is true', or 'is not' as
*is false'.

p. 217 1017*35. T h i s paragraph introduces, not a further sense of 'be', but a

further w a y of classifying the cases of being already given ( c f Θ 10.
1051^35-^1). 'Sees' (in English ' c a n see') is ambiguous between 'has
sight' a n d *is using sight'; c f Θ 6. 1048^12-14, Topics I 15. 106^15-18,
a n d Philonous' remark i n the first ofBerkeley's Three Dialogues "sensible
things are those only w h i c h are inunediately perceived by sense".
Aristotle justifiably uses tiiis fact to infer that 'is one-that-sees' is also
ambiguous; but he does not explain how, in the latter case, the ambiguity
is transferred from 'sees' to 'is'. T h e two succeeding examples do not
even employ the verb 'be' ('be-at-rest' is a single word in the G r e e k ) , but
Aristotle would presumably argue, as i n 1017a27-30, that 'x knows' and
'y is-at-rest' c a n always be paraphrased into 'x is one-that-knows' and 'y
is one-that-is-at-rest'. H i s treatment o f " t h e h a l f o f a l i n e " a s a s u b s t a n c e

146
J7 NOTES ιοΐ7·35
m a y be, as Ross suggests, a concession to mathematically minded
Platonists. O r could *substance' here m e a n *what is i n its o w n right*? I f
so, the examples of sight, knowledge, a n d r c s t w i U correspond to the
first paragraph of the chapter, so that 'CalUas is one-that-sees' asserts,
ambiguously, the coincidental existence o f the seeing Callias. Aristotle's
point i n the penultimate sentence is that a future idol must already exist
i n the unworked stone, because the very same stuflf w h i c h wiU become
the idol exists already (ought he not to a d d that the past idol must stiU
exist i n its dispersed chippings?).

CHAPTER 8

T h i s chapter is too sketchy to give a satisfactory account of the difficult

a n d important notion of a substance. M o r e is to be learnt from Categories
2 a n d 5, a n d from Aristotle's discussion o f the question *which kinds of
things are substances?' w h i c h takes u p most of Z a n d H. A thdrough
understanding would call for exanunation of the related concepts what-
it'iS'to-be, coincidenceyform, subject,

1017^10. T h e simple bodies are the elements earth, a i r , fire, a n d

water; see Δ 3 a n d A 3. 984*5-16. " D e i t i e s " : stars {E 1. 1026*17-18).
" T h i n g s not said of a s u b j e c t " : see note o n 1017^23. W i t h this p a r a ­
graph c f Z 2. 1028^8-15.

1017**14. I t is usually matter w h i c h is described as a constituent cause

( J 2. 1013^24-5), but the example shows that Aristotle's reference here
is to forms, w h i c h are sometimes said to be " i n " things (e.g. Z 11. 1037*
29) although not " i n a subject" according to the idiom of Categories 2.
1a20_bg ( c f 5. 3*14). A thing's form is саше of its being because its
existence is dependent o n its having that form. According to Categories
2. 1*20-2 forms are said of a subject; 5. 2*14-19 groups them with
genera as "secondary substances", but Z 7. 1032^1-2 makes them p r i ­
mary. T h e s e inconsistencies m a y reflect ambiguity i n the word *eidos*,
w h i c h i n the Categories is closer to 'species' t h a n to *form'.

1017^17. Does Aristotle think that planes, lines, a n d numbers meet

the conditions laid down i n this paragraph for being substances? I n B
(5. 1001^27) one o f t h c problems was " w h e t h e r bodies a n d planes a n d
points are kinds ofsubstance or n o t ? " A f 2 . 1077*31 answers " a body is a
kind ofsubstance, for i n a w a y it already possesses completion; but how
can lines be substances?" T h e latter chapter is a response to Platonists
of a Pythagorean bent; this suggests that " a s some assert" a n d " i s
thought b y s o m e " refer to such Platonists a n d convey Aristotle's dis­
sent from them. H e has reasons for dissenting. Ifbodies were made out
of numbers, as the Pythagoreans say, they would have to be without

47
ioi7^i7 METAPHYSICS J8

weight {N 3. 1090*3cH-5); a n d Physics V I 1. 231*21-^18 argues that " i t

is impossible for something continuous to be [made] out of indivisibles,
as for instance a line out of p o i n t s " . B u t these reasons seem insufficient
to deny planes etc. the status of substances under the criterion of this
paragraph. F o r (i) though not composed of planes, bodies cannot exist
unless planes do. (ii) " D e f i n e " might m e a n *bound' or *provide a
principle of individuation*. I f the former, it is indeed true that not aU
" s u c h t h i n g s " , v i z . animals a n d stars a n d so o n , are bounded b y plane
surfaces, b u t " p l a n e " m a y w e l l b e a slip for 'surface'; cf. Physics I I I
5. 204b5-7. I f the latter, it seems neither false nor unhelpful to say that
e.g. star A a n d star B are two " t h i s e s " (particulars) j u s t i f their surfaces
are discontinuous, a n d their surfaces are discontinuous j u s t i f no line
forms part o f b o t h . Aristotle is o n firmer ground i f h e denies that numbers
define bodies, i n either sense. I n view of these considerations, i t is
probably Aristotle's belief that the conditions laid d o w n i n the p a r a ­
graph are not sufficient, although some have thought them sufficient, to
make a thing a substance; cf. Z 2. 1028^18-21. T h e s u m m a r y beginning
1017b23 does not mention this sense.

1017^21. T h i s is the sense i n w h i c h it is natural to translate 'ousia' as

*essence' rather t h a n *substance'. A thing's what-it-is-to-be is the same
as its form {Z 7. 1032^2 a n d often), or its "substance without m a t t e r "
{Z 7. 1032b14).

1017^23. T h e sununary onnts the third sense a n d identifies the second

w i t h the fourth, leaving 'substance' to m e a n ( A ) 'body', i.e. " t h e concrete
whole m a d e out o f f o r m a n d m a t t e r " {Z 11. 1037*29-30) a n d (B) 'form'
or 'essence'.
(A) " U l t i m a t e s u b j e c t " picks u p 1017^13-14 *4hey are not said o f a
subject but the rest are said of t h e m " . I t is not clear that these expressions
c a n bear the weight w h i c h Aristotle puts o n them. A n vdtimate subject is
'said of' nothing else; a n d 'T is said of nothing else' appears to m e a n
' ''X is T" is true only i f X a n d T are the same thing'. B u t it seems w e
m a y say both that 'this matter is Socrates' c a n be true a n d that Socrates
is not the same thing as his m a t t e r ; a n d i f s o , (i) Socrates wiU not be a n
ultimate subject, (ü) his matter w U l be. Aristotle wrestles w i t h these
problems i n Z 3, insisting a t 1029*27 that it is impossible that matter
should be substance. F u r t h e r m o r e , there might be particular qualities,
quantities, etc., w h i c h are said o f n o subject. Aristotle acknowledges this
at Categories 2.1 *23-9, instancing " t h e individual knowledge-of-grammar"
a n d " t h e individual p a l e " . A s a result he there stipulates that particular
substances are neither said o f a subject nor i n a subject, a n d defines 'is i n a
subject' to imply " c a n n o t exist separately from w h a t it is i n " . B u t the
separability requirement is onutted here.

148
J8 NOTES 1017b23

(В) I n the case of form-substances the present passage does, hke

Categories 2, require separability. But how c a n a form be "separable a n d a
t h i s " ? I t is separable i n thought (Physics I I 2. 193^33-34), but seemingly
inseparable i n the Categories sense, since it cannot exist—or rather,
perhaps, be i d e n t i f i e d — " a p a r t from sensible substances" (M 1. 1076*11).
Usually the concrete particular, as opposed to what is said of a subject,
is a " t h i s " (Z 8. 1033^21-2); but the description is sometimes applied to
forms (e.g. H I . 1042*29), presumably o n the grounds that form is what
makes matter into countable units ("the cause by w h i c h matter is a some-
t h i n g " , Z 17. 1041b7-8).
A t H I . 1042*26-31, De Anima I I 1. 412* 6-9, Aristotle divides sub­
stance into three kinds, form, matter, a n d " w h a t is [made] out o f t h e s e " .

C H A P T E R 9

T h e r e are other discussions of 4 h e same' at Topics I 7 a n d at / 3. 1054* PP· 208-14

32-^3. Sameness i n number, form, a n d genus (cf. Δ 6. 1016^31) are
distinguished at Topics V I I 1. 152^30-3, but not i n this chapter.

1017^27. Whereas two things c a n be one, v i z . make u p a unity, it is

impossible for two things to be the same thing, as Aristotle acknowledges
at the end of the next paragraph. T h e r e is therefore no excuse for his
saying that e.g. m a n a n d artistic are the same thing because one o f t h e m
coincides i n the other. Instead, his explanation of coincidental sameness
requires h i m to distinguish two senses of *the artistic', *the pale', etc. (as
i n J 6).
" T h e artistic [is] a m a n because it coincides i n the m a n " contributes
nothing to the analysis of sameness a n d is, doubtless for that reason,
omitted by one good M S . ; disagreement o f g e n d e r forbids construing
" t h e artistic m a n " i n it as a single phrase, but that must be the phrase to
which " t h e l a t t e r " i n the next sentence purports to refer.
O n the meaning of *coincidental' see notes o n J 6. 1015^16.
Aristotle's objection to *every m a n is the same as the artistic' shows
that he takes that sentence not i n the strong sense *some (one a n d the
same) artistic thing is the same as every m a n ' , w h i c h is false because there
is more than one m a n , but i n the weaker sense *every m a n is the same as
some artistic thing or other', w h i c h is false because not every m a n is
artistic. H e argues that i f every m a n were artistic, *artistic' would hold
good of m e n i n their o w n r i g h t ; hence no m a n would be coincidentaUy
artistic; hence the identity o f a m a n a n d a particular artistic thing would
not be coincidental. T h i s argument follows Posterior Analytics I 4. 73^26-
28 i n ignoring the distinction between *universal' a n d *in its o w n right'
implied ati5 4. i o o o * i , PriorAnalytics I 1. 24*18, a n d i b . I 8.

49
1017b37 METAPHYSICS J9

W h a t docs Aristotlc m c a n w h e n he says " i n the case ofparticulars we

do so speak baldly'*? ( i ) T h c contrast m a y be between thc universal
proposition *every m a n a n d thc artistic are the same', w h i c h is false, and
thc particular proposition *Socrates a n d the artistic (Socrates) are the
same' w h i c h , like the particular proposition i n ^29 *man a n d artistic are
the same', could b c true. But (i) this interpretation docs not explain the
w o r d " b a l d l y " , a n d (ii) thc contrast it indicates does not turn o n the
alleged fact that *wc do not say " e v e r y S o c r a t e s " as w c do " e v e r y m a n " '.
(2) T h e w o r d " b a l d l y " suggests that Aristotle's contrast is between u n i ­
versal a n d particular и^ofώ. W c m a y say that some m a n a n d the artis­
tic are t h c s a m e ; but i f w c speak baldly, i.e. without such a n additional
*somc', we shall imply the falsehood that every m a n a n d thc artistic
are the same. T h e implication is not carried w h e n , instead of *man', we
use the particular word *Socrates'. T h e objections to this inteφretation
a r c (i) that i n ^29 ( " m a n a n d artistic") Aristotle has violated his own ban
o n *speaking b a l d l y ' ; (ii) he maintains elsewhere that indefinite, i.e.
unquantified, propositions do not imply their universal closure (Prior
Analytics I 7. 29*27-9, c f I 1. 24*16-22, De Interpretatione 7. 17^8).
T h e r e is further ambiguity i n the reason w h i c h Aristotle gives for its
being true to say baldly that Socrates a n d the artistic Socrates are the
same, ( i ) H e might mean that " ' S o c r a t e s ' does not apply to a number
of things" at once; that is, with proper names there is no idiom corre­
sponding to that b y w h i c h we may use the singular of a general name to
speak o f m o r e than one thing, as i n *the F r e n c h m a n is excitable'. T h i s is
a good point, but its truth is independent of the question whether we do
or do not " s a y 'every Socrates' " ; rather it requires—and this may be the
sense of the G r e e k — t h a t w e never mean *every Socrates' by *Socrates'.
(2) Aristotle m a y be making thc bolder, a n d false, c l a i m that *Socrates'
is the name ofjust one thing a n d the expression *every Socrates' has no
use; c f De Interpretatione 7. 17*39-^1.

1018*4. "Sameness is a kind of oneness" because any statement using

*same' c a n be rephrased using *one'. Aristotle appears to maintain that
the converse is also true w h e n he says that things are called the same i n
their own right " i n as m a n y w a y s " as they are called one (the word
translated *in as m a n y w a y s ' is a modern emendation, but the corrupt
readings of our M S S . convey the same sense). I t is h a r d to see the
justification for the latter claim, (i) Nothing here or elsewhere corre­
sponds to the sense of *one' at J 6. 1016^11-17. (ii) T h e sameness of
things whose matter is one i n form answers t o J 6. 1016*17-24, but the
correspondence is not exact, because at least part of the paragraph i n
J 6 treated *one' as a one-place predicate (i.e. gave a type 1 criterion of
u n i t y ; see note o n J 6. 1015^16, paragraph ( B ) ) . (iii) T h e s a m e o b j e c t i o n
faces Ross's suggestion that the sameness of things whose matter is one

150
J9 NOTES ioi8>4

in number corresponds with oneness i n continuity, Δ 6. i o 15^36-1016· 17.

" W h o s e substance is o n e " answers not to the same phrase a t J 6. ioi6^
8-9, but to "indivisible i n f o r m u l a " , 1016*32-^6.
A t the end of the paragraph Aristotle distinguishes two m a i n types of
sameness. O n e " s a y s that a thing is the same as itself", i.e. uses *same' i n a
way w h i c h implies *same thing': example *same m a n ' . T h e other m a y be
truly predicated of two different things: example 'same colour'. E v e n the
former type "treats one thing as more than o n e " because it requires two
subject-expressions; i.e. 'is the same as', unlike 'is one', is always a two-
place predicate.

1018*9. A s Ross notes, the three senses of 'other' do not answer

exactly to the senses of 'the same' i n the previous paragraph. Ross is
surely wrong i n thinking that otherness i n form " r e d u c e s " to otherness
in form of matter: for example, a bronze sphere and a bronze cube are
other i n form but the same i n form of matter.

1018*12. T h e paragraph lists four senses of 'differing'. Aristotle's

account of the first is obscure. I f w e construe the phrase " n o t only i n
n u m b e r . . . analogy" as explanatory o f " s a m e something", we m a y e i t h e r
( i ) follow Alexander a n d Ross i n understanding " n o t o n l y " i n the sense
'only not', w h i c h is difficult, or (2) rely on the fact that it is possible that
over a n interval o f t i m e x a n d ^ should be the same even i n number a n d
yet other e.g. i n quality. (3) Alternatively, the phrase might be construed
as explanatory of " o t h e r " . I f so, "other [ i n number] while being the
same something" would not imply the theory that identity is relative,
since according to Aristotle's usage it is permissible to say of two m e n
that they are different m e n but (because both men) the same animal (see
Δ 6. 1016*28-32). H e does not say whether it is possible for two things to
be other and not the same anything, but probably he would have thought
that this is true of items i n different categories. " O t h e r n e s s i n their
substance": i.e. in some quality i n w h i c h they differ essentially—a differ­
entia i n Aristotle's technical sense.

1018*15. A different classification of senses of'similar' is given at / 3 .

1054^3-13; see Ross.

CHAPTER 10

1018*20. T h e first four of Aristotle's five types of opposites recur i n

Categories 10, w h i c h is a fuller treatment of the same subject, a n d else-
wliere. " T h e points from w h i c h a n d to w h i c h comings to be a n d de­
structions ultimately l e a d " might be opposite ends of a process, e.g. the
state of being (or stuffwhich is) in a quarry and the state o f h a v i n g been
IOl8>20 METAPHYSICS J 10

(or the same stuff w h i c h later has been) formed into a house. B u t i f so,
w h y does Aristotie omit processes of non-substantial change? »22-5 do
not introduce a sixth type of opposite, but argue that 'x is capable of
being F a n d of being G but not of being both simultaneously' is not, as
we might suppose, sufficient to make F a n d G opposites. F o r grey a n d
white are so related, but grey, being m a d e out of the contraries white a n d
black (or pale a n d d a r k ) , is w h a t Categories 10. 12*17 calls a n " i n t e r ­
mediate".

1018*25. Contraries are attributes such as paleness a n d darkness, not

substances such as white m e n a n d negroes {Categories 5. 3^24-7); but the
latter a r e derivatively " s o called from possessing . . . or from being
recipient o f " contrary attributes (»31-2). / 4 , a fuller treatment, defines
contrariety as "greatest difference" (1055*4-5) or " c o m p l e t e difference"
(1055*16). T h e implication that " o n e thing cannot have more than one
c o n t r a r y " (1055*19-20) is allowed to have exceptions at Categories 11.
13^36-14*6. O f the senses given here, the first is exemplified b y j u s t i c e a n d
injustice {Categories 11. 14*22-3), the second by up a n d down {Categories
6. 6*17), the fourth by one a n d m a n y (objects of the same " c a p a c i t y " ,
viz. discipline, see / 4 . 1055*31-2). W h a t is the difference between the
second a n d third senses (genus a n d recipient)? T h e r e are four possi­
bilities, ( i ) N0 difference is intended. Ross refers to De Somno 1. 453^27-9
w h i c h says that a l l contraries are i n the same recipient; but o n a n y
interpretation the analysis i n the present passage is more complex. (2)
Contraries present " i n the same r e c i p i e n t " have to be compatible. B u t
Aristotle's usage elsewhere suggests that contraries are necessarily i n c o m ­
patible, so that " i n the same r e c i p i e n t " cannot i m p l y 'simultaneously'
a n d "differ m o s t " must imply 'incompatible'. (3) Pale a n d dark are
capable ofbeing i n the same recipient, but are not the most different " i n
the same g e n u s " because their possessors do not have to belong to differ­
ent species or forms ( M a i e r , referring to / 9). I f this is right, the second
sense is narrower. (4) O d d a n d even are i n the same genus, but not both
capable of being i n the same recipient, e.g. i n the number thirteen
(Alexander). I f this is right, the third sense is narrower. T h e first sense
is omitted i n / 4. 1055*22-3 a n d apparently excluded at De Generatione
et Corruptione I 7. 324*2; b u t see Ross. Ross regards the fifth sense (*30-1)
as a s u m m a r y of the first four (or rather three), but neither "things
whose difference is greatest . . . b a l d l y " nor "things whose difference is
greatest . . . i n respect o f g e n u s " have appeared earlier.

1018*35. Oi,r2, 1003^34-1004*2 a n d 1004*22-4. " P r e d i c a t i o n " : or

category.

1018*38. Otherness i n form is treated at greater length i n / 8 a n d 9.

" N o t subordinate to one a n o t h e r " : the subject has to be properties or

152
J10 NOTES xoi8*38

kinds of thing, not individuals. O n l y Aristotle's third sense mentions

contrariety, but different kinds of things cannot be i n the same genus, as
required under the first two senses, without either being contraries (e.g.
pale a n d dark) or having contrary properties " i n their substance", i.e.
essentially (Alexander cites fire a n d w a t e r ) : see / 8 . 1058*8-17. T h e third
sense is nevertheless independent of the other two, since fire a n d night
are other i n form i n that sense alone, pale a n d dark i n the first two senses
alone. Contrariety is no longer required i n the last sense, ^6-7.

CHAPTER 11

T h e first three a n d the fifth paragraphs answer roughly to priority i n

(i) time ( c o m i n g t o b e ) , (2) acquaintance, (3) formula, a n d (4) substance
(nature, form), w h i c h are distinguished byAristotle i n m a n y places: Θ
8 ( I , 3, 4), Л 8. 989"15-16 ( I , 4), Z I . 1028*31-b2 ( I , 2, 3), Z 13. 1038b
2 7 - 8 ( 1 , 3 ) , M 2 . 1077*19 (1,4) a n d ^2-/^{^,^),PhysicsVlllη,2^o^ιη-ι^
( i , 4) a n d 261*13-14 ( i , 4). I n the discussion ofpriority a n d posteriority
i n Categories 12, the first two senses there distinguished correspond to
(i) ( " t i m e " ) a n d (4) ("not reciprocating as to implication o f e x i s t e n c e " ) .
F o r further parallels see Ross.

1018^9. W h a t is prior or posterior must be a member of a series.

According to Aristotle every series must have a n " o r i g i n " , w h i c h is
either its first member or something outside the series (as a parent is the
origin of the developing stages ofhis child). W h a t the origin o f a member
of a series is wiU depend either o n the nature of the member or on other
things: e.g. the number 15 has 1 as origin b y its nature as a number
{Δ 6. 1016^17-21), but would have 'love' as its origin i n a game of tennis,
10 as its origin i n the m i n d of someone testing how m u c h further t h a n
10 feet he c o u l d j u m p . T h e " m i d d l e " a n d " e n d " are perhaps the centre
and circumference of a circle. I n the temporal order, only the present
can fill the role of origin; w i t h the consequence that, to avoid the
absurdity of counting the Persian W a r s as prior to the T r o j a n because
nearer Aristotle's (or our) present, Aristotle is obliged to m a i n t a i n — n o
less a b s u r d l y — t h a t what is prior i n time is i n some C2isesfurther from its
origin. T h e " t h i n g s " arranged i n order of priority i n respect of capacity
are (although Aristotle expresses himself i n the neuter) more a n d less
powerful m e n , as Alexander says.

1018^30. Priority i n acquaintance is more adequately discussed at

Z 3. 1029b3-12 a,ndPhysics I 1.

1018^31. T h i s paragraph makes two claims about priority i n formula:

153
1018b31 METAPHYSICS Jll

(A) that universals are prior to partictdars i n formula but posterior i n

perception, a n d (B) that e.g. the artistic is prior to the artistic m a n i n
formula but not prior i n substance or being.
(A) Priority i n formula a n d i n perception are presented as forms of
priority i n acquaintance ( " a m o n g these", ^31), a n d so correspond to the
distinction of Physks I 1. 184*16-^1^ ^ n d PosteriorAnalytics I 2. 71^33-
72*4 between what is more intelUgible i n its nature a n d w h a t is more
intelligible to u s ; the latter passage says that " r e l a t i v e to us, things
nearer to perception are prior a n d more intelligible". W h e n Aristotle
claims that the latter form of priority belongs to particulars, the former
to universals, it is unclear whether his contrast is between species (forms)
a n d genera, or between individuals a n d species-cum-genera (cf. Δ 2.
1014*17, 21). W h a t he says about formulae only fits the contrast be­
tween species a n d genera: for the formula of a n individual, say Callias, is
the same as, a n d so could not be posterior to, that o f i t s species, m a n ;
whereas the formula of the species m a n is posterior to, as containing, that
ofits genus a n i m a l (cf. Θ 8. 1049^16-17). W h a t he says about perception,
on the other h a n d , only fits the contrast between individuals a n d species-
cum-genera: for perceptual discrimination of specific characters comes
within h m n a n powers, he thinks, later than perceptual discrimination of
generic characters, a n d so could not be prior {Physics I 1. 184^12-14,
children caU aU m e n daddies); whereas the objects perceived b y means of
discrimination o f s u c h characters, whether specific or generic characters,
are i n the prinaary case individuals {De Anima I I 5. 417^22, Posterior
Analytics I I 19. i o o * i 6 - b j ^ ^ g. 982*25).
(B) Aristotle next states w h a t is reaUy the same sense of *prior i n
formula' i n terms not of prior acquaintance but of one formula's being
contained i n another; cf. Z 1. 1028* 35-6 "for i n the formula ofany thing
it is necessary that the formula of a substance be a constituent". T h i s
linguistic formulation invites the further contrast conveyed i n the final
sentence: whereas the contained w o r d *artistic' c a n exist o n its own, the
non-verbal item artistry (the artistic) cannot. A n alternative translation
ofthe last sentence is " i t is not possible for artistic to bewithoutsomething
artistic b e i n g " , w h i c h would make the artistic not only not prior but
posterior to the artistic something.

1018^37. A box might be caUed straight, meaning *straight-edged',

but it would be the box's edge, a line, that was straight " i n its o w n
r i g h t " ; see Δ 18. 1022*29-35. I t is not clear i n what sense Aristotle holds
that a Une is prior to a surface. *Surface' c a n be defined as 'line that has
moved' {De Anima 1 4. 409*4), but equaUy *line' c a n be defined as *limit
o f a surface' (cf. Topics V I 4. 141^22). L i n e s are " m o r e intelUgible, speak­
ing b a l d l y " than planes. Topics V I 4. 141^6—^we are not told w h y . Qua
parts of surfaces lines would be prior according to the next paragraph,

154
J l l NOTES 1018b37

but they are not parts w h i c h surfaces are " m a d e out o f " (cf. Physics V I
I . 231*24).

1019*1. T h e reference to Plato has not been identified; see Ross.

A subject is that o f w h i c h something else is said or predicated. A c c o r d ­
ing to Aristotle, a predicate is something else than its subject only i f
it belongs to the subject coincidentaUy (see Г 4. 1007*20). Therefore
subjects are prior i f a n d only i f everything is prior to its coincidental
predicates. Aristotle appears to have regarded such predicates as i n d i ­
v i d u a l to their subjects: e.g. *Callias is pale', i f true, predicates Callias'
pallor of Callias (see Categories 2. 1*20ff., but the passage is taken other­
wise by G . E . L . O w e n , Phronesis 1965). H e n c e (i) Callias c a n exist
without his pallor a n d (ii) his pallor cannot exist without h i m . T h i s
establishes that subjects are prior, a n d thus that substances are prior
w h e n they are subjects. B u t (iii) substances are always subjects {Categories
5. 2*11-14, Z 3. 1029*1-2, but contrast Z 3. 1029*23-4), a n d (iv) every
other category of thing is a (possible) predicate of (some) substance
{Z 3. 1029*23-4); therefore substances are always prior to everything
else, i.e. are primary ( Z i . 1028*29-36).
T h e end of the paragraph is difficult, chiefly because of the use w h i c h
Aristotle makes i n it of the thesis ( T ) that parts of a thing c a n survive
dissolution of the thing, a n d the matter of a substance the dissolution of
that substance. According to the definition of *prior i n nature a n d sub­
stance', T shows that parts a n d matter are prior. A n d so they are, A r i s ­
totle says, but only " i n respect of c a p a c i t y " or potentiality: for w h e n
the whole exists, the parts exist only potentially {Physics V I I 5. 250*24-5),
a n d w h e n the substance exists, its matter exists only potentially. T h u s
parts a n d matter are prior i n the sense that their potential-or-actual
existence is implied by, but does not imply, the existence of their wholes
a n d substances. So far there is nothing puzzling except that Aristotle
overlooks a n equally good argument for saying that parts a n d matter are
neither prior nor posterior i n respect of capacity: v i z . that the merely
potential existence of a part both implies a n d is implied by the actual
existence of its whole. B u t now comes a major difficulty: i n what sense
are parts posterior to wholes, a n d matter to substance, " i n complete
r e a l i t y " ? I n Z 10 we are told that " p a r t s as matter, i.e. into w h i c h a
thing is divided as into m a t t e r " are posterior (1035^11-12); a n d Aristotle
supports this assertion by the claim that w h e n a man's finger is cut off*
the m a n remains a m a n but the finger does not remain a finger except i n
n a m e (1034^28-30, 1035^23-5). T h a t claim cannot be used, however, i n
the present argument; for it makes the posteriority o f p a r t s a n d matter
depend on something inconsistent w i t h T , not, as the present argu­
ment unexpectedly asserts, on T itself. N o r c a n *posterior' here m e a n
*after' (although " w h e n the others are dissolved" translates a participial

155
ioi9*i METAPHYSICS J l l

construction w h i c h might have temporal force); for ifit d i d , (i) since matter
a n d at least certain parts c a n exist i n complete reality not only after b u t
also before the coming to be oftheir wholes, they would be both prior a n d
posterior i n respect of complete reality; (ii) Aristotle presents posteriority
i n respect ofcomplete reality as a species of posteriority i n substance a n d
nature, but *posterior' i n the latter sense implies 'cannot exist without',
w h i c h is incompatible w i t h ' c a n exist after'. F o r further discussion of
senses i n w h i c h parts are prior or posterior to wholes see Z i o .
T h e r e remains the question w h y Aristotle justifies his contentions i n
this p a r a g r a p h b y reference to the doctrine that " t h e r e are m a n y ways of
b e i n g " ( c f Δ η), ( i ) 'Prior i n respect o f c a p a c i t y ' a n d 'prior i n respect of
complete reality' are doubtless to be analysed i n terms of 'being poten­
tially' a n d 'being actually', (ii) I n connectionwithsubjectsandsubstances
the point seems to be that A c a n be prior to B i n " n a t u r e a n d substance"
only i f A a n d B ' a r e ' i n different w a y s ; i.e. that priority i n nature is
priority i n the manner of existence. I n the same spirit Z i tells us that
substance is not only " t h e p r i m a r y thing-that-is" (1028*14) but also
" t h a t - w h i c h - i s i n the p r i m a r y w a y " (1028*30). Aristotle seems to infer
from ' 5 ' s existence is dependent o n ^ ' s ' to ' 5 ' s existence cannot be under­
stood without reference to ^4's'; but he does not defend the inference.

1019*11. A part can be without the whole i n respect ofdestruction, i.e.

w h e n the reason for the absence of the whole is that the whole has been
destroyed: for instance, a half-apple c a n be without the other half, w h e n
the other h a l f h a s been eaten. A whole must be without its parts i n respect
of coming to be, because the 'absence' of the parts (in the sense of their
not being actual) is implied by the whole's h a v i n g come to be. T h i s is a
more promising contrast t h a n Aristotle made i n the previous p a r a g r a p h ,
but stiU not adequate to show that wholes are i n any w a y prior to their
parts. Aristotle does not explain how the earlier senses of 'prior' are to be
reduced to "these l a s t " .

C H A P T E R 12

Dunamis (capacity) a n d to dunaton (the capable, possible) are also dis­

cussed i n De Interpretatione 12 a n d 13, a n d more fully i n Metaphysics Θ.
Most o f t h e present chapter except 1019^22-33 is concerned w i t h c a p a ­
cities to change (active capacities, abiUties) or be changed (passive
capacities, liabilities). Aristotle barely considers capacities tofunction i n
particular ways, e.g. to lead a n a r m y {DeAnima I I 5. 417^32) or to think
about a problem or view a scene (cf. Θ 6. 1048^18-35, where these
activities are distinguished from 'changes'); or capacities to be i n
particular states, e.g. to be healthy or noiseless. A t Θ 1. 1045^35-1046*2

156
J12 NOTES x019*15

he acknowledges that there are more kinds of capacity than "those so

caUed i n respect o f c h a n g e " , but treats the latter as " m o s t fundamentaUy"
so caUed. 1019*23-6 mentions walking a n d speaking, w h i c h are not prop­
erly changes.
Aristotle's choice of the word 'arche' (translated " w h a t originates")
is unhelpful. Doubtless a teacher's capacity to teach a n d his pupil's to
learn c a n properly be regarded as origins o f c h a n g e i n the p u p i l ; but so
equaUy could the teacher hinwelf, his willingness or decision to teach,
the pupU's willingness or decision to attend, or the lesson imparted. T h e
chapter furnishes a n example of the limited value of paraphrase i n the
explanation o f m e a n i n g ; Aristotle would have done better to bring out
the meanings o f c a p a c i t y ' merely by giving examples ofcapacities, as he
saw elsewhere (Θ 6. 1048*36-^^).

1019*15. " O r qua o t h e r " : i.e. i n the changer qua other. T h e doctor m a y
doctor himsetf, but not qua himseU*. I s he doctored qua other ( i ) because
his patient does not have to be himseU*or (2) because his patient does not
have to be a doctor? "Qttfl being doctored" suggests (2); but i n the places
where Aristotle discusses the k i n d of capacity here set aside he seems to
have i n m i n d changes w h i c h таш&Х be worked o n the changer hinwelf (see
Θ 6. 1048*27-30, 8. 1049b8-10).
1019*19. " O r qua o t h e r " is again elliptical, i n its first occurrence for 'or
i n the changer qua other', i n its second for *or b y the changed thing qua
other'. T h e point of the second sentence is evidently not to introduce a
sense of *capable' corresponding to passive capacities (that comes at
1019*35), but to indicate that each passive capacity is matched to a n
active one. Aristotle adds that passive capacities sometimes exist only i f
the affection " i s for the b e t t e r " : e.g. flexibiUty is a dunamis i n a w a t c h -
spring but not i n a pit-prop.

1019*26. 'Dunamis' c a n m e a n *strength', *resistance to change*. I f a

girder is incapable ofbeing bent, or h a r d to bend, it is natural to say that
that is " o w i n g to a c a p a c i t y " — i t s strength—^rather than to a n incapacity.
T w o points are compressed into the p a r a g r a p h : resistance to change is
a true capacity; liability to impairment is not a true capacity. Aristotle
does not deny that liability to improvement is a true capacity.

1019*33. T h e five senses of 'capable' correspond to the first, second,

second again, fourth, a n d third senses of *capacity'. " W h a t c a n keep a
thing the s a m e " : i.e. prevent it changing. H a v i n g remarked that passive
capacities include resistances to change, Aristotle adds that things c a n be
actively capable i n the same w a y ; e.g. creosote has the capacity (not to
improve wood but) to prevent wood rotting. T h i s ntight stiU be regarded
as a capacity to effect change—the creosote prevents the wood from
changing i n one w a y by d i a n g i n g it i n another. Nevertheless the

157
xoi9*33 METAPHYSIGS J12

description of something as preventing a change does not describe it as

changing anything. I f this is Aristotle's thought, the w o r d h e substitutes
for *change', translated *alter' {'metabalUin\ Ross 'change'), is h a r d l y
helpful. ( H e often uses 'metaballein' a n d 'kineisthaV as s y n o n y m s ; w h e n they
are distinguished, the latter usually excludes coming to be a n d destruc­
tion, seePhysics V i . 225*34ff., 5. 229*31, ^14.)
T h e parenthesis reverts to the point m a d e at 1019*26-33 that a thing
being destroyed or damaged exhibits a weakness or incapacity rather
t h a n a capacity. T h o u g h weak, such a thing must be capable of being
destroyed, since *is being' implies *can be'. T w o solutions a r e offered:
either to be capable is to possess something, be it a capacity or a n i n ­
capacity ("the lack of t h i s " ) ; or Ч0 be capable' has two senses, *to possess
a capacity' a n d *to lack a capacity'.

1019^15. " I n the same w a y " : i.e. for the same reason. T h e reason w h y
the boy a n d the e u n u c h a r e incapable of begetting is given i n their
descriptions as boy a n d e u n u c h ; not so w i t h the grown m a n . Aristotle
thinks that *boy' a n d *eunuch' are different sorts ofreason too: *because
he is a boy' amounts to *because of his age' ( " t i m e " not " c h a r a c t e r i s t i c
ofits possession").

1019^21. " I n respect of this i n c a p a c i t y " : or rather *in respect of

incapacity', see 1019^35.

1019^22. O n this sense compare Prior Armlytics 1 13, De Interpretatione

12,13. T h e " f a l s e h o o d " i n the second sentence m m t be that a diagonal be
commensurable (sc. w i t h its s i d e ) — c a l l it S—^not that it is impossible
that S, w h i c h is a truth. W h a t , then, is S's contrary? A c c o r d i n g to De
Interpretatione 7. 17^22-3 a n d 10. 20*16-19 contraries cannot be true
together but c a n be false together (let us call these strict contraries).
Since it is not possible for a necessary proposition to be false even o n its
own, it foUows that no necessary proposition c a n be the strict contrary of
a n y proposition; b u t Aristotle says that the contrary oiS is " n o t only
true but also a necessity". T h i s suggests that " c o n t r a r i e s " are to include
contradictories as w e l l as strict contraries. W e should presumably
understand " a d i a g o n a l " i n S to m e a n *any (some) diagonal' not *every
diagonal' (although the latter too would make S a falsehood a n d i m ­
possible). I f s o , S is o f a f o r m w h i c h i n a n y case has no conesponding strict
contrary, but has a contradictory i n *no diagonal is commensurable'.
Aristotle holds, then, that since the latter is necessary, S is impossible.
H i s general c l a i m is that i f a n y propositions are contraries i n the loose
sense (cannot be true together), a n d i f o n e o f t h e m is necessary, the other
is impossible; cf. the thesis a t J 5. 1015^6-9. H e is not setting out that
part of the square of opposition for m o d a l words (see De Interpretations
13. 22*24-31) w h i c h states that necessarily-/? a n d impossible-that-j& a r e

158
J 12 NOTES 1019b22

(strict) contraries; for the (contradictory) contraries S a n d not-S do not

themselves include modal words.
H a v i n g shown that the falsehood S is impossible, Aristotle draws the
inference (^26) that it is necessarily a falsehood. I n the fourth sentence
" t h i s " is no longer S, whose (contradictory) contrary has been shown to
be (not only possible but) necessary, but impossible-that-6', w h i c h has
possible-that-5' for its (again contradictory) contrary: here Aristotle
makes a modest contribution to the square of opposition for modal
words. T h e m a i n part of the sentence is a n attempt to define the possible
i n terms of*necessary', *contrary', a n d Talse'. T h e " c o n t r a r y " mentioned
second is not the contrary of possible-that-/> but, as Aristotle's example
shov^, the (contradictory) contrary of p, v i z . not-/> ( " h e is not sitting
d o w n " ) . So the definition runs: it is possible that/> i f it is not necessary
that it is false that not-j&, that is, i f it is not necessary that p, T h i s defini­
tion is both incomplete, since the possible includes also the necessary,
a n d wrong, since the non-necessary m a y be impossible (see diagram).

true false
^ ^ ^
necessary ^ impossible
contmgent

possible

*possible', 1019b28-9

Aristotle's example does no more than show that possible-that-/> a n d not-

necessary-that-/> are compatible (they are, of course, sub-contraries—
one of them must be true). T h e same error occurs at De Interpretatione
13. 22*27, a n d is corrected at 22^10-28. I f w e supplement Aristotle's
definition w i t h the point made i n the first half of the sentence, that
*possible' a n d *impossible' are contraries, his contention will be that
*possible' is equivalent to *neither necessary nor impossible'. T h i s is one
o f t h e senses w h i c h he gives to 'admissible' 2itPriorAnalytics I 13. 32*18¬
21; it is also the sense sometimes given to the E n g l i s h *contingent'.
H a v i n g wrongly defined *possible-that-/>' as *not necessary that the-
contrary-of-j& (i.e. not-p) i s f a l s e ' , A r i s t o d e n e x t (^31-2) disconcertingly
misreports a n d corrects his o w n error: *possible-that-/>' signifies *not
necessary that p is false'. "Signifies what is t r u e " : as before, this c a n be
defended only if it means to convey that *possible-that-j5>' a n d 4rue-that-/>'
are compatible; c f De Interpretatione 13. 23*7-9. " A n d i n another what
adnaits of being t r u e " : the participle translated *what admits of' (or
*admissible', 'endechomenon') is elsewhere rendered b y *possible', but here
a verbal difference is caUed for. Does Aristotle mean that *admissible'
shares the meanings of *possible' already given i n the paragraph, or that
it specifies a new one? I n the table at De Interpretatione 13. 22*24-31 the

159
IOI9^22 METAPHYSICS J12

two words are made equivalent, a n d PriorAnalytics I 13. 32*18-21 gives

as senses of *admissible' both of the senses of *possible' (*not impossible'
a n d *neither impossible n o r necessary') confused b y Aristotle i n the
present paragraph. Aristotle's actual usage, i f it contrasts *admissible'
w i t h *dunaton' at aU, contrasts it w i t h senses outside the p a r a g r a p h (i.e.
w i t h *capable', not *possible').

1019^33. Greek mathematicians used 'dunamis' to m e a n *square' (of

a n u m b e r ) . F o r other uses see H e a t h , Mathematics in Aristotle, 207-8.

1019^34. " T h e s e things" are not, of course, squared numbers b u t

*possible' things. Aristotle contends that the senses of *capable', as
opposed to *possible', aU derive from *actively capable'. T o be liable to
change is to have something else (or yourself) capable of changing y o u ;
to be resistant to change is to have nothing capable of changing y o u at aU,
or easily (but this equivalence cannot explain the existence of the sense
*resistant to c h a n g e ' ) ; to be (we might say) amenable to change is to have
something capable of changing y o u for the better ("possessing it i n a
particular w a y " ) . T h e summary omits the derivative sense of *actively
capable' given at 1019^11-15.

CHAPTER 13

Q u a n t i t y is treated more fully i n Categories 6. H e r e as there Aristotle

employs not the abstract noun corresponding to *quality' b u t a n a d ­
jective ( L a t i n *quantum') w h i c h c a n also m e a n *how m u c h ? ' , a n d m a i n l y
discusses the possessors of quantitative properties, not the properties
themselves.

1020*7. T h e constituents o f a quantity must have " t h e nature o f a one

a n d a t h i s " : hence, as Alexander says, they do not include forms (the
constituents o f a genus), w h i c h are not continuous a n d so not one {Δ
6. i o i 6 ^ i o ) , nor properties (the constituents o f a state ofaffairs), w h i c h
are not 4hises' (cf. Δ 25).
Aristotle proposes two criteria to distinguish pluralities from m a g n i ­
tudes: the former are (i) countable—i.e. admit p l u r a l s — a n d (ii) i n d i ­
visible into continuous parts. T h e second of these (cf. Categories 6.
4^20-5*14) is not helpful: 'continuous' means 'having a c o m m o n
boundary' {Categories 6. 4^25-6, Physics V 3. 227*22), but w e are not
given rules for deciding whether e.g. adjoining rooms have a common
boundary or two boundaries w h i c h touch. Aristotle does not acknow­
ledge, but neither does h e deny, that the same thing m a y be both a
plurality a n d a magnitude.
" B r e a d t h s . . . d e p t h s " : i.e. areas a n d volumes—Aristotle has no

160
 NOTES i030*7

separate words. Since length etc. are being treated as possessors of

quantitative properties, it is appropriate to say that a line is a length,
etc. B u t a length o f w h a t sort? ( i ) E u c U d defines 'line' as 'length without
breadth', but Aristotle's definition is certainly different from that, (a)
Alexander takes 'line' etc. simply as answers to the questiom 'what is a
length?' etc. (3) Ross supposes that 'linuted' goes not only w i t h 'pltffaUty'
but also with 'length', 'breadth', a n d 'depth'. I f s o , w h a t is its meaning?
(i) I n E u c l i d , 'finite'; but it is reasonable to say 'lines must be finite,
lengths need not be' orüy i f lengths are regarded as properties of Unes,
not items of w h i c h lines are examples, (ii) I f 'limited' means 'discrimin­
able (sc. not only i n respect of quantity)', the mearung nught be that
e.g. the prime meridian a n d the 10' meridian oflongitude are the same
length but not the same l i n e ; but the facts to w h i c h this interpretation
draws attention are a good reason for refusing to identify a n y length with
any line.
A separate problem arises over the defiiution of a number as a linuted
plurality, (i) A g a i n , this might m e a n that 'infinite ntmiber' is a contra­
diction, while 'infiiute plurality' is not, but such a n J e is not hinted i n
a n y of the other Aristotelian definitions of number to w h i c h Ross refers.
(ii) ' T w o different numbers m a y be the same plurality.' T h i s seems
absurd, (iii) / 6. 1057*3 defines a number as " a plurality measurable b y
o n e " ( c f N I . 1088*5-6, A 6. 1016^17-21). H e n c e the point nught be
that a number has to be a n integer, a plurality need not b e ; or (iv) that
e.g. a sentence c a n be plural i n more than one w a y — s i x words, ten
syllables—but orüy one of these w i U be the number of the sentence,
depending whether words or syUables or something else are the units of
w h i c h sentences are made {cf.A 6. 1016^17-31). (v) T h e r e is a difference
in meaning between ' a plurality ofso-and-so's' a n d 'n so-and-so's', where
n is a n u m b e r ; but this ofcourse does not establish a difference between
' a plurality ofso-and-so's' a n d ' a number ofso-and-so's'.

1020*14. See notes on 1020*26.

10ao*17. Largeness, narrovraess, etc., differ from a line i n two w a y s :

(i) they are affections of a line a n d so not substances (strictly speaking
lines are not substances either: M 3); (ii) their definitions, although i n ­
cluding the w o r d 'quantity' (for they are "affections of a quantity i n their
o w n r i g h t " ) , do not include the whole expression ' a certain quantity'.
Aristotle divides "affections of a quantity i n their o w n r i g h t " into two
types: those represented b y comparative adjectives, e.g. 'larger', are
" s o caUed . . . w i t h reference to one a n o t h e r " , those represented b y
positive adjectives, e.g. 'large', 'broad', are " s o caUed . . . i n their o w n
right". T h i s is verbaUy inconsistent w i t h Categories 6. 5^16-18, " n o t h i n g
is caUed large or small just i n its o w n right, but wdth reference to some­
thing e l s e " ; but Aristotle's point is clear e n o u g h — ' C a U i a s is large' is,

161
I020^I7 METAPHYSIGS J13

but *Gallias is larger' is not, intelligible without knowledge of context

(see A c k r i l l o n Categories 7. 6*36). U n l i k e the Metaphysics, the Categories
classifies both these types as relatives rather t h a n quantities, a n d 6.
5^11-29 defends that classification of the second type.
W h a t are the " o t h e r things", sc. than quantities, to w h i c h "these
names are also transferred"? ( i ) T h e y might be such things as a n agree­
ment, to w h i c h w e c a n apply the predicate *broad' but w h i c h is not a
quantity because we cannot apply e.g. 4wo-foot broad', 4wice as broad'
(cf. Ross a n d the contention i n Categories 6. 5^11-29 a n d 6*19-25 that
quantitative affections have no contraries a n d admit of no degrees, true
of 4wo-foot broad' but not of ' b r o a d ' ) . (2) Aristotle m a y simply m e a n
that quantitative affections c a n be predicated of coincidental quantities
('the artistic'), not only ofquantities i n their own right (4he b o d y ' ) ; this
would be true of both *broad' a n d *two-foot broad'.

1020*26. A n F i s a quantity coincidentaUy if either (i) some but not aU

F s have quantitative properties (or it is possible but not necessary for a n
F to have such a property, 1020*14-17, *26-8), or (ii), where *Q,' is a
quantitative property w o r d , 'Fs are uses 'Q,' i n a derivative sense
(«28-32). (ii) recurs at Categories 6. 5*38-^10, but i n that chapter time is
treated as a p r i m a r y quantity. T h e clause beginning " I m e a n " explains
" a f f e c t i o n s " , v i z . derivatives. Non-derivative quantities are " a m o u n t s
by w h i c h " a given thing is changed: e.g. the twenty miles w h i c h the
train has covered, the twenty problems w h i c h the computer has solved.
T h e train's actual movement or the computer's activity (change) is a
derivative quantity w h i c h must be defined i n terms ofthese; a n d lengths
o f t i m e must i n t u m be defined i n tenns ofmovement or change (see the
difficult chapters o n time i n the Physics, I V 10, 11).

GHAPTER 14

T h e more ambitious treatment of qualities i n CategoHes 8 classifies them

quite differently.

1020*33. " Q u a l i f i c a t i o n " : i.e. (here) quaUty, but Aristotle's w o r d is

the adjective ( L a t i n *quale') w h i c h at Categories 8.10*27-9 be distingiUshes
from the abstract n o u n ( L a t i n *qualitas'), a n d I have followed A c k r i l l
i n reserving *quality' for the latter. Although differentiae are here d e ­
scribed as qualities, a n d qualities i n the " m o s t f u n d a m e n t a l " sense
(1020^14), Categories 5. 3*22 tells us that they, like substances, are " n o t i n
a s u b j e c t " , from w h i c h i t noight be thought to foUow that they are not
qualities. Aristotle is i n a dUemma: differentiae answer the question
'quaUs?' b u t also reveal, o r a t least help to reveal, essence (thus at
Z 12. 1038*19 he says " t h e last differentia is the substance of the actual

162
Δ 14 NOTES 1030*33

t h i n g " , but contrast Topws I V 2. 122^16, V I 6. i43'32). T h c right

description is 'essential quality', but the system ofcategories, with its odd
contrast between substance (or essence) a n d quaUty, cannot easily
accommodate that notion. " T h e differentia i n respect o f s u b s t a n c e " : i.e.
differentia i n the technical sense, not mere difference.

1020^2. I n Aristotle's day numbers were often represented b y m u l t i ­

dimensional arrangements of dots: thus 4 was a square a n d also tetra-
hedral number, 6 a triangular a n d also oblong number. A number with
two prime factors {"x t i m e s 7 " , literally 'so m a n y times so much') could
be represented by a rectangle, a number with three ("x times_y times z")
by a rectangular-sided solid; see H e a t h , Mathematics in Aristotle, 208-^.
I f 'without angles' states the differentia of circle (1020*35), it might b c
argued that 'solid' states the differentia of a certain k i n d ofnumber, a n d
in general that aU the qualities of changeless things (second paragraph)
are differentiae (first paragraph). A t 1020^15-17 Aristotle accepts
the conclusion of this argument, b u t elsewhere h e offers two replies:
(i) some invariable conjunctior^ of properties do not delinüt any form
or species, e.g. 'male h u m a n ' discussed i n / 9 (this is one version of the
doctrine o f n a t u r a l k i n d s ) ; (ii) even ifsolid numbers did m a k c a species,
solidity could be a coincident, a n d so not a differentia, of that specics,
just as having three sides is a n " i n v a r i a b l e " coincident of a triangle {Δ
30. 1025*30-4). A t Aristotle generalizes his examples into the c l a i m
that any "comtituent of the substance", i.c. necessary property, w h i c h
like solidity i n numbers is a non-quantitative property, must be a quality.
W h a t has happened to the other categories? T h e final parenthesis seems
intended to explain w h y the qualification " a p a r t from q u a n t i t y " is
necessary even i n the case ofnumbers: for nvunbers do have quantitative
properties (e.g. being 2 X 3) distinct from their substance (being 6).

i o 2 0 * 8 . " A f f e c t i o n " m a y be m c a n t to cover aU the types of quality

distinguished i n the Categories as states, dispositions, natural capacities,
affective qualities, affections, figure, a n d shape. Modification (tradition­
ally 'alteration') is regularly defined b y Aristotie as change of quality or
affection: e.g. Categories 14. 15^12.

1020^12. I n the Nicomachean Ethics, I I 5. 1106*n-12 exccUcnce a n d

badness (virtue a n d vice) are classified as states; cf. Δ ao. 1022^10-12.

1020*13. "Affections ofchangeables qua c h a n g i n g " : i.e. properties i n

respect of w h i c h they do change. " W h a t is capable of changing or
functiorüng i n thu w a y is g o o d " : even ifthis proposition were a n analytic
truth it would estabUsh o i J y that excellence is a capacity, not a n affec­
tion. NuomacheanEthics I I 5. 1105b1q-1106*13 rejects both classifica­
tions, taking 'affection' i n the seme 'feeling'. I n the last sentence it is

163
I02obl3 METAPHYSICS J14

not clear whether Aristotle means that the goodness of inanimate things
is not properly said to be a quality (Alexander) or (as Ross prefers) that
inanimate things are not properly said to be good.

C H A P T E R 15

x020^26. T h i s chapter has Httle i n сопмпоп w i t h the longer treatment of

relatives i n Categories 7. T h e threefold classification that it foUows fails to
cover some things, e.g. large a n d larger, acknowledged as relatives i n the
Categories.

1020^32. T h i s p a r a g r a p h is discussed by H e a t h , Mathematics in Aristotle,

209-11. Aristotle's " n u m e r i c a l relatives" are terms ofratios. {a) Double
a n d h a l f (sc. its haU") stand i n the ratio 2 :1, {b) multiple a n d (its) sub­
multiple i n the ratio n:i, {c) one-and-a-half a n d its reciprocal i n the
ratio 3:2 a n d {d) one-and-a-bit a n d its reciprocal i n the ratio n + i:n. I n
{a) a n d {c) the tenns are " r e l a t e d by a definite n u m b e r " , i.e. the ratios
consist o f a c t u a l n u m b e r s ; i n (b) a n d {d) the ratios are algebraic. I n {a)
a n d {b) the terms are related by " a number relative to o n e " , i.e. by a
ratio expressible as a n integer; i n {c) a n d {d) the ratios are not expressible
as integers but are numbers " r e l a t i v e to themselves", i.e. to numbers (one
not being a number, Δ 6. 1016^17).
I n the fourth type of ratio the term is " r e l a t e d to its reciprocal by a n
indefinite n u m b e r " (according to the emendation adopted i n the O C T ;
Ross's alternative emendation gives "indefinite relative to its r e c i p r o c a l " ) .
A t *3 Aristotle adds a fifth type {e), i n w h i c h one term is " n u m e r i c a l l y
whoUy indefinite relative t o " the other. T h e words suggest a ratio even
more algebraic than {b) a n d {d), e.g. m:n, but Aristotle's example shows
that he has i n m i n d a qmte different point: that w h e n one term exceeds
another—is " s o m u c h a n d something a g a i n " — t h e two m a y be i n ­
commensurable. I n w h a t sense are such relations indefinite? H e m a y
m e a n that the formvda 'so m u c h a n d something again:so m u c h ' is even
less specific than *n+1: n', i f the *something again' need not be c o m ­
mensurable w i t h the *so m u c h ' (in *8 " e i t h e r equal or u n e q u a l " does not
correctly express w h a t must be his meaning, viz. 'either conunensurable
or incommensurable'). I n that case he ignores definite irrational relations,
such as ^2:1. O r he m a y m e a n that irrational relations are indefinite
i n the quite new sense 'not nvmierically specifiable', given that irrationals
are not to be deemed numbers ( " n u m b e r is not said ofthe non-commen­
s u r a b l e " ) . I n that case the fifth type embraces not only indefinites i n the
first sense, like фг:п, but also definites i n the first sense, Uke ^2:1.

Z 0 2 i * 9 . T h i n g s e q u a l , sinülar, a n d the same are n u m e r i c a l relatives

because (i) their definitions bring i n the word 'one' a n d (ii) one is the

164
Δ 15 NOTES io3i*9

measure of a n u m b e r . F o r t h e l a t t e r s e e J 6. 1016b17-31. E q u a l s stand

i n the ratio 1 :1 but, since one is not itself a number, this does not make
them numerical relatives i n the first sense.
" W h o s e substance is o n e " is ambiguous (cf. 1016^5-6 a n d 8-9). ( i )
Since the quaUty οΐχ andy is one when both are e.g. pale, it ought to b c
that their substance is one w h e n both are e.g. men, even i f not the same
m a n . (2) B u t i f * and_;> m a y count as the same even w h e n they are not the
same i n number, it is not clear w h y their being e.g. the same colour is tu>t
allowed to count as a case of sameness, but only of s i m l a r i t y . T h i s
suggests that 'one i n substance' here mearw 'one i n number'. C f . B
4. 999b20-2, Z 13. 1038b9-10, 14-15, Z 16. 1040^17.

1021*14. T h e same verb 'poiein' is translated 'act' a t the beginning of

the paragraph (as it was at 1020^30) but 'produce' at the end. T h e sense i n
w h i c h numerical relatives have " a c t u a l functionings" is not described
i n a n y of Aristotle's surviving works. Probably he means that e.g. a n
architrave whose length stands i n the ratio i-6i8 to the height of a
column does not act on the c o l u m n (actual functioning " i n v o l v i n g
c h a n g e " , c f Θ 6) but does actually stand i n that r a t i o ; see Ross. I t is not
clear h o w far Aristotle w o u l d wish to extend the classes of capacity-
relatives a n d functioning-relatives to cover expressions w h i c h , like
'father', are not granMnatically derivative from verbs. H o w about 'slavc',
for instance?

10a1"26. T h i s is a difficult p a r a g r a p h . Aristotle states four propositions

about the correlative pairs measurable/measure, knowable/knowledge,
thinkable/thought, a n d visible/sight:

{a) although the thinkable (etc.) is a relative, it is not caUed thiiJ^-

able relative to, or of, something (*26-30)
{b) a thought is caUed a thought relative to, o r of, something (ib.)
(c) 'thiIжable' signifies 'that of w h i c h some thought is a thought'
(see *31)
{d) a thought is not relative to that o f w h i c h it is a thought (*32).

T w o preliminary remarks m a y be made. ( A ) Categories 7 agrees that the

knowable a n d the perceptible a r e not " c a U e d just w h a t they a r e of
something e l s e " , but offers a different reason for classing them none the
less as relatives, namely that " w h a t is knowable is knowable by k n o w ­
l e d g e " etc. (6^34-6), a n d " w e call relatives aU such things as are said
to b e j u s t what they are, of or t h a n other things, or in some other way r e ­
lative to something else" (6*36-7). (B) L a t e r i n the same chapter (7^22-
8*12) Aristotle makes the good point that knowledge a n d thc knowable
(and perception a n d the perceptible) are not "simultaneous by n a t u r e " ,
that is, they do not "reciprocate as to implication ofexistence" {Categories

165
1021*26 METAPHYSICS J15

13. 14^27-8). T h e appUcation of this point to the present p a r a g r a p h is

unfortunately spoilt b y Aristotle's (i) defining the thinkable as " t h a t of
w h i c h there is thought" (not *can b e ' ) , a n d (ii) arguing at De Anima I I I
2. 425^26-426*26 that i n that sense the visible a n d sight are simulta­
neous by nature, because without sight there is nothing w h i c h is seen
as e.g. pale or dark.
T h e paragraph raises two questions: w h a t is the meaning a n d justi­
fication of {d), a n d h o w does {d) support {a)? I f w e apply the definition
i n {c) to (b)y w e get the proposition that ' a thought is called a thought
relative to that o f w h i c h it is a thought'. I n spite of {d), Aristotle's ob­
jection to this proposition is not that it is false (see ^ i ) but that it " s a y s
the same thing t w i c e " . E v i d e n t l y , then, the contention inexactly con­
veyed by {d) is that its denial is a n unhelpful thing to say, a n d cannot
serve to identify anything as a thought, sight, etc. B u t h o w c a n this
contention support or explain proposition {a) ? I t w o u l d be no less u n ­
helpful to say *a slave is the slave of that of w h i c h he is a s l a v e ' ; yet that
of w h i c h something is a slave, viz. a master, is certainly a master relative
to, a n d of, something. O n e difference between these two cases is as
foUows. lfA is a master a n d B his slave, it is possible to identify A as B*s
master or B as ^ ' s slave (although not at the same time to identify each
by his relation to the other). B u t i f C is a thought a n d D its subject-
matter, it is not possible to identify D as the subject-matter o f C , because
it is necessary to identify C as the thought of Z>. Thoughts, sights, e t c . —
a n d measures, i.e. units of m e a s u r e — a r e thus relative i n a strong sense:
not j u s t that the description of something as a thought invites the
question *of w h a t ? ' but that any description of a thought identifies it
only b y reference to a different entity, its subject-matter. I n this strong
sense thinkables etc. are not relatives; a n d i f they were, w e should be
condemned to the hopelessly circular method of identification by means
of propositions such as the contradictory of (d),
I f Aristotlc means to convey this distinction, his exposition of it is
faulty i n three ways, (i) H e does not distinguish the thesis that thinkables
etc. are not relatives i n the strong sense from the purely grammatical
point, rightly s p u m e d i n the Categories, that thinkables are not caUed
thinkables of anything, (ü) H e holds that of the four kinds of relatives
exemplified by thoughts, masters, slaves, a n d thinkables, only the last are
relatives i n a special w a y ( " f r o m the other thing being relative to t h e m " ) ,
w h e n i n fact it is the first k i n d that are special, being relatives i n a
stronger sense t h a n the other three, (iii) Proposition {d) fails to bring out
the difference between other correlatives a n d those w i t h w h i c h the
p a r a g r a p h deals.

1021^3. Aristotle is no doubt influenced by the argument: knowledge

is a relative; doctoring is a (form of) knowledge; therefore doctoring is

166
A 15 NOTES 1021b3

a relative. But Categories 8. 11*24-32 rejects this conclusion: "for know­

ledge, a genus, is calledjust w h a t it is, ofsomething else (it is called know­
ledge of something); but none of the particular cases is caUed just w h a t
it is, of something else . . . T h u s the particular cases [e.g. grammar,
music] are not among the relatives." Aristotle is i n a dUemma. O n the
one h a n d , he often describes categories as " g e n e r a ofthe things-that-are",
a n d the species of a species must be a species of its genus {Categories 3.
1^10-12). O n the other h a n d , his linguistic tests for membership of the
category of relatives resist this treatment (rendering 'x is i n the category
of relatives' a referentially opaque context, i f 'x* has to range over
things rather than words). See Ackrill's note on Categories 8. 11*20.

io2i**8. Aristotle implies that possessors of doubleness are double i n

their own right (sc. qua being possessors ofdoubleness, not essentially).

CHAPTER 16

T h e three senses of 'teleios' correspond pretty exactly to the E n g l i s h

*entire', *perfect', a n d *complete'.

1021^14. A thing c a n be perfect, a n d therefore good, ofits kind, yet

b a d because its k i n d is b a d : it is a bad thing to be a good thief

1021^23. T h e w o r d for " f u l f i l m e n t " is 'telos% from w h i c h 'teleios*

derives. T h e traditional translation *end' suppresses this connection, a n d
obscures the point that life's end, or death, teUute, is a telos i n a subsidiary
sense only ( " b y transference"), because it is not commonly regarded as
the zenith of life (at Laws 834 c 2 Plato speaks of full-grown horses as
"possessing their telos'*; cf. t h e j o k e at Physics 11 2. 194*30-3). O n the
other h a n d , Aristotle is w r o n g to treat the expression *completely d e ­
stroyed' as a case oftransference: the thing destroyed is *completed' i n a
degenerate sense, but its destruction is completed i n the full sense. A task
is no less completed w h e n its fulfilment is undesirable.

1021^30. I n his recapitulation Aristotle attempts to reduce the three

senses h e has found to two, but his definitions reaUy fit nothing but thc
two subdivisions of sense 2, *complete doctor' a n d *complete thief'.
" T h e rest are made so b y virtue of t h e s e " : Alexander instances a
perfecteducation—one that makes the pupil perfect; a complete b o o k —
one whose contents are complete; a perfect spear—one that only t h c
perfect Achilles c a n wield.

167
 I022*4 METAPHYSICS Δ17

CHAPTER 17

O f the four senses listed by Aristotle only the first a n d third are found i n
his o w n writings. T h e purported sense *substance' is possibly suggested
by Plato's distinction between limit a n d the unlimited (Philebus 23ff.).
T h e " l i m i t o f a c q u a i n t a n c e " is perhaps its origin ( s e e J 1. i o i 3 * i 4 - i 6 ) ,
perhaps its completion (cf. Z 1. 1028*36-7 " w e hold that w e know each
thing most o f a l l w h e n w e are acquainted w i t h what a m a n , or fire, i s " ) .
" I . e . that for w h i c h " : i n the O C T the brackets close before this phrase,
making it explanatory of Tulfilment'.
Aristotle does not attempt to justify his principle that i f a n A is a k i n d
ofB every sense of *Л' is a sense of 'B\ I t is not i n fact true that origins i n
aU of the senses distinguished i n Δ 1 are limits.

CHAPTER 18

1022*14. " B y " translates the preposition 'kata'; for the phUosophical
senses i n the first paragraph Ъу virtue of' or *in respect of' would be more
natural E n g l i s h , a n d those renderings have mostly been adopted else­
where i n the translation. F o r the senses of *cause' see Δ 2.

p. 207 1022*24. *Ii^ its ovm right' is the rendering mostly used i n the trans­
lation, i n preference to the more literal *by itself. T h i s a n d the next
paragraph paraUel Posterior Analytics 1 4. 73^5-10. * " A n i m a l " is part of
CaUias' formula' expresses i n the formal mode what ' a n i m a l (i.e. being
a n animal) is part of CaUias' substance' expresses i n the material mode
(see Δ 24. 1023*35-6): v i z . that CaUias is essentially a n animal. T h e
idiom used i n *26-7, 'F is what it is to be CaUias', conveys that F (i.e.
being F) is the whole of CaUias' essence—i.e. the predicate ' F ' a n d
those w h i c h it entails are the only essential predicates of Callias. E v e n i f
Aristotle is right to cite the name *Callias' itself as such a predicate (*27),
it is surprising that he does not cite ' m a n ' , w h i c h i n his opinion gives a
true, a n d at the same time non-trivial, answer to the question 'what is it
to be C a U i a s ? ' " A n d w h a t it is to be C a U i a s " («»27) m a y m e a n ' a n d
anything else that answers the question " w h a t is it to be C a l l i a s ? " ' ;
otherwise " a n d " must have the force 'i.e.'.
T h e senses of ' i n its own right' i n this paragraph a n d the next evidently
correspond to the first sense of 'by' i n 1022*14. Callias is good by (virtue
of) good itself, viz. the affection goodness; o n the other h a n d he is a m a n
a n d a n a n i m a l by himself, a n d so (it is implied) not by virtue of the
affecüons manhood a n d animality. I n this implied denial Aristotle m a y
be influenced by the foUowing considerations. Because *good' is a n

168
J18 NOTES 1022^24

adjective, G r e e k usage permits its employment not only to describe

possessors of the affection or property goodness but also as a name of
goodness itseH"(Z 6. 1031^22-8). T h i s being so, it is easy to suppose that
the relation between goodness o n the one h a n d a n d good m e n , good
apples, good arguments, etc., o n the other h a n d is a relation between two
types of good thing, the first type good Ъу itsetf', the second type good
by having the first somehow i n it ( c f Categories 5. 2*29-34; i n a similar
w a y he wiU argue i n 1022*29-32 that a m a n is pale by h a v i n g a pale
surface somehow i n h i m ) . T h u s i n place ofthe analysis familiar to us, *an
apple is good by being something that has good(ness)', Aristotle's analysis
is *an apple is good by having something that is good (in the primary
sense)'. T h i s analysis is not available i n the case ofnouns like ' m a n ' a n d
'apple', w h i c h do not ever function as names ofproperties a n d so cannot
be thought of as applying to concrete individuals i n a merely secondary
sense. Socrates is a m a n not by having something i n h i m that is a m a n ,
but simply by being a m a n — " b y himseU"". B u t (i) this argument does
not prove it false to say that Gallias is a m a n b y (virtue of) humanity,
since it is wrong to suppose that that means ' b y virtue of humanity's
being a m a n ' ( c f L o c k e , Essay I I . 23. 3); a n d (ii) the argument does not
lead to the conclusion that Aristotle wants, for it would make Gallias
e.g. a bachelor ' b y himseU'' (since 'bachelor' is a noun) although he is not
a bachelor essentially: noun-predicates do not coincide w i t h essential
predicates. T h e r e seems, therefore, to be a n error embodied i n Aristotle's
choice of the expression ' b y itseU"' or ' i n its o w n right' as a m a r k of
essential predication.

1022*27. See preceding note.

1022*29. T h i s paragraph parallels Posterior Analytics I 4. 73*37-^3.

T h e p r i m a r y pale thing is 'the pale itself, i.e. the affection pallor (cf.
Z 6. 1031b22^). B u t among the " r e c i p i e n t s " of this affection, i.e. what
we should call pale things, some are again prior to others; for instance a
m a n is pale ' b y virtue o f his skin's being pale (1022*16-17). T h i s second
contrast leads Aristotle to say that the man's skin, a n d i n general surfaces,
are i n a n e w sense pale ' b y virtue of themselves'. T h e priority here
invoked is conceptual: it is not possible to understand what it is for a m a n
to be pale without understanding what it is for a surface to be pale, but
the converse is possible. T h i s sense of ' i n its own right' is opposed to the
sense of 'coincidentaUy' used e.g. at Categories 6. 5*38-^4.
Aristotle's contention that a m a n is alive i n his o w n right because a
part o f h i m , his soul, is the p r i m a r y recipient oflife should be contrasted
w i t h Physics V 1. 224*21-34 a n d V I I I 4. 254^7-12, where he says that a
thing wiU not change or be changed i n its own right i f i t does so by having
a portion of itself do so.

169
I022*33 METAPHYSICS J18

1022*32. T h e reference to the multiplicity of *causes' of a m a n seems

at once to disqualify that example from displaying the sense of *in its o w n
right' that the p a r a g r a p h defines. Y e t it must be intended to display i t ;
so the meaning must be that although m a n y cavises (or reasons) can be
given to explain w h y something is a m a n , no reason need be given to explain
w h y a man is a m a n . *Why should the tallest a n i m a l o n this farm be a
m a n ? ' makes good sense (answer: *it's a pig f a r m ' ) , but not *why should
a m a n be a m a n ? ' nor, less trivially, *why should a poet be a m a n ? ' T h e
latter are, i n a w a y , self-explanatory. T h i s sense of *in its o w n right' is
opposed to the sense of*coincidentally' discussed a t J 30. 1025*21-30. I t
reappears atPosteriorAnaljytics I 4. 73^10-16.

1022*35. T h e text a n d punctuation ofthis final sentence are altogether

dubious. T h e translation adopted foUows Ross's reading, though without
any strong conviction that it c a n bear the meaning that h e gives i t , v i z .
*an F is i n its o w n right w h a t holds good of F s alone, a n d of aU F s ' . I f
this is right, the sense of *in its o w n right' is that ^ e d a t J 30. 1025*30-4:
a triangle is i n its o w n right a figure whose angles a d d u p to two right-
angles, because aU a n d only triangles are such ( c f Posterior Arw,lytics I
4. 73^25-74*3).

C H A P T E R 19

I t is doubtful whether Aristotle's w o r d , a n y more t h a n its E n g l i s h

equivalent *disposition', necessarily connoted a n arrangement of parts.
I n Categories 8 Aristotle himself uses it (Ackrill *condition') to cover a n y
qualities that " a r e easily changed or quickly c h a n g i n g " , a n d although
the examples given there, hotness a n d chUl a n d sickness a n d health,
w o u l d have been thought b y h i m to involve arrangements of the parts of
the bodies that have them, that fact is not mentioned i n the definition.
A n arrangement b y capacity is perhaps a n ordering of (non-spatial)
parts w i t h respect to strength or importance, as i n the dispositions of
democracy a n d daydreaming. Ross suggests that a n arrangement b y
form is the classification of forms or species under a genus, i.e. a tree of
P o r p h y r y ; i f so, the genus w o u l d be the whole whose disposition c o n ­
sisted o f t h e arrangement ofits species as parts (cf. Δ 25. .1023^17).

C H A P T E R 20

'Hexis*, the v e r b a l n o u n from 'echein% to have or possess, has three m a i n

senses, ( i ) possession, (2) state, (3) wearing. I n this chapter Aristotle
ignores sense ( i ) , i n w h i c h 'hexis' is opposed to *lack' or *privation' (see
Categories 10. 12*26-13*17). Sense (2) arises from the c o m m o n G r e e k

170
A20 NOTES 1022b4

construction of the cognate verb *echein' with a n adverb, whereby for

instance *I have welF means *I a m i n a good state' (cf. Nicomachean
Ethics I I 5. 1105^25-6). I n his second paragraph Aristotle explains
sense (2) by means of the w o r d *disposition', although i n Categories
8. 8^26-9*13 states a n d dispositions are distinguished as different kinds of
quality, a n d the former include knowledge, w h i c h is not a matter of
being disposed well or iU.
Aristotle m a y have been prompted to give prominence to the special
sense (3) i n w h i c h 'hexis' means *wearing' by a passage i n Plato's Theae­
tetus. Both Plato a n d Aristotle make play with the distinction between
possessing a n d using (Plato, Euthydemus 280 b d , Aristotle, Nicomachean
Ethics I 8. 1098^31-1099^7) a n d between being i n a certain state, e.g.
being virtuous, a n d functioning i n accordance with that state, e.g. doing
something virtuous {NicomacheanEthics X 6. 1176^33-^2). C o m m o n l y the
word 'hexis' is used for the first member of this p a i r ; but at Theaetetus
197 b c the hexis of knowledge is compared w i t h wearing a coat*and
having hold ofbirds i n a n aviary, a n d contrasted with the mere possession
{ktesis) of these things. I n his first paragraph Aristotle draws attention to
this deviant sense, i n w h i c h 'hexis' goes o n the same side ofthe distinction
as *functioning'.
W h e n he says that " i t is not possible to possess this k i n d ofpossession",
he implies not that other *functionings' c a n be possessed, but that other
kinds of hexis can be. T h u s one c a n speak of having or possessing a state
of health, but not of possessing the wearing of clothes or the exercise of
sight (or indeed the possession o f s i g h t ) . B u t the reason Aristotle gives
for this restriction is inadequate, for the possibility of possessing a wearing
would generate a regress only i f that possession itself h a d the sense
' w e a r i n g ' ; yet w e do not wear a state of health. T h e truth is that his
regress argument explains only the impossibility of possessing a genuine
possession, a case w h i c h he does not mention.

CHAPTER 21

T h i s brief chapter does not give a satisfactory survey of the senses i n

w h i c h Aristotle actually uses the w o r d *affection' {'pathos'), the chief of
w h i c h are: ( i ) state or condition, (2) property, (3) coincident, i.e. n o n ­
essential property, (4) quality, (5) feeling, (6) happening, (7) misfortune.
*Modification' (traditionally *alteration') is defined by Aristotle as
" c h a n g e i n respect of q u a l i t y " {PhysicsW2. 226*26-7) a n d as " a l t e r a t i o n
i n respect of affection" {A 2. 1069^9-12). Qualities " i n respect of w h i c h
it is possible to be modified" c a n be contrasted, therefore, only w i t h
qualities i n respect of w h i c h no change is possible at aU. T h e s e are
differentiae, i.e. essential qualities (cf. Topics I V 5. 128*26-8, Physics
V 2. 226^27-9). I n the discussion of quality i n Categories 8 affections

i71
1022^15 METAPHYSIGS J21

are treated as a species of quality (9*28-9), but later distinguished from

qualities (10*9-10).
I t is characteristic of Aristotle's slipshod semantics that he uses three
different types of expression, 4 h e pale', *sweet', *heaviness', as names of
qualities. *The pale' c a n serve to name the quality pallor, but also means
*that w h i c h has p a l l o r ' ; *pale' a n d *sweet' o n their o w n are not names at
all.
F o r this chapter's second sense, " a c t u a l modifications", Aristotle
sometimes prefers the term 'pathesis' {Physks I I I 3. 202*22-4).
\

GHAPTER 22

'Steresis\ *lack' (traditionally *privation'), derives from a verb whose m a i n

passive sense is *be deprived of', *lose'; but the sense *loss' (1022^31, c f
Δ 5. 1015*24) is m u c h less common t h a n 'lack' i n Aristotle. Possession
a n d lack are also discussed at Categories 10. 12*26-13*17; see also Θ 1.
1046a31-5.
1022^22. Г 2. 1004*10-16 describes a lack, b y contrast w i t h a denial,
as implying " a certain nature [or characteristic] . . . as the subject of
w h i c h the lack is s t a t e d " : i.e. for aU F there is some nature G w h i c h
everything lacking F must possess. T h u s what lacks sight must possess
(presumably) eyes; so that the Г 2 sense differs from that of the present
paragraph, i n w h i c h plants are said to lack sight. Indeed, i f we c a n
assume that every possession is characteristic of something or other, the
sense given i n the present paragraph ignores the distinction between
lack a n d denial altogether. Unfortunately it is not clear whether or not
that assumption m a y be made, since Aristotle does not indicate the
criteria for being " c h a r a c t e r i s t i c " . W h e n he says i n the next paragraph,
for instance, that sight is characteristic of the genus within w h i c h moles
are a species, does he m e a n that most members ofthe genus see or that aU
members of the genus have the apparatus for seeing, even if not i n work­
ing order?
1022^27. " A n d equaUy too i f . . . " : does Aristotle m e a n *only i f ' (e.g.
that a m a n without uncharacteristic visual powers—eyes i n the back ofhis
head, the ability to see through fog or to discern the stars i n daytime—is
not said to lack sight), or *if' (e.g. that a m a n w i t h rwrie but such u n ­
characteristic powers is satid to lack sight) ?
1022^32. According to PriorAnalytics I 46. 51^25-8 predicates such as
*is invisible', *is footless', are not equivalent to 'is not visible', 'is not
footed', since the former " h a v e a certain s u b j e c t " , i.e. imply some
positive property. I n the language of the present chapter, w h a t is i n ­
visible must be characteristically visible, a n d sounds, for instance, are
neither visible nor invisible. Gf. / 4. 1055^8-11.

172
Δ 22 NOTES io23*a

1023*2. C f . J 12. 1019"23-6.

i 0 2 3 * 4 · F o r intermediate states see Categories 10. ιι^38-ΐ2»25 a n d

12^26-13*17.

CHAPTER 23

T h i s is the ordinary word for *have' {'echein'); it is mostly rendered b y

*possess' i n this translation. Aristotle's four senses answer to *control',
*harbour', 'contain', a n d *restrain'. H i s words for *wear' {'ampechesthaV),
*contain' {'penecheirC), a n d 'fuse' {'sunechein') are aU compounds of
'echeirC, T h i n g s fused are continuous {suneches, see Δ 6. 1015^36 a n d
Physics V 3. 227*11-12). F o r the treatment ofthe Atlas m y t h by " a u t h o r i ­
ties o n n a t u r e " see De Caelo I I 1. 284*20-6.
W e must understand Aristotle to m e a n that being in is the converse
relation to possessing: the city is i n ( ? t h e power of) the tyrant, the
disease i n the body, the liquid i n the bucket, a n d the overlying weights
i n (Pbearing on) the columns. T h e first a n d last of these senses seem
highly strained.
T h e r e is a n independent discussion of the senses of 'possessing' i n
Categories 15.

CHAPTER 24

1023*26. " E v e r y t h i n g meltable is out o f w a t e r " : a t J 4. 1015*10 the same

proposition is stated tentatively, w h i c h encourages the opinion that what
Aristotle means b y it is that meltables are composed entirely of water.
Meteorologica F V , possibly not by Aristotle, holds that true of bronze ( I V
10. 389*8), but states it of the whole cl2iss of meltables only w i t h q u a l i ­
fications ( I V 8. 385*25-33) a n d exceptions; e.g. iron, w h i c h is meltable
( I V 6. 383^27-30), is compounded of water a n d earth (383*32). I t is
impossible to be sure, then, whether Aristotle is saying that the "first
genus" of a thing's matter is any element w h i c h the matter contains, or
only that element (ifany) o f w h i c h the matter is solely composed—so that
a n iron statue is not " o u t o f " water, although it contains water. I f the
latter is right, Aristotle m a y i n t e n d — b u t this is also u n c l e a r — a similar
restriction o n the " l a s t f o r m " or *infima species' o f a thing's matter, to the
effect that a partly iron statue is not " o u t o f " iron, although it contains
iron.

1023*31. A ball m a y be (A) m a d e out o f b r o n z e ; i n another sense, not

separately listed, it is (B) out of bronze plus sphericity (its matter plus
shape). Differently again, the top haU*ofthe ball is ( C ) a part out o f w h a t

173
I023*3I METAPHYSIGS J24

is itseH* (В) out of certain matter a n d shape. H a v i n g slipped from talking

of matter to talking of matter-plus-shape, Aristotie feels it necessary to
justify his implied c l a i m that only the latter is a " w h o l e " : the shape
brings " f u l f i l m e n t " to the matter.

1023*35. A s the w o r d *man' is equivalent to a n expression i n w h i c h

*two-footed' is a part, so, according to Aristotle's primitive theory of
meaning, the form (? concept) man has the concepts two-footed a n d animal
as parts; a n d two-footed a n d animal are the non-perceptible matter of the
form man (cf. Δ 25. 1023^22-4). V e r b a l " e l e m e n t s " , i.e. sounded letters,
are both the perceptible a n d the non-perceptible matter of syllables: the
elements themselves are constituents of the syUables, a n d the word
*element' is a constituent ofthe definition of*syUable' ( Z 10. 1034^25-6).

1023^3. W h y does Aristotle say that plants are made out of " a certain
p a r t " of earth? H e cannot m e a n that plants do not, either severally or
coUectively, ш е u p aU the earth there is, for that is true of, say, the
relation ofstatues to bronze; nor that plants contain i n them some but
not aU of the elements of earth, for h e believes that earth is itself an
element.

1023^5. Night is " o u t o f " d a y not only i n the sense of succeeding it

("successive i n respect o f t i m e " ) but i n the stronger sense o f r e p l a c i n g it
( " h a v i n g alteration into one a n o t h e r " ) .

GHAPTER 25

Although Aristotie undertakes to explain the word 'meros\ *part', he

substitutes i n two places a diminutive form, translated *portion'.

1023^12. " W h a t is subtracted from a quantity qm q u a n t i t y " is any­

thing whose subtraction involves a dintinution ofsize; Alexander remarks
that qualities like sweetness a n d heat are not parts of things though
subtractable from things.

1023^15. T h e parts " w h i c h give the measure of a t h i n g " arie those

w h i c h divide it without remainder, its aliquot parts or factors.

1023^17. Aristotle's treatment of genera a n d forms as both n o n -

quantitative a n d related as whole to part is indicative of his vagueness
about their status. Being negro is not a part of being human, a n d humanity
has parts only because the w o r d is taken to denote the h u m a n race,
w h i c h is a quantitative set or class.

1023^19. T h e bronze cube a n d the bronze baU are examples of

"something that possesses a f o r m " . Being quantities, they wiU also have
J25 NOTES 1023b19

parts i n the first sense; but the cube's bronze nMterial a n d angular shape
are 'parts' i n a new sense, because thcy cannot be "subtracted from a
quantity gua quantity"—^without them i t would be, perhaps, a copper
cube or a bronzc p y r a c u d , not a smaller bronze cube.

1033*33. T h e division o f a form into parts, mentioned at 10a3*20, is

not illustrated iuitil *22: i f 'is a G ' is part of the definition of 'is a n F',
being a G is part of the form being an F (cf. J 24. i023"35-^2).

CHAPTER 26

1033*26. I n the first clause (^26-7) Aristotlc explains what morc is

required of a whole than containing parts, viz. that i t contain aU the
parts of that whole. H e might have added that anything containing
parts could be caUed a whole something (but not always " n a t u r a U y " ? )
e.g. a period containing nine months is a whole nine-month though nof
a whole year.
I n the rest of the paragraph from ^27 h c distinguishes two relations
that m a y hold between a whole a n d its parts. T h c first (b29-32) is thf
relation of a class ("vmiversal") to its members. Nothing i n Δ 25 corre­
sponds to this section (1023^17-19 concerned class-inclusion), i n which
Aristotle avoids thc w o r d 'part' i n favour of the vaguer 'contents'. T h e
word for 'whole' {'holos') b cognatc with his technical t c r m 'uiuvcrsal
{'katholou'). A universal whole is distinguished b y being predicated ofits
contents severally, as the genus animal is predicated of m e n , horses, and
gods: they are " a U — e a c h o n e — o n c t h i n g " , i.e. o n c a n d t h c same
thing (the words " e a c h o n e " mark off this relationship from that
of, say, a shoe to its parts, w h i c h are collectively b u t not severally a
shoe).
After b32 Aristotle turns to the relation more naturaUy described as
that o f w h o l e to part, a n d treated briefiy under the first scmc o f ' p a r t ' i n
Δ 25. H i s opening remark impUes that he takes this new relation to be
confined to wholes that are "continuous a n d linüted". H e m a y think
that such discontinuous wholes as t h c (unlimited) genus animal o r a
(limited) barrrful of animals have been disposed of i n thc preceding lines;
but that is not so, since those lines exanuned a particular relation o f s u c h
groups a n d classes, v i z . to their members, a n d they arc related to their
sub-groups i n a different w a y (it is not true that each kind o f a n i m a l is a n
a n i m a l ) . Aristotle also ignores such discontinuous wholes as the parts of
a n unassembled toy ('does that box contain the whole aeroplane?').
T h e s e cannot be intended b y thc "potential constituents" of b34, w h i c h
he teUs tM are (i) the constituents of something continuous a n d (ii) t h c
prime case o f a continuous a n d limited whole. O n the contrary, "potential

175
юзз^зб METAPHYSICS Δ26
constituents" seem to be parts w h i c h could be b u t never have been
separated: e.g. a n island is less of a whole i f a n a r m of the sea used to
divide i t into two. T h a t would explain w h y he thinks artefacts less good
examples of wholes than most natural objects (^34-6, cf. Δ 6. i o i 6 ^ i i ) .
I t remains puzzling that h e confines his attention, i n this section, to
quantities that are continuous a n d limited.

1024*1. T h i s paragraph tacitiy drops the requirement of continuity,

but seems to imply still that wholes must be limited (if Alexander is
right i n so interpreting "possessing a beginning, a middle, a n d a n e n d " ) .
A t any rate we are not told under w h a t conditions, i f a n y , quantities not
of that k i n d are caUed wholes. Aristotle proceeds to distinguish two types
of limited quantity; type 1 are those i n w h i c h position (sc. of the parts)
makes no diflference, type 2 those i n w h i c h i t makes a diflference. T h e
cliunsy expression " a n y that admit b o t h " i n effect subdivides the latter
into type 2Л quantities, i n w h i c h position makes a difference to the shape
but not the nature, a n d type 2b quantities, i n w h i c h it makes a difference
to the nature. T y p e s 2a a n d 2b are *wholes', types 1 a n d 2a *alls'. T h e r e is
a distinction d r a w n i n the passage between the singular a n d plural of the
Greek w o r d for * a l l ' ; for the sake of clarity i n the translation, Aristotle's
singular has been rendered by the singular *all' (as i n *all the w i n e ' ) , his
plural by 'every' (as i n *every house', i.e. *all (the) houses'); but i n *3 the
Greek plural means *instances of (the singular) a l l ' , hence " a l l s " . I n
E n g l i s h we m a y use the singular *all' without restriction (*all the wine',
*all the house'), but *whole' only of coutitables (not *the whole w i n e ' ) .
Aristotle's usage i s , rightly or wrongly, different: h e disallows e.g.
*aU the house' under 2b a n d allows 4he whole w a x ' under 2a, a n d
the distinctions that he bases o n these words have nothing to do with
countability.
W h e n he says that position makes a difference to the shape of w a x he
presumably means that the shape of a piece of w a x c a n be changed by
rearrangement of its parts. T y p e 1 quantities a r e therefore those i n
w h i c h this is not possible, i.e. fluids. B u t Aristotle's example, water, is a n
Mwlimited quantity, a n d should be replaced b y ponds, cupfuls of water,
etc. T y p e 2a quantities are homogeneous non-fluids, b u t (i) it is a
mystery how Aristotle imagines that his second example, a cloak, is
something whose nature wiU survive transposition a n d (ii) w e might
expect, foUowing Δ 6. 1016^11-17, that nothing whosenaturesurvives
transposition would be a whole. N0 examples are given of the third type,
2b; apart from ordinary objects like shoes a n d houses we m a y instance,
w i t h Ross, the (discontinuous) musical scale mentioned i n the next
chapter(1024*21).

1024*8. T h e purpose of the final sentence is perhaps to sanction the

use of *every' ('all' plural) w i t h words before w h i c h *all' (singular) has

176
Δ26 NOTES 1024^8

been disallowed. I t is unclear whether every case of a singular *all' c a n ,

like *all this number', be paraphrased i n terms of 'every'.

CHAPTER 27

T h i s word, 'kolobos% is h a r d l y used by Aristotle outside the zoological

w o r b , a n d the reason for its inclusion here is a mystery. T h e chapter,
alone i n Δ, does not distinguish more than one sense; but it is a n a d ­
mirable example of a definition b y division. Perhaps it started life as a n
exercise, set by or to Aristotle, w h i c h was commended a n d preserved as
a model answer.
A kolobos thing is truncated, stunted, lopped, docked, or otherwise
abbreviated.
" I t must stiU be a c u p " (*15): or perhaps, as the argument requires,
*the [sc. same] c u p must stiU be'.

GHAPTER 28

1024*29. Aristotle's a w k w a r d definition of the everyday sense i n w h i c h

'genos' means *fanuly' or *clan' is intended to convey that a family is a n
unbroken succession of offspring a n d that e.g. *Hellen's family stiU
exists' means *the succession of offspring is unbroken from Hellen's time
to the present d a y ' . " T h i n g s possessing the same f o r m " : thus H o m e r a n d
the I l i a d do not m a k e a fanüly.

1024*31. " B r o u g h t them into existence": the verb here translated

*bring', 'kinein', is elsewhere *effect change', as at 1024^8 w h i c h summarizes
this paragraph. F o r the opinion that mothers contribute matter a n d not
form to their offspring see De Generatione Animalium I I 4. 740^24-5 a n d
other references i n Ross.

1024*36. T h i s a n d the next paragraph are sunrniarized together at

1024^8-9, w h i c h defines i n a third w a y the same sense as they define i n a
first a n d a second.

1024^6. T h e " s u b j e c t " of a differentia (e.g. two-footed) is w h a t it

divides (animal) rather t h a n w h a t it constitutes (man) or is otherwise
predicated of (Socrates). Aristotle seems to picture it as the dough from
w h i c h a cutter cuts shapes, a picture w h i c h fits both the relation ofgenus
to differentia a n d form a n d the relation of matter to form a n d concrete
individual. T h i s sinularity m a y be sufficient to justify use of the w o r d
*matter' to cover genus, but genera (such as animal) r e m a i n quite a

8720270 177 N
I024^6 METAPHYSIGS J28

different k i n d o f m a t t e r from stuffs (such as flesh); cf. Z 12. 1038*6-8,

/ 8 . 1058*23-4.

1024^9. " F i r s t s u b j e c t s " ; both words are difficult, ( i ) I f they mean

*highest genera', i.e. figures of predication or categories, the proviso
about analysis is redundant (see 1024^15-16). (2) I f they m e a n 'last
forms' ('infimae species'), the proviso about analysis cannot be satisfied.
(3) Alexander plausibly construes the proviso as applying not to forms or
genera b u t to kinds of matter, such as gold a n d silver, w h i c h have a
c o m m o n 'analysis' because both are composed of water {Meteorologica I V
10. 389*7); hence *last matter'. So Ross, a n d compare the first ofthe two
senses of 'first matter' at Δ 4. 1015*7-10, a n d 'last matter' at Δ 6. ioi6*
19-20, 1017*5, Δ 24. 1023*27-8. (4) Butsense(3) w i l l n o t f i t t h e f i r s t s u b -
jects qf form a n d matter; i n those cases, as Alexander says, the meaning
must be " t h e parts of their defining f o r m u l a e " .

1024^12. G f J 7. 1017*22-31.

GHAPTER 29

Aristotle commonly prefers the noun 'falsehood' to the neuter singular

adjective 'fake', although some of the things discussed i n this c h a p t e r —
e.g. d r e a m s — w o u l d not naturaUy be caUed falsehoods b y us. H i s
classification is surprising. W e should expect the m a i n distinction to be
between false objects or events—i.e. things that are not as they seem, like
dreams, stage scenery, a n d false b e a r d s — a n d false statements a n d
beliefs. Instead, h e ignores statements a n d beliefs i n favour of states of
affairs (a diagonal's being commensurable w i t h its side, your sitting
down), w h i c h he groups, together with things that are not as they seem,
as false " a c t u a l t h i n g s " . I n the second paragraph he deals not primarily
w i t h false statements (like ' a circle is a plane figure contained b y three
straight lines') but w i t h descriptions (like ' a plane figure contained by
three straight lines') w h i c h are false qf one thing, true of another. T h e
examination o f t r u t h a n d falsity i n Θ 10 likewise promises a treatment of
" a c t u a l t h i n g s " , b u t i n fact discusses beliefs a n d statements, w i t h
particular attention to a special problem not raised i n the present chapter.
See also Έ14.

1024^17. T h e treatment of false states of a f f a i r s i s not adequate.

Aristotle says that your sitting d o w n is false w h e n it is " n o t c o m p o u n d e d " ,
i.c. w h e n y o u a n d sitting down do not combine. I f t h e y do not, there is no
such state of affairs: a false state of affairs is one that does not exist ( " i n
this w a y these things are not things-that-are"). T h i s ignores the real
problem, for false statements are not non-existent. W e might say that a

178
J29 NOTES 1024b17

false statement describes what is actually non-existent, or not combined,

as existent or combined (cf. Θ 10. 1051^3-5), so reveaHng the crucial
fact, carefully expounded by Plato i n the Sophist (260 c-264 b ) , that every
statement must describe something as something, say something about
something. B u t if so, talk of things combining into states of affairs does
not help to explain falsehood. *You are sitting down', if false, is false
not because it describes as compounded what is not compounded or as
existent what is not existent, but because it describes as sitting dbwn what
is not sitting down. I f the latter explanation is not already intelligible,
the former will not make it so.

1024^26. *The formula {logos) of so-and-so' is commonly used by

Aristotle to m e a n *the logos that says what it is to be so-and-so*. A n
individual such as Socrates cannot have more than one logos of this k i n d ,
i n the sense that true answers to the question *what is it to be (that)
Socrates?' cannot have more than one meaning. O n the other h a n d ,
more than one non-synonymous description c a n be truly given ofthe same
Socrates. I n this second " w a y " the hgos *artistic' wiU be true of Socrates
provided that "Socrates a n d artistic Socrates are . . . the s a m e " , i.e.
Socrates is artistic. W h y does Aristotie say that false logoi are ''qua false,
of things that are n o t " a n d " t a k e n baldly, the logos o f n o t h i n g " ? A logos
m a y be a complete statement, true or false, or a predicative description,
true or false of something. Aristotle insists on both taking some logoi to be
false, not false of so-and-so (hence complete statements), a n d taking aU
logoi to be logoi ofsomething (as descriptions are). H e n c e false statements
must describe false states of affairs w h i c h , as we have seen, are "things
that are n o t " .
T h e objection to Antisthenes is a bit muddled. Aristotle attributes to
h i m the restrictive theory that a description of Socrates must describe
what it is to be Socrates ( c f Sophist 251 a c ) , a n d claims that this " r e s u l t s "
i n denial ofthe possibility of contradiction. B u t the result does not foUow
unless descriptions of what it is to be Socrates have to be correct; and i f
they have to be correct, it foUows without the restriction, since even i f
there were m a n y descriptions of Socrates they would not contradict one
another so long as none of them were false. Ajistotle sees, of course, that
Antisthenes' mistake h a d something to do w i t h confusion about false­
hood («32 " t h a t is w h y " ) . Antisthenes must have denied the possibUity
offalsehood, a n d his argument must have been: since false descriptions
describe "things that are n o t " , things that are, like Socrates, cannot b c
falsely described. T h e argument's faUacy depends on the ambiguity of
*describe things that are not': *Socrates is handsome' describes a state of
affairs that is not, but also describes Socrates, who is. B u t Aristotic does
not comment on this ambiguity. Instead, he merely asserts that a p a r ­
ticular thing m a y be described {Ugesthai) by the logos of (what it is to be)

179
 I024^26 METAPHYSIGS J29

something else. I f t o be two (e.g. miles) is to be double (one mile), to be

eight or nine miles is not to be double one m i l e ; but we m a y describe
eight miles as double (truly, for it is double four miles), a n d we m a y
describe nine miles as double (falsely, for it is not double any whole
number of miles). I t is probable that "another's f o r m u l a " means 4he
expression that says (truly) what it is to be something else'. I f so, Aristotle
leaves no room for falsehoods like *Socrates speaks' or *Theaetetus flies',
for i n his view *speaks' a n d *flies' do not express what it is to be anything
(speech does not speak at a l l , a n d speakers are not identified by the
activity).

1025*1. Aristotle's false m a n is not the m a n prone to error but the

deceiver or impostor. I n Plato's Hippias Minor Socrates argues that the
" w i l y " Odysseus must be capable oftruth as well as falsity, a n d so true as
well as false (369 b 3 - 7 ) . W i t h AristoUe's reply Ross compares Nicomachean
Ethics I V 7. 1127^14: " i t is not the capacity that makes the boaster,
but the c h o i c e . " H i p p i a s doubtless h a d the same point i n m i n d when he
objected against Socrates (370 e 8 - 9 ) that Odysseus was false willingly
or intentionally (the Greek 'hekon* covers both). A t «8 Aristotle digresses
to deal with Socrates' response to this objection. H i s cavil at the example
oflimping implies that Socrates is entitled to conclude only that pretended
failures are better than involuntary failures. T h i s is unfair, for Socrates'
*induction' used a number of examples not involving pretence; i n
general Socrates argued that excellence i n a n activity is compatible with
intentional, but not w i t h unintentional, miscarriages.

GHAPTER 30

pp. 219-21 O n *coincidence' see also Г 1 , F 4 . 1007*20-^18, a n d ^ ^ 2 .

1025*14. A t Topics I 5. 102^4-7 Aristotle defines a coincidence as

" w h a t is neither a definition nor a peculiarity nor a genus, a n d yet holds
good of the actual thing; a n d what admits of holding good of any one
a n d the same thing a n d ofnot holding good o f i t " . T h e words w i t h w h i c h
*coincidence' is contrasted i n the first part of this definition apply
primarily to kinds of thing rather than to individuals; e.g. being recipient
ofgrammatical knowledge (102*18-20) is a peculiarity of men, not o f a n y
one m a n . Similarly i n the present chapter, the finding of treasure is a
coincidence for a trench-digger because trench-diggers as a class do not
usually find treasure when digging; it does not matter if the p a r d c u l a r
trench-digger i n question has always or usually been successful. T o that
extent the expression 'is coincidentaUy . . . ' works like 'is i n most cases . . . '
or 'is on average . . .'. B u t whereas the predicate 'has on average 2 3 4
children' cannot intelligibly be affirmed or denied of a parücular parent,

180
J30 NOTES 1025*14

Aristotle implies that *coincidentally found treasure' c a n be intelligibly

predicated of a particular trench-digger, not only of the class. T h e price
of this extended usage is to make the context *. . . is coincidentaUy F'
referentially opaque, i n the sense that it m a y be a coincidence that some
trench-digger found treasure but not that some archaeologist d i d , even
though the trench-digger a n d the archaeologist are one a n d the same
person. I n Aristotle's own language, it was a coincidence for the trench-
digger 'qm trench-digger' or 'qm h i m s e l f that he found treasure, but not
qua archaeologist'. ( T h i s is not exactly Aristotle's w a y of putting the
point: instead of 'x is coincidentaUy F qua x' he would say simply 'x is
coincidentaUy F\ a n d instead of 'y is non-coincidentally F qm y' he
would say simply 'y is Fquay, or qua h i m s e l f ; cf. * 2 8 - 9 . )
I n this chapter coincidence is opposed to necessity ( c f Topics 1
5. 102^6 " w h a t admits o f h o l d i n g good . . . a n d o f n o t holding g o o d " ) ,
but i n E 2 the contrast is also w i t h what happens always: see notes on
E 2. 1 0 2 6 ^ 2 4 . Another inconsistency calls for comment here: i n the
Topics passage it is implied that anything that holds good non-necessarily
ofits subject is a coincidence, but here a n d 1nE2 coincidence is opposed
also to w h a t holds good for the most part (or as a rule, or usually). T h e
reason for the extra restriction seems to be that Aristotle thought of a
coincidence as something standing i n need of explanation. T h u s 4he
doctor healed Coriscus' is self-explanatory, whereas *the housebuilder
healed Coriscus' needs extra information to explain w h y it is true, e.g.
that the housebuilder i n question is also a doctor {E 2. 1 0 2 6 ^ 3 7 - 1 0 2 7 * 2 ;
this example clarifies the cryptic fourth sentence—that he healed holds
good of the housebuilder, but not because the subject, time, or place are
as specified). Aristotle was surely right i n holding, by the time he came to
write this chapter a n d E 2 , that occurrences c a n often be explained by
reference to usual, though non-constant, conjunctions: doctors do not
always heal their patients, yet a man's being a doctor explains his
success w i t h Coriscus' Ulness. B u t he was wrong i n holding that occur­
rences c a n always be explained by, a n d never without, such usual or
constant conjunctions—which are certainly not sufiicient (e.g. sufferers
from muscular dystrophy usually die before thirty, but this victim died at
2 2 through faUing out of a n aeroplane) a n d probably not necessary (e.g.
victims of rudeness do not usually commit murder, but the reason w h y
this m a n murdered was that he h a d been rudely addressed). Aristotle
m a y have confused explanation or giving the *ratio essendi' w i t h
confirmation or giving the *ratio cognoscendi'.It is possibly the case that,
for every x, the truth of 'Fx' is some reason for thinking that Gx if a n d only i f
F s are at least usually G; but not the case that, for every x, the truth of
'Fx' is the reason why Gx either i f o r only i f F s are at least usually G .
W h e n Aristotle says ( * 2 4 - 5 ) that " t h e r e is no definite cause of the
coincidental, but a chance one, a n d that is indefinite" he does not m e a n

i8i
1025*14 METAPHYSICS J30

that it is impossible to give a reason for a coincidental occurrence, but

only that i n the statement ofthe coincidence the reason is left " i n d e f i n i t e " ,
i.e. not specified; c f E 2. 1027*5-8. T h e example ofthe sailor who visited
Aegina is somewhat puzzling. Visits to Aegina m a y be divided between
the intentional a n d the unintentional; but that division does not corre­
spond to one between the usual a n d the unusual, or between the self-
explanatory a n d the non-self-explanatory. Perhaps Aristotle means that
among landfalls i n general the intentional are more usual than the u n ­
intentional, but even that might be false of landfalls o n some barren
island. N o r would a n intentional visit explain itself, although the fact
that it was intentional might be held to explain it to the extent ofmeeting
the question *why did he land there?' w i t h the partial answer *because he
wanted, or meant, to'. Aristotle's point i n the last sentence is that this
unintentional landfall is usual, or self-explanatory, only under another
description {*'qua another t h i n g " ) , viz. Ъе was prevented by a storm from
reaching his destination'.

1025*30. I n what sense does the possession of two right angles (i.e. of
angles having that sum) hold good of a triangle " i n its o w n r i g h t " ? O f
the senses of that expression listed i n Δ 18 the first two (1022*24, 27) are
ruled out by the qualification " w i t h o u t being i n its substance". T h e
third (1022*29, *as p r i m a r y recipient') does not always demand a n e w
sense of 'coincidence', for m a n y p r i m a r y recipients of Fness are coinci­
dentaUy F i n the first sense—e.g. surfaces are coincidentaUy pale because
neither always nor usually pale. T h e fourth (1022*32, *self-evidently')
fits well, since what is seU*-explanatory cannot be coincidental i n A r i s ­
totle's first sense of *coincidental' a n d would have to be accommodated
under a new sense. A n d the fifth (1022*35, see note), i f i n d e e d it c a n be
extracted from the corrupt text of Δ 18, is tailor-made for Aristotle's
example here.
I n any case the new sense of*coincidental' w h i c h the example demands
is *non-essential'. Although this new sense applies, as Aristotle's example
shows, i n some cases where the first sense does not, it also applies i n every
case where the first does. I t is odd that Aristotle does not make this clear,
a n d that he gives so little space to the newsense, w h i c h i s p r o m i n e n t i n h i s
own use of the word.
" E l s e w h e r e " : for references see Bonitz, Index Aristotelicus 713Ч3-
714*3. A m o n g the Aristotelian senses of *coincidental' not mentioned i n
this chapter the most important is *in a derivative sense'; see Posterior
Analytics I 4. 7 3 4 " 5 ^^^y ^or possible examples of its use. Categories
6. 5 * 3 9 , J 7. 1017*7-22,J 13. 1020*26-32.

182
 NOTES

METAPHYSICS BOOK EPSILON

CHAPTER 1

T h i s chapter has always given difficulty to Aristotle's readers. L i k e Г i ,

it seeks to define the scope ofmetaphysics, the " p r i m a r y discipline" of
1026*15-16 a n d the " p r i m a r y [or first] philosophy" of 1026*24; but it
contains two at first sight incompatible definitions, one of w h i c h makes
metaphysics the non-particular (1025^7-10) study of aU things-that-are
(metaphysica generalis), while the other restricts it to the " m o s t estim­
able genus" (1026*21, metaphysica specialis). T h e last paragraph o f t h e
chapter attempts to reconcile these two conflicting conceptions.

10a5^3. I n the notes o n Г 1 wiU be found comments o n "origins a n d

causes", *'qua thing-that-is", " d i s c i p l i n e " a n d " a particular g e n u s " . T h e
disciplines w h i c h are not " t h i n k i n g , or thought-partaking" are presmn-
ably the humbler practical skills like knitting. Aristotle asserts that even
more exalted special sciences do not attempt to say either (A) what their
subject-matter is or (B) whether it is, i.e. exists.
(A) H e does not m e a n to deny that e.g. zoologists know a n d indicate
that their subject studies animals, but only to deny that they indicate
w h a t animals are, i.e. define them. Y e t this is strange^ for it seems neither
impossible that a zoologist should " p r o d u c e a statement o f " what animals
are, nor particularly appropriate that a metaphysician should do it for
h i m (there is nothing o f t h e sort i n Aristotle's Metaphysics). " T h e r e is no
demonstration of substance" is ambiguous, ( i ) Aristotle m a y m e a n that
scientists' attempts to demonstrate definitions are unsuccessful, because
they rely o n " i n d u c t i o n " w h i c h achieves only " s o m e o t h e r " — i n a d e ­
q u a t e — " m a n n e r ofindicating i t " . Induction, i n Aristotle, is the citation
of observed instances either to establish or more conmionly just to illus­
trate {PosteriorAnalytics 1 1. 71^9-10) a n d d r a w attention to a general
truth. I n other places he contrasts this method with syllogism (e.g. Prior
Analytics 1 25. 42*3-4) a n d v ^ t h demonstration (e.g. Posterior Armlytics
I 18. 8i» 40-b1)^ W e jnay agree that definitions cannot be established
by induction, even i f other truths c a n be, but three problems r e m a i n :
(i) c a n any method demonstrate a definition (see Posterior Analytics I I
3. 90^24-5, " t h e principles of demonstrations are definitions, a n d of
them we have already shown that there are no demonstrations") ? (ii) I f
it c a n , w h y should not zoologists a n d others adopt it? a n d (iii) w h y
should metaphysicians undertake the task? (2) Alternatively w e m a y
take Aristotle to m e a n that demonstration of definitions is impossible a n d
needs to be replaced by " s o m e other m a n n e r " . So m u c h is made " o b ­
v i o u s " by the scientists' actual methods; yet their preliminary induction,

183
1025^3 METAPHYSIGS El

w h i c h indicates their subject-matter by perception or renders credible

some hypothesis as to what it i s , i s not the " m a n n e r " needed: it cannot,
even non-demonstratively, " p r o d u c e any statement of w h a t the genus
i s " . T h i s solves (i) but stiU fails to explain w h y scientists cannot, a n d
metaphysicians should, give definitions. Parts of Aristotle's Metaphysics
do exhibit a non-demonstrative method of arriving at definitions, but
not definidons ofthe subject-matter ofthe special sciences. Aristotle needs
the distinction made at Posterior Analytics 1 9 , 10 between common a n d
special principles. ( 3 ) K 7. 1 0 6 4 ^ 8 - 1 0 glosses " T h a t is w h y it is also plain
from such a n induction [sc. from this review of the special sciences] that
there is no demonstration ofsubstance a n d of what a thing i s " . Whether
or not this gives Aristotle's general meaning, it certainly misunderstands,
or changes, the force of " f r o m such a n i n d u c t i o n " .
(B) I n saying that " t h e same thinking wiU indicate w h a t a thing is and
whether it i s " , Aristotle does not m e a n that these are the same question
(see Posterior Analytics I 1. 7 1 * 1 1 - 1 7 a n d I I 7. 9 2 ^ 4 - 1 1 ) , but that the
answers to both questions are principles (cf. PosteriorAnalytics I 1 0 . 7 6 ^
3 1 - 6 , where it is inferred that neither kind c a n be " s h o w n " ) .

1025^18. T h e division of " t h i n k i n g " a n d disciplines into theoretical,

practical, and productive recurs at Topics V I 6 . 1 4 5 * 1 5 . Theoretical
questions are of the form *is this the case?' practical questions of the
form *should this be done, or how should this be done?' a n d productive
quesdons of the form *how should this be m a d e ? ' Practical knowledge
is not, for Aristotle, the same as useful knowledge; whereas we might say
nowadays that knowledge of F r e n c h is more practical than knowledge
of L a t i n , a n d of F r e n c h history than R o m a n history, i n Aristotle's usage
both of the first two are practical, both of the last two theoretical.
Furthermore, his theoretical statements include the particular a n d
descriptive as well as explanatory generalizations. T h e cognate verb
4heorein' is rendered *study' i n this translation. Ross, in his book Aristotle
(p. 6 2 ) , seems to interpret the threefold scheme differently:

Aristotle's classification ofthe sciences . . . divides them into the theoretical,

which aim at knowledge for its own sake, the practical, which aim at know­
ledge as a guide to conduct, and the productive, which aim at knowledge to
be used in making something useful or beautiful.

T h e s e words imply that the same item ofknowledge, e.g. how the Polish
border r a n i n 1 7 4 0 , could be theoretical for one m a n (the academic
historian), practical for another (the diplomat who wishes to avoid old
mistakes), a n d productive for a third (the maker o f a historical atlas). O n
the other interpretation that item of knowledge is theoretical however
it is used, a n d even practical a n d productive knowledge m a y be pursued
for their o w n sakes, i n the sense that one m a y e.g. decide what to do

184
El NOTES 1025b18

merely i n order to do it (but not merely i n order to know what to do, see
A 2. 982^27-8, α I . 993^20-1).
Aristotle argues that {a) physics studies things " i n w h i c h the origin of
change a n d of keeping-the-same is i n themselves" (i.e. self-changing
things; for the double use o f " i n " cLA 4. 1014^18-20 a n d note); {b) acts
a n d products have their origins i n the doer a n d the p r o d u c e r ; ( i )
therefore physics does not deal w i t h acts a n d products; (d) therefore
physics is theoretical. T h e argument is weak, because there is nothing
to stop a discipline w h i c h deals w i t h self-changing substances dealing
eo ipso w i t h the acts a n d products of those substances. Aristotle has not
demonstrated the theoretical status of e.g. anthropology, w h i c h studies
h u m a n behaviour but is not p r a c t i c a l — o n the first interpretation,
because it answers questions ofthe form *what is the case?' on the second
because it need not be studied as a guide to conduct. T h e " s i n c e " clause
i n ^18-20 does not support Aristotle's c l a i m that physics is theoretical,
but perhaps is meant to explain w h y the question is raised.
" S u b s t a n c e as i n a f o r m u l a " : i.e. form, for, according to Aristotle, the
formula saying w h a t a particular thing is c a n specify only of w h a t sort
the thing is a n d cannotmention the matter w h i c h individuates the thing
from others of the same sort (cf. A 6. 1016^32-3). T h e r e are thus two
kinds of substance: " o n e is the f o r m " (Z 11. 1037*29) or formula {Z
15· 1039b20; c f Δ 2. 1013*26-7 for the lax identification of these t w o ) ;
the other is the concrete thing, " f o r m u l a taken w i t h m a t t e r " {Z 15.
1039^21-2). " Y e t not separable substance" is difficult. I n one sense
(employed i n the paragraph beginning 1026*6) no forms, but only
concrete things, are separable, v i z . separately identifiable. A thing's
form is often, however, "separable i n respect of f o r m u l a " . Physics I I
I . 193^4-5, i.e. a definitionwill mention it alone. T h e thesis is, then, that
although physics deals with forms or sorts of things as revealed i n defini­
tions {Z I I . 1037^16-17), there is a w a y — e x p l a i n e d i n the next p a r a ­
g r a p h — i n w h i c h matter usually also features i n those definitions. ( A n
alternative reading, *only as not separable', is adopted by Ross, giving
m u c h the same sense. T h e Greek is a w k w a r d i n either case.)

1025^28. Although definitions cannot mention individuating matter,

they c a n " i n c l u d e m a t t e r " i n a different w a y . T h e definition of a nose
or a n a n i m a l wiU, i f complete, specify the kind of matter w h i c h the form
must inform: the nose, for instance, must amongst other things be made
of flesh. T h e same is true of " t h e s n u b " because only noses are snub.
W h e n Aristotle says that, b y contrast, " c o n c a v i t y is independent of
perceptible m a t t e r " it is unclear whether he means ( i ) concave objects
do not have to be o f a n y particular kind ofmatter, (2) concave objects do
not have to be material at aU, since they m a y be geometrical figures (cf.
Z 10. 1036^3-5), (3) the property concavity is not material {De Generatione

185
t025^28 METAPHYSICS El

et Corruptione I 4. 320*2-5). T h e last would require " t h e s n u b " to be

understood i n the sense *snubness', but since snubness is not made of
flesh, the w a y i n w h i c h snubness contains matter i n its definition is quite
different fi:om the w a y i n w h i c h noses, flesh, or animals do (and con­
cavity does not). F r o m confiision of these two ways Aristotle elsewhere
developed a n argument that the snub cannot be defined at a l l , since, if it
were, its definition would be *concave nose', a n d *snub nose' would
accordingly have to m e a n *concave nose nose' (Z 5. 1030^29-1031*1). I n
De Sophisticis Elerwhis (31. 182*4-6) he came to see that this argument is
w r o n g ; 'snub' means, not *concave nose' but *(having) concavity ofthe
nose', so that *snub nose' means *nose having concavity of the nose'
w h i c h is " n o t i n the least a b s u r d " .
Aristotle concludes that " t h e manner i n w h i c h we need to investigate
a n d define what a thing is i n the case of naturally existing things is
p l a i n " . I t is not, however, plain whether a l l naturally existing things
need, as noses do, definitions that mention a particular kind of matter,
or merely, as perhaps does the concave, definitions that make them out
to be changeable a n d so material ( c f Z 11. 1036^3-7).

1026*5. See De Anima I 1. 403^16-25 a n d I I I 4. 429*24-7.

1026*6. T h e first two paragraphs ofthe chapter appeared to argue for

a difference between physics a n d metaphysics on the ground that
physics deals w i t h a genus or part of what-is. Y e t now the search is for
another part of what-is to be the subject-matter of metaphysics. T w o
solutions seem possible: ( i ) that the chapter is a clumsy recension of
material composed at different periods i n Aristotle's life, a n d reflects a
change i n his conception of metaphysics; or (2) that he was aware that
the argument implied i n 1025^18-19 is inadequate a n d needs replacing
if, as 1026^23-32 wiU contend, the same discipUne c a n have both a part
a n d the whole ofwhat-is for its province (see notes on 1026^23)—whence,
that this paragraph should be read as saying that metaphysics includes,
but is not confined to, the study of the changeless a n d separable (this was
the solution o f N a t o r p , who stressed the " a l s o " i n * i 6 — w h e r e others have
more awkwardly translated 'deal w i t h things both separable a n d change­
less'; i n addition Natorp interpreted * i o - i i 'it faUs to a theoretical
discipline to ascertain whether there is anything . . .', a h d excised the
later references to theology, the former a not impossible proposal, the
latter, as Ross says, " v i o l e n t " ) .
Some mathematics studies its objects "qua s e p a r a b l e " ; that is, although
numbers a n d lines a n d the like cannot be detached from changeable
things (and so are not substances), they are abstracted b y the m a t h e m a ­
tician {Physics I I 2. 193^33-4, " t h e y are separable from change i n
t h o u g h t " ) . Aristotle does not here commit himself on the question whether
the objects of mathematics are actually separable, although he contends

186
El NOTES I026^6

a t *14-15 t h a t t h o s e s t u d i e d b y " c e r t a i n p a r t s o f m a t h e m a t i c s " a r c not

(does h e m e a n , a r e not e v e n abstracted, i n contrast w i t h "some m a t h e ­
m a t i c s " ? ) . Might not metaphysics, then, be identified w i t h a n y r e m a i n i n g
p a r t o f m a t h e m a t i c s w h i c h does study changeless a n d separable objects?
Aristotle's a n s w e r , n o t g i v e n u n t i l M 2. 1076^11-3. 1078*9, i s t h a t t h e r e
i s n o such f u r t h e r p a r t : aU m a t h e m a t i c s d e a l s w i t h t h e p e r c e p t i b l e qva
lengths, planes, etc., a n d so witii the changeless only b y abstraction.
" I f t h e r e is a n y t h i n g i n v a r i a b l e a n d c h a n g e l e s s a n d s e p a r a b l e " ( * i o -
11): a p r o o f i s o f f e r e d a t Л 6. 1071^5-9: i f aU s u b s t a n c e s w e r e d e s t r u c ­
tible, everything w o u l d b e destructible; b u t time a n d change are inde­
structible.
We are n o t told w h y the discipline whose objects a r e changeless and
s e p a r a b l e m u s t b e p r i o r t o aU o t h e r s . I t i s p r i o r b e c a u s e i t s o b j e c t s a r c
prior; its objects a r e prior because (i) o n l y c o n c r e t e s u b s t a n c e s are
s e p a r a b l e {Physics I 2. 185*31), (ii) s u b s t a n c e s a r e p r i o r t o aU o t h e r t h i n g s
{Z I. 1028*29-^2), a n d (Ui) c h a n g e l e s s s u b s t a n c e s a r e p r i o r t o o t h e r
substances (changers a r e prior to t h e things they change, therefore
c h a n g e l e s s c h a n g e r s to e v e r y t h i n g else, Л 8. 1073*23-36).
A t *13-14 t h e M S S . r e a d " t h e s t u d y o f n a t u r e d e a l s w i t h t h i n g s t h a t
a r e z72separable b u t n o t c h a n g e l e s s " . T h e s u r r o u n d i n g ш e s o f * s e p a r a b l e '
r e q u i r e t h i s t o m e a n , n o t * i n s e p a r a b l e f r o m m a t t e r ' , i.e. c o n c r e t e , w h i c h
w o u l d i n a n y case d e m a n d ''and n o t c h a n g e l e s s " , b u t *not s e p a r a t e l y
i d e n t i f i a b l e ' , i.e. n o n - c o n c r e t e . Physics w a s d e s c r i b e d a s t h e s t u d y o f
n o n - c o n c r e t e fornas a t 1025^27-8 ( c f Z 11.1037*16-17), butformscannot
b e i n t e n d e d h e r e , b e c a u s e t h e y are changeless—whence t h e e m e n d a t i o n
*separable', adopted i n the translation.
The difference b e t w e e n m a t h e m a t i c s a n d physics is also discussed a t
Physics I I 2. 193^22-194*12, De Anima I I I 7. 431^12-16, De Caelo I I I
I . 299*11-17, K3. i o 6 i * 2 8 - 4 ·
1026*16. " I n v a r i a b l e " t r a n s l a t e s t h e a d j e c t i v e f r o m * a l w a y s ' a n d m a y
t h e r e f o r e m e a n e i t h e r * e v e r l a s t i n g ' o r 41ways t h e s a m e ' ( c f E 2 . 1027*19).
The f o r m e r m a k e s Aristotle's r e m a r k a b s u r d ( a l t h o u g h h e d i d believe
t h a t some c a u s e s a r e e v e r l a s t i n g , f o r o t h e r w i s e t h e r e w o u l d b e e i t h e r a
b e g i n n i n g o f t i m e , d e n i e d a t Physics V I I I 1. 251^19-23, o r a n i n f i n i t e
c h a i n ofcauses, denied at α 2. 994*1-^27). T h e l a t t e r c r e d i t s h i m w i t h t h e
o p i n i o n t h a t c a u s a t i o n is u n i v e r s a l i z a b l e , p e r h a p s i n the sense t h a t i f A
causes E a n d B does not there m u s t b e s o m e further difference between
t h e m ( f o r t h e c o n n e c t i o n b e t w e e n c a u s e a n d u n i v e r s a l i t y c f . A 1. 981*
24-30 with B 6. 1003*14-15). T h e " d i v i n i t i e s o b v i o u s t o u s " are the
4. 196*33-4), c a u s e d , i.e. m o v e d ,
h e a v e n l y b o d i e s {Physics I I b y the
spheres (Л 7. 1072*19-8. 1073^3).

1026*18. T h e c l a i m t h a t " t h e d i v i n e is a c o n s t i t u e n t " o f a l l c h a n g e l e s s

s u b s t a n c e s , i f o f a n y t h i n g , e x p l a i n s w h y t h e s t u d y o f t h e m is caUed

187
 I026>l8 METAPHYSICS Е\

theology (a description used only here a n d i n the corresponding passage

at K 7. 1064b3), but is not itselfanywherejustified by Aristotle. T h e o r e t i ­
cal disciplines are the most estimable because philosophy, sc. theoretical
philosophy, " a l o n e exists for its own s a k e " {A 2. 982^27-8: we want to
know what to do i n order to do it, but we w a n t — o r c a n w a n t — t o know
what is the case just i n order to know i t ) .

pp. 201-3 1026*23. Aristotle n o w attempts to resolve the conflict between the
conception of metaphysics as universal i n subject-matter (on w h i c h see
notes on Г 1) a n d the conception of it as confined to the study of change­
less substances. C a n *metaphysica generalis' be identified w i t h *meta-
physica specialis' ? (i) T h e comparison w i t h mathemadcs speaks only for
the existence of general metaphysics (more fully argued i n Г 2; o n
universal mathematics see K 7. 1064^8-9, M 2 1077^9-10, ^17-22,
PosteriorAnalytics 1 5. 74*17-25, H e a t h , Mathematics in Aristotle, 222-4).
(ii) I n his next sentence Aristotle pleads for the recognition of changeless
substances, but this, while telling against the identification of special
metaphysics w i t h physics, does nothing to support the identification of
special with general metaphysics, (iii) A U the work of reconciling these
two conceptions is left to the words " t h a t [sc. the study of changeless
substancesJ is . . . universal i n this way, because p r i m a r y " . Space permits
only a brief and dogmatic explication of these much-discussed words.
Metaphysics is p r i m a r y , or first, philosophy because it studies changeless
substances, the p r i m a r y existents; a study of primary existents will also
be universal, i.e. a study o f a l l existents, i f i t aims to establish propositions
w h i c h reveal the ways i n w h i c h other existents depend on, a n d " o w e
their being called w h a t they a r e " {Г2, 1003^16-17) to, p r i m a r y existents.
S u c h propositions constitute what we m a y call a n ontology. So far as
metaphysics is ontology, it is therefore both general a n d special. H o w ­
ever, Aristotle makes his metaphysics seem wider than ontology i n one
w a y , a n d it is wider i n another. H e makes it seem wider w h e n he implies
that any truth about p r i m a r y existents will be " u n i v e r s a l " a n d so part of
metaphysics; yet such truths will contribute to ontology only w h e n they
are investigated for the purpose ofrelating the primary existents to other
existents. T h e words " u n i v e r s a l . . . because p r i m a r y " ignore this r e ­
stricting condition, but the actual discussions of substance i n ZH a n d of
changeless substances i n Л 4-10 m a y be thought to satisfy it. Aristotle's
metaphysics is, however, wider than ontology i n another way, because it
embraces the discussion of concepts like unity, i n A a n d / , a n d of the
principles ofnon-contradiction a n d excluded middle, i n r * 3 - 8 . H e seems
to have been aware that these topics do not give a n y special place to
substance (see Г 2. 1004^6-9), let alone changeless substance. O n one
inteφretation (see notes опГ i ) , the phrase " t h a t w h i c h is qua thing-that-

188
El NOTES 1036*23

i s " is meant to cover this non-ontological part of the subject; but i f so,
it covers the ontological part also, for both parts are " u n i v e r s a l " .
I f there were no changeless substances, physics would be the p r i m a r y
discipline a n d therefore universal: i.e. ontology would be counted as a
part of physics. E v e n as it is, Aristotle seems content to include within
physics m u c h that we would regard as philosophical.

GHAPTER 2

T h e division of senses of *be' differs i n two ways from that given i n

Г 2. 1003*33-b10: the kinds of beingdistinguished i n Г 2 are here
collected together under the heading "figures of predication" a n d
separated from three further kinds, as ΊηΔ 7; a n d the purpose ofthe new
division is not, as ίηΓα, to raise a possible objection to the study of hieta-
physics but to delimit its scope. T h i s purpose was shared by the latter
part ofF 2; but the conclusions ofthe two chapters are i n conflict at one
point, a n d they have the general difference that whereas Г 2 argued for
inclusions, E 2 demands exclusions.

1026*33. F o r comments on the fourfold division of senses of Ъе' see

notes onA 7; " w a s " i n «34 m a y refer back to that chapter. T h e present
passage adds t o J 7 the point that aU four ways are ways of saying that
a thing is " b a l d l y " . T h i s word c a n m e a n *without qualification' ( c f
1027*5) or *without addition' (Topics 11 11. 115^29-35); the latter sense
occurs elsewhere w i t h Ъе' as a w a y of distinguishing existence from
copulative being {PosteriorAnalytics 11 1. 89^33, " I m e a n whether it is or
is not baldly, not whether [it is] pale or n o t " ; cf. I I 2. 90*3-4). T h a t the
word is meant to distinguish existence here has been doubted on two
grounds, (i) Aristode elsewhere asserts (Z 1. 1028*30-1) that only sub­
stances c a n Ъе' baldly. I t is h a r d to see how this objection works, since
that assertion is i n any case contradicted i n the present passage. T h e
truth is that Aristotle h a d no settled opinion as to whether *x is' must be
elliptical w h e n x is a non-substance: see Z 4. 1030*32-^4, w h i c h first
suggests that we assert being of non-substances " w i t h a n addition a n d
[ofsubstances?] w i t h a subtraction", but then substitutes as " c o r r e c t "
the alternative theory adumbrated i n Г 2. 1003*33-^10 that the being
of non-substances is not elliptical but derivative, (ii) I t has also been
objected that Aristotle's second division of being, being as truth, is not a
type of existence. But i n Δ 29 Aristotle treats falsehood as a property not
of propositions or sentences but of " a c t u a l things", sc. states of affairs,
a n d infers that a false state of affairs is one that *is not'. T h i s doctrine
does i n effect propose that truth and falsehood are forms of existence a n d

189
ioa6*33 METAPHYSICS E2

non-cxistencc. W e m a y conclude that even i f there are places i n w h i c h

Aristotle restricts *being baldly' to substances, h e does not do so i n ^ 2 ;
a n d that i n E 2 *is baldly' means 'exists'.

1026^2. T h i s paragraph is discussed under six headings: ( A ) the

example ofhousebuilding; (B) the conflict between bio-24 ^ 2;
( C ) the example ofthe triangle; ( D i ) , (D2) the sophistic puzzles; ( E ) the
coincidental a n d w h a t is not.
(A) Aristotie c a n h a r d l y m e a n t o deny either that housebuilders c a n
produce pleasing houses, or that they c a n do so deliberately; his con­
tention seems to be that such outcomes a n d such aims are not attributable
to their owners' knowledge or skiU лу hoшebгülders. W e m a y agree that
the *discipline' ofhousebuilding is simply the ability to build houses, not
houses of any particular k i n d . B u t (i) Aristotle's choice of *pleasing' a n d
*beneficial' is unfortunate, for although these a r e examples o f coinci­
dental predicates of a house, so that it is possibU to build houses without
building pleasing or beneficial ones, we naight still be inclined to say that
a m a n does not have the knowledge how to build houses until h e knows
how to make them pleasing or beneficial or at any rate i n some w a y good,
(ii) A house m a y have other coincidental properties, e.g. curvature, the
ability to produce w h i c h is a part, although not a necessary part, of the
housebuilder's skiU. T h e construction of, for example, bow windows is
not " n e g l e c t e d " i n the discipline ofhousebxtilding. (iü) O t h e r coinciden­
tal properties of houses, such as their location, although not studied i n
the art ofhousebuilding, come within the scope of different disciplines—
those of the landscape artist a n d the zoning officer. T h u s Aristotle
entirely fails to show that no study deals w i t h w h a t is coincidental to a
house.
(B) T h e example o f geometry is no more convincing; i n addition it
contradicts w h a t A r i s t o t i c says elsewhere. T h e question reserved from
treatment b y geometers, " w h e t h e r a triangle a n d a triangle possessing
two right angles (i.e. angles having that sum) are different" is one whose
answer turns on the conditions of application ofthe quite general concept
of difference. I n this respect i t is like the question mentioned at Г 2.
1004b2-3 " w h e t h e r Socrates a n d Socrates sitting down are the s a m e "
w h i c h , because i t s a n s w e r demands examination of the general concept
of sameness, is there included within the sphere of metaphysics. Y e t
according to the present passage it w o u l d be excluded not only from
geometry but by implication from metaphysics a n d every other discipline.
T h i s contrast is heightened i n the lines that foUow. Г 2. 1004^22-3
asserted that "sophistic . . . ranges over the same genus as philosophy"
sc. that w h i c h is, but at 1026^14-15 we are told that " i n a w a y " sophistic
deals w i t h w h a t is not; a n d the sophistical questions listed at ^16-21 a r e
even closer to those included within metaphysics at Г 2. 1004^2-4. T h e

190
E2 NOTES 1026b2
thought w h i c h finds no connection between sophistic a n d metaphysics
seems, Uke m u c h οΐΕ, more prinütive a n d muddled than that οίΓ a.
( C ) Although Aristotle does not tell us what sense of 'coincidental'
he is employing, we are presvmiably to understand h i m with reference
to the definition i n the next paragraph ("this w a y " : sc. i n contrast to
A 30. 1025"30). Y e t on that definition it t u r m o u t ^ a c i R o s s — t h a t l i e i n g
the same as a triangle with two right angles is not a coincidental prop­
erty of a triangU; for aU triangles are like that. W e have to apply the
definition i n another w a y ; a n d the aUeged fact w h i c h Aristotle seems to
have i n m i n d is that the predicates 'triangle' a n d 'triangle with two right
angles', although equivalent, are not freely substitutable for one another;
since the occurrence of 'triangle' within the longer phrase cannot be
replaced by the longer phrase without repetition, a n d such repetition, or
" b a b b l i n g " , seemed to h i m nonsensical ( c f De Sophisticis Elenchis,
chapters 13, w h i c h states, a n d 31, w h i c h solves, similar problems
concerned with babbling—but not precisely this one). Aristotle takes
this aUeged restriction on substituting the one expression for the other as
a reason for saying that the things designated by the expressioi« are not
the same. B u t since the substitution is not always barred, we c a n say that
the things are sometimes the same, though sometimes different—hence
coincidentaUy the same a n d coincidentaUy different.
( D i ) I t is rather simpler to explain what is coincidental about the
difference between "artistic a n d literate, a n d artistic Coriscus a n d
C o r i s c u s " . Whereas the substitution of 'triangle having two right
angles' for 'triangle' wiU sometimes, according to Aristotle's doctrine of
babbling, produce nonsense, the substitution of 'artistic' for 'literate'
wiU sometimes produce falsehood. I n the example invented b y A l e x a n d e r
a n d repeated by Ross, it is assumed that Socrates is both Uterate a n d
artistic, Aristarchus literate but not artistic. T h e n the artistic is the same
as the literate—i.e. the one expression c a n be substituted for the o t h e r —
i n reference to Socrates but not to Aristarchus. T h e two are sometimes
the same, sometimes different, hence coincidentaUy the same a n d dif­
ferent. 'Artistic Coriscus' a n d 'Coriscus' m a y fail of substitutability i n
either of the two preceding ways. I f Coriscus changes from inartistic to
artistic or vice versa, 'artistic Coriscus' wiU sometimes be a n d sometimes
not be a correct variant for 'Coriscus'. E v e n i f he does not change,
'artistic Coriscus' cannot without babbling be substituted for 'Coriscus'
in the phrase 'artistic Coriscus' itsetf.
(D2) T h e argument referred to i n bj8-20 must have sought to prove pp. 211
a n absurd implication by т е а г к of the principle that what " i s but [w£w]
not always, has come to b e " . Ross's speculative but elegant reconstruc­
tion m a y be paraphrased thus:
(1) Someone, being artistic, has come to be literate;
(2) so, being artistic, h e is Uterate;

i91
I026b2 METAPHYSICS E2

(3) so, being literate, he is artistic;

(4) but it was not always the case that, being literate, he was artistic;
(5) so, being literate, he has come to be artistic.
Ross does not explain so clearly h o w Aristotle would have us solve the
puzzle. T h e actual fallacy is of a type recognized i n De Sophisticis Elenchis
4. 166*22-32 under the title *composition', i.e. bracketing: Hne (5) ought
to read *he has come to be, being literate, artistic'. B u t i n De Sophisticis
Elenchis Aristotle rightly distinguished that type from the puzzles about
reference w h i c h he put under the heading *coincidence' (ib., chapters 5
a n d 24). H e r e , it seems, his eagerness to display the vagaries of sophistic
arguments has led h i m a w a y from the thesis w h i c h those arguments
were supposed to illustrate.
( E ) Aristotle's assimilation ofhis own view that sophistic deals with the
coincidental to Plato's that it deals w i t h w h a t is not (i.e. that its pro­
positions are false, Sophist 254 a, c f N 2. 1089*20-1) c a n be explained as
follows: i f the sophist maintains, without qualification, that e.g. the
literate a n d the artistic are the same (or are different), w h a t he maintains
is both false a n d sometimes, i.e. coincidentaUy, true. " W h a t is coinci­
dental is close to w h a t is n o t " i n the sense that being the c a s e j u s t some­
times is a w a y of not—sc. not a l w a y s — b e i n g the case. T h e point turns
on the fact that a sophist who said that p would be taken, i n the
absence of explicit qualification, to m e a n that p always and i n all cases.
" L i k e a mere n a m e " emends the M S S . reading w h i c h , i f i t means a n y ­
thing, means 4 h e coincidental exists as it were i n name only'. T h e
reference to names is not pursued, a n d m a y be due to corruption. T h e
emendation adopted c a n hardly, as Ross supposes, convey the sense that
puzzles about coincidences depend upon ambiguities of language, a
suggestion w h i c h is i n a n y case not borne out by Aristotle's examples;
the meaning is either *x is F coincidentaUy' is like 'x is F i n name a n d
not i n fact' or *coincidental properties attach no more closely than
names'. " M o r e than a n y t h i n g " : the Greek could m e a n *more than a n y ­
one's'.
F o r the contention that things that are coincidentaUy are never i n
process o f c o n u n g to be or ofdestruction see notes onE 3.

1026^24. T h e definition w h i c h Aristotle now gives of*coincidental'

(*non-regular') is more fully discussed i n the notes o n J 30. 1025*14, a
chapter i n w h i c h it appears alongside another definition ('non-essential',
1025*30). I t was i n the latter of these senses that Aristotle distinguished
coincidental being at the beginning of E 2; w e m a y therefore complain
at the revelation that his case against a science of coincidental being
assumes the former sense, for if there is no science of the non-regular it
does not foUow that there is no science of the non-essential.
W h e n , i n this a n d the foUowing chapter, he says that something ' i s '

192
El NOTES 1026b24

coincidentaUy, or always, or of necessity, or for the most part, he should

be understood to speak of the manner ofexistence ofsome state ofaffairs.
T h e " c a u s e " i n ^26 a n d the " o r i g i n a n d c a u s e " i n ^31 are the proof
that coincidental states ofaffairs occur: viz. i f it is only i n most cases that
p, it must be true i n some but less than most cases that not-j&. T h e proof
is repeated at 1027*8-13 a n d 1027*15-17. ( I n 1027*13-15, however,
" c a u s e " seems to m e a n 'explanation'.)
F o r the conjunction of 'always' with 'of necessity' see De Generatione
et Corruptione I I 11. 337^35-338*2; ^32 wiU define 'coincidence' i n terms
of 'always', but Δ 30 used 'of necessity'. B y this equation Aristotle seems
to leave no room for contingent regularities, such as might be expressed
i n the belief that honey-water always benefits the bilious; the omission
doubtless helped his assinulation of the two senses i n w h i c h 'coinciden­
tal' means 'non-essential' a n d 'non-regular'. T h e temporal element i n
'always' should not be stressed; Aristotle quite often uses the word to
m e a n ' i n aU cases' (as i n 'flowers always die i n the end', c f Δ 2. 1013^33,
Δ 6. 1016^35, but contrast Posterior Analytics I 4. 73*28-34). O n senses of
'necessary' see Δ 5.
A m o n g Aristotle's examples of the coincidental, the last two do not
obviously illustrate his definition. A r e we to infer that a housebuilder
heals his patients coincidentaUy even i f housebuilders do so for the most
part (it is not relevant to ask whether the particular housebuilder does so
for the most part or n o t ; see notes o n J 30. 1025*14), or would their
having such regular success be enough to make healing " c h a r a c t e r i s t i c "
of housebuilders? Aristotle seems to waver between a criterion based
purely o n frequencies a n d one based o n considerations as to w h a t it is to
be 2i doctor a n d a housebuilder; the latter would allow us to say a priori
that the housebuilder's success was a coincidence, arguing from the
conceptual truth that nobody heals qua housebuilder. Aristotle makes his
healing housebuilder a doctor also; but it cannot be a truth, either con­
ceptual or empirical, that aU who heal are doctors. W h e n he says at *5
that the cook produces health " i n a w a y " , he means ' w i t h a q u a l i ­
fication', viz. coincidentaUy or (perhaps) qua dietician.

1027*5. Both Ross a n d Jaeger emend, but the text adopted, w h i c h

foUows two good M S S . , seems to make tolerable sense. " F o r " introduces
a division ofcoincidental products: some o f t h e m are like health, w h i c h
is sometimes coincidentaUy produced b y cookery, sometimes n o n -
coincidentally produced b y medical a r t ; others are like pleasantness
(1026^7), never produced except coincidentaUy.

1027*13. T h e thought seems to be: i n looking for the explanation o f a

coincidental state of affairs such as a stormy dog-day or a restorative
cook, w e must examine that class of things, dog-days or men, w h i c h are

193
1027*13 METAPHYSICS E2

capable of being stormy or restorative although they are not so usually.

Aristotle does not m e a n that a coincidental state of affairs is a material
t h i n g ; its " m a t t e r " is the k i n d o f t h i n g it i s — t h e class determined b y the
properties w h i c h it must retain through every change, as the dog-day
must r e m a i n a dog-day even i f ceasing to be stormy; cf. Δ 28. 1024^8.
F o r the idea that matter c a n actually be a n explanation or " c a u s e " see
Δ 2. 1013^18.

1027*15. T h i s reads like a marginal gloss o n *8-13, the argument of

w h i c h it repeats. A negative answer to the " o r i g i n a l question" (literally
*origin') provides the premiss for deducing that coincidental states of
affairs occur.

1027*17. " I n v a r i a b l e " , i.e. invariably connected; G e w i r t h , Philo-

50phical Review 1953, 585 n . 20. T h e forward reference is apparently to
the proof of changeless substances i n Л 6-8.

1027*19. T h i s p a r a g r a p h gives a new a n d better—but c r y p t i c —

defence of the thesis already argued for at 1026^2-24. T h e argument is
this. I f the proposition *fever-patients benefit from honey-water' is to
contribute to the " d i s c i p H n e " of fevers, a n d if, i n particular, it is to be
used i n teaching somebody h o w to cure fever-patients or w h e n to expect
cures, it must be understood as a proposition about {a) aU or {b) most
cases. A s a n example of a proposition not meeting these conditions
Aristotle takes " t h e exception" to the latter, viz. {c) *in just a few cases
fever-patients do not benefit from honey-water'. T h e example is u n ­
fortunate, for (as Aristotle knew) knowledge of {b) implies knowledge of
(c). B u t his general point is that propositions of the form ' i n some cases
{at least a few cases) it is true that p' fail, by reason of indefiniteness, to be
contributions to science. T h e exception c a n be " d e f i n e d " or specified by
putting *at new m o o n ' i n place of *in some cases'; but " t h a t w h i c h is at
new moon is a k o either always or for the most p a r t " — b y w h i c h Aristotle
means not that there are no irregularities at n e w moon, nor that the
unqualified proposition *at new moon fever-patients do not benefit from
honey-water' necessarily implies *in aU or most cases', but that the u n ­
qualified proposition must be understood w i t h that implication 7 / i t is to be
used i n learning or teaching. " T h e exception . . , cannot be s t a t e d " , sc.
without abandoning indefiniteness.
T h i s argument calls for four conunents. ( A ) I t s acceptability depends
on construing " d i s c i p l i n e " i n a narrower sense than 'knowledge'. I t is
possible to know both particular a n d indefinite truths, as that honey-
water has benefited this patient a n d that it sometimes benefits some patients.
A t A I . 981*7-29 Aristode describes the former k i n d of knowledge as
" e x p e r i e n c e " ; but he concedes that it is a form of knowledge w h e n
he adds (using the ordinary Greek verb for 'know', 'eidenaV) " m e n of

194
E2 NOTES 1037*19

experience know that, but do not know why". Knowledge i n the sense of
discipline or science {'episteme'), o n thc other h a n d , has to be knowledge
of causes {E 1. 1025^5-7) a n d of universals (sc. universal truths, B
6. ioo3'i4-i5).
(B) T h e requirement that the truths of a science or discipline be u n i ­
versal is normally, as here, relaxed b y Aristotle, w h o saw that partial
generaUzations of the form ' i t is for the most part the case that p' (cf.
Posterior Analytics I 30. 87^19-22) c a n be both explanatory a n d pre­
dictive.
( C ) T r u t h s of the form 'the F's being G is a coincidental thing-that-is',
i.e. ' i t is j u s t sometimes the case than a n F is G', do not feature as the
established propositions of any discipline, but that does not mean that
they are not studied b y a n y discipline. F o r although the student will not
learn from the proposition ' i t is sometimes the case that p', he will l e a m
by studying the question w h i c h that proposition raises, viz. 'when is i i the
case that p?' I n the paragraph beginning at 1026^2 Aristotle overlooked
this distinction. I t is a coincidence that a house is pleasing, because
houses are so only sometimes; it follows that the housebuilder's art wiU
not teach h i m that houses are sometimes pleasing; but it wiU teach h i m
when they are, i.e. under w h a t producible conditions. Similarly, it is a
coincidence that the artistic a n d the Uterate are the same, because these
expressions are only sometimes interchangeable salva veritate; but w h a t
makes the sophist unscholarly is not his attention to this fact but his
inattention •to the further question 'under w h a t conditions are they
(always) interchangeable?' T h e failure of sophistic to be a discipline is
due not, as 1026^2-24 claimed, to the natureofthesophist'sproblems,but
to his not attempting—allegedly—to solve them.
( D ) Commenting o n i027"25^ Ross writes that " i t is perhaps the
only place i n w h i c h Aristotle implies the view that there is nothing w h i c h
is objectively a c c i d e n t a l " . Aristotle makes two points: that the fact o f
honey-water's being only sometimes beneficial to fever-patients is
compatible with there being certain conditions under w h i c h it is never
(or rarely or always or usually) s o ; and that he who wishes to learn must
seek to specify those conditions. Ross would a d d a third, that failure to
specify them must be due to subjective c a u s e s — h u m a n obtuseness—
never to their non-existence; but nothing i n the passage requires ш to
attribute this extra point to Aristotle.

CHAPTER 3

T h e distinction between conüng to be, or being destroyed, and being i n pp^ »11 i i
process of coming to be, or of being destroyed, is part of a more general
distinction between changes that are instantaneous and those that take
time (both are different from exercises or activities that last for a time

195
 ioa7*ag METAPHYSICS E3

but cannot stop without finishing; scc Θ 6. 1048b18-36). Although this

dbtinction admits of no intermediate degrees, its application is not
always clear. L e a r n i n g the alphabet is obviously a time4:0nsuming
change, finishing learning the alphabet a n instantaneous o n e ; but w h a t
ofmastering the alphabet? I s a m a n born, a n d does he die, over a period
or at a n instant? Aristotle's applications of the distinction i n E are two.
A t J ^ 2 . 1026b22-4 he says that the onset a n d cessation of a coincidental
state of affairs must be instantaneous. T h i s seems false. I t is a coincidence,
we are told {E 2. 1026^35-6), that a m a n is pale, yet he probably took
time to become so. Aristotle should have restricted himselfto the narrower
c l a i m , appropriate i n the immediate context i n E 2, that coincidental
identities have a n instantaneous onset; for although Coriscus c a n be i n
course of becoming pale it seems unnatural to say that he is i n course of
beconung the same as the pale Coriscus. Secondly, Aristotle now adds
that origins a n d causes c a n come to be a n d cease to be instantaneously,
pp, 222-5 T h e purpose ofE 3 is to prove this second contention, but its argument
is very obscure. W e m a y begin w i t h a point of some certainty: Aristotle
uses the notion of necessity i n such a w a y that it makes sense to say of a n
individual state ofaffairs that it becomes necessary or " i s not y e t " necessary
( ^ i o - i 1). I t follows that i n this chapter *necessary' is not always opposed
to *coincidental', a n d a state of affairs might be coincidentaUy necessary
— i . e . sometimes or i n some cases unavoidable; a n d w h e n Aristotle
accepts as a d a t u m that not everything is necessary ( ^ i o - i 1), he prob­
ably means that

(0) not everything is necessary aU the time.

Next we must determine Aristotle's view about the connection between
necessity a n d causation. I n the second paragraph he assumes that causes
precede their effects i n time; thus i f we trace the causal lineage of a
finitely future state ofaffairs " w e shall come to the present" (bi) or " t o
something that has come to b e " (^3-4). I n ^5-10 he seems to accept that
a n y future state of affairs whose causal lineage does i n this w a y stretch
back to the present or past is already necessary; at a n y rate, ifeverything
is such, " e v e r y t h i n g that wiU be w i U be of necessity" (cf. Rhetoric I I I
17. 1418*5 " f o r w h a t has come to be possesses necessity"). T h e foUowing
lines then concede that even a state of affairs w h i c h is " n o t y e t " necessary,
for instance the death of a m a n by disease, " r u n s as far as some o r i g i n " .
I f this origin (say, the eating of a toadstool) is not to make the death b y
disease already necessary, it must, seemingly, be i n the future. Since
Aristotle adds that the origin runs " n o further to anything e l s e " , i t
foUows that the whole causal lineage of the death b y disease is i n the
future, i.e. that that manner ofdeath has asyet no cause. T h u s w h a t is not
yet necessary has as yet no cause, a n d i n general
(b) w h a t has a cause at a n y timc is necessary at that timc.

196
 E3 NOTES 1027*29

Aristotle goes o n to say at b12-13 that the future *origin' (the e a t i n g o f

the toadstool) is " t h e origin ofwhatever m a y c h a n c e " , w h i c h appears to
contradict (b) by asserting that the death by disease wiU stUl be *chancc',
i.e. notnecessary, when it gets its origin or cause. T h e contradiction is
avoided if we take Aristotle to m e a n that the toadstool-eating wiU be the
origin o f w h a t is until then a chance outcome. F r o m {a) a n d (b) it foUows
that
{c) some things at some times have no cause.
Moreover, the future origin of a sometime chance state of affairs must
itself be uncaused at all times before its onset (otherwise it wiU not r u n
" n o further to anything e l s e " ) ; this is perhaps conveyed by ^13-14 " a n d
nothing else is the cause of its coming to b e " . H e n c e , by the assumption
that causes must precede their effects, it foUows that
(d) some causes at no time have any cause.
W e must now return to the first paragraph of the chapter. A t *31-2
Aristotle hypothesizes that anything whose coming to be takes time " h a s
some cause non-coincidentally". I t seems on the whole easiest i f w e take
this highly ambiguous phrase to m e a n
{ei) anything whose coming to be takes time has a cause at aU times
('non-coincidentar meaning *necessary' i n the sense *always', not
*unavoidably'). (^i) wiU imply that the causal lineage of a house, for
instance, goes back i n time without limit. T h e merit of this inter­
pretation is that {d) a n d (^i) immediately entail
(/) there are some causes whose coming to be does not take time
w h i c h , with the further assumptionthat these causes are not eternal (cf.
E I . 1026*16 a n d note), i n turn entails that their coming to be is instan­
taneous, the thesis o f the chapter. But the interpretation has its diffi­
culties. First, the qualification " n o n - c o i n c i d e n t a l l y " is redundant i f the
argument depends on {d), w h i c h says, i n effect, that some causes have not
even coincidental causes. If, on the other h a n d , we take Aristotle to have
relied on {c) rather than {d), that wouldjustify the conclusion that there
are some things whose coming to be does not take time, but not some
causes. T h e second difficulty is that it is unclear w h y Aristotle should have
thought (^i) true.
Alternatively, then, the ambiguous sentence at *31-2 might be taken
to m e a n
{e2) any X whose coming to be takes time has a cause of a sort w h i c h
always (or necessarily) causes things of ^*s sort.
or
(«3) . . . w h i c h always causes things of ^'s sort, or nothing.
O n e or other of these senses seems intended b y Ross's paraphrase

197
1027*29 METAPHYSIGS E3

"necessarily p r o d u c e s " ; but the passages w h i c h Ross cites do not support

the view that Aristotle believed either (^2) or (^3). Moreover, i n order to
m a k e either of them fit into the argument as a whole, w e should have to
r e a d the second p a r a g r a p h as distinguishing necessitating from n o n -
necessitating causes; whereas, as Ross admits, *32-^10 are most n a t u r ­
aUy taken as conunitting Aristotle to the opinion that aU causes necessitate,
i.e. to {b), T h e chapter h a s not yet received a satisfactory interpreta­
tion. (Other relevant passages are De Interpretatione 9, A " 8 . 1065*6-21,
De Partibus Animalium I 1. 640*1-9, De Generatione et Corruptione I I 11.
337*34-338*4.)
A t ^5, " o f necessity he wiU die or not d i e " is ambiguous. De Inter­
pretatione 9, w h i c h examines the relation between necessity a n d the
future i n far greater detail, employs a similar phrase explicitly w i t h the
force *necessarily {p or not-^') (19*30). B u t here the context demands
*(necessarily p) or (necessarily not-/>)': the present facts—compare the
past facts i n ^6-ιο—necessitate a future outcome, but which outcome
depends on w h i c h a r e the present facts. Although this is certainly
Aristotle's meaning, it is more t h a n he should have said, for even i f the
example shows that the m a n ' s eating something salty necessitates his
death by violence, it is far from showing that his not eating it necessitates
his not dying b y violence.

1027**14. T h e " r e d u c t i o n " leads to a n original cause w h i c h runs " n o

further to anything e l s e " . Aristotle's omission of formal causes suggests
that he regards still-avoidable states of affairs as examples ofcoincidental
beings, w h i c h have no form or essence i n the full sense {Z 4 ) . B u t the
general connection between this chapter a n d the last is obscure.

GHAPTER 4

1027*17. Aristotle's c l a i m that, taken together, a truth a n d a fakehood

are " c o n c e r n e d . . . w i t h the apportionment of a c o n t r a d i c t i o n " seems to
i m p l y that i f ^ a n d B are the members o f a contradictory p a i r , one must
be true a n d the other f a l s e — a thesis examined i n detail i n De Inter­
pretatione 7-11 a n d there found to have at least one exception. Г 7.
1011^26-7 indicates h o w the apportionment is made. T h e other haU*of
Aristotie's c l a i m about truth a n d fabehood, that they " a r e concerned
w i t h composition a n d d i v i s i o n " , is repeated at De Interpretatione 1.
16*9-18 i n the same words (translated b y A c k r i U *combination a n d
separation'); a n d there explained b y distinguishing true a n d fake
thoughts (e.g. the thought or belief that Gallias is pale) from thoughts
" t h a t are without composition a n d d i v i s i o n " (e.g. the thought or c o n ­
ception of GaUias). H e r e , however, it is not thoughts—or w o r d s — b u t

198
£4 NOTES 10a7bx7

things (e.g. CaUias a n d paUor) w h i c h are or are not compounded; cf.

Δ 29. 1024b18-19.
T h e problem raised i n the parenthesis is not how to distinguish
(a) a thought of pallor succeeded by a thought of Callias
from
{b) a thought of pale Callias,
but how to distinguish (a) from
(c) t h e thought that Callias is pale.
Aristotle's mode of expression invites a confusion between {b) a n d (c)
from w h i c h , perhaps, he was not h i n M e l f free. ( T h i s sentence shows that
the words rendered 'think' a n d 'conceive' i n the present translation do
not mark our distinction between belief a n d conception). Another
confusing f e a t u r e of the parenthesis is its twe of 'together' (the Greek
word is 'hama', elsewhere i n this translation rendered 'simultaneously')
in contrast with 'separately,' w h i c h makes it seem as i f o n e who thought
two things separately would have to think them i n succession. I n fact, of
c o u r s e , by thinking separately Aristotle here meaiw having a negative
belief, e.g.
{d) the thought that Callias is not pale,
w h i c h , like (c), contrasts w i t h (a). Aristotle fails to make it clear that he
has two contrasts on his hands, that between affirmative a n d negative
beUefs a n d that between beliefs a n d conceptions. E v e n negative beliefs
a n d statements " i n v o l v e c o m b i n a t i o n " {Categories 2. i'i6) i n the sense
that they connect thoughts a n d words i n a more unifying w a y than do
thought or spoken lists like ' m a n , pale' (cf. Categories 4. 2*8-10, 10.
13b10-11, Be Interpretatione 4. 16^26-7, Topics I 4. 101^23^); but they
" a r e concerned w i t h d i v i s i o n " i n the different sense that they declare
the things signified by those thoughts a n d words to be disjoined. T h e
difference between belief a n d conception is discussed i n De Anima I I I
chapters 2, 6, a n d 7.

1027^25. Aristotle's example shows w h y he says that falsehood and

truth a r e " n o t i n actual t h i n g s " . B u t are they " i n thought"? We might
agree that beliefs (and statements) are the p r i m a r y recipients of the
predicates 'true' a n d 'false', but i n Δ 29 Aristotle firmly describes as
falsehoods another k i n d o f " a c t u a l t h i n g " , v i z . states of affairs like C a U i a s '
being pale (a result w h i c h wUl foUow i f w e take 'CaUias' being pale' as
equivalent to 'that Callias is pale' a n d construe the latter as subject of
'it is false that Callias is p a l e ' ) . Nevertheless Aristotle ofTers one good
reason for dismissing truths and falsehoods as less " f u n d a m e n t a l " (con­
trast Θ 10. 1051^1) t h a n substances, qualities, a n d the members of
the other categories: CaUias' being pale is a state of affairs w h i c h

199
1027b25 METAPHYSICS E4

connects, a n d his not being pale a state of affairs w h i c h divides, two

simple items, Callias a n d p a l l o r ; the existence ofthe former is thus to be
explained i n terms ofthat of the latter, a n d truths a n d falsehoods " d o not
indicate the existence of a n y extra (sc. independent) nature of thing-
t h a t - i s " . " T h e cause of the other is a certain affection of t h o u g h t " :
Aristotle does not m e a n that beliefs cause facts (see Θ 10. 1051^6-9) but
that beliefs are the recipients (subject-matter) of truth. O n simples see
Θ 10.

1028*2. " T h a t w h i c h is itself": i.e. that w h i c h is i n its o w n right; cf.

Δ 7. 1017*22. "Омл thing-that-is" should be taken as modifying the verb
" i n v e s t i g a t e " , not as a gloss on " i t s e l f " ; see notes on Г 1.

1028*4. T h e sentence would be more appropriate as a n introduction

to the following book Z, w h i c h i n our M S S . begins w i t h similar words.
T h e reference is to Δ 7.

200
 F U R T H E R C O M M E N T S (1992)
Т н Е appearance o f a second e d i t i o n o f t h i s v o l u m e offers me an o p p o r t u n i t y
o f c o m m e n t i n g o n some o f the extensive l i t e r a t u r e a b o u t Aristotle's
metaphysics t h a t has been published i n recent years, a n d o f using t h a t
l i t e r a t u r e to d r a w together, a n d sometimes reassess, the views expressed
i n the Notes o n issues t h a t are p a r t i c u l a r l y pervasive or contentious i n
books Г, A, a n d E. I n some places I shall quote f r o m the Notes, i n order
to r e a f f i r m w h a t stiU seems to me i m p o r t a n t l y r i g h t or w i t h d r a w w h a t
seems i m p o r t a n t l y w r o n g , l o o k i n g back f r o m 1 9 9 2 to 1 9 7 1 . I n default
o f a completely new c o m m e n t a r y — w h i c h i n any case w o u l d have
come better f r o m another p e n — I hope t h a t new readers w i U find this
re-treatment h e l p f u l . N u m b e r s i n square brackets refer t o t h e B i b l i o g r a ­
phy.

ri,^i:METAPHYSICS

W h a t a c c o r d i n g to A r i s t o t l e is metaphysics? T h e question is ambiguous.

I t m i g h t m e a n , ' H o w d o e s — o r w o u l d — A r i s t o t l e characterize the i n q u i r ­
ies gathered i n the fourteen books w h i c h we k n o w u n d e r the ancient b u t
a p p a r e n t l y n o n - A r i s t o t e l i a n title Metaphysics?' O r i t m i g h t m e a n to ask
a b o u t A r i s t o t l e ' s c h a r a c t e r i z a t i o n o f w h a t he culls the " p r i m a r y [or: first]
p h i l o s o p h y " ( 1 0 2 6 ^ 2 4 ) , the " p r i m a r y d i s c i p l i n e " {episteme, 1 0 2 6 ^ 1 5 - 1 6 ,
2 9 ) or the " p r i m a r y science" (sophia, 1 0 0 5 ^ 1 ) . Perhaps the t w o questions
are the same: i f the collection o f o u r fourteen books i n t o a single treatise
was Aristotle's o w n w o r k , or was the w o r k o f editors f o l l o w i n g his wishes,
or at least w o u l d have been a p p r o v e d b y h i m , perhaps its r a t i o n a l e i s t h a t
a l l these books deal w i t h First Philosophy. B u t we have no statement o f
t h a t view, o n l y evidence for i t . T h e evidence is contained chiefly w i t h i n the
g r o u p o f b o o k s translated i n the present v o l u m e , a n d is as foUows. (i) E 1.
1 0 2 6 ^ 1 8 - 2 9 claims t h a t First Philosophy is T h e o l o g y , t h a t is, the study o f
changeless substances, o f w h i c h " t h e d i v i n e is a c o n s t i t u e n t " , (ii) Book Г
begins b y a r g u i n g for the existence o f " a discipline w h i c h studies t h a t w h i c h
is qua thing-that-is [to on hei on, t r a d i t i o n a l l y 'being qua b e i n g ' ] a n d those
things t h a t h o l d good o f t h i s i n its o w n r i g h t " ( 1 0 0 3 ^ 2 1 - 2 ) . ( i i i ) T h e l a t t e r
p a r t o f b o o k Г, w h i c h investigates the principles o f n o n - c o n t r a d i c t i o n a n d
excluded m i d d l e , m i g h t be expected to be an exercise i n this discipline, a n d
is so represented i n the N o t e o n 1005^8. (iv) T h e final half-sentence o f . E 1
makes T h e o l o g y embrace the discipline whose existence was defended i n Г:
" a n d i t w o u l d fall to [the study ofchangeless substance] to study t h a t w h i c h
is qua t h i n g - t h a t - i s , b o t h w h a t i t is a n d the things t h a t h o l d good o f i t qua
thing-that-is" (1026^31-2).

201
 METAPHYSICS

T h i s evidence bears o n t w o questions: w o u l d A r i s t o t l e w i s h to b r i n g a l l

the fourteen books u n d e r his c o n c e p t i o n o f T h e o l o g y ; a n d i f s o , w o u l d he
be r i g h t to d o so? O n b o t h questions the final half-sentence o f E i is
c r u c i a l . I t c o u l d be a n i n t e r p o l a t i o n , i n t r o d u c e d b y a n e d i t o r i n o r d e r to
m a k e a c o n n n e x i o n n o t elsewhere e x p l i c i t i n Aristotle's text. I f so, the
e d i t o r was i n t e l l i g e n t ; for Aristotle's i m m e d i a t e l y p r e c e d i n g r e m a r k is
t h a t T h e o l o g y is " u n i v e r s a l i n this w a y , because p r i m a r y " ( 1 0 2 6 ^ 3 0 - 1 ,
c f Г 3 . 1 0 0 5 ^ 3 5 ) , a n d the " d i s c i p l i n e " o f T h a d been suspect, a n d so i n
need o f defence, precisely because o f its c l a i m to a c e r t a i n sort o f
u n i v e r s a l i t y . A s s u m i n g t h e n t h a t the half-sentence is n o t a n i n t e r p o l a t i o n ,
A r i s t o t l e sees the science described i n b o o k Г as faUing u n d e r T h e o l o g y .
H o w m u c h m a t t e r f r o m the Metaphysics comes i n t o T h e o l o g y u n d e r this
r u b r i c ? ( V e r y l i t t l e comes i n w i t h o u t i t , perhaps o n l y Л 6-10 o n ' u n ­
m o v e d movers', i.e. changeless changers.) T h e i n t r o d u c t o r y N o t e o n p .
122 argues t h a t Δ earns its admission to the " d i s c i p l i n e " defined i n Г 1.
A s i m i l a r case c o u l d be m a d e , I t h i n k , for most or aU o f the others o f o u r
fourteen books, a n d the Notes presume s o m e t h i n g like this case b y
d e n o m i n a t i n g the " d i s c i p l i n e " metaphysics.
C o n t i n u i n g the assumption, is A r i s t o t l e r i g h t to assert the half-
sentence—to assert, t h a t is, t h a t T h e o l o g y , the seemingly d e p a r t m e n t a l
study o f changeless substances, embraces the universal study envisaged
b y the o p e n i n g sentence o f Г 1, o f " t h a t w h i c h is qm t h i n g - t h a t - i s " ? A t
this p o i n t the curious '^Ma'-phrase needs i n t e r p r e t a t i o n . O n e defence o f
A r i s t o t l e ( [ 3 4 ] ) begins b y i n t e r p r e t i n g i t i n a w a y — a K a n t i a n w a y —
w h i c h makes the envisaged study d e p a r t m e n t a l too: t h a t w h i c h is qm
t h i n g - t h a t - i s , ' b e i n g qua b e i n g ' , is a c h u n k o r genus o f w h a t there is, a
c h u n k w h i c h E 1 w i U later i n v i t e us to i d e n t i f y w i t h changeless substances.
T h e defence t h e n proceeds b y a r g u i n g t h a t a k i n d o f u n i v e r s a l i t y is
conferred o n the study o f t h a t genus b y the r e l a t i o n s h i p o f its members to
e v e r y t h i n g else ( [ 3 4 ] ) , o r o f the study o f i t to every other study ( [ 7 ] ,
[ 4 9 ] ) . T h e a u t h o r oiMetaphysics K, w h o m a y or m a y n o t be A r i s t o t l e
himself, seems t o j o i n this p a r t y w h e n i n his e p i t o m e οΐΕ i he paraphrases
the question " w h e t h e r the p r i m a r y p h i l o s o p h y reaUy is u n i v e r s a l "
( 1 0 2 6 ^ 2 3 - 4 ) i n the words " w h e t h e r one reaUy should c o u n t the discipline
concerned w i t h [ l i t e r a l l y : o f ] t h a t w h i c h is qua t h i n g - t h a t - i s as universal
o r n o t " {K 7. 1 0 6 4 ^ 7 - 8 ) . A g a i n s t these p r o p o s a l s t h e r e are n o w p o w e r f u l
objections i n T h o r p [ 5 1 ] , w h o rejects t h e i r justifications o f ' u n i v e r s a l '
(pp. 1 1 3 - 2 1 ) , a n d S t e v e n s o n [ 4 6 ] (see also the Notes o n p p . 7 7 - 8 ) , w h o
shows t h a t they m i s u n d e r s t a n d the syntax o f the '^Mo'-phrase, w h i c h
restricts the k i n d o f s t u d y , n o t the subject studied. A different suggestion,
close to [ 4 4 ] a n d [ 5 0 ] , is s u p p o r t e d i n the N o t e o n 1026^23: t h a t the
p r i m a r y substances, w h a t e v e r they are—changeless i n Aristotle's o p i n i o n ,
b u t the same w o u l d be t r u e i f they were the changeable things o f
' n a t u r e ' , so t h a t " t h e discipline concerned w i t h n a t u r e " , i.e. physics,

202
 FURTHER COMMENTS

" w o u l d be p r i m a r y " , ( 1 0 2 6 ^ 2 8 - 9 ) — t h e s e p r i m a r y substances explain

e v e r y t h i n g , a n d a c c o r d i n g l y u n d e r s t a n d i n g {episteme) o f t h e m delivers
u n d e r s t a n d i n g o f e v e r y t h i n g . Such a project o f e x p l a i n i n g e v e r y t h i n g b y
reference to basic existents is w h a t the N o t e o n 1026^23 calls o n t o l o g y .
T h e w o r d ' o n t o l o g y ' has other senses as w e l l (see [ 4 9 ] ) ; b u t even i n the
Note's sense i t contains a n i n d e t e r m i n a c y , for u n d e r s t a n d i n g a n d explana­
t i o n o f a t h i n g can be o f what the t h i n g is, its n a t u r e , o r o f why i t exists, its
cause. B o t h o f these come u n d e r Aristotle's conception, broader t h a n
ours, o f cause; a n d w h e n he says at 1 0 0 3 ^ 1 6 - 1 7 t h a t the " p r i m a r y
[ o b j e c t ] " o f a discipline is " t h a t o n w h i c h the others depend a n d to
w h i c h they owe t h e i r b e i n g [ l i t e r a l l y : t h a t because o f w h i c h they are]
called [ w h a t they a r e ] " , he appears to leave i t open w h i c h o f t h e s e t w o
kinds o f e x p l a n a t i o n is to be d e r i v e d f r o m the study o f p r i m a r y sub­
stances.
T h e suggestion is, t h e n , t h a t A r i s t o t l e identifies T h e o l o g y w i t h the
" d i s c i p l i n e " o f Г 1 b y i d e n t i f y i n g each o f t h e m w i t h ontology, i n the
i n d e t e r m i n a t e sense o f ' u n d e r s t a n d i n g e v e r y t h i n g b y reference to basic
existents'. I f t h i s is Aristotle's manoeuvre, however, i t can be criticized i n
the w a y suggested i n the N o t e , o n the g r o u n d t h a t neither i d e n t i f i c a t i o n
is j u s t i f i e d . O n the one h a n d T h e o l o g y w i l l embrace some questions—as
i n Л 6 - 1 0 — w h i c h are n o t o n t o l o g i c a l . O n the other h a n d m u c h o f
Aristotle's ' u n i v e r s a l ' i n q u i r y i n the Metaphysics, for example his e x a m i n a ­
t i o n o f the l o g i c a l principles i n Г 3 - 8 , a n d the conceptual analysis o f Δ
a n d / i f t h o s e books c o u n t u n d e r J " s r u b r i c , is n o t o n t o l o g i c a l either.

Г4: C O N T R A D I C T I O N

As p a r t o f the study envisaged i n Г 1 " w e also have to find the first

causes o f t h a t w h i c h is qua t h i n g - t h a t - i s " ( 1 0 0 3 ^ 3 1 - 2 ) . These first causes
are o r i n c l u d e archai, w h i c h the T r a n s l a t i o n gives sometimes as O r i g i n s ' ,
sometimes as 'principles'. T h e firmest p r i n c i p l e o f a l l is: " f o r the same
t h i n g to h o l d good a n d n o t to h o l d good simultaneously o f the same
t h i n g a n d i n the same respect is impossible . . . " ( 1 0 0 5 ^ 1 9 - 2 2 ) . T h e
Notes call this the p r i n c i p l e o f n o n - c o n t r a d i c t i o n ( P N C ) ; Aristotle's o w n
w o r d 'antiphasis\ m e a n i n g ' c o n t r a d i c t o r y p a i r ' , is used i n the f o r m u l a t i o n
at 1 0 1 i ^ i 6 . A r i s t o t l e argues t h a t such a p r i n c i p l e cannot be demonstrated
(see [ 8 5 ] p. 2 1 n . 1 ) ; b u t i t can be " d e m o n s t r a t e d i n the m a n n e r o f a
r e f u t a t i o n " ( 1 0 0 6 ^ 1 0 - 1 2 ) against a n o p p o n e n t w h o is w i l l i n g to start
discussion g o i n g . T h e m e t h o d is to find something w h i c h shall be
d e m a n d e d , o r 'begged', f r o m the o p p o n e n t as his i n i t i a l c o n t r i b u t i o n
(not to ask the o p p o n e n t to "choose the premiss", as the Notes say o n
p. 9 1 ) . I n a l l such dialectic i t is a f a u l t to beg w h a t is " o r i g i n a l l y at issue"
( t r a d i t i o n a l l y translated 'begging the question'; b u t n o t aU commentators

203
 METAPHYSICS

notice t h a t b e g g i n g the question is n o t the same as assuming w h a t is at

issue—Aristotle's opponent can assume w h a t is at issue b u t c a n n o t beg the
question, for he begs n o t h i n g ) . I n the peculiar circumstances o f the
p e r m i t t e d m a n n e r o f d e m o n s t r a t i n g P N C the o n l y safe w a y o f a v o i d i n g
t h a t f a u l t , A r i s t o t l e t h i n k s , is to beg no m o r e t h a n t h a t the o p p o n e n t
"signify s o m e t h i n g b o t h to h i m s e l f a n d to someone else" ( 1 0 0 6 ^ 2 1 - 2 ) ; so
m u c h can p r o p e r l y be begged, since w i t h o u t signifying there w i l l be no
saying or statement {logos^ 1 0 0 6 ^ 2 2 - 3 ) . F r o m this s t a r t i n g p o i n t i t w o u l d
seem t h a t A r i s t o t l e proposes to argue transcendentally, d e m o n s t r a t i n g
n o t the t r u t h o f P N C , b u t t h a t w i t h o u t its t r u t h the o p p o n e n t c o u l d n o t
be d o i n g w h a t he is d o i n g i n acceding to w h a t is begged o f h i m ; a n d this
m e t h o d is perhaps m o r e n a t u r a l l y called a r e f u t a t i o n t h a n a demonstra­
t i o n , as A r i s t o t l e w i s h e s — a t any rate i t is n o t a d e m o n s t r a t i o n of PNC
(for other ways o f u n d e r s t a n d i n g ' i n the m a n n e r o f a r e f u t a t i o n ' see [ 8 5 ]
P· 7 5 , [ 3 6 ] p. 1 4 4 , [ 8 7 ] n . 2 5 ) .
T h e Posterior Analytics tells us t h a t o n l y w h a t can be d e m o n s t r a t e d is
episteton^ i.e. can be u n d e r s t o o d , i.e. is p a r t o f a " d i s c i p l i n e " or science
( 7 1 ^ 1 6 - 1 9 w i t h i o o ^ i o - i i ) , a n d this raises the question: i f P N C c a n n o t
be demonstrated, w h a t are these chapters a b o u t i t d o i n g i n a treatise o n
the " d i s c i p l i n e " o f metaphysics? T h e r e are three m a i n possibilities, (i)
Perhaps b y the t i m e he w r o t e Metaphysics Г A r i s t o t l e h a d a b a n d o n e d his
earlier v i e w t h a t o n l y w h a t can be d e m o n s t r a t e d is episteton, a n d h a d
come to believe t h a t ' o b j e c t i v i t y ' is also a t t a i n a b l e b y a k i n d o f dialectic.
T h i s is the thesis o f [ 3 7 ] a n d the u n i f y i n g theme o f [ 3 8 ] . (ii) A r i s t o t l e
h a d n o t a b a n d o n e d his v i e w t h a t sciences m u s t proceed b y demonstra­
t i o n ; so since the principles o f n o n - c o n t r a d i c t i o n a n d excluded m i d d l e are
i n d e m o n s t r a b l e , Aristotle's t r e a t m e n t o f t h e m i n Г 3 - 8 is n o t a n exercise
i n the science o f metaphysics as the N o t e o n 1005^8 asserts, b u t a
p r o l e g o m e n o n o r a digression, ( i i i ) Г 3 - 8 a i m at d e m o n s t r a t i o n , b u t
w h a t they a i m to demonstrate is theses about the principles o f n o n ­
c o n t r a d i c t i o n a n d excluded m i d d l e , n o t those i n d e m o n s t r a b l e principles
themselves; see [ 3 9 ] . I t w i l l be clear t h a t the preceding p a r a g r a p h adopts
(iii) as solving the p r o b l e m ; b u t t w o f u r t h e r comments are i n order.
First, this solution w o u l d n o t prevent (i) a n d the conclusion o f (ii) f r o m
b e i n g t r u e also. Secondly, i t is i m p o r t a n t to notice t h a t Г 3 - 8 are
exceptional w i t h i n the Metaphysics—indeed w i t h i n the A r i s t o t e l i a n
c o r p u s — i n a i m i n g at d e m o n s t r a t i o n , or at least sketching a m e t h o d o f
d e m o n s t r a t i o n . Aristotle's usual procedure, w h i c h is n o t to offer d e m o n ­
strations b u t r a t h e r to i n v i t e his audience to j o i n h i m i n i n q u i r y , is q u i t e
c o m p a t i b l e w i t h his b e l i e v i n g t h a t the results o f successful p h i l o s o p h i c a l
i n q u i r y can be set o u t i n demonstrations, so q u a l i f y i n g as sciences even
a c c o r d i n g to the canons o f t h e Posterior Analytics.
Aristotle's attempts i n Г 4 - 6 to a p p l y his procedure against the
opponents o f P N C c o n t a i n m u c h t h a t is d i f f i c u l t , a n d I shall n o t ofFer

204
 FURTHER COMMENTS
any general survey o f t h e m here. I propose o n l y to select t w o p a r t i c u l a r
passages for f u r t h e r c o m m e n t , n a m e l y (according to the w a y the chapter
is d i v i d e d u p i n the Notes) parts I a n d И1 o f the second a r g u m e n t i n
Г4.

2nd argument, Part I (1006^31—^34). E v e n the structure o f t h i s passage is

unclear, b u t here is a possible o u t l i n e o f the stages i n i t b y w h i c h
A r i s t o t l e hopes to b r i n g his o p p o n e n t to give w a y o n P N C . ( 1 ) I f y o u
(use a w o r d to) signify, y o u signify a d e t e r m i n a t e n u m b e r o f t h i n g s ; ( 2 )
each t h i n g y o u signify (e.g. m a n ) is the being o f certain things ( m e n ) ; ( 3 )
i f y o u signify something w h i c h is the being o f certain things, they m,ust be
as i n the f o r m u l a o f t h a t something ( m e n must be b i p e d animals); ( 4 ) i f
so, they cannot also n o t be such (men cannot also n o t be b i p e d animals);
therefore ( 5 ) i f y o u signify, there is something y o u signify ( m a n ) a n d
some things o f w h i c h t h a t is the being (some men) such t h a t i t cannot be
true to say t h a t the l a t t e r <are a n d are n o t as i n the f o r m u l a o f the
former, n o r therefore t h a t they> are a n d are n o t the former.
I wish to make six comments s u p p l e m e n t i n g the N o t e o n p p . 9 3 - 9 . (i)
I f the a r g u m e n t works, i t shows t h a t signifying requires the t r u t h n o t o f
P N C b u t o n l y o f a n instance o f i t . T h r e e verdicts are possible: this is a
f a u l t ( [ 8 i ] ) ; or, P N C can be reached f r o m the instance b y universal
generalization ( [ 8 5 ] p p . 1 1 2 - 1 4 ) ; or, Aristotle's project is n o t to reach
P N C , b u t o n l y to defeat a n o p p o n e n t w h o m a i n t a i n s t h a t all contradic­
tions are t r u e , or at least m a i n t a i n s {b) o n p. 102 o f t h e Notes. T h e last is
w i d e l y i m p l i e d elsewhere i n Г 4 ( 1 0 0 6 ^ 3 0 - 1 , 1007^18, 1008^36, ^ i o - i i ,
^ 3 1 - 2 ) a n d a p p a r e n t l y defended as adequate at 1 0 0 8 ^ 7 - 1 2 (see [ 8 5 ] p p .
5 9 - 6 1 ) . I f Aristotle's project is this l i m i t e d one, the " m a j o r a t t r a c t i o n o f
type I i n t e r p r e t a t i o n s " (Notes p. 9 3 ) falls away.
(ii) Code ( [ 3 6 ] p. 1 4 4 , f o l l o w i n g [ 8 7 ] p. 1 0 4 ) says t h a t "Aristotle's
i n t e n t is to show t h a t adherence to the P N C is a prerequisite for
significant t h o u g h t a n d discourse", a n d thereby to e x p l a i n w h y every­
b o d y does adhere to i t ( [ 3 9 ] p. 3 5 6 ) ; b u t i f t h e a r g u m e n t is transcendental
as I have suggested, 'adherence t o ' should be replaced b y 'the t r u t h of'.
(iii) I n o w have no d o u b t t h a t Aristotle's 'signify', 'semainein% does n o t
m e a n ' m e a n ' or 'be explicated b y ' b u t expresses a r e l a t i o n w h i c h words
bear to things, better translated 'signify' or ' i n d i c a t e ' t h a n 'denote' (p.
9 4 ) ; see [ 9 0 ] , a n d [ 6 4 ] p p . 1 8 6 - 9 0 w h i c h shows h o w the type 2 i n t e r p r e t a ­
t i o n is n o t k i l l e d b y this change.
(iv) T h e T r a n s l a t i o n o f 1006^32—4 has: " W h a t I mean b y 'signifying
one t h i n g ' is this: i f t h a t t h i n g is a m a n , then i f a n y t h i n g is a m a n , t h a t
t h i n g w i l l be to be a m a n . " A m o n g the interpretations o f this t h a t are
canvassed i n the N o t e o n p. 9 4 the most l i k e l y is ' i f the one t h i n g t h a t
" m a n " signifies is [ a ] m a n , t h e n , i f a n y t h i n g is a m a n , the one t h i n g t h a t
" m a n " signifies is to be a m a n ' . I f t h a t is the r i g h t i n t e r p r e t a t i o n , t h e n

205
 METAPHYSICS
A r i s t o t l e identifies the one signification o f ' m a n ' i n t w o different ways, as
man, a n d as to be a man; a n d a c c o r d i n g l y these t w o are the same as one
another: m a n is to be a m a n . B u t n o w we face t w o problems n o t
satisfactorily addressed i n the N o t e : (i) w h a t c o u l d be m e a n t b y ' m a n is
to be a man'? a n d ( i i ) w h a t is the purpose o f the second c o n d i t i o n i n the
sentence, " i f a n y t h i n g is a m a n " ? These problems can be solved together
i f t h e t r a n s l a t i o n o f t h e last f o u r w o r d s i n the sentence is revised f r o m " t o
be a m a n " i n t o 'for a m a n to be'. I n [ 7 6 ] this l a t t e r is u r g e d as the
correct r e n d e r i n g over a range o f A r i s t o t e l i a n constructions i n c l u d i n g the
one here, 'to anthröpöi einai'; a n d a l t h o u g h I a m n o t convinced t h a t
A r i s t o t l e separated the t w o alternatives i n his o w n m i n d , i t makes better
sense here. A c c o r d i n g t o the revision, the t w o things i d e n t i f i e d as the sole
signification o f ' m a n ' , a n d therefore i n effect i d e n t i f i e d w i t h each other,
are man a n d , for a m a n , to be; a n d t h a t is to say t h a t , for m e n , t h e i r being
is or consists i n t h e i r being men. " I f a n y t h i n g is a m a n " n o w emphasizes
t h a t this account oibeing applies to e v e r y t h i n g t h a t is a m a n .
(v) 1 0 0 6 ^ 1 3 - 2 8 , i g n o r e d i n the above o u t l i n e , argues f r o m " ' m a n "
signifies one t h i n g ' to " ' m a n " a n d " n o t - m a n " d o n o t signify the same'.
( D ) o n p p . 9 8 - 9 comments i n a d e q u a t e l y o n the purpose o f this section,
w h i c h I n o w t h i n k replies to a n unstated objection o f the f o r m i n d i c a t e d
b y [ 1 0 ] sec. 6 2 1 ( q u o t e d i n [ 8 6 ] p. 1 6 9 ) a n d [ 8 5 ] p p . 5 0 - 1 .
(vi) T h e f o r m u l a t i o n above o f (4) preserves the scope a m b i g u i t y
discussed i n ( A ) o n p. 9 8 . H e r e A r i s t o t l e applies the r u l e 'is necessarily φ'
=> 'can't n o t be φ' a n d possibly also the r u l e ' c a n ' t n o t be φ' => ' c a n ' t be
a n d n o t be φ' ( I use '<^' a n d 'φ' h e n c e f o r w a r d as schematic letters open to
replacement b y c o m m o n nouns, w i t h or w i t h o u t a n i n d e f i n i t e article i n
E n g l i s h , a n d adjectives). B o t h moves raise the suspicion t h a t he has
overlooked the fact t h a t inference rules, as w e l l as s t a r t i n g points, can be
i m p r o p e r l y begged i n a d e m o n s t r a t i o n .

2ud argument, Part I I I ( 1 0 0 7 ^ 2 0 - ^ 1 8 ) . T h e N o t e o n 1007^20 (p. 1 0 0 )

adopts a n i n t e r p r e t a t i o n w h i c h n o w seems to me mistaken o f Aristotle's
d i s t i n c t i o n between w h a t i t calls essential p r e d i c a t i o n as i n 'Callias is a
m a n ' a n d c o i n c i d e n t a l p r e d i c a t i o n as i n 'Callias is pale'. Aristotle's
conclusion i n P a r t I I I is t h a t the f o r m e r " k i n d o f f o r m u l a " (^30) w i U n o t
be available to opponents w h o " p r e d i c a t e c o n t r a d i c t o r i e s " ( ^ i 8 ) . F o r
" i t " , sc. ' m a n ' or 'is a m a n ' , signifies, for Callias, t h a t his b e i n g is
" n o t h i n g else" (^27), sc. t h a n b e i n g a m a n ; yet " i t s " denials, sc. 'is a n o t -
m a n ' a n d 'is n o t a m a n ' , w o u l d signify t h a t his b e i n g is something else.
So the a r g u m e n t is: ( i ) each o f ' m a n ' a n d ' n o t - m a n ' gives a complete
account o f t h e b e i n g o f w h a t e v e r they a p p l y to; ( 2 ) they give different
accounts; therefore they c a n n o t a p p l y to the same things. ( 2 ) appears to
rely o n 1006^13, w h i c h should n o w , perhaps, be translated " i t is n o t
possible t h a t ' t o be' for a m a n should signify j u s t w h a t ' n o t to be'

206
 FURTHER COMMENTS

[signifies] for a m a n " . B u t w h y ( i ) ? W h y are essential predications

exclusive? T h e N o t e seeks i l l u m i n a t i o n f r o m Posterior Analytics I 2 2 .
8 3 ^ 2 4 - 3 2 , w h i c h ends: " n o t h i n g can be pale w h i c h is n o t a c e r t a i n other
t h i n g t h a t is p a l e " , me einai ti leukon ho ouch heteron ti on leukon estin. B y
glossing this " w h a t is pale c a n n o t be i d e n t i c a l w i t h the pale t h a t i t i s " ,
i.e. subjects o f t r u e c o i n c i d e n t a l p r e d i c a t i o n are different f r o m t h e i r
predicates, the N o t e arrives at the d o c t r i n e t h a t " a p a r t i c u l a r m a n . . . is
i d e n t i c a l w i t h m a n " , i.e. essential predications are statements o f i d e n t i t y .
I n t h a t w a y ( i ) is sustained, b u t at the price o f a " d u b i o u s theory o f
p r e d i c a t i o n " (p. 1 0 0 ) .
T h e dubious theory was ascribed to A r i s t o t l e b y O w e n [ 5 4 ] as p a r t o f a n
e x p l a n a t i o n o f A r i s t o t l e ' s m a n n e r o f e v a d i n g the T h i r d M a n objection to
Plato's T h e o r y o f F o r m s . I n fact h o w e v e r ( t h o u g h I c a n n o t argue i t here),
the evasion does n o t r e q u i r e t h a t the f o r m m a n should be i d e n t i c a l w i t h
p a r t i c u l a r m e n such as Callias, b u t o n l y t h a t b e i n g a m a n should a p p l y to
p a r t i c u l a r m e n d i r e c t l y a n d n o t b y v i r t u e of, or t h r o u g h , its a p p l y i n g to
s o m e t h i n g else. T h e N o t e o n p . 1 6 8 proposes this l a t t e r as the first sense w h i c h
A r i s t o t l e gives i n Δ 18 to ' i n its o w n r i g h t ' (or ' b y v i r t u e o f i t s e l f ) : a c c o r d i n g
to this i n t e r p r e t a t i o n Callias is pale n o t i n his o w n r i g h t b u t t h r o u g h having
something, the pale i n h i m , t h a t is pale; a n d o n l y the l a t t e r is pale i n its o w n
r i g h t . T h e d i s t i n c t i o n e m p l o y e d here between the t w o relations oihaving a n d
beingis h e l p f u l l y associated b y recent c o m m e n t a t o r s ( [ 6 4 ] p p . 1 8 2 - 3 , [ 7 5 ] p.
1 0 3 ) w i t h t w o converse relations, being in a n d being said of, distinguished b y
A r i s t o t l e at Categories 2. 1^20-^9 a n d a p p l i e d at 5. 2 ^ 1 9 - 3 4 : i n t h a t
t e r m i n o l o g y , pale is n o t saidofC2L\\i?LS, b u t is said of something t h a t is in h i m .
Such is the d o c t r i n e ascribed to A r i s t o t l e ( a n d criticized) i n the Notes to A 18,
I t is possible to i n t e r p r e t the sentence already cited f r o m the Posterior
Analytics as b e a r i n g a s i m i l a r , b u t u n f o r t u n a t e l y n o t the same, sense.
First, we must take i t as i m p o r t i n g the general idea t h a t a p a r t i c u l a r
such as Callias is pale i n d i r e c t l y or d e r i v a t i v e l y , through something: this
even creeps i n t o [ i 6 ] ' s t r a n s l a t i o n o f the sentence: " t h e r e c a n n o t be
a n y t h i n g <pale> w h i c h is n o t <pale> t h r o u g h being something differ­
e n t " , sc. f r o m pale. Secondly, we should u n d e r s t a n d ' [ i s ] different f r o m
pale' as ascribing s o m e t h i n g other t h a n pale r a t h e r t h a n , as i n the Notes,
ascribing otherness-than-pale. T h i s f u r t h e r 'something' w i l l doubtless be
a substantial f o r m such as m a n , w h i c h a c c o r d i n g to the language o f the
Categories is suitable to be said of Callias. P u t t i n g the t w o results together
we get: pale is n o t said of Callias, b u t is in something t h a t is said o / h i m .
F r o m aU this three r i v a l accounts emerge o f the d i s t i n c t i o n between
w h a t the Notes call essential a n d c o i n c i d e n t a l ( t r a d i t i o n a l l y 'accidental')
p r e d i c a t i o n , t h a t is, between t w o ways i n w h i c h one i t e m can be ( t r u l y )
predicated of, or holds g o o d of, a n o t h e r i t e m or itself A c c o r d i n g to the
first o f these accounts, w h e n pale holds good o f CaUias i t holds good o f
h i m i n the second w a y ( t h a t is, coincides i n h i m ) because

207
 METAPHYSICS

{a) Callias is pale b u t is different f r o m t h a t pale.

A c c o r d i n g t o the o t h e r t w o , the e x p l a n a t i o n is t h a t he is pale n o t i n his
o w n r i g h t , w h i c h m a y m e a n either t h a t
{b) Callias is pale t h r o u g h having s o m e t h i n g (his pale) t h a t is pale,
or t h a t
(c) Callias is pale t h r o u g h being s o m e t h i n g (a m a n ) t h a t has pale.

(a), {b), a n d (c), therefore, are accounts o f w h a t A r i s t o t l e means b y 'pale

coincides i n Callias'. {a) n o w seems to me i m p l a u s i b l e . Between {b) a n d
{c) we need n o t choose at present; I r e t u r n to t h e m later. N o t i c e too t h a t
{b) a n d {c) c o u l d be c o m b i n e d , g i v i n g 'Callias is pale t h r o u g h being
s o m e t h i n g (a m a n ) t h a t has s o m e t h i n g (the pale) t h a t is pale'.
T h e r e is, however, a price to be p a i d for this d e p a r t u r e f r o m the
Notes. T h e Notes i n v o k e account {a), the " d u b i o u s t h e o r y o f p r e d i c a ­
t i o n " , for the purpose o f d e f e n d i n g ( i ) i n P a r t I I I o f A r i s t o t l e ' s second
a r g u m e n t i n defence o f P N C , viz. 'each o f " m a n " a n d " n o t - m a n " gives a
complete account o f t h e b e i n g o f w h a t e v e r they a p p l y t o ' . I t is n o t clear
to me h o w ( i ) can be sustained o n either o f the o t h e r t w o accounts,
a l t h o u g h W e i d e m a n n i n [ 7 6 ] t h i n k s t h a t i t — h i s " f i r s t premiss" o n p . 7 9 —
c a n be.
( T h e labels 'essential p r e d i c a t i o n ' a n d ' a c c i d e n t a l (or c o i n c i d e n t a l )
p r e d i c a t i o n ' come f r o m the t r a d i t i o n , b u t A r i s t o t l e h i m s e l f usually means
s o m e t h i n g different b y 'predicate c o i n c i d e n t a l l y ' . I n the t r a d i t i o n , as we
have seen, a p r e d i c a t i o n is c o i n c i d e n t a l w h e n its predicate coincides i n its
subject, t h e r e b y a f f o r d i n g the subject o n l y a d e r i v a t i v e t i t l e to be called by
thatpredicate. I n A r i s t o t l e , a p r e d i c a t i o n is c o i n c i d e n t a l w h e n i t has o n l y a
d e r i v a t i v e t i t l e to be called a predication; see Prior Analytics I 2 7 . 4 3 ^ 3 3 - 6 ,
PosteriorAnalytics I 2 2 . 8 3 ^ 1 4 - 1 7 , [ 1 6 ] p p . 1 1 6 - 1 7 . T h e same p r e d i c a t i o n
m i g h t ofcourse be c o i n c i d e n t a l i n b o t h senses.)

Z l 6 , 9 ; ^ 2 : I D E N T I T Y

I shall n e x t look at Aristotle's views o n i d e n t i t y , w h i c h , a l t h o u g h they

are n o t v e r y pervasive i n books Г, A, a n d E o f the Metaphysics, have
a t t r a c t e d c o m m e n t i n the recent l i t e r a t u r e t h a t has spread o u t , as we
shall see, over n e i g h b o u r i n g relevant topics. I centre m y discussion o n a
question t o w h i c h I shall e v e n t u a l l y give the answer ' N o ' . I t is:
D i d A r i s t o t l e reject the L a w o f the I n d i s c e r n i b i l i t y o f Identicals? L e t
us h e n c e f o r w a r d use 'a', 'b\ etc. as schematic letters open t o replacement
b y a n y o f t h e types ofexpression t h a t A r i s t o t l e , i n his relaxed w a y , allows
h i m s e l f a s g r a m m a t i c a l subjects o f u n q u a n t i f i e d sentences, e.g. 'Coriscus',
' m a n ' , 'the m a n ' , 'the pale', 'the pale Coriscus', 'pale m a n ' , a n d so o n .

208
 FURTHER COMMENTS
T h e n the L a w states t h a t i f a a n d b are i d e n t i c a l , e v e r y t h i n g t r u e o f a is
t r u e o f b—they share the same properties a n d history. I t m u s t be
distinguished f r o m the S u b s t i t u t i v i t y L a w , t h a t i f t w o expressions 'a' a n d
'b' refer to the same t h i n g , either can be replaced b y the other w i t h o u t
affecting t r u t h value, salva veritate. T h e S u b s t i t u t i v i t y L a w makes a c l a i m
a b o u t language, w h i c h m i g h t be t r u e o f one language b u t n o t o f
another. As stated, i t is n o t t r u e o f E n g l i s h or Greek; see [ 1 0 5 ] .
B u t the L a w o f the I n d i s c e r n i b i l i t y o f Identicals, w h i c h f o l l o w i n g
c u r r e n t p h i l o s o p h i c a l usage I shall c a l l Leibniz's L a w ( L e i b n i z himself
m a y r a t h e r have i n t e n d e d some q u a l i f i e d version o f S u b s t i t u t i v i t y ) is n o t
a b o u t expressions o f a language b u t a b o u t i d e n t i t y . I t has generally been
regarded as a n a l y t i c , constitutive o f the concept o f i d e n t i t y (thus [ i o i ]
a n d [ 1 0 4 ] see the question w h e t h e r A r i s t o t l e accepts Leibniz's L a w as
the question w h e t h e r he 'has the concept o f i d e n t i t y ' ) . I n its f o r m u l a t i o n
the w o r d ' i d e n t i c a l ' c o u l d be replaced b y 'the same'; for i t is a n i l l u s i o n
to suppose t h a t s t a n d a r d E n g l i s h , even scholarly English, firmly d i s t i n ­
guishes these L a t i n a t e a n d A n g l o - S a x o n adjectives—certainly the earliest
w r i t e r s o f p h i l o s o p h i c a l E n g l i s h , slowly l e a r n i n g n o t to t h i n k i n L a t i n ,
chose either expression indifferently. H o w e v e r , i t is w e l l k n o w n t h a t care
o f a different k i n d is needed i n i n t e r p r e t i n g , or a p p l y i n g , the L a w . W e
are n o t m e a n t to suppose t h a t i f a a n d b are, for example, the same age,
or c o m p o s i t i o n — i . e . i d e n t i c a l i n age or c o m p o s i t i o n — t h e n they must
share aU the same properties a n d history. T h e L a w applies o n l y w i t h a
c e r t a i n range o f qualifications after 'same', such as ' m a n ' or 'table' or,
d e f i n i t i v e l y , ' t h i n g ' . This d i s t i n c t i o n , f a m i l i a r t h o u g h n o t m a r k e d b y any
simple linguistic test, we have l e a r n t to label as the d i s t i n c t i o n between
q u a l i t a t i v e a n d n u m e r i c a l i d e n t i t y , using t e r m i n o l o g y i n h e r i t e d u l t i ­
m a t e l y f r o m A r i s t o t l e h i m s e l f O n c e we are equipped w i t h the t e r m i n o l ­
ogy we can express Leibniz's L a w u n a m b i g u o u s l y i n the f o r m : if a and b
are numerically the same {or numerically identical), a and b share all the same
properties and history.
A l t h o u g h A r i s t o t l e i n the Topics m o r e t h a n once recommends use o f a
p r i n c i p l e at least s i m i l a r to Leibniz's L a w , a p p a r e n t exceptions to the
L a w were w e l l k n o w n to h i m . A n e x a m p l e is the r o a d f r o m Thebes to
Athens a n d the r o a d f r o m Athens to Thebes, presented at Physics I I I 3 .
2 0 2 ^ 1 3 - 1 6 as n o t h a v i n g " a U the same things h o l d i n g good o f t h e m " .
Aristotle's c o m m e n t is t h a t " o n l y i n the case o f t h i n g s whose b e i n g is the
same {hois to einai to auto), n o t things t h a t are the same i n any o l d w a y , d o
aU the same things h o l d good o f t h e m " (compare the somewhat s i m i l a r
f o r m u l a t i o n at De Sophisticis Elenchis 2 4 . 1 7 9 ^ 3 8 - 4 0 ) , f r o m w h i c h he must
infer t h a t the b e i n g o f the roads is n o t the same. Some commentators see
this as i m p l y i n g a r e s t r i c t i o n o n Leibniz's L a w ( [ 1 9 ] p p . 6 6 - 7 1 , [ i o i ]
p p . 1 7 9 - 8 0 ) , b u t Aristotle's example w i U n o t a d m i t a real exception to
the L a w unless he regards the roads as n u m e r i c a l l y the same even

209
 METAPHYSICS

t h o u g h t h e i r b e i n g is n o t the same. C o u l d he suppose t h a t to be possible?

I n some places 'same i n b e i n g ' is replaced b y the a p p a r e n t l y synonymous
'same i n f o r m u l a , logos\ here " . . . i n the f o r m u l a saying w h a t i t is to
b e " ( 2 0 2 ^ 1 2 ) . I t is t r u e t h a t each o f t h e roads i n question has a different
f o r m u l a , its o w n . Y e t i f they were n u m e r i c a l l y the same, they w o u l d also
have each other's f o r m u l a , a n d so n o t differ i n being. W e have to assess
the p r o b a b i l i t y t h a t A r i s t o t l e recognized this i n f e r e n c e — o r the feasibility
o f h i s c a r r y i n g o n w i t h o u t i t . C e r t a i n l y at Δ 6 . 1 0 1 6 ^ 9 - 1 1 (cited b y [ 1 1 0 ]
p. 7 1 ) he says t h a t " t h i n g s w h i c h . . . d o n o t have one f o r m u l a we i n
fact reckon as m o r e t h a n one t h i n g " ; a n d see De Sophisticis Elenchis 2 4 .
1 7 9 ^ 1 - 4 . O n the other h a n d De Generatione et Corruptione I 5. 3 2 0 ^ 1 3 - 1 4
contrasts 'one i n f o r m u l a ' w i t h 'one i n n u m b e r ' . T h e m a t t e r remains
controversial (see B i b l i o g r a p h y ) ; b u t i t is possible to argue t h a t A r i s t o t l e
deals w i t h a l l a p p a r e n t exceptions to the L a w b y d e n y i n g t h a t they are
real exceptions, a n d i n p a r t i c u l a r t h a t his regular s o l u t i o n is t h a t p u r p o r t ­
edly i d e n t i c a l items whose properties differ are i d e n t i c a l i n some other
way than numerically ( [ i 10] pp. 6 6 - 7 4 ) .
W h a t m i g h t be a suitable w a y for the roads f r o m Thebes to Athens
a n d Athens to Thebes to be the same, yet n o t n u m e r i c a l l y the same? W e
can easily suppose, a l t h o u g h the Physics does n o t tell us, t h a t the roads
are examples o f w h a t Metaphysics Δ 9 calls c o i n c i d e n t a l (or 'accidental')
sameness; a n d there is some evidence t h a t c o i n c i d e n t a l sameness fills the
b i l l . F o r i n the same chapter οίΔ A r i s t o t l e says t h a t " m a n a n d artistic
[are the same coincidentaUy, fauta kata sumbebekos] because one o f t h e m
coincides i n the o t h e r " ( 1 0 1 7 ^ 2 9 ) . T h e r e m a r k m a k e s i t look like A r i s t o t e ­
l i a n d o c t r i n e t h a t two things can be the same coincidentaUy.
T h e N o t e o n 1017^27 dismisses this p u t a t i v e d o c t r i n e as a n inexcusable
error; b u t m o r e needs to be said. W h a t e v e r we t h i n k a b o u t m a n a n d
artistic (mousikos; others translate ' m u s i c a l ' or 'educated' or ' c u l t i v a t e d ' )
b e i n g t w o a n d yet also the same, there is n o d i f f i c u l t y i n u n d e r s t a n d i n g
h o w things m i g h t be t w o a n d yet also one; for they m i g h t be t w o elements
w h i c h m a k e u p a single composite, as for example t w o (or more) flowers
m a y m a k e u p a single b u n c h . I n Δ 6 A r i s t o t l e gives examples o f this type
o f u n i t y , w h i c h the Notes c a l l type 1. T h e Notes argue t h a t he treats
c o i n c i d e n t a l u n i t y as o f type 1; for example, he sees m a n a n d artistic, or
Coriscus a n d artistic Coriscus, as two items t h a t are coincidentaUy one
because they m a k e u p one composite, i n the l a t t e r case b y the peculiar
process o f o n e o f t h e m b e i n g i n c l u d e d i n the other.
T h e Notes o n Δ 6 propose t h a t A r i s t o t l e uses this t h e o r y o f c o m p o s i t i o n
to e x p l a i n a m a n n e r — c o i n c i d e n t a l sameness—in w h i c h items such as
Coriscus a n d artistic Coriscus are numerically the same, a n d t h a t he
exploits for the purpose a regular Greek a m b i g u i t y b y w h i c h expressions
like 'the artistic' can refer either to a r t i s t r y (an 'affection') or to its
bearers. T h e e x p l a n a t i o n is t h a t Coriscus is n u m e r i c a l l y the same as the

210
 FURTHER COMMENTS

artistic, sc. some bearer o f a r t i s t r y , b y reason o f h a v i n g the artistic, sc.

a r t i s t r y , c o i n c i d i n g i n h i m ; this is f u r t h e r developed b y [ i o i ] . I n support
o f the v i e w t h a t coincidental i d e n t i t y is, after aU, a k i n d o f n u m e r i c a l
i d e n t i t y we m a y adduce Topics I 7. 1 0 3 ^ 2 4 - 3 1 , where A r i s t o t l e speaks o f
w h a t is the same " f r o m a coincidence", apo tou sumbebekotos, as being
"one i n n u m b e r " .
T h e Notes find this d o c t r i n e confused (p. 1 3 4 ; b u t something o n the
same Unes is defended i n [ 5 7 ] p p . 3 6 7 - 8 a n d [ 1 1 1 ] ) . W e m i g h t hope to
absolve A r i s t o d e o f the confusion b y s u b s t i t u t i n g the different i n t e r p r e t a ­
t i o n o f this p a r t οΐΔ 6 w h i c h has been proposed b y Code [ 1 0 3 ] a n d
M a t t h e w s [ 1 0 4 ] , a n d w h i c h m i g h t be caUed inflationist o n the g r o u n d
t h a t i t has the appearance, at least, o f m a k i n g A r i s t o t l e postulate extra
entities t h a t " b l o a t " his o n t o l o g y ( t h o u g h b o t h [ 1 0 3 ] p. 1 7 8 a n d [ 1 0 4 ] p.
2 3 7 t h i n k they can a c q u i t h i m o f b l o a t i n g ) . A t Physics I 7. 1 9 0 ^ 1 9 - 2 0
A r i s t o t l e says t h a t w h e n a m a n becomes artistic, " t h e inartistic does n o t
persist". Since i t is clear t h a t i n a r t i s t r y m a y persist, i n some other m a n ,
" t h e i n a r t i s t i c " o u g h t here to refer e i t h e r 4 o this man's i n a r t i s t r y or to the
bearer o f i t w h i c h existed w h i l e this m a n was inartistic. Suppose i t refers
to the latter; t h e n Aristotle's theory sees the bearer as a stage or space-
t i m e p a r t ( [ 1 0 3 ] ) . T h e i n a r t i s t i c a n d the m a n are one t h i n g i n the sense
o f m a k i n g u p one t h i n g ; b u t the one t h i n g is m a d e u p n o t b y the
inherence o f affections, as the Notes suggest, b u t by the concurrence o f
such parts. W h e n the m a n becomes artistic one such p a r t , a bearer o f
i n a r t i s t r y , goes o u t o f existence a n d another, a bearer o f artistry, comes
i n t o existence; w h i l e a t h i r d , the m a n , persists. Y e t the last o f these parts
wiU be the same as one o f the others over the t i m e o f i t s concurrence w i t h
t h a t other, m u c h as t w o m e r g i n g roads can be the same over their
c o m m o n stretch.
O p i n i o n s m a y differ as to the merits o f this ontology: [ 1 0 3 ] is h a p p y
w i t h t e m p o r a l stages, whereas [ 1 0 4 ] p. 2 2 4 calls accidental unities
" k o o k y objects". O n e v i r t u e the i n t e r p r e t a t i o n does however possess: i t is
able to reconcile Aristotle's a p p a r e n t l y conflicting claims t h a t coinciden­
tal i d e n t i t y is [Topics I 7. 1 0 3 ^ 2 4 - 3 1 ) a n d is n o t [Metaphysics Δ 9 . 1017^29)
a k i n d o f n u m e r i c a l i d e n t i t y . I t b o t h is a n d is not. Absolutely speaking,
artistic a n d m a n are t w o , because one o f t h e m coincides i n the other; b u t
they are one for a t i m e , the t i m e o f their concurrence. W e are shown a
w a y i n w h i c h two things can be numerically the same t h i n g .
T h e case for i n f l a t i o n i s m i n [ 1 0 3 ] a n d [ 1 0 4 ] rests heavily o n a certain
r e a d i n g o f Aristotle's account o f change i n Physics I 7. I n o w t u r n to
Metaphysics E 2, 1 0 2 6 ^ 1 5 - 2 4 , where an i n t e r p r e t a t i o n has been offered by
W i l l i a m s [ 1 2 6 ] w h i c h seems to me superior to w h a t is proposed i n the
Notes (as also to t h a t i n [ 1 0 3 ] a n d [ 1 0 4 ] ) a n d w h i c h w o u l d , i f c o r r e c t ,
u n d e r m i n e t h a t reading.
E 2. 1 0 2 6 ^ 1 8 - 2 0 reports sophists as posing the c o n u n d r u m " w h e t h e r

2II
 METAPHYSICS

e v e r y t h i n g t h a t is, b u t n o t always, has come to be, so t h a t i f someone,

b e i n g artistic, has come to be l i t e r a t e , he has also, b e i n g l i t e r a t e , [come
to be] a r t i s t i c " . W i l l i a m s proposes the f o l l o w i n g r e c o n s t r u c t i o n : suppose
a n artistic person has come t o be l i t e r a t e ; t h e n there is a n artistic l i t e r a t e
person; b u t t h a t has n o t always been so; therefore the artistic l i t e r a t e
[ p e r s o n ] , w h i c h is the same as to say the l i t e r a t e artistic [ p e r s o n ] , has
come to be; therefore a l i t e r a t e person has come to be artistic. ( T h e
a u t h o r o f K ' s r e p o r t at 1 0 6 4 ^ 2 3 - 6 is q u i t e close to this.) T h e s o l u t i o n i n
the Notes, d r a w n f r o m De Sophisticis Elenchis, is n o longer a p p r o p r i a t e .
W i l l i a m s finds Aristotle's s o l u t i o n i n the lines o f E 2 w h i c h f o l l o w ,
1 0 2 6 ^ 2 1 - 4 , where the T r a n s l a t i o n has, " F o r w h a t is c o i n c i d e n t a l is
obviously close to w h a t is n o t , as is p l a i n also f r o m arguments such as
this: t h a t w i t h things-that-are i n a n o t h e r sense there is [a process o f ]
c o m i n g to be a n d d e s t r u c t i o n , b u t w i t h things [ t h a t are] coincidentaUy
there is n o t . " P o i n t i n g o u t t h a t Aristotle's text has 'for w i t h things t h a t
are . . .', W i l l i a m s takes 'such' to refer to the p r e c e d i n g sophistic a r g u ­
ments, a n d the second 'for . . .' clause to e x p l a i n w h a t is w r o n g w i t h the
a r g u m e n t i n ^ 1 8 - 2 0 b y contesting its use o f t h e assertion t h a t " e v e r y t h i n g
t h a t is, b u t n o t always, has come to b e " . T h e assertion was used to infer
t h a t the artistic l i t e r a t e has come to be, speaking b a l d l y {hapl5s, t r a d i t i o n ­
a l l y 'simpliciter% here m e a n i n g w i t h o u t a c o m p l e m e n t after ' b e ' ) ; b u t the
artistic l i t e r a t e is coincidentaUy, a n d w h a t is coincidentaUy does n o t
come to be, speaking b a l d l y . A c c o r d i n g l y W i U i a m s amplifies the Greek o f
1026^23 i n t o ' c o m i n g t o be [ b a l d l y ] . . .', n o t ' [ a process o f ] c o m i n g to
be . . .'.
T h i s is a n i m p r o v e m e n t . I t presents A r i s t o t l e w i t h a u n i f i e d , i f under­
s t a n d a b l y hesitant, d o c t r i n e a b o u t things t h a t are coincidentaUy, such as
literate Coriscus o r the l i t e r a t e artistic: n a m e l y t h a t (i) they d o n o t come
or cease to be, speaking b a l d l y ; ( i i ) they n o t even are c e n t r a l l y , b u t are
"close to w h a t is n o t " ; a n d ( i i i ) they are " l i k e mere n a m e s " i n t h a t t h e i r
names h a r d l y succeed i n r e f e r r i n g to a n y t h i n g . B y contrast there is, as
the Notes c o m p l a i n (p. 1 9 6 ) , n o p l a u s i b i l i t y i n the t h o u g h t i m p u t e d b y
the T r a n s l a t i o n t h a t coincidences take n o t i m e i n c o m i n g a n d ceasing to
be.
B u t the result alters o u r r e a d i n g o f Aristotle's account o f change i n
Physics I 7. T h e i n a r t i s t i c w h i c h "does n o t persist" w h e n a m a n becomes
artistic is coincidentaUy. Therefore i t does n o t cease to be, speaking
b a l d l y , w h e n the m a n ceases to be artistic; a n d s i m i l a r l y the artistic does
n o t come to be, speaking b a l d l y ; "does n o t persist" is s h o r t h a n d for 'does
n o t r e m a i n i n a r t i s t i c ' . A s i m i l a r deflationist r e a d i n g — r e v i s i n g [ 1 8 ] p p .
1 0 0 - 2 — m a y be possible at De Generatione et Corruptione I 4 . 3 1 9 ^ 2 5 - 9 .
M a t t h e w s ' k o o k y objects a n d Code's t e m p o r a l stages are n o t yet o n the
scene.
P l e n t y o f obstacles r e m a i n , however, i n the w a y o f attempts (such as

212
 FURTHER COMMENTS

t h a t outUned b y W i l l i a m s i n [ i o 6 ] ) to purge A r i s t o t l e o f these excres­

cences. I e n d this section b y d r a w i n g o n the somewhat different i n f l a t i o n ­
ist i n t e r p r e t a t i o n i n Lewis [ 1 0 9 ] a n d M a t t h e n [ 7 2 ] i n o r d e r to a m p l i f y
and a m e n d the N o t e o n Δ 6 . 1 0 1 5 ^ 1 6 f f
De Sophisticis Elenchis 2 4 presents solutions t o fallacious refutations
d e p e n d i n g o n coincidence. A t 1 7 9 ^ 2 - 4 A r i s t o t l e has this c o m m e n t o n one
such fallacy: "so i t is n o t the case t h a t i f I k n o w Coriscus b u t a m
i g n o r a n t o f the [ m a n ] a p p r o a c h i n g , I k n o w a n d a m i g n o r a n t o f the same
[ m a n ] . " I t is supposed t h a t the t h i n g a p p r o a c h i n g is Coriscus. B u t
A r i s t o t l e proposes to escape inference to ' I a m i g n o r a n t o f Coriscus' b y
means o f the f o l l o w i n g p r i n c i p l e : " i t is n o t necessary t h a t w h a t is t r u e
o f w h a t coincides be t r u e also o f the a c t u a l t h i n g [ i t coincides i n ] "
( 1 7 9 ^ 3 6 - 7 ) . T h i s is n o t a r e s t r i c t i o n o n S u b s t i t u t i v i t y {pace [ 1 0 8 ] p. 1 4 5 ;
perhaps i t should b e ) , because for A r i s t o t l e things, n o t words, are ' t r u e
of'. N o r is i t a r e s t r i c t i o n o n Leibniz's L a w , for the example shows h i m
denying that the approaching [man]—"what coincides"—and
C o r i s c u s — " t h e a c t u a l t h i n g " — a r e the same. H i s diagnosis is t h a t an
a p p a r e n t exception to Leibniz's L a w is n o t a real exception because (1)
Coriscus a n d ( 2 ) the a p p r o a c h i n g [ m a n ] are two. I t is a short step f r o m
this to w h a t [ 1 0 9 ] calls A c c i d e n t a l C o m p o u n d s , such as ( 3 ) a p p r o a c h i n g
Coriscus. Aristotle's " t h e o r y is t h a t j u s t as the expression 'the artistic
Coriscus' is a complex o f the words 'the artistic' a n d 'Coriscus', so w h a t
i t designates is a complex o f t w o n o n - l i n g u i s t i c items, the artistic a n d
Coriscus" (Notes p. 1 3 4 ; artistic m a n is called a " w h o l e " at Δ 1 1 .
1 0 1 8 ^ 3 4 - 5 ) . Y e t the t w o items ( i ) a n d ( 2 ) a n d the t w o items ( i ) a n d ( 3 )
each have a c e r t a i n k i n d o f sameness, c o i n c i d e n t a l sameness, w h i c h holds
w h e n ( 3 ) exists, i.e. w h e n Coriscus is a p p r o a c h i n g (see Metaphysics Z 12.
1 0 3 7 ^ 1 4 - 1 7 , [ 7 2 ] p . 1 2 4 ) . M a t t h e n [ 7 2 ] makes comparisons w i t h Plato
and perhaps Parmenides, a n d Lewis [ 1 0 9 ] works o u t the logic o f s u c h
relationships (not q u i t e these ones, because Lewis identifies ( 2 ) a n d ( 3 ) ,
and takes ( 3 ) as c o m p o u n d i n g Coriscus w i t h the affection o f a p p r o a c h i n g ,
not w i t h ( 2 ) , its bearer). Lewis also shows h o w the theory makes sense o f
Г 4 . 1 0 0 7 ^ 3 3 - ^ 1 8 , o n w h i c h I shall offer m y o w n fresh comments below.
W h a t should we say o f such a theory? (i) I t treats c o i n c i d e n t a l
sameness as n o t a k i n d o f n u m e r i c a l sameness, despite Topics I 7. 1 0 3 ^ 2 4 -
3 1 a n d c o n t r a r y to the N o t e o n Δ 9 . 1017^27. Hence i f a l l a p p a r e n t
exceptions to Leibniz's L a w are cases o f c o i n c i d e n t a l sameness, t h e n
Leibniz's L a w has no exceptions, (ii) T h e r e is a good chance o f represent­
i n g A r i s t o t l e as h o l d i n g t h a t a l l a p p a r e n t exceptions to Leibniz's L a w are
cases o f c o i n c i d e n t a l sameness; for example, the r o a d f r o m Thebes to
Athens w i l l be a c o m p o u n d consisting o f the r o a d a n d the [ t h i n g ]
l e a d i n g f r o m Thebes to Athens (or i f [ 1 0 9 ] is r i g h t , consisting o f t h e r o a d
and the d i r e c t i o n f r o m Thebes to A t h e n s ) , ( i i i ) T h e theory is surely
v u l n e r a b l e to the c r i t i c i s m o n p. 145 o f t h e Notes: " i t is w r o n g to regard

213
 METAPHYSICS

the [ c o m p o u n d ] designated 'artistic m a n ' as a c o m b i n a t i o n o f t w o items,

the m a n a n d the artistic (either 'the a r t i s t i c ' means 'someone a r t i s t i c ' , i n
w h i c h case there is o n l y one i t e m , o r i t means ' a r t i s t r y ' , i n w h i c h case we
face the impossible question 'is the m a n - e l e m e n t i n the c o m p l e x artistic
or n o t ? ' ) " — a c c i d e n t a l c o m p o u n d s are i n d e e d k o o k y . ( L a t e r ancient
Sceptics p r o v e d the kookiness o f a s i m i l a r A r i s t o t e l i a n excrescence,
relatives—see [ 9 8 ] ; b u t the c o m p l a i n t as a w h o l e is rejected i n [ 7 2 ]
p. 125 a n d n . 2 4 . ) (iv) T h e s o l u t i o n i n t o w h i c h , i f [ 1 2 6 ] is r i g h t , A r i s t o t l e
was goaded b y the sophistic c o n u n d r u m o f E 2. 1 0 2 6 ^ 1 8 - 2 0 , offers
promise o f a different, deflationist account o f these c o m p o u n d s , agreeable
to those w h o l i k e t h e i r A r i s t o t l e to be commonsensical ( s o m e t h i n g I
h a n k e r e d after w h e n I w r o t e the Notes). B u t A r i s t o t l e d i d n o t w o r k i t
o u t ; a n d h a d he done so, at least m a n y o f his p r o b l e m s a b o u t i d e n t i t y ,
most o f w h i c h I have n o t t o u c h e d o n , w o u l d have needed new
s o l u t i o n s — p e r h a p s t h r o u g h the d i s t i n c t i o n , still t o d a y n o t firmly grasped
b y a l l c o m m e n t a t o r s , between L e i b n i z ' s L a w so called a n d the L a w o f
Substitutivity.

Δ y,E2: B E I N G

A r i s t o t l e says, " B u t t h a t w h i c h is, w h e n b a l d l y [haplos) so called, m a y be

so called i n several ways. O n e ofthese was t h a t [ w h i c h is] coincidentaUy,
a n o t h e r t h a t [ w h i c h is] as t r u e . . . A p a r t f r o m these are the figures o f
p r e d i c a t i o n . . . a g a i n a p a r t f r o m a l l these, t h a t [ w h i c h is] p o t e n t i a l l y
a n d a c t u a l l y " {E 2. 1 0 2 6 ^ 3 3 - ^ 2 ) . Discussing the w o r d ' b a l d l y ' the N o t e
o n 1026^33 concludes t h a t " i n E 2 'is b a l d l y ' means 'exists' " .
B u t t h a t w o n ' t do. I n the first place, 'is b a l d l y ' is n o t s o m e t h i n g t h a t
gets said o r w r i t t e n ; one says 'is' b a l d l y , n o t 'is b a l d l y ' . Secondly, the
E n g l i s h v e r b 'exist' must be said b a l d l y — i t a d m i t s no c o m p l e m e n t ; b u t
there is n o evidence t h a t A r i s t o t l e , o r any o t h e r ancient Greek t h a t I
k n o w of, perceived a sense o f the v e r b 'einai' i n w h i c h i t m u s t be said
b a l d l y . W h e n they h e a r d a b a l d use, as 'Socrates is', they h e a r d i t always
as a d m i t t i n g , t h o u g h n o t necessarily i n v i t i n g , a c o m p l e m e n t ; they h e a r d
i t as we hear ' c h i l d r e n l e a r n ' n o t as we hear ' c h i l d r e n g r o w ' ( i n l e a r n i n g ,
c h i l d r e n l e a r n things, b u t i n g r o w i n g they d o n o t g r o w t h i n g s — I owe
the example a n d the p o i n t i t makes to B r o w n [ 8 o ] ) . A d m i t t e d l y the
Greek 'einai' gets used b a l d l y n o t o n l y , as w i t h E n g l i s h 'be', w h e n a
p a r t i c u l a r c o m p l e m e n t is u n d e r s t o o d f r o m the context, t y p i c a l l y i n
answer to a question as i n 'Yes, he is', b u t also where E n g l i s h w o u l d use
'exist', for w h i c h the ancient languages h a d n o separate w o r d (this l a t t e r
is w h a t the N o t e means b y n o n - e l l i p t i c a l b e i n g ) . B u t the syntactical fact
t h a t i t always a d m i t s a c o m p l e m e n t debars 'einai' f r o m ever meaning the
same as 'exist', w h i c h does n o t a d m i t a c o m p l e m e n t .

214
 FURTHER COMMENTS

Despite this syntactical difference there are f a i r l y close similarities i n

the logical b e h a v i o u r o f ' e x i s t ' a n d οί'βιηαΓ used b a l d l y . I c o u n t these
three: each implies 'is something'; i t is a defensible philosophical thesis
t h a t i n any occurrence o f each there is some φ such t h a t the occurrence
implies 'is <^'; b u t neither o f the reverse i m p l i c a t i o n s holds w i t h o u t
restriction (however, the restrictions m a y w e l l diverge; see [ 7 1 ] ) . I f w e
a d d t h a t Aristode's phrase " t h a t w h i c h is, w h e n b a l d l y so c a l l e d " is
p r e s u m a b l y i n t e n d e d n o t to cover occurrences where the context supplies
a c o m p l e m e n t , I t h i n k we can save the substance o f the conclusion i n the
Notes, viz. t h a t the phrase indicates cases i n w h i c h English 'exist' w o u l d
be a f a i r l y n a t u r a l paraphrase o f t h e Greek 'einai'—and, one m i g h t a d d ,
i n w h i c h E n g l i s h 'be' w i t h o u t supplement w o u l d n o t be n a t u r a l at a l l .
T h e same f o u r f o l d d i v i s i o n appears i n Δ 7, b u t a p p l i e d to " t h a t w h i c h
is" w i t h o u t e x p l i c i t restriction to w h a t is "so called b a l d l y " . T h e Notes
argue t h a t the restriction is already i m p l i c i t there, at least i n the
t r e a t m e n t o f w h a t is i n its o w n r i g h t ( i n t e r p r e t a t i o n ( 2 ) , p p . 1 4 1 - 3 ) ,
p r o b a b l y i n the t r e a t m e n t o f w h a t is coincidentaUy ( i n t e r p r e t a t i o n ( 3 ) ,
p. 1 4 4 ) . These conclusions have n o t f o u n d favour. B o t h are rejected i n
[ 7 1 ] , a n d also b y T h o r p i n [ 6 9 ] , w h o is satisfied t h a t " t h e r e is n o t h i n g i n
the d e p l o y m e n t o f the four uses o f einai i n Δ 7 w h i c h looks remotely like
the existential use" (p. 2 5 4 ) ; according to T h o r p ' w h a t is b a l d l y ' does n o t
" m e a n 'existential b e i n g ' " i n E 2 either, t h o u g h i t does i n some other
places (p. 2 5 5 ) . Grice takes the same v i e w a b o u t 1 0 1 7 ^ 2 7 - 3 0 ( " a t least at
< t > h i s p o i n t " , [ 6 4 ] p. 1 8 0 ) .
I t m a y nevertheless be w o r t h w h i l e to d r a w o u t some f u r t h e r conse­
quences o f this disfavoured i n t e r p r e t a t i o n , w h i c h gets i t s e l f — o r
A r i s t o t l e — i n t o a m a j o r d i f f i c u l t y n o t acknowledged i n the Notes.
Coincidental being. A c c o r d i n g to the Notes A r i s t o t l e means us to find
c o i n c i d e n t a l b e i n g b y r e a d i n g his Δ 7 examples i n a w a y t h a t demands
the E n g l i s h w o r d order i n 'the j u s t artistic is\ 'the m a n artistic is\ a n d
'the artistic m a n is'; a n d he means us to account for i t as derivative f r o m
the b e i n g o f something else—here a m a n — i n w h o m the artistic, or the
j u s t a n d the artistic, coincide. T h e statement i n the Notes o f this
i n t e r p r e t a t i o n , " ' c o i n c i d e n t a l ' means ' d e r i v a t i v e ' " (p. 1 4 4 ) is sloppy, i f
not worse; r a t h e r 4s coincidentaUy' means 4s, b y a coincidence o f itself
or its parts or one o f its parts i n something else t h a t is'. Aristotle's
p a r e n t h e t i c a l example o f t h e not-pale, so far f r o m " h a v i n g n o t h i n g to do
w i t h c o i n c i d e n t a l b e i n g " as the tortuous N o t e o n p. 146 complains, fits
the i n t e r p r e t a t i o n w i t h o u t d i f f i c u l t y (and could be attached to the
general account o f b e i n g , said b a l d l y , i n [ 7 1 ] , as o n [ 7 1 ] p. 4 3 1 ) .
Being in its own right. T h e things "said to be i n their o w n r i g h t " are " a U
things w h i c h signify the figures o f p r e d i c a t i o n " , exemplified amongst
others b y w h a t signifies " a q u a n t i t y " a n d " a q u a l i f i c a t i o n " ( 1 0 1 7 ^ 2 2 - 6 ) .
As the Notes later e x p l a i n ( p p . 1 6 0 , 1 6 2 ) , these translations foUow [ 1 3 ]

215
 METAPHYSIGS

i n using abstract nouns where A r i s t o t l e has t w o adjectives, 'posorC a n d

'poM. T h e m e a n i n g o f the adjectives m a y be i n t e r r o g a t i v e ( ' o f w h a t
amount', 'how qualified') or indefinite ('ofsome amount', 'qualified').
Aristotle's choice o f t h e m is n o stylistic q u i r k , for he possesses abstract
nouns 'posotes' a n d 'poiotes' ( f r o m w h i c h descend o u r ' q u a n t i t y ' a n d
' q u a l i t y ' ) a n d means s o m e t h i n g different b y t h e m : thus Δ 14. 1 0 2 0 ^ 1 7 -
2 4 tells us t h a t w h i l e poiotetes i n c l u d e such things as excellence a n d
badness, a n d i n general affections, " t h e good a n d the b a d signify q u a l i f i c a ­
t i o n {to poionY'; instances o f a poiotes are at best w h a t Categories 1. 1^12-
15 has called p a r o n y m s o f i n s t a n c e s o f (or o f answers to the question)
poion, t h a t is, things whose names are d e r i v e d f r o m t h e i r names—e.g.
hotness is a p a r o n y m o f the h o t . W e should infer t h a t the things said to be
i n t h e i r o w n r i g h t are such things as the good a n d the j u s t a n d the pale
a n d the b i g , n o t such things as excellence a n d justice a n d p a l l o r a n d size.
A w h o l e new a r m y o f k o o k y objects seems to l o o m .
I make n o c o m m e n t o n this a p p a r i t i o n , b u t t u r n instead to a p a r t i c u l a r
consequential d i f f i c u l t y t h a t the Notes d o n o t address. 1 0 1 7 ^ 1 8 - 1 9 has
j u s t t o l d us t h a t the not-pale is coincidentaUy, n o t i n its o w n r i g h t ; a n d i f
the i n t e r p r e t a t i o n preferred i n the Notes is r i g h t , the same applies to the
j u s t artistic, a n d the artistic m a n . W h y t h e n should the pale, a n d other
things signi{yingpoion etc., be i n t h e i r o w n right? Categories 4 . 1^25 offers a
clue, b y specifying the " t h i n g s w h i c h signify . . . poson . . . poion . . . " etc.
as " s a i d w i t h o u t a n y c o m b i n a t i o n " . I n Δ 7 we are g i v e n a different
specification, " t h i n g s p r e d i c a t e d " , b u t perhaps i t means the same: per­
haps things p r e d i c a t e d have to be said i n single w o r d s , n o t i n phrases like
' n o t - p a l e ' {'me leukon') or 'pale m a n ' .
T h i s is n o t very c o n v i n c i n g , because i t is h a r d to see h o w the existence
o f a o n e - w o r d designation should m a k e the difference, a m o n g things t h a t
coincide i n substances, between those t h a t are i n t h e i r o w n r i g h t a n d
those whose v e r y being is c o i n c i d e n t a l . B u t let us suppose the difference
does exist, h o w e v e r i t is made; t h e n t w o dogmas o f t r a d i t i o n a l A r i s t o t e l i a n -
ism disappear f r o m the Metaphysics o f A r i s t o t l e h i m s e l f O n e is the d o g m a
t h a t accidents, i n the sense o f things t h a t are coincidentaUy, are the same
as accidents o f a substance, i n the sense o f things t h a t coincide i n a
substance. O n the above i n t e r p r e t a t i o n this is w r o n g i n a n y case, because
things t h a t are coincidentaUy i n c l u d e items such as artistic m a n , o n l y
p a r t o f w h i c h coincides i n a substance (the other p a r t ) ; i t w i l l n o w be
w r o n g i n the other d i r e c t i o n too, because things t h a t signify poson, poion,
etc. w i U coincide i n a substance w i t h o u t being coincidentaUy. T h e other
d o g m a is t h a t Aristotle's categories comprise substances a n d accidents.
O n the c o n t r a r y , i f b y 'categories' we m e a n (as t r a d i t i o n a l l y ) the " t h i n g s
said w i t h o u t a n y c o m b i n a t i o n " oiCategories 4 . 1^25, w h i c h surely are or
at least i n c l u d e the " t h i n g s w h i c h signify the figures o f p r e d i c a t i o n " i n Δ
7, these things exclude some accidents o f substances, such as the

216
 FURTHER COMMENTS

n o t - p a l e , a n d they exclude a l l o t h e r accidental beings, such as artistic m a n .

( I t is often t a k e n t h a t the categories comprise—besides substances—
q u a n t i t i e s , qualities, relations, a n d so o n . T h a t too is r u l e d o u t b y
Aristotle's d i s t i n c t i o n between poion etc. a n d poiotes etc.; a n d he does n o t
even have a w o r d for ' r e l a t i o n ' . Q u a l i t i e s , poiotetes, are a m o n g the things
m e n t i o n e d i n Г 2 . 1 0 0 3 ^ 3 3 - ^ 1 0 as b e i n g called things t h a t are because
they are s o m e t h i n g o r o t h e r of a substance; they are so caUed " w i t h
reference t o " t h a t " o n e p a r t i c u l a r n a t u r e " . Accidents i n the sense o f
accidental beings, the things t h a t are coincidentaUy, m u s t n o t be c o u n t e d
i n this g r o u p , because E 2. 1 0 2 6 ^ 2 - 4 w i U exclude t h e m f r o m the discipline
whose subject m a t t e r the g r o u p d e l i m i t s . )
T h e N o t e o n 1017^35 agrees w i t h A r i s t o t l e t h a t 'sees', a n d hence 'is
one-that-sees', are ambiguous; b u t complains t h a t " h e d o e s n o t e x p l a i n
h o w , i n the l a t t e r case, the a m b i g u i t y is transferred f r o m 'sees' to ' i s ' "
( p . 1 4 6 ) . F o r a possible e x p l a n a t i o n see [ 6 2 ] ,

A 30: C O I N C I D E N C E

T h e n o t i o n o f coincidence, t r a d i t i o n a l l y 'accident', pervades every p a r t

o f the A r i s t o t e l i a n corpus. I n these books o f the Metaphysics i t plays a
role, as we h a v e s e e n above, i n Aristotle's attack i n Г o n the opponents
o f P N C ; several chapters o(A, especially 6 , 7, a n d 9 , i n v o k e i t i n m a k i n g
t h e i r m a i n d i v i s i o n a m o n g the various " w a y s o f b e i n g caUed" w h a t e v e r
t h a t p a r t i c u l a r chapter is a b o u t ; A 3 0 has coincidence as its o w n subject
m a t t e r ; a n d E 2 argues t h a t " t h e r e is no study t h a t deals w i t h t h a t
[ w h i c h is] c o i n c i d e n t a U y " ( 1 0 2 6 ^ 3 - 4 ) . T h e m e a n i n g o f A r i s t o t l e ' s v e r b
'sumbebekenaV is discussed o n p p . 7 6 - 7 o f t h e Notes, a n d there are o t h e r
r e l e v a n t Notes o n p p . 101-2, 133-5, 1 4 3 - 6 , 1 8 0 - 2 , and 190-5. Here I
shall t r y to d r a w some threads together. (Aristotle's t r e a t m e n t i n Topics I
5 is p u z z l i n g l y different, a n d p u z z l i n g i n itself; b u t , as i n the Notes, I w i U
c o n t i n u e to i g n o r e i t , despite t h e j u s t c o m p l a i n t i n [ 1 1 7 ] n . 6 . )
T h e r e are three constructions to consider:
(1) a coincides i n b {tode toide sumbebeke)
( 2 ) a is c o i n c i d e n t a l , or a coincidence {sumbebekos)
(3) a is Ф c o i n c i d e n t a U y {kata sumbebekos)

L e t us first l o o k at ( i ) ; ( 2 ) a n d ( 3 ) w i U b e definable i n terms o f i t .

T h r o u g h o u t i t is i m p o r t a n t to r e m e m b e r t h a t (i) some c o m m e n t a t o r s o n
A r i s t o t l e , n o t a b l y S o r a b j i i n [ 1 1 2 ] c h . 1 , use 'coincidence' not t o translate
'sumbebekos' b u t i n its m o d e r n E n g l i s h sense, a n d ( i i ) m a n y use o t h e r
w o r d s to translate 'sumbebekos', e.g. 'accident', ' i n c i d e n t a l ' , 'concurrence',
'concomitant'.

( I ) W e l e a r n a good deal a b o u t the r e l a t i o n o f coinciding in f r o m Г 4 .

217
 METAPHYSICS

1 0 0 7 ^ 3 3 - ^ 1 8 . ^34-^1 says: " ' c o i n c i d e n t a r always signifies a p r e d i c a d o n

a b o u t a c e r t a i n subject". I n Aristotle's usage subjects, hupokeimena, m u s t
be o t h e r t h a n w h a t is p r e d i c a t e d o f t h e m : therefore coinciding in is a n
irreflexive r e l a t i o n . ^ 2 - 4 continues: " t h e c o i n c i d e n t a l is n o t c o i n c i d e n t a l
i n the c o i n c i d e n t a l , unless because b o t h coincide i n the same t h i n g " .
T h i s m a y l o o k like a statement o f t r a n s i t i v i t y (the N o t e o n p. 1 0 1
nonsensically refers to i t as " t h e transitive case"), b u t Aristotle's m e a n i n g
is different. As his f o l l o w i n g example shows, he has i n m i n d the case i n
w h i c h a a n d b coincide i n one another; this can h a p p e n , b u t o n l y i f — a n d
because—each coincides i n some t h i r d t h i n g c. ( I a m a f r a i d t h a t the
same N o t e misrepresents this as " a predicate o f X ' s predicate must be a
predicate o f X " , a different a n d contentious thesis o n w h i c h see especially
[ 5 7 ] . ) Since m u t u a l coincidence o f a a n d b is possible, the r e l a t i o n is n o t
a s y m m e t r i c a l ; a n d since i t is irreflexive, i t follows t h a t i t c a n n o t be
transitive. ^ 5 - 6 t h e n distinguishes t w o " w a y s " o f c o i n c i d i n g , one the w a y
i n w h i c h the pale coincides i n the artistic, the other t h a t i n w h i c h the
pale coincides i n Socrates. T h e f o r m e r occurs "because b o t h coincide i n
the same" t h i r d t h i n g , a n d is t h e r e f o r e — t h o u g h A r i s t o t l e does n o t say
s o — a secondary w a y o f c o i n c i d i n g , explained t h r o u g h the other p r i m a r y
w a y ( i n the language developed u n d e r ( 3 ) below, he c o u l d have said t h a t
the pale coincides coincidentaUy i n the a r t i s t i c ) . N e x t we are t o l d ( ^ 9 - 1 0 )
t h a t there c a n n o t be "some other t h i n g " , sc. t h a n the pale, c o i n c i d i n g
i n the pale Socrates. T h e upshot is t h a t the pale can coincide, i n the
p r i m a r y w a y , o n l y i n a substance (^17). Since c o i n c i d i n g i s " p r e d i c a t i o n
a b o u t a c e r t a i n subject", i t m i g h t seem t h a t substances c a n n o t themselves
coincide, unless perhaps i n m a t t e r . B u t A r i s t o t l e does n o t stick to this.
F o r example Metaphysics Z 5 . 1 0 3 0 ^ 2 0 - 1 speaks o f " . . . [ a ] w h i t e i n
w h i c h b e i n g a m a n coincides"; a n d A 2. 1 0 1 3 ^ 3 6 - 1 0 1 4 ^ 1 ( = Physics I I 3.
1 9 5 ^ 3 4 - 5 ) says t h a t w h e n Polyclitus makes a statue, t h a t is because
" b e i n g Polyclitus coincides i n a [ l i t e r a l l y : t h e ] statuemaker". ( W e m a y
note t h a t , because the converse also holds, viz. statuemaker coincides i n
Polyclitus, this is another p r o o f t h a t A r i s t o t l e c a n n o t a l l o w the r e l a t i o n o f
coinciding in to be b o t h irreflexive a n d transitive.)
O n the other h a n d , we never hear o f m a n c o i n c i d i n g i n Callias, or
a n i m a l i n m a n . T h e N o t e o n p . 1 0 1 explains this as resting o n Aristotle's
d o c t r i n e t h a t 'Callias is a m a n ' a n d ' m a n is a n a n i m a l ' express identities;
b u t since I n o w t h i n k A r i s t o t l e h e l d no such d o c t r i n e , we must look for
another e x p l a n a t i o n . I t has already been offered. M a n holds good o f
Callias " i n his o w n r i g h t {kath' hauton)'\ whereas w h a t coincides i n a
t h i n g holds good o f i t d e r i v a t i v e l y , i n the sense:

(C) Ф ( e q u i v a l e n t l y the φ, or being φ) coincides i n b w h e n b is φ

t h r o u g h s o m e t h i n g else b e i n g φ.

E a r l i e r I elaborated (C) i n terms o f A r i s t o t l e ' s d i s t i n c t i o n i n the Categories

218
 FURTHER COMMENTS

between b e i n g said of a n d b e i n g in\ thus φ w o u l d coincide i n b w h e n b is

Ф either t h r o u g h <^'s b e i n g i n s o m e t h i n g said o f b, o r t h r o u g h <^'s being
said o f s o m e t h i n g i n b or ( c o m b i n i n g t h e m ) t h r o u g h <^'s being said o f
s o m e t h i n g i n s o m e t h i n g said o f b. B u t this e l a b o r a t i o n w i U n o t fit
u n t y p i c a l examples such as t h a t o f PolycUtus i n Δ 2, because being
Polyclitus is n o t in a n y t h i n g , n o r is i t said of a n y t h i n g i n the statuemaker
(it is p r e s u m a b l y said o f m a n , w h i c h is said of, n o t i n , the statuemaker).
So I n o w suggest the b r o a d e r a n d less technical f o r m u l a ( C ) . E v e n so, o f
course, q u i t e a strange result foUows for Polyclitus, n a m e l y t h a t the
statuemaker is Polyclitus t h r o u g h s o m e t h i n g else b e i n g Polyclitus. Y e t
this strangeness is exactly w h a t Aristotle's entanglement w i t h kooky
objects w i U deliver: b e i n g Polyclitus holds p r i m a r i l y o f a c e r t a i n sub­
stance, a c e r t a i n m a n ; the statuemaker is something else t h a n — o n l y
coincidentaUy the same a s — t h a t m a n . ( T h e fact t h a t (C) makes a
contrast w i t h w h a t holds good o f a t h i n g i n its o w n r i g h t is perhaps
sufficient to j u s t i f y the Notes i n g i v i n g the corresponding sense o f ' c o i n c i ­
d e n t a l ' as "non-essential".)

( 2 ) 'a is coincidentaV, or ' . . . a coincidence' ( b o t h words translate 'sumbebe­

kos') is sometimes foUowed b y 'for b'; i n t h a t case i t s i m p l y means 'a
coincides i n b'. I f the context supplies no b, the m e a n i n g m a y be ' 0
coincides i n x' w i t h the context s u p p l y i n g some q u a n t i f i e r i n 'x' (e.g.
'a coincides i n something'); b u t m o r e u s u a l l y — a n d indeed most u s u a l l y —
'a is a coincidence' has an absolute m e a n i n g , w h i c h c o u l d loosely be
expressed as 'a is the resultant o f some c c o i n c i d i n g i n some d'. I n this
sense a coincidence is a k i n d o f c o m p l e x e n t i t y , composed o f t w o
elements one o f w h i c h coincides i n the other.
W e have already i d e n t i f i e d some o f these complex entities as [ 1 0 9 ] ' s
A c c i d e n t a l C o m p o u n d s : the artistic Coriscus, the artistic m a n , the artistic
pale. Δ 3 0 brings us four m o r e examples, a n d at the same t i m e offers, i n
effect, t w o f u r t h e r accounts o f w h a t a coincidence is. T h e first example i n
Δ 3 0 is a relative case where 'is c o i n c i d e n t a l for' can be understood
w i t h o u t d i f f i c u l t y as m e a n i n g 'coincides i n ' : " t h e finding o f treasure is
. . . c o i n c i d e n t a l for h i m w h o is d i g g i n g the t r e n c h " ( 1 0 2 5 ^ 1 6 - 1 7 ) . T h e
second is i n t r o d u c e d b y means o f a complete sentence, i n effect 'the
artistic is pale', w h i c h gives the impression t h a t the i t e m referred to {"this
comes to be . . . we c a l l it c o i n c i d e n t a l " , 1025^21) is a c o n j u n c t i v e event
or state o f affairs, t h a t o f something's b e i n g b o t h artistic a n d pale, r a t h e r
t h a n a n A c c i d e n t a l C o m p o u n d , the artistic pale. I n the t h i r d example
this impression is strengthened. T h e case is again a relative one; indeed
the T r a n s l a t i o n , " i t was a coincidence for someone to visit A e g i n a " ( ^ 2 5 - 6 ) ,
masks the fact t h a t A r i s t o d e a c t u a l l y uses the finite v e r b 'coincided
i n ' , albeit i n the aorist {sunebe) a n d n o t , as usually, the perfect tense. B u t
this example does n o t i d e n t i f y for us the two elements w h i c h , a c c o r d i n g

219
 METAPHYSICS

to the account o f coincidence so far given, c o m b i n e together to p r o d u c e

a n A c c i d e n t a l C o m p o u n d . W e should expect s o m e t h i n g like ' v i s i t i n g
A e g i n a coincided i n Plato' or '. . . i n the traveller to Piraeus'—they visited
A e g i n a t h r o u g h s o m e t h i n g (allegedly) 'else', the captive-of-Aeginetan-
pirates or the hugger-of-an-Aeginetan-lee-shore, v i s i d n g the place;
whereas w h a t A r i s t o t l e a c t u a l l y says is t h a t v i s i t i n g A e g i n a coincided i n
"someone". Perhaps we can s u p p l y the e x t r a element for ourselves; b u t
because o f Aristotle's m a n n e r o f expression i t is once a g a i n r a t h e r m o r e
n a t u r a l t o t h i n k o f the coincidence i n v o l v e d as a less k o o k y — o r at any
rate a m o r e f a m i l i a r — k i n d o f e n t i t y , a c o n j u n c t i v e event (see also the
example at Posterior Analytics I 4 . 7 3 ^ 1 1 - 1 2 , " i t l i g h t e n e d w h e n he w e n t
for a w a l k " ) .
A r i s t o t l e comments o n his first three examples t h a t there is " n o
definite cause o f t h e c o i n c i d e n t a l , b u t a chance o n e " ( 1 0 2 5 ^ 2 4 - 5 ) ; a n d
this looks like a new account o f w h a t i t is for one t h i n g , the φ, to coincide
i n another t h i n g , b, so t h a t t h e i r resultant is a coincidence: t h a t happens
not, o r n o t o n l y , w h e n b is φ because s o m e t h i n g else is φ, b u t also w h e n b
is Ф because o f a chance cause.
Since Aristotle's views a b o u t chance causes belong t o Physics b o o k I I
(see [ 1 5 ] , [ 1 2 0 ] ) , o n l y three f u r t h e r remarks are i n place here, (i) I t m a y
have been less clear t o A r i s t o t l e t h a n i t is to us t h a t there is a difference
between a n A c c i d e n t a l C o m p o u n d such as the Piraeus-bound A e g i n a -
stranded, a n d a c o n j u n c t i v e event such as its h a p p e n i n g t h a t someone is
b o t h b o u n d for Piraeus a n d stranded i n A e g i n a . ( i i ) T h e 'new' account
o f coincidences is n o t so different f r o m the o l d one t h a t they m i g h t n o t be
w o r k e d i n t o some sort o f u n i t y , ( i i i ) T h e r e is a f u r t h e r 'new' account o f
coincidences i n Δ 30's i n i t i a l statement o f w h a t " w e call c o i n c i d e n t a l " ,
viz. " w h a t holds good o f s o m e t h i n g . . . b u t neither o f necessity n o r for
the most p a r t " . I t seems best to r e g a r d this official statement as derived
f r o m the first 'new' account, b y means o f the f o l l o w i n g three f u r t h e r
assumptions: {a) w h e n b is φ n o t because o f a chance cause, i t is φ because
i t is b, i.e. ^'s b e i n g φ is w h a t the N o t e calls " s e l f - e x p l a n a t o r y " (p. i 8 i ) ;
{b) causes can always be g i v e n by, a n d never w i t h o u t , c i t i n g " u s u a l or
constant c o n j u n c t i o n s " ( i b i d . ) ; a n d {c) w h a t happens constantly happens
necessarily. T h e N o t e o n 1025^14 makes some c r i t i c a l remarks a b o u t
these assumptions; see f u r t h e r [ 1 1 9 ] a n d [ 1 2 0 ] .
I n Δ 30's f o u r t h example o f a coincidence, 1 0 2 5 ^ 3 0 - 2 , A r i s t o t l e says
t h a t "possessing t w o r i g h t angles" is a coincidence because i t holds good
o f a t r i a n g l e " i n [ t h e triangle's] o w n r i g h t w i t h o u t being i n [ t h e t r i ­
angle's] substance". H e r e we are m e a n t to u n d e r s t a n d 'coincidence for
the t r i a n g l e ' (indeed " [ d o e s ] f o r " m i g h t be better t h a n the T r a n s l a t i o n ' s
" [ d o e s ] o f " at 1025^32, m a k i n g us supply "is a coincidence" r a t h e r t h a n
" h o l d s g o o d " as the unexpressed v e r b ) ; so the sense is once m o r e
'coincides i n ' — " n o n - e s s e n t i a l " as the N o t e has i t . O n the o t h e r h a n d a

220
 FURTHER COMMENTS

new kind o f coincidence appears i n this example, a n i n - i t s - o w n - r i g h t

coincidence, t r a d i t i o n a l l y 'per se accident'. T h i s k i n d is i m p o r t a n t to
A r i s t o t l e as w h a t Barnes calls " t h e staple o f d e m o n s t r a t i o n " ( [ i 6 ]
p. 1 1 5 ) ; for i n the conclusion o f a n y d e m o n s t r a t i o n the predicate (i) holds
good o f the subject i n its o w n r i g h t (because u n d e r s t a n d i n g , episteme, is
always o f w h a t holds good o f something i n its o w n r i g h t , a n d w h a t is
demonstrated is understood) a n d (ii) coincides i n the subject (because i t
holds good o f i t t h r o u g h something else, viz. a m i d d l e t e r m ) . T h e sense
o f ' i n its o w n r i g h t ' here is m o r e difficult; see [ 1 1 8 ] .

( 3 ) 'Coincidentaliy is used i n the T r a n s l a t i o n to render the a d v e r b i a l

phrase 'kata sumbebekos% m o r e l i t e r a l l y 'by, or i n v i r t u e of, a coincidence'
( t r a d i t i o n a l l y 'per accidens'), for as A r i s t o t l e tells us at Δ 18. 1 0 2 2 ^ 1 9 - 2 0
'kata' indicates some k i n d ofcause. W h e n a is φ b y v i r t u e o f a coincidence,
the coincidence 'causes' a to be φ b y b e i n g the m e d i u m t h r o u g h w h i c h a's
b e i n g Ф is d e r i v e d f r o m something else's being φ. T h e coincidence thus
explains a's b e i n g φ, a n d contrasts w i t h the case where a's being "can­
not be m a d e p l a i n separately", sc. f r o m a {Metaphysics Z 5. 1 0 3 0 ^ 2 4 - 5 ) .
T h e simplest k i n d o f d e r i v a t i o n occurs w h e n a is φ t h r o u g h c o i n c i d i n g
i n some b t h a t is φ; b u t t w o complications m a y arise, i n d i v i d u a l l y o r
together, φ m a y be r e l a t i o n a l , i.e. φ to some c: this adds the f u r t h e r
possibilities t h a t a is φ to c coincidentaUy t h r o u g h b e i n g φ to some d t h a t
c coincides i n , or ( c o m b i n i n g the t w o ) t h a t a is φ to c t h r o u g h c o i n c i d i n g
i n some b t h a t is φ to some d t h a t c coincides i n . T h e other c o m p l i c a t i o n
is t h a t a itself m a y be a coincidence, call i t a^a^: t h e n a^a^ m a y be φ
t h r o u g h a^'s c o i n c i d i n g i n a^ a n d a^'s b e i n g φ, or conversely, or t h r o u g h
each oia^ a n d a^'s c o i n c i d i n g i n some b t h a t is φ; a n d o f c o u r s e the same
t h i n g can h a p p e n w i t h c, or w i t h b o t h a a n d c.
These books o f the Metaphysics present us w i t h four chief instances o f
being Ф coincidentaUy: {a) being, {b) u n i t y , {c) i d e n t i t y , a n d {d) causa­
t i o n , {a) I n the simple c a s e — a c c o r d i n g to the N o t e s — a is coincidentaUy
w h e n a coincides i n some b t h a t is. T h e r e m a y be the c o m p l i c a t i o n t h a t a
is itself a coincidence, (b) C o i n c i d e n t a l u n i t y introduces b o t h o f the
complications, for as the N o t e shows (p. 1 3 3 ) Aristotle's examples seem
to i n c l u d e cases where the predicate is r e l a t i o n a l {a is one with c ) , cases
where the subject is a coincidence {ab is one), a n d cases h a v i n g b o t h
features {ab is one with c). I t has to be a d m i t t e d t h a t i n c o p i n g w i t h these
examples he does n o t always e x p l a i n c o i n c i d e n t a l u n i t y t h r o u g h coinci­
dence i n s o m e t h i n g that is one. {c) C o i n c i d e n t a l sameness w i U a d m i t m a n y
o f the same analyses as c o i n c i d e n t a l u n i t y since, i f the remarks above are
correct, we can a p p l y to this k i n d o f sameness, a n d to relational oneness,
Aristotle's c o n t e n t i o n t h a t "sameness is a . . . oneness . . . o f the b e i n g o f
m o r e t h a n one t h i n g " {Δ 9 . 1 0 1 8 ^ 7 - 8 ) . {d) C o i n c i d e n t a l causation
exhibits a s i m i l a r v a r i e t y (see [ 1 2 0 ] p. 7 9 ) . F o r example the statuemaker

221
 METAPHYSICS

Polyclitus coincidentaUy m a d e a n eyesore i f b e i n g Polyclitus coincides i n

a statuemaker w h o m a d e a statue i n w h i c h a n eyesore coincides. I t is
w o r t h n o t i c i n g t h a t these accounts o f Aristotle's m e a n i n g w o u l d p e r m i t
h i m t o say t h a t the pale, or the pale Socrates, is coincidentaUy a
substance; b u t he never says such things.

^ 3 : DETERMINISM

Metaphysics E chapter 3 is short, d i f f i c u l t , a n d n o t clearly connected t o its

context. I t deals w i t h , or is relevant t o , d e t e r m i n i s m i n a l l the c o m m o n
acceptations o f t h a t n a m e (often assumed w i t h o u t a r g u m e n t t o be
e q u i v a l e n t ) , viz. 4 h e thesis t h a t e v e r y t h i n g is always necessary ( i t was
never possible t h a t i t should have been otherwise)', 4 h e thesis t h a t
e v e r y t h i n g has a n antecedent cause', a n d 'the thesis t h a t e v e r y t h i n g has
a n antecedent necessitating cause ( w h i c h makes i t necessary f r o m the
t i m e w h e n the cause is necessary)'. T h e Notes conclude ( p . 1 9 8 ) t h a t
" t h e chapter has n o t yet received a satisfactory i n t e r p r e t a t i o n " ; a n d
despite the appearance o f m a n y new studies (see B i b l i o g r a p h y ) since
those words were w r i t t e n , I believe they are still t r u e i n 1 9 9 2 .
Recent w o r k has tended to agree w i t h the Notes t o the extent o f
f a v o u r i n g a n i n t e r p r e t a t i o n o f the a r g u m e n t w h i c h embodies \e) below;
b u t I a m n o w m o r e sceptical. A c c o r d i n g to the Notes, the "thesis o f the
c h a p t e r " (p. 1 9 7 ) is t h a t there are causes that take no time in coming and ceasing
to be; a n d the chapter's first p a r a g r a p h argues t h a t otherwise, since
{e) e v e r y t h i n g t h a t does take t i m e i n c o m i n g a n d ceasing to be is itself
n o n - c o i n c i d e n t a l l y caused,

i n d e t e r m i n i s m i n the sense
(a) n o t e v e r y t h i n g is necessary a l l the t i m e (p. 1 9 6 )
w o u l d be r u l e d o u t , i.e. d e t e r m i n i s m i n the sense o f the n e g a t i o n o f {a)
w o u l d foUow. A l t h o u g h aU t h a t w o u l d foUow d i r e c t l y is t h a t every cause
is itself n o n - c o i n c i d e n t a l l y caused, the Notes show h o w to d i s t i l f r o m the
second p a r a g r a p h a f u r t h e r premiss
{b) e v e r y t h i n g is necessary f r o m the t i m e w h e n i t is caused,
w h i c h p e r m i t s a n o n w a r d step towards d e t e r m i n i s m . I t m a y d o so i n
either o f t w o ways: if, as the Notes suggest, 'non-coincidentaUy' i n {e)
means 'at aU times', t h e n {e) a n d {b) y i e l d t h a t e v e r y t h i n g caused w o u l d
be necessary at aU times; perhaps m o r e p l a u s i b l y , the same result can be
reached i f {b) is replaced b y
{bi) e v e r y t h i n g is necessary f r o m the t i m e w h e n i t has a causal
ancestor.

222
 FURTHER COMMENTS
w h i c h l i k e {b) foUows f r o m principles t h a t A r i s t o t l e m a y accept ('every­
t h i n g is necessary f r o m the t i m e i t happens', a n d 'causes necessitate', i.e.
any necessity they possess at a t i m e is t r a n s m i t t e d to their effects at t h a t
same d m e ) . A l t h o u g h this result is stiU n o t the universal d e t e r m i n i s m
w h i c h A r i s t o t l e claims to extract, a n d rejects as absurd b y (fl), i t is at
least u n c o m f o r t a b l y close to t h a t a b s u r d i t y . F o r the shape o f the a r g u ­
m e n t see [ 1 2 7 ] p p . 4 9 - 5 0 .
M o r e pressing t h a n quibbles a b o u t the merits o f this proof, m o r e
pressing, i n i t i a U y , even t h a n doubts whether i t captures Aristotle's mean­
i n g , is the question w h a t its p o i n t is. M a n y commentators have discerned
a h i d d e n message i n the chapter, different f r o m the 'thesis' i t announces.
T h e a u t h o r o f K 8 , a d m i t t e d l y a f l o u n d e r i n g i n t e r p r e t e r o f E 3 as a
whole, begins his s u m m a r y b y representing the " o r i g i n s a n d causes"
u n d e r e x a m i n a t i o n as causes o f " w h a t is coincidentaUy" {KS. 1065^6, c f
[ 1 2 3 ] p . 1 2 9 ) ; a n d this has generated the suggestion ( [ 7 ] 4 5 3 . 1 2 - 1 3 )
t h a t the chapter's business is to prove Aristotle's statement i n E 2.
1 0 2 7 ^ 7 - 8 t h a t " o f things t h a t are or are c o m i n g to be coincidentaUy the
cause is also coincidentaUy". A c c o r d i n g to Sorabji however, " t h e promise
made at t h e b e g i n n i n g " o f t h e chapter is " t o show t h a t accidents can
serve as causes" ( [ 1 1 2 ] p. 9 ) ; w h i l e W i U i a m s ( [ 1 2 6 ] p. 1 8 3 ) sees " t h e
chapter as a w h o l e [as] concerned to r e f u t e " causal d e t e r m i n i s m , i.e.
"every event is d e t e r m i n e d b y some p r i o r cause." These judgements
share, i f U t d e else, a n urge to p l a y d o w n the n o d o n o f taking time to come or
cease to be. Y e t the chapter's thesis appears to e m b o d y t h a t n o t i o n .
T h i s d i f f i c u l t y i n discerning a purpose for E 3 has helped to m o t i v a t e
several different i n t e r p r e t a t i o n s o n w h i c h I n o w c o m m e n t .
First, some have p o i n t e d o u t t h a t {b) is w h a t makes {a) h a r d to
sustain, a n d , they say, {b) is n o t A r i s t o t e l i a n ( [ 1 2 4 ] ) . B u t I stick b y the
c l a i m i n the Notes t h a t {b) is present i n this chapter, however successful
A r i s t o t l e m a y have been i n resisting its seductions elsewhere i n his works
(see e.g. [ 1 1 2 ] ch. 2 ) .
Secondly, {e) is a hypothesis o f u n c e r t a i n status, i n t r o d u c e d b y ' i f ' b u t
expressing no m o o d because c o n t a i n i n g no finite verb. Instead o f assum­
i n g , as the Notes d o , t h a t A r i s t o d e endorses i t ('if, as surely is the case'),
we m i g h t therefore construe i t as a consequence o f rejecting the chapter's
thesis ('if, as w o u l d t h e n be the case'), y i e l d i n g the structure: otherwise
{e) w o u l d foUow, a n d d e t e r m i n i s m i n the sense o f d e n y i n g {a) w o u l d
foUow f r o m t h a t (see [ 1 2 3 ] ) . O f the t w o inferences thereby separated, the
second comes easily v i a ( ^ i ) , i f {e) amounts to saying t h a t e v e r y t h i n g
(interesting) has some antecedent cause; b u t i t is q u i t e unclear h o w this,
asserting the existence o f causes, c o u l d be i m a g i n e d to foUow f r o m
rejection o f the chapter's thesis, w h i c h itself asserts the existence o f c e r t a i n
kinds ofcause.
T h i r d l y , then, we m i g h t foUow WiUiams [ 1 2 6 ] i n d e n y i n g t h a t the

223
 METAPHYSICS

chapter deals w i t h the n o t i o n o f t a k i n g t i m e or b e i n g i n process. W i l l i a m s '

assault o n this t r a d i t i o n (a t r a d i t i o n also affecting e.g. Metaphysics Z 15.
1 0 3 9 ^ 2 3 - 4 , H 3 . 1 0 4 3 ^ 1 4 - 1 6 ) is i n three movements, ( i ) H e is able, as
we have seen, t o offer a n a t t r a c t i v e a l t e r n a t i v e i n E 2. 1 0 2 6 ^ 2 1 - 4 . ( 2 )
Hypothesis [e), i f A r i s t o d e endorses i t , must somehow restrict the things
t h a t have a cause, since {a) a n d {b 1 ) together e n t a i l t h a t n o t a l l d o . B u t
the r e s t r i c t i o n need n o t be, as {e) assumes, to things t h a t take t i m e i n
c o m i n g or ceasing to be; for Aristotle's w o r d order allows or even favours
a different t r a n s l a t i o n , w h i c h puts " n o n - c o i n c i d e n t a l l y " w i t h the subject,
giving

(g4) e v e r y t h i n g t h a t comes a n d ceases to be n o n - c o i n c i d e n t a l l y has a

cause

( w h a t these things are t h a t come a n d cease t o be n o n - c o i n c i d e n t a l l y we

do n o t k n o w , b u t possibly the same as E 2's items t h a t come a n d cease to
be, speaking b a l d l y ) . ( 3 ) I f {e) i n the f o r m o f (^4) were still t o j o i n w i t h [a)
and ( ^ i ) i n p r o v i n g Aristotle's thesis i n the chapter b y the r o u t e so far
defended, t h a t thesis must n o w assert t h a t some causes come a n d cease to
be o n l y coincidentaUy. W i l l i a m s ' r e a d i n g o f the chapter's first sentence is
not this, however, b u t , i n effect, 'some causes come a n d cease to be'.
Stressing the " a b l e " i n m y t r a n s l a d o n o f the Greek adjectives 'able-to-
come-to-be' a n d 'able-to-be-destroyed', a feature w h i c h need n o t be read
i n t o t h e i r m e a n i n g a n d w h i c h the Notes i n the event ignore, W i l l i a m s
takes the sentence to assert t h a t there are causes w h i c h , at some times,
are m e r e l y p o t e n t i a l a n d n o t actualized. Aristotle's route to the chapter's
thesis, so understood, w i U n o w be as foUows: otherwise, since e v e r y t h i n g
is necessary f r o m the t i m e i t happens, aU causes w o u l d be necessary at aU
times, whence b y {b) e v e r y t h i n g caused w o u l d be necessary at aU times,
whence b y (^4) e v e r y t h i n g t h a t comes to be n o n - c o i n c i d e n t a l l y w o u l d be
necessary at aU times, w h i c h , i f stiU n o t exactly the d e t e r m i n i s m t h a t
A r i s t o t l e claims to extract, is once m o r e close enough to i t to be
t h o r o u g h l y embarrassing.
W h a t , t h e n , is the message o f t h e chapter? ( ^ i ) c o m m i t s A r i s t o t l e to
the v i e w t h a t there are times w h e n c o n t i n g e n c i e s — I m e a n , w h a t e v e r is
not always necessary—do n o t yet have causal ancestors; a n d {a) c o m m i t s
h i m to the existence ofcontingencies. T h u s a l t h o u g h a contingency is n o t
d e b a r r e d f r o m h a v i n g a cause or a c h a i n o f causes, nevertheless because
every cause acquires necessity at the t i m e i t occurs, a n d at the same
t i m e , b y ( ^ i ) , transmits t h a t necessity t o its f u t u r e descendants, a n y
c h a i n l e a d i n g to a contingency must have a first m e m b e r , itselfuncaused,
and the first m e m b e r (like the others) must at some earlier t i m e have
been p o t e n t i a l b u t n o t a c t u a l , i.e. must have come to be. M o r e o v e r , i f
(^4) is r i g h t , aU uncaused things t h a t come a n d cease to be must d o so
coincidentaUy; so contingencies, i f caused at aU, must have as t h e i r first

224
 FURTHER COMMENTS

cause s o m e t h i n g t h a t comes a n d ceases to be coincidentaUy. Pseudo-

Alexander's discussion o f the example o f thirst l e a d i n g to v i o l e n t d e a t h i n
Aristotle's second p a r a g r a p h alludes to one Nicostratus w h o feU i n t o
enemy hands w h e n he w e n t o u t o f t o w n to get a d r i n k ( [ 7 ] 4 5 4 . 3 5 - 8 ) . I t
m a y be r i g h t to use this as grounds for i m p u t i n g to E 3 itself the v i e w
t h a t o n l y chance happenings like the encounter w i t h enemy soldiers can
be the first causes o f c o n t i n g e n t outcomes, a n d t h a t the t h i n k e r w h o
traced Nicostratus' d e a t h back to a decision to eat salty food is presented
b y A r i s t o t l e as mistaken. C o i n c i d e n t a l causes p l a y no p a r t i n the a r g u m e n t
so i n t e r p r e t e d , a l t h o u g h i t is doubtless t e m p t i n g to concede o n Aristotle's
b e h a l f t h a t Nicostratus' decision to eat salty food d i d cause his d e a t h i n a
w a y , viz. coincidentaUy, i n the sense t h a t i t caused something, his
presence outside the t o w n , w h i c h b y coincidence w i t h something else
caused his d e a t h . A U this deserves the c o m m e n t , however, t h a t i f i t is
Aristotle's view, he needs to p o n d e r f u r t h e r ; for some events can be traced
back to a n agent's earUer decision to b r i n g t h e m about, yet these events
are c o n t i n g e n t too (see [ 1 1 2 ] p p . 2 2 8 - 9 , 2 3 8 ) — i n d e e d they are n o t
necessary even after the t i m e o f the decision w h i c h causes t h e m , c o n t r a r y
to {b). T h e r e are objections o f a different k i n d i n [ 1 2 5 ] p p . 3 1 4 - 1 6 .
I n any case I i n c l i n e to agree w i t h those commentators w h o say t h a t
the m a i n lesson o f the chapter is t h a t uncaused causes must exist, i f there is
to be a n escape f r o m d e t e r m i n i s m : " i t [sc. the causal ancestry o f the
specimen d e a t h ] runs as far as some o r i g i n , b u t this no f u r t h e r to
a n y t h i n g else" ( 1 0 2 7 ^ 1 1 - 1 2 ) . T h e thesis o f t h e chapter w i U e m b o d y this
result i f " o r i g i n " i n i t means 'uncaused cause'. T o be sure, the thesis goes
f u r t h e r , i n its obscure c h a r a c t e r i z a t i o n o f these origins. W e wish the
c h a r a c t e r i z a t i o n either to m a k e t h e m i m p e r m a n e n t , h a v i n g t h e i r o w n
c o m i n g ( a n d ceasing) to be, or, i n p a r t i c u l a r , to make t h e m things t h a t
come a n d cease to be coincidentaUy; a n d perhaps i t can be i n t e r p r e t e d
so as to i m p o r t one o r b o t h o f those features. A t any rate, the former at
least o f t h e m is i n fact, o n Aristotle's assumptions i n the chapter, p a r t o f
the price ofescape.

225
 BIBLIOGRAPHY
T h i s list, revised a n d enlarged for the second e d i t i o n , retains most o f the
items i n the B i b l i o g r a p h y o f the first e d i t i o n , i n c l u d i n g a l l works referred
to i n the Notes.

T E X T S AND T R A N S L A T I O N S

T h e text translated is i n :
[ i ] W . J a e g e r (ed.), Aristotelis Metaphysica, O x f o r d Classical Texts, Claren­
d o n Press, O x f o r d , 1 9 6 7 .

Sir D a v i d Ross's e d i t i o n , c o n t a i n i n g i n t r o d u c t i o n , text, analysis, a n d

c o m m e n t a r y , is indispensable:
[ 2 ] W . D . Ross (ed.), Aristotle's Metaphysics, C l a r e n d o n Press, O x f o r d , ist
edn. 1 9 2 4 , corrected 1 9 5 3 .

Ross h i m s e l f t r a n s l a t e d the Metaphysics i n v o l . v i i i ( 2 n d edn. 1 9 2 8 ) of:

[ 3 ] J . A . S m i t h a n d W . D . Ross (edd.), The Works of Aristotle translated
into English, C l a r e n d o n Press, O x f o r d , 1 9 1 0 - 5 2 .

Ross's t r a n s l a t i o n is reissued w i t h revisions i n v o l . i i of:

[4] J . Barnes (ed.), The Complete Works of Aristotle, Princeton U n i v e r s i t y
Press, Princeton, 1 9 8 4 .

A l t h o u g h Ross has n o t been superseded, a good alternative translation

is:
[ 5 ] H . G . Apostle, Aristotle's Metaphysics, trans, w i t h commentaries a n d
glossary, I n d i a n a U n i v e r s i t y Press, B l o o m i n g t o n a n d L o n d o n , 1 9 6 6 .

COMMENTARIES

T h e ancient commentators o n these books are edited i n :

[ 6 ] Commentaria in Aristotelem Graeca, R e i m e r , B e r l i n , 1 8 8 2 - 1 9 0 9 ,

w h i c h contains:

227
 METAPHYSIGS

[ 7 ] A l e x a n d e r o f A p h r o d i s i a s ( л в c. 2 0 0 ) , ln Aristotelis Metaphysica Commen­

taria, ed. M . H a y d u c k , [ 6 ] v o l . i , 1 8 9 1 ;
[8] Asclepius (ло 6 t h c e n t . ) , ln Aristotelis Metaphysicorum Libros A-Z
Commentaria, ed. M . H a y d u c k , [ 6 ] v o l . v i . 2 , 1 8 8 8 ;
[9] Syrianus ( d . AD 4 3 7 ) , In Aristotelis Metaphysica Commentaria, ed, J.
K r o l l , [ 6 ] vol. v i . i , 1902.

T r a n s l a d o n s i n t o E n g l i s h o f the p o r d o n s o f Alexander's c o m m e n t a r y o n
Г a n d Δ are i n progress u n d e r the e d i t o r s h i p o f R . R . K . S o r a b j i .
T r a n s l a t i o n s o f [ 8 ] a n d [ 9 ] are p l a n n e d i n the same series.

T h e c o m m e n t a r y o f T h o m a s A q u i n a s ( f . 1 2 2 5 - 7 4 ) is available i n :
[ 1 0 ] A q u i n a s , In xii Libros Metaphysicorum Aristotelis Expositio, ed. M.-R.
G a t h a l a a n d R . M . Spiazzi, M a r i e t t i , T u r i n , 1 9 5 0 ;
a n d is translated as:
[11] Aquinas, Commentary on the Metaphysics of Aristotle, trans. J . P.
R o w a n , H e n r y ^ g n e r y Go., Ghicago, 1 9 6 1 .

O f m o d e r n commentaries a p a r t f r o m [ 2 ] , a recent appearance is:

[12] B. Gassin a n d M . N a r c y , La decision du sens: le livre Gamma de la
Metaphysique d'Aristote: introduction, texte, traduction et commentaire, Vrin,
Paris, 1 9 8 9 .

T h e r e is relevant m a t e r i a l i n some o t h e r volumes o f the Glarendon

A r i s t o t l e Series:
[ 1 3 ] J . L . A c k r i l l , Aristotle's Categories and De Interpretatione, trans, w i t h
notes, G l a r e n d o n Press, O x f o r d , 1 9 6 3 ;
[ 1 4 ] D . W . Yi2iYalyn,Aristotle'sDeAnimaBooksIIandIII, trans, w i t h i n t r o d u c -
t i o n a n d notes, 1 9 6 8 , reissued w i t h a d d i t i o n a l m a t e r i a l b y G. Shields
1993;
[ 1 5 ] W . G h a r l t o n , Aristotle's Physics BooksIand I I , trans, w i t h i n t r o d u c t i o n
a n d notes, 1 9 7 0 , reissued w i t h a d d i t i o n a l m a t e r i a l 1 9 9 2 ;
[ 1 6 ] J . Barnes, Aristotle's Posterior Analytics, trans, w i t h notes, 1 9 7 5 , 2 n d
edn. 1 9 9 3 ;
[ 1 7 ] J . E. A n n a s , Aristotle's Metaphysics Books M and N, trans, w i t h
i n t r o d u c t i o n a n d notes, 1 9 7 6 ;
[ 1 8 ] C. J . F. W i U i a m s , Aristotle's De Generatione et Corruptione, trans, w i t h
notes, 1 9 8 2 ;
[ 1 9 ] E, L . Hussey, Aristotle's Physics Books IIIand IV, trans, w i t h i n t r o d u c -
t i o n a n d notes, 1 9 8 3 ;
[ 2 0 ] D . Bostock, Aristotle's Metaphysics Books Z and H, trans, w i t h i n t r o d u c ­
tion and commentary, forthcoming.

228
 BIBLIOGRAPHY

GENERAL

A m o n g general books o n A r i s t o t l e , consult especially:

[ 2 1 ] J . L . A c k r i l l , Aristotle the Philosopher, O x f o r d U n i v e r s i t y Press, O x f o r d ,
1981;
[ 2 2 ] J . Barnes, Aristotle, Past Masters, O x f o r d U n i v e r s i t y Press, O x f o r d ,
1982.

A n older book is:

[ 2 3 ] W . D . Ross, Aristotle, M e t h u e n , L o n d o n , 5 t h edn. 1 9 4 9 .

Indispensable fbr the student o f A r i s t o t l e w h o has Greek:

[ 2 4 ] H . B o n i t z , IndexAristotelicus, B e r l i n , 1 8 7 0 , reissued D a r m s t a d t , 1 9 5 5 .

C O L L E C T I O N S

GoUections c o n t a i n i n g i m p o r t a n t articles w h i c h bear o n ΓΔΕ include:

[ 2 5 ] J . Barnes, M . Schofield, a n d R. R. K . Sorabji ( e d d . ) , Articles on
Aristotle, i . Science, D u c k w o r t h , L o n d o n , 1 9 7 5 ;
[ 2 6 ] J . Barnes, M . Schofield, a n d R. R. K . Sorabji (edd.), Articles on
Aristotle, i i i . Metaphysics, D u c k w o r t h , L o n d o n , 1 9 7 9 ;
[ 2 7 ] D . J . 0 ' M e a r a (ed.), Studies in Aristotle, Gatholic U n i v e r s i t y o f
A m e r i c a Press, W a s h i n g t o n D G , 1 9 8 1 ;
[ 2 8 ] M . Schofield a n d M , G. N u s s b a u m (edd.), Language and Logos,
G a m b r i d g e U n i v e r s i t y Press, G a m b r i d g e , 1 9 8 2 ;
[ 2 9 ] F. J . Pelletier a n d J . K i n g - F a r l o w (edd.), New Essays on Aristotle,
Canadian Journal of Philosophy, suppl. v o l . x, U n i v e r s i t y o f Galgary
Press, G a l g a r y , 1 9 8 4 ;
[ 3 0 ] J . W i e s n e r (ed.), Aristoteles: Werk und Wirkung, i . Aristoteles und seine
Schule, de G r u y t e r , B e r l i n a n d N e w Y o r k , 1 9 8 5 ;
[ 3 1 ] G . E. L , O w e n , Logic, Science and Dialectic, Duckworth, London,
1986;
[32] M . Frede, Essays in Ancient Philosophy, Glarendon Press, O x f o r d ,
1987;
[ 3 3 ] R. L , J u d s o n (ed.), Aristotle's Physics: A Collection of Essays, Glarendon
Press, O x f o r d , 1 9 9 1 .

T H E N A T U R E A N D SCOPE OF METAPHYSICS

O n the n a t u r e o f A r i s t o t e l i a n metaphysics see:

[ 3 4 ] P. M e r l a n , O n the T e r m s " M e t a p h y s i c s " a n d "Being-qua-being"',

229
 METAPHYSIGS

The Monist l i i , 1968, pp. 174-94; repr. i n his Kleine philosophische

Schriften, Hildesheim, 1976, pp. 2 3 8 - 5 8 ;
[ 3 5 ] G. H . K a h n , ' O n t h e I n t e n d e d I n t e r p r e t a t i o n o f A r i s t o t l e ' s M e t a ­
physics', i n [ 3 0 ] ;
[ 3 6 ] A . Gode, 'Metaphysics a n d L o g i c ' , i n M . M a t t h e n (ed.), Aristotie
Today: Essays on Aristotle's Ideal of Science, Academic Printing and
Publishing, E d m o n t o n , 1987.

T h e Status ofmetaphysics as a 'science' ( " d i s c i p l i n e " ) is e x a m i n e d i n :

[ 3 7 ] Τ . Η . I r w i n , 'Aristotle's Discovery o f Metaphysics', Review of Meta-
physicsxxxi, 1977-8, pp. 210-29,

a n d f u r t h e r i n ch. 8 of:
[ 3 8 ] T . H . I r w i n , Aristotle's First Principles, G l a r e n d o n Press, O x f o r d ,
1988.

I r w i n ' s [ 3 7 ] led to:

[ 3 9 ] A . Code, 'Aristotle's I n v e s t i g a t i o n o f a Basic L o g i c a l P r i n c i p l e :
W h i c h Science Investigates the P r i n c i p l e o f N o n - C o n t r a d i c t i o n ? ' ,
Canadian Journal of Philosophy xvi, 1986, pp. 3 4 1 - 5 7 ;
[ 4 0 ] S. M . C o h e n , ' A r i s t o t l e o n t h e P r i n c i p l e o f N o n - C o n t r a d i c t i o n ' ,
Canadian Journal of Philosophy x v i , 1986, pp. 359—70;
[ 4 1 ] M . F u r t h , ' A N o t e o n Aristotle's P r i n c i p l e o f N o n - C o n t r a d i c t i o n ' ,
Canadian Journal of Philosophy x v i , 1986, pp. 3 7 1 - 8 1 .

Recent debate o n i n t e r p r e t i n g ' b e i n g qua b e i n g ' c a n be followed i n [ 3 4 ]

and:

[ 4 2 ] J . O w e n s , The Doctrine of Being in the Aristotelian 'Metaphysics': A Study

in the Greek Background of Mediaeval Thought, Pontifical Institute of
M e d i a e v a l Studies, T o r o n t o , 1 9 5 1 , 2 n d edn. revised 1 9 6 3 ;
[ 4 3 ] P. M e r l a n , " ^ 6v u n d πρώτη ονσία: Postskript z u einer Be-
sprechung'',PhilosophischeRundschauvn, 1959,pp. 148-53;
[ 4 4 ] G . Patzig, 'Theologie u n d O n t o l o g i e i n d e r " M e t a p h y s i k " des
Aristoteles', Kant-Studien l i i , 1 9 6 0 - 1 , p p . 1 8 5 - 2 0 5 ; t r a n s . J , M . a n d J .
Barnes as ' T h e o l o g y a n d O n t o l o g y i n Aristotle's Metaphysics', with
postscript, i n [ 2 6 ] ;
[ 4 5 ] W . Leszl, Aristotle's Conception of Ontology, A n t e n o r e , Padua, 1 9 7 5 ;
[ 4 6 ] J . G . Stevenson, 'Being qua B e i n g ' , Apeiron i x , 1 9 7 5 , p p . 4 2 - 5 0 ;
[47] K. Brinkmann, Aristoteles Allgemeine und Spezielle Metaphysik, de
Gruyter, Berlin, 1979;
[ 4 8 ] G . R e a l e , The Concept of First Philosophy and the Unity of the Metaphysics
ofAristotle, t r a n s . J . R . G a t a n , S U N Y Press, A l b a n y , 1 9 8 0 ;
[ 4 9 ] J · Owens, ' I s there A n y O n t o l o g y i n Aristotie?', Dialogue xxv, 1986,
pp. 697-707;

230
 BIBLIOGRAPHY

[ 5 0 ] Μ . Frede, ' T h e U n i t y o f General a n d Special Metaphysics: A r i s t o t ­

le's C o n c e p t i o n o f M e t a p h y s i c s ' , i n [ 3 2 ] , p p . 8 1 - 9 5 ;
and especially:
[ 5 1 ] J . W . T h o r p , 'Does P r i m a c y Confer U n i v e r s a l i t y : L o g i c a n d T h e o ­
logy i n A r i s t o t l e ' , Apeiron x x i i , 1 9 8 9 , p p . 1 0 1 - 2 5 .

T H E T H I N G S - T H A T - A R E AND T H E I R
INTERRELATIONS

I m p o r t a n t c o n t r i b u t i o n s t o u n d e r s t a n d i n g Aristotle's metaphysical views

i n Г are contained i n :

[ 5 2 ] G . E . L . O w e n , ' L o g i c a n d Metaphysics i n Some E a r l i e r W o r k s o f

A r i s t o t l e ' , i n I . D ü r i n g a n d G . E . L . O w e n (edd.), Aristotle andPlato in
the Mid-fourth Century, G ö t e b o r g , 1960; repr. i n [ 2 6 ] a n d [ 3 1 ] ;
[ 5 3 ] G . E . L . O w e n , ' A r i s t o t l e o n t h e Snares o f O n t o l o g y ' , i n R .
B a m b r o u g h (ed.), New Essays on Plato and Aristotle, Routledge &
K e g a n P a u l , 1 9 6 5 ; repr. i n [ 3 1 ] ;
[54] E . L . O w e n , ' T h e Platonism o f A r i s t o t l e ' , Proceedings of the British
Academy l i , 1 9 6 5 ; repr. i n P. F. Strawson (ed.), Studies in the Philosophy of
Thought andAction, O x f o r d U n i v e r s i t y Press, O x f o r d , 1 9 6 8 , a n d i n [ 2 5 ]
and [ 3 1 ] ;
and developed i n :
[ 5 5 ] C. H , K a h n , ' W h y Existence does n o t Emerge as a D i s t i n c t
Concept i n Greek Philosophy', Archiv für Geschichte der Philosophie l v i i i ,
1976, pp. 3 2 3 - 3 4 ;
[ 5 6 ] Μ . Τ . Ferejohn, ' A r i s t o t l e o n Focal M e a n i n g a n d the U n i t y o f
Science', Phronesis xxv, 1980, pp. 117-28;
[57] R. M . Dancy, ' O n Some o f Aristotle's First T h o u g h t s about
Substances', PhilosophicalReviewlxxxiv, 1975,pp. 338-73.

Some foundations h a d been l a i d i n the presumably earlier Categories a n d

De Interpretatione, o n w h i c h see [ 1 3 ] . F o r the linguistic b a c k g r o u n d as
seen b y a philosopher:
[58] C. H . K a h n , The Verb 'Be' in Ancient Greek, part v i o f J . W . M .
V e r h a a r (ed.), The Verb 'Be' and its Synonyms, R e i d e l , D o r d r e c h t , 1 9 7 3 .

O n the several senses o f ' b e i n g ' :

[59] F. B r e n t a n o , Von der mannigfachen Bedeutung des Seienden nach Aristo­
teles, F r e i b u r g , 1 8 6 2 , trans. R . George as On the Several Senses of Being in
Aristotle, U n i v e r s i t y o f C a l i f o r n i a Press, Berkeley a n d Los Angeles,
1975;

231
 METAPHYSIGS

[ 6 o ] M . J . L o u x , ' A r i s t o t l e o n the Transcendentals', Phronesis x v i i i , 1 9 7 3 ,

pp. 2 2 5 - 3 9 ;
[ 6 1 ] D . W . H a m l y n , ' F o c a l M e a n i n g ' , Proceedings of the Aristotelian Society
l x x v i i i , 1 9 7 7 - 8 , PP· i - i 8 ;
[ 6 2 ] M . M a t t h e n , 'Aristotle's Semantics and a Puzzle concerning
Change', i n [ 2 9 ] ;
[ 6 3 ] D . R. M o r r i s o n , ' T h e Evidence for Degrees o f B e i n g i n A r i s t o t l e ' ,
Classical Quarterly x x x v i i , 1 9 8 7 , p p . 3 8 2 - 4 0 1 ;
[ 6 4 ] H . P. G r i c e , ' A r i s t o d e o n the M u k i p l i c i t y o f B e i n g ' , Pacific Philo­
sophical Quarterlylxix, 1988, pp. 175-200,

the last m a r r e d b y a n n o y i n g m i s p r i n t s . M o r e generally o n A r i s t o t e l i a n

homonymy:

[ 6 5 ] K . J . J . H i n t i k k a , ' A r i s t o t l e a n d the A m b i g u i t y o f A m b i g u i t y ' ,

Inquiry i i , 1 9 5 9 , p p . 1 3 7 - 5 1 ; repr. w i t h revisions as ch. 1 o f h i s Time and
Necessity: Studies in Aristotle's Theory of Modality, Oxford University
Press, O x f o r d , 1 9 7 3 ;
[ 6 6 ] T . H . I r w i n , ' H o m o n y m y i n A r i s t o t l e ' , Review of Metaphysics xxxiv,
1980-1,pp. 523-44·

Different views οίΔ η, especially 1 0 1 7 ^ 2 2 - 3 0 , are p u t f o r w a r d i n :

[ 6 7 ] R. A . G o b b , ' T h e Present Progressive Periphrasis a n d the M e t a ­

physics o f A r i s t o t l e ' , Phronesis x v i i i , 1 9 7 3 , p p . 8 0 - 9 0 ;
[ 6 8 ] R. K . Sprague, ' A r i s t o t e l i a n Periphrasis: A R e p l y to M r C o b b ' ,
Phronesisxx, 1975, pp. 7 5 - 6 ;
[ 6 9 ] J . W . T h o r p , 'Aristotle's Use o f G a t e g o r i e s ' , Phronesis x i x , 1 9 7 4 , p p .
238-56;
[ 7 0 ] M . Frede, 'Categories i n A r i s t o t l e ' , i n [ 2 7 ] , p p . 1 - 2 4 , repr. i n [ 3 2 ] ,
p p . 2 9 - 4 8 : see p p . 1 5 - 1 9 (41-4);
[ 7 1 ] R. M . D a n c y , ' A r i s t o t l e a n d Existence', Synthese l i v , 1 9 8 3 , p p . 4 0 9 ¬
42.

See also [ 6 4 ] p p . i 8 o - i , a n d c o m p a r e De Generatione et Corruptione I 3-4

a n d the notes ad loc. i n [ 1 8 ] .

O n b e i n g as t r u t h {Δ 7. 1 0 1 7 ^ 3 1 - 5 a n d E / ^ ) see:

[ 7 2 ] M . M a t t h e n , 'Greek O n t o l o g y a n d the " I s " o f T r u t h ' , Phronesis

xxviii, 1983, pp. 113-35;
[ 7 3 ] J · ^ · T h o r p , ' A r i s t o t l e o n B e i n g a n d T r u t h ' , De Philosophia i i i , 1 9 8 2 ,
pp. 1-9.

T h e v i e w t h a t A r i s t o t l e regarded essential predications as statements o f

i d e n t i t y comes f r o m [ 5 4 ] a n d is defended i n the N o t e o n Г 4 . 1007^20
and i n :

232
 BIBLIOGRAPHY

[ 7 4 ] Μ . J . Woods, 'Substance a n d Essence i n A r i s t o t l e ' , Proceedings of the

AristotelianSocietylxxv, 1 9 7 4 - 5 , pp. 167-80;

b u t see the notes o n 73^24ff. a n d 83^37 fF. i n [ 1 6 ] , a n d the dissent i n :

[ 7 5 ] A . Gode, ' O n the O r i g i n s o f Some A r i s t o t e l i a n Theses about
P r e d i c a t i o n ' , i n J . Bogen a n d J . E. M c G u i r e (edd.), How Things Are:
Studies in Predication and the History of Philosophy and Science, Reidel,
Dordrecht, 1985;
[ 7 6 ] H . W e i d e m a n n , ' I n Defence o f Aristotle's T h e o r y o f Predication',
Phronesisxxv, 1980, pp. 76-87.

O t h e r articles referred to:

[77] W. V. Quine, 'On W h a t T h e r e I s ' , i n From a Logical Point of View,
H a r v a r d U n i v e r s i t y Press, G a m b r i d g e , Mass., 1 9 5 3 ;
[ 7 8 ] A . G e w i r t h , 'Aristotle's D o c t r i n e o f Being', Philosophical Review lxii,
1953, PP-577-89;
[ 7 9 ] G . E.. L . O w e n , 'Inherence', Phronesis x, 1 9 6 5 , p p . 9 7 - 1 0 5 ; repr. i n
[31];
[ 8 0 ] L . B r o w n , 'Being i n the Sophist: a Syntactical I n q u i r y ' , Oxford Studies
in Ancient Philosophy i v , 1 9 8 6 , p p . 49-70.

THE PRINCIPLE OF NON-CONTRADICTION

A n early m o d e r n t r e a t m e n t is:
[ 8 1 ] J . Lukasiewicz, ' Ü b e r den Satz des Widerspruchs bei Aristoteles',
Bulletin international de FAcademie des Sciences de Cracovie, 1910, pp. 15¬
3 8 ; trans. V . W e d i n as ' O n the Principle o f G o n t r a d i c d o n i n A r i s t o d e ' ,
Review of Metaphysics x x i v , 1 9 7 1 , p p . 4 8 5 - 5 0 9 ; retrans. J . Barnes as
' A r i s t o t l e o n the L a w o f G o n t r a d i c t i o n ' i n [ 2 6 ] .

O n Г 3 see the g r o u p ofarticles [ 3 7 ] , [ 3 9 ] - [ 4 1 ] , a n d f r o m an earlier date:

[ 8 2 ] J . Barnes, ' T h e L a w o f G o n t r a d i c t i o n ' , Philosophical Quarterly xix,
1969, PP· 3 0 2 - 9 ,
w h i c h is criticized i n :
[ 8 3 ] J . N u t t a l l , 'BeUef, O p a c i t y a n d G o n t r a d i c t i o n ' , Philosophical Quarterly
xxviii, 1978, pp. 2 5 3 - 8 .

T h e 'type 2 ' i n t e r p r e t a t i o n was p u t f o r w a r d b y A n s c o m b e i n :

[ 8 4 ] G . E. M . Anscombe a n d P. T . Geach, Three Philosophers, Blackwell,
Oxford, 1961, pp. 3 9 - 4 5 .

F o r a careful a n d sensitive t r e a t m e n t 0 f r 4 . 1 0 0 5 ^ 1 1 - 1 0 0 7 ^ 1 8 , w i t h text

a n d t r a n s l a t i o n , see:

233
 METAPHYSIGS

[ 8 5 ] R. Μ . D a n c y , Sense and Contradiction, Reidel, Dordrecht, 1975.^

T h e same passage is also e x a m i n e d i n [ 3 6 ] , [ 3 8 ] ch. 9 , a n d :

[ 8 6 ] H . W . Noonan, ' A n A r g u m e n t of Aristotle on Non-Gontradiction',
Analysis xxxvii, 1 9 7 6 - 7 , pp. 1 6 3 - 9 ; repr. i n The Philosopher's Annual,
vol. i i , Blackwell, O x f o r d , 1978, pp. 1 0 7 - 1 5 ;
[ 8 7 ] J . Lear, Aristotle and Logical Theory, G a m b r i d g e U n i v e r s i t y Press,
C a m b r i d g e , 1 9 8 0 , ch. 6 ;
[ 8 8 ] M . V . W e d i n , ' A r i s t o d e o n the R a n g e o f the P r i n c i p l e o f N o n -
C o n t r a d i c t i o n ' , Logique et analyse x x v , 1982, pp. 87-92;
[89] E. Halper, ' A r i s t o t l e o n the Extension o f N o n - C o n t r a d i c t i o n ' ,
History of Philosophy Quarterly i , 1 9 8 4 , p p . 3 6 9 — 8 0 .

T h e n a t u r e o f s i g n i f i c a t i o n is explored i n :
[ 9 0 ] T . H . I r w i n , 'Aristotle's C o n c e p t o f S i g n i f i c a t i o n ' , i n [ 2 7 ] .

T h a t Aristotle's a r m o u r withstands the blows o f m o d e r n critics is argued

in:
[ 9 1 ] T . V . U p t o n , 'Psychological a n d M e t a p h y s i c a l Dimensions o f N o n -
C o n t r a d i c t i o n ' , Review of Metaphysics x x x v i , 1 9 8 2 - 3 , pp. 591-606.

SCEPTICISM A N D R E L A T I V I S M

O n Г 5 see ch. 9 o f [ 3 8 ] a n d :
[ 9 2 ] J . D . G. Evans, ' A r i s t o t l e o n R e l a t i v i s m , ' Philosophical Quarterly x x i v ,
1974, pp. 193-203;
[ 9 3 ] A- J · P· K e n n y , ' T h e A r g u m e n t f r o m I l l u s i o n i n Aristotle's Meta­
physics {Г 1 0 0 9 - 1 0 ) ' , A f m u f l x x v i , 1 9 6 7 , p p . 184-97,

on w h i c h there are comments i n :

[ 9 4 ] M . C. Scholar, ' A r i s t o t l e Metaphysics IV 1 0 1 0 ^ 1 - 3 ' , Mindlxxx, 1971,
pp. 266-8.

See also [ 1 4 ] a n d :
[ 9 5 ] K . Lycos, ' A r i s t o t l e a n d Plato o n " A p p e a r i n g " ' , Mindlxxiii, 1964,
pp. 496-514·

T h e sceptics w h o are Aristotle's target are i d e n t i f i e d as M e g a r i a n precur­

sors o f P y r r h o i n :
[ 9 6 ] E. B e r t i , ' L a c r i t i c a allo Scetticismo nel I V L i b r o della Metafisica',
i n G. G i a n n a n t o n i (ed.), Lo Scetticismo antico, B i b l i o p o l i s , Naples, 1 9 8 1 ,
vol. i , pp. 61-79;

and Aristotle's influence o n later Greek scepticism is traced i n :

234
 BIBLIOGRAPHY

[ 9 7 ] Α . Α . L o n g , ' A r i s t o t l e a n d the H i s t o r y o f G r e e k Scepticism', i n [ 2 7 ] .

A g a i n s t A r i s t o t e l i a n relatives {Δ 15) see p p . 1 9 - 3 1 of:

[ 9 8 ] J . Barnes, 'Scepticism a n d R e l a t i v i s m ' , Philosophical Studies xxxii,
1 9 8 8 - 9 0 , pp. 1-31.

NUMBER, UNITY, IDENTITY

O n numbering:
[99] Frege, Die Grundlagen der Arithmetik, K o e b n e r , Breslau, 1 8 8 4 ;
repr. w i t h E n g l i s h trans. b y J . L . A u s t i n as The Foundations of Arithmetic,
B l a c k w e l l , O x f o r d , 2 n d edn. 1 9 5 3 ;
[ 1 0 0 ] K . R. Popper, ' T h e P r i n c i p l e o f I n d i v i d u a t i o n ' , Proceedings of the
Aristotelian Society, s u p p l . v o l . x x v i i , 1 9 5 3 , p p . 9 7 - 1 2 0 .

O n n u m e r i c a l relatives (Δ 15) there are references o n p. 4 0 o f [ 1 7 ] .

Rejection o f t h e L a w o f t h e I n d i s c e r n i b i l i t y o f I d e n t i c a l s (see p p . 2 0 8 - 1 4 ) ,
a n d thereby i n f i r m grasp o n the n o t i o n o f i d e n t i t y , are i m p u t e d to
Aristotle i n :
[ioi] N . P. W h i t e , ' A r i s t o t l e o n Sameness a n d Oneness', Philosophical
Reviewlxxx, 1971,pp. 177-97,

w h i c h is c r i t i c i z e d i n :
[102] F. D . M i l l e r , J r . , ' D i d A r i s t o t l e H a v e the Goncept o f i d e n t i t y ? ' .
Philosophical Review l x x x i i , 1 9 7 3 , p p . 4 8 3 - 9 0 ;
[103] A . Code, 'Aristotle's Response to Quine's Objections to M o d a l
L o g i c ' , Journal of Philosophical Logic v, 1 9 7 6 , p p . 159-86;
[ 1 0 4 ] G . B. M a t t h e w s , ' A c c i d e n t a l U n i t i e s ' , i n [ 2 8 ] ;
to w h i c h useful b a c k g r o u n d is:
[105] R. Cartwright, 'Identity and Substitutivity', i n M . K. Munitz
(ed.), Identity and Individuation, N e w Y o r k U n i v e r s i t y Press, N e w York,
1971; repr. i n his Philosophical Essays, M I T Press, C a m b r i d g e , Mass.,
1987,
a n d [ 5 7 ] p p . 3 6 1 - 8 . Russell is r e c r u i t e d against the i m p u t a t i o n i n :
[ 1 0 6 ] C . J . F. W i l l i a m s , 'Aristotle's T h e o r y o f D e s c r i p t i o n s ' , Philosophical
Review x c i v , 1 9 8 5 , p p . 6 3 - 8 0 ,

a n d to s i m i l a r purpose the d i s t i n c t i o n i n [ 1 0 5 ] is i n v o k e d i n :
[ 1 0 7 ] F. J . Pelletier, 'Sameness a n d Referential Opacity i n Aristotle',
JVous x i i i , 1 9 7 9 , p p . 2 8 3 - 3 1 1 ;
[ 1 0 8 ] I . A n g e l e l l i , 'Friends a n d O p p o n e n t s o f the S u b s t i t u t i v i t y o f
Identicals i n the H i s t o r y o f L o g i c ' , i n M . S c h i r n (ed.), Studien zu Frege,

235
 METAPHYSICS

i i . Logic und Sprachphilnsophie, Frommann, Stuttgart-Bad Cannstatt,

1 9 7 6 , p p . 1 4 1 - 6 6 (seep. 1 4 6 ) ;

w h i l e a n i n t e r p r e t a t i o n nearer to [ 1 0 3 ] a n d [ 1 0 4 ] is f a v o u r e d i n :

[ 1 0 9 ] F. Д . L e w i s , ' A c c i d e n t a l Sameness i n A r i s t o t l e ' , Philosophical Studies

xlii, 1982, pp. 1-36.

Close t e x t u a l analysis is d e p l o y e d against the i m p u t a t i o n i n :

[ i i o ] M . M i g n u c c i , *Puzzles a b o u t I d e n t i t y : A r i s t o t l e a n d his Greek

Commentators', i n [ 3 0 ] .

A n a t t e m p t to e x p l a i n a n d defend rejection o f the L a w o f the I n d i s c e r n -

i b i U t y o f I d e n t i c a l s a n d its converse is m a d e i n :

[ 1 1 1 ] L . S p e l l m a n , ' R e f e r e n t i a l O p a c i t y i n A r i s t o t l e ' , History of Philosophy

Quarterlyw'u, 1990, pp. 17-32.

CAUSATION, NECESSITY, COINCIDENCE

O n Δ 2 see the p a r a l l e l notes i n [ 1 5 ] , a n d o n A r i s t o t e U a n causation m o r e

generally:

[ 1 1 2 ] R . R, K . S o r a b j i , Necessity, Cause and Blame: Perspectives on Aristotle's

Theory, D u c k w o r t h , L o n d o n , 1 9 8 0 , esp. c h . 2;
[ 1 1 3 ] D . 0 . M . Charles, Aristotle's Philosophy of Action, Duckworth, 1984,
esp. p p . 4 4 - 5 5 ;
[ 1 1 4 ] M . J . W h i t e , 'Causes as Necessary C o n d i t i o n s : A r i s t o t l e , A l e x a n d e r
ofAphrodisias a n d J . L, Mackie', i n [ 2 9 ] , pp. 157-89;
[ 1 1 5 ] J . M . E. M o r a v c s i k , ' W h a t M a k e s R e a l i t y I n t e l l i g i b l e ? Reflections
o n Aristotle's T h e o r y oiAitia', i n [ 3 3 ] , pp. 3 1 - 4 7 ;
[ i i 6 ] C. A . F r e e l a n d , ' A c c i d e n t a l Causes a n d R e a l E x p l a n a t i o n s ' , i n
[ 3 3 ] , pp. 4 9 - 7 2 .

O n Aristotle's t r e a t m e n t o f c o i n c i d e n c e i n the Topics, see:

[ 1 1 7 ] T . E b e r t , 'Aristotelischer u n d t r a d i t i o n e l l e r Akzidenzbegriff'', i n G .
Patzig, Е. Scheibe, a n d W . W i e l a n d ( e d d . ) , Logik, Ethik, Theorie der
Geisteswissenschaften, Felix M e i n e r Verlag, H a m b u r g , 1977;

and on Δ 3 0 . 1025^30:

[ 1 1 8 ] J . E. Tiles, ' W h y the T r i a n g l e has T w o R i g h t Angles Kath' Hauto',

Phronesisxxviii, 1983, pp. i - i 6 .

O n E 2 see [ 1 6 ] p . 1 2 7 , a n d :

[ 1 1 9 ] M . M i g n u c c i , "Ώς inl т6 πολν et necessaire dans l a c o n c e p t i o n

236
 BIBLIOGRAPHY

aristoteIicienne de la science", i n E, B e r d (ed.), Aristotle on Science: The

Posterior Analytics, A n t e n o r e , Padua, 1 9 8 1 ;
[ 1 2 0 ] R. L . J u d s o n , 'Ghance a n d " a l w a y s or for the most p a r t " ', i n [ 3 3 ] ,
pp. 73-99.

A n u m b e r o f a t t e m p t s have been made to understand E 3. Besides [ 1 1 2 ]

ch. I , a n d [ 1 1 6 ] , see:

[121] K . J . J . H i n t i k k a , w i t h U . Remes a n d S. K n u u t t i l a , Aristotle on

Modality and Determinism, Acta Philosophica Fennica xxix, 1977, esp. pp.
101-17;
[ 1 2 2 ] G, Fine, ' A r i s t o t l e o n D e t e r m i n i s m : A R e v i e w o f R i c h a r d Sorabji's
Necessity, Cause, and Blame', Philosophical Review xc, 1 9 8 1 , pp. 5 6 1 - 7 9 ;
[ 1 2 3 ] A . M a d i g a n , 'Metaphysics E 3: A M o d e s t Proposal', Phronesis x x i x ,
1984, pp. 123-36;
[ 1 2 4 ] D . Frede, ' A r i s t o t l e o n the L i m i t s o f D e t e r m i n i s m : A c c i d e n t a l
Causes i n Metaphysics E 3 ' , i n A . G o t t h e l f (ed.), Aristotle on Nature and
Living Things, Bristol Classical Press, 1 9 8 5 , p p . 2 0 7 - 2 5 ;
[ 1 2 5 ] R. E. H e i n a m a n , ' A r i s t o t l e o n Accidents', Journal of the History of
Philosophy xxiii, 1985, pp. 311-24;
[ 1 2 6 ] C. J . F. W i l l i a m s , 'Some C o m m e n t s o n A r i s t o t l e Metaphysics E 2,
3 ' , Illinois Classical Studies x i , 1 9 8 6 , p p . 181-92;
[ 1 2 7 ] H . W e i d e m a n n , 'Aristoteles u n d das P r o b l e m des kausalen Deter­
minismus ( M e t . E 3 ) ' , Phronesis x x x i , 1 9 8 6 , p p . 2 7 - 5 0 ;
[ 1 2 8 ] P. L . D o n i n i , Ethos: Aristotele e il determinismo ( C u l t u r e antiche,
S t u d i e testi 2 ) , Edizione d e l l ' O r s o , Alessandria, 1 9 8 9 , ch. 2.

O T H E R

T h e place οίΔ 8 i n Aristotle's t h i n k i n g about substance is charted i n :

[129] R. Polanski, 'Aristode's T r e a t m e n t oiousia i n Metaphysics V 8',
Southern Journal of Philosophy x x i , 1 9 8 3 , p p . 57-66.

O n Δ 11 see:
[130] J· J · C l e a r y , Aristotle on the Many Senses of Priority, Southern Illinois
U n i v e r s i t y Press, Carbondale, 1 9 8 8 ;
and o n Δ 2 7 :
[131] R. K . Sprague, ' A r i s t o t l e o n M u t i l a t i o n ' , Syllecta Classica 2, 1 9 9 0 ,
pp. 17-22.

O t h e r works referred to:

[132] T . L . H e a t h , Mathematics in Aristotle, C l a r e n d o n Press, O x f o r d ,
1949;

237
 METAPHYSICS
[ 1 3 3 ] C. S. L e w i s , Studies in Words, C a m b r i d g e U n i v e r s i t y Press, C a m ­
bridge, 1960;
[ 1 3 4 ] J . L o c k e , Essay concerning Human Understanding, London, 1 6 9 0 ; ed.
P. H . N i d d i t c h , C l a r e n d o n Press, O x f o r d , 1 9 7 5 ;
[ 1 3 5 ] J . L o c k e , Epistola de Tolerantia, 1 6 8 9 ; trans. W m . Popple as A Letter
concerning Toleration, London, 1689; in TL· Second Treatise of Civil
Government, and A Letter concerning Toleration, ed.J. W . Gough, Blackwell,
Oxford, 1948.

a38
 GLOSSARY

άγνο€Ϊν etc.: be mistaken ctc., ignor­ €νΒ€χ€σθαι; see pp. 156, 159, i6o;
ant 1010^13. 8υνάμ€ΐ: in capacity, potentially.
dci; always, in every case. €ί8ος: form.
ά^ιος: invariable, sec p. 187. €к: out of, from, made up of, see
αίσθάν€σθαι, αΙσθητός, etc.: perccivc, p· 173·
perceptible, etc., see pp. 108, i i i . 4vavTios etc.: contrary ctc., see p.
aiTto, aiTtos: causc, reason, sec p.
152.
124. €νΒ€χ€σθαι: bc possible, admit, see
άΧηθής etc.: true etc.
p. 1 5 9 ; c f Βννασθαι.
άλλοίωσίς etc.: modification ctc., see
€ν4ργ€ΐα etc.: (actual) functioning
p. 108.
etc.; cVcpyet^i: actually.
άντικ€Ϊσθαι etc.: be opposed, opposite,
€W€Ä€X€iV · complete reality,
seep. 151.
c f i ; : state, possession, see pp. 170-1.
άντιφάναί etc.: contradict etc.
€ηιστημη ctc.: disciplinc, knowledge,
afio?, άξιονν: legitimate; ask (for),
ctc., sec p. 76.
count. €Τ€ρος: other, different.
anXovSy άπλώς: simple, baldly. €χ€ΐν: possess, have, attain, bc in a
άποφάναι etc.: deny etc. state, include 1026*3, bc equipped
άριθμός: number {άριθμ€ΐν: reckon with 1005b15, 17, sec p. 173.
1016*' 1 0 ; αριθμητός: countable i: qua, scc p. 76.
1020*9). ήρ€μία, ήρίμησις, ήρ€μ€Ϊν: rest, be at
άρχή: origin, beginning, principle, rcst.
see p. 123. θ€ωρ€ΐν, θ€ωρητικός: study, theoreti­
άφαφ€ΐν: subtract, remove. cal.
βέβαιος: firm. Γδιο;: distinct, distinctive, special.
γίγν€σθαι, γ(ν€σις, etc.: come to be, καθόλου: univcrsal(ly).
(processof) coming to be, etc. καί: and, i.c.
γιγνώσκ€ΐν, γνωστός, etc.: recognize, καταφάναι etc.: affirm etc.
be acquainted with. κατηγορίΐσθαι etc.: bc predicated ctc.
γνωρίζ€ΐν, γνώριμος: understand, Kiv€iv, κιν€Ϊσθαι, ctc.: (effect) change
(make) intelligible, certain, sce etc., move 1010*13, 1012^35.
p, 8 8 . κύριος: fundamental, authoritative.
6ιάθ€σις, ^ιακ€Ϊσθαι: disposition, bc in λ€γ€ΐν: call, so call, say, state, speak of,
a condition, see p. 170. describe, mcan.
διάνοια etc.: thought, thinking, ctc. λόγος: formula, statement, thcsis^
6ιαφορά etc.: difference, differentia, argument, see pp. 9 2 , 179.
ctc. μ€ρος, μόριον, c t c . : part, portion, ctc.,
6iopi^6iv etc.: specify, distinguish. «5cp. 174.
6oKciv: be thought, μ€ταβάΧλ€ΐν c t c . : alter ctc., scc p. 158.
δό^α etc.: opinion ctc. μ€ταξύ: in thc middle (of, between),
bvvaa0ai etc.: bc capable, bc possible, between, as a mcans to.
capacity, etc., havc force 1011*7; c f vo€iv etc.: conccivc ctc., scc p. 199.

239
 METAPHYSICS

vovs: i n t e l l i g e n c e ; νοΰν €χαν: be sane. συμπλ€Κ€ΐν e t c . : c o m b i n e etc.

οΪ€σθαι: consider. συ¥€χης e t c . : continuous etc., see p p .
οίκ€Ϊοζ: p r o p e r , o w n p r o p e r . Ϊ36, 173.
ολος, δλως: w h o l e , w h o U y , i n g e n e r a l , συγκ€ΐσθαι, συνθεσις, e t c . : be c o m ­
8Cep.175. posed, b c c o m p o u n d e d , c o m p o s i ­
dpt^ctv, ώρισμίνος, etc.: define, tion, ctc.
definite, etc. теАаоу: complete, see p . 167.
ούσία: substance, see p p . 1 4 7 - 9 . τ4λος: fulfilment, final state i o i 6 ^ 2 0 ,
πάθος e t c . : affection etc., see p . 171. see p . 167.
noteiv: p r o d u c e , d o , a c t , m a k e , see τ€χνη, τ€χνίτης: s k i l l , man-of-skill.
p . 1 6 5 ; ηοιητικός: p r o d u c t i v e . τρόπος: sense, m a n n e r , see p . 122.
ποιόν, ποιότης: q u a l i f i c a t i o n , q u a l i t y , ΰλη: m a t t e r .
sce p . 162. ύπάρχ€ΐν (τινί): h o l d good (of some­
ποσόν: q u a n t i t y , see p . 160. t h i n g ) , be, be present, h a v e r e a c h e d
πραγμα: a c t u a l t h i n g (subject, fact), i 0 2 i ^ 2 3 ; ύπάρχ€ΐν €v τινι, €νυπάρ-
oiDJect. χ€ΐν τινί: be a constituent of some­
προαίρ€σις e t c . : decision, choice; thing.
deliberately. ύποκ€ίμ€νον: subject, see p . i o i .
πρός: relative to, w i t h reference to, i n ύπολαμβάν€ΐν: believe.
r e l a t i o n to, i n response t o ; πρός τι φάναι e t c . : assert etc.
relative. φαίνεσθαι e t c . : be i m a g i n e d etc.,
πρώτος: first, p r i m a r y , i n i t i a l . e v i d e n t ( l y ) , see p . 105.
σημαίν€ΐν: signify, sec p . 9 4 . φθ€ΐρ€σθαι, φθορά: be destroyed, (pro­
σκ€7ττ€σθαι, σκοπεΐν, e t c . : e x a m i n e , cess of) destruction.
investigate, etc. φρόνησις: w i s d o m , see p . 108.
σοφία: science. φύσις e t c . : n a t u r e etc., see p . 1 2 9 ;
στάσις e t c . : k e e p i n g t h c s a m e etc. πεφυκός e t c . : c h a r a c t e r i s t i c ( a l l y ) ,
στ€ρησις: l a c k , see p . 172. see p . 8 7 .
συλλογισμός c t c . : r e a s o n i n g , c a l c u l a ­ χωριστός: separable.
tion, etc., s c c p . 8 7 . φ€υΒής e t c . : false, be i n e r r o r , see
συμβ€βηκός e t c . : c o i n c i d e n c e , c o i n c i ­ p p . 104, 178.
d e n t a l , etc., sec p p . 7 6 - 7 . ώς eni то πολύ: for the most p a r t .

240
 INDEX LOCORUM
T h i s is a n i n d e x , b y page, t o passages o f A r i s t o t l e referred t o i n the Notes
a n d F u r t h e r C o m m e n t s . I t excludes the m a i n c o m m e n t i n the Notes o n
each passage i n Metaphysics books Г, J , a n d E.

CATEGORIES 9^28-9 172

1^6-12 10^9-10 172
I 94 27-9 162
6-7 96 II^20 167
12-15 216 24-32 167
2 16 199 10
20ff. 96, 155 11^38-12^25 173
20-^9 147, 207 11^38-12^9 Ii9
23-9 148 i2*9-25 "9
3 10-12 167 17 152
16-17 129 26-13^7 170, 172
4 25-2Ч 140 ^26-13^7 173
25 216 13 IG-II 199
2^2 143 29-33 118
4-10 "9 I I 36-14^6 152
8-10 199 i4^22-3 152
5 11-14 155 12 153
12-13 140 26 129
14-19 147, 148 14 b
163
19-34 207 15 12
15 173
29-34 99, 169
3'14 147
22 162 DE INTERPRETATIONE
3i-2 152
24-32 85 I i6Vi8 198
24-7 119> 152 12-18 119
6 160 16-18 94
4^20-5^14 160 2 99
25-6 160 3 22-5 93
5'зВ-^о 80, 162 24 ^ 94
звЛ 145, 169 4 26-7 199
.39 182 28-30 92
11-29 85, 162 6 17^33-7 88
i6-i8 l6l 35 94
6^7 152 7-11 198
19-25 162 7 39-'i 150
7 164 ^8 150
162, 165 22-3 158
b28-7b14 "5 8 118
34-6 165 18^8-27 95
7^22-8^2 165 25 94
35-8 112 9 117, 118, 198
8 162,170 19^23-7 131
8^26-9^13 171 32 117, 118

241
 INDEX LOCORUM

5-20 3 118 4 73^28-g4 193

19-30 97 37- 3 169
20*13 94 "4-5 182
16-19 158 5-10 168
^15-19 135 5-8 100
21V14 ΙΟΙ 10- 16 170
24-8 ii8 11- 12 220
144 i3i, 170
25-8 156, 158 25- 74^3
26-8 149
141 76
V i o 28-9
13 156, 158 9 184
22*24-31 158,159 184
h^7 159 10 76*31-6 184
^IO-28 159
159 86
23^7-9.
14 32-'7
89 23- 34 88, 123
24 90
11 77*30 86
PRIOR ANALYTICS 18 81*40-^1 183
83*14-17 208
I 24*16-22 150
24- 32 100, 207
18 149 25 86^3-4
^l8-20 87 30 87^19-22 104
7 29*27-9 150 195
89'33 189
8 149 90^3-4
9 30^37-Ί 127 189
^24-5 183
10 Я1-3, 38-40 132 9114-19
I I 3i'7-8 98 132
92'5-8 94
12-20 127 14 82
13 158 94*20-36 124
32*18-21 159, 160 21-7 132
15 34^5-7 132 36-N 125
7, 17,21 112 126
23 40 23-5 132 ^8-26
19 IOO*l6-^I 154
25 42^3-4 183
'2-3 129
b,^ 204
27 43^33-6 208
37 49^6-7 141
50*16-28 TOPICS
44 132
46 51^22-5 118 I 4 101^23-8 199
25-34 84 29-33 92
25-8 172 5 217
36-52*14 97 102*18-20 180
II 16 64b34-8, 40 91 V7 180
6 181
7 149
POSTERIOR ANALYTICS 103 24-31 211, 213
9 138
I 71 9-10 183 Ni-3 141
Il-17 184 15 io6^i5-i8 146
71 16-19 204 29-107*2 80
33-72Ч 154 107^19-26 129
120
72*1I-I2
86
II I I 115^29-35 189
14-18 IV 121^11-13 129
132 I
'13
242
 INDEXLOCORUM
2 I22^l6 163 16-18 125
5 128*26-8 171 24-33 124
8 124
V 5 135^32-^1 91
126
VI 4 1 4 1 Ч 22 154 126
199^33-I
6 i43^32 163 ^26-8 126
144 12-30 129
III 3 202*22-4 172
i45^i5 184
^I2 210
VII I 152^0-33 149 13-16 209
VIII 159^25-37 85 5 204^-7 148
5
7 207b7 139
IV 8 2i5^l8-i9 129
SOPHISTICIS ELEN' 10 162
II 162
2 l65*38-'7 85
106 V I 224*21-34 169
110 225*34ff. 158
4 25
166*22-32 192 2 226*26-7, 27-9 171
5 l67*7 94 3 227*II-I2 136, 173
,23-7 88 17 130
i7iVi2 22 160
I I 85
4 227^20-228*23 139
9 85
I72*25-8 228*21 139
85
5 229*31, ^14 158
13 191
24 179*26ff. 101 VI I 23i*2i-^l8 148
36-7 213 155
38-40 209 5 2356-30 121
^-4 210 9 240*19-26 109
2-4 213
VII 5 250*24-5 155
31 191
182*4^ 186 VIII I 25Л9-23 187
3 253^6-254*I 121
4 2547-12 169
PHYSICS 7 26017-19 153
26l*i3-i4 153
153 9 139
184*16-^14 154
^12-14 154
185*31 85, 187 DE CAELO
'5 133
189^16 II I 284*20-6 173
128
212 III 128
19019-20 211 I 299*11-17 187
220 IV 128
1 192^8-33 130
193^^30 131
185
V5 DE GENERATIONE E T
187
2 22-194*12 CORRUPTIONE
149, 186
33-4 167 I 4 319 25-9 212
194^30-3
28-31, 124-8 320-2-5 186
3 ^23-195^21
218 5 '13-14 210
195^34-5
'23 125 6 32229-33 80
4 196^33-4 187 7 324*2 152
7 198*14-16 124 10 32722 85

243
 INDEXLOCORUM
32Q25 DE MEMORIA
335^6 124
450*12 110
131
337^34-338^ 198
337 35-338^2 193
DE SOMNO

453 27-9 152

METEOROLOGICA

IV 3S2^13 137
383*27-30,32 173 DE INSOMNIIS
383'23 137
I 459*14 IIO
384'3-5 137
15 137
385^25-33 173
137 DE PARTIBUS ANIMALIUM
388*32
34-'ii 137 I I 640*1-9 198
389^ 178
8 i73
DE G E N E R A T I O N E
ANIMALIUM
DE ANIMA

403*16-25 186 74017-19 123

409*4 '24-5 177
154
412*6-9 149
.9 125 METAPHYSICS
415 8-10 124
142 1 981*7-29 194
13
124 24- 30 187
15-21
125 2 982*25 154
20-1
126 b27-8 185,188
416*14
108 3 983*24- I 124
417V16
154 25- 6 78,123
22 156
32 25 125
418*12 110, I I I
31-2 126
III 33-'i 124
199
425^26-426*26 166 Ni 129
427^12 I 10 984*5-16 147
16-21 IIO
6 987^32 109
428*24-^9 I 10 8 989*15-16 153
^i8 110 1 993'20^1 185
429*24-7 186 2 9 9 4 ^ i - 27 187
430*10-25 125
2 996^26-33 86
199
30 89
'29 I 10
33-997^2 86
199
997*2-1I 86
431*4-8 108
187 34 78
^I2-l6
3 998*20- 14 129
433'29 IIO
b22-7 82
4 999 20-2 165
20 96
DE SENSU
IOOO*I 149
442^8 6 1003*14-15 187, 195
 INDEXLOCORUM
180, 183, 18-21 99
188, 200, 21-2, 22-3 204
201-3 30-1 205
1003*21-2 201 31-1007*20 107
31-2 203 31-'34 121, 205
140, 188 31 98
ЯЗ-'io 189, 217 32-4 205
^6-io 142 , 3 4 - 13 98
12-13 79 '11-13 92
13-14 79 13-28 98, 206
16-17 78, 188, 203 13 100, 206
21 77 18-20 89
24-5 123 28-30 88
33-1004*2 122,152 1007*6 96
1004*10-20 122, 172 7-14 92
20, 23 88 8-20 104
22-4 152 20-^18 180, 206
32-3 122 20 83, 135, 155
u33 92 21-2 134
'6-9 188 21 96
22-3 82, 190 24 97
24 122 26 120
25-6 106 27 206
26 88 28-9 97
1005*11-18 78 206
3° h
203-8 33-'i8 128, 213,
12-17 122 218
23-4 90 33 h 96
28 88 34-'i 218
77, 81, 202 '2-4 218
201 3 77
3-4 91 9-10 218
5 76 10 135
8-34 116, 201 17 218
8 204 18-1008*2 121
16-17 86 18-19 89, 105
19-22 203 18 106, 205,
19-20 94, 102 206
22-4 116 26-9 107
23-32 105 1008*7-15 106
23-5 121 7-12 205
23-4 89 17 88
25-32 106 18 104
29 89 28-30 121
116, 203-8, 34-^2 121
217 89, 205
35-1006*1I 113 ^ i o - i i 205
35 78, 87 27,30 76
ioo6*i 89 5 116
5-12 86 31-1009*5 104
5-6 91 31-2 205
5 "3 1009*6 116
10-12 203 7-10 109
14, 15 76 7 89, 146
18-26 120 16-22 120

245
 INDEX LOCORUM
l6 Ii3 s 147
22-30 106 1014^^17, 21 127.154
39-8 106 19-20 128
32-8 109 26-35 137
36-8 112 ^8 123
38-12 106, 109, 1 18-20 185
Ii4 19. 27, 35 •3"
'I2-33 106 124
12 I 1 I 6-7 131
'7-i5 106 7-10 .78
8 121 10 '73
106 11-13 87
25-^ 112 13-15 87
'l-IOI 1*2 io6 •3 130, 131
I-19 4 3
Ϊ 193
14-19 110, 114 172
25^ 99 "6-9 158
25 88 6 77. 84, 145,
106, 116 149,208-
106 •4.217
106 16 ' 83, 4 5 . 149.
1 i-i6 i'3 150
I7-^I2 113 25, 33 122
I 10 36-1016^17 •51
34 I 10 36 •73
b'
'i6 96, 203 I 0 I 6^6 •39
87,126 9-17 138
25-9 121 14 135
25 104 17-24 140,150
26-7 198 19-20 178
•2-5 121 24,32 •39
17-24 106,116 28-32 •51
21-4 92, 121 • 38
ь
24-8 116 32-^6 151
2Q-^2 121 ' 5 - 6 , 8-9 •65
'5-8 92, 120 6-ii 80
7-8 104 6-8 •35
13-22 104 8-9 •51
84, 86, 122 y^M 140, 210
168 ·«11-17 160
• 50. 176
34 79
17-31 123, i 6 i ,
'14-16 168
16-17 122 •65
78, 123, 168, 17-21 • 53. 161
219 '7 164
31-5 •38
24-5 47
26-7 185 31 •49
b 32-3 • 85
'li-i6 126
17, 20 33-4 129
125
18 194 35 •93
1017'5-6 140, 178
25, 27fr. 126
29-30 43 7 77, 156, 189,
33 '93 214-17,
34 128 2^5.217
7-22 141, 182
36-IOi4^ : 218

246
 INDEXLOCORUM
18-19 216 32 96
22-31 178 176
22-6 215 28 ^8 194
22 80, 200 9 •37
24-7 122,215 12 •39
27-30 146,215 29 •89, 199
94, 140 15-.6 178
h
35-'9 107, 217 18-19 •99
'3 76 24, 26 •05
8 •3-14 148 26-1025^1 92
21-3 100 •05
23 •47 30 77. •93.
9 145,208-14 217-22
217 14-30 125, 192,
27-1018'4 210,213 193, 220
29 210,211 16-17 219
1018'7-8 221 21-30 170
10 114 21 219
r I '34-5 213 24-5 220
12 1019^17-18 76 25-6 219
19-23 •25 30-4 170
23-6 • 57 30-2 220
.35 •57 30 •9^> •92
•22-33 156 32 220
22-7, 31-2 132
Ж I 76, 77, 201-3
35 •58 ^-7 •95
13 1020^26-32 182 7-10 183
Ч 33 •*2 109
•5 122
8-12 109 18-21 87
17-24 216 18-19 186
30 •65 20-1 126, 130
15 76 27-8 •87
I02I 9-12 83 •ort'6 185
11 138 15-16 183, 201
_ h
26-^3 I 12 16 78, 197
i6 •39 17-18 •47
i8 76, 127, 207 18-29 201
1022^19-20 221 18-19 83
22 87 21 183
24, 27 182 23-32 77, 84, 186,
29. 32, 35 182 202, 203
25-9 127 23-4 202
29-35 •54 24 183, 201
20 10-12 163 27-30 87
22 84 28-9 203
=3 127 29 201
i023*2i-3 •36 30-1 76, 202
24 26-7 •37 31-2 201
27-9 •31. •78 t 77, •40, •44,
35- 2 •75 180, 208-
35-6 168 14, 214-17
25 160 33-^2
b 214
17 170 128
22-4 •74 2-24 84, 194, 195
26 •27

247
 INDEXLOGORUM
2 -4 217 28-30 155
3-4 217 1035^11-20,23-5 155
14- 15 85 1036*3-5 185
15- ^4 2 I I 9 131
l8-20 211,214 11 ^3-7 186
21-4 212,224 1037*16-17 185,187
22-4 196 29-30 147, 148,
^3 212 185
24-31 124 12 ^IO-12 135
24 i8i 14-17 213
27-33 131 1038*6-8 178
31-2 144 162
h'^
35-6 196 13 9-io,14-15 165
37-1027*2 127, 181 27-8 153
1027*5-8 182 15 1039**20, 21-2 185
7-8 223 23-4 224
8 - 13 193 29-30 125
193 16 104017 165
15-17
187 17 1041*10-11 125
19
140, 192, b7-8 149
222-5 Η 81, 140, 147,
10- I I 131 188
11- I 2 225 138
1 1042 3-15
140, t78 26-31 149
17-23 "9 2 25-6 143
1028*4-6 122 224
3 1043 i 4 - i 6
81, 140, 147, 4 1044*36- 1 124
188,200
156
1 14,30 156 140
i-9
2C^g 76
I 1045 35-1046*2 156
29-2 187
1046*4-11 80
29- 36 155 .3^-5 172
30- I 189 132
4 1047 14-16
3fr-7 168 6 165
31- 2 153 1048*27-30 157
80,81, 154
J5-6 ,з^\ 157
2 ^8-15 147
12-14 146
18-21 148
18-36 156, 196
3 1029*1-2 155 8 153
23-4 155 10498-10 157
27 16-17 154
V 1 2
140, 178,
153
200
27-8 95 1051 199
1030*32-^4 80, 189 3-5 179
5 20-1 218 ^9 200
24- 5 ^ 221 9-17 132
29-10311 186
6 1031^22-8 84, 86, 188
134» 169
7 1032^1-2 147 133
148 105219-21 136
b''' 22-8, 2^34 138
8 1033 21-2 149
156 '15-1053'8 139
10
174 16^27 139
1034'25-6

248
 INDEXLOGORUM
io54 13-19 8 2 , 135 19 153
32- 3 149 147
b^^
2-4
3-13 151 153
Il6 17-22 188
1055 4 - 5 151 3 161
16, 1 9 - 2 0 151 N 78
2 2 - 3, 3 1 - 2 152 , 1087*29-4 85
^8-11 172 1088*5-6 l6l
1057V4 139 2 1089*20-1 192
3 г6. 3 1090*30-5 148
152 6 109318-20 140
1058*8-17 153
23- 4 178

9 152 NICOMACHEAN ETHICS

i-8 122
I 6 1096*23-9 80, 85
3 1060^31-4 79 23-4 142
33-5 79 80
^29-34
io6i*8-io 77 '26-9 80
28-^4 187
7 1097*25-8 126
5 1062*2-4 9 0 , 132 8 1098 31-1099^7 171
3 91
II 5 1105^19-1106*13 163
6-9 91
11-14 25-6 171
91
17-18 IIo6*II-I2 163
93
.3&-'7 104 III 4 126
6 24-30 107 II14*32 110
5
7 1064*8-10 184
IV 7 Ii27^i4 180
^ 2 8 , 33 77
188 VI 5 Ii4O^5-6 108
'3
7-8 202 X 5 1176*33-^2 171

8-9 188
8 23-6 212
1065*6-21 198 EUDEMIAN ETHICS
6 223
I 8 1217^26-34 80
12 1069*5-12 130
26-7 142
b
171
2 9-12 34 79
4-10 188
VII 2 1236*16-32 80
4 io7o°ii-3o,3o-5 124
'17-27 80
22-6 129
6-10 87, 202, 203
6-8 194
POLITICS
6 1071^5-9 187
7 121 IV 1292^5-10 123
5
1072*19-8.1073^3 187
21-3 121
8 1073*23-36 187 RHETORIC
10 1075*28-34 85
I I 1355'^17-21 85
Μ 78
I 1076*11 149
2 1-3. 1078*9 187
POETICS
1077^10 188
20 1456'38-1457^10 94
 INDEX O F NAMES AND SUBJECTS
References are to pages of the Notes and Further Comments, except that
those i n square brackets are to numbers i n the BibUography, and the few
in parentheses are to pages of the Translation. Bold-face indicates the
subjects of the chapters ofA.

accident, see coincidence by, by virtue of 16ft-70

Ackrill,J. L . [13], [21], 162, 167, 170, byvirtueofitself 136, 140-2, 150-1,
198 168-9, i8i, 182
affection 112, 163, 171-2
Alexander, pseudo-Alexander [7], 83, capacity, capable, see possibility
84, 86, 87, i i o , 114, 115, 117, Gartwright,R. [105]
120, 151, 152, 153, 160, i6i, 164, Gassin,B. [12]
191,225 categories 80, 140-3, 152, 189, 215-17
alteration, see change and modification cause 78, 124-8, 131-2, 170, 181-2,
analytics 87 183, 187, 193, 196-8, 203, 220,
Anaxagoras 102, 107, 108, 121 221-2, 222-5
AngelelH,I. [108] final, see fulfilment
Annas,J. E . [17] formal, see form
Anscombe, G. E . M. [84], 96 material, see matter
Antisthenes 179-80 motive (efficient), see principle
Apostle,H.G. [5] certain 88, 104, 105, 123, 154
appearance, see imagination chance 220
Aquinas [10], [11] change 107-8, 109,118-19,121,
Aristotle, see Index Locorum 151-2, 156-8, 163-4, 195-6,
Asclepius [8] 211-12
axiom 86, 88, 91, 93, 94, 96 changelessthings 107-8,109,121,
125,186-9, 201-3
Barnes,J. [4],[16],[22],[25],[26], Gharles,D. [113]
[82], [98],90,221 Gharlton,W. [15]
begging the question 91-2, 203, 206 choice 123-4, 180, 190
being, that which is, what is 76, 77, Gleary,J.J. [130]
78, 79, 117, 122, 140-7, 183, 189¬ G o b b , R . A . [67]
90, 201-3, 214-17, 221 Gode, A. [36], [39], [75], [103], 205,
s e n s e s o f ^ ' ('efvat') 81,84, 102-3, 211, 212
140-7, 156 Gohen,S.M. [40]
being baldly 144, 189-90, 212, 214¬ coincidence, coincidental 76-7, 100,
15 i o i , 126-7, 133-4, 143-6, 162,
Berkeley,G i i i 169- 70, 180-2, 190-5, 198, 206-8,
Berti, E . [96] 210, 212, 213, 215, 217-22, 222-5
Bonitz, H . [24], 114, 120, 182 complete 139, 167
Bostock,D. [20] contingency 224-5
Brentano, F . [59] continuous 135-6, 175-6, 177
Brinkmann, K . [47] contradiction, see non-contradiction
Brown, L . [80], 214 contradictory 97, 203

251
 INDEXOFNAM ; AND SUBJECTS
contrary 85,97, 112, ii6, 119, 152, formula, see 'logos'
158 fulfilment 124, 126, 167, 173-4
Cratylus 109 Furth, M. [41]

D a n c y , R . M . [57],[71],[85] G e a c h , P . T . [84]
definition 92, 104, 120, 121, 148, 183, genus 137, 139-40, 174, 175-6, 177-8
185-6 geometry 129, 190
Democritus 108 Gewirth, A. [78], 194
demonstration 86,90,91, 113, 123, God, divinities 121, 147, 187-8, 201-3
132, 183, 203-4 G r i c e , H . P . [64],215
denial 80, 84, 103, ii6, ii8, 172, 199
determinism 222-5 Halper,E. [89]
dialectic 84-5, 88, 204 H a m l y n , D . W . [14],[61], i i i
differentia 84, 151, 162-3, 177-8 Heath, T . L . [132], 86, 139, 160, 163,
discipline (science, knowledge, 164, 188
understanding) 76,83,86,122, H e i n a m a n , R . E . [125]
183, 184-5, 194-5, 201-3, 204 Heraclitus 89, 106, 109, 121
disposidon 170 H i n t i k k a , K . J . J . [65],[121]
Donini, P. L . [128] hold good 78
dreaming 109-10, 113 Homer (18)
homonymy 79-80,95, 122
Ebert,T. [117] Hume, D. 106
element 128-9, 136-7, 147, 174 Hussey, E . L . [19]
Empedocles (18), 131 hypothesis 88
Epicharmus 109
essence, essendal predication 75, 100¬ identity, see same
131> 135) 148» 168, 182, 193, imagination 105, i i o , 113-15
206-8, 218-19, 220 individual, see particular
Euclid 86, 161 inference rules 206
E v a n s , J . D . G . [92] in its own right, see by virtue ofitself
Evenus (34) intelligible, see certain
excluded middle, principle of I r w i n , T . H . [37],[38],[66],[90]
(PEM) 75,86,87, 116-21, 188,
204 Jaeger, W. [ i ] , 84, 115, 139, 193
existence 79-80,82, 141-2, 178-9, Judson, R. L . [33], [120]
189-90, 214-15 just what 96, 100

falsehoood 104, 117, 119, 121, 178-80, Kahn, G. H . [35], [55], [58]
189, 198-200 Kenny, A . J . P. [93], I I I
Ferejohn,M.T. [56] King-Farlow,J. [29]
Fine, G . [122] knowledge 105, 165
firm 87-8, 89-90, 116 and see discipline
for the most part 180-2, 195, 220 Knuuttila, S. [121]
Frede,D. [124]
Frede,M. [32],[50],[70] lack 8o-i, 84, ii6, 170, 1 7 a - 3
Freeland,G.A. [ii6] L e a r , J . [87]
Frege,G [99], 136 L e i b n i z , G . W . 209-10,213,214
form 79, 82, 83, 124, 125, 131, 138-9, Leszl,W. [45]
139-40, 147-8, 174 Lewis, G. S. [133], 129

252
 INDEX OF NAMES AND SUBJECTS
L e w i s , F . A . [109],213 origin, see principle
limit 168 out of 1 7 3 - 4
L o c k e J . [134], [135], 123, 169 O w e n , G . E . L . [31],[53],[54],[55],
4ogos' 81, 92, 95, 97, 134, 138, 139¬ [79], i o i , 155, 207
40, 153-4, 168, 179-80, 183-4, Owens,J. [42],[49]
185, 204, 210
L o n g , A . A . [97] Parmenides (18)
L o u x , M . J . [60] part, portion 155, 156, 169, 174, 174¬
Lukasiewicz,J. [81] 5, 175-7,211
Lycos, K . [95], 110 particular 128, 148, 150, 154
Patzig,G. [44]
Madigan,A. [123] Pelletier,F.J. [29], [107]
Maier, H . 152 perception 108, i i o , 112, 165
mathematics 83, 186-9 philosophy 83,85, 122, 188-9
matter 124, 125, 130-1, 136-7, 148, primary (first) philosophy 76, 188,
155,174, 185-6 201-3
Matthen, M . [62], [72], 213 physics, physicists 87, 185-9, 202-3
Matthews,G.B. [104],211,212 Plato, Platonists 83, 84, 109, 147, 155
means 126 Euthydemus 171
Merian, P. [34], [43] HippiasMinor 180
metaphysics 75, 76, 77, 78, 79, 82, 84, Laws 167
183-9, 201-3, 204 Philebus 168
Mignucci, M. [ i i o ] , [119] Republic 86, 88, 94
M i l l e r , F . D . [102] Sophist 87, 92, 179, 192
modification 171-2 Theaetetus 93, 104, 105, 108, 109,
M o r a v c s i k , J . M . E . [115] 110, 113, 171
Morrison, D. R. [63] Polanski,R. [129]
Popper, K . R. [100], 134
Narcy, M . [12] possession, possess 170-1, 173
Natorp,P. 186 possibiHty,potentiahty 103, 107, 128,
nature 87, 129-31 132, 146-7, 156-60, 165
necessity 98, 99, 112, 131-2, 135, 159¬ principle 86, 87, 88, 113, 123-4, 125¬
60, 193, 196-8, 220, 222-5 6, 153, 183, 193, 196-8, 201, 203,
non-contradiction, principle of 225
(PNC) 75,86,87-116, 188, prior and posterior 81, 112, 153-6,
203-8 187, 188-9
Noonan,H.W. [86] privation, see lack
not-being, that which is not 8o-i, 117, Protagoras 102, 105-16
178-9, 189-90, 192, 215, 216 Pythagoreans 147-8
not-man, not-equal, etc. 84, 97, 99
number 120, 136, 139, i6i, 163, 164¬ quaHty,qualification 80, 162-4, ^Ί^~
5 2, 215-17
Nussbaum, M. C. [28] quantity 138,160-2,215-17
Nuttall,J. [83] Q u i n e , W . V . [77],82

0 ' M e a r a , D . J . [27] Reale,G. [48]

one 82, 83, 84, 85, i o i , 133-40, 150- refutation 91, 113, 121, 203-4
I , 165, 188, 210, 221 relative 115-16,164-7,214
opposite 83, 151-3 Remes,U. [121]

253
 I N D E X O F NAM AND SUBJECTS
R o s s , W . D . [2],[3],[23],75,84,86, Syrianus [9]
87,90, 94, 104, 109, 113, 115,
117, 122, 128, 129, 130, 139, 147, that which is, thing that is, see
151, 152, 153, 155, i6i, 162, 164, being
170, 176, 178, 180, 185, 186, 191, that which is not, see not-being
193,195,197-8 theology 186-9, 201-3
T h o r p , J . W . [51],[69],[73],202,
same 83, 84, 134, 139-40, 149-51, 190, 215
207-8, 208-14, 221 T i l e s , J . E . 118
scepticism 105, 108 transcendental argument 204, 205
Schofield, M. [25], [26], [28] Tredennick, H . 110
S c h o l a r , M . C . [94] true 104, 117, 119, 121, 146, 189,
science, see discipline 198-200
sense(ofanexpression) 79,122 truncated 177
separable 85,149, 185, 186-7, 199¬
200 understanding, see discipline
signify 93-8, 140, 204, 205-6 unity, see one
Smith,J.A. [3] universal, aU 135, 150, 175-7
sophist, sophistic 85, 190-2 U p t o n , T . V . [91]
S o r a b j i , R . R . K . [25],[26],[112],
217, 223 W e d i n , M . V . [88]
Spellman, L . [111] Weidemann, H . [76], [127], 208
S p r a g u e , R . K . [68],[131] what is, see being
statement, see 'logos' W h i t e , M . J . [114]
stateofaffairs 80, 178-9 White, N. P. [ i o i ]
Stevenson,J. G. [46], 202 whole 1 7 5 - 7
subject 136-7, 147, 148, 178 Wiesner,J. [30]
subject-neutral 78 WiUiams, C . J . F . [18], [106], [126],
substance 78, 100, 147, 147-9, 155, 211-12, 223-4
165,185, 189-90, 201 Woods,M.J. [74]
syllogism 87
synonymy 94, 96 Xenophanes (19)

254