Hare, R M - Sorting Out Ethics

SORTING
OUT
ETHICS
BY
R. M.
HARE
CLARENDON PRESS
1997
OXFORD
Oxford University Press. Great Clarendon Street. Oxford 0x2 6DP

Oxford New York
Athens Auckland Bangkok Bogota Bombay
Buenos Aires Calcutta Cape Town Dar es Salaam
Delhi Florence Hong Kong Istanbul Karachi
Kuala Lumpur Madras Madrid Melbourne
Mexico City Nairobi Paris Singapore
Taipei Tokyo Toronto Warsaw
and associated companies in
Berlin lbadan
Oxford is a trade mark of Oxford University Press
Published in the United States by
Oxford University Press Inc.. New York
R. M. Hare 1997
PREFACE
T h e c o r e o f this b o o k i s m y A x e l H a g e r s t r m L e c t u r e s , given i n
Uppsala in 1 9 9 1 . 1 had planned to incorporate these, together with revisions of o t h e r papers, into a full-length book giving my considered
views on ethical theory. It was to have been given as t h e J o s Ferrater
M o r a Lectures at G i r o n a in Catalonia. B u t this too a m b i t i o u s project
was defeated by a series of strokes, w h i c h rendered me incapable, n o t
only of typing with m o r e t h a n one hand, but of thinking book-length
t h o u g h t s . I was very s o r r y to have to c a n c e l my visit to Catalonia, to
which I had been looking forward with pleasure.
All rights reserved. No part of this publication may be reproduced,

stored in a retrieval system, or transmitted, in any form or by any means.
without the prior permission in writing of Oxford University Press.
Within the UK. exceptions are allowed in respect of any fair dealing for the
purpose of research or private study, or criticism or review, as permitted
under the Copyright. Designs and Patents Act. I'jSH. 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.
Formerly, w h e n writing a book, I used to hold t h e whole of it in my

head from start to finish. T h i s is the only way to avoid repetitions and
even c o n t r a d i c t i o n s . B u t I c a n no longer do it. So I h a v e h a d to c o m promise, with t h e helpful advice of t h e Oxford University Press, and
publish t h e lectures with t h r e e m a j o r additions. T h e first of these is
an a t t e m p t to justify t h e whole enterprise of applying philosophy of
language to ethics. It is a revised version of my contribution to the De
G r u y t e r landbuch Sprachphilosophie, a n d gives a c o n s p e c t u s of my
entire thinking. T h e second is an introduction to my lecture c o u r s e in
British Library Cataloguing in Publication Data

Data available
library of Congress Cataloging in Publication Data
Hare. R. M. (Richard Mervyn)
Sorting out ethics I R. M. Hare
Includes bibliographical references and index.
1. Ethics
1. Title.
BJ1012.H31352
1997
170dcn
97-8001
ISBN
0-19-K2J727-X
1 5 5 7 9 10 8 6 4 2
Typeset by Invisible Ink
Printed in Great Britain
on acid-free paper by
Biddies ltd, Guildford and King's Lynn
Oxford a n d Florida, omitted from t h e five l e c t u r e s given in Uppsala

because o f lack o f time.
T h e A x e l H a g e r s t r m L e c t u r e s follow. T h e y w e r e delivered originally under t h e title. 'A T a x o n o m y of Ethical T h e o r i e s ' . T h e first and
second o f t h e s e w e r e mostly n e w ; t h e rest h a d m a n y s o u r c e s . T h e y
a r e partly a distillation of l e c t u r e s given over t h e y e a r s in Oxford,
Florida, a n d e l s e w h e r e , m u c h revised, c o n d e n s e d , a n d , I h o p e , improved. My practice h a s been to give lectures shaped round a nucleus
which r e m a i n e d basically t h e s a m e , to w h i c h I added o t h e r lectures
from time to time. M a n y of these additions were intended to illustrate
t h e uses of e t h i c a l t h e o r y by applying it to p r a c t i c a l problems. T h e y
have mostly been collected into volumes and published already. I hope
t h a t o n e o t h e r s u c h v o l u m e will appear. B u t t h e n u c l e u s , giving m y
PREFACE
PREFACE
latest considered t h o u g h t s , could n o t be published while I w a s still
with different p a g i n a t i o n s ; a n d t h e s a m e applies t o t h e n u m e r o u s

t r a n s l a t i o n s of my w o r k s . In t h e c a s e of r e f e r e n c e s to older writers
who are published in m a n y editions it has usually been best to cite the
section or c h a p t e r , or, in t h e c a s e of P l a t o . Aristotle, a n d K a n t , t h e
pages of the standard editions.
vi
l e c t u r i n g . T h i s formed t h e m a i n p a r t o f t h e Axel H a g e r s t r o m
Lectures. I have to t h a n k the very intelligent audience at Uppsala for
giving ( h e m s u c h a s t i m u l a t i n g r e c e p t i o n . I am printing t h e m as
delivered, with a few afterthoughts, but retaining the style of an oral
presentation.
Last. I have reprinted my paper 'Could K a n t have b e e n a Utilita r i a n ? ' from lltilitas 5. which h a s also appeared in Kant and Critique,
edited by R. M. Dancy. It w a s given, a m o n g o t h e r o c c a s i o n s , in
S t o c k h o l m on t h e s a m e visit to S w e d e n . I owe so m a n y of my o w n
ideas to Kant, and my interpretation of h i m as a quasi-utilitarian is so
unorthodox (though it now has supporters), that I thought it worth
reprinting here.
It will be obvious t h a t a b o o k so s t r u c t u r e d is b o u n d to c o n t a i n
overlaps. For example, points are m e n t i o n e d briefly in Chapter 1 but
taken up in m o r e detail in Chapters 3 to 7: a n d my interpretation of
K a n t figures in m a n y of t h e e a r l i e r c h a p t e r s before being fully e x plored in Chapter 8. This is unavoidable if the chapters are to be read
independently. S o m e people m a y w a n t to read j u s t C h a p t e r 1 as a
s u m m a r y of my ideas; but o t h e r s m a y find this too difficult a n d skip
on to Chapter 2, w h i c h is m u c h easier. And s o m e m a y n o t be interested in q u e s t i o n s of K a n t i a n exegesis. For these r e a s o n s I have decided to put up with some overlaps; but these are clearly signposted.
I have to t h a n k others besides the Swedes for c o m m e n t s on various
versions of these lectures. T h e y are too m a n y to list; but I have given
the n a m e s of those whose writings I found of most help with the K a n t
chapter in the bibliography. T h i s h a s been expanded into a full list of
my philosophical writings, as an aid to t h o s e w h o wish to study my
ideas, with t h e addition of a b s t r a c t s of my m o r e i m p o r t a n t r e c e n t
papers. I owe a lot to t h e e x c e l l e n t b i b l i o g r a p h y c o m p i l e d by Ulla
Wessels for the two volumes of 7.um moralisvhcn Darken (H 1 9 9 5 ) . the
proceedings of a conference on my work.
I have used an a u t h o r - d a t e system of reference, b e c a u s e it avoids
footnotes: but I have not t h o u g h t it n e c e s s a r y to cite the page n u m b e r s in c a s e s w h e r e it is e a s y to find t h e p a s s a g e referred to. T h e
r e a s o n is t h a t m a n y of t h e a r t i c l e s h a v e appeared in several p l a c e s
vii
I offer this t a x o n o m y of ethical theories to all those w h o are lost in

the moral maze, including m a n y of my philosophical colleagues.
T h e y a r e lost b e c a u s e , like m o s t of t h o s e w h o hold forth on m o r a l
q u e s t i o n s i n t h e m e d i a , t h e y have n o m a p o f t h e m a z e . T h i s i t h a s
been my aim to provide.
CONTENTS
PART I.
THE E N T E R P R I S E OF MORAL
PHILOSOPHY
1. Philosophy of L a n g u a g e in Ethics
2. Defence of the Enterprise
i
29
PART II. THE AXEL H A G E R S T R O M L E C T U R E S :

A T A X O N O M Y OF ETHICAL T H E O R I E S
3. Taxonomy
43
4 . Naturalism
63
5. Intuitionism
82
6. Emotivism
7. Rationalism
103
126
PART III.
KANT
8. Could K a n t have been a Utilitarian?
147
References and Bibliography
167
Index
187
PART I
THE E N T E R P R I S E OF MORAL
PHILOSOPHY
I
P H I L O S O P H Y OF LANGUAGE IN E T H I C S
1.1.
E T H I C S , or m o r a l philosophy, is the point at w h i c h philosophers
c o m e closest to practical issues in morals and politics. It t h u s provides

a m a j o r p a r t of t h e p r a c t i c a l j u s t i f i c a t i o n for d o i n g p h i l o s o p h y (H
1971c: 9 8 ) . If, therefore, philosophy of l a n g u a g e c a n be s h o w n to have
a c r u c i a l c o n t r i b u t i o n t o m a k e t o e t h i c s , this g r e a t l y i n c r e a s e s t h e
p r a c t i c a l r e l e v a n c e o f t h e discipline. B u t i t i s v e r y i m p o r t a n t t o b e
clear about w h a t t h e contribution is.
T h e following p r o g r a m m e looks p r o m i s i n g a t first s i g h t . P h i l o sophy of l a n g u a g e is c o n c e r n e d above all w i t h t h e study of t h e c o n cept of meaning in t h e various senses of t h a t word. B u t t h e m e a n i n g s
of moral words and sentences, in at least s o m e senses, determine t h e
,
logic of inference;} in w h i c h they appear. So a study of t h e m e a n i n g s

of moral words or s e n t e n c e s , or of w h a t people m e a n w h e n they utter
them, o u g h t to e n a b l e us to investigate t h e logical properties)of w h a t
t h e y say, a n d t h u s decide w h e t h e r w h a t t h e y s a y is self-consistent,
w h a t it implies, and in general w h i c h a r g u m e n t s (in t h e sense of r e a s o n i n g s ) a r e good o n e s a n d w h i c h a r e n o t . , S o p h i l o s o p h y o f l a n guage, applied to m o r a l language, o u g h t to be able to provide a logical
structure for o u r m o r a l thinking. And since our m o r a l thinking often
founders for lack of s u c h a structure, t h a t would be no small gain.
Revised from H 19960.
THE E N T E R P R I S E OF M O R A L P H I L O S O P H Y
1.1, i
1.1.1
T h e r e are a g r e a t m a n y pitfalls to be avoided in c a r r y i n g o u t this
have in r e c e n t times usurped the n a m e ; t h e y would be better called
p r o g r a m m e ; but I shall a r g u e t h a t it is in principle a feasible one. So
'mephistics', b e c a u s e they are attempts, like t h a t of Mephistopheles
let us first consider some possible objections to it. I shall be in danger
in Faust, to get philosophers to sell their souls for fantasies.
of being misunderstood if I do not m a k e clear at the start t h a t philo-
I wish to consider two possible objections to the p r o g r a m m e I pro-
sophy of language is not the s a m e as linguistic philosophy. T h e former
jected at the beginning. The first says. T a c t s about particular lan-
is a b r a n c h of philosophy, c o - o r d i n a t e with philosophy of s c i e n c e ,
g u a g e s , i n c l u d i n g facts a b o u t h o w people u s e words i n p a r t i c u l a r
philosophy of law, philosophy of history, e t c . To say t h a t a philo-
cultures, a r e c o n t i n g e n t facts. T h e y therefore c a n n o t be used to e s -
sopher is doing philosophy of l a n g u a g e does not presuppose that he is
tablish n e c e s s a r y t r u t h s s u c h as we are looking for in e t h i c s . We do
doing it by any particular method, or in a c c o r d a n c e with the tenets of
not w a n t to be told h o w p a r t i c u l a r people or c u l t u r e s use the m o r a l
any particular school. Philosophers of language can be realists or the
words; we w a n t to be s h o w n what is right or wrong, and to be shown
opposite, intuitionists or t h e opposite, and so on. If anybody were to
by secure r e a s o n i n g t h a t this is necessarily the case.'
say, like Plato on some interpretations, that words have m e a n i n g be-
T h e second objection is related to t h e first: it says, 'Moral r e a s o n -
c a u s e they stand for e t e r n a l l y e x i s t i n g n o n - s e n s i b l e entities up in
ing h a s to be c o n c e r n e d with m o r a l facts, w h i c h a r e facts n o t a b o u t
Heaven, he would still be doing philosophy of l a n g u a g e , b u t would
words but about t h e worldfacts about the existence of m o r a l values
obviously not be a linguistic philosopher. But see H 1 9 8 2 0 , esp. c h . 4,
i n t h e world. T h e study o f words c o u l d n e v e r yield s u c h facts.'
for a more 'linguistic' interpretation of Plato.
Answers to b o t h these objections c a n be given. For t h e first, consider
A linguistic philosopher is s o m e o n e w h o believes in a p a r t i c u l a r
t h e position of o r d i n a r y logic. It would be a m i s t a k e to suppose t h a t
way of doing philosophy (any kind of philosophy, not just philosophy
logic discovers only c o n t i n g e n t truths about l a n g u a g e ; b u t it is also a
of language), n a m e l y t h a t w h i c h consists in studying t h e m e a n i n g s
mistake to think t h a t logic is independent of t h e study of language. It
of words t h a t present philosophical problems, and so unravelling t h e
is a necessary truth that, in one c o m m o n m e a n i n g of 'all' and the
problems. He will advocate, like C a r n a p ( 1 9 3 2 ) , an ' b e r w i n d u n g der
other words used, if all t h e books on the top shelf are by Wittgenstein,
Metaphysik durch logische A n a l y s e der S p r a c h e ' . To m a k e my o w n
and this is a book on the top shelf, then this is by Wittgenstein. B u t in
position clear, I am a linguistic philosopher of a sort, but not of such
order to establish that this is a necessary truth, we have to be assured
an extreme sort as Carnap. I believe t h a t metaphysics does n o t have
that the words are being used and understood in the senses t h a t m a k e
to be overcome, nor even superseded; as inherited from Aristotle, it is
it so. Logic is, at least in part, t h e study of t h e words w h i c h people use
a respectable and c e n t r a l b r a n c h of philosophy, and o n l y c e r t a i n
in t h e i r discourse, to a s c e r t a i n w h i c h of t h e t h i n g s t h e y say are, as
bogus impersonations of it are suspect, liver s i n c e Aristotle and be-
they use the words, necessary truths.
fore, it has used linguistic m e t h o d s . A great m a n y p r o b l e m s w h i c h
This does n o t m a k e the truths of logic contingent. It is of course a
are called 'ontological' are in fact to be resolved by careful a t t e n t i o n
c o n t i n g e n t fact t h a t people do use certain sounds with c e r t a i n senses.
to the words which give rise to t h e m ; and this is true above all in
B u t to ask in w h a t senses they use t h e m is to ask a c c o r d i n g to w h a t
ethics. But I regard this, not as a way of overcoming metaphysics, but
rules or conventions, logical and semantical, they use t h e m . And it is
as a way of doing it c o m p e t e n t l y o f mastering it, if we m a y so mis-
n o t a c o n t i n g e n t fact, but a tautology, t h a t a n y o n e w h o is using t h e
t r a n s l a t e 'iiberwinden'; and I believe t h a t this way of d o i n g it h a s
words in those senses will be committing logical errors if he does not
yielded results when practised by all t h e great metaphysicians up to
observe those rules. To take t h e s a m e example: it is a c o n t i n g e n t fact
t h e present day. So I am not against metaphysicsonly against s o m e
t h a t s o m e o n e is using 'all' in the sense t h a t he is. B u t it is n o t a c o n -
wholly spurious 'philosophical' a n d ' t h e o l o g i c a l ' activities w h i c h
t i n g e n t fact t h a t , if he is u s i n g it in that s e n s e ( n a m e l y t h e s e n s e in
THF, F N T H R l ' R I S F . OF MORA!, P H I L O S O P H Y
I. I. i
which the above hypothetical is necessarily true), the hypothetical is

necessarily true. W h a t makes the sense that sense is that it is the sense
which makes that hypothetical necessarily true.
1 . 2 . Words, including words like 'all', have their m e a n i n g determined by the conventions according to which we use them. And the
c o n v e n t i o n s are in part logical ones, which determine w h a t implies
what, what we can consistently say. etc. One is not being a c o n v e n tionalist in any bad sense if one states the obvious truth that studying
what the conventions are for the use of words like 'all' (i.e. w h a t logical rules they a r e governed by. as people use t h e m ) is the basis of
the discovery of these logical rules.
To this it may be objected that people do not have to use words in
a c c o r d a n c e with t h o s e r u l e s . Humpty Dumpty was quite right
(Carroll 1 8 7 2 : 19(1). 'All' could have m e a n t the s a m e as ' s o m e ' does
n o w w h i c h is to say that the rules which determine its m e a n i n g and
the implications of propositions c o n t a i n i n g it might have been different, and like those w h i c h n o w d e t e r m i n e the m e a n i n g and implications of 'some'. And Englishmen. F r e n c h m e n , Germans, and Chinese
use different sounds to express the same things. And the inventors of
artificial languages like C a r n a p have a considerable liberty to invent
new uses of words and symbols, and to invent, pari passu with this,
new rules and conventions for their use. Here too, however, it has to
be said that if a word is being used in any language (natural or artificial) to express the s a m e m e a n i n g as a word in some other language,
it is bound by the s a m e logical r u l e s . If it were b o u n d by different
rules, it would n o t express the s a m e m e a n i n g . A word in C h i n e s e is
not the equivalent of 'all' unless, w h e n used in the c o r r e s p o n d i n g
Chinese hypothetical about Wittgenstein, it makes it necessarily true.
So. if logic as a whole involves the study of words in this way. the
s a m e will be true of t h a t b r a n c h of logic called t h e o r e t i c a l e t h i c s . I
call theoretical ethics a b r a n c h of logic b e c a u s e its principal aim is
the discovery of ways of d e t e r m i n i n g what a r g u m e n t s about moral
questions are good ones, or how to tell sound from unsound reasoning in this a r e a . It is, in particular, a b r a n c h of modal logic. 'Ought',
which we may take as the simplest e x a m p l e of a word used typically
in m o r a l d i s c o u r s e (a m o r a l word, for s h o r t ) , e x p r e s s e s a d e o n t i c
I. 1 . 2
modality, and this is s h o w n by the fact that deontic logics c a n be syst e m a t i z e d w h i c h a r e in all or n e a r l y all r e s p e c t s a n a l o g o u s to t h e
other kinds of modal logic (Prior 1955: HI. i. 6 ) . T h e s a m e is even more
clearly t r u e of t h e word ' m u s t ' : its use to express m o r a l s t a t e m e n t s
like 'I must not tell h e r a lie' is analogous in most ways to its use to express alethic modal statements.
If, as is beginning to happen, viable systems of deontic logic can be
discovered w h i c h are adequate models of ordinary m o r a l l a n g u a g e ,
they will do as m u c h for the u n d e r s t a n d i n g of m o r a l a r g u m e n t s as
ordinary logic does for the understanding of o t h e r a r g u m e n t s . So, alt h o u g h it is of c o u r s e a c o n t i n g e n t fact t h a t English uses 'ought' to
express t h e m e a n i n g t h a t it does, it is n o t c o n t i n g e n t t h a t any l a n guage that has an equivalent s e n t e n c e i . e . a way of expressing the
s a m e t h o u g h t w i l l be bound by the s a m e rules of r e a s o n i n g . A n d
what the rules are. as the word is normally used, is discovered by asking how it is normally used.
As before, we do n o t have to use it in t h a t way. B u t w h e n we are arguing about moral problems we are arguing about w h e t h e r to accept
or reject c e r t a i n m o r a l judgements. Clearly, w h e t h e r an a r g u m e n t is
a good a r g u m e n t for accepting or rejecting a c e r t a i n j u d g e m e n t will
depend on what the j u d g e m e n t is. B u t what it is depends on what the
words used in e x p r e s s i n g it a r e being u n d e r s t o o d to m e a n . If t h e y
were being understood to m e a n something different, it would be a different j u d g e m e n t . B u t o n c e we are committed to discussing w h e t h e r
to a c c e p t or r e j e c t that j u d g e m e n t (i.e. t h e j u d g e m e n t w h i c h t h o s e
words express w h e n they are taken in that way) we are committed to
following the rules of r e a s o n i n g which t h a t way of t a k i n g t h e m det e r m i n e s . To l a k e t h e words in t h a t way is to accept t h a t the judgem e n t (with or w i t h o u t additional premisses) logically implies s u c h
and such o t h e r judgements, is inconsistent with s u c h a n d such other
j u d g e m e n t s , a n d so o n . So the s e n s e of the words, as before, determines which a r g u m e n t s about the questions we are asking are sound
o n e s . T h e r e f o r e , in order to d e t e r m i n e w h e t h e r t h e y a r e sound, we
have to e x a m i n e the senses of the words, i.e. the rules for their use in
arguments.
We c a n of course, as before, use words as we wish. B u t if we decide
THE E N T E R P R I S E OF MORAL P H I L O S O P H Y
L I. 2
I.
1.2
to use words differently from h o w we were using t h e m when we posed
T h e answer to t h e second objection m e n t i o n e d at t h e beginning is
our original problem, we shall no longer be posing the s a m e problem.
thus that, b e c a u s e the concepts studied by ethics are formal, there do
We a r e free to pose different problems: a n d t h a t is what we shall be
not have to be m o r a l facts in t h e world in o r d e r for us to develop a
doing if the words m e a n something different. To revert to our original
t h e o r y o f m o r a l r e a s o n i n g , a n y m o r e t h a n t h e r e h a v e t o b e logical
e x a m p l e : if w h a t we had been asking h a d been, not w h e t h e r all t h e
facts t o s u b s t a n t i a t e logical r e a s o n i n g . T h e n e c e s s i t i e s w h i c h c o n -
books were by Wittgenstein, but w h e t h e r some of them were, it would
strain o u r r e a s o n i n g are formal n e c e s s i t i e s w h i c h does n o t m e a n ,
not have been a reason for answering 'No' that o n e of the books was
any m o r e t h a n it does in logic a n d m a t h e m a t i c s , t h a t t h e y c a n n o t in
not by Wittgenstein. So if. w h e n we said 'all', we had been using t h e
conjunction with substantial n o n - m o r a l information a b o u t t h e world,
word in t h e s a m e s e n s e as ' s o m e ' u s u a l l y h a s . the r e a s o n i n g we
help us in deciding m o r a l q u e s t i o n s of s u b s t a n c e . How this is to be
should have had to use in a n s w e r i n g o u r question would have b e e n
done, we shall see later.
different. In the s a m e way, if 'ought' m e a n s to us w h a t it does w h e n
1 . 3 . It is n o w time to ask how we c a n discover w h a t these formal
we a r e asking our m o r a l q u e s t i o n s , we shall have in our m o r a l r e a -
properties are. T h e first step requires us to a n a t o m i z e l a n g u a g e as a
s o n i n g to follow t h e rules (of i m p l i c a t i o n , c o n s i s t e n c y , e t c . ) deter-
whole in order to see where in t h e a n a t o m y such words as 'ought' be-
mined by that m e a n i n g of t h e word (by the fact that it is that question
long. T h e most perspicuous way of doing this is by speech a c t theory.
we are asking, and not a different question which would be asked by
T h e t e r m s p e e c h a c t ' w a s b r o u g h t i n t o c u r r e n c y b y } . L . Austin
s o m e o n e w h o uttered the s a m e sounds but w a s using 'ought' in a n -
( 1 9 6 2 : 4 1 , 1 4 9 ) , t h o u g h h e does not himself use t h e t e r m very m u c h ,
o t h e r s e n s e ) . It is t h e r e f o r e in order, if we wish to d e t e r m i n e w h a t
preferring m o r e specific expressions. He c a n justly be regarded as the
rules we have to follow, to ask in w h a t sense the word was being used
founder of speech a c t theory: but the idea that n o t all speech acts are
in our question. Indeed, to ask in what sense it is being used is to ask
o f t h e s a m e k i n d o r o b e y t h e s a m e rules h a s b e e n used before a n d
w h a t the rules are.
after him by Wittgenstein, Ryle, Searle, Habermas, a n d m a n y others.
All this is peculiarly true of words like 'ought', one of the most gen-
In order to divide off speech acts of different kinds from o n e another,
eral t e r m s used in a s k i n g m o r a l q u e s t i o n s . S u c h words, like o t h e r
w e need t o a r t i c u l a t e t h e s e n t e n c e s t h a t a r e used t o perform t h e m .
modal words, express formal c o n c e p t s , in t h e sense that the rules for
T h e m a i n purpose of this is, if possible, to isolate t h e features of sen-
t h e i r use a r e e x h a u s t e d by t h e i m p l i c a t i o n s and other logical prop-
tences which perform the various functions n e c e s s a r y for a complete
erties t h a t they give to t h e propositions c o n t a i n i n g them. T h i s is n o t
speech act. T h e n we c a n see which features of a s e n t e n c e are peculiar
t r u e of all words: for e x a m p l e , t h e formal logical properties of t h e
to a particular kind of speech act. and so m a r k t h e u t t e r a n c e of it as a
words 'blue' and 'red' are the same, but 'red' does not mean the s a m e
performance of that kind of speech act; and which features are c o m -
a s ' b l u e ' . S o t h e i r formal l o g i c a l properties c a n n o t e x h a u s t t h e i r
m o n to a n u m b e r of different kinds of s p e e c h a c t . T h e best k n o w n
m e a n i n g . B u t if ' o u g h t ' is a purely formal word, then we should be
marker of this sort is t h e sign of mood (e.g. indicative or imperative)
able to discover all t h e r e is to be k n o w n a b o u t its m e a n i n g a n d t h e
which (to speak generally at first) m a r k s off s t a t e m e n t s from imper-
rules for its use by studying its logical properties. If true, this is, as we
a t i o n s (if we m a y u s e t h a t expression for s p e e c h a c t s typically e x -
s h a l l see. of f u n d a m e n t a l i m p o r t a n c e for e t h i c s . It m e a n s t h a t , a l -
pressed in the imperative).
t h o u g h in a sense it h a s s e m a n t i c a l properties as well (its 'descriptive
We also need to be clear t h a t the division of speech a c t s into kinds
m e a n i n g ' ) , these are not part of its m e a n i n g in the n a r r o w sense (H
takes t h e form of a t r e e w i t h g e n e r a , species, s u b - s p e c i e s , e t c . It
1 9 8 6 c ) . a n d do n o t affect at all profoundly t h e rules for r e a s o n i n g
c a n n o t be a s s u m e d , for e x a m p l e , t h a t t h e r e a r e no further subdivi-
about what we ought to do.
sions w i t h i n t h e c l a s s e s o f s t a t e m e n t s a n d i m p e r a t i o n s , n o r t h a t
I. 1 . 3
imperations may not belong, perhaps with moral judgements, within
ferent logical implications (II 1996b): t h e latter implies t h a t you are
a larger class of prescriptions. Nor c a n it be assumed t h a t a kind of
going to leave this place: the former does not, b e c a u s e a c o m m a n d is
speech act has to belong to o n e or o t h e r of these classes and c a n n o t
not a prediction of its o w n fulfdment. Next, we have to distinguish
belong to m o r e than one. The species and genera may not be m u t u -
the c o n t e n t of the speech act (for example w h a t in particular is being
ally e x c l u s i v e : p e r h a p s m o r a l j u d g e m e n t s s h a r e s o m e o f t h e prop-
stated to be t h e case, or c o m m a n d e d to be m a d e t h e c a s e ) . T h u s the
e r t i e s both of s t a t e m e n t s a n d of p r e s c r i p t i o n s . All this h a s to be
c o m m a n d s 'Open t h e door' and ' S h u t t h e w i n d o w ' h a v e t h e s a m e
investigated by the study of speech and l a n g u a g e (I use these words
tropic but different phrastics (using that term to denote t h e feature of
to mark the distinction made famous by Saussure 1916).
the sentence, not necessarily a separate part of it. that indicates what
A further necessary clarification can conveniently be made at this
is being e.g. stated or c o m m a n d e d ) . In a c o m p l e t e l y a n d p e r s p i c u -
point. Austin used the term 'illocutionary force' to c o n n o t e the prop-
ously articulated l a n g u a g e these functions would be assigned to dif-
erty which distinguishes o n e speech act from another. T h u s the state-
ferent parts of the sentence.
ment thai you are going to shut the door has a different illocutionary
T h e r e m a i n i n g two functions, which do not need to be discussed
force from the c o m m a n d that you shut the door. But different writers
here, are those w h i c h would be expressed in a fully a r t i c u l a t e d lan-
s i n c e Austin have interpreted this d i s t i n c t i o n in different w a y s .
guage by the clistic, or sign of completeness, of the sentence, and the
Consider the two c o m m a n d s , that you open t h e door, and that you
ncustic or sign of subscription to a speech act by a speaker or writer.
shut the window. Do these have the s a m e illocutionary force, in t h a t
T h e s e signs are controversial, and m a n y writers have denied t h e ne-
they are both c o m m a n d s , or different i l l o c u t i o n a r y forces, b e c a u s e
cessity of the latter in particular; but I shall n o t need to defend t h e m
they are different c o m m a n d s ? It will make no difference to any a r g u -
for the purposes of the present a r g u m e n t (see H 1 9 8 9 a ) . Nevertheless,
m e n t , provided t h a t we a r e c l e a r a b o u t o u r use of the t e r m s : but in
it is very i m p o r t a n t to distinguish between these different functions,
w h a t follows I shall m y s e l f adopt t h e s e c o n d of these uses. I s h a l l
as m a n y writers (including myself in early days) have n o t (H 1971c:
speak of these two c o m m a n d s as h a v i n g two different i l l o c u t i o n a r y
21 IT.). In p a r t i c u l a r t h e tropic or mood-sign h a s to be distinguished
forces, t h o u g h they b e l o n g to t h e s a m e type of i l l o c u t i o n a r y force,
from the neustic or sign of subscription, because o n e c a n m e n t i o n or
namely the imperative. Similarly I can make two different statements,
embed an indicative or imperative sentence, including its mood-sign,
which have different illocutionary forces because their c o n t e n t is dif-
or use it m i m e t i c a l l y ( 6 . 4 , H 1 9 8 9 0 ) e . g . on t h e s t a g e w i t h o u t
ferent, but have t h e s a m e type of i l l o c u t i o n a r y force, n a m e l y w h a t
making a s t a t e m e n t or giving a c o m m a n d .
Austin called the constative ( 1 9 6 2 : 6 n . ) . T h i s will be b r o u g h t o u t if
It will be asked at this point w h e t h e r mood, as I am using t h e word,
the s e n t e n c e s are articulated in such a way (as they are in most lan-
is a logical or a g r a m m a t i c a l term. T h e a n s w e r is t h a t it is both, but
guages) as to distinguish the feature which marks the mood from the
t h a t w e have t o u n d e r s t a n d t h e difference b e t w e e n w h a t a r e n o w
rest: the two c o m m a n d s 'Open the door' and ' S h u t the window' s h a r e
often called surface and deep g r a m m a r , and used to be called g r a m -
this feature, by which we recognize t h e m as imperatives: but o t h e r -
m a t i c a l form a n d logical form. If there is a difference b e t w e e n these
wise they differ.
two ways of m a k i n g t h e distinction, it will n o t affect w h a t I am n o w
T h e articulation of sentences, or the speech acts that they express,
going to say. In history, g r a m m a r and logic g r e w up together, and
h a s to distinguish at least four f u n c t i o n s (H 1 9 8 9 a ) . T h e first is t h e
metaphysics with t h e m : and it h a s proved difficult to draw c l e a r dis-
mood, already mentioned. I shall call t h e sign of mood the tropic. T h a t
t i n c t i o n s b e t w e e n t h e s e t h r e e . Even s u c h diverse t h i n k e r s as Hegel
mood is, or c a n be, part of m e a n i n g is evident from the fact t h a t t h e
and C a r n a p found it h a r d to distinguish b e t w e e n logic a n d m e t a -
Latin expressions ' and 'ibis ('Go' and 'You are going to go') have dif-
physics (Hegel a s s i m i l a t i n g t h e former to t h e latter, and C a r n a p . in
THE E N T E R P R I S E OF MORAI, P H I L O S O P H Y
IO
1.1- 3
I. 1 . 4
ii
effect, t h e r e v e r s e t h o u g h he reserved t h e n a m e ' m e t a p h y s i c s ' for
the prescriptive (i . 6 ) . All kinds of ordinary s t a t e m e n t s will belong to
w h a t 1 have called ' m e p h i s t i c s ' ) . And similarly deep g r a m m a r a n d
t h e former, a n d all s p e e c h a c t s w h i c h are typically expressed in t h e
logic are so intimately bound up with e a c h other that it would be fool-
imperative to t h e latter. We must not presuppose t h a t n o t h i n g except
ish to try to prise t h e m apart. T h e difference between logic a n d sur-
i m p e r a t i o n s b e l o n g s t o t h e latter g e n u s . W e m u s t n o t e v e n presup-
face g r a m m a r is what has made people think that there is a difference
pose that in order to give a c o m m a n d it is n e c e s s a r y to use t h e imper-
between g r a m m a r and logic as a whole.
ative. B u t let us n o w ask in a p r e l i m i n a r y w a y w h e t h e r m o r a l
T h e r e are indeed g r a m m a t i c a l distinctions that have no logical sig-
j u d g e m e n t s (for e x a m p l e those expressed with ' o u g h t ' ) a r e prescrip-
nificance, like that between strong and weak forms of the past tense
tive or descriptive speech acts. T h e a n s w e r is t h a t they a r e both, but
( 3 . 3 ) . But mood is not like this: the distinction between the mood-sign
t h a t the distinction needs to be carefully preserved, b e c a u s e o t h e r -
and t h e rest of a s e n t e n c e is as i m p o r t a n t logically as t h a t b e t w e e n
wise we s h a l l n o t be able to u n d e r s t a n d t h e different f e a t u r e s of
subject a n d predicate. T h e s e two have been both g r a m m a t i c a l a n d
'ought'-sentences w h i c h link them to the two genera.
logical terms, and rightly, b e c a u s e the g r a m m a r is a way of express-
O u g h t ' - j u d g e m e n t s a r e prescriptive, a n d in this r e s p e c t like im-
ing t h e logic. In order to speak g r a m m a t i c a l l y we have to be able to
perations, b e c a u s e in their typical uses a g r e e m e n t with t h e m , if gen-
make, at a n y rate implicitly, t h e logical distinction; and w h e n s t r u c -
uine, r e q u i r e s a c t i o n in c o n f o r m i t y with t h e m , in s i t u a t i o n s w h e r e
tural linguists c o n s t r u c t their ' t r e e s ' (which in my s c h o o l days w a s
the action required is an action of the person agreeing. I deliberately
called ' p a r s i n g ' ) , t h e y a r e using t h e logical distinction in o r d e r to
say in their typical uses', b e c a u s e , as is well k n o w n , t h e r e are o t h e r
m a r k off noun-phrases from verb-phrases.
uses, w h i c h h a v e g e n e r a t e d a vast l i t e r a t u r e . S u c h a r e u s e s b y t h e
T h e r e are complications here into which I shall not be able to g o
weak-willed person, ' a c r a t i c ' or 'backslider' w h o does not do w h a t he
for example, the false thesis held by many, including Aristotle (An. Pr.
agrees he o u g h t to, b e c a u s e he very m u c h w a n t s n o t to (H 1 9 6 3 a : c h .
4 3 3 o ) . t h a t there are t e r m s which c a n occupy either subject or pre-
5, 19924": ii. 1 3 0 4 ) , a n d by t h e 'satanist' w h o does w h a t he a g r e e s he
dicate places in propositions at will. T h e truth is that in 'Red is a pri-
o u g h t n o t to. j u s t b e c a u s e i t i s w h a t h e o u g h t n o t t o ( H 1 9 9 2 ^ : 9 8 ) .
m a r y c o l o u r ' and ' T h e book is red', t h e word 'red' m e a n s different
This is not t h e place to add to this literature; the point here is just that
things, as is shown by the fact t h a t we could rewrite the lirst s e n t e n c e
typical and c e n t r a l uses of 'ought' require c o m p l i a n c e if t h e y are to
' T h e colour red is a primary colour', but could not rewrite the second
c o u n t as sincere. By contrast, constative speech a c t s require only a c -
' T h e book is the colour red'. Similarly, in 'Callias is a m a n ' we c a n sub-
cordant belief.
stitute ' h u m a n ' for 'man'; but in 'Man is an a n i m a l ' we c a n n o t . As we
I lowever, m o r a l judgements are not just like ordinary imperations.
have seen, if we a l t e r t h e m o o d of a s e n t e n c e , then by m a k i n g t h e
T h e y s h a r e with c o n s t a t i v e s p e e c h a c t s a very i m p o r t a n t feature,
g r a m m a t i c a l c h a n g e we a l t e r both its m e a n i n g and its logical prop-
namely that when I say 'I o u g h t to do that', I have to say it b e c a u s e of
erties: a n d this is e n o u g h to s h o w t h a t m o o d is both a logical a n d a
something about t h e act t h a t I say I o u g h t to do. T h i s is a feature of all
g r a m m a t i c a l category, without in this context distinguishing the two
uses of 'ought', and not just of moral uses. It is true t h a t imperations
functions.
too are n o r m a l l y issued for r e a s o n s . B u t they do n o t h a v e to be. If a
1 . 4 . I t i s time t o t u r n b a c k t o t h e q u e s t i o n o f w h a t place m o r a l
drill serjeant is trying to see w h e t h e r a n e w recruit will obey him, he
j u d g e m e n t s occupy in the a n a t o m y of language, presuming t h a t we
may say to h i m 'Right turn', and m a y have no r e a s o n at all for saying
have an adequate one. If it is adequate, it will at least distinguish be-
this r a t h e r t h a n 'Left t u r n ' . B u t with 'ought' it is different. To take a
tween two genera of speech acts t h a t I shall call t h e descriptive a n d
n o n - m o r a l e x a m p l e : suppose t h a t instead t h e y a r e doing a t a c t i c a l
THE E N T E R P R I S E 0 1 - MORAL P H I L O S O P H Y
I. 1.4
I. 1.4
1?
e x e r c i s e and the i n s t r u c t o r says 'You o u g h t to a t t a c k on t h e r i g h t ' .
officer could n o t say t h a t t h e r e m i g h t be a n o t h e r t a c t i c a l s i t u a t i o n
T h e r e has to be a reason in the facts of the situation why they o u g h t
just like this o n e in w h i c h they ought to attack on the left r a t h e r t h a n
to attack on the right rather than on the left (VR 3.3).
on the right. If the facts are just the same, they would supply a reason
It is hard for G e r m a n s to a p p r e c i a t e this point, b e c a u s e t h e
for m a k i n g t h e s a m e n o r m a t i v e j u d g e m e n t . T h i s is the basis for t h e
G e r m a n word 'soli' c a n be used to t r a n s l a t e both t h e English 'is t o '
feature of n o r m a t i v e j u d g e m e n t s called universalizability (II 1 9 6 3 :
(which c a n be equivalent to an imperative), and the English ' o u g h t
ch. 2). and moral j u d g e m e n t s share this feature ( 1 . 7 ) .
to' (which is a moral or o t h e r n o r m a t i v e expression). S y s t e m s of de-
1 . 5 . Before a s s i g n i n g t o m o r a l j u d g e m e n t s t h e i r p l a c e o n this
ontic logic have sometimes been set up which fail to make this distinc-
anatomy, there is an important distinction to be made, which in spite
tion, using a single symbol (for example '()") for both 'ought' and t h e
of a very clear statement of it by Austin ( J 9 6 2 : c h s . 9, t o ) , is still neg-
imperative. Since the logical behaviour of these is different (for e x a m -
lected by many, especially in c o n n e c t i o n with i m p e r a t i o n s . It is e n -
ple a 'square of opposition' which works for 'ought' does not work for
couraged by a too easy use of t h e term ' p r a g m a t i c s ' ( 6 . 5 ) , a n d of the
19(171/). s u c h s y s t e m s start on t h e w r o n g foot.
VVitlgensteinian linking of m e a n i n g to use, by those w h o are not very
Confusion on this point c a n s o m e t i m e s lead to treating the fact t h a t
c l e a r a b o u t what exactly they m e a n by 'use' (see Austin 1 9 6 2 : 1 0 4 ) .
o n e is c o m m a n d e d to do s o m e t h i n g (one is to or soil a c t in a c e r t a i n
Austin distinguished between illocutionary and p e r l o c u t i o n a r y a c t s
way) as showing that o n e outfit to act in that way. This c a n have grave
( 6 . 4 ) , the first being w h a t we are doing in saying s o m e t h i n g (in locu-
imperatives
II
political consequences (II K ) ^ / ' ) .
tione). a n d the s e c o n d what we are doing or seeking to do by saying
B e c a u s e m o r a l j u d g e m e n t s h a v e to he made for r e a s o n s , t h e
s o m e t h i n g {per loattionem). T h e ' p r a g m a t i c s ' a n d t h e ' u s e ' of utter-
r e a s o n s being the facts of t h e s i t u a t i o n , it is irrational to issue o n e
a n c e s are easily taken to mean the latter, especially in the c a s e of im-
having no regard for the facts ( c o n t r a s t the Serjeant's c o m m a n d in
peratives: a n d so people slip into thinking t h a t their m e a n i n g c a n be
t h e above e x a m p l e , w h i c h in no way c o n v i c t s the s e r j e a n t of i r r a -
fully explained by giving t h e i r p r a g m a t i c s or use, u n d e r s t a n d i n g by
tionality). It is indeed t r u e that the c h o i c e s expressed by imperative
this their intended perlocutionary effect.
speech a c t s are n o r m a l l y required to be m a d e for r e a s o n s if t h e
Besides t h e temptation just mentioned, there are others. M a n y lo-
c h o o s e r is not to be called irrational (H 1971)). and that even in this
g i c i a n s still hold t h e view, in spite of Austin a n d W i t t g e n s t e i n , t h a t
u n u s u a l c a s e t h e s e r j e a n t has a r e a s o n for saying what he says
t h e r e is only o n e kind of l a n g u a g e - g a m e or s p e e c h act t h a t is r e -
(namely the intention to test the obedience of the recruit). But in this
spectable e n o u g h to be worthy of their attention, n a m e l y the c o n s t a -
c a s e he could have said 'Left t u r n ' instead of 'Right turn' with equal
tive. T h e y s o m e t i m e s cite Aristotle in their support (De Int. i 6 3 3 ff-)-
rationality. It is the privilege of Serjeants not to have reasons for this
Others are so a t t a c h e d to truth-table and similar methods for setting
kind of choice.
up a logic that t h e y c a n n o t see how o n e could be set up t h a t dealt
Moral and other normative j u d g e m e n t s by contrast c a n n o t be ar-
with a n y t h i n g but true-or-false propositions. O t h e r s wish to define
bitrary in this way. T h e y have to be m a d e b e c a u s e of t h e facts. T h i s
'valid inference' as 'inference of such a form that no inference of t h a t
does not m e a n that the m o r a l j u d g e m e n t follows logically from t h e
form c a n have true premisses and a false c o n c l u s i o n ' .
l a d s (II 19(1 5/1: sec. 8). T h e facts do not fonc us logically to m a k e o n e
S u c h writers exhibit the s a m e sort of prejudice as has been in evid-
moral judgement rather t h a n a n o t h e r : but. if we make o n e about o n e
e n c e in c o n n e c t i o n with the truth-condition t h e o r y of m e a n i n g . But
situation, we c a n n o t , while admitting that t h e facts are t h e s a m e in
t h e r e are m a n y o t h e r ways of setting up logics, in p a r t i c u l a r t h a t
a n o t h e r situation, in the s a m e b r e a t h m a k e a conflicting o n e a b o u t
which starts from the notion of inconsistency. If we k n e w h o w to tell
the second situation. In the non-moral tactical example just used, the
which speech acts were inconsistent with which, we could c o n s t r u c t
14
I. i. 5
I. 1. S
15
a logic for those kinds of speech acts. And imperations c a n c e r t a i n l y
tionary force of o u r u t t e r a n c e s explicit a n d t h u s disambiguating t h e
b e i n c o n s i s t e n t with o n e a n o t h e r (for e x a m p l e ' S h u t t h e door' a n d
s e n t e n c e . We c a n do this by saying e i t h e r T w a r n you t h a t t h e ice is
'Do not s h u t t h e d o o r ' ) . T h e i n c o n s i s t e n c y lies h e r e w i t h i n w h a t I
thin' or T affirm t h a t the ice is thin'.
have called the phrastic. which the imperative shares with its c o r r e s -
Be t h a t as it may, t h e l o c u t i o n a r y a n d i l l o c u t i o n a r y a c t s lie to-
ponding indicative; so t h e s o u r c e of i n c o n s i s t e n c y is t h e s a m e for
gether on the other side of an important divide from the perlocution-
both, and therefore so is t h e n a t u r e of the logical fault. In this c a s e ,
ary. For perlocutionary acts there c a n be no logic in a strict sense. T h e
t h o u g h not always (LM 2 . 3 , Searle and Vanderveken 1 9 8 5 : 1 5 2 ) , t h e
r e a s o n is t h a t , as we h a v e s e e n , logic is d e t e r m i n e d by t h e r u l e s or
sign of negation is part of t h e phrastic. But there is n o t h i n g h e r e to
c o n v e n t i o n s for t h e use of words, and p e r l o c u t i o n a r y a c t s ( w h a t we
m a k e us banish imperative speech acts from logic. Indeed, the rules of
are doing or trying to do by saying things) need n o t be controlled by
logic itself, for example formation rules and rules of inference, are im-
any rules or c o n v e n t i o n s of a logical sort (cf. Austin 1 9 6 2 : 1 1 8 ) . It is
perations, and they have to be consistent.
true t h a t w h a t we c a n do by saying s o m e t h i n g depends on w h a t t h e
But the greatest temptation to this way of thinking a b o u t imper-
something isi.e. on w h a t we are doing in saying itbut it depends
a t i o n s (that they have only p r a g m a t i c s and no logic) is a confusion
on m u c h else; we have to size up the situation a n d think w h a t would
between illocutionary and perlocutionary acts. Here it is n e c e s s a r y to
be t h e likely effects of c e r t a i n utterances. Telling s o m e o n e t h a t the ice
depart from Austin's view. He distinguished between three, n o t j u s t
is thin m a y be a way of getting him not to go on t h e ice; but if he is a
two. kinds of act. the third being the locutionary (Austin 1962: J O 8 ) .
daredevil w h o does not fear cold water it may be a way of getting h i m
But if he t h o u g h t that only the l o c u t i o n a r y act had m e a n i n g a n d I
to go on it. If he is a n o r m a l person w h o trusts us, it m a y be a way of
have argued elsewhere that this is a misinterpretation (H 19711': 115
getting him to believe t h a t it is thin; if he is untrusting or c o u n t e r s u g -
ff.)he w a s clearly w r o n g ; for, as we have seen, mood is part of
gestible, it m a y be a way of getting h i m to believe t h a t it is n o t thin.
m e a n i n g ( ' G o ' a n d ' Y o u a r e going t o go' d o not m e a n t h e s a m e ) .
And similarly with imperatives. Say 'Go on the i c e ' to a trusting child,
Therefore, in order to understand w h a t somebody meant, we have to
a n d he m a y go; b u t say it to an u n t r u s t i n g or r e b e l l i o u s o n e , a n d it
k n o w w h a t m o o d his s p e e c h a c t was in. And this is to k n o w s o m e -
m a y m a k e h i m do t h e opposite. T h u s the s a m e illocutionary act with
t h i n g about its i l l o c u t i o n a r y force. It is therefore i n c o h e r e n t to say,
the s a m e m e a n i n g m a y have different perlocutionary effects, and this
both t h a t l o c u t i o n a r y a c t s are the sole repositories of m e a n i n g , a n d
in itself shows t h a t the perlocutionary effect or intended effect is not
that o n e c a n specify the l o c u t i o n a r y act without m e n t i o n i n g its illo-
part of the m e a n i n g .
c u t i o n a r y force. M e a n i n g is. in part, i l l o c u t i o n a r y a c t p o t e n t i a l
W h a t m a y b e called t h e 'verbal shove' t h e o r y o f t h e m e a n i n g o f
(Alston 1 9 6 4 : 37 IT.). T h i s does not necessarily imply t h a t o t h e r ele-
imperatives h a s therefore to be rejected (LM 1.7, II 1971c: 91 ff., 6 . 3 ) .
ments in the illocutionary force c a n n o t extend beyond t h e locution-
If ' p r a g m a t i c s ' is t a k e n confusedly to cover b o t h i l l o c u t i o n a r y and
ary act as specified. It h a s been alleged, for example, t h a t we c o u l d
perlocutionary a c t s , we c a n say that to study t h e m e a n i n g of imper-
k n o w w h a t a person m e a n t w h e n he said ' T h e ice is thin', a n d t h u s
atives is to study their pragmatics; but only t h e i l l o c u t i o n a r y p a r t of
know what locutionary act he performed, without knowing w h e t h e r
their pragmatics at the most. If we stray beyond this, we are no longer
he intended it with t h e i l l o c u t i o n a r y force of a w a r n i n g or a m e r e
studying t h e i r m e a n i n g at all. O n c e we realize this, we shall n o t in-
statement of fact. 1 would dispute this, but it would need too long an
clude as imperations speech a c t s which are clearly s t a t e m e n t s , s u c h
e x c u r s u s into such notions as w a r n i n g to settle the matter. I deal with
as T h e r e is dust on t h e table' said by a demanding lady to her house-
' w a r n ' briefly in 3 . 3 . It c a n at a n y r a t e be g r a n t e d t h a t , as A u s t i n
maid. It h a s b e e n alleged t h a t this is really an imperation, b e c a u s e it
( 1 9 6 2 : 3 2 . 6 9 ) pointed out. there are often ways of making the illocu-
is intended to get t h e h o u s e m a i d to dust the table. It m a y indeed be so
if.
TIIH E N T E R P R I S E OF M O R A L P H I L O S O P H Y
I. 1 . 5
I. 1 . 6
17
intended: but that does not m a k e it an imperation. It is a s t a t e m e n t ,
perlocutionary a c t with t h e logic-governed illocutionary act. ( 6 . 3 f.)
which, in c o n j u n c t i o n with an assumed standing order of the house
So they not only t h o u g h t without good reason t h a t there could be no
( w h i c h is an i m p e r a t i o n ) t h a t w h e n tables a r e dusty s h e is to dust
logic of i m p e r a t i o n s , but. b e c a u s e of this c o n f u s i o n , t a i n t e d m o r a l
t h e m , entitles t h e h o u s e m a i d to infer t h e i m p e r a t i o n t h a t s h e is to
j u d g e m e n t s w i t h the s a m e irrationality. I have even heard it argued
dust the table. So. if the housemaid is both logical and obedient, saying
t h a t , b e c a u s e m o r a l j u d g e m e n t s a r e m a t e r i a l for r a t i o n a l t h o u g h t
this will get her to dust the table. But s h e h a s understood t h e m e a n -
and imperatives a r e not. m o r a l j u d g e m e n t s c a n n o t b e imperatives.
ing of t h e u t t e r a n c e perfectly well even if s h e is not obedient, a n d
B u t the boot is on t h e o t h e r foot. B e c a u s e i m p e r a t i o n s have to obey
even if she has not heard of the standing order, and even if she is too
logical rules, t h e fact t h a t m o r a l j u d g e m e n t s s h a r e s o m e o f t h e i r
stupid to think that there might be one. If she is stupid e n o u g h , s h e
properties is no o b s t a c l e at all to t h e r a t i o n a l i t y of m o r a l t h i n k i n g .
may not dust the table even if t h e t o n e of her mistress is m e n a c i n g .
Therefore rejections of non-descriptivist ethical theories by aspiring
S h e will not know what to do. because she has not been told that.
rationalists on the ground that moral judgements could not be
The relevance of all this to ethics is this. Moral j u d g e m e n t s
rational unless they were s t a t e m e n t s in the n a r r o w s e n s e o r c o n -
are. in a sense to be explained later, prescriptive, and therefore akin in
stative. to use Austin's term ( 1 9 6 2 : 6 n . ) m i s s the point entirely. It
some respects to imperations. T h e school of moral philosophers called
can be allowed that in certain senses moral judgements c a n be called
emotivists (further discussed in Chapter 6) realized this. B u t . infected
true or false (H 1 9 7 6 b ) ; b u t even if t h e y could n o t , t h e i r r a t i o n a l i t y
with the confusion about pragmatics that 1 have just been exposing,
would not be i m p u g n e d . We shall see later t h a t t h e prescriptivity of
they were led into the e r r o r of t h i n k i n g that the m e a n i n g of m o r a l
m o r a l j u d g e m e n t s , so far from being a b a r to t h e i r rationality, is a
j u d g e m e n t s had to be explained in terms of their perlocutionary ef-
vital ingredient in it ( 1 . 8 ) .
1.6.
fect (Urmson 1 9 6 8 : 29 ff.). T h i s is evident from the title of the part of
B u t before s h o w i n g this, it is t i m e to a s k in w h a t s e n s e m o r a l
S t e v e n s o n ' s Ethics and language, ' P r a g m a t i c Aspects of M e a n i n g '
j u d g e m e n t s a r e prescriptive, and h o w their prescriptivity c o m b i n e s
( 1 9 4 s : 17). w h i c h sets t h e t o n e for t h e w h o l e book. B u t t h e s a m e
with their o t h e r features. And this c a n n o t be clarified until we have
t h o u g h t is to be found in Ayer ( 1 9 3 6 : c h . 6 ) , and seems to be implicit
explained w h a t prescriptivity is. We have a l r e a d y used t h e word to
in Carnap ( 1 9 3 5 : 2 3 ) . It led people to look for the source of the m e a n -
describe the genus of speech a c t s to w h i c h imperations belong; t h e y
ing of imperations, and therefore of part of that of moral judgements,
are the paradigm of it. T h e simplest way of characterizing this genus'
in their power of getting people to do things. B u t the p e r l o c u t i o n a r y
is to say that a speech a c t is prescriptive if s o m e o n e w h o assents to it!
act of getting them to do something is a quite different thing from the
is not being sincere if he does not act accordingly (i.e. at t h e time and j
1 9 s i n ) . As we have seen,
in the way specified), when he is the person w h o m it c h a r g e s with ful-1
the latter may be a m e a n s of a c h i e v i n g the former; but this does not
filling it, and is physically and psychologically able to do so (LM 2 . 2 ) . \
illocutionary act of telling t h e m to do it (II
m a k e t h e m the same act in the sense relevant here. In particular, the
B u t t h e r e a r e s o m e a m b i g u i t i e s h e r e w h i c h need t o b e unravelled.
illocutionary act of telling to is subject to logical control, just like t h e
Expressions like ' t h e s u b j e c t ' a n d 'the addressee' (of an i m p e r a t i o n )
illocutionary act of telling that. In telling to. one must not contradict
c a n m e a n t h r e e different t h i n g s . T h e y m a y d e n o t e t h e p e r s o n t o
oneself, a n y m o r e t h a n in telling t h a t : o t h e r w i s e o n e is n o t telling
w h o m an i m p e r a t i o n is spoken or w r i t t e n . Or they m a y d e n o t e t h e
people to do a n y t h i n g t h a t they c a n do. B u t in getting to, including
person o r t h i n g t o w h i c h t h e g r a m m a t i c a l s u b j e c t o f t h e s e n t e n c e
getting to believe that, one may contradict oneself if that is the most
used refers. O r t h e y m a y m e a n t h e person c h a r g e d w i t h c o m p l y i n g
effective way of doing it.
with the imperation. T h e s e m a y all be different persons or t h i n g s . If
T h e emotivists thus confused the essentially irrational or arational
the gmnde dame in our previous example says to her butler ' T h e table
IS
I. 1.6
I. 1. 7
19
l ,
is to be dusted', t h e g r a m m a t i c a l subject refers to the table; t h e per-
h a l f a book to resolving the S o c r a t i c problem {Eth. Nic. i i 4 5 2 i ff.).
son spoken to is the butler; and the person who is charged with c o m -
b e c a u s e he. like S o c r a t e s , h a s to explain h o w o n e c a n a c c e p t a pre-
plying is the housemaid (butlers do not dust tables).
scription a n d t h e n n o t a c t on it. His e x p l a n a t i o n , t h o u g h n o t c o m -
In the present context it is the person charged that interests us. Let
pletely adequate, is m o r e subtle t h a n t h a t of S o c r a t e s . It consists in
us call her, not the addressee or the subject, but the c h a r g e e . A pre-
pointing out that the prescription in question is universal (the house-
scriptive speech act is one such that, if I am the chargee, and I assent
maid k n o w s she o u g h t to dust t h e table b e c a u s e s h e k n o w s t h e uni-
to t h e speech a c t . I c a n n o t be a s s e n t i n g sincerely if I do n o t act a c -
versal rule o f t h e h o u s e h o l d , and all h o u s e h o l d s t h a t a r e well
cordingly. For e x a m p l e , if t h e above c o m m a n d is addressed to t h e
ordered, that dusty tables ought to be dusted, and k n o w s the particu-
h o u s e m a i d , w h o knows t h a t she is the person charged with dusting
lar fact t h a t this t a b l e is d u s t y ) . T h o u g h his e x a m p l e is different,
tables w h e n t h e y a r e to be dusted, and s h e a s s e n t s by s a y i n g 'Very
Aristotle could say that she c a n backslide from t h e universal rule be-
good, m a d a m ' , she is not assenting sincerely if. though she could dust
c a u s e she is tired and w a n t s to go to bed, a n d therefore i g n o r e s t h e
the table, she at o n c e slinks off to bed without dusting it.
particular fact, even though it is evident e n o u g h . This s u m m a r y does
1 . 7 . Are m o r a l j u d g e m e n t s prescriptive in this sense? Certainly
not do justice to the subtlety of his solution to the problem, and I have
not all are. T h e housemaid c a n assent to t h e judgement (even t a k e n
myself suggested a m o r e c o m p l e x s o l u t i o n ( F R c h . 5, H 1 9 9 2 c : ii.
in a moral sense) that she ought to dust the table, and still slink off to
1 3 0 4 ) . B u t the important thing is that there is a problem, which there
bed. T h e q u e s t i o n is rather, 'Is t h e r e an i m p o r t a n t c l a s s of m o r a l
would not be if we were descriptivists.
j u d g e m e n t s w h i c h is prescriptive, a n d if so w h a t is t h e r e l a t i o n b e -
If, in spite of this alleged difficulty, we recognize t h a t c e n t r a l cases
tween t h o s e that are and t h o s e that a r e n o t ? ' It can be argued (but
of moral j u d g e m e n t s a r e prescriptive, we still have to recognize also
not here) that Plato (see H 1 9 8 2 a : 5 6 , 6 6 ) . Aristotle {Eth. Nic. H 4 3 8 ,
that they are not purely prescriptive. T h a t indeed is t h e m a j o r part of
n 4 7 " 2 5 ff.), Hume (1739: III. 1 . 1 ) , Kant (Gr B A 3 6 f. = 4 1 2 f.) and Mill
the more complex solution to the problem of acrasia. As Aristotle
( 1 8 4 3 : last chapter) all thought t h a t moral judgements were typically
{Eth. Nic. U 4 7 3 i ) a n d K a n t (Gr s e c . 2 , p a r a . 3 1 ) b o t h saw. m o r a l
prescriptive, t h o u g h probably n o n e of t h e s e t h o u g h t t h a t all were,
j u d g e m e n t s a r e n o t merely prescriptive but universally prescriptive.
nor that this exhausted their meaning, any more than I do (H 1 9 9 8 a ) .
A n d t h e u n i v e r s a l i t y of t h e m o r a l prescription easily i n t r o d u c e s a
I have a r g u e d e l s e w h e r e t h a t t h e r e is a prescriptive use of m o r a l
n o n - p r e s c r i p t i v e e l e m e n t i n t o its m e a n i n g . To e x p l a i n t h i s : if t h e
judgements, and that this is c e n t r a l in two senses. T h e first is that, if
h o u s e m a i d a c c e p t s t h e universal rule t h a t dusty tables o u g h t t o b e
this use is explained, the o t h e r s c a n be explained in terms of it a n d
dusted, this rule will a s s u m e for her (obedient girl as she normally is)
fall into place (1M ch. 11). T h e second is that, as I shall be saying later,
t h e status of fact. T h a t is. if ever s h e is tempted (as n o w ) to n e g l e c t
t h e i r prescriplivity is a vital ingredient in m o r a l r e a s o n i n g ( 1 . 7 :
her duty, she will not be able to avoid thinking of t h e possibility t h a t
A/T6.1).
her mistress o r t h e b u t l e r will n o t i c e t h e o m i s s i o n a n d p u n i s h h e r :
It w a s his r e c o g n i t i o n , i n h e r i t e d from S o c r a t e s a n d P l a t o , t h a t
and. if they do, that is a real e n o u g h fact. And so is the fact t h a t she is
m o r a l and o t h e r n o r m a t i v e j u d g e m e n t s are prescriptive, that m a d e
frightened by t h e t h o u g h t . S o m e people's attitude to morality is like
a c r a s i a or w e a k n e s s of will a problem for Aristotle. If they a r e pre-
that of t h e h o u s e m a i d to the butler. Lven w h e n t h e h o u s e m a i d h a s
scriptive, how could the h o u s e m a i d a s s e n t to o n e and t h e n slink off
left that (or all) e m p l o y m e n t and has a house of her o w n e v e n w h e n
to bed? If Aristotle had been a pure descriptivist, as s o m e of his pre-
t h e r e is no l o n g e r a grande dame and t h e b u t l e r is o u t of w o r k s h e
tended modern followers seem to themselves to be. there would have
will not be able to escape the feeling of guilt caused in her by t h e sight
been no problem for him in the h o u s e m a i d ' s backsliding. He devotes
of a dusty table for which she is responsible.
2(.
Till- E N T E R P R I S E OF MORAL P H I L O S O P H Y
I. I. 7
II is e a s y for t h e irreligious to proceed from this a n a l o g y to t h e

thought that God does not exist, and that therefore everything is per-
Li.7
21
figure. T h u s the descriptive m e a n i n g of 'good figure' is different in the

two societies.
mitted. T h e y should reflect on two things. T h e lirsl is that, God or no
B e c a u s e t h e s t a n d a r d s or criteria for c o m m e n d a t i o n v a r y from
God. the attitudes that make us revere the laws of morality are a so-
society to society and from c e n t u r y to century, w h e t h e r we are speak-
cial necessity; we could not live in c o m m u n i t i e s without t h e m . Kant
ing o f m o r a l o r o f o t h e r kinds o f c o m m e n d a t i o n , t h e descriptive
may have carried this reverence to excess, and his moral law was no
m e a n i n g of words like 'good', 'right', 'wrong', a n d 'ought' c a n be re-
doubt too simple and rigid. But society would collapse unless children
lied on only within a c e r t a i n circle; but within t h a t circle it is reliable
were brought up to feel bad when they do bad things; and we should
e n o u g h . Other evaluative and normative words have their descriptive
not let psychologists c o n v i n c e us o t h e r w i s e w i t h o u t empirical evid-
m e a n i n g so firmly tied to them that it is hard to use t h e m in c o m m u n -
e n c e to the contrary. T h e second is that a reflective critical morality
ication between different societies; so that, if we were confined to the
c a n justify these laws or rules or principles and our attitudes to t h e m .
latter c l a s s of words (for e x a m p l e ' b l a s p h e m o u s ' a n d ' c r u e l ' ) , we
So even if t h e r e were no grande dame we would have to invent her.
might not be able to talk about values to those w h o did not substan-
Critical moral thinking can also amend the principles if they are seen
tially share our own values. We should have to light o n e another. It is
to be unsuitecl to our situation ( M T 3 . 3 ) .
the existence of shared general value-words like ' o u g h t ' t h a t m a k e s
T h e inescapable factuality or descriplivity of moral principles h a s
peaceful discussion between cultures possible (H 1 9 8 6 c . 19933, 6 . 9 ) .
a logical as well as a psychological basis (Af7'ch. 2 ) . Moral judgements
Moral j u d g e m e n t s a c q u i r e a descriptive m e a n i n g , even w i t h o u t
are like factual statements in m a n y respects (on the face of it. they re-
butlers to enforce t h e m , b e c a u s e of an i m p o r t a n t logical feature t h a t
semble each other more than either of them resembles imperations).
they s h a r e with o t h e r value judgements, called univcrsalizability ( F R
It is easy, therefore, to think that they are like them in all respects. It is
2 . 2 ) . One way of a p p r o a c h i n g this is to say t h a t all s u c h j u d g e m e n t s
made easier still by the existence of a large class of moral judgements,
are m a d e for r e a s o n s : t h a t is, b e c a u s e of something about t h e s u b j e c t
referred to above, which are not prescriptive. T h e similarity is so great
of t h e j u d g e m e n t . T h e girl's figure c o u l d n o t be good if it were n o t
that I have thought it right to follow Stevenson ( 1 9 4 s : 62 ff.) in using
good b e c a u s e of s o m e t h i n g about her m e a s u r e m e n t s . A m a n c a n n o t
t h e t e r m descriptive m e a n i n g ' for t h e e l e m e n t in t h e m e a n i n g of
be a good m a n , if not b e c a u s e of the sort of m a n he is. An a c t c a n n o t
moral j u d g e m e n t s that makes them like constative speech a c t s . This
be wrong, if not b e c a u s e of something about it. T h e y c a n n o t be good
is not t h e s a m e as the phrastic referred to above ( 1 . 3 ) ; t h a t is s o m e -
or wrong just b e c a u s e t h e y a r e good or wrong; t h e r e m u s t be prop-
t h i n g else, which would indeed be part of moral j u d g e m e n t s even if
erties other t h a n their goodness or w r o n g n e s s w h i c h m a k e t h e m so.
they were plain imperations, which they are not.
This feature of value judgements is sometimes called 'supervenience'.
T h e element I am calling descriptive m e a n i n g can best be indicated
C a u s a l j u d g e m e n t s have it too: if an event c a u s e s a n o t h e r , t h e r e
by a n o n - m o r a l e x a m p l e borrowed from Urmson ( 1 9 6 8 : r 3 3 ) . If you
could not be a qualitatively identical s i t u a t i o n in w h i c h t h e c o r r e s -
are meeting a girl at the station and do not know her by sight, I m a y
ponding e v e n t s w e r e n o t c o n j o i n e d and c a u s a l l y linked. T h i s is t h e
e n a b l e you to recognize her by saying, a m o n g other things, that s h e
basis o f t h e s o - c a l l e d ' c o v e r i n g l a w ' t h e o r y o f c a u s a l e x p l a n a t i o n
h a s a good figure. To say this is to describe her. and my purpose h a s
(Hempel 1 9 6 5 : 3 4 5 ff.). And t h e n o t i o n h a s o t h e r a p p l i c a t i o n s t o o .
nothing to do with prescribing the acquisition of such ligures. We all
B u t moral philosophers should not be misled by philosophers of mind
k n o w w h a t in our society c o u n t s as a good figure, so you will k n o w
a n d o t h e r s w h o h a v e b o r r o w e d t h e word a n d used i t i n a n o t h e r
what to look for. If your informant were a m e m b e r of a society t h a t
m e a n i n g w h i c h they have not made clear (H 1 9 8 4 b ) .
thought fat girls more attractive, you would look for a different sort of
T h a t m o r a l properties supervene o n n o n - m o r a l properties m e a n s
I. I. 7
I. 1 . 8
23
simply t h a t a c t s . e t c . . have t h e m o r a l properties b e c a u s e t h e y h a v e
respect of a universal relational property, b e c a u s e in one t h e liar and
the n o n - m o r a l properties ('It is wrong b e c a u s e it was an a c t of inflict-
the person to be lied to are related as m o t h e r to child, and in t h e other
ing pain for fun'), although the moral property is not the s a m e prop-
not.
erty as the non-moral property, nor even entailed by it. S o m e o n e w h o
Examples like this force us to make clear w h a t the thesis m e a n s by
said that it was an act of inflicting pain for fun but not wrong would
'universal property'. A simple, but for our present purposes sufficient,
n o t be c o n t r a d i c t i n g himself, t h o u g h m o s t of us would call h i m im-
definition is t h e following. A property is u n i v e r s a l if, in o r d e r to
moral. Logic does not forbid the adoption of different moral standards
specify it, it is n o t n e c e s s a r y to mention any individual (for an appar-
by different people; it simply prohibits a single person from adopting
ent exception, in w h i c h t h e expression referring to t h e individual is
inconsistent standards at the s a m e time, and says that they will be in-
preceded by 'like' or its equivalent, see 5.8 and FR 2 . 2 ) .
c o n s i s t e n t i f h e says c o n f l i c t i n g t h i n g s a b o u t s i t u a t i o n s w h i c h h e
agrees to be identical in their universal properties.
It is s o m e t i m e s c l a i m e d t h a t the thesis of universalizability is inc o n s i s t e n t w i t h t h e principle of the identity of indiscernibles. For it
It has been disputed whether the universalizabilily of m o r a l
claims that, if there were two situations identical in all their univer-
j u d g e m e n t s is a l o g i c a l feature of t h e m , or e m b o d i e s a s u b s t a n t i a l
sal properties, t h e s a m e m o r a l j u d g e m e n t s would h a v e t o b e m a d e
moral principle. A ground for holding the former view is that we react
a b o u t both; b u t t h e principle of t h e identity of i n d i s c e r n i b l e s holds
to b r e a c h e s of the principle in the s a m e sort of way as to b r e a c h e s of
that there c a n n o t be two situations, numerically different, b u t ident-
logical principles. If s o m e o n e says t h a t there are two situations iden-
ical in all t h e i r u n i v e r s a l properties. However, it h a s b e e n c o n v i n -
tical in all their universal n o n - m o r a l properties, but says he t h i n k s
cingly argued t h a t in this extreme form the principle of t h e identity of
1.8.
t h a t the protagonist in one ought to tell a lie. but the protagonist in
indiscernibles is n o t t r u e (e.g. S t r a w s o n 1 9 5 9 : 1 1 9 ) . It is true in less
the other o u g h t not, we are likely to be as nonplussed as if he had said
e x t r e m e forms, e.g. if it c l a i m s only t h a t things identical in all their
t h a t he t h o u g h t that a rotating disc was both stationary and not sta-
u n i v e r s a l p r o p e r t i e s and in t h e i r r e l a t i o n s to individuals m u s t be
t i o n a r y (cf. P l a t o , Rep. 4 3 6 d ) . In e i t h e r c a s e t h e r e c o u l d be an e x -
numerically identical; but this obviously causes no trouble to the uni-
planation. In the second he might m e a n that the axis of rotation was
versalizability thesis.
stationary, so t h a t t h e disc c o n t i n u e d to o c c u p y t h e s a m e r e g i o n of
space, but that within this region it moved around its axis.
I n t h e first c a s e t h e r e c o u l d b e m a n y e x p l a n a t i o n s , b u t n o n e o f
T h e r e is a further problem about w h e t h e r being a c t u a l as opposed

to merely possible or hypothetical is a universal property ( M T 6 . 4 ) . If
it were, a form of s p e c i a l pleading would b e c o m e possible in m o r a l
t h e m would impugn the universalizabilily thesis. T h e protagonists in
reasoning, by which an aggressor could claim t h a t he would never be
t h e two c a s e s m i g h t t h e m s e l v e s have different c h a r a c t e r i s t i c s . B u t
a c t u a l l y in t h e position of his victim, a n d t h a t t h i s difference w a s
when the thesis speaks of identical situations, it must be understood
morally r e l e v a n t . It is perhaps best to follow t h o s e (e.g. Lewis 1973:
as ruling out this difference too. A n o t h e r possibility is that in o n e c a s e
8 5 ) w h o c l a i m t h a t t h e a c t u a l world c a n n o t b e d i s t i n g u i s h e d from
t h e person to be lied to is t h e m o t h e r of t h e protagonist, a n d in t h e
possible worlds w i t h o u t a reference to individuals, n a m e l y those w h o
o t h e r not. O n e c a n only h a v e o n e ( g e n e t i c ) mother, and it m i g h t be
are actual; but not to follow t h e m into t h i n k i n g t h a t possible worlds
t h o u g h t t h a t this m a k e s a difference, b e c a u s e to tell lies to o n e ' s
have s o m e real e x i s t e n c e in limbo. In any c a s e it s e e m s t h a t m a k i n g
m o t h e r is worse t h a n if s o m e o n e else tells t h e m to a person (perhaps
moral d i s t i n c t i o n s o n t h e g r o u n d o f a c t u a l i t y would b e r e j e c t e d o n
even t h e s a m e person) w h o is n o t his mother, however similar t h e sit-
logical grounds as we use words like 'ought'. If s o m e o n e said T o u g h t
uations. But relations c a n be universal properties ( 5 . 8 ) . and the rela-
in t h e a c t u a l case, but s o m e o n e else o u g h t n o t in an identical hypo-
tion being the mother of is o n e s u c h . T h e s i t u a t i o n s a r e different in
thetical c a s e ' , we should n o t understand w h a t m o r a l principle he was
24
T i l l ' E N T E R P R I S E OF M O R A L I'll I LOSOPHY
I. 1 . 8
I.
1. 8
invoking, b e c a u s e a moral principle w h i c h applied to the a c t u a l c a s e
language considerations to an a c c o u n t of substantial morality. Here
but not to hypothetical cases exactly like it would not be c o u n t e d by
we must simply note t h a t formal considerations are only o n e e l e m e n t
us as a moral principle, whatever o u r substantial moral views were,
in moral a r g u m e n t s . Others are the facts about situations, w h i c h are
nor as any other sort of normative principle. This problem has a n a l o -
substantial, and in particular facts about people's wills, to use Kant's
gies with the old one of whether existence is a property.
word: and these facts too are substantial ( 8 . 5 f.).
T h o s e who think that the universalizability thesis is a substantial
Let us try out this essentially Kantian method m o r e clearly, and re-
moral principle and not a logical doctrine will by this time be getting
late it to its basis in philosophy of language. If m o r a l judgements are
restive. T h e y will think that we have fixed the logic so as to enable us
prescriptive, as h a s b e e n argued, t h e n in m a k i n g o n e . I am a s k i n g
to r e a c h s u b s t a n t i a l c o n c l u s i o n s in m o r a l a r g u m e n t s . We must ask
that it be acted on. and. if sincere, must will this. B u t if they are also
them to be patient until we have explained how the a r g u m e n t s work.
universalizable, I a m , in m a k i n g o n e . implicitly m a k i n g i d e n t i c a l
Until t h e n we c a n only point o u t t h a t we would object to t h e above
j u d g e m e n t s for all s i t u a t i o n s identical in their universal properties,
c o n j u n c t i o n of moral judgements about the actual and hypothetical
no matter what role particular individuals, including myself, occupy
cases even if we knew nothing whatever about the substantial moral
in t h e m . T h e question of w h a t m o r a l prescriptions I am prepared to
opinions of the person w h o made them: so it c a n n o t be anything sub-
issue thus resolves itself into t h a t of w h a t I am prepared to will for all
s t a n t i a l t h a t w e a r e o b j e c t i n g to. T h e o b j e c t i o n must therefore b e
situations of a given kind, no matter w h a t role I occupy. T h u s to issue
logical. Suppose, even, that he also says that on oilier g r o u n d s he
a moral prescription I must accept the c o n s e q u e n c e s (even the hypo-
believes in complete impartiality between people, himself and others.
thetical c o n s e q u e n c e s ) of its being obeyed whatever role I occupy.
It is n o t i n c o n s i s t e n t to believe in i m p a r t i a l i t y between people, a n d
How constrictive this is will depend on w h a t I will should be done
still try to call the difference between actual and hypothetical morally
to myself, were I in those various roles. T h e roles include t h e fact t h a t
relevant; for if it were relevant it could be used impartially between
t h e wills of t h e people in t h e m are w h a t they a r e . If I w e r e in t h o s e
people. So we c a n n o t be i n t r o d u c i n g a s u b s t a n t i a l m o r a l principle
situations, my will would in e a c h c a s e be t h e s a m e as t h e present will
r e q u i r i n g impartiality b e t w e e n people by insisting t h a t a c t u a l i t y is
of t h e p e r s o n w h o is n o w in it, s i n c e t h e willing is p a r t of t h e situ-
not a morally relevant feature. On the problem of moral relevance in
ation. So the question resolves itself into t h a t of w h a t I n o w will (NB
general see H 1978b: 73,MT 3.9.
n o t w h a t I would will) s h o u l d be done to me in t h o s e s i t u a t i o n s , in
We have found reasons for thinking that the universalizability thesis
which I willed w h a t they n o w will.
is a logical and not a substantial moral doctrine. T h e main ground on
B u t h e r e a n o t h e r factor e n t e r s , also o b t a i n a b l e from t h e logic o f
which people have thought otherwise is that the thesis does seem to
o u r l a n g u a g e . B y a n a r g u m e n t which does n o t invoke universaliz-
have i m p l i c a t i o n s of a s u b s t a n t i a l s o r t for m o r a l a r g u m e n t s , a n d
ability, we c a n see t h a t I m u s t have as m u c h regard to w h a t I would
there is some suspicion of a conjuring t r i c k o f producing a substan-
will in those situations, as I do to what I n o w will. For if I do not, I am
tial moral rabbit out of a logical hat. Moral philosophers have so often
either not representing t h e situations fully to myself, or else not think-
attempted similar tricks t h a t o n e is right to be suspicious. For e x a m -
ing of the person in t h e m as myself (7.3, MT 5 . 4 ) . To think of h i m as
ple they have sought to attribute a certain m e a n i n g on logical or c o n -
myself is to identify myself with his will. This is part of w h a t we mean
c e p t u a l g r o u n d s t o p h r a s e s s u c h a s ' h u m a n needs', a n d h a v e t h e n
by ' m y s e l f . Reflection on the m e a n i n g of ' m y s e l f should c o n v i n c e us
gone on to extract substantial moral principles from these definitions
of this. T h e c a s e is a n a l o g o u s to w h a t I t h i n k a b o u t future s t a t e s of
( 4 . 6 ) . How we c a n allay this suspicion will n o t be clear until we have
myself w h i c h I e x p e c t to be a c t u a l . If I k n o w w h a t I s h a l l t h e n will,
set out m o r e fully the a r g u m e n t from formal logical or philosophy-of-
a n d am really t h i n k i n g of t h e person in t h e situation as myself, and
2h
THE E N T E R P R I S E OF MOR AI, P H I L O S O P H Y
I. 1 . 8
do n o t i r r a t i o n a l l y discount t h e future, my will m u s t be as strongly
I. 1 . 9
27
in practice we m a n a g e to find a level of moral thinking m o r e suited to
e n g a g e d as t h a t of t h e future p e r s o n w h o will be me. If a n y b o d y
us h u m a n s t h a n the s o m e w h a t demanding level in which Holy Wills
doubts this, he should a r r a n g e for himself to be whipped, and reflect
can engage.
on his s t a t e of mind just before it h a p p e n s (cf. Aristotle. Ulh. Nk.
It was Kant's predicament in between these levels (dare we say his
1 1 1 5 2 4 ) . Failure to engage my will in this way is always due either to
insufficient grasp of an important difference between the levels?) that
a failure of representation of the situation of t h e person t h a t I shall
led h i m to try to justify what are only simple, general, prima facie in-
be. or to a failure to think of him (or her) as myself.
tuitive principles (suitable to our h u m a n condition) directly by appeal
S i n c e for moral a r g u m e n t hypothetical situations are as relevant
t o t h e C a t e g o r i c a l I m p e r a t i v e ; and this n o t o r i o u s l y got h i m i n t o
as a c t u a l . I have to will that t h e s a m e should be done to me in t h e m
trouble ( 8 . 4 ) . T h e right way to try to justify t h e m would have been to
too. T h e y will include all the situations in which I would occupy t h e
show t h a t a Holy Will (perhaps God, w h o m K a n t would have liked to
roles of t h o s e affected by proposed a c t i o n s of m i n e . 1 am therefore
believe in) would, by a use of t h e Categorical Imperative as it would
laced with the problem of finding a universal prescription for situa-
be used by s u c h a supremely rational will, select t h e s e simpler prin-
tions like that which I am presently in. which I can accept equally for
ciples for t h e g u i d a n c e o f less r a t i o n a l wills t h a n his o w n . B u t 'we
all the identical situations that I could be in, in different roles. T h i s in
have no intuition of the divine perfection, b u t c a n only deduce it from
effect gives equal weight in my moral thinking to the wills of all those
our own c o n c e p t i o n s ' (Gr B A 9 2 = 4 4 3 ) . We have no direct a c c e s s to
affected by my actions. T h e Kingdom of Fnds is not really a kingdom,
what a good God would will, so we have recourse to o u r own imper-
b u t a d e m o c r a c y with e q u a l i t y before t h e law. But if all wills h a v e
fect reason as t h e best m e a n s available to us.
e q u a l weight in proportion to t h e i r s t r e n g t h (for obviously h o w
In c o n c l u s i o n , we have to ask, in deference to an earlier objector,
s t r o n g t h e y are m u s t m a k e a difference) t h e n t h e problem of doing
w h e t h e r this development of K a n t i a n ideas relies on resources lying
justice between all these wills is to be resolved by choosing the moral
beyond t h e philosophy of language, a n d in particular on a n t e c e d e n t
prescription which m a x i m a l l y realizes t h e fulfilment of t h e m , treat-
s u b s t a n t i a l m o r a l ideas a n d i n t u i t i o n s . K a n t c a l l e d his m o s t - r e a d
ing all impartially and giving them weight according to their strength
work on this subject Grundlegung zur Metaphysik der Sitten. W h a t has
(H 1996c).
been s k e t c h e d in t h i s c h a p t e r is a kind of Grundlegung zur Logik der
1.9.
This development of K a n t ' s ideas t h u s t u r n s into a kind of
Sitten; and, as we have seen ( 1 . 3 ) , logic a n d m e t a p h y s i c s are h a r d to
rational-will utilitarianism (see Chapter 8 ) . He is. admittedly, selective
tell a p a r t . It c e r t a i n l y does n o t s e e m as if we h a v e relied on e x t r a -
with regard to t h e kinds of will t h a t he is prepared to e n f r a n c h i s e :
logical premisses. A n y o n e w h o doubts this should look for t h e m . T h e
they have to be rational; but m a n y utilitarians accept this. This shows
a r g u m e n t h a s been generated using the following elements: first, the
the superficiality of the c o m m o n l y accepted dogma that Kant a n d the
prescriptivity of m o r a l judgements; secondly, their universalizability;
u t i l i t a r i a n s need to be at odds. If t h e two d o c t r i n e s are s y m p a t h e t -
a n d thirdly, t h e t h e s i s t h a t fully to r e p r e s e n t a n o t h e r ' s s i t u a t i o n to
ically formulated, they are in a g r e e m e n t . T h e disagreement r e m a i n -
oneself o n e m u s t c o m e to have a will similar to his (or hers) for a situ-
ing is o n e within utilitarianism, as to whether any kinds of will are to
ation in w h i c h o n e o c c u p i e s his role. T h e last of these e l e m e n t s w a s
be excluded from consideration. And such a formulation involves t h e
obtained by c o n s i d e r i n g w h a t full r e p r e s e n t a t i o n of a h y p o t h e t i c a l
use of i n s i g h t s from t h e philosophy of l a n g u a g e . T h e r e is no s p a c e
situation m e a n s , a n d w h a t it m e a n s to think of a person in it as my-
h e r e to develop these insights further, n o r to deal with o t h e r o b j e c -
self. All these are c o n c e p t u a l or logical moves, n o t involving appeals
tions a n d difficulties. T h i s must be left until later, and to my writings
to substantial m o r a l intuitions. A l t h o u g h , therefore, they c a n all be
on the philosophy of education (e.g. H 1992d), in order to show how
disputed, t h e disputes will be w i t h i n t h e p h i l o s o p h y of l a n g u a g e ,
28
I. 1.9
s i n c e the t h e s e s t h e m s e l v e s belong to it. So at least we c a n c l a i m to

have s h o w n t h e r e l e v a n c e o f philosophy o f l a n g u a g e t o e t h i c s . B u t
see 5.8 for further discussion of universalizability.
DEFENCE OF THE E N T E R P R I S E
2 . 1 . T h e best way t o understand what moral philosophy is. and why

a n y b o d y s h o u l d wish to study it. will be to t a k e a p r a c t i c a l m o r a l
problem a n d find t h e points at which we seem to be raising philosophical questions in our discussion of it. If my experience is a n y t h i n g to
go by, o n e c a n n o t discuss a n y serious m o r a l problem for m o r e t h a n
about half an h o u r without some philosophical tangle emerging. I
am not g o i n g to discuss in depth a n y p r a c t i c a l m o r a l issue; I h a v e
done t h a t in m a n y of my o t h e r writings, but t h e r e is n o t space for it
here. I want just to show how philosophical questions arise; later we
shall see how the various kinds of ethical theory try to deal with t h e m
(1 give my o w n a n s w e r s in 7 . 9 ) . For this purpose a very sketchy e x a m ple will do: but this is merely by way of illustration. I have done my bit
for practical m o r a l philosophy elsewhere.
The best e x a m p l e to take may be the o n e t h a t got me myself into
moral philosophy: the question of whether it is w r o n g to light in wars
and kill people. In 1 9 3 8 - 9 I had to face the problem, a n d A m e r i c a n s
had to face it at t h e t i m e of t h e V i e t n a m War. It is a p r o b l e m a b o u t
which a n y b o d y w h o t h i n k s seriously a b o u t m o r a l q u e s t i o n s h a s t o
m a k e up his m i n d . I have written a b o u t it in H 1 9 8 5 b , MT 1 0 . 2 , a n d
e l s e w h e r e . T h e A m e r i c a n e x p e r i e n c e illustrates m o r e o f t h e points
that I want to raise t h a n my own experience before t h e S e c o n d World
War. b e c a u s e in t h e c a s e of V i e t n a m o n e c o u l d hold a position a c cording to which, t h o u g h not in general a pacifist, o n e h a d moral objections to fighting in this particular war, a n d s u c h objections were not
in t h e United S t a t e s , a n d would not have been in B r i t a i n , allowed as
reasons for exemption from military service.
Let us t h e n t h i n k a b o u t t h e p r o b l e m s f a c i n g a y o u n g A m e r i c a n
about to be drafted a n d sent to fight in V i e t n a m . Let us suppose t h a t
he is i n c l i n e d t o w a r d s , b u t n o t c o m m i t t e d to, pacifist views; t h a t .
I. 2. 1
w h e t h e r or not he accepts pacifism, he is quite sure that there is somet h i n g w r o n g ( s o m e t h i n g very h a r d to specify e x a c t l y ) with his
c o u n t r y ' s policies and actions in Vietnam: t h a t however, even if it be
agreed that A m e r i c a is doing s o m e t h i n g morally wrong in V i e t n a m ,
t h a t by no m e a n s e n t a i l s t h a t he o u g h t to refuse d o w n r i g h t to be
drafted, or take evasive a c t i o n . For it m i g h t be that, a l t h o u g h his
c o u n t r y was doing wrong, his duty was to his country, right or
wrong. Or, if that be thought too e x t r e m e or too old-fashioned a position, he might think that he was not in a position to judge of the c o m plexities of world strategy, and t h a t it was his duty to leave t h e
decision to those who were belter informed than himself. Or he might
think that although, if one considers Vietnam in particular. America
was doing something wrong, to rebel against one's country would be
a greater evil than acquiescing in this degree of moral evil committed
by o n e ' s g o v e r n m e n t . He m i g h t be quite ready to admit that, if t h e
A m e r i c a n government were to b e c o m e like the Nazis and e m b a r k on
a policy of wholesale genocide (massacring all blacks, for example), it
would be his duty to rebel a g a i n s t it; b u t he m i g h t be in d o u b t
w h e t h e r the United States' government's actions in Vietnam were of
sufficient wickedness to justify him in refusing his n o r m a l duty as a
citizen. But what is one's n o r m a l duty as a citizen?
2 . 2 . I have done enough, perhaps, t o illustrate the complexity o f
the issues that arise in s u c h a c h o i c e - s i t u a t i o n . T h e r e is no r e a s o n
why I should have stopped where I did; I could easily have shown that
in fact the issues are even more complex. But let us stop for a m o m e n t
and try to sort out the complexities that we have encountered so far.
The first group of problems that we have to consider is that raised
by the pacifist position. W h a t is supposed to be wrong about going to
w a r and fighting? We might feel inclined at o n c e to answer t h a t w a r
a n d fighting a r e prima facie w r o n g b e c a u s e they involve killing or
wounding people (to say n o t h i n g of the e c o n o m i c loss that often results from w a r s and the p r e p a r a t i o n for t h e m ) . Most people would
a g r e e t h a t in general o n e o u g h t n o t to kill or wound people. B u t
and h e r e is the difficultymost people would also a g r e e t h a t t h e r e
are p a r t i c u l a r c a s e s in which it is legitimate to kill or w o u n d people
(in self-defence, for example). It is true, however, and important, t h a t
1.2.2
people could be found w h o would dissent from e i t h e r of t h e s e two

propositions to which I have said that most people would agree. T h e r e
have been people (Nietzscheans, for e x a m p l e ) w h o have argued t h a t
fighting is a good t h i n g j u s t b e c a u s e it results in the e l i m i n a t i o n of
the weak a n d t h e i r d o m i n a t i o n by t h e s t r o n g , so t h a t t h e r a c e is
t h e r e b y improved: and t h e killing and w o u n d i n g of people is an
essential part of the process. And there have been others (the followers of Tolstoy, for example, and certain Indian sects), w h o have m a i n tained that absolutely all violence is illegitimate.
However, leaving aside for the present these two e x t r e m e positions,
even the m o r e moderate o n e s in between face us with e n o u g h problems. For if you t h i n k t h a t in general killing or w o u n d i n g people is
wrong, you have o n y o u r h a n d s the problem o f w h a t distinguishes
the classes of c a s e where it is legitimate from those in w h i c h it is not.
And I do n o t see how, in principle, o n e could set o u t to a n s w e r this
question w i t h o u t raising the prior question ' W h a t in general m a k e s it
w r o n g to kill or w o u n d ? ' For only if we k n o w w h a t is w r o n g a b o u t
killing or wounding in general shall we be able to say in w h a t particular cases this general wrongness of killing is either absent, or else outweighed by other considerations which are present in those cases.
2 . 3 . If I m a y allow myself a n o t h e r piece of autobiography: it was
when I saw, l o o m i n g behind the p a r t i c u l a r question, 'Is it w r o n g to
kill people in w a r s ? ' , t h e m u c h m o r e g e n e r a l q u e s t i o n , ' W h y is it
wrong to kill people anyway?', t h a t I really took to m o r a l philosophy
in earnest. T h e r e are, after all, a whole lot of c i r c u m s t a n c e s in which
it would be. to say t h e least, c o n v e n i e n t to kill people. In my y o u t h I
was m u c h addicted to murder stories, and these provide plenty of illustrations of cases in which, for particular people, it would be highly
c o n v e n i e n t t o kill o t h e r p a r t i c u l a r people. B u t t h o s e w e r e n o t t h e
cases I was primarily thinking of. I was thinking more of c a s e s where
even ordinary bien peasants citizens might be tempted to think that it
would be convenient to get certain people out of the way.
Let us s t a r t w i t h s o m e very t e m p t i n g c a s e s . I w a s on a w o r k i n g
party m a n y years ago w h i c h was discussing t h e problem, w h i c h still
vexes the media, and on which there have been important legal decisions r e c e n t l y t h e problem of w h e t h e r a person should in t h e fol-
32
1.2.3
I. 2. 3
33
lowing c i r c u m s t a n c e s be allowed to die: he had an injury to his brain
T h e m a i n p o i n t t o n o t i c e w i t h all t h e s e p r o b l e m s is: u n l e s s o n e
such t h a t it could be safely predicted t h a t he would never recover c o n -
k n o w s w h a t it is t h a t m a k e s killing a n o r m a l adult h u m a n being
s c i o u s n e s s : but the lower c e n t r e s of his brain were in perfectly good
w r o n g , o n e is unlikely to be able to a n s w e r w i t h a n y a s s u r a n c e t h e
order, so t h a t he could be kept alive indefinitely by intravenous artifi-
question of w h e t h e r killing a foetus, or killing a defective n e w - b o r n
cial feeding. T h e question was, ' W a s it legitimate to stop the artificial
baby, or killing at his or h e r r e q u e s t a t e r m i n a l p a t i e n t in agony, in
feeding?'
these exceptional c i r c u m s t a n c e s , shares t h o s e features w h i c h make it
I m a y m e n t i o n that, a long time after t h e working party, my o w n
wrong to kill in t h e n o r m a l case.
sister died of the effects of a stroke, and the s a m e problem might eas-
T h i s w a s essentially S o c r a t e s ' point when, again a n d again in t h e
ily have arisen (she w a s in fact u n c o n s c i o u s for a m o n t h before s h e
dialogues of Plato, he would not allow a n y b o d y to say t h a t he k n e w
died), but mercifully it did n o t . It is i n t e r e s t i n g t h a t Pope P i u s X I I ,
t h a t some particular thing was good or bad or right or w r o n g or any-
w h o s e views were in general r a t h e r conservative, pronounced in an
I h i n g else, until he h a d b e e n given a c l e a r a n s w e r to t h e q u e s t i o n
allocution that in such c a s e s it was legitimate to cease to keep a pa-
' What is it to be good or bad, etc.?' However, we m u s t be c l e a r e r t h a n
tient alive artificially (Acta Apostolicae Sedis x x x i x ( 1 9 5 7 ) : 1 0 2 7 - 3 3 ) .
p e r h a p s S o c r a t e s w a s a b o u t t h e d i s t i n c t i o n b e t w e e n two different
But this probably did not apply to artificial feeding. T h e working party
questions. T h e o n e I have been asking t h r o u g h o u t is ' What is wrong
on which I was sitting consisted of theologians, a distinguished e c c l e -
about, for e x a m p l e , killing people?' T h i s m i g h t be r e p h r a s e d in t h e
siastical lawyer, s o m e well-known doctors, a n d some philosophers.
form ' W h a t is it a b o u t killing people that makes it, in general, w r o n g ? '
Its report w a s supposed to provide g u i d a n c e to c h u r c h people a n d
We must distinguish this from t h e quite different though related ques-
o t h e r s (especially the bishops in t h e House of fords) when s u b j e c t s
tion. ' H o w do we know, or h o w c a n we prove, t h a t it is w r o n g ? '
like this c o m e up.
Somebody might be quite sure about the answer to the question
I have sat on m a n y such working parties, sponsored by the C h u r c h
' W h a t m a k e s it w r o n g to kill people?', b u t be still u n a b l e to say h o w
of England and lay institutions, including a m o r e recent o n e on e u -
he knew this. Both questions can, unfortunately, be expressed by
t h a n a s i a proper. We had o n e on abortion w h i c h , 1 think, h a d s o m e
m e a n s of the a m b i g u o u s formulation ' W h y is it w r o n g to kill people?'
influence, and was helpful in securing the liberalization of t h e law in
T h i s might m e a n ' W h a t is it that makes it w r o n g ? ' , or it m i g h t imply a
1 9 6 7 . 1 shall not discuss a n y of t h e s e subjects at length n o w : I have
question about h o w we k n o w t h a t it is wrong. T h e confusion between
m y s e l f published a r t i c l e s on t h e m as well as s u b s c r i b i n g to t h e r e -
these two q u e s t i o n s h a s got m o r a l philosophers i n t o a lot of trouble.
ports of the working parties (e.g. H 1975c. d, \yHXd, 1993d)- On abor-
B o t h q u e s t i o n s a r e . o f c o u r s e , very i m p o r t a n t o n e s , a n d t h e y a r e
tion, some people are prepared to argue as follows: it is always w r o n g
closely related to e a c h other, but all t h e s a m e distinct. And b o t h are
to kill an i n n o c e n t h u m a n being: but an abortion is killing an i n n o -
distinct from the question ' W h a t does wrong mean?', t h o u g h this too
c e n t h u m a n being; therefore abortion is always wrong. I do not think
is related.
a n y of us on t h e working party were prepared to a c c e p t this simple
2 . 4 . O n c e o n e gets talking about e u t h a n a s i a o f t h e incurably ill,
a r g u m e n t . In our discussions, we of c o u r s e considered the s o m e w h a t
o n e i s n a t u r a l l y led o n t o t h i n k a b o u t t h e p u t t i n g a w a y o f o t h e r
analogous c a s e of the Belgian m o t h e r w h o killed her infant child w h o
people w h o are not actually ill, but merely socially a n u i s a n c e . This is
w a s b o r n deformed as a r e s u l t of t h a l i d o m i d e t a k e n by h e r w h e n
a n e x a m p l e o f t h e 'slippery slope', o r t h e ' t h i n end o f t h e wedge',
p r e g n a n t . And m o r e r e c e n t l y I h a v e t a k e n p a r t in d i s c u s s i o n s a n d
w h i c h h a s figured p r o m i n e n t l y in a r g u m e n t s on all t h e s e questions.
published an article on the t r e a t m e n t of spina bifida cases which raise
People w h o u s e s u c h a r g u m e n t s usually do n o t see w h a t t h e trouble
a similar problem (H 1974/').
is. T h e trouble arises precisely b e c a u s e they have not considered w h a t
THE E N T E R P R I S E OF MORA I, P H I L O S O P H Y
34
I. 2. 4
I. z. 4
35
I called the prior question of w h a t makes killing wrong in general. So
ramifications of t h e question of w h a t is w r o n g with killing people. I
they are naturally at a loss w h e n it c o m e s to drawing the line between
t h i n k t h a t in t h e n e x t few y e a r s we s h a l l h a v e to devote a lot of
c a s e s w h e r e killing is l e g i t i m a t e a n d w h e r e it is not. S u c h people
t h o u g h t to this problem. I w a n t to show h o w m o r a l philosophy enters
s h o u l d b e r e c o m m e n d e d t o a n s w e r t h e prior q u e s t i o n . T h e n t h e y
into this s o r t of d i s c u s s i o n t o i n d i c a t e w h a t is its b e a r i n g on this
might find it easier to draw the line, and find a footing on the slippery
problem about killing people.

2 . 5 . A n y moral problem o n e cares t o take i s bound t o b e divisible
slope.
T h e r e h a s been a great deal of controversy recently, especially in
into the following elements. T h e r e are first of all questions of fact. To
t h e United States, about the death penalty. It may be t h a t in times to
take t h e e x a m p l e I have j u s t b e e n discussing: t h e question, w h e t h e r
c o m e people will think the terms in which this controversy h a s been
t h e psychologists are right w h o say t h a t it is possible to identify gen-
c o n d u c t e d terribly old-fashioned,
and in p a r t i c u l a r t h e word
etic e l e m e n t s in t h e c a u s e s of crime, is a question of fact, w h i c h c a n
'penalty'. We think it was b a r b a r o u s of o u r n i n e t e e n t h - c e n t u r y a n -
be investigated empirically. In most practical m o r a l problems it will be
cestors to h a n g people for sheep-stealing or for destroying the heads
found t h a t t h e h u g e majority of the q u e s t i o n s w h i c h h a v e to be set-
of fishponds. M a n y people now think it b a r b a r o u s to h a n g or o t h e r -
tled before we c a n solve t h e m are factual o n e s . T h i s h a s tempted s o m e
wise execute murderers. B u t suppose we forget about all this old-fash-
p h i l o s o p h e r s t o t h i n k t h a t t h e only q u e s t i o n s t h a t h a v e t o b e a n -
ioned talk of crimes a n d penalties, a n d look at the m a t t e r in an
swered before w e c a n solve t h e m a r e o f this s o r t t h a t o n c e all t h e
e x t r e m e l y practical way. W e spend a n e n o r m o u s a m o u n t o f m o n e y
facts are k n o w n , n o further problem will r e m a i n ; t h e a n s w e r t o t h e
on prisons: t h e people in t h e m do n o t h a v e a n i c e time: s o m e t i m e s
moral q u e s t i o n will be obvious. T h i s is, however, n o t so, as we s h a l l
they escape and endanger the public; a great many of them are either
see i n due c o u r s e . B u t c e r t a i n l y t h e f a c t u a l q u e s t i o n s a r e t h e o n e s
m e n t a l l y a b n o r m a l or for s o m e o t h e r r e a s o n unlikely ever to fit into
t h a t c a u s e 99 per c e n t of the trouble. We c a n see this if we study a n y
society as useful citizens. So why n o t start systematically weeding out
two people a r g u i n g a b o u t a m o r a l q u e s t i o n . We s h a l l n e a r l y always
t h e hopeless cases? We should save a lot of m o n e y and effort, w h i c h
find t h e m disputing e a c h other's facts. To revert for a m o m e n t to t h e
might then be spent on greater endeavours to help the cases t h a t are
problem of t h e draftee w h o h a s to decide w h e t h e r to go into t h e army:
not hopeless.
most of his problem is to find o u t w h a t is a c t u a l l y h a p p e n i n g in, for
T h i s train of t h o u g h t c a n go even further. T h e r e is some, t h o u g h
example, Vietnam, and what the actual consequences of various
not c o n c l u s i v e , e m p i r i c a l e v i d e n c e for t h e thesis t h a t a s u b s t a n t i a l
courses of action, w h e t h e r on his or his government's part, are likely
part of t h e causes which m a k e people take to crime are genetic. If, as
in fact to be.
is not unlikely, we b e c o m e able to identify these criminal factors early
N e v e r t h e l e s s , it is fairly obvious t h a t o n e m i g h t find o u t all t h e
(for e x a m p l e , by w a t c h i n g people's b e h a v i o u r at s c h o o l , t r u a n c y
facts t h a t anybody wanted to adduce, a n d still be in doubt w h a t o n e
being an obvious bad sign, which is said to be linked quite closely with
o u g h t to do. We c a n see this more clearly if we suppose t h a t there are
l a t e r d e l i n q u e n c y ) , why n o t weed o u t t h e s e n o t very hopeful speci-
two draftees a n d t h e y are a r g u i n g with o n e a n o t h e r a b o u t t h e ques-
m e n s early, a n d c o n c e n t r a t e o u r e d u c a t i o n a l efforts o n t h o s e b o y s
tion. It is obvious t h a t they could agree, for example, t h a t if they went
and girls w h o have a good c h a n c e of turning into the sort of men and
into the armed forces and obeyed their orders, they would find them-
w o m e n we want to have in society?
selves killing a lot of civilians in the c o u r s e of a t t a c k s on military ob-
I am not. needless to say. a c t u a l l y advocating such a policy, a n d 1
jectives. O n e o f t h e m m i g h t t h i n k i t m o r a l l y indefensible t o kill
s h a l l say why later ( 7 . 9 ) . I h a v e strayed r a t h e r far from t h e pacifist
civilians in t h e c o u r s e of fighting (especially if the civilians h a d noth-
question with which we started. I wanted us to see how wide are t h e
ing to do with t h e fighting, b u t were i n n o c e n t bystanders). T h e other
I. 2. 5
m i g h t t h i n k t h a t this, a l t h o u g h in itself an evil, had to be d o n e if

necessary in order to secure some greater good. One c a n agree about
a fact, but disagree about its bearing on a moral issue.
However, it is not at all c l e a r what follows from this. S o m e philosophers have gone straight from this premiss to the c o n c l u s i o n t h a t
t h e r e are ineliminable j u d g e m e n t s of value which are logically unrelated to questions of fact, so that o n e can agree about the facts but
still disagree about these questions of value. And these people usually
go on to say that one c a n n o t a r g u e about questions of value. All the
a r g u m e n t one c a n do on a moral question consists in establishing the
facts: o n c e these are established one may still differ about questions of
value. And then there is nothing one c a n do about it but agree to differ, or try to bring n o n - r a t i o n a l m e a n s of persuasion to bear on o n e
another, or. in the last resort, fight one another.
T h e r e m a y be some e l e m e n t of truth in w h a t these people sayI
shall be asking in Chapter 6 w h e t h e r there is. But I hope that it will be
agreed that it is m u c h too early to r e a c h this c o n c l u s i o n . For we do
not yet know how moral a r g u m e n t is supposed to proceed. If t h e
opponents of the position 1 have just outlined maintain that, on the
contrary, t h e r e are a r g u m e n t s w h i c h , s t a r t i n g from agreed facts as
premisses, lead to value j u d g e m e n t s as c o n c l u s i o n s , we obviously
c a n n o t decide between t h e m and their o p p o n e n t s unless we investigate the forms of a r g u m e n t by which, it is suggested, we c a n r e a c h
these c o n c l u s i o n s . And in investigating these forms of a r g u m e n t we
shall be doing moral philosophy.
2 . 6 . How does o n e investigate forms o f a r g u m e n t ? T h i s i s supposed to be the task of logic: b u t w h a t is logic? And c a n t h e r e be a
b r a n c h of logic w h i c h deals with m o r a l and o t h e r evaluative s t a t e ments? By 'evaluative s t a t e m e n t s ' or 'value judgements' I m e a n , for
t h e time being, the c l a s s o f s t a t e m e n t s w h i c h includes m o s t m o r a l
j u d g e m e n t s , or at a n y rate a c e n t r a l and typical class of t h e m , and
also o t h e r s t a t e m e n t s or j u d g e m e n t s in w h i c h words like ' o u g h t ' ,
'right'.'good', and the like occur. This is of c o u r s e an entirely v a g u e
and unsatisfactory c h a r a c t e r i z a t i o n of the class of evaluative statem e n t s : it is also not c o m p r e h e n s i v e e n o u g h . For a m o r e s e r i o u s attempt to define evaluative', see FR 2 . 8 .
I. 2 . 6
37
If we said t h a t t h e r e c a n be a logic w h i c h deals with e v a l u a t i v e

statements, we should of course be begging the question at issue between those w h o say t h a t there c a n be a r g u m e n t a b o u t questions of
value and t h o s e w h o say t h a t there c a n n o t . For if t h e r e is a kind of
logic w h i c h c a n deal w i t h t h e m , t h e r e c a n o b v i o u s l y b e a r g u m e n t
about them.
But h o w do we decide w h e t h e r there c a n be a kind of logic which
deals with a certain class of statements? Let us take a simple example.
How do I k n o w t h a t there is a kind of logic t h a t enables me to go from
the premiss ' I f p t h e n q; and p to the conclusion 'Therefore q? Is it bec a u s e o n e c a n find t h a t kind of inference in all the logic books? B u t
surely an appeal to the a u t h o r i t y of logic books is not e n o u g h . T h e y
might be wrong. I c a n n o t l a u n c h out now into a discourse on the n a ture of logical validity: but I will say very briefly w h a t I t h i n k a b o u t
this. We satisfy ourselves t h a t the modus ponens form of inference is
valid (modus ponens is t h e form of i n f e r e n c e t h a t I h a v e j u s t m e n tioned, from ' I f p then q; and // to 'Therefore q)we satisfy ourselves
that modus ponens is valid by satisfying ourselves t h a t t h a t is indeed
how we use t h e word ' i f . T h a t is to say, we satisfy o u r s e l v e s t h a t a
person w h o a d m i t t e d t h a t i f p t h e n q , a n d t h a t p , b u t denied t h a t q
would be misusing t h e word ' i f . To admit t h a t if p t h e n q is to admit
the propriety of affirming q, o n c e one has the additional information
that p . I f o n e then denies t h a t q , although o n e admits t h a t p . o n e can
reasonably be asked w h e t h e r one really meant 'If p t h e n q.
To take an even simpler e x a m p l e : suppose I say ' T h e r e is a dog in
the g a r d e n ' , b u t t h e n go on to deny t h a t t h e r e is a n y a n i m a l in t h e
garden. I c a n r e a s o n a b l y be asked 'How were you u s i n g t h e word
"dog", t h e n ? ' For 'dog' means one kind of a n i m a l . T h e validity of the
inference ' T h e r e ' s a dog in the garden, therefore there's an a n i m a l in
the garden' rests, plainly, on the m e a n i n g of the word 'dog'. In general, establishing the validity of logical inferences is establishing that
we use the words in t h e m in such a way that the c o n c l u s i o n s really do
follow from the premisses.
If, therefore, we are going to decide the question of w h e t h e r there
c a n b e a r g u m e n t s h a v i n g m o r a l j u d g e m e n t s a s their c o n c l u s i o n s o r
c o n s t i t u e n t s , we have, inescapably, to ask w h e t h e r t h e m e a n i n g s of
T1I11 E N T E R P R I S E O F M O R A L P H I L O S O P H Y
1.2.6
t h e m o r a l words like 'good' and ' o u g h t ' a r c s u c h a s t o m a k e a r g u m e n t s of this sort valid or possible. T h e study of logic leads on inevitably to t h e study of l a n g u a g e . So in my first book I w a s rash
e n o u g h to define ethics as 'the logical study of the language of morals'
(LM Preface). For this I was taken to task, b e c a u s e it was t h o u g h t t h a t
I was a b e t t i n g t h e diversion of t h e activities of m o r a l p h i l o s o p h e r s
from s u b s t a n t i a l q u e s t i o n s of m o r a l i t y to w h a t were called verbal
q u e s t i o n s . B u t I hope it will be c l e a r by n o w that, if we a r e going to
have a hope of answering t h e s u b s t a n t i a l questions with any assurance, we shall have to tackle these verbal questions. For unless we understand fulhi what we. or what the opponents in a moral argument, or in a
theoretical argument about morals, are saying, we shall never be able to decide rationally any of the questions that arise.
S o . therefore, alongside t h e factual q u e s t i o n s t h a t have to be a n swered before we c a n make any progress with a moral problem, there
h a s to be put a n o t h e r class of q u e s t i o n : q u e s t i o n s about t h e m e a n ings of words. I have given the theoretical reason for this, namely t h a t
all argument, depends on logic, and what is or is not logically valid depends on what words m e a n . B u t I could equally well have quoted e m pirical e v i d e n c e . I f o n e looks a t a l m o s t a n y m o r a l a r g u m e n t , for
example those conducted in the correspondence columns of the
newspapers, o n e c a n n o t help noticing, interspersed a m o n g t h e factual a r g u m e n t s t h a t are b r o u g h t forward, frequent i n s t a n c e s w h e r e
t h e disputants are at cross-purposes owing to ambiguities in t h e use
of words. One of them, it may be, thinks t h a t some fact which he h a s
e s t a b l i s h e d proves s o m e m o r a l c o n c l u s i o n : his o p p o n e n t does n o t
t h i n k it proves a n y t h i n g of t h e kind. T h i s m a y be a sign merely that
they were understanding words in different senses.
So in t r y i n g to solve a m o r a l p r o b l e m we h a v e to get t h e facts
straight, and we have to be c l e a r about the m e a n i n g s of the words we
are using, including the m o r a l words. Only when we have done that
will it be clear w h e t h e r there are o t h e r questions that have to be a n swered w h i c h do not fall into either of these two classes. In particular,
only t h e n will it be clear w h e t h e r there is a residual class of ultimate
questions of value which are neither questions of fact nor questions
about the m e a n i n g s of words, and on which we c a n go on disagree-
1.2.6
ing even w h e n we have agreed about t h e facts and a b o u t t h e m e a n ings of the words we are using.
So, really, investigation of the m e a n i n g s of the moral words plays a
key part in t h e study of m o r a l problems. It is only by u n d e r t a k i n g it
t h a t we s h a l l u n d e r s t a n d w h a t it is t h a t we a r e a r g u i n g a b o u t in a
m o r a l a r g u m e n t . A n d it is only by u n d e r t a k i n g it t h a t we shall find
out w h a t steps in t h e a r g u m e n t , if any, are valid. T h u s m o r a l philos o p h y t h e logical study o f t h e l a n g u a g e o f m o r a l s h a s a n indispensable p a r t to play in p r a c t i c a l m o r a l a r g u m e n t s . B u t it is also of
g r e a t i m p o r t a n c e t o e s t a b l i s h , a s only m o r a l p h i l o s o p h y c a n do,
w h e t h e r any m o r a l a r g u m e n t s a r e c o g e n t w h e t h e r , t h a t is to say,
m o r a l j u d g e m e n t s a r e t h e sort o f things o n e c a n a r g u e a b o u t a t all.
And this too c a n only be done by studying the moral words and their
logical properties.
2 . 7 . All this is so clearly true t h a t it really is surprising t h a t m a n y
writers have a t t a c k e d r e c e n t m o r a l philosophers for d i s c u s s i n g t h e
moral words, as if they o u g h t to have been discussing s o m e t h i n g else.
Certainly S o c r a t e s s t a r t e d t h e subject off by insisting on a study of
the m o r a l words, as 1 h a v e already m e n t i o n e d . Aristotle says of h i m
that h e was 'busying himself with m o r a l questions . . . and directing
the mind for t h e first time to definitions' {Met. 9 8 7 i ff.).
b
We m i g h t feel inclined to retort to those w h o a t t a c k m o r a l philosophy in this way. t h a t they dislike our studying t h e m o r a l words a n d
their m e a n i n g s b e c a u s e they do not want us to u n d e r s t a n d w h a t we
are saying w h e n we engage in moral a r g u m e n t t h a t they think that,
in moral matters, there is safety in obscurity. Undoubtedly there are a
lot of people going around in this area who positively prefer obscurity
to clarity. B u t to m a k e this a g e n e r a l a c c u s a t i o n would be unfair.
T h e r e are others w h o a t t a c k modern moral philosophy for a m o r e res p e c t a b l e r e a s o n t h o u g h n o t a n entirely c o g e n t o n e . T h e y t h i n k ,
rightly, t h a t t h e r e a r e i m p o r t a n t m o r a l q u e s t i o n s o f s u b s t a n c e t h a t
we have to answer, and t h a t moral philosophers o u g h t to be helping
us to answer t h e m . W i t h this we c a n agree. B u t t h e n they go on to say
that therefore moral philosophers ought to go straight on to t h e questions o f s u b s t a n c e , a n d n o t get side-tracked i n t o q u e s t i o n s a b o u t
m e a n i n g . T h e i r m i s t a k e i s n o t t o see t h a t t h e m o r a l p h i l o s o p h e r ' s
<>
T H E E N T E R P R I S E OF M O R A L P H I L O S O P H Y
I. 2. 7
distinctive c o n t r i b u t i o n t o t h e d i s c u s s i o n o f t h e s u b s t a n t i v e m o r a l
questions is the investigation of the words and concepts, and thus the
logic, t h a t are being employed. If t h e y ask t h e m o r a l philosopher to
leave this c o n c e p t u a l discussion and get on to t h e substantial issues,
they a r e asking him to stop being a m o r a l philosopher. But I believe
that the conceptual discussion can c o n t r i b u t e to the practical discussion, and that I have shown this in my writings on practical issues. I
shall try to placate these opponents of modern moral philosophy by
discussing the theoretical issues always in relation to their bearing on
practical questions. I hope t h a t we shall end up seeing that t h e o r y is
relevant to practice.
TAXONOMY OF ETHICAL THEORIES

l. Descriptivism
i.i Naturalism
1.2 Intuitionism
/
i.nObjectivistic
,.
naturalism
1.12 Sub......
icctivistic
' ,
..
naturalism
2. Non-descriptivism
P A R T II
THE AXEL HAGERSTROM LECTURES
2.1 Emotivism 2.2 Rationalistic

non-descriptivism
,
.
2.21 Universal
....
prescriplivism
A T A X O N O M Y OF E T H I C A L
THEORIES
^~ ..
2.22?
1. Descriptivism: Meanings of moral statements are wholly determined by

syntax and truth conditions.
1.1 Naturalism: 'truth conditions of moral statements are non-moral properties.
1.11 Objeetivislic naturalism:'these properties are objective.
1.12 Subjeetivistic naturalism: These properties are subjective.
1.2 Intuitionism: Truth conditions of moral statements are sui generis moral
properties
2. Non-descriptivism: Meanings of moral statements are not wholly determined by syntax and truth conditions.
2.1 Emotivism: Moral statements are not governed by logic.
2.2 Rationalistic non-descriptivism: Moral statements are governed by logic.
2.21 Universal prescriplivism: The logic which governs moral statements is
the logic of universal prescriptions.
2.22 ?
TAXONOMY
3.1.
I MUST s t a r t by s a y i n g h o w happy I a m t o be a d d r e s s i n g a
Swedish a u d i e n c e again on a topic in ethical theory. It is never difficult in Sweden, as it h a s b e c o m e in m a n y parts of t h e world, to find
s e r i o u s p h i l o s o p h e r s w h o a r e able t o discuss t h e s e q u e s t i o n s w i t h
clarity and rigour. I am also p a r t i c u l a r l y delighted to be giving l e c tures dedicated to t h e m e m o r y of Axel H a g e r s t r o m . He c o u l d justly
be called t h e pioneer, in r e c e n t times, of e t h i c a l non-descriptivism,
t h o u g h in fact views of this sort have a long history (see H 1 9 9 8 ) . He
REQUIREMENTS FOR AN ADEQUATE ETHICAL

T H E O R Y (see pp. 11 _ H'.l
Objectivistic
naturalism
1. Neutrality
2. Practicality
i,. Incompatibility
4. Logicality
5 . Arguability
(S. Conciliation
X
X
-/
J
X
X
Subjeetivistic
naturalism
J
X
X
J
X
h a s been in this c e n t u r y . And I am delighted also to be giving t h e s e

lectures in the h o m e town of Linnaeus, the pioneer of scientific t a x onomy.
Intuitionism Emotivism
V
x
J
J
thus made t h e most important b r e a k t h r o u g h in t h e subject t h a t there
J
J
J
X
X X
X X
I must begin, t h o u g h , by explaining t h e title of these lectures. This

involves saying w h a t I m e a n by 'Ethical T h e o r y ' , a n d w h a t I m e a n by
' T a x o n o m y ' . T h e first is t h e m o r e difficult task, b e c a u s e t h e expression 'Ethical T h e o r y ' h a s been used, and abused, in so m a n y different
ways. I am going to use it a good deal m o r e narrowly t h a n m a n y write r s o t h e r w i s e it would b e c o m e a s u b j e c t t h a t c o u l d n o t be covered
in five lectures. I m e a n by it t h e study of the m o r a l concepts, that is, of
A T A X O N O M Y OF E T H I C A L T H E O R I E S
44
II. 3- i
II. 5. 1
TAXONOMY
45
o u r use of the moral wordsif you like, of their m e a n i n g in a broad
is a genuine dispute at all, turns fairly rapidly into a dispute w h i c h is
sense, or of what we are doing w h e n we ask moral questions. S i n c e ,
not ontological but conceptual, and that there is no way of clearly
as I have argued ( r . i f.), an important part, at least, of the m e a n i n g of
f o r m u l a t i n g t h i s supposed dispute a b o u t w h e t h e r t h e r e really a r e
all words, including moral words, is determined by their logical prop-
moral facts or m o r a l properties in rerum natura without translating it
erties, this study of m e a n i n g s leads inescapably to the study of those
into a dispute a b o u t h o w m o r a l words get their m e a n i n g . So we are
logical properties. And that is why t h e s u b j e c t h a s practical impor-
only w a s t i n g o u r t i m e ( a s H u m e m i g h t h a v e put it. ' a m u s i n g o u r -
t a n c e . For o n e o f t h e c h i e f t h i n g s t h a t i s d e m a n d e d o f t h e m o r a l
selves') if we argue about w h e t h e r moral facts exist without first rais-
philosopher is that he (or she) should do something to help us discuss
ing the conceptual issues on which any solution to t h a t problem h a s
moral questions rationally; and this requires obedience to the logical
to depend. Even if we were to talk about real moral properties in rerum
rules governing the c o n c e p t s . Unless we follow these rules, we shall
natura (and I c a n n o t forbid people to talk in this way if they w a n t to)
never be able to a r g u e rationally a b o u t m o r a l questions. T h e prime
we shall be only thereby affirming our subscription to the m o r a l state-
task of philosophy, since S o c r a t e s started the business, is the study of
ments or principles that we accept. T h e question r e m a i n s of what we
a r g u m e n t s : a n d the prime task of m o r a l philosophy is t h e study of
are doing when we so subscribe to them.
moral arguments, to learn how to tell good from bad ones. In this task
If we ask what we are doing, we shall have to do s o m e c o n c e p t u a l
ethical theory, which reveals the logic of the moral concepts, is an es-
analysis, a n d t h e result of it is likely to be t h a t all forms of descrip-
sential tool.
tivism fail to give an a d e q u a t e a c c o u n t of t h e m a t t e r ; t h e r e is an
It may help confine the subject within bounds if I go on to say w h a t
e s s e n t i a l prescriptive e l e m e n t i n t h e m e a n i n g o f m o r a l s t a t e m e n t s
ethical theory, as I am using the term, is not. Many writers now use
which goes beyond their descriptive m e a n i n g (1.7, and see Chapters
the expression 'moral theory'. I am never sure quite what they m e a n
4 ff.). If we w a n t to be realists about the prescriptive element, we c a n
by it; it s e e m s to cover a vast a r e a of i n d e t e r m i n a t e size, but at least
if we wish speak of real prescriptive properties in a c t i o n s ; b u t t h a t is
includes the views of the writers on a lot of substantial m o r a l q u e s -
simply not illuminating.
tions, systematized often into a n u m b e r of moral principles, such as
I shall be using the expression 'ethical theory', then, in the n a r r o w
R a w l s ' s ' P r i n c i p l e s of J u s t i c e ' . T h u s a m o r a l t h e o r y c a n n o t be. as I
sense of 'theory about the m e a n i n g and logical properties of the
hope e t h i c a l t h e o r y will be in my h a n d s , a purely formal discipline
m o r a l words'. I h a v e a l r e a d y said why I t h i n k t h a t it is a n e c e s s a r y
dealing only in logical and c o n c e p t u a l studies. Kant was very insis-
study if we are to distinguish good from bad a r g u m e n t s a b o u t moral
tent on this distinction between formal a n d substantial theses ( 8 . 5 ) . I
q u e s t i o n s . It is surprising, therefore, how m a n y m o r a l philosophers
am not for a moment denying the i m p o r t a n c e of using rational argu-
try to p e r s u a d e us t h a t we do not need to study e t h i c a l t h e o r y (e.g.
ment to decide on substantial moral p r i n c i p l e s . T h a t is the a m b i t i o n
Rawls 1971: 5 1 ) . One r e a s o n why people say this m a y be t h e follow-
of all s e r i o u s m o r a l p h i l o s o p h e r s . B u t t h e r e is a prior task: t h a t of
ing. T h e y have e x a m i n e d various ethical theories t h a t have been put
finding the rules governing the a r g u m e n t . W i t h o u t those rules, a n y -
forward, a n d have (often after insufficient study) decided t h a t t h e s e
thing goes.
will not do. T h e y have therefore concluded too hastily t h a t no ethical
I shall not in these l e c t u r e s be doing m o r a l t h e o r y in t h a t wide
theory is adequate. One of t h e things I shall be doing in these lectures
sense, t h o u g h I have in other places devoted a lot of attention to prac-
is to take the various possible ethical theories in turn a n d say w h a t is
tical moral issues. Nor shall I be doing a n y t h i n g that could be called
w r o n g with e a c h of t h e m . B u t I shall also go on, unlike t h e writers I
'ontology'. I have argued elsewhere (II 19850) that an ontological dis-
am speaking of, to say w h a t is right about e a c h of t h e m . T h e y all re-
pute like the supposed dispute between realists and anti-realists, if it
veal different parts of t h e truth about morality. Instead of jumping to
II. 3 . 1
II. 3. 2
TAXONOMY
47
t h e conclusion, b e c a u s e all ethical theories you know of have faults,
division that we m a k e in the course of our taxonomy, we c a n ask, not
t h a t t h e r e is no a d e q u a t e e t h i c a l theory, so you had better give up
merely what species of ethical theory there have been, but what
looking for o n e . t h e moral philosopher w h o is less of a defeatist will
theories there could be. So, instead of going to the A m a z o n j u n g l e to
go on to try to find a t h e o r y w h i c h preserves t h e t r u t h s in e a c h of
look for new species, we c a n think t h e m out for ourselves. T h a t is the
these theories but avoids their errors. And t h a t is w h a t I shall be doing
way the subject progresses.
in these lectures. If as a result I get branded as an eclectic, so be it (H

1994M.
So w h a t I shall be a t t e m p t i n g in these lectures will be very a m b i tious (perhaps dangerously so). I shall try to show, in t h e c a s e of e a c h
An important part of the s e a r c h will be an attempt to make a list of
division t h a t I m a k e in t h e taxonomy, t h a t t h e division is exhaustive.
the requirements that an adequate ethical theory has to satisfy. T h e n
T h e easiest way of doing this, w h i c h for the m o s t p a r t I s h a l l be fol-
we can look at e a c h theory in turn and see which of the requirements
lowing, is by m a k i n g e a c h division a dichotomythat is, a division
it satisfies, and which it fails to satisfy. T h u s we may be able to c o r r e c t
into just two classes w h i c h between t h e m e x h a u s t t h e genus. T h i s c a n
and improve them, and end up with a theory thai satisfies all the re-
be achieved by m a k i n g t h e differentia of e a c h species t h e n e g a t i o n of
quirements.
the differentia of the other. I shall be giving an illustration of this in a
3 . 2 . S o m u c h , t h e n , for t h e e x p r e s s i o n ' E t h i c a l T h e o r y ' . W h a t
m o m e n t , w h e n 1 m a k e t h e m a i n division of ethical theories into t h e
about ' T a x o n o m y ? This is a good deal easier, because I shall be using
two genera, descriptivism and non-descriptivism. If it were possible to
the term in m u c h the s a m e way as the b o t a n i s t s . I was interested to
realize this ambition and m a k e the t a x o n o m y exhaustive, t h e n at the
see that Hagerstrom himself published a dialogue called The Botanist
end of t h e day we s h o u l d have a c o m p l e t e classification of possible
and the Philosopher: On the Necessity of Epistemoloyy; but since it ap-
ethical theories, with a d e m o n s t r a t i o n t h a t these were the only pos-
peared only in Swedish I have b e e n u n a b l e to read it to see w h e t h e r
sible ones. If it t u r n e d out t h a t all the possible ones were inadequate,
his view of the relations between the two disciplines was the s a m e as
then we really should have to give up all hope of finding an adequate
mine. W h e n your great naturalist Linnaeus set out to classify plants,
one. B u t I am m o r e optimistic.
he followed Aristotle in classifying them per uenus et differentiam. B u t
T h e main division of ethical theories t h a t 1 shall m a k e is into two
since his classification had m a n y m o r e t h a n the two levels of genus
genera, which I shall call 'descriptivism' and 'non-descriptivism'. Our
and species. L i n n a e u s introduced o t h e r words for the i n t e r m e d i a t e
first task, therefore, is to give t h e differentia b e t w e e n t h e s e . On this
levels: for example, 'family'. I do not think that Aristotle would have
there h a s been m u c h confusion. Terms like 'realist' and 'anti-realist',
quarrelled with this, for he c e r t a i n l y did not want to classify only at
'cognitivist' and 'non-cognitivist'. and others have been widely used,
two levels. T h e term ' s p e c i e s ' is still in use: e a c h species is distin-
as if they all m a r k e d t h e s a m e distinction. In a paper I have already
guished within a genus or subordinate class by stating the difference
referred to I h a v e a r g u e d that t h e pair of t e r m s 'descriptivist' a n d
t h a t m a r k s it out from t h e other species.

However. I shall perhaps be closer to Aristotle than to Linnaeus in
'non-descriptivist' is t h e clearest way of m a k i n g t h e distinction, a n d

t h a t the others, as soon as they are clarified, collapse into it. B u t t h e
o n e respect. L i n n a e u s was doing an empirical study: he had to take
position is worse t h a n that, because, when those w h o e n g a g e in these
the various plants as he found t h e m , and put them into a c l e a r and
disputes try to give t h e differentia between their positions, t h e y c o m -
consistent classification. But in philosophy we can do more than this.
m o n l y c h o o s e a m i s l e a d i n g o n e . n a m e l y w h e t h e r , a c c o r d i n g to a
B e c a u s e t h e e n q u i r y is a formal o n e , it is l e g i t i m a t e to ask. nol j u s t
given theory, m o r a l statements c a n be true or false. I shall be arguing
w h a t t h e o r i e s we c a n find in t h e world, but w h a t theories we could
in 2.(1 that no relevant dispute is marked out by this question, nor by
find. This question ought to be answerable a priori. In the c a s e of e a c h
the question of whether we c a n know them to be true, nor by the
II- 3- 2
H. 3- 3
TAXONOMY
49
q u e s t i o n of w h e t h e r m o r a l facts or m o r a l properties exist in t h e
m e a n i n g wholly from their truth-conditions. T h e a n s w e r seems to be
world. T h i s is because there are perfectly good senses in which a non-
t h a t they do not. It h a s been c o m m o n , w h e n discussing m e a n i n g , to
descriptivist like me c a n allow t h a t m o r a l s t a t e m e n t s c a n be t r u e or
distinguish b e t w e e n s e m a n t i c s and s y n t a c t i c s . I will leave t h e third
false, that we can know s o m e of t h e m to be true, and that there a r e
m e m b e r o r supposed m e m b e r o f this triad, ' p r a g m a t i c s ' , till m u c h
moral properties (II 1976/), 19X50. 1995/)). If 1 demur to the claim that
later ( 6 . 5 ) . T h e distinction has been made in various and often incon-
t h e r e a r e m o r a l facts in t h e world, it is b e c a u s e I do not like s a y i n g
sistent ways; b u t I s h a l l use ' s e m a n t i c s ' n o t , as s o m e do, widely to
that there are uny facts in the world. T h e world consists of things, not
cover anything to do with meaning, but narrowly to include only t h a t
facts. Hut that is a n o t h e r story (H 1985).
part of t h e m e a n i n g of s e n t e n c e s which is determined directly or in-
3 . 3 . First, let me give you my way of distinguishing descriptivism
directly b y t r u t h c o n d i t i o n s . T h i s leaves, a s a n o t h e r c o n s t i t u e n t o f
from non-descriptivism: and t h e n I will tell you why I t h i n k o t h e r
m e a n i n g , t h e s y n t a c t i c a l properties of s e n t e n c e s . For e x a m p l e , if a
ways lead to confusions. My differentia relies on the notion of t r u t h
statement is of t h e subject-predicate form, t h a t partly determines its
conditionsbut not in the simple way that some people might think.
meaning; and we can know this before we k n o w w h a t its truth condi-
It is c o m m o n l y thought that m e a n i n g depends in some way on truth
tions are.
conditions (H 19910, 1993,0. 1995b). This was the basis of the old veri-
Not all g r a m m a t i c a l distinctions are relevant to m e a n i n g . For ex-
fication theory of m e a n i n g which m a n y logical positivists embraced,
ample, a s w e h a v e s e e n , t h e distinction b e t w e e n s t r o n g a n d w e a k
but which is now in disrepute in its early simple form which claimed
verbs is not ( 1 . 3 ) . If I say 'the sun shined' instead of 'the sun s h o n e ' , I
that ' t h e m e a n i n g of a s e n t e n c e is t h e m e t h o d of its verification'.
speak u n g r a m m a t i c a l l y but my m e a n i n g is still clear, a n d t h e s a m e .
However, it is even now c o m m o n to claim that truth conditions have
But s o m e a r e r e l e v a n t . T h e clearest e x a m p l e i s t h e d i s t i n c t i o n b e -
a part to play in d e t e r m i n i n g m e a n i n g : and I a g r e e with t h i s .
tween t h e indicative and imperative verb-forms ( L . 3 , H 1 9 9 6 b ) . T h e
' M e a n i n g ' has here to be understood as including both sense and ref-
transformation w h i c h c h a n g e s the Latin 'ibis', m e a n i n g 'You will (or
e r e n c e . T h i s was how Austin used it ( 1 9 6 2 : 1 0 0 ) . To be a c c u r a t e . I
a r e going t o ) g o ' , i n t o t h e imperative ' f , m e a n i n g ' G o ' , a l t e r s t h e
must also explain that I am thinking here of the m e a n i n g of a token
m e a n i n g . S o m e t i m e s the g r a m m a t i c a l or syntactical properties affect
s e n t e n c e as used by a particular speaker on a particular occasion. We
the logical properties. To lake the s a m e example, there is a valid infer-
might make a first approximation to explaining what descriptivism is
e n c e from t h e future indicative 'You will go' to t h e future indicative
by s a y i n g that it is the view that m e a n i n g is wholly d e t e r m i n e d by
'You will not stay here', but not from the imperative 'Go' to the indic-
truth conditions. If this is held to be t r u e of the m e a n i n g of all sen-
ative You will n o t stay here'. S o m e c o m m a n d s are n o t obeyed.
tences, then that is descriptivism tout court. T h e r e have perhaps been
Confining ourselves, therefore, to syntactical or g r a m m a t i c a l prop-
people who have thought this, the victims of what Austin called 'the
erties which do affect m e a n i n g , we c a n say that they are a part of the
descriptive fallacy' ( 1 9 6 1 : 2 3 4 : 1962: 3 ) . I shall not say anything about
meaning-determining properties of sentences which are independent
this very sweeping view. Ethical descriptivism, as a first approxima-
of a n y p a r t i c u l a r t r u t h c o n d i t i o n s . So it is n o t t r u e , on a n y at all
tion, is the view that the m e a n i n g of a moral statement is wholly de-
plausible theory, t h a t all m e a n i n g is determined by t r u t h conditions.
t e r m i n e d by its t r u t h - c o n d i t i o n s , that is. by t h e c o n d i t i o n s u n d e r
So our proposed differentia, which said t h a t according to descriptivism
which it would correctly be said to be true.
t h e m e a n i n g of m o r a l s e n t e n c e s is entirely d e t e r m i n e d by t h e t r u t h
On this view, moral statements get their m e a n i n g in just the s a m e
conditions of s t a t e m e n t s expressed by them, h a s to be refined.
way as ordinary factual statements. B u t we have to ask whether even
T h e position is r a t h e r this. T h e s y n t a c t i c a l or formal properties of
in the case of ordinary factual statements it is true that they get their
a s e n t e n c e (those of t h e m t h a t affect m e a n i n g ) determine w h a t kind
II. 3. 3
H. 3- 3
TAXONOMY
5i
of s e n t e n c e it is. T h e y do this by d e t e r m i n i n g its i n t e r n a l s t r u c t u r e .

For example, they may make it into a subject-predicate sentence, apt
for a s c r i b i n g a property to an o b j e c t . What property to what o b j e c t ,
t h e y do not d e t e r m i n e . That is t h e role of t h e s e n t e n c e ' s s e m a n t i c s ,
not of its syntactics.
T h e r e are plenty of o t h e r examples in the literature, but n o n e of t h e m
Truth conditions belong to semantics. Thai a statement has to
sign s a y i n g ' Y o u h a v e b e e n w a r n e d ' . Now t h e y a r e s o m e t i m e s fol-
have truth conditions is determined when it h a s been specified that it
lowed by a sign s a y i n g ' B e w a r n e d ' . T h e r e m u s t be t w o different
is a statement. S t a t e m e n t s are speech acts which can be true or false.
s e n s e s o f ' w a r n ' h e r e , b e c a u s e o n e c a n hardly b e i n s t r u c t e d t o b e
If a putative statement has no truth conditions, it is no genuine state-
w a r n e d if o n e h a s already been warned. In o n e sense ' w a r n ' m e a n s
has convinced me t h a t illocutionary force is n o t part of m e a n i n g .

T w a r n y o u ' is s o m e t i m e s used as a supposed e x a m p l e of t h e impossibility of distinguishing illocutionary from p e r l o c u t i o n a r y a c t s .
But this too is a m b i g u o u s . Road hazard signs used to be followed by a
m e n t . T h i s does not stop it being a meaningful speech act; for t h e r e
'address a w a r n i n g t o ' . B u t in a n o t h e r s e n s e it m e a n s 'put on o n e ' s
a r e m a n y kinds of m e a n i n g f u l s p e e c h act w h i c h do not h a v e t r u t h
guard by m e a n s of a w a r n i n g ' . ' B e w a r n e d ' uses t h e latter sense, in
c o n d i t i o n s , b e c a u s e t h e y just c a n n o t b e t r u e o r false. I m p e r a t i v e
w h i c h t h e speech a c t is n o t successful unless t h e p e r l o c u t i o n a r y a c t
speech acts (or imperafions as we have called them) are an example.
h a s been effective: but 'You have been warned' simply reports t h e per-
Austin and his disciples have distinguished between m e a n i n g and

illocutionary force. William Alston, by c o n t r a s t , has included in t h e
meaningful e l e m e n t s in s e n t e n c e s w h a t he calls 'illocutionary force
f o r m a n c e of an i l l o c u t i o n a r y act, w h e t h e r or n o t t h e addressee h a s
actually been put on his guard.
3.4.
I f the syntactical o r grammatical properties o f sentences
indicating devices' ( 1 9 6 4 : 37 ff.); see also Searle 1969: 62 and S e a r l e
include their i l l o c u t i o n a r y force indicating devices (of w h i c h m o o d -
and Vanderveken 1985: 7.1 have called these in 1.3. H 19890 tropics. I
signs are an e x a m p l e ) , then we c a n restate our differentia in a clearer
think Alston and Searle are right to say that there is a wide sense of
way. A descriptivist is s o m e o n e w h o thinks, n o t merely t h a t a m o r a l
' m e a n i n g ' in which these contribute to the m e a n i n g of sentences. An
s t a t e m e n t h a s t r u t h c o n d i t i o n s (for non-descriptivists c a n a g r e e t o
example is the feature which distinguishes imperative from indicative
this, as we shall see); n o r that a moral statement's m e a n i n g is wholly
s e n t e n c e s , such as exists in m o s t l a n g u a g e s . I am inclined to d o u b t
d e t e r m i n e d by its t r u t h c o n d i t i o n s (for, as we h a v e s e e n , this is n o t
w h e t h e r Austin himself would have dissented from this (see II 1971c:
true of any s e n t e n c e s ) ; n o r that the syntactical or g r a m m a t i c a l prop-
100 If.): but some of his disciples seem to.
erties o f s e n t e n c e s e x p r e s s i n g m o r a l s t a t e m e n t s m a k e t h e i r i l l o c u -
If c a n readily be admitted t h a t s e n t e n c e s of t h e s a m e form a n d
t i o n a r y forces such t h a t t h e y have to have truth c o n d i t i o n s , a n d are
content can sometimes be used to perform different speech acts with
therefore s t a t e m e n t s i n t h e s e n s e just used (this t o o t h e n o n -
different illocutionary forces. For example 'You will go' could express
descriptivist can agree to); but further, that these truth conditions are
a prediction, but it could (at least in t h e British Army) express a c o m -
all t h a t is needed in addition to d e t e r m i n e t h e m e a n i n g of t h e sen-
m a n d . B u t this m a y be like a n y o t h e r ambiguity. Just as 'I will m e e t
tences. A non-descriptivist, then, will be s o m e o n e w h o denies this last
you at t h e b a n k ' could be r e f e r r i n g to t h e river b a n k or t h e p l a c e
clause; he thinks that moral statements, although they may have
where you get money, so. equally, the word 'will' could be the sign of
t r u t h c o n d i t i o n s , do n o t depend for t h e i r m e a n i n g wholly on t h o s e
a prediction, or of a promise (two different kinds of speech a c t ) . All
truth c o n d i t i o n s , n o r even wholly on t h e i r s y n t a c t i c a l features plus
we need say is t h a t somebody w h o took t h e sentence in a way differ-
their truth conditions, b e c a u s e their s y n t a c t i c a l features allow t h e m
ent from that in which the speaker intended it would have misunder-
t o b e used with t h e s a m e m e a n i n g , a l t h o u g h t h e t r u t h c o n d i t i o n s
stood his meaning: the speaker was intending to perform one kind of
may vary ( 7 . 3 , H 1 9 9 3 9 , 1 9 9 5 b ) -
speech act. but the hearer took h i m to be performing a different kind.
T h i s is a difficult idea to grasp, so I m u s t try to e x p l a i n it m o r e
A T A X O N O M Y 0 1 - E T H I C A L THI-ORIKS
H. 3- 4
II. 3. 4
TAXONOMY
53
simply. I c a n do so by using a t e r m first introduced, so far as I know,
have made this s t a t e m e n t b e c a u s e the person h a s c e r t a i n descriptive
b y S t e v e n s o n ( 1 9 4 5 : 6 2 ) . H e d i s t i n g u i s h e d b e t w e e n the descriptive
qualities; they were, for me, the truth conditions of the s t a t e m e n t that
m e a n i n g of moral statements and their emotive meaning. I shall later
I made. T h a t is, if the person had not had these features, I would not
be discarding the idea of emotive m e a n i n g ; I shall substitute for it the
have made the statement, and if she had them, my existing moral
term 'evaluative m e a n i n g ' (see LM c h . 7). S o m e t i m e s I shall say 'pre-
standards required me to m a k e it. So, according to my existing stand-
scriptive m e a n i n g ' , but the difference between these two expressions
ards, having the features was both a n e c e s s a r y a n d a sufficient condi-
need not n o w c o n c e r n us. T h i s will e n a b l e me to leave b e h i n d t h e
tion for m a k i n g the s t a t e m e n t . T h e features m i g h t be partly positive
' p r a g m a t i c s ' which play such a large part in Stevenson's theory, and
and partly negative: they might have included, for example, t h a t s h e
which I think are flawed ( 1 . 5 . 6 . 5 , H 1996b). So I shall distinguish be-
is kind a n d g e n e r o u s , a n d does n o t c h e a t at c a r d s . If s h e c h e a t e d , I
tween t h e descriptive a n d t h e e v a l u a t i v e m e a n i n g o f m o r a l s t a t e -
would not call her a good person, and if she were n o t kind a n d gener-
m e n t s . T h e descriptive m e a n i n g is really the s a m e thing as the truth
ous I would not either. And of course we have to add to these qualities
c o n d i t i o n s , plus t h e r e q u i r e m e n t , laid on a m o r a l s t a t e m e n t by its
all t h e o t h e r positive a n d negative qualities s h e h a s to h a v e or lack,
having the illocutionary force of a statement, that it has to have truth
and these m a y include disjunctions of alternative qualities.
conditions in order to have m e a n i n g (H 1993.0). T h e descriptive m e a n -
B u t n o w suppose t h a t m y s t a n d a r d s c h a n g e . P e r h a p s I h a v e b e -
ing is also the s a m e thing as the s e m a n t i c s of the statement. It deter-
c o m e more hard-bitten, and now think it is all right to c h e a t at cards
mines to what the descriptive terms in the statement c a n correctly be
a n d t h a t k i n d n e s s a n d g e n e r o s i t y are a sign of w e a k - m i n d e d n e s s . I
applied, and to what objects the referring expressions used in it must
shall now say t h a t s h e is not a good person just b e c a u s e of t h e very
be taken as referring. T h u s the descriptive m e a n i n g does uniquely de-
s a m e properties t h a t m a d e me call her a good person before. So am I
termine the truth conditions of t h e statement.
still using t h e phrase 'good person' with the s a m e m e a n i n g as before,
B u t a n d here is t h e i m p o r t a n t point for o u r differentia b e t w e e n
or am I not? I w a n t to say t h a t in o n e sense I a m , a n d in a n o t h e r I am
descriptivists and n o n - d e s c r i p t i v i s t s b o t h t h e truth conditions and
not. 1 am still using it with the s a m e evaluative m e a n i n g : to call some-
the descriptive m e a n i n g of a moral statement can vary, without t h e
o n e a good person is still to c o m m e n d h e r (or h i m ) . It follows t h a t I
m e a n i n g of the statement varying totally. This is because the evaluat-
have changed my mind. W h a t I am now saying contradicts w h a t I was
ive m e a n i n g , the o t h e r c o n s t i t u e n t in t h e m e a n i n g , can r e m a i n t h e
saying before. It is therefore impossible for a n y b o d y c o n s i s t e n t l y to
s a m e . In o t h e r words, t h e c r u c i a l differentia between a descriplivist
a g r e e both with w h a t I am saying n o w a n d with w h a t I w a s saying
and a non-descriplivist is this: the descriptivisl thinks that if the truth
before. To say t h a t both are right would be to c o m m i t a logical error.
c o n d i t i o n s of a m o r a l s t a t e m e n t h a v e c h a n g e d , its m e a n i n g as a
T h i s would n o t be so if t h e m e a n i n g of my words h a d e n t i r e l y
whole must have c h a n g e d : but a non-descriplivist holds that this is
changed; for then w h a t I am now saying would not be the negation of
not so. He thinks it possible for a m o r a l s t a t e m e n t to retain t h e s a m e
w h a t I t h e n said. B u t I am using the words n o w with a different de-
evaluative m e a n i n g , while c h a n g i n g its descriptive m e a n i n g , and its
scriptive m e a n i n g t h a t is, in a c c o r d a n c e w i t h different standards,
truth conditions and its s e m a n t i c s . T h i s is b e c a u s e there is an e x t r a
or different t r u t h c o n d i t i o n s . Kxamples like t h i s s h o w q u i t e clearly
bit of input that goes into the m a k i n g of a moral statement which is
that t h e r e a r e these two e l e m e n t s in t h e m e a n i n g s of evaluative e x -
not present in t h e m a k i n g of an o r d i n a r y purely descriptive s t a t e -
pressions like 'good'. Only philosophers with axes to grind deny this.
ment.
It will be n o t i c e d t h a t in giving t h e descriptive m e a n i n g or t r u t h
An e x a m p l e m a y m a k e this clearer. S u p p o s e t h a t I have called a
conditions of t h e expression 'good person' in my e x a m p l e I used t h e
w o m a n a good person, t h u s m a k i n g a m o r a l s t a t e m e n t a b o u t her. I
words ' k i n d ' a n d ' g e n e r o u s ' a n d ' c h e a t ' . I n c a s e a n y o n e w a n t s t o
54
II. 3- 4
II-3-5
TAXONOMY
object that these too are evaluative expressions, so that the s t a t e m e n t
from t h e i r s y n t a c t i c a l features ( w h i c h m a y d e t e r m i n e t h e i r i l l o c u -
of the truth conditions is itself not wholly descriptive, I must say t h a t
t i o n a r y force a m o n g o t h e r things), the only additional d e t e r m i n a n t
these words too, which belong to a c l a s s of what I shall call ' s e c o n -
of m e a n i n g for m o r a l s t a t e m e n t s is t h e i r t r u t h c o n d i t i o n s . T h i s is
darily evaluative words', and w h i c h o t h e r s have called 'thick e t h i c a l
w h a t t h e n o n - d e s c r i p t i v i s t denies. H e t h i n k s , o n t h e c o n t r a r y , t h a t
c o n c e p t s ' , c a n be treated in a similar way to 'good', except t h a t their
there is a n o t h e r element in the meaning of these statements, the
e v a l u a t i v e m e a n i n g is s e c o n d a r y to t h e i r descriptive: b u t t h a t will
evaluative or prescriptive, which c a n remain t h e s a m e a l t h o u g h the
have to be left till later ( 3 . 8 . H 1996(f).
truth c o n d i t i o n s c h a n g e , and which m a k e it t h e s a m e s t a t e m e n t , in
The important point 1 wish to m a k e now is that, a l t h o u g h
the sense that it still makes the s a m e evaluation of the s a m e act, per-
evaluative (including m o r a l ) s t a t e m e n t s do indeed have truth condi-
son, etc.. a l t h o u g h its truth conditions have c h a n g e d . T h i s is a thing
tions, these c a n c h a n g e without the entire m e a n i n g of the s e n t e n c e s
t h a t c o u l d n e v e r h a p p e n with o r d i n a r y descriptive o r f a c t u a l s t a t e -
1.5.
which express them c h a n g i n g (11 1993., 1 9 9 6 c ) . This has crucial c o n -
ments. In their case, if the truth conditions c h a n g e , it is altogether a
s e q u e n c e s for ethical theory. II we c h a n g e t h e truth conditions of a
different s t a t e m e n t . W i t h m o r a l s t a t e m e n t s , by c o n t r a s t , ' S h e is a
moral statement, we c h a n g e its descriptive m e a n i n g . But if the evalu-
good person' c a n be affirmed by o n e speaker but denied by a n o t h e r
ative m e a n i n g remains the same, we have, in making this c h a n g e , al-
b e c a u s e they use different standards a n d different t r u t h c o n d i t i o n s ,
tered our moral standards. We are appealing to different reasons, for
and yet be. in respect of its evaluative m e a n i n g , t h e s a m e s t a t e m e n t .
example, for calling an a c t wrong, but we are calling it wrong in the
This too h a s i m p o r t a n t c o n s e q u e n c e s , as we shall see when I c o m e to
s a m e sense, evaluatively speaking. We are still condemning it by call-
discuss subjectivism ( 5 . 5 ) ; it m e a n s that these two speakers really are
ing it wrong.
contradicting o n e another, which would not be t h e c a s e on a subject-
This m e a n s that a statement of the truth conditions of moral
ivist theory w h i c h held t h a t they were just m a k i n g s t a t e m e n t s about
statements, w h i c h may signal a c h a n g e in moral standards, is not it-
their own respective attitudes of approval and disapproval, and would
self morally neutral. So there c a n be no question of there being a first
also n o t b e t h e c a s e , a s w e s h a l l see, o n a n o b j e c t i v i s t i c n a t u r a l i s t
stage in t h e c o n s t r u c t i o n of an e t h i c a l t h e o r y in which we give a
t h e o r y w h i c h held t h a t different m o r a l s t a n d a r d s e n t a i l different
morally neutral general formulation of the truth conditions of moral
m e a n i n g s for the moral words ( 4 . 3 ) .
statements, and then a second stage in which we use this general for-
It is c l e a r from these explanations (which I fear have been compli-
mulation to determine which moral statements in particular are true.
c a t e d a n d h a r d to g r a s p w h i c h is why so few people g r a s p t h e m )
In the general formulation, we will already have sold the pass by m a k -
that my main division between descriptivist and non-descriptivist eth-
ing s o m e s u b s t a n t i a l m o r a l c l a i m s w h i c h is w h a t you a r e a l w a y s
ical t h e o r i e s is an e x h a u s t i v e division. T h e first kind of theory, as I
doing when you are giving the truth conditions of moral statements.
have j u s t said, affirms w h a t t h e s e c o n d denies, n a m e l y , t h a t a p a r t
In o t h e r and simpler words, it is no use thinking that the standards by
from t h e i r s y n t a c t i c a l features, t h e only a d d i t i o n a l d e t e r m i n a n t o f
w h i c h we assess the truth of m o r a l s t a t e m e n t s are morally n e u t r a l .
m e a n i n g for moral s t a t e m e n t s is their truth conditions. I hope t h a t by
T h e y a r e t h e very s a m e s t a n d a r d s a s t h o s e b y w h i c h w e m a k e t h e
differentiating the genera in this dichotomous way I have m a d e it the
moral statements themselves, a n d so incorporate a substantial moral
c a s e that there c a n n o t be a t h e o r y which falls into n e i t h e r of the two
s t a n c e . I n o u r e x a m p l e , i f you say t h a t c h e a t i n g a t c a r d s does n o t
genera. So if. as I hope to do, I c a n s h o w t h a t descriptivism, in all its
m a k e a person a bad p e r s o n , you a r e m a k i n g a s u b s t a n t i a l m o r a l
different forms, is i n a d e q u a t e , I s h a l l t h e r e b y h a v e s h o w n t h a t o n e
claim.
1 said earlier that a descriptivist is s o m e o n e who thinks that, apart
has to be s o m e kind of non-descriptivist. After rejecting an u n t e n a b l e

kind, I shall advocate a kind which I think is m o r e tenable; but I shall
II.
II. .V 6
TAXONOMY
57
leave it open w h e t h e r there may also be o t h e r kinds that are tenable
c o u r t . T h e two disagree about t h e standards to be used in assessing
too.
the goodness of people. And both may say t h a t they k n o w t h a t their
3 . 6 . I hope that w h a t 1 have said h a s s h o w n how little g r a s p of
respective standards are the right ones. And about the word 'right' as
these issues those people have w h o think (as many beginner students
used in t h e s e c l a i m s , all t h e s a m e t h i n g s c a n be said as I h a v e said
are taught to think) that it is sufficient to distinguish between w h a t
about the word 'true' (H 197b/). 199K?). So nothing is gained by intro-
they call cognitivisl and n o n - c o g n i t i v i s t e t h i c a l theories by s a y i n g
ducing t h e word ' k n o w ' into this discussion: a n d it is misleading be-
that t h e y give opposing a n s w e r s to t h e q u e s t i o n 'Can m o r a l s t a t e -
c a u s e it suggests that what is known c a n n o t be disputed: but it will be
m e n t s be true or false?' T h e answer to this question is that they c a n .
disputed.
but that the important issue between descriplivisls and non-descriptivists is not settled thereby (H 1995/). 1 9 9 M .
T h e i m p o r t a n t q u e s t i o n . I said, w a s w h e t h e r t h e r e a r e good a n d
bad w a y s of r e a s o n i n g a b o u t all t h e s e m a t t e r s : a b o u t w h e t h e r t h e
The terms cognitivisl' a n d ' n o n - c o g n i t i v i s t ' are misleading for a
standards and truth conditions that are being used are t h e ones t h a t
further reason. T h e etymology of these words seems to imply that a c -
o u g h t to be being used, and so about which of t h e s t a t e m e n t s m a d e
cording to cognitivists o n e can know t h a t a moral statement is true,
by o u r two opposing parties we ought, at t h e end of t h e day. to call
but a c c o r d i n g to non-cognitivists o n e c a n n o t . T h i s is quite mislead-
t r u e . T h i s c o m e s t o t h e s a m e a s a s k i n g h o w w e c a n rightly r e a s o n
ing. T h e important question is w h e t h e r o n e can think rationally about
about what our moral principles c o n c e r n i n g kindness and generosity
moral questions. In o t h e r words, are there ways of doing o u r moral
and c h e a t i n g at cards are to be (H 1 9 9 3 3 ) . T h a t question is just waved
r e a s o n i n g well or badly? This i m p o r t a n t question is c o n c e a l e d by
aside by t h o s e w h o speak in the way I have been c o m p l a i n i n g of. We
those w h o speak of cognitivism and non-cognilivism, and of k n o w -
shall return to it ( 7 . 8 ) .
ing that moral statements are true.

1 c a n perhaps show this by taking t h e word ' k n o w ' and doing t h e
3 . 7 . Before I end t h i s p a r t of t h e d i s c u s s i o n I w a n t to say a bit

more about the word 'true'. Up to now I have spoken as if it m e a n t no
s a m e with it as I have just been doing with t h e word ' t r u e ' . You re-
m o r e t h a n 'satisfying the truth conditions, w h a t e v e r t h o s e a r e ' . B u t
m e m b e r my example of s o m e o n e w h o said that a person was a good
the word 'true' h a s also certain formal properties which we must not
person, and said this b e c a u s e , a m o n g o t h e r t h i n g s , the person was
ignore. In giving an a c c o u n t of these. 1 am m u c h indebted to Crispin
kind and g e n e r o u s and n e v e r c h e a t e d at c a r d s . I am sure that this
Wright ( 1 9 9 2 ) . An example of these formal properties is t h e T a r s k i a n
s p e a k e r would c l a i m t h a t w h a t he said was true, and t h a t he k n e w
thesis t h a t i f p , t h e n i t i s t r u e t h a t p , and v i c e v e r s a . I suppose t h a t
that it was true. T h e linguistic p h e n o m e n a are not in doubt. He knew,
s o m e o p p o n e n t of m i n e m i g h t seek to c o n t r o v e r t w h a t I have been
t h a t is. that the person was kind a n d g e n e r o u s and never c h e a t e d .
saying by c l a i m i n g that t h e s e formal properties b l o c k t h e road to a
And this made his statement true according to the standards or truth
non-descriptivisl a c c o u n t of moral statements, or at least to the non-
c o n d i t i o n s t h a t he was using. As to t h e standards, he had no doubt
descriptivist's right to use 'true' of them. B u t they do not.
learnt these standards a n d n o t forgotten t h e m . He knew t h a t people
To explain this. I have to say s o m e t h i n g about t h e endorsing func-
w h o a r e kind and g e n e r o u s c a n be c a l l e d , so far as that goes, good
tion of the word 'true', which was first, I think, brought into the open
people, a n d t h a t people w h o c h e a t a t c a r d s c a n n o t b e called good
by Slrawson a long lime ago ( 1 9 4 9 , 1 9 5 0 ) . T h o u g h we do not say all
people. If a n y b o d y does n o t k n o w this, he would say. his e d u c a t i o n
t h a t c a n be said a b o u t t h e word ' t r u e ' in saying t h a t it is a word we
h a s been n e g l e c t e d . B u t t h e m o r e h a r d - b i t t e n person w h o c o n t r a -
use for endorsing w h a t s o m e o n e h a s said, it does have this function;
dicted him in t h e e x a m p l e could not, all t h e s a m e , be ruled o u t of
a n d this f u n c t i o n is by itself e n o u g h to a c c o u n t for t h e T a r s k i
II. 3- 7
II. 3. 7
TAXONOMY
59
p h e n o m e n o n . T h e r e are s o m e differences b e t w e e n the words ' t r u e '
out their b e c o m i n g different statements). If I say ' T h e sky is blue' and
and 'right', both of which are used for endorsing: but these I have dis-
s o m e o n e else says 'No, the sky is not blue', then we are indeed c o n t r a -
cussed elsewhere (II mjhb).
dicting e a c h other; but it must be the c a s e either that we are disagree-
If I say that ;> (some statement) is true. I thereby endorse it. But it is
ing a b o u t t h e descriptive state of t h e sky, or t h a t we a r e using t h e
obvious that if I say that p. I c a n n o t then, in the same breath, reluse
word ' b l u e ' , or o n e of the o t h e r words in t h e s e n t e n c e s , in different
to endorse the statement that p (the statement that I have just made).
s e n s e s . W e c a n n o t c o n s i s t e n t l y a g r e e a b o u t t h e descriptive state o f
T h i s is not merely a m a t t e r of p r a g m a t i c inconsistency, like that ol
the sky and use the words in the same senses and still contradict e a c h
the statement 'pbul I don't believe that /;'. If I said '/>. but it is not true
other. T h a t is, if we a g r e e about the descriptive state of the sky, and
that / / . I should be a c t u a l l y c o n t r a d i c t i n g myself (II 1 9 % " :
2.72).
agree in o u r use of t h e words, there is n o t h i n g left for us to disagree
Similarly, if I e n d o r s e t h e s t a t e m e n t that p. but refuse to affirm t h e
a b o u t . B u t in t h e 'good p e r s o n ' c a s e it m i g h t be t h a t we a g r e e d e x -
s a m e s t a t e m e n t . I contradict myself. And this is so. even t h o u g h all
actly about h o w t h e person behaved (what s h e did), and a b o u t t h e
the things I said earlier about the variability of the truth conditions of
(evaluative) m e a n i n g of 'good', but were c o n t r a d i c t i n g o n e a n o t h e r
moral statements still hold. A statement has. indeed (like some o t h e r
b e c a u s e we were e v a l u a t i n g differently people w h o did t h a t , or be-
speech acts besides statements), to have the formal property that it is
haved like that. By 'behaved like that' I m e a n that, for example, they
something which one can endorse, and which, if one can endorse it.
were kind and g e n e r o u s , and did not c h e a t at c a r d s . And by t h a t I
o n e h a s . on pain of self-contradiction, to be prepared to m a k e . B u t
mean that they, for example, gave m u c h of their m o n e y to relieve dis-
this could be so. even though different people might be using different
tress and did not hide cards in their sleeves in order to win the game.
standards or truth conditions when making this kind of statement.
T h e s e differences between the two kinds of speech a c t are readily
We can admit that in this respect (namely theTarski p h e n o m e n o n )
explained by t h e fact (hat moral s t a t e m e n t s have an e l e m e n t in their
moral statements behave just like any other kind of statement; but we
m e a n i n g w h i c h purely descriptive u t t e r a n c e s like ' T h e sky is blue' do
could go on to say that they differ in other respects. In particular, they
not have. T h i s is the evaluative element. Purely descriptive utterances
differ in that the truth conditions being used by one speaker may dif-
have ( 1 ) t h e s y n t a c t i c a l element, which in turn d e t e r m i n e s ( 2 ) their
fer from t h o s e being used by a n o t h e r , without the m e a n i n g s of t h e
i l l o c u t i o n a r y force ( t h a t they a r e descriptive s t a t e m e n t s ) , w h i c h in
two moral statements made by t h e m differing in all respects. If I say
turn requires ( 3 ) t h a i they have truth c o n d i t i o n s : a n d they have ( 4 )
of s o m e o n e ' S h e is a good person', and s o m e o n e else says 'No, she is
these particular truth conditions. Evaluative statements, by contrast,
not', then we are c o n t r a d i c t i n g o n e a n o t h e r , even ( h o u g h we a r e
have an additional e l e m e n t . T h e y have, as before, ( 1 ) t h e s y n t a c t i c a l
using different truth conditions; and this is because our evaluations,
element, which in turn determines ( 2 ) their illocutionary force (that
conveyed by the evaluative m e a n i n g s of our two utterances, are logic-
they are evaluative s t a t e m e n t s ) , which in turn requires ( 3 ) t h a t they
ally i n c o n s i s t e n t with one a n o t h e r . He is refusing to e n d o r s e w h a t I
have truth conditions; and they have ( 4 ) these particular truth condi-
have said. So he could h a v e said 'No. t h a t is not true". All s u c h
tions; bul in addition the illocutionary force requires (5) that they be
p h e n o m e n a will survive my c l a i m t h a t truth c o n d i t i o n s c a n v a r y
e v a l u a t i o n s : and this in turn m e a n s t h a t they c a n go on having this
without the entire m e a n i n g varying. Moral statements will still be. in
evaluative i l l o c u t i o n a r y force even if t h e t r u t h c o n d i t i o n s c h a n g e .
Crispin Wright's term, 'minimally truth-apt' ( 1 9 9 2 : 141 f f ) .
Thai is how describing is different from evaluating (for example c o m -
We may contrast this with what happens with ordinary purely de-
m e n d i n g ) . S i n c e e v a l u a t i n g is always a c c o r d i n g to standards, there
scriptive statements, whose truth conditions c a n n o t vary without the
will always be t r u t h conditions: but the m e a n i n g is n o t e x h a u s t e d by
niemiiiui of the s e n t e n c e s that express them c h a n g i n g (that is. with-
t h e truth c o n d i t i o n s , and s o w h a t r e m a i n s o f t h e m e a n i n g ( t h e
do
II. 3- 7
II. v S
TAXONOMY
6i
evaluative element) is e n o u g h to bring about a contradiction between
c o m m o n l y esteemed qualities, without himself e s t e e m i n g t h e m (LM
the two parties even t h o u g h they are using the words with different
7 o . FR 10.2).
descriptive m e a n i n g s . This is (lie extra bit of input that I m e n t i o n e d
I am quite confident that the s a m e treatment could be given to any
earlier. One of the parties is c o m m e n d i n g the person and the other is
e x a m p l e of a ' t h i c k ' or s e c o n d a r i l y e v a l u a t i v e c o n c e p t that w a s al-
refusing to assent to t h e c o m m e n d a t i o n . T h u s their s t a t e m e n t s a r e
leged to have descriptive and evaluative m e a n i n g s t h a t c a n n o t be dis-
mutually inconsistent. As Stevenson would have put it. there is a dis-
e n t a n g l e d . O n this, see M i l l g r a m ' s c o m m e n t s ( 1 9 9 5 ) o n W i l l i a m s
agreement in attitude which survives the agreement in belief.
( 1 9 X 5 : 1 4 0 ff.). O n e p a r t i c u l a r a r g u m e n t o f t h o s e w h o c l a i m this
People w h o d i s a g r e e with m y a n a l y s i s o f e v a l u a t i v e s e n -
should perhaps be m e n t i o n e d . It is often said t h a t if we h a d just t h e
t e n c e s often say t h a t it is not possible in all c a s e s to d i s e n t a n g l e t h e
descriptive m e a n i n g of 'kind' we might, indeed, be able to recognize
3.8.
evaluative from the descriptive element in their meanings. B u t this is
examples of kind people in the existing descriptive sense of the word,
wishful t h i n k i n g on t h e i r part. I have been in m a n y d i s c u s s i o n s of
but would be unable to extend or extrapolate its use to n e w and per-
this topic, and in them these descriplivisls have often brought up e x -
haps slightly different examples. T h i s seems to me to be simply false.
amples in which they say this disentangling is impossible. But I have
S u p p o s e t h a t I am the hard-bitten person f m e n t i o n e d earlier, a n d
always been able to achieve it fairly easily.
can recognize t h e qualities that people call kind and e s t e e m , but do
Here is an e x a m p l e to be going on with. A descriptivist m a y say
not myself esteem t h e m . And suppose t h a t some n e w e x a m p l e is pro-
that we c a n n o t d i s e n t a n g l e t h e e v a l u a t i v e from the descriptive ele-
duced of a person w h o does not have exactly t h o s e qualities, but h a s
m e n t s in t h e m e a n i n g of t h e word 'kind'. But this is really not very
qualities very like them, so that people w h o do esteem them are likely
difficult. Certainly to call somebody kind is normally to c o m m e n d him
to esteem that person too. and call him kind. I c a n see no difficulty in
(or h e r ) . It is to c o m m e n d h i m a c c o r d i n g to a certain standard. T h e
my predicting that this is what they will do. In order to m a k e this pre-
t r u t h c o n d i t i o n s of s t a t e m e n t s c o n t a i n i n g t h e word a r e fairly well
diction I do n o t myself have to e s t e e m t h e q u a l i t i e s or t h e person; I
known, although admittedly not precise. Suppose now that s o m e o n e
only have to be confident that they will. I find it surprising that people
gives much of his money to relieve distress. Nearly all of us would say
should rely on this very weak a r g u m e n t .
that s u c h a person was kind. But t h e r e might be s o m e o n e w h o
I have elsewhere (H 1 9 8 6 c = 1989: 116 ff.) gone into a lot of detail
t h o u g h t that it was not a c h a r a c t e r i s t i c of a good person to do this.
about t h e b e h a v i o u r o f t h e s e s e c o n d a r i l y e v a l u a t i v e o r t h i c k c o n -
This person could agree that s o m e o n e did this (namely gave m u c h of
cepts; so I do not need to do it again now. T h e motives of those w h o
his m o n e y to relieve distress), but might condemn his doing this, l i e
m a k e s u c h play with t h e s e c o n c e p t s are fairly e a s y to divine. T h e y
would then not be able to use 'kind' as a term of c o m m e n d a t i o n . But
were not a c t u a l l y t h e first to discover the distinction between thick
he might well be able to recognize t h e sort of people t h a t the o t h e r s
and thin concepts: see LM 7.5. FR 2 . 7 . But they found them attractive
called kind. So he would k n o w well t h e descriptive m e a n i n g they at-
b e c a u s e they seem to impugn the distinction between descriptive and
t a c h e d to t h e word. But he would not use it. b e c a u s e it c a r r i e d an
e v a l u a t i v e e x p r e s s i o n s . T h e i r useful feature, for descriptivist a r g u -
evaluative m e a n i n g to which he could not subscribe. He m i g h t stop
ments, is t h a t they have a descriptive m e a n i n g which is fairly securely
using the word altogether ( F R I O . J n . ) . or he might use it 'in inverted
attached to them. If o n e does not recognize as kind the sort of actions
c o m m a s ' , to signify t h a t a p e r s o n h a d t h e descriptive q u a l i t i e s e x -
that kind people do, t h e n o n e might be said not to k n o w t h e m e a n i n g
pected by most people in those called kind: he would be able to use the
of the word. Yet they are also, in their n o r m a l use. undeniably evalu-
word kind' purely descriptively, to signify the possession of t h o s e
ative, in t h a t s o m e o n e w h o called a person k i n d would be t a k e n by
h2
A TAXONOMY OF ETHICAL THEORIES
H-3-
nearly everyone to be c o m m e n d i n g h i m . So it is easy to see why descriptivists latched on to these words in the hope of casting doubt on
non-descriptivist theories. B u t a little m o r e attention to t h e analysis
of these concepts would have shown them, if they were willing to be
NAT UR A L I S M
s h o w n , how the two e l e m e n t s in the m e a n i n g of these words are related and how they can be distinguished.
4 . 1 . I N t h e preceding c h a p t e r I made t h e m a i n division o f e t h i c a l

theories info two genera, which I called descriptivist and non-descriptivist t h e o r i e s . I said t h a t t h e differentia b e t w e e n t h e s e w a s t h a t descriptivist t h e o r i e s affirm, but t h a t non-descriptivist t h e o r i e s deny,
that, apart from s y n t a c t i c a l features, the m e a n i n g s of m o r a l s t a t e m e n t s are determined entirely by their truth c o n d i t i o n s . In this and
the next c h a p t e r I am going to look at t h e different possible kinds of
descriptivist theories. 1 shall divide these up in the first place a c c o r d ing to the kind of truth conditions that they say determine the m e a n ings of m o r a l s t a t e m e n t s . T h e first division to be m a d e is into t h o s e
theories w h i c h hold t h a t the truth conditions a r e t h e possession, by
the actions, people, etc. about which the moral statements are made,
of what I shall call, following tradition, natural properties. T h i s is not
an entirely helpful term, and I shall have to explain it. But, again following tradition, I shall call theories that use this kind of truth condition in giving t h e m e a n i n g s of moral statements naturalist theories (H
1996).
I shall c o n t r a s t t h e m with theories w h i c h hold t h a t the truth c o n ditions w h i c h d e t e r m i n e t h e m e a n i n g s o f m o r a l s t a t e m e n t s are t h e
possession by a c t i o n s , people, e t c . of specifically m o r a l ( s o m e t i m e s
classified as ' n o n - n a t u r a l ' ) properties. They are sometimes called 'sui
generis' properties. I shall call such theories (still following tradition)
intuitionist theories. T h i s term too is unhelpful unless explained, and
h a s been used in different ways, especially recently. I might have used
instead the expression 'non-naturalist theories', and this would have
had the advantage of making explicit the d i c h o t o m o u s n a t u r e of my
classification. B u t I avoid it, because a non-naturalist theory might be
taken to m e a n any theory that rejected naturalism: and this would be
misleading, b e c a u s e all non-descriptivist theories too reject natural-
A T A X O N O M Y OF F i l l I t ' A I, T I I H O R I F S
II. 4-
II. 4. 1
N ATT IK A l . I S M
ism a l o n g with descriptivist theories in general. I am in this s e n s e a
digress into t h e s u b j e c t of modalities. Let me simply say t h a t we c a n
n o n - n a t u r a l i s t : I r e j e c t n a t u r a l i s m . B u t if I called myself a n o n -
make a d i c h o t o m o u s division of descriptivist theories into those, t h e
naturalist I might he thought to be allying myself with the intuition-
naturalist o n e s , w h i c h say t h a t the truth c o n d i t i o n s of moral state-
ists. that is. with the kind of tlescriptivisls who reject naturalism (such
ments c a n be specified without using any moral terms, and those, the
as Moore and I'richard). I do not want to give Ibis impression. So at
intuitionisl ones, which say that they c a n n o t , f shall now try to show
least, instead of ' n o n - n a t u r a l i s t ' . I should have to say n o n - n a t u r a l
that both these kinds of descriptivism. the naturalist kind and the in-
descriptivist'. and that would be intolerably cumbrous. So I ask you to
tuitionisl kind, get into trouble, and that t h e t r o u b l e is t h e s a m e in
pardon me if I go on using the term 'intuitionisl'. in spite of its ambi-
both c a s e s . T h e t r o u b l e is t h a t they both collapse into relativism (H
guity. What 1 mean by it will b e c o m e clear in due course.
19.861, 199.33). T h i s t e r m too I shall have to e x p l a i n . S i n c e t h e m a i n
I lowever. il does seem possible to divide descriptivist ethical theor-
purpose of most of those w h o e m b r a c e descriptivism is to avoid rela-
ies in this way by looking at t h e different kinds of properties w h i c h
tivism, this is a surprising result, and shows t h a t s o m e t h i n g has gone
they say have to go into a formulation of the truth c o n d i t i o n s of
seriously wrong. W h a t it is, we shall shortly discover.
m o r a l s t a t e m e n t s . Moore held so m u c h difficulty in s a y i n g w h a t he
I have put t h e distinction between naturalism a n d intuitionism in
m e a n t by non-natural property' that in the end he gave up the term.
terms of the different kinds of truth-conditions they impose on moral
Perhaps there are no such properties. But an intuitionisl. at any rate,
s t a t e m e n t s . My distinction is therefore broad e n o u g h to cover b o t h
h a s to think that there are these siii generis properties like goodness
the old-style 'refutation of naturalism' due to M o o r e ( 1 9 0 3 ) , and the
and wrongness which people, actions, etc. can have. T h e only defini-
new-style n a t u r a l i s m whose c h i e f habitat is Cornell. T h e old and t h e
tion that can be given of them is negative: we can say that a natural-
n e w n a t u r a l i s m s a r e n o t so different as is c o m m o n l y supposed;
ist is s o m e o n e w h o t h i n k s t h a t t h e truth c o n d i t i o n s of m o r a l
Morgan a n d T i m m o n s ( 1 9 9 2 ) h a v e adapted M o o r e ' s o p e n q u e s t i o n
s t a t e m e n t s require t h e possession by t h e i r s u b j e c t s of properties
a r g u m e n t to refute the new naturalism (H 1996(f). B u t t h e way I have
which can be defined, or their m e a n i n g explained, without using any
put t h e d i s t i n c t i o n will m a k e it apply to b o t h t h e old a n d t h e n e w
specifically moral t e r m s , but that an intuitionisl is s o m e o n e w h o
n a t u r a l i s m s . I d i s c u s s in t h e s a m e paper t h e w h o l e q u e s t i o n of
t h i n k s that in order to give the truth conditions of these s t a t e m e n t s
whether, as Pigden ( 1 9 9 1 ) appears to think, the fashionable new
we have to use specifically moral terms. T h e r e would be nothing very
metaphysics of P u t n a m and others can help the naturalists: and this
s c a n d a l o u s in this. P h i l o s o p h e r s c o n t i n u e to a r g u e about w h e t h e r
must be my excuse for not venturing on il here.
nwilal t e r m s like 'possible' and ' n e c e s s a r y ' c a n be defined w i t h o u t
4 . 2 . Let us then look first at naturalist theories. T h e y c a n in turn
using other modal terms in the delinilion. In both the moral and the
be subdivided. Given that (hey have to specify the truth conditions of
modal cases the issue is w h e t h e r we can ever break out of the circle.
moral s t a t e m e n t s without using any m o r a l t e r m s , they still have a
T h i s r e s e m b l a n c e is no accident, b e c a u s e the affinity between m o r a l
c h o i c e as to the kind of terms they w/// use in specifying t h e truth c o n -
and modal t e r m s is very strong, as is s h o w n by the e x i s t e n c e of t h e
ditions. S o m e kinds of naturalist specify the truth conditions without
subject called 'deontic logic'. Deontic modalities like 'ought' have a lot
reference to the attitudes, etc. of the speakers of the statements, or of
in c o m m o n with logical, alethic. or causal modalities like n e c e s s a r y ' ;
their society. I am going to call this kind of n a t u r a l i s t an objectivist
and the different sorts of modal logic which deal with these kinds of
n a t u r a l i s t . O t h e r kinds o f n a t u r a l i s t specify t h e t r u t h c o n d i t i o n s o f
modality therefore bear a striking resemblance to each other. It would
moral s t a t e m e n t s i n t e r m s o f t h e a l t i t u d e s , e t c . o f t h e people w h o
not be e x t r a o r d i n a r y if in all t h e s e c a s e s t h e r e were a c i r c l e o u t of
make t h e m . 1 shall call s u c h people subjectivist naturalists. 1 am going
which we could not break (H I 9 9 b d : .354). ' shall not. however, n o w
to deal n o w with objectivist naturalism, and c o m e b a c k to subjectivist
hh
A T A X O N O M Y OH E T H I C A L T H E O R I E S
II. 4- 2.
II. 4. 2
NATURALISM
67
n a t u r a l i s m later, after I have dealt with iniuitionism. T h e reason for
applied to without appealing to a n y t h i n g e x c e p t t h e o b s e r v a b l e lin-
this p o s t p o n e m e n t is t h a t t h e r e a r e very strong r e s e m b l a n c e s
guistic behaviour of the speakers and the observable properties of the
between intuitionism and subjectivist naturalism, although these
a c t i o n s . If we were to appeal, for example, to our own a s s e s s m e n t of
resemblances are usually ignored or repudiated by intuilionists. And
the a c t i o n s as right or wrong, that would vitiate t h e r e s e a r c h . Eor if
for short I shall in what follows call objectivist naturalism simply 'nat-
we did this, then w h a t we should discover is not w h a t actions in par-
u r a l i s m ' . It is by far t h e most i m p o r t a n t variety of n a t u r a l i s m , and
ticular the speakers were applying the words to. but r a t h e r w h e t h e r
exhibits very clearly the dangers of this kind of theory.
their assessment of the actions corresponded to our own. We need to
T h e truth conditions of moral s t a t e m e n t s are determined by the
be able to specify the sorts of actions to which they apply the words in
c o r r e c t application of moral predicates like 'right', ' w r o n g ' , 'good',
a morally neutral way; otherwise we shall not be doing t h e research
and ' b a d ' to a c t i o n s or people. T h i s is t r u e on all theories, b o t h de-
in the way t h a t a truly naturalistic theory h a s to. We have to establish
scriptivist and. as we shall later see. non-descriptivist (7.X. II 1995/)).
(hat these a r c the (neutrally specified) sorts of a c t i o n s to which t h e
Suppose, therefore, that we were to try to find out what these truth
words are applied by native speakers.
conditions are by discovering to what a c t i o n s or people these predi-
If this is t h e way our research goes, we shall have achieved some-
cates are correctly applied. If our enquiry were successful, we should
thing. We shall have discovered a rule for the application of the predi-
(hen have established J/JC truth conditions of the statements. But how
c a t e ' w r o n g ' s u c h that, if we follow it. we shall c o n f o r m perfectly to
do we discover to what objects predicates are correctly applied? In the
the usage, as r e g a r d s t h e application of this word, of t h o s e w h o s e
c a s e of predicates in g e n e r a l , we do it by e x a m i n i n g t h e l i n g u i s t i c
usage we were studying, that is, of native speakers of English. And
usage of native speakers of the language; and I can see no reason for
this is indeed what we could quite safely do with ordinary descriptive
thinking (at least no reason why a naturalist should think) t h a t it is
predicates. If we w a n t e d to find a rule for t h e c o r r e c t application of
any different with moral predicates. Indeed, this seems to be the only
the English word 'red', for example, and t h u s find out the truth condi-
way a naturalist could d e t e r m i n e t h e i r use in c o m m o n p a r l a n c e .
tions of s t a t e m e n t s c o n t a i n i n g it. we could safely do it by seeing to
Suppose, therefore, that, taking ' w r o n g ' as an example, we e x a m i n e
what sorts of things native speakers of English applied this adjective.
t h e use of this predicate by native speakers of English (one could do
If we then applied it to those and only those sorts of things, we should
the s a m e in Swedish, but I will not try. b e c a u s e to my regret 1 do not
be applying it correctly, and our statements would be true.
speak it). We shall discover that native speakers of English apply the
B u t if we try to follow this p r o c e d u r e with t h e word ' w r o n g ' , we
predicate 'wrong' to certain sorts of actions, and refuse to apply it to
shall at o n c e get into difficulties. Not all native s p e a k e r s of English
o t h e r sorts. Can we therefore say that the truth conditions of state-
apply t h e predicate ' w r o n g ' to the s a m e sorts of t h i n g s , not even in
m e n t s c o n t a i n i n g the predicate ' w r o n g ' (in its moral use) are these:
England, let a l o n e in A m e r i c a or A u s t r a l i a . P e r h a p s , if I had been
the statements are true if in them t h e predicate 'wrong' is applied to
doing this in S w e d i s h . 1 should not have got i n t o so m u c h t r o u b l e ,
t h e sorts of a c t i o n s to w h i c h native s p e a k e r s of English do apply
b e c a u s e S w e d e n h a s a fairly h o m o g e n e o u s m o r a l c u l t u r e , a n d it
them, and false if in them the predicate 'wrong' is applied to the sorts
might be. though I rather doubt it. that all native speakers of Swedish
of actions to which native speakers of English do not apply them?
apply the word ' w r o n g ' to the very s a m e sorts of things. But even in
But there is a snag here which we m u s t be careful to avoid. W h a t
Sweden, can it be the c a s e that there are no i n s t a n c e s of divergence
do we m e a n by 'the sorts of actions'? In order for this piece of linguis-
in the use of the Swedish word for 'wrong'? I am sure t h a t in fact it is
tic r e s e a r c h to give an o b j e c t i v e result, we shall have to be able to
not the c a s e . S u p p o s e we are talking a b o u t e a t i n g n o n - h u m a n a n i -
r e c o g n i z e and specify t h e s o r t s of a c t i o n s that the words a r e b e i n g
mals. 1 am quite sure that I shall tind m a n y Swedes w h o say that this
A T A X O N O M Y 0 1 ' K I T I I C A 1 , T111- O R I L S
W-4-2
is not wrong, but some w h o say that it is wrong. With English this is
even more obvious. There are a great m a n y kinds of thing which sonic
English speakers call wrong but others do not. Think, for example, of
abortion, or of lighting in wars ( 2 . 2 . 6 . 9 ) . So we shall not lind a single
rule for the correct application of the word (a set of truth conditions
for statements containing it) which will do for us what the rule for the
use of 'red' did. We shall not find a rule, that is. by c o n f o r m i n g to
w h i c h we c a n be sure of m a k i n g true moral s t a t e m e n t s . Rather, we
shall find a great m a n y rules, i n c o n s i s t e n t with one a n o t h e r , and
shall simply not know, by this method, how to use the word.
I want to ask. what is the status of these conflicting rules for the
use of w r o n g ? f o l l o w i n g the n a t u r a l i s t s , we have been a s s u m i n g
t h a t what we were discovering was a rule for the c o r r e c t application
of the word, and n o t h i n g m o r e , lint we now see that t h a t was not
what we were discovering. At least, if we were discovering a rule for
the correct application of the word, it was not a purely linguistic rule.
It was in fact a substantial moral rule ( 1 9 9 6 J ) . If one lot of people say
that abortion is wrong and a n o t h e r lot say it is not wrong, they a r e
not differing merely in their linguistic usage. They are expressing different m o r a l opinions. T h i s s h o w s very clearly w h a t is w r o n g with
n a t u r a l i s m . W h a t is w r o n g is that it pretends that w h a t a r e in fact
substantial moral principles are n o t h i n g more than linguistic rules.
Naturalism confuses learning morals with learning a language. Hut
the two are very different. If I have grown up thinking that abortion
is w r o n g . I have a c q u i r e d m o r e t h a n a m e r e linguistic skill. 1 have
acquired a moral principle, an attitude to abortion.
4. J. Now I think you will be able In see why the naturalist kind of
descriptivism leads inevitably to relativism (II 1995.'/)- T h e r e a r e in
most languages words which we translate 'wrong'. These words are,
as they are used, rough equivalents to one another. But the cultures
that use these words call quite different things wrong, hi one culture,
for example, it may be t h o u g h t wrong not to light for o n e ' s country,
in a n o t h e r more pacific culture it may be thought wrong to fight. T h e
important t h i n g to get hold of is t h a i , a l t h o u g h the people in t h e s e
cultures hold different opinions about the wrongness of lighting, they
may be using the word 'wrong', or its equivalents, in the s a m e sense.
M.4.3
NATURALISM
Otherwise they would not be contradicting o n e another, which they

clearly a r e . T h e people in o n e c u l t u r e a r e s a y i n g t h a t fighting is
wrong, and the people in the other are saying t h a t it is not wrong, in
the same sense of 'wrong', so far as its evaluative m e a n i n g goes (cf. LM
6 . 6 , FR 6 . 5 , M T 4 . 2 ) . But if we follow the naturalists, we shall have to
say that the senses of the word in the two cultures are entirely different. T h i s will have t h e c o n s e q u e n c e t h a t they a r e n o t c o n t r a d i c t i n g
o n e a n o t h e r : for fighting m i g h t be w r o n g in the s e n s e of the word
used by o n e culture, but not wrong in the sense of the word used by
the other. T h e people in each culture will be right in their own sense
of the word ' w r o n g ' . If we distinguish the senses by using different
subscripts, we c a n say t h a i o n e of the c u l t u r e s t h i n k s fighting is
w r o n g , . but that the other thinks it is not w r o n g , . But these two opinions may be mutually consistent, if the two senses of 'wrong' are different.
'There would be no h a r m in this if all they were doing were describing the act of lighting. 'They would just be attributing various descriptive properties to the ad of lighting. T h e trouble starts when we begin
using ' w r o n g ' for t h e purpose for w h i c h it a c t u a l l y is used in l a n guage, n a m e l y for condemning a c t s . T h e n a t u r a l i s t , in a c c o r d a n c e
with his descriptivism. c a n n o t include this purpose in his a c c o u n t of
the m e a n i n g of 'wrong'. But it is very natural, since this is actually its
use, to think t h a t the people in the two cultures are, respectively, c o n d e m n i n g and refusing to c o n d e m n the a c t of lighting. T h e n they are
c o n t r a d i c t i n g e a c h other. B u t a c c o r d i n g t o the n a t u r a l i s t they m a y
both be right in what they say. 'There is no contradiction. 'The naturalist s e e m s to be led to the c o n c l u s i o n thai it is both right for o n e culture to c o n d e m n lighting, and right for t h e o t h e r c u l t u r e n o t to
condemn it. And this is a relativist position. I shall answer below ( 4 . 6 )
the objection that the example of attitudes to fighting is unfair, as not
sufficiently 'basic', and t h a t we should have taken as an example attitudes to ' h u m a n flourishing'.
But first we m u s t e x a m i n e a possible escape route for the naturalist
from w h a t we h a v e said so far. T h e n a t u r a l i s t m i g h t s e e k to e s c a p e
this c o n c l u s i o n by s a y i n g (as in c o n s i s t e n c y w i t h his position he
must) t h a t in calling acts wrong one is not c o n d e m n i n g them. He is in
A T A X O N O M Y OF F T H I C A F THHORIFS
7<>
II. 4 3
II. 4. 4
NATURALISM
71
fact in a dilemma. Either he says t h a t to call an act w r o n g is to c o n -
O n c e this is e x p l a i n e d , I c a n see no difficulty in d i s t i n g u i s h i n g
demn it. in which case his theory lands him in relativism. Or he says
evaluative words or uses of words from descriptive ones. A simple test
that to call an act wrong is not to condemn it. in which c a s e it is very
is provided by M o o r e ' s w e l l - k n o w n 'open q u e s t i o n ' a r g u m e n t (H
hard to say what he thinks ' w r o n g ' does m e a n . And even if he says
i 9 9 6 d ) . For any predicate P. if it is possible to ask Agreed, it is P, but is
this, he is involved in a kind of relativism; for he is left s a y i n g t h a t
it wrong?' or 'It is P, but is it a good one?', and if a negative answer is
people who call the act of fighting w r o n g and people w h o call it not
not self-contradictory, then P is a purely descriptive predicate. If a def-
wrong are both right. They could both be right if the word m e a n s dif-
inition of an evaluative word is to be naturalistic, t h e n t h e dejiniens
ferent things in the two cultures. But this loo is a kind of relativism.
has to be purely descriptive in this sense.
At any rate, he is likely to be left in the position embraced by Professor
4 . 5 . There are various other objections that a naturalist might
M a c l n t y r e ( 1 9 8 4 ) . that people in different c u l t u r e s simply c a n n o t
make to my a r g u m e n t , with w h i c h I must now deal. T h e first two of
c o m m u n i c a t e with o n e a n o t h e r , b e c a u s e they lack t h e l i n g u i s t i c
them c o n c e r n a m a t t e r which I slid over too quickly earlier. I said, you
m e a n s of doing so. But s i n c e I have written at length a b o u t
r e m e m b e r , that t h e sort of a c t i o n s that people applied words like
M a c l n t y r e ' s position e l s e w h e r e ( 1 1 1 9 8 6 c ) . I will not say a n y t h i n g
'wrong' to had to be specifiable in a morally neutral way. T h a t was all
about it now.
right as far as it went. But I w e n t further: 1 said t h a t t h e y had to be
4 . 4 . People w h o incline t o n a t u r a l i s m s o m e t i m e s say t h a t i n ar-
specifiable without appealing to a n y t h i n g but observable behaviour
g u m e n t s like the one I have just been setting out it is simply assumed
of the speaker and observable properties of the a c t i o n s . But, it might
that a distinction c a n be m a d e b e t w e e n evaluative and descriptive
be objected, a great m a n y words are s u c h t h a t we c a n n o t say w h a t
words, but that in fact no s u c h distinction can be drawn: the words
things people apply them to without appealing to m o r e t h a n observ-
we call 'evaluative' are simply o n e kind of descriptive word. We m a y
able b e h a v i o u r and t h e observable properties of o b j e c t s . It might be
reply that at any rate they are a special kind of word, which is distin-
c l a i m e d t h a t all words for so-called ' s e c o n d a r y q u a l i t i e s ' , s u c h as
guishable from other kinds. T h e i r distinguishing feature is t h a t t h e y
colour words, fall into this class. For example, h o w do we tell w h a t ob-
are used for e v a l u a t i n g s o m e t h i n g , that is for c o m m e n d i n g or c o n -
jects people apply the word 'red' to? Redness is on the face of it an ob-
demning it.
servable property. But, it might be said, if we are to say correctly w h a t
It has to be admitted that even purely descriptive words am be used
people who know t h e l a n g u a g e are applying it to, we have to say that
for c o m m e n d i n g . To cite a c o m m o n n o n - m o r a l e x a m p l e (H 1 9 9 6 c :
they are applying it to things that appear in a certain way to t h e m . For
2 6 1 ) : one might c o m m e n d a certain hotel by saying that it faced t h e
people a r c s o m e t i m e s mistaken about w h a t things a r e red. T h e y m a y
sea. But there is a difference between saying that the hotel faces the
be white t h i n g s in a red light, for e x a m p l e . Or t h e people m a y have
sea and saying that it is a good hotel, as we c a n easily see. W h e t h e r
suddenly gone colour-blind. In either of t h e s e c a s e s , t h e y are using
the fact that the hotel faces the sea c o m m e n d s it to someone depends
the word correctly to describe objects which they mistakenly think are
on whether he likes hotels t h a t face the sea. A person w h o did not like
red. It is t h e i r o b s e r v a t i o n t h a t is at fault, n o t t h e i r u s e of t h e l a n -
such hotels could without contradiction say that the hotel faced the
guage.
sea but was not for t h a t r e a s o n a good o n e . But he could not a g r e e
If. accordingly, we a r e to distinguish g e n u i n e linguistic m i s t a k e s
that it was a good hotel and still m a i n t a i n that it was not a good one.
from faults in observation, we shall have, in t h e c a s e s of t h e colour-
To call it a good hotel has to be to c o m m e n d it. whatever one's stand-
blind and of those w h o are in a bad light, to distinguish i n c o r r e c t uses
ards of goodness in hotels, u n l e s s o n e is using 'good' (as of c o u r s e
of the word 'red' (that is, uses of it with a different m e a n i n g from that
one c a n ) in some 'inverted c o m m a s ' or off-colour sense.
which it standardly h a s ) , from i n c o r r e c t applications of it (that is. ap-
A T A X O N O M Y OF MT'i 1I (.'A L I'll F O R I MS
II-4-S
II.4. 5
NATURALISM
7?
plications of it to objects to which it is not standardly applied). T h i s is
one of the things called wrong, they c a n simply write the deviant off
obviously no place to go into this very difficult question. We can avoid
as misusing the word 'wrong'. So the correctness of the application of
it by stipulating that in our linguistic research we take into a c c o u n t
moral words h a s to be assessed relative to t h e c u l t u r e within which
only standard uses of the words we are investigating: that is. only uses
they are being used, and it becomes by definition impossible to preach
which are free from both linguistic and observational mistakes. And
moral reform. If we w a n t to avoid this c o n c l u s i o n , we s h a l l h a v e to
the only way. so far as I c a n see. to tell which uses are standard and
give up saying that moral properties are like secondary qualities such
free from both kinds of mistakes is to select a class of speakers w h o .
as 'red'.
we decide, m a k e n e i t h e r kind of m i s t a k e (at leasl on t h e o c c a s i o n s
4 . 6 . So m u c h for what I shall call the 'objection from s e c o n d a r y
when we e x a m i n e t h e m ) and to record their usage. We shall then no
q u a l i t i e s ' . Next, we h a v e to c o n s i d e r t h e o b j e c t i o n t h a t 1 have been
longer have, for our present purposes, to distinguish between use and
unfair in my c h o i c e of examples of applications of moral words. You
application.
will r e m e m b e r t h a t I used e x a m p l e s like a b o r t i o n a n d fighting. It
T h e effect of this stipulation will be that we shall get a c l a s s of
might be o b j e c t e d that t h e m o r a l s t a t e m e n t s that a b o r t i o n , or that
speakers who all apply the word 'red' to the same objects, 'the s a m e '
lighting, is w r o n g a r e not sufficiently basic, in a s e n s e that I m u s t
being defined objectively in terms of standard conditions of use. And,
e x p l a i n . It might be alleged t h a t if people call t h e s e kinds of a c t s
we may rightly add in defence of the stipulation, there h a s to be this
wrong, they call t h e m so, not because of w h a t they are in themselves,
kind of standard use. if ' l a n g u a g e is to be a m e a n s of c o m m u n i c a -
but b e c a u s e t h e y a r e infractions o f s o m e h i g h e r a n d m o r e g e n e r a l
tion' (Wittgenstein 1953: sec. 2 4 2 : see 11 1 9 9 M . This, we may admit
principle which determines, in combination with certain factual pre-
at least for the sake of a r g u m e n t , is a condition for the successful use
misses, that they are wrong. If this were so. then there would not, at
of words like 'red' in c o m m u n i c a t i o n . We shall see later that this by
the fundamental level, be the sort of moral d i s a g r e e m e n t on which I
no m e a n s applies to words like ' w r o n g ' , in spite of t h e efforts of
have been trading in my a r g u m e n t . T h e parties might disagree about
descriplivists to persuade us t h a t it does. But of words like 'red' it is
t h e m o r a l i t y o f fighting o r o f a b o r t i o n , b u t o n l y b e c a u s e t h e y dis-
perhaps true, though it has been argued (Lewis and Woodfield 19X5)
agreed a b o u t t h e facts. T h e y might agree, t h e o b j e c t i o n goes on, on
that the claim is not without its difficulties. So let us ask what would
the w r o n g n e s s of doing what results in the diminution of happiness
be the c o n s e q u e n c e s for e t h i c a l t h e o r y if it were true of words like
or h u m a n nourishing, or in the failure to meet fundamental h u m a n
'wrong'. T h e r e would then be a 'standard use' of these words, and all
needs: they differ only in that one side thinks that abortion (or light-
uses which deviated from this standard would be simply incorrect. So
ing) would have this result, and the other thinks it would not.
what I said about naturalism collapsing into relativism would be sub-
Let us l a k e first t h e ' f u n d a m e n t a l h u m a n n e e d s ' f o r m u l a t i o n ,
stantiated, and the objection we are considering would fail. For it is
which brings out well the difficulties of sustaining Ibis kind of objec-
obvious that different c u l t u r e s have different standard applications
tion. I said earlier that, in order to be a true e x a m p l e of n a t u r a l i s m ,
for the word 'wrong', as I showed earlier. And in each of the cultures
an ethical theory had to specify in morally neutral terms t h e applica-
c o n f o r m i t y to their differing s t a n d a r d uses would be e n o u g h to
tions of a m o r a l word w h i c h were to c o u n t as c o r r e c t . We c a n now
secure correctness in moral judgement.
see how n e c e s s a r y this c o n d i t i o n w a s . T h e q u e s t i o n at issue is
Let me cite an e x a m p l e w h i c h I have used before (H 1 9 8 6 c ) .
w h e t h e r e x p r e s s i o n s like ' f u n d a m e n t a l h u m a n n e e d s ' c a n ever b e
Suppose t h a t s o m e deviant says t h a t it is w r o n g n o t to love o u r e n -
morally neutral. If they c a n n o t , the naturalist will again turn into a
emies. If the people in his society have a standard application for t h e
relativist, as we s h a l l see. B u t before t h a t I h a v e to m a k e s o m e
word wrong', according to which this (not loving our enemies) is not
remarks, not for the first time, about the word 'needs' (see H 1979/1)-
74
II. 4- 6
T h e r e i s a dispute b e t w e e n t h o s e w h o t h i n k t h a t n e e d s c a n b e
absolute a n d those w h o t h i n k t h a t all needs are relative to s o m e end.
T h a t is. do all things needed have to be needed for s o m e purpose, or
c a n they be just needed? Certainly s o m e are needed for a purpose. For
example, I need transport to get to Stockholm; if I were n o t going a n y where. I would not need transport. Are there things needed w h i c h we
do not need in order to do anything, or for any purpose? Etymology is
on the side of those who deny this. T h e word 'need' in languages akin
to G e r m a n is closely linked with words for necessity. 'Need' in G e r m a n
is 'Not', and this is c o g n a t e with 'notwendig' and 'ntig', both m e a n i n g
' n e c e s s a r y ' . T h e s a m e is true of Latin. This seems to indicate t h a t for
something to be a need is for it to be a necessary condition for obtaining something else.;
On t h e o t h e r side, t h e r e a r e c e r t a i n l y i n s t a n c e s in w h i c h we say
t h a t s o m e b o d y needs s o m e t h i n g , b u t i n w h i c h the q u e s t i o n ' W h a t
for?' s e e m s out of place. If s o m e b o d y is 'in dire n e e d ' b e c a u s e of
poverty, it is easy to think t h a t he (or she) 'just needs' food, or help of
s o m e other kind. But this is a rather superficial argument. Obviously
he needs food, say, in order to survive. So far the need is relative to an
e n d . B u t . it will be said, everybody needs to survive; so t h e r e is an
a b s o l u t e need to survive lying b e h i n d the relative n e e d for food.
However, this is a mistake. Not everybody needs to survive. S o m e terminal patients in pain do not w a n t to live, and would not say t h a t they
needed to live. Living is for t h e m not a n e c e s s a r y condition for a n y thing else that they want.
One might try to get over this difficulty by putting, in place of 'survival', s o m e m o r e g e n e r a l t e r m like ' h u m a n flourishing'. T h i s w a s
first introduced, I think, as a translation of Aristotle's key term 'eudaimonid', c o m m o n l y b u t i n c o r r e c t l y t r a n s l a t e d h a p p i n e s s ' . I h a v e
a l r e a d y m e n t i o n e d b o t h t e r m s . B u t A r i s t o t l e h i m s e l f n o t e s t h e indeterminacy of m e a n i n g of such phrases. He says:
since all knowledge and choice aims at some good, what is it at which we say
politics aims, and what do we say is the highest of goods achieved by action?
In name, at any rate, we might almost say that it is agreed by the great majority. For both the many and the better sort say it is eudaimoni, and they un-
II. 4. 6
NATURALISM
75
derstand 'living well' and 'doing well' as the same as this. But about what eudaimonia is, they disagree, and the many do not give the same answer as the
wise. (Eth. Nic. K>95 i4 ff.)
a
And Aristotle is obviously also aware t h a t the t e r m is n o t evaluatively

n e u t r a l , a s i s s h o w n b y t h e e q u i v a l e n t s h e gives, 'living well' a n d
'doing well'. T h e latter is a notoriously a m b i g u o u s expression widely
used by Plato (e.g. in the last two words, as well as in the s e c o n d book,
of the Republic). It c a n m e a n either 'acting well' or 'faring well'. Even
the prefix of eudaimonid gives the game away; it is the s a m e as is translated 'well' in t h e two other expressions. Perhaps t h e best literal t r a n s lation of eudaimonid is ' h a v i n g a good daimon' (we m i g h t say, 'good
fortune', a person's daimon being his private deity, b e n i g n or m a l i g n
as the c a s e m i g h t b e ) .
T h o s e descriptivists w h o wish t o insist o n a n a b s o l u t e s e n s e o f
'need' c a n n o t therefore appeal to Aristotle. For if people need food in
order to flourish, there will be disputes about w h a t c o u n t s as flourishing. And even if t h e r e were not, the move would n o t succeed. For alt h o u g h we m i g h t say t h a t s o m e o n e needs food, clothing, shelter, etc.
in order to flourish, we c a n n o t say t h a t he needs to flourish?Everyone,
perhaps, wants to flourish (though they m a y differ about w h a t c o u n t s
as flourishing). Indeed, it is perhaps an analytic truth t h a t everybody
w a n t s to flourish, b e c a u s e T w a n t n o t to flourish' sounds at least logically odd. B u t it is a misuse of l a n g u a g e to say t h a t s o m e o n e needs
to flourish. If we understand this at all, we shall be tempted to ask, in
perplexity, ' W h a t for?'; and it is n o t at all clear w h a t the a n s w e r could
be. T h i s in itself is an i n d i c a t i o n t h a t needs h a v e to be relative to a
purpose.
It follows from this that the naturalist move I have been discussing
is bound to fail. For, as we saw, it is a condition for being a real naturalist that one gives t h e truth conditions of m o r a l s t a t e m e n t s in terms
of properties w h i c h a r e determinate, and c a n indeed be determined
by o b s e r v i n g t h e s t a n d a r d application of m o r a l words. B u t suppose
t h a t the n a t u r a l i s t n o w c l a i m s t h a t t h e r e is a s t a n d a r d application:
' m o r a l l y w r o n g ' , for e x a m p l e , is applied to a c t i o n s w h i c h d e n y to
people t h e i r f u n d a m e n t a l h u m a n needs. W e h a v e only t o point o u t
76
II. 4. 6
II. 4. 7
NATURALISM
77
t h a t t h e r e is no s t a n d a r d a p p l i c a t i o n for this latter e x p r e s s i o n a t
the kind of non-naturalistic, non-descriptivistic utilitarianism w h i c h
least n o n e t h a t helps the descriptivist. Admittedly, 'needs' in its rela-
I myself favour; so t h e differences will be instructive. I could have said
tive s e n s e is standardly applied in c a s e s w h e r e t h e q u e s t i o n ' W h a t
' . . . maximizes t h e furtherance, in sum, of t h e interests of all affected
for?" h a s an answer. But our naturalist is not appealing to those cases.
parties'; but I prefer t h e formulation in terms of preferences, b e c a u s e
He is appealing to alleged cases w h e r e t h e question is inappropriate.
the word 'interests' is a very unclear one. It could be a c c u s e d of being
T h a t is. the needs to which he is appealing have to remain needs what-
evaluative, a n d t h u s of spoiling the naturalistic credentials of t h e def-
ever anybody's aims or purposes are. But if w h a t I have just been say-
inition: but 'preferences' is all right, b e c a u s e it is clearly a m a t t e r of
ing is true, t h e r e a r e no s u c h a b s o l u t e needs. Even if everybody,
fact w h a t people do prefer ( a l t h o u g h t h e i r p r e f e r e n c e is itself an
analytically, w a n t s to flourish, different people will c o u n t different
evaluation on their part).
kinds of life as flourishing; so t h e t r u t h c o n d i t i o n s of m o r a l s t a t e -
T h i s kind of n a t u r a l i s m could be said to be subjeetivistic, b e c a u s e
ments (the application rules for m o r a l words) will vary from o n e per-
preference is a subjective state. B u t I am n o t going to hold t h a t against
s o n or at least from o n e c u l t u r e to a n o t h e r , and the n a t u r a l i s t will
t h e definition; it is w o r t h while e x a m i n i n g it all t h e s a m e . We m u s t
again turn out to be a relativist.
d i s t i n g u i s h this definition from t h a t w h i c h s a y s t h a t ' r i g h t a c t i o n '
4 . 7 . I am conscious t h a t m a n y people will be dissatisfied with my
m e a n s 'action w h i c h maximizes t h e satisfaction of the speaker's or the
a r g u m e n t so far. T h e y will complain that although 1 have refuted par-
agent's preferences'. T h e s e definitions are versions of egoism, a n d I do
t i c u l a r kinds of n a t u r a l i s m , I h a v e n o t precluded t h e r e b e i n g s o m e
n o t t h i n k t h a t a n y b o d y would a g r e e t h a t t h a t is w h a t t h e y mean by
kind w h i c h escapes these refutations. I did indeed set o u t a perfectly
' r i g h t a c t i o n ' i f t h a t i s t a k e n i n its m o r a l s e n s e . T h e definition a l s o
general a r g u m e n t to show t h a t naturalism is bound to result in rela-
runs up against Moore's a r g u m e n t already mentioned; it is clearly an
tivism. B u t t h e suspicion will r e m a i n t h a t there might be s o m e ver-
open question w h e t h e r an action which m a x i m i z e s t h e satisfaction of
sion of it which could avoid this c o n s e q u e n c e .
t h e speaker's or t h e a g e n t ' s preferences is t h e right a c t i o n . B u t if we
T h e only way I c a n think of to m e e t this objection is to choose, as a
say 'all affected parties' t h e definition b e c o m e s a bit m o r e plausible.
candidate for a viable naturalistic theory, a theory which seems to me
S u c h a definition is n o t so obviously relativistic as s o m e we h a v e
most likely to avoid collapsing into relativism, and discuss it in m o r e
discussed; for if we take into a c c o u n t all t h e preferences of all t h e af-
detail. T h e most plausible candidate for a delinition of 'right a c t i o n ' is
fected parties, we shall get a unique a n s w e r to t h e question ' W h a t is
a utilitarian one. S u c h a definition h a s sometimes been attributed to
t h e r i g h t a c t i o n ? ' w h o e v e r i s a n s w e r i n g it, a n d w h a t e v e r his indi-
]. S. Mill; but wrongly. T h e famous s t a t e m e n t of his view at t h e begin-
vidual preferences are. I would therefore t h i n k this t h e m o s t accept-
ning of his Utilitarianism, that actions are right in proportion as they
able form of n a t u r a l i s m . It is s o m e w h a t s i m i l a r to David B r i n k ' s (H
lend to promote happiness, etc. ( 1 8 6 1 : ch. 2 ) , is not intended as a def-
1 9 9 6 c ) . B u t t h e r e a r e still s o m e t h i n g s w r o n g w i t h it, a n d I d o u b t
inition but as a s u b s t a n t i a l c l a i m a b o u t w h a t a c t i o n s are right. His
w h e t h e r in t h e end it escapes relativism. For if we i m a g i n e two people,
view about t h e meanings of m o r a l words is a clearly prescriptivist o n e .
o n e a u t i l i t a r i a n a n d o n e not, w h o for t h a t r e a s o n say o n e o f t h e m
and is set oul in Mill 1 8 4 3 , last chapter.
t h a t an action is right and the other t h a t it is not, the s a m e difficulties
However, let us try o u t a u t i l i t a r i a n n a t u r a l i s t i c definition, in a
will arise as before. If t h e utilitarian says t h a t his view is established
s o m e w h a t more up-to-date form. I will formulate it as follows: ' R i g h t
by t h e very meaning of right, his opponent will reply t h a t he himself
a c t i o n ' m e a n s ' a c t i o n w h i c h m a x i m i z e s t h e satisfaction, i n s u m , o f
does n o t m e a n t h e s a m e by it; a n d t h e n t h e y will be left b o t h saying
t h e preferences of all affected parties.' T h i s formulation h a s t h e ad-
w h a t is true in their own different senses of 'right' (H I 9 9 6 d ) - S u c h an
v a n t a g e for my purposes t h a t it s e e m s to c o m e as close as possible to
impasse c a n only be avoided, I think, by r e i n t r o d u c i n g a prescriptive
II. 4-7
II. 4. 7
NATURALISM
79
e l e m e n t into t h e m e a n i n g of 'right'; they will then be really c o n t r a -
think is t h e m o s t acceptable form, c a n hardly survive as an explana-
dicting one another, and c a n begin to discuss their dispute in a c o m -
tion of t h e meaning of m o r a l words.
m o n language. How this is to be done, I shall leave until later (7.4 ff.).
4 . 8 . A t this point I should p e r h a p s s a y s o m e t h i n g a b o u t r e l a -
T h e r e are also o t h e r difficulties w i t h this utilitarian n a t u r a l i s t i c
tivism itself, into w h i c h I have claimed t h a t all t h e usual forms of nat-
definition. T h e first of t h e s e is t h a t , a l t h o u g h it sits e a s i e r w i t h
uralism collapse. Relativism is not in my n a r r o w sense an ethical
Moore's open question a r g u m e n t t h a n the egoistic version, it is still
t h e o r y t h a t is, a t h e o r y about the m e a n i n g s of m o r a l words, or the
n o t very happy. It does not seem self-contradictory to give a negative
nature of the m o r a l concepts. I have said t h a t c e r t a i n ethical theories
a n s w e r to the question ' T h e act would m a x i m a l l y satisfy t h e prefer-
(naturalism in t h e present instance) collapse into relativism; b u t t h a t
ences of all affected parties, but is it right?' 1 have in my MT set o u t a
is b e c a u s e they try to incorporate into their e t h i c a l theories theses of
two-level version of u t i l i t a r i a n i s m w h i c h does. I hope, e n a b l e us to
s u b s t a n c e w h i c h d o n o t b e l o n g t h e r e . T h e n a t u r a l i s t , for e x a m p l e ,
avoid the counter-intuitive c o n s e q u e n c e s that are c l a i m e d to afflict
treats moral principles of s u b s t a n c e as if they were no m o r e t h a n lin-
t h e utilitarian; but this simple naturalistic version seems to me to r u n
guistic rules. As we saw, the m o m e n t we start treating t h e m as m o r a l
h e a d l o n g into t h e m . O n t h e face o f it, t h e r e a r e plenty o f a c t i o n s
principles o f s u b s t a n c e , w h i c h c o n d e m n o n e kind o f b e h a v i o u r a n d
w h i c h would m a x i m i z e preference-satisfaction in sum, b u t w h i c h it
c o m m e n d a n o t h e r , t h e naturalist gets c a u g h t in relativism, b e c a u s e
would not be self-contradictory to call wrong.

My o w n t h e o r y does n o t say t h a t 'right a c t i o n ' means ' a c t i o n
he is saying t h a t we have in a given culture to follow t h e rules of application of t h e m o r a l words in use in t h a t culture; a n d h e n c e , if t h e
which maximally satisfies preferences'. Rather, it explains the m e a n -
rules are m o r a l principles of s u b s t a n c e (as they a c t u a l l y are, in spite
ing of such words as 'right' and ' w r o n g ' and 'ought' as equivalent to
o f w h a t t h e n a t u r a l i s t s a y s ) , t h e m e m b e r s o f e a c h c u l t u r e will b e
v a r i o u s kinds of u n i v e r s a l i z a b l e prescriptions or p r o h i b i t i o n s , a n d
right in their m o r a l opinions, however m u c h these differ from culture
only arrives at a utilitarian m o r a l system by applying the logical prop-
to culture. B u t a l t h o u g h ethical theories c a n stray i n t o relativism by
erties of the words as so explained, in combination with c e r t a i n o t h e r
leaving the limits of ethical t h e o r y in the n a r r o w sense, relativism it-
conceptual theses, to the world as we actually have it, and in particu-
self, as I shall be using t h e term, is a substantial m o r a l thesis: it says
lar to a world in which people have certain preferences. My version of
t h a t w h a t e v e r a n y b o d y says is w r o n g is w r o n g , a n d t h e s a m e for
utilitarianism has therefore both a formal and a substantial e l e m e n t .
'right'. T h e r e are of c o u r s e o t h e r senses of t h e word 'relativism', b u t
T h e formal element is provided, in part, by the universal-prescriptive
this is the sense w h i c h I shall be using in w h a t follows.
definition of 'ought' and o t h e r m o r a l words; but that is n o t t h e only
A l t h o u g h , however, relativism does n o t c o m e i n t o o u r t a x o n o m y
element. T h e r e is also a substantial element which emerges in the ap-
of ethical theories, it itself needs some t a x o n o m y of its own, b e c a u s e
plication of this definition to the world. Prescriptivity plays an essen-
it h a s to be divided into species. T h e main division is into w h a t I shall
tial p a r t in this c o n s t r u c t i o n of a u t i l i t a r i a n system. S i n c e this
call 'cultural relativism' a n d w h a t I shall call 'individual relativism'.
prescriptivity is n o t available to a n a t u r a l i s t , he could n o t a r r i v e at
T h e first m a k e s w h a t is right and w r o n g relative to t h e opinions held
such a system. T h a t is why my own system cannot rightly be accused
in a given culture. T h e second makes it relative to the opinions of indi-
of being naturalistic (MT 1 2 . 6 ) .
viduals, even within cultures. However, t h e m a i n a r g u m e n t s against
B u t to explore further the question of h o w I myself avoid n a t u r a l -
relativism apply to b o t h species. T h e r e are, no doubt, m a n y practical
ism would be to digress from our present a r g u m e n t . I will c o n t e n t my-
reasons why we should not e m b r a c e relativism; but I shall be c o n -
self with saying that even t h e utilitarian form of naturalism, w h i c h I
c e r n e d m o r e w i t h t h e t h e o r e t i c a l troubles it gets i n t o . In p r a c t i c e , if
we were relativists, we should stop being able to say t h a t people w h o
8o
II. 4- 8
t h o u g h t that there was a duty to b u r n people whose religious views

they disagreed with were w r o n g to t h i n k this: we c o u l d n o t say this
even il' the people they were b u r n i n g were ourselves. I shall n o t say
m u c h about these obvious practical difficulties.
But this example brings out a theoretical difficulty as well. If I am
tied to the stake and being b u r n t , I have, according to relativism, to
say that t h e people w h o a r e b u r n i n g me do right to do t h i s , j u s t
II. 4. 8
NATURALISM
8i
I have given an e x a m p l e in w h i c h it is two cultures t h a t are in disa g r e e m e n t a b o u t the w r o n g n e s s of an action. B u t t h e s a m e difficulty

would obviously arise in an even m o r e a c u t e form if it were two individuals w h o disagreed. If m o r a l s t a t e m e n t s a r e prescriptivethat is.
if t h e intention of t h o s e w h o m a k e t h e m is t h a t we should a c t accordi n g l y t h e n t h e adoption o f relativism really w o u l d p r e v e n t m o r a l
l a n g u a g e b e i n g a ' m e a n s of c o m m u n i c a t i o n ' b e t w e e n people in dif-
b e c a u s e t h e y t h i n k il right. B u t 1 s h a l l also w a n t , on my o w n part
ferent c u l t u r e s ( a s , indeed, M a c l n t y r e h a s said i t does, 1 9 8 4 ; s e e H
(together with my coreligionists), to say t h a t they do w r o n g to b u r n
1 9 8 6 c ) . In the assumption that we do need to have moral language
me: and because I t h i n k this. I shall have to say. since I am a relativist,
for t h e p a r t i c u l a r kind of c o m m u n i c a t i o n t h a t we use it for (H 1 9 8 7 0 ) ,
that 1 am right to t h i n k it. 1 s h a l l t h e r e f o r e be saying t h a t t h e y a r e
I c o n c l u d e t h a t relativism h a s to be rejected, a n d therefore t h a t a n y
right to think it right, and also t h a t I am right to think it wrong. B u t
ethical theory, like naturalism, w h i c h collapses into relativism h a s to
according to the logic of the words 'right' a n d 'wrong' as we actually
be rejected too.
use t h e m , this is self-contradictory. For it is a logical property of t h e

word 'right' as ordinarily used t h a t I c a n n o t w i t h o u t c o n t r a d i c t i o n
say that two people w h o say, o n e t h a t an a c t is (all things considered)
right, and t h e o t h e r t h a t t h e s a m e a c t i s (all t h i n g s c o n s i d e r e d )
wrong, are both right.
As I said earlier, no h a r m m i g h t c o m e of this kind of relativism at
t h e t h e o r e t i c a l level. T h e r e l a t i v i s t would b e left s a y i n g s o m e t h i n g
w h i c h a c t u a l l y is s e l f - c o n t r a d i c t o r y if t h e words a r e used in t h e i r
ordinary senses: but he might retort that he is not m u c h conc e r n e d with o u r o r d i n a r y use of words, a n d he is r e c o m m e n d i n g a
n e w use a c c o r d i n g to w h i c h t h e s t a t e m e n t in q u e s t i o n is not selfcontradictory. He m i g h t have difficulty in saying just w h a t this n e w
use was: but that would be his business.
But he is doing m o r e t h a n r e c o m m e n d i n g a c h a n g e in l i n g u i s t i c
usage. He is implicitly still p r o p o s i n g to use t h e words ' r i g h t ' a n d
' w r o n g ' to c o m m e n d a n d c o n d e m n a c t i o n s . So he will be left h a v i n g
t o a g r e e b o t h with s o m e o n e (himself) w h o c o n d e m n s t h e b u r n i n g ,
a n d with s o m e o n e else ( t h e b u r n e r s ) w h o c o m m e n d it. B u t if t h e s e
are really substantial prescriptions for action (as in ordinary p a r l a n c e
they a r e ) , w h a t shall we m a k e of his statement? How shall we k n o w
what, if we a g r e e with it, we a r e e n j o i n e d to do? We a r e a p p a r e n t l y
both enjoined to burn me and enjoined not to burn me. And what
sort of prescription is that?
II. 5. 1
5
INTUITIONISM
5 . 1 . IN this chapter I have to deal with the second of the two possible
types of descriptivism. w h i c h for w a n t of a better n a m e I am calling
'inluitionism'. You will r e m e m b e r t h a t I distinguished this from natu r a l i s m , the o t h e r type of descriptivism. N a t u r a l i s m , I said, is t h e
view that the truth conditions of m o r a l statements, which according
to descriptivism determine their m e a n i n g , have to be the possession
by actions, people, etc. of n o n - m o r a l propertiesthat is, of properties
specifiable in morally n e u t r a l terms. By contrast, inluitionism is t h e
view t h a t they are t h e p o s s e s s i o n of specifically m o r a l , sui generis
properties which c a n n o t be defined without introducing some moral
term into the dejiniens.
This m e a n s that the intuitionist is faced with a difficulty w h i c h the
n a t u r a l i s t e s c a p e s . How is he to specify t h e s e properties, or t h e s e
truth conditions, if he is forbidden to do so in non-moral terms? To revert to the project of linguistic research which I described in the last
chapter: suppose that we are trying to determine what are the truth
conditions, even within a single culture using a single language, of a
m o r a l s t a t e m e n t . If we were t r y i n g to be naturalists, we could proceed as we do with o r d i n a r y n o n - m o r a l words. We could, t h a t is to
say, look and see to w h a t things people in the culture applied t h e m ,
and then say t h a t t h a t was their c o r r e c t application. T h i s would involve being able to recognize the things the words were applied to as
belonging to a determinate class. And the naturalist (at least the objectivist n a t u r a l i s t t h a t we considered in C h a p t e r 4) would have to
c l a i m t h a t this could be done objectively. We could, I argued, do this
with the word 'red' if t h e r e were a standard application of the word
in the language. But if we a r e intuitionists. h o w are we to recognize
the class of acts that a word like ' w r o n g ' is applied to by speakers of
the language? For on the face of it people speaking the s a m e l a n g u a g e
INTUITIONISM
83
apply the word ' w r o n g ' to different and indeed i n c o n s i s t e n t kinds of

acts, as we saw. I cited as an example t h a t some people call the a c t of
fighting for o n e ' s c o u n t r y wrong, w h e r e a s o t h e r s call this act right,
and the act of not fighting w h e n called on to do so wrong. This might
be true even of a p a r t i c u l a r a c t of fighting in m i n u t e l y specified circ u m s t a n c e s on whose description the parties were agreed. T h e naturalist c a n make an attempt (unsuccessful, as we saw) to get out of this
difficulty by suggesting t h a t the word ' w r o n g ' is being used by these
people in different senses, b e c a u s e the applications a r e different. But
what is the intuitionist to do?
It s e e m s t h a t , s i n c e he is n o t allowed to appeal to t h e o b s e r v a b l e
n o n - m o r a l properties of objects, there is n o t h i n g he c a n do except to
say t h a t t h e r e s e a r c h e r j u s t c a n r e c o g n i z e t h e c l a s s o f a c t i o n s t h a t
people call wrong. How is he to recognize t h e m ? Only, it would seem,
by having the ability or capacity to recognize them. Modern intuitionists often deny t h a t they are committed to any special 'faculty of intuition'; they deny t h a t they are moral sense theoristsand we c a n well
u n d e r s t a n d why they deny this; for s u c h a m o r a l s e n s e is obviously
suspect. And it m a y be true, since the word 'intuitionist' is used in so
m a n y different s e n s e s t h e s e days, t h a t s o m e i n t u i t i o n i s t s , i n s o m e
senses of t h e word, c a n avoid positing s u c h a faculty. B u t if so, they
will have to explain how a linguistic r e s e a r c h e r could d e t e r m i n e the
truth conditions of m o r a l s t a t e m e n t s . At any rate, it should be c l e a r
by now t h a t an intuitionist, in the sense in w h i c h I am using the t e r m
( t h a t is, a descriptivist w h o says that, s y n t a x a p a r t , t h e m e a n i n g of
moral statements is wholly determined by their truth conditions, and
that the truth c o n d i t i o n s c a n n o t be specified in t e r m s of n o n - m o r a l
properties)that an intuitionist, so defined, c a n n o t do without s u c h
a faculty. For unless a linguistic r e s e a r c h e r possessed s u c h a faculty,
the class of a c t i o n s t h a t people called w r o n g would be quite indeterm i n a t e , a n d so his r e s e a r c h would be quite i n c o n c l u s i v e : t h e t r u t h
conditions of m o r a l statements might be almost anything.
5 . 2 I n o w w a n t t o c o n s i d e r t h e v i e w w h i c h i s a t first s i g h t
plausiblethat t h e r e is indeed s u c h a f a c u l t y t h a t is, t h a t we c a n ,
most of us, recognize actions t h a t are wrong and a c t i o n s t h a t are not.
Let us take a very obvious example. I have j u s t filled up at a self-ser-
II. 5- 2
vice petrol (gas) station t h a t h a s n o a u t o m a t i c m a c h i n e t o e x a c t t h e

c a s h before one fills up, and am wondering whether to go and pay the
c a s h i e r for my petrol, or j u s t to drive away w i t h o u t paying. T h e
c a s h i e r is not looking, n o r is a n y b o d y else. If I am like m o s t people,
w h e n I c o n t e m p l a t e doing this, I get a quite easily recognizable e x perience. Let us call it the t h o u g h t (even the conviction) t h a t it would
be wrong to do it. So here, at a n y rate, we seem to have a clear c a s e of
recognizing a (proposed) act as wrong. So an intuitionist might claim
t h a t there is this faculty by w h i c h we c a n recognize wrong acts.
W i t h this, t a k e n in o n e s e n s e , few will disagree. M o r a l p h i l o s o phers of all persuasionsbe they descriptivists or non-descriptivists,
objectivists or subjectivists, even emotivistswill at o n c e recognize
s o m e t h i n g that is going on in the mind of the person in this predicam e n t . I c e r t a i n l y r e c o g n i z e it myself. T h e y will call it by different
n a m e s . An intuitionist will call it the t h o u g h t or conviction t h a t t h e
act would be wrong. A subjectivist naturalist is likely to call it a feeling
of disapproval of the proposed act. An emotivist will probably use t h e
s a m e expression; and I myself see no h a r m in using it. So it looks as if
they all agree t h a t this experience o c c u r s , and disagree only in w h a t
they call it. Is there then only a verbal difference between w h a t the intuitionists say about the experience, and w h a t the others say?
If there is no experiential difference between the feeling or t h o u g h t
t h a t the person h a s according to the intuitionists, and that w h i c h he
has according to, say, an emotivist, w h a t other difference could there
be? We might suggest t h a t there is a logical differencethat is, a difference in the logical properties attributed by these different thinkers
to the statement t h a t this feeling or t h o u g h t (whichever it is) o c c u r s ,
and to the m o r a l s t a t e m e n t w h i c h is m a d e on the ground t h a t it o c curs (namely the statement t h a t the a c t would be w r o n g ) . To this suggestion we shall recur later in this chapter. B u t for the m o m e n t I w a n t
t o point out t h a t t h e r e a r e o t h e r m o r a l s i t u a t i o n s w h i c h a r e m u c h
m o r e difficult for the intuitionist. In the petrol station case, we all, or
n e a r l y all. a g r e e t h a t t h e a c t would b e w r o n g . T h e r e m a y b e s o m e
r e p r o b a t e s w h o do n o t a g r e e t h a t it would be w r o n g ; b u t t h e y a r e
probably few e n o u g h for us to i g n o r e t h e m , j u s t as, in studying t h e
s t a n d a r d use of an o r d i n a r y descriptive word like 'red', we i g n o r e
II. 5. 2
INTUITIONISM
85
those few w h o u s e it incorrectly. B u t all c a s e s a r e n o t like t h e petrol

station case.
T h e intuitionist is on the firmest ground in c a s e s w h e r e we nearly
all agree. B u t in m a n y c a s e s (those which give us trouble, a n d w h i c h
m o r a l philosophers o u g h t to be helping us with) t h e r e is no s u c h general agreement. W h e n we are thinking about fighting for o n e ' s c o u n try, o r a b o u t a b o r t i o n , o r a b o u t e a t i n g m e a t , s o m e o f u s h a v e t h e
r e c o g n i z a b l e t h o u g h t o r feeling a n d s o m e d o n o t . I t w o u l d n o t b e
m u c h of a victory for the intuitionist to show that, in c a s e s where we
all t h i n k w e k n o w t h e a n s w e r s t o m o r a l q u e s t i o n s , w e all h a v e t h e
s a m e t h o u g h t o r feeling, i f i n m a n y o t h e r c a s e s a n d t h o s e w h i c h
c a u s e u s m o s t t r o u b l e i n o u r m o r a l t h i n k i n g w e h a v e different
t h o u g h t s o r feelings. F o r i t i s t h e l a t t e r c l a s s o f c a s e s i n w h i c h w e
really n e e d t o b e told t h e t r u t h c o n d i t i o n s o f m o r a l s t a t e m e n t s , s o
t h a t we c a n find out w h i c h of t h e m are true.
Reverting, t h e n , to o u r linguistic r e s e a r c h e r : it looks as if t h e r e is
n o t h i n g he c a n do, if he wishes to l e a r n to recognize the class of a c tions t h a t are being called wrong, except to employ his own faculty of
m o r a l intuition. In a w a y this m a k e s his t a s k easier. I n s t e a d of, as a
c o n s c i e n t i o u s n a t u r a l i s t would have to, laboriously c a t a l o g u i n g , in
n o n - m o r a l t e r m s , t h e classes of a c t i o n s t h a t other people call w r o n g ,
he c a n perhaps forget a b o u t other people, and j u s t look at the actions
t h a t he h i m s e l f calls w r o n g . For t h e object of his r e s e a r c h is to give
the truth conditions of m o r a l statements. S i n c e he h a s this faculty, he
k n o w s already w h a t t h e truth conditions are. T h e t r u t h condition o f
a statement t h a t an a c t is wrong is t h a t it should produce in h i m this
recognizable r e a c t i o n . Can he then stop studying w h a t o t h e r people
say?
5 . 3 . T h e i n t u i t i o n i s t m i g h t reply t h a t this i s g o i n g t o o fast.
Suppose, he m i g h t say, t h a t I discover t h a t m o s t o t h e r people apply
the word 'wrong' to s o m e kind of a c t to w h i c h I do n o t apply it. S h a l l I
n o t t h e n begin to t h i n k t h a t my faculty of m o r a l intuition is faulty?
S h a l l I n o t even b e g i n to change my perception of w h a t is r i g h t a n d
w r o n g , s o a s t o c o n f o r m w i t h the o t h e r s , a t least i f t h e y a r e people
w h o m in general I respect? So, a l t h o u g h I go on saying t h a t t h e truth
conditions of m o r a l statements are t h a t acts, etc. should be perceived
86
IL 5- I
II. 5- 3
INTUITIONISM
87
by me as right or w r o n g , what is perceived by me as right or w r o n g
only t h a t there are i m p o r t a n t elements t h a t are c o m m o n to t h e moral-
will c h a n g e , to b e c o m e m o r e like w h a t other people call right or wrong.
ities of different c u l t u r e s . T h a t m a y well be t r u e : m o s t c u l t u r e s c o n -
This process is even clearer in t h e moral education of children, who,
d e m n murder, for e x a m p l e ( t h o u g h w h a t t h e y c o u n t a s m u r d e r
let us suppose, start off w i t h o u t any m o r a l opinions or perceptions,
varies). However, t h e intuitionist e x p l a n a t i o n o f t h e s e c o m m o n ele-
and w h o therefore have to get all their m o r a l opinions initially from
ments is not t h e only possible one. It m a y be t h e c a s e (I think it is) t h a t
o t h e r people, their elders or later their c o n t e m p o r a r i e s , w h o m t h e y
t h e e x i s t e n c e o f a c o m m o n m o r a l ' g r a m m a r ' o r ' l o g i c ' h a s led t o
respect. This is indeed, we m a y agree, h o w the mores of a given c u l -
m o r a l t h i n k e r s i n all c u l t u r e s c o m i n g t o t h e s a m e c o n c l u s i o n s . T h i s
ture b e c o m e to some degree h o m o g e n e o u s .
would be consistent with my own a c c o u n t of m o r a l thinking w h i c h I
S o m e intuitionists m i g h t dissent from this, and c l a i m t h a t t h e r e
shall be s u m m a r i z i n g later. It by no m e a n s establishes t h a t there a r e
are moral opinions which are i n n a t e , or which at least develop inde-
c o m m o n m o r a l perceptions w h i c h all cultures s h a r e . T h e c o m m o n el-
pendently of the moral opinions of o t h e r people. If so. it is pretty ob-
e m e n t s m i g h t be arrived at by reason and n o t by intuition.
vious t h a t different m o r a l o p i n i o n s a r e i n n a t e in t h e m e m b e r s of
5 . 4 . I return then to our programme of linguistic research into
different c u l t u r e s , a t least o n m a t t e r s w h i c h are disputed b e t w e e n
the truth conditions of moral statements, or the application condi-
t h o s e c u l t u r e s . T h i s would be h a r d to explain genetically. However,
tions of m o r a l predicates. T h e intuitionist view, stated baldly, is t h a t
w e m i g h t a g r e e a t least t h a t t h e r e i s a n i n n a t e disposition t o t h i n k
w r o n g n e s s , for e x a m p l e , is a c o m m o n property s h a r e d by m a n y a c -
morallya disposition which does not determine the content of a per-
tions, w h i c h is discernible by those w h o have t h e n e c e s s a r y power of
son's morality, but at least partly determines its form. This would tally
discernment. We considered the objection t h a t if this were so, I could
with Chomsky's view that there are 'universals' (as they are called) of
simply ignore t h e opinions of other people, and rely on my o w n power
l a n g u a g e w h i c h are genetically determined and c o m m o n t o all c u l -
of discernment. This objection we answered on behalf of the intu-
t u r e s ( 1 9 6 5 : 3 5 ) . A c c o r d i n g to this s u g g e s t i o n , there is a c o m m o n
itionist by suggesting t h a t even if we t h o u g h t we h a d this power, we
s t r u c t u r e of moral l a n g u a g e , with its g r a m m a r and its logic, w h i c h
m i g h t c o m e t o d o u b t its reliability i f o u r m o r a l o p i n i o n s c o n f l i c t e d
we a r e all g e n e t i c a l l y disposed to l e a r n , and therefore l e a r n m o r e
with those of o t h e r people. This might, we suggested, lead us to change
easily t h a n if we had no such g e n e t i c predisposition. S u c h a view is
our perceptions of right and wrong to conform to those of other
c o n s i s t e n t with my o w n , b u l I e x p r e s s no opinion on w h e t h e r it is
people t h a t we respected.
right; t h a t would have to be d e t e r m i n e d by empirical r e s e a r c h . T h e
B u t now look where this is leading the intuitionist, if he takes this
i m p o r t a n t thing is that even if the form of morality is innate, t h a t is
line. It t u r n s o u t t h a t we h a v e to rely, n o t on o u r o w n power of dis-
c o n s i s t e n t with morality h a v i n g very different contents in different
c e r n m e n t , but on a c o n s e n s u s between t h e deliverances of t h e i n t u -
c u l t u r e s , just as t h e view t h a t g r a m m a r and even logic is i n n a t e is
itions of people w h o m we respect. We shall presumably respect those
consistent with t h e m e m b e r s of those cultures having very different
people whose m o r a l opinions in general we s h a r e . We shall be willing
factual opinions about w h a t goes on in e a r t h or heaven.
to adapt particular o p i n i o n s to c o n f o r m to t h e o t h e r s : b u t if a n y o n e
For this reason an intuitionist could not draw m u c h comfort from
had m o r a l opinions differing radically and over a wide a r e a of moral-
t h e existence, if it did exist, of a c o m m o n or (in the linguists' s e n s e )
ity from o u r o w n , we should be unlikely to respect h i m (or h e r ) . T h e
u n i v e r s a l m o r a l l a n g u a g e w i t h its logip. For he w a n t s to c l a i m n o t
intuitionist s e e m s likely to h a v e to say t h a t o u r s o u r c e for t h e t r u t h
only that the form of morality is c o m m o n between cultures, but t h a t
c o n d i t i o n s o f m o r a l s t a t e m e n t s i s n o t o u r o w n individual faculty o f
its c o n t e n t is. T h i s is in a n y c a s e a highly implausible claim. Perhaps
r e c o g n i t i o n o f m o r a l properties, b u t r a t h e r a c o n s e n s u s o f like-
t h e intuitionist would be willing to draw in his h o r n s a bit, and c l a i m
minded people in recognizing t h e m .
88
II-5-4
INTUITIONISM
II. 5. 4
89
T h a t intuitionists might be led to take this line is suggested by w h a t
indignantly m a i n t a i n t h a t they have been well educated. How c a n we
h a s frequently h a p p e n e d i n a r g u m e n t s b e t w e e n t h e m a n d t h e i r
adjudicate b e t w e e n t h e m ? Obviously n o t by c a l l i n g in further intu-
opponents. T h e r e is a well-worn a r g u m e n t against intuitionism called
itions about w h a t is a good education.
' t h e a r g u m e n t from m o r a l d i s a g r e e m e n t ' . T h e a r g u m e n t goes like
You will now see, I hope, why I said in Chapter 4 t h a t intuitionism,
this. T h e r e undoubtedly are c a s e s in w h i c h people's m o r a l o p i n i o n s
like n a t u r a l i s m , collapses inevitably into relativism. T h e point is that
disagree. Therefore the two parties c a n n o t both be right. S o , if m o r a l
the general c o n s e n s u s on moral questions w h i c h is likely to exist in a
i n t u i t i o n s are a reliable s o u r c e of m o r a l t r u t h , o n e or o t h e r p a r t y
given c u l t u r e is t h e result of a c o m m o n m o r a l e d u c a t i o n . In closed
must lack this source. T h a t party, therefore, has an intuition which is
a n d morally h o m o g e n e o u s c u l t u r e s this c o n s e n s u s is likely to cover
faulty (or, if you refuse to call it an intuition if it is faulty, does n o t
all or n e a r l y all m o r a l q u e s t i o n s . B u t even in a pluralist s o c i e t y like
h a v e t h e power o f i n t u i t i o n i t does n o t m a t t e r for t h e a r g u m e n t
o u r o w n it is likely to cover a g r e a t m a n y q u e s t i o n s , s o m e of t h e m
which way we put it). But t h e intuitionist h a s given us no way of de-
f u n d a m e n t a l . However, if any of t h e s e c o m m o n m o r a l o p i n i o n s is
termining which of the two parties h a s a faulty intuition. It would be
challenged by a m o r a l reformer, it is no use appealing to t h e c o n s e n -
obviously and viciously circular to try to settle the matter by having a
sus itself to v a l i d a t e t h e o p i n i o n s . If I m a y q u o t e a telling p a s s a g e
further intuition t h a t o n e of t h e parties h a s a reliable intuition a n d
from Dryden:
the other not; for this further intuition could in turn be challenged by
t h e party w h o h a d been put down by it. And it would be no better to
call in t h e intuitions of third or fourth parties, for they too could be
c h a l l e n g e d in t h e s a m e way. S o , t h e objection goes on, i n t u i t i o n i s m
By education most have been misled;

So they believe, because they were so bred.
The priest continues what the nurse began:
And thus the child imposes on the man. (1637, pt. 3, 389)
will yield no determinate answer to disputed moral questions.

It h a s been c o m m o n for intuitionists to say, in reply to this o b j e c -
I n t u i t i o n s a r e relative to c u l t u r e s . As I h a v e said, I do n o t deny for a
tion, t h a t not all intuitions (or supposed intuitions) are reliable, b u t
m o m e n t t h a t i n t u i t i o n s will be found w h i c h a r e c o m m o n to m o s t or
only those of 'thoughtful and well-educated people' (W. 1). Ross 1930:
even to nearly all cultures, like t h a t forbidding m u r d e r ( t h o u g h , as I
4 1 ) . We should follow their intuitions in preference to those of people
also said, m u r d e r is n o t defined in the s a m e way in all c u l t u r e s ) . B u t
w h o have not been well educated. B u t recall what I said earlier about
e v e n if t h i s is so, if a n y b o d y were to c h a l l e n g e t h i s c o n s e n s u s , we
t h e relation between i n t u i t i o n s a n d m o r a l e d u c a t i o n . It is perfectly
could not rule h i m out of c o u r t by appealing to t h e c o n s e n s u s . True,
true that people's intuitions (that is their moral convictions) will vary
m o s t people h a v e t h e i n t u i t i o n s , a n d w e say t h a t t h o s e w h o d o n o t
according to how they have been educated. B u t what are we claiming
have t h e m were not well brought up. B u t we say this only b e c a u s e we
w h e n we c l a i m t h a t only t h e intuitions of well-educated people a r e
ourselves h a v e been b r o u g h t up in the way t h a t we have been. If we
reliable? I said t h a t we are likely to m a k e our own m o r a l opinions or
had been b r o u g h t up in a different way, we m i g h t h a v e a g r e e d with
perceptions c o n f o r m t o t h o s e o f people w h o m w e respect, a n d t h a t
t h e dissident. P e r h a p s , if he is successful in his m o r a l reform, future
this w a s especially evident in t h e m o r a l education of children, w h o
generations m a y be brought up in his way r a t h e r t h a n in ours. This is
start, we supposed, w i t h no d e t e r m i n a t e m o r a l o p i n i o n s . If t h i s is
unlikely to happen with murder, b e c a u s e t h e r e are good reasons (not
true, and if the intuitionist says we should respect only t h e m o r a l in-
based on i n t u i t i o n ) for c o n d e m n i n g murder. How we s h o u l d r e a s o n
tuitions of well-educated people, a n o t h e r obvious circularity appears
about such questions, I shall be explaining later. B u t t h e good reasons
in his a r g u m e n t . For w h o is to c o u n t as well educated? Suppose that,
do not consist in t h e fact t h a t there is a c o n s e n s u s . A n d w h e n we c o m e
as may well happen in a dispute about, say. meat-eating, both parties
to fighting or to m e a t - e a t i n g or to abortion, t h e r e is no c o n s e n s u s to
9(>
A T A X O N O M Y OF KTHICAL T H E O R I E S
H- 5- 4
II-
5-
INTUITIONISM
91
appeal to, and we have to find a w a y of r e a c h i n g one by a r g u m e n t ,
istic n a t u r a l i s m a n d w h e n I was discussing i n t u i t i o n i s m t h e fact of
n o t by i n t u i t i o n . I n t u i t i o n by itself is no prophylactic a g a i n s t r e l a -
m o r a l d i s a g r e e m e n t played a c r u c i a l role in my a r g u m e n t . T h i s will
tivism.
be the c a s e with subjectivism too. T h e r e c u r r e n c e of this fact of moral
5 . 5 . I a m n o w going o n . a s I promised, t o discuss t h e variety o f
d i s a g r e e m e n t in a r g u m e n t s against all these theories is no accident.
n a t u r a l i s m called s u b j e c t i v i s m . We h a v e n o w r e a c h e d a p o i n t at
B e c a u s e they are all in essence relativistic theories, it is inevitable t h a t
w h i c h it c a n easily be m a d e c l e a r h o w very close i n t u i t i o n i s m is to
t h e fact of m o r a l d i s a g r e e m e n t should play a p a r t in a r g u m e n t s to
subjectivism. Indeed, we c a n see t h a t w h e n its pretended objectivist
show this. I shall be discussing shortly t h e relation between t h e roles
trappings have been stripped off, intuitionism is a kind of subjectiv-
of moral d i s a g r e e m e n t in a r g u m e n t s against these different theories.
ism. It is not surprising, therefore, t h a t intuitionism collapses into rel-
B u t its role in a r g u m e n t s against subjectivism, at least, is familiar. It is
ativism. I said in t h e p r e c e d i n g c h a p t e r t h a t relativism is n o t an
this. If I say t h a t s o m e act is w r o n g and you say t h a t it is n o t wrong,
e t h i c a l t h e o r y in my n a r r o w s e n s e , b e c a u s e it is a b o u t m a t t e r s of
then, according to subjectivism, I am m a k i n g a s t a t e m e n t of psycho-
m o r a l s u b s t a n c e ( a b o u t w h a t w e o u g h t t o do) and n o t a b o u t w h a t
logical fact a b o u t my own m e n t a l state or attitude, and you are m a k -
moral words m e a n . Subjectivism is a theory about what m o r a l words
ing a s t a t e m e n t o f p s y c h o l o g i c a l fact a b o u t y o u r s . B u t t h e s e
m e a n . Its relation to relativism is that it makes relativism analytically
statements are quite consistent with e a c h other; w h e r e a s t h e original
true. Subjectivism is the view t h a t w h e n I am m a k i n g a m o r a l state-
s t a t e m e n t s t h a t t h e a c t i s w r o n g , and t h a t t h e a c t i s n o t w r o n g , a r e
ment. I am saying simply that, as a matter of psychological fact, I (the
n o t c o n s i s t e n t w i t h e a c h other. T h e s u b j e c t i v i s t m u s t t h e r e f o r e b e
speaker) approve or disapprove of s o m e act or person. T h e r e is an al-
mistaken about w h a t t h e statements m e a n .
ternative version according to w h i c h w h a t 1 am saying is t h a t people
This a r g u m e n t is so familiar, going b a c k to Moore ( 1 9 1 2 : c h . 3) and
in my society approve or disapprove. T h i s version I shall leave aside for
indeed to Sidgwick and to s o m e older moralists, t h a t I do n o t need to
the present, although similar objections can be made to it.
dwell on it. It was S t e v e n s o n ' s attempt to avoid this objection to sub-
S u b j e c t i v i s m is a form of n a t u r a l i s m , b e c a u s e it gives t h e t r u t h
j e c t i v i s m t h a t led h i m t o his variety o f n o n - d e s c r i p t i v i s m h e said
conditions of moral statements without introducing any moral terms
t h a t there was a disagreement in attitude t h o u g h n o t in belief ( 1 9 4 2 ,
into the defmiens. T h e statement that as a matter of psychological fact
1945: 3 ) . It is a rather elementary mistake to think that the same
I approve of an act c o n t a i n s no moral terms. It is an empirical state-
a r g u m e n t c a n be turned against non-descriptivism. T h i s is part of t h e
m e n t verifiable e i t h e r by i n t r o s p e c t i o n or by o b s e r v a t i o n of my
general confusion, still too c o m m o n , between subjectivism and n o n -
b e h a v i o u r . It is i m p o r t a n t not to be confused here. To approve of
descriptivism; but t h a t will have to be left until later.
s o m e t h i n g may be to have a moral opinion. But the statement that I
5 . 6 . For the present I w a n t just to draw your attention to the dif-
approve is n o t itself a m o r a l s t a t e m e n t , even if m a d e by myself.
ferent ways in which the fact of moral disagreement figures in the ar-
S o m e o n e , or even I myself, could describe my moral opinions, feelings,
g u m e n t s a g a i n s t objectivistic n a t u r a l i s m , subjectivistic n a t u r a l i s m ,
attitudes, etc. without saying a n y t h i n g moral. It is confusion on this
and intuitionism respectively. In t h e c a s e of objectivistic n a t u r a l i s m
point that h a s m a d e m a n y people mix up c e r t a i n forms of n o n -
t h e crucial point was that, since people disagree in their m o r a l opin-
descriptivism with subjectivism, as we shall see. Subjectivism, in t h e
ions, any attempt to establish a single set of truth conditions for moral
s e n s e I am using, is o n e kind of n a t u r a l i s t i c descriptivism; it c a n
s t a t e m e n t s by looking at w h a t a c t i o n s , e t c . t h e y apply m o r a l predic-
therefore not be any kind of non-descriptivism. It falls on t h e opposite
ates to will founder on this d i s a g r e e m e n t . We simply s h a l l n o t get a
side from non-descriptivism of t h e m a i n divide between ethical t h e o r -
c o n s i s t e n t set of t r u t h conditions. At best, we shall be left giving dif-
ies t h a t I set out in C h a p t e r 3. B o t h w h e n I was discussing o b j e c t i v -
ferent sets of t r u t h conditions for different c u l t u r e s . If t h e naturalist
92
II. 5. 6
tries to escape this objection by saying t h a t the words m e a n different

t h i n g s w h e n used in different c u l t u r e s (or even by different individuals within a single culture), then he will be open to a very similar
o b j e c t i o n t o t h a t w h i c h w e have j u s t b e e n m a k i n g a g a i n s t t h e s u b jectivist: he will be admitting t h a t the words do not m e a n the s a m e in
the m o u t h s of these opposing parties, and that therefore their m o r a l
statements do not contradict o n e another, which obviously they do.
In the case of intuitionism, the trouble is basically the same. If we
think that we c a n establish the truth conditions of moral s t a t e m e n t s
by appeal to a faculty of m o r a l i n t u i t i o n b y saying, t h a t is. t h a t
t h o s e m o r a l s t a t e m e n t s a r e t r u e w h i c h a r e certified as t r u e by this
f a c u l t y t h e n we shall again get different a n s w e r s , a c c o r d i n g to
whose intuitions we e x a m i n e . T r u e , there is a difference between the
c a s e s of objectivist naturalism and intuitionism. In the c a s e of intuitionism. there are not different sets of truth conditions. T h e r e is only
o n e set. t h a t is. c o n f o r m i t y w i t h i n t u i t i o n . B u t since the i n t u i t i o n s
themselves conflict, we shall still get variation in the truth values of
particular moral statements, depending on who is doing the judging.
So the end result is the same, n a m e l y relativism. Subjectivism h a s the
v i r t u e of displaying this fault, w h i c h all t h e s e t h e o r i e s c o m m i t , in
stark clarity. A c c o r d i n g to it, the person w h o m a k e s a m o r a l s t a t e m e n t is simply reporting on his own psychological slate. This, like intuitionism. yields a single set of truth conditions: if the psychological
state is that which in fact the speaker has. then the statement is true.
B u t a n d this too is like i n t u i t i o n i s m w e shall get no c o n s i s t e n t
a n s w e r about the truth of p a r t i c u l a r moral statements, b e c a u s e the
answers will depend on w h o m we ask and what altitude he has.
5 . 7 . 1 said that it was no accident t h a t all these theories are in the
s a m e trouble, or at least in closely related t r o u b l e s . I w a n t n o w to
bring this out even m o r e clearly if I c a n , by examining the cases of int u i t i o n i s m and s u b j e c t i v i s m , and s h o w i n g how similar they really
are. This will be repugnant to intuitionists, w h o often think of themselves as objectivists. indeed as model or paradigm objectivists. B u t
we c a n bring out the similarity by asking w h a t intuitionists think the
difference is between having an intuition and having a feeling or attitude of approval or disapproval. I gave an example earlier of a person
II- 5- 7
INTUITIONISM
93
w h o was thinking of driving away from a petrol station without paying. W h a t , I asked, is the difference, o t h e r t h a n a merely verbal one,
between w h a t intuitionists say about this situation and w h a t subjectivists say? Certainly t h e experience t h a t t h e y a r e b o t h a t t r i b u t i n g to
the person is the s a m e . It is the having of an attitude of disapproval,
or the having of a conviction t h a t the act would be wrong; b u t w h a t
is the difference? I c a n see no objection to saying t h a t those w h o have
moral convictions have moral attitudes, and vice versa. It o u g h t to be
a g r e e d o n all sides t h a t t h e p e r s o n i n m y e x a m p l e h a s b o t h t h e s e
things, or r a t h e r t h a t they are the s a m e thing.
T h e i n t u i t i o n i s t s a r e relying, in order to give us k n o w l e d g e of
moral truths, on a certain experience, which they call the having of a
m o r a l intuition. B u t the trouble i s t h a t s u c h e x p e r i e n c e s a r e s o m e t h i n g subjective. If I have this e x p e r i e n c e , t h e n I have it; t h e r e is a b solutely n o t h i n g t h a t c a n be appealed to, outside the experience itself,
w h i c h could s h o w w h e t h e r it was really so or n o t . If I have this e x perience. I c a n n o t be mistaken in thinking t h a t I have it. This, indeed,
is the a t t r a c t i o n , in o n e way, of the intuitionist theory, j u s t as it was
the attraction of the sense-datum theories t h a t used to be so popular
in epistemology. Here is s o m e t h i n g t h a t c a n n o t be disputed: I have
the experience called ' a n intuition t h a t a c e r t a i n a c t would be wrong',
and t h a t is all t h e r e is to be said. W h a t e v e r m a y h a p p e n to anybody
else, I have this e x p e r i e n c e , and, on t h e s t r e n g t h of it, a c c o r d i n g to
the intuitionists. I am entitled to say t h a t the act would be wrong.
But for this indisputability of t h e i n t u i t i o n we pay t o o h e a v y a
price. For i f n o t h i n g outside t h e e x p e r i e n c e c a n c o u n t a g a i n s t t h e
existence of the intuitionif, t h a t is to say, the m e r e having of the experience is t h e g u a r a n t e e t h a t it e x i s t s t h e n , by t h e s a m e token, it
c a n n o t tell us a b o u t a n y t h i n g outside the a c t u a l experience. All t h a t
we c a n be c e r t a i n of, by having this experience, is t h a t we have it.
W h a t this c o m e s to is that, for all the sound a n d fury t h a t went on
in the battle between the people w h o called themselves 'objectivists'
(that is, the intuitionists) and those w h o m they called 'subjectivists',
t h e r e was n o t h i n g o f s u b s t a n c e t h a t really divided t h e m . T h e intuitionists thought that, a c c o r d i n g to their theory, to say t h a t an a c t is
w r o n g was n o t j u s t to report on a subjective fact. B u t they were n o t
94
II. 5- 7
justified in saying this. For according to their theory the m e r e o c c u r r e n c e o f t h e e x p e r i e n c e w h i c h t h e y called a n intuition ( a n d w h i c h
II- 5- 7
INTUITIONISM
95
implying t h a t anybody who says otherwise is n o t right, t h a t is, is mistaken.
t h e subjectivists called a feeling or attitude of disapproval) is the guar-
We c a n see, therefore, that the alleged self-guaranteeingness of in-
a n t e e that a c e r t a i n m o r a l s t a t e m e n t is true. T h a t is the t r u t h condi-
tuition is really incompatible with its alleged objectivity. B u t t h e self-
tion o f t h e m o r a l s t a t e m e n t . B u t i f so. t h e n t h e m o r a l s t a t e m e n t
g u a r a n t e e i n g n e s s is w h a t is really c h a r a c t e r i s t i c of intuition. T h e r e
c a n n o t say any m o r e t h a n t h a t t h e experience o c c u r s . If the m e r e o c -
a r e o t h e r kinds of o b j e c t i v i s m . I h a v e a l r e a d y talked a b o u t o b j e c t -
c u r r e n c e o f t h e e x p e r i e n c e g u a r a n t e e s t h e truth o f t h e m o r a l state-
ivistic n a t u r a l i s m , a n d w h e n I c o m e o n i n l a t e r c h a p t e r s t o n o n -
m e n t , then there c a n n o t b e a n y m o r e t o m a k i n g the s t a t e m e n t t h a n
descriptivism. I shall s h o w t h a t o n e c a n have an objectivist but n o n -
there is to saying that the experience o c c u r s .
descriptivist t h e o r y (my o w n ) which m a i n t a i n s , at least, t h a t of two
T h e intuitionists did n o t see this, b e c a u s e t h e y w a n t e d t o m a i n -
people w h o disagree a b o u t a m o r a l s t a t e m e n t b o t h c a n n o t be right,
t a i n , at t h e s a m e t i m e as t h e position I h a v e just b e e n d i s c u s s i n g
or t h a t it is impossible consistently to agree with b o t h . S i n c e t h e r e are
( n a m e l y that the mere o c c u r r e n c e o f t h e experience g u a r a n t e e s t h e
non-intuitionist but still objectivist theories, it m u s t be t h e self-guar-
t r u t h of t h e m o r a l s t a t e m e n t ) , a n o t h e r position w h i c h is really in-
a n t e e i n g n e s s o f m o r a l intuitions t h a t intuitionists h a v e t o h a n g o n
compatible with it. T h e y wanted to m a i n t a i n , as well, t h a t t h e m o r a l
to, if t h e y a r e going to r e t a i n w h a t is e s s e n t i a l in i n t u i t i o n i s m . B u t
statement was 'objective'. W h a t this m e a n t . I am not sure; but in this
since this is i n c o m p a t i b l e with their objectivism, t h e y have, in order
c o n t e x t to call a moral s t a t e m e n t objective entails at least t h e follow-
to be consistent, to give up the objectivism. And this m e a n s , as I said,
ing: t h a t if two people make, o n e of t h e m a certain moral s t a t e m e n t ,
that the only way in which intuitionists c a n r e m a i n intuitionists and
and the other its negation (for example, that an act is wrong, a n d t h a t
avoid self-contradiction is to e m b r a c e s o m e form of subjectivism.
it is not w r o n g ) , t h e n they c a n n o t b o t h be right. It is impossible c o n -
5 . 8 . T h i s i s perhaps a n appropriate place t o m e n t i o n very briefly
sistently to agree with both their statements. I hope you will note t h a t
a topic w h i c h I have dealt with at length e l s e w h e r e (H 1 9 5 5 0 , 1 9 9 4 b ,
in this sense I am myself an objectivist.
FR 2.2 ff.) a n d w h i c h needs to be cleared up if we a r e to have an ade-
I n t u i t i o n i s t s . in fact, very frequently insisted on this, t h a t b o t h
q u a t e t a x o n o m y of e t h i c a l t h e o r i e s . It c o n c e r n s a division b e t w e e n
c a n n o t be right. But n o w you c a n see t h a t this is i n c o n s i s t e n t w i t h
ethical theories w h i c h cuts across the classification we have been out-
the other thesis they maintained, n a m e l y that the mere o c c u r r e n c e of
lining. T h i s m e a n s that, a l t h o u g h it does affect intuitionism, it also af-
t h e e x p e r i e n c e g u a r a n t e e s t h e t r u t h o f t h e m o r a l s t a t e m e n t . For i f
fects a n u m b e r of o t h e r e t h i c a l theories, both descriptivist a n d n o n -
this latter thesis is true. t h e n , as we have seen, the person w h o h a s
descriptivist. T h i s is t h e division between w h a t I shall call particularist
t h e experience, and says, accordingly, t h a t an act is wrong, is really
and w h a t I shall call universahst theories. It is very easily illustrated by
saying no more than that he has the experience; and he c a n n o t be
talking of intuitionists in particular. A b o u t what do t h e y t h i n k t h a t
m i s t a k e n a b o u t that. If he w e r e s a y i n g m o r e , t h e n t h e m e r e o c c u r -
moral s t a t e m e n t s have, in the first instance, to be made? T h a t is, w h a t
r e n c e o f the experience could n o t g u a r a n t e e t h e truth o f this m o r e .
are the objects of m o r a l intuitions? We find a m o n g intuitionist writ-
However, on the 'objectivist' thesis just mentioned, the speaker m u s t
ers s o m e w h o stress t h e need to assess particular, t h a t is individual,
be saying more. For he is saying at least that, if anybody else t h i n k s
actions morally, b u t o t h e r s w h o think t h a t m o r a l s t a t e m e n t s have to
t h a t t h e act is n o t w r o n g , he is mistaken to t h i n k this. If t h e o b j e c -
be made, not about individual actions, but about kinds of action. T h i s
tivist thesis is true, this m u s t be so: if. of two people w h o m a i n t a i n ,
ought to m a k e a big difference to h o w we a p p r o a c h m o r a l thinking.
o n e of t h e m t h a t an act is w r o n g , a n d t h e o t h e r t h a t it is n o t w r o n g ,
One a p p r o a c h is to look at individual datable a c t i o n s by individual
both c a n n o t be right, then, in saying that an act is w r o n g . I m u s t be
identifiable people, and to ask whether they are right or wrong. T h e n
II. 5. 8
II. 5. 8
INTUITIONISM
97
we c a n arrive at m o r e g e n e r a l m o r a l principles by an inductive
I think t h a t m o r a l statements are always m a d e about actions, etc. be-
process. If lying h a s been perceived to be wrong in a lot of instances,
c a u s e of universal properties t h a t they have. T h i s by no m e a n s entails
for example, we may then generalize a n d form the hypothesis t h a t all
t h a t t h o s e p r o p e r t i e s have t o b e d e s c r i b a b l e i n very g e n e r a l t e r m s
actions of that kind, namely lies, are wrong. T h e other approach is to
(that is. in unspecific, simple terms which do not go into great detail).
start by considering types of a c t i o n s , a n d decide w h e t h e r a c t i o n s of
So, if I were an intuitionist. I would support t h e universalist version of
that type are wrong, t h u s forming for ourselves general m o r a l prin-
the theory. O n c e universality is distinguished from generality, most of
ciples: after t h a t w e c a n d e t e r m i n e w h e t h e r p a r t i c u l a r a c t i o n s a r e
the plausibility of particularism disappears. W h a t particularists a r e
w r o n g by asking whether they fall under those principles. For e x a m -
after is u s u a l l y specificity: t h e y do n o t w a n t us to m a k e o u r m o r a l
ple, we first determine t h a t lying is wrong, and then infer that to say a
s t a t e m e n t s on the basis of very general descriptions of actions, s u c h
particular thing would be wrong, b e c a u s e it would be a lie.

Hut this s t a t e m e n t of t h e difference is oversimplified. I used t h e
a s lying', b u t w a n t t o b e allowed t o t a k e a c c o u n t , i n t h e i r m o r a l
t h i n k i n g , o f t h e details o f c a s e s , w h i c h may, t h e y t h i n k , b e h i g h l y
word 'general', which has led a lot of moral philosophers into confu-
r e l e v a n t . W i t h t h i s I a g r e e : it is often n e c e s s a r y to d i s c u s s c a s e s in
sion, t h r o u g h being used as if it were s y n o n y m o u s with 'universal',
c o n s i d e r a b l e detail before we p r o n o u n c e on t h e m . B u t this does n o t
t h o u g h in fact t h e r e a r e two q u i t e different c o n c e p t s to be distin-
stop my s a y i n g t h a t it is still t h e u n i v e r s a l , t h o u g h h i g h l y specific,
guished (see 7.7. II 1 9 7 2 , 1994/), MT 2 . 5 ) . Generality is the opposite
properties of t h e cases that are the grounds of o u r m o r a l statements,
of specificity and is a matter of degree. Universality contrasts, rather,
and not the mere fact that those individuals are involved in t h e m .
with particularity, a n d is n o t a m a t t e r of degree. T h e two prescrip-
I do n o t need here to enlarge on the division of e t h i c a l theories into
tions 'One ought never to tell lies' and 'One ought never to tell lies to
particularist and universalist theories. B u t I had to m a k e this distinc-
business p a r t n e r s ' are b o t h equally universal, in that any a c t w h i c h
tion in order to m a k e my t a x o n o m y complete. It is perhaps worth
falls under the description 'lies' or 'lies told to business partners' (and
mentioning, as I have elsewhere (Ft 1955a, 1 9 9 4 b ) , t h a t t h e possibility
note that these descriptions a r e in universal terms) is prohibited by
of m a k i n g m o r a l s t a t e m e n t s a b o u t f i c t i o n a l c h a r a c t e r s is a s t r o n g
the respective prescriptions. Hut the first is much more general, m u c h
a r g u m e n t a g a i n s t particularism; for fictional c h a r a c t e r s c a n only be
less specific, t h a n the second.
described in universal terms; they do not exist to be pointed to as indi-
A particularist has to decide w h e t h e r he is objecting to our m a k i n g
viduals.
moral statements about general types of cases, and asking us to m a k e
A s h o r t way with particularism is to say t h a t there c a n n o t be any-
t h e m only, in the first i n s t a n c e , a b o u t very specific cases: or w h e t h e r
thing about an a c t i o n which makes it wrong, or a b o u t a person which
he is insisting that no universal terms, even highly specific ones, are to
m a k e s h i m bad, e x c e p t s u c h features as a r e specifiable in universal
be used in t h e d e s c r i p t i o n s of t h e c a s e s we are m a k i n g o u r m o r a l
t e r m s . A n y f e a t u r e w h i c h w a s n o t so specifiable would h a v e to be
s t a t e m e n t s a b o u t . T h a t is. i s h e insisting t h a t all m o r a l s t a t e m e n t s
s o m e individual e s s e n c e o r h a e c c e i t y w h i c h w a s n o t describable e x -
have to be, in t h e first i n s t a n c e , a b o u t individual a c t s identified in
cept by saying 'this person' or 'this a c t ' , a n d r e m a i n i n g silent t h e r e -
some other way than by describing t h e m in universal terms, however
after. B u t this is not to describe an a c t or a person at all. T h e only way
specific: or is he a l l o w i n g us to m a k e t h e m a b o u t kinds of c a s e s
to describe a person or a c t is to attribute universal properties to him
described in u n i v e r s a l t e r m s , provided t h a t t h e t e r m s a r e h i g h l y
or it. And these have to include the reasons for calling t h e person bad
specific?
or t h e a c t w r o n g . T h e particularist c a n n o t reply by substituting, for
1 am a universalis!, t h o u g h not an intuitionist universalist. T h a t is.
This person' or 'this a c t ' , 'exactly like this person' or 'exactly like this
A TAXONOMY OF ETHICAL THEORIES
n.5.8
II.
5.8
INTUITIONISM
99
a c t ' , b e c a u s e this would t u r n t h e description into one in terms of uni-
T h e fact t h a t r e l a t i o n s t o individuals m a y b e u n i v e r s a l qua r e l a -
versal properties after all (H 1955a, FR 2 . 2 ) . But this is not the place to
tions (they are two-or-more-place universal predicates, a l t h o u g h the

individuals t h e m s e l v e s a r e n o t u n i v e r s a l ) o p e n s t h e door to a g r e a t
enter into such metaphysical tangles.

However, this way is too short a way. It is too short, b e c a u s e t h e r e
m a n y duties a b o u t w h i c h we m a y be in d o u b t w h e t h e r to call t h e m
are universal properties w h i c h are relational, c o n n o t i n g relations to
moral duties at all. For example, are my duties to my country, b e c a u s e
a n individual, s u c h a s ' m o t h e r o f , a n d 'lover o f . I n t h e e x p r e s s i o n
it is my country, m o r a l duties, or only, to u s e an expression of S i m o n
' m o t h e r of J a m e s ' . ' J a m e s ' denotes an individual. But all the s a m e t h e
B l a c k b u r n ' s ( 1 9 8 4 : 1 8 6 ) , shmoral duties? In o t h e r words, are they du-
duties o n e m a y have t o o n e ' s m o t h e r m a y b e universal p r o p e r t i e s
ties w h i c h c a n be fully universalized over all recipients in all s i t u a -
shared with anybody w h o is a child, or a child of a specific kind.
tions, or only over all agents. And if I have m o r a l duties to my country,
T h a t is why prudential self-interested judgements are universaliz-
have I m o r a l duties to my family, or my tribe, or my sex, or my species
able after a fashion. T h e y are j u d g e m e n t s about relations to an indi-
(for short, to my o w n set) w h i c h are not owed to o t h e r people's sets? I
vidual, n a m e l y oneself. If I o u g h t , prudentially speaking, to do a
c a n c e r t a i n l y have a m o r a l duty to keep my promises, a n d n o t o t h e r
c e r t a i n t h i n g in a c e r t a i n precise situation, then anybody w h o is in
people's promises (H 1992c: ii. 1 2 5 9 ) . This would obviously endanger
the s a m e situation would be well advised, in prudence, to do t h e s a m e
one of the m a i n props of the type of m o r a l a r g u m e n t t h a t I advocate.
thing. For example, if it would be in my interest to tell the truth ( m a k e
T h e K a n t i a n solution to this problem is p r e s u m a b l y (if I m a y put
a c l e a n breast of it), t h e situation being w h a t it is, then it would be in
words into Kant's m o u t h ) that I have only s u c h duties to my family as
the interest of any similar person in a similar situation to do the s a m e ,
I am willing to a l l o w o t h e r s i m i l a r people to h a v e to their families.
t h a t is, speak the truth. T h e s a m e holds if for speak the truth' we sub-
And the s a m e is t r u e if for 'family' we substitute a n y of t h e o t h e r re-
stitute 'tell a lie', t h o u g h in t h a t c a s e t h e m o r a l and t h e p r u d e n t i a l
stricted sets 1 have mentioned. B l a c k b u r n h a s suggested to me in c o n -
j u d g e m e n t s m a y diverge. If it is in t h e interest of one person to tell a
versation t h a t t h e difference b e t w e e n h i m s e l f a n d m e c a n b e put i n
lie, it would be in t h e interest of a n y precisely similar person in pre-
t e r m s of t h e difference b e t w e e n H u m e a n d K a n t . I t h i n k I am still a
cisely the s a m e situation to tell a lie.
follower o f Kant, and B l a c k b u r n o f H u m e . T h e a d v a n t a g e t h a t K a n t
However, we a r e c o n c e r n e d now, n o t with p r u d e n c e (i.e. w h a t I
h a s over H u m e is t h a t H u m e c a n n o t m a k e t h e move I have j u s t made.
ought to do in my own interest), but. with what I ought to do b e c a u s e
He relies on human sympathy as the foundation of morality, whereas
of my r e l a t i o n s w i t h other individuals, for e x a m p l e , duties to my
Kant relies on t h e wills of all rational beings.
m o t h e r because she i s m y mother. J . K . Hare ( 1 9 9 b : 1 5 )
s r
'ghHy
distinguished between different types of universalizability. T h e r e is
In this c o n n e c t i o n it m a y be useful to recall points m a d e by Peter

Singer and by Derek Parlit. Singer has suggested ( 1 9 8 r: c h . 4) t h a t t h e
universalizability over all a g e n t s , but t h e r e is also universalizability
ability to r e a s o n , itself g e n e t i c a l l y useful a n d t h e r e f o r e fostered by
over all r e c i p i e n t s (e.g. v i c t i m s ) ; a n d t h e r e a r e o t h e r s e n s e s t o o in
evolution, c a n , as it were, take hold and push us beyond w h a t t h e in-
w h i c h judgements may be universalizable. A judgement m a y be uni-
terest of o u r genes requires. T h u s we c a n find no good r e a s o n for stop-
versalizable in one of these senses but not others. For example, a pru-
ping at t h e i n t e r e s t s of o u r o w n village or t r i b e . It is in o u r o w n
d e n t i a l j u d g e m e n t is u n i v e r s a l i z a b l e over a g e n t s , b u t n o t over
interest and t h a t of o u r genes to preserve m e m b e r s of o u r o w n tribe;
recipients. It relates to my interest as agent, but not to my interest as
but it is hard to stop there. And so reason e n c o u r a g e s us to go further,
recipient: I can treat other recipients just as 1 please, provided t h a t my
and seek to promote the interests of other tribes and even other
own interest is secured by my action.
species. I h a v e m y s e l f suggested t h a t w h a t is t r u e of t h e faculty of
IOO
II. S. 8
r e a s o n m a y be t r u e of t h e gift of l a n g u a g e w h i c h is its v e h i c l e (H
1 9 8 1 b ) . So perhaps m o r a l l a n g u a g e is after all superior to s h m o r a l
language.
Parlit loo ( 1 9 8 4 : c h s . 6 ff.) argues for an extension of our c o n c e r n
beyond n a r r o w self-interest. H e says t h a t I h e s a m e a r g u m e n t s b y
II. 5. 8
INTUITIONISM
101
be rationally acceptable to agents, as conducive to their interests and

those of their o w n sets, may be expected not to be acceptable to their
victims, w h o c a n n o t , as Kant put it, s h a r e t h e ends of t h e prescribed
actions ( 8 . 2 ) .
lem of shmoralities. They c a n count as moralities only if they are gov-
This is o n e reason why I prefer to use the m o r a l language. A n o t h e r

reason is prudential in the broadest sense. I said in MT 1 1 . 2 ff. t h a t if
we were b r i n g i n g up a child with no t h o u g h t of a n y t h i n g b u t t h e
child's o w n interest, we should bring h i m up to use m o r a l l a n g u a g e
and follow m o r a l prescriptions. Having seen t h e terrible t h i n g s t h a t
have been done t h r o u g h o u t the ages, and right up to t h e present day,
in the c a u s e of s h m o r a l systems t h a t are n o t m o r a l in t h e full sense, I
feel an urge to solve our problems by appeal to morality. T h a t perhaps
is the only way, especially for dealing with problems of o u r relations
to other species.
erned by such a universal rule. But shmoralities are not ruled out by
f o r clarity we m u s t distinguish between two q u e s t i o n s : t h e ques-
logic, any more t h a n prudence is. I admitted in MT 11.2 ff. t h a t c o n -
tion of w h a t t h e logic of t h e moral, as opposed to t h e s h m o r a l , l a n -
sistent amoralism is a viable option, and the same is true of s h m o r a l -
guage requires, and the question of how to motivate people to use it.
which prudence prevails over t h e 'present aim theory' c a n be used to

give the victory to universal m o r a l i t y over prudence. If we m a y e x tend what he says a little, they may also be used to give the victory to
K a n t over Hume. T h a t is to say, a l t h o u g h s i n g u l a r terms a r e admissible in m o r a l j u d g e m e n t s , t h e y h a v e to be governed by a u n i v e r s a l
rule w h i c h allows anybody so related to an individual or set to h a v e
similar duties to that individual or set.
If l'arlit and Singer are right, then we have a solution to the prob-
ism. And in MT 1.5 1 admitted t h a t alternative languages are available
If I am r i g h t in my a n s w e r to the first q u e s t i o n , t h e s e c o n d still r e -
for those w h o are unwilling to use t h e m o r a l language. T h e e x t r e m e
m a i n s . B u t this is n o t a q u e s t i o n for e t h i c a l t h e o r y in t h e n a r r o w
c a s e is t h e plain imperative, t h e l a n g u a g e expressing simple indi-
s e n s e . I c a n n o t so far see t h e s o l u t i o n to t h e s e p r o b l e m s , a n d h a v e
vidual desires or 'present aims'. B u t there are other kinds of s h m o r a l
therefore to leave s o m e unfinished business, as I did in FR 7 . 4 and MT
l a n g u a g e in between this and t h e l a n g u a g e of morals. T h e question
5 . 6 . 1 h a v e s i n c e t h e n tied up s o m e loose e n d s , a n d I h o p e t h a t I or
is. ' W h y should we use t h e m o r a l l a n g u a g e in preference to t h e s e ? ' I
others will tie up more. T h a t is o n e reason why I have left a c o r n e r of
am sure that I myself prefer to use moral language, and can therefore
my t a x o n o m y open for t h o s e t h a t c o m e after ( 6 . 7 ) . B u t at present I
follow M o o r e ( 1 9 0 3 : f>) a n d say t h a t in t h e s e n s e t h a t I u s e it. t h e
feel inclined to support Kant against Hume.
moral judgements in it are universalizable.

I am equally sure t h a t B l a c k b u r n a n d I shall find o u r s e l v e s in
a g r e e m e n t in most of our moral judgements, though it is not clear to
me how he would a r g u e for t h e m . He would probably invoke Hume,
whose views do not diverge from Kant's as m u c h as is c o m m o n l y supposed, either in epistemology or in ethics, although their c h a r a c t e r s
were very different. We s h a l l see in 7.3 t h a t arguabilily. and with it
t h e possibility of r e c o n c i l i n g c o n f l i c t i n g m o r a l positions, a r e a r e q u i r e m e n t for a s a t i s f a c t o r y e t h i c a l theory. T h i s is w h a t r e n d e r s
objective moral prescriptions, acceptable to all rational thinkers, possible (H 1 9 9 3 3 ) . S h m o r a l prescriptions, by contrast, t h o u g h they m a y
5 . 9 . W e are n o w a t the end o f m y discussion and classification o f

descriptivist theories. We found t h a t they all have s o m e t h i n g w r o n g
with t h e m : t h e c o m m o n fault t h a t t h e y s h a r e i s t h a t t h e y all inevitably, if fully e x a m i n e d , collapse into relativism. T h i s , as I say, is a
surprising result, b e c a u s e the m a i n motive of m o s t descriptivists h a s
been the desire to avoid relativism. B u t they have gone t h e wrong way
about avoiding it. It is in fact only by a b a n d o n i n g descriptivism t h a t
we c a n a t t a i n a kind of o b j e c t i v i t y in o u r m o r a l s t a t e m e n t s , as we
shall see. I think t h a t Kant understood this. It is b e c a u s e w h e n we are
t h i n k i n g m o r a l l y we a r e looking for prescriptions for a c t i o n , n o t descriptions of actions, t h a t our thinking is constrained (H 1 9 9 6 c ) . T h e r e
J02
IL 5- 9
are c e r t a i n maxims (to use K a n t ' s word) w h i c h we c a n n o t will to b e c o m e universal laws; a n d m a x i m s are a kind of prescriptions.
To a n t i c i p a t e w h a t I s h a l l be s a y i n g later: t h e r e a s o n why a pre-
scriptivist t h e o r y c a n avoid collapsing into relativism is t h a t t h e prescriptive element in the m e a n i n g of m o r a l statements, and especially
EMOTIVISM
its form, c a n be shared between cultures with different mores, as t h e

descriptive m e a n i n g c a n n o t . It is b e c a u s e all the different cultures are
prescribing, and prescribing in a universal form (they share t h a t p a r t
6 . 1 . IN Chapter 5 I finished, for the time being, talking about descrip-
o f t h e m e a n i n g o f t h e i r m o r a l s t a t e m e n t s ) t h a t they a r e all c o n -
tivist theories. We saw t h a t they were all destined to collapse into rel-
strained in their r e a s o n i n g by t h e formal logical properties of w h a t
ativism, w h i c h is the reverse of w h a t most of their supporters wish. I
t h e y a r e saying, w h i c h a r e t h e s a m e w h a t e v e r t h e c o n t e n t o f t h e i r
am going in Chapters 6 and 7 to talk about non-descriptivist theories,
moral opinions. But you may not understand this until I c o m e to out-
a n d ask w h e t h e r t h e y c a n avoid this c o l l a p s e i n t o r e l a t i v i s m .
line my own theory.
S u r p r i s i n g l y (to s o m e people), we s h a l l discover t h a t it is a n o n -
In t h e n e x t c h a p t e r I s h a l l be g o i n g over to the o t h e r side of t h e
descriptive e l e m e n t i n t h e m e a n i n g o f m o r a l s t a t e m e n t s w h i c h c a n
m a i n division o f e t h i c a l t h e o r i e s , a n d classifying non-descriptivist
e n a b l e a non-descriptivist t h e o r y to avoid r e l a t i v i s m ( 7 . 3 ) . B u t this
theories. I shall start with emotivism, of which Axel Hagerstrom was
e l e m e n t w a s n o t well c h a r a c t e r i z e d b y t h e f i r s t o f t h e n o n -
a pioneer, and discuss its merits and faults, t h e main fault being t h a t
descriptivist k i n d s of t h e o r y I s h a l l discuss, n a m e l y e m o t i v i s m .
it leads to irrationalism and m a k e s any at all fundamental m o r a l rea-
However, t h e p r o p o n e n t s o f e m o t i v i s m . o f w h o m A x e l H a g e r s t r o m
s o n i n g impossible. T h e n I s h a l l p r e s e n t a t h e o r y w h i c h avoids t h i s
( 1 9 1 1 ) was t h e first in m o d e r n times, m a d e t h e i m p o r t a n t step of sug-
fault, and also the faults of descriptivism.
gesting t h a t there is a n o t h e r element in t h e m e a n i n g of m o r a l statem e n t s besides their s y n t a x and their truth conditions. If they h a d n o t
m a d e this step, t h e l a t e r a d v a n c e s t o w a r d s a n o b j e c t i v i s t e t h i c a l
t h e o r y would h a v e been impossible; for d e s c r i p t i v i s m h a s to be
rejected before this step c a n be made.
I shall in w h a t follows be criticizing emotivism in general, a n d n o t
any particular emotivist, and certainly not H a g e r s t r o m . S i n c e m a n y
of the modern emotivists make errors which are n o t essential to e m o tivism itself, it will be best if I c o n s t r u c t my o w n version of an e m o tivist theory w h i c h brings out most clearly t h e virtues and t h e faults
in emotivism. I do n o t intend this as a c a r i c a t u r e n o r as an Aunt Sally.
I intend it to r e p r e s e n t t h e best t h a t e m o t i v i s m c a n do. An e x a m p l e
will show w h a t I am up to. Charles S t e v e n s o n produced in 1 9 4 5 t h e
fullest exposition of an emotivist t h e o r y t h a t t h e r e h a s b e e n (unless
w e i n c l u d e A l l a n G i b b a r d ( 1 9 9 0 ) , w h o w a s o b v i o u s l y deeply influe n c e d by S t e v e n s o n ; G i b b a r d c a l l s h i m s e l f a ' n o r m - e x p r e s s i v i s t ' ) .
S t e v e n s o n ' s b o o k is m a d e very c o n f u s i n g by h i s i n c l u s i o n in his
,o
H. 6. i
theory of subjectivist elements (drawn perhaps from W e s t e r m a r c k or

from a misinterpretation of h i m ) . T h u s his m o s t famous analysis of
moral judgements was of the form T approve of .v; do so as well'. But
t h e first h a l f of this analysans s o u n d s undeniably like a m e r e s t a t e m e n t of p s y c h o l o g i c a l fact a b o u t t h e speaker, and so would be
subjectivist, in the sense of Chapter 5. It would be a form of subjectivist naturalism, and h e n c e a form of descriptivism. And t h e addition
of t h e imperative part of t h e analysans. 'Do so as well', does n o t do
e n o u g h to remove the confusion.
T h e r e is e v i d e n c e t h a t m a n y people were a c t u a l l y misled by this
formulation. For example Ewing ( 1 9 5 9 ) called his chapter criticizing
non-descriptivism ' T h e New S u b j e c t i v i s m ' . And S t e v e n s o n h i m s e l f
called his p a t h - b r e a k i n g e a r l i e r paper ' M o o r e ' s A r g u m e n t s a g a i n s t
C e r t a i n F o r m s o f Ethical N a t u r a l i s m ' ( 1 9 4 2 ) . t h u s s u g g e s t i n g t h a t
w h a t he, Stevenson, was defending was subjeetivistic naturalism. So
it was easy to be confused. We m a y note that when Stevenson wrote,
Ayer, a n o t h e r famous emotivist. had already ( 1 9 3 6 : ch. 6) e m p h a t i c ally dissociated h i m s e l f from this kind of s u b j e c t i v i s m , and cited
a r g u m e n t s against it, drawn from M o o r e ( 1 9 1 2 : 5 7 ff.): so S t e v e n s o n
had no excuse for this confusion. It would have been better if he had
used only his s e c o n d p a t t e r n o f a n a l y s i s , w h i c h h a s t h e m e r i t o f
bringing out clearly the distinction between the two e l e m e n t s in t h e
m e a n i n g of moral statements, the descriptive and the evaluative, t h e
descriptive e l e m e n t being t h e s t a n d a r d o f application o f t h e m o r a l
words, and the evaluative being t h e expression of an attitude. By c o n -
11.6. 1
EMOTIVISM
tivists rightly distinguished. I will call these two aspects t h e expressive

and the causative, and start by discussing the first of these.
6 . 2 . Emotivists t h o u g h t t h a t w h e n I m a k e a m o r a l s t a t e m e n t , I
am expressing an attitude of my own to an act. person, etc. Note c a r e fully that expressing an attitude is different from stating that I have it.
T h a t i s o n e w a y o f s t a t i n g t h e i m p o r t a n t difference b e t w e e n e m o tivism and subjectivism on which I have been laying so m u c h stress.
Earlier emotivists said, instead of 'attitude', 'feeling'; but 'attitude' is
preferable, for r e a s o n s which Stevenson gave. I c a n say t h a t as a m a t ter of fact I h a v e a c e r t a i n attitude or feeling, w i t h o u t expressing it.
Contrast two people, o n e of w h o m says', in a c a l m tone of voice, T am
very a n g r y with you for what you have done', and t h e o t h e r of w h o m
says ' Y o u b l i t h e r i n g idiot!' T h e first is s t a t i n g t h a t he h a s a feeling
(anger): the second is expressing it.
It is i m p o r t a n t to understand t h a t there is n o t h i n g wrong, in o n e
sense, with saying t h a t w h e n we m a k e a m o r a l s t a t e m e n t we are e x pressing an a l t i t u d e . For e x a m p l e , if I said ' M e a t - e a t i n g is w r o n g ' I
should c e r t a i n l y b e e x p r e s s i n g a n a l t i t u d e t o m e a t - e a t i n g . T h e
trouble s t a r t s b e c a u s e of an unclarity or a m b i g u i t y in t h e word 'express'. Let us t h e r e f o r e look at this word m o r e closely. T h e s e n s e in
w h i c h s o m e e m o t i v i s t s w e r e using it is i n d i c a t e d by Ayer's u s e of
' e v i n c e ' as a s y n o n y m for it. If I evince anger, I am angry, and s h o w
it. So the impression we get is t h a t t h e emotivists t h o u g h t t h a t w h e n
we make a m o r a l statement, we have an attitude (for e x a m p l e of disapproval), and s h o w it.
trast, the 'descriptive e l e m e n t ' in T approve of x: do so as well' does
B u t even L this is a possible sense of 'express', it is certainly not the
not give a standard for the application of moral words: it is really an
only o n e . H e r e is a n o t h e r . English e x p r e s s e s n e g a t i o n by t h e word
irrelevancy in the analysis, and should be replaced by an expression,

not a description, of the speaker's attitude.
I want to avoid having to expose at tedious length w h a t I think are
simply mistakes in t h e emotivists' formulation of their theories, like
this one. So I am going to give you a simplified version of emotivism
which incorporates both the merits and the defects w h i c h I think are
essential to it. W h a t then is the element that the emotivists wanted to
add to t h e analysis of the m e a n i n g of moral statements, so as to m a k e
t h e m n o l o n g e r purely descriptive? I t had two aspects, w h i c h e m o -
n o t ' . Russell and W h i t e h e a d expressed t h e s a m e o p e r a t i o n b y t h e

tilde s i g n , ' - ' . M a t h e m a t i c i a n s express addition by t h e 'plus' sign (the
sign shaped like a c r o s s . ' + ' ) . Notice h o w odd it would be to say t h a t
m a t h e m a t i c i a n s evince addition by this sign, or t h a t English people
evince negation by saying 'not'. It is not a question of having an attitude or feeling and showing it; it is. rather, a question of having s o m e thing to say, and using this word in order to say it. W h e n e v e r we say
anything, we are expressing our m e a n i n g , and expressing it correctly
if we use t h e appropriate words. For this r e a s o n , philosophers often.
II. 6.
w h e n they w a n t to talk about s o m e word or phrase, refer to it as 'the

expression ". . . " ' , followed by the word or phrase in quotation m a r k s .
A b s o l u t e l y a n y word or p h r a s e in t h e l a n g u a g e is an e x p r e s s i o n in
(his sense.
I hope you will notice t h a t the distinction between expressing a n d
stating that survives into this different sense of 'express'. If I am writing s o m e t h i n g on a piece of paper, and s o m e o n e asks me w h a t I am
writing. I may reply '1 am negating the s t a t e m e n t t h a t S t o c k h o l m is
in Sweden'. In that case, if w h a t I am writing is ' S t o c k h o l m is n o t in
Sweden', w h a t I say with my m o u t h is true; but what I am writing is
false. So the negation I am expressing is false, but the statement that I am
expressing it is true. So expressing c a n n o t be the same as stating that.
Let us c o n t r a s t the use of ' n o t ' to express negation with the use,
say, of 'Hell!' to express a n n o y a n c e . T h e r e is one important difference
t h a t I should like you to n o t i c e . T h e word 'Hell!' c a n be used to e x press a n n o y a n c e b e c a u s e in its literal s e n s e 'Hell' is t h e n a m e of a
place, supposed to exist, w h i c h is an extremely nasty place to be in,
where 'their w o r m dieth n o t and their fire is not q u e n c h e d ' . T h e use
of 'Hell!' as an expression of a n g e r is a m e t a p h o r i c a l or transferred
use. This is not so with 'not' as an expression of negation. In this respect expressions of moral attitudes resemble 'not' more t h a n they do
'Hell!'. 'Not' expressing n e g a t i o n does not s e e m to be a t r a n s f e r r e d
use; where could it be transferred from? It just is the word we have in
English for negating. And similarly ' w r o n g ' just is one of t h e words
we have in English for expressing disapproval. T h e linguistic convention whereby the sound ' n o t ' is the way we have in English for n e g ating, or for expressing negation, is, in a sense, immediate, not derived
or transferred; and the s a m e is true of 'wrong' and disapproval.
It would be a good thing, therefore, if we put aside the associations
of the word 'evince', and treated 'wrong' as a word for expressing disapproval in just the same way as we treat 'not' as a word for expressing n e g a t i o n . B o t h approval and n e g a t i o n a r e kinds o f l i n g u i s t i c
operation which have their appropriate expressions. B u t of c o u r s e we
have not said m u c h about the m e a n i n g of 'wrong' when we have said
t h a t it is used to express disapproval. We need to go on to say w h a t
disapproval is.
II.
6. 2
EMOTIVISM
107
It is obviously t h e opposite of approval; b u t w h a t is approval? It,

like disapproval, is primarily a linguistic operation. T h e Oxford English
Dictionary, under 'approval', gives the m e a n i n g 'the action of declaring to be good'; a n d under 'approve' it gives the m e a n i n g ' p r o n o u n c e
to be good'. It says n o t h i n g there about feelings. B u t I must admit that
it does define 'approbation' as 'approval expressed or felt'; so evidently
t h e r e c a n he a feeling of approval. Still, t h e impression t h a t we get
from this d i c t i o n a r y is t h a t approval is primarily a speech a c t , n o t a
feeling or attitude. But to say this gets us no n e a r e r to understanding
what speech act.
Stevenson was n o t far wrong about this. He said, if I m a y s u m m a r ize his view, t h a t the attitude of approval is a disposition to act in the
w a y approved of, a n d to e n c o u r a g e o t h e r s to a c t in t h e s a m e way.
T h u s , if m o r a l words a r e expressions of approval or disapproval, the
expressive aspect of their m e a n i n g joins up, as we m i g h t say, with the
causative aspect. To have an attitude of approval is to be disposed to
do a c e r t a i n kind of act, and disposed to want or prescribe t h a t others
have t h e s a m e disposition. T h i s w a n t i n g is, I suppose, a feeling, so
feeling h a s n o t got left out of the picture entirely; but it h a s assumed a
s u b o r d i n a t e role. So it looks as if we c o u l d b e s t u n d e r s t a n d t h e e x pressive aspect of the m e a n i n g of moral s t a t e m e n t s by e x a m i n i n g the
c a u s a t i v e a s p e c t . T h i s part of t h e e m o t i v e t h e o r y holds t h a t it is a
function of m o r a l statements to induce feelings or attitudes or to influence conduct.
6 . 3 . W e c a n best discuss this c a u s a t i v e f u n c t i o n i n c o n n e c t i o n
with the a s s i m i l a t i o n o f m o r a l s t a t e m e n t s t o i m p e r a t i v e s n o t b e c a u s e the m e a n i n g of imperatives lies in the function of inducing attitudes or g e t t i n g people to do t h i n g s . T h a t , as we s h a l l see, is a
mistake. B d t s i n c e it w a s a m i s t a k e t h a t emotivists g e n e r a l l y m a d e ,
the investigation of it will shed light on their t h e o r i e s . For if we c a n
see what is wrong with saying that the m e a n i n g of imperatives is to be
explained by saying t h a t they are used to get people to do things, we
shall be in a better position to see what is wrong with the very similar
theory about m o r a l statements.
It is extremely n a t u r a l to think t h a t o n e c a n explain t h e m e a n i n g
of the imperative mood by saying that it is the mood o n e uses for get-
io8
IL 6. 3
ting people to do things. I have called this the 'verbal shove' theory of
the m e a n i n g of imperatives (1.5. II 1996/n. We find traces of views of
this sort in m a n y w r i t e r s (e.g. A. Ross 1 9 6 8 : 6 8 , see H 1969/): von
W r i g h t 1 9 6 3 : 1 4 9 f ; C a s t a e d a 1 9 7 4 : 4 5 . see H 1 9 7 6 c : S e a r l e a n d
Vanderveken 1985: 5 2 ) . I have been controverting t h e m t h r o u g h o u t
my career, but t h e m i s t a k e is very e a s y to m a k e (H T949. LM 12 f f .
1971/): s.f.). We often do use imperatives for getting people to do
things. A little reflection, however, will show that we c a n n o t explain
their m e a n i n g in this way. For, first of all, s e n t e n c e s in o t h e r m o o d s
a r e used to get people to do t h i n g s ; a n d secondly, i m p e r a t i v e s a r e
sometimes used with other purposes than to get people to do the Ihing
c o m m a n d e d or requested (the thing specified in the imperative). B u t
m e a n i n g has lo be something essential to the utterance of a s e n t e n c e .
If it is being used for s o m e other purpose t h a n shoving, that purpose
c a n n o t give its meaning, at least on this occasion of use. For example,
if I say 'Keep quiet', w h a t I am doing, on the 'verbal shove' theory, is
II. 6. 3
EMOTIVISM
109
trying to get t h e m , a n d wanting them, to talk, is strictly irrelevant to

an a c c o u n t of the meaning of w h a t he said.
A great m a n y o t h e r examples could be given of imperative uttera n c e s which, a l t h o u g h their m e a n i n g is clear, are n o t intended to get
the people addressed to do the thing specified. Getting people to do the
t h i n g specified is a function w h i c h imperatives very frequently a n d
typically h a v e a n d t h e r e is a reason, indeed, in their m e a n i n g why
they should typically have this function: but this function c a n n o t be
used in explanation of their meaning. It is a consequence of their having the m e a n i n g t h a t they have: the m e a n i n g explains this function,
and n o t t h e o t h e r w a y r o u n d . I am h e r e s u m m a r i z i n g a long a r g u m e n t ; w h a t I h a v e said is n o t c o n c l u s i v e as it s t a n d s , but I h a v e no
time to go at length into the matter. Perhaps it will help if I say a little
in general about why this sort of explanation of m e a n i n g in terms of
intended function will not do.
6.4
J . L . Austin ( 1 9 6 2 ) , a s w e have seen ( 1 . 5 ) , distinguishes
trying lo get the person addressed to keep quiet, and this is the m e a n -
between three things that he calls the locutionary act, t h e illocution-
ing of my u t t e r a n c e . T h e view would be refuted if we found an
ary act, and the perlocutionary act. T h e distinction between the first
i n s t a n c e of somebody saying 'Keep quiet', and m e a n i n g t h e n o r m a l
two need n o t c o n c e r n us now, even if it c a n be s u s t a i n e d , w h i c h I
t h i n g by it, b u t n o t t h e r e b y t r y i n g to gel t h e person or p e r s o n s
doubt (see H 197TC: 100 f f ) . B u t t h e distinction between t h e first two
addressed to keep quiet.
taken together a n d t h e third is of great i m p o r t a n c e (see also U r m s o n
Here is such an i n s t a n c e , w h i c h 1 r e m e m b e r first using in 1 9 4 9 .

T w o s c h o o l m a s t e r s in old-fashioned b o y s ' s c h o o l s both say to t h e i r
1 9 6 8 : c h . 1 1 ) . To understand w h a t is w r o n g with emotivism o n e has

to grasp this distinction.
respective c l a s s e s 'Keep q u i e t while I am o u t of t h e r o o m ' . O n e of
T h e perlocutionary effect of an u t t e r a n c e is w h a t you are doing or
t h e m is really w a n t i n g a n d t r y i n g to get the boys to keep quiet. B u t
trying to do by m a k i n g it {per locutionem). It h a s , says Austin, to be
the oilier, as soon as he h a s shut t h e door, puts his ear to the keyhole
distinguished from w h a t you are doing in saying w h a t you say (in lo-
and, when the boys start lo talk, as he hoped they would, (lings open
cutione), t h e i l l o c u t i o n a r y a c t . And, even m o r e , il h a s to be distin-
t h e door, gets out his stick a n d proceeds to indulge h i m s e l f B o t h of
guished from the m e a n i n g of your u t t e r a n c e , f o r example, to revert
these schoolmasters m e a n t t h e s a m e by their words. It is not the c a s e
to our sadistic s c h o o l m a s t e r : w h a t he was doing in saying 'Keep quiet'
that the sadistic o n e really m e a n t 'Talk while I am out of t h e r o o m ' .
w a s telling t h e b o y s to keep quiet; t h a t w a s w h a t his words m e a n t .
For if t h a t were what he had m e a n t , t h e boys would n o t have been
But what he was trying to do by saying it was to get t h e m to talk, and
disobeying h i m , and h e would n o t h a v e had a n e x c u s e for b e a t i n g
so expose themselves to his eccentric amours. T h e reason why it is im-
t h e m . In order for his e x c u s e to work, a n d for h i m n o t to get i n t o
possible in principle to e x p l a i n m e a n i n g in t e r m s of p e r l o c u t i o n a r y
trouble with t h e headmaster, he h a s to have told t h e m to keep quiet.
effect is that m e a n i n g , in t h e relevant sense, a n d illocutionary force if
And this is indeed w h a t h a s happened. T h e fact t h a t by telling t h e m
t h a t is different, is s o m e t h i n g t h a t belongs by convention to an utter-
to keep quiet he w a s ( k n o w i n g boys to be n a t u r a l l y i n s u b o r d i n a t e )
a n c e of a c e r t a i n type m a d e in a c e r t a i n kind of situation. T h u s t h e
11(>
II. 6. 4
m e a n i n g of the u t t e r a n c e 'I promise to pay you a thousand kronor to-
II. 6 . 4
EMOTIVISM
in
a r e c o n c e r n e d w i t h , a n d n o t t h e getting. N o r is it t h e t r y i n g to get
m o r r o w ' , uttered in a n o r m a l situation (and not, for example, on t h e
which constitutes t h e telling. You c a n try to get h i m to keep quiet by
s t a g e i .3, H 1 9 8 9 ) is determined by t h e convention that the sounds
telling h i m to; but these are different things. To give an analogy; I c a n
T promise, e t c ' a r e t h e s o u n d s used in English for performing t h e
try to loosen the top of the j a m j a r by heating it; but if o n e wanted to
speech act which we call 'promising to pay the addressee S w K r 1 , 0 0 0
explain w h a t h e a t i n g was, o n e could n o t do it by s a y i n g t h a t it w a s
on t h e following day'. We c a n n o t give a series of sounds m e a n i n g , in
trying to loosenpartly b e c a u s e one could h e a t for m a n y o t h e r pur-
t h e r e l e v a n t s e n s e of t h a t word, w i t h o u t h a v i n g a c o n v e n t i o n t h a t
poses, a n d partly b e c a u s e t h e r e could b e o t h e r w a y s b y w h i c h o n e
t h a t is h o w they are to be usedi.e. that that is the speech act t h a t
could try to loosen.
t h e y express, or of w h o s e e x p r e s s i o n t h e y form part. I l l o c u t i o n a r y
T h e r e is, of course, a reason why telling people to do things is nor-
force, if it is something different from m e a n i n g , is subject to t h e s a m e
mally a way of getting t h e m to do them. T h i s will b e c o m e clearer if I
condition: we c a n n o t give to t h e sounds T promise, e t c ' t h e power of
say m o r e positively w h a t o n e is doing w h e n o n e tells s o m e o n e to do
c a r r y i n g the illocutionary force of promising without having a c o n -
s o m e t h i n g . How, for e x a m p l e , do we distinguish w h a t we a r e doing
vention that that is the speech act which they express.
when we tell s o m e o n e to keep quiet from w h a t we are doing w h e n we
But we could not, in principle, have a convention that a c e r t a i n se-
tell h i m t h a t he is going as a m a t t e r of fact to keep quiet? W h a t , in
ries of s o u n d s was used for g e t t i n g people to do t h i n g s . We do, of
g e n e r a l , is t h e difference b e t w e e n typical i m p e r a t i v e s a n d t y p i c a l
course, have a convention t h a t to utter a certain series of sounds is to
future indicatives or declaratives with t h e s a m e c o n t e n t ? I s h a l l n o t
tell s o m e o n e to do a c e r t a i n t h i n g t h a t , for e x a m p l e , to u t t e r t h e
be able to explain this at all fully; b u t perhaps I c a n m a k e a s t a r t in
words 'Keep quiet' is to tell the people addressed to keep quietto per-
this way. Suppose t h a t I am speaking to an ideally c o m p l a i s a n t per-
form t h e speech a c t of telling t h e m . B u t telling t h e m is n o t getting
s o n a person w h o is disposed to accept, a g r e e with, a n d in g e n e r a l
them, nor even (as we saw in the s c h o o l m a s t e r example) trying to get
a s s e n t to e v e r y t h i n g t h a t I say. If I tell h i m t h a t J a n e is in t h e n e x t
t h e m . T h e reason why telling is a conventional activity, w h e r e a s get-
r o o m , he will believe me w i t h o u t q u e s t i o n . If I tell h i m to s h u t t h e
ting or trying to get is not and c a n n o t be, is t h a t to tell somebody to
door, he will do it without question. T h e difference between the m e a n -
do something all you have to do is follow the appropriate convention,
ings of indicatives and imperatives c a n be b r o u g h t out by saying that,
and say 'Keep quiet' if you are speaking English. 'Chup mho if you are
in the c a s e of an indicative, t h e complaisant (or, as I have sometimes
speaking Hindi, and so on.
called it, t h e a c c o r d a n t ) response is believing whatever is said, whereas
But in order to get s o m e o n e to keep quiet, it is, perhaps, no use just

performing speech acts according to the conventions. If you are going
in the c a s e of an imperative, the complaisant or a c c o r d a n t response is

doing it.
to get someone to keep quiet, he has to be disposed to keep quiet. O n e
People a r e n o t a l w a y s c o m p l a i s a n t ; s o m e t i m e s t h e y a r e n o t dis-
of the ways of so disposing him is to tell him to keep quiet; but t h e lin-
posed to a c c e p t or a s s e n t to w^iat we say. If we k n o w t h a t t h e y a r e
guistic part of the procedure, t h e speech act, is over w h e n you have
c o u n t e r s u g g e s t i b l e , like t h e bqys in t h e s c h o o l m a s t e r e x a m p l e , we
told h i m t h a t is. w h e n you h a v e uttered t h e appropriate words in
may say one thing in the hope t h a t they will believe, or do, the oppos-
a c c o r d a n c e with the linguistic conventions. You have done this, and
ite. B u t n o r m a l l y we a s s u m e t h a t our h e a r e r s are, for o n e r e a s o n or
given the words their meaning, w h a t e v e r he subsequently does. T h e
a n o t h e r , sufficiently c o m p l a i s a n t to do or believe w h a t we say.
getting is an effect of the telling (an effect which may be produced by
Otherwise co-operation would be difficult, if n o t impossible. So we do
o t h e r m e a n s , such as doping or gagging him, or just frightening h i m
not normally ask or tell people to do things unless we think t h a t they
s p e e c h l e s s ) . W h e n you a r e discussing m e a n i n g , it is t h e telling you
are. at least s o m e w h a t , disposed to do w h a t we tell or ask t h e m t o :
1 12
II. 6.4
ll. 6.
EMOTIVISM
a n d similarly we do n o t n o r m a l l y m a k e s t a t e m e n t s to people unless
liefs: they do not have m e a n i n g in the s a m e way as ordinary descrip-
we think that they are at least s o m e w h a t disposed to believe whatever
tive s t a t e m e n t s h a v e m e a n i n g . T h e r e f o r e t h e i r m e a n i n g m u s t b e
we state to be the e a s e . B u t , j u s t as in t h e second c a s e it would be a
s o u g h t i n t h e i r p r a g m a t i c s . B u t b e c a u s e h e failed t o d i s t i n g u i s h
mistake to try to explain t h e m e a n i n g of the indicative mood by say-
between illocutionary and perlocutionary acts, he plunged headlong |j
ing t h a t it is t h e mood y o u u s e for t r y i n g to get people to believe
into i r r a t i o n a l i s m . M e a n i n g c a n b e illocutionary, a n d therefore c a n
things, so we must not step from the true explanation of the m e a n i n g
be c o n s t r a i n e d by l o g i c a l rules, even t h o u g h it is n o t g o v e r n e d by
of imperatives (namely, t h a t an imperative is the kind of speech a c t ,
truth conditions. T h e mistake was to t h i n k t h a t b e c a u s e m o r a l state-
the c o m p l a i s a n t or a c c o r d a n t response to w h i c h is to do, or b e c o m e
m e n t s do n o t h a v e t h e i r m e a n i n g d e t e r m i n e d wholly by t h e i r t r u t h
disposed to do, the thing specified) to the false notion that an imperat-
c o n d i t i o n s , t h e r e c a n b e n o m o r a l a r g u m e n t , o r o n l y v e r y limited
ive sentence is, essentially, an attempt to get someone to do t h e thing
sorts of it. T h e word ' p r a g m a t i c s ' w a s , I t h i n k , m a i n l y to b l a m e for
specified, and that this is an explanation of its m e a n i n g . Usually it is
this mistake. T h e s a m e confusion h a s been e n c o u r a g e d by s o m e fol-
an attempt to get the thing done, but not essentially.
lowers o f W i t t g e n s t e i n b y i n d i s c r i m i n a t e b a n d y i n g a b o u t o f t h e
6 . 5 . T h i s i s perhaps t h e best point a t w h i c h t o say s o m e t h i n g
expression 'the use of s e n t e n c e s ' , which could m e a n either their illo-
m o r e about the expression ' p r a g m a t i c s ' which h a s c a u s e d s o m u c h
c u t i o n a r y o r t h e i r p e r l o c u t i o n a r y use ( 1 . 5 ) . A u s t i n m e n t i o n s this
confusion (1.5 f., H 1996/)). If anybody uses it. one can be almost c e r -
source of confusion too ( 1 9 6 2 : 1 0 0 ) .
tain t h a t he is going to c o n f u s e i l l o c u t i o n a r y with p e r l o c u t i o n a r y
If it is a m i s t a k e to t r y to e x p l a i n t h e m e a n i n g of imperatives irr
acts. 'Pragmatics' was one of a triad (the other two being 'syntactics'
t e r m s of ( h e i r p e r l o c u t i o n a r y effect, it is obviously even m o r e of a
a n d ' s e m a n t i c s ' , w h i c h 1 have m e n t i o n e d a l r e a d y ) . T h e s e t h r e e e x -
mistake to do this w i t h m o r a l s t a t e m e n t s . It is even m o r e absurd to ;
pressions were introduced by Charles Morris ( 1 9 3 8 ; J 9 4 6 : 2 1 6 f.) in a
say t h a t the essential function of moral s t a t e m e n t s w h a t gives t h e m "
l a u d a b l e a t t e m p t t o bring s o m e c l a r i t y into t h e g e n e r a l a n d very
their m e a n i n g i s to get people to do things t h a n it is to say this about
vague notion of ' m e a n i n g ' . I do not want you to interpret me as say-
imperatives. O p p o n e n t s of emotivism h a v e often pointed this out. If
ing t h a t t h e r e is only o n e kind of m e a n i n g o n l y one s e n s e of t h e
s o m e o n e h a s j u s t b e e n drafted i n t o t h e a r m y a n d , h a v i n g pacifist
word. T h e r e is even a sense of ' m e a n i n g ' in which perlocutionary ef-
leanings, asks me w h e t h e r he o u g h t to obey t h e call-up a n d join the
fect is part of m e a n i n g . All I am a s k i n g is that the different kinds of
army, and I say to h i m 'Yes, you o u g h t ' , I m i g h t n o t be trying to get
m e a n i n g should be carefully distinguished, and those w h i c h have to
h i m to join t h e army. He might think it an impertinence, or at least an
do with logic and rules for use separated from those which have not.
T h e t r o u b l e c a u s e d b y t h e word ' p r a g m a t i c s ' ( w h i c h predates
Austin's distinction), c o m e s out. very clearly when people say, for e x ample, t h a t the m e a n i n g of imperatives is constituted by their pragmatics.
S t e v e n s o n even said this sort o f
u n w a r r a n t e d i n t e r f e r e n c e in a p e r s o n a l d e c i s i o n , to do a n y s u c h
thing as trying to get him to join the army. He asked for advice, not influence or i n d u c e m e n t .
However, o p p o n e n t s of emotivism often, h a v i n g pointed this out,
thing about moral
go on to infer from it t h a t m o r a l judgemerjts, s i n c e t h e y a r e n o t at-
s t a t e m e n t s . H e called o n e o f t h e m a i n s e c t i o n s o f his b o o k ( 1 9 4 5 )
tempts to get people to do t h i n g s , c a n n o t be a n y t h i n g like imperat-
' P r a g m a t i c Aspects of M e a n i n g ' . If he had m e a n t by this s o m e t h i n g
ives, because these are attempts to get people to do things. As we have
to do with illocutionary forces. I would have applauded him. But a c -
seen ( 1 . 6 ) . this h a s b e e n used as an a r g u m e n t for a r e t u r n to s o m e
tually, b e c a u s e of the confusion c a u s e d by the word ' p r a g m a t i c ' , he
kind of descriptivism. naturalistic or intuitionistic. B u t t h e a r g u m e n t
s e e m s t o h a v e a r g u e d a s follows: m o r a l s t a t e m e n t s (or e t h i c a l s e n -
h a s a false premiss. It is wrong to say t h a t even imperatives are essen-
tences as he called t h e m ) do not (at least do not primarily) express be-
tially attempts to get people to do things. O n c e this mistake about im-
ii4
II. 6. 5
peratives h a s b e e n noticed, we a r e saved from a g r e a t m a n y e r r o r s
II. 6. 6
EMOTIVISM
t i m e t h o u g h t o t h e r w i s e . T h e emotivists h a d this w r o n g view a b o u t
w h i c h have infected r e c e n t m o r a l philosophy. T h e a r g u m e n t I h a v e
imperatives t h a t I h a v e b e e n a t t a c k i n g : a n d t h e y t h o u g h t , c o n s e -
just mentioned, since t h e first of its premisses is false, does n o t even
quently, that, in assimilating moral statements, as they did, to imper-
prove t h a t m o r a l s t a t e m e n t s a r e n o t imperatives. It is n o t a c t u a l l y
atives, they were saying something about their p e r l o c u t i o n a r y effect.
t r u e t h a t they are imperatives, a n d I have never said t h a t t h e y were
B u t this would b e q u i t e useless a s a n e x p l a n a t i o n o f t h e i r m e a n i n g ,
( L M 1.1), although I have often been a c c u s e d of doing this. My view
for the reasons I have given.
is r a t h e r t h a t they share with imperatives a very i m p o r t a n t feature,
T h e opponents of emotivism shared, as I said, this false view about
w h i c h I shall be c a l l i n g prescriptivity. It is c r u c i a l , therefore, to see
imperatives. T h e y therefore thought that in order to s h o w t h e ration-
t h a t to be prescriptive, even in t h e c a s e of imperatives, is n o t the s a m e
ality of moral thinking they had to reject w h a t they loosely called 'the
as to have t h e essential function of getting people to do t h i n g s . T h e
imperative t h e o r y ' (this i s o n e o f t h e c h a p t e r - h e a d i n g s i n S t e p h e n
theory that moral judgements are prescriptive is therefore not open to
Toulmin's early book An Examination of the Place of Reason in Ethics
the attack I have just mentioned.
( 1 9 5 0 ) , and by it he m e a n t emotivism). And they included in this all
6 . 6 . I want you to see h o w i m p o r t a n t this is. Perhaps I c a n help
forms of prescriptivism. So the whole controversy between t h e e m o -
you see this by a bit of autobiography, which I hope you will pardon.
tivists a n d t h e i r o p p o n e n t s was c o n d u c t e d o n t h e w r o n g basis, a n d
W h e n I started doing moral philosophy immediately after t h e S e c o n d
m o s t o f t h e c o n f u s i o n s t h a t h a v e plagued m o r a l philosophy ever
World War. the emotivists were at t h e height of fashion, and t h e m a i n
since, right up to now, have been t h e effect of this mistake. T h e c o n -
c o n t r o v e r s y was between t h e m a n d t h e i r opponents (H 1 9 9 5 b ) . T h e
troversy was t h o u g h t of as a battle between, on t h e one side, rational-
c h i e f thing t h a t seemed to divide t h e parties was that t h e emotivists
ist descriptivists, and, on the other, irrationalist non-descriptivists. It
denied that moral thinking could be a rational activity, whereas their
was taken for granted t h a t rationalism was inseparable from descrip-
o p p o n e n t s insisted t h a t it could be. For this reason, e m o t i v i s m w a s
tivism. and non-descriptivism from irrationalism. T h a t explains why I
frowned on by all the good and great. Indeed, t h a t was w h a t m a d e it
have had s u c h a h a r d time getting my views u n d e r s t o o d . For I h a v e
so popular a m o n g the young. W h e n I entered this scene, I was an op-
been m a i n t a i n i n g a rationalist kind of non-descriptivism. A n d I c a n
p o n e n t of e m o t i v i s m . b e c a u s e 1 did w a n t to show, if I could, t h a t
d o t h a t b e c a u s e o f m y a v o i d a n c e o f t h e m i s t a k e a b o u t imperatives
moral thinking could be rational. B u t I soon b e c a m e convinced of the
that I have been pointing out.
fallaciousness of the usual a t t a c k s on emotivism, which were all from
If one thinks t h a t imperatives, and prescriptive speech a c t s gener-
a descriptivist s t a n d p o i n t . It b e c a m e c l e a r to me t h a t w h a t w a s
ally, have m e a n i n g in virtue of t h e i r use to get people to do t h i n g s ,
needed was a non-descriptivist ethical theory which was at the s a m e
then one is trying to explain their m e a n i n g in t e r m s of their perlocu-
time rationalist. For 1 was quite c e r t a i n that the emotivists were right
tionary effect. B u t perlocutionary effect h a s n o t h i n g essentially to do
in their non-descriptivism, b u t equally certain t h a t they were w r o n g
with c o n v e n t i o n s or rules for t h e c o r r e c t use of expressions. T h a t in-
in thinking that there could n o t be rational argument about even t h e
deed is why in principle it could n o t be used to explain m e a n i n g . B u t
most fundamental moral questions.
logic, as applied to a class of expressions, owes its existence and valid-
T h e key to discovering a r a t i o n a l i s t kind of non-descriptivism is
ity to t h e s e rules and c o n v e n t i o n s governing t h e use of expressions.
this: lo say that moral statements are prescriptive is to say s o m e t h i n g
For e x a m p l e , as we have seen (1.1 f.) the modus ponens form of argu-
about t h e i r c h a r a c t e r as i l l o c u t i o n a r y a c t s : it is to say s o m e t h i n g
m e n t ('If p t h e n q; andp: so q') owes its validity to t h e rules governing
about their illocutionary force (in Austin's term), and n o t about their
t h e u s e o f t h e e x p r e s s i o n ' i f a n d t h e o t h e r words i n t h e s e n t e n c e s .
perlocutionary effect. Both the emotivists and their opponents at t h a t
B u t an e x p l a n a t i o n of t h e m e a n i n g s of m o r a l words in t e r m s of per-
A T A X O N O M Y OF E T H I C A L THKORIES
II. 6. 6
II. 6. 7
EMOTIVISM
l o c u t i o n a r y effect c a n n o t g e n e r a t e rules for their use, and therefore
turned to non-descriptivist theories, and considered t h e earliest vari-
c a n n o t generate a logic. A theory which relies on it is therefore bound
ety of these, e m o t i v i s m . I found in this s o m e virtues b u t o n e serious
to be irrationalisl. However, o n c e we see that the correct explanation
fault, t h a t it could find no place for r a t i o n a l m o r a l a r g u m e n t a b o u t
of t h e m e a n i n g s of both m o r a l words and imperatives is in t e r m s of
f u n d a m e n t a l m o r a l q u e s t i o n s . T h i s gives us t h e differentia w h i c h
their illocutionary force, not their perlocutionary effect, we see, also,
divides non-descriptivism into its two main varieties. T h e variety t h a t
that it is possible to say that moral statements and imperatives are dif-
f h a v e d i s c u s s e d so far. e m o t i v i s m , is an i r r a t i o n a l i s t s o r t of n o n -
ferent varieties of the kind of speech act called prescribing, and that,
descriptivism. I shall go on in Chapter 7 to set out a rationalist sort of
s i n c e their m e a n i n g c a n be t h u s characterized in terms of their illo-
non-descriptivism, which will also yield a kind, t h o u g h n o t a descrip-
c u t i o n a r y force, it does determine rules for their use, and thus gener-
tivist kind, of objectivity for moral statements. For if m o r a l t h i n k i n g
ates a logic. So there c a n be r a t i o n a l m o r a l a r g u m e n t even t h o u g h
c a n be s h o w n to be rational, then we c a n expect rational thinkers to
moral judgements are prescriptive.
agree in their moral opinions o n c e they are in possession of t h e facts
1 h o p e 1 have n o w c o n v i n c e d you t h a t , s i n c e imperatives c a n be
and think clearly. So I wish to divide non-descriptivism into its ration-
governed by logical rules arising out of their m e a n i n g and illocution-
alist and i r r a t i o n a l i s t varieties. I do not c l a i m t h a t my t a x o n o m y of
ary force, it would be possible to be even an imperativist in e t h i c a l
non-descriptivism is complete. T h a t is, there m a y be (I am sure t h a t
theory (that is. to assimilate moral statements completely to ordinary
t h e r e a r e ) f u r t h e r subdivisions o f t h e s e two kinds o f n o n - d e s c r i p -
imperatives) without being an irrationalist. I am not. and never have
tivism. In t h e c a s e of descriptivism, I showed, I hope, t h a t all its pos-
been, an imperativist, b e c a u s e I think t h a t moral s t a t e m e n t s s h a r e
sible varieties were i n a d e q u a t e . In t h e c a s e of n o n - d e s c r i p t i v i s m I
only one feature with imperatives, their prescriptivity, and have o t h e r
have not claimed so m u c h . We shall in Chapter 7 look at a variety of
features which they do not s h a r e with imperatives, and w h i c h m a k e
rationalist non-descriptivism which I think is t h e most adequate eth-
them m o r e like indicatives (in particular t h e fact that they c a n be true
ical t h e o r y so far devised. B u t there m a y be o t h e r varieties of ration-
or false a n d have t r u t h c o n d i t i o n s ) . B u t s i n c e this one feature, pre-
alist non-descriptivism w h i c h will do better. I am n o t trying to close
scriptivity, stops us calling m o r a l s t a t e m e n t s purely descriptive, it is
t h e door to n e w a n d improved t h e o r i e s , b u t s h a l l , as I said, leave a
very important to see t h a t they c a n have it, without making rational
c o r n e r open. B u t I am quite sure t h a t the only o n e s with a n y promise
a r g u m e n t about moral q u e s t i o n s impossible. I have tried to s h o w in
will have t h e i r place on t h e non-descriptivist side of t h e t a x o n o m y ,
my b o o k s how s u c h a r g u m e n t s c a n be c o n d u c t e d , a n d I will s u m -
and in its rationalist segment.
marize my view on this in Chapter 7.
Even in t h e c a s e of e m o t i v i s m it is possible t h a t i m p r o v e m e n t s
6 . 7 . I am now in a position to c o m p l e t e t h e main framework of
could be suggested w h i c h m a k e it no longer irrationalist. For e x a m -
my t a x o n o m y of ethical theories, t h o u g h I shall deliberately leave a
ple, as we have seen, Allan Gibbard, calls himself a norm-expressivist
c o r n e r of it open ( 5 . 8 ) . I divided e t h i c a l theories into t h e species de-
( w h i c h sounds very S t e v e n s o n i a n ) , and goes on in t h e latter p a r t of
scriptivist and non-descriptivist, t h e differentia being t h a t t h e former
his excellent book ( 1 9 9 0 ) to claim t h a t in his t h e o r y a kind of objectiv-
affirmed, and the latter denied, t h a t t h e m e a n i n g of moral statements
ity c a n be achieved in m o r a l s t a t e m e n t s . His title is significant: Wise
is wholly determined, apart from s y n t a c t i c a l features, by their truth
Choices, Apt l-'eelings. A l t h o u g h his l a n g u a g e often suggests t h a t he is
conditions. 1 then divided descriptivist theories into naturalism, in its
an emotivist, we should probably n o t classify h i m as an irrationalist.
objectivist and subjectivist varieties, and intuitionism, and showed
B u t he is w i t h o u t doubt a non-descriptivist, and h a s s o m e telling crit-
that all these forms of descriptivism are bound to collapse in one way
icisms of r e c e n t descriptivists s u c h as J o h n McDowell. So perhaps we
or a n o t h e r into relativism, which I showed to be unacceptable. I t h e n
should classify h i m as a rationalist non-descriptivist like m e . I shall
n8
IL 6. 7
II. 6. 8
EMOTIVISM
119
n o t h a v e t i m e i n this b o o k t o e x a m i n e his c o m p l e x t h e o r y i n detail;
ditions of m o r a l s t a t e m e n t s which is at t h e s a m e t i m e naturalist (that
b u t I like to t h i n k of h i m as b e i n g in t h e s a m e c a m p as myself, a n d
is, w h i c h formulates t h e m in terms of n o n - m o r a l properties) is b o u n d
look on its publication as a sign t h a t t h e descriptivist tide m a y have
to introduce substantial moral stipulations into the theory; and if
turned.
anybody does n o t like t h e stipulations he will reject t h e theory.
6 . 8 . W h a t I am going on to do in t h e rest of this chapter is to state,

as briefly as I c a n , w h a t I think are t h e essential features t h a t an ethical t h e o r y h a s to have if it is to be a d e q u a t e ; t h a t is, t h e features of
m o r a l l a n g u a g e a n d its logic, as we have t h e m , w h i c h a t h e o r y m u s t
do j u s t i c e to if it is to be t e n a b l e . T h i s will provide us w i t h a s o r t of
sieve t h r o u g h w h i c h we c a n put a n y e t h i c a l theory; if it fails to pass
t h r o u g h the sieve b e c a u s e it does n o t do j u s t i c e to a n y o n e of t h e s e
features, it h a s to be rejected. I shall t h e n m a k e a m e n d s by drawing
a t t e n t i o n to t h e good points of e a c h of t h e t h e o r i e s I have discussed
(the features of moral t h o u g h t a n d l a n g u a g e to w h i c h it does do just-
(2) Secondly, no e t h i c a l theory is going to be of a n y u s e for practice if

it leads only to m o r a l c o n c l u s i o n s of w h a t I s h a l l call t h e ' S o w h a t ? '
sort. By this I m e a n t h a t if, at the end of a m o r a l a r g u m e n t , o n e of t h e
disputants is forced to a g r e e to a m o r a l c o n c l u s i o n , but c a n t h e n say
'Yes, it would be w r o n g to do that; so what?', t h e n t h e system of m o r a l
a r g u m e n t is a fraud. I give an e x a m p l e of s u c h a failure in MT 4 . 3 .
T h i s r e q u i r e m e n t I shall call t h e requirement of practicality. It is failed
by all forms of descriptivism, b e c a u s e they leave o u t t h e prescriptive
element in t h e m e a n i n g of moral statements.
ice). T h e n we shall be in a position to try to draw together these good
(3) Next, a n a c c o u n t o f t h e m e a n i n g s o f m o r a l words h a s t o b e s u c h
features i n t o o n e theory, w h i l e r e j e c t i n g t h e b a d points. A n d t h a t is
t h a t t h e m o r a l d i s a g r e e m e n t s t h a t w e find going o n r e a l l y a r e dis-
w h a t I h o p e to do. T h u s my t h e o r y will be an e c l e c t i c o n e in a good
a g r e e m e n t s . W e saw t h a t this requirement was n o t m e t b y t h e t h e o r y
sense (H 1 9 9 4 b ) .
called subjeetivistic naturalism. According to it, if I call an a c t w r o n g
T h e r e are, I t h i n k , six f e a t u r e s o f m o r a l s t a t e m e n t s t h a t w o u l d
a n d you call it n o t w r o n g , we a r e stating, respectively, t h a t I h a v e a
m a k e m e w a n t t o r e j e c t a n y t h e o r y t h a t failed t o d o j u s t i c e t o t h e m .
certain feeling or attitude a n d t h a t you have a c e r t a i n opposite feeling
Most of t h e m I have m e n t i o n e d already. I have given my sieve to c a t c h
o r attitude; a n d w e a r e t h e r e f o r e n o t s a y i n g t w o t h i n g s w h i c h a r e
inadequate ethical theories in the table on p. 4 2 . It shows w i t h a cross
i n c o m p a t i b l e w i t h e a c h other. I s h a l l c a l l this t h e incompatibility-
which of the theories, in my view, fail to satisfy which requirements.
requirement. So far as I c a n see, it is failed only by subjeetivistic n a t -
(1) First of all, no ethical t h e o r y t h a t is, no a c c o u n t of t h e m e a n i n g
u r a l i s m , t h o u g h if, as I h a v e c l a i m e d , t h e r e is no r e a l difference
o f t h e m o r a l words a n d t h e logic o f m o r a l a r g u m e n t w h i c h t h a t
b e t w e e n i n t u i t i o n i s m a n d subjectivism, i n t u i t i o n i s m fails it t o o . B u t
brings with itcan do a n y t h i n g for a c t u a l moral a r g u m e n t s unless it
t h e i n t u i t i o n i s t s c e r t a i n l y did n o t t h i n k t h e y failed it; t h e y t h o u g h t
c a n be accepted by both parties to t h e a r g u m e n t s . This m e a n s t h a t it
t h a t there could be real disagreement about w h e t h e r an a c t possessed
is a l w a y s fatal to t r y to s m u g g l e m o r a l o p i n i o n s of s u b s t a n c e i n t o
or did not possess t h e objective moral property of w r o n g n e s s . So I will
o n e ' s ethical theory in t h e guise of m e r e definitions or explanations of
allow t h e m to pass t h i s r e q u i r e m e n t . I t h i n k t h a t it w a s S t e v e n s o n ' s
m e a n i n g , as in effect t h e objectivistic n a t u r a l i s t s do. If o n e p a r t y to
big c o n t r i b u t i o n t o show, i n h i s m i s n a m e d a r t i c l e ' M o o r e ' s A r g u -
t h e a r g u m e n t does n o t like t h e c o n c l u s i o n s t o w h i c h h e i s t h e r e b y
m e n t s against Certain F o r m s o f Ethical Naturalism' ( 1 9 4 2 ) t h a t n o n -
forced, he will reject t h e theory, a n d we will be b a c k where we started.
descriptivist theories c a n satisfy this requirement.
I shall call this r e q u i r e m e n t t h e r e q u i r e m e n t of neutrality. I t h i n k t h a t
I m u s t add t h a t , as I said in 4 . 3 , objectivistic n a t u r a l i s m would fail
objectivistic n a t u r a l i s m is t h e only theory, of those e x a m i n e d , w h i c h
this r e q u i r e m e n t if, in order to escape t h e a r g u m e n t I t h e r e advanced
fails this test. It fails it b e c a u s e an objectivist a c c o u n t of t h e t r u t h c o n -
against it, its proponents took refuge in saying t h a t different cultures
120
II. 6. 8
w h o have different m o r e s a r e u s i n g t h e m o r a l words i n different

senses. T h e r e would t h e n really be no disagreement between the cultures other t h a n a merely verbal one.
( 4 ) Fourthly, and closely c o n n e c t e d with the incompatibility requirem e n t (indeed, it is a kind of generalization of it), there must be a place
in the theory for logical relations between moral statements. T h e inc o m p a t i b i l i t y o f t h e s t a t e m e n t t h a t a n a c t i s w r o n g with t h e s t a t e m e n t that it is not wrong is an e x a m p l e of a logical relation. B u t it is
n o t t h e only sort of logical r e l a t i o n t h a t is required. P e r h a p s all
logical r e l a t i o n s a r e r e d u c i b l e to r e l a t i o n s of incompatibility. F o r
example, the relation that we call e n t a i l m e n t or deducibility c a n be so
reduced: a proposition p entails a n o t h e r proposition q if and only if p
is incompatible with not-o. Any ethical theory h a s to admit of logical
r e l a t i o n s of the following s o r t : t h a t t h e two propositions t h a t it is
always wrong to tell lies, and t h a t to say so-and-so would be to tell a
lie. are conjointly incompatible with the proposition that it would n o t
be wrong to say so-and-so. I am n o t going at the m o m e n t to ask what
logical relations hold between m o r a l propositions, or between t h e m
and other propositions; all I insist on is t h a t some should. I say this n o t
only b e c a u s e w i t h o u t s u c h logical relations m o r a l a r g u m e n t would
be impossible (that I am c o m i n g to in a m o m e n t ) , but b e c a u s e it is
q u i t e evident t o a n y o n e w h o k n o w s t h e l a n g u a g e t h a t w e d o u s e
words in s u c h a way t h a t some m o r a l s t a t e m e n t s are i n c o m p a t i b l e
with at least some other moral statements. Let us call this the requirem e n t of logicality. As we saw, it is n o t fully satisfied by various forms
of the emotive theory, though s o m e of them allow subsumptive a r g u m e n t s in moral thinking (1 shall return to this point shortly).
R e v e r t i n g for a m o m e n t to r e q u i r e m e n t ( 2 ) , t h a t of p r a c t i c a l i t y :
having accepted the requirement of logicality, we c a n now put the requirement of practicality in a s o m e w h a t clearer and more convenient
form b y saying t h a t a t l e a s t s o m e m o r a l s t a t e m e n t s h a v e t o h a v e
logical relations with prescriptive speech acts of some sort (e.g. imperatives). I shall not, however, insist for the m o m e n t on this: the looser
way I put it earlier will suffice.
(5) If we put tciiether r e q u i r e m e n t s ( 3 ) and ( 4 ) (incompatibility a n d
II. 6. 8
EMOTIVISM
121
logicality) we are led to a further requirement. T h i s is t h a t o u r ethical

t h e o r y should do s o m e t h i n g to resolve m o r a l d i s a g r e e m e n t s by t h e
use of a r g u m e n t . I deliberately do not say t h a t it should m a k e it possible to resolve all m o r a l d i s a g r e e m e n t s by a r g u m e n t . If we look at
w h a t goes o n i n m o r a l a r g u m e n t s , w e see (if m y e x p e r i e n c e i s a n y
guide) t h a t s o m e disagreements are resolved by a r g u m e n t a n d s o m e
are not. An ethical theory could be wrong in two ways: either by m a k ing it impossible to r e a c h a g r e e m e n t by a r g u m e n t in c a s e s where it is
possible, or by c l a i m i n g t h a t it is possible to prove t h i n g s in m o r a l
a r g u m e n t w h e r e it is n o t possible. We must avoid b o t h of these opposite errors. You m a y r e m e m b e r t h a t in FR 8.1 f. I said t h a t t h e form of
a r g u m e n t I w a s a d v o c a t i n g t h e r e did n o t e n a b l e us to a r g u e a b o u t
ideals w h e r e no o t h e r people's interests a r e affected; if I w a s right,
this m a y be an e x a m p l e of a m a t t e r w h i c h cannot be settled by a r g u m e n t . O n t h e o t h e r h a n d , I have a r g u e d t h a t w h e r e o t h e r people's
interests are affected, c o g e n t a r g u m e n t s a b o u t m o r a l q u e s t i o n s a r e
available (MT pt. 2. H 1 9 9 3 3 ) . So let us call our moderate requirement,
t h a t the t h e o r y s h o u l d do something to resolve m o r a l d i s a g r e e m e n t s
by the use of a r g u m e n t , the arguability requirement.
A theory which fails to satisfy the requirement of logicality c a n n o t
satisfy t h a t o f arguability, i f b y ' a r g u m e n t ' w e m e a n ' l o g i c a l a r g u m e n t ' . S o m e emotivists (Ayer for e x a m p l e , a n d S t e v e n s o n ) do allow
t h e r e t o b e limited forms o f a r g u m e n t a b o u t m o r a l q u e s t i o n s ; b u t
they a r e limited to t h e subsumption of p a r t i c u l a r m o r a l s t a t e m e n t s
under m o r e general ones, a n d in a n y case it is n o t c l e a r w h e t h e r this
is for S t e v e n s o n a m a t t e r of logical derivation or m e r e l y of c a u s i n g
attitudes to c h a n g e by invoking m o r e general attitudes. I t h i n k t h a t
we do allow there to be a r g u m e n t s about m o r a l questions w h i c h are
m o r e ambitious t h a n t h i s a r g u m e n t s w h i c h c a n r e a c h a conclusion
even b e t w e e n people w h o do n o t s h a r e any initial s u b s t a n t i a l m o r a l
opinions; indeed, I s h a l l be s h o w i n g in C h a p t e r 7 h o w this c a n be
done.
As is clear from w h a t I said in Chapter 5 about w h a t happens w h e n
i n t u i t i o n s d i s a g r e e , i n t u i t i o n i s m does n o t satisfy t h e a r g u a b i l i t y
requirement. In fact, intuitionists are in no better position t h a n e m o tivists w h e n i t c o m e s t o a r g u m e n t . T h e d i s p u t a n t s c a n d o n o m o r e
122
II. 6. 8
II. 6 . 9
EMOTIVISM
123
t h a n oppose their intuitions to o n e a n o t h e r . T h e r e c a n be s u b s u m p tion o f p a r t i c u l a r m o r a l s t a t e m e n t s u n d e r m o r e g e n e r a l o n e s ; b u t

even emotivists c a n do that. Naturalism fails this test too, b e c a u s e it
fails to satisfy the n e u t r a l i t y r e q u i r e m e n t ; as I said, if t h e n a t u r a l i s t
proposes an a c c o u n t of the m e a n i n g of a moral word which he thinks
will settle the dispute between two parties, t h e party t h a t is defeated
will a t o n c e r e j e c t t h e n a t u r a l i s t ' s a c c o u n t . T h e n a t u r a l i s t h a s n o
neutral standpoint from which to adjudicate between t h e m .
c h a n g e in t h e m e a n i n g s of t h e m o r a l words. T h i s , as I have said, is a
6 . 9 . So then, all the theories we have so far discussed fail to m e e t

o n e or other of these five requirements. I wish to add to these a sixth
requirement. T h i s is of a s o m e w h a t different c h a r a c t e r to t h e others,
being a practical rather t h a n a theoretical requirement.
does c h a n g e with t i m e ; for e x a m p l e , t h e universalizability of m o r a l
m i s t a k e to w h i c h descriptivists are prone, a n d it leads to relativism.

People c a n c h a n g e their moral opinions, even quite radically, without
c h a n g i n g t h e m e a n i n g , a p a r t from t h e descriptive m e a n i n g , o f t h e
moral words that they use. I gave as an example t h e Christian precept
t h a t we should love o u r e n e m i e s ; to a c c e p t this is to a l t e r o u r m o r a l
c o n v i c t i o n s radically, b u t it does n o t entail a shift in t h e m e a n i n g of
'should'. But all the s a m e t h e structure and logic of m o r a l l a n g u a g e
s t a t e m e n t s , w h i c h is now, I am sure, a l o g i c a l f e a t u r e of t h e m o r a l
words, was n o t always so. Probably it h a s b e c o m e so in t h e c o u r s e of
h i s t o r y as a r e s u l t of C h r i s t i a n t e a c h i n g , a n d t h e w o r k of philosophers like Kant. It is a frequent p h e n o m e n o n in l a n g u a g e t h a t sen-
( 6 ) An adequate ethical t h e o r y h a s to m a k e it possible for m o r a l dis-
t e n c e s w h i c h used t o e x p r e s s s y n t h e t i c s t a t e m e n t s c h a n g e t h e i r
c o u r s e and m o r a l t h o u g h t in g e n e r a l to fulfil t h e purpose t h a t t h e y
meaning, so that the statements they express b e c o m e analytic. For
have in society. This is to e n a b l e those in society who disagree a b o u t
e x a m p l e , t h e s e n t e n c e 'water is c o m p o s e d of two p a r t s of hydrogen
what t h e y should do, especially in m a t t e r s w h i c h affect their diver-
and one of oxygen' o n c e expressed a synthetic discovery; b u t n o w o n e
gent interests, to reach a g r e e m e n t by rational discussion. I shall call
(but only o n e ) of the senses of 'water' is defined by dictionaries in t h a t
this r e q u i r e m e n t , t h a t m o r a l i t y a n d t h e m o r a l l a n g u a g e s h o u l d b e
way, t h u s m a k i n g t h e s t a t e m e n t t h a t water i s H 0 , i n t h e n e w sense
enabled by our ethical theory to preserve their function of reconciling
of t h e word 'water', analytically true (H 1 9 8 4 b , 1996a"). T h i s p h e n o m -
conflicting interests, the conciliation requirement.
e n o n was well d o c u m e n t e d by von Wright ( 1 9 4 1 : c h . 3 ) .
Moral language, whose m e a n i n g ethics tries to elucidate, is o n e of
T h e function of this remarkable language, t h e l a n g u a g e of morals,
t h e most r e m a r k a b l e i n v e n t i o n s o f t h e h u m a n r a c e , c o m p a r a b l e a t
i s t o help u s s o r t o u t c e r t a i n difficulties w h i c h a r e b o u n d t o a r i s e
least with m a t h e m a t i c a l language. It is not such an a n c i e n t invention
w h e n people live in communities, and in which, therefore, conflicts of
as is sometimes thought. Perhaps il is comparable with m a t h e m a t i c a l
interest inevitably occur. People have desires a n d needs w h i c h c a n n o t
l a n g u a g e in this respect too. t h a t we c a n watch its development dur-
all b e realized b e c a u s e t h e y c o n f l i c t w i t h t h e desires a n d n e e d s o f
ing t h e c o u r s e o f r e c o r d e d history. J u s t a s t h e a n c i e n t G r e e k s h a d
o t h e r people. M o r a l i t y a n d t h e m o r a l l a n g u a g e a r e an invention for
a r i t h m e t i c and Euclidean g e o m e t r y and their languages, but did not
dealing with this situation. I used the word 'invention' in this c o n t e x t
have the c a l c u l u s and its l a n g u a g e , so you may. if you look carefully
long before J o h n M a c k i e used it in t h e title of his v e r y good b o o k
at h o w people talked at various times in history, see t h a t t h e Greeks
Ethics: Inventing Right and Wrong ( 1 9 7 7 ) ; but I agree with h i m that it is
did n o t have a m o r a l l a n g u a g e as fully developed as o u r s , a n d t h a t
an invention, t h o u g h we disagree about s o m e o t h e r things.
o u r present-day moral l a n g u a g e h a s features which were not fully de-
It m i g h t be asked why. if m o r a l l a n g u a g e h a s t h e s e wonderful
veloped (though of c o u r s e there were m o r e primitive forms of t h e m )
properties, it h a s not enabled us already to sort o u t all o u r moral dis-
until perhaps the time of K a n t or even of Mill.
a g r e e m e n t s . T h e a n s w e r i s twofold. First, m a n y o f t h e s e d i s a g r e e -
I am not agreeing here with those who think that a mere alter-
m e n t s are rooted in disagreements about t h e facts, w h i c h in a n y at all
ation in mores (in the m o r a l principles generally accepted) involves a
intractable moral problem are bound to be extremely complex and
124
IL 6. 9
h a r d to establish. But, m o r e importantly, not m a n y people are able to

think clearly about moral questions with an understanding of the
words they are using. T h e y may therefore be expected to get confused,
and indeed examples of such confusions c a n be observed by anybody
w h o reads the newspapers, especially their correspondence c o l u m n s .
And in any case m a n y people do not t h i n k morally at all. but at most
shmorally.
If, as I have c l a i m e d , t h e l a n g u a g e in its fully developed form is a
recent growth, these failings are even easier to understand. And t h e
prevalence of descriptivism a n d other philosophical errors, w h i c h are
bound to some extent to infect public discussions, does not help. I am
sure that if we had better moral philosophy, we should have less public perplexity a n d confusion a b o u t m o r a l questions. B u t 1 am n o t at
all hopeful t h a t this will a c t u a l l y h a p p e n ; t h e r e are t o o m a n y b a d
m o r a l philosophers t h r o w i n g dust in our eyes, a n d all too few good
ones clarifying the issues.
Unlike the others, the conciliation requirement is more of a practic a l r e q u i r e m e n t t h a n a logical o n e , as I said; a n d this is i m p o r t a n t ,
b e c a u s e if I c a n show t h a t t h e t h e o r y I am going to propose meets it
in practice, t h a t will do. It will n o t be an objection to the t h e o r y if it
c a n be argued that there logically could be communities in which t h e
requirement would not be satisfied.
It is, I think obvious t h a t n o n e of the theories 1 have so far considered c a n satisfy this requirement, b e c a u s e they all fell down on o n e or
other of the r e q u i r e m e n t s 1 have listed: and in particular all of t h e m
failed to satisfy t h e a r g u a b i l i t y r e q u i r e m e n t . C o n c i l i a t i o n t h r o u g h
moral reasoning will clearly be impossible between people who do not
k n o w how to a r g u e morally.
S o t h e n w e have t h e s e six r e q u i r e m e n t s for a n a d e q u a t e e t h i c a l
theory. T h e y are only the o n e s w h i c h seem to me the most important:
people might bring forward o t h e r requirements and think t h e m m o r e
important. In this c o n n e c t i o n I m i g h t mention the so-called publicity
r e q u i r e m e n t by w h i c h Rawls a n d o t h e r s set store. Il is not a requirem e n t for an ethical t h e o r y (Rawls does n o t have an e t h i c a l theory in
my sense), but rather a r e q u i r e m e n t for a substantial m o r a l principle,
namely thru ii o sulci be openly avowed without defeating its object. I
F I
- ^
EMOTIVISM
am not sure t h a t this is a requirement for a m o r a l principle; b u t since

we are do.ng e t h i c a l theory not 'moral t h e o r y ' in Rawls's sense I shall
not discuss it. I t h i n k t h a t the ethical theory I am going to put before
you satisfies all my six requirements, and all o t h e r r e q u i r e m e n t s t h a t
am a w a r e o f - w h i c h is not to say that it is t h e last word in e t h i c a l
t h e o r i e s , b e c a u s e , as always, problems r e m a i n . B u t I t h i n k it is t h e
most adequate e t h i c a l theory I have c o m e across so far.
II. 7.1
RATIONALISM
127
t h a t badly. T h e r e is a n o t h e r very i m p o r t a n t t r u t h he h a s got hold of.

H e h a s g r a s p e d t h a t m o r a l s t a t e m e n t s a r e m a d e a b o u t a c t i o n s for
7
RATIONALISM
reasons, n a m e l y t h a t t h e a c t i o n s have c e r t a i n n o n - m o r a l properties.

An act was w r o n g , for example, because it w a s an a c t of h u r t i n g somebody for fun. T h i s property of moral s t a t e m e n t s , their supervenience
on non-moral statements, is crucial to an understanding of them.
B u t t h e objectivistic naturalist h a s m i s u n d e r s t o o d t h e n a t u r e o f t h e
7 . 1 . Up to n o w this part of my book h a s been mostly devoted to fault-
' b e c a u s e ' . He mistakes supervenience for e n t a i l m e n t , a n d t h u s m a k e s
finding. I am n e x t going to m a k e a m e n d s by telling you w h a t I t h i n k
into analytically true statements what are really substantial moral
are t h e virtues of t h e theories I have discussed. T h i s is not just in order
principles. T h a t it is w r o n g to h u r t people for fun is n o t an a n a l y t i c
to be fair, nor just in order to s h o w my good nature. I have two ulter-
statement. B u t still t h e a c t is wrong because it was t h a t sort of act. So
ior m o t i v e s . T h e first is self-protective. T h e best way of p r o t e c t i n g
the objectivistic naturalist h a s hit on, t h o u g h he h a s n o t fully under-
o n e ' s own theory is to incorporate into it all the truths t h a t upholders
stood, the supervenience or consequentiality of m o r a l properties, a n d
of rival theories insist on (H I 9 9 4 & ) - T h e n they are less likely to a t t a c k
t h u s i s o n t h e t r a c k o f t h e universalizability o f m o r a l s t a t e m e n t s
o n e . and will not be successful if they do. T h e second is constructive.
which lies at its root (on supervenience see 1 . 7 , H 1 9 8 4 b , 1996(f). T h i s
If. as I believe, n e a r l y all e t h i c a l t h e o r i e s c o n t a i n s o m e e l e m e n t s of
i m p o r t a n t feature o f m o r a l s t a t e m e n t s w e h a v e t o i n c o r p o r a t e i n t o
truth, t h e best way of c o n s t r u c t i n g a viable one is to pick out t h e true
our o w n theory.
e l e m e n t s in e a c h a n d build t h e m i n t o o n e ' s o w n theory. I advise all
Next t h e subjeetivistic naturalists. T h e y let slip t h e important t r u t h
t h o s e w h o w a n t to m a k e a c a r e e r in philosophy to do t h i s . A good
I have j u s t found in objectivistic naturalism. For all t h a t they say, it is
politician tries to steal his e n e m i e s ' c l o t h e s , and a good p h i l o s o p h e r
a sufficient r e a s o n for saying t h a t an a c t is w r o n g t h a t you disapprove
does t h e s a m e . He looks carefully at all t h e theories t h a t h a v e b e e n
of it; it does n o t have to be w r o n g b e c a u s e of a n y t h i n g a b o u t it except
put forward and asks himself w h a t is t r u e in e a c h of them; if he c a n
t h a t . B u t n e v e r t h e l e s s t h e subjectivist h a s got hold o f a n i m p o r t a n t
t h e n lay hold o n t h e s e t r u t h s a n d avoid t h e e r r o r s w h i c h t h e r e a r e
truth, n a m e l y t h a t something in the attitudes of t h e speaker goes into
also likely to be, he will have a defensible theory. Veritati omnia consen-
t h e m a k i n g of a m o r a l s t a t e m e n t . Subjectivists h a v e n o t understood
tiunt. T h i s is of c o u r s e difficult, b e c a u s e in m o s t theories t h e t r u t h s
very well w h a t this s o m e t h i n g is; t h e emotivists understooddtbetter,
are closely meshed with t h e errors, and it is hard to take t h e m apart.
a n d t h e prescriptivists better still; b u t subjectivism w a s a promising
T h e a d h e r e n t s o f t h e theories, w h o have n o t seen t h a t t h e t r u t h s d o
beginning.
n o t entail t h e errors, will always resist this treatment. B u t if o n e c a n
T h e i n t u i t i o n i s t s also h a d hold o f s o m e i m p o r t a n t t r u t h s , b o t h
achieve this kind of benign eclecticism, one will be a successful philo-
negative a n d positive. T h e y insisted, a g a i n s t t h e n a t u r a l i s t s , o n t h e
sopher.
non-analyticity of m o r a l principles, while upholding also their c o n s e -
I will start with t h e t r u t h s in objectivistic n a t u r a l i s m . I have dis-
quentiality. T h e expression 'consequential property' c o m e s , I believe,
cussed t h e m in MT c h . 4, s o l c a n be brief. T h e first is t h a t it grasps t h e
from t h e intuitionists, as does t h e expression 'supervenience', t h o u g h
essential point of ethical theory, w h i c h is, by examining t h e l a n g u a g e
I have n o t b e e n able to locate it in their writings. A l t h o u g h t h e intu-
o f m o r a l s , t h e m o r a l c o n c e p t s a n d t h e i r logic, t o s h o w h o w w e c a n
itionists m i s t o o k c o n d e m n i n g an a c t i o n for perceiving a property of
r e a s o n correctly a b o u t m o r a l questions. S o t h e objectivistic n a t u r a l -
w r o n g n e s s in it, t h e y were right about m a n y of t h e logical properties
ist's project is t h e right o n e , t h o u g h he e x e c u t e s it badly. B u t n o t all
of m o r a l statements. T h e y rightly insisted t h a t T o u g h t ' contradicts 'I
128
II. 7. 1
o u g h t not', a n d t h a t therefore it w a s impossible consistently to a g r e e

with b o t h . T h e r e w a s also a n o t h e r t h i n g right a b o u t t h e m w h i c h I
shall not be able to explain until later, when I have discussed the two
levels of moral thinking, the intuitive and the critical ( 7 . 8 ) . Intuition
h a s an i m p o r t a n t place in m o r a l t h i n k i n g , t h o u g h it is n o t an ultim a t e c o u r t o f appeal a s t h e i n t u i t i o n i s t s t h i n k . And m o s t o f w h a t
they say is correct about the intuitive level of moral thinking.
T h e emotivists. as I have said, took a very i m p o r t a n t step forward
i n e t h i c a l t h e o r y w h e n t h e y r e j e c t e d descriptivism. A l t h o u g h t h e y
m a d e a serious error in trying to explain w h a t moral s t a t e m e n t s do,
they were clear t h a t they do n o t j u s t describe the world. Moral statem e n t s do more than this, but it was left to others to explain w h a t this
m o r e was.
7 . 2 . Taking stock, therefore, o f o u r present position, w e have t h e
following t r u t h s g l e a n e d from t h e t h e o r i e s so far considered w h i c h
h a v e to be i n c o r p o r a t e d i n t o a m o r e a d e q u a t e theory. First, it m u s t
show, by an e x a m i n a t i o n of t h e m e a n i n g s a n d logic of m o r a l words,
h o w we c a n reason a b o u t m o r a l q u e s t i o n s . T h e place of logic in t h e
t h e o r y will b e c r u c i a l , for w i t h o u t i t t h e r e c a n b e n o r e a s o n i n g .
Secondly, it m u s t show how we c a n m a k e moral statements because of
t h e n o n - m o r a l properties o f t h e a c t i o n s , e t c . t h a t w e are s p e a k i n g
a b o u t . In o t h e r words, it m u s t do j u s t i c e to t h e c o n s e q u e n t i a l i t y or
s u p e r v e n i e n c e of m o r a l properties, w h i c h is linked to t h e universalizability of m o r a l s t a t e m e n t s . Thirdly, it m u s t do j u s t i c e to t h e f a c t
t h a t in m a k i n g a m o r a l s t a t e m e n t t h e speaker is himself c o n t r i b u t i n g
something. Morality is not a passive perception of the world. T h e sub-
II-7 2
RATIONALISM
129
t h e r e a l i z a t i o n t h a t i t does n o t conflict w i t h t h e i r logicality. T h a t i s

w h a t I h a v e b e e n after ever s i n c e I s t a r t e d d o i n g m o r a l philosophy,
and I still think it is t h e most important thing we have to understand
if we w a n t to m a k e s e n s e of m o r a l t h i n k i n g a n d m o r a l a r g u m e n t . I
am going in w h a t follows to show why it is prescriptive logic, if I m a y
so call it, t h a t m a k e s rationality in m o r a l t h i n k i n g possible, a n d not,
or at least n o t only, t h e fact t h a t m o r a l s t a t e m e n t s have t r u t h conditions. T h e y have t r u t h conditions, indeed, but if t h o s e were all we h a d
to rely on, we could n o t escape relativism. It is t h e fact t h a t w h e n we
are adopting a m o r a l principle we a r e prescribing t h a t gives r a t i o n a l
m o r a l a r g u m e n t its teeth (H 1 9 9 6 c ) . B e c a u s e , as I have said, t h e prescriptivity o f m o r a l s t a t e m e n t s , unlike t h e i r descriptive m e a n i n g s ,
c a n be a c u l t u r a l l y i n v a r i a n t e l e m e n t in t h e m , it e n a b l e s t h e m to be
used in rational discussion between cultures. I will n o w try to explain
this m o r e clearly.
We saw earlier t h a t the t r u t h conditions of a m o r a l s t a t e m e n t are
inevitably relative to a c u l t u r e a n d its m o r e s a n d l a n g u a g e . If a c e r tain c u l t u r e a c c e p t s c e r t a i n m o r a l principles, t h e y will be e n s h r i n e d
in both its l a n g u a g e a n d its m o r a l e d u c a t i o n , a n d a n y t h e o r y w h i c h
looks for t r u t h conditions in l a n g u a g e or in intuition will be trapped
inside t h e c u l t u r e . W h a t is needed is a w a y of criticizing t h e m o r a l
principles of a c u l t u r e : a w a y of d i s c u s s i n g in r a t i o n a l a r g u m e n t
w h e t h e r w e should o r should n o t a c c e p t t h o s e principles. M a n y c u l tures have m o r a l principles that we o u g h t to reject; but t h e m e m b e r s
o f t h o s e c u l t u r e s , i f t h e y a r e descriptivists, will h a v e n o r e a s o n t o
reject t h e m .
jectivists were h a l f r i g h t a b o u t this, but b e c a u s e t h e y w e r e still
Il is t h e r e q u i r e m e n t to prescribe in a c c o r d a n c e w i t h t h e s e prin-
descriptivists they t h o u g h t that, since in a moral statement o n e is n o t
ciples w h i c h m a k e s us reject some of them. Here 1 am following K a n t
d e s c r i b i n g t h e world, o n e m u s t b e d e s c r i b i n g oneself. Lastly, w h i l e
( 8 , 6 f.). I s u p p o s e m o s t people w h o study K a n t , b e i n g t h e m s e l v e s
rejecting descriptivism like t h e emotivists. and insisting t h a t t h e r e is
descriptivists like P r i c h a r d a n d Ross, read t h e i r o w n prejudices into
something extra to the making of a moral statement beyond the
K a n t a n d do n o t n o t i c e t h a t he is n o t h i m s e l f a descriptivist. In t h e
describing of an a c t i o n or a person in a c c o r d a n c e with t r u t h condi-
m o s t f a m o u s o f h i s f o r m u l a t i o n s o f his C a t e g o r i c a l Imperative, h e
tions, a n adequate e t h i c a l t h e o r y m u s t give a n a c c o u n t o f this e x t r a
says t h a t we are so to act t h a t we c a n will t h e m a x i m of o u r action to
ingredient in m o r a l s t a t e m e n t s w h i c h is c o n s i s t e n t with their b e i n g
b e c o m e a u n i v e r s a l law. T h e will is a prescriptive, n o t a descriptive
subject to logical control.
faculty. In this it is like Aristotle's phronesis. Aristotle himself is half a
This extra ingredient is the prescriptivity of moral statements, a n d
prescriptivist (H 1 9 9 2 c : ii. 1 3 0 4 ; 1 9 9 8 a ) . He c o n t r a s t s phronesis, or
i o
3
H- 7- 2
p r a c t i c a l wisdom, with synesis or u n d e r s t a n d i n g , and says t h a t t h e
IL 7- 3
RATIONALISM
131
of c o n t r a d i c t i o n . For some bad m a x i m s c o u l d slip t h r o u g h even this
former is epitdtike (which actually means 'prescriptive') b u t t h e latter
n e t . For e x a m p l e , I c a n w i t h o u t c o n t r a d i c t i o n will t h a t everybody
is krltike mownit only judges, n o t prescribes. T h e distinction c o m e s
s h o u l d s e e k his o w n selfish a d v a n t a g e a n d p a y n o a t t e n t i o n t o t h e
from Plato's Statesman ( 2 6 0 b ) . B o t h Plato and Aristotle were in part,
needs of o t h e r s ; a n d I t h i n k t h a t s o m e people a c t u a l l y follow this
like K a n t and Mill, prescriptivists, following S o c r a t e s in this (H
m a x i m in their actions, so it is not self-contradictory to say t h a t s u c h
3 9 9 8 a ) . This c o m e s out also in t h e fact t h a t in Aristotle's practical syl-
a m a x i m h a s b e e n obeyed. Of c o u r s e it goes w i t h o u t s a y i n g t h a t we
logism t h e c o n c l u s i o n is an a c t i o n or prescription for a c t i o n . If it is.
have to avoid contradiction in our m a x i m s ; b u t t h a t is n o t e n o u g h to
and if the syllogism is valid, t h e n o n e of t h e premisses h a s to be pre-
keep us on t h e m o r a l rails.
scriptive, and this premiss is obviously t h e first, w h i c h is a m o r a l or
I c o n c l u d e t h a t the ' c a n ' in the Categorical Imperative is not j u s t a
o t h e r normative statement. So Aristotle realized that normative state-
logical ' c a n ' . Is it t h e n a psychological ' c a n ' ? It is said t h a t t h e r e a r e
m e n t s are prescriptions. It also c o m e s out w h e n , at t h e very b e g i n -
also s o m e very bad m a x i m s that some people could bring themselves,
n i n g of t h e Nicomachean Ethics, he says t h a t t h e good is w h a t all
p s y c h o l o g i c a l l y speaking, to will to be u n i v e r s a l laws, at a n y r a t e if
t h i n g s seek; in this too he is following P l a t o . B u t P l a t o ' s a n d
their c i r c u m s t a n c e s were s u c h t h a t they would never be t h e victims
Aristotle's prescriplivism was heavily overlaid with descriptivist ele-
of t h e a c t i o n s prescribed. Could n o t a h a r d - h e a r t e d p e r s o n , w h o
ments; so most c o m m e n t a t o r s have not noticed it.
could be confident that he would never be in t h e position of his v i c -
7 . 3 . T h i s i s obviously not t h e place t o expound Plato o r Aristotle

in detail. I will, however, say a little m o r e a b o u t Kant, additional to
tim, will to go on torturing h i m for fun? S o m e t h i n g m o r e seems to be

needed t h a n logical, and t h a n psychological, possibility.
w h a t 1 say in 8 . 2 I f , b e c a u s e he gives us s o m e i m p o r t a n t c l u e s on
At this point I am going to suggest my o w n solution (already sum-
h o w to discipline m o r a l t h i n k i n g , even t h o u g h it is prescriptive.
marized in 1.8) to this problem. I think I have found it in Kant, t h o u g h
Descriptivists think they c a n discipline it by insisting that it obey t r u t h
it is n o t t h e only possible interpretation of his text, and indeed differ-
c o n d i t i o n s : b u t as we h a v e seen this o n l y lands t h e m in r e l a t i v i s m .
ent passages c a n be interpreted in different ways. T h i s is no place for
K a n t speaks very seldom, if ever, a b o u t t h e truth or the t r u t h condi-
detailed e x e g e s i s o f K a n t , a n y m o r e t h a n o f A r i s t o t l e o r P l a t o . I
t i o n s o f m o r a l s t a t e m e n t s , o r a b o u t m o r a l facts. H e speaks a b o u t
return to K a n t in Chapter 8. B u t w h a t I think he ought to have said is
w h a t we c a n will as a universal law.
this. If we have to will our m a x i m s as universal laws, we have to will
T h e will, as I said, is a prescriptive faculty. T h e n a t u r e of t h e disci-
that they should be observed in all situations resembling o n e a n o t h e r
pline imposed on it by t h e C a t e g o r i c a l Imperative will be revealed if
in t h e universal features specified in t h e m a x i m . T h e s e features will
we can understand w h a t K a n t m e a n t by 'can will'. W h a t sort of pos-
include features of t h e psychological states of t h e people in t h e situa-
sibility is he thinking of? Unfortunately he is n o t entirely c o n s i s t e n t
tions. For e x a m p l e , if we a r e speaking of v i c t i m s of torture, t h e fact
on this, and he h a s at least two a c c o u n t s of the matter, o n e of w h i c h
t h a t they badly w a n t t h e torture to stop is a feature of t h e i r psycho-
does n o t help m u c h . T h i s is t h e a c c o u n t w h i c h says t h a t t h e restric-
logical states, a n d therefore of their situations. We h a v e t h e n to will
tion on the will is simply t h a t its m a x i m s have to be logically consist-
t h a t o u r m a x i r i l s s h o u l d b e observed w h a t e v e r individuals w e r e i n
ent, in the sense t h a t to say t h a t the m a x i m s had been obeyed would
t h e s e s t a t e s , even o u r s e l v e s . T h i s helps t o e x p l a i n t h e ' c a n ' i n t h e
not involve a self-contradiction. As m a n y c o m m e n t a t o r s have pointed
Categorical Imperative.
o u t . t h e r e a r e s o m e very w i c k e d m a x i m s t h a t logically c o u l d b e
C a n I will t h a t if I were being t o r t u r e d t h e t o r t u r e r should go on
obeyed, so t h a t this discipline for m o r a l thinking is inadequate. It will
torturing me? Let us suppose t h a t no o t h e r considerations c o m e into
n o t even do to say t h a t the universalized form of a m a x i m h a s to be free
my t h i n k i n g . F o r e x a m p l e , it is n o t t h e c a s e t h a t I t h i n k I deserve
II. 7- 3
punishment, or think I should deserve it if I were in that situation

o n e in which, say. I have c o m m i t t e d a heinous crime. Suppose it just
is the case that the torturer enjoys torturing me. I do not think t h a t I
c a n will this. T h e reason is n o t just psychological impossibility. For 1
suppose that o n e might find somebody who, empirically speaking, did
will t h a t he should h i m s e l f go on b e i n g t o r t u r e d if he were ever in
t h a t situation. I am not speaking of masochists: they are supposed to
w a n t to be t o r t u r e d , w h i c h is n o t t h e position of the v i c t i m in o u r
example. I dare say I might w a n t to be tortured just to see w h a t it was
like: but that is not our present case, b e c a u s e the torture h a s already
s t a r t e d and I am asking w h e t h e r I c a n will that if 1 were b e i n g t o r tured the torturer should go on torturing me. I repeat: I do not think I
c a n will this. It is n o t j u s t t h a t I happen n o t to like being t o r t u r e d .
Torture is by definition a c a u s e of suffering: if it is not a c a u s e of suffering it is not torture. And suffering is s o m e t h i n g that by definition
the sufferer wants to stop: if he does not w a n t it to stop (other things
being equal, of c o u r s e ) it is not suffering. So at any rate at the tune the
sufferer c a n n o t will that the torture should go on. other things being
equal.
B u t c a n I in advance, and for a hypothetical situation, will t h a t the
torture should go on? This h a n g s on a tricky question about personal
identity (i . 8 ) . I am inclined to the view that, if I will t h a t the t o r t u r e
should go on in the hypothetical situation, I am not t h i n k i n g of t h e
victim as myself. As I said in MT, there are several different criteria of
personal identity which in nearly all cases coincide, so t h a t no probl e m arises: t h e y c o m e a p a r t i n p h i l o s o p h e r s ' e x a m p l e s , a n d s o m e times in rare brain disorders, and then we do not know w h a t to say.
O n e of t h e s e features ( t h o u g h n o t e x a c t l y a criterion) of p e r s o n a l
identity is this: I have, if I am t h i n k i n g of a possible future person as
myself, to identify with h i m (or h e r ) to the e x t e n t of preferring t h a t
his preferences should be satisfied. T h a t is, it is part of the c o n c e p t of
personal identity that each person h a s an interest in his own future. If
s o m e o n e h a s lost interest in his own future, he is to t h a t e x t e n t n o t
thinking of the future person as himself.
From this I conclude that unless 1 will, now, that the torture should
n o t go on in the hypothetical situation, I am not thinking of the tor-
11
7 3
RATIONALISM
l u r e v i c t i m as me. To this add the idea, put forward in detail in M T ,

that unless I fully represent to myself w h a t it is like for me. to be in that
situation, I am not in full possession of the facts a b o u t the situation.
T h e a r g u m e n t c a n then get going. It follows t h a t if I am in full possession of the facts about the situation (and of course, if I am not, I c a n
be faulted for i g n o r a n c e of the facts), t h e n I shall n o t be able to will
t h a t the torture go on in the hypothetical c a s e in w h i c h I am the victim. This is the first part of the explanation of the ' c a n ' as it figures in
my v e r s i o n of K a n t ' s C a t e g o r i c a l I m p e r a t i v e . I c a n n o t will t h a t I
should be treated like t h a t if I were the victim. B u t this does n o t take
us the whole way. T r u e , I will that I should n o t go on being tortured if
I were in t h a t s i t u a t i o n . B u t this says n o t h i n g a b o u t w h a t I c a n or
c a n n o t will s h o u l d b e d o n e t o t h e o t h e r p e r s o n w h o i s a c t u a l l y t h e
victim. For all t h a t we have said so far I c a n will t h a t he should go on
being tortured.
7 . 4 . I have been merely summarizing, so far, the a r g u m e n t of MT
c h . 5. My o b j e c t is twofold: first, to s h o w the absolutely c r u c i a l part
t h a t prescriptivity plays i n t h e e n t i r e a r g u m e n t : a n d secondly, t o
relate my own a r g u m e n t to K a n t ' s . I hope you will have noticed t h a t
a descriptivist, e v e n if he believed in universalizability, as m a n y
descriptivists do, could n o t use the a r g u m e n t I have b e e n using so far.
T h e q u e s t i o n ' C a n I will?' h a s b e e n c e n t r a l ; a n d willing is a kind of
prescribing. It does n o t e n t e r into the descriptivist's vocabulary. For
me, as for K a n t , the point is n o t t h a t a c e r t a i n kind of a c t i o n c a n n o t
be described, or even described as universally o c c u r r i n g , without selfcontradiction, but t h a t it c a n n o t be willed or prescribed universally.
But we have not yet brought universalizability into the a r g u m e n t .
Many people w h o have read MT carelessly have supposed t h a t it plays
an e s s e n t i a l p a r t in t h e a r g u m e n t of MT c h . 5. T h i s is n o t so. T h a t
chapter establishes something about prescriptivity, n o t about universalizability. Universalizability only e n t e r s e s s e n t i a l l y into t h e a r g u m e n t in MT c h . 6. B u t if the ground had n o t been prepared in MT c h .
5 by e s t a b l i s h i n g t h e thesis t h a t t h e r e is s o m e t h i n g t h a t we c a n n o t
prescribe, n a m e l y t h a t we ourselves should be t o r t u r e d in t h e hypot h e t i c a l s i t u a t i o n (indeed s o m e t h i n g t h a t w e s h a l l n e c e s s a r i l y prescribe, n a m e l y t h a t we o u r s e l v e s s h o u l d not be t o r t u r e d ) , t h e
134
A TAXONOMY
OF
ETHICAL THEORIES
II.
7- 4
I I . 7.4
RATIONALISM
^35
a r g u m e n t from universalizability in MT c h . 6 would get no grip.
tailored to o u r o w n interest; a n d it is also said, m o r e generally, t h a t it
Descriptivists therefore could n o t deploy it. You will now, I hope, see
does n o t stop us advocating m a x i m s t h a t we all find o u t r a g e o u s .
t h e tactics of my general a r g u m e n t . In Chapters 4 and 51 showed t h a t
An e x a m p l e of t h e first kind of a r g u m e n t is this: suppose s o m e o n e
descriptivism of all sorts collapsed into relativism and could n o t yield
says t h a t he prescribes universally t h a t people with six fingers on o n e
objectivity in moral statements. T h e n I expressed the hope t h a t a n o n -
h a n d should be given special privileges, a n d t h a t he h a s six fingers on
descriptivist theory could yield this objectivity. By 'objectivity' I m e a n
o n e h a n d . His m a x i m is formally impeccable; it is a universal prescrip-
n o t ' c o r r e s p o n d e n c e w i t h t h e facts' or a n y t h i n g like that. I leave all
tion in t h e fullest sense of 'universal'. T h e m a x i m is n o t in itself c o n -
t h a t to t h e descriptivists; it is a dead end. I m e a n , rather, by ' o b j e c t -
t r a d i c t o r y ; so, i f w e t o o k t h e line o f s o m e i n t e r p r e t e r s o f K a n t , w e
ive', 'such as a n y rational thinker in possession of the non-moral facts
could n o t fault it. B u t t h e question is, 'Can he adopt this m a x i m if he
must agree to'. If we could show that some moral statements have
is in full c o g n i z a n c e of t h e facts, w h i c h he c a n o n l y be if he h a s fully
this property, I s h o u l d be c o n t e n t . In this sense, t h o u g h n o t in t h e
represented to himself t h e situations of those w h o would be adversely
sense t h a t Mackie denied the possibility of them, I shall m a i n t a i n t h a t
affected by his adoption of this m a x i m ? ' As we h a v e seen, full repres-
there c a n be objective prescriptions ( H 1 9 9 3 0 ) -
e n t a t i o n o f a n o t h e r ' s s i t u a t i o n involves f o r m i n g p r e f e r e n c e s a s t o
If we add to t h e foregoing a r g u m e n t t h e requirement to universal-
w h a t should h a p p e n to oneself, were o n e in t h a t situation; a n d this,
ize our prescriptions, t h e a r g u m e n t c a n be completed. It will be obvi-
c o m b i n e d w i t h t h e r e q u i r e m e n t t o universalize o n e ' s m a x i m s , will
ous to you h o w it will go. If I c a n n o t will t h a t I should be treated in
c a u s e h i m to reject his proposed m a x i m . For he will form preferences
that way in that situation, t h e n I c a n n o t will universally t h a t whoever
that, were he in t h e positions of all those w h o will suffer if he gets his
is in t h a t situation should be treated in t h a t way. S o . if I am going to
privileges, t h e privileges should be withdrawn. A n d these preferences
m a k e a m o r a l s t a t e m e n t a b o u t t h e situation, a n d m o r a l s t a t e m e n t s
are incompatible with t h e retention of his m a x i m .
are universal prescriptions, t h e n h e r e is o n e t h a t I c a n n o t make. I c a n
Here is an e x a m p l e of t h e second kind of a r g u m e n t , w h i c h is very
of c o u r s e assent to singular prescriptions: I c a n want to go on tortur-
similar. It is said t h a t t h e prescription to keep all b l a c k people in sub-
ing my victim; but I c a n n o t say t h a t I o u g h t to.
j e c t i o n is formally universal, a n d internally c o n s i s t e n t , a n d so is n o t
Here, as you will realize, I should have to deal with t h e position of
ruled o u t b y t h e C a t e g o r i c a l Imperative. B u t t h e point is: c a n s o m e -
t h e a m o r a l i s t w h o will n o t m a k e a n y m o r a l s t a t e m e n t s a t all a b o u t
body w h o h a s fully r e p r e s e n t e d K> h i m s e l f t h e s i t u a t i o n of b l a c k
t h e s i t u a t i o n ; b u t I h a v e discussed h i m e n o u g h in MT 1 0 . 7 f f , H
people w h o are kept in subjection go on willing t h a t they should be so
19890", a n d 1 9 9 6 c . I have a l s o discussed (in 5.8) the position of t h e
treated? For i f h e h a s fully r e p r e s e n t e d t h i s t o himself, h e will h a v e
shmoralist. Leaving t h e m aside, a n d confining our attention to t h o s e
formed a p r e f e r e n c e t h a t he should n o t be so t r e a t e d if he is a b l a c k
who want to make some moral statement about the situation, we
person; a n d t h i s is i n c o n s i s t e n t w i t h t h e u n i v e r s a l form of t h e pro-
have at least ruled out one, n a m e l y t h a t I o u g h t to go on with t h e tor-
posed m a x i m . T h e r e i s o f c o u r s e t h e problem o f t h e f a n a t i c a l b l a c k -
ture. T h i s is a conclusion with w h i c h all rational thinkers m u s t agree,
h a t e r w h o is prepared to prescribe t h a t t h e m a x i m should be followed
t h a t is, an objective conclusion.
even if he h i m s e l f w e r e a b l a c k person. I h a v e discussed t h e c a s e of
In w h a t I have just been saying there lies the answer to two r a t h e r
this f a n a t i c at l e n g t h in my books (e.g. MT 1 0 . 3 ) , a n d I t h i n k I h a v e
w e a k a r g u m e n t s t h a t o n e often c o m e s a c r o s s . T h e y a r e a r g u m e n t s
s h o w n t h a t m y t h e o r y c a n deal with him; but t h e r e i s n o t i m e n o w t o
against t h e kind of t h e o r y I am putting forward. Sometimes it is said
go into t h a t problem. At any rate t h e K a n t i a n move c a n be used in ar-
t h a t i f all t h a t i s r e q u i r e d i s t h a t w e s h o u l d prescribe s o m e m a x i m
g u m e n t s w i t h ordinary non-fanatical people.
universally, this does n o t stop u s a d v o c a t i n g m a x i m s w h i c h a r e
7.5.
I have perhaps taken things in t h e w r o n g order by telling you
i 6
3
H. 7- 5
h o w o n e could a r g u e on t h e basis of t h e theory I propose w i t h o u t first

telling you w h a t the t h e o r y is, except in so far as I said it was an adaptation of Kant's Categorical Imperative. Let me now, therefore, try to
formulate the theory m o r e clearly. T h e first thing I need to say a b o u t
it is t h a t it gives a completely formal a c c o u n t of the m e a n i n g s of t h e
moral words. By this I m e a n t h a t it defines t h e m purely on t h e basis of
their logical properties. C o n t r a s t n a t u r a l i s m , which defines t h e m or
explains their m e a n i n g in terms of substantial non-moral properties.
RATIONALISM
II- 7- 5
137
scriptivist theories. B e c a u s e the a c c o u n t it gives of m o r a l l a n g u a g e is

formal, a n d in p a r t i c u l a r b e c a u s e it incorporates t h e formal features
of prescriptivity and universalizability, its a c c o u n t c a n be accepted by
different c u l t u r e s with different moralities. T h e s e formal features c a n
be c o m m o n to all their languages, even t h o u g h t h e m a t e r i a l features
o f their moralities, a n d with t h e m t h e descriptive m e a n i n g s o f their
moral words, and t h e truth conditions of their m o r a l s t a t e m e n t s , differ widely.
Let us take ' o u g h t ' as an e x a m p l e . 'Ought', I w a n t to say, is a logical
S i n c e there is a deontic logic, t h e t h e o r y satisfies r e q u i r e m e n t ( 4 ) ,
word. It is a deontic m o d a l operator. Its logical properties a n d func-
logicality. A n d in particular it satisfies r e q u i r e m e n t ( 3 ) , incompatibil-
tion are closely a n a l o g o u s to t h o s e of t h e o t h e r modal operators like
ity. Just as t w o people w h o say, o n e of t h e m t h a t s o m e proposition is
'it is n e c e s s a r y that'. T h e difference is that, whereas t h e o t h e r m o d a l
necessarily true, a n d t h e other that it is n o t necessarily true, a r e really
o p e r a t o r s govern descriptive s t a t e m e n t s , ' o u g h t ' governs p r e s c r i p -
contradicting e a c h other, so two people w h o say. o n e of t h e m t h a t an
tions ( M T 1.6). ' O u g h t ' - s t a t e m e n t s entail imperatives with t h e s a m e
a c t i o n is obligatory, a n d t h e o t h e r t h a t it is n o t obligatory, a r e really
c o n t e n t , j u s t a s s e n t e n c e s b e g i n n i n g 'It i s n e c e s s a r y t h a t . . . ' e n t a i l
contradicting e a c h other. I have already begun to show that the
indicative statements with t h e s a m e c o n t e n t . So we could s u m m a r i z e
t h e o r y satisfies r e q u i r e m e n t ( 5 ) , arguability, b y s h o w i n g h o w a r g u -
my a c c o u n t of ' o u g h t ' by s a y i n g t h a t it is t h e modality s t a n d i n g to
ments on K a n t i a n lines, using the universalizability and t h e prescrip-
prescriptions as 'necessary' stands to descriptive statements.
tivity o f m o r a l s t a t e m e n t s , c a n b e c o n d u c t e d . I h a v e a l s o a l r e a d y
It seems very n a t u r a l to say t h a t 'ought' is a deontic modality. T h i s
shown t h a t t h e t h e o r y g u a r a n t e e s prescriptivity.
m a k e s r a t h e r implausible t h e a r g u m e n t used by s o m e descriptivists
B u t it g u a r a n t e e s universalizability t o o . If ' o u g h t ' b e h a v e s like a
t h a t , s i n c e 'His a c t w a s w r o n g ' s o u n d s like a n o r d i n a r y s u b j e c t -
n e c e s s i t y o p e r a t o r g o v e r n i n g imperatives, i t will b e t h e c a s e t h a t
predicate s e n t e n c e , its s u r f a c e g r a m m a r supports t h e v i e w t h a t
' o u g h t ' - s t a t e m e n t s will b e universalizable, j u s t a s s t a t e m e n t s o f
w r o n g n e s s is a p r o p e r t y in t h e o r d i n a r y sense, and t h u s s u p p o r t s
necessity are. One c a n n o t say that in such-and-such a c a s e some-
e t h i c a l r e a l i s m . T h e fact t h a t 'His a c t was w r o n g ' m e a n s m u c h t h e
thing is necessarily so, but t h a t there could be a n o t h e r identical c a s e
s a m e as 'He o u g h t not to have done w h a t he did', which h a s a totally
in which it was n o t necessarily so. T h i s is t r u e about logical necessity;
different surface structure, o u g h t to m a k e us at least wonder w h e t h e r
if one s t a t e m e n t is t r u e by logical necessity, t h e n a n y o t h e r s e n t e n c e
t h e surface structure of t h e first s e n t e n c e is a good guide to its m e a n -
of t h e s a m e logical form will also be n e c e s s a r i l y t r u e . It is a l s o t r u e
ing and logic.
a b o u t c a u s a l necessity. If o n e event follows a n o t h e r by c a u s a l n e c e s -
If 'ought' (which I shall take as typical of moral words) is a deontic
sity, then an exactly similar event in identical c i r c u m s t a n c e s m u s t by
modality governing imperatives, a n d behaves like the necessity oper-
c a u s a l necessity be followed by an e x a c t l y similar event. T h a t is pre-
ator, t h e n 'ought'-sentences will entail imperatives, and h e n c e be pre-
cisely a n a l o g o u s t o w h a t t h e thesis o f universalizability holds t o b e
scriptive, and h e n c e satisfy my requirement ( 2 ) , practicality. S i n c e t h e
t r u e o f ' o u g h t ' - s t a t e m e n t s ( H 1 9 8 4 f t ) . S o t h e t w o m a i n props o f
properties ascribed to m o r a l words by my t h e o r y a r e purely formal,
K a n t i a n a r g u m e n t s , universalizability a n d prescriptivity, a r e b o t h
logical properties, it will also satisfy requirement (1), neutrality. Logic
provided by the theory.
is neutral between substantial opinions or claims. This, you will realize, is how my theory escapes t h a t relativism t h a t is t h e fate of all de-
7.6.
It is n o w time to revert to t h e s u b j e c t of t r u t h conditions, a n d
to c o n n e c t t h e m w i t h t h e feature of universalizability w h i c h I h a v e
13
II. 7- 6
j u s t been discussing. You will r e m e m b e r t h a t earlier I spoke of an ele-
II. 7.6
RATIONALISM
139
a r e two levels o f m o r a l t h i n k i n g , t h e c r i t i c a l a n d t h e intuitive, o f
m e n t in t h e m e a n i n g of m o r a l s t a t e m e n t s w h i c h I called, following
w h i c h the s e c o n d is a necessity for h u m a n s (e.g. MT 2 . 1 f t ) . T h e fact
S t e v e n s o n , t h e descriptive m e a n i n g . I also said that t h e objectivistic
t h a t m o s t of o u r m o r a l thinking is at t h e intuitive level will help e x -
naturalists h a d hold o f a n i m p o r t a n t t r u t h about this. T h o u g h t h e y
plain why we c a l l m o r a l j u d g e m e n t s true or false. To t a k e H a r m a n ' s
w e r e w r o n g i n t h i n k i n g t h a t a p p l i c a t i o n c o n d i t i o n s for t h e m o r a l
example ( 1 9 7 7 : 4 ) , faced with some boys w h o have poured petrol over
words gave the meaning of m o r a l s t a t e m e n t s , they were right to hold
a c a t a n d set it a l i g h t , j u s t for fun, we h a v e no h e s i t a t i o n in s a y i n g
t h a t m o r a l words do have application c o n d i t i o n s , and t h a t t h e s e do
t h a t they have done w r o n g . Or to take t h e e x a m p l e I used myself, we
d e t e r m i n e t r u t h c o n d i t i o n s for m o r a l s t a t e m e n t s . T h e y w e r e a l s o
do not hesitate to say t h a t it would be w r o n g to slip a w a y from a petrol
right, against t h e intuitionists a n d t h e subjectivists, to hold t h a t these
station w i t h o u t paying ( 5 . 2 ) . We do n o t feel t h e n e e d to criticize t h e
a p p l i c a t i o n c o n d i t i o n s c a n b e given i n n o n - m o r a l , objective t e r m s .
s t a n d a r d s or principles w h i c h m a k e us say these things, or to ask why
T h e y were wrong, however, to t h i n k t h a t to say w h a t the application
t h e s e a c t s a r e w r o n g . So it is easy to say t h a t it is obviously true t h a t
conditions were was a m e r e m a t t e r of definition; it is in fact to state a
t h e y are w r o n g . T h a t is w h a t gives i n t u i t i o n i s m its plausibility, a n d
moral principle of substance.
h a s got it t h e n a m e of 'the m o r a l philosophy of t h e m a n in t h e street'.
It should be evident by n o w t h a t t h e s a m e a n i m a l is h e r e appear-
It also gives s o m e plausibility to o b j e c t i v i s t i c n a t u r a l i s m . S u c h a
ing in different m e t a m o r p h o s e s . It does n o t m a k e a n y difference
n a t u r a l i s t m i g h t say, ' I f you d o n ' t k n o w t h a t t h o s e a c t s a r e w r o n g ,
w h e t h e r we speak of criteria of application for a m o r a l word (for e x -
c a n you really u n d e r s t a n d t h e m e a n i n g o f " w r o n g " ? ' W h a t h a s h a p -
a m p l e ' w r o n g ' ) , o r a b o u t t h e word's descriptive m e a n i n g , o r a b o u t
pened in s u c h c a s e s is t h a t t h e descriptive m e a n i n g h a s t a k e n over,
t h e t r u t h c o n d i t i o n s of s t a t e m e n t s c o n t a i n i n g it, or a b o u t a m o r a l
almost. It h a s n o t taken over completely, b e c a u s e e v e n a person w h o
s t a n d a r d or universal m o r a l principle. To e n d o r s e any of t h e s e is to
thinks it 'obviously w r o n g ' to do these things will also n o r m a l l y agree
m a k e a synthetic, substantial, m o r a l statement. T h e c r u c i a l question
t h a t s o m e o n e w h o t h i n k s an a c t w r o n g will t h i n k t h a t t h a t is a reason
for ethical theory is how we are to set about deciding rationally which
for n o t d o i n g it. S o t h e prescriptivity, t h o u g h s u b m e r g e d b y t h e
criteria, or t r u t h conditions, or standards, or principles, to e n d o r s e .
descriptive m e a n i n g , is still t h e r e . B u t if t h e m o r a l s t a t e m e n t is u n -
As we have also seen, different c u l t u r e s will have different o n e s . B u t
questioned, it is easy to be either an intuitionist or an objectivistic
we n o w have a way, a r e c o g n i z a b l y K a n t i a n way, of a d j u d i c a t i n g
naturalist. If o n e is n o t a philosopher, o n e will n o t even ask which of
between t h e m . So o u r theory no longer leads to relativism. And it Was
t h e s e o n e is. I hope y o u will see h o w n a t u r a l it is to say, a b o u t s u c h
t h e introduction of prescriptivity a n d its logic t h a t made this possible.
cases, t h a t t h e m o r a l statements made are true.
B u t why is it appropriate to speak of truth in this context? We shall
7.7.
T h e r e is a p r o b l e m w h i c h troubled S o c r a t e s a n d Aristotle,
not understand this until we have looked quite deeply into t h e h u m a n
and is t h o u g h t to c r e a t e difficulties for m o d e r n prescriptivists a n d in-
c i r c u m s t a n c e s t h e social e n v i r o n m e n t i n which we do o u r m o r a l
t e r n a l i s t s like m e . A n i n t e r n a l i s t , w h o t h i n k s t h a t t o h o l d a m o r a l
thinking. All the K a n t i a n a p p r o a c h immediately yields are maxims or
opinion is to be motivated to a c t on it, or to w a n t o t h e r s to a c t on it,
universal prescriptions. T h e r e is no obvious reason yet why we should
and a prescriptivist, w h o thinks t h a t moral s t a t e m e n t s entail imperat-
call these true or false. Could we not, in our m o r a l thinking, m a k e do
ives, a r e t h o u g h t to be in difficulties a b o u t people w h o do w h a t t h e y
just with such universal imperatives, a n d forget about t r u t h a n d false-
a c k n o w l e d g e t o b e w r o n g . T h e r e a r e s o m e people w h o d o w h a t t h e y
h o o d ? T h e a n s w e r i s t h a t a r c h a n g e l s p e r h a p s could d o t h i s , b u t
think w r o n g b e c a u s e they are in pursuit of o t h e r ends t h a t they c a n -
humans cannot.
not attain w i t h o u t doing wrong; a n d there a r e o t h e r s w h o s e end is to
I have in my o t h e r writings suggested, following Plato, t h a t t h e r e
do wrong, just b e c a u s e it is wrong. Let us call t h e first kind of person
i o
4
H. 7- 7
II. 7. 7
RATIONALISM
141
t h e acratic a n d t h e s e c o n d t h e satanist. I h a v e dealt at l e n g t h w i t h
times, like t h e t i m e o f S o c r a t e s a n d like o u r o w n time, m o r a l i t y c a n
t h e s e c h a r a c t e r s i n o t h e r p l a c e s ( F R c h . 5 , M T 3 . 7 , H 19920': c h . 6 ,
actually be in d a n g e r b e c a u s e of the failure to a s k a n d a n s w e r t h e m .
19920: ii. 1 3 0 4 , 1995ft, I 9 9 6 e ) . B u t I should like to point out here t h a t
S o c r a t e s t h o u g h t t h a t t h e way to answer t h e m w a s by a n e w kind of
t h e existence of these two levels of moral thinking makes the problenv
thinking: w h a t I am calling critical thinking.
m u c h easier to solve. If m o r a l statements have firm descriptive m e a n -
T h e t a s k of critical t h i n k i n g is to e x a m i n e t h e v a r i o u s standards,
ings and truth conditions, it is easy to see how someone could have a
o r a p p l i c a t i o n c o n d i t i o n s , o r criteria, o r t r u t h c o n d i t i o n s , o r prin-
grasp of t h e t r u t h of a m o r a l s t a t e m e n t but act c o n t r a r y to t h e pre-
ciples, t h a t we find in a given culture, and see w h e t h e r t h e y c a n be de-
scription contained in it.
fended. I n c r i t i c a l t h i n k i n g there c a n b e n o appeal t o i n t u i t i o n s , n o r
Suppose t h a t t h e person at t h e petrol station is sorely tempted to
to descriptive m e a n i n g s . T h e y are w h a t are being e x a m i n e d . Reliance
slip a w a y w i t h o u t paying, b e c a u s e he w a n t s to keep his m o n e y . He
on t h e m will always land us in relativism. T h a t , in t h e end, is why we
might say that he knows t h a t it would be wrong, or that it is true t h a t
have to reject all forms of descriptivism. T h e move w h i c h e n a b l e s us
it would be wrong. His intuition assures h i m of this. He h a s the exper-
to e x a m i n e t h e m objectively, w i t h o u t being trapped in o u r o w n c u l -
ience of 'knowing it would be w r o n g ' which intuitionists call ' m o r a l
ture, is t h e K a n t i a n move, t h e introduction of prescriptivity, a n d in
intuition'. How easy it is for s o m e b o d y in s u c h a situation to i g n o r e
p a r t i c u l a r of universal prescriptivity. T h i s formal r e q u i r e m e n t , c o m -
the prescriptivity of the s t a t e m e n t and so do the act which he 'knows
m o n t o all c u l t u r e s t h a t ask m o r a l q u e s t i o n s , i s w h a t c o n s t r a i n s u s
is wrong'! Again, suppose t h a t the boys w h o b u r n the c a t k n o w t h a t
objectively. It is w h e n we ask 'Can I prescribe, or will, t h a t this m a x i m
burning the cat satisfies t h e truth conditions accepted in their c u l t u r e
should b e c o m e a universal law?' t h a t we are on firm g r o u n d in o u r
for the statement t h a t it would be wrong. T h a t may be w h a t m a k e s it
moral thinking.
attractive for them, if they are m a l c o n t e n t s or rebels or alienated from
If this d i s c u s s i o n c o u l d go on longer, I w o u l d add s o m e t h i n g to
t h e v a l u e s o f t h e c u l t u r e . B u t i n v i r t u e o f t h e descriptive m e a n i n g
c o n n e c t up t h e formal a r g u m e n t I have been setting o u t with its prac-
and accepted truth conditions of the statement that it would be
tical c o n s e q u e n c e s . T h e s e are very great. In my b o o k s I h a v e s h o w n
wrong, they know it to be true. So, like the m a n at the petrol station,
h o w this formal a r g u m e n t yields rules o f m o r a l r e a s o n i n g t h a t will
t h e y i g n o r e t h e prescriptivity of t h e s t a t e m e n t . B u t for a detailed
bring us to m o r a l principles w h i c h are t h e s a m e as o n e kind of utili-
t r e a t m e n t of such cases I must refer you to my other writings cited.
t a r i a n would c o m e to. I have even been bold e n o u g h to call my o w n
I hope I have explained t h e s e n s e in w h i c h m o r a l s t a t e m e n t s c a n
t h e o r y of m o r a l a r g u m e n t a utilitarian theory, a l t h o u g h it does n o t
be t r u e , or even obviously t r u e . B u t , as I have explained e a r l i e r at
c o n t a i n a n y 'principle of utility', but only a rational method for arriv-
g r e a t l e n g t h , w e c a n n o t stop t h e r e . F o r t h e 'obvious t r u t h ' o f s u c h
ing at these particular moral principles ( 1 9 9 6 c : s.f.).
statements is relative to a culture. Our ancestors did not think it obvi-
In Chapter 81 shall try to s h o w t h a t t h e r e is n o t h i n g paradoxical
ously true t h a t bear-baiting w a s wrong. Spaniards n o w do n o t t h i n k
in a r r i v i n g at u t i l i t a r i a n principles v i a a K a n t i a n m e t h o d . T h o u g h
it obviously t r u e t h a t bullfighting is w r o n g , and their a n c e s t o r s did
K a n t was n o t a u t i l i t a r i a n , t h e r e w a s n o t h i n g i n his t h e o r y o f t h e
n o t t h i n k it obviously t r u e t h a t b u r n i n g people was w r o n g , if t h e y
Categorical Imperative t h a t prevented h i m being one, a n d perhaps he
were heretics. T h e R o m a n s did not think it obviously true t h a t it was
would h a v e been o n e b u t for two t h i n g s . T h e first w a s his strict up-
w r o n g to burn people, if they were Christians. So it does m a k e sense
bringing on highly rigorist principles, w h i c h he n e v e r s h o o k off. He
to ask why moral s t a t e m e n t s , w h i c h we all t h i n k are obviously true,
t h o u g h t he h a d to defend these principles ( s u c h as t h e absolute duty
are true, or even w h e t h e r t h e y really a r e true. Unless we a r e able to
to enforce capital punishment, the absolute a n d exceptionless wrong-
ask and a n s w e r s u c h q u e s t i o n s , o u r m o r a l i t y will b e v u l n e r a b l e . A t
ness of lyingeven t h e sinfulness of m a s t u r b a t i o n ) , by appeal to his
142
II. 7- 7
theory. T h i s h a d r e g r e t t a b l e effects o n t h e w a y h e e x p o u n d e d t h e
theory. B u t in itself t h e t h e o r y is consistent with the adoption of utilitarian principles.
All this would not have been so bad if it had not been for a second
t h i n g t h a t misled h i m . H e s e e m s t o h a v e t h o u g h t t h a t m o r a l principles had to be simple. You m a y r e m e m b e r t h a t in 5.8 I spoke of t h e
confusion that m a n y people still m a k e between universality and generality. T h i s confusion goes b a c k to Aristotle's use of t h e t e r m 'kath'
holou for both c o n c e p t s . I t h i n k K a n t was a victim of this confusion.
It m a y have led h i m to insist t h a t m o r a l principles should be h i g h l y
general (that is. simple), w h e n all they had to be was universal (which
is consistent with their being, if need be, quite specific).
I have tried to a m e n d this defect in K a n t ' s exposition of his t h e o r y
by, in my o w n writings, d i s t i n g u i s h i n g b e t w e e n two levels of m o r a l
thinking, the critical and t h e intuitive. At the intuitive level our m o r a l
II. 7. 8
RATIONALISM
143
ical t h i n k i n g . W e t h e n need t o c o n d e n s e this v a s t q u a n t i t y o f inform a t i o n i n t o a simpler set of guidelines or intuitive principles for use
in our daily life. T h e task might seem impossible b u t for o n e thing: we
are n o t t h e f i r s t t o address it. People t h r o u g h o u t m a n y g e n e r a t i o n s
have faced t h e problem, and arrived at their solutions. We m a y expect
m o s t of t h e s e s o l u t i o n s to be wise ones, b e c a u s e t h e y h a v e b e e n devised by people w i t h a lot of e x p e r i e n c e of s i m i l a r p r o b l e m s , a n d of
the c o n s e q u e n c e s of pursuing various solutions to t h e m .
B u t some of these solutions may have been bad ones. T h e r e is
nothing infallible about the wisdom of t h e ages. We c a n alter o u r intuitive principles, t h o u g h with difficulty, if we wish. T h a t is t h e task of
critical thinking. B u t we should be cautious. T h o s e w h o have t h r o w n
over t h e a c c u m u l a t e d intuitions wholesale have often c o m e to regret
it. T h e r e is u s u a l l y m o r e to learn t h a n to discard from t h e legacy of
the past.
t h i n k i n g h a s indeed to c l e a v e to g e n e r a l principles ( t h o u g h n o t , I
T h e question of w h e t h e r we should b e c o m e pacifists, with which I
hope, quite so g e n e r a l as t h o s e w h i c h inspired Kant's p a r e n t s w h e n
started ( 2 . 2 ) . is fairly easily settled (H 1985b). T h e c o n s e q u e n c e of n o t
t h e y were b r i n g i n g h i m u p ) . B u t a t t h e c r i t i c a l level a t w h i c h w e
e n o u g h people s t a n d i n g up for j u s t i c e a n d d e c e n c y in i n t e r n a t i o n a l
assess our intuitive principles, a n d possibly reject or a m e n d t h e m , our
relations would be that those who rejected t h e m would have their
thinking c a n deal in principles as specific as we need. So t h e require-
way, with disastrous c o n s e q u e n c e s for a l m o s t everybody. W h a t just-
m e n t t o universalize o u r m a x i m s does n o t c o m p e l u s t o adopt v e r y
ice and d e c e n c y require is a further question, w h i c h is to be solved by
general m a x i m s at this level. All we have to do is to treat similarly all
a n o t h e r application of critical thinking. T h e a n s w e r is provided by a
c a s e s having t h e s a m e universal properties, however specific, includ-
set o f g u i d e l i n e s for i n t e r n a t i o n a l politics s u c h t h a t a d h e r e n c e t o
ing c a s e s in which t h e individuals c h a n g e roles, and in w h i c h t h e r e -
t h e m does t h e best, all in all, for t h o s e affected. T h e s e guidelines a r e
fore we m a y find o u r s e l v e s in t h e position of victim. S u c h c r i t i c a l
not so hard to find (see e.g. J. E. Hare a n d C. J o y n t 1 9 8 2 ) .
thinking may, indeed, lead us to adopt quite general principles for use
T h e questions of abortion a n d e u t h a n a s i a , a n d o t h e r questions in
at t h e intuitive level: b u t t h e t h i n k i n g w h i c h goes to t h e i r adoption
medical ethics, have been widely discussed, a n d I have nothing to add
does n o t itself rely on t h e s e g e n e r a l intuitive principles, b u t only on
to w h a t I have said about t h e m (e.g. H 1974b. 1 9 7 5 c , d, 1 9 8 8 $ . Public
t h e requirement to universalize o u r m a x i m s .
opinion s e e m s to be c o m i n g r o u n d to a utilitarian solution to t h e m ,
7 . 8 . I will end on a v e r y p r a c t i c a l n o t e . We s t a r t e d off w i t h a
with w h i c h I would agree.
n u m b e r of down-to-earth m o r a l issues on which I claimed t h a t m o r a l
T h e question of w h a t to do about youth crime, to w h i c h I proposed
philosophy could shed light. Now it is t i m e to say w h a t light it h a s
in 2 . 4 a tongue-in-cheek solution, is m u c h m o r e difficult, a n d h a s re-
shed. T h e general procedure for settling moral questions should n o w
ceived a lot of media attention recently. T h e g e n e r a l impression seems
be clear. It consists in e x a m i n i n g in a factual way the c o n s e q u e n c e s of
to be that, in spite of w h a t some politicians have said, prison does n o t
alternative actions a n d policies, a n d asking whether we are prepared
work w i t h y o u n g offenders. S h o u l d w e t h e n h a n g t h e m , o r a t least
to prescribe universally their implementation. T h i s is the task of crit-
birch t h e m ? S h o u l d we adopt s h a r y a law like t h e S a u d i A r a b i a n s , or
144
II.
7. 8
RATIONALISM
145
r e v e r t t o t h e d r a c o n i a n p u n i s h m e n t s practised i n S i n g a p o r e ? T h e
a l o n g with it. We c a n n o t pre-judge t h e issue of w h e t h e r it would be
a n s w e r lies in a careful e x a m i n a t i o n of t h e c o n s e q u e n c e s of s u c h
for t h e best if t h e y did. We have to look at t h e c o n s e q u e n c e s of their
punishments, not merely on t h e offenders, but on society in general.
so doing. It is fairly obvious that they would be disastrous. Our whole
Flogging of violent c r i m i n a l s , ordered by c o u r t s , has been abolished
system of j u s t i c e is founded on the premiss t h a t nobody is to be pun-
within living m e m o r y in m a n y societies, and physical p u n i s h m e n t of
ished, let a l o n e killed, for offences t h a t they have n o t yet c o m m i t t e d .
s c h o o l c h i l d r e n in s t a t e s c h o o l s h a s been prohibited o n l y q u i t e
It would take an inconceivable shift in opinion to a b a n d o n this prin-
recently in Britain. It is still permitted in private schools in the United
ciple, a n d t h e c o n s e q u e n c e s of its a b a n d o n m e n t would be dire. T h e
K i n g d o m , t h o u g h t h e c o u r t s m a y c o m e t o b a n it, especially t h e
'slippery slope' a r g u m e n t , often abused by anti-abortionists, is really
European Court. And there is still a lot of support, both in Britain and
cogent here; if there were a danger of our sliding down this slope, we
in the United States, for t h e death penalty.
o u g h t to dig in o u r heels fast. B u t there is in fact no danger, b e c a u s e
It is h a r d to believe t h a t all t h e a r g u m e n t s in favour of s u c h
we have l e a r n t t h a t safeguards of this sort really a r e n e c e s s a r y to c o n -
c h a n g e s in the law are good o n e s , t h o u g h some probably are. T h e r e
strain the a d m i n s t r a t i o n of justice. (For justice a n d p u n i s h m e n t , see
are also some good a r g u m e n t s on t h e other side. T h e whole question
further MT9.6 ff., H 1 9 7 8 ^ 1 9 8 6 / . )
needs going into m o r e carefully, in order to find out, if we c a n , w h a t
I h a v e used this e x t r e m e e x a m p l e in o r d e r to i l l u s t r a t e h o w to
the effects on society as a whole would be of various treatments of of-
a r g u e a b o u t m o r a l questions. I n less e x t r e m e e x a m p l e s there c a n b e
fenders. So long as we are in t h e dark about the facts (which we a r e ) ,
legitimate differences of opinion which are n o t so easy to resolve. B u t
we s h a l l n o t be able to d e t e r m i n e t h e best policies. It is n o t t h e
t h e s a m e p r o c e d u r e c a n be used in resolving t h e m . First we h a v e to
p r o v i n c e of t h e m o r a l p h i l o s o p h e r to find o u t t h e facts, b u t o n l y to
work o u t t h e c o n s e q u e n c e s o f a l t e r n a t i v e policies, a n d t h e n find
probe bad a r g u m e n t s put forward on t h e basis of the facts, of w h i c h
guidelines w h i c h will on t h e whole, if generally adopted, lead to t h e
there are plenty. T h e present situation, in w h i c h appeals to m o r a l i t y
best courses of action. A n d t h e best courses of a c t i o n are those w h i c h
a r e m a d e in a l m o s t t o t a l i g n o r a n c e of w h a t it is and h o w to a r g u e
do the best, all in all, for people in society, c o u n t i n g e a c h for o n e a n d
rationally about it, is not conducive to the adoption of sound policies.
nobody for m o r e t h a n onei.e. treating e a c h individual as an end. In
B u t t h e t o n g u e - i n - c h e e k proposal t h a t I m a d e w a s n o t t h a t w e
short, we have to c o m b i n e the lessons which we o u g h t to have learnt
should h a n g or flog offenders, b u t s o m e t h i n g m o r e r a d i c a l : t h a t we

should try to c a t c h t h e m before they offended and weed t h e m out. T h i s
proposal at least we c a n reject on the basis of already k n o w n facts. To
begin with, juries would not c o n v i c t on such grounds, unless public
opinion c h a n g e d more drastically t h a n it is likely to. We should have
t o abolish t h e j u r y s y s t e m i n favour o f s o m e m o r e s u m m a r y w a y o f
administering justice. Even an inquisitorial system, s u c h as is in use
on the Continent, is unlikely to allow us to weed out potential offenders before they have offended. We should have to go to s o m e t h i n g like
t h e system o f ' j u s t i c e ' practised b y t h e KGB. A n d t h e r e a r e o b v i o u s
a n d s o u n d utilitarian a r g u m e n t s for not doing that.
Suppose, however, t h a t public opinion, prompted by a crime wave,
c a m e to favour the weeding-out policy. T h e n perhaps juries would go
from K a n t and Mill.
PART III
KANT
8
COULD KANT HAVE B E E N A
UTILITARIAN?
. . . the supreme end, the happiness of all mankind
(KrV A851 = B 8 7 9 = 549)
The law concerning punishment is a Categorical Imperative: and woe
to him who rummages around in the winding paths of a theory of
happiness, looking for some advantage to be gained by releasing the
criminal from punishment or by reducing the amount of i t . . .
( R / A i 9 6 = B226 = 3 3 i )
8 . 1 . M y a i m i n this c h a p t e r i s t o ask a question, n o t t o a n s w e r it. T o
a n s w e r it with confidence would require m o r e c o n c e n t r a t e d study of
K a n t ' s text t h a n I have yet h a d time for. I have r e a d his m a i n e t h i c a l
works, a n d formed s o m e t e n t a t i v e c o n c l u s i o n s w h i c h I s h a l l diffidently s t a t e . I h a v e a l s o read s o m e of his E n g l i s h - s p e a k i n g disciples
and would-be disciples, but not, I must admit, a n y of his G e r m a n expositors except L e o n a r d Nelson. B u t my purpose in raising t h e question is to enlist t h e help of others in answering it.
To m a n y the a n s w e r will seem obvious; for it is an accepted dogma
t h a t K a n t a n d t h e utilitarians stand at opposite poles of m o r a l philosophy. T h i s idea h a s been the c u r r e n t o r t h o d o x y at least since, in the
Revised from H 1993a.
KANT
14
III. 8.
III. 8 . 1
COULD K A N T HAVE BEEN A U T I L I T A R I A N ?
149
early twentieth century, Prichard a n d Ross, deontologists themselves,
calls 'ordinary r a t i o n a l knowledge of m o r a l i t y ' , a n d t h r o u g h o u t his
t h o u g h t t h e y h a d found a f a t h e r in K a n t . J o h n R a w l s . in t u r n , h a s
w r i t i n g s i s h a p p y w h e n c o m m o n m o r a l c o n v i c t i o n s s u p p o r t his
been deeply influenced by t h e s e intuitionist philosophers, a n d does
views, the title of t h e first chapter shows t h a t he is e n g a g e d in a 'tran-
not think it necessary to d o c u m e n t very fully the Kantian p a r e n t a g e
sition' from this to 'philosophical knowledge'. T h e s e c o n d c h a p t e r is
of their views. As a result, t h e story that Kant and utilitarians have to
called, likewise, ' T r a n s i t i o n from P o p u l a r M o r a l P h i l o s o p h y t o
be at odds is now r e g u l a r l y told to all b e g i n n e r students of m o r a l
M e t a p h y s i c o f M o r a l s ' . K a n t would n o t h a v e b e e n c o n t e n t , a s
philosophy.
P r i c h a r d was and as m a n y of our c o n t e m p o r a r i e s a r e , a n d as R a w l s
But is it true? My own hesitant answer would be that it is not. T h e
a l m o s t is. t o rely o n o u r o r d i n a r y m o r a l c o n v i c t i o n s a s d a t a , even
position is m o r e c o m p l i c a t e d . K a n t , I shall argue, could have b e e n a
after reflecting on t h e m . Instead, he developed a highly complex and
utilitarian, though he was not. His formal theory c a n certainly be in-
sophisticated a c c o u n t of moral reasoning: t h e 'Metaphysic of Morals'.
terpreted in a way t h a t allows h i m p e r h a p s even requires h i m t o
I n this h e w a s right. Moral philosophy, w h i c h P r i c h a r d t h o u g h t
be one kind of utilitarian. To that extent what J. S. Mill says a b o u t the
rested o n a m i s t a k e ( 1 9 1 2 : title), b e g a n w h e n S o c r a t e s a n d P l a t o ,
c o n s i s t e n c y of his o w n views with K a n t ' s Categorical Imperative is
faced with a collapse of popular morality b e c a u s e of t h e inability of
well founded ( 1 8 6 1 : c h . 5 , middle). B u t K a n t ' s rigorous p u r i t a n i c a l
its adherents to provide reasons for thinking as they did, set out in t h e
upbringing had imbued h i m w i t h s o m e m o r a l views w h i c h n o util-
s e a r c h for these r e a s o n s . K a n t is in this tradition; P r i c h a r d and Ross
i t a r i a n i n d e e d , w h i c h few m o d e r n t h i n k e r s o f a n y p e r s u a s i o n
are not, and Rawls, in some respects their follower, is h a l f in and h a l f
would be likely to e n d o r s e : a b o u t c a p i t a l p u n i s h m e n t , for e x a m p l e ,
out of it. He is only h a l f a rationalist, and h a l f an intuitionist, in t h a t
a n d a b o u t suicide, a n d e v e n a b o u t lying. T h e s e rigoristic v i e w s h e
he relies on intuitions altogether too m u c h (H 1 9 7 3 a ) . T h i s c h a p t e r is
does his best (unsuccessfully in the view of most expositors) to justify
the beginning of an attempt to rescue K a n t from s o m e of his modern
by appeal to his theory.
'disciples'.
I s h a l l be looking at s o m e of t h e s e a r g u m e n t s . To d e o n t o l o g i s t s
8.2.
I w a n t first t o draw a t t e n t i o n t o s o m e p a s s a g e s i n t h e
w h o seek lo shelter under Kant's wing they give small comfort; for if
Groundwork w h i c h b e a r on my question. I will s t a r t with t h e famous
his theory is c o n s i s t e n t with o n e kind of utilitarianism ( w h a t kind, I
passage, beloved of anti-utilitarians, a b o u t t r e a t i n g h u m a n i t y as an
shall be explaining), it does not do t h e m m u c h good if some of his ar-
end. In full it runs: 'Act in s u c h a way t h a t you always treat humanity,
g u m e n t s which most people would n o w reject are anti-utilitarian in
w h e t h e r in your o w n person or in t h e person of a n y other, never sim-
tendency. K a n t was, indeed, a deontologist, in t h e s e n s e t h a t he a s -
ply as a m e a n s , but always at t h e s a m e time as an end' (Gr B A 6 6 f. =
signed a primary place to duty in his a c c o u n t of moral thinking. B u t
4 2 9 ) . To understand this we have to know w h a t K a n t m e a n s by 'treat
he was not an intuitionist of t h e s t a m p of Prichard and Ross. He did
as an end'. He gives us some important clues to this in t h e succeeding
n o t believe, with Prichard, t h a t 'If we do doubt whether there is really
passage, b u t unfortunately he seems to be using t h e expression in at
an obligation to o r i g i n a t e A in a situation B, t h e remedy lies n o t in
least two different senses. Broadly speaking, t h e first a n d third of his
any process of general thinking but in getting face to face with a par-
e x a m p l e s , t h o s e c o n c e r n e d w i t h duties t o oneself, a r e i n c o n s i s t e n t
ticular instance of the situation B, and then directly appreciating t h e
with a utilitarian interpretation, but the second and fourth, those
obligation to o r i g i n a t e A in t h a t s i t u a t i o n ' ( 1 9 1 2 : s./.). K a n t w o u l d
c o n c e r n e d w i t h duties t o o t h e r s , a r e c o n s i s t e n t w i t h it. A s w e s h a l l
h a v e called this 'fumbling a b o u t w i t h t h e aid of e x a m p l e s ' (Tappen
see, this difference is no accident.
vermittelst der Beispiele, Gr B A 3 6 = 4 1 2 ) .

On t h e contrary, t h o u g h in t h e Groundwork he respects w h a t he
I will t a k e t h e s e c o n d a n d fourth e x a m p l e s first. T h e s e c o n d c o n c e r n s false promises. He c o m b i n e s this w i t h s i m i l a r e x a m p l e s about
III. 8. 2
KANT
ISO
'attempts on the freedom and property of others'. T h e fault in all s u c h

a c t s lies, he says, in 'intending to m a k e use of a n o t h e r m a n merely as
means to an end he does not s h a r e {in sich enthalte). For t h e m a n
w h o m I seek to use for my o w n purposes by such a promise c a n n o t
possibly agree with my way of b e h a v i n g to him, and so c a n n o t h i m self s h a r e t h e end of t h e a c t i o n ' . O t h e r people 'ought always at t h e
s a m e t i m e to be t r e a t e d as e n d s t h a t is. only as beings w h o m u s t
themselves be able to s h a r e in the end of the very same action'.
T h e fourth example I will quote in full:
III. 8. 2
others as ends, by saying that the ends of o t h e r s w h i c h we are to treat

as our o w n ends h a v e to be not immoral (Tgl A 1 1 9 = 4 5 0 : 'die Pflkht,
anderer ihre Zwecke (so fern diese nur nicht unsittlich sind) zu den meinen
zu machen)'. S o m e utilitarians, for e x a m p l e H a r s a n y i , t a k e a s i m i l a r
line a n d r u l e o u t i m m o r a l o r a n t i - s o c i a l e n d s from c o n s i d e r a t i o n
( 1 9 8 8 c : 9 6 ) . I am tempted to say, in the light of t h e similarity between
the views of these utilitarians and Kant, and of t h e passages we have
been discussing, t h a t he was a sort of utilitarian, n a m e l y a r a t i o n a l will utilitarian. For a utilitarian too c a n prescribe t h a t we should do
w h a t will c o n d u c e t o satisfying people's r a t i o n a l p r e f e r e n c e s o r
Fourth, as regards meritorious duties to others, the natural end which all
men seek is their own happiness. Now humanity could no doubt subsist if
everybody contributed nothing to the happiness of others but at the same
time refrained from deliberately impairing their happiness. This is. however,
merely to agree negatively and not positively with humanity as an end in itself
unless every one endeavours also, so far as in him lies, to further the ends of
others. For the ends of a subject who is an end in himself must, if this conception is to have its full effect in me, be also, as far as possible, my ends.
wills-for-endsends of which happiness is t h e s u m .

W e m a y n o t i c e i n passing t h a t this s a m e p a s s a g e i n K a n t ( G r
B A 6 9 = 4 3 0 ) provides an answer to self-styled K a n t i a n s w h o use w h a t
h a s been o n e of their favourite objections to utilitarianism, t h a t utilita r i a n s do n o t 'take seriously t h e distinction b e t w e e n persons' (Rawls
1 9 7 1 : 2 7 : see M a c k i e and Hare in H 1 9 8 4 0 : 1 0 6 , R i c h a r d s a n d Hare in
H 1 9 8 8 c : 2 5 6 ) . It is h a r d to understand precisely w h a t t h e objection
is. Clearly utilitarians are as aware as anybody else t h a t different a n d
1 interpret this as m e a n i n g that, in order to fulfil this version of t h e
distinct persons are involved in most situations a b o u t w h i c h we have
Categorical Imperative, I have to t r e a t o t h e r people's ends (i.e. w h a t
t o m a k e m o r a l j u d g e m e n t s . Probably w h a t t h e o b j e c t o r s a r e a t t a c k -
t h e y will for its o w n s a k e ) as my e n d s . T h e y m u s t be able to do t h e
ing is t h e idea t h a t we have, w h e n m a k i n g a m o r a l decision a b o u t a
same, i.e. share the end. In the Tugendlehre Kant explains the relation
s i t u a t i o n , t o t r e a t t h e i n t e r e s t s , ends, o r p r e f e r e n c e s o f different
b e t w e e n an end and t h e will as follows: 'An end is an o b j e c t of t h e
people affected by o u r a c t i o n s as of e q u a l i m p o r t a n c e , s t r e n g t h for
power of c h o i c e (Willkr) (of a rational being), through t h e t h o u g h t
strength. T h i s is t h e s a m e as to show equal c o n c e r n a n d respect for all
of which c h o i c e is determined to an action to produce this object' (Tgl
( a n o t h e r slogan o f t h e objectors, w h i c h s e e m s i n c o n s i s t e n t with t h e
A4 = 381). We shall be e x a m i n i n g later t h e distinction between 'Wille'
one we are considering). In other words, I am to t r e a t t h e interests of
and 'Willkr, and the alleged distinction between will and desire. On
t h e o t h e r s on a p a r w i t h my o w n . T h i s , a c c o r d i n g to u t i l i t a r i a n s , is
this, see esp. Tgl A 49 = 4 0 7 . w h e r e Wille is both distinguished from
w h a t is involved in b e i n g fair to all t h o s e affected. It is to o b e y
Willkr, and identified with a kind of desire: 'nicht der Willkr, sondern
B e n t h a m ' s i n j u n c t i o n 'Everybody to c o u n t for o n e , n o b o d y for m o r e
des Willens,
t h a n o n e ' (ap. Mill 1 8 6 1 : last c h a p t e r ) . A n d if we t r e a t e q u a l prefer-
gesetzgebendes
der ein mit der Regel,

Begehrungsvermgen
ist,
die er annimmt,
und eine
zugleich allgemein-
solche allein
kann
zur
ences as of e q u a l weight, utilitarianism is t h e result.
Tugend gezhlt werden' ('not a quality of the power of c h o i c e , but of
B u t t h a t is precisely w h a t K a n t is telling us to do in this passage, as
t h e will, which is o n e with t h e rule it adopts and which is also t h e ap-
Mill observes (ibid.). For if I m a k e the ends of o t h e r s my ends, I shall,
petitive power as it gives universal law. Only such an aptitude c a n be
in adjudicating b e t w e e n t h e m w h e n t h e y conflict, t r e a t t h e m in t h e
called virtue').
s a m e way as I would my o w n ends. In so doing I am n o t failing to dis-
E l s e w h e r e K a n t qualifies this e x p l a n a t i o n of w h a t it is to t r e a t
t i n g u i s h b e t w e e n different people, but, a s j u s t i c e d e m a n d s , giving
i 2
KANT
III. 8 . 2
equal weight to their a n d my e q u a l interests (the ends w h i c h they a n d
III. 8. 3
153
end (whatever nature intended) to live like t h e S o u t h S e a Islanders of
I seek with e q u a l s t r e n g t h of will), j u s t as I give equal weight to my
w h o m K a n t h a s e a r l i e r spoken slightingly; a n d h e c o u l d c e r t a i n l y
o w n e q u a l interests. S o , if t h e o b j e c t i o n did u n d e r m i n e u t i l i t a r i a n -
s h a r e this e n d with himself, and a g r e e to it. So t h e s e n s e of 'treat as
ism, it would undermine K a n t too.
an end' used in the second and fourth examples would provide no ar-
8 . 3 . B u t now w e have t o t u r n t o K a n t ' s f i r s t and third e x a m p l e s .
g u m e n t at all a g a i n s t his 'devoting his life solely to idleness, indul-
In the first, he is against suicide b e c a u s e it involves 'making use of a
gence, procreation, and in a word, to e n j o y m e n t ' (Gr B A 5 5 = 4 2 3 ) . In
person merely as a m e a n s to m a i n t a i n a tolerable state of affairs till
the sense used in the second and fourth examples, treating h u m a n i t y
the end of his life'. B u t this is n o t t h e s a m e sense of use as a m e a n s '
in myself as an end would n o t preclude my l o t u s - e a t i n g , a n y m o r e
as that which contrasts with 'treat as an end' in the second and
t h a n it would preclude suicide.
fourth examples. I might have as an end t h e saving myself from intol-
I s h o u l d like to m e n t i o n h e r e t h a t in my o w n a d a p t a t i o n of t h e
e r a b l e pain. Obviously t h e r e is no difficulty in my s h a r i n g this e n d
Kantian form of a r g u m e n t in FR ch. 81 specifically excluded from its
with myself, or a g r e e i n g with my way of b e h a v i n g to myself. K a n t
scope personal ideals not affecting other people, a n d said t h a t a b o u t
must therefore be h e r e using 'use as m e a n s ' and 'treat as an end' in
these o n e could n o t a r g u e in this way. So my view on t h e s e first and
s o m e different s e n s e . I s h a l l n o t h e r e investigate w h a t it is; b u t it
third e x a m p l e s of K a n t is t h a t he is going a s t r a y t h r o u g h t r y i n g (in
s e e m s to be s o m e t h i n g like ' r e g a r d (or n o t r e g a r d ) a h u m a n b e i n g
order to buttress his inbred convictions) to use a r g u m e n t s from u n i -
(myself) as at my own disposal to do w h a t I like with for my o w n pur-
versalizability outside t h e i r proper field, w h i c h is duties to o t h e r
poses'.
people.
B u t this objection to suicide, if valid at all, is different from t h o s e
T h e r e is a possible objection to t h e assimilation of wills to prefer-
to promise-breaking a n d n o n - b e n e f i c e n c e . To treat myself as at my
e n c e s t h a t I h a v e j u s t made: t h a t a preference, being s o m e t h i n g e m -
o w n disposal is n o t to frustrate t h e ends t h a t I will. Perhaps K a n t is
pirical, is n o t t h e s a m e as a will, w h i c h is, in t h e p u r e K a n t i a n
h e r e h a r k i n g b a c k t o s o m e t h i n g h e heard when young, t h a t m a n i s
doctrine, s o m e t h i n g n o u m e n a l (cf. KpV A 7 4 f. = 4 3 ) . To this objection
created as a h u m a n being to fulfil an end ordained by God, and there-
I shall return (8.8).
fore o u g h t n o t to a c t c o n t r a r y to God's will by not fulfilling God's
8 . 4 . B u t n o w w e m u s t t u r n t o a n o t h e r f a m o u s passage, t h e for-
ends. B u t to argue thus would be to follow a principle of h e t e r o n o m y
mulation of the Categorical Imperative w h i c h r u n s : 'Act only on t h a t
such as he later rejects (Gr B A 9 2 = 4 4 3 ) . It c a n n o t be turned into an
m a x i m t h r o u g h w h i c h you c a n a t t h e s a m e t i m e will t h a t i t should
a u t o n o m o u s principle by simply substituting ' m y s e l f for 'God'. For if
b e c o m e a universal law' (Gr B A 5 2 = 4 2 1 ) .
it is not God's will but my will t h a t is in c o m m a n d , then it c a n , within
T h i s version too is c o n s i s t e n t with utilitarianism. If we are going
a c o n s i s t e n t set of ends, c h o o s e suicide in t h e s e special c i r c u m -
to will t h e m a x i m of o u r a c t i o n to be a u n i v e r s a l law, it m u s t be, to
stances.
use t h e j a r g o n , universalizable. I have, t h a t is, to will it n o t o n l y for
T h e s a m e could b e said a b o u t t h e third e x a m p l e c o n c e r n i n g t h e
the present situation, in which I occupy t h e role t h a t I do, but also for
cultivation of o n e ' s t a l e n t s . For a full s t a t e m e n t of t h e e x a m p l e we
all situations resembling this in their universal properties, including
have to refer b a c k to Gr B A 5 5 = 4 2 3 . 1 shall discuss this earlier use of
those in w h i c h I occupy all t h e other possible roles. B u t I c a n n o t will
t h e e x a m p l e shortly. Here it is to be n o t e d t h a t K a n t s p e a k s of ' n a -
this unless I am willing to u n d e r g o w h a t I s h o u l d suffer in all those
t u r e ' s purpose for h u m a n i t y in o u r p e r s o n ' (Gr B A 6 9 = 4 3 0 ) , t h u s
roles, a n d of c o u r s e a l s o get t h e good t h i n g s t h a t I s h o u l d e n j o y in
a g a i n betraying t h e theological a n d h e t e r o n o m o u s s o u r c e o f his a r -
others of t h e roles. T h e upshot is t h a t I shall be able to will only such
g u m e n t h e r e . A person c o u l d c e r t a i n l y with c o n s i s t e n c y will as his
m a x i m s as do t h e best, all in all, impartially, for all those affected by
KANT
154
III. 8 . 4
my a c t i o n . A n d this, a g a i n , is u t i l i t a r i a n i s m . To link it up w i t h t h e
III. 8. 4
155
whenever you please', we shall probably opt for t h e former. B u t there
o t h e r formula a b o u t t r e a t i n g people as ends: if I am to universalize
are m a n y less simple m a x i m s in between t h e s e e x t r e m e s w h i c h most
my m a x i m , it must be consistent with seeking the ends of all the o t h e r
of us would will in preference to either of t h e m : for e x a m p l e 'Preserve
people on equal terms with my own.
people's lives w h e n t h a t is in their interests' ( a n d p e r h a p s we would
T h i s formulation of t h e Categorical Imperative is followed by a n -
wish to add o t h e r qualifications). As we have s e e n (8.1) m o r a l prin-
o t h e r r a t h e r similar o n e : A c t as if t h e m a x i m of your action were to
ciples do n o t have to be as simple and general as K a n t s e e m s to have
b e c o m e t h r o u g h y o u r will a u n i v e r s a l law of n a t u r e ' (Gr BA 52 =
thought, a n d t h e y c a n still be universal all t h e s a m e ( H 1 9 7 2 , 1 9 9 4 b ) .
4 2 1 ) . After this, Kant illustrates these two formulations with t h e s a m e
As regards cultivation of talents, Kant is also on s h a k y ground. It
examples as we have been discussing in c o n n e c t i o n with t h e ' h u m a n -
is perfectly possible to will t h a t those w h o a r e in t h e f o r t u n a t e posi-
ity a s a n e n d ' f o r m u l a t i o n . H e r e a g a i n t h e promise-keeping a n d
tion of b e i n g able to live like t h e S o u t h S e a I s l a n d e r s s h o u l d do so;
beneficence examples fit well with a utilitarian interpretation, but t h e
and this could b e c o m e a law of n a t u r e if n a t u r e were as benign every-
suicide and cultivation-of-talents e x a m p l e s do not. In the promising
w h e r e as it is said to be in Tahiti. T h e best a r g u m e n t a g a i n s t lotus-
c a s e , he uses a form of a r g u m e n t u s u a l l y n o w called by E n g l i s h -
eating is a utilitarian one, which K a n t does n o t use t h o u g h he could
s p e a k i n g writers u t i l i t a r i a n g e n e r a l i z a t i o n ; he asks ' H o w w o u l d
have; n a m e l y t h a t o n e person's indolence may, in t h e actual state of
things stand if my m a x i m b e c a m e a universal law?', and answers t h a t
nature, h a r m o t h e r s w h o m he might be helping if m o r e industrious,
p r o m i s e s would b e c o m e ' e m p t y s h a m s ' . T h i s is n o t a s t r o n g a r g u -
and w h o therefore c a n n o t s h a r e his ends.
m e n t , b e c a u s e o n e m i g h t will as a universal law t h a t people should
8 . 5 . T h e s c o r e a t this point i s t h a t K a n t ' s theory, i n t h e formula-
b r e a k promises i n precisely o n e ' s o w n present s i t u a t i o n , w h e n o n e
tions of the Categorical Imperative we have considered, is compatible
c a n get away with it and t h e institution of promising would survive.
with utilitarianism, and so are some a r g u m e n t s t h a t he uses, or could
( R e c e n t work on t h e difficulty of d r a w i n g a line b e t w e e n a c t - a n d
have used c o n s i s t e n t l y with t h e theory, in s o m e of his e x a m p l e s . By
rule-utilitarianism is relevant here; cf. FR 1 3 0 ff., Lyons 1 9 6 5 : c h . 3 ) .
my r e c k o n i n g t h e first e x a m p l e (suicide) is t h e only o n e t h a t c a n n o t
T h e a r g u m e n t against promise-breaking we considered earlier, w h i c h
be h a n d l e d in a u t i l i t a r i a n way in a c c o r d a n c e w i t h t h e C a t e g o r i c a l
says t h a t the victim c a n n o t s h a r e t h e end of the promise-breaker, is
Imperative in t h e s e t h r e e formulations, a l t h o u g h K a n t h i m s e l f does
m u c h stronger, and is similar to o n e I would myself, as a utilitarian,
h a n d l e both this a n d t h e third e x a m p l e in a n o n - u t i l i t a r i a n way. So,
rely on (H 19640*: s./.).
as 1 said at t h e b e g i n n i n g , K a n t could h a v e b e e n a utilitarian, in t h e
Kant's a r g u m e n t here against non-beneficence comes to m u c h t h e
s e n s e t h a t his t h e o r y is compatible with u t i l i t a r i a n i s m , b u t in s o m e
s a m e as the one I discussed earlier, and one which I should myself, as
of his p r a c t i c a l m o r a l j u d g e m e n t s his inbred rigorism leads h i m
a utilitarian, employ, a n d I h a v e no t i m e to a n a l y s e it further. T h e
into bad a r g u m e n t s w h i c h his theory will n o t really support. I do n o t
a r g u m e n t against suicide is again very weak. I could c e r t a i n l y with-
t h i n k t h a t t h i s s c o r e o u g h t t o give m u c h c o m f o r t t o m o d e r n a n t i -
o u t c o n t r a d i c t i o n will u n i v e r s a l l y t h a t t h o s e w h o would o t h e r w i s e
utilitarians w h o usurp Kant's authority.
have to endure intolerable pain should kill themselves. This could in-
It does, however, e m e r g e from his d i s c u s s i o n of t h e e x a m p l e s in
deed become a universal law of nature, and 1 could act as if it were to
t h e Groundwork t h a t t h e r e is a t e n s i o n in K a n t ' s t h o u g h t b e t w e e n
b e c o m e so t h r o u g h my will. K a n t t h i n k s it is a good a r g u m e n t o n l y
utilitarian and non-utilitarian elements. How this tension is to be
b e c a u s e h e t h i n k s ( p e r h a p s o w i n g t o his rigorist u p b r i n g i n g ) t h a t
resolved b e c o m e s a little c l e a r e r in t h e Doctrine of Virtue. T h e r e , a
m a x i m s h a v e to be v e r y simple. If we h a v e a c h o i c e b e t w e e n t h e
m a i n division is m a d e between duties to oneself a n d duties to others.
simple m a x i m s Always preserve h u m a n life' and 'Destroy h u m a n life
T h i s distinction and other related ones are laid o u t in Tgl A 3 4 = 3 9 7 ,
KANT
i 6
5
III. 8. 5
III. 8. 5
157
in t h e top h a l f of a t a b l e h e a d e d ' T h e M a t e r i a l E l e m e n t of D u t y of
one of form, n o t of content. To explain this: a morally perfect c h a r a c -
Virtue'. 'My own end, w h i c h is also my duty' is said to be 'my o w n per-
ter, or good will, as he sees it. is one formed by its o w n framing of uni-
fection; and 'the end of others, the promotion of which is also my duty'
versal laws in a c c o r d a n c e with the Categorical Imperative. In seeking
is said to be 'the happiness of others'.
moral perfection, we are seeking to make o u r wills good in this sense.
T h e immediate impression we get from this is t h a t there is a util-
I f this i s w h a t K a n t m e a n s , t h e n t h e u t i l i t a r i a n a n d t h e n o n -
itarian part of Kant's theory, and a non-utilitarian part. T h e utilitar-
utilitarian part of his morality at o n c e c o m e together again. For a will
ian part prescribes duties to o t h e r s , a n d t h e s e are c o m p a t i b l e w i t h
t h a t wills universally must, as we have seen, be a will t h a t treats t h e
utilitarianism (qualified by the requirement, as above, that we have to
ends of other people's wills on equal terms with its o w n ends; and this
advance others' ends only in so far as they are consistent with moral-
is a n o t h e r way of expressing the practical love t h a t we h a v e already
ity). B u t the other part (duties to oneself) seems to be not utilitarian at
found to be required by our duties to others. In o t h e r words, t h e m o r a l
all. but perfectionist. However, these impressions are too superficial.
perfection of a good will is a perfection of form, a n d t h e form is t h e
T h i s b e c o m e s a p p a r e n t if ( t a k i n g a h i n t from what he says a g a i n s t
form of practical love, which is utilitarian, in t h a t it seeks to advance
perfectionism in Gr B A 9 2 = 4 4 3 ) we ask, first, in what the perfection
t h e ends of all impartially. T h e ' m a t e r i a l e l e m e n t ' , referred to in t h e
is supposed to consist: and secondly, w h a t 'consistent with m o r a l i t y '
title of t h e table, all c o m e s e i t h e r directly or indirectly from this
is to m e a n . As we a n s w e r t h e s e q u e s t i o n s we shall see t h a t t h e t e n -
source.
sion between t h e utilitarian and non-utilitarian e l e m e n t s in K a n t ' s

theory begins to ease.
T h e s a m e happens when we ask w h a t it m e a n s to say t h a t t h e ends

of others w h i c h we seek impartially to a d v a n c e have to be consistent
Obviously the perfection that K a n t is after is moral perfection. It
with morality. Here we have to look in passing at w h a t K a n t says later
c o n s i s t s in t h e a c q u i s i t i o n of virtue. P a r t of this virtue will c l e a r l y
in the Groundwork about the Kingdom (or R e a l m ) of Ends. A good will
consist in the disposition to fulfil t h e duties to others laid down on t h e
h a s to be o n e t h a t c a n be a l a w m a k i n g m e m b e r of s u c h a r e a l m (Gr
utilitarian side of t h e table. B u t w h a t is the other part? T h a t is. w h a t
B A 7 7 - 9 = 4 3 5 f ) . T h i s i s Kant's way o f e n s u r i n g t h a t t h e moralities o f
content does moral perfection have, for Kant, over and above the util-
all r a t i o n a l b e i n g s will b e c o n s i s t e n t w i t h o n e a n o t h e r . T h e law-
itarian c o n t e n t consisting in practical love for other people. (For t h e
makers in t h e R e a l m of Ends will legislate unanimously, because e a c h
notion of 'practical love' see Gr B A i 3 = 3 9 9 and Tgl A n 8 f. = 4 4 8 f.).
is constrained by t h e universal form of t h e legislation.
It begins to look as if m o r a l perfection, if it sought a n y t h i n g beyond
T h e effect of this is t h a t t h e ends of others, w h i c h we have a duty
this p r a c t i c a l love, would be c h a s i n g its o w n tail. As he says in Gr
t o a d v a n c e impartially, a r e t h o s e o n l y w h i c h a r e m o r a l , i.e. w h i c h
B A 9 2 = 4 4 3 , ' | t h e o n t o l o g i c a l c o n c e p t o f p e r f e c t i o n | s h o w s a n in-
they would retain if they were legislating universally, or forming uni-
evitable t e n d e n c y to go r o u n d in a circle and is unable to avoid pre-
versal m a x i m s i n a c c o r d a n c e with t h e e a r l i e r f o r m u l a t i o n s o f t h e
supposing the morality it h a s to explain'. T h e r e would be nothing else
C a t e g o r i c a l I m p e r a t i v e . B u t i f t h e s e m a x i m s , a s t h e y m u s t , express
in t h e duty to m a k e ourselves perfect, except t h e duty to m a k e o u r -
practical love, they too will be consistent with utilitarianism. For util-
selves disposed to make ourselves perfect. It would still not have been
itarianism is, simply, the morality w h i c h seeks t h e ends of all in so far
d e t e r m i n e d w h a t t h e perfection, o r t h e p e r f o r m a n c e o f t h e duty t o
as all c a n seek t h e m consistently in a c c o r d a n c e w i t h universal m a x -
promote it, would consist in.
ims. If a utilitarian tried to promote ends w h i c h were n o t consistent
B u t we must be careful h e r e to distinguish between form and c o n -
with s u c h a morality, he would r u n up a g a i n s t t h e o b s t a c l e t h a t t h e
tent. It could be that Kant's view is this: t h e perfection we are after is
ends he was promoting would be such as o t h e r s could n o t 'share', as
KANT
III. 8. 5
HI. 8.6
159
K a n t puts it (see above); a n d so his entire m o r a l system would c o m e
' t h a t v i r t u e s h o u l d b e its o w n end a n d also, b e c a u s e o f t h e m e r i t i t
apart. It is part of the requirements for a consistent utilitarian m o r a l -
h a s a m o n g m e n , its o w n reward', a n d Tgl A 3 3 = 3 9 7 : ' t h e w o r t h of
ity that it should be able to be shared by all.
virtue itself, as its o w n end, far exceeds t h e v a l u e of a n y utility a n d
We thus see that even t h e apparently non-utilitarian part of K a n t ' s

doctrine of virtue, and of his entire system, t u r n s into utilitarianism
a n y e m p i r i c a l e n d s a n d advantages t h a t v i r t u e may, after all, bring

about.'
at o n e remove. It does so b e c a u s e even t h e apparently non-utilitarian
8 . 7 . W h y i s t h e suggestion t h a t K a n t could h a v e b e e n a utilitar-
virtue of perfection requires a s p i r a n t s to it to perfect t h e m s e l v e s in
ian t h o u g h t so bizarre? It h a s been held t h a t he could n o t have been
practical love.
for, i n t h e m a i n , two i n a d e q u a t e r e a s o n s . T h e f i r s t i s t h a t h e often
8 . 6 . T h e objection might b e made that, whereas for K a n t h u m a n
stresses t h a t t h e Groundwork of the Metaphysic of Morals, as he calls
perfection is an end in itself, for t h e utilitarian it is a mediate end, t h e
his book, c a n n o t appeal t o a n y t h i n g c o n t i n g e n t a n d empirical; a n d
ultimate end being the furtherance of the ends of all. This objection is
desires a n d preferences a r e of this sort. B u t h e r e we h a v e to be very
a n a l o g o u s to o n e which h a s been made against my own theory, t h a t
careful to distinguish, as K a n t insists on o u r doing, between t h e e m -
by dividing moral thinking into two levels I have demoted o u r ordin-
pirical and t h e rational parts of moral philosophy. He certainly thinks
ary intuitive convictions a n d prima facie principles into a merely in-
t h a t it h a s b o t h t h e s e parts. He says, a b o u t t h o s e w h o fail to distin-
strumental role. For me. it is said, the real moral thinking takes place
guish t h e two roles, ' W h a t (such a procedure) t u r n s out is a disgust-
at t h e c r i t i c a l level a n d is u t i l i t a r i a n ; w h a t goes on at t h e intuitive
ing h o t c h - p o t c h (Mischmasch) of s e c o n d - h a n d o b s e r v a t i o n s a n d
level is only a means to help us fulfil, maximally and on the whole, o u r
s e m i - r a t i o n a l principles o n w h i c h t h e e m p t y - h e a d e d r e g a l e t h e m -
utilitarian duties as determined by critical thinking. We a r e to m a k e
selves, b e c a u s e this is s o m e t h i n g t h a t c a n be used in t h e c h i t - c h a t of
ourselves into good people, a n d fulfd o u r duties, not for its o w n sake
daily life. M e n of insight, on t h e o t h e r h a n d , feel confused by it a n d
but because t h a t will conduce to the greatest good. It is further alleged
avert t h e i r eyes w i t h a dissatisfaction w h i c h , however, t h e y a r e u n -
(e.g. by B e r n a r d Williams, 1 9 8 8 : 1 8 9 ff.) t h a t if we took s u c h an atti-
able to c u r e ' (Gr B A 3 1 = 4 0 9 , cf. BAiv = 3 8 8 ) .
tude to o u r c o m m o n m o r a l c o n v i c t i o n s , they would s o o n 'erode'; if
T h e i m p o r t a n t point to get hold of is t h a t his strictures on bringing
(hey are to retain their force for us, we have to treat them as ultimate.
in e m p i r i c a l c o n s i d e r a t i o n s apply only to w h a t he is doing in t h i s
It has always seemed to me t h a t this objection, whether to my o w n
book: only, t h a t is, to the Metaphysic of Morals, a n d indeed only to its
t h e o r y or to K a n t as I have interpreted h i m , will not be sustained by
G r o u n d w o r k . I t h i n k it is l e g i t i m a t e to r e g a r d t h e Groundwork as a
a n y o n e w h o has experience even of trying to live a morally good life.
purely l o g i c a l e n q u i r y into t h e n a t u r e o f m o r a l r e a s o n i n g , a n d a s
It is perfectly possible at t h e intuitive level to treat m o r a l d u t y or
s u c h i t o f c o u r s e m u s t n o t c o n t a i n appeals t o e m p i r i c a l facts, a n y
virtue as ultimate and give t h e m the 'reverence' that Kant d e m a n d s ,
m o r e t h a n a n y o t h e r kind of logic. T h i s is t h e c h i e f t h i n g , as I said,
while at the s a m e time to recognize t h a t to establish t h a t those traits
that distinguishes K a n t from some of his m o d e r n self-styled disciples.
of c h a r a c t e r really do constitute virtue, and that those intuitive m o r a l
Let us t h e n look at t h e K a n t i a n p r o g r a m m e , or at this interpreta-
principles really a r e t h e o n e s w e should observe, r e q u i r e s m o r e
tion of it, in m o r e detail. It rests on a metaphysical or logical enquiry
t h o u g h t t h a n t h e m e r e intuition t h a t this is so. I am sure t h a t K a n t
into the n a t u r e of t h e moral concepts. T h i s h a s to be t h e basis of any
would have agreed, a l t h o u g h h e m a k e s his a c c o u n t o f t h e r e l a t i o n
system of moral reasoning. We have to do it by considering the
between virtue a n d duty m u c h m o r e obscure by failing to clarify t h e
n a t u r e of t h e c o n c e p t s only, not a n y t h i n g empirical. K a n t believed in
distinction between levels of m o r a l t h i n k i n g (see below). It is in this
t h e synthetic a priori, a n d indeed calls his Categorical Imperative the
s e n s e t h a t we should u n d e r s t a n d p a s s a g e s s u c h as Tgl A 3 2 = 3 9 6 :
' p r a c t i c a l s y n t h e t i c a priori' (Gr B A 5 0 = 4 2 0 ) . B u t he explains later
KANT
III. 8. 7
III. 8. 8
161
t h a t the question h o w s u c h a synthetic a priori proposition is possible
W h a t t h e n did he m e a n ? I think that w h a t he m e a n t was this. Our
and n e c e s s a r y lies outside t h e bounds of a metaphysic of m o r a l s (Gr
will is initially free to will whatever we will. We a r e not constrained to
B A 9 5 = 4 4 0 ) . T h e first two chapters of the Groundwork (those we have
will this or t h a t b e c a u s e of w h a t this or t h a t is. T h e will is constrained
been c o n c e r n e d with), are 'merely analytic' (Gr B A 9 6 = 4 4 5 ) ; he h a s
only by w h a t K a n t calls 'the fitness of its m a x i m s for its o w n m a k i n g
been 'developing t h e c o n c e p t of morality as generally in v o g u e ' . At
of universal law' (Gr B A 8 8 = 4 4 1 ) . This is w h a t is implied in t h e 'auto-
any rate he would, I am sure, have rightly excluded from this p a r t of
n o m y ' f o r m u l a t i o n of t h e Categorical I m p e r a t i v e . T h a t is, it is only
his enquiry any empirical data, w h e t h e r about what actually goes on
t h e universal form of w h a t we a r e going to will t h a t c o n s t r a i n s us,
in people's minds or a b o u t a n y t h i n g else, including any a n t e c e d e n t l y
and not any c o n t e n t . T h e content gets put in by t h e will itself. T h e will
held substantial moral judgements; for the only source of these could
c a n a c c e p t o n l y s u c h c o n t e n t s o r o b j e c t s o f its volition a s c a n b e
be s o m e t h i n g that goes on in people's minds, t h a t is, intuitions. T h a t
willed universally. T h i s is the s a m e doctrine as I have myself expressed
we h a v e a c e r t a i n i n t u i t i o n is an e m p i r i c a l fact, and as s u c h is e x -
by saying t h a t m o r a l judgements have to be universal prescriptions.
cluded from this part of t h e enquiry, for the s a m e r e a s o n as desires
So interpreted, t h e doctrine of a u t o n o m y would exclude as hetero-
t h a t we contingently have are excluded. Kant explicitly rejects moral
n o m o u s m a n y of t h e principles advocated by s o m e m o d e r n so-called
sense theories (Gr BA91 f. = 4 4 2 ) , and would equally have rejected in-
K a n t i a n s : for t h e y do seek to constrain t h e will n o t j u s t formally b u t
tuitionism of the sort expressed in the quotation from Prichard t h a t I
substantially by saying t h a t it has to have c e r t a i n objects. S u c h i n t u -
gave earlier. O r d i n a r y people u n d e r s t a n d , indeed, t h e concepts of
itionists n o t only appeal, t h o u g h they do n o t call it that, to something
morality, b u t this is no m o r a l s e n s e a p p r e h e n d i n g t h e substance of
empirical, n a m e l y t h e c o n t i n g e n t fact t h a t we have c e r t a i n intuitions
morality.
or c o n v i c t i o n s , b u t seek to c o n s t r a i n t h e will a n d bind it to t h e sub-
8 . 8 . T h e elements of K a n t ' s metaphysic of morals that I find most
stantial c o n t e n t of these convictions. T h i s is most u n - K a n t i a n .
c e n t r a l a r e its r e l i a n c e o n t h e p u r e will, a n d its i n s i s t e n c e t h a t i n
Returning, then, to the objection we are considering to calling
moral reasoning we have to will universally. W h a t does 'pure' m e a n ,
K a n t a utilitarian: t h e objection says t h a t this c a n n o t be so, b e c a u s e
a n d w h a t does ' r e l i a n c e ' m e a n ? To understand this we have to c o n -
utilitarians appeal to desires or preferences, which are something em-
sider Kant's doctrine of the a u t o n o m y of the will. This, he says, is 'the
pirical, and therefore excluded by Kant. To this t h e a n s w e r is first, t h a t
property t h e will h a s of being a law to itself (independently of every
they are excluded only from t h e formal p a r t of his enquiry, b u t have
property belonging to the objects of volition)' (GYBA87 = 4 4 0 ) .
to be admitted into any application to c o n c r e t e situations of t h e form
Here it is very easy to go astray in one's interpretation of Kant, and
of m o r a l r e a s o n i n g w h i c h t h e e n q u i r y generates; a n d secondly, t h a t
attribute to him a n o n s e n s e . One way of taking this doctrine would
there is n o t h i n g to prevent a utilitarian from dividing up his enquiry
be to say t h a t to be a u t o n o m o u s t h e will h a s to h a v e no r e g a r d to
in the s a m e K a n t i a n way, as for clarity he should, and as I do myself.
w h a t in p a r t i c u l a r it is willing. S o . for example, w h e n I am deciding
A utilitarian system also h a s a pure formal part, w h i c h (in my view)
w h e t h e r to will to tell an u n t r u t h , I have to have no r e g a r d to t h e
needs to rely only on t h e logical properties of t h e m o r a l c o n c e p t s . It
property of this proposed o b j e c t of my volition, n a m e l y t h a t w h a t I
operates, indeed, with the concept of preference (and w h e t h e r this is a
should be saying would be u n t r u e . Or, if I am c o n t e m p l a t i n g killing
different c o n c e p t from t h a t of will needs further d i s c u s s i o n ) ; b u t it
someone, I am not to pay attention to the property of my action t h a t
does not a s s u m e t h a t preferences have a n y p a r t i c u l a r c o n t e n t . What
it would consist in bringing about his death. I c a n n o t believe t h a t this
people prefer is an e m p i r i c a l m a t t e r ; it h a s to be a s c e r t a i n e d o n c e
is w h a t K a n t m e a n t , b e c a u s e he c e r t a i n l y t h o u g h t it r e l e v a n t to t h e
we start to apply o u r system of r e a s o n i n g , b u t in order to set up the
morality of actions that they were lies or murders.
s y s t e m we do n o t n e e d to a s s u m e t h a t people prefer o n e t h i n g or
162
KANT
III. 8. 8
III. 8. 8
163
a n o t h e r : that is, in setting up t h e system we look merely at t h e form of
sent problem; for utilitarianism could easily be expressed in terms of
people's preferences, not at their c o n t e n t .
rational will.
It has to be asked w h e t h e r Kant's wills are any different in this re-
Secondly, w h e n K a n t draws, as he often does, a c o n t r a s t between
spect. Gr B A 6 4 = 4 2 7 would suggest t h a t they are not: ' P r a c t i c a l prin-
rational will a n d inclination (Neigung), it is often, t h o u g h not always,
ciples are formal if they a b s t r a c t from all subjective ends'; a n d this is
selfish i n c l i n a t i o n t h a t he h a s in mind. An e x a m p l e is Gr B A 8 = 4 9 6 .
equally true of t h e 'Principle of Utility' in those utilitarians w h o have
We a r e n o t to follow o u r desires in so far as t h e y a r e desires for our
one, especially if it is expressed in terms of the formal notion of pref-
own a d v a n t a g e ; t h a t would n o t be to t r e a t o t h e r s ' e n d s as o u r o w n
erence-satisfaction. It is an empirical fact t h a t a person wills this or
ends. B u t of c o u r s e a utilitarian could agree w i t h this insistence that
that, just as it is an empirical fact t h a t he prefers this or that. B u t t h e
the desires t h a t determine our moral j u d g e m e n t have to be universal
form of the will or preference c a n be t h e s a m e w h a t e v e r he wills or
and impartial.
prefers, provided t h a t for c a t e g o r i c a l or m o r a l imperatives, as b o t h

the utilitarians and K a n t c a n agree, t h e form is universal.
Thirdly, K a n t , t h o u g h he m a k e s a c l e a r d i s t i n c t i o n b e t w e e n will
a n d i n c l i n a t i o n (Neigung), does n o t in fact always distinguish desire
T h a t , for both K a n t a n d t h e u t i l i t a r i a n s , is the only formal c o n -
(Begierde) in t h e r e l e v a n t s e n s e from will, t h o u g h he does in Gr
straint on the will. However, for both there are material c o n s t r a i n t s ,
B A 1 2 4 = 4 6 1 . In m o r e t h a n o n e place he identifies t h e m . In t h e pref-
in the c o n c r e t e situation in which we are doing the willing. S u c h c o n -
a c e a n d t h e introduction to t h e second Critique t h e r e a r e two defini-
straints are, for example, t h a t if I were to say what I am proposing to
tions, o n e of t h e faculty of desire (Begehrungsvermogen), a n d t h e o t h e r
say, I should be speaking falsely, or that if I were to pull t h e trigger I
of will, which are in almost identical terms (KpV A 1 7 n. = 9 n A 2 9 =
should be killing someone. I have to be able to will this universally for
15). Later in t h e s a m e work he speaks of 'the faculty of desire w h i c h is
all similar cases, and this c o n s t r a i n s me b e c a u s e of the empirical fact
therefore called t h e will, or t h e pure will in so far as t h e pure under-
that in that situation t h e person I should be lying to does n o t w a n t , or
standing ( w h i c h in s u c h a c a s e is called r e a s o n ) is practical t h r o u g h
will, to be deceived (as K a n t m i g h t put it, he and I c a n n o t ' s h a r e ' t h e
the m e r e c o n c e p t i o n of a law' ( A 9 6 = 5 5 ) . F r o m KU BAxxiii = 1 7 8 n.
will t h a t he should b e ) , a n d t h e person I should be killing does n o t
(different versions in different editions) a n d Rl A B l ff. = 2 1 1 ff., it looks
want, or will, to be killed. Given t h a t this is the will or preference of
a s i f K a n t c a m e t o see t h a t t h e r e a r e different t h i n g s t h a t c o u l d b e
the other party, I am constrained by this, and by the form of t h e r e a -
called 'desire', 'inclination', etc. (as indeed t h e r e a r e ) . If so, it m a y be
soning, to treat h i m as an end by making what he wills my end, or in
t h a t w h a t m o d e r n u t i l i t a r i a n s call ' p r e f e r e n c e ' m i g h t b e e x c l u d e d
o t h e r words to treat his preference as if il were my own. Otherwise I
from his b a n on t h e empirical, and assimilated m o r e to his Willkur, or,
shall not be able to universalize my m a x i m .
if rationally universalizing, to his Wille.
II m a y be objected t h a t for K a n t t h e distinction between will a n d
8 . 9 . O n c e w e h a v e distinguished pure from applied e t h i c s , this
m e r e preference or desire is f u n d a m e n t a l . To this t h e r e a r e t h r e e
first objection to enrolling K a n t as a kind of utilitarian collapses. B u t
replies. T h e first is t h a t for K a n t there is an important distinction be-
now we are able to deal with the second objection, t h a t K a n t c a n n o t
tween t h e will w h i c h is ' n o t h i n g but p r a c t i c a l r e a s o n ' (Gr BA}f> =
have been a consequentialist, but utilitarians have to be. O n c e c o n s e -
4 1 2 ) i . e . the rational willand the will that is the source of m a x i m s
quentialism is properly formulated, it is h a r d to see h o w anyone, K a n t
w h e t h e r good or bad, r a t i o n a l or i r r a t i o n a l . He calls t h e l a t t e r
included, could fail to be a c o n s e q u e n t i a l i s t . T h e d o c t r i n e gets a bad
'Willkur (sometimes translated ' c h o i c e ' ) . His Latin equivalent for this
n a m e only b e c a u s e its opponents, t h r o u g h their o w n confusions, for-
is liberum arbitrium, a n d it is t h e possession of this t h a t gives us free
mulate it incorrectly ( 1 . 8 , 7 . 8 , H 1 9 9 3 c : 1 2 3 . 1 9 9 8 b ) .
will or autonomy. B u t this distinction is not m u c h relevant to o u r pre-
Let us confine ourselves for the present to m o r a l j u d g e m e n t s which
r64
KANT
III. 8 . 9
are on, or about, acts; for these are the judgements about w h i c h c o n sequentialists and anti-consequentialists are supposed to be disagreeing. To a c t is to m a k e a difference to the c o u r s e of events, a n d w h a t
the act is, is determined by w h a t difference. To revert to my previous
examples (hackneyed o n e s . I am afraid): if I am wondering w h e t h e r
to pull the trigger, the m a i n morally relevant consideration is that, if I
did. the m a n t h a t my g u n is pointing at would die. Killing, w h i c h is
the m o r a l l y w r o n g a c t , is causing d e a t h , thai is. doing s o m e t h i n g
which h a s death as a c o n s e q u e n c e . Similarly, w h a t is w r o n g a b o u t
lying is t h a t it is c a u s i n g s o m e o n e else to be deceived (to hold a false
o p i n i o n ) b y o n e s e l f s a y i n g s o m e t h i n g false. T h e intended c o n s e q u e n c e is w h a t m a k e s it w r o n g . It would not be lying if it were n o t
intended to have this c o n s e q u e n c e .
1 am not saying t h a t all t h e c o n s e q u e n c e s of acts a r e m o r a l l y
relevant. Nor does any utilitarian have to say this. Many will be irrele v a n t . W h i c h are r e l e v a n t depends on w h a t m o r a l principles apply
t o the situation (the r e l e v a n t c o n s e q u e n c e s a r e those w h i c h t h e
principles forbid or r e q u i r e o n e to bring a b o u t ) . So w h a t t h e a n t i c o n s e q u e n t i a l i s t s o u g h t to be saying is s o m e t h i n g t h a t c o n s e q u e n tialists w h o u n d e r s t a n d t h e issue c a n also say: t h a t t h e r e a r e s o m e
c o n s e q u e n c e s which are morally relevant, and that we o u g h t to bring
about, or not bring about, those c o n s e q u e n c e s regardless of the other
c o n s e q u e n c e s w h i c h a r e m o r a l l y i r r e l e v a n t . T h u s J o u g h t to s p e a k
the truth and so inform the o t h e r party of it. even t h o u g h there will
also be the c o n s e q u e n c e t h a t I am disadvantaged thereby. It is still the
i n t e n t i o n t o bring a b o u t t h e c o n s e q u e n c e t h a t h e i s m i s i n f o r m e d
which makes telling a lie wrong. Kant could not have disagreed.
A further point of objection is related but slightly different. S o m e
of the consequences of actions are intended and some not. W h e n we
are speaking of the 'moral worth of the agent', or wondering w h e t h e r
to blame him, it is of course relevant w h e t h e r he intended the c o n s e quences or not. We can say, with Kant, that the only good thing without qualification is a good will (Gr BA1 = 393), m e a n i n g t h a t people
are judged by their intentions and not by the actual c o n s e q u e n c e s .
B u t let us for the present leave aside these post eventum j u d g e m e n t s
and consider the situation of s o m e o n e w h o is trying to decide w h a t to
III. 8. 9
165
do. He is trying to decide what to do intentionally, i.e. w h a t intention

to form; for we c a n n o t decide to do s o m e t h i n g u n i n t e n t i o n a l l y (if it
were u n i n t e n t i o n a l , we could n o t speak of o u r h a v i n g decided to do
it). W h e n w e a r e w o n d e r i n g w h a t i n t e n t i o n t o form, the i n t e n t i o n s
that are the possible candidates are all intentions to bring about c e r tain c o n s e q u e n c e s ; t h a t is, to do certain a c t i o n s or to m a k e the course
of events different in c e r t a i n ways. So t h e will itself, w h i c h is b e i n g
formed in this deliberative process, is a will to bring about certain c o n s e q u e n c e s . T h e y a r e w h a t is willedthe o b j e c t s of volition, as K a n t
calls t h e m . S o , a l t h o u g h the only good thing w i t h o u t qualification is
a good will, w h a t makes it a good will is what is willed (autonomously,
universally, rationally, and impartially), and t h a t is the c o n s e q u e n c e s
t h a t are intended.
Clearly I have been able only to scratch the surface of my question.
T h e r e are m a n y further points of difficulty in interpreting K a n t t h a t I
have not had r o o m to raise, let a l o n e discuss. T h e limit of my a m b i tion has been to get intuitionists, deontologists, a n d c o n t r a c t u a l i s t s ,
who are so sure t h a t Kant was on their side against utilitarianism, to
look m o r e carefully at his (admittedly o b s c u r e ) t e x t . I am confident
that, like me. they will at least find m a n y utilitarian elements in it.
R E F E R E N C E S AND B I B L I O G R A P H Y
i.
Complete Bibliography of Writings of R. M.
Hare
References in the text of the form 'H 1971a: 100' are to this part of the bibliography, the last figure being the page except where otherwise indicated. Dates
from 1997 are conjectural. References to The Language of Morals (1952b),
Freedom and Reason (19630), and Moral Thinking (1981a) take the form of the
letters 'LM', ' F R ' , and ' M r , respectively, followed by the section number.
References of the form '5.3' are to sections of this volume. The author is heavily indebted to Ulla Wessels. whose bibliography appears in H 1995a. Reprints
and translations into other languages are included where I have records of
them, but these are incomplete. Abstracts are given of the more important
recent papers.
1949. 'Imperative Sentences', Mind 58. Repr. in 1971c.
1950a. Review of E. W. Hirst, Morality and God, Philosophy 25.
1950b. Review of H. A. Prichard, Moral Obligation and Knowledge and
Perception, Oxford Magazine (15 June).
1950c. 'Theology and Falsification', University 1. Repr. in A. G. N. Flew and
A. Maclntyre, eds.. New Essays in Philosophical Theology (London: SCM
Press, 1955); in R. E. Santoni. ed., Religious Language (Bloomington, Ind.:
Indiana University Press, 1968): in 19920": and in other collections. Italian
translation in collection ed. by G. Gava (Liviana, 1972).
1951a. 'Freedom of the Will', Proceedings of the Aristotelian Society, supp. vol.
25. Repr. in 1972b.
1951 b. Review of G. (.'. Field, The Philosophy of Plato, Mind 6 0 .
1951 <:. Review of S. E. Toulmin, An Examination of the Place of Reason in Ethics,
Philosophical Quarterly 1.
19510". Review of R. Lepley, ed.. Value: A Cooperative Inquiry, Mind 60.
1952a. Review of H. D. I,ewis, Morals and Revelation, Philosophy 27.
1952/). The Language of Morals (Oxford: Oxford University Press). Italian translation, Rome: Ubaldini, 1968: German, Frankfurt a. M.: Suhrkamp, 1972:
Japanese, Tokyo: Keiso Shobo, 1981; Portuguese, So Paolo: Martins
Fontes, 1996. Extracts in Danish translation, Copenhagen: Gyldendalike,
1976.
1954. Review of E. W. Hall, What is Value?, Mind63.

1954b. Review of The FAhics of Aristotle, trans. J. A. K. Thompson. Oxford
Magazine 62 (25 Feb.).
1954c Review of J. Wisdom. Philosophy and Psychoanalysis, Philosophy 29.
1955c Universalisability'. Proceedings of the Aristotelian Society 55. Repr. in
1972ft. German translation in G. Grewendorf and G. Meggle. eds.. Seminar:
Sprache und Ethik (Frankfurt a. M.: Suhrkamp, 1974): Spanish in E. Rabossi
and F. S a l m e r n , eds., Etica;/ Anlisis 1 (Mexico City: Mexico University
1960/7. 'A School for Philosophers', Ratio 2 (also in German edition). Repr. in
168
Press. 1976).
1955/).'Ethics and Politics' (two articles and letters). Listener (Oct.). Spanish
translation in Revista Universidad de San Carlos 33. First article repr. in
I972. in M. Wakin, ed.. War, Morality and the Military Profession (Boulder,
Colo.: Westview, 1979), and in P. Werhane, ed.. Ethical Issues in Business
(Englewood Cliffs. NJ: Prentice-Hall. 1991).
19560. Review of P. H. Nowell-Smith. FAhics. Philosophy 31.
1956b. Review of U. Scarpelli, Filosofia analtica e giurisprudenza, Mind 65.
1 9 5 7 0 . 'Geach: Good and Evil'. Analysis 1 7 . Repr. in P. Foot, ed., Theories of
Ethics (Oxford: Oxford University Press. 1967) and in 1 9 7 2 / ) . Spanish
translation in E. Rabossi and F. S a l m e r n , eds.. Etica y Anlisis 1 (Mexico
City: Mexico University Press. 1 9 7 6 ) .
1957/7. 'Oxford Moral Philosophy' (letters). Lisiencr (21 Feb. and 28 Mar.).
1957c. Review of A. J. Ayer, The Problem of Knowledge and B. Russell, Logic and
Knowledge. Spectator (4 Jan.).
1957/. Are Discoveries about the Uses of Words Empirical?', journal of
Philosophy 34. Full version in 19600.
1957c. Review of J. 0. Urmson. Philosophical Analysis and A. J. Ayer et al. The
Revolution in Philosophy, Philosophische Rundschau 5 (in German).
1 9 5 7 / . 'Religion and Morals' in B. G. Mitchell, ed., Faith and Logic (London:
Allen & Unwin).
1 9 5 9 . 'Broad's Approach to Moral Philosophy', in P. Schilpp, ed.. The
Philosophy of C. 1). Broad (New York: Tudor). Repr. in 1 9 7 1 c and in 1 9 7 K / .
19600. 'Philosophical Discoveries' (the full version of 19570"), Mind 69. Repr.
in A. Sesonske and N. Fleming, eds.. Plato's Meno (Belmont. Calif.:
Wadsworth. T965): R. Rorty, ed.. The Linguistic Turn (Chicago, 111.: Chicago
University Press, 1967); C. Lyas, ed.. Philosophy and Linguistics (London:
Macmillan. 1971): N. Bowie, ed., The Tradition of Philosophy (Belmont,
Calif: Wadsworth, 1986): and in 1971c. German translation in G.
Grewendorf and G. Meggle, eds., Linguistik und Philosophie (Frankfurt a.
M.: A t h e n u m . 1974).
I97TD.
1960c. Review of F. E. Sparshott, An Enquiry into Goodness, Philosophical

Quarterly 10.
1960t/. '"Rien n'a d'importance": l ' a n a n t i s s e m e n t des valeurs est-il pensable?', also discussion of other papers, in L. Beck. ed.. La Philosophie analytique (Paris: Minuit). English version. 'Nothing Matters', in 19720". Repr. in
T. Beauchamp, ed.. Death and Dying (Englewood Cliffs. NJ: Prentice-Hall.
J978): E. Klemke, ed., The Meaning of Life (New York: Oxford University
Press. 1981); J. Halberstam. ed., Virtues and Values (Englewood Cliffs, NJ:
Prentice-Hall. 1987). and other collections. German translation (abridged)
in C. Fehige and G. Meggle. eds., Der Sinn des Lebens (Munich: 1995).
r96oe. 'Ethics', in J. 0. Urmson, ed., Encyclopedia of Western Philosophy and
Philosophers (London: Rainbird). Repr. in M. Lipman, ed., Growing Up with
Philosophy (Philadelphia, Pa.: Temple University Press, 1978), and in
1972ft.
1962. Review of M. Singer. Generalization in Ethics, Philosophical Quarterly 12.

196 30. Freedom and Reason (Oxford: Oxford University Press). Italian translation. Milan: II Saggiatore, 1971: German, Dsseldorf: Palmos. 1975, republished Frankfurt a. M.: Suhrkamp, 1983.
1963/7. 'Descriptivism', Proceedings of the British Academy 49. Repr. in W. D.
Hudson, ed., The Is-Ought Question (London: Macmillan. 1969) and in
1972/7. German translation in G. Grewendorf and G. Meggle. eds., Seminar:
Sprache und Ethik (Frankfurt a. M.: Suhrkamp, 1974); Spanish in E. Rabossi
and F. S a l m e r n , eds., Etica y Anlisis 1 (Mexico City: Mexico University
Press. 1976); Serbo-Croat in collection ed. by J. Babi-Avdispabic (Sarajevo:
Svjetlost, T987).
1963c. Letter in Times Literary Supplement (26 Apr.) on review of Freedom and
Reason.
19640. 'Pain and Evil', Proceedings of the Aristotelian Society, supp. vol. 38.
Repr. in J. Feinberg. ed.. Moral Concepts (Oxford: Oxford University Press.
1969) and in 1972/7.
1964/1. 'Adolescents into Adults', in T. C. B. Hollins, ed., Aims in Education

(Manchester: Manchester University Press). Repr. in 1972a and 1992/.
1964c. A Question about Plato's Theory of Ideas', in M. Bunge, ed.. The
Critico/ Approach: Essays in Honor of Karl Popper (Glencoe. III.: Free Press of
Glencoe). Repr. in 19710".
1 9 6 4 a . 'The Promising Game'. Revue Internationale de Philosophie 70. Repr. in
P. Foot, ed.. Theories of Ethics (Oxford: Oxford University Press. 1967):
I70
REFERENCES
AND
REFERENCES
BIBLIOGRAPHY
W. D. Hudson, ed., The ls-Ought Question (London: Macmillan, 1969); and

in 1989b. Spanish translation in E. Rabossi and F. S a l m e r n , eds.. Etico y
Anlisis i (Mexico City: Mexico University Press, 1976), Polish in Etyka
(1979)1 9 6 4 c 'Wat is Leven?', Elseviers Weekblad (in Dutch). English version in 1965b.
1964/. 'The Objectivity of Values', Common Factor 1.
1965a. Review of G. H. von Wright. Norm and Action, Philosophical Quarterly 15.
1965b. 'What is Life?', Crucible, (English version of 1964c). Repr. in 19720".
1965c. 'Plato and the Mathematicians', in R. Banibrough, ed.. New Essays on
Plato and Aristotle (London: Routledge). Repr. in 19710".
1966. 'Peace', RSA Lecture, Australian National University, Canberra, privately printed. Repr. in J. Rachels, ed., Moral Problems (New York: Harper
& Row, 1970), and in 19720".
19670. 'The Lawful Government', in P. Laslett and W. G. Runciman, eds.,
Philosophy, Politics and Society (Oxford: Blackwell). Repr. in J. Rachels and
F. Tillman, eds., Moral and Social Problems (New York: Harper & Row, 1971),
and 19720".
1967b. 'Conventional Morality', 'Decision', 'Deliberation'. 'Ethics', 'Intention', and 'Right and Wrong', in J. Macquarrie, ed., Dictionary of Christian
Ethics (London: SCM Press).
1967c. Review of S. Hampshire, Freedom of the Individual, Philosophical Review 76.
1967a. 'Some Alleged Differences between Imperatives and Indicatives', Mind
76. Repr. in 1971c.
19680. Review of G. J. Warnock, Contemporary Moral Philosophy, Mind 77.
1968b. Review of R. S. Peters, The Concept of Education, Mind 77.
19690. 'Practical Inferences', in V. Kruse, ed., Festskrift til Alf Ross (Copenhagen: Juristforbundets Vorlag). Repr. in 1971c.
1969b. Review of A. Ross, Directives and Norms, Mind 78.
1969c. 'Community and Communication', in S. Verney, ed., People and Cities
(London: Fontana). Repr. in i972d.
1969a. Review of B. G. Mitchell, Law, Morality and Religion, Philosophy 4 4 .
1970a. 'Meaning and Speech Acts', Philosophical Review 79. Repr. in 1971c.
1970b. 'Condizioni intellettuali per la sopravvivenza dell' uomo', Proteus 1.
1970c. Reply to R. S. Katz, 'Liberals, Fanatics and Not-so-innocent
Bystanders', )owett Papers, 1 9 6 8 - 1 9 6 9 , ed. B. Y. Khanbhai et al. (Oxford:
Blackwell).
i97od. General Introduction and Introduction to Meno in paperback edition
of The Dialogues of Plato, trans. B. Jowett, ed. R. M. Hare and D. A. F. M.
Russell (London: Sphere Books).
AND
BIBLIOGRAPHY
171
1971a. Review of L. van der Post, The Prisoner and the Bomb, New York Review
of Books 17 (20 May).
1971b. 'Wanting: Some Pitfalls', in R. Binkley et al, eds., Agent, Action and
Reason (Toronto, O n t : Toronto University Press). Repr. in 1971c. German
translation in A. Beckermann and G. Meggle, eds.. Analytische Handlungstheorie (Frankfurt a. M.: Suhrkamp, 1985).
1971c. Practical Inferences, containing 1 9 4 9 , 1 9 6 9 0 . 1 9 7 0 0 , 1 9 7 1 b , 'Austin's
Distinction between Locutionary and Illocutionary Acts', and new appendices (London: Macmillan).
i97id. Essays on Philosophical Method, containing 1959a, 19600, 1960b,
1964c, 1965c, 'The Practical Relevance of Philosophy", and 'The Argument from Received Opinion' (London: Macmillan). Italian translation,
Armando, 1977.
i 9 7 i e . 'Drugs and the Role of the Doctor', and other contributions to I. T.
Ramsey and R. Porter, eds., Personality and Science (Edinburgh: Churchill
Livingstone).
1972a. 'Principles', Proceedings of the Aristotelian Society 73. Repr. in 1989b.
1972b. Essays on the Moral Concepts, containing 1951a (part), 1955a, 19570,
1 9 6 0 c 1963b, 1964a, and 'Wrongness and Harm' (London: Macmillan).
1972c. 'Rules of War and Moral Reasoning', Philosophy and Public ffairs 1.
Repr. in M. Cohen et al., eds., War and Moral Responsibility (Princeton, NJ:
Princeton University Press, 1974), and in 1989c.
i972d. Applications of Moral Philosophy, containing 1955b (first article), 19600"
(English version), 1964b, 1965b, 1 9 6 6 , 1 9 6 7 0 , 1 9 6 9 c , 'Reasons of State',
and 'Function and Tradition in Architecture' (London: Macmillan).
Japanese translation, Tokyo: Keiso Shobo, 1981.
1972. Review of G. J. Warnock, The Object of Morality, Ratio 14 (also in
German edition).
1972/. 'Wissenschaft und praktische Philosophie', in A Diemer, ed., Proc. of 9.
Deutscher Kongress fr Philosophie und Wissenschaft (Meisenheim: Hain).
1973a. Critical Study, 'Rawls' Theory of Justice' I and II, Philosophical
Quarterly 23. Repr. in N. Daniels, ed., Reading Rawls (Oxford: Oxford
University Press, 1975), and in 1989b.
1973b. 'The Simple Believer', in G. Outka and J. P. Reeder, eds., Religion and
Morality (New York: Anchor Press). Repr. in T. Beauchamp et al., eds.,
Philosophy and the Human Condition (Englewood Cliffs, NJ: Prentice-Hall,
1980), and in i992d.
i97 c.'Language and Moral Education', in G. Langford and D. J. O'Connor,
eds., New Essays in the Philosophy of Education (London: Routledge). Repr.
3
i 2
7
in 19790" with criticism by G. J. Warnock and reply, and in 19920" with reply
only.
1973d. 'Sad Moralny (Moral Judgment)', Etyka 11 (Warsaw), with abstracts in
English and Russian.
1974a. Comment on R. Edgley, 'Reason and Violence', in S. Krner, ed..
Practical Reason (Oxford: Blackwell).
7974b. 'The Abnormal Child: Moral Dilemmas of Doctors and Parents',
Documentation in Medical Ethics ~\. Repr. as 'Survival of the Weakest' in S.
Gorovitz, ed.. Moral Problems in Medicine (Englewood Cliffs, NJ: PrenticeHall, 1976), and in 1993c. German translation in A. Leist, ed.. Um lieben
und Tod (Frankfurt a. M.: Suhrkamp. 1990).
1974c. 'Platonism in Moral Education: Two Varieties'. Monist 58. Repr. in
i992d. French translation in M. Canto-Sperber. ed.. volume of essays on
Plato's Meno (Paris: Jacob, 1990).
i974d. 'What Use is Moral Philosophy', TV discussion with A. J. P. Kenny, in
Philosophy in the Open (Milton Keynes: Open University).
19750. 'Contrasting Methods of Environmental Planning', in R. S. Peters, ed.,
Nature and Conduct, Royal Institute of Philosophy Lectures (London:
Macmillan). Repr. in K. Goodpaster and K. Sayre, eds., Ethics and Problems
of the 21st Century (Notre Dame: University of Notre Dame Press, 1979),
and in 1989c.
/
1975b. Autonomy as an Educational Ideal', in S. C. Brown, ed.iPhilosophers
Discuss Education (London: Macmillan). Repr. in I992d.
1975c. 'Abortion and the Golden Rule', Philosophy and Public Affairs 4. Repr. in
R. Baker and F. Elliston, eds., Philosophy and Sex (Buffalo, NY: Prometheus,
1975): in Moral Problems, ed. J. Rachels (New York: Harper & Row, 1978);
in D. Goldberg, ed., Ethical Theory and Society (New York: Holt Rinehart,
1987); and in 1993c. German translation in A. Leist, ed., Um Leben und
Tod (Frankfurt a. M.: Suhrkamp. 1990); Italian in G. Ferranti and S.
Maffettone, eds., Introduzione alia biotica (Naples: Liguori, 1991).
I975d. 'Euthanasia: A Christian View', Philosophie Exchange 2 (Proceedings of
Center for Philosophie Exchange). Repr. in ig92d.
19760. 'Ethical Theory and Utilitarianism', in H. D. Lewis, ed., Contemporary
British Philosophy 4 (London: Allen & Unwin). Repr. in A. K. Sen and
B. A. Williams, eds., Utilitarianism and Beyond (Cambridge: Cambridge
University Press, 1982); in D. Barnett, ed.. Western Moral Philosophy
(Mountain View, Calif.: Mayfield, 1994): and in 1989b.
1976b. 'Some Confusions about Subjectivity', in J. Bricke, ed.. Freedom and
Morality (Lawrence. Kan.: University of Kansas Press). Repr. in 1989b.
173
1976c. 'Political Obligation', in T. Honderich, ed., Social Ends and Political

Means (London: Routledge). Repr. in 1989c.
1 9 7 6 a . 'Value Education in a Pluralist Society: A Philosophical Glance at the
Humanities Curriculum Project', in R. S. Peters, ed., Proceedings of the
Philosophy of Education Society of Great Britain 10. Repr. in M. Lipman, ed..
Growing Up with Philosophy (Philadelphia, Pa.: Temple University Press,
1978), and in 19920".
1 9 7 6 c Review of H.-N. Castaeda, The Structure of Morality, Journal of
Philosophy 73.
19770. 'Medical Ethics: Can the Moral Philosopher Help?', in S. Spicker andT.
Engelhardt, eds., Philosophy and Medicine 3: Philosophical Medical Ethics, its
Nature and Significance (Dordrecht: Reidel). Repr. in 1993c.
1977b. 'Geach on Murder and Sodomy' (about 'is'-'ought' derivations),
Philosophy 52.
1977c. 'Opportunity for What? Some Remarks on Current Disputes on
Equality in Education', Oxford Review of Education 3. Repr. in 19920".
19770". 'Sprawiedliwosc i rownosc', Etyka 15 (Warsaw) (in Polish, with abstracts in English and Russian). English version in 19780", repr. in 1989c.
Discussed in 1979/1.
19780. 'Prediction and Moral Appraisal', Midwest Studies in Philosophy .
Repr. in 1993c.
1978b. 'Relevance', in A. Goldman and J. Kim, eds., Values and Morals
(Dordrecht: Reidel). Repr. in 1989b.
1978c. 'Moral Philosophy', TV interview with B. Magee in his Men of Ideas
(London: BBC).
19780". Justice and Equality', in J. Arthur and W. Shaw, eds., Justice and
Economic Distribution (Englewood Cliffs, NJ: Prentice-Hall). Revised
English version of 19770". Repr. in J. Sterba, ed., Justice: Alternative Political
Perspectives (Belmont, Calif.: Wadsworth, 1991); in W. Kymlicka, ed.,
Justice in Political Philosophy (Aldershot: Elgar, 1992): in L. Pojman and R.
Westmoreland, eds., Equality (New York: Oxford University Press, 1997);
in M. Hajdin, ed., The Notion of Equality (forthcoming); and in 1989c.
Discussed in 1979/1.
3
1979a. 'What Makes Choices Rational?', Review of Metaphysics 32. Repr. in

1989b.
1979b. 'What is Wrong with Slavery', Philosophy and Public Affairs 8. Repr. in
J. Arthur, ed., Morality and Moral Controversies (Englewood Cliffs, NJ:
Prentice-Hall, 1979); in P. Singer, ed., Readings in Applied Ethics, (Oxford:
Oxford University Press, 1986); in T. Lott, ed.. Slavery and Social Philosophy
174
(Totowa, NJ: Rowman & Littlefield, forthcoming); and in 1989c. German

translation in Conceptus 15 (1981).
1979c. 'Non-descriptivist Ethics' and 'Utilitarianism' in W. Reich, ed.,
Encyclopedia of Bioethics (New York: Free Press).
1979a. Reprint of 1973c with criticism by G. J. Warnock and reply, in D. B.
Cochrane et ah. The Domain of Moral Education (Toronto. Ont.: Paulist
Press). Repr. without Warnock's criticism in 1992a.
1979c. 'Universal and Past-Tense Prescriptions: A Reply to Mr. Ibberson',
Analysis 39.
!979/- On Terrorism', Journal of Value Inquiry 13. Repr. in 1989c.
1979a. 'Utilitarianism and the Vicarious Affects', in E. Sosa, ed., The Philosophy of Nicholas Resetter (Dordrecht: Reidel). Repr. in 1989b.
1979/1. 'Justice and Equality: Some Comments on the 1979 Symposium in
Warsaw', Dialectics and Humanism 6 (Warsaw). Contains discussion of
19780".
19791. 'Behaviour Therapy and Moral Responsibility', Midwife, Health Visitor
and Community Nurse (London).
1980a. 'Moral Conflicts', in S. McMurrin, ed.. The Tanner Lectures on Human
Values vol. i (Salt Lake City, Utah: University of Utah Press and Cambridge:
Cambridge University Press). Revised in 1981a, chs. 2 f.
x
1980b. Review of T. Nagel. Mortal Questions, Philosophical Books 21.
/
1981a. Moral Thinking: lis Levels, Method and Point, with bibliography of R. "M.
Hare's writings 1 9 7 1 - 8 0 (Oxford: Oxford University Press). Italian translation, Bologna: II Mulino, 1989; Chinese. Hong Kong: Philosophia, 1992;
German, Frankfurt a. M.: Suhrkamp. 1992: Japanese, Tokyo: Keiso, 1994;
Swedish, Stockholm: Thales, 1995.
1981b. Review of P. Singer, The Expanding Circle, New Republic (7 Feb.).
1981c. 'On a Misunderstanding of Geach's', Analysis 41.
19810". 'On the Possibility of a Rational Foundation of Norms', in A. S. Skiadas,
ed., Scientific and Extra-scientific Rationality (Proceedings of conference of
Ellenike Anthropistike Etaireia at Portaria, Thessaly, 1978) (Athens).
1982a. Plato (Oxford: Oxford University Press). German translation, Stuttgart: Reclam, 1990; Spanish, Madrid: Alianza, 1992; Malaysian. Kuala
Lumpur: DBP, 1993; Romanian. Bucharest: Ilumanitas. 1997; Polish,
Warsaw: Urbanski, forthcoming; Portuguese, Lisbon: P r e s e n a , forthcoming: Indonesian, Jakarta: Graflti. forthcoming. Repr. in R. M. Hare, J.
Barnes, and H. Chadwick, Founders of Thought (Oxford: Oxford University
Press, 1991), Czech translation, Prague: Svoboda, 1994; Chinese translation forthcoming
175
1982b. 'Utilitarianism and Double Standards: A Reply to Dr. Annas', Oxford

Review of Education 8.
1982c. 'Moral Philosophy: Some Waymarks' (in Hebrew), in A. Kasher and
S. Lappin, eds.. New Trends in Philosophy (Tel Aviv: Yachdav).
1982a". Interview with Carl Rudbeck, Svenska Dagbladet, Feb. (in Swedish).
1983a. 'Philosophical Introduction', in S. Bloch and P. Chodoff, eds..
Psychiatric Ethics (Oxford: Oxford University Press). Repr. as 'The
Philosophical Basis of Psychiatric Ethics', in 1 9 9 3 c
1983b. 'On the Scientific Justification of Norms', in A. Diemer, ed.. 16.
Weltkongress fur Philosophie (Frankfurt a. M.: Peter Lang).
1984a. 'Do Agents Have to be Moralists?', in E. Regis, Jr., Gewirth's Ethical
Rationalism (Chicago, 111.: University of Chicago Press).
1984b. 'Supervenience', Proceedings of the Aristotelian Society, supp. vol. 56.
Repr. in 1989b.
1984c. 'Utility and Rights: Comment on David Lyons' Paper', Nomos 2 4 :
Ethics, Economics and the Law. Repr. in 1989c.
19840". 'Arguing about Rights', Emory Law Journal 33. Repr. in 1989c.
1984c 'Liberty and Equality: How Politics Masquerades as Philosophy', Social
Philosophy and Policy 2. Repr. in 1989c.
1984/. 'Some Reasoning about Preferences: A Response to Essays by Persson,
Feldman and Schueler', Ethics 95. Repr. in 1989b.
1984a. 'Rights, Utility and Universalization: a Reply to John Mackie', in R.
Frey, ed., Utility and Rights (Minneapolis, Minn.: University of Minnesota
Press). Repr. in 1989c.
1985a. 'Ontology in Ethics', in T. Honderich, ed., Morality and Objectivity:
Tribute toJ.L. Mackie (London: Routledge). Repr. in 1989b.
1985b. 'Philosophy and Practice: Some Issues about War and Peace', in A. P.
Griffiths, ed.. Philosophy and Practice (Royal Institute of Philosophy
Lectures 19, supp. to Philosophy 59) (Cambridge: Cambridge University
Press). Repr. in 1989c.
1985c. 'Little Human Guinea-Pigs?', in M. Lockwood, ed.. Moral Dilemmas in
Modern Medicine (Oxford: Oxford University Press). Repr. in 1993c.
19850". 'The Ethics of Experimentation on Human Children', in R. B. Marcus
et al. eds., Logic, Methodology andPhilosophy of Science 7 (Proceedingsof 7th
International Congress of Logic, Methodology and Philosophy of Science,
Salzburg 1983) (Amsterdam: North Holland). No overlap with 1985c.
1 9 8 5 c 'Come decidere razionalmente le questioni morali', in L. Gianformaggio and E. Lecaldano. eds., Etica e diritto: le vie delta giustificazione
razionale (Rome: Laterza). English version in 19861, repr. in 1989b.
1 7
1985/. Comment on T. Beauchamp, 'Manipulative Advertising', Business and

Professional Ethics Journal 3.
19860. A Kantian Utilitarian Approach', in G. Ezorsky, ed., Moral Rights in the
Workplace (New York: State University of New York Press). Excerpt from
1989c, ch. 11.
1986b. 'Why Do Applied Ethics?', in R. M. Fox and J. P de Marco, eds., New
Directions in Ethics (London: Routledge). Repr. in R. Shehadi and D. M.
Rosenthal, eds.. Applied Ethics and Ethical Theory (Salt Lake City. Utah:
University of Utah Press, 1988). and in 1989b. Italian translation in M.
Mori, ed., Questioni di biotica (Rome: Riuniti, 1988).
1986c. A Reductio ad Absurdum of Descriptivism', in S. Shanker. ed.,
Philosophg in Britain Today (London: Croom Helm). Repr. in 1989b.
19860". 'Health'. Journal of Medical Ethics 12. Repr. in 1993t'1986c. 'Liberty, Equality and Fraternity in South Africa?', in South African
Journal of Philosophy 5. Also in Philosophical Forum 8 (1986). Repr. in P.
Collins, ed., Thinking about South Africa (Hemel Hempstead: Harvester
Wheatsheaf, 1990). and in 1989c.
1986/. 'Punishment and Retributive Justice', Philosophical Topics 2. Papers on
Ethics, ed. J. Adler and R. N. Lee. Repr. in J989C.
1986.0. 'Universalizability' and 'Utilitarianism', and articles listed under
1967b. in J. Childress and J. Macquarrie, eds.. New Dictionary of Christian
Ethics (London: SCM Press), also published as Westminster Dictionary of
Christian Ethics (Philadelphia, Pa.: Westminster Press).
1986/1. 'Warunkowe i bezwarunkowe obowiazywanie norm moralnych' ('The
Conditional and Unconditional Validity of Moral Norms') (reply to Ija
Lazari-Pavlowska's paper of the same name). Etyka 22 (Warsaw).
19861. 'How to Decide Moral Questions Rationally', Crtica 18 (English version of 1985c). Repr. in 1989b.
1986/'. Comment on Putnam, Critica 18, Appendix.
1987a. An Ambiguity in Warnock'. Bioethics 1.
1987b. 'Moral Reasoning about the Environment'. Journal of Applied
Philosophy 4. Repr. in 1989c.
1987c. Review of B. Mayo, The Philosophy of Right and Wrong. Philosophical
Quarterly 37.
19870". 'In Vitro Fertilization and the Warnock Report', in R. F. Chadwick, ed.,
Ethics. Reproduction and Genetic Control (London: Croom Helm). Repr. in
1993'"1987c 'Why Moral Language?', in P. Pettit et al., eds.. Metaphysics and Morality: Essays in Honour of ]. J. C. Smart (Oxford: Blackwell). Repr. in 19920".
177
1987/ Comment on J. Kaufman, 'Hamlethics in Planning', Business and

Professional Ethics Journal 6.
1987a- 'Embryo Experimentution: Public Policy in a Pluralist Society',
Bioethics News 7. Repr. in K. Dawson and J. Hudson, eds.. IVF: The Current
Debate (conference proceedings) (Clayton, Vic.:i987); in H. Kuhse et al,
eds.. Embryo Experimentation (Cambridge: Cambridge University Press,
1990): in K. W. M. Fulford . ed.. Medicine and Moral Reasoning (Cambridge:
Cambridge University Press 1994): and 1993c. German translation in
H.-M. Sass, ed., Medizin und Ethik (Stuttgart: Reclam, 1989).
1987. La enseanza de la tica mdica: la contribucin de la filosofa', JANO 33.
1988a. 'When Does Potentiality Count? A Comment on Lockwood', Bioethics
2. Repr. in 1993c and in N. Fotion and J. C. Heller, eds., Contingent Future
Persons (Dordrecht: Kluwer, forthcoming).
1988/). 'Possible People'. Bioethics 2. Repr. in 1993c.
1988c. Comments on R. B. Brandt, W. K. Frankena. A. Gibbard. J. Griffin, J. C.
Harsanyi, W. D. Hudson. T. Nagel, D. A. J. Richards, T. M. Scanlon, P.
Singer, J. O. Urmson. Z. Vendler, and B. Williams, in D. Seanor and N.
Fotion, eds.. Hare, and Critics (Oxford: Oxford University Press).
19880". 'A Kantian Approach to Abortion' in M. Bayles and K. Henley, eds.,
Right Conduct: Theories and Applications (New York: Random House). Repr.
in S. Luper-Foy and C. Brown, eds., The Moral Life (New York: Holt
Rinehart, 1991), and i n i 9 9 c , and with commentary and reply by R. B.
Brandt in Social Theory and Practice 15 (1989). Spanish translation in
Dianoia 36 (1990).
1988c. 'The Poverty of Ideas', Guardian (11 Oct.). Repr. in Political Studies
Association (June 1990).
1989a. 'Some Sub-atomic Particles of Logic', Mind 98.
3
Four constituents in the expression of speech acts are distinguished: (1) the sign of
mood (indicative, imperative, etc.) or tropic: (2) the sign of subscription or neustic
(l-rege's judgement-stroke); (3) the sign of completeness or clistic: (4) the indication
of the content of the speech act. or phrastic. All contribute to the meaning and logical properties of speech acts. Various ordinary-language signs performing these
functions are noticed, and (2) defended against the usual objections of Wittgenstein
and others. It is asked which of these particles include which in their scopes, and in
what senses truth-value attaches to phrastics with or without the other particles.
1989ft. Essays in Ethical Theory (Oxford: Oxford University Press). Contains
19640*, 19720,1973a, 1976a, 1976"), 1978b, 1979a, 1979a, 1984b, 1984/, 1985a,
1986b, 19861'. and 19940". Italian translation, Milan: II Saggiatore, 1992.
1989c. Essays on Political Morality (Oxford: Oxford University Press). Contains
1972c, 1 9 7 5 0 , 1 9 7 6 c 19780", 1979b. 1979/. 1984c, 19840". 1984c 19843.
i 8
7
1985b, 1986c 1986/, 1987b, 'The Role of Philosophers in the Legislative Process', 'Rebellion', and 'The Rights of Employees: The European
Court of Human Rights and the Case of Young, James and Webster'.
Italian translation, Milan: II Saggiatore. 1995: Arabic (in part), Beirut:
Saqi, 1996.
i989<f. 'Brandt on Fairness to Happiness', with reply by R. B. Brandt, Social
Theory and Practice 15.
1989c. 'Una a p r o x i m a c i n kantiana a la poltica sanitaria', Agora 8. Repr. in
part as 'Health Care Policy: Some Options' in 1993c
1989/. Replies to Persson. Rabinowicz, Sandoe, and Wetterstrm, Theoria 55.
1989g. Interview with E Apsden. Times Higher Education Supplement (June).
1991a. 'Universal Prescriptivism', in P. Singer, ed., A Companion to Ethics
(Oxford: Blackwell). German translation in C. Fehige and G. Meggle, eds.,
Zum moralischen Denken (Frankfurt a. M.: Suhrkamp, 1995).
1991b. 'Are there Moral Authorities?' in D. R. Bromham et al, eds.,
Reproductive Medicine (Berlin: Springer). Repr. in 1 9 9 2 a .
199IC. 'Kant utilitarista?', Materiali per una storia della cultura giuridica 21.
English version in 1993a, German in C. Fehige and G. Meggle, eds., Zum
moralischen Denken (Frankfurt a. M.: Suhrkamp, 1995).
1992a. 'Morality. Moral Theory and Applied and Professonal Ethics: Reply to
Bernard Gert', Professional Ethics 1.
1992b. 'One Philosopher's Approach to Business and Professional Ethics',
Business and Professional Ethics Journal 11. Repr. in C. Cowton and R. Crisp,
eds.. Business Ethics (Oxford: Oxford University Press, forthcoming).
1992c. 'What are Cities For? The Ethics of Urban Planning', in C. C. W. Taylor,
eds.. Ethics and the Environment (Oxford: Corpus Christi College). German
translation in 1995a: Italian in S. Moroni, ed.. collection on planning and
social justice (Milan: Franco Angeli, forthcoming).
1 9 9 2 . Essays on Religion and Education (Oxford: Oxford University Press).
Contains 1950c. 1957/. 1964b. 1973b. 1973c. I 9 7 4 C J975'- >975d. 197^''.
1977c, 1 9 7 9 a . 1987c 'How did Morality Get a Bad Name?', and 'Satanism
and Nihilism'.
1 9 9 2 C 'Moral Terms'. 'Prescriptivism', 'Slavery', 'Universalizability', and
"Weakness of Will', in L. Becker, ed., Encyiopedia of Ethics (New York:
Garland).
1992/. 'Utilitarianism and Moral Education: Comment on S. Levy's Paper',
Studies in Philosophy and Education 11.
1993a Could Kant have been a Utilitarian?', Utilitas 5. Also in R. M. Dancy,
ed., Kant and Critique (Dordrecht: Kluwer, 1993). Repr. in 1998c.
179
1993b. 'The Ethics of Medical Involvement in Torture: Comment on R.

Downie's Paper', Journal of Medical Ethics 19.
1993c Essays on Bioethics (Oxford: Oxford University Press). Contains 1974b,
1975c, 1977a, 1978a, 1983a. 1985c, 19860". 19870", 19873,1988a. 1988b,
19880", English version of 1989c 'Moral Problems about the Control of
Behaviour', and 'Why I am only a Demi-Vegetarian'.
1993d. Guest Editorial, 'Is Medical Ethics Lost?', and letter, Journal of Medical
Ethics 19.
1993c 'Utilitarianism and Deontological Principles', in R. Gillon, ed..
Principles of Health Care Ethics (Chichester: Wiley).
1993/- 'Brandt's Methods of Ethics', in B. Hooker, ed.. Rationality, Rules and
Utility: New Essays on Richard Brandt's Moral Philosophy (Boulder, Colo.:
Westview).
19939- 'Objective Prescriptions', in E. Villanueva, ed.. Naturalism and
Normativity: Philosophical Issues 4 (Atascadero, Calif.: Ridgeview). Also in
A. P. Griffiths, ed., Etm'cs (Royal Institute of Philosophy Lectures 1992/3)
(Cambridge: Cambridge University Press, 1993).
Objectivity is distinct from factuality. Prescriptions are objective, if all rational
thinkers would agree to them. Attempts, whether intuitionistic or naturalistic, to
attain moral objectivity via facts and truth conditions collapse into relativism, because these vary with cultures, whether the facts invoked are clear and empirical,
or appeal to elusive notions like needs or human flourishing. Objectivity can be
achieved, as Kant saw, only by following the culturally invariant logic of the moral
concepts, and seeking prescriptions or maxims that rationally must be assented to.
These will generate secure descriptive meanings for moral statements, and thus
moral truths invariant through cultures.
1994a. Applied Philosophy and Moral Theory: R. M. Hare talks to Philosophy
Today', Philosophy Today 38.
1994b. 'Methods of Bioethics: Some Defective Proposals', Monash Bioethics
Review 13. Repr. in J. W. Sumner and J. Boyle, eds., Philosophical Perspectives
on Bioethics (Toronto, Ont.: University of Toronto Press, 1996).
The right kind of eclecticism in moral philosophy consists in picking out the good
points in all theories and discarding the bad, provided that this leaves one with a
consistent theory. Four defective theories are considered: situation ethics, caring
ethics, virtue ethics and rights-based ethics, and it is shown how to frame a theory
which combines their virtues but avoids their defects.
1994c. 'Philosophic et Conflit', Revue de Metaphysique et Morale 99. English
version in 1997b.
I994d- 'The Structure of Ethics and Morals', in P. Singer, ed.. Ethics (Oxford:
Oxford University Press).
1995a. Replies to Birnbacher, Corradini. Fehige. Hinsch, Hoche, Kusser,
19960". 'A New Kind of Ethical Naturalism?', in P. French et al, eds., Moral
181
Kutschera, Lampe. Leist, Lenzen, Lumer, Millgram, Morscher, Nida-
Concepts, Midwest Studies in Philosophy 20 (Notre Dame, Ind.: Notre Dame
R m e l i n , Rohs. Schaber, S c h n e - S e i f e r t . Spitzley. Stranzinger, Trapp,
University Press).
Vogler. Wimmer. and Wolf, in C. Fehige and G. Meggle. eds.. Zum moralis-
There is not much new about the 'new realism'. Ethical naturalists in particular
should not draw support from recent developments in metaphysics. Putnam's attempt, with his twin-earth argument, to establish a metaphysical necessity distinct
from logical or conceptual and from causal necessity fails, as can be shown by arguments derived from Von Wright and Sidelle. But even if it be accepted, Horgan
and Timmons (1992) have shown that Moore's open question argument can be
adapted to refute the new naturalism. This move depends on a distinction similar to
mine between evaluative and descriptive meaning.
chen Denken (Frankfurt a. M.: Suhrkamp). Contains also German translations of 1991a. 1992c, and 199 }fl.
1995/). 'Off on the Wrong Foot', in J. Couture and K. Nielsen, eds., On the
Relevance
of
Metaethics:
New
Essays
on
Metaetliics.
Canadian
Journal
of
Philosophy Supp. 21 : a reply to P. R. Foot, 'Does Moral Subjectivism Rest on

a Mistake?'. Oxford journal of Legal Studies 15 ( 1995)Foot shows her misunderstanding of myself and the issues by calling me a subjectivist and a non-cognitivist. A non-descriptivist like me can give a clear sense to
'true' and know' as applied lo moral statements, but their truth-conditions may
vary with cultures: so descriptivism. unlike my Kantian prescriptivism. cannot
yield objectivity (II 1991. 19939). Foot's way, following (leach and Anscombe. of
explaining the action-guidingness of moral judgements, via 'specific goods' is
Hawed, as is her account of amoralism.
1996(1. 'Philosophy of Language in Ethics', in M. Dascal et al, eds., Handbuch
Sprachphilosphie (Berlin: De Gruyter). Repr. revised in 1998c.
1996ft. ' I m p r a t i f s , prescriptions et leur logique', in M. Canto-Sperber,
Dictionnaire de philosophie morale (Paris: Presses Universitaires de France).
Normatives and imperatives are different species of prescriptions. Prescriptions have
to contain phrastics. tropics, neustics and clistics (H 1989a). Prescriptions, as illocutionary not perlocutionary acts ( 1.5). are subject to logic, and this makes moral
reasoning possible. Claims that the logic of imperatives is different from that of indicatives are sometimes ill-founded (H 19670"). Probably normative logic is a form of
modal logic, analogous to the usual form (MT 1.6): but that of imperatives is different (no square of opposition). Supervenience applies to normatives but not to imperatives. Normatives have truth-conditions, unlike imperatives, but these vary
with cultures.
1996c. 'Foimdationalism and Coherentism in Ethics', in W. SinnottArmstrong and M. Timmons, eds., Moral Knowledge: New Readings in
Moral Epistemologg (Oxford: Oxford University Press).
This paper attempts to mediate between the advocates of foundationalism and coherentism. Kantian foundationalism (as in the title of the Grundlegung) is viable;
Cartesian foundationalism is not. The latter requires there to be 'foundations'
which are both necessary and substantial: but there cannot be. Kantian foundationalism requires only that we adopt a logically consistent set of maxims or prescriptions; and this we can do. These have to be cotenable by our wills in the world
as it is. There is a unique set of such maxims, consistent with utilitarianism, for our
relations with other people.
1996c. Internalism and Externalism in Ethics', in J. Hintikka and K. Puhl,

eds.,
Proceedings
of
18th
International
Wittgenstein
Congress
(Vienna:
Hlder-Pichler-Tempsky).
Brink's externalism is the view that one can fully endorse a moral judgement without any corresponding motivation. Brink holds that moral judgements could not be
prescriptive unless they were thought to be true; but this would be like saying that
(e.g.) imperatives cannot be prescriptive. The natural properties of actions, etc.. are
linked only contingently, nbt conceptually or metaphysically, to their moral properties. The link is a deliverance of our autonomous will. Brink's argument for realism from the existence of amoralism misfires. But I can support his utilitarianism
by better arguments.
1996/. 'Hare: A Philosophical Self-Portrait', in T. Mautner, ed., A Dictionary of
Philosophy (Oxford: Blackwell).
1997a. 'Preferences of Possible People', in C. Fehige and U. Wessels, eds..
Preferences (Proceedings of Conference in S a a r b r c k e n , 1 9 9 2 ) (Berlin: De
Gruyter).
Hajdin's view in Dialogue 29 (1990), excluding from moral relevance external and
now-for-then preferences, is attractive. But it does not affect my argument for including as relevant the preferences of possible people. This claims that because actual people prefer, if happy, to exist, universalizability requires the extension of this
consideration lo possible people in identical situations. It supports a liberal view on
abortion: if the best family planning and population policy is being followed and
the number of procreated children is determinate, the non-procreation of this child
will make room for another child.
r997ft. Philosophy and Conflict', in 0. Neumaier et al.. eds., Applied Ethics in a
Troubled World (Proceedings of aborted 15th International Wittgenstein
Congress) (Dordrecht: Kluwer). The English version of 1994c.
1997c. Sorting Out Ethics, containing A Taxonomy of Ethical Theories' (the
Axel H g e r s t r m Lectures given in Uppsala in 1991), 'Defence of the
Enterprise', I993. 1996a, and bibliography of R. M. Hare's writings
1949-98 (Oxford: Oxford University Press).
1998a. 'Prescriptivism'. in E. Craig, ed., Routledge Encyclopedia of Philosophy

(London: Routledge).
( I 9 3 5 ) - Philosophy and Logical Syntax (London: Routledge).

(1942). Introduction to Semantics (Cambridge, Mass.: Harvard University
Press).
CARROLL, Lewis (C. L. Dodgson) (1872). Through the Looking Glass (London:
Macmillan). Cited from Nonesuch edn. (London, 1939).
CASTAEDA, H . (1974)- The Structure of Morality (Springfield, Mass.: Thomas).
CHOMSKY, N. (1965). Aspects of the Theory of Syntax (Cambridge, Mass.: MIT
Press).
CUMMISKEY, D. (1990). 'Kantian Consequentialism'. Ethics 100.
DRYDEN, J. (1637). The Hind and the Panther.
EWING, A. C. (1959). Second Thoughts in Moral Philosophy (London: Routledge).
GIBBARD, A. ( 1 9 9 0 ) . Wise Choices, Apt Feelings (Oxford: Oxford University
Press).
I82
Prescriptivism holds that moral judgements contain an element of meaning which

serves to prescribe or direct actions. The history of prescriptivism includes Socrates,
Aristotle, Hume, and Mill, and it has been influential also in recent times. Moral
judgements also contain a factual or descriptive element, which differs between
persons and cultures: but the prescriptive element remains constant. Prescriptivism
can allow for moral disagreement and explain moral weakness. It can also explain
better than other theories the rationality and objectivity of moral thinking.
1 9 9 8 b . A Utilitarian Approach to Ethics', in H. Kuhse and P. Singer, eds., A

Companion to Bioethics (Oxford: Blackwell).
1 9 9 8 c . 'Towards Objectivity in Morals', in Ouyang Kang and S. Fuller, eds.,
Contemporary British and American Philosophy and Philosophers (Beijing:
People's Press). Based on I 9 9 4 d , with additional autobiographical material. In Chinese; English version planned.
I 9 9 8 d . Interview, 'Die Vernunft kommt zuerst', in Borchers et al, eds.,
Einfoden 211m Denken (Vienna: Holder-Pichler-Tempsky).
HAGERSTROM, A. (1911). Om moraliska forestallningars sanning (On the Truth of

Moral Propositions) (Inaugural Lecture, University of Uppsala). Translated
with other writings by R. T. Sandin in Hagerstrom, Philosophy and Religion
(Allen & Unwin, 1964).
\
E. (1996). The Moral Gap (Oxford: Oxford University Press).
and JOYNT, C. (1982). Ethics and International Affairs (New York: St.
Martin's Press).
HARMAN, G. (1977). The Nature of Morality (New York: Oxford University
Press).
HARE. J.
2.
Other Writings
References in the text of the form 'Alston, W. P. (1964:100)' are to this part of
the bibliography, the last n u m b e r being t h e page, unless otherwise indicated. Abbreviated titles of works in brackets, as used in text.
AI.EXY, R. ( 1 9 7 9 ) . 'R. M. Hares Regeln des moralischen Argumentierens und
L. Nelsons Abwegungsgesetz', inP. Schrder, ed.. Vernunft, Erkenntnis,
Sittlichkeit (Hamburg: Meiner, 1 9 7 9 ) .
ALSTON, W. P. ( 1 9 6 4 ) . Philosophy of Language (Englewood Cliffs, NJ: PrenticeHall).
Prior Analytics (An. Pr.). LV Inierpretatione (De Int.), Metaphysics
(Met.). Nicomachean Ethics (Eth. Nie). References are to Bekker pages.
AUSTIN J. L. ( 1 9 6 1 ) . Philosophical Papers (Oxford: Oxford University Press).
( 1 9 6 2 ) . How lo Do Things with Words (Oxford: Oxford University Press).
ARISTOTLE.
AIIXTKR,
T. ( 1 9 8 2 ) . Kant's Moral Teleology (Macon. Ca.: Mercer University
Press).
AYER, A. J. ( 1 9 3 6 ) . Language, Truth and Logic (London: Goilancz).
BLACKBURN. S ( 1 9 8 4 ) . Spreading the Word (Oxford: Oxford University Press).
CARNAP, R ( 1 9 3 2 ) . 'berwindung der Metaphysik durch logische Analyse der
Sprache', Erkenntnis 2.
HARRIS, N. G. E. (1992). 'Kantian Duties and Immoral Agents', Kant-Studien

83.
HARRISON, J . (1985). 'Utilitarianism, Universalization, Heteronomy and
Necessity or Unkantian Ethics', in N. Potter and M. Timmons, eds.,
Morality and Universality (Dordrecht: Reidel).
HARSANYI, J . C. (1988). 'Problems with Act-Utilitarianism and with
Malevolent Preferences', in Seanor and Fotion (1988).
HEMI'EL, C. G. (1965). Aspects of Scientific Explanation (New York: Free Press).
HORGAN, T. and TIMMONS, M. (1992). 'Troubles for New Wave Moral
Semantics: The Open Question Argument Revived', Philosophical Papers
21.
HUME, D. (1739). A Treatise of Human Nature (London: Noon).
KANT, I. References are to pages of earliest editions and the Royal Prussian
Academy edition.
(1781). KritiJc der reinen Vernunft (KrV) (Riga: Hartknoch). Translation
by N. K. Smith, lmmanuel Kant's Critique of Pure Reason (London:
Macmillan, 1933).
!84
KANT, I. (1785). Grundlegung zur Metaphysik der Sitten (Gr) (Riga: Hartknoch).
Translation by H. J. Paton, The Moral Law (London: Hutchinson. 1948).
References are to this work unless indicated.
( 1788). Kritik der praktischen Vernunft (KpV) (Riga: Hartknoch). Translation by L. W. Beck. Critique of Practical Reason (Indianapolis. Ind.: BobbsMerrill. 1956).
( 1790). Kritik der Urteilskraft, pt. T (KU) (Berlin: Lagarde). Translation by
J. C. Meredith, Kant's Critique of Aesthetic Judgement (Oxford: Oxford
University Press. 1911 ).
(1797). Metaphysische Anfangsgrnde der Rechtslehre (Rl) (Knigsberg:
Nicolovius). Translation by J. Ladd. The Metaphysical Elements of Justice
(Indianapolis. Ind.: Bobbs-Merrill. 1965)(1797). Metaphysische Anfangsgrnde der Tugendlehre (Tgl) (Knigsberg:
Nicolovius). Translation by M. Gregor, 77ie Doctrine of Virtue (New York:
Harper & Row, 1964).
LEWIS. D. K. (1973). Counterfactuals (Cambridge: Cambridge University Press).
LEWIS, H. A. and WOODIIELD, A. (1985). 'Content and Community'.
Proceedings of the Aristotelian Society supp. vol. 59.
Lo, P.-C. (1981 ). A Critical Rvaluation of the Alleged "Empty Formalism" of
Kantian Ethics'. Ethics 91.
LYONS, D. (1965). Forms and Limits of Utilitarianism (Oxford: Oxford University
Press).
MACINTYRE, A. (1984), 'Relativism. Power and Philosophy', Proceedings of the
American Philosophical Association 59 (1985).
MACKIE. J. Ii. (1977). Ethics: Inventing Right and Wrong (Harmondsworth:
Penguin).
(1984). 'Rights, Utility and Universalization', with reply by Hare, in R.
Frey. ed.. Utility and Rights (Minneapolis, Minn.: University of Minnesota
Press).
MILL, ). S. (1843). A System of Logic (London: Parker).
(1861). Utilitarianism. Eraser's Magazine. Reprinted London 1863.
MILLGRAM, E. (1995). 'Inhaltsreiche ethische Begriffe und die Unterscheidung zwischen Tatsachen und Werte', in C. Fehige and G. Meggle,
eds.. Zum moralischen Denken vol. i (Frankfurt a. M.: Suhrkamp).
MOORE. G . E. (1903). Principia Ethica (Cambridge: Cambridge University
Press).
(1912). Ethics (London: Oxford University Press).
MORRIS. C. W. (1938). 'Foundations of the Theory of Signs', International
Encyclopedia of Unified Science i. 2 (Chicago, III.: Chicago University Press).
185
(1946)- Signs, Language and Behavior (New York: Prentice-Hall).

(1984). Reasons and Persons (Oxford: Oxford University Press).
PIC.DKN. C. (1991). 'Naturalism', in P. Singer, ed., A Companion to Ethics
(Oxford: Blackwell).
PIPER. A. M. S . ( 1 9 8 2 ) . A Distinction without a Difference'. Midwest Studies in
Philosophy 7.
Puis XII, POPE (1957). Allocution, Acta Apostolicae Sedis xxxxix. 1027-33.
PLATO. Republic, Statesman. References are to Stephanus pages.
PRICHARD. H. A. (1912). 'Does Moral Philosophy Rest on a Mistake?', Mind 21.
Repr. in his Moral Obligation (Oxford: Oxford University Press, 1949).
PRIOR, A. N. (1955). Formal Logic (Oxford: Oxford University Press).
RAWLS, J . (1971). A Theory of Justice (Cambridge, Mass.: Harvard University
Press).
/'
PARFIT. D.
RICHARDS. D. A. J. (1988). 'Prescriptivism, Constructivism and R i g h t s f i n

Seanor and Fotion (1988).
\
Ross. A. (1968). Directives and Norms (London: Routledge).
Ross. W. D. (1930). The Right and the Good (Oxford: Oxford University Press).
SAUSSURE, F. DE (1916). Cours de linguistique generale (Paris: Payot, 1916).
Translation by Wade Baskin, Course in General Linguistics (New York:
Philosophical Press, 1959).
SEANOR, D. and FOTION. N.. eds. (1988). Hare and Critics (Oxford: Oxford
University Press).
SEARLE. J. (1969). Speech Acts (Cambridge: Cambridge University Press).
and VANDERVEKEN, D. (1985). Foundations of Illocutionary Logic
(Cambridge: Cambridge University Press).
SIDELLE, A. (1989). Necessity. Essence and Individuation (Ithaca. NY: Cornell
University Press).
SINGER, P. (198T). The Expanding Circle (New York: Farrer, Straus, Giroux).
STEVENSON. C. L. (1942), 'Moore's Arguments against Certain Forms of
Ethical Naturalism', in P. Schilpp. ed.. The Philosophy of G. E. Moore
(Evanston. 1 1 1 . : Northwestern University Press).
(T945). Ethics and Language (New Haven, Conn.: Yale University Press).
STRAWSON, P. F. (1949). 'Truth', Analysis 9.
(1950). 'Truth'. Proceedings of the Aristotelian Society, supp. vol. 24.
(1959)- Individuals (London: Methuen).
TOUI.MIN, S. E. (1950). Au Examination of the Place of Reason in Ethics
(Cambridge: Cambridge University Press).
URMSON, J . 0 . (1968). The Emotive Theory of Ethics (London: Hutchinson.
Paperback. New York: Oxford University Press, 1969).
REFERENCES
AND
BIBLIOGRAPHY
VON WRIGHT, G . H. (1941). The Logical Problem of Induction, Acta Philosophica

Fennica 3 (Helsinki: Soc. Philosophica).
(1963). Norm and Action (London: Routledge).
INDEX
W I I J . I A M S . B. A. 0 . (1985). Ethics and the Limits of Philosophy ( L o n d o n :

Fontana/Collins).
(1988). 'The S t r u c t u r e of Hare's T h e o r y ' , in S e a n o r and Fotion (1988).
WITTGI:NSTI;IN. L. (1953)- Philosophical Investigations (Oxford: Blackwell).
W R I G H T , C. (1992). Truth and Objectivity ( C a m b r i d g e , M a s s . : H a r v a r d
University Press).
abortion 32, 68, 73, 85. 89,143

accordant see complaisance
acratic 11.18 f., 139 f.
addressee of imperation 17 f.
Alston, W. 14, 50
alternative prescriptive languages 100
amoralist see shmoralist
analytic vs. empirical 160
application of words see use
approval 84, 90, 92-4, 104-7.127
arguability requirement 44 f., 90.100,
121 f., 137
arguments, moral see reasoning
Aristotle 2.10,13, 18 f., 26, 39,46, 74 f.,
129-31.142
attitudes 90-4,104-7, " 9 . 1 2 1 , 1 2 7
Austin, J. 7 f., 13 f., 17, 48, 50, 109 f.,
112-14
autonomy of will 152,160-2
Ayer, A. J. 16,104 f.
backslider see acratic
beneficence 150-7.163
Bentham, J. 151
Blackburn. S. 99 f.
Brink. D. 77
Carnap, R. 2,4, gf., 16
Categorical Imperative 26,129-31,
135 f.. 141.147-50,153-9,162
change of mind, of moral standards
80 f.. 85 f., 123
chargee 18
choice 150, T62
Chomsky, N. 86
citizen, duty of 30
classification, classes 7 f., 46 f., 63.95,
101 f.
clistics. signs of completeness 9
cognitivism 56
commending and condemning 69 f.,

79 f- 127
complaisance 111 f.
conciliation requirement 100,122,124
consequentialism 163-5
consequentiality see supervenience
constative 8, n, 13,17, 20
contradiction 58. 69 f., 78,80,92
in will 130 f.. 154
conventions, conventionalism 15,106,
109 f., 115
convictions, moral 84, 88,93,123,149.
153,158,161
countersuggestibility 15,108, i n
covering law theory 21
culture, moral 67-72, 8r f., 86 f 89,
91 f., 102, 141
death penalty 34,141,144,147
delinquent children 34 f., 143 f.
deontic logic, deontic modalities 4 f., 12,
64 f.
deontologists 148,165
descriptive and prescriptive 10 f., 18-21
descriptive fallacy 48
descriptive meaning 45, 52-5, 58-61,
102,104,113,116,123,129,
137-41
descriptivism 48, 52,63-5,68-77,
82-4,101-4,113 f 117-19,123 f
128-30.133-7
desire 150,160-3
dichotomy 47,63, 65
disagreement, moral 85, 88, 91.95,
119-23
disapproval see approval
distinction between persons 151
Dryden, J. 89
duties to oneself, to others
eclecticism
118.126
149 f., 153-8
education, moral 86. 88 f.. 129

emotivists 16 f, 52. 8 4 , 1 0 2 - 2 8
empirical
vs. conceptual 86. 90. IS9
162 fvs. rational 159
end. humanity as 1 4 7 - 5 7 . >2 f.
ends:
immoral 151. 1
kingdom of 26. 157
shared km. 150. 1S3-N- 162
equal concern and respect 151. 154. 157
essence, individual 97
ethical theory 43-6. 54. 72. 79- 90. 100.
116-28.138
vs. moral theory 4 4 . 1 2 s
eiulaiiiionui 74 f.
euthanasia 32 f., 143
evaluative vs. descriptive meaning 36.
52-5. 5 8 - 6 1 . 6 9 - 7 1 . 75. 104
evincing 105 f.
Ewing, A. 104
examples 148
expression:
senses of 'expression' 105 f.
vs. causation 105. 107
vs. reporting 104-6
Ir
haecceity see individual essence

Hgerstrm. A. 43, 4 6 , 1 0 2 f.
happiness 74. 147. isof.. 156
Hare. J. K. 98. 143
Harsanyi. J. C. IS I
Hegel. G. VV. I'. 9
'Hell' 106
Hempel. C. 21
heteronomy see autonomy
holy will 27
Horgan.T. 6s
human flourishing 73-6
Hume. D. t8. 45. 99-101
hypothetical cases 23 f.. 132 f.
ideals 15 3
illocutionary act potential 14
illocutionary and perlocutionary acts 8.
13-17. 109. 112-1 s
illocutionary force 8. 14. 50-2. 55. 59.
110-16
impartiality 24-6, 153, 157. 163-5: see
also universalizabilily
imperatives, imperalions 10-17. too.
104. 107-16, 120. 136-9
inclination see desire
incompatibility requirement 58 f.. 119 f..
facts 3. 7. 12 I'., 25. 35 f- 38 f.. 7.V 90.

117. 123, 133-5. '44. irio-2
moral 7. 45.48
fanatic 135
feelings, moral 84 f.. 90. 92. 94- 105.
107. 119
fighting sec killing
form vs. content of moral judgements
86. 102. 13s. 150' 1^'
formal concepts, properties 6 f., 49, 57.
102.136 f.. 141. 157
formal enquiries, theory 4 4 - 6 , 148
t37
inconsistency 6, 13 f., 22, 58-60, 83, 135
indicatives 14.49 f.. 111 f.. 116, 136; see
(ikoconstative
indiscernibles, identity of 23
inference 3 7 , 4 9 , 9 6 . 1 1 5
intuitions, intuitionism 63-6, 82-95,
116. 119-22, 127-9, 138-43. 148,
161; sir also moral sense
intuitive level see levels of moral thinking
general see universal

getting to vs. telling to 15 f.. 1 0 7 - n . 114
(libbard. A. 103. 117
God 20. 27. 152
good will 157. 164 f.
Kant. I. 18 f.. 25 f.. 44. 99-102. 122 f..

129-38. 141 f.. 145-65
killing, morality of 29-36. 83. 85. 89.
154. 160-4: sir also pacifism
Habermas. J. 7
INDEX
INDEX
188
justice 144 f.. 151
language, philosophy of, vs. linguistic

philosophy 2
languages, artificial 4
learning morals vs. learning a language
68
levels of moral thinking 27. 128, 139 f.,
142. 158
f,ewis. A. 72
Lewis. 1). 23
'like' 23. 97 f.
linguistic research 66-8. 71 f., 80. 82 f..
85. 87
Linnaeus 43. 46 f.
locutionary acts 14 f.. 109
logic 36-9, 44 f.. 49, 80. 101. 1 1 2 - 3 1 ,
136-8, 159. 161
and grammar 9 f., 86 f., 101
and language 4, 13, 25
logicality requirement 58 f.. 120 f.. 129.
137
love 72. 123. 156-8
lying 148. 162-4
McDowell.]. 117
Maclntyre. A. 70,81
Mackie.). 123, 134, 151
mathematics 105. 122
maxims 102. 130 f.. 1 3 5 , 1 4 2 , 1 5 3 - 7 ,
161 f.
meaning 1-28, 3 7 - 9 , 44 f., 48 f., 63.
76 f., 82, 103-19. 122 f.
and illocutionary force 14 f.. 50-2,
109 f 1 1 2 - 1 4 . 6
descriptive and evaluative 6, 20 f..
4 5 , 4 8 - 5 5 , 58-62, 6 9 , 1 0 2 - 4 , 113,
116, 123, 129, 137-41
meat-eating 85, 88 f.. 105
mephistics 3, 10
metaphysics 2. 9 f.. 98
and logic 9
of morals 27. 149, 159 f.
Mill. J. S. 18, 76. 122. 130, 145, 148. 151
Millgram. E. 61
mimesis 9 . 1 1 0
modal terms 136
modus ponens 3 7 . 1 1 5
mood and meaning 7 - 1 0 , 14. 5 1 , 1 0 7 f..
112
1 J
189
Moore. G. E. 64. 7 1 , 77 f., 9 1 , 1 0 0 , 1 0 4 ,

119
moral sense 83, 160; see also intuitionism
moral worth 164
mores sec culture
Morris. C. W. ri2
'myself 2 5 - 7 . 132, 152
natural properties 63 f.
naturalism 55, 6 3 - 8 2 . 9 0 - 2 , 9 5 , 1 0 4 ,
1 1 6 - 2 2 . 1 2 6 f.. 1 3 6 - 9
nature 1 5 2 - 5
needs, human 7 3 - 5 . 1 2 3 . 1 3 1
negation 14. 47. 53, 94. ros f.
Nelson. L. 147
neustics. signs of subscription 9
neutrality requirement 67 f.. 7 1 - 5 . 118.
122,136
Nietzcheans 31
non-descriptivism 17, 43. 47 f.. 5 r f.,
55-7- "2 f.. 77. 9 " f - 9 5 . 102-4.
1 1 4 f.. 117. 119, 1 34
non-naturalism 63 f.. 77
non-prescriptive moral judgements see
acratic
normative vs. imperative i2f.
noumena 153
objectivism, objectivity 8 2 - 4 , 9 0 - 5 ,
1 0 0 - 3 , 1 1 6 - 1 9 . T26 f., 134,138 f.,
141
observable properties 67, 7 1 , 83, 90
ontology 2 , 4 4 f., 156
open question argument 7 1 , 77 f.
opposition, square of 12
'ought', 4 - 7 , 11 f., 113, 127 f., 136 f., 164
pacifism 29 f 68. 113. 143
Parfit. D. 99 f.
particularism 9 5 - 8 . 1 4 8
perlocutionary see illocutionary
perfectionism 1 5 6 - 8
phrastics 9 , 1 4 . 20
Pigden.C. 65
Pius XN. Pope 32
Plato 22, 33, 7 5 . 1 3 0 f.. 138, 149
INDEX
INDEX
practicality requirement 119 C. 136
pragmatic inconsistency 58
pragmatics 1 3 - 1 6 , 4 9 , 52, 112 f.
preferences, preference-ulililarianism
76--K. 1 U- 1 iS. I S 1 - ? . iS9-f>3
prescriptions X. 11. 16-20. 25-7. 45. 52,
55. 76-81. 96.101. 107. 1 1 4 . 119 f.,
128 f.. 133-42
prescriptivism, universal 102. 1 1 5 , 1 3 0 ,
1.59
Prichard. H. A. 6 4 . 1 2 9 . 148 f.. 160
promising 1 1 0 . 1 4 9 - 5 4
prudence 98. 100 f.
psychological possibility 131
publicity requirement 124
punishment, capital 34. 144. 147 f.
rationalism 1 1 4 - 1 7 . 126-45. 149

rationality sec reasoning
rational-will utilitarianism 26. 1 5 1 , 1 6 3
Rawls, J. 124 f.. 148-51
realism 136
reasoning:
moral 37-40. 57, 102. 124-8. 141,
149. 159-61
rules of 5 f. 141
reasons for normative Judgements
1 1 - 1 3 , 89. 97. 127- 149
relativism 69 f.. 77, 7 9 - 8 1 , 8 8 - 9 2 ,
1 0 1 - 3 , 1 1 6 , 1 2 3 . 1 2 9 f., T 3 4 - 8 , 1 4 1
representation, full 2 6 , 1 3 3 - 5
Richards. I). A. J. 151
rigorism. Kant's 2 0 , 1 4 1 , 1 5 5
Ross. A. 108
Ross.W.D. 88. 129. 148 f.
Uylo.C. 7
sadistic schoolmaster 108 f.
satanist 1 1 . 1 4 0
Saussure. F. de 8
Searle. |. 7. 14. 50, 10S
secondarily evaluative words see thick
and thin moral concepts
secondary qualities 7 1 - 3
semantics 6 f.. 4 9 - 5 2 . 1 1 2
sense-datum theories 93
shmoral duties, shmoralist 99, 101, 124,

134
Sidgwick. H. 91
sieve for ethical theories 118
Singer. P. 99 f.
slippery slope 3 3 1 . 1 4 5
Socrates 18 f.. 33. 39. 44. 1 3 0 , 1 3 9 , 141.
149
speech acts 7 - 1 7 . 20, 50 f., 58 f.. 107,
112, 1 1 7 - 1 9 . 121. 138
standards 70. 104. 138 f.. 141
standing orders 16
Stevenson, C. I.. 16. 20. 60. 9 1 , 1 0 3 f.,
1 0 7 . 1 1 2 . 117. 1 2 1 , 1 3 8
Strawson. P. F. 23. 57
strong and weak verbs 10, 49
subject and predicate ro
subjectivism 65 f.. 90-2. 95, 104, 116,
1 1 9 . 1 2 7 f.. 138
substantial vs. formal 6 f.. 24 f., 38,40,
44. 68, 76. 7 8 , 8 0 , 1 1 9 , 1 2 4 , 1 3 6 ,
138.160 f.
suicide 148, 152-5
supervenience 21 f., 127 f.
surface structure 136
syntactics 1 1 2 . it6
synthetic a priori 159 f.
talents, cultivation of 1 5 2 - 5
taxonomy 4 3 , 4 6 f.. 79, 9 5 . 1 0 1 , 1 1 6 f.
thick and thin moral concepts 21, 54, 61
Timmons. M. 65
Tolstoy. N. 31
Toulmin.S. 115
tropics, signs of mood 8 f.
'true', endorsing function of 57 f., 138:
see also expression vs. reporting:
lying
truth conditions 48-60. 63-8, 75 f..
82 f., 85-7. 90-2. 9 4 . 1 0 3 . 1 1 3 ,
1 1 6 - 1 8 . 128-30. 1 3 7 - 4 0
truth of moral judgements 3. 13.48, 56.
77. 92-4. 116. 138. 140
truth tables 13
'universal' vs. 'general' 96 f., 142
universal relational properties 23. 98 f.
universalism see particularism

universalizability 13, 2 2 - 7 , 98,123.
1 2 7 - 3 1 . 1 3 3 - 8 , 1 5 3 f, 162 f.
universais of language see Chomsky
llrmson,). 0. 16. 20, 109
use of sentences, words. 13, 66-9, 7 1 - 3 ,
75 f-. 79-85
utilitarianism 7 6 - 8 , 1 4 1 - 4 , 1 4 7 - 6 5
verbal shove theory 15 f., 108
Vietnam 29!.. 35
virtue 156-8
its own reward 159
volitions see wills
191
Von Wright. G. H. 123

warning 14, 51
'water' 123
Westermarck. K. 104
Williams. B. 158
Willkur 150.163
wills 2 5 - 7 , 9 9 . 1 2 9 - 3 5 , 1 4 1 . 1 5 0 - 4 . 1 5 7 .
160-5: see also autonomy
Wittgenstein, 1* 7 , 1 3 , 7 2
Woodtield, A. 72
Wright. C. 57
'wrong' 6 6 - 7 3 , 78-80. 82-7, 91, 93-7.
105 f.. 119 f., 1 3 6 - 4 1 . 1 6 4
Hare, R. M. Sorting Out Ethics.

Oxford: Clarendon Press, 1997. Pp. 191. $18.95 (cloth).
This short, clear, and very readable book is probably Hare's most important
contribution to ethicsor, more strictly, metaethicsin recent years. There are,
I think, three reasons for this.
First, the book sets out by far the clearest account I know, outside The
Language of Morals and Freedom and Reason, of Hare's universal prescriptivism,
ill main tenets, and its genesis in the philosophy of language as an account of
the meanings and logical properties of the moral terms. The book's first two
chapters"Philosophy of Language and Ethics" and "Defence of the Enterprise"both show why we need an account of the meanings and logical properties of the moral terms (not least because they make possible moral argument)
and where we are to look for such an account. Here, Hare sets up the ultimate
division of his taxonomy of metaethical theories into descriptivist and nondescriptivist varieties, according to whether they hold that the meaning of moral
statement* is exhaustively given by their syntax and truth conditions. Everything
subsequently takes place within the confines of this ultimate division. Of course,
to some critics, this very division is problematic, and it is important, I think, to
realize how crucial to Hare's case in ethics his views in the philosophy of language
about meaning are. Were these views to be false, and he tries here to show why
they are true, Hare's case in ethics would collapse.
Second, Hare then turns to a consideration of the strengths of universal
prescriptivism by way of considering it in juxtaposition to other (types of) theories usually seen in one or more regards as competitors to it. These chapterson naturalism, intuitionism, emotivism, and rationalismconsist in the
Axel Hagerstrom Lectures, given in Uppsala, Sweden, and they form the core
of the book. Their interest lies not only in the fact that they touch on issues of
great moment in contemporary metaethics (e.g., the extent to which moral
judgments are fact-stating, true, and objective, with these notions explicated in
terms of the possession of descriptive, natural properties), but they show Hare
trying to move his prescriptivism along in the light of some contemporary objections to that theory. Here figure some of the realism-antirealism debates, the
concern with forms of Blackburn's projectivism, with Gibbard's expressivism,
and with the whole question of the extent to which rational argument is possible
in ethics and, to a Harean, in what it consists. Hare takes something from each
of the positions he rejects and tries to show that what U salvaged is already
yielded by his own position. This material is very illuminating with regard to
how Hare sees himself as situated in current metaethical controversies. Among
other things, for example, Hare concedes that there is a perfectly clear sense
in which moral judgments make objective claims, something which critics, by
not fully appreciating how universalizability works in Hare, have often failed to
see that he endorses. Hare reprints, as the final chapter of the book, his wellknown paper of Kantian exegesis on whether Kant could have been a utilitarian,
and the paper is interesting in the present context because it shows how Hare
thinks his use of universalizability can confer a kind of objectivity on moral
judgments. Is that kind of objectivity the kind that moral realists want? No, but
that kind of objectivity, which turns moral expressions into purely descriptive,
lact-stating ones, runs afoul of Hare's claims about descriptive and prescriptive
meaning, which realists will reject. Yet Hare's sense of objectivity is not foreign,
I would imagine, to a kind of nondescriptivist (noncognitivist) expressivism, with
which Hare shows some sympathy.
Third, I doubt that there can be a clearer statement of where so much of
the trouble lies between Hare and his descriptivist critics than what appears in
this book. For the book's very framework forces the critic to conduct his argument in terms of rejecting Hare's account of the meaning and logical properties of the moral terms, wherein the meaning of statements containing those
terms is for Hare not exhausted by their syntax and truth conditions. Many
critics may accept this ground, play by Hare's rules, and reject the whole distinction between descriptive and prescriptive meaning. But other critics may not
want to be so constricted; for some may want to hold, for example, that moral
judgments are prescriptive to the extent that we think they are true, and this
in essence violates Hare's framework, since it makes prescriptivity depend on
truth and not be independent of il.
In sum, Hare has provided the best account of Hare on metaethics, whether
for adherents or critics of his universal prescriptivism.
R. G. FREY
Bowling Green State University
Harris, John, and Holm, Soren, eds. The Future of Human Reproduction: Ethics,
Choice, and Regulation.
Oxford: Clarendon Press, 1998. Pp. 254. $ 2 4 . 9 5 (paper).
The Future of Human Reproduction is a compilation of fourteen essays addressing
current and near-future reproductive techniques. One essay is on abortion; the
remainder deal with aspects of assisted reproduction. Contributors represent
diverse fields, including law, philosophy, social science, religion, and medical
professions. Their work is generally accessible, not requiring specialized background knowledge.
Each essay is independent, though some cluster around issues or themes.
With a few exceptions, the various contributors do not stake out and defend
opposing claims. When two or more essays address the same topic, they tend
to emphasize different approaches to it. Though more internal debate might
have been profitable, this technique works surprisingly well. The essays complement each other without much repetition.
* Chapter 1, by John Harris, is a concise survey of ethical issues relating to
a wide variety of contemporary methods of assisted/artificial reproduction. The
chapter provides a good setup for the rest of the book, as many of the topics
Harris briefly introduces here are explored at greater length in other essays.
Harris neady critiques several bad arguments. F o r example, ( 1 ) he points
out that in the case of a woman who wants to be implanted with her daughter's
fertilized egg, incredulously repeating the fact that the woman would simultaneously be the mother and grandmother of this child does not constitute an
argument against the procedure (p. 9 ) ; ( 2 ) he notes that those who criticize
BOOK
HOOK
REVIEWS
The Philosophical Review, Vol. 109, No. I ( J a n u a r y 2 0 0 0 )
REVIEWS
1: able to do at least s o m e t h i n g to resolve, by way of a r g u m e n t , m o r a l

disagreements even a m o n g people who share no "[relevant] initial sub-
SORTING OUT ETHICS. By R. M. HARE. Oxford: O x f o r d University Press,

Clarendon Press, 1 9 9 7 . Pp. x, 1 9 1 .
stantial moral opinions" ( 1 2 1 ) . Naturalists will ask, however, if t h e r e a r e

or could even be such d i s a g r e e m e n t s . T h e answer depends, in part, on
what theory of r e f e r e n c e lor moral predicates naturalists c h o o s e . Many
T h e bulk of this volume consists of a somewhat revised version of the Axel
naturalists c o n t e n d that r e f e r e n c e theories a r e available which s u p p o r t
H a g e r s t r o m L e c t u r e s given in Uppsala, Sweden in 1991 ( a t t e n d e d by the
their claim that p r o p o n e n t s of different m o r a l theories use the m o r a l pred-
present reviewer). It also contains previously published papers on the rel-
icates with the same m e a n i n g , and thus that 1 lare's arguability r e q u i r e m e n t
evance of philosophy of language to ethics and the interpretation of Kant's
can be set aside.
m o r a l philosophy. T h e latter, in particular, deserves c o m m e n t , but space
It is therefore u n f o r t u n a t e that Hare's discussion of naturalism is so in-
considerations force me to devote my attention to the H a g e r s t r o m L e c -
conclusive. Sophisticated forms of naturalism, Hare's critics a r g u e , a r e im-
tures, entitled "A T a x o n o m y of Ethical T h e o r i e s . "
m u n e to objections of the kind he has offered in the past. However, instead
By an "ethical t h e o r y " H a r e means a " t h e o r y about the m e a n i n g and

logical properties of the moral words" ( 4 5 ) . He distinguishes between two
main categories: descriptivism (the " ( m j e a n i n g s of m o r a l statements are
of taking on the critics, he refers r e a d e r s to a n o t h e r r e c e n t publication

(HWri) lot his a r g u m e n t s against these sophisticated naturalisms. It would
have been heller if that material had been included in the present book.
wholly d e t e r m i n e d by syntax and truth conditions" ( 4 2 ) ) a n d nondescrip-
T u r n i n g to nondescriptivism, we find that H a r e contrasts his own "ratio-
tivism (the negation of descriptivism). Within the descriptivist c a m p he
nalist" view with emotivism. which is consequently irrationalist. In the sum-
distinguishes between naturalism ( " f t j r u t h conditions of m o r a l statements
mary on 42 he says that the difference between the two is that his view,
a r e non-moral p r o p e r t i e s " ( 4 2 ) ) and intuitionism (truth conditions a t e sui
unlike emotivism, implies that moral statements a r e "governed by logic."
generis p r o p e r t i e s ) . Within the nondescriptivist c a m p he separates emotiv-
However, later (1201.) he allows that some emotivist theories allow "sub-
ism ( " [ m j o r a l statements a r e not g o v e r n e d by logic" ( 4 2 ) ) from rational-
sumptive a r g u m e n t s " in ethics. Now, his "logicality r e q u i r e m e n t " is that a
istic nondescriptivism (the negation of emotivism). He also sets up six re-
satisfactory ethical theory must entail that "some moral statements a r e in-
quirements which he claims any satisfactory ethical t h e o r y must meet
compatible with at least s o m e o t h e r moral statements" ( 1 2 0 ) . And any
( 1 1 8 - 2 5 ) , arguing that only rationalistic nondescriptivism meets all six.
theory that allows siibsimiptive a r g u m e n t s will satisfy this r e q u i r e m e n t . So
I will focus on what I take to be Mare's main c o n c e r n : rationality in

ethics. H a r e ' s main claim seems to be that only his own position, a m o n g
actual c o n t e n d e r s , c a n explain how moral conflicts a r e rationally resolvable. Emotivism is by definition incapable of this feat, but what about des-
maybe the really important difference between emotivism and H a r e ' s theory, in I laic's eyes, is thai his theory, unlike emotivism, implies that even
fundamental moral d i s a g r e e m e n t s c a n be resolved rationally. And crucially,
he holds, this difference tells d e a r l y in favor of his view over emotivism.
criptivism? Hare's c h a r g e against the latter theory rests on his assumption
Hare here employs a very narrow, or formal, conception of rationality. Per-
that speakers in different cultures use moral terms with different descrip-
meating his approach to ethics is the idea that moral rationality consists in
tive meaning. F o r if it should turn out, f)ace H a r e , that we all use 'ought',
following the logic of the moral words ( 1 2 9 ) . This logic in turn rests on facts
etc. with the same descriptive meaning, then all moral disputes will turn
about linguistic usage (37). Indeed, setting aside irrational nonmoral beliefs,
out to be at bottom n o n m o r a l disputes, and these can be resolved ratio-
a speaker can be accused of irrationality in morals only if he contradicts him-
nally using the p r o c e d u r e s that work for all such disputes. But this as-
self. Specifically, I would be mmn/ly irrational only if I both (i) prescribe uni-
sumption is precisely what m a n y naturalists reject: they would say that, at
versalh that everylwidy do A in (universally characterized) circumstances C
least prima facie, the g r e a t majority of moral conflicts can be resolved
and (ii) prescribe particularly that I (or someone else) do not-A in circum-
rationally, t h o u g h p e r h a p s s o m e moral conflicts are merely a p p a r e n t , due
stances C. Moreover, all interesting moral disputes can be settled, at least in
to equivocation. So if naturalism has a problem h e r e , it has less to do with
principle, Hare thinks, by showing either that one of the parties lo the dispute
its ability to explain how moral conflicts can be resolved rationally than
is irrational in this way or ( m o r e probably) that one of the parties is ignorant
with its ability to explain how t h e r e a r e conflicts in the first placethat is,
of die tactsmost interestingly by having failed to "represent fully" lo himself
its ability to explain how we all, despite a p p e a r a n c e s , use m o r a l terms with
lire situations of certain relevant individuals.
the s a m e m e a n i n g .
A related point c o n c e r n s H a r e ' s "arguability r e q u i r e m e n t " ( 1 2 2 ) . I think
we should i n t e r p r e t it (see 1 2 1 ) as follows: a satisfactory ethical t h e o r y must
1 should point out that I lare's method of resolving irroral disputes has been
rtmndlv ciitie i?ed and is indeed implausible. See especially (libbard 1988.
BOOK
HOOK
REVIEWS
IVVIEWS
The Philosophical Hnew, Vol. 109, No. 1 ( J a n u a r y 2 0 0 0 )
Now, if H a r e is right, no o t h e r kind of moral irrationality is possible, but

he fails to give us reasons for accepting such a "formalistic" conclusion.
And o n e may w o n d e r why we should a c c e p t it in the first place. F o r instance, suppose s o m e o n e , without i g n o r a n c e , a n d without accepting any
THREE METHODS OF ETHICS: A DERATE. By MARCIA W. BARON, PHILIP

PK II 11, and MICIIAF.I. S I . O I K . O x f o r d : Blackwell, 1997. Pp. vi, 2 8 5 .
relevant irrational beliefs, claims that o u r most i m p o r t a n t m o r a l obligation

is to p r o d u c e a n d preserve objects of g r e a t aesthetic value. Many of us
In 'The Methods of Ethics, Sidgwick took seriously egoism, utilitarianism, a n d
would say his view is irrational, quite apart from any possible self-contra-
comnioiiscnsc morality. Virtue ethics was treated as part of c o m m o n s e n s e
diction a la H a r e . It is, we might say, "substantially" irrational. Indeed,
morality. 'Three Methods, reflecting r e c e n t tastes, considers Kant, c o n s e q u e n -
perhaps it is possible for a moral j u d g m e n t not irrational in H a r e ' s sense
tialism, and virtue ethics. Oddly, it d o e s not reflect t h e m a j o r d e v e l o p m e n t
to be substantially i r r a t i o n a l . And now, if we g r a n t that a m o r a l view may
since Sidgwickthe revival of < o n t r a c t u a l i s m .
be substantially irrational, then even the most inveterate emotivist might
Marcia Baron defends Kitnt l i o m conscquentialists and virtue theorists.
still have a way of resolving moral disagreements rationally: all substantially
She closes by raising, without solution, familiar worries a b o u t the C a t e g o r -
irrational positions should be rejected.
ical Imperative.
H a r e would respond to this line of thought, I surmise, by claiming that
Kant values liter < .i|>ac iiy lo sei ends"humanity." ('.onscquciiiialists r e -
it falls foul of ( p e r h a p s an e x t e n d e d version of) his "neutrality require-
quire thai what is valuable !>< p r o m o t e d . Kant can reply that s o m e values
m e n t " ( 1 1 8 f . ) . T h a t is, since by whatever standards of rationality the pro-
call for h o n o r i n g r a t h e r than p r o m o t i o n . (I h o n o r r a t h e r than p r o m o t e
p o n e n t of the "aesthetic" principle accepts, this principle is not irrational,
(say) honesty when I prefer honesty to dishonesty, even when dishonesty
we do not have a non-question-begging a r g u m e n t against its rationality.
would maximize the a m o u n t of honesty in t h e world.) Humanity calls for
But this response in turn raises questions. F o r o n e thing, "substantial"
honoring; p r o m o t i o n may be possible and even desirable, but is subject to
criteria of rationality a r e employed outside ethics, but it seems that a ver-
honoring.
sion of Hare's "neutrality a r g u m e n t " would hamstring t h e m as efficiently
Virtue ethicists c h a r g e that Kant disregards feelings, privileges the motive
as it does their cousins in ethics. A n d it is u n c l e a r how H a r e could avoid
of duty, and finds i n n e r moral conflict as admirable as no conflict. In reply,
this undesirable conclusion. Secondly, will any old s t a n d a r d s of rationality
B a r o n has two strategies. By relying on texts o t h e r than the Groundwork,
d o ? H a r e is known for c o m b a t i n g the idea that we c a n dismiss certain
she presents a Kant who finds s o m e i n n e r conflicts u n a c c e p t a b l e , finds
peculiar m o r a l j u d g m e n t s by saying that they a r e n o t really moral at all.
s o m e feelings i m p o r t a n t , and requires only that thoughts of duty be suf-
But c a n we dismiss peculiar rationality j u d g m e n t s for that reason? This
ficient ( r a t h e r than that feelings be a b s e n t ) . W h e r e differences between
question leads to a m o r e general one: how should rationality j u d g m e n t s
Kant and virtue ethics r e m a i n , B a r o n argues that Kant is superior. We find
themselves be u n d e r s t o o d ? Should we be descriptivists or nondescriptivists
s o m e i n n e r conflicts a d m i r a b l e . M o r e importantly, the motive of duty is
about rationality j u d g m e n t s ? Until these and o t h e r questions have been
necessary, for this provides a n e e d e d c h e c k on feelings.
answered H a r e ' s formalistic a p p r o a c h remains in doubt.
Baron is impressive in presenting a Kant a m e n a b l e or superior to virtue

DAVID ALM
University
of Arizona
ethics. B u t
( 1 ) S o m e virtue ethicists c a n avoid the a r g u m e n t for duty. S u p p o s e that
all duties are derived from universal benevolence. T h e motive of universal
b e n e v o l e n c e then needs no c h e c k from a separate thought of duty; it is a
reliable motive for p r o d u c i n g dutiful actions. Baron makes a c h e c k s e e m
References
necessary by taking the motive to be merely local b e n e v o l e n c e a n d by

assuming that not all duties can he derived f r o m universal b e n e v o l e n c e .
Gibbard, A. 1 9 8 8 . " H a r e ' s Analysis of 'Ought' a n d its Implications." In
( 2 ) B a r o n does n o t show the rationality of h o n o r i n g . It does n o t suffice
Hare and Critics, ed. D. S e a n o r and N. Fotion, 5 7 - 7 2 . O x f o r d : O x f o r d
to offer cases in which the value c a n n o t be p r o m o t e d ; this fails to defend
University Press.
H a r e , R. M. 1 9 9 5 . "A New Kind of Ethical Naturalism?" In Moral Concepts:
the side-constraints on p r o m o t i o n that Kant a n d B a r o n want. In t h e case

of p r o m o t a b l e humanity, B a r o n notes that meddling with o t h e r s could
Midwest Studies in Philosophy 2 0 , ed. P. F r e n c h . T. U e h l i n g and H. Wett-
increase their humanity, but seems merely to stipulate that properly valuing
stein, 3 4 0 - 5 6 . N o t r e D a m e : N o t r e D a m e University Press.
humanity precludes meddling.

Hare, R M - Sorting Out Ethics

Diunggah oleh

Informasi Dokumen

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

Hare, R M - Sorting Out Ethics

Diunggah oleh

Hak Cipta:

Format Tersedia

SORTING

Oxford University Press. Great Clarendon Street. Oxford 0x2 6DP

All rights reserved. No part of this publication may be reproduced,

Formerly, w h e n writing a book, I used to hold t h e whole of it in my

British Library Cataloguing in Publication Data

Oxford a n d Florida, omitted from t h e five l e c t u r e s given in Uppsala

latest considered t h o u g h t s , could n o t be published while I w a s still

with different p a g i n a t i o n s ; a n d t h e s a m e applies t o t h e n u m e r o u s

I offer this t a x o n o m y of ethical theories to all those w h o are lost in

PART II. THE AXEL H A G E R S T R O M L E C T U R E S :

8. Could K a n t have been a Utilitarian?

References and Bibliography

E T H I C S , or m o r a l philosophy, is the point at w h i c h philosophers

c o m e closest to practical issues in morals and politics. It t h u s provides

logic of inference;} in w h i c h they appear. So a study of t h e m e a n i n g s

T h e r e are a g r e a t m a n y pitfalls to be avoided in c a r r y i n g o u t this

have in r e c e n t times usurped the n a m e ; t h e y would be better called

p r o g r a m m e ; but I shall a r g u e t h a t it is in principle a feasible one. So

'mephistics', b e c a u s e they are attempts, like t h a t of Mephistopheles

let us first consider some possible objections to it. I shall be in danger

in Faust, to get philosophers to sell their souls for fantasies.

of being misunderstood if I do not m a k e clear at the start t h a t philo-

I wish to consider two possible objections to the p r o g r a m m e I pro-

sophy of language is not the s a m e as linguistic philosophy. T h e former

jected at the beginning. The first says. T a c t s about particular lan-

is a b r a n c h of philosophy, c o - o r d i n a t e with philosophy of s c i e n c e ,

g u a g e s , i n c l u d i n g facts a b o u t h o w people u s e words i n p a r t i c u l a r

philosophy of law, philosophy of history, e t c . To say t h a t a philo-

cultures, a r e c o n t i n g e n t facts. T h e y therefore c a n n o t be used to e s -

sopher is doing philosophy of l a n g u a g e does not presuppose that he is

tablish n e c e s s a r y t r u t h s s u c h as we are looking for in e t h i c s . We do

doing it by any particular method, or in a c c o r d a n c e with the tenets of

not w a n t to be told h o w p a r t i c u l a r people or c u l t u r e s use the m o r a l

any particular school. Philosophers of language can be realists or the

words; we w a n t to be s h o w n what is right or wrong, and to be shown

opposite, intuitionists or t h e opposite, and so on. If anybody were to

by secure r e a s o n i n g t h a t this is necessarily the case.'

say, like Plato on some interpretations, that words have m e a n i n g be-

T h e second objection is related to t h e first: it says, 'Moral r e a s o n -

c a u s e they stand for e t e r n a l l y e x i s t i n g n o n - s e n s i b l e entities up in

ing h a s to be c o n c e r n e d with m o r a l facts, w h i c h a r e facts n o t a b o u t

Heaven, he would still be doing philosophy of l a n g u a g e , b u t would

words but about t h e worldfacts about the existence of m o r a l values

obviously not be a linguistic philosopher. But see H 1 9 8 2 0 , esp. c h . 4,

i n t h e world. T h e study o f words c o u l d n e v e r yield s u c h facts.'

for a more 'linguistic' interpretation of Plato.

Answers to b o t h these objections c a n be given. For t h e first, consider

A linguistic philosopher is s o m e o n e w h o believes in a p a r t i c u l a r

t h e position of o r d i n a r y logic. It would be a m i s t a k e to suppose t h a t

way of doing philosophy (any kind of philosophy, not just philosophy

logic discovers only c o n t i n g e n t truths about l a n g u a g e ; b u t it is also a

of language), n a m e l y t h a t w h i c h consists in studying t h e m e a n i n g s

mistake to think t h a t logic is independent of t h e study of language. It

of words t h a t present philosophical problems, and so unravelling t h e

is a necessary truth that, in one c o m m o n m e a n i n g of 'all' and the

problems. He will advocate, like C a r n a p ( 1 9 3 2 ) , an ' b e r w i n d u n g der

Metaphysik durch logische A n a l y s e der S p r a c h e ' . To m a k e my o w n

and this is a book on the top shelf, then this is by Wittgenstein. B u t in

position clear, I am a linguistic philosopher of a sort, but not of such

order to establish that this is a necessary truth, we have to be assured

an extreme sort as Carnap. I believe t h a t metaphysics does n o t have

to be overcome, nor even superseded; as inherited from Aristotle, it is

it so. Logic is, at least in part, t h e study of t h e words w h i c h people use

a respectable and c e n t r a l b r a n c h of philosophy, and o n l y c e r t a i n