Virtues and Vices and Other Essays in

Moral Philosophy
Philippa Foot

Great Clarendon Street, Oxford

Oxford University Press is a department of the University of Oxford
It furthers the University's objective of excellence in research, scholarship, and
education by publishing worldwide in Oxford New York Auckland Bangkok Buenos
Aires Cape Town Chennai Dar es Salaam Delhi Hong Kong Istanbul Karachi Kolkata
Kuala Lumpur Madrid Melbourne Mexico City Mumbai Nairobi São Paulo Shanghai
Taipei Tokyo Toronto

Oxford is a registered trade mark of Oxford University Press in the UK and in certain
other countries

Published in the United States by Oxford University Press Inc., New York
© in this edition, Philippa Foot 2002
The moral rights of the authors have been asserted
Database right Oxford University Press (maker)
First published 1978 by Blackwell Publishers and University of California Press

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, or as expressly permitted by law, or under terms
agreed with the appropriate reprographics rights organization. Enquiries concerning
reproduction outside the scope of the above should be sent to the Rights Department,
Oxford University Press, at the address above

You must not circulate this book in any other binding or cover and you must impose
this same condition on any acquirer

British Library Cataloguing in Publication Data

Data available
 Philippa Foot

Library of Congress Cataloging in Publication Data

Foot, Philippa.
Virtues and vices and other essays in moral philosophy/Philippa Foot.
p. cm.
Originally published: Berkeley: University of California Press, 1978. With new pref.
and index.

ISBN 0199252858
To The Memory Of Iris Murdoch
Contents
Preface to 2002 Edition.............................................................................................................5
Preface .......................................................................................................................................6
Introduction (1977) ...................................................................................................................6
Acknowledgements .................................................................................................................9
I Virtues and Vices ................................................................................................................. 10
I ............................................................................................................................................ 10
II ........................................................................................................................................... 15
III .......................................................................................................................................... 20
II The Problem of Abortion and the Doctrine of the Double Effect ................................... 24
III Euthanasia .......................................................................................................................... 34
IV Free Will as Involving Determinism................................................................................ 55
V Hume on Moral Judgement ............................................................................................... 64
VI Nietzsche: The Revaluation of Values ............................................................................. 69
VII Moral Arguments ............................................................................................................ 80
VIII Moral Beliefs ................................................................................................................... 90
I ............................................................................................................................................ 90
II ........................................................................................................................................... 98
IX Goodness and Choice ...................................................................................................... 106
X Reasons for Action and Desires ....................................................................................... 118
Postscript ........................................................................................................................... 124
XI Morality as a System of Hypothetical Imperatives ...................................................... 125
XII A Reply to Professor Frankena ..................................................................................... 137
XIII Are Moral Considerations Overriding? ...................................................................... 142
XIV Approval and Disapproval .......................................................................................... 148
 Preface to 2002 Edition

Preface to 2002 Edition

These essays, originally published between 1958 and 1977, were written at a time when
emotivist and prescriptivist views reigned in analytic moral philosophy, and when
individual virtues and vices were still given relatively little attention. Things have
changed since then, and I am conscious that some of the papers have dated at least in
their idiom, given that at the time they appeared ‘man’ was taken to mean ‘human
being’ rather than ‘male person’, so that the intrusively corrective ‘she’ or the cumbrous
‘he or she’ were considered unnecessary. But I believe that the path that I was taking in
most of the papers towards an objectivist theory of ethics is still serviceable today, while
other essays such as the one on euthanasia are fairly constantly in demand.
In general I still agree with what I wrote in younger days. But I can claim little
consistency over the years in respect of one crucial problem in moral philosophy: that of
the relation between practical rationality and moral virtue. The problem is to see how
for every person and in every case it can be rational to follow moral edicts—in
particular the demands of justice and charity—when these seem to clash with self-
interest or desire. We want to be able to say that to act as justice or charity demands is
to act rationally in every case, even in the tight corner. But how is this possible?
Ever preoccupied by this problem, I have tried now this now that way out of the
difficulty. In ‘Moral Beliefs’ I argued that only the coincidence of virtue and self-interest
could give the desired rationality to each and every person whatever his or her
contingent desires. Later, when writing ‘Morality as a System of Hypothetical
Imperatives’, the problem of the ‘tight corner’ made me dissatisfied with this supposed
solution, after which, in a despairing mood, I was even ready to deny that for everyone,
always, it would be rational to act morally. This was often identified as ‘Foot's position’,
long after I myself had abandoned it and was working my way around, very slowly, to
the quite different position that I first took up in the eighties and have held ever since.
In a 1995 lecture called ‘Does Moral Subjectivism Rest on a Mistake?’, which is
substantially Chapter 1 of my 2001 book Natural Goodness, as well as in Chapter 4 of that
book, I argued that the whole direction of my earlier efforts (and of those of many
others) had been mistaken. We had tried to reconcile the demands of virtue with an
independently determined canon of practical rationality such as conduciveness to the
fulfilment of desire or to rational self-interest. And that was a mistake. A tip-off from my
UCLA colleague the late Warren Quinn had allowed me to see that the very concept of
practical rationality might not be independent of that of practical goodness as describable
in a really general account of the virtues: indeed that it might have to fit in with them
rather than them with it. 1 Therefore, the appearance of a clash between the

1See ‘Putting Rationality in its Place’, in Warren Quinn, Morality and Action, (Cambridge University Press,
1993).
 Philippa Foot

requirements of moral virtue and practical rationality should put in question not charity
or justice but rather the theory of reasons for action that gave that result. I am only sad
that Quinn did not have time to develop this idea himself.
P.R.F.
December 2001

Preface
It is hard to thank all those who have helped me with my work over the years in which
these essays were written. I owe most, perhaps, to Donald MacKinnon who first taught
me moral philosophy, but also much to Somerville College where I was supported most
generously through many unproductive years. I was most fortunate to have Elizabeth
Anscombe as my colleague at Somerville, and I know that some of the ideas for these
essays came out of our lunchtime discussions in the Senior Common Room.
More recently I have been helped, in different but always important ways, by Rogers
Albritton, John Giuliano, Rosalind Hursthouse, Jerrold Katz, John McDowell, Warren
Quinn, David Sachs, Barry Stroud, and Michael Sukale, while special thanks must go to
Derek Parfit who has worked on drafts of my papers with a speed and skill that amaze
me, and a kindness to warm the heart. I am also most grateful to the Philosophy
Department at the University of California at Los Angeles, and to the efficient and
generous office staff, with Louise Nunnink as its guiding spirit.
P.R.F.
November 1977

Introduction (1977)
Two of the papers in this volume, ‘Virtues and Vices’ and ‘Are Moral Considerations
Overriding?’, are published here for the first time. The others were written between
1957 and 1976, and are mostly reprinted as they originally appeared. I have, however,
added footnotes here and there; and as well as a short Postscript to ‘Reasons for Action
and Desires’ there is a long footnote to ‘Morality as a System of Hypothetical
Imperatives’ containing material from an earlier version of that essay. Furthermore I
have made some changes in the paper called ‘Approval and Disapproval’ which
originally appeared as part of the Festschrift for Herbert Hart. I included this last essay
after some hesitation, and I hope it may be read not as a finished product but rather as
an indication of the direction in which I should like to go after the publication of the
present volume. I am sad that even after much work I did not have something more
complete to offer for the Festschrift.
The volume falls into two parts. Papers I–VI are on various topics and authors, and not
presented here in the order in which they were written. Papers VII–XIV do, however,
 Introduction (1977)

represent the development of a certain line of thought on the theory of moral

judgement, and the General Editor of this series has suggested that I should say
something here about the way in which my views changed over the period of time in
which these papers were written. Two themes run through many of the essays:
opposition to emotivism and prescriptivism, and the thought that a sound moral
philosophy should start from a theory of the virtues and vices. It was reading Aquinas
on the individual virtues that first made me suspicious of contemporary theories about
the relation between ‘fact’ and ‘value’; and I have thought about virtues and vices on
and off for the last twenty years. Meanwhile I was moving, a step at a time, away from
the theories of philosophers such as C. L. Stevenson and R. M. Hare.
When I wrote ‘Moral Arguments’ I was still very much under the influence of these
theories, and supposed, even if only for the sake of the argument, that my opponents
were right in thinking that evaluations involved emotions, attitudes, the acceptance of
first person imperatives, the recognition of reason for action or something of the sort.
What I argued was that no such ‘evaluative element’ could explain the whole meaning of
moral terms such as ‘right’ and ‘wrong’, and that there were in fact logical limits to the
considerations that could be used to back up moral judgements. In my opponents'
terminology, which I never found perspicuous, I was saying not just that moral terms
must have some ‘descriptive’ meaning but that they had a more or less determinate
‘descriptive’ meaning. I was not prepared to say just what this meaning should be, but I
was sure that it would not do to suppose that, for instance, someone might have a
morality in whose ultimate principle was that it was wrong to run round trees right
handed or to look at hedgehogs in the light of the moon. And if there were limits to the
considerations that could be used to support a moral judgement, it followed that my
emotivist and prescriptivist opponents were wrong in their account of moral argument.
For they thought they could see, without enquiring into the details, that moral
arguments ‘might always break down’, because there were invulnerable positions to
which each side might retreat in case of need. I argued that no such general account of
moral disputes could be given, and that for all we knew there might be a quite
determinate starting point provided by the concept of morality. I supposed that moral
considerations were necessarily related in some way to human good and harm, though
I did not at any time accept a utilitarian theory of ethics, and have argued against it in
some of the essays on substantial moral problems reprinted here.
In ‘Moral Beliefs’ I took up a position more radically opposed to the emotivists and
prescriptivists. There I queried the existence of a separate ‘evaluative element’ in the
meaning of ethical terms, logically unrelated to the descriptive element in moral
judgement; and I argued that the facts are such as to provide all the connexion required
between moral judgement and the will. In the actual conditions of human life—the
conditions shared by all men—anyone has reason to cultivate the virtues. I thought it
was necessary for me to show this, because at that stage it had simply not occurred to
 Philippa Foot

me to question the often repeated dictum that moral judgements necessarily give
reasons for acting to each and every man. Nor was it too hard to argue that any man has
reason to choose the virtues of courage, wisdom and temperance; since courage,
wisdom and temperance are necessary for a man's well being whatever his particular
interests and desires. And as self-interest seemed to me to be though not indeed the
only reason for acting the only one that could be appealed to universally, I tried to show
that there was a necessary connexion between virtues and self-interest. Where I came to
grief was, predictably, over justice. It seems obvious that a man who acts justly must on
occasion be ready to go against his own interests; but so determined was I to think that
every man must have reason to act morally that I was prepared to doubt that justice is a
virtue rather than give up that idea.
By the time I came to write ‘Goodness and Choice’ a few years later it had come to seem
important that in non-moral contexts a predication of goodness was not necessarily
related to the choices of the individual speaker. It was possible for someone who did
not share standard desires to have no reason to choose the good cars or the good knives,
or to choose to be a good rider, a good patient and all the rest. Might it not be the same
with good actions and good character traits? In ‘Morality as a System of Hypothetical
Imperatives’ I argued this explicitly, finally getting away from the presuppositions of
emotivism and prescriptivism, and also from the widely accepted neo-Kantian
supposition that immoral action must somehow be irrational. This argument seems to
me to be undoubtedly correct; but whether I was wise to adopt the Kantian terminology
of categorical and hypothetical imperatives is another matter. If a categorical imperative
is understood as one that any man must automatically have reason to obey, then moral
judgements are in my opinion not categorical imperatives. But I do not suppose that the
means—ends model of Kant's hypothetical imperatives fits all cases of reasons, and I
now wish that I had attacked Kant without taking over his terminology.
To sum up. The essays discussed in this introductory note are ones in which I was
making a painfully slow journey, upward or downward according to your opinion,
away from theories that located the special character of evaluations in each speaker's
attitudes, feelings, or recognition of reason for acting. So far as I know the earlier essays
are incomplete rather than mistaken, except for the arguments about virtue and self-
interest in the second part of ‘Moral Beliefs’. It is to this that the footnote on pp. 130–131
of this volume relates, and I explain there my change of mind about the necessary
relation between virtue and happiness: it is not that I have given up thinking that there
is a close connexion between the two, but that I no longer have to say that justice and
advantage coincide, because I no longer think that each man, whatever his desires and
whatever his situation, necessarily has reason to be just.
I should add that while some people take my position to be inimical to morality I
myself do not. Considerations of justice, charity and the like have a strange and
powerful appeal to the human heart, and we do not need bad arguments to show that
 Acknowledgements

no one could be indifferent to morality without error. If I am right that cannot be

shown, but morality may be stronger rather than weaker if we look this fact in the face.

Acknowledgements
The author and publisher wish the thank the original publishers for permission to
reprint the following essays: ‘The Problem of Abortion and the Doctrine of the Double
Effect’, from the Oxford Review, No. 5 (1967); ‘Euthanasia’, from Philosophy and Public
Affairs, Vol. 6, No. 2 (Winter 1977); ‘Free Will as Involving Determinism’, from The
Philosophical Review, LXVI, No. 4 (October 1957); ‘Hume on Moral Judgement’, from
David Pears (ed.), David Hume (London, Macmillan, 1963); ‘Nietzsche: The Revaluation
of Values’, from R. C. Solomon (ed.), Nietzsche, A Collection of Critical Essays (New York,
Doubleday, 1973); ‘Moral Arguments’, from Mind, 67 (1958); ‘Moral Beliefs’, from
Proceedings of the Aristotelian Society, Vol. 59 (1958–1959); ‘Goodness and Choice’, from
The Aristotelian Society Supplementary Volume 1961; ‘Reasons for Actions and Desires’,
from The Aristotelian Society Supplementary Volume 1972; ‘Morality as a System of
Hypothetical Imperatives’, from The Philosophical Review, Vol. 81, No. 3 (July 1972); ‘A
Reply to Professor Frankena’, from Philosophy, 50 (1975); and ‘Approval and
Disapproval’, from P. M. S. Hacker (ed.), Law, Morality, and Society: Essays in Honour of
H. L. A. Hart, (Oxford: Clarendon Press, 1977).
 Philippa Foot

I Virtues and Vices

Philippa Foot
Abstract: Virtues and vices, often neglected in analytic philosophy, are discussed here, drawing on the
works of Aristotle, St. Thomas Aquinas, and Immanuel Kant. Virtues are first described as beneficial
characteristics that consist in goodness of the will, but a discussion of the virtue of wisdom shows a virtue
may also require knowledge. Virtues are seen as correctives of harmful human passions and other
general temptations, but this does not mean that a particular act of virtue must always be difficult to
perform. Lastly, it is argued that a virtue does not operate as a virtue when turned to a bad end, as for
instance when someone needs daring to commit a murder.

I
For many years the subject of the virtues and vices was strangely neglected by moralists
working within the school of analytic philosophy. The tacitly accepted opinion was that
a study of the topic would form no part of the fundamental work of ethics; and since
this opinion was apparently shared by philosophers such as Kant, Mill, G. E. Moore, W.
D. Ross, and H. A. Prichard, from whom contemporary moral philosophy has mostly
been derived, perhaps the neglect was not so surprising after all. However that may be,
things have recently been changing. During the past ten or fifteen years several
philosophers have turned their attention to the subject; notably G. H. von Wright and
Peter Geach. Von Wright devoted a not at all perfunctory chapter to the virtues in his
book The Varieties of Goodness 1 published in 1963, and Peter Geach's book called The
Virtues 2 appeared in 1977. Meanwhile a number of interesting articles on the topic have
come out in the journals.
In spite of this recent work, it is best when considering the virtues and vices to go back
to Aristotle and Aquinas. I myself have found Plato less helpful, because the individual
virtues and vices are not so clearly or consistently distinguished in his work. It is
certain, in any case, that the most systematic account is found in Aristotle, and in the
blending of Aristotelian and Christian philosophy found in St. Thomas. By and large
Aquinas followed Aristotle—sometimes even heroically—where Aristotle gave an
opinion, and where St. Thomas is on his own, as in developing the doctrine of the
theological virtues of faith, hope and charity, and in his theocentric doctrine of
happiness, he still uses an Aristotelian framework where he can: as for instance in

I am indebted to friends in many universities for their help in forming my views on this subject; and
particularly to John Giuliano of UCLA, whose unpublished work on the unity of the virtues I have
consulted with profit, and to Rosalind Hursthouse who commented on a draft of the middle period.
1 G. H. von Wright The Varieties of Goodness (London, 1963)

2 Peter Geach, The Virtues (Cambridge, 1977)

 I Virtues and Vices

speaking of happiness as man's last end. However, there are different emphases and
new elements in Aquinas's ethics: often he works things out in far more detail than
Aristotle did, and it is possible to learn a great deal from Aquinas that one could not
have got from Aristotle. It is my opinion that the Summa Theologica is one of the best
sources we have for moral philosophy, and moreover that St. Thomas's ethical writings
are as useful to the atheist as to the Catholic or other Christian believer.
There is, however, one minor obstacle to be overcome when one goes back to Aristotle
and Aquinas for help in constructing a theory of virtues, namely a lack of coincidence
between their terminology and our own. For when we talk about the virtues we are not
taking as our subject everything to which Aristotle gave the name areté or Aquinas
virtus, and consequently not everything called a virtue in translations of these authors.
‘The virtues’ to us are the moral virtues whereas areté and virtus refer also to arts, and
even to excellences of the speculative intellect whose domain is theory rather than
practice. And to make things more confusing we find some dispositions called moral
virtues in translations from the Greek and Latin, although the class of virtues that
Aristotle calls aretaiethikai and Aquinas virtutes morales does not exactly correspond with
our class of moral virtues. For us there are four cardinal moral virtues: courage,
temperance, wisdom and justice. But Aristotle and Aquinas call only three of these
virtues moral virtues; practical wisdom (Aristotle's phronèsis and Aquinas's prudentia)
they class with the intellectual virtues, though they point out the close connexions
between practical wisdom and what they call moral virtues; and sometimes they even
use areté and virtus very much as we use ‘virtue’.
I will come back to Aristotle and Aquinas, and shall indeed refer to them frequently in
this paper. But I want to start by making some remarks, admittedly fragmentary, about
the concept of a moral virtue as we understand the idea.
First of all it seems clear that virtues are, in some general way, beneficial. Human beings
do not get on well without them. Nobody can get on well if he lacks courage, and does
not have some measure of temperance and wisdom, while communities where justice
and charity are lacking are apt to be wretched places to live, as Russia was under the
Stalinist terror, or Sicily under the Mafia. But now we must ask to whom the benefit
goes, whether to the man who has the virtue or rather to those who have to do with
him? In the case of some of the virtues the answer seems clear. Courage, temperance
and wisdom benefit both the man who has these dispositions and other people as well;
and moral failings such as pride, vanity, worldliness, and avarice harm both their
possessor and others, though chiefly perhaps the former. But what about the virtues of
charity and justice? These are directly concerned with the welfare of others, and with
what is owed to them; and since each may require sacrifice of interest on the part of the
virtuous man both may seem to be deleterious to their possessor and beneficial to
others. Whether in fact it is so has, of course, been a matter of controversy since Plato's
time or earlier. It is a reasonable opinion that on the whole a man is better off for being
 Philippa Foot

charitable and just, but this is not to say that circumstances may not arise in which he
will have to sacrifice everything for charity or justice.
Nor is this the only problem about the relation between virtue and human good. For
one very difficult question concerns the relation between justice and the common good.
Justice, in the wide sense in which it is understood in discussions of the cardinal virtues,
and in this paper, has to do with that to which someone has a right—that which he is
owed in respect of non-interference and positive service—and rights may stand in the
way of the pursuit of the common good. Or so at least it seems to those who reject
utilitarian doctrines. This dispute cannot be settled here, but I shall treat justice as a
virtue independent of charity, and standing as a possible limit on the scope of that
virtue.
Let us say then, leaving unsolved problems behind us, that virtues are in general
beneficial characteristics, and indeed ones that a human being needs to have, for his
own sake and that of his fellows. This will not, however, take us far towards a
definition of a virtue, since there are many other qualities of a man that may be
similarly beneficial, as for instance bodily characteristics such as health and physical
strength, and mental powers such as those of memory and concentration.
What is it, we must ask, that differentiates virtues from such things?
As a first approximation to an answer we might say that while health and strength are
excellences of the body, and memory and concentration of the mind, it is the will that is
good in a man of virtue. But this suggestion is worth only as much as the explanation
that follows it. What might we mean by saying that virtue belongs to the will?
In the first place we observe that it is primarily by his intentions that a man's moral
dispositions are judged. If he does something unintentionally this is usually irrelevant
to our estimate of his virtue. But of course this thesis must be qualified, because failures
in performance rather than intention may show a lack of virtue. This will be so when,
for instance, one man brings harm to another without realising he is doing it, but where
his ignorance is itself culpable. Sometimes in such cases there will be a previous act or
omission to which we can point as the source of the ignorance. Charity requires that we
take care to find out how to render assistance where we are likely to be called on to do
so, and thus, for example, it is contrary to charity to fail to find out about elementary
first aid. But in an interesting class of cases in which it seems again to be performance
rather than intention that counts in judging a man's virtue there is no possibility of
shifting the judgement to previous intentions. For sometimes one man succeeds where
another fails not because there is some specific difference in their previous conduct but
rather because his heart lies in a different place; and the disposition of the heart is part
of virtue.
Thus it seems right to attribute a kind of moral failing to some deeply discouraging and
debilitating people who say, without lying, that they mean to be helpful; and on the
other side to see virtue par excellence in one who is prompt and resourceful in doing
 I Virtues and Vices

good. In his novel A Single Pebble John Hersey describes such a man, speaking of a
rescue in a swift flowing river:
It was the head tracker's marvellous swift response that captured my admiration at first,
his split second solicitousness when he heard a cry of pain, his finding in mid-air, as it
were, the only way to save the injured boy. But there was more to it than that. His
action, which could not have been mulled over in his mind, showed a deep, instinctive
love of life, a compassion, an optimism, which made me feel very good . . .
What this suggests is that a man's virtue may be judged by his innermost desires as well
as by his intentions; and this fits with our idea that a virtue such as generosity lies as
much in someone's attitudes as in his actions. Pleasure in the good fortune of others is,
one thinks, the sign of a generous spirit; and small reactions of pleasure and displeasure
often the surest signs of a man's moral disposition.
None of this shows that it is wrong to think of virtues as belonging to the will; what it
does show is that ‘will’ must here be understood in its widest sense, to cover what is
wished for as well as what is sought.
A different set of considerations will, however, force us to give up any simple statement
about the relation between virtue and will, and these considerations have to do with the
virtue of wisdom. Practical wisdom, we said, was counted by Aristotle among the
intellectual virtues, and while our wisdom is not quite the same as phronesis or prudentia
it too might seem to belong to the intellect rather than the will. Is not wisdom a matter
of knowledge, and how can knowledge be a matter of intention or desire? The answer is
that it isn't, so that there is good reason for thinking of wisdom as an intellectual virtue.
But on the other hand wisdom has special connexions with the will, meeting it at more
than one point.
In order to get this rather complex picture in focus we must pause for a little and ask
what it is that we ourselves understand by wisdom: what the wise man knows and
what he does. Wisdom, as I see it, has two parts. In the first place the wise man knows
the means to certain good ends; and secondly he knows how much particular ends are
worth. Wisdom in its first part is relatively easy to understand. It seems that there are
some ends belonging to human life in general rather than to particular skills such as
medicine or boatbuilding, ends having to do with such matters as friendship, marriage,
the bringing up of children, or the choice of ways of life; and it seems that knowledge of
how to act well in these matters belongs to some people but not to others. We call those
who have this knowledge wise, while those who do not have it are seen as lacking
wisdom. So, as both Aristotle and Aquinas insisted, wisdom is to be contrasted with
cleverness because cleverness is the ability to take the right steps to any end, whereas
wisdom is related only to good ends, and to human life in general rather than to the
ends of particular arts.
Moreover, we should add, there belongs to wisdom only that part of knowledge which
is within the reach of any ordinary adult human being: knowledge that can be acquired
 Philippa Foot

only by someone who is clever or who has access to special training is not counted as
part of wisdom, and would not be so counted even if it could serve the ends that
wisdom serves. It is therefore quite wrong to suggest that wisdom cannot be a moral
virtue because virtue must be within the reach of anyone who really wants it and some
people are too stupid to be anything but ignorant even about the most fundamental
matters of human life. Some people are wise without being at all clever or well
informed: they make good decisions and they know, as we say, ‘what's what’.
In short wisdom, in what we called its first part, is connected with the will in the
following ways. To begin with it presupposes good ends: the man who is wise does not
merely know how to do good things such as looking after his children well, or
strengthening someone in trouble, but must also want to do them. And then wisdom, in
so far as it consists of knowledge which anyone can gain in the course of an ordinary
life, is available to anyone who really wants it. As Aquinas put it, it belongs ‘to a power
under the direction of the will’. 3
The second part of wisdom, which has to do with values, is much harder to describe,
because here we meet ideas which are curiously elusive, such as the thought that some
pursuits are more worthwhile than others, and some matters trivial and some important
in human life. Since it makes good sense to say that most men waste a lot of their lives
in ardent pursuit of what is trivial and unimportant it is not possible to explain the
important and the trivial in terms of the amount of attention given to different subjects
by the average man. But I have never seen, or been able to think out, a true account of
this matter, and I believe that a complete account of wisdom, and of certain other
virtues and vices must wait until this gap can be filled. What we can see is that one of
the things a wise man knows and a foolish man does not is that such things as social
position, and wealth, and the good opinion of the world, are too dearly bought at the
cost of health or friendship or family ties. So we may say that a man who lacks wisdom
‘has false values’, and that vices such as vanity and worldliness and avarice are contrary
to wisdom in a special way. There is always an element of false judgement about these
vices, since the man who is vain for instance sees admiration as more important than it
is, while the worldly man is apt to see the good life as one of wealth and power.
Adapting Aristotle's distinction between the weak-willed man (the akrates) who follows
pleasure though he knows, in some sense, that he should not, and the licentious man
(the akolastos) who sees the life of pleasure as the good life, 4 we may say that moral
failings such as these are never purely ‘akratic’. It is true that a man may criticise
himself for his worldliness or vanity or love of money, but then it is his values that are
the subject of his criticism.

3 Aquinas, Summa Theologica, 1a2ae Q.56 a.3.

4 Aristotle, Nicomachean Ethics, especially bk. VII.
 I Virtues and Vices

Wisdom in this second part is, therefore, partly to be described in terms of

apprehension, and even judgement, but since it has to do with a man's attachments it
also characterises his will.
The idea that virtues belong to the will, and that this helps to distinguish them from
such things as bodily strength or intellectual ability has, then, survived the
consideration of the virtue of wisdom, albeit in a fairly complex and slightly attenuated
form. And we shall find this idea useful again if we turn to another important
distinction that must be made, namely that between virtues and other practical
excellences such as arts and skills.
Aristotle has sometimes been accused, for instance by von Wright, of failing to see how
different virtues are from arts or skills; 5 but in fact one finds, among the many things
that Aristotle and Aquinas say about this difference, the observation that seems to go
the heart of the matter. In the matter of arts and skills, they say, voluntary error is
preferable to involuntary error, while in the matter of virtues (what we call virtues) it is
the reverse. 6 The last part of the thesis is actually rather hard to interpret, because it is
not clear what is meant by the idea of involuntary viciousness. But we can leave this
aside and still have all we need in order to distinguish arts or skills from virtues. If we
think, for instance, of someone who deliberately makes a spelling mistake (perhaps
when writing on the blackboard in order to explain this particular point) we see that
this does not in any way count against his skill as a speller: ‘I did it deliberately’ rebuts
an accusation of this kind. And what we can say without running into any difficulties is
that there is no comparable rebuttal in the case of an accusation relating to lack of
virtue. If a man acts unjustly or uncharitably, or in a cowardly or intemperate manner, ‘I
did it deliberately’ cannot on any interpretation lead to exculpation. So, we may say, a
virtue is not, like a skill or an art, a mere capacity: it must actually engage the will.

II
I shall now turn to another thesis about the virtues, which I might express by saying
that they are corrective, each one standing at a point at which there is some temptation to
be resisted or deficiency of motivation to be made good. As Aristotle put it, virtues are
about what is difficult for men, and I want to see in what sense this is true, and then to
consider a problem in Kant's moral philosophy in the light of what has been said.
Let us first think about courage and temperance. Aristotle and Aquinas contrasted these
virtues with justice in the following respect. Justice was concerned with operations, and
courage and temperance with passions. 7 What they meant by this seems to have been,

5 von Wright op. cit. chapter VIII.

6 Aristotle op. cit. 1140 b 22–25. Aquinas op. cit. 1a2ae Q.57 a.4.
7 Aristotle op. cit. 1106 b 15 and 1129 a.4 have this implication; but Aquinas is more explicit in op. cit.

1a2ae Q.60 a.2.

 Philippa Foot

primarily, that the man of courage does not fear immoderately nor the man of
temperance have immoderate desires for pleasure, and that there was no corresponding
moderation of a passion implied in the idea of justice. This particular account of
courage and temperance might be disputed on the ground that a man's courage is
measured by his action and not by anything as uncontrollable as fear; and similarly that
the temperate man who must on occasion refuse pleasures need not desire them any less
than the intemperate man. Be that as it may (and something will be said about it later) it
is obviously true that courage and temperance have to do with particular springs of
action as justice does not. Almost any desire can lead a man to act unjustly, not even
excluding the desire to help a friend or to save a life, whereas a cowardly act must be
motivated by fear or a desire for safety, and an act of intemperance by a desire for
pleasure, perhaps even for a particular range of pleasures such as those of eating or
drinking or sex. And now, going back to the idea of virtues as correctives, one may say
that it is only because fear and the desire for pleasure often operate as temptations that
courage and temperance exist as virtues at all. As things are we often want to run away
not only where that is the right thing to do but also where we should stand firm; and
we want pleasure not only where we should seek pleasure but also where we should
not. If human nature had been different there would have been no need of a corrective
disposition in either place, as fear and pleasure would have been good guides to
conduct throughout life. So Aquinas says, about the passions
They may incite us to something against reason, and so we need a curb, which we name
temperance. Or they may make us shirk a course of action dictated by reason, through
fear of dangers or hardships. Then a person needs to be steadfast and not run away
from what is right; and for this courage is named. 8
As with courage and temperance so with many other virtues: there is, for instance, a
virtue of industriousness only because idleness is a temptation; and of humility only
because men tend to think too well of themselves. Hope is a virtue because despair too
is a temptation; it might have been that no one cried that all was lost except where he
could really see it to be so, and in this case there would have been no virtue of hope.
With virtues such as justice and charity it is a little different, because they correspond
not to any particular desire or tendency that has to be kept in check but rather to a
deficiency of motivation; and it is this that they must make good. If people were as
much attached to the good of others as they are to their own good there would no more
be a general virtue of benevolence than there is a general virtue of self-love. And if
people cared about the rights of others as they care about their own rights no virtue of
justice would be needed to look after the matter, and rules about such things as
contracts and promises would only need to be made public, like the rules of a game that
everyone was eager to play.

8 Aquinas op. cit. 1a2ae Q.61 a.3.

 I Virtues and Vices

On this view of the virtues and vices everything is seen to depend on what human
nature is like, and the traditional catalogue of the two kinds of dispositions is not hard
to understand. Nevertheless it may be defective, and anyone who accepts the thesis that
I am putting forward will feel free to ask himself where the temptations and
deficiencies that need correcting are really to be found. It is possible, for example, that
the theory of human nature lying behind the traditional list of the virtues and vices puts
too much emphasis on hedonistic and sensual impulses, and does not sufficiently take
account of less straightforward inclinations such as the desire to be put upon and
dissatisfied, or the unwillingness to accept good things as they come along.
It should now be clear why I said that virtues should be seen as correctives; and part of
what is meant by saying that virtue is about things that are difficult for men should also
have appeared. The further application of this idea is, however, controversial, and the
following difficulty presents itself: that we both are and are not inclined to think that
the harder a man finds it to act virtuously the more virtue he shows if he does act well.
For on the one hand great virtue is needed where it is particularly hard to act
virtuously; yet on the other it could be argued that difficulty in acting virtuously shows
that the agent is imperfect in virtue: according to Aristotle, to take pleasure in virtuous
action is the mark of true virtue, with the self-mastery of the one who finds virtue
difficult only a second best. How then is this conflict to be decided? Who shows most
courage, the one who wants to run away but does not, or the one who does not even
want to run away? Who shows most charity, the one who finds it easy to make the good
of others his object, or the one who finds it hard?
What is certain is that the thought that virtues are corrective does not constrain us to
relate virtue to difficulty in each individual man. Since men in general find it hard to
face great dangers or evils, and even small ones, we may count as courageous those few
who without blindness or indifference are nevertheless fearless even in terrible
circumstances. And when someone has a natural charity or generosity it is at least part
of the virtue that he has; if natural virtue cannot be the whole of virtue this is because a
kindly or fearless disposition could be disastrous without justice and wisdom, and
because these virtues have to be learned, not because natural virtue is too easily
acquired. I have argued that the virtues can be seen as correctives in relation to human
nature in general but not that each virtue must present a difficulty to each and every
man.
Nevertheless many people feel strongly inclined to say that it is for moral effort that
moral praise is to be bestowed, and that in proportion as a man finds it easy to be
virtuous so much the less is he to be morally admired for his good actions. The dilemma
can be resolved only when we stop talking about difficulties standing in the way of
virtuous action as if they were of only one kind. The fact is that some kinds of
difficulties do indeed provide an occasion for much virtue, but that others rather show
that virtue is incomplete.
 Philippa Foot

To illustrate this point I shall first consider an example of honest action. We may
suppose for instance that a man has an opportunity to steal, in circumstances where
stealing is not morally permissible, but that he refrains. And now let us ask our old
question. For one man it is hard to refrain from stealing and for another man it is not:
which shows the greater virtue in acting as he should? It is not difficult to see in this
case that it makes all the difference whether the difficulty comes from circumstances, as
that a man is poor, or that his theft is unlikely to be detected, or whether it comes from
something that belongs to his own character. The fact that a man is tempted to steal is
something about him that shows a certain lack of honesty: of the thoroughly honest
man we say that it ‘never entered his head’, meaning that it was never a real possibility
for him. But the fact that he is poor is something that makes the occasion more tempting,
and difficulties of this kind make honest action all the more virtuous.
A similar distinction can be made between different obstacles standing in the way of
charitable action. Some circumstances, as that great sacrifice is needed, or that the one to
be helped is a rival, give an-occasion on which a man's charity is severely tested. Yet in
given circumstances of this kind it is the man who acts easily rather than the one who
finds it hard who shows the most charity. Charity is a virtue of attachment, and that
sympathy for others which makes it easier to help them is part of the virtue itself.
These are fairly simple cases, but I am not supposing that it is always easy to say where
the relevant distinction is to be drawn. What, for instance, should we say about the
emotion of fear as an obstacle to action? Is a man more courageous if he fears much and
nevertheless acts, or if he is relatively fearless? Several things must be said about this. In
the first place it seems that the emotion of fear is not a necessary condition for the
display of courage; in face of a great evil such as death or injury a man may show
courage even if he does not tremble. On the other hand even irrational fears may give
an occasion for courage: if someone suffers from claustrophobia or a dread of heights he
may require courage to do that which would not be a courageous action for others. But
not all fears belong from this point of view to the circumstances rather than to a man's
character. For while we do not think of claustrophobia or a dread of heights as features
of character, a general timorousness may be. Thus, although pathological fears are not
the result of a man's choices and values some fears may be. The fears that count against
a man's courage are those that we think he could overcome, and among them, in a
special class, those that reflect the fact that he values safety too much.
In spite of problems such as these, which have certainly not all been solved, both the
distinction between different kinds of obstacles to virtuous action and the general idea
that virtues are correctives will be useful in resolving a difficulty in Kant's moral
philosophy closely related to the issues discussed in the preceding paragraphs. In a
passage in the first section of the Groundwork of the Metaphysics of Morals Kant
notoriously tied himself into a knot in trying to give an account of those actions which
have as he put it ‘positive moral worth’. Arguing that only actions done out of a sense
 I Virtues and Vices

of duty have this worth he contrasts a philanthropist who ‘takes pleasure in spreading
happiness around him’ with one who acts out of respect for duty, saying that the
actions of the latter but not the former have moral worth. Much scorn has been poured
on Kant for this curious doctrine, and indeed it does seem that something has gone
wrong, but perhaps we are not in a position to scoff unless we can give our own
account of the idea on which Kant is working. After all it does seem that he is right in
saying that some actions are in accordance with duty, and even required by duty,
without being the subjects of moral praise, like those of the honest trader who deals
honestly in a situation in which it is in his interest to do so.
It was this kind of example that drove Kant to his strange conclusion. He added another
example, however, in discussing acts of self-preservation; these he said, while they
normally have no positive moral worth, may have it when a man preserves his life not
from inclination but without inclination and from a sense of duty. Is he not right in
saying that acts of self-preservation normally have no moral significance but that they
may have it, and how do we ourselves explain this fact?
To anyone who approaches this topic from a consideration of the virtues the solution
readily suggests itself. Some actions are in accordance with virtue without requiring
virtue for their performance, whereas others are both in accordance with virtue and
such as to show possession of a virtue. So Kant's trader was dealing honestly in a
situation in which the virtue of honesty is not required for honest dealing, and it is for
this reason that his action did not have ‘positive moral worth’. Similarly, the care that
one ordinarily takes for one's life, as for instance on some ordinary morning in eating
one's breakfast and keeping out of the way of a car on the road, is something for which
no virtue is required. As we said earlier there is no general virtue of self-love as there is
a virtue of benevolence or charity, because men are generally attached sufficiently to
their own good. Nevertheless in special circumstances virtues such as temperance,
courage, fortitude, and hope may be needed if someone is to preserve his life. Are these
circumstances in which the preservation of one's own life is a duty? Sometimes it is so,
for sometimes it is what is owed to others that should keep a man from destroying
himself, and then he may act out of a sense of duty. But not all cases in which acts of
self-preservation show virtue are like this. For a man may display each of the virtues
just listed even where he does not do any harm to others if he kills himself or fails to
preserve his life. And it is this that explains why there may be a moral aspect to suicide
which does not depend on possible injury to other people. It is not that suicide is
‘always wrong’, whatever that would mean, but that suicide is sometimes contrary to
virtues such as courage and hope.
Let us now return to Kant's philanthropists, with the thought that it is action that is in
accordance with virtue and also displays a virtue that has moral worth. We see at once
that Kant's difficulties are avoided, and the happy philanthropist reinstated in the
position which belongs to him. For charity is, as we said, a virtue of attachment as well
 Philippa Foot

as action, and the sympathy that makes it easier to act with charity is part of the virtue.
The man who acts charitably out of a sense of duty is not to be undervalued, but it is the
other who most shows virtue and therefore to the other that most moral worth is
attributed. Only a detail of Kant's presentation of the case of the dutiful philanthropist
tells on the other side. For what he actually said was that this man felt no sympathy and
took no pleasure in the good of others because ‘his mind was clouded by some sorrow
of his own’, and this is the kind of circumstance that increases the virtue that is needed
if a man is to act well.

III
It was suggested above that an action with ‘positive moral worth’, or as we might say a
positively good action, was to be seen as one which was in accordance with virtue, by
which I mean contrary to no virtue, and moreover one for which a virtue was required.
Nothing has so far been said about another case, excluded by the formula, in which it
might seem that an act displaying one virtue was nevertheless contrary to another. In
giving this last description I am thinking not of two virtues with competing claims, as if
what were required by justice could nevertheless be demanded by charity, or
something of that kind, but rather of the possibility that a virtue such as courage or
temperance or industry which overcomes a special temptation, might be displayed in
an act of folly or villainy. Is this something that we must allow for, or is it only good or
innocent actions which can be acts of these virtues? Aquinas, in his definition of virtue,
said that virtues can produce only good actions, and that they are dispositions ‘of which
no one can make bad use’, 9 except when they are treated as objects, as in being the
subject of hatred or pride. The common opinion nowadays is, however, quite different.
With the notable exception of Peter Geach hardly anyone sees any difficulty in the
thought that virtues may sometimes be displayed in bad actions. Von Wright, for
instance, speaks of the courage of the villain as if this were a quite unproblematic idea,
and most people take it for granted that the virtues of courage and temperance may aid
a bad man in his evil work. It is also supposed that charity may lead a man to act badly,
as when someone does what he has no right to do, but does it for the sake of a friend.
There are, however, reasons for thinking that the matter is not so simple as this. If a
man who is willing to do an act of injustice to help a friend, or for the common good, is
supposed to act out of charity, and he so acts where a just man will not, it should be
said that the unjust man has more charity than the just man. But do we not think that
someone not ready to act unjustly may yet be perfect in charity, the virtue having done
its whole work in prompting him to do the acts that are permissible? And is there not
more difficulty than might appear in the idea of an act of injustice which is nevertheless
an act of courage? Suppose for instance that a sordid murder were in question, say a

9 Aquinas op. cit. 1a2ae Q.56 a.5.

 I Virtues and Vices

murder done for gain or to get an inconvenient person out of the way, but that this
murder had to be done in alarming circumstances or in the face of real danger; should
we be happy to say that such an action was an act of courage or a courageous act? Did
the murderer, who certainly acted boldly, or with intrepidity, if he did the murder, also
act courageously? Some people insist that they are ready to say this, but I have noticed
that they like to move over to a murder for the sake of conscience, or to some other act
done in the course of a villainous enterprise but whose immediate end is innocent or
positively good. On their hypothesis, which is that bad acts can easily be seen as
courageous acts or acts of courage, my original example should be just as good.
What are we to say about this difficult matter? There is no doubt that the murderer who
murdered for gain was not a coward: he did not have a second moral defect which
another villain might have had. There is no difficulty about this because it is clear that
one defect may neutralise another. As Aquinas remarked, it is better for a blind horse if
it is slow. 10 It does not follow, however, that an act of villainy can be courageous; we are
inclined to say that it ‘took courage’, and yet it seems wrong to think of courage as
equally connected with good actions and bad.
One way out of this difficulty might be to say that the man who is ready to pursue bad
ends does indeed have courage, and shows courage in his action, but that in him
courage is not a virtue. Later I shall consider some cases in which this might be the right
thing to say, but in this instance it does not seem to be. For unless the murderer
consistently pursues bad ends his courage will often result in good; it may enable him
to do many innocent or positively good things for himself or for his family and friends.
On the strength of an individual bad action we can hardly say that in him courage is not
a virtue. Nevertheless there is something to be said even about the individual action to
distinguish it from one that would readily be called an act of courage or a courageous
act. Perhaps the following analogy may help us to see what it is. We might think of
words such as ‘courage’ as naming characteristics of human beings in respect of a
certain power, as words such as ‘poison’ and ‘solvent’ and ‘corrosive’ so name the
properties of physical things. The power to which virtue-words are so related is the
power of producing good action, and good desires. But just as poisons, solvents and
corrosives do not always operate characteristically, so it could be with virtues. If P (say
arsenic) is a poison it does not follow that P acts as a poison wherever it is found. It is
quite natural to say on occasion ‘P does not act as a poison here’ though P is a poison
and it is P that is acting here. Similarly courage is not operating as a virtue when the
murderer turns his courage, which is a virtue, to bad ends. Not surprisingly the
resistance that some of us registered was not to the expression ‘the courage of the
murderer’ or to the assertion that what he did ‘took courage’ but rather to the
description of that action as an act of courage or a courageous act. It is not that the

10 Aquinas op. cit. 1a2ae Q.58 a.4.

 Philippa Foot

action could not be so described, but that the fact that courage does not here have its
characteristic operation is a reason for finding the description strange.
In this example we were considering an action in which courage was not operating as a
virtue, without suggesting that in that agent it generally failed to do so. But the latter is
also a possibility. If someone is both wicked and foolhardy this may be the case with
courage, and it is even easier to find examples of a general connexion with evil rather
than good in the case of some other virtues. Suppose, for instance, that we think of
someone who is over-industrious, or too ready to refuse pleasure, and this is
characteristic of him rather than something we find on one particular occasion. In this
case the virtue of industry, or the virtue of temperance, has a systematic connexion with
defective action rather than good action; and it might be said in either case that the
virtue did not operate as a virtue in this man. Just as we might say in a certain setting ‘P
is not a poison here’ though P is a poison and P is here, so we might say that
industriousness, or temperance, is not a virtue in some. Similarly in a man habitually
given to wishful thinking, who clings to false hopes, hope does not operate as a virtue
and we may say that it is not a virtue in him.
The thought developed in the last paragraph, to the effect that not every man who has a
virtue has something that is a virtue in him, may help to explain a certain discomfort
that one may feel when discussing the virtues. It is not easy to put one's finger on what
is wrong, but it has something to do with a disparity between the moral ideals that may
seem to be implied in our talk about the virtues, and the moral judgements that we
actually make. Someone reading the foregoing pages might, for instance, think that the
author of this paper always admired most those people who had all the virtues, being
wise and temperate as well as courageous, charitable, and just. And indeed it is
sometimes so. There are some people who do possess all these virtues and who are
loved and admired by all the world, as Pope John XXIII was loved and admired. Yet the
fact is that many of us look up to some people whose chaotic lives contain rather little of
wisdom or temperance, rather than to some others who possess these virtues. And
while it may be that this is just romantic nonsense I suspect that it is not. For while
wisdom always operates as a virtue, its close relation prudence does not, and it is
prudence rather than wisdom that inspires many a careful life. Prudence is not a virtue
in everyone, any more than industriousness is, for in some it is rather an over-anxious
concern for safety and propriety, and a determination to keep away from people or
situations which are apt to bring trouble with them; and by such defensiveness much
good is lost. It is the same with temperance. Intemperance can be an appalling thing, as
it was with Henry VIII of whom Wolsey remarked that
rather than he will either miss or want any part of his will or appetite, he will put the
loss of one half of his realm in danger.
Nevertheless in some people temperance is not a virtue, but is rather connected with
timidity or with a grudging attitude to the acceptance of good things. Of course what is
 I Virtues and Vices

best is to live boldly yet without imprudence or intemperance, but the fact is that rather
few can manage that.
 Philippa Foot

II The Problem of Abortion and the Doctrine of the Double Effect 1

Philippa Foot
Abstract: Abortion is considered in relation to the doctrine of the double effect, which distinguishes
between the intentions behind an action and consequences foreseen but not desired. This doctrine, often
used to defend the position of the Catholic Church on abortion, is here rejected. Alternative distinctions
are drawn, between what we do and what we allow to happen, and between the infringement of
‘positive’ and ‘negative’ duties. These distinctions are applied to the problem of abortion and to other
moral dilemmas.

One of the reasons why most of us feel puzzled about the problem of abortion is that we
want, and do not want, to allow to the unborn child the rights that belong to adults and
children. When we think of a baby about to be born it seems absurd to think that the
next few minutes or even hours could make so radical a difference to its status; yet as
we go back in the life of the foetus we are more and more reluctant to say that this is a
human being and must be treated as such. No doubt this is the deepest source of our
dilemma, but it is not the only one. For we are also confused about the general question
of what we may and may not do where the interests of human beings conflict. We have
strong intuitions about certain cases; saying, for instance, that it is all right to raise the
level of education in our country, though statistics allow us to predict that a rise in the
suicide rate will follow, while it is not all right to kill the feeble-minded to aid cancer
research. It is not easy, however, to see the principles involved, and one way of
throwing light on the abortion issue will be by setting up parallels involving adults or
children once born. So we will be able to isolate the ‘equal rights’ issue and should be
able to make some advance.
I shall not, of course, discuss all the principles that may be used in deciding how to act
where the interests or rights of human beings conflict. What I want to do is to look at
one particular theory, known as the ‘doctrine of the double effect’ which is invoked by
Catholics in support of their views on abortion but supposed by them to apply
elsewhere. As used in the abortion argument this doctrine has often seemed to non-
Catholics to be a piece of complete sophistry. In the last number of the Oxford Review it
was given short shrift by Professor Hart. 2 And yet this principle has seemed to some
non-Catholics as well as to Catholics to stand as the only defence against decisions on

1 ‘The Problem of Abortion and the Doctrine of the Double Effect’ originally appeared in the Oxford
Review, Number 5, 1967.
2 H. L. A. Hart, ‘Intention and Punishment’, Oxford Review, Number 4, Hilary 1967 . Reprinted in Hart,

Punishment and Responsibility (Oxford, 1968) . I owe much to this article and to a conversation with
Professor Hart, though I do not know whether he will approve of what follows.
 II The Problem of Abortion and the Doctrine of the
Double Effect11F

other issues that are quite unacceptable. It will help us in our difficulty about abortion if
this conflict can be resolved.
The doctrine of the double effect is based on a distinction between what a man foresees
as a result of his voluntary action and what, in the strict sense, he intends. He intends in
the strictest sense both those things that he aims at as ends and those that he aims at as
means to his ends. The latter may be regretted in themselves but nevertheless desired
for the sake of the end, as we may intend to keep dangerous lunatics confined for the
sake of our safety. By contrast a man is said not strictly, or directly, to intend the
foreseen consequences of his voluntary actions where these are neither the end at which
he is aiming nor the means to this end. Whether the word ‘intention’ should be applied
in both cases is not of course what matters: Bentham spoke of ‘oblique intention’,
contrasting it with the ‘direct intention’ of ends and means, and we may as well follow
his terminology. Everyone must recognise that some such distinction can be made,
though it may be made in a number of different ways, and it is the distinction that is
crucial to the doctrine of the double effect. The words ‘double effect’ refer to the two
effects that an action may produce: the one aimed at, and the one foreseen but in no
way desired. By ‘the doctrine of the double effect’ I mean the thesis that it is sometimes
permissible to bring about by oblique intention what one may not directly intend. Thus
the distinction is held to be relevant to moral decision in certain difficult cases. It is said
for instance that the operation of hysterectomy involves the death of the foetus as the
foreseen but not strictly or directly intended consequence of the surgeon's act, while
other operations kill the child and count as the direct intention of taking an innocent
life, a distinction that has evoked particularly bitter reactions on the part of non-
Catholics. If you are permitted to bring about the death of the child, what does it matter
how it is done? The doctrine of the double effect is also used to show why in another
case, where a woman in labour will die unless a craniotomy operation is performed, the
intervention is not to be condoned. There, it is said, we may not operate but must let the
mother die. We foresee her death but do not directly intend it, whereas to crush the
skull of the child would count as direct intention of its death. 3
This last application of the doctrine has been queried by Professor Hart on the ground
that the child's death is not strictly a means to saving the mother's life and should
logically be treated as an unwanted but foreseen consequence by those who make use of
the distinction between direct and oblique intention. To interpret the doctrine in this
way is perfectly reasonable given the language that has been used; it would, however,
make nonsense of it from the beginning. A certain event may be desired under one of its
descriptions, unwanted under another, but we cannot treat these as two different

3For discussions of the Catholic doctrine on abortion see Glanville Williams, The Sanctity of Life and the
Criminal Law (New York, 1957) ; also N. St. John Stevas, The Right to Life (London, 1963)
 Philippa Foot

events, one of which is aimed at and the other not. And even if it be argued that there
are here two different events—the crushing of the child's skull and its death—the two
are obviously much too close for an application of the doctrine of the double effect. To
see how odd it would be to apply the principle like this we may consider the story, well
known to philosophers, of the fat man stuck in the mouth of the cave. A party of
potholers have imprudently allowed the fat man to lead them as they make their way
out of the cave, and he gets stuck, trapping the others behind him. Obviously the right
thing to do is to sit down and wait until the fat man grows thin; but philosophers have
arranged that flood waters should be rising within the cave. Luckily (luckily?) the
trapped party have with them a stick of dynamite with which they can blast the fat man
out of the mouth of the cave. Either they use the dynamite or they drown. In one
version the fat man, whose head is in the cave, will drown with them; in the other he
will be rescued in due course. 4 Problem: may they use the dynamite or not? Later we
shall find parallels to this example. Here it is introduced for light relief and because it
will serve to show how ridiculous one version of the doctrine of the double effect would
be. For suppose that the trapped explorers were to argue that the death of the fat man
might be taken as a merely foreseen consequence of the act of blowing him up. (‘We
didn't want to kill him . . . only to blow him into small pieces’ or even ‘ . . . only to blast
him out of the cave.’) I believe that those who use the doctrine of the double effect
would rightly reject such a suggestion, though they will, of course, have considerable
difficulty in explaining where the line is to be drawn. What is to be the criterion of
‘closeness’ if we say that anything very close to what we are literally aiming at counts as
if part of our aim?
Let us leave this difficulty aside and return to the arguments for and against the
doctrine, supposing it to be formulated in the way considered most effective by its
supporters, and ourselves bypassing the trouble by taking what must on any reasonable
definition be clear cases of ‘direct’ or ‘oblique’ intention.
The first point that should be made clear, in fairness to the theory, is that no one is
suggesting that it does not matter what you bring about as long as you merely foresee
and do not strictly intend the evil that follows. We might think, for instance, of the
(actual) case of wicked merchants selling, for cooking, oil they knew to be poisonous
and thereby killing a number of innocent people, comparing and contrasting it with that
of some unemployed gravediggers, desperate for custom, who got hold of this same oil
and sold it (or perhaps they secretly gave it away) in order to create orders for graves.
They strictly (directly) intend the deaths they cause, while the merchants could say that
it was not part of their plan that anyone should die. In morality, as in law, the
merchants, like the gravediggers, would be considered as murderers; nor are the
supporters of the doctrine of the double effect bound to say that there is the least

4 It was Professor Hart who drew my attention to this distinction.

 II The Problem of Abortion and the Doctrine of the
Double Effect11F

difference between them in respect of moral turpitude. What they are committed to is
the thesis that sometimes it makes a difference to the permissibility of an action involving
harm to others that this harm, although foreseen, is not part of the agent's direct
intention. An end such as earning one's living is clearly not such as to justify either the
direct or oblique intention of the death of innocent people, but in certain cases one is
justified in bringing about knowingly what one could not directly intend.
It is now time to say why this doctrine should be taken seriously in spite of the fact that
it sounds rather odd, that there are difficulties about the distinction on which it
depends, and that it seemed to yield one sophistical conclusion when applied to the
problem of abortion. The reason for its appeal is that its opponents have often seemed to
be committed to quite indefensible views. Thus the controversy has raged around
examples such as the following. Suppose that a judge or magistrate is faced with rioters
demanding that a culprit be found for a certain crime and threatening otherwise to take
their own bloody revenge on a particular section of the community. The real culprit
being unknown, the judge sees himself as able to prevent the bloodshed only by
framing some innocent person and having him executed. Beside this example is placed
another in which a pilot whose aeroplane is about to crash is deciding whether to steer
from a more to a less inhabited area. To make the parallel as close as possible it may
rather be supposed that he is the driver of a runaway tram which he can only steer from
one narrow track on to another; five men are working on one track and one man on the
other; anyone on the track he enters is bound to be killed. In the case of the riots the
mob have five hostages, so that in both the exchange is supposed to be one man's life
for the lives of five. The question is why we should say, without hesitation, that the
driver should steer for the less occupied track, while most of us would be appalled at
the idea that the innocent man could be framed. It may be suggested that the special
feature of the latter case is that it involves the corruption of justice, and this is, of course,
very important indeed. But if we remove that special feature, supposing that some
private individual is to kill an innocent person and pass him off as the criminal we still
find ourselves horrified by the idea. The doctrine of double effect offers us a way out of
the difficulty, insisting that it is one thing to steer towards someone foreseeing that you
will kill him and another to aim at his death as part of your plan. Moreover there is one
very important element of good in what is here insisted. In real life it would hardly ever
be certain that the man on the narrow track would be killed. Perhaps he might find a
foothold on the side of the tunnel and cling on as the vehicle hurtled by. The driver of
the tram does not then leap off and brain him with a crowbar. The judge, however,
needs the death of the innocent man for his (good) purposes. If the victim proves hard
to hang he must see to it that he dies another way. To choose to execute him is to choose
that this evil shall come about, and this must therefore count as a certainty in weighing up
the good and evil involved. The distinction between direct and oblique intention is
 Philippa Foot

crucial here, and is of great importance in an uncertain world. Nevertheless this is no

way to defend the doctrine of double effect. For the question is whether the difference
between aiming at something and obliquely intending it is in itself relevant to moral
decisions; not whether it is important when correlated with a difference of certainty in
the balance of good and evil. Moreover we are particularly interested in the application
of the doctrine of the double effect to the question of abortion, and no one can deny that
in medicine there are sometimes certainties so complete that it would be a mere quibble
to speak of the ‘probable outcome’ of this course of action or that. It is not, therefore,
with a merely philosophical interest that we should put aside the uncertainty and
scrutinise the examples to test the doctrine of the double effect. Why can we not argue
from the case of the steering driver to that of the judge?
Another pair of examples poses a similar problem. We are about to give a patient who
needs it to save his life a massive dose of a certain drug in short supply. There arrive,
however, five other patients each of whom could be saved by one-fifth of that dose. We
say with regret that we cannot spare our whole supply of the drug for a single patient,
just as we should say that we could not spare the whole resources of a ward for one
dangerously ill individual when ambulances arrive bringing in victims of a multiple
crash. We feel bound to let one man die rather than many if that is our only choice. Why
then do we not feel justified in killing people in the interests of cancer research or to
obtain, let us say, spare parts for grafting on to those who need them? We can suppose,
similarly, that several dangerously ill people can be saved only if we kill a certain
individual and make a serum from his dead body. (These examples are not over-
fanciful considering present controversies about prolonging the life of mortally ill
patients whose eyes or kidneys are to be used for others.) Why cannot we argue from
the case of the scarce drug to that of the body needed for medical purposes? Once again
the doctrine of the double effect comes up with an explanation. In one kind of case but
not the other we aim at the death of an innocent man.
A further argument suggests that if the doctrine of the double effect is rejected this has
the consequences of putting us hopelessly in the power of bad men. Suppose for
example that some tyrant should threaten to torture five men if we ourselves would not
torture one. Would it be our duty to do so, supposing we believed him, because this
would be no different from choosing to rescue five men from his torturers rather than
one? If so, anyone who wants us to do something we think wrong has only to threaten
that otherwise he himself will do something we think worse. A mad murderer, known
to keep his promises, could thus make it our duty to kill some innocent citizen to
prevent him from killing two. From this conclusion we are again rescued by the
doctrine of the double effect. If we refuse, we foresee that the greater number will be
killed but we do not intend it: it is he who intends (that is strictly or directly intends) the
death of innocent persons; we do not.
 II The Problem of Abortion and the Doctrine of the
Double Effect11F

At one time I thought that these arguments in favour of the doctrine of the double effect
were conclusive, but I now believe that the conflict should be solved in another way.
The clue that we should follow is that the strength of the doctrine seems to lie in the
distinction it makes between what we do (equated with direct intention) and what we
allow (thought of as obliquely intended). Indeed it is interesting that the disputants
tend to argue about whether we are to be held responsible for what we allow as we are
for what we do. 5 Yet it is not obvious that this is what they should be discussing, since
the distinction between what one does and what one allows to happen is not the same
as that between direct and oblique intention. To see this one has only to consider that it
is possible deliberately to allow something to happen, aiming at it either for its own sake
or as part of one's plan for obtaining something else. So one person might want another
person dead and deliberately allow him to die. And again one may be said to do things
that one does not aim at, as the steering driver would kill the man on the track.
Moreover there is a large class of things said to be brought about rather than either
done or allowed where either kind of intention is possible. So it is possible to bring about
a man's death by getting him to sea in a leaky boat, and the intention of his death may
be either direct or oblique.
Whatever it may, or may not, have to do with the doctrine of the double effect, the idea
of allowing is worth looking into in this context. I shall leave aside the special case of
giving permission, which involves the idea of authority, and consider the two main
divisions into which cases of allowing seem to fall. There is firstly the allowing which is
forbearing to prevent. For this we need a sequence thought of as somehow already in
train, and something that an agent could do to intervene. (The agent must be able to
intervene, but does not do so.) So, for instance, he could warn someone, but allows him
to walk into a trap. He could feed an animal but allows it to die for lack of food. He
could stop a leaking tap but allows the water to go on flowing. This is the case of
allowing with which we shall be concerned, but the other should be mentioned. It is the
kind of allowing which is roughly equivalent to enabling; the root idea being the
removal of some obstacle which is, as it were, holding back a train of events. So
someone may remove a plug and allow water to flow; open a door and allow an animal
to get out; or give someone money and allow him to get back on his feet.
The first kind of allowing requires an omission, but there is no other general correlation
between omission and allowing, commission and bringing about or doing. An actor
who fails to turn up for a performance will generally spoil it rather than allow it to be
spoiled. I mention the distinction between omission and commission only to set it aside.

5See, e.g., J. Bennett, ‘Whatever the Consequences’, Analysis, January 1966 , and G. E. M. Anscombe's
reply in Analysis, June 1966. See also Miss Anscombe's ‘Modern Moral Philosophy’ in Philosophy, January
1958.
 Philippa Foot

Thinking of the first kind of allowing (forbearing to prevent), we should ask whether
there is any difference, from the moral point of view, between what one does or causes
and what one merely allows. It seems clear that on occasions one is just as bad as the
other, as is recognised in both morality and law. A man may murder his child or his
aged relatives by allowing them to die of starvation as by giving poison; he may also be
convicted of murder on either account. In another case we would, however, make a
distinction. Most of us allow people to die of starvation in India and Africa, and there is
surely something wrong with us that we do; it would be nonsense, however, to pretend
that it is only in law that we make the distinction between allowing people in the under-
developed countries to die of starvation and sending them poisoned food. There is
worked into our moral system a distinction between what we owe people in the form of
aid and what we owe them in the way of non-interference. Salmond, in his
Jurisprudence, expressed as follows the distinction between the two.
A positive right corresponds to a positive duty, and is a right that he on whom the duty
lies shall do some positive act on behalf of the person entitled. A negative right
corresponds to a negative duty, and is a right that the person bound shall refrain from
some act which would operate to the prejudice of the person entitled. The former is a
right to be positively benefited; the latter is merely a right not to be harmed. 6
As a general account of rights and duties this is defective, since not all are so closely
connected with benefit and harm. Nevertheless for our purposes it will do well. Let us
speak of negative duties when thinking of the obligation to refrain from such things as
killing or robbing, and of the positive duty, e.g., to look after children or aged parents. It
will be useful, however, to extend the notion of positive duty beyond the range of
things that are strictly called duties, bringing acts of charity under this heading. These
are owed only in a rather loose sense, and some acts of charity could hardly be said to
be owed at all, so I am not following ordinary usage at this point.
Let us now see whether the distinction of negative and positive duties explains why we
see differently the action of the steering driver and that of the judge, of the doctors who
withhold the scarce drug and those who obtain a body for medical purposes, of those
who choose to rescue five men rather than one man from torture and those who are
ready to torture the one man themselves in order to save five. In each case we have a
conflict of duties, but what kind of duties are they? Are we, in each case, weighing
positive duties against positive, negative against negative, or one against the other? Is
the duty to refrain from injury, or rather to bring aid?
The steering driver faces a conflict of negative duties, since it is his duty to avoid
injuring five men and also his duty to avoid injuring one. In the circumstances he is not
able to avoid both, and it seems clear that he should do the least injury he can. The
judge, however, is weighing the duty of not inflicting injury against the duty of

6 J. Salmond, Jurisprudence, 11th edition, p. 283.

 II The Problem of Abortion and the Doctrine of the
Double Effect11F

bringing aid. He wants to rescue the innocent people threatened with death but can do
so only by inflicting injury himself. Since one does not in general have the same duty to
help people as to refrain from injuring them, it is not possible to argue to a conclusion
about what he should do from the steering driver case. It is interesting that, even where
the strictest duty of positive aid exists, this still does not weigh as if a negative duty
were involved. It is not, for instance, permissible to commit a murder to bring one's
starving children food. If the choice is between inflicting injury on one or many there
seems only one rational course of action; if the choice is between aid to some at the cost
of injury to others, and refusing to inflict the injury to bring the aid, the whole matter is
open to dispute. So it is not inconsistent of us to think that the driver must steer for the
road on which only one man stands while the judge (or his equivalent) may not kill the
innocent person in order to stop the riots. Let us now consider the second pair of
examples, which concern the scarce drug on the one hand and on the other the body
needed to save lives. Once again we find a difference based on the distinction between
the duty to avoid injury and the duty to provide aid. Where one man needs a massive
dose of the drug and we withhold it from him in order to save five men, we are
weighing aid against aid. But if we consider killing a man in order to use his body to
save others, we are thinking of doing him an injury to bring others aid. In an interesting
variant of the model, we may suppose that instead of killing someone we deliberately
let him die. (Perhaps he is a beggar to whom we are thinking of giving food, but then
we say ‘No, they need bodies for medical research.’) Here it does seem relevant that in
allowing him to die we are aiming at his death, but presumably we are inclined to see
this as a violation of negative rather than positive duty. If this is right, we see why we
are unable in either case to argue to a conclusion from the case of the scarce drug.
In the examples involving the torturing of one man or five men, the principle seems to
be the same as for the last pair. If we are bringing aid (rescuing people about to be
tortured by the tyrant), we must obviously rescue the larger rather than the smaller
group. It does not follow, however, that we would be justified in inflicting the injury, or
getting a third person to do so, in order to save the five. We may therefore refuse to be
forced into acting by the threats of bad men. To refrain from inflicting injury ourselves
is a stricter duty than to prevent other people from inflicting injury, which is not to say
that the other is not a very strict duty indeed.
So far the conclusions are the same as those at which we might arrive following the
doctrine of the double effect, but in others they will be different, and the advantage
seems to be all on the side of the alternative. Suppose, for instance, that there are five
patients in a hospital whose lives could be saved by the manufacture of a certain gas,
but that this inevitably releases lethal fumes into the room of another patient whom for
some reason we are unable to move. His death, being of no use to us, is clearly a side
effect, and not directly intended. Why then is the case different from that of the scarce
 Philippa Foot

drug, if the point about that is that we foresaw but did not strictly intend the death of
the single patient? Yet it surely is different. The relatives of the gassed patient would
presumably be successful if they sued the hospital and the whole story came out. We
may find it particularly revolting that someone should be used as in the case where he is
killed or allowed to die in the interest of medical research, and the fact of using may
even determine what we would decide to do in some cases, but the principle seems
unimportant compared with our reluctance to bring such injury for the sake of giving
aid.
My conclusion is that the distinction between direct and oblique intention plays only a
quite subsidiary role in determining what we say in these cases, while the distinction
between avoiding injury and bringing aid is very important indeed. I have not, of
course, argued that there are no other principles. For instance it clearly makes a
difference whether our positive duty is a strict duty or rather an act of charity: feeding
our own children or feeding those in faraway countries. It may also make a difference
whether the person about to suffer is one thought of as uninvolved in the threatened
disaster, and whether it is his presence that constitutes the threat to the others. In many
cases we find it very hard to know what to say, and I have not been arguing for any
general conclusion such as that we may never, whatever the balance of good and evil,
bring injury to one for the sake of aid to others, even when this injury amounts to death.
I have only tried to show that even if we reject the doctrine of the double effect we are
not forced to the conclusion that the size of the evil must always be our guide.
Let us now return to the problem of abortion, carrying out our plan of finding parallels
involving adults or children rather than the unborn. We must say something about the
different cases in which abortion might be considered on medical grounds.
First of all there is the situation in which nothing that can be done will save the life of
child and mother, but where the life of the mother can be saved by killing the child.
This is parallel to the case of the fat man in the mouth of the cave who is bound to be
drowned with the others if nothing is done. Given the certainty of the outcome, as it
was postulated, there is no serious conflict of interests here, since the fat man will perish
in either case, and it is reasonable that the action that will save someone should be
done. It is a great objection to those who argue that the direct intention of the death of
an innocent person is never justifiable that the edict will apply even in this case. The
Catholic doctrine on abortion must here conflict with that of most reasonable men.
Moreover we would be justified in performing the operation whatever the method
used, and it is neither a necessary nor a good justification of the special case of
hysterectomy that the child's death is not directly intended, being rather a foreseen
consequence of what is done. What difference could it make as to how the death is
brought about?
Secondly we have the case in which it is possible to perform an operation which will
save the mother and kill the child or kill the mother and save the child. This is parallel
 II The Problem of Abortion and the Doctrine of the
Double Effect11F

to the famous case of shipwrecked mariners who believed that they must throw
someone overboard if their boat was not to founder in a storm, and to the other famous
case of the two sailors, Dudley and Stephens, who killed and ate the cabin boy when
adrift on the sea without food. Here again there is no conflict of interests so far as the
decision to act is concerned; only in deciding whom to save. Once again it would be
reasonable to act, though one would respect someone who held back from the appalling
action either because he preferred to perish rather than do such a thing or because he
held on past the limits of reasonable hope. In real life the certainties postulated by
philosophers hardly ever exist, and Dudley and Stephens were rescued not long after
their ghastly meal. Nevertheless if the certainty were absolute, as it might be in the
abortion case, it would seem better to save one than none. Probably we should decide in
favour of the mother when weighing her life against that of the unborn child, but it is
interesting that, a few years later, we might easily decide it the other way.
The worst dilemma comes in the third kind of example where to save the mother we
must kill the child, say by crushing its skull, while if nothing is done the mother will
perish but the child can be safely delivered after her death. Here the doctrine of the
double effect has been invoked to show that we may not intervene, since the child's
death would be directly intended while the mother's would not. On a strict parallel
with cases not involving the unborn we might find the conclusion correct though the
reason given was wrong. Suppose, for instance, that in later life the presence of a child
was certain to bring death to the mother. We would surely not think ourselves justified
in ridding her of it by a process that involved its death. For in general we do not think
that we can kill one innocent person to rescue another, quite apart from the special care
that we feel is due to children once they have prudently got themselves born. What we
would be prepared to do when a great many people were involved is another matter,
and this is probably the key to one quite common view of abortion on the part of those
who take quite seriously the rights of the unborn child. They probably feel that if enough
people were involved one must be sacrificed, and they think of the mother's life against
the unborn child's life as if it were many against one. But of course many people do not
view it like this at all, having no inclination to accord to the foetus or unborn child
anything like ordinary human status in the matter of rights. I have not been arguing for
or against these points of view but only trying to discern some of the currents that are
pulling us back and forth. The levity of the examples is not meant to offend.
 Philippa Foot

III Euthanasia 1

Philippa Foot
Abstract: A problem in medical ethics is examined in this discussion of euthanasia, which starts with a
rejection of use of the term by the Nazis to describe their killing of unwanted mental patients. By
definition, euthanasia must be supposed to be a good to the person killed. But it is difficult to decide
when death is better than life, perhaps because we see a conceptual connexion between life and good.
Distinctions are drawn between active and passive, voluntary and non-voluntary euthanasia. A general
legalization of euthanasia is opposed because of the dangers of abuse, but it is suggested that Living Wills
and related contracts about treatment between doctors and patients should be allowed.

The widely used Shorter Oxford English Dictionary gives three meanings for the word
‘euthanasia’: the first, ‘a quiet and easy death’; the second, ‘the means of procuring this’;
and the third, ‘the action of inducing a quiet and easy death’. It is a curious fact that no
one of the three gives an adequate definition of the word as it is usually understood. For
‘euthanasia’ means much more than a quiet and easy death, or the means of procuring
it, or the action of inducing it. The definition specifies only the manner of death, and if
this were all that was implied a murderer, careful to drug his victim, could claim that
his act was an act of euthanasia. We find this ridiculous because we take it for granted
that in euthanasia it is death itself, not just the manner of death, that must be kind to the
one who dies.
To see how important it is that ‘euthanasia’ should not be used as the dictionary
definition allows it to be used, merely to signify that a death was quiet and easy, one
has only to remember that Hitler's ‘euthanasia’ programme traded on this ambiguity.
Under this programme, planned before the War but brought into full operation by a
decree of 1 September 1939, some 275,000 people were gassed in centres which were to
be a model for those in which Jews were later exterminated. Anyone in a state
institution could be sent to the gas chambers if it was considered that he could not be
‘rehabilitated’ for useful work. As Dr. Leo Alexander reports, relying on the testimony
of a neuropathologist who received 500 brains from one of the killing centres,
In Germany the exterminations included the mentally defective, psychotics (particularly
schizophrenics), epileptics and patients suffering from infirmities of old age and from
various organic neurological disorders such as infantile paralysis, Parkinsonism,
multiple sclerosis and brain tumors. . . . In truth, all those unable to work and
considered nonrehabilitable were killed. 2

1 ‘Euthanasia’ originally appeared in Philosophy and Public Affairs, Volume 6, Number 2, Winter 1977.
I would like to thank Derek Parfit and the editors of Philosophy & Public Affairs for their very helpful
comments.
2 Leo Alexander, ‘Medical Science under Dictatorship’, New England Journal of Medicine, 14 July 1949, p. 40
 III Euthanasia17F

These people were killed because they were ‘useless’ and ‘a burden on society’: only the
manner of their deaths could be thought of as relatively easy and quiet.
Let us insist, then, that when we talk about euthanasia we are talking about a death
understood as a good or happy event for the one who dies. This stipulation follows
etymology, but is itself not exactly in line with current usage, which would be captured
by the condition that the death should not be an evil rather than that it should be a good.
That this is how people talk is shown by the fact that the case of Karen Ann Quinlan
and others in a state of permanent coma is often discussed under the heading of
‘euthanasia’. Perhaps it is not too late to object to the use of the word ‘euthanasia’ in this
sense. Apart from the break with the Greek origins of the word there are other
unfortunate aspects of this extension of the term. For if we say that the death must be
supposed to be a good to the subject we can also specify that it shall be for his sake that
an act of euthanasia is performed. If we say merely that death shall not be an evil to
him, we cannot stipulate that benefiting him shall be the motive where euthanasia is in
question. Given the importance of the question, For whose sake are we acting? it is good
to have a definition of euthanasia which brings under this heading only cases of opting
for death for the sake of the one who dies. Perhaps what is most important is to say
either that euthanasia is to be for the good of the subject or at least that death is to be no
evil to him, thus refusing to talk Hitler's language. However, in this paper it is the first
condition that will be understood, with the additional proviso that by an act of
euthanasia we mean one of inducing or otherwise opting for death for the sake of the
one who is to die.
A few lesser points need to be cleared up. In the first place it must be said that the word
‘act’ is not to be taken to exclude omission: we shall speak of an act of euthanasia when
someone is deliberately allowed to die, for his own good, and not only when positive
measures are taken to see that he does. The very general idea we want is that of a choice
of action or inaction directed at another man's death and causally effective in the sense
that, in conjunction with actual circumstances, it is a sufficient condition of death. Of
complications such as over-determination, it will not be necessary to speak.
A second, and definitely minor, point about the definition of an act of euthanasia
concerns the question of fact versus belief. It has already been implied that one who
performs an act of euthanasia thinks that death will be merciful for the subject since we
have said that it is on account of this thought that the act is done. But is it enough that
he acts with this thought, or must things actually be as he thinks them to be? If one man
kills another, or allows him to die, thinking that he is in the last stages of a terrible
disease, though in fact he could have been cured, is this an act of euthanasia or not?
Nothing much seems to hang on our decision about this. The same condition has got to
enter into the definition whether as an element in reality or only as an element in the
agent's belief. And however we define an act of euthanasia culpability or justifiability
 Philippa Foot

will be the same: if a man acts through ignorance his ignorance may be culpable or it
may not.3
These are relatively easy problems to solve, but one that is dauntingly difficult has been
passed over in this discussion of the definition, and must now be faced. It is easy to say,
as if this raised no problems, that an act of euthanasia is by definition one aiming at the
good of the one whose death is in question, and that it is for his sake that his death is
desired. But how is this to be explained? Presumably we are thinking of some evil
already with him or to come on him if he continues to live, and death is thought of as a
release from this evil. But this cannot be enough. Most people's lives contain evils such
as grief or pain, but we do not therefore think that death would be a blessing to them.
On the contrary, life is generally supposed to be a good even for someone who is
unusually unhappy or frustrated. How is it that one can ever wish for death for the sake
of the one who is to die? This difficult question is central to the discussion of
euthanasia, and we shall literally not know what we are talking about if we ask whether
acts of euthanasia defined as we have defined them are ever morally permissible
without first understanding better the reason for saying that life is a good, and the
possibility that it is not always so.
If a man should save my life he would be my benefactor. In normal circumstances this is
plainly true; but does one always benefit another in saving his life? It seems certain that
he does not. Suppose, for instance, that a man were being tortured to death and was
given a drug that lengthened his sufferings; this would not be a benefit but the reverse.
Or suppose that in a ghetto in Nazi Germany a doctor saved the life of someone
threatened by disease, but that the man once cured was transported to an extermination
camp; the doctor might wish for the sake of the patient that he had died of the disease.
Nor would a longer stretch of life always be a benefit to the person who was given it.
Comparing Hitler's camps with those of Stalin, Dimitri Panin observes that in the latter
the method of extermination was made worse by agonies that could stretch out over
months.
Death from a bullet would have been bliss compared with what many millions had to
endure while dying of hunger. The kind of death to which they were condemned has
nothing to equal it in treachery and sadism. 4
These examples show that to save or prolong a man's life is not always to do him a
service: it may be better for him if he dies earlier rather than later. It must therefore be
agreed that while life is normally a benefit to the one who has it, this is not always so.

3 For a discussion of culpable and nonculpable ignorance see Thomas Aquinas, Summa Theologica, First
Part of the Second Part, Question 6, article 8, and Question 19, articles 5 and 6.
4 Dimitri Panin, The Notebooks of Sologdin (London, 1976), pp. 66–7
 III Euthanasia17F

The judgement is often fairly easy to make—that life is or is not a good to someone—
but the basis for it is very hard to find. When life is said to be a benefit or a good, on
what grounds is the assertion made?
The difficulty is underestimated if it is supposed that the problem arises from the fact
that one who is dead has nothing, so that the good someone gets from being alive
cannot be compared with the amount he would otherwise have had. For why should
this particular comparison be necessary? Surely it would be enough if one could say
whether or not someone whose life was prolonged had more good than evil in the extra
stretch of time. Such estimates are not always possible, but frequently they are; we say,
for example, ‘He was very happy in those last years’, or, ‘He had little but unhappiness
then’. If the balance of good and evil determined whether life was a good to someone
we would expect to find a correlation in the judgements. In fact, of course, we find
nothing of the kind. First, a man who has no doubt that existence is a good to him may
have no idea about the balance of happiness and unhappiness in his life, or of any other
positive and negative factors that may be suggested. So the supposed criteria are not
always operating where the judgement is made. And secondly the application of the
criteria gives an answer that is often wrong. Many people have more evil than good in
their lives; we do not, however, conclude that we would do these people no service by
rescuing them from death.
To get around this last difficulty Thomas Nagel has suggested that experience itself is a
good which must be brought in to balance accounts.
. . . life is worth living even when the bad elements of experience are plentiful,
and the good ones too meager to outweigh the bad ones on their own. The
additional positive weight is supplied by experience itself, rather than by any of
its contents. 5
This seems implausible because if experience itself is a good it must be so even when
what we experience is wholly bad, as in being tortured to death. How should one
decide how much to count for this experiencing; and why count anything at all?
Others have tried to solve the problem by arguing that it is a man's desire for life that
makes us call life a good: if he wants to live then anyone who prolongs his life does him
a benefit. Yet someone may cling to life where we would say confidently that it would
be better for him if he died, and he may admit it too. Speaking of those same conditions
in which, as he said, a bullet would have been merciful, Panin writes,
I should like to pass on my observations concerning the absence of suicides
under the extremely severe conditions of our concentration camps. The more that
life became desperate, the more a prisoner seemed determined to hold onto it. 6

5 Thomas Nagel, ‘Death’, in James Rachels, ed., Moral Problems (New York, 1971), p. 362
6 Panin, Sologdin, p. 85.
 Philippa Foot

One might try to explain this by saying that hope was the ground of this wish to survive
for further days and months in the camp. But there is nothing unintelligible in the idea
that a man might cling to life though he knew those facts about his future which would
make any charitable man wish that he might die.
The problem remains, and it is hard to know where to look for a solution. Is there a
conceptual connexion between life and good? Because life is not always a good we are
apt to reject this idea, and to think that it must be a contingent fact that life is usually a
good, as it is a contingent matter that legacies are usually a benefit, if they are. Yet it
seems not to be a contingent matter that to save someone's life is ordinarily to benefit
him. The problem is to find where the conceptual connexion lies.
It may be good tactics to forget for a time that it is euthanasia we are discussing and to
see how life and good are connected in the case of living beings other than men. Even
plants have things done to them that are harmful or beneficial, and what does them
good must be related in some way to their living and dying. Let us therefore consider
plants and animals, and then come back to human beings. At least we shall get away
from the temptation to think that the connexion between life and benefit must
everywhere be a matter of happiness and unhappiness or of pleasure and pain; the idea
being absurd in the case of animals and impossible even to formulate for plants.
In case anyone thinks that the concept of the beneficial applies only in a secondary or
analogical way to plants, he should be reminded that we speak quite straightforwardly
in saying, for instance, that a certain amount of sunlight is beneficial to most plants.
What is in question here is the habitat in which plants of particular species flourish, but
we can also talk, in a slightly different way, of what does them good, where there is
some suggestion of improvement or remedy. What has the beneficial to do with
sustaining life? It is tempting to answer, ‘everything’, thinking that a healthy condition
just is the one apt to secure survival. In fact, however, what is beneficial to a plant may
have to do with reproduction rather than survival of the individual member of the
species. Nevertheless there is a plain connexion between the beneficial and the life-
sustaining even for the individual plant; if something makes it better able to survive in
conditions normal for that species it is ipso facto good for it. We need go no further, and
could go no further, in explaining why a certain environment or treatment is good for a
plant than to show how it helps this plant to survive. 7
This connexion between the life-sustaining and the beneficial is reasonably
unproblematic, and there is nothing fanciful or zoomorphic in speaking of benefiting or
doing good to plants. A connexion with its survival can make something beneficial to a

7Yet some detail needs to be filled in to explain why we should not say that a scarecrow is beneficial to
the plants it protects. Perhaps what is beneficial must either be a feature of the plant itself, such as
protective prickles, or else must work on the plant directly, such as a line of trees which give it shade.
 III Euthanasia17F

plant. But this is not, of course, to say that we count life as a good to a plant. We may
save its life by giving it what is beneficial; we do not benefit it by saving its life.
A more ramified concept of benefit is used in speaking of animal life. New things can be
said, such as that an animal is better or worse off for something that happened, or that it
was a good or bad thing for it that it did happen. And new things count as benefit. In
the first place, there is comfort, which often is, but need not be, related to health. When
loosening a collar which is too tight for a dog we can say, ‘That will be better for it.’ So
we see that the words ‘better for it’ have two different meanings which we mark when
necessary by a difference of emphasis, saying ‘better for it’ when health is involved. And
secondly an animal can be benefited by having its life saved. ‘Could you do anything
for it?’ can be answered by, ‘Yes, I managed to save its life.’ Sometimes we may
understand this, just as we would for a plant, to mean that we had checked some
disease. But we can also do something for an animal by scaring away its predator. If we
do this, it is a good thing for the animal that we did, unless of course it immediately
meets a more unpleasant end by some other means. Similarly, on the bad side, an
animal may be worse off for our intervention, and this not because it pines or suffers
but simply because it gets killed.
The problem that vexes us when we think about euthanasia comes on the scene at this
point. For if we can do something for an animal—can benefit it—by relieving its
suffering but also by saving its life, where does the greater benefit come when only
death will end pain? It seemed that life was a good in its own right; yet pain seemed to
be an evil with equal status and could therefore make life not a good after all. Is it only
life without pain that is a good when animals are concerned? This does not seem a crazy
suggestion when we are thinking of animals, since unlike human beings they do not
have suffering as part of their normal life. But it is perhaps the idea of ordinary life that
matters here. We would not say that we had done anything for an animal if we had
merely kept it alive, either in an unconscious state or in a condition where, though
conscious, it was unable to operate in an ordinary way; and the fact is that animals in
severe and continuous pain simply do not operate normally. So we do not, on the
whole, have the option of doing the animal good by saving its life though the life would
be a life of pain. No doubt there are borderline cases, but that is no problem. We are not
trying to make new judgements possible, but rather to find the principle of the ones we
do make.
When we reach human life the problems seem even more troublesome. For now we
must take quite new things into account, such as the subject's own view of his life. It is
arguable that this places extra constraints on the solution: might it not be counted as a
necessary condition of life's being a good to a man that he should see it as such? Is there
not some difficulty about the idea that a benefit might be done to him by the saving or
prolonging of his life even though he himself wished for death? Of course he might
have a quite mistaken view of his own prospects, but let us ignore this and think only of
 Philippa Foot

cases where it is life as he knows it that is in question. Can we think that the prolonging
of this life would be a benefit to him even though he would rather have it end than
continue? It seems that this cannot be ruled out. That there is no simple incompatibility
between life as a good and the wish for death is shown by the possibility that a man
should wish himself dead, not for his own sake, but for the sake of someone else. And if
we try to amend the thesis to say that life cannot be a good to one who wishes for his
own sake that he should die, we find the crucial concept slipping through our fingers. As
Bishop Butler pointed out long ago not all ends are either benevolent or self-interested.
Does a man wish for death for his own sake in the relevant sense if, for instance, he
wishes to revenge himself on another by his death? Or what if he is proud and refuses
to stomach dependence or incapacity even though there are many good things left in
life for him? The truth seems to be that the wish for death is sometimes compatible with
life's being a good and sometimes not, which is possible because the description
‘wishing for death’ is one covering diverse states of mind from that of the determined
suicide, pathologically depressed, to that of one who is surprised to find that the
thought of a fatal accident is viewed with relief. On the one hand, a man may see his life
as a burden but go about his business in a more or less ordinary way; on the other hand,
the wish for death may take the form of a rejection of everything that is in life, as it does
in severe depression. It seems reasonable to say that life is not a good to one
permanently in the latter state, and we must return to this topic later on.
When are we to say that life is a good or a benefit to a man? The dilemma that faces us
is this. If we say that life as such is a good we find ourselves refuted by the examples
given at the beginning of this discussion. We therefore incline to think that it is as
bringing good things that life is a good, where it is a good. But if life is a good only
because it is the condition of good things why is it not equally an evil when it brings
bad things? And how can it be a good even when it brings more evil than good?
It should be noted that the problem has here been formulated in terms of the balance of
good and evil, not that of happiness and unhappiness, and that it is not to be solved by
the denial (which may be reasonable enough) that unhappiness is the only evil or
happiness the only good. In this paper no view has been expressed about the nature of
goods other than life itself. The point is that on any view of the goods and evils that life
can contain, it seems that a life with more evil than good could still itself be a good.
It may be useful to review the judgements with which our theory must square. Do we
think that life can be a good to one who suffers a lot of pain? Clearly we do. What about
severely handicapped people; can life be a good to them? Clearly it can be, for even if
someone is almost completely paralysed, perhaps living in an iron lung, perhaps able to
move things only by means of a tube held between his lips, we do not rule him out of
order if he says that some benefactor saved his life. Nor is it different with mental
handicap. There are many fairly severely handicapped people—such as those with
Down's Syndrome—for whom a simple affectionate life is possible. What about
 III Euthanasia17F

senility? Does this break the normal connexion between life and good? Here we must
surely distinguish between forms of senility. Some forms leave a life which we count
someone as better off having than not having, so that a doctor who prolonged it would
benefit the person concerned. With some kinds of senility this is however no longer
true. There are some in geriatric wards who are barely conscious, though they can move
a little and swallow food put into their mouths. To prolong such a state, whether in the
old or in the very severely mentally handicapped is not to do them a service or confer a
benefit. But of course it need not be the reverse: only if there is suffering would one
wish for the sake of the patient that he should die.
It seems, therefore, that merely being alive even without suffering is not a good, and
that we must make a distinction similar to that which we made when animals were our
topic. But how is the line to be drawn in the case of men? What is to count as ordinary
human life in the relevant sense? If it were only the very senile or very ill who were said
not to have this life it might seem right to describe it in terms of operation. But it will be
hard to find the sense in which the men described by Panin were not operating, given
that they dragged themselves out to the forest to work. What is it about the life that the
prisoners were living that makes us put it on the other side of the dividing line from
that of most of the physically or mentally handicapped and of some severely ill or
suffering patients? It is not that they were in captivity, for life in captivity can certainly
be a good. Nor is it merely the unusual nature of their life. In some ways the prisoners
were living more as other men do than the patient in the iron lung.
The idea we need seems to be that of life which is ordinary human life in the following
respect—that it contains a minimum of basic human goods. What is ordinary in human
life—even in very hard lives—is that a man is not driven to work far beyond his
capacity; that he has the support of a family or community; that he can more or less
satisfy his hunger; that he has hopes for the future; that he can lie down to rest at night.
Such things were denied to the men in the Vyatlag camps described by Panin; not even
rest at night was allowed them when they were tormented by bed-bugs, by noise and
stench, and by routines such as body-searches and bath-parades—arranged for the
night time so that work norms would not be reduced. Disease too can so take over a
man's life that the normal human goods disappear. When a patient is so overwhelmed
by pain or nausea that he cannot eat with pleasure, if he can eat at all, and is out of the
reach of even the most loving voice, he no longer has ordinary human life in the sense
in which the words are used here. And we may now pick up a thread from an earlier
part of the discussion by remarking that crippling depression can destroy the
enjoyment of ordinary goods as effectively as external circumstances can remove them.
The suggested solution to the problem is, then, that there is a certain conceptual
connexion between life and good in the case of human beings as in that of animals and
even plants. Here, as there, however, it is not the mere state of being alive that can
determine, or itself count as, a good, but rather life coming up to some standard of
 Philippa Foot

normality. It was argued that it is as part of ordinary life that the elements of good that
a man may have are relevant to the question of whether saving his life counts as
benefiting him. Ordinary human lives, even very hard lives, contain a minimum of
basic goods, but when these are absent the idea of life is no longer linked to that of
good. And since it is in this way that the elements of good contained in a man's life are
relevant to the question of whether he is benefited if his life is preserved, there is no
reason why it should be the balance of good and evil that counts.
It should be added that evils are relevant in one way when, as in the examples
discussed above, they destroy the possibility of ordinary goods, but in a different way
when they invade a life from which the goods are already absent for a different reason.
So, for instance, the connexion between life and good may be broken because
consciousness has sunk to a very low level, as in extreme senility or severe brain
damage. In itself this kind of life seems to be neither good nor evil, but if suffering sets
in one would hope for a speedy end.
This, admittedly inadequate, discussion of the sense in which life is normally a good,
and of the reasons why it may not be so in some particular case, completes the account
of what euthanasia is here taken to be. An act of euthanasia, whether literally act or
rather omission, is attributed to an agent who opts for the death of another because in
his case life seems to be an evil rather than a good. The question now to be asked is
whether acts of euthanasia are ever justifiable. But there are two topics here rather than
one. For it is one thing to say that some acts of euthanasia considered only in
themselves and their results are morally unobjectionable, and another to say that it
would be all right to legalise them. Perhaps the practice of euthanasia would allow too
many abuses, and perhaps there would be too many mistakes. Moreover the practice
might have very important and highly undesirable side effects, because it is unlikely
that we could change our principles about the treatment of the old and the ill without
changing fundamental emotional attitudes and social relations. The topics must,
therefore, be treated separately. In the next part of the discussion, nothing will be said
about the social consequences and possible abuses of the practice of euthanasia, but
only about acts of euthanasia considered in themselves.
What we want to know is whether acts of euthanasia, defined as we have defined them,
are ever morally permissible. To be more accurate, we want to know whether it is ever
sufficient justification of the choice of death for another that death can be counted a
benefit rather than harm, and that this is why the choice is made.
It will be impossible to get a clear view of the area to which this topic belongs without
first marking the distinct grounds on which objection may lie when one man opts for
the death of another. There are two different virtues whose requirements are, in
general, contrary to such actions. An unjustified act of killing, or allowing to die, is
contrary to justice or to charity, or to both virtues, and the moral failings are distinct.
Justice has to do with what men owe each other in the way of non-interference and
 III Euthanasia17F

positive service. When used in this wide sense, which has its history in the doctrine of
the cardinal virtues, justice is not especially connected with, for instance, law courts but
with the whole area of rights, and duties corresponding to rights. Thus murder is one
form of injustice, dishonesty another, and wrongful failure to keep contracts a third;
chicanery in a law court or defrauding someone of his inheritance are simply other
cases of injustice. Justice as such is not directly linked to the good of another, and may
require that something be rendered to him even where it will do him harm, as Hume
pointed out when he remarked that a debt must be paid even to a profligate debauchee
who ‘would rather receive harm than benefit from large possessions’. 8 Charity, on the
other hand, is the virtue which attaches us to the good of others. An act of charity is in
question only where something is not demanded by justice, but a lack of charity and of
justice can be shown where a man is denied something which he both needs and has a
right to: both charity and justice demand that widows and orphans are not defrauded,
and the man who cheats them is neither charitable nor just.
It is easy to see that the two grounds of objection to inducing death are distinct. A
murder is an act of injustice. A culpable failure to come to the aid of someone whose life
is threatened is normally contrary, not to justice, but to charity. But where one man is
under contract, explicit or implicit, to come to the aid of another injustice too will be
shown. Thus injustice may be involved either in an act or an omission, and the same is
true of a lack of charity; charity may demand that someone be aided, but also that an
unkind word not be spoken.
The distinction between charity and justice will turn out to be of first importance when
voluntary and nonvoluntary euthanasia are distinguished later on. This is because of
the connexion between justice and rights, and something should now be said about this.
I believe it is true to say that wherever a man acts unjustly he has infringed a right, since
justice has to do with whatever a man is owed, and whatever he is owed is his as a
matter of right. Something should therefore be said about the different kinds of rights.
The distinction commonly made is between having a right in the sense of having a
liberty, and having a ‘claim-right’ or ‘right of recipience’. 9 The best way to understand
such a distinction seems to be as follows. To say that a man has a right in the sense of
liberty is to say that no one can demand that he do not do the thing which he has the
right to do. The fact that he has a right to do it consists in the fact that a certain kind of
objection does not lie against his doing it. Thus a man has a right in this sense to walk
down a public street or park his car in a public parking space. It does not follow that no

8 David Hume, Treatise, book III, part II, section 1.

9 See, for example, D. D. Raphael, ‘Human Rights Old and New’, in D. D. Raphael, ed., Political Theory and
the Rights of Man (London, 1967) , and Joel Feinberg, ‘The Nature and Value of Rights’, The Journal of Value
Inquiry 4, no. 4 (Winter 1970): 243–57 . Reprinted in Samuel Gorovitz, ed., Moral Problems in Medicine
(Englewood Cliffs, New Jersey, 1976)
 Philippa Foot

one else may prevent him from doing so. If for some reason I want a certain man not to
park in a certain place I may lawfully park there myself or get my friends to do so, thus
preventing him from doing what he has a right (in the sense of a liberty) to do. It is
different, however, with a claim-right. This is the kind of right which I have in addition
to a liberty when, for example, I have a private parking space; now others have duties
in the way of noninterference, as in this case, or of service, as in the case where my
claim-right is to goods or services promised to me. Sometimes one of these rights gives
other people the duty of securing to me that to which I have a right, but at other times
their duty is merely to refrain from interference. If a fall of snow blocks my private
parking space there is normally no obligation for anyone else to clear it away. Claim
rights generate duties; sometimes these duties are duties of noninterference; sometimes
they are duties of service. If your right gives me the duty not to interfere with you I
have ‘no right’ to do it; similarly, if your right gives me the duty to provide something
for you I have ‘no right’ to refuse to do it. What I lack is the right which is a liberty: I am
not ‘at liberty’ to interfere with you or to refuse the service.
Where in this picture does the right to life belong? No doubt people have the right to
live in the sense of a liberty, but what is important is the cluster of claim-rights brought
together under the title of the right to life. The chief of these is, of course, the right to be
free from interferences that threaten life. If other people aim their guns at us or try to
pour poison into our drink we can, to put it mildly, demand that they desist. And then
there are the services we can claim from doctors, health officers, bodyguards, and
firemen; the rights that depend on contract or public arrangement. Perhaps there is no
particular point in saying that the duties these people owe us belong to the right to life;
we might as well say that all the services owed to anyone by tailors, dressmakers, and
couturiers belong to a right called the right to be elegant. But contracts such as those
understood in the patient—doctor relationship come in an important way when we are
discussing the rights and wrongs of euthanasia, and are therefore mentioned here.
Do people have the right to what they need in order to survive, apart from the right
conferred by special contracts into which other people have entered for the supplying of
these necessities? Do people in the underdeveloped countries in which starvation is rife
have the right to the food they so evidently lack? Joel Feinberg, discussing this question,
suggests that they should be said to have ‘a claim’, distinguishing this from a ‘valid
claim’, which gives a claim-right.
The manifesto writers on the other side who seem to identify needs, or at least
basic needs, with what they call ‘human rights’, are more properly described, I
think, as urging upon the world community the moral principle that all human
needs ought to be recognized as claims (in the customary prima facie sense)
worthy of sympathy and serious consideration right now, even though, in many
cases, they cannot yet plausibly be treated as valid claims, that is, as grounds of
any other people's duties. This way of talking avoids the anomaly of ascribing to
 III Euthanasia17F

all human beings now, even those in pre-industrial societies, such ‘economic and
social rights’ as ‘periodic holidays with pay’. 10
This seems reasonable, though we notice that there are some actual rights to service
which are not based on anything like a contract, as for instance the right that children
have to support from their parents and parents to support from their children in old
age, though both sets of rights are to some extent dependent on existing social
arrangements.
Let us now ask how the right to life affects the morality of acts of euthanasia. Are such
acts sometimes or always ruled out by the right to life? This is certainly a possibility; for
although an act of euthanasia is, by our definition, a matter of opting for death for the
good of the one who is to die, there is, as we noted earlier, no simple connexion
between that to which a man has a right and that which is for his good. It is true that
men have the right only to the kind of thing that is, in general, a good: we do not think
that people have the right to garbage or polluted air. Nevertheless, a man may have the
right to something which he himself would be better off without; where rights exist it is
a man's will that counts not his or anyone else's estimate of benefit or harm. So the
duties complementary to the right to life—the general duty of noninterference and the
duty of service incurred by certain persons—are not affected by the quality of a man's
life or by his prospects. Even if it is true that he would be, as we say, ‘better off dead’, so
long as he wants to live this does not justify us in killing him and may not justify us in
deliberately allowing him to die. All of us have the duty of noninterference, and some
of us may have the duty to sustain his life. Suppose, for example, that a retreating army
has to leave behind wounded or exhausted soldiers in the wastes of an arid or
snowbound land where the only prospect is death by starvation or at the hands of an
enemy notoriously cruel. It has often been the practice to accord a merciful bullet to
men in such desperate straits. But suppose one of them demands that he should be left
alive? It seems clear that his comrades have no right to kill him, though it is a quite
different question as to whether they should give him a life-prolonging drug. The right
to life can sometimes give a duty of positive service, but does not do so here. What it
does give is the right to be left alone.
Interestingly enough we have arrived by way of a consideration of the right to life at the
distinction normally labelled ‘active’ versus ‘passive’ euthanasia, and often thought to
be irrelevant to the moral issue. 11 Once it is seen that the right to life is a distinct ground
of objection to certain acts of euthanasia, and that this right creates a duty of
noninterference more widespread than the duties of care there can be no doubt about
the relevance of the distinction between passive and active euthanasia. Where everyone

10Feinberg, ‘Human rights’, Gorovitz, Moral Problems in Medicine, p. 465.

11See, for example, James Rachels, ‘Active and Passive Euthanasia’, New England Journal of Medicine 292,
no. 2 (9 Jan. 1975): 78–80
 Philippa Foot

may have the duty to leave someone alone, it may be that no one has the duty to
maintain his life, or that only some people do.
Where then do the boundaries of the ‘active’ and ‘passive’ lie? In some ways the words
are themselves misleading, because they suggest the difference between act and
omission which is not quite what we want. Certainly the act of shooting someone is the
kind of thing we were talking about under the heading of ‘interference’, and omitting to
give him a drug a case of refusing care. But the act of turning off a respirator should
surely be thought of as no different from the decision not to start it; if doctors had
decided that a patient should be allowed to die, either course of action might follow,
and both should be counted as passive rather than active euthanasia if euthanasia were
in question. The point seems to be that interference in a course of treatment is not the
same as other interference in a man's life, and particularly if the same body of people
are responsible for the treatment and for its discontinuance. In such a case we could
speak of the disconnecting of the apparatus as killing the man, or of the hospital as
allowing him to die. By and large, it is the act of killing that is ruled out under the
heading of noninterference, but not in every case.
Doctors commonly recognise this distinction, and the grounds on which some
philosophers have denied it seem untenable. James Rachels, for instance, believes that if
the difference between active and passive is relevant anywhere, it should be relevant
everywhere, and he has pointed to an example in which it seems to make no difference
which is done. If someone saw a child drowning in a bath it would seem just as bad to
let it drown as to push its head under water. 12 If ‘it makes no difference’ means that one
act would be as iniquitous as the other this is true. It is not that killing is worse than
allowing to die, but that the two are contrary to distinct virtues, which gives the
possibility that in some circumstances one is impermissible and the other permissible.
In the circumstances invented by Rachels, both are wicked: it is contrary to justice to
push the child's head under the water—something one has no right to do. To leave it to
drown is not contrary to justice, but is a particularly glaring example of lack of charity.
Here it makes no practical difference because the requirements of justice and charity
coincide; but in the case of the retreating army they did not: charity would have
required that the wounded soldier be killed had not justice required that he be left
alive. 13 In such a case it makes all the difference whether a man opts for the death of
another in a positive action, or whether he allows him to die. An analogy with the right
to property will make the point clear. If a man owns something he has the right to it
even when its possession does him harm, and normally we have no right to take it from
him. But if one day it should blow away, maybe nothing requires us to get it back for

12Ibid.
13It is not, however, that justice and charity conflict. A man does not lack charity because he refrains from
an act of injustice which would have been for someone's good.
 III Euthanasia17F

him; we could not deprive him of it, but we may allow it to go. This is not to deny that
it will often be an unfriendly act or one based on an arrogant judgement when we
refuse to do what he wants. Nevertheless, we would be within our rights, and it might
be that no moral objection of any kind would lie against our refusal.
It is important to emphasise that a man's rights may stand between us and the action we
would dearly like to take for his sake. They may, of course, also prevent action which
we would like to take for the sake of others, as when it might be tempting to kill one
man to save several. But it is interesting that the limits of allowable interference,
however uncertain, seem stricter in the first case than the second. Perhaps there are no
cases in which it would be all right to kill a man against his will for his own sake unless
they could equally well be described as cases of allowing him to die, as in the example
of turning off the respirator. However, there are circumstances, even if these are very
rare, in which one man's life would justifiably be sacrificed to save others, and ‘killing’
would be the only description of what was being done. For instance, a vehicle which
had gone out of control might be steered from a path on which it would kill more than
one man to a path on which it would kill one. 14 But it would not be permissible to steer
a vehicle towards someone in order to kill him, against his will, for his own good. An
analogy with property rights again illustrates the point. One may not destroy a man's
property against his will on the grounds that he would be better off without it; there are
however circumstances in which it could be destroyed for the sake of others. If his
house is liable to fall and kill him that is his affair; it might, however, without injustice
be destroyed to stop the spread of a fire.
We see then that the distinction between active and passive, important as it is
elsewhere, has a special importance in the area of euthanasia. It should also be clear
why James Rachels' other argument, that it is often ‘more humane’ to kill than to allow
to die, does not show that the distinction between active and passive euthanasia is
morally irrelevant. It might be ‘more humane’ in this sense to deprive a man of
property that brings evil on him, or to refuse to pay what is owed to Hume's profligate
debauchee; but if we say this we must admit that an act which is ‘more humane’ than its
alternative may be morally objectionable because it infringes rights.
So far we have said very little about the right to service as opposed to the right to
noninterference, though it was agreed that both might be brought under the heading of
‘the right to life’. What about the duty to preserve life that may belong to special classes
of persons such as bodyguards, firemen, or doctors? Unlike the general public they are
not within their rights if they merely refrain from interfering and do not try to sustain
life. The subject's claim-rights are two-fold as far as they are concerned and passive as

14For a discussion of such questions, see my article ‘The Problem of Abortion and the Doctrine of the
Double Effect’, Oxford Review, no. 5 (1967); reprinted in Rachels, Moral Problems, and Gorovitz, Moral
Problems in Medicine
 Philippa Foot

well as active euthanasia may be ruled out here if it is against his will. This is not to say
that he has the right to any and every service needed to save or prolong his life; the
rights of other people set limits to what may be demanded, both because they have the
right not to be interfered with and because they may have a competing right to services.
Furthermore one must enquire just what the contract or implicit agreement amounts to
in each case. Firemen and bodyguards presumably have a duty which is simply to
preserve life, within the limits of justice to others and of reasonableness to themselves.
With doctors it may however be different, since their duty relates not only to preserving
life but also to the relief of suffering. It is not clear what a doctor's duties are to his
patient if life can be prolonged only at the cost of suffering or suffering relieved only by
measures that shorten life. George Fletcher has argued that what the doctor is under
contract to do depends on what is generally done, because this is what a patient will
reasonably expect. 15 This seems right. If procedures are part of normal medical practice
then it seems that the patient can demand them however much it may be against his
interest to do so. Once again it is not a matter of what is ‘most humane’.
That the patient's right to life may set limits to permissible acts of euthanasia seems
undeniable. If he does not want to die no one has the right to practise active euthanasia
on him, and passive euthanasia may also be ruled out where he has a right to the
services of doctors or others.
Perhaps few will deny what has so far been said about the impermissibility of acts of
euthanasia, simply because we have so far spoken about the case of one who positively
wants to live, and about his rights; whereas those who advocate euthanasia are usually
thinking either about those who wish to die or about those whose wishes cannot be
ascertained either because they cannot properly be said to have wishes or because, for
one reason or another, we are unable to form a reliable estimate of what they are. The
question that must now be asked is whether the latter type of case, where euthanasia
though not involuntary would again be nonvoluntary, is different from the one
discussed so far. Would we have the right to kill someone for his own good so long as
we had no idea that he positively wished to live? And what about the life-prolonging
duties of doctors in the same circumstances? This is a very difficult problem. On the one
hand, it seems ridiculous to suppose that a man's right to life is something which
generates duties only where he has signalled that he wants to live; as a borrower does
indeed have a duty to return something lent on indefinite loan only if the lender
indicates that he wants it back. On the other hand, it might be argued that there is
something illogical about the idea that a right has been infringed if someone incapable
of saying whether he wants it or not is deprived of something that is doing him harm
rather than good. Yet on the analogy of property we would say that a right has been

15George Fletcher, ‘Legal Aspects of the Decision not to Prolong Life’, Journal of the American Medical
Association 203, no. 1 (1 Jan. 1968): 119–22 . Reprinted in Gorovitz.
 III Euthanasia17F

infringed. Only if someone had earlier told us that in such circumstances he would not
want to keep the thing could we think that his right had been waived. Perhaps if we
could make confident judgements about what anyone in such circumstances would
wish, or what he would have wished beforehand had he considered the matter, we
could agree to consider the right to life as ‘dormant’, needing to be asserted if the
normal duties were to remain. But as things are we cannot make any such assumption;
we simply do not know what most people would want, or would have wanted, us to do
unless they tell us. This is certainly the case so far as active measures to end life are
concerned. Possibly it is different, or will become different, in the matter of being kep
alive, so general is the feeling against using sophisticated procedures on moribund
patients, and so much is this dreaded by people who are old or terminally ill. Once
again the distinction between active and passive euthanasia has come on the scene, but
this time because most people's attitudes to the two are so different. It is just possible
that we might presume, in the absence of specific evidence, that someone would not
wish, beyond a certain point to be kept alive; it is certainly not possible to assume that
he would wish to be killed.
In the last paragraph we have begun to broach the topic of voluntary euthanasia, and
this we must now discuss. What is to be said about the case in which there is no doubt
about someone's wish to die? Either he has told us beforehand that he would wish it in
circumstances such as he is now in, and has shown no sign of a change of mind, or else
he tells us now, being in possession of his faculties and of a steady mind. We should
surely say that the objections previously urged against acts of euthanasia, which it must
be remembered were all on the ground of rights, had disappeared. It does not seem that
one would infringe someone's right to life in killing him with his permission and in fact
at his request. Why should someone not be able to waive his right to life, or rather, as
would be more likely to happen, to cancel some of the duties of non-interference that
this right entails? (He is more likely to say that he should be killed by this man at this
time in this manner, than to say that anyone may kill him at any time and in any way.)
Similarly someone may give permission for the destruction of his property, and request
it. The important thing is that he gives a critical permission, and it seems that this is
enough to cancel the duty normally associated with the right. If someone gives you
permission to destroy his property it can no longer be said that you have no right to do
so, and I do not see why it should not be the same with taking a man's life. An objection
might be made on the ground that only God has the right to take life, but in this paper
religious as opposed to moral arguments are being left aside. Religion apart, there
seems to be no case to be made out for an infringement of rights if a man who wishes to
die is allowed to die or even killed. But of course it does not follow that there is no
moral objection to it. Even with property, which is after all a relatively small matter, one
might be wrong to destroy what one had the right to destroy. For, apart from its value
 Philippa Foot

to other people, it might be valuable to the man who wanted it destroyed, and charity
might require us to hold our hand where justice did not.
Let us review the conclusion of this part of the argument, which has been about
nonvoluntary euthanasia and the right to life. It has been argued that from this side
come stringent restrictions on the acts of euthanasia that could be morally permissible.
Active nonvoluntary euthanasia is ruled out by that part of the right to life which
creates the duty of noninterference, though passive nonvoluntary euthanasia is not
necessarily ruled out, except where the right to life-preserving action has been created
by some special condition such as a contract between a man and his doctor. Voluntary
euthanasia is another matter: as the preceding paragraph suggested, no right is
infringed if a man is allowed to die or even killed at his own request.
Turning now to the other objection that normally holds against inducing the death of
another, that it is against charity, or benevolence, we must tell a very different story.
Charity is the virtue that gives attachment to the good of others, and because life is
normally a good, charity normally demands that it should be saved or prolonged. But
as we so defined an act of euthanasia that it seeks a man's death for his own sake—for
his good—charity will normally speak in favour of it. This is not, of course, to say that
charity can require an act of euthanasia which justice forbids, but if an act of euthanasia
is not contrary to justice—that is, it does not infringe rights—charity will rather be in its
favour than against.
Once more the distinction between nonvoluntary and voluntary euthanasia must be
considered. Could it ever be compatible with charity to seek a man's death although he
wanted to live, or at least had not let us know that he wanted to die? I have argued that
in such circumstances active euthanasia would infringe his right to life, but passive
euthanasia would not do so, unless he had some special right to life-preserving service
from the one who allowed him to die. What would charity dictate? Obviously when a
man wants to live there is a presumption that he will be benefited if his life is
prolonged, and if it is so the question of euthanasia does not arise. But it is, on the other
hand, possible that he wants to live where it would be better for him to die: perhaps he
does not realise the desperate situation he is in, or perhaps he is afraid of dying. So, in
spite of a very proper resistance to refusing to go along with the man's own wishes in
the matter of life and death, someone might justifiably refuse to prolong the life even of
someone who asked him to prolong it, as in the case of refusing to give the wounded
soldier a drug that would keep him alive to meet a terrible end. And it is even more
obvious that charity does not always dictate that life should be prolonged where a
man's own wishes, hypothetical or actual, are not known.
So much for the relation of charity to nonvoluntary passive euthanasia, which was not,
like nonvoluntary active euthanasia, ruled out by the right to life. Let us now ask what
charity has to say about voluntary euthanasia both active and passive. It was suggested
in the discussion of justice that if of sound mind and steady desire a man might give
 III Euthanasia17F

others the right to allow him to die or even to kill him, where otherwise this would be
ruled out. But it was pointed out that this would not settle the question of whether the
act was morally permissible, and it is this that we must now consider. Could not charity
speak against what justice allowed? Indeed it might do so. For while the fact that a man
wants to die suggests that his life is wretched, and while his rejection of life may itself
tend to take the good out of things he might have enjoyed, nevertheless his wish to die
might here be opposed for his own sake just as it might be if suicide were in question.
Perhaps there is hope that his mental condition will improve. Perhaps he is mistaken in
thinking his disease is incurable. Perhaps he wants to die for the sake of someone else
on whom he feels he is a burden, and we are not ready to accept this sacrifice whether
for ourselves or others. In such cases, and there will surely be many of them, it could
not be for his own sake that we kill him or allow him to die, and therefore euthanasia as
defined in this paper would not be in question. But this is not to deny that there could
be acts of voluntary euthanasia both passive and active against which neither justice nor
charity would speak.
We have now considered the morality of euthanasia both voluntary and nonvoluntary,
and active and passive. The conclusion has been that nonvoluntary active euthanasia
(roughly, killing a man against his will or without his consent) is never justified; that is
to say, that a man's being killed for his own good never justifies the act unless he
himself has consented to it. A man's rights are infringed by such an action, and it is
therefore contrary to justice. However, all the other combinations, nonvoluntary passive
euthanasia, voluntary active euthanasia, and voluntary passive euthanasia are
sometimes compatible with both justice and charity. But the strong condition carried in
the definition of euthanasia adopted in this paper must not be forgotten; an act of
euthanasia as here understood is one whose purpose is to benefit the one who dies.
In the light of this discussion let us look at our present practices. Are they good or are
they bad? And what changes might be made, thinking now not only of the morality of
particular acts of euthanasia but also of the indirect effects of instituting different
practices, of the abuses to which they might be subject and of the changes that might
come about if euthanasia became a recognised part of the social scene?
The first thing to notice is that it is wrong to ask whether we should introduce the
practice of euthanasia as if it were not something we already had. In fact we do have it.
For instance it is common, where the medical prognosis is very bad, for doctors to
recommend against measures to prolong life, and particularly where a process of
degeneration producing one medical emergency after another has already set in. If
these doctors are not certainly within their legal rights this is something that is apt to
come as a surprise to them as to the general public. It is also obvious that euthanasia is
often practised where old people are concerned. If someone very old and soon to die is
attacked by a disease that makes his life wretched, doctors do not always come in with
life-prolonging drugs. Perhaps poor patients are more fortunate in this respect than rich
 Philippa Foot

patients, being more often left to die in peace; but it is in any case a well recognised
piece of medical practice, and a form of euthanasia.
No doubt, the case of infants with mental or physical defects will be suggested as
another example of the practice of euthanasia as we already have it, since such infants
are sometimes deliberately allowed to die. That they are deliberately allowed to die is
certain; children with severe spina bifida malformations are not always operated on
even where it is thought that without the operation they will die; and even in the case of
children with Down's Syndrome who have intestinal obstructions the relatively simple
operation that would make it possible to feed them is sometimes not performed. 16
Whether this is euthanasia in our sense or only as the Nazis understood it is another
matter. We must ask the crucial question, ‘Is it for the sake of the child himself that the
doctors and parents choose his death?’ In some cases the answer may really be yes, and
what is more important it may really be true that the kind of life which is a good is not
possible or likely for this child, and that there is little but suffering and frustration in
store for him. 17 But this must presuppose that the medical prognosis is wretchedly bad,
as it may be for some spina bifida children. With children who are born with Down's
Syndrome it is, however, quite different. Most of these are able to live on for quite a
time in a reasonably contented way, remaining like children all their lives but capable of
affectionate relationships and able to play games and perform simple tasks. The fact is,
of course, that the doctors who recommend against life-saving procedures for
handicapped infants are usually thinking not of them but rather of their parents and of
other children in the family or of the ‘burden on society’ if the children survive. So it is
not for their sake but to avoid trouble to others that they are allowed to die. When
brought out into the open this seems unacceptable; at least we do not easily accept the
principle that adults who need special care should be counted as too burden-some to be
kept alive. It must in any case be insisted that if children with Down's Syndrome are
deliberately allowed to die this is not a matter of euthanasia except in Hitler's sense.
And for our children, since we scruple to gas them, not even the manner of their death
is ‘quiet and easy’; when not treated for an intestinal obstruction a baby simply starves
to death. Perhaps some will take this as an argument for allowing active euthanasia, in
which case they will be in the company of an S.S. man stationed in the Warthgenau who
sent Eichmann a memorandum telling him that ‘Jews in the coming winter could no
longer be fed’ and submitting for his consideration a proposal as to whether ‘it would
not be the most humane solution to kill those Jews who were incapable of work through

16 I have been told this by a paediatrician in a well-known medical centre in the United States. It is
confirmed by Anthony M. Shaw and Iris A. Shaw, ‘Dilemma of Informed Consent in Children’, The New
England Journal of Medicine 289, no. 17 (25 Oct. 1973): 885–90 Reprinted in Gorovitz.
17 It must be remembered, however, that many of the social miseries of spina bifida children could be

avoided. Professor R. B. Zachary is surely right to insist on this. See, for example, ‘Ethical and Social
Aspects of Spina Bifida’, The Lancet, 3 Aug. 1968, pp. 274–6 . Reprinted in Gorovitz.
 III Euthanasia17F

some quicker means.’ 18 If we say we are unable to look after children with handicaps we
are no more telling the truth than was the S.S. man who said that the Jews could not be
fed.
Nevertheless if it is ever right to allow deformed children to die because life will be a
misery to them, or not to take measures to prolong for a little the life of a newborn baby
whose life cannot extend beyond a few months of intense medical intervention, there is
a genuine problem about active as opposed to passive euthanasia. There are well-
known cases in which the medical staff has looked on wretchedly while an infant died
slowly from starvation and dehydration because they did not feel able to give a lethal
injection. According to the principles discussed in the earlier part of this paper they
would indeed have had no right to give it, since an infant cannot ask that it should be
done. The only possible solution—supposing that voluntary active euthanasia were to
be legalised—would be to appoint guardians to act on the infant's behalf. In a different
climate of opinion this might not be dangerous, but at present, when people so readily
assume that the life of a handicapped baby is of no value, one would be loath to support
it.
Finally, on the subject of handicapped children, another word should be said about
those with severe mental defects. For them too it might sometimes be right to say that
one would wish for death for their sake. But not even severe mental handicap
automatically brings a child within the scope even of a possible act of euthanasia. If the
level of consciousness is low enough it could not be said that life is a good to them, any
more than in the case of those suffering from extreme senility. Nevertheless if they do
not suffer it will not be an act of euthanasia by which someone opts for their death.
Perhaps charity does not demand that strenuous measures are taken to keep people in
this state alive, but euthanasia does not come into the matter, any more than it does
when someone is, like Karen Ann Quinlan, in a state of permanent coma. Much could
be said about this last case. It might even be suggested that in the case of
unconsciousness this ‘life’ is not the life to which ‘the right to life’ refers. But that is not
our topic here.
What we must consider, even if only briefly, is the possibility that euthanasia, genuine
euthanasia, and not contrary to the requirements of justice or charity, should be
legalised over a wider area. Here we are up against the really serious problem of abuse.
Many people want, and want very badly, to be rid of their elderly relatives and even of
their ailing husbands or wives. Would any safeguards ever be able to stop them
describing as euthanasia what was really for their own benefit? And would it be
possible to prevent the occurrence of acts which were genuinely acts of euthanasia but
morally impermissible because infringing the rights of a patient who wished to live or
whose wishes were unknown?

18 Quoted by Hannah Arendt, Eichmann in Jerusalem (London, 1963), p. 90

 Philippa Foot

Perhaps the furthest we should go is to encourage patients to make their own contracts
with a doctor by making it known whether they wish him to prolong their lives in case
of painful terminal illness or of incapacity. A document such as the Living Will seems
eminently sensible, and should surely be allowed to give a doctor following the
previously expressed wishes of the patient immunity from legal proceedings by
relatives. 19 Legalising active euthanasia is, however, another matter. Apart from the
special repugnance doctors feel towards the idea of a lethal injection, it may be of the
very greatest importance to keep a psychological barrier up against killing. Moreover it
is active euthanasia which is the most liable to abuse. Hitler would not have been able
to kill 275,000 people in his ‘euthanasia’ programme if he had to wait for them to need
life-saving treatment. But there are other objections to active euthanasia, even voluntary
active euthanasia. In the first place it would be hard to devise procedures that would
protect people from being persuaded into giving their consent. And secondly the
possibility of active voluntary euthanasia might change the social scene in ways that
would be very bad. As things are, people do, by and large, expect to be looked after if
they are old or ill. This is one of the good things that we have, but we might lose it, and
be much worse off without it. It might come to be expected that someone likely to need
a lot of looking after should call for the doctor and demand his own death. Something
comparable could be good in an extremely poverty-stricken community where the
children genuinely suffered from lack of food; but in rich societies such as ours it would
surely be a spiritual disaster. Such possibilities should make us very wary of supporting
large measures of euthanasia, even where moral principle applied to the individual act
does not rule it out.

19Details of this document are to be found in J. A. Behnke and Sissela Bok, eds., The Dilemmas of
Euthanasia (New York, 1975) , and in A. B. Downing, ed., Euthanasia and the Right to Life: The Case for
Voluntary Euthanasia (London, 1969)
 IV Free Will as Involving Determinism36F

IV Free Will as Involving Determinism 1

Philippa Foot
Abstract: There is a widely accepted view that free will is compatible with determinism, and some have
even argued that free will implies determinism. The arguments of R. E. Hobart, David Hume, and A. J.
Ayer, which support the latter view, are analysed and rejected.

The idea that free will can be reconciled with the strictest determinism is now very
widely accepted. To say that a man acted freely is, it is often suggested, to say that he
was not constrained, or that he could have done otherwise if he had chosen, or
something else of that kind; and since these things could be true even if his action was
determined it seems that there could be room for free will even within a universe
completely subject to causal laws. Hume put forward a view of this kind in contrasting
the ‘liberty of spontaneity . . . which is oppos'd to violence’ with the nonexistent ‘liberty
of indifference . . . which means a negation of necessity and causes’. 2 A. J. Ayer, in his
essay ‘Freedom and Necessity’ 3 was summing up such a position when he said, ‘from
the fact that my action is causally determined . . . it does not necessarily follow that I am
not free’4 and ‘it is not when my action has any cause at all, but only when it has a
special sort of cause, that it is reckoned not to be free’. 5
I am not here concerned with the merits of this view but only with a theory which
appears more or less incidentally in the writings of those who defend it. This is the
argument that so far from being incompatible with determinism, free will actually
requires it. It appears briefly in Hume's Treatise and was set out in full in an article by R.
E. Hobart.6 P. H. Nowell-Smith was expressing a similar view when he said of the idea
that determinism is opposed to free will that ‘the clearest proof that it is mistaken or at
least muddled lies in showing that I could not be free to choose what I do unless
determinism is correct. . . . Freedom, so far from being incompatible with causality
implies it’. 7 Ayer has taken up a similar position, arguing that the absence of causal
laws governing action ‘does not give the moralist what he wants. For he is anxious to
show that men are capable of acting freely in order to infer that they can be morally

1 ‘Free Will as Involving Determinism’ originally appeared in The Philosophical Review, Volume LXVI,
Number 4, October 1957.
2 Treatise, bk. II, pt. III, sec. 2.

3 Polemic, no. 5 (1946) ; reprinted in his Philosophical Essays (London, 1954)

4 Philosophical Essays, p. 278.

5 Ibid., p. 281.

6 ‘Free will as Involving Determinism’, Mind, XLIII (1934), 1–27

7 ‘Free will and Moral Responsibility’, Mind, LVII (1948), 46

 Philippa Foot

responsible for what they do. But if it is a matter of pure chance that a man should act in
one way rather than another, he may be free but he can hardly be responsible.’ 8
This argument is not essential to the main thesis of those who use it; their own account
of free will in such terms as the absence of constraining causes might be correct even
though there were no inconsistencies in the suggestion put forward by their libertarian
opponents. But if valid the argument would be a strong argument, disposing of the
position of anyone who argued both that free will required the absence of determining
causes and that free will was a possibility. That the argument is not valid, and indeed
that it is singularly implausible, I shall now try to show. It is, I think, surprising that it
should have survived so long; this is perhaps because it has not had to bear much
weight. In any case the weapons which can be used against it are ones which are in
general use elsewhere.
In discussing determinism and free will it is important to be clear about the sense which
is given in this context to words such as ‘determined’ and ‘caused’. Russell gave this
account:
The law of universal causation . . . may be enunciated as follows: There are such
invariable relations between different events at the same or different times that, given
the state of the whole universe throughout any finite time, however short, every
previous and subsequent event can theoretically be determined as a function of the
given events during that time. 9
This seems to be the kind of determinism which worries the defender of free will, for if
human action is subject to a universal law of causation of this type, there will be for any
action a set of sufficient conditions which can be traced back to factors outside the
control of the agent.
We cannot of course take it for granted that whenever the word ‘determined’ or the
word ‘cause’ is used this is what is implied, and what is intended may be in no way
relevant to the question of free will. For instance, an action said to be determined by the
desires of the man who does it is not necessarily an action for which there is supposed
to be a sufficient condition.
In saying that it is determined by his desires we may mean merely that he is doing
something that he wants to do, or that he is doing it for the sake of something else he
wants. There is nothing in this to suggest determinism in Russell's sense. On the whole
it is wise to be suspicious of expressions such as ‘determined by desire’ unless these
have been given a clear sense, and this is particularly true of the phrase ‘determined by
the agent's character’. Philosophers often talk about actions being determined by a
man's character, but it is not certain that anyone else does, or that the words are given
any definite sense. One might suppose that an action was so determined if it was in

8 Philosophical Essays, p. 275.

9 ‘On the Notion of Cause’, in Our Knowledge of the External World (London, 1914), p. 221
 IV Free Will as Involving Determinism36F

character, for instance the generous action of a generous man; but if this is so we will
not have the kind of determinism traditionally supposed to raise difficulties for a
doctrine of free will. For nothing has been said to suggest that where the character trait
can be predicated the action will invariably follow; it has not been supposed that a man
who can truly be said to be generous never acts ungenerously even under given
conditions.
Keeping the relevant sense of ‘determinism’ in mind, we may now start to discuss the
view that free will requires determinism. The first version which I shall consider is that
put forward by R. B. Hobart, who suggests that an action which is not determined
cannot properly be called an action at all, being something that happened to the agent
rather than something he did. Hobart says, ‘In proportion as it [the action] is
undetermined, it is just as if his legs should suddenly spring up and carry him off
where he did not prefer to go.’ To see how odd this suggestion is we have only to ask
when we would say that a man's legs were carrying him where he did not prefer to go.
One can imagine the scene: he is sitting quietly in his chair and has said that he is going
to go on reading his book; suddenly he cries, ‘Good heavens, I can't control my legs!’
and as he moves across the room, he hangs on to the furniture or asks someone else to
hold him. Here indeed his legs are carrying him where he does not want to go, but what
has this to do with indeterminism, and what has the ordinary case, where he walks
across the room, to do with determinism? Perhaps Hobart thinks that when a man does
something meaning to do it, he does what he wants to do, and so his action is
determined by his desire. But to do something meaning to do it is to do it in a certain
way, not to do it as the result of the operation of a causal law. When one means to do
something, one does not call out for help in preventing the movement of one's limbs; on
the contrary, one is likely to stop other people from interfering, saying, ‘I want to do
this’. It is by such factors that walking across the room is distinguished from being
carried off by one's legs. It is to be explained in terms of the things said and done by the
agent, not in terms of some force, ‘the desire’, present before the action was done and
still less in terms of some law telling us that whenever this ‘desire’ is found it will be
followed by the action. The indeterminist has no difficulty in distinguishing an action
from something that happens to the agent; he can say exactly the same as anyone else.
Nowell-Smith seems to be thinking along somewhat the same lines as Hobart when he
attacks C. A. Campbell for saying that free will requires indeterminism:
The essence of Campbell's account is that the action should not be predictable from a
knowledge of the agent's character. But, if this is so, can what he does be called his
action at all? Is it not rather a lusus naturae, an Act of God or a miracle? If a hardened
criminal, bent on robbing the poor-box, suddenly and inexplicably fails to do so, we
should not say that he chose to resist or deserves credit for resisting the temptation; we
should say, if we were religious, that he was the recipient of a sudden outpouring of
Divine Grace or, if we were irreligious, that his ‘action’ was due to chance, which is
 Philippa Foot

another way of saying that it was inexplicable. In either case we should refuse to use the
active voice. 10
It is hard to see why a man who does something inexplicably does not really do it. Let
us suppose that the hardened criminal's action really is inexplicable; we can only say,
‘He just turned away’, and not why he did so; this does not mean that he did it by
accident, or unintentionally, or not of his own free will, and I see no reason for refusing
to use the active voice. In any case, to explain an action is not necessarily to show that it
could have been predicted from some fact about the agent's character—that he is weak,
greedy, sentimental, and so forth. We may if we like say that an action is never fully
explained unless it has been shown to be covered by a law which connects it to such a
character trait; but then it becomes even more implausible to say that an action must be
explicable if we are to admit it as something genuinely done. In the ordinary sense we
explain the criminal's action if we say, for instance, that a particular thought came into
his mind; we do not also have to find a law about the way such thoughts do come into
the minds of such men.
A rather different version of this argument runs as follows. We hold responsible only a
man who is a rational agent; if someone were always to do things out of the blue,
without having any reason to do them, we should be inclined to count him as a lunatic,
one who could not be held responsible for his actions, so that even if he did things he
would do things for which he could not be held responsible. And is it not through being
determined by motives that actions are those of a rational agent whom we can praise or
blame?
It certainly would be odd to suppose that free will required the absence of motives for
action. We do not of course expect that everything that the rational man does should be
done with a motive; if he gets up and walks about the room he need not be doing so in
order to take exercise; it is quite usual for people to do this kind of thing without any
particular purpose in view, and no one is counted irrational for doing so. And yet we
do expect a man to have a motive for a great number of things that he does, and we
would count anyone who constantly performed troublesome actions without a motive
as irrational. So it looks as if a moral agent is a man whose actions are in general
determined, if determinism is involved in ‘having a motive’ for what he does.
What does it mean to say that someone had a motive for doing what he did? Often this
particular expression means that he did it with a particular intention, so we should first
say something about intentions and the sense in which they can be said to determine
action. We say that a man had a certain intention in acting when he aimed at a certain
thing, and ‘his motive for such and such’ often means ‘his aim in doing such and such’,
for instance, ‘His motive for going to the station was to take a train to London’. But
where motives are intentions it is clear that they cannot be determining causes; for

10 Ethics (London, 1954), pp. 281–2

 IV Free Will as Involving Determinism36F

intending to do x and being ready to take the steps thought necessary to do x are
connected not empirically but analytically. A man cannot be said to have an intention
unless he is reconciled to what he believes to be the intermediate steps. We cannot
speak as if the intention were something which could be determined first, and ‘being
ready to take the necessary steps’ were a second stage following on the first.
It might be objected that this does not cover the case of ‘doing y because one wants x’
where ‘wanting x’ does not imply trying to get x. In one sense of ‘want’ it is possible to
say, ‘He wants x’ without knowing whether he is prepared to take steps to get it. (One
might, for instance, want to go to London but not be prepared to spend the money to
get there.) So that wanting seems here to be a separate condition, which might in certain
cases be connected by an empirical law to the adoption of particular courses of action.
Certainly wanting is not an event, but one gets rid of wanting as a determining factor
too easily if one merely says that desires are not causes because they are not
occurrences.
We say ‘He wants’ in this sense where he would adopt certain policies if there were no
reasons for not doing so. We can say, ‘He wants to get to London’, even when he is not
prepared to take the necessary steps to get to London, provided he can say, ‘Trains are
too expensive’, or ‘Hitchhiking is too uncomfortable’. If we offered him a spare railway
ticket or otherwise disposed of his reasons against going, and he still did not go, we
would have to say, ‘He didn't really want to go after all’. So wanting in this sense is
being prepared to act under certain conditions, though not being prepared to act under
the given conditions. It is a description which could be applied to a man before we
knew whether he was ready to act in a given situation, and it seems that there might
then be a causal relation between the wanting and the acting where the latter took
place. This is quite true; there could be a law to the effect that when the description ‘He
wants x’ applied at t 1 , the description ‘He is taking the necessary steps to get x’ applied
at t 2 . It would be possible to say this without making a mistake about what it is to want
and inventing a hidden condition of body or mind. One could say, ‘Wanting in this
sense just is being prepared to act under some conditions’, and still maintain that there
could be an empirical law connecting wanting with acting under a particular set of
conditions. The mistake lies not in the idea that such laws are possible but in the thought
that there is a reference to them in the statement that a man did one thing because he
wanted something else.
So far we have been dealing only with cases in which a question about a motive was
answered by specifying something aimed at or wanted. Now we should turn to the
cases in which the motive is said to be kindness, vanity, ambition, meanness, jealousy,
and so on, to see whether determinism is involved.
It is easy to show that a motive is not a cause in Russell's sense, for it is clearly not an
antecedent cause. Professor Gilbert Ryle has pointed out that a man who acts out of
vanity is not a man who had a feeling of vanity immediately before he acted, and if it is
 Philippa Foot

objected that the vanity which preceded the action need not have manifested itself in a
feeling, one may ask what else would count as the vanity which was causing him to act.
A man's motives are not given by what was happening to him immediately before he
started to act. Nor do we discover some independent condition contemporaneous with
the action and a law linking the two, for again there is nothing which would count as
vanity except the tendency to do this kind of thing.
So much is implied in what Ryle says about acting out of vanity, but his own account of
what it is to do so still uses something which is objectionably like a causal model. The
analogy which he thinks apt is that between saying a man acted out of vanity and
saying a piece of glass broke because it was brittle: ‘To explain an act as done from a
certain motive is not analogous to saying that the glass broke because a stone hit it, but
to the quite different type of statement that the glass broke, when the stone hit it,
because the glass was brittle.’ 11 The positive part of this statement seems to me
mistaken. Acting out of vanity is not so closely connected with being vain as Ryle must
suppose it to be. Let us suppose that his account of what it is to be vain is entirely
correct; to say that a man is vain is to say that he tends to behave in certain ways, to feel
aggrieved in particular situations, and so on. 12 It does not follow that ascribing vanity as
a motive for an action is bringing this action under the ‘lawlike’ proposition that the
agent is a man who tends to do these things. For it makes sense to say that a man acts
out of vanity on a particular occasion although he is not in general vain, or even vain
about this kind of thing. It cannot therefore be true that when we speak of an agent's
motive for a particular action we are explaining it in terms of his character, as Ryle
suggests; we are not saying ‘he would do that’. It is of course possible to give a motive
and to say that the agent has the character trait concerned, but the latter cannot be
included in an account of what it is to assign a motive to a particular action.
The explanation of why Ryle says what he does seems to lie in the fact that he has taken
a false example of explaining an action by giving a motive. He considers as his example
the explanation, ‘He boasted because he is vain’, which is not in fact an explanation of
the right type; considered as a statement assigning a motive to a particular action it
would be uninformative, for except in very special cases boasting is acting out of vanity.
It is not surprising that this particular sentence has a different function—that of relating
this act of vanity to the character trait. What Ryle says about the example is correct, but
it is not an example of the kind of thing he is trying to describe.
It might seem as if we could reformulate the theory to meet the objection about the man
who acts out of vanity on one occasion by saying that a man's acting out of vanity is like
glass breaking because of a brittleness which could be temporary. ‘He acted out of
vanity’ would then be explained as meaning that at that particular time he tended to

11 Concept of Mind (London, 1949), pp. 86–7

12 Ibid., p. 86.
 IV Free Will as Involving Determinism36F

react in the ways described by Ryle. (If he finds a chance of securing admiration and
envy of others, he does whatever he thinks will produce this admiration and envy.) This
is wrong because, whereas glass which is even temporarily brittle has all the reactions
which go by this name, a man who is temporarily acting out of vanity is not liable to do
other things of this kind. To find concepts which this model would fit one must turn to
such descriptions as ‘a boastful mood’, ‘a savage frame of mind’, or ‘a fit of bad temper’.
Assigning a motive to an action is not bringing it under any law; it is rather saying
something about the kind of action it was, the direction in which it was tending, or
what it was done as. A possible comparison would be with the explanation of a
movement in a dance which consisted in saying what was being danced. Often in
diagnosing motives we should look to purposes—to what the action was done for. This
we should discover if we found out what the agent was prepared to go without and
what he insisted on having; the fact that visitors are made to admire a garden even in
the rain is strong evidence that they were invited out of vanity rather than kindness. In
other cases finding the motive will be better described as finding what was being
done—finding, for instance, that someone was taking revenge. We should take it that a
man's motive was revenge if we discovered that he was intentionally harming someone
and that his doing so was conditional on his believing that person had injured him. In
the same way we should take it that someone was acting out of gratitude if he (1)
intended to confer a benefit and (2) saw this as called for by a past kindness. The fact
that it is only the character of the particular action which is involved shows how far we
are from anything which could involve motives as determining causes.
We have now considered two suggestions: (1) that an undetermined action would not
be one which could properly be attributed to an agent as something that he did and (2)
that an undetermined action would not be the action of a rational agent. A third version,
the one put forward by Hume, suggests that an undetermined action would be one for
which it would be impossible to praise or blame, punish or reward a man, because it
would be connected with nothing permanent in his nature.
'Tis only [Hume says] upon the principles of necessity, that a person acquires any merit
or demerit from his actions. . . . Actions are by their very nature temporary and
perishing; and where they proceed not from some cause in the characters and
disposition of the person, who perform'd them, they infix not themselves upon him,
and can neither redound to his honour, if good, nor infamy, if evil. The action in itself
may be blameable. . . . But the person is not responsible for it; and as it proceeded from
nothing in him, that is durable and constant, and leaves nothing of that nature behind
it, 'tis impossible he can, upon its account, become the object of punishment or
vengeance. 13

13 Treatise, bk. II, pt. III, sec 2.

 Philippa Foot

Hume is surely wrong in saying that we could not praise or blame, punish or reward, a
person in whose character there was nothing ‘permanent or durable’. As he was the first
to point out, we do not need any unchanging element in order to say that a person is the
same person throughout a period of time, and our concept of merit is framed to fit our
concept of personal identity. We honour people as well as nations for what they have
done in the past and do not consider what has been done merely as an indication of
what may be expected in the future. Moreover, it is perfectly rational to punish people
for what they have done, even if there is no reason to think that they would be likely to
do it again. The argument that it will be a different me who will be beaten tomorrow
carries no weight, for ‘different’ or not the back which will be beaten is the one about
which I am concerned today. So we have no reason to invent something durable and
constant underlying the actions which we punish or reward. And it is not in fact our
practice to pick out for praise or blame only those actions for which something of the
kind can be found. It would be possible, of course, that we should do this, punishing
the cruel action of the cruel man but not that of one usually kind. But even in such a
situation there would be no argument against the man who said that moral
responsibility depended upon indeterminism; for a motive is not a determining cause,
nor is an habitual motive. If we say that a man constantly acts out of cruelty, we no
more say that his actions are determined than if we say that he acts out of cruelty on a
particular occasion. There could of course be a law to the effect that no one who has
been cruel for thirty years can turn to kindness after that, and this would throw
responsibility back from the later to the earlier acts. But it is clear that this is a special
assumption in no way involved in the statement that cruelty is a ‘durable and constant’
element in someone's character.
I have already mentioned Ayer's argument that moral responsibility cannot be
defended on the basis of indeterminism and will now consider his version in detail.
Ayer says that the absence of a cause will not give the moralist what he wants, because
‘if it is a matter of pure chance that a man should act in one way rather than another, he
may be free but he can hardly be responsible’. 14 To the suggestion that ‘my actions are
the result of my own free choice’, Ayer will reply with a question about how I came to
make my choice:
Either it is an accident that I choose to act as I do or it is not. If it is an accident,
then it is merely a matter of chance that I did not choose otherwise; and if it is
merely a matter of chance that I did not choose otherwise, it is surely irrational to
hold me morally responsible for choosing as I did. But if it is not an accident that
I chose to do one thing rather than another, then presumably there is some causal
explanation in my choice: and in that case we are led back to determinism. 15

14 Philosophical Essays, p. 275.

15 Ibid.
 IV Free Will as Involving Determinism36F

The ‘presumably’ seems to be the weak link in the argument, which assumes a
straightforward opposition between causality and chance that does not in general exist.
It is not at all clear that when actions or choices are called ‘chance’ or ‘accidental’ this
has anything to do with the absence of causes, and if it has not we will not be saying
that they are in the ordinary sense a matter of chance if we say that they are
undetermined.
When should we say that it was a matter of chance that a man did what he did? A
typical example would be the case in which a man killed someone with a bullet which
glanced off some object in a totally unforseeable way; here he could disclaim
responsibility for the act. But in this instance, and that of something done ‘by accident’,
we are dealing with what is done unintentionally, and this is not the case which Ayer
has in mind. We may turn, as he does, to the actions which could be said to have been
‘chosen’ and ask how the words ‘chance’ and ‘accident’ apply to choices. Ayer says,
‘Either it is an accident that I choose to act as I do, or it is not’. The notion of choosing by
accident to do something is on the face of it puzzling; for usually choosing to do
something is opposed to doing it by accident. What does it mean to say that the choice
itself was accidental? The only application I can think of for the words ‘I chose by
accident’ is in a case such as the following. I choose a firm with which to have dealings
without meaning to pick on one run by an international crook. I can now rebut the
charge of choosing a firm run by an international crook by saying that I chose it by accident.
I cannot be held responsible for this but only for any carelessness which may have been
involved. But this is because the relevant action—the one with which I am being
charged—was unintentional; it is for this reason and not because my action was
uncaused that I can rebut the charge. Nothing is said about my action being uncaused,
and if it were, this could not be argued on my behalf; the absence of causes would not
give me the same right to make the excuse.
Nor does it make any difference if we substitute ‘chance’ for accident'. If I say that it
was a matter of chance that I chose to do something, I rebut the suggestion that I chose
it for this reason or for that, and this can be a plea against an accusation which has to do
with my reasons. But I do not imply that there was no reason for my doing what I did,
and I say nothing whatsoever about my choice being undetermined. If we use ‘chance’
and ‘accident’ as Ayer wants to use them, to signify the absence of causes, we shall have
moved over to a totally different sense of the words, and ‘I chose it by chance’ can no
longer be used to disclaim responsibility.

Additional note: The work by R. B. Hobart referred to on p. 64 is ‘Freewill as Involving Determinism and
Inconceivable Without It’, Mind, LXIII (1934), pp. 1–27.
 Philippa Foot

V Hume on Moral Judgement 1

Philippa Foot
Abstract: David Hume's view of virtue as agreeable and useful differs from the views of philosophers
such as Immanuel Kant and Jean Jacques Rousseau who see virtue as sublime and noble. Hume does not
differentiate between virtues, skills, and talents that, as such, all arouse approbation or pleasing
sentiments in others. Hume's view commits him to subjectivism in his theory of ethics.

Some philosophers talk about morality in an elevated tone; and they seem to be entirely
sincere, finding virtue a sublime and noble subject, the pursuit of virtue an inspiring
life's work. So it is, for instance, with Kant, who writes in one place about the moral law
within and the starry heavens above filling the mind with ever increasing awe and
admiration, the oftener and more steadily we reflect on them. It came quite naturally to
Kant, as it did to Rousseau, to talk about the sublimity of our nature in its higher aspect,
and of reverence for the moral law. ‘Duty!’ he says. ‘Sublime and mighty name.’ But
Hume speaks with a very different voice.
And as every quality which is useful or agreeable to ourselves or others is, in common
life, allowed to be a part of personal merit; so no other will ever be received, where men
judge of things by their natural, unprejudiced reason, without the delusive glosses of
superstition and false religion. Celibacy, fasting, penance, mortification, self-denial,
humility, silence, solitude, and the whole train of monkish virtues; for what reason are
they everywhere rejected by men of sense, but because they serve to no manner of
purpose; neither advance a man's fortune in the world, nor render him a more valuable
member of society; neither qualify him for entertainment of company, nor increase his
power of self-enjoyment? We observe, on the contrary, that they cross all these desirable
ends; stupify the understanding and harden the heart, obscure the fancy and sour the
temper. We justly, therefore, transfer them to the opposite column, and place them in
the catalogue of vices; nor has any superstition force enough among men of the world,
to pervert entirely these natural sentiments. A gloomy, hare-brained enthusiast, after
his death, may have a place in the calendar; but will scarcely ever be admitted, when
alive, into intimacy and society, except by those who are as delirious and dismal as
himself. 2
Where Kant, or Rousseau, close to Hume in time, but at a great distance in mental
space, saw virtue as inspiring, Hume found it useful and agreeable, fitting a man for
business and society. Indeed he actually identified a sense of virtue with a pleasing
sentiment of approbation, which, he thought, men find within themselves on the

1 ‘Hume on Moral Judgement’ originally appeared in David Pears, ed., David Hume (London, 1963)
2 An Enquiry Concerning The Principles of Morals, sec. IX, pt. I.
 V Hume on Moral Judgement51F

contemplation of certain actions and qualities of mind. He defines virtue to be

‘whatever mental action or quality gives to a spectator the pleasing sentiment of
approbation; and vice the contrary’. Moreover, enquiring into the common
characteristic of those actions or qualities which have this pleasant effect upon the
spectator, he decides that they are all agreeable, or useful, to ourselves or other men. So
a sense of virtue is itself a kind of pleasure, and this pleasure arises on the
contemplation of what is useful or agreeable to mankind.
Hume's account of the common characteristics of the qualities called virtues is, it must
be said, bad. It leads him to class with such things as honesty, justice, benevolence, and
courage, not only cleanliness, which ‘naturally renders us agreeable to others, and is a
very considerable source of love and affection’, but also such things as wit and
eloquence. One does not find in Hume an account of the difference between skills or
talents and virtues and he even says that there is no reason to consider virtue as
something distinct. I suppose it is partly due to Hume's influence that this important
topic, which was splendidly treated by Aristotle and Aquinas, is hardly discussed by
modern moral philosophers.
But there is a much more serious charge to be laid at Hume's door. Even supposing that
he had been right in saying that the things which we call virtues are the qualities useful
and agreeable to ourselves and others—if these were their common characteristics—
Hume's account of the status of the proposition ‘virtues are qualities useful or agreeable,
etc.’ might still be attacked. For he seems to think that we find out by observation that
these are the qualities which happen to arouse in a spectator the pleasing sentiment of
approbation called the sense of virtue. We find that, as a matter of fact, people do feel
the peculiar sentiment of approbation when they contemplate just these actions and
qualities: but it might have been otherwise. He speaks as if we first identify the special
sentiment of approbation, moral approval, and then look around to see what can be
said about the things that arouse this feeling in us, thus implying that there is no
difficulty in discovering that people are feeling approval before it is known what beliefs
they hold about the things in question. Mr. Gardiner pointed out in his essay 3 that
Hume held a similar theory about the identification of the internal sensation of pride,
and objected that one could not really say that it was pride that someone was feeling
unless he had the right thoughts about the thing of which he was supposed to feel
proud. (He must, I suppose, see it as something like an achievement, and as in some
way related to himself.) A feeling of pride is not identified like a tickle, but requires a
special kind of thought about the thing of which one feels proud. Now I should say,
though I do not know whether Mr. Gardiner would, that it is just as bad to try to
identify a feeling as a feeling of approval, whether moral approval or any other, without
its particular objects as it is to try to identify pride without talking about the only kinds

3 Patrick Gardiner, ‘Hume's Theory of the Passions’: David Pears (ed.), David Hume.
 Philippa Foot

of things about which one can logically feel proud. (I do not mean, of course, that one
would be illogical in feeling pride towards something which one did not believe to be in
some way splendid and in some way one's own, but that the concept of pride does not
allow us to talk like that.) Similarly for the concept of approval, though the reader will
kindly excuse me from giving an account of what exactly a man must believe of those
things of which he can logically approve. Anyone who doubts this point about approval
should ask what it would be to have this feeling when contemplating an object one did
not see as useful, beautiful, efficient or anything like that. Does it make sense to suppose
that one might wake up one morning feeling approval of something believed to be an
ordinary, unnecessary, unbeautiful speck of dust? Hume was, I think, making a mistake
when he tried to explain what it meant to say that an action or quality was virtuous in
terms of a special feeling; for the explanation of the thought comes into the description
of the feeling, not the other way round.
Now this theory of Hume's about moral sentiment commits him to a subjectivist theory
of ethics. He could not consistently maintain both that a man calls qualities virtues
when he happens to feel towards them this peculiar sentiment, and that statements
about virtue and vice are objective. For if they were objective, like ordinary statements
of fact, there would have to be some method of deciding, in case of disagreement,
whether one man's opinion or another's was correct—as the opinion that the earth is flat
can be shown to be mistaken by a voyage round the globe. But since Hume has denied
all logical connexions—all connexions of meaning—between moral approval and the
objects of moral approval, and would have to allow anyone to assert any kinds of
actions he chose to be virtuous on the strength of the supposed feeling of approbation, it
follows that no one could get at an opponent who professed weird ‘moral views'. And
Hume himself, though he sometimes modifies his theory and talks about the sentiments
of the majority, in most places accepts this subjectivism with ease, and even with relish.
Take any action allowed to be vicious: Wilful murder for instance. Examine it in all
lights, and see if you can find that matter of fact, or real existence, which you call vice. In
whichever way you take it, you find only certain passions, motives, volitions and
thoughts. There is no other matter of fact in the case. The vice entirely escapes you, as
long as you consider the object. You can never find it, till you turn your reflexion into
your own breast, and find a sentiment of disapprobation, which arises in you, towards
this action. Here is a matter of fact; but ’tis the object of feeling, not of reason. It lies in
yourself, not in the object.4
Such a theory does not look at all plausible. We are not inclined to think that when a
man says that an action is virtuous, or vicious, he is talking about his own feelings
rather than a quality which he must show really to belong to what is done. It seems
strange to suggest that he does not have to bring forward any special fact about the action

4 Treatise, bk. III, pt. I, sec. 2.

 V Hume on Moral Judgement51F

in order to maintain what he says. So, of course, what one wants to know is why Hume
adopted this strange theory: what drove him to say that the virtuousness of an action
could not be a plain provable matter of fact. Sometimes, indeed, he suggests that one
simply cannot find any such fact, but since he himself claims that all things called
virtues are qualities agreeable or useful to mankind why should he not say that it is in
this that their virtuousness consists? Why does he have to bring in his feeling of
approbation, and by making this the essential part of moral judgement anchor
statements about virtue to the sentiments of the observer and not to the facts?
Hume would reply that one must distinguish a judgement about morality from the
whole class of what he calls ‘conclusions of reason’—that is, ordinary provable
propositions about what is the case. For, as he says, no such factual proposition could
ever have a necessary connexion with the will of the man who accepted it, while it is an
essential fact about a moral judgement that it does have this practical force.
‘Morals excite passions, and produce or prevent actions. Reason of itself is utterly
impotent in this particular. The rules of morality, therefore, are not conclusions of our
reason.’ Hume, who is not usually repetitious, says this kind of thing over and over
again. ‘As long as it is allowed, that reason has no influence on our passions and
actions, 'tis in vain to pretend, that morality is discover'd only by a deduction of reason.’
‘Reason is wholly inactive and can never be the source of so active a principle as
conscience, or a sense of morals.’
What does Hume mean when he says that reason is wholly inactive? He argues that
whenever a man is led by a judgement of reason to some action it is not reason alone
which impels him, but reason with the co-operation of desire. Reason can tell us, for
instance, that a certain action will have a particular effect, or, again, that a certain object
exists within our reach; but if the effect is indifferent to us, the thing not wanted, the
discovery of fact which Hume calls ‘a conclusion of reason’ will make no difference to
what we do. And so, he says, reason alone is ‘perfectly inert’; it may of course influence
actions, but only when we happen to have certain desires. Thus conclusions of reason
have a merely contingent connexion with action, whereas the propositions of morality
are necessarily practical, going beyond the ‘calm and indolent judgements of the
understanding’. Between these calm and indolent judgements and the assertion that
something should be done there is, Hume thinks, the famous gap between is and ought.
I cannot forbear adding to these reasonings an observation, which may, perhaps, be
found of some importance. In every system of morality, which I have hitherto met with,
I have always remarked, that the author proceeds for some time in the ordinary way of
reasoning, and establishes the being of a God, or makes observations concerning human
affairs; when of a sudden I am surpriz'd to find, that instead of the usual copulations of
propositions, is, and is not, I meet with no proposition that is not connected with an
ought or an ought not. This change is imperceptible; but is, however, of the last
consequence. For as this ought, or ought not expresses some new relation or affirmation,
 Philippa Foot

'tis necessary that it shou'd be observed and explain'd; and at the same time that a
reason should be given, for what seems altogether inconceivable, how this new relation
can be a deduction from others, which are entirely different from it. 5
Hume thought that he himself had hit on the perfect solution to the problem. The new
element in a proposition about virtue was the reference to a special sentiment of
approbation: nothing new in the object, but something in ourselves. At a blow he
seemed to have put an end to the hunt for mysterious extra properties, and also to have
shown the necessary connexion between morality and the will. For the moral sentiment,
the special feeling which we call approbation, was a pleasurable sentiment, by which
we were inclined towards those actions whose contemplation gave rise to it. ‘To know
virtue is to love it.’ This, Hume might have said, is a logical truth.
This extraordinarily interesting theory has been a great influence in contemporary
ethics. Many modern moral philosophers have taken up Hume's argument and, starting
from his premise about the necessarily practical nature of morality, assert his conclusion
about the gap between is and ought. Indeed they often make the connexion between
action and the propositions of morality even closer than he did, suggesting that an
actual rule of conduct rather than a mere sentiment is required for moral judgement. ‘To
know virtue is to be prepared to follow it’ is what they might say. And like Hume,
having anchored moral judgement to the will of the judger, they have cut it loose from
the world.
Such theories are, I believe, wrong, and the mistake can be traced back to the
interpretation of the crucial premise: morality is necessarily practical. It is not that this is
false, but that one may easily insist on too close a connexion between moral judgement
and the will. I do not know quite what sense ought to be given to the proposition that
morality is necessarily practical, but two things at least can be said. In the first place we
take it as part of the meaning of what we call ‘moral terms' that they are in general used
for teaching particular kinds of conduct; though nothing follows about what any
particular individual who uses the terms must feel or do. Secondly, since moral virtues
are qualities necessary if men are to get on well in a world in which they are frightened,
tempted by pleasure and liable to hurt rather than help each other, they need to make
common plans. This general connexion between such things as courage, temperance,
and justice and human good is quite enough to explain why people are often influenced
by considerations of morality. They are not necessarily influenced, as Hume must have
known; but they are concerned to teach and practise virtue in so far as they have taken
this thought for their own and the common good. It is, therefore, unnecessary to posit a
special sentiment to explain why observations about virtue have an influence on the
will, and the raison d’être of Hume's subjectivist theory of ethics disappears.

5 Ib. sec. 1.
 VI Nietzsche: The Revaluation of Values56F

VI Nietzsche: The Revaluation of Values 1

Philippa Foot
Abstract: Few philosophers in recent years have attempted to refute Friedrich Nietzsche's attack on
Christian and other moralities. Nietzsche sees the morality derived from Christianity as harmful because
it is slavish, rooted in weakness, fear, malice, and a desire for punishment of oneself and others. He sees
the preoccupation with others through pity and charity as a sign of spiritual ill health and argues that we
should value the strong; hence his concept of the Übermensch, or Superman. The author criticizes these
views.

This problem of the value of pity and of the morality of pity . . . seems at first sight to be
merely something detached, an isolated question mark; but whoever sticks with it and
learns how to ask questions here will experience what I experienced—a tremendous
new prospect opens up for him, a new possibility comes over him like a vertigo, every
kind of mistrust, suspicion, fear leaps up, his belief in morality, in all morality, falters—
finally a new demand becomes audible . . . we need a critique of moral values, the value
of these values themselves must just be called in question . . . (GM Preface 6). 2
What Nietzsche expresses here, his sense of the fearful strangeness of his thoughts, is
something intensely felt and not unfitting given the facts of the case. For in his lonely,
highly daring mental voyage he had come to a view of life which was quite unlike that
of any of his contemporaries, and which brought him to challenge ways of thought and
behaviour centuries old. He was ready, he said, to call in question Christian morality
and even all morality, and when he had questioned he condemned. Yet Nietzsche saw
as clearly as anyone that morality could fascinate and inspire. ‘Thou shalt’ he said is the
name of a great dragon ‘sparkling like gold’ (Z I ‘On the Three Metamorphoses’ 10). He
knew that what he was doing was almost unthinkable; he was branding as evil what
seemed most certainly good.
Now one would expect that such a challenge from an undoubted genius must either be
defeated or else shake the world. But neither of these things have happened. It is true
that Nietzsche's theories (or a travesty of them) played a brief and inglorious part on the
world's stage when he was proclaimed as a prophet by the Nazis, but by and large he
has neither been accepted nor refuted, and this seems a remarkable fact. How is it, one
may ask, that philosophers today do not even try to refute Nietzsche, and seem to feel
morality as firm as ever under their feet? Why do we not argue with him as we argue
with other philosophers of the past? Part of the answer seems to be that a confrontation

1 ‘Nietzsche: The Revaluation of Values’ originally appeared in R. C. Solomon, ed., Nietzsche, A Collection
of Critical Essays, New York, 1973.
2 Translations, unless otherwise stated, by Walter Kaufmann. A list of abbreviations follows.
 Philippa Foot

with Nietzsche is a difficult thing to arrange. We find it hard to know where we could
meet him because of the intrinsically puzzling nature of a project such as his. Nietzsche
had demanded a critique of moral values and announced that he was calling in
question ‘the value of these values themselves’. But how can one value values? The idea
of such a thing is enough to make one's head spin. It is, therefore, with a rueful sense of
the difficulties that I shall try, in this essay, to confront Nietzsche, or at least to help to
prepare the ground for a confrontation.
A problem arises at the outset. In the passage I quoted he spoke of the morality of pity
but also of ‘all morality’. Which shall we take his target to be? I shall consider first
Nietzsche's special objection to Christian morality, with its teaching of the virtues of
humility and compassion, and its rejection of ‘the world’.
Nietzsche wanted to show Christian morality as a ‘slave morality’ rooted not in
anything fine or admirable but rather in weakness, fear, and malice; these were its
origins and to these origins its present nature conformed. In this morality the good man
is the humble and compassionate man, the one who is not to be feared. But originally,
he insists, it was quite otherwise. In the beginning it was the strong, noble, privileged
aristocrat who called himself good, and called those who lacked his own qualities bad.
These old concepts were turned on their heads when the perspective of the weak
prevailed. For then the contrast of what was good and bad (schlecht) gave way to the
contrast between what was good and evil (böse); the weak branded those they feared
evil, and praised the ‘propitiatory’ qualities natural to men like themselves who were
incapable of aggression. Where the old valuation had been positive the new was
negative; the ‘members of the herd’ must first brand the enemies they feared as evil
before they could see themselves as good. Moreover Nietzsche detects a large amount
of malice under the professions of Christian humility and goodwill. When the weak call
the strong evil the move is not merely defensive; it is also an expression of that peculiar
malice which Nietzsche referred to as ressentiment. Those who cultivate humility and
the other propitiatory virtues to cloak their weakness nourish an envious resentment
against those stronger than themselves. They want revenge for their inferiority and
have a deep desire to humiliate and harm. The wish to punish seems to Nietzsche one
of the most evident signs of this hidden malice, and he sees the idea of free will, and
accountability, as invented by those who desired to inflict punishment. Nor is
punishment always directed outward; the man of self-sacrificing virtue is resentful and
venomous also towards himself. ‘But thus I counsel you, my friends: Mistrust all in
whom the impulse to punish is powerful’ (Z II ‘On the Tarantulas’).
The man professing Christian virtues is, Nietzsche insists, a sick individual, deeply
malicious to himself and others. He has been taught to reject life as it is, to despise his
own sensuality, and to torment himself and others in the name of his ideals. Even these
ideals are inimical to health, since what is preached is compassion, and this Nietzsche
sees as a kind of sickness in itself. Pity, he says, is a temptation to be resisted at all costs;
 VI Nietzsche: The Revaluation of Values56F

he thinks of it as a kind of poison to the compassionate man, who becomes infected by

the sufferings of others. ‘The suffering of others infects us, pity is an infection’ (WP
368). 3 Nor does he believe that pity relieves suffering. Now and then it may do so, but
more often the object of our compassion suffers from our intervention in his affairs. He
suffers first from the fact that we are helping him. ‘Having seen the sufferer suffer, I
was ashamed for the sake of his shame; and when I helped him I transgressed
grievously against his pride’ (Z II ‘On the Pitying’). ‘It seems to me that a human being
with the very best intentions can do immeasurable harm, if he is immodest enough to
wish to profit those whose spirit and will are concealed from him . . . ’ wrote Nietzsche
in a letter to his sister in 1885 (quoted in Walter Kaufmann [ed.], The Portable Nietzsche,
New York, Viking, 1956, p. 441). Nor did Nietzsche think that good motives lay behind
most charitable acts. Charitable and helpful people ‘dispose of the needy as of
possessions. . . . One finds them jealous if one crosses or anticipates them when they
want to help’ (BGE 194). Concern for others often betrayed a man's dissatisfaction with
himself; men who were dull tried to cheer themselves up with the sight of their
neighbour's misfortunes, while those who had a low opinion of themselves would try to
buy back a better opinion from those on whom they had conferred a benefit. Nietzsche
saw the preoccupation with others as an evasion, and a sign of spiritual ill-health; what
is important is to love oneself ‘so that one can bear to be with onself and need not roam’
(Z III ‘On the Spirit of Gravity’ 2). The man who loves himself will be the one who most
truly benefits others; in his own rejoicing he will forget how to contrive pain for them.
What shall we say of Nietzsche's attack on Christian morality as it has so far been
described? With what weapons is he attacking, and on what ground? Several different
lines of attack can be discerned. In the first place he is suggesting that what is praised as
Christian virtue is largely a sham, and that true goodwill would be produced not by
teaching the morality of compassion but rather by encouraging ‘a healthy egoism’.
Secondly he is saying that judged by its own aims this morality is bad. Men suffer pity
as a sickness, and by their pity they do more harm than good. Each of these charges
would be damaging if it could be shown to be supported by the facts. But what of
Nietzsche's account of the origins of Christian morality, and his insistence that it

3Translation by Walter Kaufmann and R. J. Hollingdale.

List of abbreviations of titles of Nietzsche's works:
A: The Antichrist.
BGE: Beyond Good and Evil.
CW: The Case of Wagner.
EH: Ecce Homo.
GM: The Genealogy of Morals.
Twilight: The Twilight of the Idols.
WP: The Will to Power.
Z: Thus Spake Zarathustra.
 Philippa Foot

represents the ascendency of the weak over the strong? If proved would this be
damaging or not? Could one reply that virtues such as compassion and justice are
indeed of special interest to those liable to misfortune and vulnerable to oppression,
and that they are none the worse for that? Such a reply would miss the point of
Nietzsche's attack. He is trying to show the ‘good and virtuous’ as representatives of a
mean and base section of mankind, as fawning, timid, incapable people who express in
hidden form the malice they are afraid to express openly. He wants to suggest that they
are both despicable and dislikable, and if he could really do this he would have struck a
most telling blow. For how could a society which came to see things in Nietzsche's
fashion have a morality of this kind? It is not, after all, enough for a moral system that
particular actions should be rewarded and punished as in a system of laws. If a certain
man is to be seen as a good man, and certain actions as good actions, then he and they
must be generally esteemed. And no one is esteemed if he is the object of scorn and
dislike. If Nietzsche could show that we have no reason either to admire the man of
Christian virtue or to be grateful to him, he would have knocked away a psychological
base without which this morality cannot stand.
In representing Christian morality as the weapon used by the weak to defend and exalt
themselves Nietzsche was trying to show it and them in a disagreeable light. But it was
much more important to him to show that in favouring the weak at the expense of the
strong Christianity was the most powerful of the forces making for the degeneration of
the human race. ‘Nothing has preoccupied me more profoundly than the problem of
decadence’ he wrote (CW Preface). And he saw as decadent the type of man
encouraged by Christian teaching, describing him as an accommodating, industrious,
gregarious individual who was mediocre, and dull. Against this portrait he set that of a
stronger ‘higher’ type of individual, bold, independent, and ready to say ‘yes’ to life.
Such a man would not be much concerned about suffering, whether his own or that of
others. Among his equals he would behave with restraint; to the weak he might be
dangerous, but if he harmed them it would be rather from disregard than from malice.
The weak man, however, is afraid of suffering for himself and preoccupied with the
misfortunes of others. He tries to build himself a safe life which shall not require too
much exertion. ‘One has one's little pleasure for the day and one's little pleasure for the
night: but one has a regard for health’ (Z I ‘Zarathustra's Prologue’ 5). He preaches the
morality of compassion, though filled with secret ill will towards others.
Much controversy has surrounded Nietzsche's writings on the ‘higher’ and ‘lower’
types of man. Is he to be taken as glorifying the cruel tyrant, the ‘beast of prey’, or did
he have some less repulsive ideal? The answer seems to be neither an unqualified yes
nor an unqualified no. There is no doubt that in the rank of men he preferred as the
more healthy type ‘even Caesare Borgia’ to the mediocre submissive modern man. And
there are some embarrassingly awful passages in which he speaks of the superior man's
ruthlessness towards his inferiors as if it could be seen as something merely pranksome.
 VI Nietzsche: The Revaluation of Values56F

But the cruel man is certainly not his ideal, and there are some to whom he refuses to
preach egoism at all.
There are the terrible ones who carry around within themselves the beast of prey and
have no choice but lust or self-laceration. And even their lust is still self-laceration. They
have not even become human beings yet, these terrible ones: let them preach
renunciation of life and pass away themselves! (Z I ‘On the Preachers of Death’).
In Nietzsche's eyes the important distinction was that of the ‘ascending’ and
‘descending’ types of men. One great question was to be asked about the history of any
race: did it represent decline or ascent? And each individual should be scrutinised ‘to
see whether he represents the ascending or the descending line of life’ (Twilight
‘Skirmishes of an Untimely Man’ 33). Nietzsche saw himself as the one, the only one,
who saw clearly the contrast between ascending and declining mankind. ‘I have a
subtler sense of smell for the signs of ascent and decline than any other human being
before me; I am the teacher par excellence for this . . . ’ (EH ‘Why I Am So Wise’ 1). It was
in this context that he preached egoism to the strong:
Every individual consists of the whole course of evolution. . . . If he represents the
ascending course of mankind, then his value is in fact extraordinary; and extreme care
may be taken over the preservation and promotion of his development. (It is concern
for the future promised him that gives the well-constituted individual such an
extraordinary right to egoism.) If he represents the descending course, decay, chronic
sickening, then he has little value: and the first demand of fairness is for him to take as
little space, force, and sunshine as possible away from the well-constituted (WP 373).
And it was in this context that Nietzsche spoke least ambiguously about the fate he
envisaged for the weak. ‘The weak and the failures shall perish: first principle of our
love of man’ (A 2).
It was no wonder then that Nietzsche had a special hatred of Christianity. He saw it as
the religion of the weak designed for their protection and glorification, and he saw it as
the most powerful influence for decadence and decline. Above all he thought Christian
morality harmful to the stronger and healthier type of man. By preserving the incapable
and ‘misbegotten’, and by insisting that they be the object of compassionate attention, it
would cause even the strong to be infected with gloom and nihilism. And even more
importantly it would lead the ‘higher’ type of man to mistrust his own nature, and
would create conditions in which it was impossible for him to find his health. To
require a peaceable benevolence from such a man, to preach humility and pity to him, is
necessarily to injure him.
To demand of strength that it should not express itself as strength, that it should not be a
desire to overcome, a desire to throw down, a desire to become master, a thirst for
enemies and resistances and triumphs, is just as absurd as to demand of weakness that
it should express itself as strength (GM I 13).
 Philippa Foot

The strong man condemned by society for doing what his nature demands will suffer
from guilt and self-hatred, and may well be turned into a criminal.
The criminal type is the type of the strong human being under unfavourable
circumstances: a strong human being made sick. . . . His virtues are ostracized by
society; the most vivid drives with which he is endowed soon grow together with the
depressing affects—with suspicion, fear and dishonour (Twilight ‘Skirmishes of an
Untimely Man’ 45).
Nietzsche does not shrink from the conclusion that for some men ruthlessness may be
the condition of health. It is the counterpart of his belief that ‘everything evil, terrible,
tyrannical in man, everything in him that is kin to beasts of prey and serpents, serves
the enhancement of the species “man” as much as its opposite does’ (BGE 44). If God is
dead nothing guarantees that evil may not be the condition of good.
If . . . a person should regard even the affects of hatred, envy, covetousness, and the lust
to rule as conditions of life, as factors which, fundamentally and essentially, must be
present in the general economy of life (and must, therefore, be further enhanced if life is
to be further enhanced)—he will suffer from such a view of things as from seasickness.
And yet even this hypothesis is far from being the strangest and most painful in this
immense and almost new domain of dangerous insights . . . (BGE 23).
It is, then, for the sake of the ‘higher’ man that the values of Christian morality must be
abandoned, and it is from this perspective that the revaluation of values takes place. Is
it Nietzsche's intention to present us with a clash of interests—the good of the strong
against that of the weak? Obviously he has this intention, but just as obviously this is
not all that he wants to suggest. A more puzzling aspect of his doctrines comes before
us when we remind ourselves of what he says about the value of the ‘higher’ type of
man.
The problem I pose is . . . what type of men shall be bred, shall be willed, for being higher
in value, worthier of life, more certain of a future (A 3).
And again Nietzsche says:
We have a different faith; to us the democratic movement is . . . a form of decay,
namely the diminution, of man, making him mediocre and lowering his value
(BGE 203, italics added).
What does he mean when he speaks of the value of one type of man as greater than that
of another? Nietzsche himself has remarked elsewhere that one can never too carefully
consider the question ‘value for what?’ And in these terms one might try to explain
what he says about the value of certain men. Perhaps he means that the contribution
they make to life in general—by their optimism and fearlessness for instance—makes
them valuable to us all. Or perhaps he is judging their value by the contribution they
make to the future. Are they not a bridge to the superior man who may come in the
future—to the Übermensch? Neither suggestion tells the whole story, and the second
simply shifts the problem. If the ‘Overman’ or ‘Superman’ is the one who gives his
 VI Nietzsche: The Revaluation of Values56F

forerunners value, this must be because he has value himself. In fact Nietzsche seems to
want to say that anyone who is strong, independent, and so on—anyone who fits his
description of the higher type of man—is one who has value in himself, and we are left
with this puzzling idea. Is Nietzsche merely talking nonsense, or can we make sense of
the word ‘value’ as it is used here? The answer seems to be that we can. For it does
make sense to say that we value strong and exceptional individuals, whether or not
Nietzsche's picture of these individuals rings true. We do find patterns of reaction to
exceptional men that would allow us to see here a valuing rather similar to valuing on
aesthetic grounds, even if it is one for which we have no special name. I am thinking of
the interest and admiration which is the common attitude to remarkable men of
exceptional independence of mind and strength of will. Such men hold our attention,
and are often willingly served. When Nietzsche says that what is at stake is whether
‘the highest power and splendor actually possible to the type man’ is ever to be attained
(GM Preface 6) it suggests that he is appealing to our tendency to admire certain
individuals whom we see as powerful and splendid. He himself even says, in one
passage, that ‘This is at bottom a question of taste and of aesthetics: would it be
desirable that the “most respectable”, i.e., most tedious, species of man should survive?’
(WP 353). But I think that the passage is untypical, and in any case does not quite
describe the facts as they are; if there is an element of respect in the common reaction to
strong and remarkable men then the analogy with an aesthetic valuation should not be
pressed too far. Perhaps what we should do is simply to suggest a similarity between
the way we attribute value (aesthetic value) to art objects and the value that Nietzsche
attributes to a certain kind of man, both resting on a set of familiar reactions, and on
reactions that have much in common.
If this were a correct account of the matter what would we conclude? Would Christian
morality, or any other, be vulnerable to Nietzsche's attack? From the comparison with
aesthetic values one would say that it might be vulnerable, not because something had
been proved against it but because men might come to care more about producing and
preserving interesting and ‘splendid’ individuals than about the ends of morality. For
consider what the implications would be were it to be discovered that the human race
would become physically uglier if morality flourished, or that justice and kindness
destroyed beauty of some other kind. This could be considered irrelevant by those for
whom moral values were more important than aesthetic values, and one might count as
similarly irrelevant the discovery that ‘the highest power and splendor’ was
inconsistent with moral ends; nevertheless morality might decline.
So far we have been considering Nietzsche's objections to one specific moral system—
that of Christianity. But he had spoken of an attack on ‘all morality’ and was ready to
call himself an immoralist. Does he really have arguments reaching so far? Some of his
arguments against Christian morality will be brought also against other moral systems.
But with others this will not be possible. Nietzsche could not, for instance, accuse
 Philippa Foot

Aristotle of preaching a morality of pity, nor of extolling humility. On the contrary

Aristotle's description of the megalopsychos who possesses the virtue of greatness of soul
and ‘deserves and claims great things’ (Nicomachean Ethics 1123 a 15) has much in
common with Nietzsche's picture of the ‘higher’ type of man. Let us ask which part of
Nietzsche's doctrines could justify us in thinking that morality—all morality—was
indeed his target.
Is it relevant, for instance, that Nietzsche had no place in his ideology for the concept of
guilt? About this he was quite explicit. He saw efforts to make men feel guilty as
expressions of malice, and rejected guilt as a reaction to anything he himself would do.
‘Not to perpetrate cowardice against one's own acts. Not to leave them in the lurch
afterwards! The bite of conscience is indecent!’ (Twilight ‘Maxims and Arrows’ 10). This
seems to prove nothing about whether Nietzsche was, as he said, rejecting all morality;
it does not seem impossible that a man should have a morality without accepting guilt
as a response to moral failure. It would have been different had he been rejecting the
aim of self-discipline, but this Nietzsche never did. Like Callicles in Plato's Gorgias
Nietzsche objects to the ‘taming’ of the strong man by society, but where Callicles urges
that the strong should throw away all restraint and allow their passions full rein
Nietzsche was scornful of such a suggestion. He does indeed oppose those who would
weaken or even destroy a man's passions, but insists that a strong will belongs only to
one who has imposed discipline and unity on his desires. So instead of objecting to
morality on the ground that it involves discipline of the passions he says that this is its
one merit. ‘What is essential and estimable in every morality is that it constitutes a long
compulsion . . . ’ (BGE 188). Obviously drawing on his own experience he insists that
what is most natural is a kind of self-discipline.
Every artist knows how far from any feeling of letting himself go his ‘most natural’ state
is—the free ordering, placing, disposing, giving form in the moment of ‘inspiration’—
and how strictly and subtly he obeys thousandfold laws precisely then . . . (BGE 188).
To this extent, then, Nietzsche is at one with the moralist: he is preaching self-discipline
and control of the passions. Nevertheless it may be argued that he is rightly to be called
an immoralist. It is relevant here to recall that the word ‘morality’ is derived from mos
with its plural mores, and that in its present usage it has not lost this connexion with the
mores—the rules of behaviour—of a society. For Nietzsche keeps some of his sharpest
vituperation for those who try to impose social rules and a code of behaviour which
shall be uniform throughout the community. He repeatedly rages against those who
preach ‘Good and evil, good and evil, the same for all’.
Let us finally consider how naive it is altogether to say: ‘Man ought to be such and such!’
Reality shows us an enchanting wealth of types, the abundance of a lavish play and
change of forms—and some wretched loafer of a moralist comments: ‘No! Man ought to
be different.’ He even knows what man should be like, this wretched bigot and prig: he
 VI Nietzsche: The Revaluation of Values56F

paints himself on the wall and comments, ‘Ecce homo!’ (Twilight ‘Morality as Anti-
Nature 6’).
What will create health in one will enfeeble another, and each man must discover the
rule of his own health. To desire that men should be virtuous
means that they should cease to be distinct
means that they should begin to resemble one another in their needs and demands—
more clearly that they should perish—
The will to a single morality is thereby proved to be a tyranny over other types by that
type whom this single morality fits: it is a destruction or a levelling for the sake of the
ruling type (whether to render the others no longer fearsome or to render them useful)
(WP 315).
It may be suggested that Nietzsche, even if he will not accept the rules of behaviour to
be taught to all men, does at least set up ideals of character valid for all. Is it not the case
that he refuses to praise anyone who is not, e.g., courageous and independent? This is
of course true, and it does give some overlap, both in form and content, between a
moral system and a set of teachings such as Nietzsche's. Nevertheless an injunction
such as ‘seek your own health’ is so neutral as to actual behaviour as to fail to
reestablish the link with social norms. And even injunctions such as this one were not
preached to everyone, since Nietzsche thought that many men were simply incapable of
health and strength. There were, as we have already seen, cruel monsters to whom
Nietzsche would not preach egoism. And as for the members of ‘the herd’ he said that
he had no wish to change them; the spirit of the herd should rule within the herd. He is
not, he insists, trying to preach his kind of virtue generally: it belongs only to the rare
and exceptional man.
These considerations should, I think, incline us to the view that Nietzsche is an
immoralist rather than a special kind of moralist. And one is led in the same direction
by the fact that he was prepared to throw out rules of justice in the interests of
producing a stronger and more splendid type of man. I suggested that this implied a
quasi-aesthetic rather than a moral set of values. Morality is necessarily connected with
such things as justice and the common good, and it is a conceptual matter that this is so.
Why then should we still have a feeling, as I think we do, that Nietzsche has a great
deal in common with the moralist and that he is not simply arguing from an
incompatible and irreconcilable point of view? I think that this is due to the fact that in
much of his work he can be seen as arguing about the way in which men must live in
order to live well. It is the common ground between his system and that of traditional
and particularly Greek morality that makes us inclined to think that he must be a
moralist after all. For while Nietzsche loathed utilitarianism, with its concern for the
greatest happiness of the greatest number, and its tendency to take pleasure and the
absence of pain as the motive of all human action, he himself was interested, one might
say, in the conditions in which men—at least strong men—would flourish. The issue is
 Philippa Foot

hard to get clear because Nietzsche, as well as introducing quasi-aesthetic criteria which
are irrelevant in this context, also appeals to an idea of human good that is opposed to
that of his opponents, and there is no concept that has proved more intractable than that
of human happiness or human good. We are inclined to say at first sight that happiness
is equivalent to contentment, and only see that this cannot be the case when we notice,
for example, that we count someone as unfortunate, not fortunate, if he suffers brain
damage and thereafter lives the life of a happy child. But this correction shows that
however little we are able to give an account of the idea of human good we have reason
to agree with Nietzsche that a man is harmed if he is taught to be content with small
pleasures and made unfit for enterprises requiring daring and independence. So,
insofar as Nietzsche is suggesting that morality in general, and Christian morality in
particular, has this effect he is at least arguing on moral ground. And of course there
were more obvious ways in which the ‘good and virtuous’ had been ‘harmed’. For
Nietzsche saw them as resentful, hating themselves and others, and without strong
purpose or desire. No man can live happily if he lives like this.
The conclusion of this discussion must be that Nietzsche's ‘revaluation of values’ is a
most complex matter, and there is no single answer to the question as to what he was
attacking or as to what the basis might be for the attack. It is not, therefore, surprising
that we should shy away from the attempt to say whether he was right. I shall,
however, try to say one or two things about this. First of all I would like to point out
that everything depends on his theories and observations of human nature. If his attack
on Christian morality and on other moralities is going to be worth anything he has got
to be right about the effect of teaching pity and justice—that it merely hides the
ressentiment of the weak while it does injury to the strong. And he would have to be
right in seeing compassion as necessarily harmful to the compassionate man and of
little use to the unfortunate. Moreover he would have to be right about the possibilities
of a ‘healthy egoism’ in the strong, even when this egoism could involve a ruthlessness
to those who are less fortunately placed. Now on some points in his psychological
observation Nietzsche undoubtedly was right; he was right for instance to teach us to be
wary of one who finds other men most satisfactory when they are in need of his help,
and to be wary also of the one who hates himself. At certain points his observation, and
his anticipation of depth psychology, shows him as a brilliant psychologist. But one
could not see Nietzsche as one who had a great knowledge of life and of the human
heart. He describes convincingly what he knew thoroughly, as he knew the life of the
lonely genius, the creative artist or thinker. It is, however, noticeable that his picture of
the strong noble man fails to carry conviction when it deviates from this model; what he
says about the conditions for this man's health seems to stem largely from his belief that
the overriding and underlying principle of human behaviour is the will to power. Now
it is notorious that general theories about the springs of action are traps for
philosophers, and Nietzsche, who tried to work on a world historical scale, is an
 VI Nietzsche: The Revaluation of Values56F

obvious victim of the delusion of having seen things whole. This charge would have to
be proved against him, but prima facie one has no reason to trust Nietzsche's views of
human nature beyond a very limited perspective, and one quite inadequate for his case.
There is no reason to suppose that we really are in the dilemma that he insists on—that
we either sacrifice the weak or else deform the strong. And in a way events have caught
up with Nietzsche. How could one see the present dangers that the world is in as
showing that there is too much pity and too little egoism around? One wonders what
Nietzsche himself, in some ways a most humane man, would have said if he were living
now, and could see inhumanity on its present scale, and in its present blatant forms.
 Philippa Foot

1
VII Moral Arguments

Philippa Foot
Abstract: Hare, Stevenson, and others have asserted that moral arguments ‘may always break down’ in a
way in which other arguments do not. This is because moral arguments must end with the assertion of a
moral principle that may simply be denied. This view is flawed by its reliance on the idea that there is no
logical connexion between facts and values. Emotivism and naturalism and their divergent approaches to
moral arguments are examined.

Those who are influenced by the emotivist theory of ethics, and yet wish to defend
what Hare has called ‘the rationality of moral discourse’, generally talk a lot about
‘giving reasons’ for saying that one thing is right, and another wrong. The fact that
moral judgements need defence seems to distinguish the impact of one man's moral
views upon others from mere persuasion or coercion, and the judgements themselves
from mere expressions of likes and dislikes. Yet the version of argument in morals
currently accepted seems to say that, while reasons must be given, no one need accept
them unless he happens to hold particular moral views. It follows that disputes about
what is right and wrong can be resolved only if certain contingent conditions are
fulfilled; if they are not fulfilled, the argument breaks down, and the disputants are left
face to face in opposition which is merely an expression of attitude and will. Much
energy is expended in trying to show that no sceptical conclusion can be drawn. It is
suggested, for instance, that anyone who has considered all the facts which could bear
on his moral position has ipso facto produced a ‘well founded’ moral judgement; in spite
of the fact that anyone else who has considered the same facts may well have come to
the opposite conclusion. How ‘x is good’ can be a well founded moral judgement when
‘x is bad’ can be equally well founded it is not easy to see.
The statement that moral arguments ‘may always break down’ is often thought of as
something that has to be accepted, and it is thought that those who deny it fail to take
account of what was proved once for all by Hume, and elaborated by Stevenson, by
Ayer, and by Hare. This article is an attempt to expose the assumptions which give the
‘breakdown’ theory so tenacious a hold, and to suggest an alternative view.
Looked at in one way, the assertion that moral arguments ‘may always break down’
appears to make a large claim. What is meant is that they may break down in a way in
which other arguments may not. We are therefore working on a model on which such
factors as shortage of time or temper are not shown; the suggestion is not that A's
argument with B may break down because B refuses for one reason or another to go on
with it, but that their positions are irreconcilable. Now the question is how can we

1 ‘Moral Arguments’ originally appeared in Mind, Volume 67, 1958

 VII Moral Arguments59F

assert that any disagreement about what is right and wrong may end like this? How do
we know, without consulting the details of each argument, that there is always an
impregnable position both for the man who says that X is right, or good, or what he
ought to do, and for the man who denies it? How do we know that each is able to deal
with every argument the other may bring?
Thus, when Hare describes someone who listens to all his adversary has to say and then
at the end simply rejects his conclusion, we want to ask ‘How can he?’ Hare clearly
supposes that he can, for he says that at this point the objector can only be asked to
make up his mind for himself. 2 No one would ever paint such a picture of other kinds
of argument—suggesting, for instance, that a man might listen to all that could be said
about the shape of the earth, and then ask why he should believe that it was round. We
should want, in such a case, to know how he met the case put to him; and it is
remarkable that in ethics this question is thought not to be in place.
If a man making a moral judgement is to be invulnerable to criticism, he must be free
from reproach on two scores: (a) he must have brought forward evidence, where
evidence is needed; and (b) he must have disposed of any contrary evidence offered. It
is worth showing why writers who insist that moral arguments may always break
down assume, for both sides in a moral dispute, invulnerability on both counts. The
critical assumption appears in different forms because different descriptions of moral
arguments are given; and I shall consider briefly what has been said by Stevenson and
by Hare.
I. Stevenson sees the process of giving reasons for ethical conclusions as a special
process of non-deductive inference, in which statements expressing beliefs (R) form the
premises and emotive (evaluative) utterances (E) the conclusion. There are no rules
validating particular inferences, but only causal connexions between the beliefs and
attitudes concerned. ‘Suppose’, he writes,
that a theorist should tabulate the ‘valid’ inferences from R's to E's. It is difficult to see
how he could be doing anything more than specify what R's he thereby resolves to
accept as supporting the various E's. . . . Under the name of ‘validity’ he will be selecting
those inferences to which he is psychologically disposed to give assent and perhaps
inducing others to give a similar assent to them. 3
It follows that disputes in which each man backs up his moral judgement with ‘reasons’
may always break down, and this is an implication on which Stevenson insists. So long
as he does not contradict himself and gets his facts right, a man may argue as he
chooses, or as he finds himself psychologically disposed. He alone says which facts are
relevant to ethical conclusions, so that he is invulnerable on counts (a) and (b): he can
simply assert that what he brings forward is evidence and can simply deny the

2 The Language of Morals, p. 69.

3 Ethics and Language, pp. 170–1.
 Philippa Foot

relevance of any other. His argument may be ineffective, but it cannot be said to be
wrong. Stevenson speaks of ethical ‘inference’ and of giving ‘reasons’, but the process
he describes is rather that of trying to produce a result, an attitude, by means of a
special kind of adjustment, an alteration in belief. All that is needed for a breakdown is
for different attitudes in different people to be causally connected to the same beliefs.
Then even complete agreement in belief will not settle a moral dispute.
II. Hare gives a picture of moral reasoning which escapes the difficulties of a special
form of inference without rules of validity. He regards an argument to a moral
conclusion as a syllogistic inference, with the ordinary rules. The facts, such as ‘this is
stealing’, which are to back up a moral judgement are to be stated in a ‘descriptive’
minor premise, and their relevance is to be guaranteed by an ‘evaluative’ major premise
in which that kind of thing is said to be good or bad. There is thus no difficulty about
the validity of the argument; but one does arise about the status of the major premise.
We are supposed to say that a particular action is bad because it is a case of stealing,
and because stealing is wrong; but if we ask why stealing is wrong, we can only be
presented with another argument of the same form, with another exposed moral
principle as its major premise. In the end everyone is forced back to some moral
principle which he simply asserts—and which someone else may simply deny. It can
therefore be no reproach to anyone that he gives no reasons for a statement of moral
principle, since any moral argument must contain some undefended premise of this
kind. Nor can he be accused of failing to meet arguments put forward by opponents
arguing from different principles; for by denying their ultimate major premises he can
successfully deny the relevance of anything they say.
Both these accounts of moral argument are governed by the thought that there is no
logical connexion between statements of fact and statements of value, so that each man
makes his own decision as to the facts about an action which are relevant to its
evaluation. To oppose this view we should need to show that, on the contrary, it is laid
down that some things do, and some things do not, count in favour of a moral
conclusion, and that a man can no more decide for himself what is evidence for
rightness and wrongness than he can decide what is evidence for monetary inflation or
a tumour on the brain. If such objective relations between facts and values existed, they
could be of two kinds: descriptive, or factual, premises might entail evaluative
conclusions, or they might count as evidence for them. It is the second possibility which
chiefly concerns me, but I shall nevertheless consider the arguments which are
supposed to show that the stronger relationship cannot exist. For I want to show that
the arguments usually brought forward do not even prove this. I want to say that it has
not even been proved that moral conclusions cannot be entailed by factual or
descriptive premises.
It is often thought that Hume showed the impossibility of deducing ‘ought’ from ‘is’,
but the form in which this view is now defended is, of course, that in which it was
 VII Moral Arguments59F

rediscovered by G. E. Moore at the beginning of the present century, and developed by

such other critics of ‘naturalistic’ ethics as Stevenson, Ayer, and Hare. We need
therefore to look into the case against naturalism to see exactly what was proved.
Moore tried to show that goodness was a non-natural property, and thus not to be
defined in terms of natural properties; the problem was to explain the concept of a
‘natural property’, and to prove that no ethical definition in terms of natural properties
could be correct. As Frankena 4 and Prior 5 pointed out, the argument against naturalism
was always in danger of degenerating into a truism. A natural property tended to
become one not identical with goodness, and the naturalistic fallacy that of identifying
goodness with ‘some other thing’.
What was needed to give the attack on naturalism new life was the identification of
some deficiency common to the whole range of definitions rejected by Moore, a reason
why they all failed. This was provided by the theory that value terms in general, and
moral terms in particular, were used for a special function—variously identified as
expressing feelings, expressing and inducing attitudes, or commending. Now it was
said that words with emotive or commendatory force, such as ‘good’, were not to be
defined by the use of words whose meaning was merely ‘descriptive’. This discovery
tended to appear greater than it was, because it looked as if the two categories of fact
and value had been identified separately and found never to coincide, whereas actually
the factual or descriptive was defined by exclusion from the realm of value. In the
ordinary sense of ‘descriptive’ the word ‘good’ is a descriptive word and in the
ordinary sense of ‘fact’ we say that it is a fact about so and so that he is a good man, so
that the words must be used in a special sense in moral philosophy. But a special
philosopher's sense of these words has never, so far as I know, been explained except by
contrasting value and fact. A word or sentence seems to be called ‘descriptive’ on
account of the fact that it is not emotive, does not commend, does not entail an
imperative, and so on according to the theory involved. This might seem to reduce the
case against naturalism once more to an uninteresting tautology, but it does not do so.
For if the non-naturalist has discovered a special feature found in all value judgements,
he can no longer be accused of saying merely that nothing is a definition of ‘good’
unless it is a definition of ‘good’ and not ‘some other thing’. His part is now to insist
that any definition which fails to allow for the special feature of value judgements must
be rejected, and to label as ‘naturalistic’ all the definitions which fail to pass this test.
I shall suppose, for the sake of argument, that the non-naturalist really has identified
some characteristic (let us call it f) essential to evaluative words; that he is right in
saying that evaluations involve emotions, attitudes, the acceptance of imperatives, or
something of the kind. He is therefore justified in insisting that no word or statement

4 W. K. Frankena, ‘The Naturalistic Fallacy’, Mind, 1939

5 A. N. Prior, Logic and the Basis of Ethics, chap. I.
 Philippa Foot

which does not have the property f can be taken as equivalent to any evaluation, and
that no account of the use of an evaluative term can leave out f and yet be complete.
What, if anything, follows about the relation between premises and conclusion in an
argument designed to support an evaluation?
It is often said that what follows is that evaluative conclusion cannot be deduced from
descriptive premises, but how is this to be shown? Of course, if a descriptive premise is
redefined, as one which does not entail an evaluative conclusion, the non-naturalist will
once more have bought security at the price of becoming a bore. He can once more
improve his position by pointing to the characteristic f belonging to all evaluations, and
asserting that no set of premises which do not entail an f proposition can entail an
evaluation. If he takes this course he will be more like the man who says that a
proposition which entails a proposition about a dog must be one which entails a
proposition about an animal; he is telling us what to look out for in checking the
entailment. What he is not so far telling us is that we can test for the entailment by
looking to see whether the premise itself has the characteristic f. For all that has yet been
shown it might be possible for a premise which is not f to entail a conclusion which is f,
and it is obviously this proposition which the non-naturalist wants to deny.
Now it may seem obvious that a non-evaluative premise could not entail an evaluative
conclusion, but it remains unclear how it is supposed to be proved.
In one form, the theory that an evaluative conclusion of a deductive argument needs
evaluative premises is clearly unwarrantable; I mention it only to get it out of the way.
We cannot possibly say that at least one of the premises must be evaluative if the
conclusion is to be so; for there is nothing to tell us that whatever can truly be said of
the conclusion of a deductive argument can truly be said of any of the premises. It is not
necessary that the evaluative element should ‘come in whole’, so to speak. If f has to
belong to the premises it can only be necessary that it should belong to the premises
together, and it may be no easy matter to see whether a set of propositions has the
property f.
How in any case is it to be proved that if the conclusion is to have the characteristic f the
premises taken together must also have it? Can it be said that unless this is so it will
always be possible to assert the premises and yet deny the conclusion? I shall try to
show that this at least is false, and in order to do so I shall consider the case of
arguments designed to show that a certain piece of behaviour is or is not rude.
I think it will be agreed that in the wide sense in which philosophers speak of
evaluation, ‘rude’ is an evaluative word. At any rate it has the kind of characteristics
upon which non-naturalists fasten: it expresses disapproval, is meant to be used when
action is to be discouraged, implies that other things being equal the behaviour to
which it is applied will be avoided by the speaker, and so on. For the purpose of this
argument I shall ignore the cases in which it is admitted that there are reasons why
something should be done in spite of, or even because of, the fact that it is rude. Clearly
 VII Moral Arguments59F

there are occasions when a little rudeness is in place, but this does not alter the fact that
‘rude’ is a condemnatory word.
It is obvious that there is something else to be said about the word ‘rude’ besides the
fact that it expresses, fairly mild, condemnation: it can only be used where certain
descriptions apply. The right account of the situation in which it is correct to say that a
piece of behaviour is rude, is, I think, that this kind of behaviour causes offence by
indicating lack of respect. Sometimes it is merely conventional that such behaviour does
indicate lack of respect (e.g. when a man keeps his hat on in someone else's house);
sometimes the behaviour is naturally disrespectful, as when one man pushes another
out of the way. (It should be mentioned that rudeness and the absence of rudeness do
not exhaust the subject of etiquette; some things are not rude, and yet are ‘not done’. It
is rude to wear flannels at a formal dinner party, but merely not done to wear a dinner
jacket for tennis.)
Given that this reference to offence is to be included in any account of the concept of
rudeness, we may ask what the relation is between the assertion that these conditions of
offence are fulfilled—let us call it O—and the statement that a piece of behaviour is
rude—let us call it R. Can someone who accepts the proposition O (that this kind of
offence is caused) deny the proposition R (that the behaviour is rude)? I should have
thought that this was just what he could not do, for if he says that it is not rude, we
shall stare, and ask him what sort of behaviour would be rude; and what is he to say?
Suppose that he were to answer ‘a man is rude when he behaves conventionally’, or ‘a
man is rude when he walks slowly up to a front door’, and this not because he believes
that such behaviour causes offence, but with the intention of leaving behind entirely the
usual criteria of rudeness. It is evident that with the usual criteria of rudeness he leaves
behind the concept itself; he may say the words ‘I think this rude’, but it will not on that
account be right to describe him as ‘thinking it rude’. If I say ‘I am sitting on a pile of
hay’ and bring as evidence the fact that the object I am sitting on has four wooden legs
and a hard wooden back, I shall hardly be described as thinking, even mistakenly, that I
am sitting on a pile of hay; all I am doing is to use the words ‘pile of hay’.
It might be thought that the two cases were not parallel, for while the meaning of ‘pile
of hay’ is given by the characteristics which piles of hay must possess, the meaning of
‘rude’ is given by the attitude it expresses. The answer is that if ‘thinking a thing rude’
is to be described as having a particular attitude to it, then having an attitude
presupposes, in this case, believing that certain conditions are fulfilled. If ‘attitudes’
were solely a matter of reactions such as wrinkling the nose, and tendencies to such
things as making resolutions and scolding, then thinking something rude would not be
describable solely in terms of attitudes. Either thinking something rude is not to be
described in terms of attitudes, or attitudes are not to be described in terms of such
things. Even if we could suppose that a particular individual could react towards
conventional behaviour, or to walking slowly up to an English front door, exactly as
 Philippa Foot

most people react to behaviour which gives offence, this would not mean that he was to
be described as thinking these things rude. And in any case the supposition is nonsense.
Although he could behave in some ways as if he thought them rude, e.g. by scolding
conventional or slow-walking children, but not turning daughters with these
proclivities out of doors, his behaviour could not be just as if he thought them rude. For
as the social reaction to conventional behaviour is not the same as the social reaction to
offensive behaviour, he could not act in just the same way. He could not for instance
apologise for what he could call his ‘rudeness’, for he would have to admit that it had
caused no offence.
I conclude that whether a man is speaking of behaviour as rude or not rude, he must
use the same criteria as anyone else, and that since the criteria are satisfied if O is true, it
is impossible for him to assert O while denying R. It follows that if it is a sufficient
condition of P's entailing Q that the assertion of P is inconsistent with the denial of Q,
we have here an example of a non-evaluative premise from which an evaluative
conclusion can be deduced.
It is of course possible to admit O while refusing to assert R, and this will not be like the
refusal to say about prunes what one has already admitted about dried plums. Calling
an action ‘rude’ is using a concept which a man might want to reject, rejecting the whole
practice of praising and blaming embodied in terms such as ‘polite’ and ‘rude’. Such a
man would refuse to discuss points of etiquette, and arguments with him about what is
rude would not so much break down as never begin. But once he did accept the
question ‘Is this rude?’ he would have to abide by the rules of this kind of argument; he
could not bring forward any evidence he liked, and he could not deny the relevance of
any piece of evidence brought forward by his opponent. Nor could he say that he was
unable to move from O to R on this occasion because the belief in O had not induced in
him feelings or attitudes warranting the assertion of R. If he had agreed to discuss
rudeness he had committed himself to accepting O as evidence for R, and evidence is
not a sort of medicine which is taken in the hope that it will work. To suggest that he
could refuse to admit that certain behaviour was rude because the right psychological
state had not been induced, is as odd as to suppose that one might refuse to speak of the
world as round because in spite of the good evidence of roundness a feeling of
confidence in the proposition had not been produced. When given good evidence it is
one's business to act on it, not to hang around waiting for the right state of mind. It
follows that if a man is prepared to discuss questions of rudeness, and hence to accept
as evidence the fact that behaviour causes a certain kind of offence, he cannot refuse to
admit R when O has been proved.
The point of considering this example was to show that there may be the strictest rules
of evidence even where an evaluative conclusion is concerned. Applying this principle
to the case of moral judgements, we see that—for all that the nonnaturalist has proved
to the contrary—Bentham, for instance, may be right in saying that when used in
 VII Moral Arguments59F

conjunction with the principle of utility ‘the words ought and right and wrong, and
others of that stamp, have a meaning: when otherwise they have none’. 6 Anyone who
uses moral terms at all, whether to assert or deny a moral proposition, must abide by
the rules for their use, including the rules about what shall count as evidence for or
against the moral judgement concerned. For anything that has yet been shown to the
contrary these rules could be entailment rules, forbidding the assertion of factual
propositions in conjunction with the denial of moral propositions. The only recourse of
the man who refused to accept the things which counted in favour of a moral
proposition as giving him a reason to do certain things or to take up a particular
attitude, would be to leave the moral discussion and abjure altogether the use of moral
terms.
To say what Bentham said is not, then, to commit any sort of ‘naturalistic fallacy’. It is
open to us to enquire whether moral terms do lose their meaning when divorced from
the pleasure principle, or from some other set of criteria, as the word ‘rude’ loses its
meaning when the criterion of offensiveness is dropped. To me it seems that this is
clearly the case; I do not know what could be meant by saying that it was someone's
duty to do something unless there was an attempt to show why it mattered if this sort
of thing was not done. How can questions such as ‘what does it matter?’, ‘what harm
does it do?’, ‘what advantage is there in . . . ?’, ‘why is it important?’, be set aside here?
Is it even to be suggested that the harm done by a certain trait of character could be
taken, by some extreme moral eccentric, to be just what made it a virtue? I suggest that
such a man would not even be a moral eccentric, any more than the man who used the
word ‘rude’ of conventional behaviour was putting forward strange views about what
was rude. Both descriptions have their proper application, but it is not here. How
exactly the concepts of harm, advantage, benefit, importance, etc. are related to the
different moral concepts, such as rightness, obligation, goodness, duty and virtue is
something that needs the most patient investigation, but that they are so related seems
undeniable, and it follows that a man cannot make his own personal decision about the
considerations which are to count as evidence in morals.
Perhaps it will be argued that this kind of freedom of choice is not ruled out after all,
because a man has to decide for himself what is to count as advantage, benefit, or harm.
But is this really plausible? Consider the man described by Hare as thinking that
torturing is morally permissible. 7 Apparently he is not supposed to be arguing that in
spite of everything torture is justifiable as a means of extracting confessions from
enemies of the state, for the argument is supposed to be at an end when he has said that
torturing people is permissible, and his opponent has said that it is not. How is he
supposed to have answered the objection that to inflict torture is to do harm? If he is

6 Principles of Morals and Legislation, chap. I, x.

7 Universalizability, Proceedings of the Aristotelian Society, 1954–1955, p. 304.
 Philippa Foot

supposed to have said that pain is good for a man in the long run, rather than bad, he
will have to show the benefits involved, and he can no more choose what shall count as
a benefit than he could have chosen what counted as harm. Is he supposed perhaps to
count as harm only harm to himself? In this case he is guilty of ignoratio elenchi. By
refusing to count as harm anything except harm to himself, he puts himself outside the
pale of moral discussion, and should have explained that this was his position. One
might compare his case to that of a man who in some discussion of common policy says
‘this will be the best thing to do’, and announces afterwards that he meant best for
himself. This is not what the word ‘best’ does mean in the context of such a discussion.
It may be objected that these considerations about the evidence which must be brought
for saying that one thing is good and another bad, could not in any case be of the least
importance; such rules of evidence, even if they exist, only reflecting the connexion
between our existing moral code and our existing moral terms; if there are no ‘free’
moral terms in our language it can always be supposed that some have been invented—
as indeed they will have to be invented if we are to be able to argue with people who
subscribe to a moral code entirely different from our own. This objection rests on a
doubtful assumption about the concept of morality. It assumes that even if there are
rules about the ground on which actions can be called good, right, or obligatory, there
are no rules about the grounds on which a principle which is to be called a moral
principle may be asserted. Those who believe this must think it possible to identify an
element of feeling or attitude which carries the meaning of the word ‘moral’. It must be
supposed, for instance, that if we describe a man as being for or against certain actions,
bringing them under universal rules, adopting these rules for himself, and thinking
himself bound to urge them on others, we shall be able to identify him as holding moral
principles, whatever the content of the principle at which he stops. But why should it be
supposed that the concept of morality is to be caught in this particular kind of net? The
consequences of such an assumption are very hard to stomach; for it follows that a rule
which was admitted by those who obeyed it to be completely pointless could yet be
recognised as a moral rule. If people happened to insist that no one should run round
trees left handed, or look at hedgehogs in the light of the moon, this might count as a
basic moral principle about which nothing more need be said.
I think that the main reason why this view is so often held in spite of these difficulties, is
that we fear the charge of making a verbal decision in favour of our own moral code.
But those who bring that charge are merely begging the question against arguments
such as those given above. Of course if the rules we are refusing to call moral rules can
really be given this name, then we are merely legislating against alien moral codes. But
the suggestion which has been put forward is that this could not be the right description
for rules of behaviour for which an entirely different defence is offered from that which
we offer for our moral beliefs. If this suggestion is right, the difference between
ourselves and the people who have these rules is not to be described as a difference of
 VII Moral Arguments59F

moral outlook, but rather as a difference between a moral and a non-moral point of
view. The example of etiquette is again useful here. No one is tempted to say that the
ruling out, a priori, of rules of etiquette which each man decides for himself when he
feels so inclined, represents a mere verbal decision in favour of our kind of socially
determined standards of etiquette. On what grounds could one call a rule which
someone was allowed to invent for himself a rule of etiquette? It is not just a fact about
the use of our words ‘rude’, ‘not done’, etc., that they could not be applied in such a
case; it is also a fact about etiquette that if terms in another language did appear in such
situations they would not be terms of etiquette. We can make a similar point about the
terms ‘legal’ and ‘illegal’ and the concept of law. If any individual was allowed to apply
a certain pair of terms expressing approval and disapproval off his own bat, without
taking notice of any recognised authority, such terms could not be legal terms. Similarly
it is a fact about etiquette and law that they are both conventional as morality is not.
It may be that in attempting to state the rules which govern the assertion of moral
propositions we shall legislate against a moral system radically opposed to our own.
But this is only to say that we may make a mistake. The remedy is to look more
carefully at the rules of evidence, not to assume that there cannot be any at all. If a
moral system such as Nietzsche's has been refused recognition as a moral system, then
we may have got the criteria wrong. The fact that Nietzsche was arguably a moralist
cannot, however, be quoted in favour of the private enterprise theory of moral criteria.
Admittedly Nietzsche said ‘You want to decrease suffering; I want precisely to increase
it’ but he did not just say this. Nor did he offer as a justification the fact that suffering
causes a tendency to absent mindedness, or lines on the human face. We are inclined to
recognise Nietzsche as a moralist because he tries to justify an increase in suffering by
connecting it with strength as opposed to weakness, and individuality as opposed to
conformity. That strength is a good thing can only be denied by someone who can show
that the strong man overreaches himself, or in some other way brings harm to himself
or other people. That individuality is a good thing is something that has to be shown,
but in a vague way we connect it with originality, and with courage, and hence there is
no difficulty in conceiving Nietzsche as a moralist when he appeals to such a thing.
In conclusion it is worth remarking that moral arguments break down more often than
philosophers tend to think, but that the breakdown is of a different kind. When people
argue about what is right, good, or obligatory, or whether a certain character trait is or
is not a virtue, they do not confine their remarks to the adducing of facts which can be
established by simple observation, or by some clear-cut technique. What is said may
well be subtle or profound, and in this sort of discussion as in others, in the field of
literary criticism for instance, or the discussion of character, much depends on
experience and imagination. It is quite common for one man to be unable to see what
the other is getting at, and this sort of misunderstanding will not always be resolvable
by anything which could be called argument in the ordinary sense.
 Philippa Foot

VIII Moral Beliefs 1

Philippa Foot
Abstract: This essay looks critically at the arguments by which moral philosophers have rejected ethical
naturalism. The author denies the supposed logical gap between assertions of fact and moral evaluations,
arguing that there is an internal logical relation between the two. The word ‘good’ no more ‘floats free’
than does, e.g., ‘dangerous’. Both have a connexion in meaning with particular facts, and through them
with reasons for action.

I
To many people it seems that the most notable advance in moral philosophy during the
past fifty years or so has been the refutation of naturalism; and they are a little shocked
that at this late date such an issue should be reopened. It is easy to understand their
attitude: given certain apparently unquestionable assumptions, it would be about as
sensible to try to reintroduce naturalism as to try to square the circle. Those who see it
like this have satisfied themselves that they know in advance that any naturalistic
theory must have a catch in it somewhere, and are put out at having to waste more time
exposing an old fallacy. This paper is an attempt to persuade them to look critically at
the premises on which their arguments are based.
It would not be an exaggeration to say that the whole of moral philosophy, as it is now
widely taught, rests on a contrast between statements of fact and evaluations, which
runs something like this: ‘The truth or falsity of statements of fact is shown by means of
evidence; and what counts as evidence is laid down in the meaning of the expressions
occurring in the statement of fact. (For instance, the meaning of “round” and “flat”
made Magellan's voyages evidence for the roundness rather than the flatness of the
Earth; someone who went on questioning whether the evidence was evidence could
eventually be shewn to have made some linguistic mistake.) It follows that no two
people can make the same statement and count completely different things as evidence;
in the end one at least of them could be convicted of linguistic ignorance. It also follows
that if a man is given good evidence for a factual conclusion he cannot just refuse to
accept the conclusion on the ground that in his scheme of things this evidence is not
evidence at all. With evaluations, however, it is different. An evaluation is not
connected logically with the factual statements on which it is based. One man may say
that a thing is good because of some fact about it, and another may refuse to take that
fact as any evidence at all, for nothing is laid down in the meaning of “good” which
connects it with one piece of “evidence” rather than another. It follows that a moral
eccentric could argue to moral conclusions from quite idiosyncratic premises; he could

1 ‘Moral Beliefs’ originally appeared in the Proceedings of the Aristotelian Society, Volume 59, 1958–1959
 VIII Moral Beliefs66F

say, for instance, that a man was a good man because he clasped and unclasped his
hands and never turned N.N.E. after turning S.S.W. He could also reject someone else's
evaluation simply by denying that his evidence was evidence at all.
The fact about “good” which allows the eccentric still to use this term without falling
into a morass of meaninglessness, is its “action-guiding” or “practical” function. This it
retains; for like everyone else he considers himself bound to choose the things he calls
“good” rather than those he calls “bad”. Like the rest of the world he uses “good’ in
connexion only with a “pro-attitude”; it is only that he has pro-attitudes to quite
different things, and therefore calls them good.'
There are here two assumptions about ‘evaluations’, which I will call assumption (1)
and assumption (2).
Assumption (1) is that some individual may, without logical error, base his beliefs about
matters of value entirely on premises which no one else would recognise as giving
evidence at all. Assumption (2) is that, given the kind of statement which other people
regard as evidence for an evaluative conclusion, he may refuse to draw the conclusion
because this does not count as evidence for him.
Let us consider assumption (1). We might say that this depends on the possibility of
keeping the meaning of ‘good’ steady through all changes in the facts about anything
which are to count in favour of its goodness. (I do not mean, of course, that a man can
make changes as fast as he chooses; only that, whatever he has chosen, it will not be
possible to rule him out of order.) But there is a better formulation, which cuts out
trivial disputes about the meaning which ‘good’ happens to have in some section of the
community. Let us say that the assumption is that the evaluative function of ‘good’ can
remain constant through changes in the evaluative principles; on this ground it could be
said that even if no one can call a man good because he clasps and unclasps his hands,
he can commend him or express his pro-attitude towards him, and if necessary can
invent a new moral vocabulary to express his unusual moral code.
Those who hold such a theory will naturally add several qualifications. In the first
place, most people now agree with Hare, against Stevenson, that such words as ‘good’
apply to individual cases only through the application of general principles, so that
even the extreme moral eccentric must accept principles of commendation. In the
second place ‘commending’, ‘having a pro-attitude’, and so on, are supposed to be
connected with doing and choosing, so that it would be impossible to say, e.g., that a
man was a good man only if he lived for a thousand years. The range of evaluation is
supposed to be restricted to the range of possible action and choice. I am not here
concerned to question these supposed restrictions on the use of evaluative terms, but
only to argue that they are not enough.
The crucial question is this. Is it possible to extract from the meaning of words such as
‘good’ some element called ‘evaluative meaning’ which we can think of as externally
related to its objects? Such an element would be represented, for instance, in the rule
 Philippa Foot

that when any action was ‘commended’ the speaker must hold himself bound to accept
an imperative ‘let me do these things’. This is externally related to its object because,
within the limitation which we noticed earlier, to possible actions, it would make sense
to think of anything as the subject of such ‘commendation’. On this hypothesis a moral
eccentric could be described as commending the clasping of hands as the action of a
good man, and we should not have to look for some background to give the
supposition sense. That is to say, on this hypothesis the clasping of hands could be
commended without any explanation; it could be what those who hold such theories
call ‘an ultimate moral principle’.
I wish to say that this hypothesis is untenable, and that there is no describing the
evaluative meaning of ‘good’, evaluation, commending, or anything of the sort, without
fixing the object to which they are supposed to be attached. Without first laying hands
on the proper object of such things as evaluation, we shall catch in our net either
something quite different such as accepting an order or making a resolution, or else
nothing at all.
Before I consider this question, I shall first discuss some other mental attitudes and
beliefs which have this internal relation to their object. By this I hope to clarify the
concept of internal relation to an object, and incidentally, if my examples arouse
resistance, but are eventually accepted, to show how easy it is to overlook an internal
relation where it exists.
Consider, for instance, pride.
People are often surprised at the suggestion that there are limits to the things a man can
be proud of, about which indeed he can feel pride. I do not know quite what account
they want to give of pride; perhaps something to do with smiling and walking with a
jaunty air, and holding an object up where other people can see it; or perhaps they think
that pride is a kind of internal sensation, so that one might naturally beat one's breast
and say ‘pride is something I feel here’. The difficulties of the second view are well
known; the logically private object cannot be what a name in the public language is the
name of.2 The first view is the more plausible, and it may seem reasonable to say that
given certain behaviour a man can be described as showing that he is proud of
something, whatever that something may be. In one sense this is true, and in another
sense not. Given any description of an object, action, personal characteristic, etc., it is
not possible to rule it out as an object of pride. Before we can do so we need to know
what would be said about it by a man who is to be proud of it, or feels proud of it; but if
he does not hold the right beliefs about it then whatever his attitude is it is not pride.
Consider, for instance, the suggestion that someone might be proud of the sky or the
sea: he looks at them and what he feels is pride, or he puffs out his chest and gestures
with pride in their direction. This makes sense only if a special assumption is made

2 See Wittgenstein, Philosophical Investigations, especially §§ 243–315.

 VIII Moral Beliefs66F

about his beliefs, for instance that he is under some crazy delusion and believes that he
has saved the sky from falling, or the sea from drying up. The characteristic object of
pride is something seen (a) as in some way a man's own, and (b) as some sort of
achievement or advantage; without this object pride cannot be described. To see that the
second condition is necessary, one should try supposing that a man happens to feel
proud because he has laid one of his hands on the other, three times in an hour. Here
again the supposition that it is pride that he feels will make perfectly good sense if a
special background is filled in. Perhaps he is ill, and it is an achievement even to do this;
perhaps this gesture has some religious or political significance, and he is a brave man
who will so defy the gods or the rulers. But with no special background there can be no
pride, not because no one could psychologically speaking feel pride in such a case, but
because whatever he did feel could not logically be pride. Of course, people can see
strange things as achievements, though not just anything, and they can identify
themselves with remote ancestors, and relations, and neighbours, and even on
occasions with Mankind. I do not wish to deny there are many far-fetched and comic
examples of pride.
We could have chosen many other examples of mental attitudes which are internally
related to their object in a similar way. For instance, fear is not just trembling, and
running, and turning pale; without the thought of some menacing evil no amount of
this will add up to fear. Nor could anyone be said to feel dismay about something he
did not see as bad; if his thoughts about it were that it was altogether a good thing, he
could not say that (oddly enough) what he felt about it was dismay. ‘How odd, I feel
dismayed when I ought to be pleased’ is the prelude to a hunt for the adverse aspect of
the thing, thought of as lurking behind the pleasant façade. But someone may object
that pride and fear and dismay are feelings or emotions and therefore not a proper
analogy for ‘commendation’, and there will be an advantage in considering a different
kind of example. We could discuss for instance, the belief that a certain thing is
dangerous, and ask whether this could logically be held about anything whatsoever.
Like ‘this is good’, ‘this is dangerous’ is an assertion, which we should naturally accept
or reject by speaking of its truth or falsity; we seem to support such statements with
evidence, and moreover there may seem to be a ‘warning function’ connected with the
word ‘dangerous’ as there is supposed to be a ‘commending function’ connected with
the word ‘good’. For suppose that philosophers, puzzled about the property of
dangerousness, decided that the word did not stand for a property at all, but was
essentially a practical or action-guiding term, used for warning. Unless used in an
‘inverted comma sense’ the word ‘dangerous’ was used to warn, and this meant that
anyone using it in such a sense committed himself to avoiding the things he called
dangerous, to preventing other people from going near them, and perhaps to running
in the opposite direction. If the conclusion were not obviously ridiculous, it would be
easy to infer that a man whose application of the term was different from ours
 Philippa Foot

throughout might say that the oddest things were dangerous without fear of disproof;
the idea would be that he could still be described as ‘thinking them dangerous’, or at
least as ‘warning’, because by his attitude and actions he would have fulfilled the
conditions for these things. This is nonsense because without its proper object warning,
like believing dangerous, will not be there. It is logically impossible to warn about
anything not thought of as threatening evil, and for danger we need a particular kind of
serious evil such as injury or death.
There are, however, some differences between thinking a thing dangerous and feeling
proud, frightened or dismayed. When a man says that something is dangerous he must
support his statement with a special kind of evidence; but when he says that he feels
proud or frightened or dismayed the description of the object of his pride or fright or
dismay does not have quite this relation to his original statement. If he is shown that the
thing he was proud of was not his after all, or was not after all anything very grand, he
may have to say that his pride was not justified, but he will not have to take back the
statement that he was proud. On the other hand, someone who says that a thing is
dangerous, and later sees that he made a mistake in thinking that an injury might result
from it, has to go back on his original statement and admit that he was wrong. In
neither case, however, is the speaker able to go on as before. A man who discovered
that it was not his pumpkin but someone else's which had won the prize could only say
that he still felt proud if he could produce some other ground for pride. It is in this way
that even feelings are logically vulnerable to facts.
It will probably be objected against these examples that for part of the way at least they
beg the question. It will be said that indeed a man can be proud only of something he
thinks a good action, or an achievement, or a sign of noble birth; as he can feel dismay
only about something which he sees as bad, or frightened of some threatened evil;
similarly he can warn only if he is also prepared to speak, for instance, of injury. But
this will limit the range of possible objects of those attitudes and beliefs only if the range
of these terms is limited in its turn. To meet this objection I shall discuss the meaning of
‘injury’ because this is the simplest case. Anyone who feels inclined to say that anything
could be counted as an achievement, or as the evil of which people were afraid, or
about which they felt dismayed, should just try this out. I wish to consider the
proposition that anything could be thought of as dangerous, because if it causes injury
it is dangerous, and anything could be counted as an injury. I shall consider bodily
injury because this is the injury connected with danger; it is not correct to put up a
notice by the roadside reading ‘Danger!’ on account of bushes which might scratch a
car. Nor can a substance be labelled ‘dangerous’ on the ground that it can injure delicate
fabrics; although we can speak of the danger that it may do so, that is not the use of the
word which I am considering here.
When a body is injured it is changed for the worse in a special way, and we want to
know which changes count as injuries. First of all, it matters how an injury comes about;
 VIII Moral Beliefs66F

e.g., it cannot be caused by natural decay. Then it seems clear that not just any kind of
thing will do, for instance, any unusual mark on the body, however much trouble a
man might take to have it removed. By far the most important class of injuries are
injuries to a part of the body, counting as injuries because there is an interference with
the function of that part; injury to a leg, an eye, an ear, a hand, a muscle, the heart, the
brain, the spinal cord. An injury to an eye is one that affects, or is likely to affect, its
sight; an injury to a hand one which makes it less well able to reach out and grasp, and
perform other operations of this kind. A leg can be injured because its movements and
supporting power can be affected; a lung because it can become too weak to draw in the
proper amount of air. We are most ready to speak of an injury where the function of a
part of the body is to perform a characteristic operation, as in these examples. We might
hesitate to say that a skull can be injured, and might prefer to speak of damage to it,
since although there is indeed a function (a protective function) there is no operation.
But thinking of the protective function of the skull we may want to speak of injury here.
In so far as the concept of injury depends on that of function it is narrowly limited, since
not even every use to which a part of the body is put will count as its function. Why is it
that, even if it is the means by which they earn their living, we would never consider
the removal of the dwarf's hump or the bearded lady's beard as a bodily injury? It will
be tempting to say that these things are disfigurements, but this is not the point; if we
suppose that a man who had some invisible extra muscle made his living as a court
jester by waggling his ears, the ears would not have been injured if this were made to
disappear. If it were natural to men to communicate by movements of the ear, then ears
would have the function of signalling (we have no word for this kind of ‘speaking’) and
an impairment of this function would be an injury; but things are not like this. This
court jester would use his ears to make people laugh, but this is not the function of ears.
No doubt many people will feel impatient when such facts are mentioned, because they
think that it is quite unimportant that this or that happens to be the case, and it seems to
them arbitrary that the loss of the beard, the hump, or the ear muscle would not be
called an injury. Isn't the loss of that by which one makes one's living a pretty
catastrophic loss? Yet it seems quite natural that these are not counted as injuries if one
thinks about the conditions of human life, and contrasts the loss of a special ability to
make people gape or laugh with the ability to see, hear, walk, or pick things up. The
first is only needed for one very special way of living; the other in any foreseeable
future for any man. This restriction seems all the more natural when we observe what
other threats besides that of injury can constitute danger: of death, for instance, or
mental derangement. A shock which could cause mental instability or impairment of
memory would be called dangerous, because a man needs such things as intelligence,
memory, and concentration as he needs sight or hearing or the use of hands. Here we
do not speak of injury unless it is possible to connect the impairment with some
 Philippa Foot

physical change, but we speak of danger because there is the same loss of a capacity
which any man needs.
There can be injury outside the range we have been considering; for a man may
sometimes be said to have received injuries where no part of his body has had its
functions interfered with. In general, I think that any blow which disarranged the body
in such a way that there was lasting pain would inflict an injury, even if no other ill
resulted, but I do not know of any other important extension of the concept.
It seems therefore that since the range of things which can be called injuries is quite
narrowly restricted, the word ‘dangerous’ is restricted in so far as it is connected with
injury. We have the right to say that a man cannot decide to call just anything
dangerous, however much he puts up fences and shakes his head.
So far I have been arguing that such things as pride, fear, dismay, and the thought that
something is dangerous have an internal relation to their object, and hope that what I
mean is becoming clear. Now we must consider whether those attitudes or beliefs
which are the moral philosopher's study are similar, or whether such things as
‘evaluation’ and ‘thinking something good’ and ‘commendation’ could logically be
found in combination with any object whatsoever. All I can do here is to give an
example which may make this suggestion seem implausible, and to knock away a few
of its supports. The example will come from the range of trivial and pointless actions
such as we were considering in speaking of the man who clasped his hands three times
an hour, and we can point to the oddity of the suggestion that this can be called a good
action. We are bound by the terms of our question to refrain from adding any special
background, and it should be stated once more that the question is about what can
count in favour of the goodness or badness of a man or an action, and not what could
be, or be thought, good or bad with a special background. I believe that the view I am
attacking often seems plausible only because the special background is surreptitiously
introduced.
Someone who said that clasping the hands three times in an hour was a good action
would first have to answer the question ‘How do you mean?’ For the sentence ‘this is a
good action’ is not one which has a clear meaning. Presumably, since our subject is
moral philosophy, it does not here mean ‘that was a good thing to do’ as this might be
said of a man who had done something sensible in the course of any enterprise
whatever; we are to confine our attention to ‘the moral use of “good” ’. I am not clear
that it makes sense to speak of ‘a moral use of “good” ’, but we can pick out a number
of cases which raise moral issues. It is because these are so diverse and because ‘this is a
good action’ does not pick out any one of them, that we must ask ‘How do you mean?’
For instance, some things that are done fulfil a duty, such as the duty of parents to
children or children to parents. I suppose that when philosophers speak of good actions
they would include these. Some come under the heading of a virtue such as charity, and
they will be included too. Others again are actions which require the virtues of courage
 VIII Moral Beliefs66F

or temperance, and here the moral aspect is due to the fact that they are done in spite of
fear or the temptation of pleasure; they must indeed be done for the sake of some real or
fancied good, but not necessarily what philosophers would want to call a moral good.
Courage is not particularly concerned with saving other people's lives, or temperance
with leaving them their share of the food and drink, and the goodness of what is done
may here be all kinds of usefulness. It is because there are these very diverse cases
included (I suppose) under the expression ‘a good action’ that we should refuse to
consider applying it without asking what is meant, and we should now ask what is
intended when someone is supposed to say that ‘clasping the hands three times in an
hour is a good action’. Is it supposed that this action fulfils a duty? Then in virtue of
what does a man have this duty, and to whom does he owe it? We have promised not to
slip in a special background, but he cannot possibly have a duty to clasp his hands
unless such a background exists. Nor could it be an act of charity, for it is not thought to
do anyone any good, nor again a gesture of humility unless a special assumption turns
it into this. The action could be courageous, but only if it were done both in the face of
fear and for the sake of a good; and we are not allowed to put in special circumstances
which could make this the case.
I am sure that the following objection will now be raised.
‘Of course clasping one's hands three times in an hour cannot be brought under one of
the virtues which we recognise, but that is only to say that it is not a good action by our
current moral code. It is logically possible that in a quite different moral code quite
different virtues should be recognised, for which we have not even got a name.’ I
cannot answer this objection properly, for that would need a satisfactory account of the
concept of a virtue. But anyone who thinks it would be easy to describe a new virtue
connected with clasping the hands three times in an hour should just try. I think he will
find that he has to cheat, and suppose that in the community concerned the clasping of
hands has been given some special significance, or is thought to have some special
effect. The difficulty is obviously connected with the fact that without a special
background there is no possibility of answering the question ‘What's the point?’ It is no
good saying that there would be a point in doing the action because the action was a
morally good action: the question is how it can be given any such description if we
cannot first speak about the point. And it is just as crazy to suppose that we can call
anything the point of doing something without having to say what the point of that is. In
clasping one's hands one may make a slight sucking noise, but what is the point of that?
It is surely clear that moral virtues must be connected with human good and harm, and
that it is quite impossible to call anything you like good or harm. Consider, for instance,
the suggestion that a man might say he had been harmed because a bucket of water had
been taken out of the sea. As usual it would be possible to think up circumstances in
which this remark would make sense; for instance, when coupled with a belief in
magical influences; but then the harm would consist in what was done by the evil
 Philippa Foot

spirits, not in the taking of the water from the sea. It would be just as odd if someone
were supposed to say that harm had been done to him because the hairs of his head had
been reduced to an even number. 3
I conclude that assumption (1) is very dubious indeed, and that no one should be
allowed to speak as if we can understand ‘evaluation’ ‘commendation’ or ‘pro-attitude’,
whatever the actions concerned.

II
I propose now to consider what was called assumption (2), which said that a man might
always refuse to accept the conclusion of an argument about values, because what
counted as evidence for other people did not count for him. Assumption (2) could be
true even if assumption (1) were false, for it might be that once a particular question of
values—say a moral question—had been accepted, any disputant was bound to accept
particular pieces of evidence as relevant, the same pieces as everyone else, but that he
could always refuse to draw any moral conclusions whatsoever or to discuss any
questions which introduced moral terms. Nor do we mean ‘he might refuse to draw the
conclusion’ in the trivial sense in which anyone can perhaps refuse to draw any
conclusion; the point is that any statement of value always seems to go beyond any
statement of fact, so that he might have a reason for accepting the factual premises but
refusing to accept the evaluative conclusion. That this is so seems to those who argue in
this way to follow from the practical implication of evaluation. When a man uses a
word such as ‘good’ in an ‘evaluative’ and not an ‘inverted comma’ sense, he is
supposed to commit his will. From this it has seemed to follow inevitably that there is a
logical gap between fact and value, for is it not one thing to say that a thing is so, and
another to have a particular attitude towards its being so; one thing to see that certain
effects will follow from a given action, and another to care? Whatever account was
offered of the essential feature of evaluation—whether in terms of feelings, attitudes,
the acceptance of imperatives or what not—the fact remained that with an evaluation
there was a committal in a new dimension, and that this was not guaranteed by any
acceptance of facts.
I shall argue that this view is mistaken; that the practical implication of the use of moral
terms has been put in the wrong place, and that if it is described correctly the logical
gap between factual premises and moral conclusion disappears.
In this argument it will be useful to have as a pattern the practical or ‘action-guiding’
force of the word ‘injury’, which is in some, though not all, ways similar to that of moral
terms.

3In face of this sort of example many philosophers take refuge in the thicket of aesthetics. It would be
interesting to know if they are willing to let their whole case rest on the possibility that there might be
aesthetic objections to what was done.
 VIII Moral Beliefs66F

It is clear I think that an injury is necessarily something bad and therefore something
which as such anyone always has a reason to avoid, and philosophers will therefore be
tempted to say that anyone who uses ‘injury’ in its full ‘action-guiding’ sense commits
himself to avoiding the things he calls injuries. They will then be in the usual difficulties
about the man who says he knows he ought to do something but does not intend to do
it; perhaps also about weakness of the will. Suppose that instead we look again at the
kinds of things which count as injuries, to see if the connexion with the will does not
start here. As has been shown, a man is injured whenever some part of his body, in
being damaged, has become less well able to fulfil its ordinary function. It follows that
he suffers a disability, or is liable to do so; with an injured hand he will be less well able
to pick things up, hold on to them, tie them together or chop them up, and so on. With
defective eyes there will be a thousand other things he is unable to do, and in both cases
we should naturally say that he will often be unable to get what he wants to get or
avoid what he wants to avoid.
Philosophers will no doubt seize on the word ‘want’, and say that if we suppose that a
man happens to want the things which an injury to his body prevents him from getting,
we have slipped in a supposition about a ‘pro-attitude’ already; and that anyone who
does not happen to have these wants can still refuse to use ‘injury’ in its prescriptive, or
‘action-guiding’ sense. And so it may seem that the only way to make a necessary
connexion between ‘injury’ and the things that are to be avoided, is to say that it is used
in an ‘action-guiding sense’ only when applied to something the speaker intends to
avoid. But we should look carefully at the crucial move in that argument, and query the
suggestion that someone might happen not to want anything for which he would need
the use of hands or eyes. Hands and eyes, like ears and legs, play a part in so many
operations that a man could only be said not to need them if he had no wants at all.
That such people exist, in asylums, is not to the present purpose at all; the proper use of
his limbs is something a man has reason to want if he wants anything.
I do not know just what someone who denies this proposition could have in mind.
Perhaps he is thinking of changing the facts of human existence, so that merely wishing,
or the sound of the voice, will bring the world to heel? More likely he is proposing to rig
the circumstances of some individual's existence within the framework of the ordinary
world, by supposing for instance that he is a prince whose servants will sow and reap
and fetch and carry for him, and so use their hands and eyes in his service that he will
not need the use of his. Let us suppose that such a story could be told about a man's life;
it is wildly implausible, but let us pretend that it is not. It is clear that in spite of this we
could say that any man had a reason to shun injury; for even if at the end of his life it
could be said that by a strange set of circumstances he had never needed the use of his
eyes, or his hands, this could not possibly be foreseen. Only by once more changing the
facts of human existence, and supposing every vicissitude foreseeable, could such a
supposition be made.
 Philippa Foot

This is not to say that an injury might not bring more incidental gain than necessary
harm; one has only to think of times when the order has gone out that able-bodied men
are to be put to the sword. Such a gain might even, in some peculiar circumstances, be
reliably foreseen, so that a man would have even better reason for seeking than for
avoiding injury. In this respect the word ‘injury’ differs from terms such as ‘injustice’;
the practical force of ‘injury’ means only that anyone has a reason to avoid injuries, not
that he has an overriding reason to do so.
It will be noticed that this account of the ‘action-guiding’ force of ‘injury’ links it with
reasons for acting rather than with actually doing something. I do not think, however,
that this makes it a less good pattern for the ‘action-guiding’ force of moral terms.
Philosophers who have supposed that actual action was required if ‘good’ were to be
used in a sincere evaluation have got into difficulties over weakness of will, and they
should surely agree that enough has been done if we can show that any man has reason
to aim at virtue and avoid vice. But is this impossibly difficult if we consider the kinds
of things that count as virtue and vice? Consider, for instance, the cardinal virtues,
prudence, temperance, courage and justice. Obviously any man needs prudence, but
does he not also need to resist the temptation of pleasure when there is harm involved?
And how could it be argued that he would never need to face what was fearful for the
sake of some good? It is not obvious what someone would mean if he said that
temperance or courage were not good qualities, and this not because of the ‘praising’
sense of these words, but because of the things that courage and temperance are.
I should like to use these examples to show the artificiality of the notions of
‘commendation’ and of ‘pro-attitudes’ as these are commonly employed. Philosophers
who talk about these things will say that after the facts have been accepted—say that X
is the kind of man who will climb a dangerous mountain, beard an irascible employer
for a rise in pay, and in general face the fearful for the sake of something he thinks
worthwhile—there remains the question of ‘commendation’ or ‘evaluation’. If the word
‘courage’ is used they will ask whether or not the man who speaks of another as having
courage is supposed to have commended him. If we say ‘yes’ they will insist that the
judgement about courage goes beyond the facts, and might therefore be rejected by
someone who refused to do so; if we say ‘no’ they will argue that ‘courage’ is being
used in a purely descriptive or ‘inverted comma sense’, and that we have not got an
example of the evaluative use of language which is the moral philosopher's special
study. What sense can be made, however, of the question ‘does he commend?’ What is
this extra element which is supposed to be present or absent after the facts have been
settled? It is not a matter of liking the man who has courage, or of thinking him
altogether good, but of ‘commending him for his courage’. How are we supposed to do
that? The answer that will be given is that we only commend someone else in speaking
of him as courageous if we accept the imperative ‘let me be courageous’ for ourselves.
But this is quite unnecessary. I can speak of someone else as having the virtue of
 VIII Moral Beliefs66F

courage, and of course recognise it as a virtue in the proper sense, while knowing that I
am a complete coward, and making no resolution to reform. I know that I should be
better off if I were courageous and so have a reason to cultivate courage, but I may also
know that I will do nothing of the kind.
If someone were to say that courage was not a virtue he would have to say that it was
not a quality by which a man came to act well. Perhaps he would be thinking that
someone might be worse off for his courage, which is true, but only because an
incidental harm might arise. For instance, the courageous man might have
underestimated a risk, and run into some disaster which a cowardly man would have
avoided because he was not prepared to take any risk at all. And his courage, like any
other virtue, could be the cause of harm to him because possessing it he fell into some
disastrous state of pride. 4 Similarly, those who question the virtue of temperance are
probably thinking not of the virtue itself but of men whose temperance has consisted in
resisting pleasure for the sake of some illusory good, or those who have made this
virtue their pride.
But what, it will be asked, of justice? For while prudence, courage and temperance are
qualities which benefit the man who has them, justice seems rather to benefit others,
and to work to the disadvantage of the just man himself. Justice as it is treated here, as
one of the cardinal virtues, covers all those things owed to other people: it is under
injustice that murder, theft and lying come, as well as the withholding of what is owed
for instance by parents to children and by children to parents, as well as the dealings
which would be called unjust in everyday speech. So the man who avoids injustice will
find himself in need of things he has returned to their owner, unable to obtain an
advantage by cheating and lying; involved in all those difficulties painted by
Thrasymachus in the first book of the Republic in order to show that injustice is more
profitable than justice to a man of strength and wit. We will be asked how, on our
theory, justice can be a virtue and injustice a vice, since it will surely be difficult to show
that any man whatsoever must need to be just as he needs the use of his hands and
eyes, or needs prudence, courage and temperance?
Before answering this question I shall argue that if it cannot be answered, then justice
can no longer be recommended as a virtue. The point of this is not to show that it must
be answerable, since justice is a virtue, but rather to suggest that we should at least
consider the possibility that justice is not a virtue. This suggestion was taken seriously
by Socrates in the Republic, where it was assumed by everyone that if Thrasymachus
could establish his premise—that injustice was more profitable than justice—his
conclusion would follow:
that a man who had strength to get away with injustice had reason to follow this as the
best way of life. It is a striking fact about modern moral philosophy that no one sees any

4 Cp. Aquinas, Summa Theologica, I–II, q. 55, Art. 4.

 Philippa Foot

difficulty in accepting Thrasymachus' premise and rejecting his conclusion, and it is

because Nietzsche's position is at this point much closer to that of Plato that he is
remote from academic moralists of the present day.
In the Republic it is assumed that if justice is not a good to the just man, moralists who
recommend it as a virtue are perpetrating a fraud. Agreeing with this, I shall be asked
where exactly the fraud comes in; where the untruth that justice is profitable to the
individual is supposed to be told. As a preliminary answer we might ask how many
people are prepared to say frankly that injustice is more profitable than justice? Leaving
aside, as elsewhere in this paper, religious beliefs which might complicate the matter,
we will suppose that some tough atheistical character has asked ‘Why should I be just?’
(Those who believe that this question has something wrong with it can employ their
favourite device for sieving out ‘evaluative meaning’, and suppose that the question is
‘Why should I be “just”?’) Are we prepared to reply ‘As far as you are concerned you
will be better off if you are unjust, but it matters to the rest of us that you should be just,
so we are trying to get you to be just’? He would be likely to enquire into our methods,
and then take care not to be found out, and I do not think that many of those who think
that it is not necessary to show that justice is profitable to the just man would easily
accept that there was nothing more they could say.
The crucial question is: ‘Can we give anyone, strong or weak, a reason why he should
be just?’—and it is no help at all to say that since ‘just’ and ‘unjust’ are ‘action-guiding
words’ no one can even ask ‘Why should I be just?’ Confronted with that argument the
man who wants to do things has only to be careful to avoid the word, and he has not
been given a reason why he should not do the things which other people call ‘unjust’.
Probably it will be argued that he has been given a reason so far as anyone can ever be
given a reason for doing or not doing anything, for the chain of reasons must always
come to an end somewhere, and it may seem that one may always reject the reason
which another man accepts. But this is a mistake; some answers to the question ‘why
should I?’ bring the series to a close and some do not. Hume showed how one answer
closed the series in the following passage:
Ask a man why he uses exercise; he will answer, because he desires to keep his health. If you
then enquire, why he desires health, he will readily reply, because sickness is painful. If you
push your enquiries farther, and desire a reason why he hates pain, it is impossible he can
ever give any. This is an ultimate end, and is never referred to any other object.
(Enquiries, Appendix I, V.)
Hume might just as well have ended this series with boredom: sickness often brings
boredom, and no one is required to give a reason why he does not want to be bored,
any more than he has to give a reason why he does want to pursue what interests him.
In general, anyone is given a reason for acting when he is shown the way to something
he wants; but for some wants the question ‘Why do you want that?’ will make sense,
 VIII Moral Beliefs66F

and for others it will not.5 It seems clear that in this division justice falls on the opposite
side from pleasure and interest and such things. ‘Why shouldn't I do that?’ is not
answered by the words ‘because it is unjust’ as it is answered by showing that the
action will bring boredom, loneliness, pain, discomfort or certain kinds of incapacity,
and this is why it is not true to say that ‘it's unjust’ gives a reason in so far as any
reasons can ever be given. ‘It's unjust’ gives a reason only if the nature of justice can be
shown to be such that it is necessarily connected with what a man wants.
This shows why a great deal hangs on the question of whether justice is or is not a good
to the just man, and why those who accept Thrasymachus' premise and reject his
conclusion are in a dubious position. They recommend justice to each man, as
something he has a reason to follow, but when challenged to show why he should do so
they will not always be able to reply. This last assertion does not depend on any ‘selfish
theory of human nature’ in the philosophical sense. It is often possible to give a man a
reason for acting by showing him that someone else will suffer if he does not; someone
else's good may really be more to him than his own. But the affection which mothers
feel for children, and lovers for each other, and friends for friends, will not take us far
when we are asked for reasons why a man should be just; partly because it will not
extend far enough, and partly because the actions dictated by benevolence and justice
are not always the same. Suppose that I owe someone money:
‘ . . . what if he be my enemy, and has given me just cause to hate him? What if
he be a vicious man, and deserves the hatred of all mankind? What if he be a
miser, and can make no use of what I would deprive him of? What if he be a
profligate debauchee, and would rather receive harm than benefit from large
possessions?’ 6
Even if the general practice of justice could be brought under the motive of universal
benevolence—the desire for the greatest happiness of the greatest number—many
people certainly do not have any such desire. So that if justice is only to be
recommended on these grounds a thousand tough characters will be able to say that
they have been given no reason for practising justice, and many more would say the
same if they were not too timid or too stupid to ask questions about the code of
behaviour which they have been taught. Thus, given Thrasymachus' premise
Thrasymachus' point of view is reasonable; we have no particular reason to admire
those who practise justice through timidity or stupidity.
It seems to me, therefore, that if Thrasymachus' thesis is accepted things cannot go on as
before; we shall have to admit that the belief on which the status of justice as a virtue
was founded is mistaken, and if we still want to get people to be just we must

5 For an excellent discussion of reasons for action, see G. E. M. Anscombe, Intention § 34–40.
6 Hume, Treatise book III, part II, sect. 1.
 Philippa Foot

recommend justice to them in a new way. 7 We shall have to admit that injustice is more
profitable than justice, at least for the strong, and then do our best to see that hardly
anyone can get away with being unjust. We have, of course, the alternative of keeping
quiet, hoping that for the most part people will follow convention into a kind of justice,
and not ask awkward questions, but this policy might be overtaken by a vague
scepticism even on the part of those who do not know just what is lacking; we should
also be at the mercy of anyone who was able and willing to expose our fraud.
Is it true, however, to say that justice is not something a man needs in his dealings with
his fellows, supposing only that he be strong? Those who think that he can get on
perfectly well without being just should be asked to say exactly how such a man is
supposed to live. We know that he is to practise injustice whenever the unjust act
would bring him advantage; but what is he to say? Does he admit that he does not
recognise the rights of other people, or does he pretend? In the first case even those who
combine with him will know that on a change of fortune, or a shift of affection, he may
turn to plunder them, and he must be as wary of their treachery as they are of his.
Presumably the happy unjust man is supposed, as in Book II of the Republic, to be a
very cunning liar and actor, combining complete injustice with the appearance of
justice: he is prepared to treat others ruthlessly, but pretends that nothing is further
from his mind. Philosophers often speak as if a man could thus hide himself even from
those around him, but the supposition is doubtful, and in any case the price in vigilance
would be colossal. If he lets even a few people see his true attitude he must guard
himself against them; if he lets no one into the secret he must always be careful in case
the least spontaneity betray him. Such facts are important because the need a man has
for justice in dealings with other men depends on the fact that they are men and not

7I was in this difficulty because I had supposed—with my opponents—that the thought of a good action
must be related to the choices of each individual in a very special way. It had not occurred to me to
question the often repeated dictum that moral judgements give reasons for acting to each and every man.
This now seems to me to be a mistake. Quite generally the reason why someone choosing an A may ‘be
expected’ to choose good A's rather than bad A's is that our criteria of goodness for any class of things are
related to certain interests that someone or other has or takes in those things. When someone shares these
interests he will have reasons to choose the good A's; otherwise not. Since, in the case of actions, we may
distinguish good and bad on account of the interest we take in the common good, someone who does not
care a damn about it may ask why he should be just, and we must take this question seriously, as Plato
did. Nor need we be frightened by the thought that there does not have to be an answer to be given, and
that the rational cooperation of others in moral practices is not to be taken for granted. Things might get
better, not worse, if we recognized that the reasons men have for acting justly and charitably depend on
contingent human attitudes, and the identification of one man with another in society. For then we would
see that it is up to us to cherish these things, and (above all) that it is no good treating people despitefully
and divisively and then demanding morality of them with an alien ‘ought’.
On these topics see ‘Morality as a System of Hypothetical Imperatives’, reprinted in this volume, and also
the Introduction, pp. xiii–xvi.
 VIII Moral Beliefs66F

inanimate objects or animals. If a man only needed other men as he needs household
objects, and if men could be manipulated like household objects, or beaten into a
reliable submission like donkeys, the case would be different. As things are, the
supposition that injustice is more profitable than justice is very dubious, although like
cowardice and intemperance it might turn out incidentally to be profitable.
The reason why it seems to some people so impossibly difficult to show that justice is
more profitable than injustice is that they consider in isolation particular just acts. It is
perfectly true that if a man is just it follows that he will be prepared, in the event of very
evil circumstances, even to face death rather than to act unjustly—for instance, in
getting an innocent man convicted of a crime of which he has been accused. For him it
turns out that his justice brings disaster on him, and yet like anyone else he had good
reason to be a just and not an unjust man. He could not have it both ways and while
possessing the virtue of justice hold himself ready to be unjust should any great
advantage accrue. The man who has the virtue of justice is not ready to do certain
things, and if he is too easily tempted we shall say that he was ready after all.
 Philippa Foot

IX Goodness and Choice 1

Philippa Foot
Abstract: Moral philosophers often say that the special meaning of the word ‘good’ is to be explained by
its connexion with the speaker's choices. This connexion is shown to be generally neither necessary nor
sufficient.

It is often said nowadays that the meaning of the word ‘good’ is to be explained by
talking about some necessary connexion between calling things good and choosing
them. This theory is thought to have the following merits. (1) It seems to distinguish
‘good’ from such terms as ‘yellow’ and ‘square’ without talking about peculiar non-
natural properties. (2) It seems to show how a value judgement in general, and a moral
judgement in particular, is, as Hume insisted, essentially practical, being supposed ‘to
influence our passions and actions, and to go beyond the calm and indolent judgments
of the understanding’. (Treatise, Bk. III, Part 1, Section 1.) (3) It seems to solve the
problem of how ‘good’ can have the same meaning when applied to many diverse
things; the relation to choice being offered as the thread on which these different uses
are all strung. Nevertheless, the subject is full of obscurities and uncertainties, as we can
see if we try to answer questions such as these. (1) Is a connexion with the choices of the
speaker ever a sufficient condition for the use of the word ‘good’, as it would be if a man
could ever call certain things (let us call them A's) good A's merely because these were
the A's which he was thereafter ready to choose? Is there any case in which he could
defend his use of the word ‘good’ by saying, for instance, ‘I committed myself to a
choice’? Or, if ‘committing himself to choosing’ is not thought quite the right condition,
will it do if he fills in more detail, e.g., about the way he feels if he fails to choose these
A's, and how he encourages others to make the same choice? (2) Is it even true to say
that a connexion with the choices of the speaker is a necessary condition of the use of
‘good’, or of the use of this word ‘in its proper evaluative sense’? Might there not be
cases in which calling an A a good A does not commit one to anything at all in the way
of choices, perhaps not even to the admission of any reason for preferring the A's which
are good? And where calling an A a good A and choosing it are connected, will they be
connected in just the same way for all the different things that A is supposed to be?
I shall first deal with the question of choice as a sufficient condition. Are facts about the
speaker's choices ever a sufficient defence for his calling a certain kind of A a good A, no
matter what that kind of A may be? No one, I think, would try to maintain such a view
quite generally; it is certain that the expression ‘a good A’ cannot always be used in this
way. Such a theory could not be right, for instance, for the use of ‘good’ in the

1 ‘Goodness and Choice’ originally appeared in The Aristotelian Society Supplementary Volume, 1961.
 IX Goodness and Choice73F

expression ‘a good knife’; the man who uses these words correctly must use them in
conjunction with particular criteria of goodness: those which really are the criteria for
the goodness of knives. No matter what he may do in the way of choosing knives which
are M he cannot say ‘M knives are good knives’ unless M is a relevant characteristic, or
unless he is prepared to show that M knives are also N knives, and N is a characteristic
of the right kind. He could not, for example, say that a good knife was the one which
rusted quickly, defending his use of the word ‘good’ by showing that he picked out
such knives for his own use. I imagine that almost everyone would agree about this,
saying that there are some cases in which the correct use of the expression ‘a good A’
requires that one set of criteria rather than another should be used for judging the
goodness of the things. But many people who would admit this think that in other cases
any criteria of goodness would be logically possible, so that for some A's the individual
calling an A a good A has to decide for himself which characteristics he will take as
counting in favour of the goodness of an A. The example which they most often have in
mind is that in which ‘A’ is ‘man’ or ‘action’; for it has seemed to them that moral
judgements must be construed in this way. Nevertheless it is supposed that the account
given of the use of ‘good’ in these expressions is one that goes for many other cases as
well; indeed the suggestion often seems to be that this is a typical use of ‘good’, and
sometimes other examples are given, as for instance by Hare when he talks about good
cacti in The Language of Morals. 2 A man is described as having imported a cactus, the
first cactus, into his own country, and it is implied that he can decide which of the
plants he shall count as good cacti, laying down criteria in respect of such things as size
and shape. There is no suggestion of any limits to the criteria which can be criteria of
goodness in cacti, and Hare obviously thinks that this is quite an ordinary case of the
use of the word ‘good’. I shall argue that, on the contrary, it is hard to find any genuine
example of this kind, so that if Hare's account of the ‘good’ in ‘good man’ were correct
then this use of the word would seem to be different from all other cases in which we
speak of a good such-and-such. My thesis is not, of course, that criteria for the goodness
of each and every kind of thing are determined in the same way as they are determined
for such things as knives, but rather that they are always determined, and not a matter
for decision. (The only qualification to this statement comes from consideration of the
range of cases which I shall call ‘competition examples’; these are discussed on pp. 140–
2 infra.)
What is peculiar about a word such as ‘knife’ is that it names an object in respect of its
function. This is not to say (simply) that it names an object which has a function, but
also that the function is involved in the meaning of the word, and I shall call such
words functional in the strong sense. That ‘knife’ is a functional word in the strong
sense is shown by describing another community supposed to have objects exactly like

2 Pp. 96–7.
 Philippa Foot

our knives but never used for cutting. To save complications let us suppose that never
in the history of this community were knife-shaped objects used as we use knives.
These people have objects which look just like knives, are made of the same material
and so on, but they only use them for some quite different purpose, such as marking
plots of land. If asked whether they had knives we should not say, ‘Yes, but they use
them for quite different purposes from us’, but rather ‘No’, explaining that they did as a
matter of fact have things just like our knives, but that these were actually markers. The
equivalent of our word ‘knife’ would not exist in their language just because they had a
word to name these objects indistinguishable by such things as shape and colour from
our knives. This is not, of course, to deny that given the use of knives in a particular
community one can have things which are, so to speak, degenerate knives, used only as
ornaments, hung upon the wall, and perhaps manufactured for this purpose; it is
another matter to suppose that in a community which used knife-like objects only for
the purpose of ornaments the word which names them would be properly translated as
‘knife’.
Where a thing has a function the primary (but by no means necessarily the only)
criterion for the goodness of that thing will be that it fulfils its function well. Thus the
primary criterion of goodness in a knife is its ability to cut well. If a man goes into a
shop and asks for a knife, saying that he wants a good knife, he can be understood as
wanting one that cuts well, and since ‘knife’ is a functional word in the strong sense
‘good knives cut well’ must be held to be some kind of analytic proposition. Moreover,
since ‘cutting’ is the name of an operation with a particular point, different, for
example, from the activity of producing noises by running steel over wood, no one is at
liberty to pick on just any kind of cutting to count as cutting well. We may compare the
activity of writing, and its connexion with the goodness of pens. The word ‘pen’ means
something used in writing, and writing is making a set of marks designed to be read; so
the minimum condition for a good pen is that it writes legibly. One would not know the
meaning of the word ‘pen’ if one did not know that a good pen had to write well, or the
meaning of ‘writing’ without understanding that good writing had to be writing which
it was possible to read. 3
Knives and pens have functions not only because we use them for a central purpose but
also because they are manufactured for a specific use. But it is obvious that there are
examples of words which, without naming manufactured things, are functional in the
strong sense: ‘eye’, for instance, and ‘lung’ are words like this. Moreover words can be
functional in the ‘strong’ sense without naming anything that we ourselves use or need.

3Things are not, of course, really as simple as this paragraph might suggest. Some people (hunters, for
instance) use their knives as much for stabbing as cutting, and cutting will not be the primary function of
their knives. There are also derivative concepts such as ‘palette-knife’ to be explained. But I do not think
that complications of this kind will affect my main point.
 IX Goodness and Choice73F

Any part of a plant or animal may have a function, and often we would refuse to call by
the same name something that played no part, or a quite different part, in the life of the
living thing. Such things as roots and claws are named in this way.
Are there, in fact, any examples of words naming things which have a function without
being functional words in the strong sense? I suppose that this would be the case where
the function of some organ was discovered, but it was not renamed. We ask, for
instance, ‘Has the appendix a function?’, and an affirmative answer would surely give
us a contingent proposition, at least for a time. A good appendix would thereafter be
judged by its ability to perform this function well, but we could not be expected to
know what, if anything, had to be done well in knowing the meaning of the word.
It has been common for philosophers to call all words which allow one to derive criteria
of goodness from their meaning ‘functional words’. This is confusing, for it suggests
either a quite improper use of the word ‘function’ or the denial of clear examples for
which the condition stated in the last sentence is fulfilled. This may be seen by
considering expressions such as ‘good farmer’, ‘good rider’ and ‘good liar’. Neither
‘farmer’, ‘rider’ nor ‘liar’ picks a man out by reference to a function, though of course
they name him in respect of something that he does. It would be comic to speak of the
function of a rider or a liar, and we can only think of a farmer as having a function if we
think of him in some special way, as serving the community. In any ordinary context
we should be puzzled if asked for the function of a farmer, thinking the questioner must
mean something very odd. But although words such as ‘farmer’, ‘rider’ and ‘liar’ are
not functional words, in either the weak or the strong sense, when joined with ‘good’
they yield criteria of goodness as functional words were seen to do. We say that a man
is a good farmer only because of his farming, while what counts as good farming must
be e.g., maintaining crops and herds in healthy condition, and obtaining the maximum
from the soil. What is good farming will naturally vary somewhat from place to place: a
nomadic race could exhaust a piece of land in farming it, though a settled race could
not. But within such limits the standards by which farming is judged depend on the
meaning of the word, since what counts as farming is only something which has a
particular point. If in another community there was an activity (let us call it ‘snodding’),
and people said that a man snodded well, and was a good snodder, when plants and
animals died in his charge, we should not consider translating ‘snodding’ as ‘farming’;
perhaps a snodder would be a black magician or something like that. Similarly, a good
liar must tell convincing or artistic untruths, and the minimum condition of good riding
is an ability to control a horse.
These examples show that the range of words whose meaning determines criteria of
goodness is much wider than that of functional words. And perhaps the range is even
greater than these examples suggest. We might consider whether, for instance,
‘daughter’, ‘father’ and ‘friend’ might not be added to the list. It might seem that since
‘daughter’ simply means ‘female offspring’ nothing is laid down in the language about
 Philippa Foot

the criteria of goodness in daughters, as nothing is laid down about the meaning of
‘good female offspring’—if this has a meaning at all. Is it, however, true that a word in
another language would be translated by our word ‘daughter’ whatever the criteria of
goodness for those named by this name? Does it make sense to say that they speak of
good daughters, but judge daughters on quite different grounds? Of course it will
depend on the structure of a particular society—its economic arrangements for instance,
and its marriage rules—as to just what is expected of daughters, and to this extent at
least what counts as being a good daughter can vary from place to place. But surely it is
wrong to suppose that even anything expected of a female offspring could be part of
what counted when the goodness of a daughter was being weighed. If it were expected,
as in Nazi Germany, that a daughter (like a son) should denounce disloyal parents to
the police, this still could not be part of being a good daughter; a word which combined
with ‘good’ to give this result would be closer to our word ‘citizen’ or ‘patriot’. Only in
the context of a belief that denunciation would lead to regeneration could this be seen
as one of the things by which the goodness of a daughter could be judged.
‘Father’ seems to be another word of roughly the same kind; for a man can only be said
to be a good father if he looks after his children as best he can. Being a good father must
have something to do with bringing up children, and more specifically caring for them.
While opinions may differ as to what is best for children, and while more or less of the
children's care may be assigned to parents in different communities, it is only within
such limits that the criteria of a good father will differ from place to place. If, in a certain
community, a man were said to be a good A in so far as he offered his children up for
sacrifice, ‘A’ could not be translated by our word ‘father’, but would be like ‘citizen’
again, or ‘provider of children for the state’. Similarly, a good friend must be one who is
well disposed towards the man whose friend he is; it makes no sense to say that he
would be a good friend in so far as he cheated the other, or left him in the lurch. One
interesting point about being a good father, daughter, or friend is that it seems to
depend on one's intentions, rather than on such things as cleverness and strength. No
one could be counted a bad father because he failed to make a living, though he tried as
hard as he could. Similarly, the best of daughters might lack the capacity to provide for
aged parents, while good friends are those who want to help us rather than those who
actually have the capacity to succeed. This is not to deny that in such a context much
may be held against people besides the things they intended to do, and that it is hard to
know where to draw the line, but if for instance someone were said to be a bad friend
on grounds of tactless or obtuse behaviour, it would be a fault of character and not mere
lack of cleverness which was involved. On investigation we might decide that ‘father’,
‘daughter’ and ‘friend’ should be called moral terms, especially if we thought that a
wholly good man could not be bad in any of these respects.
We have now described two classes of words whose meaning determines criteria of
goodness for the things they name: functional words such as ‘knife’, ‘pen’, ‘eye’ and
 IX Goodness and Choice73F

‘root’, and non-functional words such as ‘farmer’, ‘liar’ and ‘father’. This is not,
however, the only way in which criteria of goodness are fixed; they can be just as
determinate, leaving the individual as little scope for decision, in cases where it would
be impossible to get them from the meaning of a word. We have mentioned one
example of this kind in speaking of things that have a function, but are not named in
respect of their function, and there will be many more. Take, for instance, the word
‘coal’. The question is ‘What could be counted as good coal?’ and the answer seems to
be that depends on what coal happens to be used for. ‘Happens to be used for’ is the
right expression here, though it would not have been had we still been talking about
knives or pens. For it makes sense to suppose that in some other community coal—the
same stuff—was mined as we mine it but used quite differently, for some magical or
ornamental purpose perhaps. Because of these different purposes, what counts as good
coal here might count as bad coal there, though within either community prospectors
and coal merchants must take these purposes into account in promising their customers
good coal. Within each society the goodness of coal is settled by the purposes for which
coal is used, while outside such a context it is not clear how anyone could talk about
coal as good or bad at all.
There will obviously be many cases of this sort, where the criteria of goodness are fixed
by the use to which the thing is usually put; but there is another range of examples
rather like this where it would be better to speak of the interest we have in something,
and what we expect from it, than the use to which it is put. We do not, for instance, use
works of literature, or not normally, and could not say that it is by their use that the
criteria for their goodness are determined. On the other hand, the interest which we
have in books and pictures determines the grounds on which their excellence is judged.
Just as we cannot consider the question ‘Is this a good piece of coal?’ without taking
into account the use which we have for coal, so we cannot consider the criteria of
goodness in books and pictures without noticing the part which literature and art play
in a civilisation such as ours. First of all, we treat works of art very seriously, and while
we may expect to enjoy reading a good novel or looking at a good picture, we are
prepared to take trouble about them. We study pictures, and may learn new languages
to read works of literature, all of which depends on the fact that we expect a lot from
such things. We expect them to interest us profoundly, and this must have something to
do with the fact that we are not allowed to give as support for a judgement of aesthetic
merit the mere fact that a book passes the time easily, or makes us cry, or cheers us up.
That when we speak about a good book or picture we do very often mean to judge it as
a work of art depends on the rôle which these things have, though perhaps only in the
lives of a minority whose word has become law. Obviously it would be possible to
imagine a society in which things would be quite different, where, for instance pictures
though hung on the walls were never looked at with concentration except by chance.
Pictures might in fact be treated rather as we usually treat wall-paper, as a cheerful
 Philippa Foot

background decoration which should not demand full attention for itself. There are thus
some similarities between ‘a good picture’ and a ‘good piece of coal’, and in neither case
can someone seize on anything he likes as the criteria of goodness and badness,
justifying his use of the word ‘good’ by pointing to his own choices. He might resolve to
read only novels which would send him to sleep, and might choose them only for this
reason, but he could not say that for him this was the characteristic which made a novel
a good novel.
This is not, however, to deny that sometimes a particular man's interests determine
what is good and bad. For it is always possible to speak of an A as being ‘a good A for
X's purposes’, when the criteria of goodness may be quite different from those by which
the goodness of A's is judged. But since it is one thing to speak of ‘a good A’ and
another to speak of ‘a good A for so-and-so's purposes’ this cannot be used to show
anything about the use of the former expression. Nor does it give us a case in which
criteria of goodness have to be decided upon by the man who uses the word ‘good’. He
can choose what he will try to do, which games he will play, and so on; but given his
purposes, and in general his interests, it is a plain matter of fact that particular A's will
be good for his purposes, or from his point of view. In such a context he will be able to
attach the word ‘good’ even to such things as pebbles or twigs, implying ‘for my
purposes’ or ‘for someone playing this game’. But even here it is not by his readiness to
pick up the pebbles that he legitimises his words.
Where then are the cases to be found in which someone can simply decide to take
anything he likes as criteria of goodness for a certain kind of thing? A perfect example
might seem to be that of the show dog pronounced by the judge to be ‘better’ than
others of his breed because of some fancy quality such as length of ears. Here the
standard may be simply conventional; it is decided, in some more or less deliberate
way, that a twist in the tail or an extra inch on the ears shall be the criterion of goodness
in a particular breed. There is usually no point in picking on these characteristics; it is
not a matter of beauty or strength in the animal, but simply what we have decided to
call ‘good’. This case belongs with a number of others which we might call ‘competition
examples’, belonging to the game of ‘Let's see who can do this’. Suppose, for instance,
that people decide, for a joke, to see who can produce something—say the longest
sentence containing only Anglo-Saxon words, or the musical instrument which will
produce the lowest note. Then, although there may be no point in having such
instruments or such sentences, there is a point in producing one once the competition
has been set, and it is by its ability to serve as a good entry in the competition that the
sentence or instrument is then judged good. That long ears can be one of the criteria of
goodness in a spaniel depends entirely on the competition element in the breeding of
show dogs. If we take away this background, and suppose that someone says ‘Good
spaniels have long ears’, it will not be clear what he means. It is true that in such cases a
man who is in a special position can lay down standards quite arbitrarily, but he must
 IX Goodness and Choice73F

be in the position of setting the competition, so that when he says ‘This is to be the mark
of a good X’ he means ‘This is the target you are to try to hit’. In one sense the man who
is allowed to choose the target decides ‘what shall be the criterion of a good shot’, but in
another sense he does not; the criterion of a good shot is that it should hit the target,
and he merely chooses what the target shall be.
What has just been said may be applied to the discussion of statements about good cacti
in The Language of Morals. 4 Hare supposes that someone has just imported the first
cactus into the country, and that since no one has ever seen them before no one has yet
called one cactus good and another bad. But other people bring them in too, and
eventually someone claims that his is a good cactus; he sets up standards for cacti, and
people may set up rival standards of their own. This is all quite possible, but what Hare
does not give us is the necessary background, and as he describes it, it is not clear how
the criteria could be criteria of goodness at all. There is no reference to the fact that a
cactus is a living organism, which can therefore be called healthy or unhealthy, and a
good or bad specimen of its kind. But in that case we must be told what the rôle of the
cacti is to be. If they are to be used as ornaments, then good cacti must possess the kind
of shape and colour that we find pleasing or curious, and the criteria of goodness are
determined by the interest which we have in the things, and not by any standard set up
by the importer. If, on the other hand, cacti are to be collected and grown for cactus
shows, standards may be set, but only by those who are in some special position of
authority. In this case we have a ‘competition example’, and this will hardly seem
suitable as a model for the use of ‘good’ in moral contexts.
The reason why it seems as if choice alone, or some condition connected with choice,
might be a sufficient condition for the use of the word ‘good’ probably lies in certain
facts about the concept of choice. Of course choosing is not just pointing at something;
for while even pointing is not simply extending one's hand in the direction of a certain
object, given our institution of pointing I can say, ‘Point to one of the windows in this
room’ without giving any idea of what if anything is to happen next. Whereas if I said
‘Choose one of the windows in this room’ the question would be ‘For what?’ It would
hardly count as choosing if people made pointing gestures, or touched particular
objects, and nothing followed, or was meant to follow. Nevertheless it may look as if
choosing could be just distinguishing, or marking, or separating out, because we are
used to games in which people are told for instance to choose a card, and they do not
necessarily know what they are choosing it for. But there are two points to notice about
this example:
(1) The context of a card trick does tell us roughly the rôle which the chosen card will
play, and children who were used to choosing sweets and toys, not knowing about
card tricks, might be puzzled when asked to choose a card.

4 Pp. 96–7.
 Philippa Foot

(1) The context of a card trick does tell us roughly the rôle which the chosen card will
play, and children who were used to choosing sweets and toys, not knowing about
card tricks, might be puzzled when asked to choose a card.
(2) Choosing for a rôle which is kept secret is itself a kind of game. But then the whole
point is that X knows the rôle but is keeping it from Y: it is not a case of choosing but
not for anything, but rather of choosing ‘for whatever X has in mind’.
So when one chooses one chooses for a rôle. One chooses something to have, or to
throw away, to sit on, or to put on the mantelpiece; a road to drive down or to have
closed. But now the question arises—will anything that happens as the intended
consequence of the distinguishing gesture do as the rôle for which one is choosing the
thing? Suppose, for instance, that someone said ‘Choose a window pane for me to put
my finger on’ or ‘Choose a pebble to have a handkerchief waved at.’ Could these things
be the object of choice? One is inclined to say that choosing is rather the kind of picking
out for a rôle, which has a ground; we choose one thing rather than another because it
has some advantage in the context, as a flat pebble is best for playing ducks and drakes,
and a heavy stone to stop a car from running downhill. There are, indeed, what we
might call peripheral cases of choosing, when we ‘just pick’ a card from the pack for
instance, or take any one of a number of objects which would serve equally well. But a
word which applies only to such activities would not properly be translated by our
word ‘choice’, and we are inclined to say in such cases that ‘there was nothing to
choose’ between the alternatives. In any case these are certainly not the examples which
we have in mind when we think that there is a very close connexion between choosing
something and calling it good. For in so far as the choosing was supposed to be
choosing of this kind it is not guaranteed that the word ‘good’ will be given an
application by the connexion with choice. However strictly a man might set himself to
choose a card in a certain position in the pack when told to ‘choose a card’ by someone
showing a card trick, we should not be able to describe him as thinking cards in this
position the best cards or the best cards to take unless he were prepared to justify his
choice. And he could do that only by asserting, for instance, that this was the best card
to take in order to beat the trick. He could never say, ‘Those just are, to my mind, the
best ones to take.’
In the beginning we raised two questions about the relation between choosing
something and calling it good, asking whether some connexion with choice (1) could
give a sufficient condition, and (2) was a necessary condition for the application of the
word. The rest of this paper will be about question (2).
It is very often said that the use of ‘good’ in its ‘proper evaluative sense’ commits a man
to choosing the things he calls good in preference to others, or at least that ‘other things
being equal’ this what he must do. Such statements are quite misleading in so far as
they suggest that a man is only at liberty to call an A a good A if he thinks that such A's
are A's he has reason to choose. Consider, for example, someone who is choosing a pen.
 IX Goodness and Choice73F

Since a good pen is one which will write well, he will have reason to choose a good pen
if he wants to write with it and to write well. But what if his purposes are different?
Suppose that he wishes to cover a piece of paper with blots? Perhaps we might say that
he is not choosing a pen to write with, and so not as a pen, hoping to keep the
connexion between good pens and the pens which a man should choose (other things
being equal) if he is choosing a pen as a pen. But there is also the case of the man who
wants a pen to write with, but has some reason for wanting to write untidily or
illegibly, or to whom it does not matter either way. We could hardly say that he was not
choosing a pen as a pen, and he would have no reason to choose a good pen, perhaps
rather the reverse. That most men must have a reason to choose good pens depends on
the purposes which we take for granted in talking about good and bad pens at all: we
cannot suppose that the standard case is that of wanting pens for the creation of blots,
or undecipherable marks without dissociating pens from writing, and changing the
concept pen. The necessary connexion lies here, and not in some convention about what
the individual speaker must be ready to choose if he uses the word ‘good’.
There seems to be an exact parallel in the case where ‘good’ is used with a word such as
‘rider’, naming a man in respect of some activity with a characteristic point. (It is only
where the activity has a point that we can speak of a good—er or a bad; there are no
good foot-wagglers, or tree-touchers, because nothing counts as doing these things
well.) Since the point of riding is pleasure, exercise, or locomotion, riders are judged
good or bad by their ability to do such things as staying in the saddle and controlling
the horse, with additional requirements of elegance and style. Anyone who shares the
typical purposes of the rider has a reason to ride well rather than badly, and will want
to be a good rider rather than a bad. But it is quite possible for someone to be riding
with different purposes in view; perhaps he wants to play the fool and make people
laugh. It is tempting to say that anyone must be called a good rider if he is achieving
what he wants, but on reflection one sees that this is not ‘the language of mankind’.
Good riding is the kind of riding likely to achieve the characteristic purposes of the
rider, not those which a particular individual may happen to have.
In such cases the point of the activity lies in something commonly wanted by the man
who engages in it, but there are other things normally done for hire. Then the one who
hires is the one who has a direct interest in what is done, and will normally want it
done well. For instance, someone going to a tailor, or engaging a teacher, will usually
have reason to choose a good one, though here again we cannot say that there is a
necessary connexion between calling an A a good A and having reasons for choice. For
someone may choose a teacher, and even choose him as a teacher—that is with an
interest in his teaching—without wanting the teaching to be done well. A wicked uncle
choosing a tutor for his nephew would be a case in point.
If this is a correct account of the connexion that there is between calling an A a good A
and choosing it where an A is something that we use, or where ‘A’ is the name of a man
 Philippa Foot

who does something that he or others want done, it is not surprising that the word
‘good’ can sometimes be used without any implications for choice. For where there is
no connexion with anything that we use, or need, or want doing, the A that is called
good may be one that no one has reason to choose. We say, in a straightforward way,
that a tree has good roots, meaning by this that they are well suited to the performance
of their function, serving the plant by anchoring it and drawing moisture out of the soil.
Our interests are not involved, and only someone in the grip of a theory would insist
that when we speak of a good root we commit ourselves in some way to choosing a root
like that. Nor do we need to look for such a connexion in order to find ‘what is
common’ to the different cases where we apply the word ‘good’. Because the root plays
a part in the life of the organism we can say ‘it has a function’, relating what it does to
the welfare of the plant. And good roots are like good eyes, good pens, and many other
things that are good, in being of the kind to perform their function well.
If someone should say that in the expression ‘a good root’ ‘good’ is not used
‘evaluatively’ this would only increase the artificiality of the notion of ‘evaluation’ as
used in moral philosophy, and it would raise a number of awkward problems as well.
For if the ‘good’ in ‘good roots’ is said to lack ‘evaluative meaning’ because good roots
are not things that we should have any reason to choose, then presumably ‘good claws’
and ‘good fangs’ are expressions which must be treated in the same way. But then we
shall be in difficulties over examples such as ‘good eyes’, ‘good muscles’ and ‘good
stomachs’. We could hardly say that the meaning of ‘good’ changes when the word is
applied to those parts of organisms which can belong to our bodies; it would seem
therefore that if ‘good roots’ and ‘good claws’ are non-evaluative expressions, ‘good
eyes’ must be the same.
Nor is it easy to see how a distinction between ‘evaluative’ and ‘non-evaluative’ uses of
‘good’ would be applied in cases such as that of the good pen. I suppose it would be
said that one who, wanting to write illegibly, rejected the pen he called a good pen,
must use the word in its ‘non-evaluative sense’. This case might seem easy to decide,
but what of a man discussing the merits of something like a pen or a motor car in the
abstract, as people often do? What criteria would we introduce for deciding whether he
was using ‘good’ evaluatively or not? Should we take into account his last choice, or his
next choice, or the choices he made in the majority of cases, or what he thought he
would choose if he were to choose just now? If he himself had no use for the kind of
thing under discussion the last clause would lack a definite sense. And in any case it
seems entirely arbitrary, and unimportant, that any particular decision should be made.
The man who calls roots good roots uses ‘good’ in a straightforward sense, which has
nothing to do with the irony or reported speech marked by the use of inverted commas.
So too does the man who calls a pen a good pen, and says that being a good one it is not
one that he has any reason to choose.
 IX Goodness and Choice73F

The conclusion seems to be that a connexion with the choices of the speaker is not a
necessary condition of the use of the word ‘good’ in its ordinary sense. If a man who
calls an A a good A has reason, other things being equal, to prefer it to other A's, this is
because of the kind of thing that an A is, and its connexion with his wants and needs. It
is clear that most people will have reason to choose good pens and good knives if they
are choosing pens and knives, and that they will have reason to try to ride well if they
are riding, and to tell good lies if they are going in for lying. But this is because the
word ‘good’ is here joined to the name of objects or activities which are
characteristically chosen with given ends in view. It is not nearly so easy to see just
when someone will have reason for choosing to be, or to have, a good A where ‘A’ is
e.g., ‘parent’ or ‘daughter’ or ‘friend’. Since the most basic needs and desires are
involved here it is not at all surprising that the word ‘good’ should have an application
in cases like these; but while there will often be obvious reason to be given for choosing
to be, or to have, a good A, when ‘A’ is a word in this range, it is impossible to say
offhand how far these reasons will extend. Anyone may properly ask why he should
choose to have, or to be, the kind of A he has called a good A, and we may or may not
be able to give him a reply.
 Philippa Foot

X Reasons for Action and Desires 1

Philippa Foot
Abstract: This essay engages with Michael Woods’ paper of the same title arguing against the theory that
all reasons for actions derive from wants or desires, relying on the case of moral reasons. It is suggested
that an argument from prudential reasons is stronger.

Mr. Woods wants to disprove a certain theory of what it is for an agent to have a reason
for acting. 2 This theory is the one that makes reasons dependent on desires, saying that
every statement of a reason, when fully spelled out, must contain a reference to
something wanted or desired by the person whose reason it is to be. As an account of
sufficient conditions this is quite implausible, as Mr. Woods shows. Nevertheless it
might truly state a necessary condition, and it is to this suggestion that Mr. Woods
addresses himself. He wants to show, by producing examples, that there are reasons for
acting not dependent on the agent's desires.
Two questions arise, first whether Mr. Woods is right in rejecting the theory he rejects,
and secondly whether his arguments are sound. It will be argued here that the answer
to the first question is ‘yes’ and the answer to the second ‘no’. I shall first give my own
counter-example and then go on to say why I reject those put forward by Mr. Woods.
Prudential reasons seem to me to provide the most obvious counter-examples to the
thesis that all reasons for action depend on the agent's desires. By ‘prudential reasons’ I
mean those having to do with the agent's interests. There are, of course, problems about
the limits of this class, but these need not concern us here. It will be enough to take
some uncontroversial example of a prudential reason.
Let us, then, consider the case of a man who knows he will go hungry tomorrow unless
he goes shopping today. We will suppose that circumstances are normal; he has no
reason for wanting to be hungry tomorrow, and his house is not on fire. He has a
prudential reason for visiting the shops. Now according to the theory we are discussing,
the statement of his reason, if fully spelled out, must contain a reference to one of his
desires. What desire will this be? The desire not to be hungry tomorrow seems the most
obvious candidate. For we may tell ourselves that this is the desire that will be present
in every case. We should pause, however, and ask why we say that the man who has
reason to go shopping today has a desire not to be hungry tomorrow. Do we find
ourselves, in such circumstances, with thoughts and feelings that allow us to speak of
‘desire’? Sometimes this is exactly how it is; the prospect of tomorrow fills us with

1 ‘Reasons for Action and Desires’ originally appeared in The Aristotelian Society Supplementary Volume,
1972
2 Michael Woods, ‘Reasons for Actions and Desires’, Aristotelian Society Supp. Vol., 1972.
 X Reasons for Action and Desires77F

anxiety and apprehension, together with a lively hope that the evil may be averted. But
in most ordinary cases we would have nothing of the kind to report; it is not on such
grounds that we can speak of a desire, and if reasons depended on such things there
would be few reasons indeed. Yet surely we cannot deny that when a man goes
shopping today because otherwise he will be hungry tomorrow he wants, or has a
desire to, avoid being hungry? This is true, but an analysis of the use of expressions
such as ‘want’ and ‘has a desire to’ in such contexts shows that these ‘desires’ cannot be
the basis of the reason for acting. Thomas Nagel, in an excellent discussion of prudence,
has explained the matter in the following way:
That I have the appropriate desire simply follows from the fact that these considerations
motivate me; if the likelihood that an act will promote my future happiness motivates
me to perform it now, then it is appropriate to ascribe to me a desire for my own future
happiness (The Possibility of Altruism (O.U.P., 1970) pp. 29–30).
What we have here is a use of ‘desire’ which indicates a motivational direction and
nothing more. One may compare it with the use of ‘want’ in ‘I want to ’ where only
intentionality is implied. Can wanting in this sense create the reason for acting? It seems
that it cannot. For in the first place the desires of which we are now speaking are to be
attributed to the agent only in case he is moved to action, or would be so moved in the
absence of counteracting reasons or causes, and it is a mistake to set corresponding
limits to the scope of reasons for acting. Moreover a false account is given even of the
cases in which action occurs. For what happens there is that a man is moved to action
by the recognition that he has reason to act. This would be impossible if there were not
reason to be recognised until the agent has been moved. 3
It seems, then, that the agent's present desire to avert the future evil cannot be the basis
of his prudential reason for acting. Would it be more plausible to suggest that it is on
future desires that prudential reasons depend? It might be said, for instance, that
suffering is something we necessarily want away at the time we are suffering, so the
thought of future suffering contains the thought of future desire. Whether or not it is
right to see a connexion as close as this between every prudential consideration and
future desire, the theory seems to put the emphasis in the wrong place. For it is not as
implying a desire, but rather as suffering, that future pain is something we have reason
to avoid. If we postulate the future desire alone it is not at all certain that a reason for
action in the present remains, as we can see by considering the case of further whims.
Suppose we belive today that tomorrow we shall want to do something, which will,
however, be a mere whim. Suppose, moreover, that we do not expect to get pleasure

3This paragraph did not make my point as clearly as it should have done. The fact that if I go to the shops
in order not to be hungry tomorrow I can ipso facto be said to have a desire not to be hungry tomorrow
cannot show that reasons are necessarily based on desires. For it seems reasonable to say that I have the
reason even in the absence of this, or any other, ground for attributing the desire.
 Philippa Foot

from the satisfaction of our whim, and do not think we will feel frustrated or annoyed if
the whim is not gratified. Have we any reason to make provision now for the
satisfaction of this future desire? We are inclined to say that we have not; at the very
least our intuitions become uncertain, and even this suggests that in stripping down to
future desire we have not reached the essential factor in prudential reasoning.
Mr. Woods himself did not discuss prudential reasons except as refuting the suggestion
that reasons must necessarily move to action. I hope he may agree with most of what I
have said up to this point. But whether or not he thinks that prudential reasons could be
used as a counter-example to the thesis that statements of reason, when fully spelled
out, must contain a reference to the desires of the person for whom they are to be
reasons, he himself relied on examples of another kind. He believes that the theory can
be refuted by showing that moral beliefs, aesthetic opinions and ‘evaluative beliefs in
general’ give reasons that are not dependent on the agent's desires. I want to show that
Mr. Woods is mistaken in thinking that evaluative beliefs give reasons in this special
way. They have a special connexion with reasons for action, but this depends on their
special connexions with interests and desires.
How then, do evaluative judgements give reasons for acting? Let us put aside moral
and aesthetic examples for the moment, and look at familiar everyday instances of
evaluation outside these fields. We may think, for instance, about good doctors and
good philosophers, good arguments and good parties, good houses and good land. If a
patient has a reason to choose a good doctor, and a tenant a good house, why is this so?
Obviously it is because the qualities by which a doctor is judged a good doctor, and a
house a good house, are those which make them useful to patients and tenants
respectively. Good qualities are exactly those which are of some interest or use, and it is
not surprising that someone has a reason to choose good F's and good G's where F's and
G's are the kind of things one can choose. In general very many people will have reason
to choose to have good ones, or, in some cases, to choose to be good ones, since the
good will be judged from some standard point of view. But not everyone will have
these reasons, since he may not have standard desires and interests, and may not be the
one whose desires and interests are taken into account. There is no magical reason-
giving force in evaluative judgements, and it would be ludicrous carefully to choose a
good F or a good G, rather than a bad F or a bad G, if one's own desires and interests
were not such as to provide a reason in one's own case.
The reason-giving force of aesthetic judgement has a similar dependence on interests
and desires. A man has reason to read an interesting book just so far as he thinks it will
interest him, or if he expects the reading to be pleasing or profitable in some other way.
The idea that one should have good works of art around just because they are good
works of art, and even if one can get nothing from them, must depend on hopes of
improvement in one's taste. This is probably bad policy, and in any case does nothing to
 X Reasons for Action and Desires77F

support the thesis that aesthetic beliefs provide reasons for action that are independent
of what the agent wants.
Are moral evaluations linked with reasons for acting only when connected with the
agent's interests and desires? There are many who will refuse to admit that it is so,
agreeing with Mr. Woods at least so far as this class of evaluations is concerned. It
seems to them essential that moral considerations give reason for acting to each and
every man, and they rightly say that such universal reasons cannot be provided for
moral action if the basis is interest and desire. Why, however, do they insist that each
and every man whatever his situation and whatever his nature, must have reason to do
what morality requires? Is there perhaps some confusion between the moral judgement,
which undoubtedly stands for any man, and the judgement about reasons which may
not? Or is it that they want to be able to call a wicked, amoral man irrational, as if by
that they can hit him where it will hurt? Some would admit that the cool, calculating,
amoral man is not to be called irrational, but nevertheless say that he has reason to act
morally whatever his interests and desires. This seems to be an inconsistent position, as
we may show in the following way.
Suppose a man to assert these propositions:
A.
1. I have a reason to do
action a.
2. I have no reason not to do
a.
3. I am not going to do a.
From these premises he must conclude
4. I am going to act
irrationally.
On the other hand the following are admitted consistent:
B.
1. I shall act immorally unless I
do a.
2. I have no reason not to do a.
3. I am not going to do a.
4. I am not going to act
irrationally.
We can show by a simple argument that when B. 1–4 are true, A. 1 is false; from which
it follows that B. 1 cannot entail A. 1.
What proof can be given of the special reason-giving force of moral considerations?
What is its source supposed to be? Mr. Woods is impressed by the fact that moral
judgement, and evaluation generally, is connected with ‘intelligible human pursuits
and practices’. But as we saw in discussing non-moral evaluations this will not
 Philippa Foot

guarantee that every man has a reason to choose what is properly called good. Perhaps
good men are those who have qualities useful to society not to themselves; perhaps it is
others who have reason to want them to be good? Or perhaps it is sometimes one and
sometimes another, with no guarantee that on each occasion each man will have reason
to do what morality requires.
Why is it, one may ask, that these particular evaluations—moral evaluations—are
supposed by so many people to be different from the rest? The reason why they should
be thought to be different is not hard to find. For moral judgements are not only
evaluative but also normative. Moral language is used not simply to pick out actions
and qualities which are of interest or use, but also to teach that these actions should be
performed, and these qualities acquired. Moral behaviour is the subject of social
teaching and social demand. There is thus no doubt that people are taught to take moral
considerations as reasons for acting, without any reference to what they want, or what
their interests are. They are to say ‘It's immoral, so I won't do it’ or ‘Justice requires that
I act in this way, so that is what I shall do’. This kind of teaching explains why it should
be thought that moral judgements give reasons for acting which are independent of
interests and desires, but it does nothing to prove that it is so. For we certainly do not
think that independent reasons could be created by social pressures, and elsewhere we
are quite ready to throw over the rules of behaviour that we were taught when we were
young. We were taught, for instance, that we should follow rules of etiquette. Later we
ask whether there is any reason to do what we were taught.
It may be suggested that we are looking for the special connexion between reason for
acting and morality in the wrong place. Perhaps it is to be found not in the form of
moral judgements but rather in their content. Morality, or a large part of morality, has
to do with a man's actions so far as these affect other men for good or ill. If it can be
shown that we have reason to aim at the good of others, as much as we have reason to
aim at our own, then we shall have reason to follow a great many of morality's laws.
Theories have, of course, been constructed along these lines. A version of the argument
was put forward by G. E. Moore in the following well known passage:
. . . The only reason I can have for aiming at ‘my own good’ is that it is good absolutely
that what I so call should belong to me . . . But if it is good absolutely that I should have it,
then everyone else has as much reason for aiming at my having it, as I have myself
(Principia Ethica (Cambridge, 1903) p. 99).
What Moore is saying is that if I have reason to promote state S then it is good that state
S comes about, and if it is good that S comes about then anyone has reason to promote S.
The argument is faulty because there is no such thing as an objectively good state of
affairs. Such constructions as ‘a good state of affairs’, ‘a good thing that p’, are used
subjectively, to mark what fits in with the aims and interests of a particular individual
or group. Divorced from such a background these expressions lose their sense, and an
argument of the type that Moore suggested will not, for this reason, go through.
 X Reasons for Action and Desires77F

Thomas Nagel, who has recently argued the case against rational egoism with much
subtlety, employs what is basically the same argument as Moore's. When anyone has a
reason for bringing anything about there has to be a reason for that thing ‘to happen’,
and Nagel says that in acting for a reason one must be able to regard oneself as
‘promoting an objectively valuable end’ (The Possibility of Altruism (O.U.P. 1970) pp. 96–
7). But if it means nothing to speak of an ‘objectively valuable end’ then there are no
reasons such as Nagel describes and I may say that another has reason to aim at his own
good without implying that I too have reason to promote this end.
It seems, then, that we lack a convincing argument for the special, automatic reason-
giving force of moral judgement, and should be prepared to think that moral
considerations give reasons for action only in ordinary ways. Can we explain how men
do have reasons for acting morally if we confine ourselves to reasons based on interests
and desires? It seems that we can do so, and that without placing too much emphasis on
self-interested reasons for doing what ought to be done. One does not want to deny a
general connexion between virtue and happiness, but no one who acts justly or
charitably only where it pays him to do so will qualify as a just or benevolent man. A
moral man must be ready to go against his interests in the particular case, and if he has
reason to act morally the reason will lie rather in what he wants than in what is to his
advantage. On this basis reasons will exist for many kind and upright men. We readily
accept private affection as giving reasons for actions without the least hint of self-
interest; why should a more extended fellow-feeling not do the same? If a man has that
basic sense of identification with others that makes him care whether or not they live
wretched lives, has he not the best possible reason for charitable action? And would it
not be misrepresentation to speak of this as a charity dependent on the feelings and
inclinations of the moment, since both public and private affections endure through
periods of coldness, and lack of inclination never destroys the reason to act?
The case of justice may be thought to present greater difficulties for such an account of
moral reasons for acting, since justice is notoriously not to be reduced to benevolence,
and a man's affections, however extended, would not necessarily lead him to be just.
Two different types of example have to be considered. There is first the case where the
life or liberty of an individual is set against the good of the majority. This seems to pose
no special problems. There are some who have a special concern for those who are
vulnerable to oppression, and one who is a lover of justice will be a man such as this.
The second case is more puzzling: it is that of Hume's profligate creditor to whom a
debt should be paid though no good is foreseen. Why does anyone think that he has
reason to be honest in such a case? It could be mere superstition. Perhaps we have been
bewitched by the idea that we just do have reason to obey this part of our moral code.
But such a suggestion may well be wrong. It seems that we cannot get on well without
the kind of justice that is ‘without reason’ in the particular case. We need just men who
are prepared to follow rules of justice even where these are not coincident with any
 Philippa Foot

benevolence, and men act justly because they believe that this is so. We should also take
into account, here and elsewhere, the desires that people have to live a certain kind of
life. Of course these desires vary greatly from person to person. One man likes to be
useful; another demands a part in some great or noble cause. Perhaps it will be said that
such people choose the life they choose because they think they ought to do so—because
this is how a man ought to live. But perhaps no such thought, with its problematic
reason-giving force, enters into the matter at all. Without any moral imperatives a man
may have such desires.

Postscript
I am sure that I do not understand the idea of a reason for acting, and I wonder whether
anyone else does either. I myself incline to the view that all such reasons depend either
on the agent's interest (meaning here what is in his interest) or else on his desires. I take
these to be independent sources of reasons for action, so that the fact that a man is
indifferent to his future welfare does not destroy the reason he has for paying attention
to it, but this particular thesis is not one on which I place any importance. Perhaps all
reasons for action are desire-dependent, even if some Humean arguments for it are
faulty.
Is it not, however, possible that if we come to understand reasons for action better we
shall find that some are dependent neither upon the agent's interests nor upon his
desires? Indeed it is possible, and then it is possible that moral considerations will be
found to be reasons of this kind. I do not claim to have shown that it is not so, but rather
that no one has offered any valid argument for the proposition that moral
considerations have an automatic reason-giving force, and that until such arguments
are given it merely expresses a pious hope.
I should add here, that I have never at any time held an egoistic theory of human action,
or thought that all reasons for action were in some sense self-interested. Nor have I
thought that a means-ends model was suitable for all cases of acting for a reason.
 XI Morality as a System of Hypothetical
Imperatives80F

XI Morality as a System of Hypothetical Imperatives 1

Philippa Foot
Abstract: It is commonly supposed that Immanuel Kant, distinguishing categorical imperatives from
hypothetical (or desire based) imperatives, proved that moral judgements give categorical imperatives.
These views are challenged.

There are many difficulties and obscurities in Kant's moral philosophy, and few
contemporary moralists will try to defend it all. Many, for instance, agree in rejecting
Kant's derivation of duties from the mere form of the law expressed in terms of a
universally legislative will. Nevertheless, it is generally supposed, even by those who
would not dream of calling themselves his followers, that Kant established one thing
beyond doubt—namely, the necessity of distinguishing moral judgements from
hypothetical imperatives. That moral judgements cannot be hypothetical imperatives
has come to seem an unquestionable truth. It will be argued here that it is not.
In discussing so thoroughly Kantian a notion as that of the hypothetical imperative, one
naturally begins by asking what Kant himself meant by a hypothetical imperative, and
it may be useful to say a little about the idea of an imperative as this appears in Kant's
works. In writing about imperatives Kant seems to be thinking at least as much of
statements about what ought to be or should be done, as of injunctions expressed in the
imperative mood. He even describes as an imperative the assertion that it would be
‘good to do or refrain from doing something’ 2 and explains that for a will that ‘does not
always do something simply because it is presented to it as a good thing to do’ this has
the force of a command of reason. We may therefore think of Kant's imperatives as
statements to the effect that something ought to be done or that it would be good to do
it.
The distinction between hypothetical imperatives and categorical imperatives, which
plays so important a part in Kant's ethics, appears in characteristic form in the following
passages from the Foundations of the Metaphysics of Morals:
All imperatives command either hypothetically or categorically. The former present the
practical necessity of a possible action as a means to achieving something else which
one desires (or which one may possibly desire). The categorical imperative would be

1 ‘Morality as a System of Hypothetical Imperatives’ originally appeared in The Philosophical Review,

Volume 81, Number 3, July 1972
2 Foundations of the Metaphysics of Morals, Sec. II, trans. by L. W. Beck.
 Philippa Foot

one which presented an action as of itself objectively necessary, without regard to any
other end. 3
If the action is good only as a means to something else, the imperative is hypothetical;
but if it is thought of as good in itself, and hence as necessary in a will which of itself
conforms to reason as the principle of this will, the imperative is categorical. 4
The hypothetical imperative, as Kant defines it, ‘says only that the action is good to
some purpose’ and the purpose, he explains, may be possible or actual. Among
imperatives related to actual purposes Kant mentions rules of prudence, since he
believes that all men necessarily desire their own happiness. Without committing
ourselves to this view it will be useful to follow Kant in classing together as
‘hypothetical imperatives’ those telling a man what he ought to do because (or if) he
wants something and those telling him what he ought to do on grounds of self-interest.
Common opinion agrees with Kant in insisting that a moral man must accept a rule of
duty whatever his interests or desires. 5
Having given a rough description of the class of Kantian hypothetical imperatives it
may be useful to point to the heterogeneity within it. Sometimes what a man should do
depends on his passing inclination, as when he wants his coffee hot and should warm
the jug. Sometimes it depends on some long-term project, when the feelings and
inclinations of the moment are irrelevant. If one wants to be a respectable philosopher
one should get up in the mornings and do some work, though just at that moment
when one should do it the thought of being a respectable philosopher leaves one cold. It
is true nevertheless to say of one, at that moment, that one wants to be a respectable
philosopher, 6 and this can be the foundation of a desire-dependent hypothetical
imperative.
The term ‘desire’ as used in the original account of the hypothetical imperative was
meant as a grammatically convenient substitute for ‘want’, and was not meant to carry
any implication of inclination rather than long-term aim or project. Even the word
‘project’, taken strictly, introduces undesirable restrictions. If someone is devoted to his
family or his country or to any cause, there are certain things he wants, which may then
be the basis of hypothetical imperatives, without either inclinations or projects being
quite what is in question. Hypothetical imperatives should already be appearing as
extremely diverse; a further important distinction is between those that concern an
individual and those that concern a group. The desires on which a hypothetical

3 Ibid.
4 Ibid.
5 According to the position sketched here we have three forms of hypothetical imperative: ‘If you want x

you should do y’, ‘Because you want x you should do y’, and ‘Because x is in your interest you should do
y’. For Kant the third would automatically be covered by the second.
6 To say that at that moment one wants to be a respectable philosopher would be another matter. Such a

statement requires a special connexion between the desire and the moment.
 XI Morality as a System of Hypothetical
Imperatives80F

imperative is dependent may be those of one man, or may be taken for granted as
belonging to a number of people engaged in some common project or sharing common
aims.
Is Kant right to say that moral judgements are categorical, not hypothetical,
imperatives? It may seem that he is, for we find in our language two different uses of
words such as ‘should’ and ‘ought’, apparently corresponding to Kant's hypothetical
and categorical imperatives, and we find moral judgements on the ‘categorical’ side.
Suppose, for instance, we have advised a traveller that he should take a certain train,
believing him to be journeying to his home. If we find that he has decided to go
elsewhere, we will most likely have to take back what we said: the ‘should’ will now be
unsupported and in need of support. Similarly, we must be prepared to withdraw our
statement about what he should do if we find that the right relation does not hold
between the action and the end—that it is either no way of getting what he wants (or
doing what he wants to do) or not the most eligible among possible means. The use of
‘should’ and ‘ought’ in moral contexts is, however, quite different. When we say that a
man should do something and intend a moral judgement we do not have to back up
what we say by considerations about his interests or his desires; if no such connexion
can be found the ‘should’ need not be withdrawn. It follows that the agent cannot rebut
an assertion about what, morally speaking, he should do by showing that the action is
not ancillary to his interests or desires. Without such a connexion the ‘should’ does not
stand unsupported and in need of support; 7 the support that it requires is of another
kind. 8
There is, then, one clear difference between moral judgements and the class of
‘hypothetical imperatives’ so far discussed. In the latter ‘should’ is ‘used
hypothetically’, in the sense defined, and if Kant were merely drawing attention to this
piece of linguistic usage his point would easily be proved. But obviously Kant meant
more than this; in describing moral judgements as non-hypothetical—that is, categorical
imperatives—he is ascribing to them a special dignity and necessity which this usage
cannot give. Modern philosophers follow Kant in talking, for example, about the
‘unconditional requirement’ expressed in moral judgements. These, they say, tell us
what we have to do whatever our interests or desires, and by their inescapability they
are distinguished from hypothetical imperatives.
The problem is to find proof for this further feature of moral judgements. If anyone fails
to see the gap that has to be filled it will be useful to point out to him that we find

7 I am here going back on something I said in an earlier article (‘Moral Beliefs’, reprinted in this volume)
where I thought it necessary to show that virtue must benefit the agent. I believe the rest of the article can
stand.
8 Op. cit., p. 119. See also Foot, ‘Moral Arguments’, p. 105.
 Philippa Foot

‘should’ used non-hypothetically in some non-moral statements to which no one

attributes the special dignity and necessity conveyed by the description ‘categorical
imperative’. For instance, we find this non-hypothetical use of ‘should’ in sentences
enunciating rules of etiquette, as, for example, that an invitation in the third person
should be answered in the third person, where the rule does not fail to apply to someone
who has his own good reasons for ignoring this piece of nonsense, or who simply does
not care about what, from the point of view of etiquette, he should do. Similarly, there
is a non-hypothetical use of ‘should’ in contexts where something like a club rule is in
question. The club secretary who has told a member that he should not bring ladies into
the smoking-room does not say, ‘Sorry, I was mistaken’ when informed that this
member is resigning tomorrow and cares nothing about his reputation in the club.
Lacking a connexion with the agent's desires or interests, this ‘should’ does not stand
‘unsupported and in need of support’; it requires only the backing of the rule. The use
of ‘should’ is therefore ‘non-hypothetical’ in the sense defined.
It follows that if a hypothetical use of ‘should’ gave a hypothetical imperative, and a
non-hypothetical use of ‘should’ a categorical imperative, then ‘should’ statements
based on rules of etiquette, or rules of a club would be categorical imperatives. Since
this would not be accepted by defenders of the categorical imperative in ethics, who
would insist that these other ‘should’ statements give hypothetical imperatives, they
must be using this expression in some other sense. We must therefore ask what they
mean when they say that ‘You should answer . . . in the third person’ is a hypothetical
imperative. Very roughly the idea seems to be that one may reasonably ask why anyone
should bother about what should (from the point of view of etiquette) be done, and that
such considerations deserve no notice unless reason is shown. So although people give
as their reason for doing something the fact that it is required by etiquette, we do not
take this consideration as in itself giving us reason to act. Considerations of etiquette do
not have any automatic reason-giving force, and a man might be right if he denied that
he had reason to do ‘what's done’.
This seems to take us to the heart of the matter, for, by contrast, it is supposed that
moral considerations necessarily give reasons for acting to any man. The difficulty is, of
course, to defend this proposition which is more often repeated than explained. Unless
it is said, implausibly, that all ‘should’ or ‘ought’ statements give reasons for acting,
which leaves the old problem of assigning a special categorical status to moral
judgement, we must be told what it is that makes the moral ‘should’ relevantly different
from the ‘shoulds’ appearing in normative statements of other kinds. 9 Attempts have

9To say that moral considerations are called reasons is blatantly to ignore the problem. In the case of
etiquette or club rules it is obvious that the non-hypothetical use of ‘should’ has resulted in the loss of the
usual connexion between what one should do and what one has reason to do. Someone who objects that
in the moral case a man cannot be justified in restricting his practical reasoning in this way, since every
 XI Morality as a System of Hypothetical
Imperatives80F

sometimes been made to show that some kind of irrationality is involved in ignoring
the ‘should’ of morality: in saying ‘Immoral—so what?’ as one says ‘Not comme il faut—
so what?’ But as far as I can see these have all rested on some illegitimate assumption,
as, for instance, of thinking that the amoral man, who agrees that some piece of conduct
is immoral but takes no notice of that, is inconsistently disregarding a rule of conduct
that he has accepted; or again of thinking it inconsistent to desire that others will not do
to one what one proposes to do to them. The fact is that the man who rejects morality
because he sees no reason to obey its rules can be convicted of villainy but not of
inconsistency. Nor will his action necessarily be irrational. Irrational actions are those in
which a man in some way defeats his own purposes, doing what is calculated to be
disadvantageous or to frustrate his ends. Immorality does not necessarily involve any
such thing.
It is obvious that the normative character of moral judgement does not guarantee its
reason-giving force. Moral judgements are normative, but so are judgements of
manners, statements of club rules, and many others. Why should the first provide
reasons for acting as the others do not? In every case it is because there is a background
of teaching that the non-hypothetical ‘should’ can be used. The behaviour is required,
not simply recommended, but the question remains as to why we should do what we
are required to do. It is true that moral rules are often enforced much more strictly than
the rules of etiquette, and our reluctance to press the non-hypothetical ‘should’ of
etiquette may be one reason why we think of the rules of etiquette as hypothetical
imperatives. But are we then to say that there is nothing behind the idea that moral
judgements are categorical imperatives but the relative stringency of our moral
teaching? I believe that this may have more to do with the matter than the defenders of
the categorical imperative would like to admit. For if we look at the kind of thing that is
said in its defence we may find ourselves puzzled about what the words can even mean
unless we connect them with the feelings that this stringent teaching implants. People
talk, for instance, about the ‘binding force’ of morality, but it is not clear what this
means if not that we feel ourselves unable to escape. Indeed the ‘inescapability’ of moral
requirements is often cited when they are being contrasted with hypothetical

moral ‘should’ gives reasons for acting, must face the following dilemma. Either it is possible to create
reasons for acting simply by putting together any silly rules and introducing a non-hypothetical ‘should’,
or else the non-hypothetical ‘should’ does not necessarily imply reasons for acting. If it does not
necessarily imply reasons for acting we may ask why it is supposed to do so in the case of morality. Why
cannot the indifferent amoral man say that for him ‘should m ’ gives no reason for acting, treating ‘should
m ’ as most of us treat ‘should e ’? Those who insist that ‘should m’ is categorical in this second ‘reason-

giving’ sense do not seem to realise that they never prove this to be so. They sometimes say that moral
considerations ‘just do’ give reasons for acting, without explaining why some devotee of etiquette could
not say the same about the rules of etiquette.
 Philippa Foot

imperatives. No one, it is said, escapes the requirements of ethics by having or not

having particular interests or desires. Taken in one way this only reiterates the contrast
between the ‘should’ of morality and the hypothetical ‘should’, and once more places
morality alongside of etiquette. Both are inescapable in that behaviour does not cease to
offend against either morality or etiquette because the agent is indifferent to their
purposes and to the disapproval he will incur by flouting them. But morality is
supposed to be inescapable in some special way and this may turn out to be merely the
reflection of the way morality is taught. Of course, we must try other ways of
expressing the fugitive thought. It may be said, for instance, that moral judgements
have a kind of necessity since they tell us what we ‘must do’ or ‘have to do’ whatever
our interests and desires. The sense of this is, again, obscure. Sometimes when we use
such expressions we are referring to physical or mental compulsion. (A man has to go
along if he is pulled by strong men and he has to give in if tortured beyond endurance.)
But it is only in the absence of such conditions that moral judgements apply. Another
and more common sense of the words is found in sentences such as ‘I caught a bad cold
and had to stay in bed’ where a penalty for acting otherwise is in the offing. The
necessity of acting morally is not, however, supposed to depend on such penalties.
Another range of examples, not necessarily having to do with penalties, is found where
there is an unquestioned acceptance of some project or rôle, as when a nurse tells us
that she has to make her rounds at a certain time, or we say that we have to run for a
certain train. 10 But these too are irrelevant in the present context, since the acceptance
condition can always be revoked.
No doubt it will be suggested that it is in some other sense of the words ‘have to’ or
‘must’ that one has to or must do what morality demands. But why should one insist
that there must be such a sense when it proves so difficult to say what it is? Suppose
that what we take for a puzzling thought were really no thought at all but only the
reflection of our feelings about morality? Perhaps it makes no sense to say that we ‘have
to’ submit to the moral law, or that morality is ‘inescapable’ in some special way. For
just as one may feel as if one is falling without believing that one is moving downward,
so one may feel as if one has to do what is morally required without believing oneself to
be under physical or psychological compulsion, or about to incur a penalty if one does
not comply. No one thinks that if the word ‘falling’ is used in a statement reporting
one's sensations it must be used in a special sense. But this kind of mistake may be
involved in looking for the special sense in which one ‘has to’ do what morality
demands. There is no difficulty about the idea that we feel we have to behave morally,
and given the psychological conditions of the learning of moral behaviour it is natural
that we should have such feelings. What we cannot do is quote them in support of the

10I am grateful to Rogers Albritton for drawing my attention to this interesting use of expressions such as
‘have to’ or ‘must’.
 XI Morality as a System of Hypothetical
Imperatives80F

doctrine of the categorical imperative. It seems, then, that in so far as it is backed up by

statements to the effect that the moral law is inescapable, or that we do have to do what
is morally required of us, it is uncertain whether the doctrine of the categorical
imperative even makes sense.
The conclusion we should draw is that moral judgements have no better claim to be
categorical imperatives than do statements about matters of etiquette. People may
indeed follow either morality or etiquette without asking why they should do so, but
equally well they may not. They may ask for reasons and may reasonably refuse to
follow either if reasons are not to be found.
It will be said that this way of viewing moral considerations must be totally destructive
of morality, because no one could ever act morally unless he accepted such
considerations as in themselves sufficient reason for action. Actions that are truly moral
must be done ‘for their own sake’, ‘because they are right’, and not for some ulterior
purpose. This argument we must examine with care, for the doctrine of the categorical
imperative has owed much to its persuasion.
Is there anything to be said for the thesis that a truly moral man acts ‘out of respect for
the moral law’ or that he does what is morally right because it is morally right? That
such propositions are not prima facie absurd depends on the fact that moral judgement
concerns itself with a man's reasons for acting as well as with what he does. Law and
etiquette require only that certain things are done or left undone, but no one is counted
as charitable if he gives alms ‘for the praise of men’, and one who is honest only because
it pays him to be honest does not have the virtue of honesty. This kind of consideration
was crucial in shaping Kant's moral philosophy. He many times contrasts acting out of
respect for the moral law with acting from an ulterior motive, and what is more from
one that is self-interested. In the early Lectures on Ethics he gave the principle of truth-
telling under a system of hypothetical imperatives as that of not lying if it harms one to
lie. In the Metaphysics of Morals he says that ethics cannot start from the ends which a
man may propose to himself, since these are all ‘selfish’. 11 In the Critique of Practical
Reason he argues explicitly that when acting not out of respect for moral law but ‘on a
material maxim’ men do what they do for the sake of pleasure or happiness.
All material practical principles are, as such, of one and the same kind and belong
under the general principle of self love or one's own happiness. 12
Kant, in fact, was a psychological hedonist in respect of all actions except those done for
the sake of the moral law, and this faulty theory of human nature was one of the things
preventing him from seeing that moral virtue might be compatible with the rejection of
the categorical imperative.

11 Pt. II, Introduction, sec. II.

12 Immanuel Kant, Critique of Practical Reason, trans. L. W. Beck, p. 133.
 Philippa Foot

If we put this theory of human action aside, and allow as ends the things that seem to
be ends, the picture changes. It will surely be allowed that quite apart from thoughts of
duty a man may care about the suffering of others, having a sense of identification with
them, and wanting to help if he can. Of course he must want not the reputation of
charity, nor even a gratifying rôle helping others, but, quite simply, their good. If this is
what he does care about, then he will be attached to the end proper to the virtue of
charity and a comparison with someone acting from an ulterior motive (even a
respectable ulterior motive) is out of place. Nor will the conformity of his action to the
rule of charity be merely contingent. Honest action may happen to further a man's
career; charitable actions do not happen to further the good of others. 13
Can a man accepting only hypothetical imperatives possess other virtues besides that of
charity? Could he be just or honest? This problem is more complex because there is no
end related to such virtues as the good of others is related to charity. But what reason
could there be for refusing to call a man a just man if he acted justly because he loved
truth and liberty, and wanted every man to be treated with a certain respect? And why
should the truly honest man not follow honesty for the sake of the good that honest
dealing brings to men? Of course, the usual difficulties can be raised about the rare case
in which no good is foreseen from an individual act of honesty. But it is not evident that
a man's desires could not give him reason to act honestly even here. He wants to live
openly and in good faith with his neighbours; it is not all the same to him to lie and
conceal.
If one wants to know whether there could be a truly moral man who accepted moral
principles as hypothetical rules of conduct, as many people accept rules of etiquette as
hypothetical rules of conduct, one must consider the right kind of example. A man who
demanded that morality should be brought under the heading of self-interest would not
be a good candidate, nor would anyone who was ready to be charitable or honest only
so long as he felt inclined. A cause such as justice makes strenuous demands, but this is
not peculiar to morality, and men are prepared to toil to achieve many ends not
endorsed by morality. That they are prepared to fight so hard for moral ends—for
example, for liberty and justice—depends on the fact that these are the kinds of ends
that arouse devotion. To sacrifice a great deal for the sake of etiquette one would need
to be under the spell of the emphatic ‘ought’. One could hardly be devoted to behaving
comme il faut.
In spite of all that has been urged in favour of the hypothetical imperative in ethics, I
am sure that many people will be unconvinced and will argue that one element
essential to moral virtue is still missing. This missing feature is the recognition of a duty
to adopt those ends which we have attributed to the moral man. We have said that he

13It is not, of course, necessary that charitable actions should succeed in helping others; but when they do
so they do not happen to do so, since that is necessarily their aim. (Footnote added, 1977.)
 XI Morality as a System of Hypothetical
Imperatives80F

does care about others, and about causes such as liberty and justice; that it is on this
account that he will accept a system of morality. But what if he never cared about such
things, or what if he ceased to care? Is it not the case that he ought to care? This is
exactly what Kant would say, for though at times he sounds as if he thought that
morality is not concerned with ends, at others he insists that the adoption of ends such
as the happiness of others is itself dictated by morality. 14 How is this proposition to be
regarded by one who rejects all talk about the binding force of the moral law? He will
agree that a moral man has moral ends and cannot be indifferent to matters such as
suffering and injustice. Further, he will recognise in the statement that one ought to care
about these things a correct application of the non-hypothetical moral ‘ought’ by which
society is apt to voice its demands. He will not, however, take the fact that he ought to
have certain ends as in itself reason to adopt them. If he himself is a moral man then he
cares about such things, but not ‘because he ought’. If he is an amoral man he may deny
that he has any reason to trouble his head over this or any other moral demand. Of
course he may be mistaken, and his life as well as others' lives may be most sadly
spoiled by his selfishness. But this is not what is urged by those who think they can
close the matter by an emphatic use of ‘ought’. My argument is that they are relying on
an illusion, as if trying to give the moral ‘ought’ a magic force. 15
This conclusion may, as I said, appear dangerous and subversive of morality. We are
apt to panic at the thought that we ourselves, or other people, might stop caring about
the things we do care about, and we feel that the categorical imperative gives us some
control over the situation. But it is interesting that the people of Leningrad were not
struck by the thought that only the contingent fact that other citizens shared their loyalty
and devotion to the city stood between them and the Germans during the terrible years
of the siege. Perhaps we should be less troubled than we are by fear of defection from
the moral cause; perhaps we should even have less reason to fear it if people thought of
themselves as volunteers banded together to fight for liberty and justice and against
inhumanity and oppression. It is often felt, even if obscurely, that there is an element of
deception in the official line about morality. And while some have been persuaded by
talk about the authority of the moral law, others have turned away with a sense of
distrust. 16

14 See, e.g., The Metaphysics of Morals, pt. II, sec. 30.

15 See G. E. M. Anscombe, ‘Modern Moral Philosophy’, Philosophy (1958) . My view is different from Miss
Anscombe's, but I have learned from her.
16 So many people have made useful comments on drafts of this article that I despair of thanking them all.

Derek Parfit's help has been sustained and invaluable, and special thanks are also due to Barry Stroud.
An earlier version of this paper was read at the Center for Philosophical Exchange, Brockport, N.Y., and
published in Philosophical Exchange (Summer 1971). Footnote 1 is mostly from this paper, and I add here
some other paragraphs that may throw light on the present paper.
 Philippa Foot

My own view is that while there are ends within ethics which one should adopt (as e.g. that one's
children get a good education) there are difficulties about saying that one ‘ought’ to take account of the
general good. For either the ‘ought’ means ‘morally ought’ or ‘ought from a moral point of view’ or else it
does not. If it does we have a tautological principle. If it does not the problem is to know what is being
said. By hypothesis a prudential ‘ought’ is not intended here, or one related to others of the agent's
contingent ends. Nor do we have the ‘ought’ and ‘ought not’ operating within the system of etiquette, or
some system of institutional rules. This ‘ought’—the one in the sentence ‘One ought to be moral’—is
supposed to be free floating and unsubscripted, and I have never found anyone who could explain the
use of the word in such a context. (They are apt to talk about the expressing of resolves, or of decisions,
but it is then not clear why we need the ‘ought’ terminology when ‘I resolve’ and ‘I've decided’ are
already in use.) My own conclusion is that ‘One ought to be moral’ makes no sense at all unless the
‘ought’ has the moral subscript, giving a tautology, or else relates morality to some other system such as
prudence or etiquette. I am, therefore putting forward quite seriously a theory that disallows the
possibility of saying that a man ought (free unsubscripted ‘ought’) to have ends other than those he does
have: e.g. that the uncaring, amoral man ought to care about the relief of suffering or the protection of the
weak. In my view we must start from the fact that some people do care about such things, and even
devote their lives to them; they may therefore talk about what should be done presupposing such
common aims. These things are necessary, but only subjectively and conditionally necessary, as Kant
would put it.
Kant would of course object that I am treating men as if, in the army of duty, they were volunteers, and
this is exactly my thought. Why does Kant so object to the idea that those who are concerned about
morality are joining together with like-minded people to fight against injustice and oppression, or to try
to relieve suffering, and that they do so because, caring about such things, they are ready to volunteer in
the cause? Kant says that there is a kind of conceit involved in such a conception. Why does he think this?
He supposes that in so viewing the matter (in seeing ourselves as volunteers) we are forgetting our
nature as human beings, members of the phenomenal as well as the noumenal world, whose will may be
determined by the moral law but also by desire, so that we do not automatically act as a being with a holy
will would act, and are beings for whom the dictates of reason take the form of a command. Let us agree
that a human being doesn't necessarily have moral ends, and that even when he does his inclination may
be stronger than his moral resolves. Both of these things being true a man could be mistaken either in
supposing that his concern for others could not fail, or in supposing that by merely following his
inclinations he could serve moral causes with no need for resolution or self-discipline. We have already
pointed out that a morality of hypothetical imperatives is not a morality of inclination; resolution and
self-discipline being at least as necessary to achieve moral ends as to achieve anything else. So let us
consider the other suggestion, that the supporter of the hypothetical imperative is failing to recognise that
the desires of even the most moral of men could always change. This charge should be denied. One who
supports the hypothetical imperative does not forget that desires might change; he has simply given up
trying to deal, in advance, with such a contingency by saying to himself that he would still be under
command. That this seems hard to accept is the fact that lies, I think, at the heart of Kantianism in ethics,
and to the neo-Kantianism of those who accept his strictures on the hypothetical imperative though
rejecting the rest of his theory. It will seem to many impossible that one should have nothing to say about
the case where moral concerns have vanished, except, of course, to note the character of the man
concerned, and in the case of other people, to take what measures one can to stop them from doing harm.
Perhaps the greatest fear is of a change in oneself; one wants as it were to make sure one is stuck with the
idea of acting morally whatever one's concerns have become. The move betrays a lack of confidence
which oddly does not often trouble people when their devotion is to causes other than those of morality.
 XI Morality as a System of Hypothetical
Imperatives80F

These suggestions are, of course, directly relevant to Kant's arguments against the hypothetical
imperative, for the same problems arise about the meaning of the things that he says. What, for instance,
does it mean to say that moral rules are categorical commands? In the first place they are not commands
at all, neither commands of men nor commands of God; or rather if they are commands of God it is not as
commanded by God that they are in Kant's sense categorical commands. (This he says explicitly.) What
we actually have are rules of conduct adopted by certain societies, and individuals within these societies;
and Kant is saying that these rules are universally valid. But when we put it this way, in terms of rules,
the difficulty of understanding the notion of universal validity is apparent. (It can no longer mean that
everyone is commanded, which shows that there is some point in denying Kant the picturesque language
of commands.) Kant's thought seems to be that moral rules are universally valid in that they are
inescapable, that no one can contract out of morality, and above all that no one can say that as he does not
happen to care about the ends of morality, morality does not apply to him. This thought about
inescapability is very important, and we should pause to consider it. It is perhaps Kant's most compelling
argument against the hypothetical imperative, and the one that may make Kantians of us all.
There is, of course, a sense in which morality is inescapable. Consider, for instance, moral epithets such as
‘dishonest’, ‘unjust’, ‘uncharitable’; these do not cease to apply to a man because he is indifferent to the
ends of morality: they may indeed apply to him because of his indifference. He is judged by the criteria of
morality when moral character is in question, and Kant is indeed right in saying that these criteria are
independent of his desires. (Contrast a word such as ‘rash’). No one can escape the application of the
moral terms by pleading his indifference. Nor can he escape them by turning to ways of life in which he
can be counted as being neither morally good or morally bad, as he can escape being a good or a bad
husband by simply not marrying, or a good or bad carpenter by refusing to take up the tools. In this
sense, then, morality is inescapable, but this can be accepted, and insisted on, by Kant's opponent, the
defender of the hypothetical imperative. The latter may also agree that the application of such epithets
will often be the vehicle for the expression of opposition, disgust or hatred. It has already been agreed
that with our present ‘non-hypothetical’ use of ‘should’ in moral contexts the application of the moral
‘should’ is also inescapable. But someone who thinks that significant concessions have now been made to
Kant must answer the following question. Has anything been said about the inescapability of morality
which could not also be said about the inescapability of etiquette? For just as a man is immoral if he does
certain things, despite his indifference to the nature and result of his actions, so he is rude or unmannerly,
or one who does what is ‘not done’, whatever his views about etiquette. Since no one says that the rules
of etiquette are categorical imperatives the task must be to explain the additional inescapability belonging
to morality.
We must return, therefore, to the difficulty of discovering what Kant can have meant by saying that
moral rules have objective necessity, are categorical commands, are universally valid, or are binding
upon the will of every free agent. Nothing that we have yet considered has given Kant what he wants,
and one cannot, I think, avoid the following conclusion. Kant's argument that moral rules have a peculiar
and dignified status depends wholly upon his attempt to link moral action with rationality through the
mere concept of the form of law and the principle of universalizability, as interpreted by him. In acting
morally Kant thinks that we do as reason dictates. In acting immorally we are acting irrationally, and if
this is not how Kant puts it, it is what he must show in order to make his point; if it could be proved, then
any man, whatever his desires, could be shown to have reason to act morally, since one has reason to do
what it is rational to do. The difficulty, as everyone knows, is to accept Kant's arguments purporting to
show that morally bad actions are those whose maxim could not belong to a universally legislative will,
and moreover that action according to such maxims is irrational action. These difficulties have been
 Philippa Foot

argued ad nauseam, and I shall not repeat the arguments here. All I would claim to have shown is that no
one who rejects Kant's attempts to derive morality from reason has been given any reason to reject the
hypothetical imperative in morals. It is commonly believed that even if Kant has not shown the
connexion between reason and morality he has at least destroyed the hypothetical imperative. I have
urged that, on the contrary, there is no valid argument against the hypothetical imperative to be found in
Kant should the argument from reason fail. (Footnote added in 1977.)
 XII A Reply to Professor Frankena96F

XII A Reply to Professor Frankena 1

Philippa Foot
Abstract: The author replies to Prof. William K. Frankena's article accusing her of an ‘attack on morality’.
Her articles on ‘Morality and Art’ and ‘Morality as a System of Hypothetical Imperatives’ are discussed,
and her criticism of the doctrine of the categorical imperative is outlined.

Professor Frankena finds himself in a state of bewilderment about my opinions,

particularly about those expressed in two recent articles, ‘Morality and Art’ 2 and
‘Morality as a System of Hypothetical Imperatives’, 3 which he discusses in an article
pointedly entitled ‘The Philosopher's Attack on Morality’. 4 To say, as he himself does,
that he finds these articles ‘somewhat unclear’ seems on the internal evidence to be
something of an understatement; he finds them full of uncertainties, contradictions,
ambiguities and qualifications, and I do not know how he thinks anyone could have
written such stuff.
I will try to be very plain in my reply. First as to the question of whether I was attacking
morality itself, or a certain view of morality: it was definitely the latter. I have on
occasions suggested that morality should be indicted for its ineffectiveness, and I
suppose it was something of this kind that Professor Hare heard, and from which he
concluded that I wanted to reject morality itself. But in the published articles it was the
fictions surrounding morality that I wanted to expose; if Professor Frankena thinks that
a morality such as I describe—a defictionalised morality—is no morality, this is an issue
between us.
That Professor Frankena signally failed to understand me is obvious, and it is not
surprising that he did. In the first place he ran together the things I said in ‘Morality and
Art’ about the relativity of some moral judgements with the things I said there and in
the later article against the doctrine of the Categorical Imperative. In ‘Morality and Art’
I remarked that while we were ready to admit that some judgements of artistic merit
were relative to the taste of an age, we were inclined to deny relativity to any of the
judgements of morality; whereas my own view is that while some moral judgements
may even be strictly provable, from the definition of morality and from the facts, others
at the other end of the spectrum depend on assumptions which we are free to make or
not to make. (I gave as an example the assumption that the human foetus is a human

1 ‘A Reply to Professor Frankena’ originally in Philosophy, Volume 50, 1975 .

2 Proc. Brit. Acad., LVI (1970) . Reprinted in Burnyeat and Honderich (eds.) Philosophy As It Is (Penguin,
1978) .
3 Reprinted in this volume.

4 Philosophy 49, 1974, pp. 345–56 .

 Philippa Foot

being from the moment of conception.) After discussing this issue I went on to what
was explicitly stated to be a further point in the contrast of our attitudes to art and
morality, namely that we tend to think of morality as having an authority over our
actions that artistic doctrines do not; it was the idea of moral judgements as
authoritative or necessarily binding that I took up, and attacked, in ‘Morality as a
System of Hypothetical Imperatives’, where nothing was said about the earlier point.
Professor Frankena must have failed to notice that the two points were different, or he
could not have written, in a purported summary of my views, ‘we pretend that moral
principles and judgements are non-hypothetical, objective, absolute, fixed, universally
applicable, inescapable, categorical, rational, and authoritative in ways or to degrees
that we are not warranted in doing’ 5 and then gone on to say that I ‘qualify’ this view
when I say that some moral judgements are strictly provable. In the first place he has
thrown the two different theses of mine together in concocting the list; and in the
second he finds qualification where there is none. To say that some moral judgements
are provable, from the concepts of morality and from certain facts, is not to qualify the
judgement that some are not so provable. At this point I was contrasting some moral
judgements with others—for instance a judgement about the wickedness of Hitler's
treatment of the Jews with a judgement about abortion. But when I attacked the
doctrine of the Categorical Imperative I was talking about moral judgements in general;
in my view none of them have the kind of authority or binding force that is claimed for
them, and I did not qualify this by saying that some of them do have it. No wonder
Professor Frankena found my articles ‘somewhat unclear’ when he treated the two
separate discussions as discussions of a single point.
This confusion is one source of Professor Frankena's bewilderment. Another and more
important reason why he failed to understand me is that he seems not to have noticed
the distinction, made on pp. 160–1 of ‘Morality as a System of Hypothetical
Imperatives’, between a non-hypothetical use of ‘ought’ (Henceforth NHU) and a non-
hypothetical imperative (NHI). I used the convenient example of etiquette to illustrate
the observation that one may have the first without the second. For we say ‘You ought
to do x’ to make a point of etiquette even if the subject convinces us that it is in no way
ancillary to his desires and interests to do so. Thus we do not withdraw the ‘ought’
statement as we would withdraw a statement, e.g. about the train a man should take, or
the food he ought to eat, in the range of examples which are typically cases of advice. In
my terminology the ‘ought’ of etiquette should therefore be labelled a non-hypothetical
use (NHU). Nevertheless we would hardly deny that judgements of etiquette express
hypothetical imperatives, in the sense that they have no automatic reason-giving force
independent of the agent's interests or desires. My challenge, which Professor Frankena
has done nothing to meet, is that morality should be shown to be different from

5 Op. cit. p. 347.

 XII A Reply to Professor Frankena96F

etiquette in this respect. Of course the two are different in many ways; for one thing
etiquette is conventional as morality is not. But what I want to hear of is a difference
relevant to my particular problem about why we should think that judgements of
etiquette are hypothetical imperatives while judgements of morality are not.
I am sure that my terminology, and particularly my invention of the expression ‘non-
hypothetical use of “ought” ’ was not happy, and I will think up some other name for it.
What I must insist, however, is that no one can convict me of unclarity by saying that it
is not clear what I want to do with ‘non-hypothetical “oughts” ’ (whether to retain or
reject them) when he himself puts the question in this form. For in terms of my
distinction between non-hypothetical uses of ‘ought’ and ‘oughts’ used to express
imperatives that are not hypothetical there are two questions here not one. Does
Professor Frankena want to know if I think we should abolish the non-hypothetical use
(NHU) in morals, or does he want to know if I think we should abolish the non-
hypothetical imperative in morals (NHI)? He should of course mean the former, for if
my thesis is correct the latter does not exist.
Moral judgements are, I say, hypothetical imperatives in the sense that they give
reasons for acting only in conjunction with interests and desires. We cannot change that,
though we could keep up the pretence that it is otherwise. To hang on to the illusion,
and treat moral judgements as necessarily reason-giving, is something I would compare
to a similar choice in the matter of etiquette; and indeed we do find some who treat the
consideration that something is ‘bad form’ or ‘not done’ as if it had a magical reason-
giving force. We may suppose such persons harmless, but can hardly hold them up as
models of rationality. Yet this is just what Professor Frankena does think rational in the
case of moral judgements. He also thinks that morality itself depends on it, to which I
should reply that if this were true, as I do not believe it is, then so much the worse for
morality. What Professor Frankena must do if he is to make a point against me is to
show that it is not an illusion to think morality different in this respect from etiquette.
All he does, however, is to assert that to take moral imperatives as non-hypothetical is
the more rational course, and that the man who does so is morally better than the one
who does not. Unless we are to think that moral goodness depends on philosophical
error, the second point must wait upon proof of the first, and once more I would point
out that none has been offered.
There is, if I am right, no question of choosing whether moral judgements shall be
hypothetical imperatives. They are hypothetical imperatives, and a decision to obey
them blindly will not alter the fact. What we can, however, choose is whether to change
our linguistic usage by dropping the non-hypothetical use of ‘ought’ or whether to
leave things as they are. This does not seem to me a very important matter, so long as
we understand the consequences of each choice. Let us set them out.
1. If we make no change we are left with some uses of ‘ought’ which do not carry with
them the implication of reasons for acting. When we say of some particular individual
 Philippa Foot

that he ought to do something, we may consistently add ‘But he has no reason to do it’.
And this we may find objectionable, once it is brought out into the open.
There are, however, ways of weakening the force of a non-hypothetical use of ‘ought’.
We may, for instance, qualify ‘You ought to do a’ by adding ‘from the point of view of
etiquette’, ‘morally speaking’, and so on. This qualification is, it should be noticed,
stronger than the one that merely removes the implication of conclusive reasons for
acting. The expression ‘By the rules of etiquette’ unambiguously makes the stronger
qualification, as ‘from the point of view of etiquette’ does not.
2. If we decide to give up the non-hypothetical use of ‘ought’ in such areas as etiquette
and morality, in order to preserve the strong connexion between ‘ought’ and reasons for
action, a different connexion will be broken—a connexion between two types of moral
judgement and between two types of judgements about etiquette. As things are we are
inclined to infer some ought-type statements from some statements of the ‘good F’
‘good G’ form; for instance to move from ‘He will act badly if he does a’ to ‘he ought not
to do a’. Will it matter if we can no longer do this? I think not for the following reason.
Moral judgements of the ‘x is a good man’ form ought to be interpreted as showing the
same pattern, so far as the connexion with reasons for action are concerned, as
judgements about good doctors, good friends, good daughters, good citizens, and the
rest; and as I have elsewhere tried to show, propositions of the ‘good F’ ‘good G’ form
do not, in general, have a direct connexion with reasons for choice. 6 One cannot, except
in special cases, infer ‘NM has a reason to do a’ from ‘NM must do a if he is to be a good
such and such’, since this particular individual may have no reason to be a good such
and such. Outside areas such as etiquette and morality there is, therefore, a gap
between judgements of the ‘good F’ type and the reason-bearing judgements about
what an individual ought to do. If judgements about good men, good actions, etc., and
judgements about what a man ought to do morally speaking are seen in their natural
alliances one does not expect a simple entailment relation between one type of
judgement and the other. This, I think, is the true gap that can open between ‘is’ and
‘ought’, but it comes within what has been called ‘evaluation’, and a different account is
to be given of the distinction between, e.g. ‘is a red F’ and ‘is a good F’.
As to the provability of certain moral judgements; what becomes of that if we choose
the second alternative and jettison the non-hypothetical use of ‘ought’? Obviously the
provability of ought judgements would be affected, at least if one were thinking, as I
was, of a proof from facts such as the intention of killing millions of innocent persons,
and not from facts among which were included the agent's interests and desires. But
even if we decided to throw out the non-hypothetical use of ‘ought’ in order to keep the
strong connexion between ‘ought’ and reasons we would not have reason to do the
same in the case of other forms of moral judgement. As I have insisted in several

6 ‘Goodness and Choice’, reprinted here.

 XII A Reply to Professor Frankena96F

articles, no one can deny that he is a wicked or otherwise morally bad man on the
ground that his immorality is not contrary to his interests and desires. And given the
pattern on which these judgements of goodness or badness are to be interpreted there is
no difficulty about this. Professor Frankena is, therefore, wrong in saying that on her
own views ‘Mrs Foot is wrong in thinking that some moral propositions can be proved
from the facts to all men’, 7 even on a restricted interpretation of ‘the facts’. But I think
that Professor Frankena's mistake may here be partly due to my lack of explicitness; I
should have spelled out the difference between judgements of different forms.
Finally I should state, for the record, that I do not hold the view of reasons that
Professor Frankena attributes to me, and I do not think I have ever said anything to
imply that I did. I do not use ‘reason’ for ‘something that tends to move to action’ as he
postulates on p. 351. I believe that a reason for acting must relate the action directly or
indirectly to something the agent wants or which it is in his interest to have, but an
agent may fail to be moved by a reason, even when he is aware of it, and he may also be
moved by something that is not a reason at all, as e.g. by the consideration that
something is contrary to etiquette. Being moved is therefore neither a necessary nor a
sufficient condition of having a reason.

7 Frankena, op. cit., p. 349.

 Philippa Foot

XIII Are Moral Considerations Overriding?

Philippa Foot
Abstract: Dewi Z. Phillips has argued that moral considerations must override other considerations. The
author distinguishes ‘evidential’ from ‘verdictive’ moral considerations, arguing that only the latter are in
any sense overriding.

It is often said that in practical reasoning moral considerations must be overriding

considerations; but I do not myself know of any good defence of this proposition, or
even an exposition which makes the thesis clear. In my article ‘Morality as a System of
Hypothetical Imperatives’ (reprinted in this volume) I said that there was nothing in
this idea that could be used in defence of the doctrine of the categorical imperative, and
that seems to me as true now as it seemed to me then. But I remember at the time
thinking that there was something interesting in the idea that moral considerations are
overriding, and that one should at least explain the pervasiveness of the thought.
I shall therefore consider two passages from a recent article by Professor D. Z. Phillips, 1
which present a weak version of the thesis, by which I mean of course not one that is
weakly defended but one whose claims are modest, and which should, therefore, be
defensible if any version is. Professor Phillips wants to show that to anyone who cares
about morality moral considerations must be ‘the most important of all considerations’.
In one passage Professor Phillips writes:
. . . moral considerations are, for the man who cares for them, the most important of all
considerations . . . we do say that moral considerations are all-important, that not
hurting a person or humiliating him in public, for example, is more important than the
formalities of greeting. (p. 150)
In another he says:
A man may, of course, succumb to temptation and put other matters before his moral
obligations. In that case, however, he feels remorse for what he has done. The demands
of etiquette are . . . conditional in so far as morality may ask us to put them aside. The
demands of morality, however, are unconditional, in that they cannot be put aside for
considerations of another kind. (p. 150)
Since the theses of these two passages differ, in a way that Professor Phillips has not,
perhaps, noticed, one must discuss them separately. The question to ask is what, in each
of the passages, is meant by ‘a moral consideration’? In the first passage we are given
the following examples of moral considerations: it is a moral consideration that if a
certain thing is done a man will be hurt, or again that he will be humiliated in public. In
the second passage, however, it is a ‘demand’ of morality that is a moral consideration,

1 ‘In Search of the Moral Must’, Philosophical Quarterly: vol. 27, no. 107 (April 1977) .
 XIII Are Moral Considerations Overriding?

so the kind of consideration that Professor Phillips now has in mind is, presumably, that
it would be immoral to do a certain thing, or morally permissible, or required. Thus in
the first case it was a piece of evidence that was in question, but here it is a verdict or
moral judgement. Now we can speak quite naturally of a moral consideration in either
case; but it is important that we know which we are talking about at any given point in
the argument, and I shall therefore invent a piece of terminology, calling the first kind
of moral consideration evidential and the second verdictive. That a promise is being
broken, or a man killed or injured, is an evidential moral consideration; that something
immoral is being done is a verdictive moral consideration.
Let us now use this terminology in discussing the passages from Professor Phillips's
article. The first passage quite clearly has to do with evidential moral considerations,
since he illustrates the assertion that ‘moral considerations are, for the man who cares
about them, the most important of all considerations . . . ’ with the example of the hand,
offered where hand-shaking is not (by the rules of etiquette) appropriate. Someone who
cares about morality takes this hand in order to avoid hurting the other's feelings, or
humiliating him in public. The consideration that he would be hurt or humiliated if his
hand were not shaken is, by implication, picked out as a ‘moral consideration’; and if it
is a moral consideration it is obviously a moral consideration of the evidential sort.
So now we must ask whether it is true that evidential moral considerations are, as
Professor Phillips thinks, always seen as more important (by anyone who cares about
morality) than considerations of any other kind. It seems to me clearly false, and I will
bring examples to show this. But it should first be remarked that the subject could be
treated with any rigour only if we first said a good deal about the distinction between
evidential moral considerations and ‘other’ considerations. It is clear for instance that
the fact that something is contrary to the rules of etiquette is not itself supposed to be a
moral consideration. And yet it could be relevant to a moral judgement; and in the right
context almost anything could be: as for instance that what was done was done on a
Friday or by a man whose name began with ‘S’. Yet we do have an intuitive idea of a
moral consideration which corresponds to the notion Professor Phillips seems to have
in mind. For while it is directly relevant to the morality of an action that it would break a
promise or bring death or injury to another, it is only indirectly relevant that it is done
on a Friday. It could, for instance, be indirectly relevant in that action done on a Friday
was apt to create a traffic hazard, and so bring death or injury. 2 Thus a moral
consideration is, roughly, a consideration through which those considerations not called
moral considerations get their relevance to moral judgements. But it would take a good
deal of apparatus before this could be made precise, and perhaps Professor Phillips
would not agree that it caught his idea of a moral consideration. So what I shall do in

2 I am here indebted to an unpublished article by Robert Brandom, of the University of Pittsburgh, which
is the best thing I have seen on the subject of the supposed overridingness of moral considerations.
 Philippa Foot

bringing counter-examples to his thesis so far as it concerns evidential moral

considerations is simply to take his example of a moral consideration, or to use
considerations which I am sure he must think of as such.
Let us concoct a variant of Professor Phillips's own example about the handshaking.
Once again, we shall say, someone is in danger of suffering hurt feelings or humiliation,
but this time the price at which the hurt may be avoided is not the breaking of a rule of
etiquette but rather the spending of a rather large sum of money; perhaps thousands of
dollars or pounds. We may further suppose that the man who would spend the money
doesn't need it, and is going to spend it on something for himself, not on some morally
good cause.
We have, therefore, on one side the moral consideration about the hurt the other man
will suffer, and on the other the non-moral consideration about the financial loss. If
Professor Phillips's thesis were correct, anyone who ‘cared about’ morality would have
to spend the money if he were so placed. But of course no one expects him to. In face of
a sizeable financial consideration a small moral consideration often slips quietly out of
sight.
To find another case serving as a counter-example to this part of Professor Phillips's
thesis we may think of a rule of etiquette which operates on most people so strongly
that it takes precedence even over a rather weighty moral consideration. There is, for
instance, a distinct resistance to the idea that a host or hostess might refuse to serve any
more drinks when the guests have had as much as is good for them given that they
must drive home. In spite of the fact that they might kill or injure someone, which is
surely a moral consideration, the host is not expected to close the bar and refuse to
serve more alcohol as soon as this point has been reached. A strong rule of etiquette
forbids such a course of action, and it is the rule of etiquette that takes precedence in
ninety-nine cases out of a hundred in circles familiar to many of us. But to say that no
one in these circles ‘cares about’ morality would be a bit stiff.
It seems, therefore, that when a moral consideration is understood in this way, as an
evidential moral consideration, the thesis that no one who cares about morality ever
gives considerations of etiquette precedence over moral considerations (or that he feel
remorse if he does so) is quite implausible. So let us turn to the other version of the
thesis. Perhaps what Professor Phillips and others like him are really thinking about is
what I have called verdictive moral considerations. Let us see whether these must be
the most important considerations for anyone who cares about morality.
The first thing to notice is that there have been very many communities in which a strict
code of behaviour incompatible with the moral judgements of that community has
existed and been strictly followed. I am thinking, for instance, of societies in which
duelling codes were taught, and followed, without being reconciled with the prevailing
morality. Dostoevsky describes such a situation in the chapter of The Brothers Karamazov
which treats of the life of Father Zosima. As a young man Zosima had provoked
 XIII Are Moral Considerations Overriding?

another officer, simply because he was jealous of him, and forced him to issue a
challenge. However, realising that his conduct has been disgraceful he waits until the
other man has shot at him and then, when it is his turn to shoot, throws away his pistol
and apologises. The seconds are furious about this breach of the rules, and the
following exchange takes place:
I stood facing them and now I addressed them all seriously: ‘Gentlemen,’ I said, ‘is it
really so surprising these days to meet a man who can admit he has done something
stupid and apologise publicly for the wrong he has done?’
‘But you cannot apologise in the middle of a duel’ my second shouted at me angrily.
Now it may be suggested that the outraged second couldn't have recognised that an
apology rather than a shot was what morality called for, or that he couldn't have ‘cared
about’ morality. Are not these things being said simply to rescue the thesis, which is
now in danger of becoming trivial? Suppose the second did know perfectly well that,
morally speaking, Zosima was right to apologise and refuse to shoot? And supposing
he did care about morality? After all duelling rules impinge rather little on most men's
time and leave plenty over for the pursuit of the moral life.
Perhaps it will be insisted that at least no one could actually say that an action was
immoral and yet that it should be done. But why not? People do say this kind of thing,
and it is only a preoccupation with the ‘should’ of moral judgement that makes anyone
deny it. In fact a stronger word than ‘should’ is often used: a man will say that although
it is wrong to do a certain thing he ‘has to’ or ‘must’ do it in order to stave off disaster to
himself, or his family, or his country. And many people are ready to do what they know
and admit to be wrong for far less than this, as for instance in spending money on
frivolities while other people starve. Professor Phillips will have to say either that such
a man cannot ipso facto ‘care about’ morality, in which case his thesis will once more
become uninterestingly trivial. Or else he will have to say that such men are
succumbing to temptation and are bound to feel remorse. But has he not noticed that it
is quite common for people who do care about morality to admit that some of the things
they do are morally indefensible and never lose a wink of sleep over it? Only a
philosopher could say the kinds of things that Professor Phillips says; and if the
Martians take the writings of moral philosophers as a guide to what goes on this planet
they will get a shock when they arrive.
In view of all this one may wonder why the proposition that moral considerations are
overriding considerations has received so much support. What is it that makes the
thought so persuasive? This is, I think, quite easy to explain. Let us go back to the
contrast between morality and etiquette, which seems to be what most people have in
mind when they argue that moral considerations are overriding. The fact is that there is
a difference between the way the two codes are taught which is of just the right kind to
explain why morality is supposed to be overriding in some significant way.
 Philippa Foot

To make the difference apparent we may think again of Professor Phillips's example
about the handshaking. We all agree, let us suppose, that his view of matter reflects the
prevailing code of conduct about such a case and that the hand is to be shaken not
refused. So the official teaching is that something we describe as contrary to etiquette is
nevertheless to be done; and it is to this way of teaching the prevailing code of
behaviour that we should attend. For what we then notice is that etiquette, unlike
morality, is taught as a rigid set of rules that are on occasion to be broken. We do not, as
we might have done, incorporate the exceptions to rules about handshaking and so on
into the code of etiquette, when it would have appeared as not contrary to etiquette to
shake the hand, where we are taught to do so. And this is why children find the whole
thing so confusing. They are taught that some things are ‘not done’ and then expected
to understand that (of course) such rules may be ignored in an emergency, or even in the
kind of case that Professor Phillips describes. We could have incorporated the
exceptions into the rule of etiquette, but we do not do so. But morality we teach
differently. Moral rules are not taught as rigid rules that it is sometimes right to ignore;
rather we teach that it is sometimes morally permissible to tell lies (social lies), break
promises (as e.g. when ill on the day of an appointment) and refuse help (where the cost
of giving it would be, as we say, disproportionate). So we tend, in our teaching, to
accommodate the exceptions within morality, and with this flexibility it is not surprising
that morality can seem ‘unconditional’ and ‘absolute’. In the official code of behaviour
morality appears as strong because it takes care never to be on the losing side.
And now we may ask whether this kind of ‘absoluteness’ and this difference between
etiquette and morality, can be used to meet the challenge thrown down in my article
‘Morality as a System of Hypothetical Imperatives’ to the doctrine of the categorical
imperative? That it cannot be used is obvious from the fact that any code of conduct
could be taught in the ‘flexible’ rather than the ‘rigid’ way. And as a matter of fact one
can point to a rather surprising example of a non-moral code that is taught in this way. I
am thinking of the way in which it is taught that some things are not to be done because
the doing of them is imprudent. For in spite of the fact that philosophers commonly talk
about prudential judgements as if these merely weighed the interests of the subject, we
do not actually speak of imprudence in this way. That something is foreseeably contrary
to the agent's interests, is not in fact enough to make it imprudent. For if it were so a
man would act imprudently whenever he went against his own interests, even if the
sacrifice were strictly required by virtue or honesty. And we might indeed have taught
that on occasions a man must act imprudently, as we teach that on occasions he must do
what is contrary to etiquette. But since we do not do so we preach against imprudence
as we preach against morality, and not as we preach against etiquette. But once again
this is simply because the exceptions are incorporated into the code, not because
judgements about what is imprudent are ‘absolute’ or ‘unconditional’ in any other way.
 XIII Are Moral Considerations Overriding?

It seems then that the thesis that evidential moral considerations are invariably taken (by
anyone who cares about morality) as more important than other considerations is
simply false. The thesis that verdictive moral considerations are invariably taken as more
important than other considerations is also false. This version of the thesis that moral
considerations are overriding is, however, given the appearance of truth by a genuine
difference in the way in which a moral code and many other codes are taught. The
exceptions to moral rules are built in to the verdictive moral system and so it is taught
that morality is always to be obeyed. With etiquette it is different, and therefore the
rules of etiquette appear as a set of ‘conditional’ commands. 3

3 I am particularly grateful to Rosalind Hursthouse and John Giuliano, for their help with this paper.
 Philippa Foot

XIV Approval and Disapproval1

Philippa Foot

Abstract: Emotivists and prescriptivists try to explain the meaning of sentences expressing moral
judgements in terms of the expression of attitudes, feelings, or resolutions as these could occur in a single
individual. It is suggested that that it is necessary to postulate a particular social setting in order to
understand the concept of approval, in moral judgement as elsewhere.

When anthropologists or sociologists look at contemporary moral philosophy they must

be struck by a fact about it which is indeed remarkable: that morality is not treated as
essentially a social phenomenon. Where they themselves would think of morals first of
all in connexion with moral teaching, and with the regulation of behaviour in and by
society, philosophers commonly take a different starting-point. What the philosopher
does is ask himself what it is to make a moral judgement, or take up a moral attitude,
and he tries to give the analysis in terms of elements such as feeling, action, and
thought, which are found in a single individual. Controversy persists between
emotivists, prescriptivists, and those who have been labelled ‘neo-naturalists’ as to just
which elements are needed and how they must be combined; all are agreed, however,
in looking to the individual for their location. We are first to find out how it must be
with him if he is to think something right or wrong, or to have an attitude such as that
of moral approval; then we may go on, if we choose, to talk of shared moral beliefs and
of the mechanism by which morality is taught. The essentials are found in the
individual; social practices come in at a later stage in the story.
In a way it is strange that so few people question the methodological assumption just
described, since we are well aware that many concepts cannot be analysed without the
mention of social facts. For example it is a commonplace that writing a cheque, or going
bankrupt, is impossible without the existence of particular social institutions such as
banks and debts; no one would think that he could say what it is to write a cheque
without giving an account of the social practices which give these marks on these bits of
paper their significance.
Examples may easily be multiplied. One recognises at once, for instance, that voting is
something that a man can do only in the right social setting. Other people come in to the
matter not just as being voted for, but also as the creators or perpetuators of the
arrangements which make voting possible. Voting requires conventions about what
counts as voting, and voting on a certain side; and a special piece of social stage-setting

1 ‘Approval and Disapproval’ originally appeared in P. M. S. Hacker, ed., Law, Morality, and Society: Essays
in Honour of H. L. A. Hart, Oxford, 1977 .
 XIV Approval and Disapproval106F

is needed for each election in so far as lists of candidates must be drawn up, or the
possible choices established in some other way.
We are, therefore, well aware that the analysis of certain concepts cannot proceed
without the description of social institutions and conventions. Why is it thought that we
can ignore such things in analysing the fundamental concepts of ethics? One answer is
that it is the attitude, not the act, of approval that we speak about in our moral
philosophy, and that it is easier to believe that acts may require an appropriate social
setting than that attitudes may do so. Another answer is that we have in mind the
analysis of emotions such as fear, and that for these concepts the individualistic
assumption seems to be justified. Whatever precisely is to be said about fear, however
such elements as feelings, desires, actions, and thoughts are involved in being afraid, it
does not seem necessary to look beyond the individual in order to understand the
concept. Other people come in as the possible objects of fear, and conventions are
needed for its verbal expression; but there seems to be no reason why a single
individual should not feel fear whatever the social setting in which he finds himself. In
this instance it is, therefore, right to start with the individual, going on, if one chooses,
to shared fears and fears that can be attributed to social groups. No fact about society is
implied by the attribution of fear to an individual as facts about society are implied by
saying that he voted or signed a cheque. There are, therefore, mental concepts for which
the individualist assumption is correct. The question is whether it is also right for the
case of approval and disapproval. Are approving and disapproving more like voting in
this respect, or more like feeling afraid?
The thesis to be put forward in this paper is that it is no more possible for a single
individual, without a special social setting, to approve or disapprove than it is for him
to vote. This will be argued for non-moral approval and disapproval; moral attitudes
will be considered briefly later on.
It is evident that there is one case of approval which is more like voting than feeling fear
so far as the relation to a specific social setting is concerned; this is the case of approval
by an inspector passing e.g. plans for buildings in a city. The distinguishing mark of this
kind of approval (apart from the fact that it is opposed to rejection rather than
disapproval) is that the approving is something done, whether by a performative use of
‘I approve’ or by some other recognised device; if told that the inspector has approved
the plans we can ask ‘When did he do so?’, asking not for the period during which he
had an attitude (which perhaps he never did) but rather for the time at which the act of
approval was performed. For such an act of approval a special social setting is
obviously required. In the first place anyone who approves plans must be appointed to
do so; the appointment may be formal or informal, but it must exist. Someone not
designated as competent in the matter cannot be said to approve the plans; if he
imagines himself to be doing so he must also imagine himself to occupy the relevant
position. Moreover to each inspecting position belong not only conditions of
 Philippa Foot

appointment but also the standards which are to be used by the inspector. He is
required to see if certain standards are met, and this is one of the things that makes his
act one of approving rather than merely giving permission. In some cases he will be
given the tests that he must apply; in others he will rather judge, in his own way, how
far particular ends are served, as a school inspector may license a school on the basis of
his judgement, however arrived at, that the pupils are well educated and cared for.
It is clear that we cannot describe this kind of approving without mentioning the social
practices which create and maintain the position of inspector. There is, it seems, no need
to prove the thesis of this paper so far as the acts of approval are concerned; the
debatable proposition concerns the attitudes not the acts. It does not, however, seem
right to argue from the features of one to the features of the other. No doubt it is not an
accident that we use the same word ‘approval’ in both cases; but who can say just
where the similarities lie?
We must, then, start afresh and ask whether the attitudes of approval and disapproval
can exist only in a determinate social setting? Why is it commonly supposed that
nothing of the kind is required? The main reason is, I think, that approving and
disapproving are thought to be rather like wanting and not wanting. The chief element
in their composition is supposed to be a readiness to work towards or away from some
result, or perhaps from some kind of result. When one approves of the thing one is
thought to favour it, which means working towards it, and the ill-defined term of art
‘pro-attitude’ is used to slide between wanting and approving.
I shall first try to show that such things as promoting and wanting are quite different
from approving, and that wanting and promoting may exist without the possibility of
approving; the same going for the opposites. For this I shall use a number of examples,
all of approval and disapproval on non-moral grounds. Let us consider, for instance,
parents who approve or disapprove of the marriage of one of their children; it is
common for parents to do so, and we understand very well what this means. We notice,
however, that a stranger cannot approve or disapprove of the marriage as the parents
can, and we must ask why this is so. Why is it that I can approve of my own daughter's
marriage, or disapprove of it, but not of the marriage of some girl who is not a relative,
or even a friend? (Perhaps I have read of her in the local paper, or heard of her from a
gossiping neighbour.) No doubt it will be objected that approval and disapproval of a
marriage is not in fact confined to parents or even relatives and friends. For surely
many white men in South Africa disapprove of the marriage of any white man's
daughter to any black man, and surely there are grounds on which even reasonable
people might disapprove of the marriage of some girl with whom they had no special
connexion, for instance because it was a case of child marriage, or a marriage that was
forced upon the girl. This is true, but it should be noticed that these examples all come
within the sphere of public manners and morals. The fact that anyone can approve or
disapprove on such grounds is something we will discuss later on. It is enough for the
 XIV Approval and Disapproval106F

present argument, which confines itself to other areas, that there are grounds on which
a parent can disapprove and a stranger cannot. Parents may disapprove of their
daughter's marriage because the man is too old for her, or not rich enough, or not well
enough connected; they may disapprove because they think that the marriage will not
work out well, or because the family's honour or pride is at stake. Now the very same
opinions may be held by someone unconnected with the family. Why is it that this other
person cannot disapprove? One might think that it is because the stranger will not care
about the girl's fortunes, or the fortunes of the family. But this may not be true. One
may very well care about what happens to a stranger, either through a kindly
disposition, or because there is some other factor at work. Then one will hope that the
marriage will not take place and be sorry if it does; but it will still be wrong to say that
one approves or disapproves. Nor is it a matter of having or not having actual power.
Perhaps the parents are not able to influence what happens, and perhaps the stranger
has this power; he decides to prevent the marriage, and by some means or other,
perhaps by an anonymous letter to one of the parties, he is able to do so.
Nor does this appear to be an isolated case, which might be thought to depend on some
ancient ritual aspect of parental approval. One can find other examples with the
greatest of ease. Suppose, for instance, that an elderly couple are thinking of retiring to
some seaside resort, and the evidence is against the success of the venture. A relative or
close friend who knows how things are likely to turn out may therefore disapprove of
the idea. But what about a stranger? Or what about one who though a neighbour is not
a friend? He too is of the opinion that if they go they will regret it, and being of a kindly
disposition he cares, and may actually care more than the friends and relations, about
whether they are unhappy or not. Moreover, he may, by chance, be in a better position
than the others to make a difference to what the old people do. Perhaps there is only
one bungalow in the seaside town, and he can buy it himself; or perhaps he can send an
anonymous letter that will frighten them off. Even with the caring, and the power, we
still cannot say of him that he disapproves of their idea unless, surreptitiously, we
introduce some new rôle for him, such as that of the family doctor or the social worker
in charge of elderly persons' welfare. If, not having such a position, he went and offered
them advice he would merely be impertinent or officious; if, however, he said that he
disapproved he would be saying something that could not, in the circumstances, be
true.
Exactly the same point, about some people being able to approve or disapprove where
others cannot, is illustrated by examples in which some but not others are involved in a
particular enterprise, like giving a dinner party, or robbing a bank, or where some but
not others belong to a particular institution like a college, or support an organisation
such as a sporting club. If my team captain chooses a certain player I may approve or
disapprove of his choice; if your team captain chooses a player I may be glad or sorry,
 Philippa Foot

but barring a special circumstance such as bribery or blackmail I cannot approve or

disapprove. 2
It seems, therefore, that there is this first difference between approval and disapproval
and, e.g., wanting and liking. What anyone can want or like is not restricted, logically
speaking, by facts about his relationship to other people, as for instance that he is the
parent or friend of one, and engaged in a joint enterprise with another. Such facts can,
however, create possibilities of approving and disapproving that would otherwise not
exist.
Another contrast lies in the fact that approval and disapproval are logically connected
with specific opinions as wanting and liking are not. It would be, at most, a requirement
of rationality that one wanted or liked things for a reason, whereas attitudes of approval
and disapproval are impossible without grounds. One could not say that one
disapproved of the marriage, the retirement plans, the nomination of a player, or the
choice of guests, but that one had no reason for doing so. And only some considerations
will count as grounds for approval or disapproval in each case. A parent cannot
disapprove of the marriage of one of his children on the grounds that a neighbour, with
no particular standing in the matter, will be annoyed by it, or that his, the parent's,
income will be somewhat reduced. If he disapproves he must hold an opinion about the
way the marriage will work out for his son or daughter, or about the way the family
fortunes will be affected by it. Similarly, the relatives or friends of the elderly couple can
disapprove of their projected move on the ground that they will be unhappy, but not
because a neighbour will be put out by the sale of their house. With special assumptions
these, too, could be grounds of disapproval of a moral kind, but for the moment we are
putting morality aside. In the case of the joint enterprise the grounds of approval and
disapproval are similarly limited but now the restriction is that they must have to do
with the success or failure of the enterprise or with other shared ends of the
participants. If we are giving a joint dinner party I may disapprove of your choice of
guests or food or wine if I maintain that these are unfavourable to the success of the
party, or, e.g., that they will ruin us financially. I may, however, have some reason of
my own for disliking and working against some idea of yours for the dinner party
which is nevertheless not a ground of disapproval. Perhaps some friend of mine in
another town will be annoyed by the fact that we did not take his advice about where to
eat, and I do not want to offend him. It does not follow that I can disapprove of what
you have in mind.
Incidentally, examples such as these help one to get away from the idea that the
grounds of disapproval must be lofty, or at least respectable. If two men are engaged

2Someone who resolutely denies what is said here might be asked if he really thinks that an American
might disapprove of the British Government's bill giving a measure of Home Rule to Scotland.
 XIV Approval and Disapproval106F

together in some murderous project one may disapprove of the other's choice of
assassins because the man chosen is not ruthless enough for the job.
The immediately preceding paragraphs have shown that approval and disapproval are
possible attitudes only in one who has an opinion from a determinate range. The
opinion itself is not, however, enough any more than the right position or relationship
would be. Everyone has opinions about what is and is not a good idea in the matter of
marrying, house-buying, the choosing of players, the running of dinner parties, and a
thousand other matters of everyday life. And each could be a ground of approval or
disapproval given the right context. We do not, however, say that everyone has
thousands of approvals and disapprovals, one to each of his everyday maxims. The
alone should give pause to anyone who wants to insist that he does disapprove of
retirement to the seaside, of putting a long and a short man in a team together, of
inviting to a dinner party guests who see each other every day, or whatever it may be
that he thinks not a good idea.
Having seen something of the part played by opinion in making approval and
disapproval possible, we may now add another to our list of the special positions which
may enable someone to approve or disapprove. So far we have said nothing about the
position of experts, but this is obviously one which can allow one to approve and
disapprove where others cannot. I am thinking, naturally, of expertise in some practical
field such as the curing of illnesses or mending of motor cars. The doctor can approve
and disapprove of treatments as the layman cannot, and the mechanic is in a similar
position vis à vis the design of an engine or the mending of a car. Anyone can, of course,
disapprove of a particular form of medical treatment if moral or other social issues are
involved: for instance of the treatment of a patient against his will. And anyone can
disapprove of the practice of dishonest mechanics who ‘mend’ a car in such a way that
it is temporarily in running order but liable to break down at any moment. But as before
we are leaving such considerations aside. What a person without medical expertise
cannot do is to disapprove of a form of treatment on purely medical grounds, any more
than one who knows little or nothing about car engines can approve or disapprove of
something done to mend a car on the grounds that the mechanic can. And the point is
not that he cannot have the opinion which they have; even if he happens to have it that
it is not enough to allow him to approve or disapprove.
It may be objected that one sometimes says to someone not in fact expert on motors ‘Do
you approve of my new car?’ or to one not versed in fashion or dress designing ‘Do you
approve of my outfit?’ But when we notice the point of doing so the examples turn out
to be ones which support the ‘special position’ thesis. ‘Do you approve?’ is, in such
circumstances, a complimentary and friendly thing to say, and the reason it is so is the
pretence that the conditions for approval are fulfilled. We are treating the person as if
he were an expert, or alternatively as a close friend to whose scrutiny purchases would
naturally be brought.
 Philippa Foot

It seems, then, that the position of an expert is one of those giving the standing
necessary for approving or disapproving in non-moral matters. If a doctor thinks a form
of treatment a bad form of treatment he can be said to disapprove of it. His disapproval
is general, and can be applied to any case without the need for a special relationship.
The position of an expert is like that of close friend or relation, or partner in a joint
enterprise, in that it is enabling with regard to approval and disapproval. It is unlike the
other positions in giving standing for approvals and disapprovals of a general kind. To
illustrate the difference we may use an example in which expertise but also public
policy is involved. An economist may, for instance, disapprove of certain kinds of taxes
both in general and in any particular case. I, on the other hand, can disapprove of a
measure of taxation only in my country. Nor is it hard to see why I should be able to do
so if we notice that public policy in my community is like policy in a joint enterprise in
which I am taking part.
Let us now ask how all this is related to the thesis that approval and disapproval is
dependent on a specific social setting? Has this been proved, at least for the range of
approvals and disapprovals to which we have so far confined our attention? In fact that
thesis has not yet been argued even for this set of cases. We have indeed seen that the
possibility of approval and disapproval may depend on what could, I suppose, be
called a social fact, as that A and B are engaged together in a joint enterprise. But this
was not the kind of thing that I had in mind in suggesting this thesis. I was thinking of
social practices that might underlie this and all other forms of approval and
disapproval.
To broach this topic let us consider again what it is to approve or disapprove, still
thinking of the non-moral cases, but mostly because they seem to present less difficulty
than the others. So far we have said something about the occasions for possible
approval or disapproval, but nothing of its consequences. It is as if we had said who is
appointed inspector, and what standards he must use in his inspection, but nothing
about the fact that, by and large, it is only the plans that he approves that are put into
effect. It is obvious that in the case of the inspector and his acts of approval it doesn't
just happen that approval has these consequences. Without the agreement that gives his
words or other actions these consequences, acts of approval would not exist. What I
want to consider is whether something of the same kind may not be true for the
attitudes of approval and disapproval? Obviously the same consequences are not in
question for the attitudes and the act. The inspector disallows plans which he refuses to
approve, and no one would suggest that this kind of command is the general rule
where the attitude is concerned. What is the general rule is rather some kind of
influence. Even this may be too strong a word; we should rather say that those able to
approve or disapprove are as a general rule taken account of or listened to. And we
must notice that ‘as a general rule’ is not put in for caution's sake, but rather to imply
that there may well be instances of the attitude which do not have any such feature,
 XIV Approval and Disapproval106F

either because it is both unexpressed and unnoticed, or for some other reason. Parents,
we said can approve or disapprove of their offspring's marriage; it does not follow that
some parents are not totally disregarded in decisions of this kind. Going back to
inspectors and acts of approval we see that there is nothing nonsensical about the idea
of an individual inspector so personally feeble or badly situated that none of his acts of
approval made any difference to what was done. Nevertheless it is impossible to
describe the concept of an inspector, or an act of approval, without talking about the
practice of building the buildings he approves rather than those he turns down.
Nobody will deny that when other people approve or disapprove of what we are
thinking of doing, or have done, we do take account of it. The question is whether it is a
contingent fact that we do so, or whether it is not. On one view of the matter we could
have had approval or disapproval—the attitudes—without any presumption that this
would make any difference to what happened, rather as we could in theory have
wanting which no one else took any account of. On the other view, which is the one for
which I am arguing, the attitudes of approval and disapproval would not be what they
are without the existence of tacit agreement on the question of who listens to whom and
about what. On this view, where we have approval or disapproval we necessarily have
such agreements, though it may not be necessary that we have just the agreements that
we do. Thus it may not be a necessary fact about approval and disapproval that, e.g.,
relations are able to approve where others are not. It could be that different classes of
persons were able to approve and disapprove, or even that anyone could approve or
disapprove of anyone in respect of anything. But the society just described would
necessarily have very different social practices from ours. For if it is true that where
there is approval and disapproval there must be the understanding that as a general
rule other people's views are taken account of there would have to be a lot more taking
account of other people's views in that society than there is in ours.
That approval and disapproval can, logically speaking, exist only against a background
of agreement about the part that other people's views shall be given in decision making,
seems to me to be correct. It seems that it is anomalous to join a supposition about
approval and disapproval with statements such as ‘no one is expected to take any notice
of that’. And if this is not obvious it is because the presumption that account will be
taken is only a presumption, and may be destroyed in an individual case. What may
destroy it is for instance the fact that some individual forfeits our respect by his
habitually foolish opinions.
It will not be surprising if many find this thesis unacceptable on account of a picture
that they have of what an attitude is. An attitude, they say, is something mental, and
who can rule out in advance the possibility of any mental occurrence or state in any
social setting? Suppose that in the ‘wrong’ social setting, where there was no
understanding that others were to be listened to, some individual just did have the
attitude of approval or disapproval? (The ‘wrong’ social setting is either that in which
 Philippa Foot

there is no presumption that anyone will listen to anyone about anything, or else one
which does not accord the right standing to someone placed as he is placed.) The idea is
that whatever the social background this might be how it was with him. The proper reply
for one who agrees with my thesis is that the hypothesis is impossible: that couldn't be
how it was with him. If they say that approving and disapproving are states of mind,
and states of mind may logically exist in an individual whatever social setting he is in,
one would reply that on that definition of ‘state of mind’ (which does not particularly
recommend itself) what is at issue is whether approving or disapproving of something
is a state of mind.
More powerful, perhaps, is the thought that approving is being for something, and
disapproving against it. Surely it cannot be argued that these attitudes too require a
tacitly recognised position? Curiously enough this is just what can be argued. If we go
back, once more, to our old examples, we find that the stranger can no more be against
the elderly couple's move to the seaside than he can disapprove of their going. If a
relative or close friend says that he is against their going we understand him. To say to
them ‘I am against your going’ may be ineffective, ill judged, or impertinent, but we
understand the attribution of the attitude as conceptually in order. He is against their
going. But what could it mean to say that the stranger was against their going, or to say
it even of the neighbour who does not want new tenants next door? Either could want
the old people to stay, but wanting-that-not is not the same as being against. 3 They
could hope that the old people would not go, and could work against the project, but
without some addition to the story we could not describe them as being against it. If the
stranger, or the neighbour not a friend, went up to them and said ‘I'm against your
going’, this would not be merely impertinent, like offering them advice; it would be
presumptuous because presuming a relationship which did not in fact exist. Nor is the
rule simply a social rule against expressing an attitude impertinently: ‘he is against it’ is
just as bad as ‘I'm against it’, said to the people concerned. If this is not obvious it is
because what is said at a distance, and not to their faces, leaves more room for fantasies
of a special relationship.
What, then, is the relation between approving and disapproving on the one hand and
on the other being for and against? It seems that approving and disapproving are
special cases of being for and against, and that the two are practically indistinguishable
in cases like the one described above. There are, however, examples of being for or
against something which are not also instances of approval or disapproval.

3The only case I know of where to say one is for or against is the same as to say that one wants
something, is in the locution ‘I'm for so and so’, meaning that one wants him to win. In this example,
which was given me by Gilbert Harman, one is simply, perhaps in imagination, taking sides, and for that
no special position or social setting is required.
 XIV Approval and Disapproval106F

Let us consider the following example. A group of people are deciding how to spend
the evening together. Should they go to a cinema? Should they go into town? If they go
to a cinema, which one should it be? Some are in favour of one thing. Some of another.
And this need not be a matter of approval or disapproval; they may simply be for or
against.
The characteristic of this class of examples is that here we do not always have shared
ends, as with the joint enterprise, but ends that may conflict. And here each man may,
though he need not, favour a course of action on the ground that it suits himself. Each
has in theory a voice in the decision even if in practice some particularly ineffective or
unpopular individual may be ignored. What we now want to ask, as in our question
about approval and disapproval, is whether the tacit agreement about how such
questions are to be settled is necessary if the attitudes of being for or against different
courses of action are to exist? It seems to me clear that it is so. That the agreement is
necessary for one use of the expressions ‘I'm in favour of X’ or ‘I'm against X’ is
obvious. For these expressions sometimes function rather like a device for registering a
vote, and it seems clear that if you are going to register a vote you must have it, and that
if you have it the relevance of what you say to the decision cannot be denied. Of course
even formal votes can be miscounted, falsified, wasted, etc. Nevertheless where the
words ‘I'm in favour of X’ are used in this way there must be some presumption that
what the speaker wants is to be taken into account. Now it may be said that this is all
very well when we are considering the use of expression ‘I'm in favour of X’ as if it
were the casting of a vote. Perhaps it is a different matter when an attitude is involved.
This seems a strange position because it drives a large wedge between two things
apparently so closely connected. One would have thought that the expression of the
attitude required rather more in the way of conditions than the ‘voting’ use. But on the
present hypothesis a social background would be presupposed by the latter but not by
the former.
If being for or against, in favour of or opposed to, are attitudes that do require a
particular social background, then even where approval and disapproval are not
themselves in question we have social conditions which are required for the attitudes. It
is not, however, to be suggested that the agreement behind each is the same. When one
approves or disapproves of something it seems to be one's opinions which are to be
taken into account, and hence it is right to speak of being listened to. In the cases of
being for or against just discussed the understanding is rather that one's wishes are to
be taken into account.
The thesis that being for and against are attitudes which, like approval and disapproval,
presuppose a determinate social background is strengthened by consideration of the
following example. There is a prison camp in which the prisoners have no say
whatsoever in the running of the place. The camp commandant takes decisions to suit
himself or his superiors, and the prisoners' wishes are considered irrelevant. Now it is
 Philippa Foot

easy to suppose that some ruling, as e.g. about the work they are to do, affects the
prisoners adversely. They do not like the new rule; perhaps they even hate it.
Nevertheless in the circumstances imagined we cannot say that they are opposed to it
or against it. This might seem to be because there is no such way in which they can
work against it, but actually this might not be so. They might, for instance, scheme to
kill the commandant in the hope that his successor would change the rule, and might
actually have more effect on things than the members of some group, in a different
setting, whose wishes were supposed to be taken into account. Nevertheless, since the
position of being against something is one that only society can create, it is the former,
not the prisoners, to whom it is available. 4
Turning back to approval and disapproval we may at last raise the question of
approving or disapproving of something on moral grounds. Does this, too, presuppose
particular social practices? My intuition is strongly that it does, and that the practices
and expectations necessary for moral approval and disapproval share at least some of
the features of those which other kinds of approval and disapproval require. It is not, of
course, that customs about who is to be listened to here establish classes of persons able
to approve or disapprove where others cannot. In matters of morality we do not need
experts, and such things as relationships are irrelevant to moral approval and
disapproval. If anyone can approve or disapprove on moral grounds, then everyone can
do so, or any sane person over a certain age. What is understood is that anyone is to
listen to anyone when considerations are brought forward which are moral
considerations.
There is, therefore, no difficulty in the idea that although no exclusive position is
needed for moral approval or disapproval determinate social practices are nevertheless
presupposed, and one might ask at this point why it is supposed that the possibility of
moral approval and disapproval must be independent of social facts? It is not, after all,
that all moral concepts must have such independence. Obviously, promising and owing
are things that require a particular social background, and it has been argued by H. L.
A. Hart that obligations are in the same boat so far as this is concerned. 5 Without the
institutions that give these concepts application a man could not truly think that he had
promised this, was owed that, or had a certain obligation. And if he, on his own,
dreamed up the ideas, he would have to dream up the social practices as well.
One cannot say that just because moral approval and disapproval are moral we can see,
ahead of any investigation, that they cannot require a determinate social setting. But
does anything, apart from the analogy with the inspector's act of approval and (if the

4 This way of putting it was suggested by Jerrold Katz, who has given me much help in the writing of this
paper.
5 ‘Legal and Moral Obligation’ in Essays in Moral Philosophy, ed. A. I. Melden (University of Washington

Press, Seattle and London, 1958) .

 XIV Approval and Disapproval106F

preceding arguments have been right) with non-moral attitudes of approval and
disapproval, suggest this link between social practices and moral attitudes? The crucial
experiment is to suppose a world in which the practices are different in the relevant
respects, and then to see what would be possible for an individual in a world such as
that.
What is it then, that we have, which may make moral approval and disapproval
possible? What we have, which we might not have had, and perhaps will not always
have, is a certain assumption about the determination of conduct within a certain area.
It is our custom, and for all I know the custom of every society in the world, to take
matters such as killing, stealing, and lying as a concern of the community. Everyone has
a voice in the matter of whether or not these things are to be done; actions within this
range are up for scrutiny, and by and large men take account of what others say about
such matters. The actions are ‘passed’ or ‘not passed’, and moreover people are
accepted or not accepted on the ground that they do or do not do them. This is, of
course, so familiar to us that it is hard even to see it as a particular social setting which
might be necessary for moral approval and disapproval. Perhaps this will be easier if
we try to think the familiar setting away.
Suppose, for instance, that no one took much notice of what anyone else did, except
when it affected him, or him and his family. Cows and horses behave like this. Suppose
that human beings did so as well? Or suppose that people were interested in each
others' actions but only aesthetically. They liked to watch rescues and not murders, or
murders and not rescues, but this had no effect on the number of murders or rescues
that were performed. In such settings people would not have a ‘say’ in what other
people did as they do in a society such as ours. No one would be taught not to murder
or ostracised for stealing; these things would be treated as the affair of the agent and his
victim.
Now let us place in this world an individual of universal benevolence, hating the
suffering that these people inflict on each other. And suppose that he sets out to try to
prevent murders and to get other people to hate them too, and to refrain from such acts.
Can we now say that his attitude to murder is one of moral disapproval? It seems to me
that we cannot; and that the reason is that we cannot think of him, situated as he is, as
refusing to ‘pass’ acts of murder. In this situation, where by hypothesis there is no social
regulation of behaviour, and no one cares what anyone else thinks of what he does, no
one has any authority to speak against murder. Then does a man who disapproves of
something on moral grounds necessarily speak with the authority of society? In one
way, of course, he need not do so, or else it would be impossible for anyone to come out
against established moral opinions. But it matters here what our moral theory allows. If
we think that there are no limits, logically speaking, to the considerations that may be
argued by a moral dissenter, that is a would-be reformer, then his moral opinions need
relate to no standards or goals accepted by others who have a morality. If, however, this
 Philippa Foot

seems on other grounds implausible, we shall think that even the most radical moral
reformer must, if he is to keep within morality, refer to more or less determinate
standards or ends. He and others must have common ground, and it is this that allows
him to speak with authority even when he is criticising those among whom he lives.
What, however, are we to say about those who altogether reject morality? Surely we
think it possible to disapprove of their actions, although they do not agree to take any
account of what we say? This is true, and it is an important fact about the phenomenon
we call ‘morality’ that we are ready to bring pressure to bear against those who reject it.
But this no more shows that moral attitudes do not depend on agreement within human
society than the possibility of asserting other kinds of authority against those who do
not accept it shows that authority requires no agreement. The position of the one man
described in the last paragraph is not at all like our position, given that we are able to
bring the pressure and authority of society against those who reject the ends of
morality. Nor do we get a position like his if we think of discovering some tribe whose
members behave in the way we described earlier, having no morality and not, initially
at least, caring whether we accept or reject them. For it matters that if we do find such a
tribe we will confront them with the confidence that we have the world with us—the
world that pays at least lip service to morality. Perhaps we would coerce them; perhaps
whatever might be their cruelties, we would not. In neither case would it be nonsense to
speak of disapproving of what they did.
Finally, I want to consider whether what was said earlier about the need for reasons for
attitudes of approval and disapproval could be true also in the moral case. Must there
not be some basic moral approvals and disapprovals for which no reasons are given? 6
To see how we should answer this question it will be useful to go back to our earlier
examples of non-moral approval and disapproval. We were talking, for instance, of
friends and relations who disapproved of an elderly couple's plan of retiring to the
seaside, and of partners in joint enterprises who disapproved of choices of dinner
guests. In one case the reason for the disapproval was the thought that the couple
would be unhappy; in the other cases what was in question was the success of the joint
enterprise, or some other shared end, these being the standards dictated by the
particular context. And we see that indeed we cannot suppose the question ‘Why do
you disapprove?’ applied to the objects specified in the standard. It is not that we could
not understand questions such as ‘Why do you disapprove when your partner makes a
choice disastrous to the enterprise in which you are both engaged?’ or ‘Why do you
disapprove when there is a plan afoot that will probably turn out badly for one of your
friends or relatives?’ But we would have to understand these questions differently from
‘Why do you disapprove of their moving to the seaside?’ or ‘Why do you disapprove of

6This problem was put to me by Bernard Gert, and I am grateful to him for making me think about it in
this context.
 XIV Approval and Disapproval106F

the choice of so and so?’ Perhaps what is being asked is why we go in for this kind of
disapproving, which is not something that everyone does. But to this question we could
answer that we didn't know and perhaps had no reason at all. Similarly, in the case of
moral approval and disapproval the question ‘Why do you disapprove?’ requires an
answer so long as the standard itself has not been reached. Here even those who have
not decided on prescriptivist or emotivist grounds that there could not be such a
standard will agree that it is much harder to say where it is reached. Perhaps there is
more than one ‘good’ to which moral standards relate, as for instance both the common
good and the liberty of the individual. But given that these are standards one does not
need a reason for applying them. The question ‘Why do you disapprove of harming
others?’ is not to be treated like ‘Why do you disapprove of homosexuality?’ One can
ask whether, and if so why, someone goes in for the approvals and disapprovals we call
moral, but if there is no answer to the question ‘Why?’ no doubt is thrown on the
attitude itself.
In conclusion, I should like to go back to the main thesis of this paper and ask about the
implications of the idea that moral approval and disapproval can exist only in a setting
in which morality is taught and heeded. Does it imply that outside this setting all moral
judgement would be impossible? The answer to this question must depend on the
connexion we see between moral judgement and approval and disapproval. If there are
any expressions—‘right’ and ‘wrong’ for instance—that serve only to express these
attitudes, these would obviously not be in the language. But it is not clear that the
whole of our moral language has an essential connexion with the expression of
attitudes, even if any part does so. When, for instance, we speak of good men and bad
men, or good actions and bad actions, we are using a form of expression (‘good A’)
which is, in general, linked to standard interests rather than speakers' attitudes. There
seems to be no reason why an individual who observes that there is a certain interest in
a given class of persons or things should not have the materials for the invention of a
new expression of this form—as he might have been the first to put ‘good’ together with
‘friend’ or ‘neighbour’ or ‘singer’ or ‘speaker’—and similarly he might introduce ‘good
man’ or ‘good action’ even into a setting in which the attitudes of moral approval and
disapproval were not possible. Whether he would then be said to have introduced
‘moral judgement’ is another matter, and one not easily decided given that the
expression ‘moral judgement’ is for the most part a term of art. What is clear is that the
expression of approval and disapproval is part of the complex phenomenon we call
‘morality’, and that if approval and disapproval are essentially social in the sense
explained here, then the morality we are trying to analyse is so too.