BERNARD WILLIAMS

MORALITY
AN INTRODUCTION
TO ETHICS

CAMBRIDGE
UNIVERSITY PRESS
Published by the Press Syndicate of the University of Cambridge
The Pitt Building, Trumpington Street, Cambridge CB2 1RP
40 West 20th Street, New York, NY 100114211, USA
10 Stamford Road, Oakleigh, Melbourne 3166, Australia

O Bernard Williams 1972 To

Preface to the Canto edition @ Bernard Williams 1993 My Mother and Father
ISBN 0 521 45729 7 paperback

First published in the USA by Harper & Row 1972

Published in Pelican Books 1973
Reissued by the Cambridge University Press 1976
Reprinted 1978 1980 1982 1987 1990
Canto edition 1993

Printed in Great Britain at the

University Press, Cambridge
 CONTENTS

Preface to the Canto edition page xi

Preface xvii
The amoralist 3
Subjectivism: First thoughts 14
Interlude: Rela tivism 20
Subjectivism: Further thoughts 26
'Good' 38
Goodnessand roles 48
Moral standards and the distinguishing mark o f a man 55
God, morality, andprudence 63
What ismorality about? 73
Utilitarianism 82
 PREFACE T O THE CANTO EDITION

THIStext was originally intended to be part of a larger

book, which was to consist of several novella-length
pieces by different writers, forming collectively a
substantial introduction to philosophy. When the
editor, Arthur Danto, invited me to write the section
on moral philosophy, he made it clear that while we
were encouraged to write in an introductory way, we
were not being asked to write merely a survey, but
rather to pursue the interests and questions that each
found most interesting or fruitful. The publishers in the
end decided not to put out the big book (which,
granted who they were, some of us had inevitably
called 'Harper's Bazaar'), and published each section
separately.
At least one of those books (Richard Wollheim's Art
and Its Objects) has grown in later life, acquiring new
sections in further editions. This one, on the other
hand, remains as it was. The main reason for this is
that I have subsequently written other books and
papers on some of the same subjects, and could see no
point in loading this text with intrusive (and probably
misleading) references to that later work. In the case of
one topic, utilitarianism, this would have been par-
ticularly inappropriate, since in what I wrote later I
tried to take account of what I had written here, and to
develop rather different points; the relevant chapter
here perhaps summarizes the central problem, as I see
it, of the utilitarian project more compactly than I tion often made, not just by moral philosophy but by
have done elsewhere. ethical reflection more generally, that we are clear
For rather similar reasons, I have not tried to provide enough about what count as 'moral' considerations
an up-to-date bibliography. Recent literature on sub- and sentiments, and that what moral philosophy must
jectivism, for instance, has obviously changed and seek is the basis and status of these considerations,
extended the questions beyond my treatment of the taken more or less as a whole. I have wanted to ask a
subject here, but to explain this so as to introduce that prior question, about what the distinction between the
literature would have involved substantial further 'moral' and the 'non-moral' is supposed to do for us;
philosophical discussion. In fact, there is no bibli- and I have suggested that considerations of the moral
ography, even an out-of-dateone, but only a handful of kind make sense only if they are related to other
references, perhaps rather idiosyncratic, to some writ- reasons for action that human beings use, and gener-
ing that I had found helpful. ally to their desires, needs and projects.
In one respect at least the book may seem dated, to Those concerns are perhaps not altogether explicit in
the extent that it starts by complaining of a situation this book, and in particular it does not observe a
which no longer exists, one in which moral philosophy certain verbal distinction which I have more recently
addressed itself to meta-ethical questions about the found useful, between a broader conception of 'the
nature of moral judgement, the possibility of moral ethical', and the narrower concerns (focused particu-
knowledge, and so forth, at the expense of discussing larly on ideas of obligation) of what may be called the
first-order ethical questions. Moral philosophy still, system of 'morality'. Others may not find this termin-
appropriately, discusses meta-ethical questions, but it ology helpful, but since I have suggested it, it is perhaps
is certainly not true any longer that first-order ques- worth mentioning, in particular, that the title and
tions are not mentioned. On the contrary, issues such sub-title of this book do not use those words in that
as abortion, feminism, and famine are now standardly way. It would surely be possible to discuss morality- in
discussed in moral philosophy courses and textbooks. I ,my current, restrictive sense - as an introduction to
must confess that some of these discussions, assuming ethics (though I doubt that it would be the best way to
as they do that ethical thought is made more rational get introduced to it); but this is not in fact what this
by deploying ethical theory, seem to me as distanced book does. Rather, it discusses, a lot of the time, ethics
from real experience as the forms of uncommitted
moral philosophy about which I complain here. The * I have pressed the later complaint, against the supposed
spirit in question is, unnervingly, the same. But the power of ethical theory, in Ethics and the Limits o f Philos-
complaint is certainly different. * ophy (London: Collins, and Harvard University Press, 1985),
There are points at which the two complaints unite. where I also discuss some peculiarities of the 'morality
I have criticized in some more recent work the assump- system', which I mention below.
as an introduction to the problems and limitations of opher's speciality, can certainly play a part in sharpen-
morality. ing perception. But the aim is to sharpen perception, to
The placing of morality in relation to other ethical make one more acutely and honestly aware of what
considerations and to the rest of life - in relation to one is saying, thinking and feeling. Philosophy invites
happiness, for instance - is in fact a theme here, us (perhaps more insistently now than when this book
although it is not expressed in those terms. Because it was written) to ask whether what we say in morality is
does contain that theme, the book has a characteristic true. One thing I felt in writing this book, and feel even
which I did not consciously have in mind when I wrote more now, is that it is vital not to forget another
it, but which was pointed out to me by a classicist who question that is to be asked both about morality and
had used it in his teaching, that the concerns from about moral philosophy, how far what we say rings
which it sets out are those more typical of the ancient true.
world and its philosophies than of modernity. In a BERNARD WILLIAMS
recent book, * I have tried to develop more reflectively Berkeley, March 1993.
and on a larger scale this interest in ethical ideas of the
ancient world (and not only of its philosophies), and a
sense of their relevance to our present situation.
Near the beginning of this book, I talk about the
problems of finding a style for moral philosophy. I still
think that these problems are real, and also that moral
philosophy involves such problems to a greater extent
than most other areas of philosophy. What I should not
want to accept now is an implication to be heard in this
discussion, that there might be one general solution to
this problem, and that once one had found it one would
know how to write moral philosophy. That cannot be
so: the problems of finding a convincing, adult, and
unmechanical way of approaching the subject must be
faced on each occasion. Sometimes literature or his-
tory can be called upon, to give some idea of the weight
or substance of ethical concepts that we use or have
been used by others; analytic argument, the philos-

* Shame and Necessity (California University Press, 1993).

-

PREFACE

WRITING about moral philosophy should be a hazard-

ous business, not just for the reasons attendant on writ-
ing about any difficult subject, or writing about
anything, but for two special reasons. The first is that
one is likely to reveal the limitations and inadequacies
of one's own perceptions more directly than in, at least,
other parts of philosophy. The second is that one could
run the risk, if one were taken seriously, of misleading
people about matters of importance. While few writers
on the subject have avoided the first hazard, very many
have avoided the second, either by making it impos-
sible to take them seriously, or by refusing to write
about anythlng of importance, or both.
This sad truth is often brought forward as a partic-
ular charge against contemporary moral philosophy of
the 'analytical' or 'linguistic' style: that it is peculi-
arly empty and boring. In one way, as a particular
charge, this is unfair: most moral philosophy at most
times has been empty and boring, and the number of
great books in the subject (as opposed to books involved
in one way or another in morality) can be literally
counted on the fingers of one hand. The emptiness of
past works, however, has often been the emptiness of
conventional moralizing, the banal treatment of moral
issues. Contemporary moral philosophy has found an
original way of being boring, which is by not discussing
moral issues at all. Or, rather, it is not so much that a
 MORALITY PREFACE

style of moral philosophy has been evolved which cuts istic problem, in the deepest sense of 'style' in which
the connection with moral issues altogether - that, if it to discover the right style is to discover what you are
were possible, would have the interest of being remark- really trying to do. How does one combine argument
able; but the desire to reduce revealed moral commit- (which is after all likely to constitute the philosopher's
ment to a minimum and to use moral arguments in the Special claim on anyone's attention) with either the
role of being uncontentiously illustrative leaves an im- longer leaps or the more concrete detail which provide
pression that all the important issues are off the page, the more interesting stuff of moral thought? Can the
somewhere, and that great caution and little imagination reality of complex moral situations be represented by
have been used in letting tiny corners of them appear. means other than those of imaginative literature? If
There are many reasons for this situation. A central not, can more schematic approaches represent enough
one is that contemporary views about morality itself of the reality? How much of what genuinely worries
leave an unclarity about what qualities of mind or anyone is responsive to general theory ?
character are particularly called upon in constructive If I knew answers to these questions, I should not
moral thought (indeed, in some accounts of morality it have to ask them now.
is not even clear that there can be such a thing as con- This essay takes a rather tortuous course, and while
structive moral thought); they hence leave one all the I have tried to signpost the major bends, it may be
more uneasy about whether those qualities are likely to worth sketching a plan in advance. I start with a figure
be the qualities of philosophers, when philosophy who has often been of interest, indeed a cause of con-
is largely a professional and academic activity calling cern, to moralists, as providing a challenge to morality
principally, though not exclusively, on discursive and and a demand for its justification: the arnoralist, who
analytical abilities. If there were such an activity as is supposedly immune to moral considerations. Some
deducing substantial moral conclusions from a priori of the most interesting questions about him, which I
premises, trained philosophers might reasonably be have barely touched on, lie not so much in what might
expected to be rather specially good at it; but there is be said to him, as in what might be said about him -
not, and the fact that if there were, then professional what the amoralist can consistently be like. From him,
philosophers would stand a specially good chance of we move to those who do not reject morality, but do
being informed about morality, is itself one of the good take certain special, and it may be almost as disquiet-
reasons for thinking that there could not be such an ing, views about its nature: subjectivists of vari-
activity. ous kinds, and an unashamedly crass (but common)
Certainly the trouble is not, as some pretend, that if kind of relativist. Here I try to examine carefully a p r e
the philosopher is not patently detached and even ject very close to the heart of much modem moral
methodological, then he must be preaching; that can- philosophy, which I have called that of defusing s u b
not possibly be the only alternative. It is rather a styl- jectivism.

xviii xix
 MORALITY PREFACE

From there, to some considerations about goodness; &le), or even without theoretical preconceptions
and, trying to get clear about some ways a man can be (which might well prove static and sterile). It is merely
good at certain things, and still more about things he that one's initial responsibilities should be to moral
can be good as, I seek to disentangle some purely logi- phenomena, as grasped in one's own experience and
cal considerations from what seem to me more s u b imagination, and, at the more theoretical level, to the
stantial issues about what men are and the connection demands of other parts of philosophy - in particular, of
of that with goodness. Two questions in particular the philosophy of mind. There is no reason why moral
emerge from the many that crowd round that area : the philosophy, or again something in some respects
relations of intellectual achievement to the standards broader, in some respects narrower, called 'value
of morality, and the question whether, if God existed, theory', should yield any interesting self-contained
that would make any difference to the situation of theory at all.
morality. This raises some important genera; questions Another way in which this is not a textbook is that
about moral and other motives. These in turn lead to it leaves out large tracts of the subject. That is, at least,
some issues about the point or substance of morality, fully obvious. But it may help to put what is here in
and whether it is ultimately all about human welfare some better perspective if I mention one or two subjects
or (more narrowly, perhaps) happiness. Lastly, the most which a larger treatment of moral philosophy should in
simple-minded way of aiming morality at happiness, my view have near its centre: what practical thought,
that of utilitarianism, is touched on, but only long and acting for a reason, are; what consistency in action
enough to suggest how special and peculiar a system, is, and in moral thought; relatedly, how moral conflict
properly understood, it is; and to point in the direction is a basic fact of morality; how the notion of a rule is
where its peculiarities are to be found. To follow them important for some, but not all, parts of morality (the
out is a task for another occasion.* present essay has nothing to say about its importance);
One of the many ways in which this essay is not a how shaky and problematical is the distinction between
textbook, even an introductory and outline textbook, the 'moral' and the 'non-moral' - above all in its most
of moral philosophy is that it offers no systematic important employment, to distinguish between differ-
theory. I am unashamed about that, since it seems to ent sorts of human excellence.
me that this subject has received more over-general and That this essay should leave out most things of im-
over-simplified systematization, while inviting it less, portance was inevitable; that it should follow a tor-
than virtually any other part of philosophy. I do not tuous path, was not. Whether it was inevitable that it
mean by that that one should approach moral philo- should fail to find an answer to the problem of how to
sophy without preconceptions (which would be impos- write about moral philosophy, I do not know.
*See A Critique of Utilitarianism, which appears in Utili-
tarianism: For and Against (Cambridge University Press, 1973).
XX xxi
 MORALITY
AN I N T R O D U C T I O N TO ETHICS
 THE AMORALIST

<WHY should I do anything?' Two of the many ways

of taking that question are these: as an expression of
despair or hopelessness, when it means something like
'Give me a reason for doing anything; eve-ng is
meaningless'; and as sounding a more defiant note,
against morality, when it means something like 'Why
is there anything that I should, ought to, do 2'
Even though we can paraphrase the question in the
first spirit as 'Give me a reason ...', it is very unclear
that we can in fact give the man who asks it a reason -
that, starting from so far down, we could argue him
into caring about something. We might indeed 'give him
a reason' in the sense of finding something that he is
prepared to care about, but that is not inducing him to
care by reasoning, and it is very doubtful whether there
could be any such thing. What he needs is help, or hope,
not reasonings. Of course it is true that if he stays alive
he will be doing something, rather than something else,
and thus in some absolutely minimal sense he has some
sort of reason, some minimal preference, for doing
those things rather than other things. But to point this
out gets us hardly anywhere; he does those things just
mechanically, perhaps, to keep going, and they mean
nothing to him. Again, if he sees his state as a reason
for suicide, then that would be to make a real decision;
as a way out of making any decisions, suicide comes
inevitably one decision too late (as Camus points out in
 MORALITY THE AMORALIST
Le Mythe de Sisyphe). But it would be no victory for us out a more formal aspect of morality, namely any dis-
or for him if it turned out there was after all just one position on his part to stand back and have the thought
decision that he was prepared to acknowledge, that that if it is 'all right' for him to act in these ways, it
one. must be 'all right' for others to act similarly against
I do not see how it could be regarded as a defeat for him. For if he is prepared to take this stance, we might
reason or rationality that it had no power against this be able to take a step towards saying that he was not a
man's state; his state is rather a defeat for humanity. man without a morality, but a man with a peculiar
But the man who asks the question in the second spirit one.
has been regarded by many moralists as providing a real However, we need a distinction here. In one way, it
challenge to moral reasoning. He, after all, acknowledges is possible for a man to think it 'all right' for everyone
some reasons for doing things; he is, moreover, like to behave self-interestedly, without his having got into
most of us some of the time. If morality can be got off any distinctively moral territory of thought at all: if,
the ground rationally, then we ought to be able to get it roughly, 'it's all right' means 'I am not going to
off the ground in an argument against him; while, in his moralize about it'. He will be in some moral temtory if
pure form - in which we can call him the arnoralist - 'all right' means something like 'permitted', for that
he may not be actually persuaded, it might seem a com- would c a w implications such as 'people ought not to
fort to morality if there were reasons which, if he were interfere with other people's pursuit of their own inter-
rational, would persuade him. ests', and that is not a thought which, as an amoralist,
We might ask first what motivations he does have. he can have. Similarly, if he objects (as he no doubt
He is indifferent to moral considerations, but there are will) to other people treating him as he treats them, this
things that he cares about, and he has some real pre- will be perfectly consistent so long as his objecting con-
ferences and aims. They might be, presumably, pleasure sists just in such things as his not liking it and fighting
or power; or they might be something much odder, such back What he cannot consistently do i's resent it or
as some passion for collecting things. Now these ends disapprove of it, for these are attitudes within the moral
in themselves do not exclude some acknowledgement of system. It may be difficult to discover whether he has
morality; what do we have to leave out to represent him given this hostage to moral argument or not, since he
as paying no such acknowledgement ? Presumably such will no doubt have discovered that insincere expressions
things as his caring about other people's interests, hav- of m t r n e n t and moral hurt serve to discourage some
ing any inclination to tell the truth or keep promises if of the more squeamish in his environment from hostile
it does not suit him to do so, being disposed to reject action.
courses of action on the ground that they are unfair or This illwtrates, as do many of his activities, the
dishonourable or selfish. These are some of the substan- 0 ~ fact0that this
~ man
~ is a parasite on the moral
tial materials of morality. We should perhaps also leave System, and he and his satisfactions could not exist as
 MORALITY THE AMORALIST

they do unless others operated differently. For, in gen- thougaa need not lead directly to moral considerations,
eral, there can be no society without some moral rules, they give a substantial footing to them, since they im-
and he needs society; also he takes more particular ad- mediately invite questions about what is so good about
vantage of moral institutions like promising and of those dispositions, and it will be difficult for him to
moral dispositions of people around him. He cannot pursue those questions very far without thinking in
deny, as a fact, his parasitic position; but he is very t- of the general interests and needs of his fellow
resistant to suggestions of its relevance. For if we try human beings, which would land him once more back
saying 'How would it be for you if everyone behaved in the world of moral thought from which he is exclud-
like that?' he will reply, 'Well, if they did, not good, I ing himself.
suppose - though in fact I might do better in the result- The temptation to think of himself as courageous is
ing chaos than some of the others. But the fact is, most a particularly dangerous one, since it is itself very
of them are not going to do so; and if they do ever get nearly a moral notion and draws with it a whole chain
round to it, I shall be dead by then.' The appeal to the of distinctively moral reflections. This man's application
consequences of an imagined universalization is an of the notion will also have a presupposition which is
essentially moral argument, and he is, consistently, not false: namely, that the more moral citizens would be
impressed by it. amoral if they could get away with it, or if they were
In maintaining this stance, there are several things not too frightened, or if they were not passively con-
he must, in consistency, avoid. One - as we noticed ditioned by society - if, in general, they did not suffer
before, in effect - is any tendency to say that the more from inhibitions. It is the idea that they are afraid that
or less moral majority have no right to dislike him, gives him the idea of his own courage. But these pre-
reject him, or treat him as an enemy, if indeed they are suppositions are absurd. If he means that if as an indi-
inclined to do so (his power, or charm, or dishonesty vidual one could be sure of getting away with it, one
may be such that they do not). No thoughts about justi- would break any moral rule (the idea behind the model
fication, at least of that sort, are appropriate to him. of Gyges' ring of invisibility in Plato's Republic), it is
Again, he must resist, if consistent, a more insidious iust false of many agents, and there is reason why : the
tendency to think of himself as being in character really more basic moral rules and conceptions are strongly
rather splendid - in particular, as being by comparison internalized in upbringing, at a level from which they
with the craven multitude notably courageous. For in do not merely evaporate with the departure of police
entertaining such thoughts, he will run a constant men or censorious neighbours. This is part of what it is
danger of getting outside the world of his own desires for them to be moral rules, as opposed to merely legal
and tastes into the region in which certain dispositions MQuirements or matters of social convention. The
are regarded as excellent for human beings to have, or of moral education can actually be to make
good to have in society, or such things; and while such people want to act, quite often, in a non-self-interested
6
 MORALITY THE AMORALIST

way, and it often succeeds in making it at least quite stress, deprivation, or scarcity (the test that Hobbes, in
difficult, for internal reasons, to behave appallingly. his picture of the state of nature, imposed), one can only
But this, he will say, is just social conditioning; ask again, why should that be the test? Apart from the
remove that, and you will find no moral motivations. unclarity of its outcome, why is the test even appro-
- We can reject the rhetoric of the word 'conditioning'; priate? Conditions of great stress and deprivation are
even if there were a true theory, which there is not, not the conditions for observing the typical behaviour
which could explain all moral and similar education in of any animal nor for observing other characteristics of
terms of behaviourist learning theory, it would itself human beings. If someone says that if you want to see
have to explain the very evident differences between what men are really like, see them after they have been
successful and intelligent upbringing, which produces three weeks in a lifeboat, it is unclear why that is any
insight, and the production of conditioned reflexes. better a maxim with regard to their motivations than it
Then let us say instead that all moral motivation is the is with regard to their physical condition.
product of social influences, teaching, culture, etc. It is If there is such a thing as what men are really like, it
no doubt true. But virtually everything else about a may be that (in these sorts of respects, at least) it is not
man is such a product, including his language, his so different from what they are actually like; that is,
methods of thought, his tastes, and even his emotions, creatures in whose lives moral considerations play an
including most of the dispositions that the amoralist important, formative, but often insecure role.
sets store by. - But, he may say, suppose we grant that The amoralist, then, would probably be advised to
anything complex, even my desires, are influenced by avoid most forms of self-congratulatory comparison of
culture and environment, and in many cases produced himself with the rest of society. The rest may, of
by these; nevertheless there are basic impulses, of a self- course, have some tendency to admire him, or those
interested kind, which are at the bottom of it all: these may who are at such a distance that he does not tread
constitute what men are really like. directly on their interests and affections. He should not
If 'basic' means 'genetically primitive', he may pos- be too ercouraged by this, however, since it is probably
sibly be right: it is a matter of psychological theory. wish-fulfilment (which does not mean that they would
But even if true in this sense, it is once more irrelevant be like him if they could, since a wish is different from
(to his argument, not to questions about how to bring a frustrated desire). Nor will they admire him, still less
up children); if there is such a thing as what men are like him, if he is not recognizably human. And this
really like, it is not identical with what very small chil- raises the question, whether we have left him enough to
dren are like, since very small children have no lan- be that.
guage, again, nor many other things which men really Does he care for anybody? Is there anybody whose
have. If the test of what men are really like is made, sufferings or distress would affect him? If we say 'no*
rather, of how men may behave in conditions of great to this, it looks as though we have produced a psycho-
8
 MORALITY THE AMORALIST

path. If he is a psychopath, the idea of arguing him into ground, even though it is unlikely in practice to get off
morality is surely idiotic, but the fact that it is idiotic the ground in a conversation with him.
has equally no tendency to undermine the basis of He gives us, I think, almost enough. For he has the
morality or of rationality. The activity of justifying notion of doing something for somebody, because that
morality must surely get any point it has from the exist- person needs something. He operates with this notion
ence of an alternative - there being something to justify in fact only when he is so inclined; but it is not itself
it against. The amoralist seemed important because he the notion of his being so inclined. Even if he helps
seemed to provide an alternative; his life, after all, these people because he wants to, or because he likes
seemed to have its attractions. The psychopath is, in a them, and for no other reason (not that, so far as these
certain way, important to moral thought; but his impor- particular actions are concerned, he needs to improve
tance lies in the fact that he appals us, and we must on those excellent reasons), what he wants to do is to
seek some deeper understanding of how and why he help them in their need, and the thought he has when
appals us. His importance does not lie in his having an he likes someone and acts in this way is 'they need
appeal as an alternativeform of life. help', not the thought 'I like them and they need help'.
The amoralist we loosely sketched before did seem to This is a vital point: this man is capable of thinking in
have possibly more appeal than this; one might picture terms of others' interests, and his failure to be a moral
him as having some affections, occasionally caring for agent lies (partly) in the fact that he is only intermit-
what happens to somebody else. Some stereotype from a tently and capriciously disposed to do so. But there is
gangster movie might come to mind, of the ruthless and no bottomless gulf between this state and the basic
rather glamorous figure who cares about his mother, dispositions of morality. There are people who need
his child, even his mistress. He is still recognizably help who are not people who at the moment he happens
amoral, in the sense that no general considerations to want to help, or likes; and there are other people who
weigh with him, and he is extremely short on fairness like and want to help other particular people in need.
and similar considerations. Although he acts for other To get him to consider their situation seems rather an
people from time to time, it all depends on how he hap- extension of his imagination and his understanding,
pens to feel. With this man, of course, in actual fact than a discontinuous step onto something quite dif-
arguments of moral philosophy are not going to work; ferent, the 'moral plane'. And if we could get him to
for one thing, he always has something he would rather consider their situation, in the sense of thinking about
do than listen to them. This is not the point (though it it and imagining it, he might conceivably start to show
is more of a point than some discussions of moral argu- some consideration for it: we extend his sympathies.
ment would lead one to suppose). The point is rather And if we can get him to extend his sympathies to
that he provides a model in terms of which we may less immediate persons who need help, we might be able
glimpse what morality needs in order to get off the to do it for less immediate persons whose interests have
 MORALITY T H E AMORALIST

been violated, and so get him to have some primitive later. I shall now turn to someone who is also found
-

grasp on notions of fairness. If we can get him all this upsetting by morality, but in a different way from the
way, then, while he has no doubt an extremely shaky amoralist. This man is content that he should have a
- .

hold on moral considerations, he has some hold on morality, but points out that other people have differ-
them; at any rate, he is not the amoralist we started ent ones - and insists that there is no way of choosing
with. between them. He is the subjectivist.
This model is not meant to sketch the outline of a
construction of the whole of morality from the possi-
bility of sympathy and the extensions of sympathy:
that would be impossible. (Even Hume, who perhaps
came nearest to it, did not attempt that. His system,
among the many interesting and valuable things that
it contains, has a distinction between the 'natural' and
the 'artificial' virtues which is relevant to this point.)
The model is meant to suggest just one thing : that if we
grant a man with even a minimal concern for others,
then we do not have to ascribe to him any fundarnen-
tally new kind of thought or experience to include him
in the world of morality, but only what is recognizably
an extension of what he already has. He is not very
far into it, and it is an extensive territory : as we saw in
drawing up the amoralist, you have to travel quite a
long way to get out of it. But the man with the extended
sympathies, the ability to think about the needs of
people beyond his own immediate involvement, is
recognizably in it.
It does not follow from this that having sympathetic
concern for others is a necessary condition of being in
the world of morality, that the way sketched is the
only way 'into morality'. It does not follow from what
has so far been said; but it is true.
Some of the considerations touched on here, about
moral and other motivations, we shall come back to
12
 SUBJECTIVISM: FIRST THOUGHTS

inadequate way it seems to gesture towards something

which is closest of all to what has concerned many
SUBJECTIVISM : FIRST THOUGHTS who have been womed about moral objectivity : the
idea that there is no moral order 'out there ' - out there,
in the world, are only the sorts of things and the sorts
of facts that science, and the more everyday processes
CONSIDERthree statements, each of which, in its of human enquiry of which science is a refinement,
different way, expresses a view that moral opinions, deal. Alles anderes ist Menschenwerk. Statement (c)
or moral judgements. or moral outlooks are 'merely can be said - using the word in an unambitious way -
subjective' : to express a metaphysical view.
I
(a) A man's moral judgements merely state (or express) The metaphysical view brings out most explicitly
his own attitudes. something that is latent in all the three statements, a
(b) Moral judgements can't be proved, established, shown distinction between fact and value. A central concern
to be true as scientific statements can; they a n matt6rs of
individual opinion.
1 of much modem moral philosophy has been the dis-
tinction between fact and vahe. One important form
(c) There are no moral facts; there are only the sorts of , that this concern has taken has been to emphasize the
facts that science or common observation can discover, and
the values that men place on those facts.
1 distinction, while rejecting the supposedly disquieting
consequences of it, by trying to show either that they
The three statements come very close to one another.
and in discussions of subjectivism and objectivism one
often finds versions of the three being used virtually
I
I

I
are not consequences, or that they are not disquieting.
Thii project of defusing subjectivism (as it might be
called) can be expressed in terms of our three subjecti-
vist statements roughly as follows: that, in so far as
interchangeably. They are, indeed, genuinely related to
one another. Yet they are significantly different. The they are defensible, they come to the same thing; and
first, (a), expresses what might be called in a broad that what they come to is both not alarming and essen-
sense a logical or linguistic view : it purports to tell us tial to the nature of morality (the point that is essential
something about what moral remarks are or do. The to the nature of morality seems sometimes to be thought,
second, @), introduces a set of notions not present in oddly, to imply just by itself the point that it is not
the first, notions connected with the concept of know- alarming).
ledge, and may be taken to express an epistemological This project we shall follow, with interruptions, for
view about moral judgements. The third statement, (c), some whiIe. It starts as follows. Statement (a), first of
is the vaguest and least tangible of the three, and shows all, is either false or harmless. It is false if it claims that
on its surface the danger of collapsing, partly or com- moral judgements state their utterer's attitudes. For if
pletely, into one or other of the first two : which is what this were so, they would be simply autobiographical
many philosophers would claim it must do. Yet in its
 MORALITY SUBJECTIVISM: FIRST THOUGHTS

remarks, replaceable without loss by statements ex- that is all there is t o be said about it. In particular the
.'
plicitly of the form 'My attitude towards this . . or 'I subjectivist force of (a) lies in a suggestion that there is
..
feel .' But if this were so, there would be no inter- no question of the attitudes expressed in moral judge
menu being right or wrong, whereas there is a ques-
personal moral disagreements; two persons expressing
what we would normally take to be conflicting views tion of the beliefs expressed in factual judgements
would not be expressing conflicting views at all, but being true or false.
would be, rather, like two persons on a boat, one of Now that the thesis has regained a distinctively s u b
whom says that he feels sick while the other says that jectivist form it will be said that it is false. For - appeal-
he, on the other hand, does not. But it is an evident fact ing once more to the ways in which moral judgements
that there are genuine moral disagreements, and -
are actually made and treated it is not true that there
that moral views can conflict. Moral judgements must is no question of moral attitudes being right or wrong.
(at least to this extent) mean what we take them to One of their distinguishing marks, as against mere ex-
mean; and what we take them to mean, the way we use pressions of taste or preference, for instance in matters
them, is such that they do not merely make autobio- of food, is that we take seriously the idea of a man's
graphical claims, but a sort of claim which is being being wrong in his moral views; indeed, the very con-
rejected by someone who utters a contrary moral judge- cept of a moral view marks a difference here, leaning as
ment. Thus they do not merely describe the speaker's it does in the direction of belief rather than of mere
own attitude. taste or preference. It is precisely a mark of morality
This argument, however, does not dispose of the pos- that de gustibus non disputandum is a maxim which
sibility that moral utterances, while they do not directly does not apply to it.
describe, may nevertheless have the function of expres- To this it might be replied that the fact that moral
sing, the utterer's attitude. This claim in itself is incon- attitudes can be called 'right' or 'wrong', and that the
testable and harmless; for in itself it comes to no question of their rightness or wrongness is taken seri-
more than saying that a man who makes a moral judge- ously, does not in any ultimate sense help to transcend
ment can (if that judgement is sincere) be said to be subjectivism. It shows not that moral attitudes are
expressing his attitude to a certain moral issue, and this more than (merely) attitudes, but that they are attitudes
does not support any distinctively subjectivist view of which we get disturbed about; that it matters to us to
moral judgements: a man who sincerely makes a fac- secure similarity of attitude within society. The use of
tual judgement may be said to be expressing his belief the language of 'right' and 'wrong' can be seen as part
about a factual issue, but that does not support a sub- of the apparatus of securing agreement, marking off
jectivist view of factual judgements The subjectivist those who disagree and so forth: it remains the case that
interpretation comes in when one says that one who all we have are people's attitudes towards different sorts
makes a moral judgement expresses his attitude, and of conduct, personality, social institutions, etc.
 MORALITY SUBJECTIVISM: FIRST T H O U G H T S

Yet this account, in its turn, seems inadequate to the only within a framework of given assumptions. If
facts: it must at least underestimate the logical com- people can argue one with another about an individual
plexity of the situation. Notably it fails to account moral issue or a question of principle, this is only be-
for the undoubted facts that a man may be in a state of cause there are moral attitudes in the background
moral doubt, which he may resolve - that a man can which are not in dispute and in the light of which the
nonarbitrarily change his mind about a moral matter, argument goes on. Putting this point in a rather stronger
not merely in the individual case, but on a general issue, form, it might be said that it is only about the applica-
and for reasons. Thus a man previously convinced that tion of accepted moral views that the argument goes
a permissive attitude towards abortion was wrong, on. So where there is no background of moral agree
might change his view on this, and not merely (for in- ment, there can be no argument. At this point the sub-
stance) because he felt lonely in a group which held the jectivist attitude can reappear, claiming now that all
permissive attitude. No doubt many writers on moral that has been shown by the considerations about ex-
philosophy overestimate the extent to which people are changing reasons is that the morality of a man or a
led by rational considerations to modify their moral society is to some degree general and systematic and
views; those writers ignore the evident extent to which that general attitudes can be applied to less general
attitudes are modified by factors such as the desire to cases. When we get outside the framework of agreed
conform with one group or nonconform with another general attitudes, there is no further argument, and no
- the groups themselves not being chosen in the light of way of showing any position to be right or wrong.
moral reasons, but rather determined by the individ- This comes very close to our second formulation, @),
ual's situation and needs. But this is, in one sense, be- of subjectivism; we have been led to it, it seems, by
side the point. For even if moral attitudes were rarely modifications of the first formulation. One might, how-
determined by reasons, and the reasons advanced in ever, want to change one element that came naturally
their support were rationalizations, our model of moral into the formulation of @), and, with that, one implica-
attitudes and moral judgements must at least be com- tion of the term 'subjectivism'. For when one turns to
plex enough to leave a place for the rationalizations. It the issue of ultimate disagreements, it is natural to take
is only if the position to which a man is led by these as the unit which holds a set of moral attitudes, the
forces satisfies some conditions of being the sort of posi- society rather than the individual - not in order to
tion to which reasons are relevant that we can under- hypostasize societies, but to draw attention to the
stand it as a moral position at all. point that there are limits to the degree of ultimate dis-
However, even if moral attitudes are of a character agreement that can exist within a society (for without
to admit of support and attack by reasons, and the some degree of moral homogeneity it would not be a
deployment of reasons in reaching a conclusion; never- society); but there are no limits, at least of that kind, on
theless it may be said that these activities are possible disagreement beween societies.
 INTERLUDE: RELATIVISM

Ashanti comes to be taken as saying that human sacrifice

INTERLUDE: RELATIVISM was right among the Ashanti, and this in turn as saying
that human sacrifice among the Ashanti was right; i.e.,
we had no business to interfere with it. But this last is
certainly not the sort of claim allowed by the theory.
The most the theory can allow is the claim that it was
LET US at this stage of the argument about subjectivism right for (i.e., functionally valuable for) our society not
take a brief rest and look round a special view or assem- to interfere with Ashanti society, and, first, this is cer-
blage of views which has been built on the site of moral tainly not all that was meant, and, second, is very dubi-
disagreements between societies. This is relativism, the ously true.
anthropologist's heresy, possibly the most absurd view Apart from its logically unhappy attachment of a
to have been advanced even in moral philosophy. In its nonrelative morality of toleration or non-interference
vulgar and unregenerate form (which I shall consider, to a view of morality as relative, the theory suffers in its
since it is both the most distinctive and the most influ- functionalist aspects from some notorious weaknesses
ential form) it consists of three propositions: that of functionalism in general, notably difficulties that
'right' means (can only be coherently understood as surround the identification of 'a society'. If 'society'
meaning) 'right for a given society'; that 'right for a is regarded as a cultural unit, identified in part through
given sodety' is to be understood in a functionalist its values, then many of the functionalist propositions
sense; and that (therefore) it is wrong for people in one will cease to be empirical propositions and become bare
society to condemn, interfere with, etc., the values of tautologies: it is tediously a necessary condition of the
another society. A view with a long history, it was p o p survival of a group-withcertain-values that the group
ular with some liberal colonialists, notably British ad- should retain those values. At the other extreme, the
ministrators in places (such as West Africa) in which survival of a society could be understood as the sur-
white men held no land. In that historical role, it may vival of certain persons and their having descendants,
have had, like some other muddled doctrines, a bene- in which case many functionalist propositions about
ficent influence, though modem African nationalism may the necessity of cultural survival will be false. When in
well deplore its tribalist and conservative implications. Great Britain some Welsh nationalists speak of the sur-
Whatever its results, the view is clearly inconsistent, vival of the Welsh language as a condition of the sur-
since it makes a claim in its third proposition, about vival of Welsh society, they manage sometimes to
what is right and wrong in one's dealings with other convey an impression that it is a condition of the sur-
societies, which uses a nonrelative sense of 'right' not vival of Welsh people, as though the forgetting of
allowed for in the first proposition. The claim that Welsh were literally lethal.
human sacrifice, for instance, was 'right for' the In between these two extremes is the genuinely inter-
 MORALITY INTERLUDE: RELATIVISM

d n g territory, a province of informative social The central confusion of relativism is to try to con-
science, where there is room for such claims as that a jure out of the fact that societies have differing atti-
given practice or belief is integrally connected with tudes and values an a priori nonrelative principle to
much more of a society's fabric than may appear on the determine the attitude of one society to another; this
surface, that it is not an excrescence, so that discourage- is impossible. If we are going to say that there are ulti-
ment or modification of this may lead to much larger mate moral disagreements between societies, we must
social change than might have been expected; or, again. include, in the matters they can disagree about, their
that a certain set of values or institutions may be such attitudes to other moral outlooks. It is also true, how-
that if they are lost, or seriously changed, the people in ever, that there are inherent features of morality that
the society, while they may physically survive, will do tend to make it difficult to regard a morality as applying
so only in a deracinated and hopeless condition. Such only to a group. The element of universalization which
propositions, if established, would of course be of first is present in any morality, but which applies under
importance in deciding what to do; but they cannot tribal morality perhaps only to members of the tribe,
take over the work of deciding what to do. progressively comes to range over persons as such. Less
Here, and throughout the questions of conflict of formally, it is essential (as was remarked earlier) to mor-
values between societies, we need (and rarely get) some ality and its role in any society that certain sorts of
mildly realistic picture of what decisions might be being reactions and motivations should be strongly interna-
made by whom, of situations to which the consid- lized, and these cannot merely evaporate because one is
erations might be practically relevant. Of various confronted with human beings in another society. Just
paradigms that come to mind, one is that of conflict, such as de gustibus non disputandum is not a maxim which
as the confrontation of other societies with Nazi Ger- applies to morality, neither is 'when in Rome do as the
many. Another is that of control, where (to eliminate Romans doD,which is at best a principle of etiquette.
further complications of the most obvious case, colon- Nor is it just a case of doing as the Romans do, but
ialism) one might take such a case as that of the rela- of putting up with it. Here it would be a platitude to
tions of the central government of Ghana to residual point out that of course someone who gains wider ex-
elements of traditional Ashanti society. In neither case perience of the world may rightly come to regard some
would functionalist propositions in themselves provide moral reaction of his to unfamiliar conduct as parochial
any answers at all. Still less will they where a major and will seek to modify or discount it. There are many
issue is whether a given group should be realistically or important distinctions to be made here between the
desirably regarded as 'a society' in a relevant sense, or kinds of thoughts appropriate to such a process in dif-
whether its values and its future are to be integrally ferent cases: sometimes he may cease to regard a cer-
related to those of a larger group - as with the case of tain issue as a moral matter at all, sometimes he may
blacks in the United States. come to see that what abroad looked the same as some-
22 23
 MORALITY INTERLUDE: RELATIVISM

thing he would have deplored at home was actually, in the owner of the house trying to murder somebody, is
morally relevant respects, a very different thing. (Per- he morally obliged not to interfere because he is tres-
haps - though one can scarcely believe it - there were passing 1
some missionaries or others who saw the men in a None of this is to deny the obvious facts that many
polygamous society in the light of seedy bigamists at have interfered with other societies when they should
home.) But it would be a particular moral view, and one not have done; have interfered without understanding:
both psychologically and morally implausible, to in- and have interfered often with a brutality greater than
sist that these adaptive reactions were the only correct that of anything they were trying to stop. I am saying
ones, that confronted with practices which are found only that it cannot be a consequence of the nature of
and felt as inhuman, for instance, there is an a priori morality itself that no society ought ever to interfere
demand of acceptance. In the fascinating book by Ber- with another, or that individuals from one society con-
nal de Diaz, who went with Cortez to Mexico, there is fronted with the practices of another ought, if rational,
an account of what they all felt when they came upon to react with acceptance. To draw these consequences is
the sacrificial temples. This morally unpretentious col- the characteristic (and inconsistent) step of vulgar rela-
lection of bravos was genuinely horrified by the Aztec tivism.
practices. I t would surely be absurd to regard this reac-
tion as merely parochial or self-righteous. It rather in-
dicated something which their conduct did not always
indicate, that they regarded the Indians as men rather
than as wild animals.
It is fair to press this sort of case, and in general the
cases of actual confrontation. 'Every society has its
own standards' may be, even if confused, a sometimes
useful maxim of social study; as a maxim of social
study it is also painless. But what, after all, is one sup-
posed to do if confronted with a human sacrifice?- not
a real question for many of us, perhaps, but a real ques-
tion for Cortez. 'It wasn't their business,' it may be
said; 'they had no right to be there anyway.' Perhaps -
though this, once more, is necessarily a nonrelative
moral judgement itself. But even if they had no right
to be there, it is a matter for real moral argument what
would follow from that. For if a burglar comes across
 SUBJECTIVISM: FURTHER THOUGHTS

protesting. If the first of these, then the inflicters of the

SUBJECTIVISM : FURTHER THOUGHTS injustice are not justified in what they are doing either,
nor is this man justified in not protesting, and these
considerations remove any basis he was supposed to
have for his indifference. In any case, the argument in
this version has surely gone too far, even for subjectiv-
THE enticements of vulgar relativism are not very ism; since subjectivism did not claim it to be impos-
tempting. But its central error is quite important; and sible to think that anything was justified, but only that
that it is an error has a significance beyond that partic- when a given man thinks something is justified, he can-
ular doctrine, concerned as that is with the relations not in the end be proved wrong. The man we are argu-
between societies. It is possible for someone persuaded ing with has somehow got from that to a position in
of subjectivist views to cease to care about moral issues. which no one (however subjectively) can think some-
(This is different from ceasing to regard something as, thing justified, and that must be a mistake; unless, of
in itself, a moral issue. Thus it is both possible and rea- course, subjectivism is inconsistent, in which case s u b
sonable to suppose that there is no distinctively sexual jectivism is false, and the argument is over anyway.
morality, in the sense of moral considerations that Let us try the second alternative, then, that because
govern sexual relationships and nothing else; while ad- subjectivism is true, he is specially not justified in pro-
mitting the extremely obvious fact that sexual relation- testing. Why should this be so? 'Well', he may say,
ships are profoundly and specially liable to give rise to 'they think they are right, and who am I to say that
moral issues, of trust, exploitation, unconcern for the they are wrong?' But the apparent force of this is en-
interests of third parties and so forth.) A man con- tirely gained from its subtly moving out of the subjec-
fronted with some monstrous political injustice, for in- tivist arena and importing the idea that there is such a
stance, may feel no confidence in protesting or fighting thing as objective rightness, only he is not sure whe-
against it because, as he says, 'Who's to judge?,' or 'It's ther these other people's actions possess it or not.
only my feelings against theirs,' or something like that. Sticking to the subjectivist path, he must recognize that
In so far as there is any traceable intellectual link be- if he chooses to think that they are wrong and that he is
tween the subjectivism and the indifference, it must right in protesting, then no one can say he is wrong
involve something like the relativist error. For the either, and he can be no less justified in protesting than
thought must be something like this : 'Because subjec- they are in doing what they are doing. Another way of
tivism is true, I am not justified in protesting.' If this is putting this point is this : 'perhaps they are right' must
right, it must be so either because, if subjectivism is be one of his moral thoughts. If he also has the thought
true, no one is justified in doing anything; or because, 'They are wrong (only I am not justified in protesting)',
if subjectivism is true, he is specially not justified in he has inconsistent moral thoughts within his own sys-
 MORALITY SUBJECTIVISM: F U R T H E R THOUGHTS

tem, and subjectivism never required him to do that. In in particular, factual thought. Indeed, subjectivism
fact, of course, the thought 'Who am I to say that they itself (see (b) and (c) of our original formulations, given
are wrong?' is one that is had, not within his own sub- above), insists on the contrast between moral attitudes
jective compound, but in mid-air above his own and the and factual beliefs, regarding the latter as 'objective' in
other people's; it tries to stand outside all moral posi- a way that the first are not. The subjectivist leaves us,
tions (including the thinker's own) and yet still be a and may mean to leave us, with an uneasy feeling that
moral thought. But this mid-air place, by subjectivism factual beliefs have got, and moral beliefs lack, some-
itself, is not a place in which anyone can have a moral thing that it is nice to have: that factual beliefs and
thought. science are somehow solider than morality.
These sorts of reasons show why the defusers of sub But still, it will be said, subjectivism leaves every-
jectivism say that it leaves everything where it was and thing where it was, so far as morality is concerned,
cannot possibly logically imply indifferentism or any though not perhaps so far as muddled feelings about
other practical attitude. (They move perhaps rather morality are concerned. Even granted the contrast of
rapidly from that to the claim that their views cannot solidity we cannot draw any practical conclusions.
encourage any such attitudes, perhaps on the question- We cannot, in particular, conclude (as some today are
able ground that I cannot be responsible for muddles obviously inclined to conclude) that since science is
that people get into as a result of what I say.) But does objective and morality is not, we are objectively justi-
subjectivism leave everything where it was? Surely not fied in devoting ourselves to science, while only sub-
everything. The dialectical skeins we just went through jectively justified in protesting against injustice. For
are, I think, perfectly valid, as showing that indifferen- devoting oneself to science is as much a practical activ-
tism could not follow from subjectivism. But the reason ity as any other, and there is no more reason why that
why this is so is that the argument for indifferentism one should be objectively justified rather than any
requires, and subjectivism forbids, the mid-air position. other. Justifications for doing objective subjects are not
-But in forbidding the mid-air position, subjectivism objective justifications for doing those subjects, any
seems to have taken something away, for we at least more than the fact that there are deductive justifica-
seem to recognize the mid-air position (for instance, in tions of the theorems of Principia Mathemqtica means
some of subjectivism's own statements). Another way that there are deductive justifications of the projects of
of putting our bafflement is that we seem in these argu- reading, rehearsing, or discovering the theorems of
ments to have been given no special reason why the Principia Mathematica. All these are instances of that
mid-air position is debarred to morality. It was said that strangely tempting fallacy, the 'fat oxen' principle:
the mid-air position was no place for a moral thought. who drives fat oxen must himself be fat.
But it does seem to be a place for some sorts of thought, Equally, though more subtly, the fact that 'prag-
matic' political policies (i.e., policies which apply
29
 MORALITY S U B J E C T I V I S M : F U R T H E R THOUGHTS

sophisticated technical considerations to the pursuit of the nature and extent of disagreement which people en-
self-interest) can be evaluated one against another with gaged in those activities may, in the nature of the case,
greater expectations of expert agreement than more encounter. So let us look more closely at what the sub-
idealistic policies often achieve, does not, even if true, jectivist has to say, in this central respect, about the
show that we are on more objective and solid ground in unsettling contrast.
adopting pragmatic rather than idealistic policies. Once He may start by saying that if we engage in factual
we have adopted them we may be on more objective or scientific enquiry, then, facts being as they are, we
ground in working them, i.e., in deciding what to do are bound to reach some agreed scientific or factual be-
next; but we never get on to more objective ground liefs; but we are not bound, because we engage in moral
with reference to the question whether we ought to be thought, and the facts are so, to reach some agreed
doing this sort of thing at all. We are merely more com- moral beliefs. There is one element in this answer which
fortable, and if among the sophisticated experts, in has to be corrected straight away. For it is of course not
work. 1
true that if the facts are so, we are bound (granted fac-
So subjectivism, even with its unsettling contrast tual thought) to reach some agreed factual beliefs: the
between morality and science, is still not logically com-
mitted to producing practical consequences. It cannot
I facts may be hidden from us. The most we can say is
that if we recognize that the facts are so, we are bound
even yield the conclusion that we are more solidly justi- to reach some agreed factual beliefs. And this is a tauto-
fied in having factual beliefs than moral ones; or on logy, since our recognizing the facts to be so entails
more objective ground in pursuing a factual question our reaching some agreed factual beliefs. Perhaps the
than pursuing a moral one; or objectively justified in subjectivist can improve on this unexciting proposi-
seeking the truth about anything; or in trying to find tion by saying something like this: that if two obser-
b
scientific explanations of phenomena rather than rest- vers are in the same observational situation, and have
ing content with superstition. For all it gave us was that L the same concepts and are not defective as observers,
factual and scientific beliefs were objective; that we etc., then they will reach the same factual beliefs about
should seek factual or scientific beliefs is not itself a that situation. If the 'etc.' in this can be handled so as to
factual or scientific belief. make the statement true, it will almost certainly emerge
Perhaps the subjectivist will readily grant these as necessarily true as well. All right, says the subjec-
points. The unsettling contrast which he insists on tivist; but this is not necessarily true, indeed is not true
between factual enquiry and moral thought is not a con- at all: that if two observers are in the same observa-
trast in the grounds for starting on or pursuing those tional situation, have the same concepts, etc., then they
activities - rather, it is a contrast in what those two are bound to reach the same moral beliefs - and there
sorts of activities are, in what goes on wher, one is en- is the contrast. But, we may say, are we sure that this
gaged in them. In particular, it is a contrast concerning latter is not true if we posit that they have the same
 MORALITY SUBJECTIVISM: FURTHER THOUGHTS

moral concepts - which is surely the fair parallel? No, nam make); nevertheless, there are eliminatory proce-
it is not true even then, the subjectivist says; for if they dures which must be respected by persons within the
have, as they might, rather minimal moral concepts, scientific framework, and even the conceptual disagree-
such as merely the concept of what one ought to do, ments may with luck yield to impersonally accepted
they can agree on all the facts and disagree morally. criteria. Yet for some disagreements between persons
This is a central position of many philosophers who or groups both engaged in recognizably moral thought,
insist on the fact/value distinction. They represent dif- no such procedures exist. The contrast - some contrast
ferent moral outlooks as all using some common, - exists.
skeletal, moral concept and giving it different fillings or But why should it not? This, finally, might be the
contents. Now we might point out that a lot of moral point at which a philosopher who had been arguing
thought does not operate with such impoverished con- with the subjectivist all this time might at last turn
cepts, that we standardly think in terms of more complex round and say: of course the contrast exists; morality
ideas of virtues, types of wrong action, etc., notions of is not just like science or factual knowledge, and it is
theft, for instance, or cowardice, or loyalty, or the duties essential that it should not be. The point of morality is
of one's job. With those more substantial concepts, there not to mirror the world, but to change it; it is concerned
is much greater hope that if we use the same concepts, with such things as principles of action, choice, respon-
we will reach agreement, or at least the sorts of dis- sibility. The fact that men of equal intelligence, factual
agreement which we may reasonably expect, and which knowledge, and so forth, confronted with the same
are less unnerving. But the subjectivist will say that situation, may morally disagree shows something about
there is this asymmetry, that in scientific or factual morality - that (roughly) you cannot pass the moral
contexts, if two observers have different concepts they buck on to how the world is. But that does not show (as
may ultimately find either that the concepts are in subjectivism originally seemed to insinuate) that there
fact equivalent, or else find reason for preferring one to is something wrong with it.
the other in terms of predictive success, explanatory Some such statement - and I have only blocked it in
power, and so forth, and there is no parallel to this in the very roughly - will express the culmination of the pro-
moral case. ject of what I called 'defusing' subjectivism. It has, I
Even if we abandon the ndive view (which some sub- think, to be granted some success. The most obvious
jectivists, in working their contrast, are attached to) ways in which somebody might be unnerved by sub-
that science 'proves' things; even if we accept that jectivism seem to have been blocked. Thus, to revert
what science does is eliminate hypotheses and that once again to the indifferentism we discussed before,
there are infinitely many hypotheses which have never the 'defuser' can arrange some of the subjectivist's
been eliminated because they are too dotty for anyone materials into the following argument. We observe that
to bother to test them (a point I have heard Hilary Put- when men of equal scientific or historical competence,
32
 MORALITY S U B J E C T I V I S M : F U R T H E R THOUGHTS

equal perceptual and intellectual abilities, etc., disagree I right,' this will naturally be taken as an expression of
strongly about some scientific or historical matter, there somewhat shaken confidence also. But now, in the fac-
is good reason for them to stop disagreeing so strongly, tual case, there is a possible thought which seems rather
and recognize something which their very disagreement, similar, but is not the same: the thought 'I am con-
granted their knowledge and skills, reveals, namely that vinced that p, but it is possible for all that that not-p,'
the matter is uncertain : it is rational for them, and third I where this is not an expression of doubt at all, but
parties, to suspend judgement. One may be tempted to rather registers the impersonal consideration that how
think that the same should obtain in cases of moral )I things are is independent of my belief; however they
disagreement; but this will be a mistake. It depends on I
are, they are, whatever I believe. We do not know exact-
first contrasting morality and factual knowledge, and ly what content to assign to this thought, but unless
then assimilating them. For the vital difference is that the we have the most drastic philosophical views, we are
disagreement in morality involves what should be done, convinced that it has a content: and the defuser will
and involves, on each side, caring about what happens; agree. But even defused subjectivism leaves no parallel
and once you see this difference, you see equally that it thought possible on the moral side: for subjectivism,
could not possibly be a requirement of rationality that however defused, there just is no content to 'I am con-
you should stop caring about these things because some- vinced that racial discrimination is intrinsically wrong,
one else disagrees with you. but it is possible for all that, that it is not,' except things
This, and the similar arguments, do seem to me to like 'How convinced am I ? ' or 'I suppose somebody
show that the defusing operation, in certain vital re- might make me change my mind.'
spects, has succeeded. Has it totally succeeded? If it has, Such a contrast (and it, and related contrasts, need
then we were wrong, some time back, in feeling uneasy proper investigation, which we cannot attempt here)
because subjectivism banned, for morality but not for might make us agree with the third subjectivist formula
factual beliefs, something we called the 'mid-air posi- we originally introduced: there are no ethical facts.
tion'. But, I think, we were not entirely wrong, in feel- Yet once more the defuser will say : this is just one more
ing that unease - perhaps we can now see why. formulation of what I said already, and of something
If I have a factual disagreement with a man, I might essential, not detrimental, to morality. For I have al-
think: 'I believe that p, he believes that not-p. Perhaps ready said that moral thought is essentially practical; it
he is right.' The natural way of taking this is as an ex- is not its business to mirror the world. - But now we
pression of doubt, of somewhat shaken confidence; the might reply: you said that it was not its business to
way I mean 'perhaps he is right' implies that while I mirror the world of empirical facts, and we agreed. But
still believe that p, I do not believe it as strongly as all did we agree that it mirrored no facts at all? And here
that. In this way of taking it, it can be painlessly paral- the locus of our dissatisfaction may become clearer in
leled for the moral case; for if I think 'perhaps he is the thought that the reason why even defused subjectiv-
 MORALITY SUBJECTIVISM: FURTHER THOUGHTS

ism seems to have left something out is that moral might have thought when we started. If subjectivism,
thinking feels as though it mirrored something, as however, defused, is true, things are with morality not
though it were constrained to follow, rather than be quite as they seemed; but the fraud, we might say, just-
freely creative. When we see further that many defus- ifies at most resentment rather than panic. We shall not
ing philosophers express the essential difference be- however lose sight of the idea of constraints on moral
tween factual and moral thought in terms of a contrast thought, the limitations on the creation of values. We
between the intellect and the will, and represent the shall come round to them by another route, through
responsibilities of morality in terms of our deciding on the idea of goodness. To talk about goodness, we shall
certain moral principles - then we have reason to be start with 'good'.
dissatisfied, either with them, or, if they are right, with
moral thought. For certainly the consciousness of a
principle of action as freely decided upon is very unlike
the consciousness of a moral principle, which is rather
of something that has to be acknowledged. If it is then
said that there is just a psychological explanation of
that - then moral thought seems a cheat, presenting
itself to us as too like something which it is not.
These remarks only gesture towards a centre of dis-
satisfaction. They leave almost everything to be done:
and not perhaps first in moral philosophy. For instance,
we need to distinguish two things run together in what
was just said, the idea of realism - that thought has a
subject matter which is independent of the thought -
and the idea of thought being constrained to certain
conclusions. Thus mathematical thought has the latter
property, but it is a deep and unsettled question in the
philosophy of mathematics how far we should or can
think in realist terms about the subject matter of math-
ematics.
Here I shall leave the direct discussion of subjectivism
and issues that it raises, with the conclusion that de-
fused subjectivism does not leave everything where it
was, but that it does leave more where it was than we
36
 way left very obscure by his theory) observation of a
'GOOD' thing's empirical characteristics was doubtless relevant
to the apprehension of goodness.
Apart from its evidently mysterious and unexplana-
tory character, there is a logical objection to Moore's
account. A very important feature of 'good' is that, in
THEuse of the word 'good' has provided a focus for many of its occurrences, it functions as an attributive
many discussions of basic issues in moral philosophy; and not as a predicative adjective (as P. T. Geach has
while it would be a mistake to think that this word, or expressed the distinction).* 'Yellow', for instance, is a
its approximate equivalents in other languages, could predicative adjective, because a sentence such as
possibly bear by itself the weight of the issues, never- That is a yellow bird
theless the consideration of it provides a useful lever
for lifting up some of them. We shall start with some admits of the analysis
logical considerations: these will lead us to things of That is a bird and it is yellow.
greater moral substance. By the same token, from the two sentences
As Aristotle observed, 'good' is used of many differ-
ent sorts of things, of things indeed in different cate- That is a yellow bird
gories. While in one way we do not mean the same A bird is an animal
when we apply it to these Merent sorts of things - in we can infer the conclusion
this sense, that what makes a general a good general is That is a yellow animal.
different from what makes a doctor a good doctor - But the sentence
nevertheless the word is not just ambiguous : we could
not tidy language up and say just what we want to say He is a good cricketer
now by replacing 'good' with a different expression in cannot be analyzed as
each of these occurrences.
More than one theory in recent times has tried to pro- He is a cricketer and he is good
vide a model to show that 'good' is genuinely unam- nor can we validly infer from
biguous. One such attempt was that of G. E. Moore, He is a good cricketer
who claimed that goodness was a simple indefinable A cricketer is a man
property like yellowness, but that, unlike yellowness,
to
it was nonnatural - that is to say (roughly), it was not
the sort of property whose presence or absence could be He is a good man.
established by empirical investigation, although (in a *P. T. Geach, 'Good and Evil'. Analysis, Vol. 17(1956).

38
 MORALITY 'GOOD'

An adjective which has this latter characteristic, that 'large' is a comparative term, and 'This is a large
it is logically glued to the substantive it qualifies, may mouse' means something like 'This is a mouse larger
be called an attributive adjective; or, more precisely, a than most mice'. By a similar analysis the conclu-
use in which it is so glued may be called an attributive sion (c) means something like 'This is an animal larger
use of it. Now Moore's account claimed that 'good' than most animals', and we can see why the inference
was like 'yellow' in standing for a simple quality, fails. What can be validly inferred from these premises,
though unlike it in that the quality was nonnatural; on the comparative analysis, is
and, mysterious as this is, it must at least imply that (d)This is an animal larger than most mice
the logical behaviour of 'good' as an adjective must be
like that of 'yellow'. But it is not, and so Moore's ac- and indeed this conclusion does follow.
count must be rejected not just as unilluminating, but Is the attributiveness of 'good' to be explained in the
as radically misguided. same way? It can hardly be that all attributiveness
Another important attributive adjective is 'real' - an could be so explained - a real Van Meegeren is not one
assertion that something is real can be understood only more real than most Van Meegerens. But it is more plau-
if we can answer the question 'a real what?'. This is sible to suggest that 'good' in 'good F' is attributive
illustrated by the situation in the art world, in which because 'good F' means something like 'better than
collectors are interested in acquiring the work of cer- most Fs'. But further consideration shows that this will
tain forgers, so that it can be in people's interests to not do. We have just seen that on the comparative ana-
forge forgeries : thus the question might arise whether lysis of 'large' we can get validly from the premises
this picture was a real Van Meegeren, everyone know- (a) and (b) to the conclusion (d). If 'good' were attribu-
ing that at any rate it was not a real Vermeer. tive because it is comparative, then similarly we could
The characteristic of attributiveness needs, however, get from
deeper exploration if we are to get an understanding of He is a good cricketer
'good'. We can see by the tests that 'large' is attribu- A cricketer is a man
tive : thus there is no valid inference from to
(a) This is a large mouse He is a man better than most cricketers
@) A mouse is an animal but this conclusion does not follow, and is just as ob-
to jectionable as the original one, 'He is a good man'. We
should get nearer to an acceptable conclusion only with
(c)This is a large animal. something like
The explanation of the failure of this inference, and He is a man better at cricket (or better as a cricketer) than
-
of the attributiveness of 'large', is clear it is that most cricketers
40 41
 MORALITY 'GOOD'

and even this is dubious, since there seems no contra- within the broad and abstract framework. In particular,
diction in the idea that, so flourishing is the state of the there can be differing or changing views of what aspects
game, most cricketers are pretty good; if this is possible, of a human activity are to have what sort of weight in
the comparative analysis in anything like its present
form disappears altogether. But apart from that, the
comparative analysis in any case has not unstuck
f the evaluation : thus one person, or one age, may take a
diBerent view from another about how important it
is for a good general to gain his victories at a minimum
'good' from its substantive; in the conclusion above, loss of life. Nevertheless, an understanding of what an
the connection persists, with 'good' now in its compar- x is does seem, in these cases, to contain a general un-
ative form. So the attributiveness of 'good' demands a derstanding of the criteria appropriate to saying that
more intimate coilnection with its substantive than is something is a good x: we are not just free to invent
demanded in the case of a merely comparative adjec- criteria of goodness. The clearest cases of all are, of
tive like 'large'. course, those of technical descriptions of artefacts; if
Since 'good' in this sort of construction is intimately someone went into an aircraft factory and said 'that's
connected with the substantive that it qualifies, the a good aerofoil', with reference to a rejected prototype
meaning of a phrase of the form 'a good x' has to which was in fact ill-designed or ill-executed, he would
be taken as a whole; and its meaning is partly deter- just have made a mistake; and if he then explained that
mined by what fills the place of 'x'. Can we go further he said this because its shape or its polish took his fancy,
than this and say that in phrases of this form, the mean- this would not make his remark any better, because
ing of the whole is essentially determined by the mean- these are not criteria of being a good aerofoil, though
ing of what takes the place of 'x'? In many cases, it they may well be appropriate to some other evaluation
looks as though we might take this further step. For if of this piece of metal, for instance as an aesthetic object.
we consider functional descriptions of artefacts, such (This illustrates once more the importance, with evalua-
as 'clock' or 'tin opener', or again descriptions of hu- tion, of the question, under what concept is the thing
man beings which refer to their roles or jobs or skilled being evaluated.)
activities, such as 'gardener' or 'general' or 'cricketer', There is a powerful tradition in contemporary philo-
it does seem that if one understands these expressions sophy of resistance to the idea that criteria of value.
(at least in the strong sense that one understands what what makes a thing of a certain sort a good thing of
a tin opener is, for instance, or what a general does), that sort, could ever be logically determined by factual
then one has understanding, within limits, of what a or conceptual truths : this is a central application of the
I
good thing of that sort is. distinction between fact and value which we have al-
This understanding may be at a very general and ab- ready referred to. This resistance was influentially en-
stract level and there will be a lot of room for disagree couraged by Moore, who invented the phrase 'the
ment, and for sophisticated comparisons of merits, 1 naturalistic fallacy ' for a mistake allegedly committed
 MORALITY

by any view which held that the goodness of a thing such as commending or prescribing on the one hand,
could be identified with some set of empirical, or in- and the meaning of sentences such as 'this is a good x'
deed metaphysical, characteristics. Moore exposed this on the other. At best, the utterance of those sentences
mistake in terms of his own view that goodness was a will constitute an act of commendation, etc., only if
nonnatural property, a view which we have already the sentences are actually asserted; but the sentence has
seen to be, in so far as it is comprehensible at all, a sad the same meaning whether it is asserted or not. Thus we
error. Many modern philosophers who agree that understand the sentence 'This is a good film' in the con-
Moore was in error in that view, nevertheless agree with text 'If this is a good film, it will get an Oscar'; but in
him that 'the naturalistic fallacy' is a genuine and im- that context it is not asserted, and no actual act of com-
portant fallacy; they give a new explanation of its na- mendation has occurred.* Thus the connection between
ture. This explanation, crudely put, comes to something meaning and commendation must at least be indirect.
like this: that the function of statements of the form Second, the theory seems too readily to assume that
'that is a good x' is to prescribe or commend, or to per- the functions of commending, etc., and of describing,
form some such linguistic purpose in the general range are exclusive of one another. Yet one and the same ut-
of the normative or evaluative, whereas merely to de- terance may carry many speech-acts at once: if I say
scribe the characteristics of x is not to perform such a 'Tomorrow will be wet', I may at once have described
purpose; and no set of statements which do not perform tomorrow's weather, made a prediction, given you a
such a purpose can logically entail any statement which warning, etc. Further, the question of whether I do, in
does. To prescribe, commend, etc., is to do something, making a certain utterance, perform any of these
which (to put it roughly) the facts by themselves cannot speech-acts, may be determined by what I say together
make us do; we must have some evaluative or prescrip with the facts of the situation. Thus if I say 'The ice is
tive attitude which favours certain characteristics, if thin'. I will have described the ice, no doubt, but also,
those characteristics are to count with us as grounds of in the light merely of your interests and purposes, may
approval. Merely knowing about the world, or under- have done something that counts as giving you a
standing concepts, cannot in itself be enough to bring warning. Rather similarly (though not exactly simi-
this about. larly), if I say, in a description of this clock, that it
A full examination of this position requires some- keeps the time exactly, needs no winding, never breaks
thing that would take us too far in this essay, an en- down, etc., I will have done something like commend-
quiry into an important and developing field in the ing it as a clock quite independently of some supposed
philosophy of language, the theory of speech-acts, the choice of criteria for clocks on my part. Of course, the
various things that we can do by making utterances. *See J. R. Searle, 'Meaning and Speech-Acts', Philosophical
Three points can be briefly made. First, there cannot Review, Vol. 71 (1962). and Speech Acts (Cambridge University
be any very simple connection between speech-acts Press, 1969),Chapter 6.

44
 MORALITY

facts about this clock, and the general nature of clocks, When we examine this question, we find cases where
cannot bring it about that 1 make the remark - I may the going is much harder, and where the factlvalue dis-
remain silent: but they determine that if I give a true tinction, or something like it, gets a better grip : where
description of this clock in these respects, then I do some deeper considerations about value emerge. What
something in the range of commending or giving a fav- much contemporary philosophy has presented as a very
ourable evaluation of the clock. general logical doctrine about the workings of 'good'
This last point brings us to the third consideration. may be seen to have its interest rather as a much more
Activities such as commending, etc., are essentially restricted, and not purely logical, doctrine about the
overt activities, connected with actual utterances; this goodness of things and, more particularly, persons, un-
is why I have said that their study belongs to the theory der certain specific descriptions.
of speech-acts. But no account of sentences containing
'good' could possibly be adequate which remained
merely at this level; for it is possible merely to think, or
believe, or reach the conclusion that something is good
of its sort, without making an utterance to that effect at
all. Hating Bloggs, I may carefully refrain from making
favourable remarks about his performance as a crick-
eter, i.e., precisely not commend him or do any such
thing; yet in my thought I may be forced to recognize
that he is a good cricketer. An adequate account must
leave room for such recognition
These extremely sketchy considerations may indi-
cate some difficulties in the revamped version of the re-
jection of the 'naturalistic fallacy'. I think, in fact, that
as a general doctrine about the workings of 'good' the
harsh distinction between fact and value that goes with
that view is mistaken : it seems clear that for many fill-
ings of 'x' in 'that is a good x', an understanding of
what an x is or does, and factual knowledge about this
x - i.e., a combination of conceptual and factual infor-
mation - is sufficient for one to determine, at least
broadly, the truth or falsity of the judgement. This, cer-
tainly, is objectivity. But is it so for all W g s of 'x'?
46
 GOODNESS A N D ROLES

father' - unless we meant that he was good at becoming

GOODNESS AND ROLES a father.
Nor do we necessarily advance towards evaluative
notions merely by introducing a reference to social in-
stitutions such as marriage. 'Wife's brother' is a well-
defined kinship relation, involving a reference to
CONSIDER the expression 'a good father'.* While it is marriage, but the idea of being a good wife's brother is
moderately clear in outline what sort of criteria go with one which, in our society, lacks any content. The con-
this expression, it is not at all clear that an understand- tent of 'a good father' is determined by the biological
ing of these is involved in understanding what it is to relationship together with certain responsibilities
be a father. Nor is this merely because the idea of being which, in our sort of society, are ascribed to people in
a good father contains a reference to certain social con- that relationship. Only when there are such responsi-
ventions; for so, indeed, does 'cricketer' or 'bank clerk' bilities can we say that 'father' refers not merely to a
- to know what a bank clerk is involves knowing a good biological relationship, but to a role; and the role can,
deal about the social fabric in terms of which the role of course, exceptionally be played by someone not in
of a bank clerk is defined - but when I understand that the biological relationship, as when someone is said to
role in those t e r n , I also understand in outline the have been a father to an orphaned child. The difference
sorts of things that a man would have to do in order to from 'bank clerk' then emerges in this : that there is no
be called a good bank clerk. The difference with the notion of 'bank clerk' which does not involve a refer-
idea of a good father is that it looks as though one can ence to the responsibilities, and the term refers to a
have a perfectly clear idea of fatherhood, and this not role which can only be explained in relation to social
in itself lead one to an understanding of the sorts of institutions which give someone with that role certain
things that make someone a good father. The explana- functions and duties. Whereas under the role-concept of
tion of this difference lies partly in the fact that the fatherhood there lies the more restricted concept of it as
idea of fatherhood which we can grasp without a purely natural relationship, a relationship which in
grasping the evaluative criteria is an idea of father- our sort of society is taken as the basis of that sort of
hood merely as a biological relationship; but it can- role.
not be merely that idea which occurs in the notion If a man is an ardent cricketer and perhaps has en-
of a good father. If we have in mind the idea of tered cricketing as a profession, then his performance
a father only as a procreator, it is not clear what as a cricketer and assessments of that performance by
we could even mean by calling someone 'a good the competent, will obviously matter to him; if he does
*See G.Cohen. 'Beliefs and Roles', Proceedings of the Aristote- poorly, he has failed. A man who modestly plays some
lian Society, Vol. 67 (I-). part on the cricket field on a Sunday afternoon will,
 MORALITY GOODNESS AND ROLES

equally obviously, tend to care less; the evident truth this if he straightforwardly thought that 'soldier' was
that he is a poor cricketer means little to him - he might a title that applied to him only because it had been
say that he wasn't a cricketer but merely someone who forcibly applied to him from the outside; cared nothing
occasionally played cricket. A man who was a very in- for the assessments that go with that title; regarded the
different bank clerk, doing barely what was expected hostility of his superiors as a blank external force like
of him, might equally, though in a rather different way, the force that had put him into the army; and felt as the
have dissociated himself from doing well in those activ- only constraint on opting out as much as possible, the
ities. Me may hate the bank, despise banking, and care fear of punishment. The straightforwardness of this
only about his friends and growing chrysanthemums. attitude may be bought at the cost of a certain despair,
The bank he regards only as the means to a living; he since it naturally goes with a feeling (as in Catch 22)
does not, in any important sense, see himself as a bank that the environment is, just as a matter of brute fact,
clerk. Yet he could hardly say that he wasn't a bank insane.
clerk (really), or if he did, he might be in danger of re- In other cases, a man may come to dissociate himself
treating into fantasy; for too much of his life is in fact from a role and the assessments that apply to it, with
bound up with this role which, like it or not, he has. which previously he has wholeheartedly or unreflect-
The unfavourable attitudes and lack of respect which ingly identified himself, because he cannot bring him-
he encounters from his superiors in the bank, while he self to do something which he is expected to do in that
may have succeeded in caring little for them, he at role. In favourable circumstances, he may be able to re-
any rate cannot dismiss (as the cricketing man can) as sign from it. If that is not possible, he may feel required
based on a misunderstanding: the contractual relation- to refrain in some other way from doing what would
ship of his job puts him in a position where this is what be expected of someone in his role: in an extreme pol-
he must expect. Though he may have reached stability itical case, possibly by covert disobedience and subver-
in this situation, by irony perhaps, the situation is not a sion. It is said of certain German generals who during
happy one, and if he came into money he might well the war were appalled by Hitler's policies that they were
(if it did not now require too much courage) get rid of for a long time inhibited from setting to work against
the role and resign from the bank him by considerations of the oath they had taken, as
Contrast with these simple prototypes, another - that officers of the Wehrmacht, to obey him. One might
of the unwilling drafted soldier. His life, very probably wonder, from outside the situation, how any oath
more miserable than that of the bank clerk, could in- taken to that man who was then in that condition and
volve a yet more uneasy and ambiguous state of mind ruling by those means could have been regarded as in-
- if, for instance, he was just unsure whether, being a extricably bound up with one's duties even as a Ger-
soldier, he should try to put up a good performance as man officer- this is to regard oaths as having, not just
a soldier. But his state of mind could be simpler than sanctity, but magical powers. But suppose obedience to
50
 MORALITY GOODNESS AND ROLES

the oath could be established as certainly part of one's refuse to make those standards the determinants of his
duties as a German officer; then what the generals life.
needed was to form the conception of things they had It does not follow from these possibilities that some
to do which were contrary to what they were supposed one who unreflectively and passively leads a life struc-
to do as German officers - that is to say, it was no longer tured by some role which he takes for granted must in
under the title, and in the role, of German officer that fact have chosen that role, as Sartre (some of whose
they had to act. And this was a conception which, it concerns are mirrored in these remarks) seems, in his
seems, some of them found it very difficult to form. classical Existentialist period, to have thought. Hence
For them, 'a German officer' was not just something his description of an unreflective condition of this sort
that they were, but what they were. as a condition of bad faith. This requires at least that
These schematic examples are supposed to illustrate every man has the possibility of reflection and choice,
ways in which men may dissociate themselves from and that this possibility is both recognized and re-
roles which they bear, roles which bring with them cer- nounced. Yet even the first step, of a real possibility of
tain sorts of assessment of their activities. This dissocia- reflection and choice, may be too much to ascribe to
tion may be, in different cases, variously defensible or men in some social and psychological situations.
indefensible, noble or ignoble, prudent or the reverse; Sartre's view is connected with his extremely nonempir-
but it is in every case intelligible, and' it is comprehen- ical notion of freedom, a notion which may also help to
sible how a man can form a coherent picture of himself explain a certain ambiguity in his thought between the
in relation to a role which he has or used to have or ideas that man has no essence, and that man has an
might be thought to have, in which he does not allow essence, which is freedom.
the standards that go with that role to be ultimate, basic, If there were some title or role with which standards
or important in the assessment of the success or excel- were necessarily connected and which, by necessity, a
lence of his life. Now this possibility seems to me a vital man could not fail to have nor dissociate himself from,
counterbalance to a fact which we have already no- then there would be some standards which a man
ticed, namely that these various sorts of title and role would have to recognize as determinants of his life, at
can conceptually carry with them broad standards of least on pain of failing to have any consciousness at all
assessment of people under those titles, as the descrip- of what he was. There is certainly one 'title' - for good
tions of artefacts can cany standards of assessment of reason, we can scarcely speak here of a 'role' - which
those artefacts. While the standards can be in this way is necessarily inalienable, and that is the title of 'man'
logically welded to the title, the title is not logically itself. So it is a central question whether 'man' is a con-
welded to the man; hence the standards are not logic- cept which itself provides standards of assessment and
ally welded to the man. Through his consciousness excellence as a man; for if it does, then it seems that
-of a given title and his relation to it, a man may they must be our standards.
52
 MORALITY

There have famously been philosophies which have

held that it did: that either directly, by reflection on MORAL STANDARDS AND THE
the concept of 'man', or more indirectly, by reflection DISTINGUISHING MARK OF MAN
on some supposed further necessary truths about what
man is, one could arrive at an understanding of what a
good man must be.
Such philosophies can helpfully be distinguished, at A PROTOTYPE of the nontranscendental type is to be
least initially, into two sorts - those that do, and those found in the philosophy of Aristotle. According to
that do not, make a transcendental appeal, that is to say, Aristotle, there are certain characteristics, in particular,
an appeal to some framework for human life which lies certain activities and powers, which are distinctive of
outside human life and the empirical world. We may man, and the life of the good man will exemplify to the
start with the nontranscendental type and come later fullest degree the development of those powers and
to an example (a religious example) of the transcen- activities. Or, more accurately, there is one distinctive
dental type. feature of man - his ability to shape his actions and
dispositions by reason - which will be manifested in
the highest degree; other of his potentialities will, under
the ordering power of reason, be realized in a balanced
way and not each maximally. Practical reason is sup-
posed to produce coherence, and reduce conflict, among
the desires of the individuaI living (as man must live) in
society. This aim of reducing conflict between desires,
while not inordinately suppressing them, is part of
what gives recognizable content to the claim that the
aim of the sort of life outlined in Aristotle's system is
happiness.
The importance of the harmonization of desires in
Aristotle, and of practical reason in securing this, is
illustrated in a backhanded way by his notable failure to
deal with one problem of reconciliation which, in his
own terms, must be important. The 'reason' that we
have so far referred to is practi~alreason, which applies
to particular actions and desires and which is the
ground of what Aristotle (or rather his translators) call
55
 MORAL STANDARDS
MORALITY
their expression is ruled out, and a less than total com-
'virtues of character' - that is to say, those dispositions
mitment is not represented as something that practical
to right action which involve motivations of pleasure
thought can rationally arrive at. This weakness in
and pain. There is also, however, theoretical reason, the
Aristotle's system is a model of a more general prob-
power of thinking correctly about abstract questions of
lem : what might possibly be called the 'Gauguin prob-
science and philosophy, which Aristotle is disposed
lem', except that that label imports rather special and
to regard as a yet higher expression of man's nature:
Romantic connotations of the claims of self-expres-
the most excellent form of human life, accordingly, is
sion. A moralist who wants to base a conception of the
one devoted in fair degree to intellectual enquiry. He
right sort of life for man on considerations about the
makes it clear that since man is man, and not a god, his
high and distinctive powers of man can scarcely dis-
life cannot solely be devoted to this, and he must also
regard the claims of creative genius in the arts or
have a life to which the virtues of character are neces-
sciences to be included pre-eminently among such pow-
sary* ers; yet he will find it hard to elicit from, or even recon-
What Aristotle does not do, however - and granted
cile with, an ideal of the development and expression
his system, cannot do - is to provide any account of
of such genius, many of the virtues and commitments
how the intellectual activities, the highest expression
which belong to morality, some of which are merely
(in his view) of man's nature, are to be brought into
more everyday, while most make demands on one's re-
relation to the citizenly activities which are regulated
lations to other people which are quite different from
by the virtues of character. Practical wisdom does not
those made by creative work.
stretch so far and could not impose a 'mean' between
Plato, it must be said, saw one half of this problem
philosophizing or doing science on the one hand and
with extreme clarity. The famous banishment of all but
being a good citizen, father, etc., on the other. It is a
celebratory art from his Republic (a state designed pre-
curious, and significant, feature of Aristotle's system
cisely as an institutionalization of morality) was the
that the highest potentialities of man have admittedly
product of his taking art seriously and his seeing per-
to compete with others for expression, but no coherent
fectly correctly that both the life of the creative artist
account can be given of how this competition is to be
and the effects on a serious public of the free explora-
regulated.
tion embodied in his creations were likely to resist the
This is one weakness of Aristotle's attempt to elicit
demands of stability made by the fully moralized soc-
the good for man out of man's nature by appealing to
iety. We may well reject the alternative Plato chose, but
a distinguishing mark of man, his intelligence and ca-
he was surely right in seeing these as alternatives: a
pacity for rational thought: a weakness that can be
society protected against moral and social change, and a
seen in the following light, that the pure or creative
society in which free creation and exploration through
aspects of intelligence would seem to be the highest
art is permitted. The reason why (as I put it) he saw only
form of these capacities, yet a total commitment to
 MORALITY MORAL STANDARDS

one half of this problem is that he did permit, indeed conflict with the unimpeded development of human
made the central activity of his ruling class, creative in- creative and intellectual aspirations; it is, for us, much
tellectual activity - and one may well wonder whether too late in the day to say that because natural science
this would have proved any more compatible in the clearly constitutes one of the highest achievements of
long run with the kind of stability he longed for. (It men, its development must have an unchallengeable
seems that he admired the intellectual achievements of claim on our moral approval. It would be as optimistic
Athens while deploring its political and moral disorder, for us to think this, as it was for Plato to think that the
and admired the political stability of Sparta while re- pursuit of these activities could coexist with the degree
gretting its unintellectual militarism. It must surely of social restriction and illiberalism which he identi-
have occurred to him to wonder whether it was just an fied with the moralized society.
accident that these qualities were distributed as they This illustrates one, rather central, objection to the
were.) Part of the explanation is that where we speak Aristotelian enterprise, with reference to the particular
of 'creative activity', Plato thought largely in terms of (and plausible) selection of intellect as the distinguish-
discovery: the philosophical activity was in his view ing mark. There are more general objections to the pro-
inherently disciplined by the a priori moral truths wait- cedure of trying to elicit unquestionable moral ends or
ing to be discovered. If we reject this picture of moral ideals from distinguishing marks of man's nature. We
philosophy and admit further that in natural science may mention three. First, a palpable degree of evalua-
there is an ineliminable creative element, these intellec- tion has already gone into the selection of the distin-
tual activities will look less like natural inhabitants of guishing mark which is given this role, such as
the kind of environment which Plato provided for them. rationality or creativity. If one approached without pre-
I am not saying (what seems to me high-minded non- conceptions the question of finding characteristics
sense) that the very activities of scientific enquiry which differentiate men from other animals, one could
require or presuppose that persons engaged in such as well, on these principles, end up with a morality
enquiry will have liberal and humane values, opposed which exhorted men to spend as much time as possible
to an authoritarian social environment. Recent experi- in making fire; or developing peculiarly human physi-
ence suggests that the most that such enquiries presup- cal characteristics; or having sexual intercourse with-
pose is a certain liberalism in the environment of out regard to season; or despoiling the environment and
science itself, which can well coexist with a cynical upsetting the balance of nature; or killing things for
indifference to many humane considerations, concern- fun.
ing (for instance) how scientific discoveries are em- Second, and very basically, this approach bears out
ployed or the source of funds which support the the moral ambiguity of distinctive human characteris-
research. But this itself is only another illustration of tics (though Aristotle paid some attention, not totally
the various ways in which demands of morality can successfully, to this point). For if it is a mark of a man
58
 MORALITY MORAL STANDARDS
to employ intelligence and tools in modifying his en- sions themselves as blind causal forces or merely animal
vironment, it is equally a mark of him to employ intel- characteristics. To be helplessly in love is in fact as dis-
ligence in getting his own way and tools in destroying tinctively a human condition as to approve rationally
others. If it is a mark of a man to have a conceptua- of someone's moral dispositions. But it is easy to see
lized and fully conscious awareness of himself as one why, in the present direction, Manicheanism looks in-
among others, aware that others have feelings like him- viting. If rationality and consistent thought are the pre-
self, this is a precondition not only of benevolence but ferred distinguishing marks of man, then even if it is
(as Nietzsche pointed out) of cruelty as well : the man admitted that man, as a whole, also has passions, the
of sadistic sophistication is not more like other animals supremacy of rational thought over them may well
than the man of natural affections, but less so. If we seem an unquestionable idea'. This is all the more so,
offer as the supreme moral imperative that old cry, 'be since it is quite obvious that gaining some such control
a man!', it is terrible to think of many of the ways in is a basic condition of growing up, and even, a t the ex-
which it could be taken literally. treme, of sanity. But to move from that into making
Here we seem to encounter a genuine dimension of such control into the ideal, rules out a priori most forms
freedom, to use or neglect the natural endowment, and of spontaneity. And this seems to be absurd.
to use it in one way or another: a freedom which must All these considerations suggest that the attempt to
cut the central cord of the Aristotelian sort of enter- found morality on a conception of the good man
prise. Nor can this freedom itself be used as the distin- elicited from considerations of the distinguishing
guishing mark of man, and the enterprise mounted marks of human nature is likely to fail. I am far from
again on the basis of that. For this freedom can surely, thinking that considerations about human nature, what
by its nature, determine no one form of life as against men are, what it is for men to live in society, do not
another - as Sartre, in virtue of a central ambiguity al- contribute to a correct view of morality. Of course they
ready mentioned, perhaps has thought. One might say: do: one could not have any conception of morality at
if there were a distinctive form of life, that of 'realizing all without such considerations. In particular, they
freedom', then there must still be a freedom to reject help to delimit the possible content of what could be
that, too. regarded as a morality. Just as obviously, differing
Third, if we revert to that particular case of the views of human nature (as, for example, some psycho-
rational as the distinguishing mark of man: there is a analytical view) must have differing effects on what
tendency for this approach to acquire a Manichean lean- views one takes of particular moral requirements and
ing and emphasize virtues of rational self-control at the norms. Not merely scientific or semi-scientific views
expense of all else. There is no reason why such an out- must have this effect but also views in the philosophy
look should inevitably follow; apart from anything of mind. Thus a proper philosophical understanding of
else, it involves a false and inhuman view of the pas- the nature of the emotions should have a discouraging
60
 MORALITY

effect on Manichean views about their management,

and philosophical considerations about the nature, in- GOD, MORALITY, AND PRUDENCE
deed the existence, of something called the will must
have a direct effect on moralities which find in the
exercise of the will (against the desires, for instance) a
central clue to moral worth.
While all this is true, and while there are very definite WE distinguished some time ago among views which
limitations on what could be comprehensibly regarded seek to elicit a notion of a good man from considera-
as a system of human morality, there is no direct route tions of human nature - those that set man in some
from considerations of human nature to a unique transcendental framework and those that did not. Hav-
morality and a unique moral ideal. It would be simpler ing just said something about the second sort, I shall
if there were fewer things, and fewer distinctively hu- now turn to look at an example of the first. In the course
man things, that men can be; or if the characters, dis- of doing this, it will be helpful to discuss a question
positions, social arrangements and states of affairs which is important apart from the present issue - the
which men can comprehensibly set value on were all, relations of the moral and the prudential.
in full development, consistent with one another. But A leading feature of this sort of theory is that it seeks
they are not, and there is good reason why they are not : to provide, in terms of the transcendental framework,
good reason which itself emerges from considerations something that man is for: if he understands properly
of human nature. his role in the basic scheme of things, he will see that
there are some particular sorts of ends which are pro-
perly his and which he ought to realize. One archetypal
form of such a view is the belief that man was created
by a God who also has certain expectations of him.
A central difficulty with this lies in the question of
which properties of God are supposed to justify the
claim that we ought to satisfy his expectations. If it is
his power, or the mere fact that he created us, analogies
with human kings or fathers (often employed in this
connection) leave us with the recognition that there are
many kings and fathers who ought not to be obeyed.
If it is urged that God has infinite power and created
everything, we point out that infinite kinghood or
creatordom does not seem evidently more worthy of
 MORALITY GOD, MORALITY, A N D P R U D E N C E

obedience but merely more difficult to disobey. If it is and the introduction of God adds nothing extra. But if
then said that in addition to these other properties God they are not moral motives, then they will be motives
is good, the objection is forthcoming (as it was from of such a kind that they cannot appropriately motivate
Kant) that this already involves a recognition of what morality at all: in particular, they are likely to be
is admirable and valuable, a recognition of the kind motives of prudence, a possibility most crudely por-
which the appeal to God was supposed to underwrite. trayed by certain evangelists (whether of belief or dis-
Such arguments, which are very familiar, may be belief) in terms of hellfire. But nothing motivated by
taken as attacking the idea that one might work out prudential considerations can be genuinely moral ac-
purely deductively and a priori the required life for man tion; genuinely moral action must be motivated by the
from a description of him as created by God. In that consideration that it is morally right and by no other
role, they seem to me successful. But such arguments consideration at all. So, taking this all together, we
tend to carry with them a larger ambition - to show reach the conclusion that any appeal to God in this
that even if God's existence were established, that fact connection either adds nothing at all, or it adds the
could not in principle supply any acceptable or appro- wrong sort of thing.
priate motive for moral conduct, of a kind which would Two questions are raised about morality and moti-
otherwise be lacking. In this role, too, the arguments vation by this sort of argument. First, whether there are
are very widely accepted, so that it is practically a really no relevant types of motivation except moral or
philosophers' platitude that even if God did exist, that prudential - that is to say, whether the distinction be-
would not, to a clear-headed and moral thinker, make tween moral and prudential is exhaustive. Second,
any difference to the situation of morality. The origins whether a policy or outlook may not be moral, while
of this view go back to a famous discussion in Plato's being in some way at the same time prudential - that
Euthyphro, but in its modern form it owes most to is to say, whether the distinction is exclusive. Let us
Kant. It owes to him in particular a clear statement take the second question first. Is it essential to morality
-
of the assumptions on which it rests assumptions to distinguish totally the moral and the prudential?
about the essential purity of moral motivation. These Here we need to make some distinctions. It is cer-
assumptions are pervasive in much moral thought, and tainly true that it is essential to morality that a dis-
their influence and importance goes a long way beyond tinction is drawn at some level between the moral and
the present issue of a religious morality. They are also the prudential. At the most primitive level, it is clear
very importantly mistaken. that any morality has to apply this distinction, or some
The argument, in simplest form, goes something like thing like it, to actions and policies; it has to be able to
this. Either one's motives for following the moral word distinguish actions and policies which are selfish and
of God are moral motives, or they are not. If they are, which minister to the gratification or safety of the
then one is already equipped with moral motivations, agent at the expense of others, from those which take
64 65
 MORALITY GOD, MORALITY, A N D PRUDENCE
the interests of others into account. If some such disc merely approve of the act and do not in any sense a p
tinction is not made, there are no moral considerations, prove of the agent; for we can say of him that he has
at all. It is clear that the religious morality we are dis- intentionally done something which it is a good thing
cussing, however crudely put, can observe this distinc- to have done, and this surely constitutes some sort of
tion at the primitive level : it will, with respect to the approval, if a limited sort, of him. The point, presum-
secular world, approve policies and actions which take ably, is that we do not morally approve of him. There
the interests of others into account and will disapprove is a very good point in saying this, but it must not be
selfish policies. taken too far. For if we insist that to act morally is
However, it may be thought that this level of draw- essentially to act from a moral motivation, we may well
ing the distinction is, by itself, too primitive, and that be tempted to add to that the innocuous-looking p r e
we need to extend the distinction from merely classi- position that all that can matter from a moral point of
fying policies and intentional actions, to discriminat- view is that people should act morally, and then con-
ing between motives. Thus one who gives money to clude (rightly, from those premises) that from the moral
charity merely to improve his reputation with the point of view any two situations of self-interested moti-
Rotary Club or to ease his own guilt, acts no more vation are indistinguishable, and it must be impossible
morally than if he had spent the money on his own from the moral point of view to prefer one to the other.
pleasures. (The example illustrates why it was appro- This is absurd. It is not perhaps logically absurd, but
priate to speak of 'intentional action' and to distin- morally absurd - the Puritan moral absurdity that the
guish this from the question of motive. The self- only morally relevant property of the world is how
interested business man who writes a cheque to famine much righteousness it contains. But since this is morally
relief, does so intentionally, and his intention is that the absurd (or, rather, since it is obviously not incoherent
money should go to famine relief: if famine is relieved to regard it as morally absurd), it follows that some-
by his action, this will not be, relative to his thoughts in thing else is logically absurd : namely a view of moral-
so acting, an accident. The point is that his motive was ity from which it follows that that is the only moral
not a concern for the relief of famine, but for his own position which is coherently tenable.
reputation or comfort.) What is, then, the point and content of saying that
If we say, as many would say, that the man who so we do not morally approve of the self-interested donor
acts acts no more morally than one who spends the to charity, or that, though he does a good thing, he
money on himself, it will not follow that what he does does not act morally 1 With what motivations, first, are
is no better than what is done by the simply selfish man; we contrasting this man's motivation? Some, such as
for famine will be, hopefully, relieved, and this is better Kant and R. M. Hare, have laid emphasis on the con-
than that another combined cocktail cabinet and TV set trast with acting from principle; roughly, doing it just
should be bought. Nor, surely, can it follow that we because one thinks one ought to. Others, such as Hume,
66
 MORALITY GOD, MORALITY, A N D PRUDENCE

have emphasized the contrast with doing something be- reasonably concerned about himself. Under the Kantian
cause one cares disinterestedly about the situation emphasis, however, this suddenly emerges as a prob-
which one's action is supposed to alter or cares about lem, since to act with regard to one's own interests, in a
the other people involved. Leaving aside the notable dif- straightforward way, is to act from a kind of motive
ferences between these two formulations, they do have which has nothing to do with morality at all and is
something in common: that the man who has a moral indeed alien to it. Since we are presumably enjoined to
motivation for doing things of the non-self-regarding maximize moral action, extremes of selfdenial would
sort, has a disposition or general motive for doing things seem to follow, as derived, indeed, from the concept of
of that sort; whereas the self-interested man has no such morality itself. At the best, doing what one simply
steady motive, for it will always only be luck if what wants to do will constitute unregulated and probably
benefits others happens to coincide with what, by the guilty departures from the moral point of view. To
limited criteria of simple self-interest, happens to cope with this problem, the Kantian tradition produces
benefit him. This must surely, as Hume said, have some- a set of 'duties to oneself', recognition of which licen-
thing to do with the point of selecting certain motives ses one to do for moral reasons some of the things one
for moral approbation: we are concerned to have would be disposed to do anyway. This absurd apparatus
people who have a general tendency to be prepared to is just the product of trying to adjust to some rather
consider other people's interests on the same footinn " as
more reasonable view of human life the awkward con-
their own, and if necessary to put other people's inter- sequences of holding three things - that morality is
ests first. concerned above all with motivation, that moral
It is perhaps worth noting, in passing, that one of the motivation is motivation of principle, and that the
(numerous) advantages of Hume's emphasis in this mat- moral point of view must be ubiquitous. To avoid these
ter, with its stress on sympathy and feeling for other particular awkward consequences, it would be enough
people's situation, as opposed to the Kantian emphasis to abandon any one of these propositions; but there are
on acting from principle, is that it introduces a simi- good reasons for abandoning all of them.
larity between the sorts of reasons one has for doing To revert, after this diversion, to our religious moral-
things for others, and the sorts of reasons one has for ist. We saw that he could quite easily draw a distinc-
doing them for oneself. Despite the mechanical charac- tion, though a primitive one, between the moral and
ter of Hume's psychological system, it makes some the prudential. I think that we can see now that he can
sense of the idea that to care about others' pain is an even draw a more refined distinction, at the level of
extension of caring about one's own: the second is in- motivation. It was suggested that one (though perhaps
deed a necessary condition of the first, and there is cer- not the only) point of distinguishing between moral and
tainly no problem (as there should not be) about why a self-interested motivations was that of picking out
man who is concerned about others may not also be general dispositions to do things of the non-self-
69
 MORALITY GOD, MORALITY, A N D PRUDENCE

interested sort. But dispositions of this sort, the man times allowed, that of inclination, that is, doing some-
with even a crude religious morality will certainly thing because you feel like doing it. Clearly the list of
admit. Indeed, perhaps what his God wants is that men examples could be extended indefinitely to include vast
should feel for the sufferings of each other and act in numbers of the special relationships in which one per-
one another's interests because they so feel. So a lot of son can stand to another. It is a grotesque product of
the time persons of this belief, if they did as God would theory and strenuous moralism to suppose that 'moral'
like them to, would act from ordinary human motives and 'prudential' sufficiently divide up the justifiable
which by most people (other than Kantians) would be motives or reasons a man can have for doing something :
regarded as themselves moral motives. If these fail,,or they leave out, in fact, almost everything. We do need
temptation to selfish action is strong, then perhaps the something over and above these particular, or more
crude believer's thoughts turn a while to hellfire, and specific, motivations, just because they are particular
this fortifies a disposition to do things of the non-self- and specific, and especially because the particular rela-
interested sort (in the worldly sense of 'self-interest', tionship I have to another person may be of a hostile
that is). While his action is in this sense prudential, it character, and there will in all probability be no other,
is not prudential in that sense in which it is essential more beneficent, particular relationship in the offing
to the concept of morality that the prudential be con- to inhibit destructive conduct. So of course we need,
trasted with the moral. Indeed, there is a special reason over and above, general motivations to control and regu-
why his actions, though prudential, are not selfish: late these particular ones; and the most general thing
namely there is presumably no effective way of aiming that is over and above is morality. But, happily for
at salvation at the expense of others. humanity, we do not have to leave it to those general
In fact, it is quite unrealistic to force onto our reli- considerations to motivate everything of a desirable
gious moralist (or anyone else) an exhaustive disjunc- kind. Some of our decent actions come not from that
tion between the prudential and the moral. Leaving motive which Christians misrepresent as our loving
aside the more general operations of sympathy of which everybody, but just from our loving somebody.
Hume wrote: what about someone who does some- The religious moralist, now, can see the general re-
thing in the interests of another, and to his own disad- quirements as stemming from a particular relation, that
vantage, because he loves that person; or, indeed, is in to God, and this relation can be represented as one of
love with them; or admires them; or respects them; or love, or awe, or respect, or whatever words are found
because they are (after all) a member of the family ? None appropriate for this bafling semantic task. And this rela-
of these reasons for acting have to be moral reasons, in tion he will rightly resist being categorized as either
any exigent or purified sense of that term; equally they moral or prudential in its practical import. I think,
are not prudential reasons. Nor, again, do they belong however, that he may have to be careful about saying
to the third class of motive moral philosophy has some- that this is an attitude towards God which anyone who
70 71
 MORALITY

knows what God is ought to have, for this might, in the

way that Kant insisted on, make morality prior to God WHAT IS MORALITY ABOUT?
again. He will have to say, rather, that this is an atti-
tude which anyone who knows what God is will inevit-
ably have; God is one whose word exacts an
unquestioning acceptance. In itself, this is still not
enough, of course; compatibly with this much, God THISlast discussion has led us sideways into certain
could be an unfailing hypnotist. So the believer will questions of what morality is about and how 'the
proceed, as always, by negation and analogy and say moral' is to be delimited. A lot has been written about
that it is not like that, but more like a loving father, and this, in search of some criteria for distinguishing the
so forth. I myself doubt whether at the end he will pro- moral from the non-moral. These recent discussions are
duce any coherent account at all. But this is because of considered by G. J. Warnock in his admirably concise,
difficulties in belief in God, not because of something lucid, and forceful book Contemporary Moral Philos-
in the nature of morality. I do not think it right - and ophy :* he rightly emphasizes the extraordinary fact
these arguments have been trying to show this - to say, that a great deal of this discussion has proceeded in a
as many do, that even if God existed, this could give no vacuum, in pursuit of a criterion which might give us
special and acceptable reason for subscribing to moral- a way of distinguishing moral and non-moral, perhaps
ity. If God existed, there might well be special, and a way which bore some rough resemblance to ways in
acceptable, reasons for subscribing to morality. The which we, now, make such a distinction, but without
trouble is that the attempt to formulate those reasons shedding any light on, or being guided by, the evidently
in better than the crudest outline runs into the impossi- more basic question of what this distinction is for, what
bility of thinking coherently about God. The trouble significant point is made by dividing up human actions, or
with religious morality comes not from morality's policies, or motives, or reasons along these lines. Some of
being inescapably pure, but from religion's being in- the remarks in the previous section touch on that point.
curably unintelligible. I shall assume as given - indeed I have already as-
sumed it earlier - a conclusion which Mr Warnock
reaches in his discussion and which must certainly be
correct, namely that any significant delimitation of the
moral must involve reference to the content of the
judgements, policies, principles, or whatever, that are
being described as 'moral'. It might perhaps surprise
people innocent of moral philosophy that anyone has
Macmillan, 1967.
73
 MORALITY

ever supposed this not to be so; but it has in fact been things as policies, institutions, dispositions, sorts of
maintained, and frequently, that moral views (as o p motive, etc., is to count as moral approval, then one
posed to non-moral ones) can be identified without must suppose that those policies, institutions, etc., min-
reference to their content, by some considerations such ister in some way to the achievement of some kind of
as their being practical maxims which are entirely uni- human well-being. The second point is that 'well-being'
versal or their being practical maxims which are ack- certainly cannot be interpreted, for these purposes, as
nowledged as ovemding other practical maxims. The meaning merely that the persons involved in these
motive for these improbable manoeuvres has been, as situations get what they actually want nor (what is not
always, the maintenance of the fact/value distinction. necessarily the same thing) that they enjoy the out-
For suppose one introduces a reference to the content come. For it must surely be possible to recognize as
of the moral and says (for instance) that moral views moral views (though utilitarians will be disposed to
essentially refer to human well-being, where 'well- think that they are mistaken moral views) outlooks
being' has itself some content and does not merely which hold that people very often want and enjoy the
mean 'whatever one thinks ought to happen to human wrong things.
beings' : then the range of possible moral views is -
Nor - though this is more contestable is it clear
seriously limited by facts and by logic, contrary to the that 'well-being' in this connection can be identified
fact/value distinction. This motive for the manoeuvres with happiness. Certainly if we think (as the English
hardly makes them any more attractive than do their language, as it now is, rather encourages us to think)
consequences. that contentment is a sufficient, though not a necessary,
The question I want to discuss here is not, then, this condition of happiness, it will not do for the present
general point, but the merits of the partial criterion I purpose; we must be able to recognize as moral views
just mentioned, of a reference to human well-being as a (though cynics will think them mistaken) outlooks
mark of a moral position; advanced by Mrs P. R. Foot
and others, this proposal is itself well regarded by Mr
1 which deplore contentment, if secured at too low a
level of consciousness and activity. Even if we break
Warnock. If this is to be defended, it is essential both this connection and refuse to count as 'happy' those
that the test be applied at a very general level and that who are content in some bovine, doped, or subdued
'well-being' be interpreted in a very general way - state - as Aristotle and, with noble inconsistency, J. S-
though not, of course, in such a general way that it be- Mill, refused to count them - and insist on a more a d -
comes vacuous. vist or i-eflective level of happiness, it is still not clear
The first point comes out in this, that certainly one that we have what the thesis requires. The idea of a
could not apply such a test to the motivation of par- man's being happy, and indeed the less episodic notion
ticular actions, and always get the right results : it must of his being a happy man, has surely something to do
rather be that if one's approval of such more general with his not suffering; or his not suffering too much;
 MORALITY

or at the limit, not suffering in the way that matters Luther's outlook, seems to me essentially to misrepre
most, as we might just about say of a man with a physi- sent it: still more, perhaps, some other Protestant out-
cally painful condition that he was nevertheless happy, looks. The point is that there is no means open to man
if he resisted despair, self-pity, and so forth, and re- towards reconciliation with God, no set of human pro-
tained interest and pleasure in other things - and he jects conceivably adequate to secure this result - the
would have to be at least some of the time moderately gap is too great, and there is merely one sign of hope,
cheerful. The views of certain philosophers of an- Jesus Christ, that God's grace will lift up the undeserv-
tiquity, that virtue was sufficient for happiness and ing. The devout man will obey the will of God, as best
that the good man could be happy on the rack, have been he can in his forlorn condition, and must retain his
rightly thought before, after, and no doubt during their consciousness of that condition, but not in order to
time to involve some paradox. But if happiness is ulti- secure for himself or anyone else salvation, which is
mately incompatible with too much, or too total, suf- at best a wild hope; and if he is rejected he can have no
fering, there can perhaps be recognizably moral out- complaint.
looks which reject the notion that happiness is the I think that this kind of Protestant outlook could be
concern of our arrangements. It is reported that Luther, called a moral outlook - certainly it purports to ex-
when someone proposed Glucklichkeit (happiness) as plain man's situation in relation to what conduct is
the end of human life, violently rejected the idea and expected of him, and it indeed speaks of what, on its
said 'leiden, leiden, Kreuz, Kreuz' ('suffering ... the view, is central to man's happiness. But that happiness
..
Cross .').Here there could be a view that man's sin is seen as so far off and man as so alienated from the
and distance from God was such that only a life of peni- source of it, that it would seem a distortion to represent
tence and consciousness of one's own and everyone's such a morality as aiming at man's happiness : the aim
evil could appropriately respond to the situation. Such is rather that life should mirror, in suffering and obed-
a view would no doubt deplore institutions, outlooks, ience, man's deplorable condition. This is surely a very
ways of life, which tended to eliminate these, the most different sort of outlook from the one that deals directly
basic of man's sufferings, and might only contingently with happiness, and yet it seems perverse not to call it
or peripherally approve those which lightened less a morality. One might perhaps say that moralities with
spiritual sufferings. a transcendental dimension have a greater logical
To this it might be replied that it onIy shows that freedom with respect to their content than moralities
Luther placed man's well-being elsewhere, in eventual which have no such dimension: their transcendental
reconciliation with God, and that sufferings here were picture speaks of men's general condition and role in
only a means to happiness elsewhere. So happiness is terms which may make partially intelligible as a moral
still the point, though happiness elsewhere. But this for- outlook attitudes which would be utterly opaque if
mulation, to the limited extent that I understand offered in a purely secular framework.
 MORALITY

Even in a secular framework, however, it may be pos- an appeal to something there in human life which has
sible to find moral outlooks for which equally 'happi- to be discovered, trusted, followed, possibly in grave
ness' seems a poor characterization of their central ignorance of the outcome.
concerns. Thus certain Romantic outlooks which speak As the last phrase indicates, I am not speaking here
in terms of a free response to life; or of 'honesty' to of such a thing as Marxist morality, which is not our
one's impulses, including destructive ones; or of the sig- present concern since it is fairly straightforwardly con-
nificance of extreme experiences - may, any of them, cerned with ultimate happiness : the well-being of men
be ineptly described by saying that they have a special which is envisaged after the destruction of capitalism
view of men's happiness. It may be that certain of these and the elimination of exploitation will not perhaps
outlooks say less about the general framework of mor- merely consist in their being happier, but it will cer-
ality than they do about certain personal ideals. These tainly include that. What I rather have in mind is, for
indeed enter into morality in the sense that for those instance, something indicated by a phrase of D. H.
who respond to such an ideal, it provides a model of life Lawrence's in his splendid commentary on the compla-
to be lived through and to which a special kind of im- cent moral utterances of Benjamin Franklin: 'Find
portance is attached, but they are less concerned with your deepest impulse, and follow that! The notion that
what rules, institutions, dispositions, etc., are required there is something that is one's deepest impulse, that
in society as a whole. But this raises large issues, since there is a discovery to be made here, rather than a de-
the relation between personal ideals and general social cision; and the notion that one trusts what is so dis-
norms is itself an important moral issue. covered, although unclear where it will lead - these,
It would be stupid to try to discuss these last issues rather, are the point. The combination - discovery,
in very general terms: the outlooks in question need trust, and risk - are central to this sort of outlook, as of
to be set out and understood in some depth, and that is course they are to the state of being in love. It is even
not something that can be attempted here. But it seems tempting to find, among the many historical legacies of
an open question whether some such outlooks may not Protestantism to Romanticism, a parallel between this
genuinely cut the link with happiness as the focus of combination and the pair so important to Luther: o b
human moral activity. A central question to be asked edience and hope. Both make an essential connection
in considering this will always be, I think, to what ex- between submission and uncertainty; both, rather than
tent the moral outlook makes, perhaps tacitly or vesti- offering happiness, demand authenticity.
gially, a transcendental appeal of some kind. And even Perhaps the outlook I have gestured towards could
where there is no transcendental appeal in the sense of not possibly constitute a complete morality, because it
a reference, such as the religious moralist makes, to has nothing, or not enough, to say about society, and
something outside human life which provides in some hence not enough to say about even one man's life as a
way a pattern for that life; nevertheless, there may be whole. Perhaps even so far as it goes it rests on an illu-
78
 MORALITY WHAT I S MORALITY ABOUT?

sion. But the very fact that it exists and has power de- left. On the other hand, in characterizing these out-
mands some response from anyone who thinks it evi- looks, one speaks of what men in fact find value in, or
dent that general happiness must be the focus of need, or want; and if someone said - obscurely enough -
morality; as does the religious morality, in so far as its that men need a world in which there is risk, uncertainty,
outlook (as in our extreme Protestant example) radi- and the possibility of despair, then a morality which
cally differs from a purely secular outlook. For granted emphasized this, as opposed to moralities which want
that its transcendental claim is false, human beings as much as possible tidied up, might still be said to be
must have dreamed it, and we need an understanding concerned with men's well-being. Something will still
of why this was the content of their dream. (Humanism be excluded by the use of this term: systems of values
- in the contemporary sense of a secularist and anti- or precepts which paid no attention at all to what we
religious movement - seems seldom to have faced fully can understand men as needing or wanting.
a very immediate consequence of its own views: that
this tenible thing, religion, is a human creation.) Men
do, as a matter of fact, find value in such things as s u b
mission, trust, uncertainty, risk, even despair and suf-
fering, and these values can scarcely all be related to a
central ideal of happiness. And if we find some explan-
ations, psychoanalytical, perhaps, or even in some cases
zoological, of such attitudes, and also come to regard
them as aberrancies which we seek to reduce, then cer-
tainly we are changing the world from the standpoint
of a certain morality, not merely making the world
more responsive to what morality unquestionably is.
'Well-being' was the point from which we started;
we have been concerned recently with taking that as
'happiness'. Perhaps it might be said that even if some
sorts of moral ideas reject happiness as the central
notion, there is still a wider, yet contentful, notion of
well-being in which they do not reject that. It is a real
question, and I do not know the answer. On the one
hand, the most extreme cases seem to leave us with a
notion of well-being which is really at no great distance
from 'being as men ought to be'. where no content is
80
 UTILITARIANISM

bad states of affairs. This very broad sense - which is

UTILITARIANISM probably better represented by the word 'consequen-
tialism' than by 'utilitarianism' - is not my concern
here; we are interested only in views of this kind which
do take happiness as the one intrinsically good thing,
at which actions and social arrangements are suppos-
IN discussing the question whether any moral outlook edly aimed. But that restriction still leaves a lot of room
must ultimately be concerned with human happiness, I for different sorts of utilitarianism.
have not supposed that question to be the same as the To discuss in a vacuum what might or might not
question whether all moral outlooks must be one or count as a recognizable form of utilitarianism would
another version of utilitarianism. Obviously they are be a purely verbal and pointless exercise. The question
not the same question if we take the narrowest sense of can only be approached by asking what the point of
'utilitarianism', which holds that there is just one moral the utilitarian outlook on morality is; and that can be
principle, to seek the greatest happiness of the greatest discovered not merely, nor principally, by consulting
number; that 'happiness' here means pleasure and the what Bentham and J. S. Mill and other classical expo-
absence of pain; and that the one moral principle - nents of the system had in mind, but by considering
since it is the one moral principle - is to be applied to what the attractions of the utilitarian outlook are for
each individual situation ('act-utilitarianism'). Obviously moral thought. I think that there are four major ones:
there are all sorts of ways in which a morality can be this is not to deny that these are, in ways worth ex-
ultimately concerned with human happiness without ploring, related to one another. First, it is non-transcen-
being identical with that. But I think also that there are dental, and makes no appeal outside human life, in
ways in which morality can be ultimately concerned particular not to religious considerations. It thus helps,
with human happiness without being identical with in particular, with the entirely reasonable demand that
utilitarianism even taken in a more extended sense. morality now should be obviously free from Christian-
A difficulty in discussing this issue is a lack of agree- ity. It can even seem to help - because of a certain con-
ment about how extensively the term 'utilitarianism' servatism which I shall consider later - with a demand
may properly or sensibly be used. The term has some- far less reasonable, indeed rightly perceived by Nietzsche
times been used to include moral outlooks which do to be idiotic, that the morality thus freed from Christian-
not have anything specially to do with happiness or ity should be very much the same as the one previously
pleasure at all; in this sense, it is used to refer to any attached to Christianity. In more radical hands, how-
outlook which holds that the rightness or wrongness of ever, utilitarianism promises more radical change.
an action always depends on the consequences of the Second, its basic good, happiness, seems minimally
action, on its tendency to lead to intrinsically good or problematical : however much people differ, surely
82 83
 MORALITY UTILITARIANISM

they at least all want to be happy, and aiming at as much quences. Moral thought becomes empirical, and on
happiness as possible must surely, whatever else gives questions of public policy, a matter of social science.
way, be a reasonable aim. Now there is a notorious This has always been found by many one of the most
problem at this point about the transition from a sup- gratifying features of utilitarianism. It is not that the
posedly indisputable aim of seeking one's own happi- calculations are thought to be easy, or even practically
ness, to a more disputable aim of seeking other people's possible in many cases; the charm lies rather in this,
happiness, and the unfortunate Mill has been repeat- that the nature of the difficulty is at least quite un-
edly beaten over the head by critics for (it is said) trying mysterious. All moral obscurity becomes a matter of
to make this transition by deductive argument. I doubt technical limitations.
whether that was what he was trying to do, but in any Fourth, utilitarianism provides a common currency
case the problem is of no special force against utili- of moral thought: the different concerns of different
tarianism - there is no reason why it, any more than parties, and the different sorts of claims acting on one
anyone else, should possess a magic formula for argu- party, can all be cashed (in principle) in terms of happi-
ing the moralist out of his amoralism. The point is ness. This provision, importantly, has the consequence
rather that utilitarianism is a minimum commitment that a certain kind of conflict, well-known to some
morality, in this as in other respects: given merely other moral outlooks, is impossible - the conflict, that
the minimum requirements for being in the moral is to say, of two claims which are both valid and ir-
world, a willingness to consider other people's wants reconcilable. Under some other systems, a man may
as well as one's own, utilitarianism can get going on this come to be in a situation in which (as it seems to him)
spot A much more interesting question is whether the whatever he does involves doing something wrong. For
'indisputable' aim of happiness can in fact be made utilitarianism, this is impossible. The various claims he
to serve utilitarian purposes. We have already seen may feel on him can be brought to the common mea-
some reason, in the previous section, for doubting sure of the Greatest Happiness Principle, and there can
whether happiness must be seen as the aim of human be no coherent idea of a right or wrong thing to do,
life at all; but even waiving those questions, it is far other than what is, or is not, the best thing to do on the
from clear that any sense in which it is (more or less) whole: and if two courses come out equal, then it
indisputably such an end, is also a sense in which utili- really cannot matter which he does. As against this,
tarianism can be made to work on it. This is a central many people can recognize the thought that a certain
issue: we shall be in a better position to consider it course of action is, indeed, the best thing to do on the
when we have looked at the third and fourth attractions whole in the circumstances, but that doing it involves
of utilitarianism. doing something wrong. This is a thought which for
Its third attraction is that moral issues can, in prin- utilitarianism must, I think, ultimately be incoherent.
ciple, be determined by empirical calculation of conse- This is one reason for saying (what is certainly true)
84 85
 MORALITY UTILITARIANISM
that for utilitarianism, tragedy is impossible; but it them together into some kind of General Happiness,
has wider, if not deeper, consequences than that. can we make the thing work. At a technical level these
The utilitarian may be able to move back a little to- problems have been the concern of such subjects as wel-
wards this type of thought, by invoking such things as fare economics and preference theory, which have
the desirable social consequences of people being a bit laboured within very artificial assumptions and with
squeamish about certain actions, even when those are, only moderate success to deal with them for economic
in the circumstances, the best available : we shall come theory. Here we are concerned with more general diffi-
back to that type of argument later on. But what he is culties. If the 'happiness* involved is to be such as to
bound to do as a utilitarian is to regard as an indisput- allow utilitarianism to deliver on its third and fourth
able general aim of moral thought, the reduction of promises, can it also be the indisputable aim which was
conflict, the elimination wherever possible of value promised in the second ?
conflicts without remainder. Here, as elsewhere, he The answer to that seems to be just 'no*. Bentham
is concerned with efficiency: the generation of con- offered an account of happiness, namely as pleasure and
flicts is a sign of inefficiency in a value system, and the absence of pain, which b-~assupposed very clearly
utilitarianism has a general device for eliminating or to deliver on all the promises at once; but even if it had
solving them. But some might wonder whether such satisfied (as of course it did not) the conditions of
efficiency was an indisputable aim. One can certainly being calculable, comparable, and additive, it failed the
reduce conflict, and make life simpler, by cutting down condition of being an indisputable objective: the more
the range of claims one is prepared to consider; but in it looked like the sort of pleasure that could conceiv-
certain cases, that might seem not so much a triumph ably be dealt with in those quasi-arithmetical terms,
for rationality, as a cowardly evasion, a refusal to see the less it looked like something that any rational man
what is there to be seen (we may ask here, once more, must evidently be aiming at - as Mill came, if uneasily,
whether defused subjectivism really leaves everything to see. If, on the other hand, the conception of happi-
where it was). ness is made generous enough to include anything that
So even the attractiveness of utilitarianism's fourth might reasonably be aimed at as a satisfying life or in-
attraction may be importantly disputed. Other difficul- gredient of a life - then it less and less looks like some-
ties crowd in when one considers what it presupposes. thing which could fit in with the third and fourth con-
For we are going to be able to use the Greatest Happi- ditions. Apart from anything else, there is the difficulty
ness Principle as the common measure of all and every- that many things which people actually include in the
body's claims, only if the 'happiness* involved is in content of a happy life are things which essentially in-
some sense comparable and in some sense additive. volve other values, such as integrity, for instance, or
Only if we can compare the happiness involved for dif- spontaneity, or freedom, or love, or artistic selfexpres-
ferent people and over different outcomes, and also put sion; and not only can they not be treated in the way
86
 MORALITY UTILITARIANISM

that the third and fourth conditions require of utilitar- with the dilemma, of either refusing to quantify the
ianism's 'happiness', but there seems, in the case of value in question, in which case it disappears from
some of them at least, an actual contradiction in think- the sum altogether, or else of trying to attach some
ing of them as something that could be so treated. quantity to it, in which case they misrepresent what
This is the first general difficulty, then, with utili- they are about and also usually lose the argument, since
tarianism. Its 'happiness' has to satisfy certain con- the quantified value is not enough to tip the scale. In
ditions, if the point of utilitarianism is to be retained; such matters, it is not that utilitarians are committed
and the condition, that it should be indisputably an to thinking that these other values do not matter; nor
aim of human aspiration, conflicts with the other con- are they confined to thinking valuable those things
ditions which it must satisfy if it is to be treated as which can presently be handled by cost-benefit analy-
utilitarianism requires it to be treated. Faced with this sis. They are perhaps not even bound to think that every
general difficulty, one way in which utilitarianism social value should eventually be handleable by some-
tends to react is to dispute the values involved in the thing like cost-benefit analysis: they might say that
more intractable conceptions of happiness, as irrational, they were not committed to the view that the common
perhaps, or as hangovers of a past age. Such arguments currency of happiness is money. But they are commit-
may involve some interesting points on the way, but ted to something which in practice has those implica-
their strategy is shamelessly circular : utilitarian ration- tions: that there are no ultimately incommensurable
ality is made the test of what counts as happiness, in values. Nor is it an accidental feature of the utilitarian
order to remove that sort of happiness which consti- outlook that the presumption is in favour of the mone-
tutes an objection to utilitarianism. All that is needed tarily quantifiable, and the other values are forced into
to counter this at the theoretical level is a suitable un- the apologetic dilemma we have just met. It is not an
willingness to be bullied. accident, because (for one thing) utilitarianism is un-
The problem, however, is not confined to the thee surprisingly the value system for a society in which
retical level : it occurs drastically at the social level, and economic values are supreme; and also, at the theoreti-
an unwillingness to be bullied may here be inadequate, cal level, because quantification in money is the only
or hard to enforce. In cases of planning, conservation, obvious form of what utilitarianism insists upon, the
welfare, and social decisions of all kinds, a set of values commensurability of value.
which are, at least notionally, quantified in terms of There is great pressure for research into techniques
resources, are confronted by values which are not quan- to make larger ranges of social value commensurable.
tifiable in terms of resources: such as the value of pre- Some of the effort should rather be devoted to learning
serving an ancient part of a town, or of contriving - or learning again, perhaps - how to think intelligently
dignity as well as comfort for patients in a geriatric about conflicts of values which are incommensurable.
unit. Again and again defenders of such values are faced These have been difficulties in satisfying utilitarian-
88 89
 MORALITY UTILITARIANISM

ism's conditions for 'happiness'. The fact that there are to certain rules which they apply usually without cal-
these difficulties does not mean, of course, that we can culation to particular cases; it is the adoption of these
never in any particular case or sort of case arrive at an rules which is assessed by appeal to the Greatest Happi-
idea of what the utilitarian solution would be, or of ness Principle, and not the choice of particular actions.
what sorts of things the utilitarian would count. Some The same idea is invoked to explain the other fact
thing would be wrong if this did follow, since clearly which otherwise presents a difficulty, that we can
we can sometimes do these things: in discussing utili- easily construct cases - for instance where the convic-
tarianism, we are discussing something, and something tion of an innocent man is necessary and sufficient to
quite often recognizable. So let us grant that in some avoid great harms - in which the utilitarian result seems
cases, at least, we do know what is meant by working to conflict with what many would regard as the morally
out what course of action would lead to the greatest right answer: as in this case with justice, so promise-
happiness all round. Granted this, we now meet two keeping and truth-telling are found to present difficul-
new difficulties. One is that the process of working out ties under the act-utilitarian interpretation. The
such consequences is itself an activity, which itself in rule-utilitarian can, it is hoped, dissolve these difficul-
various circumstances possesses various degrees of ties by claiming that all that has to be shown is that
utility, and this has to go into the sum. The other is the rules or practices of justice, promise-keeping or
that the answer reached by utilitarian calculation of truth-telling possess positive utility over the alterna-
the particular case seems in certain cases to be morally tives.
the wrong answer. There exists a kind of utilitarianism, This is only a sketch. Many importantly different
called 'ruleutilitarianism', whose aim is precisely to things can be covered by the term 'ruleutilitarianism',
solve both these difficulties at once, by one and the same and different things need to be said about them.' All I
device. will try to do here is to suggest one or two points about
The first problem is that any actual utilitarian cal- how far the utilitarian can consistently go in the rule
culation will take place under conditions of consider- direction; and argue that either he cannot go far enough
able uncertainty and very partial information, so that to solve the second difficulty, or else he has to go so far
its results are likely to be unreliable. Moreover, the that he (and everyone else) ceases to be a utilitarian.
business of calculation itself takes time; and the dis- It is certainly possible for a utilitarian, without in-
position to calculate in each case has psychological consistency, to adopt a general practice for dealing with
features which may as a matter of fact impede things a certain kind of case, even though some particular
which are utilitarianly desirable, such as resolute ac- applications of the practice produce a result different
tion. These things being so, it is suggested that better *For a detailed and subtle discussion of the issues involved,
consequences may follow from the practice, not of see David Lyons's important book, The Forms and Limits of
agents calculating each action, but of their subscribing Utilitarianism (Oxford : Clarendon Press, 1969).
90
 MORALITY UTILITARIANISM

from what would have been reached by individual cal- seems to be getting further and further away from the
culation of those instances. The paradigm of this is the original utilitarian advantages. For, first, the supposed
accounting system of many public utilities, who may calculation of the utility of people's thinking in terms
occasionally send out a bill for some very small sum, of imagined consequences - as against their working
even though each bill costs more than that to process : out particular cases; or working out the consequences
the point being, that it is actually cheaper to send out of more specific rules; or taking the local morality as a
all bills when due, however small the amount, rather going concern; or many other possibilities - this calcu-
than to interrupt the processes to extract a few bills. lation begins to look more and more like bluff. How
Let us call this the 'gas bill model'. does he know what the consequences of these various
Now the gas bill model deals in actual consequences : practices might be? Second, he does know one thing,
the actual consequences of merely applying a rule, on at least: that the more general the provision to which
the one hand, and of making a particular interference, the utilitarian calculation is attached, the more cases
on the other. This sort of model, at least, cannot render there will be in which particular calculation in that
palatable to a consistent utilitarian a form of argument case would have produced a different result, so the more
which invokes neither the actual consequences of a par- tactical disutility he is licensing in his pursuit of strat-
ticular choice, nor the actual consequences of the egic utility. In view of the doubtfulness of the strategic
general following of a rule, but the hypothetical conse- utility, this ought to worry him: one of the motiva-
quences of an imagined following of a rule. Thus the tions of utilitarianism, after all, was a hard-headed
familiar pattern of moral argument, 'how would it be injunction to think in terms of calculable consequences,
if everyone did that?' cannot have any effect on a con- and not just to rely on tradition, received practice and
sistent utilitarian unless his action really will have the so forth.
effect of making everyone do it, which is usually pretty The more one considers rule-utilitarianism, the more
implausible. A-purely imaginary consequence can no pressing this sort of point becomes. Turning once more
more figure in a utilitarian calculation, than the happi- to the gas bill model, we can recall that what princi-
ness or unhappiness of purely imaginary persons can. pally made the uniform practice sensible was the cost
So the gas bill model, at least, cannot in itself get us on of interfering with it. The analogy to this in ordinary
the road to that kind of generalization argument. moral deliberation is the disutility of calculating par-
If the utilitarian wants to justify the use of the gen- ticular consequences. But the effect of that argument
eralization argument which deals in imagined conse- is cancelled out if we consider a case in which the par-
quences, he will have to take a further step away from ticular calculation has already been made: and this is
the actual consequences of particular choices, and deal so in the morally disquieting cases which presented the
in t e r n of the actual consequences of people's think- second kind of difficulty rule-utilitarianism was sup-
ing in terms of imagined consequences. But now he posed to deal with. If calculation has already been
92
 MORALITY UTILITARIANISM

made, and the consequences of breaking the rule are utilitarianism, and not, as they surely should be, doubts
found better than those of keeping it; then certainly no about whether public executions might not be reintro.
considerations about the disutility of calculation could duced. This is an absurd case. But more generally all the
upset that result And, indeed, it is very difficult to see many human qualities which are valued and yet resist
how anything, for a consistent utilitarian, could upset utilitarian treatment, such as an unaccommodating
that result. Whatever the general utility of having a passion for justice; certain sorts of courage; spon-
certain rule, if one has actually reached the point of taneity; a disposition to resist such things as useful ex-
seeing that the utility of breaking it on a certain occa- periments on senile patients or the use of napalm on
sion is greater than that of follo&ng it, then surely it some people to secure (as it is supposed) the happiness
would be pure irrationality not to break it? of more people, often elicit from utilitarian theorists
This consequence has indeed been drawn by some attempts to accommodate utilitarianism to those values
tough utilitarians, such as J. J. C. Smart. If utilitarian- rather than condemnation of such values as irrational
ism can be got going at all, then I am sure theirs must legacies of a pre-utilitarian era. This is no doubt a trib
be the right attitude to it: it is a special doctrine, not Ute to the decency and imagination of those utilitarians
necessarily coincident with contemporary Western but not to their consistencyor their u t i l i t a r i a
moral ideas in all respects, and one must expect it to Rule-utilitarianism, as the enterprise of trying to hold
have what may well seem unpalatable conclusions. on to something distinctively utilitarian, while knock-
Contrary to this, one feature of much modem utilitar- ing the rougher edges off it, seems to me a failure. This
ian theory is that it is surprisingly conformist. Bentham middle ground is not logically habitable. As opposed to
and Mill regarded the Greatest Happiness Principle as this, one might, on the one hand, take the line of Smart
an instrument of criticism, and thought that by appeal and others, and pursue act-utilitarianism modified only
to it {theycould show that many Victorian moral b e within the recognizable limits of the gas bill model.
liefs were mistaken and irrational, as indeed they were. This is at least consistent with the misguided aims of
But, except for the wellestablished areas of sexual and utilitarianism, and the fact that it yields some distinc-
penal reform, themselves inherited from Bentham and tive and (to many) unpalatable particular moral results
Mill, modem utilitarian theorists tend to spend more should be a matter of no surprise. If, on the other hand,
effort in reconciling utilitirianism with existing moral you desert this territory and start to apply the utili-
beliefs than in rejecting those beliefs on the strength of tarian principle to more and more general practices and
utilitarianism. One recent writer, for instance, has habits of thought, what you come out with is unlikely
taken great and honest pains to show that public execu- to have any distinctively utilitarian content at all.
tions could not, as might seem, be justified on utilitarian This capacity of utilitarianism, once detached from
grounds. He is left with some frank doubts; but these the ground level, to annihilate itself, can be illustrated
are doubts about the application and formulation of by a brief argument, with which I shall end. Its empiri-
94 95
 MORALITY UTILITARIANISM

cal premises are not, perhaps, beyond doubt. But they would have been if there had been no villains - but
are certainly at least as plausible as most of those gen- that, certainly, is Utopian. What looks more hope-
erally used by utilitarians in such matters. ful is a state of affairs in which enough people are
One disturbing effect of people being active and con- resistant to continuing the rot: resistant, for instance,
scious utilitarians is that it tends to debase the moral by there being a range of things that they cannot con-
currency : a Gresham's Law operates, by which the bad sider doing, or bring themselves to do, or put up with
acts of bad men elicit from better men acts which, in being done, whatever other people do or may do. There
better circumstances, would also be bad. There is a is a limit to their pre-emptive activities. Enough people,
simple reason for this: a utilitarian must always be enough of the time, it seems, have to be prepared to stick
justified in doing the least bad thing which is necessary at doing various things, whatever the consequences may
to prevent the worst thing that would otherwise happen be. That means that enough people, enough of the time,
in the circumstances (including, of course, the worst do not have to think as utilitarians; they have, quite
thing that someone else may do) - and what he is thus definitely, to think as non-utilitarians. Nor will it do for
justified in doing may often be something which, taken them to preserve at the back of their mind the utili-
in itself, is fairly nasty. The preemptive act is built in tarian rationale in coexistence with the required moral
to utilitarian conceptions, and certain notions of nega- bloody-mindedness. For they have to be able to resist
tive responsibility (that you are as responsible for what utilitarian temptation in the most difficult circum-
you fail to prevent, as much as for what you do) are by stances, when much obvious harm will follow from re-
the same token characteristic of it. This being so, it is sisting it, and for that their non-utilitarianism has to
empirically probable that an escalation of pre-emptive be very deeply engrained.
activity may be expected; and the total consequences of Some utilitarians have reached, if not quite for these
this, by utilitarian standards themselves, will be worse reasons, something rather like this conclusion, and
than if it had never started. have thought that what it showed was that the truth of
The utilitarian who is immersed in the system, how- utilitarianism could be known to a responsible Clite,
ever, cannot do anything about this; he must think in but should not be too widely spread among the masses.
terms of actual consequences, and nothing in the realm Such a proposal is both personally and socially hope-
of actual consequences (at least, nothing helpful) will less. Personally, since the state of mind ascribed to the
now be effected by some gesture of principle - there is reflective utilitarian, and the attitude to others it in-
no way in which, from where he is, he can lead a dash volves could be honestly held, if at all, only by a very
to morally higher ground. Stepping back in reflection, innocent man (as perhaps Sidgwick was), and no reflec-
however, he can consider how utilitarian aims might tive man in our age can be that innocent. Socially, be-
have been better realized than they have been in a world cause the educational and other institutions required
of utilitarians interspersed with villains. No doubt they to embody such a view would have to be quite different
96
 MORALITY

from anything we could now expect or tolerate, or that

utilitarianism itself could want.
If all this is true, then the world which the reflective
utilitarian must finally settle on as most likely to yield
the outcomes he wants, is a world in which the Gres-
ham's Law is defeated because enough people enough of
the time are deeply disposed against thinking in a utili-
tarian fashion. It is not possible -that this disposition
should coexist with believing in utilitarianism; nor is
it acceptable or socially possible that most should have
this disposition while others, the utilitarian glite, should
believe in utilitarianism. All that is left is that the world
which would satisfy the utilitarian's aspirations would
be a world from which belief in utilitarianism as an
overall moral doctrine was totally absent, excf pt per-
haps as a minor and ineffectiveeccentricity.
So, if utilitarianism is true, and some fairly plausible
empirical propositions are also true, then it is better
that people should not believe in utilitarianism. If, on
the other hand, it is false, then it is certainly better that
people should not believe in it. So, either way, it is better
that people should not believe in it.