Mortal Questions

Mortal

Questions

THOMAS NAGEL

Canto is a paperback imprint which offers a broad range of titles, both classic and more recent, representing some of the best and most enjoyable of Cambridge publishing.

J\IIORTAL QUESTIONS Thomas Nagel

DCAMBRIDGE ~ UNIVERSITY PRESS

PUBLISHED BY THE PRESS SYNDICATE OF THE UNIVERSITY OF CAMBRIDGE

The Pitt Building, Trumpington Street, Cambridge, United Kingdom CAMBRIDGE UNIVERSITY PRESS

The Edinburgh Building, Cambridge CB2 2RU, UK 40West 20th Street, New York, NY 10011-4211, USA 477 Williamstown Road, Port Melbourne, VI C 3207, Australia Ruiz de Alarc6n 13,28014 Madrid, Spain Dock House, The Waterfront, Cape Town 8001, South Africa http:/ /www.cambridge.org

© Cambridge University Press 1979 This book is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 1979 Reprinted 1979, 1980,1981,1982,1983 (twice), 1985, 1987, 1988 (twice), 1990 Canto edition 1991 Reprinted 1992, 1993, 1996, 1997, 2000, 2002 Printed in the United Kingdom at the University Press, Cambridge

British Library Cataloguing in Publication data Nagel, Thomas Mortal questions. 1. Ethics I. Tide 170 Library

of Congress

Cataloguing in Publication data Nagel, Thomas Mortal questions. Includes bibliographical references and index. 1. Life-addresses, essays, lectures. 2. Ethics-Addresses, essays, lectures. I. Tide. BD431.N32 170 78-58797 ISBN 0 521 40676 5 paperback

Cover illustration: The Penitent Magdalen (detail). Georges de la Tour. New York, Metropolitan Museum of Art, gift of Mr and Mrs Charles Wrightsman, 1978.

to my father

WALTER NAGEL pessimist and skeptic

CONTENTS

Preface ix

1

2 3 4 5 6 7 8 9 10 11

Sources xv

Death The Absurd

11

Moral Luck

24 39 53 75 91 106 128 142

Sexual Perversion War and Massacre Ruthlessness in Public Life The Policy of Preference Equality The Fragmentation of Value Ethics without Biology Brain Bisection and the Unity of Consciousness

12 What is it like to be a bat? 13 Panpsychism 14 Subjective and Objective lt1dex

147 165 181 196 215

PREFACE

Philosophy covers an immense range of topics, but part of its concern has always been with mortal life: how to understand it and how to live it. These essays are about life: about its end, its meaning, its value, and about the metaphysics of consciousness. Some of the topics have not received much attention from analytic philosophers, because it is hard to be dear and precise about them, and hard to separate from a mixture of facts and feelings those questions abstract enough for philosophical treatment. Such problems must be attacked by a philosophical method that aims at personal as well\as theoretical understanding, and seeks to combine the two by incorporating theoretical results into the framework of self-knowledge. This involves risk. Large, relevant questions too easily evoke large, wet answers. Every theoretical field faces a contest between extravagance and repression, imagination and rigor, expansiveness and precision. Fleeing from the excesses of the one, it is easy to fall into the excesses of the Nher. Attachment to the grand style can produce an impatience with demands for rigor and may lead to a tolerance for the unintelligible. Since the defects of a tradition tend to reflect its virtues, the problem in ar;alytic philosophy has been the reverse. It is not exactly correct to say that AngloAmerican philosophy avoids the big questions. For one thing, there are no problems deeper or more important than those in metaphysics, epistemology, and philosophy of language that lie at the center of the field. For another, the analytic establishment has been quite hospitable to recent attempts to explore

X

Priface

unfamiliar territory. Nevertheless, the fear of nonsense has had a powerful inhibiting effect. Long after the demise of L�i partly on causal factors: do these dispositions result only when normal development has been prevented? Even the form in which this question has been posed is circular, because of the: word 'normal'. We appear to need an independent criterion for a distorting influence, and we do not have one. It may be possible to class sadism and masochism as perver­ sions because they fall short of interpersonal reciprocity. Sadism concentrates on the evocation of passive self-awareness in others, but the sadist's engagement is itself active and requires a retention of deliberate control which may impede awareness of himself as a bodily subj ect of passion in the required sense. De Sade claimed that the object of sexual desire was to evoke involuntary responses from one's partner, especially audible ones. The infliction of pain is no doubt the most efficient way to accomplish this, but it requires a certain abrogation of one's own exposed spontaneity. A masochist on the other hand imposes the same disability on his partner as the sadist imposes on himself The masochist cannot find a satisfactory embodiment as the object of another's sexual desire, but only as the object of his control. He is passive not in relation to his partner's passion but in relation to his nonpassive agency. In addition, the subjection to one's body characteristic of pain and physical restraint is of l very different kind from that of sexual excitement: pain causes people to contract rather than dissolve. These descriptions may not be generally accurate. But to the extent that they are, sadism and masochism would be disorders of the second stage of awareness - the awareness of oneself as an object of desire. Homosexuality cannot similarly be classed as a perversion on phenomenological grounds. Nothing rules out the full range oi interpersonal perceptions between persons of the same sex. The issue then depends on whether homosexuality is produced by distorting influences that block or displace a natural tendency to heterosexual development. And the influences must be more

Sexual perversion

51

distorting than those which lead t o a taste for large breasts o r fair hair or dark eyes. These also are contingencies of sexual prefer­ en ce in which people differ, without being perverted. The question is whether heterosexuality is the natural expres­ on si of male and female sexual dispositions that have not been distorted. It is an unclear question, and I do not know how to ap p roach it. There is much support for an aggressive-passive dis tin ction between male and female sexuality. In our culture the male's arousal tends to initiate the perceptual exchange, he usually makes the sexual approach, largely controls the course of the act, and of course penetrates whereas the woman receives. When two men or two women engage in intercourse they cannot both adhere to these sexual roles. But a good deal of deviation from them occurs in heterosexual intercourse. Women can be sexually aggressive and men passive, and temporary reversals of role are not uncommon in heterosexual exchanges of reasonable length. For these reasons it seems to be doubtful that homosexuality must be a perversion, though like heterosexual­ ity it has perverted forms. Let me close with some remarks about the relation of perversion to good, bad, and morality. The concept of perversion can hardly fail to be evaluative in some sense, for it appears to involve the notion of an ideal or at least adequate sexuality which the pervcrions in some way fail to achieve. So, if the

concept is viable, the judgment that a person or p ractice or desire is perverted will constitute a sexual evaluation, implying that better sex, or a better specimen of sex, is possible. This in itself is a very weak claim, since the evaluation might be in a dimension that is of little interest to us. (Though, if my account is correct, that will not be true. ) Whether it is a moral evaluation, however, is another question entirely - one whose answer would require more understanding of both morality and perversion than can be deployed here. Moral evaluation of acts and of persons is a rather special and very complicated matter, and by no means all our evaluations of persons and their activities are moral evaluations. We make judgments about people's beauty or health or intelligence which are evaluative without being moral. Assessments of their sexual­ ity may be similar in that respect.

52

Mortal questious

Furthermore, moral issues aside, it is not clear that unper­ verted sex is necessarily preferable to the perversions. It may be that sex whiCh receives the highest marks for perfection as sex i s less enjoyable than certain perversions; and if enjoyment is considered very important, that might· outweigh considerations of sexual perfection in determining rational preference. That raises the question of the relation between the evaluative content of judgments of perversion and the rather common geueral distinction between good and bad sex. The latter distinc­ tion is usually confined to sexual acts, and it would seem, within limits, to cut across the other: even someone who believed, for example, that homosexuality was a perversion could admit a distinction between better and worse homosexual sex, and might even allow that good homosexual sex could be better sex than not very good unperverted sex. If this is correct, it supports the position that, if judgments of perversion are viable at all, they represent only one aspect of the possible evaluation of sex, even qua sex . Moreover it is not the only important aspect: sexual deficiencies that evidently do not constitute perversions can be the object of great concern. Finally, even if perverted sex is to that extent not so good as it might be, bad sex is generally better than none at all. This should not be controversial: it seems to hold for other important matters, like food, music, literature, and society. In the end, one must choose from among the available alternatives, whether their availability depends on the environment or on one's own constitution. And the alternatives have to be fairly grim before it becomes rational to opt for nothing.

5 War and Massacre

From the apathetic reaction to atrocities committed in Vietnam by the United States and its allies, one may conclude that moral restrictions on the conduct of war command almost as little sympathy among the general public as they do among those. charged with the formation ofU.S. military policy. t Even when restrictions on the conduct of warfare are defended, it is usually on legal grounds alone: their moral basis is often poorly understood. I wish to argue that certain restrictions are neither arbitrary nor merely conventional, and that their validity does not depend simply on their usefulness. There is, in other words, a moral basis for the rules of war, even though the conventions now officially in force are far from giving it perfect expression. elaborate moral theory is required to account for what is wrong in cases like the Mylai massacre, since it did not serve, and was not intended to serve, any strategic purpose. Moreover, if the participation of the United States in the Indo-Chinese war . is entirely wrong to begin with, then that engagement is incapable of providing a j ustification for a u y measures taken in No

its pursuit - not only for the measures which are atrocities in

every war, however just its ai ms. But this war has revealed attitudes of a more general kind, which influenced the conduct of earlier wars as well. After it has ended, we shall still be faced with the problem of how warfare 1

This essay was completed in 1 97 1 . Direct U.S. military involvement in the Vietnam War lasted from 1 % 1 to 1 973. Hence the present tense.

54

Mortal questious

may be conducted, and the attitudes that have resulted in th e specific conduct of this war will not have disappeared. Moreover, similar problems can arise in wars or rebellion s fought for very different reasons, and against very different opponents. It is not easy to keep a firm grip on the idea of wha t is not permissible in warfare, because while some military actions are obvious atrocities, other cases are more difficult to assess, and the general principles underlying these judgments remain obscure. Such obscurity can lead to the abandonment of sound intuitions in favor of criteria whose rationale may be more obvious. If such a tendency is to be resisted, it will require a better understanding of the restrictions than we now have. I propose to discuss the most general moral problem raised by the conduct of warfare: the problem of means and ends. In one view, there are limits on what may be done even in the service of an end worth pursuing - and even when adherence to the restriction may be very costly. A person who acknowledges the force of such restrictions can find himself in acute moral dilemmas. He may believe, for example, that by torturing a prisoner he can obtain information necessary to prevent a disaster, or that by obliterating one village with bombs he can halt a campaign of terrorism. If he believes that the gains from a certain measure will clearly outweigh its costs, yet still supects that he ought not to adopt it, then he is in a dilemma produced by the conflict between two disparate categories of moral reason: categories that may be called utilitarian and absolutist. Utilitarianism gives primacy to a concern with what will happe11 . Absolutism gives primacy to a concern with what one is doing . The conflict between them arises because the alternatives we face are rarely just choices between total outcomes : they are also choices between aiternative pathways or measures to be taken. When one of the choices is to do terrible things to another person, the problem is altered fundamentally; it is no longer merely a question of which outcome would be worse. Few of us are completely immune to either of these types of moral intuition, though in some people, either naturally or for doctrinal reasons, one type will be dominant and the other suppressed or weak. But it is perfectly possible to feel the force of both types of reason very strongly; in that case the mora l dilemma in certain situations of crisis will be acute, and it may

War and massacre

55

a pp ear that every possible course of action or inaction is unac ceptable for one reason or another. II Alth ou gh it is this dilemma that I propose to explore, most of the discussion will be devoted to its absolutist component. The utilitarian component is straightforward by comparison, and has a natural appeal to anyone who is not a complete skeptic about ethics. Utilitarianism says that one should try, either individu­ ally or through institutions, to maximize good and minimize evil (the definition of these categories need not enter into the schematic formulation of the view), and that if faced with the possibility of preventing a great evil by producing a lesser, one should choose the lesser evil. There are certainly problems about the formulation of utilitarianism, and much has been written about it, but its intent is morally transparent. Nevertheless, despite the additions and refinements, it continues to leave large portions of ethics unaccounted for. I do not suggest that some form of absolutism can account for them all, only that an examination of absolutism will lead us to see the complexity, and perhaps the incoherence, of our moral ideas. Utilitarianism certainly justifies some restrictions on the con­ duct of warfare. There are strong utilitarian reasons for adhering to any limitation which seems natural to most people - particu­ larly if the limitation is widely accepted already. An exceptional measure which seems to be justified by its results in a particular conflict may create a precedent with disastrous long-term effects.2 It may even be argued that war involves violence on such a scale that it is never justified on utilitarian grounds - the consequences of refusing to go to war will never be as bad as the war itself would be, even if atrocities were not committed. Or in a more sophisticated vein it might be claimed that a uniform policy of never resorting to military force would do less harm in the long run, if followed consistently, than a policy of deciding each case on utilitarian grounds (even though on occasion particular applications of the pacifist policy might have worse results than a specific utilitarian decision). But I shall not 2 S t raightforward considerations of national interest often tend in the same

direction: the inadvisability of using nuclear weapons seems to be overdetermined in this way.

56

Mortal questions

consider these arguments, for my concern is wi th reasons of a different kind, which may remain when reas ons of utility and interest fail.3 In the final analysis, I believe that the dilem ma cannot always be resolved. While not every conflict between absolutism and utilitarianism creates an insoluble dilemma, and while it seems to me certainly right to adhere to absolutist restrictions unless the utilitarian considerations favoring violation are overpoweringly weighty and extremely certain - nevertheless, when that special condition is met, it may become impossible to adhere to an absolutist position. What I shall offer, therefore, is a somewhat qualified defense of absolutis m. I believe it underlies a valid and fundamental type of moral j udgment - which cannot be reduced to or overridden by o th e r princi p les. And while there may be other principles just as fundamental, it is particularly important not to lose confidence in our absolutist intuitions, for they are often the only barrier before the abyss of utilitarian apologetics for large-scale murder.

III One absolutist position that creates no problems of interpreta­ tion is pacifism: the view that one may not kill another person under any circu mstances, no matter what good would be achieved or evil averted thereby. The type of absolutist position that I am going to discuss is different. Pacifism draws the conflict with utilitarian considerations very starkly. But there arc: other views according to which violence may be u n dert a ken even on a large scale, in a clearly j ust cause, so long as certain absolute restrictions on the character and direction of ' that violence are observed. The line is drawn somewhat closer to the bone, but it exists. The philosopher who has done most to advance contempor­ ary philosophical discussion of such a view, and to explain it to those unfamiliar with its extensive treatment in Roman Catholic ,

3

These reasons, moreover, have special importance in that they are available even to one who denies the appropriateness of utilitarian considerations in international matters. He may acknowledge limitations on what may be done to the soldiers and civilians of other countries in pursuit ofhis nation's military objectives, while denying that one country should in general consider the interests of nationals of other countries in determining its policies.

War and massacre

57

rno ral theology, is G. E.M. Anscombe. In 1 958 Miss Anscombe pu bl is hed a pamphlet entitled Mr. Truman 's Degree, 4 on the occasion of the award by Oxford University of an honorary do cto rate to Harry Truman. The pamphlet explained why she had opposed the decision to award that degree, recounted the story of her unsuccessful opposition, and offered some reflec­ tions on the history of Truman's decision to drop atom bombs on Hiro shima and Nagasaki, and on the difference between · rnurd er and allowable killing in warfare. She pointed out that the policy o f d eliberatel y killing large numb ers o f civilians either as a rneans or as an end in itself did not originate with Truman, and was common practice among all parties during World War II for some time before Hiroshima. The Allied area bombings of German cities by conventional explosives included raids which killed more civilians than did the atomic attacks; the same is true of certain fire-bomb raids on Japan. The policy of attacking the civilian population in order to induce an enemy to surrender, or to damage his morale, seems to have been widely accepted in the civilized world, and seems to be accepted still, at least if the stakes are high enough. It gives evidence of a moral conviction that the deliberate killing of ' non combatants- women, children, old people-is permissible if enough can be gained by it. This follows fro m the more general position that any means can in principle be j ustified if it leads to a sufficiently worthy end. Such an attitude is evident not only in the more spectacular current weapons systems but also in the day-to-day conduct of the nonglobal war in Indo-China: the indiscriminate destructiveness of antipersonnel weapons, napalm, and aerial bombardment; cruelty to prisoners; massive relocation of civilians; destruction of crops; and so forth. An absolutist position opposes to this the view that certain acts 4

(Privately printed.) See also her essay 'War and Murder', in Nuclear Weapous aud Christia 11 Corrscience, ed. Walter Stein (London: The Merlin Press, 1 961). The present paper is much indebted to these two essays throughout. These and related subjects are extensively treated by Paul Ramsey in The just War (New York: Scribners, 1 968) . Among recent writings that bear on the moral problem are Jonathan Bennett, 'Whatever the Consequences', Arralysis, XXVI, no. 3 (1 966), 83-102; and Philippa Foot, 'The Problem of Abortion and the Doctrine of the Double Effect', Oxford Review, v (1 967), 5-1 5. Miss Anscombe's replies are ' A Note on Mr. Bennett', Aualysis, XXVI, no. 3 (1 966), 208 and ' Who is Wronged?', Oxford Review, v (1 967), 16-17.

Mortal questio11s

58 cannot

be

justified no ma tter what the conseq uences . .A mon g

those acts is murder - the deliberate killing of the harmless: civilians, prisoners of war, and medical personnel. In the present war such measures are sometimes said to be regrettable, but they are generally defended by reference to military necessity and the i mportance of the long-term consequ­ ences of success or failure in the war. I shall pass over the inadequacy of this consequentialist defense in its own terms . . (That is the dominant form of moral criticism of the war, for it is · part of what people mean when they ask, 'Is it worth it?') I a m concerned rather to account for the inappropriateness of offering any defense of that kind for such actions. Many people feel, without being able to say much more about it, that something has gone seriously wrong when certain measures are admitted into consideration in the first place. The fundamental mistake is made there, rather than at the point where the overall benefit of some monstrous measure is judged to outweigh its disadvantages, and it is adopted. An account of absolutism might help us to understand this. If it is not allowable to do certain things, such as killing unarmed prisoners or civilians, then no argument about what will happen if one does not do them can show that doing them would be all right. Absolutism does not, of course, require one to ignore the consequences of one's acts. It operates as a limitation on utilitarian reasoning, not as a substitute for it. An absolutist can be expected to try to maximize good and minimize evil, so lo n g as this does not require him to transgress an absolute prohibition like that against murder. But when such a conflict occurs, the prohibition takes complete precedence over any consideration of consequences. Some of the results of this view are clear enough. It requires us to forgo certain potentially useful military meas­ ures, such as the slaughter of hostages and prisoners or indis­ criminate attempts to reduce the enemy civilian population by starvation, epidemic infectious diseases like anthrax and bubonic plague, or mass incineration. It means that we cannot delibe rate on whether such measures are justified by the fact that they will avert still greater evils, for as intentional measures they cannot be justified in terms of any consequences whatever. Someone unfamiliar with the events of this century might i magine that utilitarian arguments, or arguments of national

War and massacre

59

in terest, would suffice to deter measures of this sort. But it has beco me evident that such considerations are insufficient to p reven t the adoption and employment of enormous anti popula­ tio n weapons once their use is consi dered a serious moral p os sibility. The same is true of the piecemeal wiping out of rural civilian populations in airborne antiguerrilla warfare. Once the do or is opened to calculations of utility and national interest, the us ual speculations about the future of freedom, peace, and econ o mic prosperity can be brought to bear to ease the consci­ ences of those responsible for a certain number of charred babies. For this reason alone it is important to decide what is wrong with the frame of mind which allows such arguments to begin. But it is also important to understand absolutism in the cases where it genuinely conflicts with utility. Despite its appeal, it is a paradoxical position, for it can require that one refrain from choosing the lesser of two evils when that is the only choice one has. And it is additionally paradoxical because, unlike pacifism, it permits one to do horrible things to people in some circums­ tances but not in others. IV Before going on to say what, if anything, lies behind the position, there remain a few relatively technical matters which are best discussed at this point. First, it is important to specify as clearly as possible the kind of thing to which absolutist prohibitions can apply. We must take seriously the proviso that they concern what we deliberately do to people. There could not, for example, without incoherence, be an absolute prohibition against brit�ging about the death of an innocent person. For one may find oneself in a situation in which, no matter what one does, some innocent people will die as a result. I do not mean just that there are cases in which someone will die no matter what one does, because one is not in a p osition to affect the outcome one way or the other. That, it is to be hoped, is one's relation to the deaths of most innocent people. I have in mind, rather,

a

case in whi ch so meone is bound

to die, but who it is will depend on what one does. Sometimes these situations have natural causes, as when too few resources (medicine, lifeboats) are available to rescue everyone threatened with a certain catastrophe. Sometimes the situati ons are man­ made, as when the only way to control a campaign of terrorism is to

60

Mortal questions

to employ terrorist ta c t i cs ag ai nst the community from which it has arisen. Whatever one does in cases such as these, some innocent peop l e will die a s a result. If the absolutist prohibitio n forbade doing what would result in the deaths of innocent people, it would have the consequence that in such cases nothing one could do would be morally permissible. This problem is avoided, however, because what absolutis m forbids is doing certain things to people, rather than bringing about certain results. Not everything that happens to others as a result of what one does is something that one has done to them. Catholic moral theology seeks to make this distinction precise in a doctrine known as the law of double effect, which asserts that there is a mor a lly relevant distinction between bringing about or p er mit tin g the death of an innocent person deliberately, either as an end in itself or as a means, and bringing it about or pe rmi t ting it a s a side effect of something else one does deliberately. I n the latter case, even if the outcome is foreseen, it is not mu rder, and does not fall under the absolute p rohibition, though of course it may still be wrong for other reasons (reasons of utility, for ex a mp le) Briefly, the principle s tates that one is sometimes permitted knowingly to bring about or p e r mit as a side-effect of one's actions something which it would be a bs ol utely imper­ missible to bring about or permit deliberately as an end or as a mea ns In application to war or revolution, the l a w of double effect permi ts a certain amount of civilian carnage as a side-effect of bombing munitions plants or attacking enemy soldiers . And even this is permi s si bl e only i f the cost is not too great to be j us tified by one's objectives. However, despite its importance and its usefulness in account­ i ng for certain plausible moral judgments, I do not belie ve that the law of double effect is a generally app l icable test for the consequences of an absolutist position. Its own app l i ca tion is not always clear, so that it introduces uncertainty where there need not be u n ce rta i nty In Indo-China, for exa m p le, there is a great deal of aerial bombardment, strafing, spraying of napalm, and employment of p ell et- or needle-spraying antipersonnel weapons a g a inst rural villages in which g u e r rill as are suspected to be hiding, or from which small-arms fire has been received. The majority of those killed and wounded in these aerial attacks are reported to be .

.

.

War and massacre

61

wo men and children, even when some combatants are caught as well. However, the government regards these civilian casualties as a regrettable side-effect of what is a legiti mate attack against an armed enemy. It might be thought easy to dismiss this as sophistry: if one bo mbs, burns, or strafes a village containing a hundred people, twenty of whom one believes to be guerrillas, so that by killing most of them one will be statistically likely to kill most of the guer rillas, then is not one's a ttack on the group of one hundred a means of destroying the guerrillas, pure and simple? If one makes no attempt to discriminate between guerrillas and civilians, as is imp ossible in an aerial attack on a s mall village, then one cannot regard as a mere side-effect the deaths of those in the group that on e would not have bothered to kill if more selective means had been available. The difficulty is that this argument depends on one particular description of the act, and the reply might be that the means used against the guerrillas is not: killing everybody in the village - but rather: obliteration bombing of the area in which the twenty guerri l l as are kn own to he located. If there are civilians in the area as well, they will' be killed as a side-effect of such action.s Because of casuistical problems like this, I prefer to stay with the original, unanalyzed distinction between what one does to people and what merely happens to them as a result of what one does. The law of double effect provides an approximation to that distinction in many cases, and perhaps it can be sharpened to the point where it does better than that. Certainly the original distinction itself needs clarification, particularly since some of the things we do to people involve things happening to them as a result of other things we do. In a case like the one discussed, however, it is clear that by bombing the village one slaughters and maims the civilians in it. Whereas by giving the only available medicine to one of two sufferers from a disease, one does not kill the other or deliberately allow him to die, even if he dies as a result. The second technical point is this. The absolutist focus on actions rather than outcomes does not merel y introduce a new,

outstanding item into the catalogue of evils. That is, it does not 5

This counter-argument was suggested by Rogers Albritton.

62

Mortal questiom

say that the worst thing in the world is the deliberate murder of an innocent person. For if that were all, then one could presu m­ ably justify one such murder on the ground that it would prevent several others, or ten thousand on the ground that they would prevent a hundred thousand more. That is a fa milia r argument. But if this is allowable, then there is no absolute prohibition against murder after all. Absolutism requires that we avoid murder at all costs, not that we prevmt it at all costs. It would also be possible to adopt a deontological position less stringent than absolutism, without falling into utilitarianism. There are two ways in which someone might acknowledge the moral relevance of the distinction between deliberate and nondeliberate killing, without being an absolutist. One would be to count murder as a specially bad item in the catalogue of evils, much worse than accidental death or nondeliberate killing. But the other would be to say that deliberately killing an innocent is i mpermissible unless it is the only way to prevent some very large evil (say the deaths of fifty innocent people). Call this the threshold at which the prohibition against murder is overridden. The position is not absolutist, obviously, but it is also not

equivalent to an assignment of utilitarian disvalue to murder equal to the disvalue of the threshold. This is easily seen. If a murder had the dis value of fifty accidental deaths, it would still be permissible on utilitarian grounds to commit a murder to prevent one other murder, plus some lesser evil like a broken arm. Worse still, we would be required on utilitarian grounds to prevent one murder even at the cost of forty-nine accidental deaths that we could otherwise have prevented. These are not in fact consequences of a deontological prohibition against murder with a threshold, because it does not say that the occurrence of a certain kind of act is a bad thing, and therefore to be prevented, but rather tells everyone to refrain from such acts, except under certain conditions. In fact, it is perfectly compatible with a deontological prohibition against murder to hold that, consi­ dered as an outcome, a murder has no more disvalue than an accidental death. While the admission of thresholds would reduce the starkness of the conflicts discussed here, I do not think it would make them disappear, or change their basic character. They would persist in the clash between any deon­ tological requirement and utilitarian values somewhat lower

War aud massacrt

63

than its threshold. fina lly, let me remark on a frequent criticism of absolutism tha t depends on a misunderstanding. It is sometimes suggested th at such prohibitions depend on a kind of moral self-interest, a p rim ary obligation to preserve one's own moral purity, to keep one's hands clean no matter what happens to the rest of the world. If this were the position, it might be exposed to the charge of self-indulgence. After all, what gives one man a right to put the purity of his soul or the cleanness of his hands above th e lives or welfare of large numbers of other people? It might be argued that a public servant like Truman has no right to put himself first in that way; therefore if he is convinced that the alternatives would be worse, he must give the order to drop the bombs, and take the burden of those deaths on himself, as he must do other distasteful things for the general good. But there are two confusions behind the view that moral se lf-interest underlies moral absolutism. First, it is a confusion to suggest that the need to preserve one's moral purity might be the source of an obligation. For if by committing murder one sacrifices one's moral purity or integrity, that can only be because there is already something wrong with murder. The general reason against committing murder cannot therefore be merely that it makes one an immoral person. Secondly, the notion that one might sacrifice one's moral integrity justifiably, in the service of a sufficiently worthy end, is an incoherent notion. For if one were justified in making such a sacri fice (or even morally required to make it), then one would not be sacrificing one's moral integrity by adopting that course: one would be preserving it. Moral absolutism is not unique among moral theories in requiring each person to do what will preserve his own moral purity in all circumstances. This is equally true of utilitarianism, or of any other theory which distinguishes between right and wrong. Any theory which defines the right course of action in various circumstances and asserts that one should adopt that course, ipso fa ct o asserts that one should do what will preserve one's moral purity, simply because the right course of action is what will preserve one's moral purity in those circumstances. Of course utilitarianism does not assert that this is why one should adopt that course, but we have seen that the same is true of absolutism.

64

Mortal questiotls v

It is easier to dispose of false explanations of absolutis m than to produce a true one. A positive account of the matter must begin with the observation that war, conflict, and aggression are relations between persons. The view that it can be wrong to consider merely the overall effect of one's action on the general welfare comes into prominence when those actions involve relations with others. A man's acts usually affect more people than he deals with directly, and those effects must naturally be considered in his decisions. But if there are special principles governing the manner in which he should treat people, that will require special attention to the particular persons toward whom the act is directed, rather than just to its total effect. Absolutist restri ctions in warfare appear to be of two types: restrictions on the class of persons at whom aggression or violence may be directed and restrictions on the manner of attack, given that the obj ect falls within that class. These can be combined, however, under the principle that hostile treatment of any person must be j ustified in terms of something about that p erson which makes the treatment appropriate. Hostility is a personal relation, and it must be suited to its target. One consequence of this condition will be that certain persons may not be subjected to hostile treatment in war at all, since nothing about them justifies such treatment. O thers will be proper objects of hostility only in certain circumstances, or when they are engaged in certain pursuits. And the appropriate manner and­ extent of hostile treat ment will depend on what is justified by the particular case. A coherent view of this type will hold that extremely hostile behavior toward another is compatible with treating him as a person - even perhaps as an end in himsel£ This is possible only if one has not automatically stopped treating him as a person as soon as one starts to fight with him. If hostile, aggressive, or combative treatment of others always violated the condition that they be trea ted as human beings, it would be difficult to make further distinctions on that score withit1 the class of hostile actions. That point of view, on the level of international relations, leads to the position that if complete pacifism is not accepted, no holds need be barred at all, and we may slaughter

War aud massacre

65

and m assacre to our hearts' content, if it seems advisable. Such a posi tio n is often expressed in discussions of war crimes. But the fact is that ordinary people do not believe this about conflicts, physical or otherwise, between individuals, and there is no more reason why it should be true of conflicts between nations. There seems to be a perfectly natural conception of the distinction between fighting clean and fighting dirty. To fight dirty is to direct one's hostility or aggression not at its proper object, but at a peripheral target which may be more vulnerable, and through which the proper object can be attacked indirectly. This applies in a fist fight, an election campaign, a duel, or a philosophical argument. If the concept is general enough to apply to all these matters, it should apply to war - both to the conduct of individual soldiers and to the conduct of nations. Suppose that you are a candidate for public office, convinced that the election of your opponent would be a disaster, that he is an unscrupulous demagogue who will serve a narrow range of interests and seriously infringe the rights of those who disagree with him; and suppose you are convinced that you cannot defeat hi m by conventional means. Now imagine that various uncon­ ventional means present themselves as possibilities: you possess information about his sex life which would scandalize the electorate if made public; or you learn that his wife is an alcoholic or that in his youth he was associated for a brief period with a proscribed political party, and you believe that this information could be used to blackmail him into withdrawing his candidacy; or you can have a team of your supporters flatten the tires of a crucial subset of his supporters on election day; or you are in a position to stuff the ballot boxes; or, more simply, you can have him assassinated. What is wrong with these methods, given that they will achieve an overwhelmingly desirable result? There are, of course, many things wrong with them: some are against the law; some infringe the procedures of an electoral process to which you are presumably committed by taking part in it; · very importantly, some may backfire, and it is in the inte rest of all political candidates to adhere to an unspoken agreement not to allow certain personal matters to intrude into a campaign. But that is not all. We have in addition the feeling that these measures, these methods of attack, are irrelevant to the issue

Mortal questions

66

between you and your opponent, that in taking them up you would not be directing yourself to that which makes him an object of your opposition. You would be directing your attack not at the true target of your hostility, but at peripheral targets that happen to be vulnerable. The same is true of a fight or argument outside the framework of any system of regulations or law. In an altercation with a taxi driver over an excessive fare, it is inappropriate to taunt hi m about his accent, flatten one of his tires, or smear chewing gum on his windshield; and it remains inappropriate even if he casts aspersions on your race, politics, or religion, or dumps the contents of your suitcase into the street.6 The importance of such restrictions may vary with the seriousness of the case; and what is unj ustifiable in one case may be justified in a more extreme one. But they all derive from a single principle: that hostility or aggression should be directed at its true object. This means both that it should be directed at the person or persons who provoke it and that it should aim more specifically at what is provocative about them. The second condition will determine what form the hostility may appropri­ ately take. It is evident that some idea of the relation in which one should stand to other people underlies this principle, but the idea is difficult to state. I believe it is roughly this: whatever one does to another person intentionally must be aimed at him as a subject, with the intention that he receive it as a subject. It should manifest an attitude to h im rather than just to the situation, and he should be able to recognize it and identify himself as its object. The procedures by which such an attitude is manifested need not be addressed to the person directly. Surgery, for example, is not a form of personal confrontation but part of a medical treatment that can be offered to a patient face to face and received by him as a response to his needs and the natural outcome of an attitude toward h im . 6

Why, on the other hand, does i t seem appropriate, rather than irrelevant, to punch someone in the mouth if he insults you? The answer is that in our culture it is an insult to punch someone in the mouth, and not just an injury. This reveals, by the way, a perfectly unobjectionable sense in which convention may play a part in determining exactly what falls under an absolutist restriction ·and what does not. I am indebted to Robert Fogelin for this point.

War and 111assacre

67

Hos tile treatrilent, unlike surgery, is already addressed to a pe rson, and does not take its interpersonal meaning from a wider co ntext. But hostile acts can serve as the expression or imp lem entation of only a limited range of attitudes to the person wh o is a ttacked. Those attitudes in turn have as objects certain real or presumed characte ristics or activities of the person which are tho ught to justify them . When this background is absent, hostile or aggressive behavior can no longer be intended for the reception of the victim as a subject. Instead it takes on the character of a purely bureaucratic operation. This occurs when one attacks someone who is not the true object of one's hostility - the true object may be someone else, who can be attacked through the victim; or one may not be manifesting a hostile attitude toward anyone, but merely using the easiest available path to some desired goal. One finds oneself not facing or addressing the victim at all, but operating on him - without the larger context of personal interaction that surrounds a surgical operation. If absolutism is to defend its claim to priority over considera­ tions of utility, it must hold that the maintenance of a direct interpersonal response to the people one deals with is a require­ ment which no advantages can justify one in abandoning. The requirement is absolute only if it rules out any calculation of what would justify its violation. I have said earlier that there may be circumstances so extreme that they render an absolutist position untenable. One may find then that one has no choice but to do something terrible. Nevertheless, even in such cases absolutism retains its force in that one cannot claim _iiJStijication for the violation. It does not become all right. As a tentative effort to explain this, let me try to connect absolutist limitations with the possibility of justifying to the victim what is being done to him. If one abandons a person in the course of rescuing several others from a fire or a sinking ship, one could say to him, ' You understand, I have to leave you to save the others. ' Similarly, if one subjects an unwilling child to a painful surgical procedure, one can say to him, 'If you could understand, you would realize that I am doing this to help you. ' One could even say, as one bayonets an enemy soldier, 'It's either you or me. ' But one cannot really say while torturing a prisoner, ' You understand, I have to pull out your finger-nails because it is

Mortal questions

68

absolutely essential that we have the names of your confede r­ ates'; nor can one say to the victims of Hiroshima, 'You understand, we have to incinerate you to provide the Japanes e government with an incentive to surrender. ' This does not take us very far, of course, since a utilitarian would presumably be willing to offer justifications of the latter sort to his victims, in cases where he thought they were sufficient. They are really justifications to the world at large, which the victim, as a reasonable man, would be expected to appreciate. However, there seems to me something wrong with this view, for it ignores the possibility that to treat someone else horribly puts you in a special relation to him, which may have to be defended in terms of other features of your relation to him. The suggestion needs much more development; but it may help us to understand how there may be requirements which are absolute in the sense that there can be no justification for violating them. If the justification for what one did to another person had to be such that it could be offered to him specifically, rather than just to the world at large, that would be a significant source of restraint. If the account is to be deepened, I would hope for some results along the following lines. Absolutism is associated with a view of oneself as a small being interacting with others in a large world. The j ustifica tions it re q ui res are pri marily interpersonal.

Utilitarianism is associated with a view of oneself as a benevo­ lent bureaucrat distributing such benefits as one can control to countless other beings, with whom one may have various relations or none. The justifications it requires are pri marily administrative. The argument between the two moral attitudes may depend on the relative priority of these two conceptions.7 VI Some of the restrictions on methods of warfare which have been adhered to from time to time are to be explained by the mutual interests of the involved parties: restrictions on weaponry, 7

Finally, I should me n t io n a different possibility, suggested by Robert Nozick: that there is a strong general presumption against benefiting from the calamity of another, whether or not it has been deliberately inflicted for that o r any other reason. This broader p rin ci ple may well lend i t s force to the absol u ti s t position.

War aud massacre

·

69

treatment of prisoners, etc. But that is not all there is to it. The conditions of directness and relevance which I have argued apply to rela tions of conflict and aggression apply to war as well. I have said that there are two types of absolutist restrictions on the conduct of war: those that limit the l egitimate targets of hostility and those that limit its character, even when the target is acceptable. I shall say something about each of these. As will become clear, the principle I have sketched does not yield an unambiguous answer in every case. First let us see how it implies that attacks on some people are allowed, but not attacks on others. It may seem paradoxical to assert that to fire a machine gun at someone who is throwing ham:) grenades at your emplacement is to treat him as a human being . Yet the relation with him is direct and straightforward.B The attack is aimed specifically against the threat presented by a dangerous adversary, and not against a peripheral target through which he happens to be vulnerable but which has nothing to do with that threat. For example, you might stop him by machine­ gunning his wife and children, who are standing nearby, thus distracting him from his aim of blowing you up and enabling you to capture him. But if his wife and children are not threatening your life, that would be to treat them as means with a vengeance. This, however, is just Hiroshima on a smaller scale. One objection to weapons of mass annihilation - nuclear, thermonuc­ lear, b iological, or chemical - is that their indiscriminateness disqualifies them as direct instruments for the expression of hostile relations. In attacking the civilian population, one treats neither the military enemy nor the civilians with that minimal respect which is owed to them as human beings. This is clearly true of the direct attack on people who present no threat at all. But it is also true of the character of the attack on those who are threatening you, i.e. , the government and military forces of the enemy. Your aggression is directed against an area of vulnerabil­ ity quite distinct from any threat presented by them which you may be justified in meeting. You are taking aim at t he m through

the mundane life and survival of their countrymen, instead of 8

Marshall Cohen once remarked that, according to my view, shooting at someone establishes an 1-thou relationship.

70

Mortal questions

aiming at the destruction of their military capacity. And o f course it does not require hydrogen bombs to com mit such cnmes. This way of looking at the matter also helps us to understan d the importance of the distinction between combatants and noncombatants, and the irrelevance of much of the criticis m offered against its intelligibility and moral significance. Accord­ ing to an absolutist position, deliberate killing of the innocent is murder, and in warfare the role of the innocent is filled by noncombatants. This has been thought to raise two sorts of problems: first, the widely imagined difficulty of making a division, in modern warfare, between combatants and noncom­ batants; second, problems deriving from the connotation of the word 'innocence'. Let me take up the latter question first. 9 In the absolutist position, the operative notion of innocence is not moral inno­ cence, and it is not opposed to moral guilt. If it were, then we would be justified in killing a wicked but noncombatant hair­ dresser in an enemy city who supported the evil policies of his government, and unjustified in killing a morally pure conscript who was driving a tank toward us with the profoundest regrets and nothing but love in his heart. But moral innocence has very little to do with it, for in the definition of murder 'innocent' means 'currently harmless', and it is opposed not to 'guilty' but to 'doing harm'. It should be noted that such an analysis has the consequence that in war we may often be justified in killing people who do not deserve to die, and unjustified in killing people who do deserve to die, if anyone does. So we must distinguish combatants from noncombatants on the basis of their immediate threat or harmfulness. I do not claim that the line is a sharp one, but it is not so difficult as is often supposed to place individuals on one side of it or the other. Children are not combatants even though they may join the armed forces if they are allowed to grow up. Women are not combatants just because they bear children or offer comfort to the soldiers. More problematic are the supporting personnel, whether in or out of uniform, from drivers of munitions trucks and army cooks to civilian munitions workers and farmers. I 9

What I s a y on this subject derives from Anscombe.

War a11d massacre

71

believe they can b e plausibly classified b y applying the condition th at the p rosecution of conflict must direct itself to the cause of danger, and not to what is peripheral. The threat presented by an army and its members does not consist merely in the fact that they are men, but in the fact that they are armed and are using their arms in the pursuit of certain objectives. Contributions to their arms and logistics are contributions to this threat; contribu­ tio ns to their mere existence as men are not. It is therefore wrong to direct an attack against those who merely serve the comba­ tants ' needs as human being s, such as farmers and food suppliers, even though survival as a human being is a necessary condition of efficient functioning as a soldier. This brings us to the second group of restrictions: those that what may be done even to combatants. These limits are mit li harder to explain clearly. Some of them may be arbitrary or conventional , and some may have to be derived from other sources; but I believe that the condition of directness and relevance in hostile relations accounts for them to a considerable extent Consider first a case which involves both a protected class of noncombatants and a restriction on the measures that may be used against combatants. One provision of the rules of war which is universally recognized though it seems to be turning into a dead letter in Vietnam, is the special status of medical personnel and the wounded in warfare. It might be more efficient to shoot medical officers on sight and to let the enemy wounded die rather than be patched up to fight another day. But someone with medical insignia is supposed to be left alone and permitted to tend and retrieve the wounded. I believe this is because medical attention is a species of attention to completely general human needs, not specifically the needs of a co mbat soldier, and our conflict with the soldier is not with his exis tence as a human being. By extending the application of this idea, one can justify prohibitions against certain particularly cruel weapons: starva­ tion, poisoning, infectious diseases (supposing they could be inflicted on combatants only), weapons designed to mai m or disfigure or torture the opponent rather than merely to stop him. It is not, I think, mere casuistry to claim that such weapons attack the men, not the soldiers . The effect of dum-dum bullets. .

,

Mortal questions

72

for example, is much more extended than necessary to cope wi th the combat situation in which they are used. They abandon any attempt to discriminate in their effects between the combatant and the human being. For this reason the use of flamethrowen and napalm is an atrocity in all circumstances that I can imagine, whoever the target may be. Bums are both extremely pain ful and extremely disfiguring - far more than any other category of wound. That this well-known fact plays no (inhibiting) part in the determination of U. S. weapons policy suggests that moral sensitivity among public officials has not increased markedlv since the Spanish Inquisition. tO Finally, the same condition of appropriateness to the true object of hostility should limit the scope of attacks on an enemy country: its economy, agriculture, transportation system, and so forth. Even if the parties to a military conflict arc considered to be not armies or governments but entire nations (which is usually a grave error), that does not justify one nation in warring against every aspect or element of another nation. That is not justified in a conflict between individuals, and nations are even more complex than individuals, so the same reasons apply. Like a human being, a nation is engaged in countless other pursuits while waging war, and it is not in those respects that it is an enemy. The burden of the argument has been that absolutism abou1 murder has a foundation in principles governing all one's relations to other persons, whether aggressive or amiable, and that these principles and that absolutism, apply to warfare as well, with the result that certain measures are impermissible no matter •

10

Beyond this I feel uncertain. Ordinary bullets, after all, can cause death, and nothing is more permanent than that. · 1 am not at all sure why we are justified in trying to kill those who are trying to kill us (rather than merely in trying to stop them with force which may also result in their deaths). It is often a rg u ed that incapacitating gases are a relatively humane weapon (when not used, as in Vietnam, merely to make people easier to shoot). Perhaps the legitimacy of restrictions against them must depend on the dangers of escalation, and the great utility of maintaining auy conventional category of restriction so long as nations are willing to adhere to it. Let me make clear that I do not regard my argument as a defense of the moral im mutability of the Hague and Geneva Conventions. Rather, I believe that they rest partly on a moral foundation, and that modifications of them should also be assessed on moral grounds.

War at1d massacre

73

wh at the consequences. I I I do not mean to romanticize war. It is su fficien tly utopian to suggest that when nations conflict they mi gh t rise to the level of limited barbarity that typically charac te rizes violent conflict between individuals, rather than wallow ing in the moral pit where they appear to have settled, surro unded by enormous arsenals. VI

described the elements of the absolutist position, we return to the conflict between it and utilitarianism. now must certain types of dirty tactics become acceptable when the if Even are high enough, the most serious of the prohibited acts, es stak r and torture, are not j ust supposed to require unusu­ like murde justification. They are supposed never to be done, strong ally quantity of resulting benefit is thought capable of cause no be jllstifying such treatment of a person. The fact remains that when an absolutist knows or believes that the utilitarian cost of refusing to adopt a prohibited course will be very high, he may hold to his refusal to adopt it, but he will find it difficult to feel that a moral dilemma has been satisfactorily resolved. The same may be true of someone who rejects an absolutist requirement and adopts instead the course yielding the most acceptable consequences. In either case, it is possible to feel that one has acted for reasons insufficient to justify violation of the opposing principle. In situations of deadly conflict, particularly where a weaker party is threatened with annihilation or enslavement by a stronger one, the argument for resorting to atrocities can be powerful, and the dilemma acute. There may exist principles, not yet codified; which would enable us to resolve such dilemmas. But then again there may not. We must face the pessimistic alternative that these two forms of moral intuition are not capable of being brought together into a single, coherent moral system, and that the world can present us with situations in which there is no honorable or Having

II

It is possible to draw a more radical conclusion, which I shall not pursue here. Perhaps the technology and organization of modern war are such as to make it impossible to wage as an acceptable form of interpersonal or even international hostility. Perhaps it is too impersonal and large-scale for that. If so, then absolutism would in practice imply pacifism, given the · present state of things. On the other hand. I am skeptical about the . unstated assumption that a technology dictates its own use.

74

Mortal questions

moral course for a man to take, no course free of guilt and responsibility for evil. 12 The idea of a moral blind alley is a perfectly intelligible one. h is possible to get into such a situation by one's own fault, an d people do it all the time. If, for example, one makes two incompatible promises or commitments - becomes engaged to two people, for example - then there is no course one can take which is not wrong, for one must break one's p romise to at least one of them. Making a clean breast of the whole thing will not be enough to remove one's reprehensibility. The existence of such cases is not morally disturbing, however, because we feel that the situation was not unavoidable: one had to do something wrong in the first place to get into it. But what if the world itself, or someone else's actions, could face a previously innocent person with a choice between morally abominable courses of action, and leave him no way to escape with his honor? Our intuitions re b e l at the idea, for we feel that the constructibility oi such a case must show a contradiction in our moral views. But it is not in itself a contradiction to say that someone can do X or not do X, and that for him to take either course would be wrong. It merely contradicts the supposition that ough t implies can since presumably one ought to refrain from what is wrong, and in such a case it is impossible to do so. t3 Given the limitations on human action, it is naive to suppose that there is a solution to every moral problem with which the world can face us. We have always known that the world is a bad place. It appears that it may be an evil place as well.

12

13

In his reply to this essay ('Rules of War and Moral Reasoning', P!Jilosop lly I, no. 2 (Winter, 1 972), 1 67), R. M. Hare pointed out the apparent discrepancy between my accep tance of such a possibility here and my earlier claim in se c tio n IV that absolutism must be formulated so as to avoid the consequence that in certain cases nothing one could do wo u ld be mo rall y permissible. The difference is that in thos e cases the moral incoherence would result from the application of a single principle. whereas the dilemmas d escribed here result from a conflict between two fundamentally di ffe re nt types of p rinciple. This was first pointed out to me by Christopher Bo orse . The point is al so made in E. ]. Lemmon's ' Moral Dilemmas', Pllilosopll ica l Review , LXX I (Ap ril 1 962), I SO. r� Public Affa irs,

,

6 Ruthlessness

tn

Public Life

The great modern crimes are public crimes. To a degree the same can be said of the past, but the growth of political power has introduced a scale of massacre and despoliation that makes the efforts of private criminals, pirates, and bandits seem truly modest. Public crimes are committed by individuals who play roles in political, military, and economic institutions. (Because religions are politically weak, crimes committed on their behalf are now rare.) Yet unless the offender has the originality of Hitler, Stalin, or Amin, the crimes do not seem to be fully attributable to the individual himself. Famous political monsters have moral per­ sonalities large enough to transcend the boundaries of their public roles; they take on the full weight of their deeds as personal moral property. But they are exceptional. Not only are ordinary soldiers, executioners, secret policemen, and bombar­ diers morally encapsulated in their roles, but so are most secretaries of defense or state, and even many presidents and prime ministers. They act as office-holders or functionaries, and thereby as individuals they are insulated in a puzzling way from what they do: insulated both in their own view and in the view of most observers. Even if one is in no doubt about the merits of the acts in question, the agents seem to have a slippery moral su rface, produced by their roles or offices. This is certainly true of several American statesmen respons­ ible for the more murderous aspects of policy during the Vi etnam War. Robert McNamara is president of the World _

76

Mortal questions

Bank. McGeorge. Bundy is president of the Ford Foundation. Elliot Richardson was secretary of defense under Nixon duri ng the completely illegal bombing of Cambodia which went on after the Vietnam peace agreements were signed. He then became attorney general and was widely acclaimed for resigning th:u office rather than comply with Nixon's request that he fire Archibald Cox for demanding the White House tapes. His highly selective sense of honor has served him well: he has since been ambassador to Britain, secretary of commerce and ambas­ sador at large, and we shall hear more of him. Kissinger is of course a highly esteemed figure, despite the Christmas bombing of 1 972 and all that preceded it. The judgments I am presupposing are controversial: not everyone agrees that A merican policy during the Vietnam War was criminal. But even those who do think so may find it hard to attach the crimes to the criminals, in virtue of the official role in which they were committed. Few old anti-war demonstrator.; would feel more than mildly uncomfortable about meeting one of these distinguished figures, unless it was j ust because we were unaccustomed to personal contract with anyone as powerful as the president of the World Bank. There is, I think, a problem about the moral effects of public roles and offices. Certainly they have a profound effect on the behavior of the individuals who fill them, an effect partly restrictive but significantly liberating. Sometimes they confer great power, but even where they do not, as in the case of an infantryman or police interrogator, they can produce a feeling of moral insulation that has strong attractions. The combination of special requirements and release from some of the usual restric­ tions, the ability to say that one is only following orders or doing one's job or meeting one's responsibilities, the sense that one is the agent of vast impersonal forces or the servant oi institutions larger than any individual - all these ideas form a heady and sometimes corrupting brew. But this would not be so unless there were something to the special status of action in a role. If roles encourage illegitimate release from moral restraints it is because their moral effect has been distorted. It will help to understand the distortion if we consider another curiosity of current moral discourse about public life: the emphasis placed on those personal restrictions

Ruthlessness itt public life

77

that comple ment the lack of official restraint - the other side of the coin of public responsibility and irresponsibility. Public

fi gu res are not supposed to use their power openly to enrich th emsel ves and their families, or to obtain sexual favors. Such pri miti ve indulgences are generally hidden or denied, and stress is lai d on the personal probity and disinterest of pub lic figures. This kind of personal detachment in the exercise of official func ti ons is thought to guarantee their good moral standing, and it lea ves them remarkably free in the p ub lic arena. No doubt p rivate transgressions are widespread, but when they are ines­ capably .exposed the penalty can be severe, for a delicate bou ndary of moral restraint that sets off the great bod y of public power and freedom has been breached. Spiro Agnew will never be head of the Ford Foundation. The exchange seems fairly straightforward. The exercise of pub lic power is to be liberated from certain constraints by the imposition of others, which are primarily personal. Because the office is supposedly shielded from the personal interests of the one who ftl.ls it, what he does in his official capacity seems also to be depersonalized. This nourishes the illusion tha t personal moralit y does not apply to it with any force, and that it cannot be strictly assigned to his moral account. The office he occupies gets between him and his depersonalized acts. Among other things, such a picture disguises the fact that the exercise of power, in whatever role, is one of the most personal forms of individual self-expression, and a rich source of purely personal pleasure. The pleasure of power is not easily acknow­ ledged , but it is one o f the most primitive human feelings proba bl y one with infantile roots. Those who have had it for years sometimes realize its i mportance only when they have to retire. Despite their grave demeanor, impersonal diction, and limited physical expression, holders of public power are person­ ally involved to an intense degree and probably enjoying it immensely. But whether or not it is consciously enjoyed, the exercise of power is a primary form of individual expression, not diminished but enhanced by the institutions and offices on which it depends. When we try, therefore, to say what is morally special about pu blic roles and public action, we must concenttate on how they alter the demands on the individual. The actions are his, whether -

78

Mortal questiom

they consist of planning to obliterate a city or only firing in response to an order. So if the moral situation is different from the case where he acts in no official capacity, it must be because the requirements are different.

II It is hard to discuss this subject in general terms, since roles and offices differ so widely. Nevertheless, the question of the natu re of the discontinuity between individual morality and public morality is in part a general one, because the answer must take one of two forms. Either public morality will be derivable from individual morality or it will not. The answer will vary greatly in detail from case to case, but if a significant element of public morality is not derivable from the moral requirements that apply to private individuals, it is probably a common feature of many different examples. To give the question content, it is necessary to say more about derivability. The interesting question is whether the special features of public morality can be explained in terms of princi­ ples already present at the individual level, which yield apparent moral discontinuities when applied to the special circu mstances of public life. If so, then public morality is in a substantive and not merely trivial sense derivable from private morality. t It emerges naturally from individual morality under the conditions that define the individual's public role. This could still yield different moral requirements in two ways. Either the general principles could imply additional constraints on public action; or the principles could be such that certain requirements would cease to apply once one assu med a public role, because the conditions for their application would have disappeared. Or the change might involve some combination of the two. In view of the second kind of change, even if public morality is derivable from private, it is possible that the moral restraints on public action are weaker than those on individual action. The alternative to derivability is that public morality is not 1

Public morality becomes trivially derivable from individual morality i i individual morality is extended to include all true propositions of the form, 'if the individual is acting in public role X, he may (or must) do Y', and so forth. This is compatible, however, with there being no connection between the grounds of the public and private req uirements.

Ruthlessness in public life

79

gr� und�d on individual morality, and that therefore people acting m certain official roles or capacities are required or per mitted to do things that cannot be accounted for on that basis. This also might take two forms. They might come under restrictions in areas left free by individual morality: public officials might be held to higher standards of concern for the general welfare, for example, than ordinary people. Or else those acting in official roles might be permitted or even required to do things which, considered from the point of view of individual morality, would be impermissible. Both derivability and non-derivability are formally suited to exp lain either the addition or the removal of restrictions in public morality; both can therefore explain the appearance of discontinuity. The only way to decide between them is to see which form of explanation can be more plausibly filled out. I shall begin with a version of the derivability hypothesis, based on familiar concepts of individual morality. But while this can explain a good deal, it also leaves something out. I shall therefore go on to say what seems to me true in the nonderivability hypothesis, and this will involve giving an account of the alternative basis on which special conditions of public morality depend. Even if public morality is not derivable from private, how­ ever, it does not mean that they are independent of one another. Both may derive from a common source that yields different results when applied to the generation of principles for action in the widely differing circumstances of private and public life. Neither private morality nor public morality is ultimate. Both result when the general constraints of morality are applied to certain types of action. Public morality would be derivable from private only if those constraints had to be applied first to the development of principles governing the conduct of persons acting individually, and could not be applied directly to public life. In that case one would have to reach the private principles fro m the general constraints of morality, and the public princi­ ples only from the private ones, as applied to public circums­ tan ces . But there is no a priori reason to think that ethics has this stru cture. If it does not, then public and private morality may share a common basis without one being derived from the other. I shall say more about this later. First I want to explore the more

!vlortal questio11s

80

direct connexions between them. Part of my aim is to give a correct account of facts that are easily distorted by those defenders of political, diploma tic or military license who cloak themselves in the responsibilities of office. Whoever denies the application of moral restrain ts to certain public decisions is making a moral claim, and a ver y strong one. But there is something to the idea of a mo ral discontinuity between private and public, and to understand the distortions we must know what this is. III Some of the moral pecul iarity of official roles can be explained by the theory of obligation. Whoever takes on a public or offi cial role assumes the obligation to serve a special function and often the interests of a special group. Like more personal obligations this limits the claim that other sorts of reasons can make on him. Recall E. M . Forster's re m ark : 'I hate the idea of causes, and if I had to choose between betraying my country and betraying my friend, I hope I should have the courage to betray my country. '2 He was not talking about public office, but similar pro b lems can arise there. In a rigi dly defined role like that of a soldier or judge or prison guard, o nl y a very restricted set of considerations is supposed to bear on what one decides to do, and nearly all general considerations are excluded. With less definition, other public offices limit their occupants to certain considerations and free them from others, such as the good of ma n k ind Public figures sometimes even say and believe that they are obliged to consider only the national or state interest in arriving at their decisions as if it would be a breach of responsibility for them to ,

.

consider an y thing else.

This apparent restriction on choice is easy to accept partly because, looked at from the other direction, it lifts restraints that might otherwise be burdensome. But any view as absolute as this is mistaken: there are no such extreme obligations or offices to which they attach. One cannot, by joining the army, under­ take an o b ligation to obey any order whatever from one's commanding officer. It is not possible to acquire an obligation to 2 'What I Believe',

1939).

in Two Chew for Democracy (London: Edward Arnold

Ruthlessness in public life kill indebte d

81

gamblers by signing a contract a s a Mafia hit man. not even possible to undertake a commitment to serve the interes ts of one's children in complete disregard of the interests of everyone else. Obligations to the state also have limits, which derive from their moral context. Every obligation or commitment reserves some portion of the general pool of motivated action for a special purpose. Life being what it is, each person's supply of time, power, and energy is limited. The kinds of obligations one may undertake, and their limits, depend on how it is reasonable to allocate this pool, and how much liberty individuals should have to allocate it in radically uneven ways. This is true for personal obligations. It applies to public ones as well. In private life some exclusivity is necessary if we are to allow people to form special relations and attachments, and to make special arrangements with each other on which they can rely. For similar reasons larger groups should be able to cooperate for mutual benefit, or to form social units that may have a geog­ raphical definition. And it is natural that the organization of such cooperative units will include institutions, roles, and offices and that the individuals in them will undertake obligations to serve the interests of the group in special ways - by promoting its prosperity, defending it against enemies, etc. To a degree, large-scale social arrangements can be seen as extensions of more individual obligations and commitments. It may be that the added power conferred by an institutional role should be used primarily for the benefit of that institution and its constituents. The interests of mankind in general have a lesser claim on it. But this does not mean that prohibitions 1gainst harming others, directly or indirectly, are correspond­ ingly relaxed. Just because the power to kill thousands of people is yours only because you are the secretary of defense of a certain country, it does not follow that you should be under no restrictions on the use of that power which do not derive specifically from your obligations to serve that country. The same reasoning that challenges private obligations that i mply too much of a free hand in carrying them out, will also disallow public commitments with inadequate restraints on their greater power. Insofar as public obligations work like private ones, there is no reason to think that individuals in public roles are

It is

82

Mortal questions

released from traditional moral re q uirements on the treatmen t of others, or that in public life , the end justifies the means. IV

Let me now say what such an account leaves out. The moral impersonality of public action may be exaggerated and abused, but there is something in it, which a general theory of o blig ation cannot explain. Such a theory fails to explain why the content of public obligations differs systematically from that of private ones. The i mpersonality suitable for public acti o n has two aspects: it implies both a heightened concern for resu l ts and a stricter requirement of impartiality. It warrants methods usua l ly excluded for private individuals, and sometimes it licenses ruthlessness. This can be explained only by a direct application of moral theory to those public institutions that create the roles to which public obligations are tied.3 To account for the difference between public and private life we must return to a point mentioned earlier: that public moralit y may be underiv­ able from private not because they come from different sources, but because each of them contains elements derived indepen­ dently from a common source .4 Morality is complicated at every level. My basic claim is that its impersonal aspects are more prominent in the assessment of institutions than in the assessment of individual actions, and that as a result, the design of institutions may in clude roles whose occupants must determi ne what to do by principles different from those that govern pri v ate individuals. This will be morally justified, however, by ultimate considerations that underlie individual morality as well. I shall present the view only in outline, and mostly without defending the moral opinions it expresses. My main contention is that the degree to which ruthlessness is acceptable in public life - the ways in which pu blic actors may have to get their hands dirty - depends on 3

4

What I say will be put in terms of the largest and most powerful institutions, the state and its agencies. But there is a wide range of pub lic institutions, including universities, political parties, charitable organizations, and revolutionary movements. Much of what I shall say about nation-states applies to these cases also in some degree. They too come under a kind of public morality. This retracts something I said at pp. 1 39-40 of 'Libertarianism without Foundations', Yale Law Journal, LXXXV (1975).

Ruthlessness in public life m oral

83

features of the institutions through which public action is

e2 rried out. Two types of concern determine the content. of morality: concern with what will happen and concern with what one is doi ng.s Insofar as principles of conduct ·are determined by the first concern, they will be outcome-centered or consequentialist, req uirin g that we promote the best overall results. Inso far as the y are determined by the second, the influence of consequ­ ences will be limited by certain restrictions on the means to be used, and also by a loosening of the requirement that one always pursue the best results. The action-centered aspects of morality include bars agains t treating others in certain ways which violate their rights, as well as the space allotted to each person for a life of h is own, without the perpetual need to contribute to the general good in everything he does. Such provisions are des cribed as action-centered because, while they apply to everyone, what they require of each person depends on his particular standpoint rather than on the impersonal consequen­ tialist standpoint that surveys the best overall state of affairs and p re scribes for each person whatever he can do to contribute to it. The interaction and conflict between these two aspects of m orality are familiar in private life. They · result in a certain balance that emphasizes restrictions against harming or interfer­ ing with others, rather than requirements to benefit them, except in cases of serious distress. For the most part it leaves us free to pursue our lives and form particular attachments to some people, so long as we do not harm others. When we apply the same dual conception to public i nstitu­ tions and activities, the results are diffe rent. There are s everal reasons for this. Institutions are not persons and do not have private lives, nor do institutional roles usually absorb completely the lives of their occupants. Public institutions are designed to s erve purposes lar g er than those of particular individuals or families. They tend to pursue the interests of masses of people (a limiting case would be that of a world government, but most actual institutions have a less than universal constituency). In addition, public acts are diffused over many actors and sub­ institutions; there is a division of labor both in execution and in decision. All this results in a different balance between the 5

I discuss this

distinction

in chapter

5.

84

Mortal questiorrs

morality of outcomes and the morality of actions. These two types of moral constraint are differently expressed in public life, and both of them take more impersonal forms. Some of the same agent-centered restrictions on means will apply to public action as to private. But some of them will be weaker, permitting the public employment of coercive, man­ ipulative, or obstructive methods that would not be allowa ble for individuals. There is some public analogue to the individual's right to lead his own life free of the constant demand to promote the best overall results, but it appears in the relations of sta tes to one another rather than in their relations to their citizens: states can remain neutral in external disputes, and can legitimately favor their own populations - though not at any cost whatever to the rest of the world. There is no comparable right of self-indulgence or favoritism for public officials or institutions vis-a-vis the individuals with whom they deal. Perhaps the most significant action-centered feature of public morality is a special requirement to treat people in the relevant population equally. Public policies and actions have to be much more i mpartial than private ones, since they usually employ a monopoly of certain kinds of power and since there is no reason in their case to leave room for the personal attachments and inclinations that shape individual lives. 6 In respect to outcomes, public morality will differ from private in according them greater weight. This is a consequence of the weakening of certain action-centered constraints and permissions already described, which otherwise would have restrictive effects. The greater latitude about means in turn makes it legitimate to design institutions whose aim is to produce certain desirable results on a large scale, and to define roles in those institutions whose responsibility is mainly to further those results. Within the appropriate limits, pu blic decisions will be justifiably more conse q uentialist th an private ones. They will also have larger consequences to take into account. 6

Would a giant with i m mense power be obliged to act primarily on impers onal grounds, if he were unique a m ong millions of ordinary people whose lives he could affect? I doubt it. He would presu mably have a personal life as well, which made some cl a ims on him. The state is the closest thing we know to such a gian t and it is not similarly encumbered. ,

Ruthlessness in public life

85

To say that consequentialist reasons will be prominent is not o t sa y what kinds of consequences matter. This is a well-worked field, and I shall avoid discussing the place of equality, liberty, autonom y and individual rights, as well as overall level of happi ness, in a consequentialist view of the good. The point to re member is that consequentialist values need not be utilitarian; assessment of social institutions can be a con sequentialist strong ly egalitarian, in addition to valuing welfare, liberty, and individuality in themselves. Moreover, giving the members of a societ y the opportunity to lead their own lives free of conse­ quentialist demands is one of the goods to be counted in a consequentialist social reckoning. But I will not try to present a complete system of public values here, for I am concerned with the more abstract claim that consequentialist considerations, together with impartiality, play a special role in the moral assessment and justification of public institutions. The effect of these two deviations of public from private morality on the assessment of public action will be complex. The reason is that the constraints of public morality are not imposed as a whole in the same way on all public actions or on all public offices. Because public agency is itself complex and divided, there is a corresponding ethical division of labor, or ethical specialization. Different aspects of public morality are in the hands of different officials. This can create the illusion that public morality is more consequentialist or less restrictive than it is, because the general conditions may be wrongly identified with the boundaries of a particular role. But in fact those boundaries usually presuppose a larger institutional structure without which they would be illegitimate. (The most conspicu­ ous example is the legitimacy conferred on legislative decisions by the limitation of constitutional protections enforced by the courts.) By this rather complex route, the balance of outcome-oriented and action-oriented morality will justify the design of public institutions whose officials can do what would be unsuitable in p rivate life. Some of the deviations will be conspicuously consequentialist: others will express the impersonality of public morality in other ways. Action-centered constraints will not be abs ent: there will still be restrictions on means. But those restrictions may be weaker in relation to the results than they are

Mortal questio11s

86

for individuals. I have simply adapted a point made by Rawls in 'Two Concepts of Rules'.? He argued that utilitarianism could justify practices that exclude utilitarian reasoning in some circums­ tances. I am arguing that a more complex morality than utilitarianism will likewise have different i mplications for human conduct when applied to its assessment directly and when applied indirectly via the assessment of institutions through which action occurs. The details of this morality cannot be explained here, but many of its features depend on an idea of moral universality different from that which underlies utilitarianism. Utilitarian assessment decides, basically, whether something is acceptable from a general point of view that combines those of all individua]s. The method of combination is basically majoritarian. The alternative is to ask whether some­ thing is acceptable from a schematic point of view that repres­ ents in essentials the standpoint of each individual. The method of combination here is a form of unanimity, since acceptability from the schematic point of view represents acceptability to each person. Both of these moral conceptions can claim to count everyone e q ually, yet they a re very d i fferent. My own opinion is that morality should be based on acceptability to each rather than on acceptability to all. The problem is to define the two points of view that express these opposed moral conceptions.s It could also be said that the separate application of these basic constraints to social institutions and to individual conduct yields a moral division of labor between the individual and society, in which individual and social ideals are inseparably linked. The impersonal benevolence of public morality is intended to pro­ vide a background against which individualism in private moral­ ity is acceptable. It is a pressing and diffi cult question whether private individualism and public benevolence are socially com­ patible, or whether the tension between them makes this an unstable moral conception and an unstable social ideal. v

Because they are specialized, not all public institutions are 7

8

Philosophical Review,

LXIV ( 1 955), 3-32. One attempt is made by Rawls in A Theory ofjustice (Cambridge, Mass. : Harvard University Press, 1 97 1 ), ch. 111. See also chapter 8 below.

RuthlesSI/ess i11 public life

87

equa lly sensitive to overall consequences. An important excep­ tio n is the judiciary, at least in a society where the courts are design ed to protect individual rights against both public and p ri vate encroachment. Neither the institution itself nor the roles it defi nes - judge, juror, prosecutor - are dominated by a concern with overall results. They act on narrower grounds. To some extent this narrowing of grounds is itself justified by consequentialist reasoning about the overall effects of such an institution. However -the courts also embody the state's action­ centered ,moral constraints - impersonal but not consequential­ ist. i ty

Very i mportantly, they are supposed to enforce its imp arti al­

in serious dealings with individual citizens. And by setting limits to the means that can be employed by other public institutions, they leave those institutions free to concentrate more fully on achieving results within those limits. To illustrate the positive claim that these limits differ from those that operate in private life, let me consider two familiar examples of public action: taxation and conscription. Both are imposed by the legislature in our society, and it may be thought that they are therefore indirectly consented to by the population. I believe it is a desperate measure to impute consent to everyone who is drafted or pays income taxes, on the ground that he votes or accepts certain public services. Consent is not needed to justify such legislative action, because the legislature is an institution whose authority to make such decisions on conse­ quentialist grounds is morally justified in other ways. Its periodic answerability to the electorate is one feature of the institution (another being the constitutional protection of rights) that contributes to its legitimacy - but not by implying each citizen's consent to its actions.9 Particularly when those actions are coercive, the defense of consent is not credible. Some would describe taxation as a form of theft and conscrip­ tion as a form of slavery - in fact some would prefer to describe taxation as slavery too, or at least as forced labor. tO Much might be said against these descriptions, but that is beside the point. For 9

10

This conception of legitimacy is found in Thomas M. Scanlon. ' Nozick on Rights, Liberty, and Property', Philosophy & Public Affairs, VI, no. 1

(1 976), 17-20.

E. g . Robert Nozick, Auarrhy, Statt, aud Utopia (New York: Basic Books, 1 974), pp. 1 69-74.

88

Mortal questio11s

within proper limits, such practices when engaged in by gov­ ernments are acceptable whatever they are called. If someon e with an income of $2000 a year trains a gun on someone with an income of $ 1 00 000 a year and makes him hand over his wallet, that is robbery. If the federal government withholds a portion of the second person's salary (enforcing the laws against tax evasion with threats of imprisonment under armed guard) and gives some of it to the first person in the form of welfare pay ments, food stamps, or free health care, that is taxation. In the first case it is (in my opinion) an i mpermissible use of coerci ve means to achieve a worthwhile end. In the second case the means are legitimate, because they are impersonally imposed by an institution designed to promote certain results. Such general methods of distribution are preferable to theft as a form of private initiative and also to individual charity. This is true not only for reasons of fairness and efficiency, but also because both theft and charity are disturbances of the relations (or lack of them) between individuals and involve their individual wills in a way that an automatic, officially imposed system of taxation does not. The results achieved by taxation in an egalitarian welfare state would not be produced either by a right of individual expropriation or by a duty of charity. Taxation therefore provides a case in which public morality is derived not from p ri vate morality, but from impersonal consequentialist considerations applied directly to public institu tions, and secon­ darily to action within those institutions. There is no way of analyzing a system of redistributive taxation into the sum of a large number of individual acts all of which satisfy the require­ ments of private morality. In the case of conscription, the coercion is extreme, and so is what one is forced to do. You are told to try to kill people who are trying to kill you, the alternative being imprisonment. Quite apart from fighting, military service involves unusual restric­ tions of liberty. Even assu ming agreement about when conscri p­ tion is acceptable and what exemptions should be allowed, this is a kind of coercion that it would be unthinkable to impose privately. A cannot force B to help him fight a gang of hoodlums who are robbing them both, if B would rather give them his money. Again, the more impersonal viewpoint of public morality gives a different result.

Ruthlessuess iu public life

89

Bu t not everything is permitted. Restrictions on the treatment continue to operate from a public point of view, be implemented entirely by the courts. One of ey cannot a nd th lines to draw in public policy is the one that defines ardest the h the end stops justifying the means. If results were the only where public morality then it would be possible to justify for basis including torture and massacre, in the service_ of ing, anyth sufficiently large interests. Whether the limits are drawn by specific constitutional protections or not, the strongest con­ strain ts of individual morality will continue to limit what can be publicl y justified even by extremely powerful consequentialist reas ons. of in divi duals

VI

This completes my discussion of the continuities and discon­ tinuities between public and private morality. I have argued that some of the special features of public morality can be explained in terms of a theory of obligation that also accounts for the steps individuals can take to restrict the grounds on which they will make certain choices. Public officials accept special obligations to serve interests that their offices are designed to advance - and to serve them in more or less well-defmed ways. In doing so, they correlatively reduce their right to consider other factors, both their personal interests and more general ones not related to the institution or their role in it. I have also argued, however, that the special character of public obligations - the weight they give both to results and to impartiality - reflects the relative impersonality of public action: its scale, its lack of individuality, its institutional structure. A theory of obligation explains only part of the change that occurs when an individual takes on a public role. It does not explain either the prominence of consequentialism or the shift in strength and character of action-centered reasons. I have tried to explain these differences as the result of a direct application of basic moral constraints to public institutions and therefore to the public functions that individuals may undertake. Both of these sources of public morality generate limits to what a public official may do in the conduct of his office, even if he is serving institutional interests. It is easy to forget about those li mits, for three reasons. First, restrictions against the use of

90

Mortal questions

public power for private gain can seem like a moral cushion that insulates wha tever else is done officially from moral reproa ch.

Second, the fact that the holder of a public office takes on a n obligation to a particular group may foster the idea that he is obliged not to consider anything except the interest of that group . Third, the impersonal morality of public institutions, and the moral specialization that inevitably arises given the complex­ ity of public actions, lead naturally to the establishment of many roles whose terms of reference are primarily consequentialist. Lack of attention to the context that is necessary to make these roles legitimate can lead to a rejection of all limits on the means thought to be justified by ever greater ends. I have argued that these are all errors. It is important to remember that they are mora l views: the opinion that in certain conditions a certain type of conduct is permissible has to be criticized and defended by moral argument. Let me return finally to the individuals who occupy public roles. Even if public morality is not substantively derivable from private, it applies to individuals. If one of them takes on a public role, he accepts certain obligations, certain restrictions, and certain limi tations on what he may do. As with any obligation, this step involves a risk that he will be required to act in ways incompatible with other obligations or principles that he accepts. Sometimes he will have to act anyway. But sometimes, if he can remember them, he will see that the limits i mposed by public morality itself are being transgressed, and he is being asked to carry out a judicial murder or a war of unjust aggression. At this point there is no substitute for refusal and, if possible, resistance. Despite the impersonal character of pu b lic morality and its complex application to institutions in which responsibility is divided, it tells us not only how those institutions should be designed but also how people iri them should act. Someone who has committed public wrongs in the exercise of his office can be just as guilty as a priv:ate criminal. Sometimes his responsibility is partly absorbed by the moral defects of the institution through which he acts; but the plausibility of that excuse is inversely proportional to the power and independence of the actor. Unfortunately this is not reflected in our treatment of former public servants who have often done far worse than take bribes.

7 The P olic y of Preference

It is currently easier, o r widely thought to be easier, to get certain j obs o r to gain ad mission to certain educational institu­ tions if one i s black or a woman than i f one is a white man. Whether or not this is true, many think it should be true, and many others think i t should not. The issue is this: I f a black p erson o r a woman is admitted to a law school or medical school, or appointed to an academic o r administrative post, in p reference to a white man who is in other respects better qualified, 1 and if this is done i n pursuit of a p referential policy or to fill a quota, is i t unjust? Can the white man complain that he has been unjustly treated? It is i m portant to i nves tigate the justice o f such practi ces, because if they are unj us t, i t is much more difficult to defend them on grounds of social utility. I shall argue that although preferential policies are not req uired by justice, they a re not seriously u nj u s t either - because the s ys tem from which they depart is already unj ust for reasons having nothing to do with racial o r sexual discri mination. 1

B y saying t h a t t h e white man is 'in other respects better qualified' I mean that i f. e.g., a black candidate with similar qualifications had been available for the position, he would have been selected in preference to the black candidate who was in fact selected; or, if the choice had been between two white male candidates of corresponding qualifications, this one would have been selected. Ditto for two white or two black wo men. (I realize that i t may not always be easy to deter mine similarity of qualificati ons, and that in some cases similarity of credentials may give evidence of a difference in qualifications - because, e.g. , one person had to overcome more severe obstacles to acquire those credentials.)

92

Mortal questio11s

In the Uni ted States we have reached the p resen t situation by th e following s teps. First, and not very long ago, i t ca me to be widely accepted that deliberate barriers against the admissio n to desirable positions of blacks and women should be abolis hed. Their abolition is by no means co mplete, and certain educat io nal institutions, for exa mple, may be able to maintain limiting quotas on the ad mission of women for some ti me. But deliberate discri mination is widely condemned. Secondly, i t was recognized that even without explicit barriers there could be discrimination, either consciously or u ncon­ sciousl y motivated, and this gave support to self-conscious efforts at i m partiality, careful consideration of candidates belonging to the class discri minated against, and attention to the proportions of blacks and women in desirable positions, as evidence that otherwise undetecta ble bias might be influencing the selections. (Another, rela ted consideration was tha t criteria which were good p redictors of performance for one group might turn out to be poor predictors of performa nce for another group, so that the continued employ ment o f those criteria might introduce a concealed i nequi ty. ) The third s tep ca me with the realization that a social s ystem may continue to deny different races or sexes equal opportunity or equal access to desirable positions even after the discri mina­ tory barriers to those positions have been lifted. Socially caused inequality in the capacit y to use available opportuni ties or to co mpete for available positions may persist, because the society systematically provides to one group m o re than to another certain educational, social, or economic adva ntages. Such advan­ tages improve one's competitive position in seeking access to jobs or places in professional schools. Where there has recently been widespread deliberate discri mination in many areas, it is not surprising if the formerly excluded group experiences relativ e difficulty in gaining access to newly opened posi tions, and it is plausible to explain the difficulty a t least partly in terms of disadvantages produced by past discri mination. This leads to the adoption of co mpensatory measu res, in the for m of s peci al training p rogra ms, or financial support, or day-care centers, or apprenticeships, or tutoring. Such measu res are designed to

The p olic y of priference

93

q ual ify those whose reduced qualifications are due to racial or se x ual di scri mination, either because they have been its direct vi ct im s, or because they a re deprived as a result of mem bership in a g rou p o r community many of whose other members have been disc riminated against. The second o f these types of influ­ en ce cove rs a great deal, and the i mportance of the social con trib ution is not always easy to establish. Nevertheless i ts effec ts typically include the loss of such goods as sel f-esteem, se lf-c onfidence, motivation, and a mbition - all o f which contri­ bu te to competitive success and none of which is easi ly restored by special training p rogra ms. Even if social inj ustice has pro­ duced such effects, it may be difficult for society to eradica te them . This type of j ustification for compensatory p rogra ms raises another question. I f i t depends on the claim that the disadvan­ tages being compensated for are the product of social injus tice, then it becomes i mportant how great the contribution of social injustice actually is, and to what extent the situation is due to social causes not involving inj u s tice, or to causes that a re not social, but biological. I f one believes that society's responsibility for compensatory measures extends only to those disadvan tages caused b y social inj us tice, one will assign political impo rtance to the degree, if any, to which racial differences in a verage I . Q. are genetically influenced, or the innate contribution, i f any, to the statistical differences in emotional or intellectual characteristics between men and women. A lso, i f one believes that a mong socially produced inequalities, there is a crucial d istinction between those tha t were produced u njustly and those that are merely the i ncidental results of just social arrangements, then it will be very i mportant to decide exactly where that line falls : whether, for example, certain intentions must be referred to in arguing that a d isadvantage has been unjustly i mposed and therefore merits compensation. But let me p u t those issues aside for the mo ment. The fou rth s tage comes when it is acknowledged that some unjustly caused disadvantages that create difficulties of access to positions formally open to all cannot be overcome b y special programs o f preparatory or remedial training. One is then faced with two alternatives. One can permit the effects of social injus tice to confer a disadvantage i n access to desirable positions

94

Mortal questiotls

filled simply on the basis of qualifications relevant to per for­ mance. Or one can institute a system of p referential selectio n that will facili tate access for those whose qualifications are lo wer ad east partly because of unj u s t discri mina tion in other situations and at other times (and possibly agains t other persons). This is a difficult choi ce, and ideally i t would be far better to use a more direct method of rectification, than to balance inequality in o ne part of the social s ystem by introducing a reverse inequality at a different point. I f the society as a whole contains serious injustices with complex effects, there is no way for a sing le institu tion within that society to adj u s t i ts criteria for competi­ tive a d mission or employ ment so that the effects of inj us tice are nullified as far as that i nsti tution is concerned. That gives appeal to the position that places should be filled solely by criteria relevant to performance, and if this tends to amplify or extend the effects of inequitable treatment elsewhere, the remedy must be found in a more direct attack on those differences in quali fication, rather than in the introdu ction of irrelevant criteria " of appointment or ad mission which will sacrifice efficiency, productivity, or effectiveness of the institution in its specific tasks. A t this fou rth stage we therefore find a broad division of opinion. Some believe that nothing fur ther can legitimately be done in the short run, once the remediable unjust inequalities of opportunity between individuals have been dealt with: the irremediable ones are unj ust, but fu rther s teps to counter­ balance them by reverse discrimination would also be unjust, because they must employ irrelevant cri teria. On the other hand, some find it unacceptable in such circu mstances to s tay with the cri teria usually related to successfu l performance, and believe that differential ad mission or hiring s tandards for worse-off groups are justi fied because they compensate in some approxi­ mate way for the i nequalities o f opportunity produced by past inj us tice. But at this point there is also a temptation to resolve the dilemma and strengthen the argu ment for preferential s tandards by proceeding to a fifth stage. One may reflect that if the criteria relevant to the prediction of performance are not inviolable it may not matter whether one violates them to compensate for disadvantages caused by i njustice or disadvantages cause d in

The policy of prefe reuce

95

other ways. One does not have to settle the q uestion of the deg ree to which racial or sexual discrepancies are socially p ro d uced, because the differentials in reward ordinarily corre­ la ted with differences in qualifications are not the result of n at ural justice. They simply arise in a competitive system when em plo yers try to fill positions and perform tasks efficiently. Ce rtain abilities may be relevant to fil ling a job fro m the point of view of efficiency, but they are not relevan t fro m the poi nt of v ie w of j u s tice, because they provide no ind ication that one des er ves the rewa rds that go with the job. The qual ities, experience, and attain ments that make success in a certain p osi tion l ikely do not in themselves merit the rewa rds that hap p en to a ttach to occupancy of that position i n a co m petitive eco nom y. Consequently i t might be concluded that i f women or black people are less qualified, for whatever reason, i n the respects that lead to success in the professions that our society rewa rds most highly; then i t would be just to compensate for this disad van­ tage, within the limits permitted by efficiency, by having suitably differen t standards for these groups, and thus b ringing their access to · desirable positions more into line with that of others. Preferenti al trea tment would not, on this view, have to be tailored to deal only with the effects of past injustice. But it is clea r that this is not a s table position. For i f we abandon the condition that to qualify for compensation an inequity must be socially caused, then there is no reason to restrict the compensatory measu res to well-defi ned racial or sexual grou ps. Co mpensatory selection procedu res would have to be applied on an individual basis, within as well as between such groups - each person, regardless of race, sex, or quali fica­ tions, being gran ted equal access to the desirable positions, within l i mits set by efficiency. This might requi re randomiza­ tion of law and medical school ad missions, for exa mple, from among all the candidates above some mini m u m standard enabl­ ing them to do the work. I f we were to act on the princi ple that different abilities do not merit different rewa rds, i t would result in much more equality than is dema nded by proponents of preferen tial treatment, There is no likelihood that such a radical cou rse will · be ad opted in the U n i ted Sta tes, but the fact that it seems to follow

96

Mo rtal

questions

naturally fro m a certain view about how to deal with racial or sexual inj ustice reveals something i mportant. When we try to deal with the inequality i n advan tages tha t results fro m a disparity in qualifications (however produced) between race s or sexes, we are up against a featu re o f the system which a t every turn exacts costs and p resents obs tacles in response to attempts to reduce the inequalities. We must face the possibility that the primary injustice with which we have to contend lies in th is fea ture i tsel f, and that some of the worst aspects of what we now perceive as racial or sexual i njustice are merel y conspicuous manifestations of the great social injusti ce of d i fferential reward. II I f differences in the capacities that any society rewards a rc visi bly correlated, for whatever reason, with other traits such as race or rel igion or social origin, then a system of liberal equality o f opportunity will give the appearance of supporting racial or religious or class i nj ustice. Where there is no such correla tion, there can be the appeara nce ofjustice through equal opportunity. But there is injustice in both cases, and i t l ies in the schedule of rewards. The l iberal idea of equal treatment demands that people receive equal opportunities if they are equally qualified by talent or education to use those opportunities. In requiring the relativ­ ization of equal treatment to characteristics in which people are very u nequal, it guarantees that the social order will reflect and probably magnify the initial distinctions produced by nature and the past. Li beralism has therefore come under increasing attack in recent years, on the ground that the familiar p rinciple of equal treat ment, with its meritocratic concep tion of relevant differ­ ences, seems too weak to combat the inequalities dispensed by nature and the ordinary workings of the social s ystem. This critici s m of the view that people deserve the rewards that accrue to them as a result of their natural talents is not based o n the idea that apart fro m social institutions no one can be said to deserve anything.2 For if no one deserves anything, then no 2 Rawls appears to regard this as the basis of his own view. He believes it

makes sense to speak of positive desert only in the context of distributions by a just system, a nd not as a pre-institutional conception that can be used to measure the justice of the system. John Rawls, A Theory ofjustice (Cambridge, Mass: Ha rvard University Press, 1 97 1 ), pp. 3 1 0-13.

1

ne policy of p reference

97

ine qu alities a re contrary to desert, and desert provides no a rgu m ent for equality. But for many benefits and disadva ntages, ce rt ain characteristics of the recipient a re relevant to what he des er ves. If people are equal in the relevant respects, that by i tsel f cons tit utes a reason to distribute the benefit to them equally.3 Th e relevant featu res will vary with the benefit or disadvan­ ta ge, a nd so will the weight of the resulting considerations of desert. Desert may so metimes, in fact, be a rather unimportant consideration in determining what ought to be done. But I do wish to claim, with reference to a central case, that differential abilities are not usually a mong the characteristics that determine whether people deserve econo mic and social benefits (though of course they determine whether people get such benefits). In fact, nearly all characteristics are i rrelevant to what people deserve in this dimension, and most people therefore deserve to be treated equally.4 Perhaps voluntary differences in effort or moral differ­ ences in conduct have so me bea ring on economic and social desert. But they a re featu res in which most people do not differ enough to justify very wide differences in reward. 5 I shall not try to defend these clai ms here, or the legitimacy o f the notion of desert itsel( If these things make no sense, neither docs the rest ·of the argument. A decision that people are equally or unequally deserving in some respect is not the end of the story. First of all, desert can be overridden, for exa mple by liberty o r even by efficiency. In some cases the presu mption of equality is rather weak, and not much is requi red to depart from it. This will be so i f the in terest 3

4

5

Essentially this view is put forward by Bernard Williams in ' The Idea of Equality', in PhiiMoph y , Politics , a u d Society (second series), ed. P. Laslett and W. G. Runciman (Oxford: Blackwell, 1 964), pp. 1 1 0-1 3 1 . This is distinct from a case in which nothing is relevant because there is no desert in the mauer. In that case the fact that people d iffered in no relevant characteristics would not create a presumption that they be treated equally. It would leave the determination of their trea tment entirel y to other considerations. I t is uot m y view that we cannot be said to deserve the remits of anything which we do not deserve. I t is true that a person does not deserve his intelligence, and I have maintained that he does not deserve the rewards that superior intelligence can provide. But neither does he deserve his bad moral character or his above-average willingness to work, yet he probably does deserve the punishments or rewa rds that flow from those qualities. For an illuminating discussion of these matters, see Robert Nozick, A uarch y , State, aud Utopia (New York: Basic Books, 1 974), ch. 7.

98

Mortal

questio11s

in question is minor or temporally circu mscribed, and does not represent an i mportant value in the subject's life. Second, it may be that although an inequality is con trary to desert, no one can benefit from i ts remova l : all that ca n be don e is to worsen the position o f those who benefit u ndeservedly from its p resence. Even if one believes that desert is a very important factor · i n determining just distributions, one need not obj ect to ineq ualities that a re to no one's disadvantage. In other words, it is possible to accep t something like Rawls' Difference Principle from the standpoint of an egalitarian view of desert.6 (I say it is possible; It may not be req uired. Some may rej ect the Difference Princi p le beca use they regard equality of treatment as a more stringent requirement. ) Third (and most significantly for the present discussion), a determination of relative desert in the distribution of a particular advantage does not even settle the question of desert in eve r y case, for there may be other advantages and disadvantages whose distribution is tied to that of the first, and the traits relevant to the deter mination of desert are not necessarily the sa me from one advantage to another. This bears on the case under consideration. I have said that people with different talents do not thereby deserve different economic and social rewards. They may, however, deserve different opportuni ties to exercise and develop those talents. 7 Whenever the distribution of two different types of benefit is connected in this way, through social or econo mic mechanisms or through natural human reactions, it may be impossible to avoid a distribu tion contrary to the conditions of desert in respect of at least one of the benefits. There will be a dilemma in which i nj ustice cannot be entirely avoided. It may then be necessary to assign justice in the distribution of one advantage priority over justice in the distribution of a nother that automati­ cally goes with it. I n the case under discussion, there appears to be a conflict between justice in the distribution of educational and p rofes­ sional opportunities and j ustice in the distribu tion of economic 6

7

Rawls, Theory ofjustice, pp. 75-80. Either because differences of ability are relevam to degree of desert in these respects or because people are equally deserving of opportunities proportional to thei r talents. More likely the latter.

The policy of prefermce

99

an d so cial rewards. Th"ere is a presu mption, based on something more than efficiency, i n favor o f giving equal opportunities to th o se equally likely to su cceed. B u t if the p resu mption in favor of e conomic equality is considerably stronger; the j usti fi cation fo r departing from i t must be stronger too. So when ' educa­ tional' justice and economic j ustice come into conflict, it will sometimes be necessary to sacrifice the former to the latter. III In thinking about racial and sexual discri mination, the view that economic j ustice has priority may tempt some to favor ad mis"'­ sion quotas p roportional to the representation of a given group in the population. Whatever explains the small n u m ber of women or blacks in the professions, i t has the result that they have less of the financial and social benefits that a ccrue to members of the professions, and what a ccou n ts for those differences cannot j ustify them. So jus tice may seem to require that more women and blacks be admi tted to the professions. The trouble with this solution is that i t does not locate the injustice accurately, but merely tries to correct the racially. or sexually skewed economic distribution which is one of its more conspicuous symptoms. We are enabled to perceive the situation as unjust when we see i t through its racial manifestations, because race is a subj ect by now associa ted i n our minds with injustice. But little is gained by merely transferri ng the sa me system of differential rewa rds, suitably adjusted to achieve comparable proportions, to the class o f blacks o r the class of women. I f i t is unj ust to rewa rd people differentially for what certain qualities enable them to do, i t is equally u njust whether the distinction is made between a white man and a black man or between two black men, or two white women, or two black women. There is no way o f attacking the u njust reward schedules of a meritocratic s ystem (if i ndeed they a re u nj ust) by attacking their racial or sexual manifestations directly. In most societies reward is a function o f demand, and many of the human traits most i n demand result largely from gifts or talents. The greatest i njustice i n our society, I believe, ·is neither racial nor sexual but i ntellectual. I do not mean that i t is u njust that some people are more intelligent than others; Nor do I mean that- society rewards people differentially si mply o n the basis of

100

Mortal questio11s

their intelligence: usually it does not. But i t p rovides on the average much larger rewards for tasks that require supe rior intelligence than for those that do not. This is how things work out in a technologically advanced society with a mark et econo my. I t does not reflect a social judg ment that smart people deserve the opportunity to make more money than dumb people. They may d eserve richer educational opportunity, but they do not therefore deserve the material wealth that goes with i t. The same could be said about society's differential reward of beau ty, a thletic ability, musicali ty, etc. B u t in telligence and its develop­ ment by education provide a particularly significant and perv a­ sive exa mple. A general reform of the current schedule of rewards, even if they are unjust, is beyond the power of individual educational o r business institu tions, working through their admissions or appointments policies. A competitive econo my is bound to reward those with superior training and abilities : the refusal to do so will put any busi ness enterprise in a poor competitive position. And those who succeeed in medical school or law school will tend to earn more than those who do not - whatever criteria of adm ission the schools adopt. It is not the p rocedures of appointment or ad missi on, based on criteria that predict su ccess, that are unjust, but rather what happens as a res u l t of success. No completely just solution is a vailable. If different factors determine what is deserved in the distribution of different benefits and disadvantages, and if the distribution of several distinct advantages is someti mes connected even though the relevan t factors are not, then inevitably there will be injusti ce in some respect, and it may be i mpossible to substitute a principle of distribution that avoids it. Justice may require that we try to red uce the a u to matic connections between material advantages, cultural opportunity, and institutional authority. B u t such changes can be brought about, i f at all, only by large alterations i n the social system, the system of taxation, and the salary structure. They will not be achieved by modifying the admissions or hiring policies of colleges and universities, or even banks, law firms, and busines­ ses. Compensatory measu res i n ad mission or appointment can be

The policy of p reference

101

d efen ded on grounds of j ustice only t o the extent that they co m pensate for specific d isadvantages that have themselves been unj ustl y caused, by factors distinct fro m the general meritocratic ch ar a cter of the system of distribution. Such contributions arc d iffic ult to veri fy or es timate; they probably vary a mong in di viduals in the oppressed group. And even where a j ustifica­ tio n for p referential treatment exists, i t may not be strong en o ugh to create an obligation, since it is doubtful that one ele m ent of a pluralistic society is obliged to adopt discri minatory measures to coun teract i nj ustice due to a no ther element, or even to the society as a whole. IV

These considerations suggest that a n argument on grounds of j ustice for the i m position of racial or sexual quotas would be difficult to construct wi thou t rather p recise assumptions about the sou rce of unequal qualifications between members of differ­ ent groups. The more speculative the ass u mp tions, the weaker the argu ment. But the issue is different if we return to the q uestion posed at the beginning o f this essay. The q uestion was not whether preferential trea tment is required by justice, but whether i t is compatible with j ustice. To that question we can give a different answer. I f the reflections about differential reward to which we have been led are correct, then p referential treatment need not be seriously unj ust, and it may be warranted not b y justice but b y considerations o f social utility. I s a y n o t serio u s ly u nj ust, to acknowledge that a depa rtu re from the standards relevant to distribution of i n tellectual opportunities p e r se is i tself a kind o f injustice. B u t i t s seriousness is lessened because the factors relevan t to the distribution o f i ntellectual opportunity are irrelevant to the distri bu tion of the ma terial benefits that go with it. Where the allocation of one benefit on relevant grounds carries with it the allocation o f o ther, more significant benefits to which those grounds a re irrelevant, the departu re fro m those grou nds need not be a serious o ffense against j ustice. This may be so for two reasons. First, the presu mption o f eq ual treatment of releva ntly equal persons in respect of the first benefit may not be very strong to begin with. Second, the fai rness of abiding by

1 02

Mortal questioiiS

· that presumption may be overshadowed by the unfai rness of th e other distribution correlated with it. Consequently, it may be acceptable to depart fro m the ' relevant' grounds for .undra matic reasons of social utility, o r to serve legiti mate institutional aims, which would not j ustify tnore flagrant and undiluted examples of unfai rness. Naturally a deviation from the usual method will appear unjust to those who a re accustomed to regarding abil ity to s ucceed as the correct cri terion, but this appearance may be an illusion. That depends on how m uch injus tice is i nvolved in the usual method, and whether the reasons for departing fro m it are good enough, even though they do not correct the injustice. The problem, of course, is to say what a good reason is. I do not want to p rod uce an argument that will justify ordinary racial or sexual discri m i na tion designed to p reserve i nternal harmony in a business, for insta nce. Even someone who thought that" the system of differen tial economic rewards fo r differen t abilities was unj us t would presu mably regard it as an additional injustice if sta ndard racial, religious, or sexual d iscri m i nation were a factor in the assignment of individ uals to highly rewarded positi ons. I can offer only a partial accou nt of what makes systematic racial or sexual discri mination so exceptionally unj ust. It has no social advantages, and it attaches a sense o f reduced worth to a feature with which people arc born . B A psychological consequ­ ence of the systematic attachment of social disadvantages to an inborn feature is that both the possessors of the feature · and others begin to regard it as -a n essential and i mportant charac­ teristic, one tha t reduces the esteem in which its possessor can he held.9 Concomitantly, those ·withou t the characteristic gain free esteem by comparison, and the a rrangement thus sacrifices the most basic personal interes ts of some for the interests of others, 8

9

For a detailed and penetrating treatment of this and a nu mber of other matters discussed here, see Owen M. Fiss, 'A Theory of Fair Employment Laws', Uuivmity of Chicago La�IJ Review, X X X VIII (Winter, 1 97 1 ), 235-314. This effect would not be produced by an idiosyncratic discriminatory practice limited to a few eccentrics. If some people decided they would have nothing to do· with anyone left-handed, everyone else, including the left-handed, would regard it as a silly objection to an i nessential fea ture. B u t if everyone shunned the left-handed, left-handedness would become a strong component of their sel f-image, and those discri minated against would feel they were being despised for their essence. What people regard as thei r essence is not independent of what they get admired and despised for.

The policy of preference

1 03

with those sacrificed being on the bottom. (Because si milar th in gs can be said about the social and econo mic disadvantages

atta chi ng to l o w intelligence, that, too, is a maj o r inj u s tice. ) Rev erse discri mina tion need not have these consequences, and

it can have soci a l advantages. Suppose, for exa mple, that a substantial increase in the nu mber of black doctors is desirable because the health needs of the black com munity are unlikely to be met otherwise. And suppose that, a t the present average level of premedical q u a l ifications a mong black a p plicants, i t would req uire a huge expansion o f total medical school enro l l ment to supply the desi rable absolute nu mber of black doctors without adopting differen tial a d mission standards. S u ch a n expansion may be unacceptable either because of i ts cos t o r because i t would produce a t o t a l supply o f doctors, black and whi te, m u ch grea ter than the s ociety requires. This is a s t rong argu ment for preferential ad missi ons,

not on grounds of j u s tice but on

grounds of social u tility. (In addi tion, there is the salutary effect on the aspira tions and expecta tions of other blacks, from the visibil ity of exemplars in formerly inaccessible position s . ) The argu men t in the other direction, fro m the point of view o f q ualified w h i t e applicants who a r e tu rned away, is n o t nea rly as strong a s the a r g u ment against standard racial discri mination. The self-es tee m of whi tes as a g roup is n o t enda ngered b y such a practice, since the situa tion a rises only beca use of thei r general social dominance, and the aim of the p ractice is only to benefit blacks, not to exclude whi tes. Moreover, a l though the in terests o f some a re being sa crificed to the in teres ts of others, it is the better placed who a re being sacrificed and the worst pl aced who are being helped . 1 0 The policy i s designed to fa vor a group whose social position is exceptionally depressed, with destruc­ tive consequences both for the self-esteem of members of the group and fo r the health and cohesion o f the society. t t So, i f a preferential ad mission o r a p p o i n t ment policy is adopted to mi tigate a grave social evil, and it favors a group in a particularly unfortu n a te social position, and i f fo r these reasons it 10 11

T his is a preferable di rection of sacri fi ce if one accepts Rawls' egalitarian assumptions about distributive justice. Rawls' Throry ofJustice, pp. 1 00-3. It is therefore not, as some have feared, the first step towa rd a n i mposition of minimal or ma ximal q uotas for all racial, religious, and ethnic subgroups of the society.

Mortal questions

1 04

diverges fro m a meritocratic s ystem that i s not i tself req u i red b y j u stice, then the prefe rential p ra ctice is proba b l y not u nj ust. 1 2 I t is not without costs, however. N o t only does it inevita bly produce resentment in the better q u a l ified who a re passed over beca use of the policy, but i t also a llows those who would in any case have failed to gain a desi red position on the basis of

their

quali fications to feel that they m a y have lost out to someone less q u a l ified because of the preferential policy. S i milarly, such

a

practice cannot do much fo r the sel f-esteem of those who know they have benefited fro m i t, and i t may threaten the self-esteem of those i n the fa vored group who would in fac t have gained thei r positions even in the a bsence of the d iscrimina tory policy, b u t who cannot be su re that they a re not a mong its beneficia ries. This has led some institu tions to lie about thei r policies, or to

hide them behind clouds of obscurantist rhetoric a b o u t the

discri m i natory cha racter of s tandard a d missions criteria. Such con cea l ment is possi ble a n d even j u s ti fied up to a point, but the costs cannot be entirely evaded, and p referential policies will

be

tolerable only so long as they contribute to the eradication of great social evils.

v When racial and sexual i nj us tice have been red u ced, we s h a l l still be left with the great i nj ustice o f the s ma r t and the dumb, who

are so d i fferently rewa rded for comparable effo rt. This would be an i njustice even if the system of differential econo mic and socia l

rewa rds had no sy stematic sexual or racial reflection. And if the social esteem a n d econ o m i c advantages attaching to d ifferent occu p a tions a n d educational a chieve ments were much more uniform, there would be little cause for concern about racial, ethnic, or sexual pa tterns i n education o r work. A t present we have no way of d ivorcing professional s t a tus fro m social esteem a n d economic reward, at least not without a gigantic i ncrease in total social control, o n the Chinese model. Perhaps so meone wi ll di s cover a way to redu ce the socially produced inequalities 1 2 Adam Morton has suggested an interesting alternative, which I shall n ot

try to develop: na mely, that the p ractice is justified not by social utility, but because it will contribute to a more just situation in the future. The practice considered in i tself may be u nj ust, but it is warranted by its greater contribution to j u s tice over the long term, through eradication of a sel f-perpetuating pattern.

The policy of preference (e sp ecially the un i ntelligent, b ea utifu l and op p ortu ni ties,

1 05

economic ones) between the i n telligent and the the talented and the untalented, or even the the ugly, without limi ting the availability of products and services, and without resort to

in cr eased coercion or decreased liberty in the choice o f work o r sty l e of life. B u t i n t h e a bsence of such a u topian solution, the fa m iliar task o f balancing liberty against equality will re main with us.

8 E quality

I t is difficult to argue for the in trinsic social value of equality withou t begging the question. Equality can be defended u p to

a

point in terms of other values like utility and liberty. B u t some of the most difficult questions .a re posed when it conflicts with these. Contemporary political debate recognizes fou r types of equal­ i t y : political, l eg � I. s ocial, and economic. The first three cannot be defined in formal terms. Political equality is n o t guara nteed by g ranting e a ch a d u l t one vote and the right to hold public offi ce. Legal equality is not guaranteed by granting everyone the right to a j u r y trial, the right to bring suit for inj u ries, and the right to counsel. S ocial equality is not p rodu ced by the a bolition o f titles and official barriers to class mobility. Great substantive inequalities i n political power, legal p rotection, social esteem and self-respect are compatible with these formal conditions. I t is a commonplace that real equality of every kind is sensitive to econo mic factors. While formal institu tions may guarantee

a

minimu m social status to everyone, big differences in wealth and income will p roduce big distinctions above tha t - distinctions that may be inheri ted as well. So the question o f economic equality cannot be detached from the others, and this complica tes the issue, because the value of the other types of equality may be of a very different kind. To p u t i t somewhat p aradoxically, their value may not be s trictly egalitarian. It m a y depend on certain rights, like the right to fair t rea tment b y the law, that must be i mpartially p rotected, and that cannot be p rotected without a measure of substantive

Equ a l ity

1 07

eq u ali ty Rights a re in an extended sense e ga l i t aria n because everyone is s upposed to h a v e them; bu t this is not a mat ter of .

,

dis t ri b u t ive j ustice. The equal protection of individual rights is u sua lly thought to b e a value i ndependent of utility and of e qua lity· in the distribution of advantages. Later I shall com ment on the rel a tion a m on g these va lu e s but fo r now· let us assume the i r distinctness . This means that the defense of e c o no m i c e q u a l i t y on t h e g r o u n d t h a t i t is neede d t o prote c t p o l i tica l , legal, a n d social equality m a y not be a defense of e q ual i t y per se e qu a l i ty in the possession of benefits in g e n e r al Yet the l a tter is a further m o ra l i dea of great i m portance. Its valid ity would pro vide an i n dep e n d e n t reason to favor econo mic equality as a good in its o wn right. If, per imposs ibile, l a rge econ o m i c ine­ qual ities did not thr e a te n politi cal, legal, and social eq u a l i ty , they wou l d · be m u ch l ess obj ectionable. But there might s ti l l be something wrong wi t h them. I n add ition to the argu ments tha t depend on its rela tion to other types o f equality, there is a t least one nonegalitarian, instrumental argument fo r ec on o m i c e q u a l i t y i tsel f, on grounds of util ity. The p r incip l e of d i minishing ma r gi n a l u tility states . that for m a n y goods, a particular fu rther increment has less value to some o n e who al read y p o ss e s s e s a signi ficant a mou n t of the good than to so m e one who has . less. I So i f the total quantity of such a g o o d and t h e nu mber of reci pients r e m a i ns constant, an equal d i s t r i b u tion of it will a l w a y s have g re a t e r total u ti l i t y than a less equal one. This mus t be b a l a n c e d against certai n costs. First, a t t e m p ts to reduce ine q u a l i ty may also reduce the t ot a l q u a n ti ty of goods ,

-

.

available, by a ffecting i ncenti ves to work and invest. For exa m­ ple, a progressive i nc o m e tax and di minishing ma rginal u t i l i ty make it

more expensive t o . p u rchase the labor of those whose a certain point, the p u r s u it of eq ua l i t y may sacri fi ce overal l u ti l i t y or even the welfare o f everyone in the society. Second, the pro mo ti o n of equality m a y r e q u i re obj ectiona ble means. T o achieve e v en moderate e q ua l i t y it i s necessary to restrict economic l i b e r t y including the free d o m to make beq u ests. Grea ter equalit y may be a ttaina b l e o n l y b y m ore

services a r e most i n demand. B e y o n d

,

,

I This

is obviousl y not true of things in w h i ch i n terest varies greatl y , l i k e

recordings of b i r d songs, o r h o r r o r c o m i c boo k s .

Mortal q u estio11s

1 08

general coerci ve techniq ues, including u l ti m a tely the assignment of work by public ad ministration instead o f private contracts. Some of these costs may be unacceptable not only on utilitar ian grounds but because they violate i n dividual righ ts. Opponen ts of the goal of e q u a l i ty may a rgue that if an unequal distribu tion o f benefits results fro m the free interactions and a g reements of persons who do not violate each other's rights, then the results are not obj ectionable, provided they do not include extreme hardship for the worst off. II S o there is much to be said abou t the instru mental value and disvalue of eq uality; the question of its intrinsic value d oes not a rise in isolation. Yet the answer to that question determines what instru mental costs a re acceptable. If equality is in i tself good, then p ro d u cing it may be worth a certain a mount of inefficiency a n d loss of li berty. There are two types of a rgu ment for the i n trinsic value of equality, com munitarian and individua listic. A ccording to the c o m m u n i tarian a rgu ment, equality is good for a society taken as a whole. It is a con d i tion of the right kind of relations a mong its members, and of the for m a tion in them of healthy fra terna l attitu d es, desi res, and s y mp a thies. This view anal yzes the value of equality i n terms of a social and individual ideal. The individualistic view, on the other hand, defends equality as a correct

distributive principle - the correct way to meet tht:

conflicting needs and i nterests of distinct people, whatever those interests may be, more or less. It does not assume the des i rability of any particul a r kinds of desi res, or any particular kinds of interpersonal relations. Ra ther it favors equality in the dis tribu­ tion of h u m a n goods, wha tever these may be - whether or not they necessarily include goods of community and fraternity.

Though the co m mu n i tarian a rgu ment is very influential, I am going to explore only the individualistic one, because that is thr type of argument that I think is more likely to su cceed. It would p rovide a moral basis for the kind o f liberal egalitariani s m that seems t o me pla usible. I do not have such an argument. This essay is a discussion o f the for m such an argument would have to take, what its s t a rting points should be, and what it mus t overcome.

Equality

A

1 09

p reference for equality is at bes t one component in a theory

of socia l choice, or choice involving numbers o f people. I ts defe nse d oes not req u i re the rejection of other values wi th w h i ch it m ay come into conflict. However, i t is excluded by theories of so cial choice w h i c h make certain o t h e r values d o m i nant. E g alitarianism m a y once have been opposed to aris tocratic theories, but now i t is opposed i n theoretical debate b y the adherents o f two nona ristocratic values: utility and i n d i vidual rights. I a m going to examine the dispute i n order to see how eq uality might be shown to have a value tha t can resist these to some extent, without replacing them. Though I a m i nteres ted i n the most general fou n d a tion for such a princi ple, I shall begin by discussing a m o re specialized egalitarian view, the position of John Rawls.2 I t applies s peci fi­ cal l y to the design of the basic social i nstitu tions, rather than to distribu tive choi ces, and perhaps it cannot be extended to other cases. But it is the most developed l i beral egal itarian view i n the fiel d, a n d much debate about equality fo cuses o n i t . So I will initially pose the oppos i tion between equality, utility, a n d rights in terms of his position. La ter I shall explain how my own egalitarian view d i ffers fro m his. Rawls' theory assigns more i mporta nce to equal protection of pol itical a n d personal liberties than to equality i n the distribution of other benefits. Nevertheless i t is s trongly egalitarian in this respect also. His principle o f distribution for general goods, once equality i n the basic l iberties is secu re, is t h a t inequalities are justified only i f they benefit the worst-off group i n the society (by yielding higher productivity and employ ment, for exa mple) . This so-cal led D i fference Principle is used not to dete r m i ne alloca tion d i rectly, b u t only for the assess ment of econ o m i c and social institutions, which i n turn i n fluence the allocation of goods . While i t i s cou n ted

a

g o o d t h i n g fo r a n yone to be made

better o ff, the value of i mproving the si tuation o f those who are

worse o ff ta kes priority over the value o f i m p ro v i n g the situation o f those who a re better o ff This is largely i n d e p e n d e n t o f the relative q u a n tities o f i m p ro ve ment i n volved, a n d also of the relative n u m bers o f person s . S o given a choice between

making a thousand poor people somewhat better off and making 2 John R a w l s .

Press,

1 97 1 )

.

A

Tluwy of}listie£"

(Ca m b r i d g e, M a ss . :

Ha rvard U n i v e r s i t y

1 10

Mortal

q u estions

two thousand middle class people considerably better off, the first choice would be p referred. I t should be added that people 's welfare for these p u rposes i s assessed i n terms of overall life p rospects, not j u s t prosperity at the m o ment. This is a very s t rong egalita rian principle, though i t is not the most radical we can i magi ne. It is constru cted by adding to the general value of i mprovement a condition of priority to the worst of( A m o re egalitarian position would hold that so me inequalities are bad even i f they benefi t the worst off, so that a situation in which

every one is worse off m a y be p referable i f the

inequali ties a re reduced enough.

So long as

the a rg u ment

remains individ ua l istic s u ch a position could seem attra ctive only

for

reasons

stemming

fro m

the connection

between

econo m i c a n d social equality.3 Later I shall discuss Rawls' argu m ents for the view, and offer some additional ones, b u t first let me say something ab ou t the two positions to which it is naturally opposed, and against which it has to be defended. They a re positions that do not a ccord intrinsic value to equality b u t a d m i t other val ues whose pursuit or protection m a y requ i re the a cceptance of considerable i nequality. Those values, as I have said, are utility and indi vidual rights . Fro m a u tilitarian point o f view, i t does n o t m a k e sense to forego g reater benefits for the sake of lesser, or benefits to more people for the sake of fewer, j u s t because the benefits to the worst off will be g reater. It is better to have m o re of what is good and less of what is bad, no m a tter how they a re distri b u ted. According to· a theory of individual rights, it is wrong to i nterfere with people's liberty to keep or bequeath what they can earn merel y in order to p revent the development of inequalities i n distribution. It may be a cceptable to limit individual liberty to p revent grave evils, b u t inequality is not one of those. Ine­ qualities are n o t wrong if they do not resul t from wrongs of one person against another. They must be accepted if the only way 3

The argument would be that i mprovements in the wel l-being of the lowest class as a result of material productivity spurred by wage differentials are only apparent: da mage to their self-respect outweighs the m aterial gains. And even inequalities that genuinely benefit the worst off may destroy nondistributive values like communit y or fraternity. Sec Christopher Ake, 'Justice as Equa)ity', Philosophy & Public Affairs, v, no. I (Fall, 1 975), 69--89, esp. 76--7 .

Equ a l it y

111

to p revent them is to abridge individual rights to the kind of free act ion tha t violates no one else's rig h ts. Both types of theory point ou t the costs o f p u rsuing distribu­ ti ve equa l i ty, and deny that it has ind epen dent value that o u tweighs these costs. More speci fically, the pursuit o f eq uality is held to req uire t he illegiti mate s a c r i fi c e of the rights or interests of s o m e i nd ivid uals to the less i m portant i n teres ts of others . These two theories are also radica l l y opposed to one an other. Together with eg a l it a r i a ni s m they fo r m a trio of fundamental ly di fferen t vi ews about how to settle confli cts am ong the i n te rests of d i fferent people.

III W h a t is t h e na ture of the d is p u te between th e m ? The u n i ts about w h i ch the problem arises are i n d i vi d u a l persons, i n d ividual hu man l i ves. Each of them has a c l a i m to considera tion. I n some s e n s e the d i s tinctness of these c l a i m s is a t the hea rt of the issue. The q u e s t i o n is whether (a) the worst off have a p rior claim, or (b) the enforce ment of that claim would ignore the greater cla i m of others n o t a mong t h e worst off, who would benefit signific­ antly mor·e i f a l ess egal itarian policy were adopted ins tea d, or (c) it would i n fringe the cl a i m s of other persons to l i berty and the protection of their rights. Now this looks l i ke a dispu te about the value o f eq uality. B u t it can a l so be viewed as a d i spute a b o u t h o w people should b e trea ted e q u a l l y , n o t about whether t h e y s h o u l d be. The th ree views s h a re an assum ption of moral equal i t y between persons, but differ in their interpretations of it. They ag ree that the mo ral claims of all persons a re, a t a sufficient l y abstract level, the sa me, b ut disagree over what t h ese a re. 4 The defender of ri g h ts locates them in the fr e e do m to do certain things without direct i n terference b y others. The utilita rian loca tes them in the requ i rement that each person's 4 T h i s way o f look i n g a t t h e problem w a s s uggested t o me by a proposal o f R a w l s (personal co m m u n ication,

January

31,

1 976) :

S u ppose we distinguish bet ween the eq u a l trea t me n t of persons and t he i r (eq u a l ) r i g h t to be t r e a t e d a s equals. (Here persons a re JIIOI'ill persons. ) The latter is more basic: S u p pose the O r i gi n a l Pos i t i o n represents t he l atter re moral p e r so n s when t h e y agree on p r i n ciples a n d s u ppose t h e y wou ld a g r e e on s o m e for m of e q u a l treatment. W h a t more is needed ?

1 12

Mortal questions

i n terests be ful l y coun ted as a com ponent in the calculation of utility u sed to decide which states of a ffa i rs a re best and whi ch acts or pol icies a re right. The egalitarian finds them in an eq ua l cl ai m to a c t u a l o r possible ad vantages. T h e i s s u e remains acute even though most social theories d o not fal l s q u a rely i n to one of these categories, b u t give primacy to one in terpretation o f moral

equality and secondary status to the others. All th ree interpretations o f moral eq uality a t tem pt to give equal weight, in essen tial respects, to each person's point of view. This might even be described as the mark of an enl igh­ tened ethic, though some theories that do not s h a re i t still qualify as ethical. If the opposition of views about distribu tive equality can be regarded as a disagreement about the proper i n terp reta­ tion of this basic requ i re ment of moral equality, that p rovides a c o m m o n reference against which the opposing positions may be measured. I t should be possible to compare the quality of their

justifica tions, instead of s i m p l y regis tering their mu tual incom­ patibility. What i t means to give equal weight to each person 's point of view depends on what is morally essen tial to that point of view, what i t is in each of us that must be gi ven equal weight. It also depends on how the weigh ts a re combined. And these two aspects of the answer are interdependent. Let us consider each of the positions fro m this point of view. IV The m o ral equality o f u tili tarianis m is a kind of majority rule : each perso n 's interests count once, b u t some may be outweighed by o thers. I t is not rea l l y a maj o rity of persons that deter mines the res ult, b u t a m aj o rity of in terests suitably weigh ted for in tensity. Persons are equal in the sense tha t each of them is gi ven a ' vote' · weighted in propo rtion to the magnitude of his in terests. Although this

means

that the i nterests of a minority can

so metimes ou tweigh the interests of a majority, the basic idea is maj o ritarian because each individual is a ccorded the same (vari­ a ble) weight and the ou tco me is determined by the largest total. In the sim plest version, all o f a perso n ' s interests o r prefer­ ences are counted, and gi ven a rela tive wei g h t depending on their weight fo r h i m . B u t various modifica tions have been s u g gested. One doubt voiced about u ti l i ta rianis m is that it

Equality

1 13

cou nts pos i t i vely the satisfaction of evil desi res (sadistic or bigoted ones, for example). Mill employed a d istinction between higher and lower pleasu res .. and gave priority to the for mer. (Coul d there be a corresponding distinction for pains?) Recently, Tho mas Scanlon has a rgued tha t a n y distribu tive principle, utili tarian o r egalitarian, must use some obj ective s tandard of interest, need , o r u rgency d i s t i n c t fro m m e r e s u bj ective p refer­ ence to avoid u n a ccep table consequences . Even if the a i m is to maxi m ize the total of some quantity of benefit over all persons, it is necessary to pick a single measu re o f that quantity that app lies fairly to everyone, and p u re preference is not a good me asu re. ' The fa ct that so meone would be willing to fo rego a de cent diet in order to build a monument to his god does not mean that his clai m on others fo r aid in his proj ect has the same strength as a cla i m fo r aid in obtaining enough to eat (even ass u m i ng t h a t the sacri fices requ i red o f others would be the sa me) . ' 5 Even i f a standard of objectivity is introduced, the range of morally relevant i n teres ts can s t i l l b e q u i te b r o a d , and i t will v ar y from person to person. T h e individual as moral clai m a n t con­ tinues to be more o r less the whole person. On the other hand, anyone's claims can in pri nciple be com pletely outvoted b y the claims o f others. In the final outcome a given i n d i v i d ual 's claims may be met hardly a t all, though they have been coun ted i n the majoritarian calcula tion used to arrive at that outcome. U t i l i tarianism ta kes a generous view o f i n d i vid ual moral claims and co mbines them aggregatively. I t appl ies the res u l tin g values to t h e assessment o f overall resu lts o r sta tes o f a ffa i rs, an d derives t h e assess ment of actions fro m t h i s as a secondary result. One i s to do what will tend to promote the results that appear bes t fro m a point of view that co mbines all individual i nteres ts. The moral equality o f u t i l i tarianism consists i n letting each perso n 's i n terests contri b u te i n the same way to determining what in sum would be best overall.

v Rights are very d i fferent, both in structu re a n d in content. They are not m aj o ritarian or in any other way aggregative, and they 5

T. M . Scanlon, ' Preference and

Urgcncy',JUYIIII/ of Philosoph y ,

19 (November 6, 1 975), 659-60.

LXXII.

no.

Mortal questio11s

1 14

do not provide an assess ment o f overall res u l ts . I ns tead, the y deter mine the acceptability of actions d i rectly. The moral eq ual­ ity of persons under this conception is their equal clai m a gai nst each other not to be i n terfered with in s pecified ways. Ea ch person m u s t be trea ted equally in certain defini te respects by

each other person.

In a sense, these claims a re not combi ned at all. They must be respected i !�dividually. What anyone may do is res tricted to what will not violate the rights of a n yone else. Since the designated aspect o f each person 's point o f view sets this limit by

itse[f, the condition is a kind o f unanimity req uirement.

Rights m a y be absolu te, o r i t m a y be permissible ·to override them when a significant th reshold is reached in the level of harm that can be p revented by doing so. But however they are defined, they must be respected in every case where they apply. They give every person a l i m i ted veto over how others m a y trea t him. This kind o f unanimity condition is possible only for rights that l i m i t what one person may do to another. There cannot in this sense be rights to have certain things - a right to medical

care, or to a decent standard of living, or even a right to life. The

language of rights is so meti mes used in this way, to indicate the special i m portance of certain h u m a n goods. But I believe that the true moral basis of s u ch claims is the priority of more u rgen t over less u rgent individual needs, and this is essen tially

an

egalitarian p rinciple. T o preserve distinctions I shall use the term ' right' only for a clai m that gives its possessor a kind of veto power, so that if everyone has the right, that places a condition o f u nani mous a cceptability, in this respect, on action. There ca n be no l i teral right to life in that sense, because there are situations in which any possible cou rse of action will lead to the death of so meone or other; and if everyone had a right to stay alive, nothing would be permissible in those situations . 6 Rights o f t h e k i n d I a m cons idering escape this problem

beca use they a re agent-centered. A right not to be kil led, for exa mple, is not a right that everyone do what is required to 6

There may be ci rcu mstances in which nothing is permissible - true mora l dilemmas in which every possible course of action is wrong. But these arise only from the clash of distinct moral principles and not from the application of one principle. Sec chapter 5, above.

Equality

115

in s u r e that you a r e not killed. It is merely a righ t not to be killed , and i t is correlated w i th o t h e r people's d u t y not to kill you. Such a n ethic does not enjoin th a t violations of rights be minimized. That would be to cou n t them merely a s p a rti cularly grave evils i n the assess men t of o u tcomes. Instead, rights l i m i t action d i rectl y : each person is fo rbidden to violate d i rectly the righ ts o f others even i f he coul d red u ce the overall n u mber of · viol a tions of rights i n d i rectly by violating a few hi msel( I t is ha rd to a ccount for s u ch agent-centered res trictions. One thing

to sa y a b o u t them b y way of i n terpretation is t h a t they rep resen t a higher degree of moral inviolability than principles req u i ring us to do wha tever will m i n i mize the v i o l a tion of rights. For if

that were the p r i n ciple, then violation of the righ t would not always be wrong. The moral c l a i m o f a righ t not to be m u rdered e ven to prevent several other mu rders is s tronger than the clai m which merely counts mu rder as a great evil, for the fo rmer prohib i ts m u rders tha t the l a tter would permit. That is true even though the l a tter might enable one to preve n t more m u rders than the former. But this does not go very fa r toward explaining agent-cen tered rights. A serious a ccou n t would h a v e to consider not only the protected in terests but the relation between the agent a n d the person he is constrai ned not to treat i n certain ways, even to a chieve very desira ble ends. The concern with wha t one is doing to whom, a s opposed to the concern with wha t h appens, is an i m portant pri m a r y sou rce o f ethics t h a t is poorly u nders tood. H a v i n g noted tha t rights yield a n assessment i n the first instance of actions rather than of o u tcomes, we ca n see that they also define i n d i v i d u a l moral c l a i ms more n a rrowly than does utilita rianism, and combine them d i fferently. The utilitarian constructs a n i m pers o n a l point o f view i n w h i ch those o f a l l i n d i v i d u a l s a re combined to g i v e j u dgments of u tility, wh ich i n tu rn are to g u i d e everyone's a ctions. For a defender o f rights, the respects i n which each person is inviolable present a d i rect and

independent l i m i t to what any other person may d o to him. There is no single combination of viewpoin ts which yields a common goal for everyone, b u t each o f us must l i m i t o u r a ctions to a r ange tha t is not un acceptable to a n yone else in certain respects. Typically, the ra nge o f what may b e done because i t violates no ri ghts is rather large.

t16

Mortal questio11s

For this reason the morality of rights tends to be a limited, even a mini mal morality. It leaves a great deal of human life ungoverned by moral restrictions or requirements. That is why, if unsupplemented, it leads naturally to political theories of limited government, and, in the extreme, to the libertarian theory of the minimal state. The justification of broad govern­ ment action to promote all aspects of the general welfare requires a much richer set of moral requirements. 7 This type oflimited morality also has the consequence that the numbers of people on either side of an issue do not count. In a perfectly unani mous morality the only number that counts is one. If moral acceptability is acceptability in a certain respect from each person's point of view, then even if in other respects one course of action is dearly more acceptable to most but not all of the people involved, no further moral requirement fol­ lows.& The moral equality of rights, then, consists in assigning to each person the same domain o f interests with respect to which he may not be directl y i nterfered with by anyone else. VI

Oddly enough, egalitarianism is based on a more obscure conception of moral equality than either of the less egalitarian theories. It employs a much richer version of each person's point of view than does a theory of rights. In that respect it is closer to utilitarianism. It also resembles utilitarianism formally, in being 7

8

The issue over the ex rem of morality is one of the deepest in ethical theory. M an y have felt it a n objec tion to utilitarianism that it makes ethics s wallow up eve ry t hing, l ea ving o nl y on e o ptim a l choice, or a small set of equally optimal alternatives, p e r m issible for any person at any ti me . Those who offer t h i s o bj ection differ over the size and shape of the range of choices that should be left to i nd iv i d ual inclination after the ethical boundaries have been drawn. John Taurek has recently defended essentially this position in his paper, 'Should the Numbers Count?', Philosoplry & Public Affairs, V I , no. 4 (Summer. 1977), 293-3 1 6. He holds that given a choice between saving one life and saving five others, one is not required to save the five: one may save either the one or the five. I believe that he holds this because there is at least o ne p oi nt of view from which saving the five is not the better choice. Taurek does believe tha t some moral requirements derive from s pecial rights and obligations, but in cases like this, where there are funda mental conflicts of interest, it is i mpossible to define a condition of universal acceptability, and the choice is th e re fore not governed by any moral re q uirement.

Equality

1 17

a pp lied first to the assessment of outcomes rather than of ac tions. But it does not combine all points of view by a majo rita rian method. Instead, it establishes an order of priority am on g needs and gives preference to the most urgent, regardless of numb ers. In that respect it is closer to rights theory. Wh at conception of moral equality is at work here, i.e. what e qu al moral claim is being granted to everyone and how are these claims combined? Each individual's claim has a complex form: it includes more or less all his needs and interests, but in an order of relative urgency or importance. This determines both wh i ch of them are to be satisfied first and whether they are to be satisfied before or after the interests of others. Something close to unanimity is being invoked. An arrangement must be accept­ able first from the point of view of everyone's most basic claims, then from the point of view of everyone's next most basic claims, etc. By contrast with a rights theory, the individual claims are not limited to specific restrictions on how one may be treated. They concern whatever may happen to a person, and in appropriate order of priority they include much more than protection from the most basic misfortunes. This means that the order of priority will not settle all conflicts, since there can be conflicts of interest even at the most basic level, and therefore unanimity cannot be achieved. Instead, one must be content to get as close to it as possible. One problem in the development of this idea is the definition of the order of priority: whether a single, objective standard of urgency should be used in construing the claims of each person, or whether his interests should be ranked at his own estimation of their relative importance. In addition to the question of objectivity, there is a question of scale. Because moral equality is equality between persons, the individual interests to be ranked cannot be momentary preferences, desires, and experiences. They must be aspects of the individual's life taken as a whole: health, nourishment, freedom, work, education, self-respect, affection, pleasure. The determination of egalitarian social policy requires some choice a mong them, and the results will be very different depending on whether material advantages or indi­ vidual liberty and self-realization are given priority. ·

But let me leave these questions aside. The essential feature of an

egalitarian priority system is that it counts improvements to

1 18

Mortal questio11s

the welfare of the worse off as more urgent than improvemen ts to the welfare of the better off. These other questions must be answered to decide who is worse off and who is better off, and how much, but what makes a system egalitarian is the priori ty it gives to the claims of those whose overall life prospects put th ent at the bottom, irrespective of numbers or of overall utility. Each individual with a more urgent claim has priority, in the simplest version of such a view, over each individual with a less urg ent claim. The moral equality of egalitarianism consists in taking into account the interests of each person, subject to the same system of priorities of urgency, in determining what would b e best overall. VII It is obvious that the three conceptions of moral equality with which we are dealing are extremely different. They define each person's equal moral claim differently, and they derive practical conclusions from sets of such claims in different ways. They seem to be radically opposed to one another, and it is very difficult to see how one might decide among them. My own view is that we do not have to. A plausible social morality will show the influence of them all. This will certainly not be conceded by utilitarians or believers in the dominance of rights. But to defend liberal egalitarianism it is not necessary to show that moral equality cam1ot be interpreted in the ways that yield rights or utilitarianism. One has only to show that an egalitarian interpretation is also acceptable. The result then depends on how these disparate values combine. Though my own view is somewhat different from that of Rawls, I shall be g i n b y considerin g his ar g u ments, in order to

explain why another account seems to me necessary.9 He gives two kinds of argument for his position. One is intuitive and belongs to the domain of ordinary moral reasoning. The other is theoretical and depends on the construction by which Rawls works out his version of the social contract and which he calls the Original Position. I shall begin with two prominent exam­ ples of the first kind of argument and then go on to a brief consideration of the theoretical construction. 9

Some of my comments are developed in ' Rawls on Justice', Pllilosophical

Review,

LXXX I I I

(1 973), 220-33.

Equality

1 19

One poin t Rawls makes repeatedly is that the natural and so cial con tingencies that influence welfare - talent, early envi­ ron ment, class background - are not themselves deserved. So differences in benefit that derive. from them are morally arbit­ rary. tO They can be justified only if the alternative would leave the l east fortunate even worse of( In that case everyone benefits from the inequalities, so the extra benefit to some is justified as a rneans to this. A less egalitarian principle of distribution, whether it is based on rights or on utility, allows social and natural contingencies to produce inequalities justified neither because everyone benefits nor because those who get more des erve more. The other point is directed specifically against utilitarianism. Rawls maintains · that utilitarianism applies to problems of social choi ce - problems in which the interests of many individuals are involved - a method of decision appropriate for one individual. t 1 A single person may accept certain disadvantages in exchange for greater benefi ts. But no such compensation is possible when one person suffers the disadvantages and another gets the benefits. So far as I can see, neither of these arguments is decisive. The first assumes that inequalities need justification, that there is a presumption against permitting them. Only that would imply that undeserved inequalities are morally arbitrary in an invidious sense, unless otherwise justified. If they were arbitrary only in the sense that there were no reasons for or against them, they would . require no justification, and the aim of avoiding them could provide no reason to infringe on anyone's rights. In any case the utilitarian has a j ustification to offer for the inequalities that his system permits: that the sum of advantages is greater than it would be without the inequality. But even if an inequality were accep table only if it benefited everyone, that would not have to imply anything as strong as the Difference Principle. More than one deviation from equality may benefit everyone to some extent, and it would require a specific egalitarian assump­ tion to prefer the one that was most favorable to the worst off. The second argument relies on a diagnosis of utilitarianism IO 11

Rawls,. · Rawls.

Theory ofjustice, pp. 74, 1 04. Theory ofJustice, pp. 27, 1 87.

1 20

Mortal qt�estions

that has recently been challenged by Derek Parfit. 12 But even if the diagnosis is correct, it does not supply an argument for equality, for it does not say why this method of summation is not acceptable for the experiences of many individua ls. I t certainly cannot be j ustified simply by extension from the individual case, but it has enough prima facie appeal to require displacement by some better alternative. It merely says that more of what is good is better than less, and less of what is bad is better than more. Someone might accept this conclusion with­ out having reached it by extending the principle of individual choice to the social case. There is no particular reason to think that the principle will be either the same or different in the two cases. In Utilitarianism intrapersonal compensation has no special significance. It acquires significance only against the background of a refusal in general to accept the unrestricted summation of goods and evils - a background to which it provides the exception. This background must be independently justified. By itself, the possibility of intrapersonal compensation neither supports nor undermines egalitarian theories. It implies only that if an egalitarian theory is accepted, it should apply only across lives rather than within them. It is a reason for taking individual human lives, rather than individual experiences, as the units over which any distributive principle should operate. But it could serve this function for anti-egalitarian as well as for egalitarian

views. This is the revers': of Rawls' argument: no special distributive principle should be applied withiu human lives because that would be to extend to the individual the principle of choice appropriate for society. Provided that condition is met, intrapersonal compensation is neutral all)ong distributive prin­ ciples. Next let me consider briefly Rawls' contractarian argument. Though he stresses that his theory is about the morality of social institutions, its general ideas about equality can I think be applied more widely. The Original Position, his version of the 12

L a c e r Sdvc:s a nd Moral Principles', in Philosopl1y [.,- Prrso11al Rtlatiolls, ed. A. Moncefiore (London: Routledge & K eg a n Paul, 1 973). Parfic suggests chat u t il it ari an ism could express the dissolution of temporall y extended

'

individuals into experiential seq uences rather chan the: conflation of separate individuals into a mass p e rso n .

1 21

Equality

social cont ract, is a constructed unanimity condition which a ttri bu tes to each person a schematic point of view that abstracts i11>m the differences between people, but allows for the main cl tegories of human interest. The individual is expected to ch oose p rinciples for the assessment of social institutions on the 155 u rnp tion that he may be anyone, but without assuming that he has an equal chance of being anyone, or that his chance of be ing in a certain situation is proportional to the number of people in that situation. The resulting choice brings out the priorities that are generally shared , and combines interests ranked by these priorities with­ out regard to the n u m hers of people invol ved. The pri n cip l es u nanimously chosen on the basis of such priorities grant to each person the same claim to have his most urgent needs satisfied pri or to the less urgent needs of anyone else. Priority is given to individuals who, taking their lives as a whole, have more urgent needs, rather than to the needs that more individuals have. There has been much controversy over whether the rational choice under the conditions of uncertainty and ignorance that pre vail in the Original Position would be what Rawls says it is, or even whether any choice could be rational under those conditions. But there is another question that is prior. Why does what it would be rational to agree to under those conditions determine what is right? Let us focus this quest i on more specifically on the features of the Original Position that are responsible for the egalitarian result. There are two of them. One is that the choice must be unani mous, and therefore everyone must be deprived

of

all

information about his conception of the good or his position in society. The other is that the parties are not allowed to choose as if they had an equal chance of being anyone in the society, because in the absence of any information about probabilities it is not, according to Rawls, rational to assign some arbitrarily, using the Princi p le of Insufficient Reason. The Original Posi­ tion is constructed by subtracting information without adding artificial substitutes. This results directly in the maximin strategy of choice, which leads to principles that favor the worst off in general and i mpose even more stringent equality in the basic liberties. Suppose Rawls is right about what it would be rational to

1 22

Mortal questions

choose under those conditions. We must then ask why � unanimous choice under conditions of ignorance, withou t �n assu mption that one has an equal chance of being anyone in the society, correctly expresses the constraints of mp rality. Other constructions also. have a claim to counting all persons as mo ral equals. What makes these conditions of unanimity under ignor­ ance the right ones? They insure that numbers do not count13 and urgency does, but that is the issue. A more fundamental type of argument is needed to settle it. V iH The

main question is whether a kind of unani mity should enter into the combination of different points of view when evaluative judgments are being made about outcomes. This is an issue between egalitarian and utilitarian theories, both of which concern themselves with outcomes. Rights theories are opposed to both, because although they use a kind of unanimity condi­ tion, it is a condition on the acceptability of actions rather than of outcomes. In defending an interpretation of moral equality in terrns of unani mity applied in the assessment of outcomes, I am therefore denying that either utilitarianism or rights theories, or both, represent the whole truth about ethics. As I have said, acceptance of egalitarian values need not imply total exclusion of the others. Egalitarians may allow u tility independent weight, and liberal egalitarians standardly acknow­ ledge the importance of certain rights, which limit the means that may be used in pursuing equality and other ends. 14 I believe that rights exist and that this agent-centered aspect of morality is very important. The recognition of individual rights is a way of accepting a requirement of unani mous acceptability when weighing the claims of others in respect to what one may do. But a theory based exclusively on rights leaves out too much 13

14

Since the Difference Pri nciple is applied not to individuals but to soci� l classes, conflicts of i n terest within t h e worst o ff or any other groups are absorbed in a set of average expectations . This means that the nu mbers count in a sense witlriu a social class, in determining which policy benefits it most on average. But numbers do not count i n determining priority among classes in the u rgency of their clai ms. That is why the problems oi. this concep tion of social justice are similar to those of a mo re individually tailored egalitarianism.

Such a view is defended by Ronald Dworkin in Takiug Riglrts Serioruly (Ca mbridge, Mass. : Harvard University Press . 1 977).

Equality

1 23

th a t is morally relevant, even if the interests it includes are � mong the most basic. A m oral view that gives no weight to the val ue of overall outcomes cannot be correct. I S So let me return to the issue of unani m ity in the assessment of outco mes. The essence of such a criterion is to try in a moral assess ment to include each person s point of view separately, so JS t o a chieve a result which is in a significant sense acceptable to each person involved or affected. Where there is conflict of interests, no r esult can be c ompl e t ely acceptable to everyone. But it is possible to assess each result from each point of view to tr y to find the one that is least unacceptable to the person to who m it is most unacceptable. This means that any other altern ative will be more unacceptable to someone than this alternative is to anyone The preferred alternative is. in that sense the least unacceptable, considered from each person s point of view separately. A radically egalitarian policy of giving absolute priority to the worst off, regardless of numbers, would result from always choosing the least unacceptable alternative, in this sense. This ideal of individual acceptability is in fund a menta l opposi­ '

.

'

tion to . the aggregative ideal, which constructs

a special moral

point of view by combining those of individuals into a single conglomerate viewpoint distinct from all of them. That is done in utilitarianism by adding them up. Both the separate and the conglomerate methods count everyone fully and equally. The difference between them is that the second moves beyond individual points of view to something more comprehensive than any of them, though based on them The first stays closer to the points of view of the individuals considered It is this ideal of a cceptability to each individual that underlies the appeal of equality. We can see how it operates even in a case involving small numbers. Suppose I have two children one of which is normal and quite happy, and the o the r of which suffers from a painful handicap. Call them respectively the first child and the second child. I am about to change jobs Suppose I must decide between movi n:g to an expensive city where the second child can receive special medical treatment and schooling, but .

.

,

.

15 I

have said more about this in 'Libertarianism without Foundations', Yalt LXXXV, (1975), a review of Robert Nozick, Auarchy, State, aud Utopia (New York: Basic· Books, 1 974). L11w ]oumal,

1 24

Morta l questions

where the family's standard of living will be lower and the neighborhood will be unpleasant and dangerous for the first child - or else moving to a pleasant semi-rural suburb whe re the first child, who has a special interest in sports and nature, can have a free and agreeable life. This is a difficult choice o n any view. To make it a test for the value of equality, I wan t to suppose that the case has the following feature: the gain to the first child of moving to the suburb is substantially greater than the gain to the second child of moving to the city. After all, the second child will also suffer from the family's reduced standard of living and the disagreeable environment. And the educational and therapeutic benefits will not make him happy but only less miserable. for the first child, on the other hand, the choice is between a happy life and a disagreeable one. Let me add as 4 feature of the case that there is no way to compensate either child significantly for its loss if the choice favoring the other child is made. The family's resources are stretched, and neither child has anything else to give up that could be converted into something of significant value to the other. If one chose to move to the city, it would be an egalitarian decision. It is more urgent to benefit the second child, even though the benefit we can give him is less than the benefit we can give the first child. This urgency is not necessarily decisive. It may be outweighed by other considerations, for equality is not the only value. But it is a factor, and it depends on the worse off position of the second child. An i mprovement in his situation is more important than an equal or somewhat greater i mprove­ ment in the situation of the first child. Suppose a third child is added to the situation, another happy, healthy one, and I am faced with the same choice in allocation of indivisible goods. The greater urgency of benefiting the second child remains. I believe that this factor is essentially unchanged by the addition of the third child. It remains just as much more urgent to benefit the second child in this case as it was when there were only two children. t6 16

Note that these thoughts do not drp md on any idea of personal identity over time. though they can emp loy such an idea. All that is needed to evoke them is a distinction between persons at a time. The i mpulse to distributive equa l i t y arises so long as we can distinguish between two experiences being had by two persons and their being had by one person. The cri teria of personal identity over time merely determine the size of

Equality

1 25

The main point about a measure of urgency is that it is done

by p airwise comparison of the situations of individuals. The

si m plest method would be to count any i mprovement in the situati on of someone worse off as more urgent than any intp ro vement in the situation of someone better off; but this is no t especially plausible. It is more reasonable to accord greater urgen cy to large improvements somewhat higher in the scale chan to very s mall improvements lower down. Such a modified principle could still be described as selecting the alternative that was least unacceptable fro m each point of view. This method can be extended to problems of social choice involving large numb ers of people. So long as numbers do not count it remains a type of unanimity criterion, defined by a suitable measure of urgency. The problem of justifying equality then becomes the problem ofjustifying the pursuit of results that are acceptable to each person involved. Before turning to a discussion of this problem, let me say why I think that even if it were solved, it would not provide the foundation for a correct egalitarian theory. It seems to me that no plausible theory can avoid the relevance of numbers com­ pletely. There may be some disparities of urgency so great that the priorities persist whatever numbers are involved. But i f the choice is between preventing severe hardship for some who are very poor and deprived, and preventing less severe but still substantial hardship for those who are better off but still struggling for subsistence, then it is very difficult for me to believe that the numbers do not count, and that priority of urgency goes to the worse off however many more there are of the better off. It might be suggested that this is a case where equality is outweighed by utility. But if egalitarian urgency is itself sensitive to nu mbers in this way, it does not seem that any form of unani mity criterion could explain the foundation of the view. Nor does any alternative foundation suggest itsel(

IX For a view of the more unco mpronusmg type, similar in structure to that of Rawls, we need an explanation of why the units over which a distributive principle operates. That, briefly, is what I think is wrong with Parfit's account of the relation between distributive justice and personal identity.

1 26

Mortal questions

individual pairwise comparison to find the individually least unacceptable alternative is a good way to adjudicate a mong competing interests. What would it take to justify this metho d of combining individual claims? I think the only way to answer this question is to ask another: what is the source of morality? How do the interests of others secure a hold on us in moral reasoning, and does this imply a way in which they must be considered in combination? I have a view about the source of other-regarding moral reasons that suggests an answer to this question. The view is not very different from the one I defended in The Possibility of Altruism , 1 7 and I will only sketch it here. I believe that the general form of moral reasoning is to put yourself in other people's shoes. This leads to acceptance of an impersonal concern for them corresponding to the impersonal concern for yourself that is needed to avoid a radical incongruity between your attitudes from the personal and impersonal standpoints, i. e. from inside and outside your life. Some considerable disparity remains, because the personal concerns remain in relation to yourself and your life: they are not to be replaced or absorbed by the impersonal ones that correspond to them. I S (One is also typically concerned in a personal way for the i nterests of certain others to whom one is close. ) But we derive moral reasons by forming in addition a parallel i mpersonal concern corresponding to the interests of all other individuals. It will be as strong or as weak, as comprehensive or as restricted, as the impersonal concern we arc constrained by the pressures of congruency to feel about ourselves. In a sense, the requirement is that you love your neighbor as yourself: but only as much as you love yourself when you look at yourself from outside, with fair detachment. The process applies separatel y to each individual and yields J set of concerns corresponding to the individual lives. There may ' be disparities between a person s o bj ec ti v e interests and his own subjectively perceived interests or wishes, but apart from this, his claims enter the i mpersonal domain of reasons unchanged, as those of an individual. They do not come detached from him and go into a big hopper with all the others. The impersonal 17 IS

Oxford: Clarendon P ress, 1 970. In this respect my p resen t view differs from the one in The Possibility

A ltruism.

of

Equality

1 27

co ncern of ethics is an i mpersonal concern for oneself and all others as individuals. It derives from the necessary generalization of a n i mpersonal concern for one's own life and interests, and the genera lization preserves the individualistic form of the original. Fo r this reason the impersonal concern that results is frag­ rnented: it includes a separate concern for each person, and it is real ized by looking at the world fro m each person's point of view separately and individually, rather than by looking at the wo rld from a single comprehensive point of view. Imaginatively one m ust split into all the people in the world, rather than turn on eself into a conglomeration of them. This, it seems to me, makes pairwise co mparison the natural way to deal with conflicting claims. There may be cases where the p olicy chosen as a result will seek to maximize satisfaction rather than equalizing it, but this will only be where all individuals have an equal chance of benefiting, or at least not a conspicuously unequal chance. t9 At the most basic level, the way to choose from many separate viewpoints si multaneously is to maintain them intact and give priority to the most urgent individual claims. As I have said, equality is only one value and this is only one method of choice. We can understand a radically egalitarian system j ust as we can understand a radical system of rights, but I ;�ssume neither is correct. U tility is a legitimate value, and the majoritarian or conglomerate viewpoint on which it d epends is an allowable way of considering the conflicting interests of numbers of diffe rent people at once. S till, the explanation of egalitarian values in terms of separate assess ment from each point of view is a step toward understanding; and if it does not imply that these values are absolute, that is not necessarily a drawback.

19

I leave aside the question when the equality of chances can be counted as

real enough to supersede the inequality of actual outcomes. Perhaps that applies only to certain kinds of outcomes, and certain ways of determining chances.

9 The Fra g mentation of Value

I want to discuss some problems created by a disparity between the fragmentation of value and the singleness of decision. These problems emerge in the form of practical conflicts, and they usually have moral components. By a practical conflict I do not mean merely a difficult decision. Decisions may be difficult for a number of reasons: because the considerations on different sides are very evenly balanced; because the facts are uncertain; because the probability of different outcomes of the possible courses of action is unknown. A difficult choice between chemotherapy and surgery, when it is uncertain which will be more effective, is nor an example of what I mean by practical conflict, because it does not involve conflict between values which are incomparable for reasons apart from uncertainty about the facts. There can be cases where, even if one is fairly sure about the outcomes of alternative courses of action, or about their probability distribu­ tions, and even though one knows how to distinguish the pros and cons, one is nevertheless unable to bring them together in a single evaluative judgment, even to the extent of fmding them evenly balanced. An even balance requires comparable quan­ tities. The strongest cases of conflict are genuine dilemmas, where there is decisive support for two or more incompatible courses of action or inaction. In that case a · decision will still be necessary, but it will seem necessarily arbitrary. When two choices are very evenly balanced, it does not matter which choice one makes, and arbitrariness is no problem. But when

The fragmentation of value

1 29

ea ch seems right for reasons that appear decisive and sufficient, arbitrariness means the lack of reasons where reasons are needed, sin ce either choice will mean acting against some reasons wi th out bei "g able to claim that they are outwighed. There· are �ve fundamental types of valuif that give rise to basi c conflict. Conflicts can arise within as well as between them, but the latter are especially difficult. (I have not included self-in terest in the group; it can conflict with any of the others. ) First, there are specific obligations to other people or institu­ tions: obligations to patients, to one's family, to the hospital or university at which one works, to one's community or one's country. Such obligations have to be incurred, either by a deli berate undertaking or by some special relation to the person or institution in question. Their existence depends in either case on the subject's relation to others, although the relation does not have to be voluntary. (Even though young children are not at liberty to choose their parents or guardian'i, parental care creates some obligation of reciprocal future concern. ) The next category is that of constraints on action deriving from general rights that everyone has, either to do certain things or not to be treated in certain ways. Rights to liberty of certain kinds, or to freedom from assault or coercion, do not depend on specific obligations that others have incurred not to interfere, assault, or coerce. Rather, they are completely general, and restrict what others may do to their possessor, whoever those others may be. Thus a doctor has both specific obligations to his patients and general duties to treat anyone in certain ways. The third category is that which is technically called utili.!y. _ This is the consideration that takes into account the effects of what one does on everyone's welfare - whether or not the components of that welfare are connected to special obligations or general rights. Utility includes all aspects of benefit and harm to all people (or all sentient beings), not just those to whom the agent has a special relation, or has undertaken a special com mit­ ment. The general benefits of medical research and education obviously come under this heading. The fourth category is that of perfectionist ends or values. B y this I mean the intrinsic value of certain achievements o r creations, apart from their value to individuals who experience or use them. Examples are provided by the intrinsic value of

1 30

Mortal questio11s

scientific discovery, of artistic creation, of space explorat ion , perhaps . These pursuits do of course serve the interests of the individuals directly invol ved in them, and of certain spect ato rs .

But typically the pursuit of such ends is not justified solel y in terms of those interests. They a re thought to have an int rin sic

value, so that it is important to achieve fundamental advances, for example, in mathematics or astronomy even if very few people come to understand them and they have no prac tica l

effects. The mere existence of such understanding, somewhere in the s pecies, is regarded by many as worth substantial sac­

rifices. Naturally opinions differ as to what has this kind of

worth. Not everyone will agree that reaching the moon or Mars

has the intrinsic value necessary to justify its current cost, or that the performance of obscure or difficult orchestral works has any

value apart from its worth to individuals who enjoy them. But

many things people do cannot be justified or understood with­ out taking i n to account such perfectionist values.

The final category is that of commitment to one's own

proj ects or undertakings, which is a value in addition to what. : ever reasons may have led to them in the first place. If you have 'set out to cli mb Everest, or translate Aristotle's master the

Well- Tempered Clavier,

Metaphysics,

or

or synthesize an a mino acid,

then the further pursuit of that project, once begun, as�ull:es

�e �a � k_able_i ll!P.Ort�nce. t It is partly a matter ofj u s ti fying earlier investment of rime and energy, and not allowing i t to have been

in vain. It is partly a desire to be the sort of person who finishes

what he begins. But whatever the reason, our proj ects make

autonomous claims on us, once undertaken, which they need not have made in advance. Someone who has determined to master ..._ the Well- Tempered Clavier may say 'T can·cgo-tothe movies, I ··-

have to practice ' ; but it would be strange for him to say that he had to master the

Well- Tempered Clavier.

These com mitments should �ot be confused with self-interest, for self-intere:;t � i ms at the i 9-tegrated fulftJ9fient over tin:te of all

cyd

oes no t one's inte�e:sts and desires .{or at least thoft desires on / e1i minate}. Spe�al commitmepfs may, in th�r pursui t, wis be i : , cal to self-int est thus defmed. They need not have been

�� ��

9f ;/r"tterested reasons, and their pursuit certainly

undertaken for self I

See Gilbert Harman, 'Practical Reasoning',

{1976), 432-63.

Review of Metaphysics,

XXIX

The fragmentation

of value

131

--

need not be controlled by self-interest. / O bligations, rights, utility, perfectionist ends, and p �vatc com mit ments - these values enter into our decisions con�antly, and co nflicts among them, and within them, arise ir:Vmedical research, in politics, in personal life, or wherever th� g'rounds of ac tion are not artificially restricted. What would �mean to give a sys tem of priorities among them? A simpler _p{oral conception might permit a solution in terms of a ,sfiort list of clear pro hibitions and injunctions, with the bal �e of decision left to pers onal preference or discretion, but th;afwill not work with so mixed a collection. One might try to 6rder them. For example: never infringe general rights, and Zu'dertake only those special oblig ations that cannot lead to/the infringement of anyone's rights; maximize utility withityfhe range of action left free by the constraints of rights and ytlligations; where utility would be eq ually served by vari9us policies, determine the choice by refere ? ce to perfecti � st ends; and finally, where this leaves _ anythmg unsettled /decide on grounds of personal com mttment or even simple p leference. Such a method of decision is absurd, not because 9( the particular order chosen but because of its absolutene�s:' The ordering [ have given is not arbitrary, for it reflects :vBegree of relative stringency in these types of values. But i ts absurd to hold that obligations can never outweigh rig s, or that utility, however large, can never outweigh ligation. However, if we take the idea of outweighing seriously, and try to think of an alternative to ordering as a method of rationalizing decision in conditions of conflict, the thing to look for seems to be a single scale on which all these apparently disparate considerations can be measured, added, and balanced. Utilitarianism is the best example of such a theory, and interest­ ing attempts have been made to explain the apparent priority of rights and obligations over utility in utilitarian terms. The sarne might be tried for perfectionist goals and personal commit­ ments. My reason for thinking that such explanations are unsuccessful, or at best partially successful, is not just that they imply specific moral conclusions that I find intuitively unaccept­ able (for it is always conceivable that a new refinement of the theory may iron out many of those wrinkles). Rather, my reason for doubt is theoretical: I do not believe that the source of value

1 32

·

Mortal questions

is unitary - displaying apparent multiplicity only in its app lica­ tion to the world. I believe that value has fundamen tally different kinds of sources, and that they are reflected in the classification of values into types. Not all values represent the pursuit of some single good in a variety of settings. Think for example of the contrast between perfectionist and utilitarian values. They areformally different, for the latter takes into account the number of people whose interests are affected, and the former does not. Perfectionist values have to do with the mere level of achievement and not with the spread either of achievement or of gratification. There is also a formal contrast between rights or obligations and any ends, whether utilitarian or perfectionist, that are defined in terms of the outcome of actions - in terms of how things are as a result. The claims represented by individual obligations begin with relations bet­ ween individuals, and although the maintenance of those rela­ tions in a satisfactory form must be part of any uti · conception of a good state of affairs, that is not the basic motiv behind claims of obligation. It may be a good thing that e keep their promises or look after their children, but the reason a person has to keep his own promises is very different from the reason he has to want other people unconnected with him to keep their promises - just because it would be a good thing, i mpersonally considered. A person does not feel bound to keep his promises or look after his children because it would be a good thing, impersonally considered. There certainly are things

= ·

we do for such reasons, but in the motive behind obligations a more personal outlook is-e�entiaf. It is yo�-nrel.iliorr-to-the' o��� erson or the institution or aimmunity that moves YO_E.

. not a detached concern for what would be best overall. ----------­ � of this kind may · be described as agept-centere_d or subjective (tho ggh the term 's�b]ective' here/�hould n6t be misunderstoo)i'- it does not mean that the general prinpples of obligation ..ate matters of subjective preference which -inay vary from pepson to person). The reasons in each case apply primarily to thdndividual involved, as reasons for him to w;�'l1t to fulfll his obligations - even though it is also"a good thin g , i mpersonally considered, for him' to do so. (;eneral rights are less personal in their claims, since a right to be free from.interference·arassawt,"for-e�ample, does not derive

The fragmentation of value

-

1 33

fr om the possessor's relation to anyone in particular: everyone is ob li ged to respect it. Nevertheless, they are agent-centered in the sense that the reasons for action they provide apply primarily to

indivi duals whose actions are in danger of infringing such rights. R i ghts mainly provide people with reasons not to do certain

thi n gs to other people - not to treat them or interfere with them in certain ways. Again, it is objectively a good thing that

pe ople's rights not be violated, and this provides disinterested parties with some reason for seeing that X's rights are not violated by Y. But this is a secondary motive, not so powerful as

the reason one has not to violate anyone's rights directly. (That �ect to

is-w.b.r_ it is reasonable for defenders of civil Iiberti

�

· ces that v · e rights of criminal police a1m o t e-policies is to prevent suspects, even wh!:Dy criminals of the rights o ii:Viat� In

fthe

�

that sense the claims deriving from general rights are agent­

centered: less so than those deriving from special obligations,

but still defini tely agent-centered in a sense in which the claims

of utility or perfectionist ends are not. Those latter claims are

impersonal or outcome-centered ; they have to do with what happens, not,

in the first instance, with wha t one does. It is the

contribution of what one does to what happens or what is achieved that matters.

This great division between personal and impersonal, or

�l!hl� !

between agent-centered and �utcome-ce� d, or 1 and objective reasons, is so basicthatttrenders i mplausible any

� let alone of practi��_!!._e�-�c:ming) �ces among these types of reasons'

reductive u nification of in g eneral. The formal

correspond to deep differences in their sources. We appreciate· the force of impersonal reasons when we detach from our

personal situation 'and our special relations to others. Utilitarian ,

considerations arise in this way when .our detach ment takes the .

fo"rm of adopting a general point of view that comprehends

everyone's view of the world within it. Naturally the results will not always be clear. But such an outlook is obviously very different from that which appears in a person's concern for his

s pecial obligations to his family, friends, or colleagues. There he

is thinking very much of his particular situation in the world.:

The two motives come fro m two different points of view, both important, but fundamentally irreducible to a common basis.

1 34

Mortal questio11s

I have said nothing about the still more agent-centered mo tive of commitment to one's own projects, but since that involves one's own life an � �ssarily any relations with others, the same points ob � ply. It is a source of reasons that ca nn o t be assi milated either to u ti l i t y, or perfectionism, or righ ts, o r obligations (except that they might be described as obligations to oneself). My general point is that the formal differences among types of reason reflect differences of a fundamental nature m tbeJr �. and that this rules out a certain kind of solution t o conflicts among these types. H_u man beings are su bject to moral and other motivational claims of ��!L�ifferent kmds. T�i�s because "they are compleX �rea.tures who c;:_��- �!._e� _ the world ftoril nlany - persp�ctives individual, relational, i mpersonal:· ict€.. 1 , etc. � and- e�cli perspecnv-e--preserit� a--different-· set ·;;c Claims. Conflii:: F can e-ii st .wit- hm oneorthese sets�- and it ma_ybe hard to r�sohre�But�h en ;;;�fli ctoccurs- between them, the proolem is still more difficult. Conflicts between personal and impersonal claims are ubiquitous. They cannot, in my view, be resolved by subsuming either of the points of view under the other, or both under a third. Nor can we simply abandon any of them. There is no reason why we should. The capacity to view the world simultaneously from the point of view of one's relations to others, from the point of view of one's life extended through time, from the point of view of everyone at once, and finally from the detached viewpoint often described as the view sub specie aeternitatis i s one of the marks of humanity. Thi� complex capacity is an obstacle to simpbhcatlon. Does this mean, then, that basic practical conflicts have no solution? The unavailability of a single, reductive method or a clear set of p riorities for settling them d oes �ot ···re move the necessity for m aking decisions in such cases. When faced with con flic ting and incommensurable claims we still have to do something - even if it is onlY. to·do nothing. And the fact that action must be unitary s�em5 to imply that unless justification is also unitary, nothipg · can be either right or wrong and all decisions unde r conflict are arbitrary. I believ.c thi� is wrong, but the alternative is hard to explain. . B riefl¥f1 contend that there can be good judgment without total jl}stifi cation, either explicit or implicit. The fact that one cannot -

1 35

Tlze fi"agmelltatiou of value

say why a certain decision is the correct one, given a par lar balance of conflicting reasons, does not mean that t claim to e process of corre ctness is meaningless. Provided one has take p ra cti cal justification as far as it will go in th ourse of arriving at the conflict, one may be able to pr eed without further justification, but without irrationalit ither. What makes this possible is Judgment - essentially t faculty Aristotle described as practical wisdom, which rev: s itself over time in individual decisions rather than in th nunciation of general principles. It will not always yield a ution: there are true practical dilemmas that have no solu n, and there are also conflicts so complex that judgmen annot operate confidently. But in many cases it can be r "ed on to take up the slack that remains beyond the of explicit rational argument. This view has sometimes been regarded as defeatist and empty since it was expressed by Aristotle. In reply, let me say two things. First, the position does not imply that we should abandon the search for more and better reasons and more critical insight in the domain of practical decision. It is just that our capacity to resolve conflicts in particular cases may extend beyond our capacity to enunciate general principles that explain those resolutions. Perhaps we are working with general princi­ ples unconsciously, and can discover them by codifying our decisions and particular intuitions. But this is not necessary either for the operation or for the development of judgment. Second, the search for general principles in ethics, or other aspects of practical reasoning, is more likely to be successful if systematic theories restrict themselves to one aspect of the subject - one component of rational motivation - than if they try to be comprehensive. To look for a single general theory. of how to decide the right . thing to do is like looking for a single theory of Tiow to decide what to believe. Such progress as we have made in the systema­ tic justification and criticism of beliefs has not come mostly from general principles of reasoning but from the understanding of particular areas, marked out by the different sciences, by history, by mathematics. These vary in exactness, and large areas of ·

·

belief a r e left out of the scope of any theory. These must be

governed by common sense and ordinary, prescientific reason­ ing. Such reasoning must also be used where the results of

Mortal questious

1 36

various more systematic methods bear on the matter at h a nd,

but no one of them determines a conclusion. In civil enginee ring problems, for example, the solution depends both on phy sical

factors

capable

of

precise

calculation

and

behavioral

psychological factors that are not. O bviously one should

or

use

exact principles a nd methods to deal wi th those aspects of

a

problem for which they are available, but someti mes there are

other aspects as well, and one must resist the temptation ei ther to

ignore them or to treat them by exact methods to which they are not susceptible.

�mcu.t� a tion f understanding and

We are familiar with this

method when it comes to belief, but we tend to resist it in the case of decision . Yet it is as i rrational to despair of systematic

ethics because one cannot find a completely general account of

what should be done as it woul d be to give up s cien tific research

because there is no general method of arriving at true beliefs. I

am not saying tha t ethics is a science, only that the relation

between ethical theo!:Y_and p ractical decisions is analogous to the

relation between scie�tific theory and beliefs about particular ·

things or events in the world.

In both areas, some problems are much purer than others, that

is, their solutions are more completely determined by factors

that a d mi t of p recise understanding. So metimes the only signifi­ cant factor i!!._ a practical decision is personal obligation, or

general utility, and then one 's reasoning can be confined to that

(however precisel y i t may be unders tood) . Sometimes a process

of decision is artificially insulated against the influence of more

than one type of factor. This is not always a good thing, but

someti mes i t is. The example

I

have i n mind is the j udicial

process, which carefully excludes, or tries to exclude, considera­ tions of utility and personal com mitment, and limits i tself to

claims of right. Since the systematic recognition of such clai ms is

very important (and also tends over the long run not to conflict

unacceptably with other values), it is worth isolating these factors for special treatment. As a resul t, legal argu ment has been

one of the areas of real progress in the understanding of a special

aspect of practical reason. S ystematic theory and the search for

��

_ general principles and methods � a.y s� c ewhere i f we accept a ragmentary approach. U uhtanan,!heory� for exa mple, has a great deal to contnbute If it is not reqiiired,.to account for

f

The fragmentation of value

1 37

ev er ything. U tility is an extremely i mportant factor in decisions, particu larly in public policy, and philosophical work on its

d efi nition, the coordination problems arising in the design of in sti tutions to promote utility, i ts connections with preference, with equality, and with efficiency, can have an impact on such

de cisions. This and other a reas can be the scene of progress even if none

of them aspires to the status of a general and complete theory of right and wrong. There will never be such a theory, in my view,

since the role of j u dgment i n resolving conflicts and appl ying dis parate claims and considerations to real life is indispensable. o dangers can be avoided i f this idea of noncomprehensive syste

·

tion is kept in mind. One is the danger of romantic

defeatism, whic

bandons rational theory because it inevitabl y

leaves many proble

unsolved. The other is the danger of

exclusionary overrational

·

n, which bars

empty all considerations that cann

as

irrelevant or

be brought within the scope

of a general system admitting explicr �efensible conclusions. This yields s kewed results by counting�l y measurable or otherwise precisely describable factors, even �

n others are in

fact relevant. The alternative is to recognize that t

grounds of decision are extremely various and underst

1t1 mate

different degrees. This has both theoretical and p ractical i mplic ..

tions.

On the theoretical side,

areas··o

I

have said that progress in particular

ics and value theory need not wait for the discovery

of a general

dation (even if there is such a thing). This is

recognized by man

by John Rawls,

ilosophers and has recently been urged

who c

·� not only that the pursuit of

substantive moral theory, for eX:H-Q.I' Ie the theory of justice, can proceed independently of views abolit -th_e fou ndation of ethics,

but that until substantive theory is further· Cl�eloped, the search for foundations may be premature.2

.

,_ _

This seems too strong, but it is certainly true o'f.m

one need not make progress at the most funda menta

field that

:vel to

make progress at all. Chemistry went through great deve

Rawls, A Th�ry of)11stict (Cambrid ge: Harva rd l:Jniversity Press. 1 97 1 ). pp . 5 1 -60. See also 'The Indep endence of Moral Theor y '. Procttdit�gs a11d AJdwm of tht A m rrica11 Philosophical Associatio11 (1 974-5) pp . 5-22.

2 John

1 38

Mortal questious

ments during the century before its basis in atomic physics ca me to be understood. Mendelian genetics was developed long 9.ef6'rc an y understanding of the molecular b as i s of heredity. rPres en t, progress in psychology must be pursued great extent independently of any idea about its b m the operation of the g1cal phenomena are ultimately brain. It may be that all ps t e theory of the central nervous system, explainable in t but our _ sent understanding of that system is too meager to �us even to look for a way to close the gap. The corresponding theoretical division in ethics need not be so extreme. We can continue to work on the foundations while exploring the superstructure, and the two pursuits should re­ inforce each other. I myself do not believe that all value rests on a single foundation or can be combined into a unified system, he Jevelo ment an d because different types of valu s r rticulation o 1 erent points of view, all of which t.com i nc o . ·

·

rodUre· s!fciS]o"� . .,F.thics is unlike phys1es,

w

1c

represents

one

point of view, that which apprehends the spatiotemporal proper­ ties of the universe described in mathematical terms. Even in this case, where i t is reasonable to seek a unified theory of all physical phenomena, it is also possible to underste � reduced to a common denominator any more th:iii tiistOry, �gy, philology, and economics can be reduced to phys­ ics. Just as the types of understanding available to us are distinct, even though they must all coexist and cooperate in our minds, so 1the types of value that move us are disparate, even though they )nust cooperate as well as they can in �etermining what we do._ With regard to practical implications, it seems to me that the fragmentation of effort and of results that is theoretically to be expected in the domain of value has implications for the strategy to be used in applying these results to practical decisions, ·

The jragmematio11 of value

1 39

esp ecially questions of public policy. The lack of a general theory of value should not be an obstacle to the employment of those areas of understanding that do exist; and we know more th an is generally appreciated. The lack of a general theory leads too easily to a false dichotomy: either fall back entirely on the unsystem atic intuitive judgment of whoever has to make a d eci sion, or else cook up a unified but artificial system like cost-b enefit analysis,J which will grind out decisions on any problem presented to it. (Such systems may be useful if their claims and scope of operation are less ambitious.) What is needed instead is a mixed strategy, combining systematic results where these are applicable with less systematic judgment to fill in the gaps . However, this requires the development of an approach to decisions that will use available ethical understanding where it is rele vant. Such an approach is now being sought by different groups working in applied ethics, with what success we shall not , know for some time. I want to suggest that the fragmentation of value provides a rationale for a particular way of looking at the task, and an indication of what needs to be done. What we need most is a method of breaking up or analyzing practical problems to say what evaluative principles apply, and how. This is not a method of decision. Perhaps in special cases it would yield a decision, but more usually it would simply indicate the points at which different kinds of ethical considera­ tions needed to be introduced to supply the basis for a respons­ ible and intelligent decision. This component approach to prob­ lems is familiar enough in connection with other disciplines. It is expected that important policy decisions may depend on economic factors, political factors, ecological factors, medical safety, scientific progress, technological advantages, military security, and other concerns. Advice on all these matters can be obtained by responsible officials if there is anyone available whose job it is to think about them. In some cases well­ established disciplines are involved. Their practitioners may vary widely in understanding of the subject, and on many issues they will disagree with one another. But even to be exposed to

)

3

Sec Lawrence Tribe, 'Policy Science: A nalysis or Ideology?', Philosophy & Affairs, 11, no. 1 (Fall, 1 972), 66-1 1 0.

P11blic

1 40

Mortal questions

these controversies (about inflation or nuclear power safety or recombinant DNA risk) is better than hearing nothing at all Moreover it is i mportant that within most serious disciplines there is agreement about what is controversial and what is not. Anyone with an important decision to make, whether he is a legislator or a cabinet officer or a department official, can get advice on different aspects of the problem from people who have thought much more than he has about each of those aspects, and know what others have said about it. The division of disciplines and a consensus about what di mensions of a problem have to be considered are very useful in bringing together the problems and such expertise as there is. We need a comparable consensus about what i mportant ethical and evaluative questions have to be considered if a policy decision is to be made responsibly. This is not the same thing as a consensus in ethics. It means only that there are certain aspects of any problem that most people who work in ethics and value theory would agree should be considered, and can be profession­ ally considered in such a way that whoever is going to make the decision will be exposed to the relevant ideas currently available. Someti mes the best ideas will not be very good, or they will include diametri cally opposed views ; but this is true every­ where, not just in ethics. It might be suggested that the best approach would be to emulate the legal system by setting up an advocacy procedure before a kind of court whose job would be to render decisions on ethically loaded policy questions. (The recent proposal of a science court shows the attractions of the legal model: its non-democratic character has great intellectual appeal. ) But I think the actual situation is too fluid for anything like that. Values are relevant to policy in too many ways, and in combina­ tion with too many other kinds of knowledge and opinion, to be treatable in this manner. Although some legal decisions are very difficult, courts are designed to decide clear, narrowly defined questions to which a relatively li mited set of argu ments and reasons is relevant. (Think of the function of a judge in striking material from the record or refusing to admit certain data or testimony in evidence: such restrictions do not in general apply to legislative or administrative deliberations.) Most practical issues are much messier than this, and their ethical dimensions are much more complex. One needs a method of insuring that

The fragmmtation of value

141

where relevant u n derstanding exists, i t i s made available, and where there is a n aspect o f the problem that no one u nderstands very well, this is u nderstood too. I have not devised such a method, but clea rly i t would have to provide that factors considered should include, a mong o thers, the fol lowing : economic, political, and personal liberty, equality, equity, privacy, p roced ural fairness, intellectual and aesthetic develop ment, community, general utility, desert, avoi d a nce of arbitrari ness, a cceptance of r i s k , t h e interests o f fu ture genera­ tions, the weight to be given to i nterests o f o ther s ta tes or cou n t ries. There is m u ch to be said about each of these. The method would have to be more orga nized to be useful, but a general position on the ways in which ethics is relevant to policy could probably be agreed on b y a wide range o f ethical theorists, from relativists to u tilitarians to K a n tians. R a d i cal disagreement abo u t the basis o f ethics is compatible with substantial agree­ ment about what the i m portant factors are in real li fe. If this consensus, which I believe already exists among ethical theorists, were to gain wider acceptance a mong the public and those who make policy, then the extensive but fragmented understanding tha t we possess i n this a rea could be put to better use than i t is now. I t would then be more difficult simply to ignore certain ques tions, and even i f the ethical considera tions, once offered, were disregarded o r rej ected, the reasons o r absence of reasons for such rej ection would become part of the basis for any decision made. There is a modicu m o f power even in being able to state a

prima facie case.

This conception of the role of moral theo ry also i mplies an answer to the question of its relation to politics, and other methods of decision. Ethics is not being reco m mended as a decision procedu re, but as an essen tial resou rce for making decisions, j us t as p h y s i cs , econ o m i c s , a n d demography are. In fu nda mental const i t u tional d e c is i o n s of t h e S u preme C o ur t , on e b r a nc h of e t hi c s p l a y s a cen tral r o l e i n a process that takes' precedence over the usual methods of po l i t i c a l and a d m i n i s tra­ tive d ecision. B u t for most of the q u estions that n eed deci di n g , eth i c a l considera tions a r e m u l t i p l e, c o m p l e x , often clou dy, a n d mixed up w i t h m a n y others. T h e y need t o b e consid ered in a s ystematic way, b u t in most cases a reason a bl e decision can be reached o n l y by sound j u dg ment, i n fo r med as wel l as possible by the best a rgu ments that any relevant d i s c i p l i nes h a ve to offer.

10 E thics wi thout Biolo g y

The usefulness of a biological approach to ethics depends on what ethics i s. If it is j u s t a certa in type o f behavioral pattern or habit, accompa nied by s o me emotional res ponses, then biologi­ cal theories can be expected to teach us a great deal about it. But i f it is a theoretical i n q u i ry that can be approached by rational methods, and that has in ternal s ta n d a rds of j ustification and criticism, the a ttempt to understand it fro m outside by means of biology will be much less valuable. This is true fo r the same reason that the search fo r a biological explanation of ma themati­ ca l or physical theories, or biological theories for that matter, would be rel a ti vely fu tile. First, we have no general biological understanding of h u m a n thought. Second, it is not a fixed set of behavioral a n d i n tellectual habits b u t a p rocess of develop ment that advances b y constant reexa mination of the total body of results to date. A being who is engaged in s u ch an open-ended process o f discovery cannot at the s a m e t i m e u n derstand it fully fro m o u tside: otherwise he would have a decision procedu re rather than a critica l method. In most i n teresting subj ects we do n o t want a decision procedure because we want to p u rsue a deeper l evel of u n derstanding than that represented by our cu rrent questions and the methods we have for a ns wering them . No one, to m y knowledge, has suggested a b i ological theory

of mathematics, yet the biologica l app roach to ethics has aroused

a g rea t deal of i n terest. There is a reaso n fo r this. Ethics exists on both the behav i o ral and the theoreti cal level. Its appearance in s o me for m in every culture and s ubcultu re as a pa ttern of conduct and j u d g ments abou t conduct i s m o re conspicuous than

Eth ics w itho u t b iology

1 43

its theoretical treatment by philosophers, political an d legal theorists, u topian anarchists, and evangelical reformers. Not only is ethical theory and the a ttempt a t ethical discovery less socially conspicuous than com mon behaviora l morality, but the amount o f disagreement about ethics a t both levels produces dou b t that it is a field fo r rational discovery at all. Perhaps there is nothing to be d iscovered about i t b y s u ch methods, and i t can be u n d erstood only as a social and psychological pec uliarity of human life. I n that case biology will p rovide a good fou ndation, though psychology and sociology will be i m portant as well. In this ess a y I want to explain the reality o f ethics as a theoretical subj ect. Its p rogress is slow and u ncertain, b ut it is i m p o rt a n t both i n i t s e l f and i n relation to t h e non-theoretical forms that ethics takes, because the two levels influence each other. The ethical commonplaces o f any period include i deas that may have been radical discoveries i n a previous age. This is true of modern conceptions o f liberty, equality, and democracy, and we a re i n the midst o f ethical debates which will probably result two hundred years hence i n a dissem i n a ted moral sensibil­ ity that people o f our time would find very u n familiar. A l though the rate of progress i s much slower, the for m of these develop­ ments is so mewhat analogous to the gradual assimilation of revol utionary s cientific discoveries into the c o mm on world­ view. As i n s ci ence, also, b y the time one advance has been widely assimilated it is being s uperseded b y the next, and fu rther develop ments use accep ted current understanding as the basis for extension a n d revision. In ethics the two l evels i n teract in both d i rections, and the division between them i s n o t sharp. Acute questions o f social policy produce widespread attempts to theorize about the basic principles of ethics. A common i dea o f progress is fou n d i n all these fields, although it i s n o t very well understood i n any o f them. I t is assu m e d that we begin, as a species, with certain pr i mitive intui tions and res ponses that may have biological sou rces. But in addition we have

a

critical capacity that has a l l o wed us, starting a

long time ago, to assess, systematize, extend, and in some cases rej ec t these p re-reflective responses. I n s tead o f estimating size and weight b y touch and vision, we develop devices of mea­ surement. Ins tead of guessing about nu merical q uan ti ties we

1 44

Mortal questions

develop mathematical reasoning. Instead of sticking to an idea of the physical world that comes directly from our senses, we have progressively asked questions and developed methods of an­ swering them that yield a picture of physical reality farther a nd farther removed from appearance. We could not have done any of these things if we had not, as a species, had some pre­ reflective, intuitive beliefs about numbers and the world. Pro g­ ress beyond this has required both the efforts of creativ e individuals and the communal activities of criticism, justifica­ tion, acceptance, and rejection. The motivating idea has been that there is always more to be discovered, that our current intuitions or understanding, even if commendable for their date, are only a stage in an indefinite developmental process. In applying this idea to ethics we must allow for the big difference that ethics is meant to govern action, not just belie( In trying to solve ethical problems we are trying to find out how to live and how to arrange our social institutions - we are not just trying to develop a more accurate picture of the world and the people in it. Therefore ethics is connected with motivation. It begins not with pre-reflective ideas about what the world is like, but with p.re-reflective ideas about what to do, how to live, and how to treat other people. It progresses by the subjection of these impulses to examination, codification, questioning, critic­ ism, and so on. As in other areas, this is partly an individual process and partly a social one. And the progress of earlier ages is included as part of the socialization of members of later ones, some of whom may make advances in turn. The development in this case is not just intellectual but motivational, and it cannot be pursued exclusively by small groups of experts, as some scientific or technical subjects can. Because the questions are about how men should live and how societies should be arranged, the answers must be accepted and internalized by many people to take effect, even if only as steps in a continuing process. Though they need not be internalized equally by everyone, this requirement makes ethics a more democratic subject than any science, and severely limits its rate of progress. The community of debate is not a set of experts, except in special institutional cases like the judicial system. S till, the premise of this view of ethics as a subject for rational development is that motives, like beliefs, can be criticized,

Eth ics without biology

1 45

j u stified, and i mp roved - in other words that there is s u ch a thing as p ractica l reason. This means that we can reason n o t only, as H u m e thought, about t h e most effective methods o f achieving w h a t we want, b u t a l s o a b o u t w h a t we should want, both for ou rselves and fo r o thers. It is o f the u t most i m p ortance that such a n i n vestigation, s uch reasoning, is i n ternal to the subj ect. It does not p roceed b y the a pplication to this s u bj ect of methods developed in rel a ti on to other s u bjects, o r o f a general method o f problem-solving and question-answering. While there are s o me extremely general conditions of rationality, they will n o t get you very far in any specific a rea o f inquiry. Whether i t is molecu lar biology, algebra, or distribu tive j u s tice, one has to develop questions, concepts, argu ments, a n d p rinciples by thinking about that field and allowing reason a n d intuition to respond to its specific cha racter. It happens again and again that the methods o f one s ubj ect a re taken as a model of i n tellectual respectability or o bjective rationality, and are then applied to a q u i te different s u bj ect for which they were not developed and for which they a re u nsuited. The results are shallow questions, nonexplanatory theories, and the ana thematization o f i m portant questions as meaningless. Fields that l ac k a wel l-developed method of their own, like the social s ciences, psychology, and ethics, are particularly vulner­ able to su ch intellectual displacement. The point i s that ethics is a subject. I t i s p u rsued b y methods tha t are continually being developed in response to the p roblems that a rise within it. O b viously the creatures who engage i n this activity are organisms about whom we can learn a g reat deal fro m biology. Moreover their capacity to perfo r m the reflective and critical tasks i nvolved i s presumably so meh o w a function of their organic structu re. But i t would be as foolish to seek a biological evolutionary explanation of ethics as i t would be to seek such an explana tion of the development of physics . The develop ment o f physics is

an

i n tellectual pro cess. P resu m a b l y

the h u m a n i n tellectual capacity t h a t has permi tted this extremely rapid p rocess to occur was i n some way an effect, perhaps only a

side-effect, o f a process of biological evolution tha t took a very

long ti me. But the l a t ter can provide no explanation o f ph ysical theories that i s not trivia l . What h u m a n b e i n g s have d i s covered in themselves is a capacity to subj ect their p re-reflective or i n n ate

1 46

Mortal questions

responses to criticism and revision, and to create new forms of understanding. I t is the exercise of that rational capacity that explains the theories. Ethics, though more primitive, is si milar. I t is the result of a h u m a n capacity to subject innate or conditioned pre-reflective motivational and behavioral patterns to critici s m and revision, and to create new forms o f conduct. The capacity to d o this pres u m a b l y has some biological foundation, even i f i t is only a side-effect of other develop ments. B u t the history of the exercise of this capacity and its continual reapplication i n criticis m and revision of i ts own products is not part of biology. Biology may tell us about percep tual and motivational starting points, but in its present state i t has little bearing on the thinking process by which these s tarting points a re trans cended. There may be biological obsta cles to the a chievement of certain kinds of moral p rogress. Withou t question there are psychological and social obstacles, and some of them may have biological causes. That does not make them insurmountable. They must be recognized and dealt wi th b y any moral theory that is not utopian. B u t this recognition does not a mount to acceptance of a biological foundation fo r ethics. It is no more than an acknowledgment that morality, like a n y other p rocess of cultural development, must reckon with its starting points and with the nature of the materials i t is a t tempting to transform.

11 B rain B isecti on and the Unit y o f Consciousness

There has been considerable opt1m1sm recently, a mong philosophers a n d neuroscientists, concern i n g the prospect for

maj o r discoveries about the !l europ h ysiol9gi/em. ( E n gelw o od C l i ffs. N .J. : P re n ti c e- H a ll , 1 97 1 ) ; H i l a r y Putn a m , ' Psychological P red i c a tes ', in Art, Miud, & Religiou , ed. W. H. Capitan and D. D. Merrill (PittSburgh : University o f Pittsburgh Press. 1 967), reprinted in Materialism , ed. R o se nt hal , as 'The N a ture of Mental States ' ; D. M . Armstrong, A Materialist Theory oft h e lllti u d (London : Routledge & Kegan Paul, 1 968); V. C. Dennett, Colllerll aud Cousciousrress (London: Routledge & Ke gan Paul, 1 969). I h ave expressed earlier doubts i n ' A rmstrong on the M i n d ' . Phil.phical R£'view, L X X I X ( 1 970). 394-403 ; a review of Dennett, joumal of Philc>sophy , L X I X ( 1 972); and cha pter I I a b o v e . See a lso Saul K ri p ke, ' N a m i n g and Necess i t y ' . in Semautics of Natural Laii}{IIO�e. ed. D. Davids.on a mi G. H a r m a n ( V o r d rech t : Rei d e l , 1 972), csp. pp. 334-42-; and M. T. Th o r n t o n, ' Ostensive Terms and Material ism', Tire• Morl i.< t , L V I ( 1 972) , 1 93-2 1 4.

1 66

Mortal questions

Every reductionist has his favorite analogy from modern science. It is most unlikely that any of these unrelated examples of su ccessful reduction will shed light on the relation of mi nd to brain. B u t philosophers share the general h u ma n weakness for expla nations of what is incomprehensible in terms suited for what is familiar and well understood, though entirely different. This has led to the acceptance of i m plausible accounts of the mental largely because they would permit fa miliar kinds of reduction. I shall try to explain w.h y � he Jl SUal exa mples do not help us to understand the relation between mind and body-why, indeed, we have at present no conception of what an explanation of the physical natu re of a mental pheno menon would be. Without consciousness the mind-body problem would be much Jess in teresting. With consciousness i t seems hopeless. The most i mportant and characteristic feature of conscious mental pheno mena is very poorly understood . Most reductionist th eories do not even try to explain it. And careful exa mination will show that no currently available concept of reduction is applicable to it. Perhaps a new theoretical form can be devised for the purpose, but such a solu tion, if it exists, lies in the distant intellectual fu ture. Conscious experience is a widespread phenomenon. It occurs at many levels of ani mal life, though we cannot be sure of its presence in the simpler organisms, and it is very difficult to say in general what provides evidence of it. (Some extremists have been prepa red to deny it even of m a m mals other than man. ) No doubt it occurs in countless forms totally uni maginable to us, on other planets in other solar systems throughout the universe. But no matter how the form may vary, the fact that an organism has conscious experience at all means, basically, that there is so mething it is like to be that organis m. There may be fu rther implications about the form o( the experience; there may even {though I doubt it) be implications about the behavior of the organism. B u t fundamentally an organism has conscious mental states if and only if there is something that it is like to be that organis m - something it is Jikefor the organism. We may call this the subj ective character o f experience. It is not captured by any of the fa milia r, recently devised reductive analyses of the mental, for all of them a re logically compatible with its absence. It is not analyzable in terms of any explanatory

What is it l ike to l1e

a ba t?

1 67

system of functional states, or intentional s tates, since these could be ascribed to ro b o ts or au t o ma t a that behaved like p e o ple though they experienced nothing.2 It is not analyzable in terms of the causal role of experiences in relation to typical hu man be h avior for similar reasons . 3 I do not d e n y that c o n s ciou s mental s tates and events cause behavior, nor that the y may be given functional characterizations. I deny only that t h i s kind of th ing exhausts their analysis. Any reductionist program has to be based on an anal ysis of what is to be re d uced . If t h e analysis le aves something out, t he probl em will be falsely posed. I t is useless to base the defense of materialism on any ana l ys i s of m en t al phenomena that fails to deal e x p lic i t ly with their subj ec­ tive character. For there is no reason to suppose that a reduction which seems p l a us i b l e when no attempt is made to account for consciousness can be extended to i n clude cons ciousness. Wi th­ ou t some idea, therefore, of wha t the subjec tive cha racter of experience is, we cannot know wh a t is requi red of phy s icalist theory. While an account of the ph ys ical basis o f mind mus t explain many thi ngs, this appears to be the m ost difficult. I t is i m po s si ble to exclude the pheno menological featu res of experience fro m a reduction in the same way that one excludes the phenomenal featu res of an ordinary su bstance fro m a phys i c al or che m ical reduction of i t - na mel y, b y explaining them as effects on the minds of human observers . 4 If physicalism is to be d efended, the pheno menological fe a t u res must t h emse l ves be given a p h ys i c a l account. But when we e xa m i n e thei r s u bjecti ve cha racter it see ms that such a res u l t is i mp oss i b l e . The reason is tha t eve r y subj ective ph e n o m e n o n is essentially c o n n ect e d w ith a single point of view, and i t seems inevitable tha t an obj ective, physical theory will abandon that point of view. -

­

there cou ld not a c w a l l y be such robots. Perhaps a n y thing enough to behave l i k e a person would have experiences. B u t that, if true, is a fact which cannot be discovered merely by analyzing the c o n c ep t of experience.

2 Perhaps

complex

is not equivalent to that about which we a re inco rrigible, b o th because we a re not incorrigible about experience and beca use ex perie n ce is p resen t i n a n i mals lacking language a n d thought, who have no belief.� a t all a b o u t

3 It

t h e i r e x p e riences.

� i ch a rd R o r t y, ' Mi n d-l:lod y Iden tity, Privacy, and Review of Metaplrysics , X I X (1 965), esp. 37-8.

4 Cf

Ca tegories ' ,

1 68

Mortal questions

Let me first try to s tate the issue so mewhat more fully than by referring to the relation between the s u bj ective and the objective, or between the pour soi and the en soi. This is far fro m easy. Facts about what it is like to be an X are very peculiar, so peculiar that some may be inclined to doubt their reality, or the significance of claims about them. To illustrate the connexion between subjectivit y and a point of view, and to make evident the importance of subj ective features, it will help to explore the matter in relation to an example that b rings out clea rly the divergence between the two types of conception, subj ective and objective. I assum e we all believe that bats have experience. A fter all, they are ma mmals, and there is no more doubt that they have experie nce than that mice or pigeons or whales have experien ce. I have chosen bats instead of wasps or flounders because if one travels too far down the phylogenetic tree, people gradually shed their faith that there is experience there a t all. Bats, although more closely related to us than those other s pecies, nevertheless present a range of activity and a sensory a pparatus so different fro m ours that the problem I want to pose is exceptionally vivid {though it certainly could be raised with other species) . Even withou t the benefit of philosophical reflection, anyone who has spent some time in an enclosed space with a n excited bat knows what it is to encounter a fundamentally alien form of life. I have said that the essence of the belief. that ba ts have experience is that there is so mething that it is like to be a bat. Now we know that most bats {the microchiroptera, to be precise) perceive the external world primarly by sonar, or echolocation, detecting the reflections, fro m objects within range, of their own rapid, subtly modulated, high-frequency shrieks. Their . brains a re designed to correlate the outgoing i mpulses with the subsequent echoes, and the information thus acquired enables bats to make precise discri minations of dis­ tance, size, shape, motion, and texture comparable to those we make by vision. But bat sonar, though clearly a for m of perception, is not similar in i ts operation to any sense that we possess, and there is no reason to suppose that it is subjectively like anything we can experience or imagine. This appears to create difficulties for the notion of what it is like to be a bat. We must consider whether any method will permit us to extrapolate

What is it like to be a bat?

1 69

to the inner life of the bat fro m our own case, 5 a nd if not, what alternative methods there may be for understanding the notion. Our own experience provides the basic material for our imagination, whose range is therefore limited. I t will not help to try to i magine that one has webbing on one's arms, which e nables one to fly around at dusk and dawn catching insects in one's mou th; that one has very poor vision, and perceives the surrounding world by a system of reflected high-frequency sound . signals; and that one spends the day hanging upside down by one's feet in an attic. Insofar as I can imagine this (which is not very fa r), it tells me only what it would be like for me to behave as a bat behaves. But that is not the question. I want to know what it is like for a bat to be a bat. Yet if I try to i magine this, I a m restricted to the resources of my own mind, and those resou rces are inadequate to the task. I cannot perform it either by imagining additions to my p resent experience, or by i magining segments gradually subtracted from i t, or by i magining some combination of additions, subtractions, and modifications. To the extent that I could look and behave like a wasp or a bat without changing my fundamental structure, my experiences would not be anything like the experiences of those animals. On the other hand, it is doubtful that any meaning can be attached to the supposition that I should possess the internal neu ro­ physiological constitution of a bat. Even if I could by gradual degrees be transformed into a bat, nothing in my present constitution enables me to imagine what the experiences of s u c h a fu ture stage of myself thus metamorphosed would be like. The best evidence would come fro m the experiences of bats, if we only knew what they were like. So if extrapolation from our own case is involved in the idea of what it is like to be a bat, the extrapolation must be mcomplet­ able. We cannot form more than a schematic conception of what it is like. For example, we m a y ascribe general types of experience on the basis of the ani mal 's structure and behavior. Thus w e describe bat sonar as a form of th ree-d i m ensional fo rward percepti on ; we b el i eve that bats feel some versions of pain, fea r , h u nger, a n d l u s t, and that they have o t h e r , m o r e fa m i l i a r types 5

By 'our own case' I do not mean j ust 'my own case ', but rather the mentalistic ideas that we apply unproblematica l l y to ourselves and other human beings.

Mortal questiom

1 70

of perception besides sonar. But we bel ieve that these experi­ ences also have in ea ch case a speci fic s u bjective chara cter, which i t is beyond our ability to con ceive. And if there is conscious life elsewhere in the u niverse, i t is l i kely that some of it will not be describable even in the most general experiential terms available to us.6 (The problem i s not con fined to exotic cases, however, fo r it exists between one person and another. The subjective character of the experience of a person deaf and blind fro m birth is not accessible to me, for exa m p le, nor p res u m ably is mine to h i m . This does not prevent us each from believing that the other's experience has such a subj ective character. ) If anyone is inclined to deny that we can b e li e ve in the existence of facts like this whose e x a ct nature we cannot possibly conceive, he should reflect that in contemplating the bats we are i n m uch the same position that int e l l igent bats or Martians7 would occupy if they tried to for m a concep tion of what it was like to be us. The structure of their own minds might make i t i m possible for t h e m t o su cceed, b u t w e k n o w t h e y wou ld b e w rong to concl u de t h a t there is n o t anything precise t h a t i t is like to be us : that on l y certain general types of mental s tate could be ascribed to us (perh a p s perception and ap petite would be concepts co mmon to us both; perhaps not). We know they would be wrong to draw s u c h a s k e p t i cal concl usion because we know what it is like to be us. And we know that while it includes an enormous a m ou n t of variation and complexity, and while we do not possess the vo c a b u lary to describe it ad eq ua tely its subj ective character is highly s pecific, and i n some respects describable in terms that can be understood only by creatures like us. The fact that we cannot expect ever to accom modate in ou r l a n guage a detailed d e s cr ip tion of Martian or bat pheno menology should not lead us to di smi s s as m ea n i ngl ess the clai m that bats and Martians have experiences fully comp ar able in richness of detail to our own. It would be fine if someone were to develop concepts and a theory that enabled us to think about those things; but such an understanding may be perma­ nently denied to us by the limits of our nature. And to deny the ,

6

7

Therefore the analogical form of the English expression 'what it is like' is misleading. It does not mean 'what (in our experience) i t resembles ', but rather 'how it is for the subject himself'. Any intelligent extraterrestrial beings totally different fro m us.

What is it like to be a bat?

1 71

reality or logical significance of what we can never describe or understand is the crudest for m of cognitive dissonance. This brings us to the edge of a topic that requires much more discussion than I can give it here : namely, the relation between facts on the one hand and conceptual schemes or systems of representation on the other. My realis m about the subjective domain in all i ts forms implies a belief in the existence of facts beyond the reach of h u man concepts. Certainly it is possible for a hu man being to believe that there are facts which hu mans never will possess the requisite concepts to represen t or com­ prehend. Indeed, it would be foolish to doubt this, given the finiteness of hu manity's expectations. A fter all, there would have been transfinite nu mbers even if everyone had been wiped out by the B lack Dea th before Cantor discovered them. But one might also believe that there are facts which could not ever be represented or comprehended by human beings, even i f the species lasted for ever - simply because our s tructure does not permit us to operate with concepts of the requisite type. This impossibility might even be observed by other beings, but it is not clear that the existence of such beings, or the possibility of their existence, is a precondition of the significance of the hypothesis that there are humanly ina ccessible facts. (After all, the nature of beings with access to hu manly inaccessible facts is presu mably itself a hu manly inaccessible fa ct. ) Reflection on what it is like to be a bat seems to lead us, therefore, to the conclusion that there are facts that do not consist in the truth of propositions expressible in a human language. We can be compelled to recognize the existence of such facts without being able to s tate or comprehend them. I shall not pursue this subject, however. I ts bearing on the topic before us (namely, the mind-body problem) is that it enables us to make a general observation about the subjective cha racter of experience. Whatever may be the status of facts about what it is like to be a human being, or a bat, or a Martian, these appear to be facts that embody a pa rticular point of view. I a m not adverting here to the alleged privacy of experience to its possessor. The point of view in q u e s ti o n is not one accessi ble onl y to a single individual. Rather it is a typ e . It is often possi ble to take u p a point of view other than one's own, so the com prehension of su ch facts is not l i m i ted to one's own case.

1 72

Mortal questions

There is a sense in which pheno menological fa cts a re perfectly obj ective: one person ca n know or say of another what the quality of the other's experience is. They are subjective, how­ ever, i n the sense that even this objective ascription of experience is possible only fo r someone sufficiently similar to the obj e c t of ascription to be ab l e to adopt his point of view - to understand the ascription in the first person as well as in the third, so to speak. The more different fro m oneself the other experiencer is, the less s uccess one can expect with this enterprise. In our own case we occupy the relevant point of view, but we will have as much difficulty u nderstan ding our own experience p roperly i f we approach it fro m another point of view as we w o u l d if we tried to understand the experience of another species without taking u p its point of view . s This bears dir e ct l y on the mind-body pro b l e m . F o r i f the fa cts of experience - facts abou t what it is like for the e x p e r i e n c i ng organism - are accessible on l y fro m one point o f view, then i t is a mys tery how the true character of experiences c o u l d b e revealed in the physical operation of that organis m. The latter is a d o m a i n of objective facts par excellence - the kind that can be observed a n d understood fro m many points of view and by individuals with differing p e r ce p tua l s ystems. There a re no comparable i maginative obstacles to the acquisition of know­ ledge a b o u t bat neurophysiology by human scientists, and intelligent bats or Martians might learn more about the h u man brain than we ever will. This is not b y i tself an argu ment against reduction. A Ma rtian 8

It may be e a sie r than I suppose to transcend inter-species barriers with the aid of the i magination. For ex a m p le, blind p eo pl e are able to detect objects nea r them by a form of sonar, u s i ng vocal c l i c k s or taps of a cane. Perhaps if one knew what that was like, one cou l d by extension i magine roughl y

what it was like to possess the much more refined sonar of a bat. The distance between onesel f and other persons and other species can fall anywhere on a continuum. Even for other persons the understanding of what i t is like to be them is only partial, and when one moves to s pecies very different from oneself, a lesser degree of partial understanding may still be a vailable. The i magination is remarkably flexible. My point, however, is not· that we cannot know what it is like to be a bat. I am not raising that epistemological problem. My point is rather that even to form a conception of what it is like to be a bat (and a fortiori to know what it is like to be a bat) one must take up the bat's point of view. If one can take it up roughly, o r partially, then one's conception will also be rough or partial. Or so it seems in our present state of understanding.

What is it like to be a bat?

1 73

scientist with no understanding of visual perception could u nderstand the rainbow, or lightning, or clouds as physical phenomena, though he would never be able to u nderstand the human concepts of rainbow, lightning, or cloud, or the place these things occupy in our pheno menal world. The objective nature of the things pi cked out by these concepts coul d be apprehended by h i m because, although the concep ts themselves are connected with a pa rticular point of view and a particular visual phenomenology, the things apprehended from that point of view are not: they are observable fro m the point of view but external to it; hence they can be comprehended fro m other points of view also, either by the same organisms or by others. Lightning has a n obj ective cha racter that is not exhausted by i ts visual appea rance, and this can be investigated by a Martian without vision. To be precise, i t has a m o re objective character tha n is revealed in i ts visual appearance. In speaking of the move from s u bj ective to obj ective characterization, I wish to remain nonco mmi ttal about the exis tence of an end point, the com­ pletely obj ective i ntrinsic nature of the thing, which one might or might not be able to reach. It may be more a ccu rate to think of obj ectivity as a d i rection i n which the understanding can travel. And i n u n derstanding a pheno menon like lightning, i t is legitimate to go as fa r a way as one can from a strictly hu man viewp oint. 9 In the case of experience, on the other hand, the connexion with a particular point of view seems much closer. It is di fficult to understand what cou l d be mean t by the o�jective character of an experience, apart from the particular point of view from which its s u bject apprehends it. After all, what would be left of what i t was like to be a bat if one removed the viewpoint of the bat? But i f experience does not ha ve, in addition to its subj ective character, an obj ective nature that can be apprehended from many different points of view, then how can it be supposed that a Martian investigating my brain might be observing physical 9

The problem I a m going to raise can therefore be posed even if the distinction between more subjective and more objective descriptions or viewpoints can itself be made only within a l arger human point of view. I do not accept this kind of conceptual rela tivism, but it need not be refu ted to make the point that psychophysical reduction cannot be acco m m odated by the subjective-to-objective model fa miliar fro m other cases.

Mortal questions

1 74 processes w h i c h w e r e

m y m e n ta l p r o c ess es ( as he m ig h t obs e r v e

p h y s i c a l processes w h i ch were b o l t s of lightning), o n l y fro m a

d i fferent p o i n t of v i e w ? How, for that m a t ter, c o u l d a h u m a n

physiologist o b s e rve them fro m another point of view ? I O We a p pear t o b e fa c e d with a g e ne r a l d i ffi c ul t y about psychophysical reduction. In other a r e a s the process of redu ction is a m ov e in the d i rection o f greater o bjectivity, t o w a r d a m o re . a ccurate view of the r e a l n a t u re of things. This is a c c o m pl i s hed b y red u ci ng ou r d e p e nd e n ce on i n d i v i d u a l o r species-specific points of view t o w a r d the o bj e c t of i n vestiga t i o n . We d e s c r i b e it n o t i n terms of the i m p re s s i o ns i t m a k e s on our senses, b u t i n t e r m s o f i t s more ge n e r a l effects a nd o f p r o p e r t ies detecta ble by means o t h e r than t h e h u m a n senses. The l ess it d e p e n d s on a s p e c i fi c a l l y h u ma n v ie w p o int , the more o bj e c t i v e is ou r d es c ri p ­ tion. It is possible to follow this path beca use a l t h o u g h the concepts and ideas we e m p l o y in t h i n k i n g about the external world are i n i t i a lly a p p l ied fro m .a point of view that i n v o l v e s our p e r ce p t u a l a p paratus, they are used by us to refer to thi n g s be y o n d themselves - toward which we have the ph en o m e n a l p oi nt of view. Therefore we can a b a nd o n it in fa vor o f a n o t her, a n d still be thin k i ng a bo u t t h e same things. Experience itself. h o we v e r , does not see m to fit t h e pattern. The i d ea of moving from a p p e a r a nc e to reality seems to make no se n s e here. What is t h e analogue in t h i s case to pu r s u i n g a more objective und erstanding of the s a m e p h eno m e na by ab a n don in g the i n i tial s ubj ective vie w po i n t toward t h e m in fav o u r of a n o t h e r that is m o re o bj e ct i ve but concerns the same thing? Certainly i t app ears u n l i k e l y that we will get closer to the rea l n a t u r e o f h u man e x p erie n c e by l e a vi ng behind t h e particularity of o u r h u m a n p oi n t o f view and s t r i v i n g fo r a description in t e r m s a ccessi ble to beings that could not i magine what it was like t o be u s. If the s u bj e c t i v e ch a ra cter of e x p er i en c e is ful l y co m p rehensible only fro m o ne p oi n t of view, t hen any s h i ft t o g r e a t e r obj e c t i v i t y - that is, less a t t a c h me n t to a s p e c i fi c view­ point - doe s not take us ne a r e r to the real nature o f the p h e n o me no n : it t a k e s us fa rther away fr o m it. 10

T he problem is not just tha t w h e n I look a t t h e MotJa Lisa , my visual ex perience has a certain quality, no trace of which is to be found by so meone looking i n to m y brain. For even i f he did observe there a tiny image of the Mot1a Lisa , he wou l d h ave no reason to identify i t with the experience.

·

Wh at

is

it like to be a bat?

1 75

In a sense, the seeds of this o bje c t i o n to the red uci bil i ty o f ex perience a re a l r ea d y detectable i n su ccessfu l c a s e s of r e d u c tion; for i n discovering sound to b e , in re a l i t y , a wa v e p h e n o m e n on in air o r other m e d i a , w e leave be hind one v i e w po i n t to take up ano ther, and th e a u d i to r y hu man or animal v i e w p o i n t that we lea ve behind r e m a i n s u n re d u ce d Members of r a d ical l y d i ffer e n t s p e ci e s may both understand the s a m e p h y s i c a l events in o bjec­ tive t e r m s and this docs not requ i re that t he y understand the pheno m en a l fo rms in wh i c h those events a p p e a r to th e s e n s e s of m e m be r s o f the other s p e c i e s Th u s it is a condition of thei r referring to a co m mo n r e a l i t y that thei r m o r e p a r t i c u l a r view­ points a re not p a r t of the co m m o n rea l i t y t h a t they bot h a p p rehen d The re d u ct i o n can s u c c e e d only i f the species-speci fi c vi e w p o i nt is o m i t t e d fro m what is to be red u ced. But while we a re r i g h t to leave this ·poi n t o f view aside in s e e ki n g a fu l l e r u n dersta n d i n g o f t h e external world, we cannot i g nor e i t p e r m a n e ntl y , s i n ce i t i s the e ss e n c e of the i n ternal world, a n d not merely a p oi n t of view on it. Most of the neobe h a v i o r i s m of recent phi losophical p s y c h olo g y res u l ts fro m t h e effort t o s u bsti tute a n o bj e c ti v e concept o f m i n d for t h e rea l th i n g in order to h a v e not h i n g left o v e r which cannot be red u c e d If we acknowl edge that a p h y s i ca l t h e o r y of m i n d m u s t a c co u n t fo r the s u bj e c t i ve character of e xp e r i e nc e we m u s t a d m i t t h a t no p re s e n t l y a vail able c on c e p ti on gives us a clue how this could be done. T h e problem is unique. I f mental p r o ces s es are i n dee d p h ysi c a l processes, then there is so m e t h i n g i t i s like, intrinsically, I I to undergo certain p h y si ca l p rocesses. W h a t it is for such a t h i n g t o be the c a s e remains a m y ster y . ,

.

,

.

.

,

.

,

11

The rel a tion would therefore not be a contingent o ne. l i k e t h a t of a cause and its d i s t i n c t effect. I t wou l d be necess a r i l y t r u e that a certa i n p l1ysical

s t a ll'" fel t a c e r t a i n way. S a u l K ri p k e i n S!"llwlllic.< of N,uural L.w.�UII)lt', (ed. D a v i dson and H a r m a n ) a rgues t h a t causal behaviorist a n d r e l a t e d a n a l y ses o f the mental fa i l because t h e y const rue. e . g ' pa i n ' as a mere l y c o n t i ngent n a m e of p a i n s . The subjective character of a n experience (' i ts i m mediate p h enomenolog i c a l quality' K ri p k e c a l l s i t (p. 340) ) is the c:ssenti a l p roperty left o u t by s u c h a n a l y s es . and the one i n v i rt u e o f w h i ch i t i s , nccessui l y . the e x p e r i e n ce it is. My v i e w is c l osel y rda ted to h i s . Like: K ripke. l fi n d the hypothesis that a certain brain s t a t e s h o u l d uea•.<sarily have a ce rt a i n s u bj e c t i v e c h a r a c t e r i n c o m preh e n s i b l e w i t h o u t fu rtl-acr e x p l a n a t i o n . No such expla n a t i on emer g es from theories which view t h e m i n d-b rain relation as c o n t i n gent, but p e r h a p s therl' a re ocher a l t e r n a t i ves, n o r yet discovered. A theory that e x p l a i ned how t h e m i n d-brain relation was necessary ..

1 76

Mortal questions

What moral should be d rawn fro m these reflections, and what should be d o ne next? It would be a mistake to conclude that physicalism must be fa lse. Nothing is p roved by the inadequacy of ph ysicalist hypotheses that assume a fau l t y obj ective analysis of mind. I t would be truer to say that physicalism is a position we cannot und erstand beca use we do not at p resent have any conception o f how i t might be true. Perhaps i t will be thought u n reasonable to require such a conception as a condition of understanding.

A fter all, it might be said,

the meaning of

physical ism is clear enough : men tal states a re sta tes o f the body; mental events are physical events. We d o not know

which

ph ysical sta tes a n d events they a re, but tha t should not p revent us from understanding the hypothesis. What could be clearer than the words ' i s ' and ' a re' ? B u t I believe it is precisely this apparent clarity of the word 'is' that is deceptive. Usually, when we are told that know

X

is Y we

how i t is supposed to be true, b u t -th a t depends on a

conceptual or theo reti cal backgrou nd and is not conveyed by the ' i s ' alone. We know how both

'X' and ' Y '

refer, and the kinds of

would still leave us with K ripke's problem of explaining why i t nevertheless appears contingent. T h a t d i ffi c u l t y seems to m e surmountable, in t h e following way. We may i magine something by representing i t to ou rselves either perceptually, s ympathetica lly, or s y m bolically. I shall not try to say how s y mbolic i magination works, but part o f what happens i n the other two cases is this. To i magine something perceptua l l y, we put ourselves in a conscious state rese mbling the state we would be in if we perceived it. To i magine something s y m pathetically, we put ou rselves i n a conscious state resembling the thing itself. (This method can be used only to imagine mental events and states - our own or another's . ) When we try to i magine a mental state occu rring without its associated bra i n state, we fi rst sympatheticall y i magine the occu rrence of the mental state: that is, we put ourselves into a state that resem bles i t mentally. A t t h e s a m e ti me, we a t t e m p t perceptually to i magine the nonoccu rrence of the associated physical state, b y putting ourselves· into a nother state unconnected with the first: one resembling that which we would be in if we perceived the nonoccu rrence of the physical sta te. Where the imagination of physical features is perceptual and the imagination of mental features is s y m pathetic. it a ppears to us that we can i magine any ex perience occu rring without its associated brain state, and vice versa. The rela tion between them will appear contingent even i f it is necessary, because of the independence of the disparate types of i magination. (Solipsism, incidentally, results if one misinterprets sympathetic i magination as i f i t worked like perceptual i magination: i t then seems i m possible to i magine any experience that is not one's own . )

What is it l ike

to

be

a bat?

1 77

things to w h i ch t hey refer, and we h a ve a rough idea h o w the tw o referential p a ths m i g h t converge on a si n g l e thing, be it an object, a person, a p rocess, a n eve n t o r wha tever. But when the two ter ms o f the iden t i fication a re ver y dispa r a te i t may not be so clear how it could be t rue. We m a y n o t h a ve even a rough idea of h o w t h e two referential pa ths could converge, or what k i n d o f thi ngs t h e y m i g h t converge on, a n d a theoretica l fra mework may h a ve to be supplied to ena ble us to u n derstand this. W i t h o u t t h e fra mework, a n a i r o f m ysticism surrounds t h e iden t i fi c a tion. This explains the magical flavor o f popular p resentations of fu n d a men tal scientific d iscoveries, g i ven o u t as p roposi tions to which one m ust subscribe with o u t real l y u n derstan d i n g them. For e x a m p l e, people a re now told a t an ear l y age that a l l matter is rea l l y energy. But despite the fact that they k n o w w h a t ' is' mea ns, m ost of them never for m a concep tion o f what ma kes this c l a i m t r ue, because they l a c k the theo retical b a c k g r o u n d . A t the p r esen t ti me the s t a t u s o f ph ysi calism is si m i l a r to t h a t w h i ch the h y p o t h esis that m a t te r is energy would have h a d i f u t tered b y a p re-Socra t i c p h i losopher. W e do n o t h a v e the beginni ngs o f a conception of how i t m i gh t be true. I n order to u n d ersta nd the h y p o thesis that a men t a l event i s a physical event, we req u i re m o re than a n understanding o f the word ' is ' . The idea o f h o w a mental a n d a p h ysi c a l term m i g h t refer to t he sa me thing is lacking, and the usua l a n a l og ies with t heo retical identifi­ cation i n other fields fail to su p p l y it. They fa i l because if we const r u e the reference of menta l terms to p h ysica l events on the usua l model, we either get a reappea r a n ce of sep a r a te subjective even ts as the effects through which mental reference to p h ysical events is secu red, o r else we get a false a ccou n t of h o w mental ter ms refer (for e x a m p l e, a causal beh a vio rist one) . S trangely enough, we m a y h a ve evid ence fo r the t r u t h o f so mething w e c a n n o t really u n derst a n d . S u p p ose a cater p i l l a r is locked in a sterile safe b y someone u n fa mi l i a r with i nsec t meta m o r p h osis, a n d weeks l a ter t h e sa fe i s reopened, revealing a b u t terfly. I f the person knows t h a t the sa fe has been sh u t the whole time, he has reason to bel ieve that the b u tterfl y is o r was once the caterpillar, w i t h o u t h a v i n g any idea i n what sense this might be so. (One p ossibility is t h a t the ca terpil l a r contained a t i n y winged pa rasi te t h a t devou red it a n d grew i n to t h e b u t­ terfl y . )

Mortal questions

1 78

I t is conce i v a ble that we a re i n s u ch a p osi t i o n with reg a r d to

p h y s i c a l i s m . D o n a l d D a v i d s o n h a s a rg u e d t h a t i f m e n t a l events have

p h ysi cal

causes a n d

descri p tions. He though

we

effects,

they

must

p h ys i c a l

have

h olds that we h a ve reason to be l i e v e t h i s even

do not - and in fa ct could not

-

have a general

ps yc h o ph ysical theory . 1 2 His argu ment appl ies to i n t e n t i o na l men ta l events, b u t I think we also h a ve s o m e reason to believe

that s e n s a t io n s a re physical p ro cess es, w i thou t being in a posi­

tion to u n d e rs t a n d h o w.

D a vids o n

'

s

p o si t i o n i s that certain

p h ys i c a l events have i rr ed u c i b l y mental p ro p e rties, and perhaps some v i e w

d es c r i b a b l e i n

w h i ch we ca n n o w fo r m

this way i s c o r r e ct . But n o t h i n g of

a c o n cept i o n c o rres p o n ds to it; n or

h a v e we a ny idea w h a t a theory would be l i ke tha t enabled us to conceive

of i t. 1 3 d o ne o n t h e b a s i c q ues tion (from b r a i n can be enti rely o m i t ted) whether any

V e r y l i ttle work has been which

m e nt i o n of the

sense can be m a d e of experien ces ' h a v i ng a n obj ective c h a racter · a t a l l . Does i t m a ke sense, i n other words, to ask what m y e x peri e n ce s a re

rea lly l i k e ,

as

o p p osed to

h o w they a ppea r to me?

We c a n n o t gen u i nely u n ders t a n d the h y p o t h e si s that the i r n a t u re

is

ca p t u red

in

a

physical

d e s c r i p t i on u n less we understa nd the

more fu n d a mental idea t h a t they

ha ve

an o bjective n a tu re (or

o bj e c t i v e p rocesses c a n h a ve a s u bj e c t i v e n a t u re) . l 4 I sh o u l d l i ke t o cl o s e wi t h

a

sp e c ul a t i v e p r o p o sa l

.

that

I t m a y be

p o s s i b l e to a p proa ch the gap between s u bj ecti ve and o bjecti ve fr o m a n other d i rection. S e t ti n g a s i d e

temporarily

be tween the mind and the bra i n, we can p u rsue

a

the relation

more o bj e c t i v e

understanding o f the mental in its own r i g h t . A t present we a re co m pletely unequi pped to thin k a b o u t the su bj ective cha ra cter o f experience w i t h o u t re l y i ng on the i m a gi n a tion - without taking u p the p o i n t o f view o f the experien t i a l s u bject. This 12

Sec ' Ment a l Even ts' in Experimce and Th rory , cd. L aw rence Foster and

]. W. S wa n s o n ( A m h e r s t : Uni versit y of M assachuse tts P r e s s 1 970) ; t h o u g h I do not understand the a r g u m ent aga i n s t p s y ch oph ysical l a ws. ,

13

14

Similar remarks a p ply to my paper ' P-h ysicalis·m · : · Phi)osophiml Revir111, L X X I V ( 1 965), 339-56, reprinted with pos tsc ript i n Modem lvfamia lism , ed. John O ' C o n n o r (New Y o r k : Ha rcourt B race J o vanovich, 1 969) . T h i s question also l i es a t t h e heart of the p r o b l e m o f other m inds, whose c l ose co nnection with t h e m i n d-bod y problem i s o ften o v e r l ooked . I f one u n derstood h o w s u bjective experience could h a ve a n obj ect ive na t u r e one would understand the e x i s t e n ce of subjects otht•r than onese l f ,

What

is it

like

to

a bat?

/Je

1 79

c h a l l e n g e to for m n e w co n ce p ts and pheno m e n o l o g y no t d ep e n den t on empathy or t h e i m a g i na t i o n T h o u g h p resu m a b l y i t s h o u l d be r e g a r d e d as

d e v ise a n ew

a

method - an o bj ective

­

.

wou l d n o t captu re e ve r y t h i n g , i ts go a l w o u l d b e t o describe, a t

l ea s t i n p a rt, t h e s u bj ective c h a ra c ter of

ex pe r i e n ces

a

in

fo r m

c o m p rehens i b l e to beings i n c a pa b l e o f h a v i n g those e x p e ri e n c e s We wo u l d h a v e to d e vel o p s u c h

the sonar experiences o f

bats;

a

.

p h e no m e n olo g y to describe

b u t i t would a ls o be

beg i n with h u mans. O n e m i g h t try, fo r

conce p ts t h a t co u l d be used to exp l a i n to

exa mple, a

p ossi b l e to

to d e v e l o p

p e r s o n b l i n d fro m

bi r t h w h a t i t was l i k e t o see. O n e w o u l d reach a b l a n k w a l l

eventu a l ly, b u t it s h o u l d b e po s s i b l e to d e v i s e a m e t h od o f e x p r es s i n g i n obj ective t e r m s m u c h more th a n we c a n a t p resent,

w i th much g re a ter p re c i s io n T h e l o o se i n ter m o d a l n a l ogi e s - fo r e x a m ple, ' Red is l i ke the s o u n d of a tru m p e t ' - w h i ch c r o p up in d iscussions o f t h is s u bj e c t a re of l i ttle use. T h a t s h ould be clear to a n y one who has b o th heard a tru m p e t a n d seen red . B u t s t r u c t u r a l fea t u res o f p e r c ep t io n m i g h t be more a c c es s i b l e to o bj e c t i ve descri p ti o n , even th o u gh s o m e t h i n g and

.

a

w o u l d b e left out. A n d concepts alternative to t h o s e we l e a r n i n

t h e fi r s t person m a y enable u s t o a r r i ve a t a k i n d o f unders tand­ ing even of o u r o w n e x p e ri e n ce w h i c h is denied us by the very

ease o f a ffo r d .

d e s c r i p t i o n and lack of d i s tance that s u bj e c ti v e concepts

A pa r t fro m i ts o w n interest, a p h e n o m e n o l o g y t h a t i s

in t h i s

s e n s e obj e c ti v e m a y p e r m i t q u e s t i o n s about the p h ys i ca J 1 5 basis

a more i n t e l li g i b l e for m . Aspects of a d m i t t e d this kind o f obj e ctive d es c r i p t io n might b e better c an d id a tes for obj e c t i v e e x p l a n a t io n s o f a m o re fa m i l i a r sore. B u t w h e t he r or not t h i s guess is correct, of ex pe ri ence to

s u bj ective

15

a ss u me

experi ence

that

I h�ve not defined the t e r m 'ph ysica l ' . Obvious l y i t does not apply just to what can be described by the concepts o f con temporar y ph y s i c s since we expect fu rther developments. Some may think there is nothing to p revent mental phenomena from eventua lly being recognized a s physica l in their own right. B u t whatever else may be said of the physical, i t has to be objective. So if our idea of the ph ysical ever expands to include mental phenomena, it will have to assign them an objective cha racter-whether or not this is done b y anal yzing them in terms of other phenomena a l ready regarded as physical. I t seems to me more likely, however, that mental-physical relations will eventually be expressed in a theory whose fundamental terms cannot be placed clea rly i n ei ther category. ,

1 80

Mortal questions

i t seems unlikely that any physical theory of mind can be contemp l a ted u ntil more thought has been given to the general problem o f subj ective and obj ective. O therwise we cannot even pose the mind-body problem without sidestepping it.

13 Panps ychi s m

By

p a n p s y ch i s m

I

mean

the v i e w

that the basic

physical

constituents o f the u niverse have m e n t a l p rope rties, whether o r n o t t h e y a re p a rts o f l i v i n g org a n i s m s . I t a p pears to fol l o w fro m a few s i m p l e p r e m ises, each of w h i c h i s m o r e p l a u s i ble t h a n i t s den i a l , though n o t perh a p s m o re p l a u s i b l e t h a n the denial o f p a n ps y c h i s m . 1 . Ma teria l compos ition

Any l i v i n g o r g a n i s m , i n c l u d i n g a h u m a n bei ng, ma terial s ys t e m .

is

a

co m p lex

I t consists o f a huge n u m be r o f p a r t i cles

combined in a special way. E a ch of us is co m posed o f m a tter t h a t had a l a rgely i n a n i mate h i s tory before fi n d in g i ts w a y o n to o u r p l a tes or those of o u r parents. It w a s o n ce p r o b a b l y p a r t o f the sun, but ma tter fro m another galaxy w o u l d d o a s well. ! f i t were brought to earth, a n d grass were grown

in i t, and m i l k fro m a

cow t h a t a te the g rass were d r u n k b y a pregn a n t w o m a n , then her chi l d ' s b ra i n would be partl y co m p osed o f t h a t m a tter. A n ything w h a tever, i f broken down fa r enough and rea rra n g e d , cou ld be i ncorpora ted into a living organis m . N o cons t i t u e n ts besides m a t ter a re needed.

2. Nonreductionism O r dina r y mental s t a tes like thoug h t, feeling, e m otion, sensation, or

d esire

beha vioral,

a re

not

physical

p h ys i o l o g i ca l ,

or

p r o p e r ti es

of

the

o t h e rwise - a n d

o r g a n is m -

they

are

not

i m p l ied b y p h ysical p r operties a lo n e. l 1

Strictly speaking, the argument req u ires only that �ome mental states are not reducible.

1 82

Mortal questions

3. Realism Nevertheless they are properties o f the organism, since there is no soul, and they are not p roperties of nothing a t alJ. 2

4. Nonemerg ence There a re no truly emergent properties o f co mplex systems. All p roperties of a co mplex system that are n o t relations between it and something else derive from the properties o f i ts consti tuents and their effects on each other when so combined. E mergence is a n epistemological condition: it means that an observed feature of the system cannot be deri ved fro m the properties cu rrently attri bu ted to its constituents. But this is a reason to conclude that either the system has fu rther constituents of which we a re not yet aware, o r the cons tituen ts of which we are a w a re have fu rther properties that we have not yet d iscovered. Panpsychis m seems to fol lo w fro m these fou r p remises. I f the mental properties of an organism are not i m plied b y any physical p roperties b u t must derive fro m properties o f the organi s m ' s constituents,

then those constituents must have

nonph ysical properties fro m which the appearance of mental p roperties fol lows when the combination is of the right kind. Since any matter can compose a n organism, all ma tter must have these p roperties. A n d since the same matter can be made into different types of organis ms with d i fferent types of mental l i fe (of which we have encoun tered only a tiny sa mple) , it must have pro perties

that

i mply

the

a ppearance

of different

mental

pheno mena when the matter is combined i n different ways. This would a mount to a kind of mental che mistry. The conclusion has its a ttractions as a general explanation of ho w conscious l i fe a rises in the universe. But there are three problems about the argu ment that I want to discuss.

1.

Why call these inferred properties of ma tter menta l ? What is mean t by a ph ysical p roperty and why does that concept not apply to them ?

2.

What view

of ca usality is i n volved in

the denial of

emergence? 2

Some of them, l ike belief and perception, are rela tional properties, bu t all involve some nonrelational aspect.

Paupsychism 3.

1 83

Do the featu res of mental phenomena that argue against reduction also a rgue against Realism?3

To deal with the first question, we must consider what makes a newl y discovered property or phenomenon physica l. Since the

class of known p h ysical properties is constantly expanding, the physical cannot be defi ned i n terms of the concep ts of con tem­ p orary ph ysics, but must be more general . N e w properties a re cou n ted as physical i f they are d iscovered b y expla natory i n ference from those already i n the class. This repeated process starts fro m a base of fa mi liar, observable spatia-temporal phenomena and p roceeds to take in mass, fo rce, kinetic energy, charge, valence, gravitational and electroma gnetic fields, quan­ tu m sta tes, an ti-particles, strangeness, charm, and wha tever physics will bring us next.4 What the a rgu ment clai ms is that a s i milar chain of explana­ tory i n ference beginning from fa miliar mental phenomena wou l d lead to general properties o f matter that would not be reached along the path of exp l a n a tory inference b y which physics is extended. Let us put aside for the moment the uneasiness that one may well feel a bout the sugges tion tha t mental phenomena should derive fro m any properties of matter at a l l . The c l a i m is t h a t i f such properties exist, t h e y a r c not p h ysical in the sense explained. N o properties o f the organism o r i ts consti tuents dis covered solely by p h ysics will be the fa miliar mental properties with their conscious or preconscious aspects, nor will they be the more basic proto-mental properties that i m p l y these; for i t will never be legiti mate to infer, as a theoreti cal explanation of phys i cal phenomena alone, a p roperty that incl u des or implies the consciousness o f its subject. We do i n fer explicitly mental ex p l a n a tions of ph ysical behavior, but these employ concepts understood i n d ependently and not intro­ d uced through physical theory. Theories constructed on the basis o f ph ysical observations and p arallels alone will not include t e r m s that i mply the consciousness o f the system. 3 4

I shall capitalize t h i s t e r m when usin g it in t h e special sense of premise 3. This is roughly equivalent to Feigl's ' ph ysical2'. Sec H. Feigl, 'The " Mental " and the "Physical" ' , Minn eso/a S1udie.< in thr Philosophy of Sdr11ce, vol. 1 1 , cd. H . Fei g l, M. Scri ven, and G. M a x w e l l (Minneapolis: U n i versit y of Minnesota Press, 1 958).

1 84

Mortal questions

It is this assu m p tion about inference that underlies the posit ion that the p h ysical will never i n clude the mental. If it is true, then i n the event that a n y properties of matter are discoverable by explana tory i n ference fro m observable mental phenomena, the y will have mental i mplications of a kind that p h ysically i n ferred p roperties will never have. In that sense the ulti mate p roperties i n ferred to explain mental p rocesses would be mental and not physi cal . Ho wever, this needs modification, for there is a third poss ibil­ ity. Perhaps there a re not two chains of i n ference, b u t one chain leading fro m the mental and the physical to a com mon sou rce. It i s con ceivable in the abstract that i f mental phenomena deri ve fro m the p roperties of matter a t all, those may be identical at some level with nonphysical properties fro m which physical phenomena also derive. This merits a brief digression. S u ch reducibility to a co m mon base would have the advan tage of explaining how there could be necessa r y causal connexions in either d i rection, b etween mental and physical phenomena. It would also make less problematic the possibility that a single event like a bodily movement could have both a mental cause and a complete p h ysical explanation. The mental cause, sufficiently analyzed, could be part o f the physical cause, s u fficiently a nalyzed . But i f. this were so, the common reducing p roperties would not be p h ysical. They could not be reached by a chain o f explanatory i n ference from physical phenomena a lone, for physica l data alone would provide no grounds for postulating explanatory theories

that also had

mentalistic consequences. The theories that physical data pro­ vide grounds fo r may take extraordinary leaps which permit the deduction o f radical physical consequences (the convertibility o f matter and energy, the deflection of light by gravi ty). But without any mentalistic evidence there is no reason to give mental content to the explanation of physical events. (Someone who infers fro m a d rought that the rain god is angry is nor basing his hypothesis on ph ysical evidence alone. He is making

a

psychological interpretation of the drought, based on fa miliarity with human motivation. Any i n ference o f this kind, reasonable or unreasonable, does not belong to physics. ) Therefore even .if there a re common ultimate p roperties underlying both the mental and the physical, they do not lie on the path of physical

Pa 1 1psych ism

discovery, the

1 85

path of e xplana t ory i n f e re nce fr o m o b s e r va bl e a re not physical p roper­

phys i c a l ph enom e na alone, and so th ey

ties.

I f th e re

w e r e s u c h p r o pert ies they would be d i s coverable only e x p l a n a to r y i n ference from both men tal and p h y si c al phenom e n a This seems in fact somewhat less i m plausible than t h a t there are two quite d i s tin c t chains of e x p lanation l ea d i n g back to t w o distinct sets of basic p r o p e r t i e s If i t were tru e, then it w o u l d be im p roper to describe the basic pr o p e rti es as men tal for the sa m e reason t h a t they cou ld not be described as p h ysi c al S t rictly, on ly w ha t is i n ferred to e x p l a i n mental p h e n o me na ( i n c l u d i ng actions) should b e called mental. This dea r l y ad mits con cep ts like rep res s i o n and u ti li ty function, or p e r h a p s uni ver­ s a l gra mmar.s They a p pea r at a level of psychological t he o r y not fa r re m o v e d fro m fa m i li ar m e n t a l processes . . B u t eve n i f by some c ri te r i on the fu nda mental particles had pr o p e r t ie s that were not m e n tal but neither mental nor p h ys i ca l the c o n c l u s i o n of th e argu ment wou l d s u r v i v e in a modified fo r m . There would b e p ro per ti es o f matter that were not physical f r o m w h i ch t h e m e n t a l pr o p e r ti es o f o rga n i c systems were derived. This c o u ld s till be cal led p a n p sy ch i s m ,

by

.

.

.

,

.

The s e c o n d q uestion is about caus ali t y and e m ergence. What

is t h e view

of causal e x p l a n a tio n fro m w h i c h it follows th a t true

em ergence is i m possible ? I have said that the p r o p e r t i e s of

a

c o m p lex

system must derive fro m the pro p e r t i es of its c o n­ sti tuents, p l u s t he way t hey a re combined. T h e a r g u m e n t assu mes that u n i fo r m correlations cannot p r ovi de a n a d e quat e basis for the e x pl a na t i o n o f c o m plex phenomena. It t h erefor e rej ects w h a t is of t e n called, i naccura tel y, a H u mean a n a l ysi s of causation. According to H u me our idea o f causal n e c e s s i ty is a kind of i l l usion, because all we ever o b serve a rc natu ral regu­ lari ties and correla tions, and never necessary connexions of cau s e and effect. H u m e d id not think t ha t our idea of c a u s e wa s tha t of a n .instance of a constant conj u n c tion in nature. ,

5

I have discussed the sense i n which such concepts o f psychological theory mental in ' Linguistics and Epistemology', in Language and Ph ilosophy, ed. Sidney Hook (New York : New York Universi t y P ress, 1 969), and in ' Freud 's A n thropomorph ism', i n Freud, ed. Richard Wollheim (New York : Doubleday, 1 974). are

Mortal questions

1 86

He was right, i n my opinion, to say that if this were all there was, then causality would be an illusion. B u t I do not believe it an illusion. True causes

is

do necessitate their effects : they make

them happen o r make them the case. Uniform correlations ar e at best evidence of such u nderlying necessi ties. This seems to

me

clearly true in elementary cases : heat causing water to boil, rocks causing glass to break, magnets induci n g electric current, the wind making wa ves. Given what heat is and what water is, it is

literally i mpossible for water to be hea ted beyond a certain point a t normal at mospheric p ressu re without boiling. Causal necessity opera tes even a t the most funda mental levels. An electron is a pa rticle with a certain charge and a certain mass. Those properties i mply that it will interact in a definite way with fields and with other obj ects. Some o f the i mplications will be probabilistic, b u t that does not a ffect the point. And si milar things are true o f other subatomic pa rticles. O rdinary physics and chemistry explain macroscopic phen o mena, so fa r as they can be explained, as the necessary consequences of the p roperties of the pa rticles (sometimes essential prope rties) and their interac­ tions. They do not rely merely on contingent correla tions. This is particularly clear when we consider the rela tion between p roperties of complex systems and p roperties of their components

at the same time. Consider the ph ysical p roperties of

a dia mond. Some of them, like shape, size, weight, a n d crystal structure, are di rectly entailed by the p h ysical p roperties anJ relations o f i ts constituents and their effects on each other when they are so combined. Others, like color, glitter, and hardness, involve i n tera ction between the d ia mond and other things, and must be explained in terms of the effects o f the dia mond 's constituents on those other things. The supposition that a dia mond or an organism should have truly (not . j u s t epistemologically) e mergent properties is that those properties appear at certain complex levels of organization but a re n o t explainable in terms of any more funda mental pro perties ; known o r un known, o f the constituents of the system. If causal connexions were nothing but instances of

contingent regularities, such a situation would be compatible with the existence of causal explanations of the emergent properties at a complex level. There would probably be many uniform psycho-physical correlations o f the form; ' Whenever an

Panpsych is111 o r ga n i s m is in

1 87

exactly p h y s i c a l s t a t e P i t is a l s o i n m e n t a l s tate M . ' of my p re s en t t o tal p h ysical and m e n t a l s t ates,

T h i s m a y be t rue

ex a mple .

ge n e r a l correlations also exist. be e n o u g h fo r M t o be ca u s a l ly expl a i ne d b y P . But i t i s not eno u g h o n a s tronger v i e w o f c a u s a t i o n . A s tronge r view req u i res t h a t P s o m e h o w necess itate M; bu t at t h i s c o m p l ex leve l , no n e ce s s a ry c o n n e x i o n s can be discovered . T h e r e is n o sense i n w h i c h m y b o d y ' s physica l s t a te by itse lf m a k es i t t h e case t h a t I a m in mental s ta te M. I t is o f cou rse obvious t h a t w h a t is g o i n g o n in m y b r a i n ca uses m y m e n ta l state, j u s t as it is o b viou s t h a t w h e n I to u ch a ho t p a n i t causes pa in . T h e r e must be some kind of necess i t y here. What we c a n n o t u n d ersta n d is how the heat, or the brain process, necessitates the sensa tion . S o long as we re m a i n a t the l eve l o f a p u rel y p h ys i c a l c o n c e p t i o n of w h a t goes on in the b r ai n , t h i s w i l l continue to a p pear i m possi b l e . The conclusion is t h a t unless we are prepared to a c c ep t the alternative t h a t the a p pearance of men t a l p ro p erties i n complex s ys t e ms has n o ca u s a l e x p l a n a ti o n a t a l l , w e m u s t take t h e cu rren t epistemological e mergence of the mental as a reason to bel ieve t h a t the constitu ents h a v e p ro p e r ties of w h i c h we a r e n o t a w a re, a n d which do necess i tate these res u l t s . T h e d e m a nd fo r a n acco u n t of how m e n t a l s t a tes necessarily a pp e a r i n p h y s i ca l or g a n i s m s cannot be sa tisfied by t h e dis c o ver y o f un i fo r m correl ations between m e n t a l s t a t e s a n d p h y s i c a l brain s t a t e s , t h ou gh that i s h o w psycho-physical laws have t ra d i t io n ­ a l l y been conceived. I n s te a d , in trinsic pr o p e rti e s o f t h e co m po­ n e n t s must be discovered from w h i ch the mental p ro p ertie s of the s ys t e m follow necessa rily. T h is may be u n a ttaina ble, but if mental p heno m e n a have a causal ex planation s u ch pro perties must exist, and t h e y will n o t be p h ysica J . 6 fo r

N o d ou b t m o r e

O n a correl a t i o n v i e w t h a t sh ou ld

T h e th i r d q u e s t i o n ,

about Rea l is m , is t h e m o s t d i ffi c u l t. W h a t deny t h a t m e n t a l p ro p e r t i e s can be e n t a i l e d b y physical ones ? I t i s certa i n l y co n c e i v a b l e t h a t t h e ph ysiological and b e h a v i o ral characteris tics of a living o r g a n is m should follow necessari l y fro m the physi c a l p r o per t ie s of funda mental pa rticles i s the reason to

6

The inference to such properties is not trivial, like the statement that opium puts people to sleep because it has a dormativc virtue. A lthough t he c a u s e s a re formula ted so as to email their effects, the r�:verse i mplication does not hold, as i t does in the joke.

1 88

Mortal questions

w h en

t hey

are combined in that way, though we can never

expect to possess more th a n fra gments of s u ch a n e x pla n a ti on

.

This is true also of functional sta tes, so called, i f they a re defined in terms of the i r relations to one a not h er , to s t i m u li, a n d to If the definition is general enough, the func t i o nal state

b eh a vi o r.

could appear in a wide va r iet y o f ph ysical sys tems, includin g

organisms whose beha vior took widely different forms. B u t i ts p resence cou l d

still b e entailed b y the physical m i cro p r operties

of any organism

in which

-

i t appe a r ed

.

A physical expl a n at i o n of beha v i o ral or fu nctional sta tes does not explain the mental because it does not explain its s u bjective

fea t u res: w h a t any conscious mental state is like for i ts possessor.

Let me say b riefly what I mean by this, t h o u g h it is too large a

topi c for proper discussion here. 7 A fe atu r e of experience is s u bj e ctive if it c a n i n pri nciple be fully u n d e rstoo d only fro m one type o f point o f view: t h a t of a being like the one h a v i n g the experience, or at least like it i n

the

relev a n t mod a lity. The

pheno menologi cal qualities o f our own experiences are s u bj e c

­

tive i n this way. The p h y s i ca l events in our brains a re not.

Hu m a n p h ysiologists may take a special i n terest in t h e m ; b u t

t hey can, i n p r i n ciple, be u n derstood j u s t as well, o r e v e n better, by creatu res t o t a lly unlike us in p h ysi c al and mental stru cture.

To u n derstand them such

c r e a tu r es need not

take u p our poi n t of

view. Physical brain p rocesses can be understood objectively, from the outside, beca use they a re not s u bj ective p h eno m en a

.

And no des c r i p t i o n or ana l ysi s of the objective nervous system,

however complete, will ever by itself i m pl y a n y thing which is

not objective,

i . e. which

ca n

be unders tood only from one kind

of viewpoi nt, that of the b ei n g whose sta tes are being descri bed . One cannot derive a p o ur soi fr o m an en

soi.

N o t all m e n t a l sta t es are conscious, b u t all of them a re capable

of producing s ta tes that are. So any derivation of the men tal

properties of a n organism fro m the p roperties of i ts components would have to show tha t s u bj ective states necessarily a rise fro m them. Of cou rse i f, as was sugges ted earlier, the explanation of

behavior lea ds u l ti mately to properties that are neither mental · nor physical , then a s u fficiently basic explanation of the p h ysic a l aspects of behavior might also explain subj ective experience as a 7

I try to give a fuller account of this idea in chapter 1 2 above.

Pa11ps yell ism

1 89

necessary part of the p rocess. B u t physical properties alone cou ld not gi ve this result; they e x plain not how t h i n gs are fro m a p articular subj e c t i v e p o int of view but how they a rc objectively, in ways that can be apprehended from different points of view and do not belong to any. This gap is logically unbridgeable. I f a bodiless god wanted to create a conscious b ei n g he cou ld not expect to do it by co mbining together i n organic form a lot of particles with none bu t physical properties. a Given a n a ccount o f the phe n o m e no logy of a particular kind of percep tion, it may be possible to d e d u c e how a particular obj e cti v e state of affairs wou l d app e ar from that point of view. B u t the su bj ec t i v e p remise se e m s essential. A n d this is no less true when the obj ective state is a physical brain s ta te, and the appearance is what i t is like to be in that brain sta te, rat her than what i t is like to observe i t. That, i n brief, is the argu ment against reductionism. B ecause of t h e way i n wh i c h it rel ies on t h e subj ectivity o f the m e ntal, I be l i eve that i t casts doubt on Realism, though I find this hard to ,

­

ex p lain.

For Realism as I ha v e defined it to be t r u e p h y s ical orga n is m s m u s t h a v e s u bj e cti v e properties . What seems una cceptable about this is tha t the organis m does not have a point of view : the person or crea ture does. I t seems absurd to try to discover the b a sis of the point o f view of the person in an atomistic breakdown o f t h e organi s m , because t h a t o bj e c t is not a possible s u bj e c t for the p o i n t of view to which the person's e x periences a ppear. And if it makes no sense to ascribe subj ective states to the co mplex whole, there will be no b asi s for ascri b i n g proto­ mental s tates to i ts constituents; so they cannot be a pp ea led to i n explanation of w h a t i t means for an organism to h a v e experi­ ences . I simply record this fee l i n g of i mpossibility because I have no more to say about it. When a mouse is frightened i t does not seem to me that a small material obj ect i s frightened. The trouble with this i n t u i t i on is that it le a d s nowhere. What is the alternative? I assume that neither I n o r the mouse has a soul to bear these mental p roperties. And even if we d i d , i t w o u l d n o t remove the p rob l e m because insofa r a s i t i s pos s i b le ,

,

,

8

C( Saul Kri pke, ' N a mi n g and Necessity', in Semamics of Natural ed. D. Davidson and G. Harman (Dod rech t : Reidel, 1 972), pp . 340-1 .

Language,

1 90

Mortal questions

to g rasp the idea of a nonmaterial thing, there is j us t as m uch difficulty in understanding how it could have a point of v i e w B u t i f th e occurrence of a subj e ct i v e e x p er i ence is not th e possession of a p r op e rty by someth ing , what is it? A n d what connexion does it have with the organis m ? Evidently in some way ex periences depend on the ma terial organism even if t hey are not s t a tes of it. T h e only view I know o f that may q ualify as a n a l t e r nati v e is fou n d in the Philosoph ical Investigations . According to Wi ttg e n s tei n as I understand h i m the person (or mouse) who is t he subj ec t of mental sta tes is not to be identified with a n organism or a soul or a nythi ng else. He holds tha t all kinds of fa miliar proposi tions about the mental s ta tes o f i n d i v i d u a l living be i ngs are true, but tha t there is al most nothing to be said about what property must be possessed by what th i n g if one of these a scri p t i o n s is to be tr u e . A l l s u ch specifications of t r u th conditions a re tri vial. What can be more fully described, h owe v er are the kinds of circu m­ stances, including evidential grounds, tha t make the a s c ri p tion a p p r o p riate : criteria rather than truth conditions. For thi rd­ p e r s o n ascriptions the grounds a re behavior, sti muli, circum­ sta nces, and testi mony (once the subject has lea r ned the relevant mental vocabulary). For self-ascriptions n o evidential grounds are needed. Although facts about the body a re a mong the criteria for ascri bing m e n t a l s ta tes to others, a n d also fo r a s c ri b in g to them an u n de rst a n di ng o f the terms they use to ascribe me n tal states to themselves, the m ental s t a tes a rc not s tates of the body. The view is not redu ctionist. Mental states a re no less r e a l than behavior, physical sti muli, and physiological p rocesses. In fact their situation with respect to one another is sym metrical, because physical p rocesses have mental (specifically observa­ tional) criteria just as mental p rocesses have physical cri teria. According to Wittg en s te i n e ve rythi n g there is must be sys­ te matically connected w i t h other th ings in a way that per m i ts pub l ic agreement, or at least public d isagreement, about whether it i s there or not. Mental pheno m ena meet this condition through thei r connexion wit h behaviour and circumstances, but they a re perfectly real in t hei r own right. They cannot be anal yzed as dispositions to b e ha vi or or properties o f t he organ­ ism, any more than p h ys ical ph e no m e na can be a n a ly z ed as .

­

,

,

Par1psychism

191

mu l tiple possibilities o f sensa tion o r o f observation. I f asked to say what any o f these kinds of thing really about them really

is, or what statements assert, we can give no reply that is not trivial.

I n some ways that is an attractive position. It does j u s ti ce to the subj ectivity of the mental, because o f the central place i t assigns to criterionless mental self-ascriptions. H o w things appear to so meone must h a n g together with h o w they appear to others to appear to h i m ; but these fa cts are i nextrica bly con­ nected with his point of view, as this can be p u b l icly identified. There is clea r s u pport for the idea that mental states a re subj ective if they a re ascribed to creatu res who can ascribe them to themselves without observation, b y other creatu res who can ascribe si milar states to

themselves i n the same way. And since it

does not seem correct to describe these sta tes o f the individual as sta tes o f the organi s m , this idea p rovides an a l ternative to Realism. M y difficulty with the view is that i t depends too heavily on our language. Essentially i ts account of mental phenomena is an account of how t h e y are ascribed, pa rticularly i n t h e fir s t person. B u t not all conscious bei ngs are capable of language, a n d that leaves the difficult problem of how this view acco m modates the subj ectivity o f their mental s tates. We ascribe experience to a n i m als on the basis of thei r beh a vi o r, structure, and circums tances, b u t we are not j u s t ascribing to t h e m behavior, s tructure, and circu mstances. So wha t are we saying? The same kind of thing we say of people when we sa y they have experiences, o f cou rse. But here the special relation between firs t- and thi rd-person ascription is not available as an ind ication o f the s u bj ectivity o f the mental. We are left with concepts that a re anchored i n their appl ication to hu mans, and that apply to other creatu res by a natural extension fro m the behavioral and contex tual criteria that operate in ordinary h um a n cases. This seems definitely unsatisfactory, because the ex periences of other creatu res are certa inly independent of the reach of an analogy with the hu man case. They have their own reali t y and their own subjectivity. They a re not, I assu me, o f indeterminate cha racter in cases where the natural extension from hu man beha vior and circu ms tances g ives no deter minate result. To take a very clear case, i f things e merged fro m a spaceship which we

Mortal questions

1 92

could not be sure were machines or conscious beings, what we were wondering about would have an answer even if the things were so different fro m a n ything we were fa miliar with that we could never discover it. It would depend o n whether there was s o m ething it was like to be them, not on whether behavioral similarities warranted our saying so. This seems true quite apart fro m the question o f what the s u bject of mental s tates is. They may not be states of the body, b u t they certainly exist in forms beyond the reach o f our language. So they cannot be anal yzed i n terms o f human cri teria for their ascription. And since h u m a n experiences have the s ame kind of reality, must n o t the same be true of the m ? What they are is not fully captured by an acco u n t of the conditions u nder which first- and third-person ascriptions o f experience are appropriate. I will mention that this raises problems about whether the concep t of experience, as I am applying it, meets basic conditi ons o f p u blicity that i t rriust meet to be well-defined a t all. I t is widely accepted that one cannot always define a type of similar­ ity or a type of thing si mply by pointing to an instance and saying ' the same as

this ' . And i t may be doubted whether

so meone who wonders whether the things co ming out of the spaceship have experience, withou t any idea o f the possibility of determining whether they do o r not, is really asking a well­ d efined question. I think that in this case the conditions of meaning are met, b u t I will not try to defend the cla i m here. Experience m u s t have systematic connexions with behavior and circu mstances in order for experiential qualities and experiential si milarity to be real.

But we need not know what these

connexions are in order to ask whether experience is present in an alien thing. I therefore seem to be drawn to a posi tion m o re ' realistic' than Wittgens tein's. This may be because I am drawn to positions more realistic than Wittgenstei n ' s a b o u t everything, not j u s t the mental. I believe that the question about whether the things coming out o f the spaceship a re conscious m u s t have an answer. Wittgenstein would presu mably say that this assu m ption reflects a grou ndless confidence that a certain picture u na mbiguously determines its own application. That is the picture o f so methin g going on in their heads (or whatever they have in place of heads)

Panpsych ism

1 93

that cannot be observed by d issection. Whatever picture I may use to represent the i dea, i t does seem to me that I know what i t mea ns to ask whether there is something i t is like to be them, and that the answer to that question is what determ ines whether they a re conscious - not the possibility of extending mental ascriptions on evidence analogous to the human case. Conscious mental sta tes a re real states of so mething, whether they are mine or those of an alien creature. Perhaps Wittgenstein's view can acco m modate this intuition, but I do not at the moment see how. Where does this leave us? I have now expressed dissatisfaction with three alternative i nterpretations of mental s tates: that they a re sta tes of the body, that they are states o f the soul, and that all we can say about their essence is to give criteria or con ditions for their ascription. B u t what is left? I f they a rc real states of s o mething i n the world , i f they depend on what is going on in the creature's body, i f they are i n ti mately connected with stimuli and behavior, and if the creat u re does not consist of a body plus something else, what can experience be but states of the organ­ ism? I f one accepts real ism i n a broa d sense abou t mental states, it is very di fficul t to avoid Realism in the more s peci fic sense that forms a premise of the argument fo r panpsychism. This of cou rse expresses t h a t fa tal s tep i n the philosophy of mind, the argument by elimination. There is no reason to think that all poss i bilities have been thought of, so there is no reason to assume that a view is correct if all cu rrentl y conceivable al terna­ tives a re even more unacceptable. Still, when a mouse or a fl y or a man co mes to exist beca use matter has been combined in certa i n ways, the res u l ting mental s ta tes seem to h a ve to belong to the orga nism fo r want of a better home. Rea l is m may be the weakest premise i n the argument, but i t i s more plausible at the moment than its denial. I therefore believe th a t panpsych i s m should be added to the current list of mutually incompatible and hopelessly unaccep t­ able solutions to the mind-bod y problem. It can be avoided by denying any of the p remises of the argument. Denial of the first results i n dualism. This still leaves p roblems about the ca usal connexions between mind and body: ei ther (a) those connexions are pure correlations and not necessa ry; or (b) the body will have properties that necessitate mental effects i n the soul and effects of

1 94

Morta l qu estions

the soul on the body; or else (c) t h e soul will have properties that enable i t to be a c ted on by the body and vice versa. I f (b), then the b ody will have mental or a t l east non-p h ysi cal properties. If (c), then the soul will have phy s ical pro pertie s as well as mental ones. D enial o f the s e c ond p remi s e is fairl y co m m o n among con­ temporary phi l o s ophers, but the onl y motive I can see for acceptin g any of the res u l t in g kinds of red u ct i o n i s m is a des i r e to make the mind-bod y problem go away. N one of them has any in trinsi c plau s ibi l i t y . Denia l of the third p remise, Realism, i s m ore a ttractive but awaits the devel o p ment o f a viable alternati ve, some way of a d m i tting the reality of mental occu rren ce s without asc r ibing them to e ither organisms or s oul s a s su bj e ct s . Denial of the fourth p re mise, no n e m ergence, involves accept­ i n g the existence o f irreducible contingent law s connecting co mplex organic states with mental states. In a s ense t h i s would mean that mental states had no causal explanation: that they were not necessi t a ted by a n ything. I do n o t believe the wor l d is like t hat, but here, as with the other p remises, one can take that escape route. It would b e useful to develop all the a l ternati ves mo re fu lly. As for panp s ychi s m, it is d i ffi c u l t to imagine how a chain of ex p lanatory i n ference cou ld ever get from the mental states of whole ani mals back to t h e proto- mental p roperties o f dead matter. It is a kind o f bn:akdown we cannot envi s ion, perhaps it is unintel l igible. Presu mably the co mponents out of which a poin t of v i ew is c o n s tructed would not themselves have to have poin t s of view. (How c ould a single sel f be composed of many selves?) Yet the y would have to be re c ombinable to form different points o f view, for not onl y can a s i ngle organism have d i fferent expe r ience s , b u t i ts m a tter can be recombined to form other organisms with tota ll y different forms of experience. The mental p roper t ie s o f all matter, therefore, would have to be not species-s pecific but universal, since they would underlie all possible forms o f consciousness. In a sense, they would be less subjecti ve than a n y of the specific forms. Panpsy c hism in this sense does not entail panpsychism i n the more fa miliar sense, accord i ng to which trees and fl owers, and perhaps even rocks, lakes, and blood cells have consciousne s s of

Paups y chis111

1 95

a kind. But we know so little about how consciousness a rises fro m matter i n our own case and that of the ani mals in which we can identify it that it would be dogmatic to assu me that it does not exist in other complex systems, or even in systems the size of a galaxy, as the result of the same basic properties of ma tter that a re responsible for us.9

9 M y ideas o n t h i s

t o p i c . especially on t h e conce p t o f the p h y s i c a l a n d t h e role of neces s i t y i n c a u , a l e x p l a n a t i o n , h a v e been s t r o n g l y i n fl u e n ced by

Rebecca G o l d s t e i n a nd William L. S t � n to n . Thei r o w n v i e w s a rc d e vel o ped in S t a n t o n ' s ' A n o m a l ous M o n i s m a n d ! ' h e M e nt a l Q u a M e n t a l ' (Ph. D . d i ssert a t i o n ,

Princeton University, 1 975)

a n d G o l d stei n ' s

' Reduction, R ea l i s m . a n d M i n d (Ph . D . d issert a t i o n , P r i n ceton '

U n i versity, 1 976).

14 Subj ective and Obj ective

1

There is a problem that e merges in several areas of philosophy whose connexion with one another is not obvious. I believe that it can be given a general form, and that some treatment of it is possible in abs traction fro m its particular ins tances - with results that can be applied to the instances eventually. This discussion is a p reliminary sketch for what I hope will be a more thorough treatment. The p roblem is one of opposition between subj ective and obj ective points of view. There is a tendency to seek an obj ective account of everything before ad mitting its reality. B u t often what appears to a more s u bj ective point o f view cannot be accounted for in this way. So either the obj ective conception of the world is incomplete, or the subj ective involves illusions that should be rej ected. Instead of trying to define these terms at the outset, I shall begin with s o me exa mples, d rawn fro m ethics and metaphysics. The parallels between them should e merge as I p roceed. Consider first a problem about the meaning of life. ! There is a way of considering hu man pursuits fro m within life, which allows j ustification o f some activities i n terms o f other s, b u t does not permit us to question the significance of the whole thing, unless we are asking, fro m within l ife, whether the allocation of energy or attention to different seg ments o f it makes sense in vi rtue o f their relative i m portance. This view comes under challenge fro m a position that regards life in detach ment from I

Sec chapter 2 above.

Su�;ective and o�iective

1 97

speci fi c or general hu man pu rposes . People, and oneself in particular, a re perceived as hav i ng no signi fi c a n c e a n d a bs u r d bec aus e the y seem to accord their lives g re at i m p or t an c e i n a c t i o n , e v e n th o ug h t h e y can also appreciate a b r o ade r po int of view fro m wh i c h th e y h a v e no i m portance. Each of the tw o p o ints o f view c l ai m s p r i o ri t y The internal view asks, wha t is the i m po r t a n ce fo r i n d i v idual l ife o f i ns ignifi ­ cance fro m an external p o int of view ? Life is lived from i nside, and issues of s i gni fi canc e are signi fica n t only if they can be raised fro m i nside. It therefore does not matter that from a poin t of view o u tside m y l i fe, m y l i fe does n o t ma tter. The external view, on the o the r ha nd , co m p r e h end s wi thin its scope of observation all the a i ms and com m i tments by re ference to which i nternal s ignifica nce is measu red. I t p r es e nt s i tself as the right way fo r the i n d ividual to look a t the world and his place i n i t : the b i g picture. He develops this kind of detach ment n a t u r al l y to cou n te r t h e ego c ent r ic distortion of a p u rely in ternal view, and to correct the parochialism e n g e n d e re d by th e con­ t i n g e nci es of his overs pe c i fi c nature and circums tances. Hut i t is not merel y c o r r e ctiv e It cla i m s a po s iti o n of dominance, as the o n l y co mp le t e concep tion of how thi ngs r e al l y a re. T h i s d o m i­ nance is not i m posed fro m ou tside, but deri ves fro m the in t r i nsi c appeal of i mpersonality to individual reflection. Life seems a b s u r d because i t see ms abs u rd to oneself, taki n g u p a point o f view that is both natural a n d appealing. The second ex a m ple to consider is the problem of free will. This p ro b l e m arises initially in the fo r m of a t h re a t to free ag en cy fr o m th e hy p ot he s i s that actions are determi ned by ant e ce d ent circu mstances. T he re have been ma n y a t tempts to anal yze agency in terms c o mp a ti b l e with deter minism by reference to inte nti o ns motives, second-order volitions, cap a c i ti es absence of obstacles o r coercion. Real advances have been made in s p e c i fy i ng necessary cond i tions o f agency, but the p os sibility that t h e s e conditions a re themsel ves determined see ms still to p r e s ent a threat to s o me element of the o rdi n a r y concep t of ac t i on 2 They may be necessary, but they do not seem s u fficient. ,

.

­

,

.

-

,

,

.

2

The litera ture on this subject is enormous. Three of the b e s t recent articles are P. F. S t rawson, ' Freedo m and Resentment', Proceedi11g.< of the British A cademy ( 1 962) ; H a rry G. Fra n k furt, ' F reedom of the Will and the Concept of a Person ' , joumal of Ph ilosop h y , t x m (January 1 4 , 1 97 1 ), 5--2 0;

Mortal questions

1 98

The next step, ho wever, is the discovery tha t fr e e a ge n c y is no t i m p l i ed by the absmce of d e t e r mi n is m, even t h o u gh i t ap pears t o be th reatened by t h e p resence of deter minism. Uncaused acts are no more attributable to the a g e n t tha n those caused by a n teced e n t ci rcu ms tances. One is th e re fo r e led to wonder what fu rther factor, in addition to the absence of determinism, is r e q u i r e d fo r fre e agency, and whether this fu r t her fac t or m i gh t not be s u ffi cient for freedom by i tsel f. The most d i fficult problem of free will is saying wha t the problem is, which seems to su rvive every attempt to specify s u ffi cient conditions for fre e action. The rece n t a t t e m pt to a n a l y ze a c t i on i n te r ms o f agmt causa­ t i on r a t h e r t h a n event ca u s a tion3 is ins tru ctive b e c a u s e i t r e v e a l s the true so u r ce o f discomfort w i th determinism. The problem is that when one views an action as an event causally con nected with other events, there is no room in the p i cture fo r so meone's doing i t But it turns out that there is no room for so meone's doing i t i f it is an event caus a l l y u n co n nect e d with other events, either. Hence some philosophers h ave tried to capture this aspect b y making an a g e n t ra the r than an ev e n t the cause. I d o not find the concept o f agent causation i n telligible, b u t I think I und e r stand its motiva tion. While i ts positive con ten t is obscu re, i ts n e g a t ive i m p l i ca ti ons arc clea r. I t removes a ction fro m the causal s e quen c e of events b y d e n yi n g tha t it is caused by antecedent ci rcu m s ta nces ; a n d b y substituting an agent as the cause, it a voids the al ternative that action is s o m eth i n g that j u s t happens . I t is a d oome d attempt to capture the doing of the action in a new kind of causa tion. But the problem is not that the i dea o f agency clashes with this o r t h a t p a r ti c u l a r conception o f what h a ppens in ac t i on viewed externally as a type of event. It is n o t predictabil ity tha t crea tes the pro blem, for I make many choices a n d do m a n y things th a t a re completel y predictable. I t is j ust that when I pick the shiny apple ins tead of the rotten one, i t is m y doing and there is no room fo r th i s i n an e x tern a l account of the event, deterministic or not. The real problem stems fro m a clash betw e e n the view of .

,

,

­

,

-

Gary Watson, ' Free Agency ', journal of Philosophy,

3

20Y-20.

LXXII

(April 24, 1 975),

Roderick M. Chisholm, ' Freedom and Action ', in Freedom and Determ inism , ed. K eith Lehrer (New York : Random House, 1 966).

1 99

Subjective and objective

action fro m inside and any view of it fro m outside. Any external view of a n act as something that happens, with o r without causal antecedents, seems to omit the doing of it: Even i f a n action is described in terms o f motives, reasons, abilities, absence of i m pediments or coercion, this does not capture the agent 's own idea o f himself as its source. His a ctions appear to him different fro m other things that happen i n the world, b ut not merely a different kind of happening, w ith different causes or none at all. They seem in s o m e indescribable way not to

happen at all (unless they a re quite out of his control),

though things happen when he does them. And i f he sees others ·as agents too, their actions will seem to have the same quality. The tendency to exp ress this conception of agency i n terms of freedom fro m antecedent causes is a mistake, b u t an understand­ able one. When the act is viewed u nder the aspect of deter mina­ tion by antecedents, its status as a n event beco mes p r o minent. B u t as appears upon further i nves tigation, no account of i t as an event is satisfactory fro m the internal viewpoint of the agent doing it. The connexion of this problem with moral responsibility is that when we view actions, our own or others ' , merely as part of the general course of events, it seems i mpossible to attribute them to individuals i n a way that makes sense of the attitudes we take toward so m eone we regard as the source of a n action. Certain a ttitudes toward the agent, rather than j us t abou t him, lose their footing. If a n individual is destructive enough we may think it would be better i f he did not exist ; b u t i f h e is just a disastrous part of the world, blame directed at h i m or guilt he di rects at himself make no sense, however causally or indeter­ ministically complex his behavior and moti ves are.4 The true nature of the third problem I want to mention - that of personal identity - is also hidden in many discussions. The problem is usually p resented as a search for the conditions that must obtain i f two experiential episodes separated i n time are to belong to a single person. Various types of continuity and similarity - p hysical,

mental, causal,

emotional - have

been

considered and they all seem to leave an aspect of personal identity unaccoun ted for. Given that any proposed set of condi4

These points a re discussed more fully in chapter

3

above.

Mortal questions

200

tions i s me t , there s t i l l seems to be a fu rther q u e s t i o n a s to whether the sa m e

su�iect

or

self i s p r e serv e d

u n der thes e condi­

tions. This fu rther q u e s t i on can be raised b y i m ag i n i n g t h a t you

h a v e the first of two e x p e ri ences a n d as k i n g a b ou t t h e o t he r ( w hich bears the candidate relation to i t ) , ' Yes, b u t w i l l i t be mil'l e?' As w i th free w i l l, t h e r e a l p ro b l e m seems to be to i d e n t i fy the p r o b l e m t h a t a l wa y s r e m ai n s no m a tter h o w i n gen i ous a solution h a s been proposed. I t may s e e m t h a t t h i s fu rther

of a

m et a ph y s i c a l

q ue s t i on i n volves the assu m p tio n

ego w h i c h pres erves p e rso n al i d e n t i t y B u t this .

would b e a m i s take, for the ego, i f i t

is

a co n t i n u i ng i n d i v i d ual

i d e n t i t y over ti me, w o u l d be j ust one more thi n g the sa me problem cou l d b e ra ised (will that ego s t i l l be me?). If on the o t h e r h a n d i t s o 1 1 l y i d e n t i t y o v e r t i me i s that o f s t i l l bei n g me, t h e n it can n o t be t h e i nd i v i d u a l w h o s e per s i s te n c e p reserves p e rso n a l i d e n t i ty For its i d en t i t y w o u l d then s i m p l y consis t i n the fact tha t e x p e r i e n ces ha d b y i t were a l l m i n e ; and

with i ts

own

a b o u t w h i ch

.

that ca n not

e x p l a i n wha t m a k es t h e m all m i ne.

The p ro bl e m s e e m s u n real when persons a re viewed as bei ng s i n the wo r ld

,

whether physical

or menta l . They p e r s i s t and

ch a n ge th r ou g h t i me, a n d t h ose

are

th e

terms in

which th e y

m u s t be descri b e d . B u t as wi t h the p roblem o f free w i l l , t he pe r si s t en t dissatisfa c t i o n w i t h cand i d a te a n a l y se s deri ves fro m is

left

a

of

this fo r m

s u b me r ged i n ternal a s pect o f the p r obl e m w h i ch

u n t ou c h e d

by

a l l e x t er n a l trea t ments. Fro m t h e po i n t o f

view of the p e r s o n hi mself, the q u est i o n o f

h i s i d e n ti t y o r

n o n i d e n t i t y with s o meone u ndergoing s o m e experience i n t h e fu ture a p p ea rs to

have a co n te n t t h a t cannot be e x ha usted b y a n y

acco u n t i n terms of m e mory, s i m i l a r i t y of c h a r a c t e r

,

or p h ys i c a l

con t i n u i t y . S u ch a n a l y ses a r e never s u ffi cient, and fro m this poi n t of view t h e y m a y a pp ea r n o t e v en to s u p p l y

n ece s sa r y

condi tions for identity. When s o meone poses i n w a r d l y the q u estion fu t u r e experience was or wi l l be

his,

whether

a past or

he has the s e n s a t i on o f

p i ck i ng out so mething w h ose i dentity over ti m e is well defined, j u s t by concentrating o n h i s p r es e n t ex pe r i e n c e a n d s p e cify in g the temporal e x t e n s i o n of its subject . The concept of the self is a psychological one, a nd it is characteristic of s u ch c on cep ts to gi v e rise to the p h i l os o ph i ca l idea t h a t their s u bj e c ti ve e ss enc e exp r e s se d m o s t clea r l y in fi rs t p e rs on a p p l i c a t i o n s , is detachable ,

-

Subjective a nd objective

20 1

fro m more obj ective accompani ments and even to a considera b le extent fro m necessary connexion with o ther psychological phenomena. (Another exa mple: the conviction that i t is a perfectly well-defined but in principle unanswerable question whether sugar tastes like this to o ther peopl e. ) This may be an il lusion. I t m a y have no sense to speak of ' the same self as this one' i n complete detach ment fro m all external conditions. But i t is s t i l l t h e i nternal i d e a o f t h e self t h a t gives r i s e to the problem of personal identity. Any a ttempt to conceive persons completely as a kind of thing in the world persisting through time will come up agai nst th is obsta cle. The sel f that appears to the subj ect seems to disappear under external analysis. My fourth example is the mind-body problem. A pa rticularly d ifficult aspect o f that problem comes fro m the subj ective cha racter of exp erience. S o long as mental states are looked at obj ectively, in their causal relations to stimuli and behavior, no special issues a rise which do not arise about the ph ysical analysis o f other natural phenomena. Even p roblems o f i ntentionality may seem to be solu ble if one puts aside their subj ective aspect, for then one may be able to describe certain kinds of computers as in tentional systems. What seems i mpossible is to include i n a p h ysical conception of the world the facts about what mental sta tes are like for the crea tu re having the m . The creature and his sta tes seem to belong to a world that can be viewed i m personally and externally. Yet subj ective aspects o f the mental can be apprehended only fro m the point of view o f the creatu re i tself (perhaps taken u p by s omeone else), whereas what is physica l is simply there, and can be externally apprehended fro m more than one point of view.6 Is there any way of i n cluding mental phenomena i n the world as well, as part o f what is simply th ere ? Here too the idea of i mpersonally comprehensible reality asserts i ts clai m to dominance. We are not fa ced only wi th the problem of the relation between mind and body, or the inclusion of the mental in the physical world. The broader issue between personal and i mpersonal, or subj ective and objective, a rises also for a d ualist theory of mind. The question o f how one can include i n the obj ective world a mental substance having subjec­ tive properties is as acute as the question how a physical substance can have subjective p roperties. (, Sec

cha pter

12 a b o ve.

202

1\1/orta/ questions

The physical i s an ideal representa t i v e fo r the o bj e c t i ve i n genera l ; therefore m u ch obscurity has been s hed on the problem by fault y ana logies between the mental-p h ys i cal relation a nd

relations between the physical and other o bj e c t i v e aspects of

re a l i t y As deter m i n i s m i s a substitute for externa l i t y or o bj e cti v .

­

i t y i n p o s i ng the p r o b l e m of free w i l l , so the physical is a

s u b s t i t u te for o bj ectivity in posing the m i n d b o d y p r o b l e m All -

.

t h e d i s p u tes over ca usal r o l e , t h eo retical identi fica tion, a nd

fu nctional realization, while of i n terest i n them sel v e s , fail to give

exp ress i on to the central issue that m akes the m i n d-body p roblem so h a rd . A n d a s with free w i l l and personal i denti ty, the

internal elemen t re mains, even if ignored, as the true sou rce of persistent dissatisfaction with a ll p h ysical o r o t h e r e x ternal theories of the mind. A t the same ti me, th e idea that persons

(a long wi th e v erything about them) must b e pa rts o f o bj ec t i v e

reality co n t i n u es to exert its powerful appeal. Obj ectivity is n a t u r a l l y l i n ked with real i t y ; it is easy to feel that a n ything has to be l o c a t e d in the obj e c t i ve world i n order to qual i fy as real , and

that i t must have a s i ts real nature some character which,

whether physical o r not, can b e regarded i m person a l l y and e x te r na l l y The

.

final e x a mp l e I want to discuss comes fro m ethics, and

con cerns the d i fference between consequentialist and agent-centered views of

r ig h t

m o re

and wrong. A fa m i l ia r type of

o bj ection to u t i l i taria ni s m and o ther conseq uentialist views cha rges them with unj u s tifiably making questions about what to do s u bordinate to q u e s t i on s a b o u t w h a t would be best overall.

Such criticisms assert that a n ethical theory should leave s o me

room for each i ndividual to p u rsue his own l i fe without having to consider at every p o i n t how he is serving m ore co m p rehen sive goals; or else they u rge the need fo r certa i n res t ri ctions or

requ i r e ments on action that a re not j u s tified by th e i r contribu­ tion to the general good. In other w o r d s , both what is permi tted and what is requi red of a person can so meti mes deviate from

what would be best. I group these two rather different ex cep­ tions to consequentialism together beca use, w h ile th ey can also be opposed to each o ther, they deviate fro m t h e consequentialist v i ew poin t in the sa me di rection. This is clea r in the case o f per­

mission to p u rsue one's own l i fe, less clear in the case of gen­ eral requirements or restrictions on action, whatever the goal.

Subjective and objective

203

Utilitaria n i s m , or a n y other p u r e ly conseq uentialist view, is

very demanding . I t req u ires you to j u s t i fy the p u rs u i t of your

own personal life a n d i nterests only as components of the genera l good, a n d does not per m i t reasons for action to end with a reference to w h a t you w a n t or a re d evoted to. Those considera­

tions a re c o m pletely en c o mpasse d by a n i m p e r son a l point o f

view which

accords

you n o speci a l posi ti o n , u n less it can b e

i m p e rsonally j u s t i fi ed .

Re sistance comes , na turally enough,

fro m the poin t o f view o f the individual , who m a y be willing to

a ccord i m perso n al conside r a tions s o m e weigh t , but who is also p ow e rfu lly motivated b y the i n d e pend e n t cl aims of his own

life - o f the view fro m where he is i n the worl d . But t h i s does not remain a con fl i c t bet w e en i m p er sonal values a n d mere

i nd i v i d u a l i n terest, because the resis tance can be ge n eraliz e d . Someone who regards conseq uenti alist r e q u i r e m e n t s as

u n a c­

cep tab l e because of their cl a i m to d o m i nance over his own p o i n t of view wi l l naturally e x ten d this obj ection to othe r s . He will

gravitat e toward a general e x cep tion to conse q u e n ti alis m in fav o r o f t h e pers o n a l vi e wpoint , and t h i s w i l l constit u t e a n a l ternative

ethic, rather than m e rely a r e sistance to ethics. Such a n eth i c n eed

be n o less u n i versal than util i ta rianis m , but i t w i l l be s u bj ective

in

a

way that conse q u en t i a l i s t positions a re not. Each person will

b e p e r mitt e d , within l i m i ts, to concentrate on the p u rs u i t o f his l i fe, a nd there will not be a si n gle , obj e ct i vely describable end b y refer en c e to which e veryone ' s actions must be j u stified .

I n this sense the d eontol o g i ca l r e q u i re ments that resist a consequentia list acc o u n t a re also subjective. C o ns trai n t s against m u rder, l y i n g , betrayal, assault, or coercion, tho u g h i n tended to a p p l y u niversa l l y, o p pose the agent ' s specific r e l a tion s to o t h e r people to the conce p tio n of a si n gl e end that everyone should e x cl u s ively p r o m o te. They a re agen t-centered , but i n a diffe rent

way. The rea l sou rce o f these restrictions, unlike tha t o f the age nt- cent e r ed p e rmis s ions , i s not the agent but the poten t i a l

victi m whose r ights are p r o tec t ed . B u t the w r ongness o f violat­ ing those righ t s i mplie s

a

cons t r a i n t on ea ch p e rson again s t

viol a t i n g them, rather than a req u i rement that he t ry to mini m­ ize thei r overall violation (even i f this means c o m mitting a few himsel f) . 7 D e o n tological req u irements a re ag e n t-c e n t e red b e ca u s e the y i n s truct e a c h p e rson t o d e t ermine t h e rightn e ss or 7

See chapter

5 above.

Mortal

204

q u estions

wrongness of his a cts solely from t he point of view of his position i n the world and his d i rect relation to others. The very idea that the basic moral concepts are right and wrong rather than good and bad entails tha t the character of one's actions rather than the world as a whole must be one's pri mary concern. a [f there is a difference in poin t of view between the two types of excep tion to conseq uen tial ism, it is that the fi rst derives s i m pl y from the standpoi nt of the individual agent, whereas the second emerges when he considers in a certain way his own point of view together with those of the persons to whom he is directly related in action. Deontologi cal cons traints are i nter­ mediate between pu rely individual moti ves and co mpletely i m personal values. There are fa m i l i a r dispu tes about whether u tilitarianis m really does have the conseq u ences a ttributed to i t by anti­ conseq uentialist critics - aspects of the wider dispute between radical and moderate interpretations o f utilitarianism. Likewise there are disputes about the formulation o f al ternative views : how a bsolutist they a rc, whether they should be stated i n terms of individual rights, or l i berty, or self-realization, or interper­ sonal commit ment. But the essence o f the conflict is clearer than the exact na tu re of the alternatives. The issue is how the individual position of the agent should enter into a decision about what he should or may do. Obviously i t cannot fail to enter in certain ways. E ven on a consequential ist view, what one shou ld do will depend on what j one is i n a position to do, and on the relative desi rabi l i t y of the \ \ possible outco mes. Nevertheless, the conseq uentialist j u d g ment that one shou ld do something is essen tially the j u dgment that i t wou l d be bes t if o n e did it - t h a t i t o u g h t to happen . T h e right thing to do is to turn oneself as fa r as possible i nto an instru ment for the rea lization of what is best sub specie aeternitatis . Agent-centered views, on the other hand, determine what is right, wrong, and permissible partly at least on the basis of the 8

A moral theory of this type is developed by Cha rles Fried in Right at1d Wrcmg (Cam bri dge, Mass. : H a rvard Uni versit y Press, 1 978). A n i n termediate v i e w has been put forward b y Sa muel Scheffler, in ' Agents and Outcomes' (Ph . D . dissertation, Princeton University, 1 977): he defends agent-centered permissions but rej ects agent-centered requirements as h a v i n g n o intelligible basis.

Su�jective and o�jective

205

individual's l i fe, his role in the world, and his relation with others. Agent-centered morality gives p ri macy to the q u estion o f what to do, a question asked b y the individual agent, and does not assume tha t the only w a y to answer it is to say what i t would be best i f he did,

sub specie aeternitatis . It may also hold that the

place for considerations o f what would be best, i n a decision a b o ut what to do, is not ob vious and must be established by analysis of agen t-centered choice and i ts grounds. The real issue, therefore, i s the r elati v e p riority, in regard to action , of two ways of looking a t the world. On the one hand the re is the p o s i t i o n that one's decisions should be tested u l t i m a tely fro m a n external point o f view, to which one appears as j u s t one p ers o n a mong others. The question then becomes,

' W h a t would be best? Which of the acts within my power would do the most good, cons idering matters fro m out here, i mp erson­ a l l y ? ' This point of view clai ms priority by virtue of greater comp rehens iveness. The agent's s i tuation is s u p posedly given i ts due in a l arger perspective.9 O n the other hand there is the position that si nce an agent lives his l i fe fro m where he is, even if he m a nages to achieve an i m pe rsonal view of his s i t u a tion, wha tever i nsights resu lt from this detachment need to b e made p a r t of a personal view befo re they c a n i n fl uence deci sion and a ction. The pu rsuit of what

seems i mperson a l l y best may be an i m porta n t aspect o f indi­ vidual life, b u t i ts place in that l i fe must be deter m ined fro m a personal standpoint, beca use life is always the l i fe of a p a rticular person, and cannot be lived sub specie aeternitatis . 1 0 The o pp osi tion looks like a stalemate beca use each of the p o i n t s o f view clai ms d o m i nance over the other, by virtue of inclusion. The i m personal standpoint takes in a world that inclu des the ind ividual a nd his personal views. The personal 9 In 10

Th e

Possibility of A ltru ism (O x fo r d : a version of this position.

O x ford U n i vers i t y P r e s s ,

1 970)

I

defended

T his pos i t i o n i s persu a s i v e l y presented b y Berna rd Williams i n ' A Critique o f U til i ta rianis m ' , i n J . J. C. S m a r t and Berna rd W i l l i a m s , Utilitaria uism For at�d Agaiu.