Oxford University Press The Analysis Committee

Oxford University Press The Analysis Committee http://www.jstor.org/stable/3327571. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at. http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Oxford University Press and The Analysis Committee are collaborating with JSTOR to digitize, preserve and extend access to Analysis. http://www.jstor.org

NAGEL, WILLIAMS, AND MORAL LUCK By JUDITH ANDRE BERNARD WILLIAMS and Thomas Nagel begin their discussions of 'Moral Luck' by contrasting morality with luck.' Morality - at least as Kant articulates it - is the sphere of life in which, no matter what our circumstances, each of us can become worthy. Moreover, moral worth is the highest worth of all, and so there is a kind of ultimate justice in the world: each person is equally able to achieve that which matters most in life. Nagel and Williams argue, however, that in practice we evaluate actions and agents partly on the basis of circumstances beyond the agent's control - on the basis of luck. Nagel argues that we ascribe moral value partly on such a basis, Williams that 'rational justification' rests partly on luck. For Nagel the consequence is an incoherence within our moral conceptual scheme. Williams reaches a similar conclusion more slowly. Since rational justification is partly a matter of luck, rational justification is not synonymous with moral justification. If this is so, then our notion of rational justification is not synonymous with that of moral justification, and morality is not the unique source of value; and if not unique - he argues - not supreme. 'If the moral were really supreme, it would have to be ubiquitous: like Spinoza's substance, if it were genuinely unconditioned, there would have to be nothing to condition it' (Williams, p. 38). But this is fundamentally different from the concept of morality we (in some sense) have now: 'one thing that is particularly important about ours is how important it is taken to be' (Williams, p. 39). For Williams, when luck enters into judgments of justification, the judgment is non-moral but competitive with - and so destructive of - moral justification. For Nagel, luck does enter into moral assessment, although our intuitions say it should not. Each writer finds destructive inconsistency - possibly incoherence - within our concept of morality. To these charges, two responses are possible. One is that we do not in fact justify actions as it is claimed we do, or at least would not if we were sufficiently reflective. A second possible response is to admit that we do justify actions partially on the basis of luck, accept the conclusion that we are not consistent Kantians, but reject the implication that our moral scheme is therefore incoherent. I shall make both of these responses here. I will discuss first the four areas where, Nagel claims, luck partially determines 'moral' value. One is constitutive luck: 'the kind of person you are.. your inclinations, capacities, and temperament'. Another is luck in consequences: 'luck in the way one's actions and 'Thomas Nagel, 'Moral Luck', in Mortal Questions (Cambridge: Cambridge University Press, 1979). Bernard Williams, 'Moral Luck', in Moral Luck (Cambridge: Cambridge University Press, 1981). 202

NAGEL, WILLIAMS, AND MORAL LUCK 203 projects turn out'. One drunken driver kills a child, but another gets home safely. A third area is 'luck in how one is determined by antecedent circumstances'. This third area involves the classic conflict between believing we are determined and believing that we are morally responsible. So does Nagel's fourth area, which he calls 'luck in circumstance'. The person who became a Nazi prison guard might simply have been a bureaucrat, had his parents emigrated to Canada when he was young - yet we hold the guard morally responsible. This kind of example illuminates particularly well the difficulty in reconciling determinism and moral responsibility. Since we evaluate these two people differently on the basis of their different actions, we must be implicitly assuming some source of agency other than character (assumed to be identical in the two men) and circumstance (assumed to be beyond the control of either). But compatibilism is not Nagel's primary subject. Since so much has been said on this issue already, and since Nagel does not advance that discussion, I would like to bracket the issue here. The other areas, however - constitutive luck and luck in consequences - raise issues that can be discussed without entering the free-will-determinism debate. Take for instance the category 'luck in consequences'. Is the drunken driver who kills a child morally worse than the drunken driver who manages not to? Is successfully rescuing someone from a burning building morally better than trying unsuccessfully to rescue him? Or than accidentally harming him during the attempted rescue - perhaps by dropping him from a high ladder? As I mentioned earlier, two kinds of response are possible. One is simply to deny the implication, to assert that these people do not differ in moral status. I'll make a more minimal claim here: the difference in moral evaluation substantially lessens upon reflection. The 'lucky' drunken driver has done something seriously wrong. Anyone who has lost a relative in an accident involving an 'unlucky' drunken driver will look with loathing upon any drunken driver. Reflection has the opposite effect upon our evaluations of lucky and unlucky careless drivers. If the carelessness is particularly common - taking one's eyes briefly off the road, for example we're - likely to say 'Anyone might have done that' and lighten our condemnation of the driver. In both cases, reflection at least narrows the apparent moral difference created by luck. A variation of this response is to admit that luck does enter our moral evaluation, but to treat this as an inconsistency of which we could and should purge ourselves. Both Nagel and Williams claim that doing so would narrow the area of moral concern to the vanishing point (Williams, p. 38; Nagel, p. 35). If we were to exclude everything which Nagel calls luck - the kind of person one is, the circumstances which confront one, the choices one makes (if these are taken to be the result of antecedent circumstance), and the

204 ANALYSIS results of those choices - then Nagel and Williams are right. But in this paper I am bracketing the issue of determinism. Consistency in the remaining areas (constitutive luck and luck in consequences) would narrow but not erase the sphere of moral concern. Most people already make consistent efforts to separate moral judgment of someone from their appraisal of that person's 'constitution' (inclinations, capacities, temperament). Phrases such as 'Only God can judge her' reflect the belief that her internal intention to do the right thing - unknowable to the rest of us - is what really counts. As for luck in circumstances, it might be possible to learn to classify all careless drivers (lucky and unlucky) as morally identical. However, a different kind of response is possible. Suppose that we do evaluate the lucky differently from the unlucky, and that we cannot (in some sense of 'cannot') change what we do. What follows from that fact? Although the remaining use of moral categories is clarly not Kantian, it does not follow that it is an inconsistent use. The word 'moral' is not nearly so precise as we sometimes feel - this is Nagel's and Williams' point - but it does not follow that the concept is internally incoherent. The most obvious non-kantian element in our moral scheme is the Aristotelian.2 Part of being moral, for most of us, is being virtuous; and being virtuous involves more than doing the right thing. It involves as well the ability to see what the right thing to do is, and the desire to do that right thing. (Not just Kant's more abstract desire- to- do- whatever- the- right- thing- turns- out- to- be.) Virtues, as Aristotle describes them, are possible only to those who have been reared in a moral community; a fortunate childhood fosters adults who feel rightly as well as acting rightly. But people cannot choose their own upbringing, and emotions are not voluntary. (We do have indirect control over our emotions, as L. A. Kosman points out, but this is limited.3 For Aristotle it is better not to have to struggle to do the right thing. For Kant the presence or absence of inner struggle is morally irrelevant. The word 'moral' has a somewhat different sense in the two writers, but there is this in common: for both the adjective refers to an excellence of character such that the moral person is praiseworthy and emulable. He or she is a model for our children, and the kind of person which we would like our communities to foster. The Kantian concept, however, unlike the Aristotelian, is closely linked to Christian ideas of reward and punishment by an all-just, omniscient Judge. Even an atheist can ask what such a Judge would do; and of course the 2 Alastair MacIntyre, After Virtue (Notre Dame: University of Notre Dame Press, 1981). Although I use the terms 'Aristotelian', 'Kantian', and 'Christian' here, I am not defending any particular historical view about the content of these traditions. Instead I use the terms roughly, as convenient labels for identifiably separate strands within our present ways of thinking. 3 L. A. Kosman, 'Being Properly Affected: Virtues and Feelings in Aristotle's Ethics', in Essays on Aristotle's Ethics, Amelie Oksenberg Rorty, ed. (Los Angeles: University of California Press, 1980).

NAGEL, WILLIAMS, AND MORAL LUCK 205 Judgment could only concern those matters over which the agent had control. Reward and punishment could not justly be allotted on the basis of upbringing or genetic endowment. We use the concept of moral in both the Aristotelian and the Kantian sense. Many of our central concepts are hybrid: we can call someone 'evil', for instance, without imputing responsibility (for she may be criminally insane). Someone may be sadistic because of the way he was raised - but he's still sadistic, and it is not a contradiction to say this is a moral fault for which he is not to blame. Open-mindedness comes more easily to some than to others; it's still (when properly limited) a moral asset. Similar remarks could be made about selfishness, generosity, cruelty, bravery - and even about the spoiling of children. Through no fault of their own, they've been made defective. Of course our condemnation increases as the child becomes adult - blameworthiness is a specific kind of criticism which implies free choice - but selfishness and petulance in children are nevertheless character defects. Is a moral framework which includes both Kantian and Aristotelian elements self-contradictory? It can look that way, as Nagel and Williams' examples make clear; but the appearance results from assuming that 'moral' can only have a Kantian sense, and then discovering cases where its application depends on circumstances beyond the agent's control. In its central sense, I contend, morality refers to excellence of character. Whenever we praise people as moral we mean they are worthy of praise and emulation; but only sometimes do we mean that they are worthy of reward. Let me apply this to Nagel's 'luck in consequences' - the cases of the lucky and the unlucky bad drivers, of the successful and unsuccessful rescue attempts. Some people persistently misjudge, and in the process hurt other people. (We might call the agents 'morally accident-prone.') They are malformed in some way; something prevents them from correctly assessing the facts before they act. Impulsiveness, or heroic fantasies, or self-absorption - whatever its source, their ineffectiveness make them less than admirable, less than a model. But to the extent that its source is beyond their reach, they are neither blameworthy nor punishable. This accounts for some of the 'moral distance' between our evaluations of the drivers, too; some people know their limitations better than other people do. Furthermore, we say that the unlucky driver is morally worse than the lucky one in part because we hold people responsible for the results of their actions. But 'responsible' has two kinds of application. One is what I've called the Christian-Kantian sense: the responsible party is subject to reward and punishment. The other sense is more prosaic: to be responsible is to have an obligation to rectify bad consequences. If I break your vase, I must replace it. I can be responsible in the second sense without being in the least blameworthy, although often the two coincide. We can also be

206 ANALYSIS blameworthy (for the risk we take) without in fact bringing about any bad consequences, and so without being obliged to rectify anything. At this point one may ask, obliged on pain of what? Certainly there is moral defect if we fail to make good what we have destroyed when it is possible to make this good; furthermore, the defect is greater when the source of the damage is not just our action but our culpable action. But what about the cases where the damage cannot be repaired? Loss of life, most obviously, cannot be put right. There is at least a great sadness in having incurred a debt which one cannot meet, and therefore a feeling of inadequacy. As a result of that there is a sense of diminished worth. This is, however, distinguishable from moral fault in the sense of deserving punishment. If these considerations are right, then part of our sense of 'moral distance' between the lucky and unlucky driver is neither derived from nor inconsistent with Kantianism. These same considerations apply, more simply and directly, to what Nagel calls 'constitutive luck'. Some people are generous, brave, and honest by inclination; others are not. The fortunate ones are so constituted by upbringing and biology. They may deserve no particular reward for what they are; but they are nevertheless worthy of admiration and imitation. I think, then, that Nagel's claim of contradiction within our moral concepts is mistaken. What about Williams' related claim that the justification of actions (as, roughly, the right thing to have done) depends partly on luck? If that's true, he claims, then rational justification is not synonymous with moral justification, and hence moral justification is neither the unique nor the supreme source of personal value. His argument is intricate, and I will address only a few strands of it here. He describes in particular the cases of Gauguin and of Anna Karenina. Each abandoned a family in order to pursue a future good (artistic accomplishment, on the one hand; passionate love on the other). Williams claims that these undertakings turn out to have been justified only if the goal is realized. This is close to the kind of situation Nagel describes as luck in consequences, and it is open, I think, to the same analysis. First, it's at least reasonable to claim that consequences have nothing to do with our judgment of the cases. Many would simply condemn both Gauguin and Karenina; others would approve their actions. Even if, as Williams suggests, we ask whether their actions were reasonable (rather than moral) it's quite possible to assess the reasonableness of risk-taking. Finally, however, if we suppose that Gauguin's success does affect our evaluation of his choice, we can account for that on the Aristotelian grounds mentioned earlier. The person who can correctly assess his or her chances of success is better-formed than the person who cannot. Since considerations of luck enter without paradox into one kind of moral evaluation, they do not indicate a non-moral, competitive

NAGEL, WILLIAMS, AND MORAL LUCK 207 realm of evaluation. It is still possible to characterize the morally good life as that which should be sought, all things considered. A final note. Not only does 'moral luck' pose considerably smaller problems than Nagel and Williams suggest, it is in fact part of our conceptual scheme which should, upon reflection, be kept rather than changed. First of all, it reminds us of moral questions different from those upon which we typically concentrate. 'What should I do in these circumstances?' is important; but 'What kind of person should I try to be, and help others to be?' is of equal importance. Secondly, 'moral luck' is sometimes illusory;we are sometimes to blame for results that at first seem beyond our control. We have more control over the kind of person we are than we sometimes think. I can, for instance, whittle away at the habits of self-deception which prevent me from seeing the world as it is. The 'morally accident-prone' can learn to be better; but they are unlikely to change if all our moral assessment is concentrated on intention and none on actual result.4 Old Dominion University, Norfolk, Virginia 23508, U.S.A. @ JUDITH ANDRE 1983 4 This paper received the 1983 Griffith Memorial Award from the Southern Society for Philosophy and Psychology, U.S.A. THE TOXIN, THE BLOOD DONOR AND THE BOMB By JANICE THOMAS I form an intention to WHEN 0 it sometimes happens that my formation of the intention has consequences independent of the effects, if any, of my 0-ing. My present intention to give blood tomorrow might inspire a new blood donor today - whether or not I go through with it. Thus, knowing that my terror of the needle makes me a persistent akratic where giving blood is concerned, I might still operate an effective scheme to recruit donors by manifesting a good strong intention to donate (and disguising my full awareness of my past weaknesses) on suitable occasions. Now Gregory Kavka ('The Toxin Puzzle', ANALYSIS 43.1, January 1983, pp. 33-6) might concede that I could inspire with my beneficent intention-forming, but would be quick to distinguish my akratic blood donor story from his tale of the toxin. For his protagonist is faced with the challenge to gain the fruitful consequences which would flow from his forming a simple intention when he has no reason to do the action intended and good reason not to. To gain the eccentric billionaire's million dollars all the protagonist has to do is form the intention to take the toxin -