1 Scanlon, T.M. ‘Contractualism and Utilitarianism,’ in Utilitarianism and Beyond, Sen, Amartya and Williams, Bernard eds. (Cambridge: Cambridge University Press 1982)Google Scholar.

2 Scanlon's discussion straddles the frequently drawn distinction between metaethics and normative ethics. The overall theory Scanlon presents addresses traditional metaethical problems about the objectivity of moral judgments but the account of objectivity Scanlon provides cannot be divorced from the substantive normative content of contractualism.

3 ‘If we could characterize the method of reasoning through which we arrive at judgments of right and wrong, and could explain why there is good reason to give judgments arrived at in this way the kind of importance moral judgments are normally thought to have, then we would, I believe have given a sufficient answer to the question of the subject matter or right and wrong as well. No interesting question would remain about the ontology of morals - for example, about the metaphysical status of moral facts’ (2).

4 See Nagel, Thomas Equality and Partiality, (New York: Oxford University Press 1991), 7.Google Scholar

5 At one point, Scanlon cites the sense of moral crisis precipitated in America by the Vietnam War as support for his view that humans are highly motivated by a concern with ensuring that ‘our lives and institutions are justifiable to others’ (163). Obviously, the interpretation of the moral significance of the divisions in American society occasioned by the war is a complex matter. But I suspect that Scanlon's interpretation is driven in part by his prior commitment to contractualism. Thus for him, the polarization of American society around Vietnam is optimistically interpreted as evidence that Americans were divided about whether the war could be justified to all affected by the war (including the Vietnamese) rather than about whether war really served the national interest. The former interpretation is more commensurate with contractualism since it supposes that humans are deeply motivated by mutual justification. Colin McGinn has also complained that Scanlon's frequent appeals to phenomenology are idiosyncratic in a problematic way. See McGinn, Colin ‘Reasons and Unreasons’ in The New Republic (24 May 24 1999), 37-8Google Scholar.

6 Of course, it is possible that the surface appearance of moral discourse is misleading. The reason-giving moral properties to which moral judgments seemingly refer may be entirely illusory. The best semantic interpretation of moral discourse may a cognitivist one that emphasizes the practicality of moral judgments without committing itself to real properties of the sort referred to in moral discourse. A broadly realist account of moral judgments, such as Scanlon's, takes moral discourse at face value and attempts to provide a plausible philosophical explanation of the subject matter of moral that leaves in tact the apparent phenomenology of moral judgment. By contrast, a broadly anti-realist account of moral judgments offers some kind of revisionist interpretation of moral judgment in which the link between ordinary moral discourse and at least some of its apparent metaphysical, semantic and epistemic commitments is severed. An error theorist, such as Mackie, accepts the cognitivist interpretation of moral discourse but claims that the action-guiding moral properties to which moral judgments purport to refer to do not exist. See Mackie, J.L. Ethics: Inventing Right and Wrong (Hammondsworth, UK: Penguin 1977)Google Scholar. Sophisticated expressivists, such as Gibbard, accept the action guiding character of moral judgment but resist any straightforward cognitivist interpretation of moral judgment. See Gibbard, Allan Wise Choices, Apt Feelings (Cambridge, MA: Harvard University Press 1990)Google Scholar. On this view, moral judgments ultimately express some kind of attitude and are not, strictly speaking, beliefs. A traditional problem that besets anti-realists accounts of morality is that they demand too much revision in our intuitive conceptions of morality. Realist accounts, by contrast, seem to fit well with our moral phenomenology but often seem to involve unattractive metaphysical or epistemological commitments. Most notoriously, realists have struggled to provide some account of moral properties that are suitably objective, ontologically tolerable, and ground the practicality of moral judgments. Scanlon thinks that contractualism can accommodate the realist intuitions without being encumbered with embarrassing metaphysical or epistemological commitments.

7 Though Scanlon holds that a person who fails to recognize (or fails to appreciate the significance of) a sound normative reason that bears upon the justifiability of her commitments need not be irrational, such a person does suffer from some kind of rational failing. She may be unreasonable or insensitive. Scanlon favors a narrow construal of irrationality according to which a person is irrational only when he or she ‘recognizes something as a reason but fails to be affected by it in one of the relevant ways’ (25).

8 On many accounts, the ends an agent has reason to adopt and pursue can be identified with the ends she would desire under certain idealized (and typically counterfactual) conditions. This sort of strategy allows theorists to link rational ends with desires while avoiding the embarrassment of merely identifying an agent's rational ends with the objects. of the desires they currently have.

9 Indeed, strictly speaking even reasons that have subjective conditions have a kind of force that is independent of subjective facts about an agent. Other things being equal, the fact that something tastes good to me is, for instance, a reason to eat it whether or not I ever recognize it as a reason. The somewhat perverse person who says ‘This tastes good but that gives me no reason whatsoever to eat it’ is insensitive to a practical reason that applies to them.

10 Sometimes we experience what Scanlon calls ‘desire in the directed attention sense.’ ‘A person has a desire in the directed attention sense that P if the thought of P keeps occurring to him or her in a favorable light, that is to say, if the person's attention is directed insistently toward considerations that present themselves as counting in favor of P’ (39). Desire in this sense is an important phenomenon and characteristic of how we often think of desires. Scanlon argues that although this form of desire can be motivationally efficacious, it is not always so since we often act in the absence of such desires and indeed in opposition to such desires.

11 See Williams, Bernard ‘Internal and External Reasons’ in Moral Luck (New York: Cambridge University Press 1981)CrossRefGoogle Scholar. Scanlon's book includes a special appendix devoted to discussing Williams's view (363-73).

12 A related but distinct issue broached by Scanlon concerns whether the standards of correctness are best interpreted in cognitivist or non-cognitivist terms. Here Scanlon contrasts his own cognitivist, ‘belief’ interpretation of reasons with the non-cognitivist, expressivist position defended by Gibbard. On the belief interpretation, the judgment that ‘X is a reason for doing A’ involves acceptance of the belief that ‘the relation “counting in favor of” holds between X and doing A’ (58). The standards of correctness for assessing the correctness of this belief are independent of the psychological dispositions of agents. Gibbard's position (and the closely allied position defended by Simon Blackburn) allows that there are standards for the evaluation of the reasons of agents. However, these standards are ultimately traceable to norms, that as a matter of psychological fact are accepted by agents. Thus Gibbard holds that ‘when a person calls something—call it R—a reason for doing X, he expresses his acceptance of norms that say to treat R as weighing in favor of doing X’ (Gibbard quoted by Scanlon 58). Somewhat misleadingly, Scanlon claims that not much turns on the difference between the belief and special attitude interpretation of judgments about reasons so long as ‘there are standards of correctness of the relevant sort’ (59). This is misleading because the expressivist position yields very different conclusions about the possible reasons of agent than does Scanlon's position. Scanlon holds that a person can have a reason to X even if the norm recommending X is not one which she actually accepts. By contrast, the expressivist, in a way parallel to Williams's position, holds that there must some relation between the actual attitudes held by an agent and the reasons she has to act. Scanlon thinks we can say, for instance, a person has a reason not to be cruel even if it is true that the person has no disposition to refrain from acts of cruelty and that cruelty would not impede the achievement of other ends of the person. Expressivists seem committed to denying the possibility of such external reasons. Blackburn contends that there is a sense in which Humean expressivists can tolerate some talk of external reasons but it falls short of encompassing the range of reasons Scanlon defends. See Blackburn, Simon Ruling Passions (Oxford: Oxford University Press 1998), 265-6Google Scholar.

13 The kind of constructivism that Scanlon seems committed to both here and with respect to the specific contractualist account of distinctively moral reasons might be described as ‘weakly regulated’ constructivism. The process for deliberation about reasons is not regulated by a tightly defined point of view. We need only think carefully about the reasons for action that seem to present themselves to us. By contrast, Rawls's form of Kantian constructivism is much stricter in that the process and point of view through which specifically moral principles can be derived are very well defined. Thus the details of the original position as well as the principles of rationality that guide the deliberations of persons behind the veil of ignorance are clearly specified. See Rawls, J. ‘Kantian Constructivism in Moral Theory,’ The Journal of Philosophy 77 (1980) 515-72Google Scholar.

14 Formally, the challenge is to come up with a representative example with the following structure. At t1, an agent A has motivational set M. From this set (and given the circumstances in which A finds herself), it is impossible to derive, through any sound deliberative route, R, a reason to do X. Moreover at t1, A believes she has no reason, R, to do X. At t1, A engages in the process of reflection outlined by Scanlon and concludes that she does have reason R to do X. That such a phenomenon of the revision of practical judgment sometimes takes place seems plausible to me. But, in light of the controversy about whether it is possible, it is disappointing that Scanlon has no clear cut example of it.

15 See Darwall, Stephen Gibbard, Allan and Railton, Peter ‘Toward Fin de siècle Ethics: Some Trends,’ The Philosophical Review 101 (1992), 115-16CrossRefGoogle Scholar for a useful overview of reactions to the ‘open question’ argument.

16 Mackie, J.L. Ethics: Inventing Right and Wrong, 37-9Google Scholar

17 See Nagel, Thomas The View from Nowhere (New York: Oxford University Press 1986), 158-60.Google Scholar

18 Here I share Nagel's skepticism about the idea that living well is always in harmony with the demands of morality. See ibid., 195-9.

19 I choose this example because it is frequently taken to be a paradigmatic example of a moral judgment. See, for instance, Harman, Gilbert The Nature of Morality (New York: Oxford University Press 1977), 4Google Scholar; Sturgeon, Nicholas ‘Moral Explanations,’ in Morality, Reason, Truth, Copp, David and Zimmerman, David eds. (Totowa, NJ: Rowman & littlefield 1985), 52Google Scholar; Hampton, Jean The Authority of Reason (Cambridge: Cambridge University Press 1998), 211CrossRefGoogle Scholar.

20 It might be suggested that the contractualist can deal with this sort of case by arguing that torturing cats is wrong because it cannot be justified to other human beings. This is a lame response because it offers a distorted account of what is wrong about torturing cats. The wrongness of such treatment is a direct consequence of the harm done to cats, not an indirect consequence of human reaction to such harm.

21 I am grateful to an anonymous referee for supplying this suggestion about a possible contractualist rationale for MP.

22 Proponents of Hobbesian contractarianism would presumably insist that Conrad, given his situation and preference structure, could reject RP and endorse MP. The Hobbesian position may strike many as repugnant yet it is difficult to see how the contractualist framework licenses any such conclusion. Scanlon must say that the Hobbesian has made an error about the nature or relative force of the reasons that are relevantly considered in this case. But it is difficult to characterize the Hobbesian position as a violation, in any straightforward sense, of the normative standards defined by contractualism. Instead, Hobbesians and Scanlon seem to have divergent judgments about how the force of conflicting generic reasons is to be evaluated.

23 Scanlon does argue that certain considerations have no weight in determining whether a principle can be reasonably rejected. Specifically, he rules out appeals to impersonal values— ‘reasons that are not tied to the well-being, claims or status of individuals in any particular position’ (219). So the fact that a principle is compatible with the destruction of a rain forest, judged to be valuable in itself, does not count as a reason for rejecting the principle.

24 Scanlon would certainly dispute the charge that contractualism lacks sufficient normative content. After all, part of the putative importance of contractualism is that it provides ‘a clear alternative to utilitarianism and other forms of consequentialism’ (229). Full consideration of Scanlon's critique of consequentialism is beyond the scope of this essay. His contention that consequentialism is defective because it is compatible with the imposition of unfair burdens on individuals or groups strikes me as quite plausible (230). However, it is not clear that a perspicuous explanation of this failure of consequentialism can be provided through sponsorship of a contractualist criterion of reasonable rejection.

25 I would like to thank Alistair Macleod, Jan Zwicky and an anonymous referee for providing me with very helpful comments on this essay. Sam Black, Avigail Eisenberg and James Young also helped by discussing various aspects of Scanlon's project with me.