¹ Stocker, Michael, ‘The Schizophrenia of Modern Ethical Theories’, Journal of Philosophy, lxxiii (1976)Google Scholar; Cocking, Dean and Oakley, Justin, ‘Indirect Consequentialism, Friendship, and the Problem of Alienation’, Ethics, cvi(1995)Google Scholar; Scanlon, Thomas, What We Owe to Each Other, Cambridge, MA, 1998, pp. 88–90Google Scholar.

² The exception is Jackson, Frank, ‘Decision-theoretic Consequentialism and the Nearest and Dearest Objection’, Ethics (1991)CrossRefGoogle Scholar.

³ Pettit, Philip and Smith, Michael, ‘Global Consequentialism’, Morality, Rules, and Consequences: A Critical Reader, ed. Hooker, Brad, Mason, Elinor and Miller, Dale E., Edinburgh, 2000, pp. 121–33Google Scholar.

⁴ Though why shouldn't the local motive consequentialists say instead that the relation is one according to which the morally best acts are those that would have been performed by people if they had had the morally best motives? Or why not say instead that the relation is one according to which the morally best acts are those that would have been performed by people if they had had the set of motives that it would have been morally best for them to try to acquire? It is because we see no principled and plausible way of answering these questions that Pettit and I reject local motive consequentialism (Pettit and Smith).

⁵ Local act consequentialists might defend the choice of this relation by citing the following remark of Thomas Nagel's. ‘[A]n impersonal morality requires of us not only certain forms of conduct but also the motives required to produce that conduct. This much is I think true of any morality properly worked out. If we are required to do certain things, then we are required to be the kinds of people who will do those things’ (, Nagel, The View from Nowhere, Oxford, 1986, p. 191Google Scholar). Needless to say, the global consequentialist denies that this is true. Rather, according to the global consequentialist, all requirements are generated by the consequentialist principle itself

⁶ Compare Philippa Foot, ‘Utilitarianism and the Virtues’, repr. Consequentialism and its Critics, ed. Samuel Scheffler, Oxford, 1988, pp. 224–42.

⁷ Pettit and Smith.

⁸ Sidgwick, Henry, The Methods of Ethics, 7th edn.London, 1907Google Scholar; Parfit, Derek, Reasons and Persons, Oxford, 1984Google Scholar; Peter Railton, ‘Alienation, Consequentialism, and the Demands of Morality’, repr. Consequentialism and its Critics, ed. Samuel Scheffler, Oxford, 1988; ‘How Thinking about Character and Utilitarianism Might Lead to Re thinking the Character of Utilitarianism’, Midwest Studies in Philosophy, vol. xiii, Ethical Theory: Character and Virtue, ed. French, Peter A., Uehling, Theodore E. Jr, and wettstein, Howard K., Dame, Norte, IN, 1988, pp. 398–416Google Scholar.

⁹ Though I make this assumption I must say that it does seem to me to be rather unrealistic. In the context of human misery on the scale that currently exists in the world it would, I think, be far more realistic to suppose that it would maximize neutral value for many of those of us who have a large number of resources at our disposal to be motivated almost exclusively to bring about as much neutral value as possible. A good number of us, at any rate, should remain friendless for the sake of the greater good. But I propose to leave this objection to one side for the purposes of the present argument.

¹⁰ Smith, Michael, The Moral Problem, Oxford, 1994, p. 95 and ch. 5Google Scholar.

¹¹ What does talk of evaluative significance, simpliciter, amount to? I take it that the intuitive idea is plain enough: it is the ranking of possible worlds that results when an evaluator who has one ranking of possible worlds in neutral terms and another ranking of possible worlds in relative terms merges those two rankings into one overall ranking. A more precise suggestion is made in note 27 below.

¹² , Wolf, ‘Moral Saints’, Journal of Philosophy, lxxix (1982)Google Scholar.

¹³ Ibid., 421 f.

¹⁴ Ibid., 436 f.

¹⁵ , Railton, ‘Moral Realism’, Philosophical Review, xcx (1986)Google Scholar.

¹⁶ Ibid., 189.

¹⁷ Ibid., 189–200.

¹⁸ , Firth, ‘Ethical Absolutism and the Ideal Observer’, Philosophy and Phenomena-logical Research, xii (1952)Google Scholar.

¹⁹ Ibid., 337.

²⁰ Ibid.

²¹ Ibid.

²² What is the wider conception of impartiality that Firth has in mind? I take it that the wider conception of impartiality is something close to what Hare has in mind when he talks of universalizability. For more on universalizability, see the discussion of Hare in the next section.

²³ , Railton, ‘Moral Realism’, 204Google Scholar.

²⁴ Ibid., 205.

²⁵ , Hare, Moral Thinking, Oxford, 1981, p. 55Google Scholar.

²⁶ , Smith, The Moral Problem, pp. 6 fGoogle Scholar.

²⁷ It is worth pointing out that we can provide a direct argument for the prescriptivity of judgements about the desirability of giving a slightly smaller benefit to a friend, over a slightly larger benefit to a complete stranger. I have argued elsewhere that judging it desirable that one acts in a certain way is, in effect, a matter of judging that one would want that oneself acts in that way if one had a set of desires that was maximally informed and coherent and unified (, Smith, The Moral Problem and ‘In Defense of The Moral Problem: A Reply to Brink, Copp and Sayre-McCord’, Ethics, cviii (1997))Google Scholar. If this is right, however, then it follows that those who make the judgement that it is desirable to provide a slightly smaller benefit to their friends rather than a slightly larger benefit to a complete stranger in effect judge that they would want that they themselves provide a slightly smaller benefit to their friends rather than a slightly larger benefit to a complete stranger if they had a set of desires that was maximally informed and coherent and unified. But those who make this judgement would then be required, by norms of coherence, to desire to provide a slightly smaller benefit to their friends rather than a slightly larger benefit to a complete stranger. They would be required to have this desire because the belief-desire pair that comprises the belief that one would want that one gives a slightly smaller benefit to one's friend rather than a slightly larger benefit to a complete stranger, on the one hand, and the preference to give one's friend that smaller benefit on the other, is a more coherent pairing of belief and desire than that which comprises the belief together with indifference or aversion to providing one's friend with that smaller benefit. And, if this is right, then it follows immediately that, absent practical irrationality - that is, absent incoherence in their psychology - those who make the judgement about the desirability of giving a smaller benefit to their friends would indeed desire accordingly. The judgement is thus prescriptive. (Note that this amounts to a more precise suggestion about the nature of evaluative significance, simpliciter (Smith, Michael, ‘Normative Reasons and Full Rationality: Reply to Swanton’, Analysis, lvi (1996)Google Scholar, and Smith, ‘In Defense’). Neutral value is a matter of what would be desired if we had a set of desires that was maximally informed and coherent and unified where the desires in question have non-indexical content. Relative value is a matter of what would be desired if we had a set of desires that was maximally informed and coherent and unified where the desires in question have indexical content. Evaluative significance, simpliciter, is a matter of the relative strength of the non-indexical and indexical desires that we would have if we had a set of desires that was maximally informed and coherent and unified.)

²⁸ , Adams, ‘Motive Utilitarianism’, Journal of Philosophy, lxxiii (1976), pp. 477 fGoogle Scholar.

²⁹ Jackson.

³⁰ Parfit, D., Reasons and Persons, Oxford, 1984, p. 24Google Scholar.

³¹ I would like to thank Frank Jackson for helpful comments on an earlier version of this paper.