¹ Social beings that we are, we need the ability to form special relationships with others. Certain virtues such as kindness and loyalty are needed for this (Foot, Philippa, Natural Goodness, Oxford, 2001, pp. 44 f.)CrossRefGoogle Scholar .

² Perhaps there are values that are non-instrumental but also non-intrinsic. Rae Langton's example is a wedding ring. But I want to focus here on the idea that virtue is intrinsically valuable.

³ For sympathetic discussion of objective pluralism about value in human life, see Rashdall, Hastings, A Theory of Good and Evil, Oxford, 1924Google Scholar ; Finnis, John, Natural Law and Natural Rights, Oxford, 1980Google Scholar ; Parfit, Derek, Reasons and Persons, Oxford, 1984Google Scholar , appendix I; Brink, David, Moral Realism and the Foundations of Ethics, Cambridge, MA, 1989, pp. 221–36CrossRefGoogle Scholar ; Nozick, Robert, The Examined Life, New York, 1989, ch. 10Google Scholar ; Scanlon, T. M., ‘Value, Desire, and the Quality of Life’, The Quality of Life, ed. Nussbaum, M. and Sen, A., Oxford, 1993, pp. 185–200CrossRefGoogle Scholar ; Hurka, Thomas, Perfectionism, New York, 1993Google Scholar ; Crisp, Roger, Mill on Utilitarianism, London, 1997, ch. 3Google Scholar; Arneson, Richard, ‘Human Flourishing versus Desire Satisfaction’, Social Philosophy and Policy, xvi (1999)Google Scholar ; and my own Ideal Code, Real World, Oxford, 2000Google Scholar , sect. 2.3.

⁴ Hooker, B., ‘Does Being Virtuous Constitute a Benefit to the Agent?’, How Should One Live? Essays on the Virtues, ed. Crisp, R., Oxford, 1996Google Scholar .

⁵ Aristotle's answer is that being virtuous is the only way to obtain ‘the noble’ (see Nicomachean Ethics, bk. IX, ch. 8). But I cannot see why there are not other ways of obtaining nobility.

⁶ I am not supposing that virtuous people are especially concerned with whether they are rewarded for their virtue. Nor am I supposing that virtue ethicists need think that virtuous people are concerned with this.

⁷ Cf. Ross, W. D., The Right and the Good, Oxford, 1930, pp. 134 f.Google Scholar ; , Hooker, Ideal Code, Real World, pp. 34–6Google Scholar .

⁸ This is one of the lessons of Hurka's, ThomasVirtue, Vice, and Value, New York, 2001CrossRefGoogle Scholar .

⁹ Slote, Michael, From Morality to Virtue, New York, 1992, pp. 130–3, 230–4Google Scholar .

¹⁰ , Hurka, Virtue, Vice, and Value, p. 229Google Scholar .

¹¹ These points are made in Ideal Code, Real World, sect. 3.1.

¹² It is also partly motivated by the thought that feelings are important as well as action.

¹³ I discuss these matters at greater length in Ideal Code, Real World, sects. 3.5, 6.2–6.4. Let me add here that in some ways even a fairly crude act-utilitarianism accepts the need for judgement and the possibility of indeterminacy. For there are quite different pleasures and pains to be weighed against one another. Still, there is a recognizable sense in which some theories endorse and other theories deny codifiability.

¹⁴ Deontology, contractualism, and rule-consequentialism ares normally framed in terms of rules rather than virtues. I think this ultimately makes no difference. Deontology, contactualism, and rule-consequentialism are interested in rule-internalization where this includes developing dispositions to behave in certain ways, to think in certain ways, and to react in certain ways. Such theories would hold that to internalize the rules they advocate would be to come to possess the virtues.

¹⁶ I shall ignore here the implausible view, sometimes associated with Socrates, that the virtues form such a tight and seamless unity that we should think of ‘them’ as an ‘it’, i.e., a singular item.

¹⁶ See Hursthouse, pp. 41 f.

¹⁷ Ideal Code, Real World, pp. 88–91, 106 f., 133–6.

¹⁸ This is widely acknowledged. A relatively recent example: ‘If we cannot count on others' keeping their word to us unless we are so far lucky as that their keeping their word to us maximizes goodness-for [i.e., maximizes aggregate well-being], then we (simply) cannot count on others' keeping their word to us. The bearing of this on the possibility of our forming a community is obvious.’ Thomson, Judith Jarvis, ‘The Right and the Good’, Journal of Philosophy, xciv (1997), 284Google Scholar .

¹⁹ Brandt, R., ‘Morality and its Critics’, American Philosophical Quarterly, xxvi (1989), 95Google Scholar .

²⁰ Brandt, R., ‘Toward a Credible Form of Utilitarianism’, Morality and the Language of Conduct, ed. Castaneda, H.-N. and Nakhnikian, G., Detroit, 1963, p. 134Google Scholar.

²¹ For an argument that Kant's ‘universal law’ formulation of the categorical imperative is not a codification of rightness but merely a moral decision procedure, see Timmons, M., ‘Decision Procedures, Moral Criteria, and the Problem of Relevant Descriptions in Kant's Ethics’, Jahrbuch fürRecht und Ethik (Annual Review ofLaw and Ethics), Band 5 (1997)Google Scholar . Timmons contends that Kant's criterion of rightness is supplied by the ‘Humanity’ formulation of the categorical imperative (‘treat humans always as ends rather than merely as means’).

²² Admittedly, other theories can be fairly uninformative until they are fully specified. For example, the act-consequentialist principle ‘An act is morally permissible if and only if no alternative act would produce more (expected) value’ needs to specify what has value. Suppose, to take the most familiar example, the act-consequentialist principle is filled out with a hedonistic value theory. Then the principle would say that an act is morally permissible if and only if it would produce more (expected) pleasure than any alternative act would. This is informative. In contrast, even when Ross-style pluralism has been fully specified – i.e., even when all its pro tanto duties are listed and explained – it can say nothing more informative than that an act is right if and only if it is required by the most important moral duty in the situation.

²³ Hursthouse, pp. 198–201.

²⁴ While rule-consequentialism could prescribe self-regarding virtues, rule-consequentialists have tended to leave purely self-regarding behaviour to be guided by natural self-interest (e.g., Brandt, ‘Morality and its Critics’).

²⁵ , Hurka, Virtue, Vice, and Value, pp. 232 fGoogle Scholar.

²⁶ See Hurka's discussion of whether this combination of claims pushes virtue ethics to be self-effacing, that is, to condemn belief in itself, ibid., pp. 246 f.

²⁷ , Hart, The Concept of Law, Oxford, 1961, p. 85Google Scholar ; , Mackie, Ethics: Inventing Right and Wrong, Harmondsworth, 1977, p. 106Google Scholar .

²⁸ On Virtue Ethics, ch. 8 and pp. 204 f.

²⁹ Compare Hursthouse's attempt to show that her criterion of virtue is compatible with moral vegetarianism (pp. 227 f.). Admittedly, for people with access to other sources of nourishment, meat-eating is not necessary for the survival of our species. Still, it seems that the moral reason not to eat meat has primarily to do with the welfare of the animals, not (even in some derivative way) from some actual or possible benefits to us. It is not clear that Hursthouse's theory can do justice to that idea. Moreover, since I guess meat-eating does provide ‘characteristic enjoyments’ for humans, I cannot see how on balance her theory can count vegetarianism a virtue. My own view is that including ‘characteristic enjoyments’ in her criterion is another mistake, since there are obviously bad enjoyments that are characteristic of humans (as Hursthouse herself notes on p. 219). For a discussion where Hursthouse takes virtually all the descriptive content out of ‘characteristic’ and leaves it as a purely normative notion, see her pp. 221–4. See also Nafsika Athanassoulis's review of , Hursthouse in Ratio (New Series), xiv (2001)Google Scholar .

³⁰ Hursthouse, pp. 214 f.

³¹ Cf. Foot's discussion of childlessness, Natural Goodness, p. 42.

³² Note the difference between evaluating various dispositions from the point of view of impartial benevolence, which is what the most familiar form of rule-consequentialism does, and holding that from this point of view impartial benevolence would be selected as the central or chief or overriding disposition, a view that no true rule-consequentialist could hold.

³³ Thomson, 282.

³⁴ Ibid.

³⁶ Ibid., 282 f.

³⁶ Ibid., 283.

³⁷ Ibid., 286.

³⁸ E.g., , Thomson'sThe Realm of Rights, Cambridge, MA, 1990Google Scholar .

³⁹ This is rule-consequentialist, not merely split-level act-consequentialism, because Thomson's suggestion is that dispositions are to be selected for their general consequences, and acts are to be evaluated in terms of whether they follow from these dispositions.

⁴⁰ Though this is not acknowledged by Thomson herself.

⁴¹ Helpful comments on this paper have been provided by Robert Audi, Tom Baldwin, Roger Crisp, Jonathan Dancy, Sarah Goldsmith, Steve Holland, Thomas Hurka, Matt Matravers, Sue Mendus, Simon Parker, Mark Smith, Mark Timmons, and Robert Wavre.