¹ Earlier versions of this paper were given at the Department of Politics at the University of Glasgow (December 1995) and University College London (April 1996). I am grateful to the participants at both for their helpful suggestions. I am also grateful to Andrew Mason and Joseph Chan for helpful Comments on an earlier draft of this paper.

² Barry, B., Justice as Impartiality, Oxford, 1995Google Scholar. Hereafter all references to this work will be inserted in the text.

³ See also ‘In Defense of Political Liberalism’, Ratio Juris, vii (1994), 326Google Scholar. For other helpful accounts of the nature of neutrality see Kymlicka, Will, ‘Liberal Individualism and Liberal Neutrality’, Ethics, ic (1989), 883–4CrossRefGoogle Scholar; Larmore, Charles, Patterns of Moral Complexity, Cambridge, 1987, pp. 43–7CrossRefGoogle Scholar; and Rawls, John, Political Liberalism, New York, 1993, pp. 190–5Google Scholar.

⁴ Ibid., pp. 214, 227–30.

⁵ Barry refers, e.g., to ‘the virtually unanimous concurrence of the human race in caring about the defensibility of actions in a way that does not simply appeal to power’, Barry, , Theories of Justice, London, 1989, p. 285Google Scholar. See also ibid., pp. 284, 363.

⁶ Andrew Mason has suggested one possible response to this line of argument. As he points out, one might make a distinction between particular ‘judgements about the good’ and ‘traditions of thought about the good’, where the former denote individual judgements about particular activities and the latter denote systems of ethical beliefs. He then suggests that scepticism about the latter (that is, bodies of thought about the good) is more plausible than scepticism about the former (that is, individual judgements about particular pursuits). See Mason, ‘Justice, Contestability, and Conceptions of the Good’, this journal, n. 8.

Two responses can be made to this suggestion. First, it is not clear to me that scepticism about all ‘bodies of thought about the good’ is a plausible position. Consider, for example, the views discussed in examples (a) and (b) above. Can we plausibly argue that affirmation of these sets of beliefs about the good is reasonable? Alternatively consider a variation on example (e): consider someone whose whole outlook on life is focused simply on acquiring power, money, and status, and mixing with the rich and famous. Now this is not simply an individual judgement but a body of beliefs about the good that governs decisions such as who one should befriend (those who can help promote your career) or marry (someone wealthy) or what one should wear, where one should holiday, what pursuits one should adopt, etc. (i.e. in short, how one should lead one's life). My suggestion is that it is implausible to claim that this is a worthwhile ‘body of thought about the good’.

Second, it is worth noting that if Barry adopts the position Mason describes then it is not clear how, if at all, his position would differ from the perfectionist position affirmed by philosophers like Joseph Raz (cf. The Morality of Freedom, Oxford, 1986)Google Scholar, since both would allow principles of justice to be informed by judgements about the good. For an instructive discussion of related issues see Chan, Joseph, ‘Aristotelian and Liberal Conceptions of Political Community: Political Perfectionism vs. Liberal Neutrality’, unpublished paper, pp. 14–19Google Scholar.

⁷ Suppose that someone denies these claims. Here it might be appropriate to observe that some of our judgements about the good are unreasonable. Judgements about the good might, for example, be unreasonable because of their reliance upon one or more of the following defects: incorrect beliefs, inadequate information, selective information, invalid logical reasoning, self-deception, or wishful thinking.

⁸ I am not committed to the claim that all the above judgements should be incorporated into a theory of justice. My claim is only that scepticism does not provide a good reason for excluding them. There may, of course, be other reasons for not including them.

⁹ See the characterization of pluralism given by Raz, Joseph, ‘Liberalism; Scepticism, and Democracy’, in Ethics in the Public Domain: Essays in the Morality of Law and Politics, Oxford, 1994, pp. 103–4Google Scholar.

¹⁰ ‘Contractualism and Utilitarianism’, in Utilitarianism and Beyond, ed. Sen, Amartya and Williams, Bernard, Cambridge, 1982CrossRefGoogle Scholar. Scanlon writes that ‘An act is wrong if its performance under the circumstances would be disallowed by any system of rules for the general regulation of behaviour which no one could reasonably reject as a basis for informed, unforced general agreement’, ibid., p. 110.

¹¹ The following discussion is indebted to the instructive but different taxonomy of different types of contractarianism outlined by Kukathas, Chandran and Pettit, Philip, Rawls: ‘A Theory of Justice’ and its Critics, Cambridge, 1990, pp. 27–35Google Scholar.

¹² Thomson, Judith Jarvis, The Realm of Rights, Cambridge, Mass., 1990, p. 30 n. 19Google Scholar. Thomson's objection is against the Scanlonian claim that moral rules are correct because no one can reasonably reject them but her objection is as forceful against the contention that rules of justice (like prohibitions of murder or torture) are correct because no one can reasonably reject them.

¹³ See, ibid., pp. 188–9 n. 5. For other objections to Scanlon's method see Dworkin, Ronald, ‘Foundations of Liberal Equality’, Tanner Lectures on Human Values, xi (1990), 28–31Google Scholar.

¹⁴ I shall assume in what follows that the principles to be discussed are concerned with the distribution of burdens and benefits. This enables me to employ the shorter statement that ‘X cannot be reasonably rejected’ in place of the more cumbersome statement that ‘X is a principle governing the distribution of burdens and benefits which cannot be reasonably rejected’.

¹⁵ See also Theories of Justice, pp. 288–9, 348–9.

¹⁶ See Ronald Dworkin, ‘Foundations of Liberal Equality’; Kymlicka, Will, Liberalism, Community and Culture, Oxford, 1989, pp. 10–12Google Scholar; Kymlicka, Will, Contemporary Political Philosophy: An Introduction, Oxford, 1990, ch. 2, § 2, esp. pp. 14–16, and ch. 6, esp. pp. 202–3Google Scholar.

¹⁷ This raises the question of why Barry does phrase it in the way that he does. The most likely explanation is that Barry wishes to counter any suggestion made by advocates of ‘justice as mutual advantage’ that the only reason we have to do something is that it furthers our self-interest. Barry's point is that we recognize other reasons for action. See Theories of Justice, pp. 284–5.

¹⁸ This account of justice, it should be noted, is compatible with a perfectionist position. A perfectionist can plausibly argue that a perfectionist state best enables people to flourish (and therefore meets condition (i)) and can do so by adopting procedures which command widespread support (and therefore meets condition (ii)).

¹⁹ Barry unpacks this moral component in Theories of Justice. He writes ‘Flatly asserting “It's contrary to my interest and that's why I oppose it” will be ruled out. It is simply not a valid move in the game. Beyond that, we must invoke a requirement of good faith: that people will not put forward frivolous or tendentious arguments for a position that they actually hold only because it is personally advantageous’, p. 352.

²⁰ For an outline of political procedures which would bestow legitimacy on the principles chosen, see Caney, Simon, ‘Anti-Perfectionism and Rawlsian Liberalism’, Political Studies, xliii (1995), 255–6Google Scholar.

²¹ Consider also Barry's, claim that ‘[t]he essential idea is that fair terms of agreement are those that can reasonably be accepted by people who are free and equal’ (p. 112: emphasis added)Google Scholar. Again this does not require that principles of justice must not rely on claims which reasonable persons reject. It is compatible with a political system in which decisions concerning matters of justice are the product of a political procedure (which reasonable people endorse) in which people may appeal to their ideals of the good (even if they are claims which could reasonably be rejected). People could accept the ‘terms of agreement’ because of the way in which the decision was made.

²² See Rawls, , Political Liberalism, pp. 75–6, 178–90Google Scholar.

²³ One suggestion (made to me by both Brian Barry and Peter Jones) is that neither of the proposals described above can accurately be described as an ‘educational’ system since the latter, by its very nature, must incorporate judgements about the good. A full neutralist can, I think, plausibly reply that this definitional move is too hasty and that we can define education in terms of providing a general preparation for life (whatever one's conception of the good) and an understanding of one's society and the world. On this broad definition of education, both of the neutralist schemes described in the text could count as educational systems.

²⁴ For (a) see Rawls, John, ‘Fairness to Goodness’, Philosophical Review lxxxiv (1975), 539Google Scholar, and Rawls, , Political Liberalism, pp. 193–4Google Scholar. For (b) see Kymlicka, , ‘Liberal Individualism and Liberal Neutrality’, pp. 884–5Google Scholar.

²⁵ I am grateful to Michael Lessnoff for suggesting the following line of argument.

²⁶ Barry, , Theories of Justice, p. 356Google Scholar.

²⁷ Ibid., p. 355.