The author wishes to express her gratitude to Les Green and Steve Newman for helpful comments on an earlier version of this essay.

1. Kymlicka, Will, Liberalism, Community and Culture (Oxford: Clarendon, 1989)Google Scholar; Raz, Joseph, The Morality of Freedom (Oxford: Clarendon, 1986).Google Scholar

2. See Walzer, Michael, “The Communitarian Critique of Liberalism,” Political Theory 18, no. 1 (1990): 6–23.CrossRefGoogle Scholar

3. Rawls, John, “Kantian Constructivism in Moral Theory,” Journal of Philosophy 77 (1980)CrossRefGoogle Scholar; Rawls, , “Justice as Fairness: Political not Metaphysical,” Philosophy and Public Affairs 14 (1985)Google Scholar; Rawls, , “The Idea of an Overlapping Consensus,” Oxford Journal of Legal Studies 7 (1987)CrossRefGoogle Scholar; Rawls, , “The Priority of Right and Ideas of the Good,” Philosophy and Public Affairs, vol. 17 (1988).Google Scholar

4. See my discussion of Rawls's, recent essays, “Justice for Our Times,” Canadian Journal of Political Science 23 (1990).Google Scholar

5. Sandel, Michael J., Liberalism and the Limits of Justice (Cambridge: Cambridge University Press, 1982), pp. 49–53Google Scholar; Taylor, Charles, Human Agency and Language: Philosophical Papers, I (Cambridge: Cambridge University Press, 1985) p. 5, pp. 29–33.CrossRefGoogle Scholar

6. Communitarians claim that underlying liberal theory is a conception of the person as identified only with the possession of autonomy. And because the identification of the person with the capacity for autonomy involves abstracting from the person her determinate characteristics, there is a serious gap or dichotomy between the conception of the person on which the theory is based and particular embodied human beings in the real empirical world. This abstraction leads liberals to engage in a discourse which does not give importance to values other than autonomy. See Beiner, Ronald, What's the Matter with Liberalism? (forthcoming), p. 47.Google Scholar This abstraction also leads the theory into difficulty in explaining why anyone would act in accordance with the principles of liberalism. This criticism is advanced by Walzer, Michael, Spheres of Justice (New York: Basic Books, 1983), p. 5Google Scholar; Williams, Bernard, Ethics and the Limits of Philosophy (London: Fontana, 1985), pp. 69–70.Google Scholar

7. Kymlicka, , Liberalism, p. 18.Google Scholar

8. Unger, Roberto M., Knowledge and Politics (New York: MacMillan, 1984), pp. 66–67Google Scholar; Jaggar, Alison, Feminist Politics and Human Nature (Totowa, NJ: Roman & Allenheld, 1983), p. 194.Google Scholar

9. Kymlicka, , Liberalism, p. 18.Google Scholar

10. Ibid., p. 13.

11. Ibid., p. 12.

12. Ibid., p. 11.

13. Raz, , Morality, p. 299.Google Scholar

14. Parfit, Derek, Reasons and Persons (Oxford, New York: Oxford University Press, 1986), p. 499CrossRefGoogle Scholar; Griffin, James, Well-Being; Its Meaning, Measurement and Moral Importance (Oxford: Clarendon, 1986), p. 107.Google Scholar

15. Raz, , Morality, p. 381.Google Scholar

16. Kymlicka, , Liberalism, p. 18.Google Scholar

17. Raz, , Morality, p. 319.Google Scholar

18. This is implicit in Kymlicka's theory, in his endorsement of the internal value argument. Notice that he does not say that autonomy confers value on things. That would have been inconsistent with his grounding in terms of the good life, which also includes substantive objective values.

19. Raz, , Morality, p. 306.Google Scholar

20. Ibid., pp. 306–307, 387.

21. Kymlicka, , Liberalism, p. 12.Google Scholar

22. Barry, Brian, “How Not to Defend Liberal Institutions,” British Journal of Political Science 20, no. 1 (1990): 5.CrossRefGoogle ScholarPubMed

23. Raz's argument is in fact more complicated than this. He distinguishes between valuing the person's capacity for autonomy and valuing a life lived autonomously. He believes that his theory is perfectionist because the ultimate, i.e., nonderivative, value of personal autonomy is not posited in the person but is an achievement to be attained. He argues that a theory which conceives of value as attaching to the capacity for autonomy is fundamentally individualist, because the source of ultimate value resides in the individual, and it is in virtue of each person's possession of this capacity that each is deemed worthy of respect. By contrast, the central concern of Raz's “perfectionist” theory is with the attainment of autonomy, and only derivatively with the capacity for autonomy. However, the distinction that Raz is relying on here is tenuous, indeed. It is difficult to see how it is possible to value the capacity for autonomy without at the same time valuing its exercise, or vice versa. The capacity for autonomy is a potential which can be actualized, and the attainment of autonomy is the achievement of the full exercise of that potential. Individualist liberals do not in fact value only the capacity, and not its exercise. That would be compatible with imprisoning people or otherwise restricting their free action, as long as one does not harm their capacity for autonomy. At the same time, the achievement of autonomy can be conceived as valuable only if the capacity is regarded as valuable, as an aspect of the person which she should develop.

24. These goods are conceived as intrinsic goods rather than ultimate goods, because they are derived from the ultimate value of personal autonomy. They are intrinsically rather than instrumentally valuable, because the range of acceptable goods from which the autonomous person chooses is part of what the exercise of personal autonomy consists in. And these intrinsic goods are public or collective in the sense that they are subject not to voluntary distribution by the individual but to collective or communal control. Raz, , Morality, pp. 198–201.Google Scholar

25. Ibid., pp. 206–207.

26. Kymlicka, , Liberalism, p. 165.Google Scholar

27. Ibid., p. 166.

28. Mandel, Michael, The Charter of Rights and the Legalization of Politics in Canada (Toronto: Wall and Thompson, 1989), p. 256.Google Scholar

29. Kymlicka, , Liberalism, p. 195.Google Scholar

30. Ibid., p. 196.

31. Ibid.

32. Ibid.

33. Ibid., p. 197.

34. Beiner, , What's the Matter with Liberalism? p. 47Google Scholar; Sandel, , Liberalism and the Limits of Justice, p. 168, 174.Google Scholar

35. Kymlicka, , Liberalism, p. 18Google Scholar; Raz, , Morality, pp. 381, 395.Google Scholar

36. Sandel, , Liberalism and the Limits of Justice, p. 178Google Scholar; Williams, , Ethics and the Limits of Philosophy, pp. 61–69, 79–80.Google Scholar

37. Raz, , Morality p. 373.Google Scholar

38. Kymlicka, , Liberalism, p. 166.Google Scholar

39. Raz, , Morality p. 6, note.Google Scholar

40. Ibid., pp. 373–374.

41. Ibid., p. 205.

42. Kymlicka, , Liberalism, p. 166.Google Scholar

43. Raz, , Morality, p. 371.Google Scholar

44. Ibid., pp. 411–412.

45. The neutral or independent starting-point is associated mainly with Rawls's A Theory of Justice, which provoked the fruitful debate between liberals and communitarians. In that book, Rawls claimed that his theory of justice was neutral among competing conceptions of the good, that the conception of the original position provided an “Archimedean point” which justified his principles in adjudicating among rival moral conceptions (or, more precisely, actions based on rival moral conceptions). Rawls justified the regulative role of the principles of justice on the grounds that the principles of justice are derived independently of all socially-acquired and so contingent moral ideals and conceptions. The principles of justice are justified because agreed to or derived from a standpoint (the original position) in which choice cannot be influenced by socially relative factors, for no feature of social circumstance, such as unequal bargaining power or conceptions of the good, can enter into deliberation and so affect the principles agreed to. Rawls, , A Theory of Justice (Cambridge, MA: Harvard University Press, 1971), pp. 260–263, p. 587.Google Scholar

46. This criticism of liberal theory is advanced by Walzer, , Spheres of Justice, p. 5Google Scholar; Williams, , Ethics and the Limits of Philosophy, pp. 69–70.Google Scholar