I am grateful to Will Kymlicka, Hilliard Aronovitch and Wayne Norman, all of whom read drafts of this paper and provided helpful comments and criticisms.

1. Charles, Larmore Patterns of Moral Complexity (Cambridge: Cambridge University Press, 1987) at 102.Google Scholar

2. Many of the quotations in this paper are taken from Will, Kymlicka Liberalism, Community and Culture (Oxford: Clarendon Press, 1989).Google Scholar Page numbers in the text refer to this book.

3. Will, Kymlicka Contemporary Political Philosophy (Oxford: Oxford University Press, 1991) at 204.Google Scholar Much of the same material from Kymlicka, supra, note 2 has been reworked and elaborated in interesting ways in this book.

4. Ibid, at 203.

5. In conversation Kymlicka has in fact expressed some sympathy for this view.

6. I use this word with some reluctance. Obviously, it carries a lot of baggage. So I should underline that I am using it in a quite particular, and perhaps controversial, way. I do not deny that there are other senses of the term which might be quite appropriate to members of tribal societies.

7. Someone might wonder about individuals like, for example, slaves who are taken by one tribal society to another and then forced to live in the latter. Are they not being forced to live by a set of values which they reject? This is a complex and difficult question. Let me just hint at what an answer might look like.

Someone might wonder about individuals like, for example, slaves who are taken by one tribal society to another and then forced to live in the latter. Are they not being forced to live by a set of values which they reject? This is a complex and difficult question. Let me just hint at what an answer might look like.

Secondly, the situation need not be as clear cut as I am making it sound. Indeed, it is the burden of my argument that moral autonomy is not something one either has or not. It admits of degrees. So, in tribal as in liberal societies, what we really want to say is that the internal/external distinction applies to the extent that (1) self- conscious, critical, moral reflection is possible; and (2) takes place in the individual in question.

I do not, therefore, see any reason to deny categorically that individuals in tribal societies canengage in actions we might sometimes want to call autonomous. There may even be room here to speak of ‘forms of autonomy’, though we must guard against diluting the concept to the point that it loses its special moral significance.

My point here is just that, in contrast to liberal societies, autonomous actions in tribal societies would be at least much rarer and the degree of autonomy much weaker. Most importantly, autonomy is a value which such societies are not designed to promote—indeed, they would generally discouraged it.

8. John, Rawls A Theory of Justice (Cambridge, Mass.: Harvard University Press, 1971) at 92–95.Google Scholar

9. I must say that this example seems to me to raise more questions than it answers. Kymlicka claims that French-Canadians constitute(d) this cultural community before and after the Quiet Revolution. But, in fact, one of the consequences of the Quiet Revolution seems just to have been a fragmenting of the French-Canadian cultural community.

Before 1960, when the Catholic church was still the most important agent of French-Canadian nationalism, the principal focus of this nationalism was language and religion. These, at bottom, constituted the distinctive elements of the identity of the French-Canadian nation. Hence at that time there was a much greater sense of solidarity between French-speaking Canadians inside and outside Quebec. Consequently, it is at least plausible to argue that at that time there was a single, more-or-Iess unified cultural community which identified itself as French- Canadian. But when the authority of the Church was undermined during the Quiet Revolution, the mantle of leadership passed to the state, i.e., the Quebec provincial government. Since then citizenship in that province seems to have become almost a necessary condition for membership in ‘the nation’.

This opened a gulf between French-speaking Canadians inside and outside Quebec. The latter have tried to cope with this new nationalism by attempting to reassert their own historical identities as e.g., franco- Ontarians, franco-Manitobans, etc. But, insofar as they have succeeded, it has been due in large part to the patronage and protection of the federal government. Unfortunately, this leaves the existence of these cultural groups tragically vulnerable to shifts in public opinion and policy.

So today there is a plurality of relatively autonomous groups of French-speaking Canadians, few of whom still think of themselves primarily as French-Canadians. Certainly nationalists in Quebec would bridle at that description: They regard themselves, first and foremost, as Québécois. So, pace Kymlicka, it seems to me that it is something of an open question just how far it makes sense to speak of French-speaking Canadians today as members of the same cultural community as the French-Canadians before 1960.

10. Allen, Buchanan “Assessing the Communitarian Critique” (1989) 99 Ethics 852 Google Scholar, points out that Michael Sandel's critique of Rawls, which makes much of this remark, misunderstands it by interpreting it out of context. This may well be. But it need not affect my point here. For I am not concerned about what Rawls actually meant in the disputed passage so much as what Sandel and others have taken him to be mean. For the particular point they are making is, I think, an important one. The extent to which it does, or does not, apply to Rawls is a matter for dispute.

11. For an careful discussion of this distinction see Robert, McRae Leibniz (Toronto: University of Toronto Press, 1976).Google Scholar

12. David, Hume A Treatise of Human Nature, Book I, Section VI: ‘Of Personal Identity’Google Scholar

13. See Immanuel, Kant’s Critique of Pure Reason, ‘Deduction of the Pure Concepts of the Understanding’ (Second Edition), Section 2, Transcendental Deduction of the Pure Concepts of the Understanding.Google Scholar

14. That Kymlicka approvingly attributes this distinction to Rawls is, I think, just more evidence that his use of the inside/outside distinction is grounded in some prior notion of, and commitment to. rational autonomy. See supra, note 2 at 5–6.

15. Notice that, once again, the argument rests on the distinction between an autonomous self which can think itself to be independent of the values it endorses and the empirical self which is simply given in introspection. Kymlicka reading of communitarians seems to assume that they either reject or do not see this distinction.

16. In fairness, communitarians often do talk in loose and provocative ways that seem—probably do—attribute to Rawls or Dworkin or Kant crude views about the self and overly abstract notions of practical reason. Sandel and Maclntyre, to name only two, seem guilty of this—as many of their critics have pointed out. Indeed, the tensions and conflicts in their positions can easily turn the interpretation of their own texts into a mug’s game where one quote is refuted with another. Perhaps the most charitable way to deal with this problem, the approach I am trying to take here, is to bear in mind that they are struggling to articulate and reconcile some deep, unorthodox and conflicting intuitions about the nature of, and relations between, the self, society and practical reason. The persistent and confusing vagueness of their principal claim that the self is ‘situated’ is at the heart of all this. Typically, in the struggle to sort things out, they have often found it convenient to contrast their views with those of opponents which then serve as a foil against which to work. But it is easy to oversimplify in the process. Communitarians have hardly been immune.

17. Kymlicka has objected here that, in the point under dispute, he was addressing Sandel, not Taylor. The latter, he says, has different—and more sophisticated—views. Perhaps. But throughout the chapter Kymlicka speaks of ‘communitarians’ in a generic sense. I also have done this in places, though not without reservation. I think it is fair, then, to show that, insofar as there is something called ‘communitarianism’, it is not without the resources to reply to Kymlicka’s charges.

18. See Charles, Taylor Sources of the Self (Cambridge, Mass.: Harvard University Press, 1989) at 29, 47,64.Google Scholar

19. Taylor is, after all, an admirer—albeit a critical one—of Nietzsche. See ibid, at c. 3, s. 2.

20. For a statement of Kymlicka’s interpretation of Rawls, see supra, note 2 c. 3.

21. Supra, note 2 c. 3, Kymlicka maintains that the argument from the Original Position in A Theory of Justice is meant only as a tool for modeling the more important ‘intuitive argument’ according to which we recognize that it is unfair that the distribution of resources be determined by our natural talents. For example, in the lottery of talents, some have been endowed with considerable intelligence, others with little. The fact that Jane is intelligent, therefore, is a morally neutral fact about her for which she should be neither praised or blamed. A liberal theory of justice should reflect our intuition that the possession of such talents is brute luck and, rather than rewarding those who possess them and punishing those who do not, should try to compensate those who through no fault of their own suffer as a result of the lack of them.

This analysis greatly downplays the significance of the Original Position while playing up the role Rawls accords to intuition. The correctness of this reading is, of course, a matter which can be settled only by looking closely at the text. But whatever the outcome, I do not think my argument here is affected. I am not denying that Rawls thinks that a deep intuitive sense of who we are as persons is guiding the whole project. I think he probably does. I am rather asserting that this intuition turns out to be one in which our status as moral agents is defined in terms of our status as autonomous, rational agents. But nor am I denying that we actually have a deep intuition of our moral life as constituted along these lines. Indeed, I am convinced that we do. My point is that, pace Kymlicka (and perhaps Rawls): (a) we also have other, equally powerful and conflicting intuitive pictures of ourselves as moral beings—one need only turn to Nietzsche or Freud to see this—and (b) there is no non-question-begging way to establish the primacy of the Rawlsian one. For a similar criticism of Rawls, see Gerald, Doppelt “Is Rawls’s Kantian Liberalism Coherent and Defensible?” (1989) 99 Ethics 815.Google Scholar

22. Bernard, Williams Ethics and the Limits of Philosophy (Cambridge, Mass.: Harvard University Press, 1985) at 63.Google Scholar

23. This point is eloquently and convincingly made by Hilliard Aronovitch in The Primacy of Politics, in manuscript.

24. Supra, note 18 at 27.

25. In a forthcoming book entitled The Morality of Political Divorce from Fort Sumpter to Lithuania to Quebec, Allen Buchanan makes a criticism of Kymlicka which picks up on this point.

26. This, it should be remembered, was the distinction which was to distinguish his own arguments from those of communitarians.

27. For an interesting discussion of this claim, see Charles, Taylor “What is human agency?” and “Self-Interpreting Animals” in Philosophical Papers, vol. 1 (Cambridge: Cambridge University Press, 1986).Google Scholar

28. Of course, they differ greatly in how they cash out the crucial but vague clause ‘in such a way’. On the one hand, it comes down to in some way accepting the idea of a ‘situated’ self. And, in a general and intuitive way, communitarians pretty much agree about this. But when it comes to working out the details, there are almost as many versions as there are communitarians. No one, as far as I can tell, has yet given a really definitive account of this idea the way, say, Kant did for transcendental theories of the self or Hume did for empirical ones.

29. See the section above ‘The inside/outside distinction’.

30. There has been much debate, for example, of Rawls’ ‘communitarian shift’. For some interesting discussions of various aspects of this see “Symposium on Rawlsian Theory of Justice: Recent Developments” (1989) 99 Ethics 695.Google Scholar