1. I use “ethical liberalism” as the name for this view rather than the “comprehensive liberalism” that Rawls favors, even though he does sometimes contrast the ethical and the political in the way I do (e.g., Rawls, John, Political Liberalism [New York: Columbia University Press, 1993, p. 78Google Scholar]). The label Rawls prefers is a bit misleading since it applies to liberalisms whose constitutive ethical doctrine is only partially comprehensive. Further, it is at least doubtful that any truly liberal ethical doctrine could be fully comprehensive since its distinctive emphases on freedom, diversity and innovation are not obviously consistent with the idea of a systematic ordering of all relevant values which a fully comprehensive ideal entails. Indeed, the degree to which some species of ethical liberalism might be open to diversity makes even the appellation “partially comprehensive” seem less than apt. See ibid., p. 13.

2. Ibid., pp. xv-xvi, 99, 199-200.

3. Ibid., p. 37.

4. Ibid., pp. 227-30

5. Ibid., pp., 37-38.

6. Ibid., pp. 11-22, 63-64.

7. Ibid., p. 19.

8. Ibid., p. 54.

9. Ibid., p. 223.

10. Ibid., p. 218.

11. Ibid., p. 55.

12. Ibid., pp. 56-58.

13. Ibid., pp 12-13.

14. Rawls says that such doctrines have three main features. They engage theoretical reason in offering a “more or less consistent and coherent” interpretation of the main aspects of human life; they involve practical reason in singling out the values that should guide our conduct; and they normally belong to a tradition of thought and doctrine. Rawls describes this account as “deliberately loose” because it is intended to be as accommodating to diversity as possible (ibid., p. 59) There is no suggestion here that the relevant exercise of theoretical or practical reason has to be even moderately competent—the most outlandish dogma might satisfy these three conditions. Rawls concedes that many doctrines we regard as “plainly unreasonable and untrue” outside public reason are nonetheless reasonable doctrines in the sense relevant to political liberalism (ibid., p. 60n). Elsewhere he does place more substantive limits on reasonable doctrines. They cannot “reject the essentials of a democratic regime” (ibid., p. xvi). For example, forms of fundamentalism which deny that faith must be freely given are not reasonable doctrines (ibid., p. 170). Of course, many varieties of fundamentalism endorse some principle of free faith.

15. ibid.

16. See Audi, Robert, “The Separation of Church and State and the Obligations of Citizens,” Philosophy and Public Affairs, 18 (1989): 259–96Google Scholar.

17. Rawls, , Political Liberalism, pp. 66–71Google Scholar.

18. Ibid., p. 66.

19. Ibid., p. 68

20. Ibid., pp. 12-13.

21. Ibid., p. 56.

22. I assume here that we have ruled out the kind of integrity-destroying compartmentalization that I discussed earlier as a way of preserving doctrines learned in the nonpublic sphere from doubts provoked by the burdens of judgment.

23. Macedo, Stephen, Liberal Virtues (Oxford: Clarendon Press, 1991), p. 265Google Scholar.

24. Rawls, , Political Liberalism, p. 63Google Scholar.

25. Nagel, Thomas, Equality and Impartiality (Oxford: Oxford University Press, 1991), p. 168Google Scholar.

26. Psalms 93:11.

27. Rawls, , Political Liberalism, p. 37Google Scholar.

28. Ibid., p. 199.

29. Ibid., p. 200.

30. My defense of this conclusion has focused exclusively on Rawls's version of political liberalism. But substantially the same conclusion would be reached if I had addressed the work of other exponents of political liberalism who have written about political education. I have in mind recent essays by Kenneth Strike and Stephen Macedo. Both Strike and Macedo are acutely aware of some of the difficulties I discuss here. Neither assumes that a conception of liberal politics can be freestanding in Rawls's sense. Yet both underestimate the porousness of the distinction between public and nonpublic spheres once the centrality of personal autonomy to liberal politics is acknowledged, and hence neither of them accepts the concomitant collapse of political into ethical liberalism. However, I cannot do justice to the subtlety of Strike's or Macedo's views in a footnote. See Strike, Kenneth, “On the Construction of Public Speech: Pluralism and Public Reason,” Educational Theory 44 (1994): 1–26CrossRefGoogle Scholar; Macedo, Stephen, “Liberal Civic Education and Religious Fundamentalism: The Case of God v. John Rawls,” Ethics 105 (1995): 468–96CrossRefGoogle Scholar and “Liberal Civic Education and its Limits: A Comment on Eamonn Callan”, Canadian Journal of Education 20 (1995): 304–14CrossRefGoogle Scholar. I offer a more detailed response to Macedo, in “Rejoinder: Pluralism and Polarization”, Canadian Journal of Education 20 (1995): 325–29Google Scholar.

31. Here is a representative example: “As we have seen, feminists (along with others) have been critical of the connotations of exaggerated individualism, detached spectatorship and Faustian desires for control that Western notions of autonomy have carried” (Farley, Margaret A., “Feminism and Universal Morality,” in Prospects for a Common Morality, ed. Outka, Gene and Reeder, John P. Jr., [Princeton, NJ: Princeton University Press, 1993], p. 182Google Scholar).

32. E.g., see Kymlicka, Will, Liberalism, Community, and Culture (Oxford: Oxford University Press, 1989), pp. 162–81Google Scholar; Friedman, Marilyn, “Feminism and Modern Friendship: Dislocating the Community,” Ethics 99 (1989): 275–90CrossRefGoogle Scholar; Dworkin, Ronald, “Liberal Community,” California Law Review 77 (1989): 478–506CrossRefGoogle Scholar; Macedo, Stephen, Liberal Virtues, pp. 203–253Google Scholar; and my own “Tradition and Integrity in Moral Education,” American Journal of Education 101 (1992): 1–28CrossRefGoogle Scholar.

33. Ackerman, Bruce, Social Justice and the Liberal State (New Haven: Yale University Press, 1980), pp. 139–67Google Scholar.

34. Gutmann, Amy, Democratic Education (Princeton: Princeton University Press, 1987), pp. 48–70Google Scholar. See also Barry, Brian, “How Not to Defend Liberal Institutions” in Liberalism and the Good, ed. Douglass, R. Bruce, Mara, Gerald R., and Richardson, Henry S. (New York: Routledge, 1990), pp. 46,54Google Scholar.

35. Galston, William, Liberal Purposes (Cambridge: Cambridge University Press, 1991), pp. 241–56CrossRefGoogle Scholar; Lomasky, Loren, Persons, Rights, and the Moral Community (New York: Oxford University Press, 1987), pp. 171–87Google Scholar.

36. The relevant question here is how we are to understand the common education that liberal politics requires on its best interpretation, and by itself that does not settle the problem of whether the state should intervene coercively when particular groups or individuals reject the required common education. The two issues are not always clearly discriminated. The distinction between the problem of identifying a liberal theory on some broad matter of policy like political education or citizenship and the problem of imposing that theory on others is well-drawn in Kymlicka, Will, “The Rights of Minority Cultures: Reply to Kukathas,” Political Theory 20 (1992): 144–45CrossRefGoogle Scholar. I have canvassed some reasons for circumspection in the imposition of liberal educational values in “Common Schools for Common Education” and “Rejoinder: Pluralism and Polarization,” Canadian Journal of Education 20 (1995): 251–71; 315–32Google Scholar.

37. Rawls, , Political Liberalism, p. 158Google Scholar.

38. Ibid., p. 159.

39. Ibid., pp. 161-62.

40. Ibid., p. 163.

41. Ibid., pp. 163-64

42. Ibid., p. 163.

43. Ibid., pp. 164-65.

44. Ibid., pp. 161-62.

45. Ibid., p. 165.

46. See my “Beyond Sentimental Civic Education,” American Journal of Education 102 (1994): 190–221CrossRefGoogle Scholar.

47. Hampshire, Stuart, “Liberalism: The New Twist,” The New York Times Review of Books, 40, 14 (1993): 47Google Scholar.

48. I am grateful to the anonymous reviewers for The Review of Politics for their very helpful comments on an earlier draft. For both criticism and encouragement, I am also indebted to Stephen Macedo, and Will Kymlicka. Research for this paper was made possible through the generous support of the Social Sciences and Humanities Research Council of Canada.