1. A quick search of the indices of journals in philosophy, in the social sciences, and in the humanities bear out this observation. See also books on the subject published in the last several years: Lindley, Richard, Autonomy (London: Macmillan, 1986);CrossRefGoogle ScholarYoung, Robert, Personal Autonomy (London: Croom Helm, 1986);Google ScholarHaworth, Lawrence, Autonomy: An Essay in Philosophical Psychology and Ethics (New Haven: Yale University Press, 1986);CrossRefGoogle ScholarDworkin, GeraldThe Theory and Practice of Autonomy (New York: Cambridge University Press, 1988);CrossRefGoogle ScholarChristman, John, ed., The Inner Citadel (New York: Oxford University Press, 1989);Google ScholarMeyers, Diana T., Self, Society, and Personal Choice (New York: Columbia University Press, 1989);Google Scholar and Wolf, Susan, Freedom Within Reason (New York: Oxford University Press, 1990).Google Scholar Gerald Dworkin argues persuasively that autonomy has been important to contemporary political philosophers John Rawls, Ronald Dworkin, and Thomas Scanlon (op. cit., pp. 3–4). While that is so, it is only more recently that autonomy has become the focus ofcontemporary moral and political philosophers.

2. Walzer, , Spheres of justice (New York: Basic Books, 1983), p. 298.Google Scholar

3. See, for example, Held, David, Models of Democracy (Cambridge: Polity Press, 1987)Google Scholar, especially the last chapter; and Graham, Keith, The Battle of Democracy (Brighton, England: Wheatsheaf Books, 1986).Google Scholar

4. Arblaster, , The Rise and Decline of Western Liberalism (Oxford: Basil Blackwell, 1984), p. 22Google Scholar; Nedelsky, , “Reconceiving Autonomy: Sources, Thoughts and Possibilities,” Yale Journal of Law and Feminism, 1, no. 1 (1989): 12.Google Scholar Along the same lines, William Sullivan comments that liberal thinkers see autonomy as isolated from sociability. See Reconstructing Public Philosophy (Berkeley: University of California Press, 1982), p. 168.Google Scholar For Lukes's, position, see Individualism (Oxford: Basil Blackwell, 1973).Google Scholar

5. Berlin, , Four Essays on Liberty (Oxford: Oxford University Press, 1969), pp. 135–36.Google Scholar

6. Jennifer Nedelsky remarks that “autonomy is so closely tied to the liberal tradition that it is often treated as symbolizing the very individualism from which I am trying to reclaim it” (“Reconceiving Autonomy,” p. 10). I am not entirely sympathetic to the view that individualism as defined by Nedelsky (ibid., p. 11) and by communitarian critics is at the heart of liberalism. It is problematic whether such a claim has purchase and is certainly unfounded that all versions of liberalism have always been or must be based on “individualism” in the pejorative sense. Fora different understanding of the liberal self, see Crittenden, Jack, Beyond Individualism:Reconstituting the Liberal Self (New York: Oxford University Press, 1992).Google Scholar

7. Taylor, , Sources of the Self (Cambridge, MA: Harvard University Press, 1989), p. 36.Google Scholar

8. Gray, , “Political Power, Social Theory, and Essential Contestability,” in The Nature of Political Theory, ed. Miller, D. and Siedentop, L. (Oxford: Oxford University Press, 1983), p. 89.Google Scholar

9. Ibid., p. 87.

10. Ibid., p. 88.

11. Ibid., p. 86.

12. Raz, , “Right-Based Moralities,” in Theories of Rights, ed. Waldron, J. (Oxford: Oxford University Press, 1984), p. 191.Google Scholar

13. Benn, , “Freedom, Autonomy and the Concept of A Person,” Proceedings of the Aristotelian Society, 76 (1975/1976): 129–30.Google Scholar My understanding of autonomy as process is similar to Gerald Dworkin's view of autonomy as “procedural independence.” See Dworkin, Theory and Practice, especially chap. 1.

14. Ibid., pp. 123–24.

15. Gray, , “Political Power, Social Theory, and Essential Contestability,” p. 24.Google Scholar

16. This distancing or separation or stepping back reflectively is not simply from one's own thought but also from oneself. To scrutinize one's own thought requires taking up a perspective that is not, or is not yet, one's own. To scrutinize one's own thought-one's own principles, actions, decisions, standards, ideals from a position outside that thought is also to scrutinize one's self constituted by such thought. Stanley Benn conveys the point through the image of a potter: “[I]n judging his own pots, knowing them to be his own... his judgment goes beyond assessing the pots, not merely to an assessment of himself as a potter but to an assessment of his life as a person, for being a potter is a central element in his being the person he is“ (Benn, , A Theory of Freedom [Cambridge: Cambridge University Press, 1988], p. 178).CrossRefGoogle Scholar So the autonomous person, in critically thinking about his thinking, asks himself what it means to be a person who acts this way, holds these views, does these things. Moreover, he asks what he would be, what his life would be like, if he held different views and lived in a different way. Unless one scrutinizes one's thought-steps back reflectively from it-and thus steps back reflectively from one's self, one cannot be autonomous. Diana Meyers argues that personal autonomy “requires living in harmony with one's... true—one's authentic—self” (Self, Society, and Personal Choice, p. 19). Self-reflectivity, on this view, is essential to understanding one's authentic self, for Meyers does not see the authentic self as static. Rather, it is the result of an introspective process whereby the person evaluates and adjusts commitments, wants, and beliefs according to the kind of self she takes herself to be. Meyers, I think, makes too much of the idea of the authentic self, and too much of socially conditioned values or interests. For even if one's interests and values are the result of socialization, the ability to introspect, to examine these ideas in light of one's self-definition (i.e., self-reflectivity) means that the person can change these values and interests even if socialized and even if self-reflection itself is socialized. What if socialization predetermines the outcome of introspection? That possibility is why role-taking-taking up the perspectives of others, especially those opposite yours, as if they were one's own—is crucial to self-reflectivity, because one introspects not by assuming an Archimedean position but by taking on the perspectives of others, of those different from oneself. Different perspectives enable one to get outside of himself and thereby assess his ends, values, and beliefs. Thus the standards of assessment are independent of, because different from, one's own nomoi, but not independent in some universal, objective sense. For another view of the importance of self-reflectivity to autonomy, see Young, Robert, “Autonomy and the ‘Inner Self,' in Christman, The Inner Citadel,” pp. 81–83.Google Scholar For the importance of self-reflectivity to autonomy from the perspective of developmental psychology, see Crittenden, Beyond Individualism.

17. Habermas, Jürgen, “Moral Development and Ego Identity,” Telos 24 (1975): 85.Google Scholar

18. Ibid., p. 86.

19. Ibid., p. 90.

20. It should be kept in mind that while the subject is free of any particular context, the subject is not free of all contexts, a point to which I shall recur.

21. While a slave might well be able to be self-reflective and thus autonomous, that hardly means that the slave can control the rules that shackle him. Yet that reality, it seems to me, is a question of physical and social freedom, not of autonomy. While not physically free, the slave is mentally free. Slavery therefore might not fetter the mind. That the slave can control. For a fuller discussion of the importance of developmental psychology to the arguments in political theory about the nature of the self or personal identity, see Crittenden, Beyond Individualism.

22. Berlin, , Four Essays, p. 136.Google Scholar

23. Rationality is actually built into self-reflectivity, for it is by principles of reason, and logic, even if misapplied, that one reflects at all. For the sake of emphasis, however, I shall continue to separate them.

24. For an argument that rationality is ineluctably tied to cultural traditions, thus creating rationalities, see Macintyre, Alasdair, Whose Justice, Which Rationality? (Notre Dame, IN: University of Notre Dame Press, 1988)Google Scholar, especially chaps. 1 and 18.

25. Taylor, , “The Person,” in The Category of the Person, ed. Carrithers, M. et al. (Cambridge: Cambridge University Press, 1985), p. 269Google Scholar. Or as Evans-Pritchard pointed out: “[O]ne cannot well express in his language objections not formulated by his culture” (Witchcraft, Oracles and Magic among the Azande [Oxford: Oxford University Press, 1976], p. 150Google Scholar); quoted in Shorter, John, Social Accountability and Selfhood (Oxford: Basil Blackwell, 1984), p. 219Google Scholar.

26. See his discussion of language use in Social Accountability and Selfhood., pp. 178–81. This issue about the framework of concepts and the horizon of evaluations raises a problem about this account of autonomy. I shall recur to the problem in the Conclusion.

27. Ibid., p. 38.

28. Taylor, , “Overcoming Epistemology,” in After Philosophy, ed. Baynes, K. (Cambridge, MA: The MIT Press, 1987), p. 471.Google Scholar

29. Joel Feinberg argues that among other traits responsibility entails “answerability,” “accountability,” and “liability to the judgment... that certain propositions are true” (“Autonomy,” in Christman, , The Inner Citadel, p. 42.)Google Scholar

30. Lukes, , Individualism, p. 128.Google Scholar

31. Shorter, , Social Accountability and Selfhood, p. 73.Google Scholar Robert Young points out that the central problem of the heteronomous person is that he has uncritically accepted or even unconsciously holds the views of others. Such a person may be enthusiastic about the views he holds, may even identify strongly with them, but he will “be at sea in furnishing explanations” ("Autonomy and ‘Inner Self “p. 45)

32. Berlin, , Four Essays, p. 131Google Scholar, emphasis added.

33. Taylor, , Sources of the Self, p. 168.Google Scholar

34. Ibid., pp. 37, 38.

35. Dworkin, , Theory and Practice, p. 17.Google Scholar

36. John Shorter says that self-consciousness, self-reflectivity, means that persons see themselves as responsible for their actions (Social Accountability and Selfhood, p. 42).

37. V. S. Naipaul provides a lucid illustration of this in his novel A Bend in the River. Salim, the protagonist, says that his family “were not fools. My father and his brothers were traders, businessmen; in their own way they had to keep up with the times. They could assess situations; they took risks and sometimes they could be very bold. But they were buried so deep in their lives that they were not able to stand back and consider the nature of their lives” (p. 16). Salim, , on the other hand, finds himself from an early age “develop[ing] the habit of looking, detaching myself from a familiar scene and trying to consider it as from a distance” (p. 15).Google Scholar What Salim, could do, which his family could not do, he says, is assess himself—his self—while his family “continued to live as we had always done, blindly” (p. 17)Google Scholar. Salim finds that he can and must “stand alone... to be master of my fate.”

38. Jürgen Habermas takes a further step. “No one,” he says, “can construct an identity independently of the identifications that others make of him… identifications not in the propositional attitude of observers, but in the performative attitude of participants in interaction. Indeed the ego [autonomous self] presents itself to itself [knows itself] in the performance of communicative actions... [T]he participants must reciprocally suppose that the distinguishing of oneself from others is recognized by those others. Thus the basis for the assertion of one's own identity is really not self-identification, but intersubjectively recognized self-identification” (Quoted in Mccarthy, T., The Critical Theory of Jürgen Habermas [Cambridge, MA: The MIT Press, 1979], p. 381Google Scholar). The upshot of the quotation is that one establishes a separate self, an independent identity, only through communication with other separate selves. Identity as an independent self can be derived only through dialogue with others whom we recognize and who recognize us as autonomous.

39. Shotter argues that in order to be intelligible even to oneself, thought must be “socially negotiated” (Social Accountability and Selfhood, p. 46). Therefore decisions and acts of any kind are not intelligible even to the actor/chooser until they are validated as rational, though not necessarily acceptable, intersubjectively. This thesis is stronger than mine, and I am not sure I can accept it. It seems to make short shrift of the idea of internalized voices.

40. What about the idea of in ternalizing voices and thus becoming self-sufficient by needing to listen, therefore, only to oneself? A problem with internalized voices is that though they, and the positions they represent, are objects of attention, as would be the positions of flesh-and-blood persons in a conversation, internalized voices are easily manipulated because they arise within us. Therefore they lack the kind of unexpected, idiosyncratic responses of others that would really test our own positions by taking us by surprise.

41. Beiner, , Political Judgment (London: Methuen, 1983), pp. 8–9.Google Scholar

42. I am indebted to Joel Feinberg's discussion of this idea in his article in Christman, The Inner Citadel, especially the section “Autonomy as Capacity.” Gerald Dworkin argues that in defining autonomy we must guard against too intellectualist a position so that “those who are less educated, or who are by nature or upbringing less reflective, are not, or not as fully, autonomous individuals” (Theory and Practice, p. 17). I certainly agree with Dworkin about guarding again too intellectualist a position, but I think his linking of “less educated” and “less reflective” is unfortunate. These two are not reliably related. In the first case, I agree that a less educated person can be as autonomous as anyone else; but in the second, I see a very good reason why a less reflective person may not be, for autonomy depends upon and is constituted in part by self-reflective thought. One must be reflective enough to think about one's own thinking and be self-conscious of that; one must be reflective enough to look rationally and critically at the foundations, meaning, and consequences of whatever social rules or roles are under scrutiny. The threshold is to stand back reflectively from one's social context and take a look from different perspectives at what its rules and roles mean. If by “less reflective” Dworkin meant that some persons do not scrutinize as many aspects of their lives as others, then I agree fully with him. A person does not need to be comprehensive in his critical reflection to be autonomous. Also, it is important to repeat that even persons who are deeply reflective—that is, analytical—are not necessarily autonomous, for reflectivity must include self-reflectivity.

43. Brown, , Rationality (New York: Routledge, 1988)Google Scholar. See also Jon Elster who argues that “rational beliefs are those which are grounded in the available evidence; they are closely linked to the notion of judgment” (Sour Grapes [New York: Cambridge University Press, 1985], p. 2Google Scholar).

44. Ibid., p. 137.

45. Ibid., p. 143ff. See also Nichomachean Ethics, Book Three, Chapter Three and Book Six, Chapter Five.

46. Ethics, 1140b25–30. While Brown argues that phronesis is a narrower concept than his own concept of judgment, it appears that the only difference between these concepts lies in the range of the capacity, not in its nature.

47. Brown, , Rationality, p. 184.Google Scholar

48. Ibid., p. 33.

49. Ibid., p. 187.

50. I have developed this line of thought in more detail in “Epistemology and Selfhood in Charles Taylor's Sources of the Self: What Happens After Epiphany?” Paper delivered at the annual meeting of the American Political Science Association, Washington, D. C., 09 1991.Google Scholar

51. Imagine a Rawlsian grass counter who decides to devote his remaining years to counting grass (A Theory of Justice [Cambridge, MA: Harvard University Press, 1971], p. 432).Google Scholar He can present reasons for his decision and the actions based on the decision: “An inheritance provides materially for my family and me, so although I do not earn a living counting grass, I do not jeopardize anyone's welfare by doing so; I might be listed in the Guinness Book of Records as the first ‘full-time grass counter'; and I am not simply passionate about grass counting, but I also have expertise. I have counted the hairs on my head.” This decision strikes us as so odd that we might think that this person is mentally disturbed. But on the evidence that we have, we must accord him the status of an autonomous person making an autonomous decision.

52. See footnote 16, above.

53. Maclntyre, , Whose Justice, Which Rationality?, p. 396.Google Scholar

54. Taylor, , “The Person,” in The Category of the Person, ed. Carrithers, M. (Cambridge: Cambridge University Press, 1985), p. 276.Google Scholar

55. Taylor, , Philosophical Papers, vol. 2: “Philosophy and the Human Sciences,” (Cambridge: Cambridge University Press, 1985), p. 54.Google Scholar

56. Taylor, , Philosophical Papers, vol. 1: “Human Agency and Language,” (Cambridge: Cambridge University Press, 1985), p 152.Google Scholar

57. Ibid., p. 40.

58. All quotations are from Taylor, , Sources of the Self, p. 491.Google Scholar

59. Note, for example, that Taylor says that the tacit background of objects can be brought to the foreground, but that we do not do so because we rely upon those objects as basic assumptions. To explore the premises of the argument that the structure of language constitutes subjectivity, see Crittenden, Beyond Individualism.

60. Shotter, , Social Accountability and Selfhood, p. 218.Google Scholar

61. Does this make autonomous persons the unencumbered selves excoriated by Michael Sandel and other communitarians? Are they transcendental selves able to stand back reflectively and scrutinize every aspect of their characters from some impartial, if not disembodied, perspective? Such a position is neither required nor possible. Autonomous persons, it seems to me, will stand back reflectively from some aspects of their lives or characters, but not simultaneously from all such aspects. They will scrutinize any aspect using perspectivism—standing in another person's shoes. That requires reflectively adopting a different, even an opposite, viewpoint; it surely does not require holding a disembodied or unencumbered or positionless position. One can transcend his or her community's rules and roles without assuming a transcendental pose.

62. In addition to the example from V. S. Naipaul (See footnote 37, above), this point was dramatized in the experiences of a friend who teaches Islamic Studies. He lived in Cairo, Egypt, for two years and traveled throughout the Middle East. While doing so, he often asked Islamic clerics the following hypothetical: “Let's assume that the Koran is not the word of God dictated to Mohammed.” He was surprised to find that in virtually every case the clerics would not respond to the hypothetical. They had the conceptual apparatus to do so, because they could, and would, respond to the hypothetical, “Let's assume Israel did not exist”; but they were embedded in their Islamic conditions and could not step back from this one. The first hypothetical was dismissed as unintelligible.

63. Alasdair Maclntyre takes a position similar to Taylor's., In After Virtue (Notre Dame, IN: University of Notre Dame Press, 1981)Google Scholar, he says we can transcend the moral limitations of our traditions “through criticism and invention” (p. 222). But he also says, in Whose Justice? Which Rationality?, that such criticism or invention only has purchase if within one's tradition. Adopting the perspective of another tradition is empty, if even possible, since the criteria for rationality and the history of their application will be incommensurable. This leaves us, I think, with only two possible, and equally unattractive, scenarios: Either “tradition relativism” or “tradition entrapment,” which I have already discussed as community entrapment. Maclntyre rejects the tradition relativism since he thinks Augustinian Thomism has clearly shown itself to be the superior tradition. Those unconvinced because they “give their allegiance to rival traditions of enquiry” (p. 403) will simply have to write rival histories. There are no tradition-independent criteria by which to show either one to be in error.

64. See the last chapter of Crittenden, Beyond Individualism, “Socrates in a Pluralistic World,” for a full discussion of this point.

65. Nedelsky, , “Reconceiving Autonomy,” p. 24.Google Scholar

66. For a discussion of how democratic procedures can be structured to create the common good, see Crittenden, Beyond Individualism, chap. 4.