I would like to thank the directors of the Centre for Philosophy and Public Affairs at the University of St. Andrews, Scotland, for affording me the time and resources, as a Visiting Fellow, for this project. Also, thanks are owed to the members of my philosophy of law seminar, especially Tom May, for their helpful comments on a much earlier draft of this paper.

1. lthough a highly controversial concept, most authorities would agree on our rough working definition, while going along different paths in elaborating it. See Feinberg, Joel, Social Philosophy (Englewood Cliffs, NJ: Prentice-Hall, 1973) at 15,Google Scholar and his Harm to Self (New York: Oxford University Press, 1986) at 27. See also Christman’s “Introduction” in Christman, John, ed., The Inner Citadel (New York: Oxford University Press, 1989) at 5 Google Scholar and May, Thomas, “The Concept of Autonomy” (1994) 31:2 Am. Phil. Quart. 133–44, especially at 139–42.Google Scholar

2. We refer here to someone exercising authority over another person or agent as compared to exercising authority over an object, i.e., a piece of tangible property. One exercises authority over tangible property by “exercising physical dominion” over it. See Brown, Roy Andrews, The Law of Personal Property, 2d ed. (Chicago: Calaghan and Co., 1955) at 13.Google Scholar This would include determining it disposition, use and condition. Contrast this with exercising authority over a rational agent by giving her reasons including second order exclusionary reasons for actions or refrainings. These two notions of authority are so different as to constitute distinct meanings of the term. But then the full meaning of this latter use and its presuppositions is the subject of this paper.

3. This “paradox” is dramatized by Robert Paul Wolff, who argues that a truly autonomous agent must be totally free of the encumbrance of any (practical) authority. This means, of course, that no autonomous person can recognize the political authority of another. So, the obvious conclusion, welcomed not resisted by Wolff, is that an autonomous person must be an anarchist, recognizing no claims or obligations of citizenship (except in an unanimous, direct democracy). See Wolff’s, In Defense of Anarchism (New York: Harper & Row, 1970) at 12–27.Google Scholar

Raz, Joseph identifies a similar paradox as one of the “paradoxes of authority”, the upshot of which is that “Since all practical questions may involve moral considerations, all practical authority denies moral autonomy and is consequently immoral”. The Authority of Law (Oxford: Oxford University Press, 1979) at 3.Google Scholar

4. Harm to Self, supra note 1 at ch. 1

5. I refer to Raz’s work as it is formulated, The Authority of Law, chs. 1 and 2. For a slightly different version see Raz, Joseph, The Morality of Freedom (Oxford: Clarendon Press, 1986)Google Scholar at 29 and 52 and passim.

6. Ibid, at 28 and 47–51.

7. It may sound like an oxymoron to call supreme authority context dependent, but the alternative is worse. It is to plow head first into the mysteries of absolute sovereignty, that fully independent, unlimited, unitary and absolutely final authority sought after by Bodin and Austin. For, if all “supreme” authority is not context dependent, then there must exist some genuinely supreme, ultimate final authority. But, this notion has foundered on constitutionalism, federalism, popular sovereignty, and the reality of various other restrictions on authority of both a normative and a de facto kind.

8. By a constitutive action I mean apart of an action that is itself an action. A constitutive action is to be distinguished from the same act individuated differently or the same act under several descriptions. This view of actions and their parts is consonant with the view taken by Davis, Lawrence H. in “Individuation of Actions” (1970) LXVII J. of Phil. 520–30CrossRefGoogle Scholar and Thomson, Judith J., “The Time of a Killing” (1971) LXVII J. of Phil. 115–32.CrossRefGoogle ScholarPubMed

9. I use “action-plan” in the sense of Goldman, Alvin in A Theory of Human Action (Englewood Cliffs, NJ: Prentice-Hall, 1970) at 56–63.Google Scholar This means a series of beliefs and other acts that issue in a given act (at 56). This does not mean that the agent necessarily deliberates or plans the projected acts (at 57).

10. I owe the example to Richard Bronaugh.

11. The notion of ability is a deep and deeply controverted one among action theorists. I have endeavored to side step controversy here by providing an admittedly somewhat superficial treatment of this notion. Since the concept does not lie directly in our path, this analysis should suffice. For more, see Thalberg, Irving, Enigmas of Agency (New York: Humanities Press, 1972) at 115–42Google Scholar and Davis, Lawrence, Theory of Action (Englewood Cliffs, NJ: Prentice-Hall, 1979) ch. 3.Google Scholar

12. See my “Davidson, Donald and Section 2.01 of the Model Penal Code” 11 Crim. Justice Ethics 31, (Winter/Spring, 1992)Google Scholar for this analysis. For a treatment without benefit of action theory that arrives at essentially the same conclusion see Williams, Glanville, Criminal Law: The General Part, 2d ed. (London: Stevens and Son, 1961) at 16–27.Google Scholar

13. Ibid, and Model Penal Code, Sec. 2.01 and Sec. 2.0

14. For a well worked out analysis of “minimal competence”, in this sense, see Haworth’s, Laurance Autonomy (New Haven: Yale University Press, 1986) chs. 1 and 2.Google Scholar

15. But see Delgado, Jose MR., Physical Control of the Mind (New York: Harper, 1969)Google Scholar especially Part III. In 1969, Delgado was doing some hair raising things with the behavior of monkeys by using electrical stimulation. It is too frightening to contemplate what Delgado or his cohorts are up to these days.

16. Working through the agency of an agent, of course, may still be causal. But, however we analyze agency, action and human behavior, we must admit that the causal story is more complicated when a person is ordered not to do X (and he refrains) than when he is knocked unconscious so that he cannot do X.

17. Such a rational reconstruction need not be easy nor without profound philosophical problems of its own.

18. See Haworth, supra note 14, and Daniel, Wikler, “Paternalism and the Mildly Retarded” in Rolf Sartorius, Paternalism (Minneapolis: University of Minnesota Press, 1983) at 83–94.Google Scholar

19. Harm to Self, supra note 1 at 28–3.

20. This is not to confuse or conflate authority, a normative notion, with a capacity to visit unpleasant consequences, a matter of fact. They are not the same thing at all as Hart, H.L.A. has shown in his critique of John Austin. See The Concept of Law (Oxford: Clarendon Press, 1961) ch. 2.Google Scholar However, often, perhaps even typically, practical authority is attended by coercive power.

21. “Discourses, ”, Bk.I, Ch. 1 in Whitney Oates, ed., The Stoic and Epicurean Philosophers (New York: Modern Library, 1957) at 225.Google Scholar

22. Hayek, F.A., The Constitution of Liberty (Chicago: The University of Chicago Press, 1960) at 133.Google Scholar

23. Of course, acting on reasons might still ultimately be reduced to natural causes. See note 16.

24. Hayek, , supra note 22 at 150.Google Scholar

25. The Province of Jurisprudence Determined (1832), Lecture I.

26. Note the similarity to the way the sovereign authority enforces the law. It sets out, through promulgation and publication, the rules. Then it assumes that the subject knows them, or can acquire knowledge of them, apply them in the appropriate context to herself and that she will be disciplined only if she fails to conform her behavior to the rule. Fuller, Lon clearly points this out in The Morality of Law, 2d ed. (New Haven: Yale University Press, 1969) at 162–63.Google Scholar We might say that the subject is the sovereign’s deputy in seeing to her own conformity to the law. Likewise, the recruit is his sergeant’s deputy in monitoring his own actions.

This second order exercise of judgement and volition reminds one of Harry Frankfort’s notion of a person as someone capable of second order volitions, i.e., effective acts of willing to change her preferences, habits or features of character. See his “Freedom of the Will and the Concept of the Person” (1971) LXVIII1 J. of Phil, at 5-20. This is not to say that our kind of autonomy requires an exact version of Frankfort’s person. But it does, in this notion of a second order view of self, tell us something very important about autonomous agents operating under rules and commands where they must stand back and assess their own performances.

27. Rescher, Nicholas, The Logic of Commands (London: Routledge and Kegan Paul, 1966) at 23.Google Scholar

25. Ibid., at 18–21.

29. Note how this resembles the fox hole digging case but how much more judgement I must presume on the part of the jeweller and, thus, how much greater the range of discretion.

30. This analysis of authority owes a great deal to Raz’s, Joseph notion of exclusionary reasons and to his account generally in The Authority of Law (Oxford: Clarendon Press, 1979) at 17–33.Google Scholar

31. See Searle, John R., Speech Acts (Cambridge: Cambridge University Press, 1969) at 50–54.CrossRefGoogle Scholar See also Hamblin, C.L., Imperatives (Oxford: Blackwell, 1987) at 5–6.Google Scholar