1. Waldron, J., Legislators' Intentions and Unintentional Legislation, in Marmor, A. (ed.), Law and Iinterpretation, Oxford University Press (1995) at 329–356.Google Scholar

2. See Marmor, A., Interpretation and Legal Theory, Oxford University Press (1992), chapter 8.Google Scholar

3. See, Austin, J., The Province of Jurisprudence Determined (First Published 1832) (1954) Weidenfeld and Nicolson, LondonCrossRefGoogle Scholar. I use “formulated” advisedly: The basic intuitions of Austin's personal conception of authority, and the resemblance of law to a command issued by one person to another, has been well known and espoused by many philosophers before him, most famously, perhaps, by Hobbes.

4. See Hart, H. L. A.. The Concept of Law, Oxford University Press (1961), mainly chapters 2–4.Google Scholar

5. See Waldron, , supra note 1, at 329.Google Scholar

6. See Raz, J., The Mortality of Freedom, Oxford University Press (1986), chapters 2–4Google Scholar. See also Raz, J., Authority, Law, and Morality, 68 Monist (1985), at 295–324.CrossRefGoogle Scholar

7. A very clear indication to my view that the expertise thesis may not hold as often as it may be thought, I gave by raising some doubts about the possibility of acknowledging expertise in the realm of moral issues. See Marmor, , supra note 2, at 181.Google Scholar

8. By the term “single author jurisprudence” Waldron does not necessarily mean single legislator. He recognizes that deference to legislative intent may be appropriate, for example, when the expertise of a particular legislator had been generally acknowledged by the legislature as a whole, and the law had passed “on the nod,” so to speak, on the basis of such an individual expertise. I believe that we would both agree, however, that such cases are quite rare.

9. Waldron, , supra note 1, at 331Google Scholar. Emphasis in the original.

10. Id. at 342.

11. Id. at 343.

12. Id. at p. 348, but cf.at 351.

13. See Nozick, R., Anarchy, State and Utopia, Blackwell (1974) at 19.Google Scholar

14. This is a conclusion that I have initially accepted, and I gave an example of how this may come about; see Marmor, , supra note 2, at 163–164.Google Scholar

18. Id. at 354.

19. As a matter of fact, Waldron's remarks here are particularly perplexing. After parsing my explicit disavowal of intentionalism based on democratic and majoritarian principles at the beginning of his article, he reintroduces the majoritarian principle on my behalf at the end: “Surely—[Marmor] will say—it matters whether the view we have in mind corresponds to the intentions of the majority or the minority in the legislature. Surely the majority view is—by reason of the rules that allow it to prevail—the more canonical.” (see p. 354) (emphasis mine). And this alleged reply of mine, Waldron claims, “is based on a deep misapprehension about the… principle of the majority rule.” Id. I certainly agree that such a view is based on a deep misapprehension indeed, but as Waldron himself acknowledges at the beginning of his article, the misapprehension is not mine.

20. Id. at 340.

21. Id. at 341.

22. A similar point is made by M. Moore, but his interest concerns interpretation in general, not so much the concept of authority. See Moore, M., Interpreting InterpretationGoogle Scholar, in Marmor, A. (ed.), Law and Interpretation, supra note 1 at 14–15Google Scholar; see also Hurd, H., Interpreting AuthoritiesGoogle Scholar, Id. at 405.

23. See Raz, , Authority, Law, and Morality, supra note 6, at 303Google Scholar, and Marmor, , supra note 2, at 104–107 and 113–118.Google Scholar

24. Wollheim, R., A Paradox in the Theory of Democracy, in Laslett, P. and Runciman, W. G. (eds.) Philosophy, Politics, and Society, Blackwell, Oxford (1969), at 75Google Scholar. Cited by Waldron, , supra note 1, at 336.Google Scholar

25. Waldron, id. at 337.

26. Id. at 342.

27. Moore, , supra note 22, at 13.Google Scholar

28. Note that Moore's point is different: His examples concern the objects of interpretation in general, not authority. Hence, the following discussion is not meant to be a criticism of his arguments concerning “texthood.”

29. See Hart, H. L. A., Essays on Bentham (1982), Oxford University PressGoogle Scholar, chapter 10, and Raz, , The Morality of Freedom, supra note 6, at 35–37.Google Scholar

30. Note that it is not the particular formulation of thesis [J] that I find wanting, nor do I claim that it is an implausible interpretation of Raz's normal justification thesis. Waldron's mistake lies in the fact that he ignores other, important elements in our conception of authority, some of which I will indicate more clearly in the sequel.

31. As Martin Stone pointed out to me, it is hard to see how the voice machine's output is based on reasons at all, in which case, the Razian dependence thesis is out of place here too.

32. See supra note 6, chapters 2 and 3.

33. Id. at 57–62.

34. For a more detailed account, see my On Convention, forthcoming in Synthese (1996).Google Scholar

35. Note that Raz was careful to formulate his justification thesis in terms of “the normal” justification, thus indicating, I believe, his own view that a distinction may exist between the legitimacy of an authoritative directive and its alleged subjects' reasons for following it On the other hand, his explanation for the need to follow, occasionally, mistaken directives, partly based on his interesting distinction between “clear” and “great mistakes,” is, I believe, not entirely persuasive. (See supra note 6, at 61–62)Google Scholar. A detailed analysis of Raz's views on these issues, however, exceeds the scope of this article. Basically, my point is that Raz has underestimated the importance of auxiliary reasons in this context.

36. Cf. Gans, C., Philosophical Anarchism and Political Disobedience (1992), Cambridge University Press, at 71.CrossRefGoogle Scholar

37. Authority, Law and Morality, supra note 6, and supra note 2, chapter 6.

38. See supra note 4, at 75, and cf. Austin, , supra note 3, at 11.Google Scholar

39. See supra note 4. at 60–61.

40. Id. at 53.

41. Id. at 42.

42. Id. at 46–48.

43. Id. at 60.

44. Perhaps the most dramatic example of this we have is in the authority of the Constitution of the United States: For most American lawyers, the present authority of the Constitution in American law does not derive from the personal authority of its long-dead framers. It derives from the fact that the American people and their lawyers regard the document, and not the framers, as the supreme authority of their law.

45. Note, however, that I do not intend to claim that role-dependence is a necessary prerequisite of content-independence of reasons. It is possible that one's reasons for action would be partly content-independent even in the absence of a structured role, such as friendship. Sarah may simply be in love with Jim, and consequently she may act on the basis of his requests, in a manner which is partly content-independent, This is intelligible even if we are reluctant to define the object of “being in love with…” in terms of a role. I am indebted to Meir Dan-Cohen for bringing this to my attention.

46. See supra note 1, at 349. Note that Waldron's formulation does not apply to the coordination branch of the normal justification thesis.

47. See supra note 4, at 43.

48. For an interesting account of role identity and its significance for legal theory, see Dan-Cohen, M., Lata, Community and Communication, 1989 Duke Law Journal, at 1654–1676.CrossRefGoogle Scholar

49. See supra note 3, at 32.