I would like to thank Wil Waluchow, Bryan Peeler, the participants of the 2004 CS-IVR meeting in Winnipeg, an anonymous CJLJ referee, and Richard Bronaugh for generous and insightful comments on versions of this paper.

1. Waldron, Jeremy, Law and Disagreement (Oxford: Oxford University Press, 1999) at 102 CrossRefGoogle Scholar [LD].

2. Ibid. at 10.

3. Ibid. at 99, citing Raz, Joseph, “Authority, Law and Morality” (1985) 68 The Monist 295 CrossRefGoogle Scholar at 303.

4. Raz, Joseph, The Morality of Freedom [MF] (Oxford: Clarendon Press, 1986) at 53.Google Scholar Note that I say that on Raz’s account, this is typically the way in which authority is justified. As I will discuss, Raz’s account is intended to explain the normal and ordinary way in which authority ought to be justified, not the only way.

5. LD, supra note 1 at 7.

6. I say this despite the fact that in some places it seems as though Waldron agrees with Raz’s account of authority, and some have read him this way. For example, Andrei Marmor makes the following statement: “Drawing on Joseph Raz’s analysis of authority, (with which Waldron basically agrees)….” Positive Law and Objective Values (Oxford: Clarendon Press, 2001) at 90. My view is that there are significant differences between Waldron’s approach and the Razian approach.

7. For an interesting critique of Raz’s account and its alleged indifference to procedure, see Scott|Hershovitz, “Legitimate Authority, Democracy, and Razian Authority” (2003) 9 Legal Theory at 201-20.

8. MF, supra note 4 at 53 [emphasis removed].

9. Ibid. at 47 [emphasis removed].

10. Ibid. at 46 [emphasis removed].

11. This point has been the subject of considerable controversy, given its place in Raz’s Authority Argument for Exclusive Legal Positivism—the view that evaluative considerations can never serve as a criterion for legal validity. Raz argues that since the law necessarily claims authority, we must be able to determine the law’s content without reference to the reasons that justify the law. On this argument, if we need to deliberate on the basis of dependent reasons to determine the law’s content, that defeats the purpose of the law, since legal authorities intend to supply reasons for action that replace disputed dependent reasons. For an Inclusive Legal Positivist response to the Authority Argument, see Waluchow, Wilfrid J., Inclusive Legal Positivism (Oxford: Clarendon, 1994) at 129–40 Google Scholar and “Authority and the Practical Difference Thesis” in (2000) 6 Legal Theory 45.CrossRefGoogle Scholar Waluchow argues that authoritative directives can constitute content-independent reasons for action without pre-empting all of the dependent reasons. On Waluchow’s argument, an authority can sometimes lay out a framework for settling a matter—a framework that requires parties to deliberate on the basis of a specified set of dependent reasons. In another response to the Authority Argument, Stephen Perry argues that authoritative directives can constitute weighty reasons for action without being fully exclusionary. See Perry, Stephen R., “Judicial Obligation, Precedent, and the Common Law” (1987) 7 Oxford J. Legal Stud. 215.CrossRefGoogle Scholar

12. One could counter Waldron by arguing that the fact that the judiciary is insulated from the political process (e.g. desire for re-election) is a reason to entrust them with the final say on matters of justice and rights, since they will be less likely to compromise on matters of principle due to political Pressures. For an example of this line of argument, see Dworkin, Ronald, Freedom’s Law: The Moral Reading of the American Constitution (Cambridge, MA: Harvard University Press, 1996)Google Scholar. Waluchow also advances this line of argument in “Constitutions as Living Trees: An Idiot Defends” (2005) XVIII Can. J. Law & Juris. 207.

13. Waldron is quite clear about the fact that the decision-procedures that we adopt should be adopted exclusively for reasons of procedural fairness, given that we disagree about the substantive merits of legislative outcomes: “It is because we disagree about what counts as a substantively respectful outcome that we need a decision-procedure; in this context, folding substance back into procedure will necessarily privilege one controversial view about what respect entails and accordingly fail to respect the others.” LD, supra note 1 at 116.

14. For example, in general, one will likely do better by obeying the posted speed limit than by trying to figure out for oneself how fast to drive. This kind of case fits the schema of the normal justification thesis quite nicely. But it is less clear how laws that deal with controversial social issues fit this schema.

15. LD, supra note 1 at 103.

16. Ibid. at 105.

17. Ibid.

18. Ibid. at 161.

19. Ibid. at 162 [emphasis added].

20. Ibid. at 159 [original emphasis].

21. For example, it may be fundamental to a libertarian’s conception of justice that it is wrong for the government, for the purpose of redistribution, to take wealth to which a subject is entitled. A libertarian may nevertheless be able to live with a tax that is intended to fund a universal health care system. If, on the other hand, the government began adopting policies that involved overt racial discrimination, most citizens would likely not put up with these policies. My point presently is that in some cases compromises and accommodation can be reasonable, even when justice is at issue. The significance of this point will become clear in what follows.

22. MF, supra note 4 at 342.

23. MF, supra note 4 at 395.

24. Raz, Joseph, Engaging Reason: On the Theory of Value and Action (Oxford: Oxford University Press, 1999) at 65.Google Scholar

25. “Incorporation by Law” (2004) 10 Legal Theory at 5.

26. This is not to say that an individual cannot also have competing wants—our wants are often far from transparent. But a different problem is faced in communal choice, since it may involve a number of individuals who each have clear and often intense wants (they know what they want, and they want it very badly) that are pitted against the wants of others.

27. It is worth noting that wants are not necessarily brute preferences. We can have reasons for wanting something. For example, a person with young children has a reason to want investment in education, while a person in poor health has a reason to want investment in health services. For Raz’s analysis of the relation between wants and reasons, see MF, supra note 4 at 140-45 and Engaging Reason, supra note 24 at 56-64.

28. “Majority-decision does not require anyone’s view to be played down or hushed up because of the fancied importance of consensus.” LD, supra note 1 at 111.

29. Ibid. at 113.

30. Ibid. at 114 [emphasis added].

31. See supra note 43.

32. Raz makes a related (though not identical) point when he argues that there can be a range of morally permissible constitutions because constitutional principles are underdetermined by morality. He argues that, if a constitution is morally acceptable, the constitutional practices and traditions of a state are self-legitimating. See Raz, Joseph, “On the Authority and Interpretation of Constitutions: Some Preliminaries” in Alexander, L., ed., Constitutionalism: Philosophical Foundations (Cambridge: Cambridge University Press, 1998) at 152.Google Scholar The point that I am arguing for presently—that political decision-making should employ accepted and agreed upon fair procedures for choosing from among a range of morally acceptable options—is meant to capture a similar spirit.

33. MF, supra note 4 at 79.

34. Thomas Aquinas, Summa Theologica, First of Second Part, Q. 95, Art. 2.

35. Also, in cases where options are incommensurable, some parties may (mistakenly) believe that their option is clearly the best option.

36. “Different people in society hold differing and opposed theories of justice; yet social decisions are reached and institutions and frameworks established which then purport to command loyalty even in the face of those disagreements, even when such loyalty seems like moral betrayal to those who hold a contrary view.” LD, supra note 1 at 2 [emphasis added].

37. In Canada, the first federal franchise in 1885 was based on property ownership and excluded women. In 1917, the Wartime Elections Act and the Military Voters Act extended the right to vote to all British subjects, male or female, who were active or retired members of the armed forces, and some 2,000 military nurses, the “Bluebirds.”

38. An authority can issue the occasional ill-advised directive. For example, if one percent of an arbitrator’s rulings are unfair, then the arbitrator can likely be considered competent, and his rulings can be accepted as authoritative. At the same time, there may be thresholds for injustice: if a decision passes a certain threshold of injustice, then it is not authoritative, notwithstanding the merits of the authority’s other decisions. I will leave open the question of how to determine the thresholds.

39. Waldron wants to make a connection between the respect that the law deserves because of its communitarian status, and the law’s authority. It is in that context that Waldron states: “Once voted on in the legislature, it is entitled to whatever respect that communitarian status confers on it, without regard to—indeed bracketing away from—the substantive merits of its content.” LD, supra note 1 at 101.

40. What Raz says about consent is instructive here: “Consent to political authority, where given, is often free. It does not follow that it is binding. It is binding only if there are good reasons to enable people to subject themselves to political authorities by their consent.” MF, supra note 4 at 89. By analogical extension, one could say that agreement is binding only if there are good reasons for people to agree.

41. See MF, supra note 4 at 76.

42. It is also important to note that the law’s authority is only one reason that people may have to obey the law. “People who are not subject to the authority of law may then obey the rules for prudential reasons, or because even though they have no duty to obey, disobedience will do more harm than good.” MF, supra note 4 at 103.

43. As Raz explains: “They [political authorities] have the power to determine that a given situation justifies or requires coordination when the general doctrine of authority determines that they do. For example, when the normal justification thesis vindicates their possession of the power, namely, when people or institutions would better conform to reason if that power were exercised by an authority than otherwise.” Joseph Raz, “Comments and Responses” in Lukas H. Meyer, Stanley L. Paulson & Thomas W. Pogge, eds., Rights, Culture and the Law: Themes from the Legal and Political Philosophy of Joseph Raz at 259.

44. LD, supra note 1 at 295.

45. Aileen Kavanaugh argues that Waldron’s arguments would also impugn participatory majori-tarianism since there is disagreement about whether it represents the best model of political decision-making. Kavanagh, Aileen, “Participation and Judicial Review: A Reply to Waldron” (2003) 22 Law & Phil. 451 Google Scholar. Similarly, Waluchow has argued that there is considerable disagreement surrounding the claim that we should adopt procedures that fulfill a condition of equal respect by giving each person an equal say, and that we should do this by adopting and accepting as authoritative the results of participatory majoritarian procedures. Waluchow further points out that Waldron could attempt to persuade us to adopt his model by offering reasons, but if disagreement truly goes all the way down, then he has undercut any basis he might have for persuading us. Waluchow, “Constitutions as Living Trees: An Idiot Defends” supra note 12. Similar points have been made by Christiano, Thomas, “Waldron on Law and Disagreement” (2000) 19 Law & Phil. 513 Google Scholar, and Estlund, David “Jeremy Waldron on ‘Law and Disagreement’” (2000) 99:1 Phil. Stud. 111 CrossRefGoogle Scholar.

46. In Canada, for instance, the adoption of The Charter of Rights and Freedoms introduced into the Canadian legal system reasons that served to limit the range eligible legislative options. Raz appeals to the idea that constitutions limit the range of available legislative options in response to Waldron’s question about why the U.S. Congress is not under the authority of the Conference of Catholic Bishops when the Bishop’s recommendations for welfare reform are regularly better an fairer than the measures adopted by Congress: “its constitution precludes it, or so I assume, from acknowledging any such authority…. The important point is that when we deal with institutions, what they can or cannot do is determined, to a large degree, by their constitution, formal or informal.” “Comments and Responses” in Rights, Culture and the Law: Themes from the Legal and Political Philosophy of Joseph Raz, supra note 43.

47. Joseph Raz “Authority, Law and Morality” supra note 3 at 303, cited by Waldron, supra note 1 at 130.

48. Ibid.

49. Ibid. at 132.

50. It is, of course, unlike a legal directive in that it is not a practical authority. In this case, we are more likely to form a true belief about our credit card balance if we treat the automated response as authoritative.

51. LD, supra note 1 at 10 [original emphasis].

52. Ibid. As I am characterising Waldron’s concern with the Razian account, the problem would be captured by “marginal” rather than “problematic”.

53. Raz, Joseph, The Authority of Law: Essays on Law and Morality (Oxford: Clarendon Press, 1979) at 105 CrossRefGoogle Scholar, quoted in Waldron, supra note 1 at 34-35.

54. See Waldron, supra note 1 at 36-38 and The Dignity of Legislation (Cambridge: Cambridge University Press, 1999) at 92-123. In Law and Disagreement, Waldron himself suggests the Aristotelian synthesis model as one way of conceiving of how a legislature can fulfil the normal justification thesis. Of the models he suggests, this seems to be the one Waldron favours.

55. It is worth noting that I am not suggesting that the two suggestions articulated here are mutually exclusive. Rather, they can be (and on my view should be) thought of as complementary.

56. “Comments and Responses,” supra note 46 at 262.

57. “Authority for Officials” in Rights, Culture and the Law: Themes from the Legal and Political Philosophy of Joseph Raz, supra 43 at 68 [original emphasis].

58. Ibid. at 68-69.

59. It is worth noting that Raz does not seem to fully understand the problem Waldron is raising: “Waldron seems to think that my account is not suitable to explain when one institution is under the legitimate authority of another. But his reasons are not clear to me.” “Comments and Responses,” Rights, Culture and the Law: Themes from the Legal and Political Philosophy of Joseph Raz, supra note 43 at 261.

60. Ibid. at 262.

61. Ibid. at 69.

62. Watkins v. Olafson, [1989] 2 S.C.R. 750 at para 14.

63. It is worth noting that in his response to Waldron, Raz discusses the principle of subsidiarity that governs relations between institutions in the European Community to argue that the cases where lower authorities defer to higher authorities can be explained neatly in terms of the normal justification thesis. See “Comments and Responses,” supra note 43 at 261-62. My claim that a division of labour among institutions can be justified in Razian terms does not depend on the idea that there must be a hierarchy of institutions, and that one institution has authority over another when it occupies a higher position in the hierarchy. My argument is intended to accommodate the kind of “dialogue” approach to understanding the relationship between Parliament and the courts described by Hogg, Peter &Bushell, Allison in “The Charter Dialogue between Courts and Legislatures” (1997) 35 Osgoode Hall L. J. 75 Google Scholar. For an interesting recent discussion of this approach, see Huscroft, Grant & Brodie, Ian, eds., Constitutionalism in the Charter Era (LexisNexis, 2004)Google Scholar.