1. Here and elsewhere, I do not take “consensus” to require strict unanimity, but something close, which I will sometimes alternatively call “convergence.” To account for the ability of a legal system to endure and change, Hart introduced a seminal distinction between primary and secondary rules. Whereas the primary rules prohibit certain acts and facilitate others, the secondary rules determine the legal status of the primary rules and the ways that status may be revised. HLA Hart, The Concept of Law, 2nd ed (Oxford: Oxford University Press, 1994) at 79-123, especially 91-99.

2. Ibid at 94-95, 100-10.

3. Dworkin, Ronald, Law’s Empire (Cambridge: Harvard University Press, 1986)Google Scholar at 3-11, 15-30, 122. No one has done more than Scott Shapiro to reconstruct Dworkin’s idea of theoretical disagreement and the problem it poses for a broadly conventionalist view of law. My ideas about both are as indebted to Shapiro’s work as to Dworkin’s. See Scott J Shapiro, Legality (Cambridge: Harvard University Press, 2011) at 282-306 (reconstructing and critiquing Dworkin’s argument that legal positivism is incompatible with judges’ theoretical disagreements about the law).

4. This can be seen in Shapiro, Legality, supra note 3 at 282-306, Brian Leiter, “Explaining Theoretical Disagreement” (2009) 76:3 U Chicago L Rev 1215, and Barbara Baum Levenbook, “Dworkin’s Theoretical Disagreement Argument” (2015) 10:1 Philosophy Compass 1. Dale Smith takes a different view, however, in “Theoretical Disagreement and the Semantic Sting” (2010) 30:4 Oxford J Legal Stud 635 at 643. He treats theoretical disagreement “as an objection to any theory of law that denies that theoretical disagreement exists.” I follow Raz and Shapiro in understanding legal positivism as defined by the claim that the content of the law does not depend on moral facts (other than, perhaps, indirectly and derivatively). Raz, Joseph, The Authority of Law: Essays on Law and Morality, 2nd ed (Oxford: Oxford University Press, 2009) at 37–38Google Scholar; Shapiro, Legality, supra note 3 at 269-71.

5. To be clear, my aim is not to convince readers definitively that any particular version of the three theses is true—let alone that three particular versions of the three respective theses are true. To do so would require much more space than a single article. My aims are more limited and conditional: to show that (a) non-trivial and plausibly correct specifications of all three theses can be true simultaneously without contradiction, and (b) identifying and holding in mind such specifications leads to a novel and ecumenical view of faithful theoretical disagreement worthy of further examination.

6. One way to partially reconcile Conventionality, Disagreement, and Fidelity is to drop Fidelity—i.e., to conclude that theoretical disagreement must reflect confusion or deception. Brian Leiter has given us the best sympathetic reconstruction of that position. See Leiter, “Explaining Theoretical Disagreement”, supra note 4 (discussing “Error Theory” and “Disingenuity” explanations of theoretical disagreement and arguing that, in any case, an explanation of theoretical disagreement is not especially important to a theory of law because theoretical disagreement is not central to the law). There are at least two reasons to search for a reconciliation of theoretical disagreement and conventionality that does not necessarily undermine judges’ lucidity and/or good faith. First, as Shapiro argues, to treat theoretical disagreement as necessarily a sign of confusion is “uncharitable in the extreme,” and it is one thing to suppose that jurists sometimes act in bad faith, quite another to suppose that they necessarily do so simply in virtue of having theoretical disagreements at all. Shapiro, Legality, supra note 3 at 290-91. As he points out, “no one, except for the legal positivist, thinks [judges] are acting in bad faith merely because they are engaged in theoretical disagreements.” Ibid at 291. Second, even if the guidance that the law gives to judges is sometimes insufficient to yield unique determinate legal answers to legal questions, surely fidelity to the law requires judges to extract as much guidance from the law as possible. So we should search painstakingly for any ways in which theoretical disagreements may be guided by and faithful to the law before concluding that such disagreements necessarily reflect deception or confusion.

7. See Lon L Fuller, “Positivism and Fidelity to Law: A Reply to Professor Hart” (1958) 71:4 Harv L Rev 630; Lon L Fuller, The Morality of Law, revised ed (New Haven: Yale University Press, 1969).

8. The phrase “grounds of law” is Dworkin’s. See Dworkin, Law’s Empire, supra note 3. The phrase “criteria of legal validity” comes from Shapiro, and Hart before him. See Shapiro, Legality, supra note 3; Hart, Concept of Law, supra note 1.

9. Shapiro, Legality, supra note 3 at 283.

10. Ibid at 293 (claiming that in Dworkin’s view, “insofar as the content of the law is dependent on which principles portray legal practice in its morally best light, genuine moral disagreements will induce genuine legal disagreements.”). On the pervasiveness of reasonable ethical disagreement in the modern world, see especially John Rawls, Political Liberalism, expanded ed (New York: Columbia University Press, 2005); Larmore, Charles, The Morals of Modernity (Cambridge: Cambridge University Press, 1996).CrossRefGoogle Scholar

11. See Finnis, John, “Law as Coordination” in Philosophy of Law: Collected Essays, vol. 4 (Oxford: Oxford University Press, 2011) 66.CrossRefGoogle Scholar

12. See Leslie Green, “Positivism and Conventionalism” (1999) 12:1 Can JL & Jur 35; Jules L Coleman, The Practice of Principle: In Defence of a Pragmatist Approach to Legal Theory (Oxford: Oxford University Press, 2001) at 94-100; Scott J Shapiro, “Law, Plans, and Practical Reason” (2002) 8:4 Legal Theory 387 at 388-94, 435-37, 441; Julie Dickson, “Is the Rule of Recognition Really a Conventional Rule?” (2007) 27:3 Oxford J Legal Stud 373; Andrei Marmor, Social Conventions: From Language to Law (Princeton: Princeton University Press, 2009) at 155-75; Shapiro, Legality, supra note 3 at 107-10.

13. On the non-arbitrariness of religious practices, see Shapiro, Legality, supra note 3 at 109 (“Consider, for example, the Ancient Hebrews …. Surely many, if not most, would not have preferred to follow a different text just in case everyone else did.”).

14. These conditions are also “common knowledge” or the object of a hierarchy of reciprocal expectations regarding conformity. David Lewis, Convention: A Philosophical Study (Oxford: Blackwell, 2002, orig 1969) at 42, 58, 76, 78-79.

15. Marmor, supra note 12 at 8. Decades earlier, Dworkin drew on Lewis to make, in substance, the same point, in his criticism of conventionalist perspectives on the law. Dworkin, Law’s Empire, supra note 3 at 144-47.

16. Marmor, supra note 12 at 11.

17. Green, supra note 12 at 40.

18. Ibid at 39-40.

19. Ibid.

20. Finnis, supra note 11 at 70-71; William S Boardman, “Coordination and the Moral Obligation to Obey the Law” (1987) 97:3 Ethics 546; Gerald J Postema, “Coordination and Convention at the Foundations of Law” (1982) 11:1 J Legal Stud 165.

21. Shapiro, Legality, supra note 3 at 107-10; Coleman, Practice of Principle, supra note 12 at 94-95; Christopher Kutz, “The Judicial Community” (2001) 11:1 Philosophical Issues 442 at 454-55.

22. Shapiro, Legality, supra note 3 at 109-10.

23. Green, supra note 12 at 50.

24. See Shapiro, “Law, Plans, and Practical Reason”, supra note 12 (drawing these technical ideas from Michael Bratman). For similar accounts, see Coleman, Practice of Principle, supra note 12 at 96-99; Kutz, supra note 21 at 455-65.

25. Shapiro, “Law, Plans, and Practical Reason”, supra note 12 at 394-441.

26. Coleman, Practice of Principle, supra note 12 at 91. See Margaret Gilbert, “Walking Together: A Paradigmatic Social Phenomenon” (1990) 15:1 Midwest Studies in Philosophy 1.

27. We may disagree fiercely about textualism and purposivism, originalism and non-originalism while still substantially agreeing about most actual legal issues. On the possibility and actuality of agreements about many legal issues notwithstanding more abstract disagreements about law and legal principles, see Cass R Sunstein, “Incompletely Theorized Agreements” (1995) 108:7 Harv L Rev 1733.

28. Green, supra note 12 at 48 (citing Postema, supra note 20 at 188-94).

29. Matthew D Adler, “Constitutional Fidelity, the Rule of Recognition, and the Communitarian Turn in Contemporary Positivism” (2006) 75:3 Fordham L Rev 1671 at 1674 (discussing the versions of legal positivism defended in Shapiro, “Law, Plans, and Practical Reason,” supra note 12; Coleman, Practice of Principle, supra note 12 at 94-100; Kutz, supra note 21).

30. Adler, “Constitutional Fidelity”, supra note 29 at 1674.

31. Hart, Concept of Law, supra note 1 at 91-99; Shapiro, Legality, supra note 3 at 79-83.

32. Dworkin, Law’s Empire, supra note 3 at 114-50.

33. Ibid at 45-113, 139-40, 247-50, 254-58.

34. Dworkin recognized “fit” as a “threshold test” in several instances in Law’s Empire. Ibid at 255, 259, 262, 374.

35. Ibid at 225.

36. Dworkin once wrote, for instance, of “questions of fit surfac[ing] … because an interpretation is pro tanto more satisfactory if it shows less damage to integrity than its rival.” Ibid at 246-47. See also ibid at 257 (arguing that a judge’s “convictions about fit … express his commitment to integrity: he believes that an interpretation that falls below his threshold of fit shows the record of the community in an irredeemably bad light….”). Note finally that Dworkin responds to the hypothetical critic who complains that “integrity is at work in [a Dworkinian judge’s] calculations just up to the point at which he has rejected all interpretations that fail the threshold test of fit” not by disclaiming the idea that concern for fit is rooted in concern for integrity but by trying (and I believe struggling) to show that integrity somehow also bears upon the choice among competing interpretations of the law that satisfy the requirement of fit equally well. See ibid at 262-63.

37. Ibid at 251-52.

38. Ibid at 163.

39. Ibid at 404-07.

40. Ibid at 266.

41. Cf Hart, “Postscript” to Concept of Law, supra note 1 at 266 (arguing that to find the “preinterpretive law” he needs for his process of interpretation, Dworkin “presupposes” “something very like a rule of recognition”).

42. In his explanation of constructive interpretation Dworkin contended that interpretations of the law must “fit” the “preinterpretive” materials of which they are interpretations. Change sufficiently the “preinterpretive” materials given by practice and our interpretations of the law must change accordingly. As Raz notes: “To deny that the criteria which determine the content of the law of this country or that have a content-independent component is to deny that there can be law-making authorities.” Joseph Raz, “Two Views of the Nature of the Theory of Law: A Partial Comparison” in Jules Coleman, ed, Hart’s Postscript: Essays on the Postscript to The Concept of Law (Oxford: Oxford University Press, 2001) 1 at 29-30.

43. Dworkin, Law’s Empire, supra note 3 at 9-11; Ronald Dworkin, “No Right Answer?” (1978) 53:1 NYU L Rev 1; Ronald Dworkin, “The Model of Rules I” in Taking Rights Seriously (Cambridge: Harvard University Press, 1977) 14 at 31-39, 44-45; and Dworkin, “The Model of Rules II” in Taking Rights Seriously 46 at 68-71.

44. Richard H Fallon, Jr, “Precedent-Based Constitutional Adjudication, Acceptance, and the Rule of Recognition” in Matthew D Adler & Kenneth Einar Himma, eds, The Rule of Recognition and the US Constitution (Oxford: Oxford University Press, 2009) 47 at 58.

45. Raz, Joseph, “Why Interpret?” in Between Authority and Interpretation: On the Theory of Law and Practical Reason (Oxford: Oxford University Press, 2009) 223 at 224.CrossRefGoogle Scholar

46. In other words, there might be what Sunstein has called an “incompletely theorized agreement” in law analogous to what Rawls called an “overlapping consensus” in a political liberal state. See Sunstein, supra note 27; Rawls, supra note 10 at 133-72.

47. Philip Bobbitt, Constitutional Fate: Theory of the Constitution (Oxford: Oxford University Press, 1982). Bobbitt also discussed an additional category, “ethical argument,” but I leave it aside here because it is not the subject of the same consensus acceptance as the other types of argument Bobbitt identified.

48. Shapiro, Legality, supra note 3 at 260 (citing Karl N Llewellyn, “Remarks on the Theory of Appellate Decision and the Rules or Canons About How Statutes Are to Be Construed” (1950) 3:3 Vand L Rev 395 at 401).

49. Dworkin, Law’s Empire, supra note 3 at 45-113, 139-40, 247-50, 254-58.

50. Shapiro, Legality, supra note 3 at 383. Shapiro’s positivist response to the problem of theoretical disagreement has the same structure as that of Jules Coleman’s. In response to Dworkin’s use of judicial disagreement to call into question legal positivism, Coleman observed, correctly, that “judges can disagree in some significant set of controversial cases, without in the process abandoning their agreement about what the rule [of recognition] is.” Coleman, Practice of Principle, supra note 12 at 116. As Dworkin argued, however, this “abstraction strategy” signals significant limits to the extent to which convergent practice determines unique answers to legal questions. Ronald Dworkin, “Thirty Years On” (2002) 115:6 Harv L Rev 1655 at 1658-62 (reviewing Coleman, Practice of Principle, supra note 12). Shapiro accepts that disagreements about interpretive methodology often will not be settled even indirectly by convergence upon principles by which to resolve such disagreements. In his view laws are plans and “in systems where meta-interpretive disputes are prevalent, proper interpretive methodologies are not plans, or even as planlike as customary norms. Meta-interpretive disputes arise precisely because no one has settled on how legal texts are to be interpreted.” Shapiro, Legality, supra note 3 at 446.

51. See Shapiro, Legality, supra note 3 at 383 (“There may be no right answer to those disputes, but there are usually wrong ones.”). See generally ibid at 307-87.

52. Ibid at 353-87.

53. Ibid at 340-42.

54. William Baude, “Is Originalism our Law?” (2015) 115:8 Colum L Rev 2349 (arguing for originalism on empirical as opposed to “conceptual” or “normative” grounds and suggesting that doing so is innovative); Stephen E Sachs, “Originalism as a Theory of Legal Change” (2015) 38:3 Harv JL & Pub Pol’y 817 at 819 (“In academic circles, positive defenses [of originalism] are relatively rare; they’re almost unheard of.”).

55. Matthew Adler puts the point modestly when he notes that leading constitutional theorists “[Bruce] Ackerman, [John Hart] Ely, [Richard] Fallon, [Thomas] Grey, [Michael] Perry, and [David] Strauss, all … ground interpretive methods in culture/tradition facts in a manner not too distant from [Dworkin’s idea of] constructive interpretation.” Matthew D Adler, “Social Facts, Constitutional Interpretation, and the Rule of Recognition” in The Rule of Recognition and the US Constitution 193, supra note 44 at 231. Even those who think that political morality is not directly implicated in legal interpretation can and often do argue for particular approaches to interpretation partly on the basis of their accordance with normative ideals and not just empirical facts. Keith Whittington, a constitutional originalist hardly Dworkinian in his view of the content of American law, has expressed a view about how to judge interpretive methods that sounds like something Dworkin might have written: “[t]he justification for adopting any particular interpretive method depends on external reasons of normative political theory. As a consequence, originalism cannot be justified by reference to the intent of the founders or by a purely historical argument.” Keith E Whittington, Constitutional Interpretation: Textual Meaning, Original Intent, and Judicial Review (Lawrence: University Press of Kansas, 1999) at 3.

56. Antonin Scalia, A Matter of Interpretation: Federal Courts and the Law, Amy Gutmann, ed (Princeton: Princeton University Press, 1997) at 40 (“The Constitution … even though a democratically adopted text, we formally treat like the common law.”).

57. In more abstract terms, Smith, supra note 4 at 642, distinguishes four types of disagreements he deems theoretical disagreement about the law: “theoretical disagreement encompasses not only disputes about what sources of law there are, and about what effect a given source has on the content of the law, but also disputes about what determines what sources of law there are and what determines what effect a given source has.” The concrete disagreements I describe here could be characterized as instances of some of the types of disagreement Smith describes, though it is not obvious to me which fall into which categories, because it is not obvious how to distinguish in practice between a source of law and a determinant of a source of law.

58. Shapiro, Legality, supra note 3 at 267-81.

59. Ibid at 271-73 (citing Raz, Authority of Law, supra note 4 at 46).

60. Shapiro, Legality, supra note 3 at 272.

61. See Sunstein, supra note 27.

62. Shapiro, Legality, supra note 3 at 280-81.

63. Ibid at 280.

64. As Jules Coleman has suggested, Raz and Shapiro have each endorsed exclusive rather than inclusive legal positivism because they have thought that (a) the law’s status as law must be able in principle to give us additional reasons for action than those already given to us by ethics or morality and (b) to the extent that the law’s status as law is itself dependent upon ethical or moral considerations, it gives us no additional reasons for action beyond those that ethics and morality already supply. See Coleman, Practice of Principle, supra note 12 at 134-48. Yet as Coleman points out, even if we accept (a), which Shapiro has called the “practical difference thesis,” “[w]hat is or must be true of the law need not be true of a law” and “it is not obvious why each rule must be conceived of as contributing to the guidance function [of law] in the same way.” Ibid at 144, 146. Shapiro discusses and defends the practical difference thesis in Scott J Shapiro, “On Hart’s Way Out” (1998) 4:4 Legal Theory 469; Scott J Shapiro, “Law, Morality, and the Guidance of Conduct” (2000) 6:2 Legal Theory 127. Scott Hershovitz’s recent criticisms of exclusive legal positivism are also intriguing. See Hershovitz, “The Model of Plans and the Prospects for Positivism” (2014) 125:1 Ethics 152. At the least, they should reinforce our sense of the demandingness and non-obviousness of exclusive legal positivists’ claims about laws’ necessary capacity for guidance.

65. In fact, Jules Coleman has put the matter in just that way. See Coleman, “The Architecture of Jurisprudence” (2011) 121:1 Yale LJ 2 at 19 (“For Dworkin, the distinctive virtue of law is not justice, but what he calls ‘integrity.’”).

66. Dworkin rightly contended that predictability is not always required by fairness, often conflicts with other values, and should not be pursued to its limits. Dworkin, Law’s Empire, supra note 3 at 140-44. But none of those considerations furnishes a decisive argument against the idea that predictability in government is morally important, nor against the idea that it is a characteristic virtue of law.

67. Ibid at 220.

68. Ibid at 255-56.

69. Ibid at 163.

70. For important philosophical analyses of the nature and values of the rule of law, or what I call “law’s distinctive virtues,” see especially Fuller, Morality of Law, supra note 7 at 33-94; Raz, Authority of Law, supra note 4 at 210-29; Jeremy Waldron, “Is the Rule of Law an Essentially Contested Concept (In Florida)?” (2002) 21:2 Law & Phil 137; Shapiro, Legality, supra note 3 at 388-400. I borrow the term “virtue” from Raz, though I do not endorse his conception of law’s virtue. Margaret Jane Radin argues that Fuller’s “internal morality of law” fundamentally requires “rules” “capable of being followed.” Radin, “Reconsidering the Rule of Law” (1989) 69:4 BU L Rev 781 at 785 (cited in Waldron, “Is the Rule of Law an Essentially Contested Concept (In Florida)?” at 154). I will assume that law’s distinctive virtues include but are not obviously limited to the virtues of generality and predictability in government. Waldron refers to “generality, clarity, and prospectivity” as “formal or structural,” aspects of the rule of law, which he believes also includes procedural aspects. See Waldron, “The Concept and the Rule of Law” (2008) 43:1 Ga L Rev 1 at 7.

71. Even the exclusive positivist need not object to saying that such virtues may dictate legally determined answers to legal questions, so long as the relevant principles—of say, generality, predictability, and the like—can be correctly understood and applied in particular cases without reference to moral reasoning.

72. It is debatable whether those standards are wholly conceptual or also partly moral. For more detailed discussion, see infra note 81 and accompanying text.

73. Cf Waldron, “The Concept and the Rule of Law”, supra note 70 at 51 (“I think Dworkin is right to observe that those who disagreed in each of [the “hard”] cases [he discussed] disagreed not just about what to do, but about what it meant to abide by the law when deciding what to do.”). See also ibid at 35 (noting that “the pretense” that judges do not create law is enabled by the fact that “[t]he process by which courts make law involves projecting the existing logic of the law into an area of uncertainty or controversy….”).

74. Raz, Authority of Law, supra note 4 at 208.

75. Ibid at 197.

76. Shapiro, Legality, supra note 3 at 248.

77. Waldron, “The Concept and the Rule of Law”, supra note 70 at 53 (citing Ronald Dworkin, “Hart’s Postscript and the Character of Political Philosophy” (2004) 24:1 Oxford J Legal Stud 1 at 24-25) (attributing this view to Dworkin and endorsing it). Similarly suggestive is David Dyzenhaus’s claim that “[t]he seemingly intractable disputes between Dworkin and legal positivists about hard cases, and within legal positivism over the same issue, are best understood in light of the puzzle of legality.” Dyzenhaus, Hard Cases in Wicked Legal Systems: Pathologies of Legality, 2nd ed (Oxford: Oxford University Press, 2010) at 179.

78. Shapiro, Legality, supra note 3 at 273-81, 309-12.

79. Ibid at 278.

80. Raz, Authority of Law, supra note 4 at 226.

81. This question is subtly but importantly different from the question of whether law’s distinctive virtues are necessarily pro tanto moral goods. On the one hand, it is possible that the law’s distinctive virtues are not pro tanto moral goods, but that the reason this is so is that partly moral reasoning about law tells us that law is purely instrumental. Raz, for instance, seems to derive his view of the law in part from the moral premise that “law must be the sort of thing of which the claim to legitimate authority could be true.” (The phrase in quotation marks comes from Jules Coleman, interpreting Raz in Coleman, “The Architecture of Jurisprudence”, supra note 65 at 43.) Yet Raz denies “that there is necessarily at least some moral value in every legal system.” Raz, Authority of Law, supra note 4 at 224. On the other hand, it is possible that law’s distinctive virtues are pro tanto moral goods, but that it is possible to understand them and their implications for legal interpretation without engaging in moral reasoning. We may, for instance, understand what predictability in governance means and implies for legal interpretation, without having to engage in moral reasoning. And this may be true even if it is also true that predictability in governance is a pro tanto moral good. Hence, even if disagreements about the content of the law are sometimes legally resolvable by reference to the ideal of the rule of law and that ideal has pro tanto moral value, this does not imply the disagreements are resolved by morality and thus inconsistent with legal positivism. Here I follow Jeremy Waldron, who has suggested that the law may depend on rule of law principles that happen to lead to the satisfaction of certain moral principles without those rule of law principles themselves being moral principles and making the existence or content of the law dependent upon morality. Waldron, “Positivism and Legality: Hart’s Equivocal Response to Fuller” (2008) 83:4 NYU L Rev 1135 at 1165-66.

Those who gravitate toward thicker and less instrumental conceptions of the rule of law are more likely to suppose we cannot understand the rule of law and its implications for legal interpretation without engaging in substantive reasoning about political morality. Suppose, for instance, we adopt Dworkin’s view that “integrity” is the distinctive virtue of law, or we conclude that ideals such as equality of treatment or due process are part of the ideal of the rule of law. Then we are unlikely to suppose we can understand the nature of the rule of law, or its implications for legal interpretation, without recourse to substantive moral reasoning.

As to the distinct question of whether law’s distinctive virtues are necessarily pro tanto moral goods, the answer is not obvious if we adopt a minimalist conception of them. Waldron, for instance, suggests that “conformity to the principles of legality does tend to mitigate certain aspects of injustice that might otherwise be present,” but “might also have the potential in some cases to aggravate injustice.” Waldron, “Positivism and Legality” at 1162. It might be thought, for instance, that the distinctively legal virtue of generality in governance is a moral good in the sense of necessarily demanding at least some degree of equal treatment of persons, even when it involves treating some people equally badly, and worse than members of some other group. Alternatively, perhaps the distinctively legal virtue of predictability in the use of sanctions is a moral good in the sense of necessarily ensuring for people at least some degree of fair opportunity to avoid punishment. See TM Scanlon, “Punishment and the Rule of Law” in The Difficulty of Tolerance: Essays in Political Philosophy (Cambridge: Cambridge University Press, 2003) 219; HLA Hart, “Legal Responsibility and Excuses” in Punishment and Responsibility: Essays in the Philosophy of Law, 2nd ed (Oxford: Oxford University Press, 2008) 28. On the other hand, those ideas are open to dispute. And it could be that the coordinative powers of generality and predictability in governance make them a particularly effective tool in service of immoral agendas (and perhaps for that reason they cannot as a general matter be considered even pro tanto moral virtues). See Waldron, “Positivism and Legality” at 1162-63.

82. As Waldron demonstrates in “The Concept and the Rule of Law”, supra note 70 at 53, this comes through most clearly in Dworkin, “Hart’s Postscript”, supra note 77 at 24-25.

83. Cf Smith, supra note 4 at 642 (“[A] dispute between exclusive legal positivists and their opponents counts as a theoretical disagreement.”).

84. For relevant discussion of Dworkin’s ideas about methodology in legal philosophy, see Dworkin, “Hart’s Postscript”, supra note 77.

85. Shapiro, Legality, supra note 3 at 391 (“Broken clocks are not diluted, peripheral, borderline clocks…. Broken clocks are real, but defective, clocks.”). Cf Raz, Authority of Law, supra note 4 at 226 (suggesting that the law “… is not of the kind unless it has at least some ability to perform its function” in the way that “[a] knife is not a knife unless it has some ability to cut.”).

86. See the discussion in supra note 64.

87. See Waldron, “The Concept and the Rule of Law”, supra note 70.

88. Ibid at 10-19. Mark John Bennett finds that some leading contemporary positivists have been willing to accept that a legal system must conform at least to a minimal extent to distinctively legal principles like Lon Fuller’s, but have denied that even minimal conformity to the rule of law must be among the standards determining which specific laws exist and what they require and allow. See Bennett, Legal Positivism and the Rule of Law: The Hartian Response to Fuller’s Challenge (unpublished SJD Thesis, University of Toronto, 2013) at 177-219, 262-316. Bennett worries about the consistency of those two stances and expresses sympathy for the idea that fidelity to law should be understood to encompass fidelity to the rule of law. I defend here a modest version of the idea—rightly attributed to Fuller—that rule of law considerations sometimes at least partially determine what judges ought to do to follow the law. Dyzenhaus notes, for instance, that “[Fuller] argues that fidelity to law is served when judges interpret particular laws in accordance with the principles of legality so that these principles inform the judicial understanding and interpretation of the law.” Dyzenhaus, supra note 77 at 21 (citing Fuller, “Positivism and Fidelity to Law”, supra note 7 at 661-69). However, I do not believe there is any obvious incoherence in the position of positivists who believe that a legal system, but not necessarily individual laws, must minimally exhibit distinctive virtues of law. For as we have already seen, “[w]hat is or must be true of the law need not be true of a law.” Coleman, Practice of Principle, supra note 12 at 144.

89. Raz, “Two Views of the Nature of the Theory of Law”, supra note 42 at 35.

90. Ibid.

91. Shapiro, Legality, supra note 3 at 390-92.

92. Fuller, “Positivism and Fidelity to Law”, supra note 7 at 659-60. For a defense of this idea based largely on analysis of real judges’ behaviour during apartheid in South Africa, see Dyzenhaus, supra note 77.

93. Coleman and Hershovitz have both expressed some openness to a similar sounding but importantly distinct claim “that law has no fundamental aim”. Hershovitz, supra note 64 at 166 (citing Coleman, Practice of Principle, supra note 12 at 113).

94. Frederick Schauer, “The Best Laid Plans” (2010) 120:3 Yale LJ 586 at 613-14 (reviewing Shapiro, Legality, supra note 3).

95. Ibid.