1. H.L.A. Hart, The Concept of Law (3d ed. 2012), at 214.

2. See, e.g., Ian Brownlie, The Rule of Law in International Affairs: International Law at the Fiftieth Anniversary of the United Nations (1998), at 3–6; Beckett, Jason A., The Hartian Tradition in International Law , 1 J. Juris. 51–83 (2008) at 56 Google Scholar; Capps, Patrick, International Legal Positivism and Modern Natural Law , in International Legal Positivism in a Post-Modern World (Kammerhofer, J. & d'Aspremont, J. eds., 2014) at 215 Google Scholar; and D'Amato, Anthony A., The Neo-Positivist Concept of International Law , 59 Am. J. Int'l L. 321–324 (1965) at 322–323 Google Scholar, though D'Amato later adopted facets of Hart's analysis of law. Sociolegal jurisprudents sometimes respond this way as well; see, e.g., Cotterrell, Roger, What Is Transnational Law? 37 Law & Soc. Inquiry 500–524 (2012) at 507 CrossRefGoogle Scholar; and Brian Tamanaha, What Is Law? Washington University in St. Louis Legal Studies Research Paper No. 15-01-01 (2015), available at SSRN, https://ssrn.com/abstract=2546370, at 54.

3. Hart, supra note 1, at 214.

4. Waldron, Jeremy, International Law: “A Relatively Small and Unimportant” Part of Jurisprudence? in Reading HLA Hart's The Concept of Law (Duarte d'Alemida, L., Edwards, J. & Dolcetti, A. eds., 2013), at 213 Google Scholar and 211. See also Waldron's description of Hart's reflections on international law as “unhelpful,” “careless,” and “indifferent” in Waldron, Jeremy, Hart and the Principles of Legality , in The Legacy of H.L.A. Hart 67–84 (Kramer, M.H. et al. eds., 2008), at 68–69 Google Scholar.

5. Payandeh, Mehrdad, The Concept of International Law in the Jurisprudence of H.L.A. Hart , 21 Eur. J. Int'l L. 967, 978 (2011).CrossRefGoogle Scholar

6. Jean d'Aspremont, Formalism and the Sources of International Law: A Theory of the Ascertainment of Legal Rules (2011), at 56. For another example of an international legal theorist who put Hart's analysis of law to work despite expressing deep dissatisfaction with Hart's discussion of international law, see G.J.H. van Hoof, Rethinking the Sources of International Law (1983).

7. Waldron, Hart, supra note 4, at 69.

8. My use of the term “governance tasks” is inspired by Kornhauser, Lewis A., Governance Structures, Legal Systems, and The Concept of Law, 79 Chi.-Kent L. Rev. 355 Google Scholar. Like Kornhauser, my reading of The Concept of Law pushes it in a social scientific direction, albeit without reducing the study of the question “What is law?” to a mere description of common features characteristic of putative legal orders that Hart seems to have thought exhaustively characterized the sociology of law.

9. Few believe that either the idea of social rules or of specialization in the performance of governance tasks or their combination suffice to distinguish law from other kinds of social orders. Beyond one or two passing remarks, I leave the question of what does so aside in this paper and use the terms “law” and (social) “rule” interchangeably.

10. No doubt many international lawyers will immediately object and point to international courts, dispute resolution panels, semiautonomous administrative rule-making bodies created by treaties, etc. There is some truth in these observations, although, as Hart emphasizes in ch. 10, we should be careful not to exaggerate it. In any case, conceding the development that international law has undergone since Hart wrote The Concept of Law does not undermine the superiority of the interpretation of his argument that I offer to that offered by his critics. Nor, more importantly, does it detract from the theoretical and moral advantages we stand to gain if we take Hart's remarks on international law to heart.

11. Hart, supra note 1, at 91.

12. Id. at 98.

13. Id. at 91.

14. Hart then immediately lays the ground for further possible confusion by briefly noting the content of the rules that must regulate even a primitive society, such as “some form [of] restrictions on the free use of violence, theft, and deception” (id. at 91). Given this characterization of a primitive community, a society will qualify as advanced if the content of its norms regulate and make possible many forms of conduct beyond those that make up what Hart later refers to as the minimum content of natural law. One perhaps too charitable interpretation of Hart's repeated insistence in ch. 10 that municipal and international law are characterized by many analogies of content would be to read him as insisting that in this vitally important respect international law resembles more closely an advanced legal system like that of a well-functioning modern state than it does the social order of a primitive society.

15. Id. at 92–93.

16. Id. at 93.

17. Id. at 93.

18. See Postema, Gerald, Custom, Normative Practice, and the Law , 62 Duke L.J. 707–738 (2012)Google Scholar; Lefkowitz, David, Sources in Legal Positivist Theories: Law as Necessarily Posited and the Challenge of Customary Law Formation , in The Oxford Handbook of the Sources of International Law 323–341 (d'Aspremont, Jean & Besson, Samantha eds., 2017)Google Scholar.

19. Hart, supra note 1, at 93 (emphasis added).

20. Id. (emphasis added).

21. Id. at 98.

22. Id. at 92.

23. Id. at 95.

24. Id. (emphasis added).

25. Id. (emphasis added).

26. Id.

27. The rule can be deployed by participants in the practice to criticize others or to defend themselves against others’ criticisms. Properly understood, however, such arguments can only insist that the other has failed to comprehend accurately some element of the practice. This is so because it is the actual practice of holding accountable, not any description of it, that makes it the case that agents enjoy particular rights, duties, powers, and immunities, the scope of which are more or (sometimes a lot) less determinate. See Postema, supra note 18, at 728; Lefkowitz, supra note 18, at 333–334.

Of course, insofar as those whose practice of holding accountable constitutes the existence of a specific rule-guided social order use (purportedly) authoritative epistemic rules of recognition to understand the practice in which they engage, such rules can influence the form the practice takes. Written texts in particular, such as a constitutional document or hornbooks used in legal education, can channel and in that way constrain a practice's development. But this is fully consistent with the claim that it is the practice of holding accountable itself that makes it the case that the rules that regulate a given society are what they are.

28. In many societies, the practice of holding accountable that constitutes the existence of law encompasses subjects or citizens as well as officials, but as Hart points out, that is not strictly necessary for the existence of a legal system; i.e., for the actual social world to approximate the one that participants in the practice aim to create through their practice of holding accountable.

29. Hart, supra note 1, at 116.

30. Id. at 117.

31. Or perhaps it is only the modifier “of obligation” that Hart so often attaches to “primary rules” that we should abandon, in which case the primary/secondary distinction becomes synonymous with the absence or presence of a division of labor in governance.

32. Waldron, International, supra note 4, at 216 (emphasis in the original). Payandeh, supra note 5, adopts a similar strategy, though he sometimes runs together claims regarding the development of secondary rules for addressing uncertainty, rigidity, and inefficiency with the development of specialization in the performance of governance tasks. See also Capps, supra note 2, at 213; Liam Murphy, What Makes Law (2014), at 147–150.

33. Waldron, International, supra note 4, at 215 and 217.

34. Id. at 215–216.

35. Id. at 217.

36. Id. at 214–215.

37. Payandeh, supra note 5, at 981. In what respect is such an analysis “more convincing”? I address this question and rebut Payandeh's assertion in the following section.

38. Hart, supra note 1, at 214 (emphasis added).

39. Id. at 92–93.

40. Waldron is right, then, to observe the parallel between actors entering into contracts within a municipal legal order and states entering into treaties within the international one and to note that the intelligibility of both types of conduct requires postulating rules of change; Waldron, International, supra note 4, at 217. But he errs in concluding that this observation undermines Hart's characterization of international law as primitive, because Hart's claim rests not on the absence of any secondary rules at all but on the absence of a division of labor in the making of rules that apply generally to members of society, i.e., independent of each (particular) member willingly submitting to those rules.

41. These appear primarily in the last section of ch. 10, where once again Hart indicates that his concern is with the existence of specialization in the performance of governance tasks, albeit in terms that invoke the now familiar conflation of the functional and specialization grounds for distinguishing primitive from advanced societies. The first two sentences of this section read as follows: “To the innocent eye, the formal structure of international law, lacking a legislature, courts with compulsory jurisdiction and officially organized sanctions, appears very different from that of municipal law. It resembles, as we have said, in form though not at all in content, a simple regime of primary or customary law”; Hart, supra note 1, at 232.

42. Id. at 97 and 93.

43. Id. at 93. Hart and Kelsen agree on this point, with the latter stating that in the international legal order “the technique of self-help, characteristic of primitive law, prevails.” See Hans Kelsen, Pure Theory of Law (1967), at 323.

44. Id. at 93 (emphasis added). Payandeh, supra note 5, at 988.

45. Hart, supra note 1, at 218–220.

46. For an excellent contemporary exposition of this approach to analyzing international law and how it contributes to the production of social order, one at which Hart only gestured, see Joel Trachtman, The Economic Structure of International Law (2008).

47. Hart, supra note 1, at 96.

48. Id. at 93.

49. See Payandeh, supra note 5, at 985–986. See also Terry Nardin, Law, Morality, and the Relations of States (1983), at 156–166.

50. Waldron shifts over the course of his discussion from describing Hart's view of international law as characterizing “international law as, in many respects, more like a system of ‘primitive’ law than like a municipal legal system”—Waldron, International, supra note 4, at 209 (emphasis added)—to describing Hart as “claim[ing] that the international order is a primitive legal system”; id. at 216 (emphasis added).

51. Hart, supra note 1, at 214 (emphasis added).

52. Id. at 236.

53. The rule is an epistemic authority, recall, because it is the actual practice of holding accountable, not a description of that practice, that determines what the law of a given community is.

54. Waldron, International, supra note 4, at 219; Payandeh, supra note 5, at 989–990.

55. Payandeh, supra note 5, at 992.

56. Id. at 991; Waldron, International, supra note 4, at 220–221. Payandeh, id. at 985–986, in particular makes much of the fact that the constitutional law of a domestic state is characterized by a degree of uncertainty similar to that present in the case of international law and that at least in some such states, there exists no specialized body charged with adjudicating such disputes. Had Hart only acknowledged that fact, Payandeh maintains, he would not have offered as “antagonistic” a divide between international and municipal law, as is suggested by the description of the former as primitive and the latter as advanced. The argument fails to persuade, however, because in a well-functioning municipal legal order, constitutional disputes typically leave much existing law settled. Whatever uncertainty and inefficiency attach to those disputes have far less impact on municipal law's ability to contribute to the production of social order than is true in the case of international law, where a far wider range of norms is subject to these defects. (Note: this claim concerns law's contribution to producing social order, not the degree to which a stable social order obtains.) If we keep in mind that the simplicity or complexity of a legal system is a matter of degree and recall that Hart was well aware of the fact that municipal legal systems sometimes suffer from constitutional crises that inhibit their value as a means for social control (Hart, supra note 1, at 122–123), then the comparison between municipal constitutional law and international law does little to weaken Hart's case for characterizing the latter as akin to a primitive social order

57. Other theorists who criticize Hart's assertion that international law lacks a rule of recognition because they mistakenly believe that what Hart has in mind when he makes this claim is an epistemically authoritative rule for identifying international law include van Hoof, supra note 6, at 55–56; Capps, supra note 2, at 214; and Kingsbury, Benedict, The Concept of Law in Global Administrative Law , 20 Eur. J. Int'l L. 23 (2009), at 28 Google Scholar.

58. Hart, supra note 1, at 117.

59. Particular individuals may be members of both sets, of course, or there may be no overlap of membership at all between the two. A complete overlap of membership marks the limiting case that I characterize in the previous section as the germ of a legal system.

60. No less an authority on the history of international law than Martti Koskenniemi testifies to the mistaken acceptance of this assumption, describing international lawyers’ “past strategy to defend international law by a domestic analogy: the assumption that treaties were a kind of legislation, peaceful settlements of disputes a type of adjudication, and war and counter-measures a primitive form of enforcement.” See Koskenniemi, M. & Leino, P., Fragmentation of International Law? Postmodern Anxieties , 15 Leiden J Int'l L. 553 (2002), at 558 CrossRefGoogle Scholar.

61. Payandeh, supra note 5, at 981.

62. The legal theorist who comes closest to describing accurately Hart's position as set out in the text is Alexander Somek, who suggests that the point would be better stated in terms of international law's constitutional deficiency. See Somek, Alexander, Kelsen Lives , 18 Eur. J. Int'l L. 409–451 (2007), at 433.CrossRefGoogle Scholar

63. Hart, supra note 1, at 98.

64. Id. at 236.

65. On this point, see also Richard Collins, The Institutional Problem in Modern International Law (2016), at 83. As Hart notes, however, a legal system that exists in virtue of secondary rules that create the office of judge and the institution of a court “is necessarily also committed to a rule of recognition of an elementary and imperfect sort” (id. at 97). One might point to the recent growth of adjudicatory panels and semiautonomous administrative rule-making bodies in various domains of international law to make the case that it is now possible to formulate a theoretically useful ontological “rule” of recognition for international law. Doing so will require responding to skepticism regarding the genuine autonomy enjoyed by these administrative rule-making bodies and rebutting the claim that the decisions of adjudicatory panels are merely epistemically authoritative claims regarding states’ practice of holding accountable but are not themselves part of that practice (i.e., part of what makes international law what it is). Whether one treats the European Union and the European Court of Human Rights as components of international law will likely also have a significant impact on the position one takes.

66. Waldron's suggestion that Hart's remarks in these paragraphs reflect a confusion on his part rests on the assumption that the rule of recognition plays a different (i.e., epistemic) role in Hart's analysis of law from that which the Grundnorm plays in Kelsen's analysis of law; Waldron, International, supra note 4, at 221. See also Murphy, supra note 32, at 147. My reading of Hart's argument suggests otherwise, with Hart countering Kelsen's transcendental account of law's normativity by pointing to (though not developing in this passage) a naturalistic account of normativity originating in the human practice of holding accountable. For a sketch of such an approach, see Lefkowitz, David, Giving up on Moral Truth Shall Set You Free: Walzer on Relativism, Criticism, and Toleration , 274 Revue Internationale de Philosophie (2015).Google Scholar

67. Hart, supra note 1, at 234.

68. Id. at 235.

69. See Postema, supra note 18; Lefkowitz, Sources, supra note 18.

70. Indeed, we can develop very sophisticated epistemic rules of this sort; see, e.g., the International Law Commission's Draft Reports on the Identification of Customary International Law, available at http://legal.un.org/ilc/guide/1_13.shtml.

71. Hart, supra note 1, at 235.

72. Murphy, supra note 32, at 151–152.

73. Hart, supra note 1, at 214.

74. Waldron, International, supra note 4, at 214. Arguably, the conclusions of a theory of society should strike as trivial the members of the type of society of which it is a theory. For if they do not recognize (some aspect of) themselves and their way of life in the theory, even if in a less flattering light than they might have expected, then that is a good reason to conclude that the theory is mistaken.

Payandeh, too, momentarily takes Hart at his word before rejecting his claim on the basis of the mistaken understanding of what Hart means when he describes international law as primitive, as described in the previous section. See Payandeh, supra note 5, at 994.

75. Hart, supra note 1, at 4.

76. Id. at 216–226.

77. Id. at 227–232.

78. Id. at 232–237.

79. Waldron, International, supra note 4, at 210–111; Waldron, Hart, supra note 4, at 68–69. Jack L. Goldsmith & Eric A. Posner, The Limits of International Law (2006).

80. One might follow Thomas Franck and read portions of The Concept of Law, including Hart's discussion of the rule of recognition, as concerned with law's legitimacy. See Franck, Thomas M., Legitimacy in the International System , 82 Am. J. Int'l L. 705 (1988), at 751–758 CrossRefGoogle Scholar. Though misguided in certain respects, this reading can be made to stand up to some extent as long as our concern is with de facto legitimacy. Yet though they sometimes claim to be engaged only in the kind of social scientific analysis in which this sense of legitimacy has its place, few doubt that Goldsmith, Posner, and Yoo actually advance a normative argument, a claim regarding when actors, or at least the United States, ought to treat international legal obligations as providing it with a reason for action.

81. See Lefkowitz, David, The Legitimacy of International Law , in Global Political Theory Today 98–116 (Held, David & Maffettone, Pietro eds., 2016).Google Scholar

82. Tamanaha, supra note 2, at 54. See also Payandeh, supra note 5, at 993.

83. Cotterrell, supra note 2, at 507, writes:

(Citations removed from quotation; emphasis in original.)

84. Contra Cotterrell, then, von Daniels is right to describe Hart as holding that “a simple regime of primary law can be identified as a legal one.” See Detlef von Daniels, The Concept of Law from a Transnational Perspective (2010), at 143.

85. Payandeh, supra note 5, at 981.

86. See, e.g., id. at 994.

87. Id. at 978.

88. Id.

89. Hart, supra note 1, at 237.

90. Suppose, as some maintain, that Hart thought that noncontingently serving to advance the minimum content of natural law is a necessary condition for a set or system of rules to count as law. If so, then one might read into Hart's talk of international law having a content closer to that of municipal law than any other social rules his implicit recognition of the fact that just as a well-functioning municipal legal order serves to restrain violence, theft, and fraud between individuals, so, too, a well-functioning international legal order serves to restrain the use of violence, theft, and fraud between states (though not between states and colonies).

91. Id. at 214.

92. See, e.g., Jeffrey L. Dunoff & Mark A. Pollack, Interdisciplinary Perspectives on International Law and International Relations: The State of the Art (2013).

93. For arguments to this effect, see David Lefkowitz, International Law, Institutional Moral Reasoning, and Secession, Law & Phil. (forthcoming 2018); Allen Buchanan, Justice, Legitimacy, and Self-Determination (2007); Steven R. Ratner, The Thin Justice of International Law (2015); Carmen Pavel, Divided Sovereignty: International Institutions and the Limits of State Authority; Weins, David, Prescribing Institutions without Ideal Theory , 20 J. Pol. Phil. 45 (2012).Google Scholar