1. See Joseph Raz, The Concept of a Legal System: An Introduction to the Theory of a Legal System, 2d ed (Clarendon Press, 1980) [Raz, The Concept].

2. See ibid at 1.

3. Joseph Raz, “Authority, Law and Morality” (1985) 68:3 The Monist 295 at 295.

4. Stanley L Paulson, “Introduction” in Hans Kelsen, Introduction to the Problems of Legal Theory, translated by Bonnie Litschewski Paulson & Stanley L Paulson (Clarendon Press, 1992) xvii at xviii [Kelsen, PTL].

5. Hans Kelsen, Hauptprobleme der Staatsrechtslehre, 2d ed (JCB Mohr, 1923) at v, cited in Paulson, supra note 4 at xx [translated by Stanley L Paulson].

6. See Joseph Raz, Practical Reason and Norms (Oxford University Press, 1999) [Raz, Practical Reason].

7. See David Dyzenhaus, The Long Arc of Legality: Hobbes, Kelsen, Hart (Cambridge University Press, 2022); David Dyzenhaus, “The Constitution of Legal Authority/The Authority of Legal Constitutions” in ibid at 149 [Dyzenhaus, “The Constitution”]; David Dyzenhaus, “Kelsen’s Contribution to Contemporary Philosophy of International Law” (2020) [unpublished], DOI: https://doi.org/10.2139/ssrn.3571343 [Dyzenhaus, “Kelsen’s Contribution”]; Lars Vinx, Hans Kelsen’s Pure Theory of Law: Legality and Legitimacy (Oxford University Press, 2007).

8. See Joseph Raz, Between Authority and Interpretation: On the Theory of Law and Practical Reason (Oxford University Press, 2009) [Raz, Between Authority].

9. See Raz, supra note 1 at 5.

10. Ibid .

11. John Austin, The Province of Jurisprudence Determined, ed by Wilfrid E Rumble (Cambridge University Press, 1995) at 166 [emphasis in original].

12. John Austin, Lectures on Jurisprudence or The Philosophy of Positive Law, 4th ed by Robert Campbell (John Murray, 1873) vol 2 at 525ff.

13. Ibid at 526 [emphasis added].

14. Raz, supra note 1 at 18.

15. An act of legislation is defined as discernable behavior expressing a wish that some person(s) will act in a certain way. See ibid at 11.

16. See ibid at 18-19.

17. Ibid at 18 [emphasis added].

18. In Raz’s words: “To establish the existence or otherwise of a sovereign power one looks for social facts, for a habit of obedience of the population, etc.” Ibid at 29.

19. See Jeremy Waldron, “‘Transcendental Nonsense’ and System in the Law” (2000) 100:1 Colum L Rev 16 at 39.

20. Raz, supra note 1 at 26.

21. This designation follows Raz’s discussion. He asks: What are the criteria which determine the system to which a given law belongs? These Raz dubs “criteria of membership,” and from them one can derive “criteria of identity,” which answer the question: “Which laws form a given system?” Ibid at 1.

22. Hence, for example, the conceptual possibility of a legal system containing no laws at all at a certain moment, e.g. because the sovereign has chosen to repeal all existing laws in a given system and replace them with new ones. See ibid at 16.

23. This fact is the basis for one of Hart’s famous critiques of Austin’s theory: it fails to account for the diachronic unity of law, namely of the fact that both particular laws and legal systems often outlast their makers. See HLA Hart, The Concept of Law, 3d ed (Oxford University Press, 2012) at 50-71.

24. See Raz, supra note 1 at 33 (“certain constitutional laws … are not laws according to the [command] theory”).

25. See e.g. WL Morison, “Some Myth About Positivism” (1958) 68:2 Yale LJ 212 at 215-16; Frederick Schauer, “The Jurisprudence of Custom” (2013) 48:3 Tex Intl LJ 523 at 526.

26. Raz, supra note 1 at 18ff.

27. Raz calls such laws “unlegislated laws.” Ibid at 32.

28. Ibid at 19.

29. See Kelsen, supra note 4 at 21.

30. Waldron, supra note 19 at 37.

31. Ibid.

32. Raz, supra note 1 at 26 [emphasis added]. See also § I.1, above.

33. See HLA Hart, “Positivism and the Separation of Law and Morals” (1958) 71:4 Harv L Rev 593 at 614-15.

34. Ibid at 604.

35. See Waldron, supra note 19 at 37ff.

36. Hart, supra note 33 at 606ff, 628-29.

37. In The Concept, this commitment is reflected in Raz’s meticulous analysis of the structure of a law, following Bentham’s exposition, which vindicates Hart’s claim that in the normal case, laws are applied by classifying particulars. See Raz, supra note 1 at 50ff. As will be seen, Raz fully subscribes to this claim, and it remains part of his commitments throughout his different works.

38. According to Paulson, this question is the legal equivalent of Kant’s question concerning the possibility of our knowledge. See Paulson, supra note 4.

39. Kelsen thought that there is no objectively valid morality, universally true; he was an avowed moral pluralist, even sceptic. Hence, it would be inaccurate to situate PTL as between the natural and moral reality.

40. See Kelsen, supra note 4 at 22ff.

41. Kelsen repeats this condition in several different places. See e.g. Hans Kelsen, Pure Theory of Law, translated by Max Knight (University of California Press, 1967) at 198.

42. See Kelsen, supra note 4 at 11.

43. Raz writes: “[Austin’s] theory, though differing in important respects from that of Kelsen, can be profitably regarded as a variant of the same kind of theory. I propose to regard their theories as two variants of what I shall call the imperative approach.” Raz, supra note 1 at 3-4.

44. See Kelsen, supra note 4 at 23. See also Kelsen, supra note 41 at 58 (“the law does not have exclusively a commanding or imperative character”).

45. Kelsen, supra note 4 at 7-8.

46. For an illuminating discussion of PTL’s Kantian presuppositions, and a critique of Kelsen as a “neo Kantian,” see Paulson, supra note 4 at xxixff.

47. See Kelsen, supra note 41 at 3ff.

48. In Kelsen’s words, positive laws “link legal condition with legal consequence.” Kelsen, supra note 4 at 23.

49. Kelsen, supra note 41 at 4.

50. Ibid .

51. Ibid at 72.

52. Ibid at 90.

53. Kelsen sometimes calls this relation “imputation,” ibid at 89. However, to avoid unnecessary confusions that need not detain us here, we can simply dub it a normative principle, bearing in mind the analogy to the causal principle.

54. See Kelsen, supra note 4 at 11.

55. Kelsen, supra note 41 at 193; and see our opening discussion in the Introduction, above.

56. Kelsen, supra note 4 at 55 [emphasis added].

57. See Paulson, supra note 4 at xviii.

58. RK Gooch, Book Review of General Theory of Law and State by Hans Kelsen, (1945-1946) 32:1 Va L Rev 212 at 214.

59. The point is discussed thoroughly in § III, below.

60. Raz, supra note 1 at 95.

61. Ibid [emphasis added].

62. See e.g. Kelsen, supra note 41 at 193 (“the question why a norm is valid … cannot be answered by ascertaining a fact, that is, by a statement that something is”) [emphasis in original].

63. Kelsen, supra note 4 at 55.

64. See ibid .

65. Kelsen, supra note 41 at 196.

66. Ibid at 198 [emphasis added].

67. See ibid at 199-200.

68. See ibid at 200.

69. See the discussion at § II.1, above.

70. See Kelsen, supra note 41 at 201.

71. Ibid at 202 [emphasis added].

72. See ibid . Kelsen’s use of this term, ‘rule of law’, is equivalent to what some legal philosophers called ‘normative propositions.’ A normative proposition describes the norm to which it relates, and it is true if and only if there is a valid norm corresponding to the description. This allowed philosophers to logically relate norms indirectly, since the norms themselves are excluded from entering logical relations directly (they are considered not to have truth value). Curiously, Kelsen had several changes of heart concerning this issue, which need not detain us here. For an incisive survey, see Stanley L Paulson, “Metamorphosis in Hans Kelsen’s Legal Philosophy” (2017) 80:5 Mod L Rev 860.

73. Kelsen, supra note 41 at 209.

74. Ibid at 209-10.

75. Ibid at 210.

76. Ibid at 210 [emphasis added].

77. See Raz, supra note 1 at 99ff, especially at 101-04.

78. See Paulson, supra note 4.

79. Paulson, in the introduction to the first edition of the Pure Theory, thinks that this flaw does not do away with the theory, because Kelsen’s analysis is not designed to answer the sceptic. See ibid .

80. Curiously, Raz himself notes this: “Of course, Kelsen claims that the basic norm is important for reasons which have nothing to do with the identity and structure of legal systems, reasons derived from his general theory of norms.” Raz, supra note 1 at 105. Nevertheless, Raz never really engages with these reasons, since, as will be seen, his naturalist commitments cannot accommodate them.

81. Raz, supra note 1 at 93.

82. Kelsen repeatedly warns of the conflation of the two, or the reduction of one to the other. See Kelsen, supra note 41 at 211ff.

83. Non-positivist coherence-based theories, e.g., Weinrib’s formalist theory, might be less dynamic than Kelsen’s. Yet they too are committed to the rejection of factual criteria as accounting for validity. See e.g. Ernest J Weinrib, “Legal Formalism: On the Immanent Rationality of Law” (1988) 97:6 Yale LJ 949 at 956. The same applies to Dworkin’s theory, which emerged precisely from the failure of criterial models to account for legal principles. See Ronald M Dworkin, “The Model of Rules” (1967) 35:1 U Chicago L Rev 14.

84. See Kelsen, supra note 4 at 56.

85. See Kelsen, supra note 41 at 230ff.

86. See ibid .

87. See Kelsen, supra note 4 at 67.

88. Ibid .

89. See Kelsen, supra note 41 at 236-37.

90. Ibid at 237.

91. Ibid at 348.

92. Ibid at 349ff.

93. Ibid at 349.

94. Kelsen, supra note 4 at 78.

95. See Kelsen, supra note 41 at 349-50.

96. According to Kelsen, when the legislature creates a statute, they implement the constitution’s provisions similarly to the way that courts create individual legal norms by implementing a statute’s provisions. The distinction between adjudication and legislation is merely quantitative, not qualitative. See Kelsen, supra note 41 at 231-32.

97. Dyzenhaus, “The Constitution”, supra note 7 at 170. See also Dyzenhaus, “Kelsen’s Contribution”, supra note 7, especially at 24-25.

98. At one point, Raz notes an “astonishing” conclusion of Kelsen’s doctrine of norm conflicts (amidst his discussion of the basic norm). Raz, supra note 1 at 65, n 4. Regrettably—and, on my view, given the contrast between the position he ascribes to Kelsen and Kelsen’s actual view—this is the only thing Raz says about this crucial issue.

99. Kelsen, supra note 41 at 267.

100. Ibid at 268.

101. Ibid .

102. Ibid at 268-69.

103. Ibid at 269.

104. Ibid at 270.

105. Ibid at 276. Kelsen sees the creative law application by courts as analogous to any other case of law creation: creating statutory norms based on the constitution, creating administrative norms based on statutes, and so on.

106. See Raz, supra note 1 at 96.

107. For several reservations in Raz’s adoption of Hart’s understanding of secondary rules, see ibid at 164ff. For his discussion of the rule of recognition itself, see ibid at 197ff. It is important to note that for Raz, legal systems may well include more than one rule of recognition. This entails that his model is potentially less centralized than Hart’s.

108. See the discussion at the end of § I.3, above, and specifically supra note 37.

109. Raz, supra note 1 at 89.

110. Ibid at 66.

111. Ibid at 1.

112. Ibid at 66.

113. See Hart, supra note 23.

114. See Raz, supra note 1 at 108; ibid at 189ff (Raz’s discussion of the importance of norm-applying institutions for the identity of the legal system); ibid at 191-92 (Raz’s explication of the criterion of recognition by courts). Again, Raz misrepresents Kelsen as part of the theorists who emphasized legislation over adjudication. See ibid at 190-91.

115. Ibid at 105.

116. Ibid at 24.

117. Ibid at 105.

118. Ibid .

119. Ibid at 105-06.

120. Ibid at 106.

121. Ibid .

122. Ibid at 107.

123. Ibid at 136ff. Raz writes: “Legislative power is simply the ability to create or repeal laws,” and he refers to his earlier discussion, relying on the concept of a basic power. Ibid at 138. But it is precisely to this derivation of ‘is’ from ‘ought’ that Kelsen objects, and for which he resorts to a presupposed Grundnorm.

124. See Raz, supra note 1 at 108.

125. Ibid at 100.

126. Ibid at 95. See also the discussion at § II.2, above.

127. Raz, supra note 1 at 107.

128. Ibid .

129. Ibid at 24.

130. Ibid .

131. See ibid at 60-69.

132. See ibid at 60.

133. See James Bryce, Studies in History and Jurisprudence (Oxford University Press, 1901); Raz, supra note 1 at 97-98.

134. See the discussion in § I.2-3, above.

135. According to Raz, a derivative norm comes into being (only) the moment that at least one appropriate set of creation-conditions is fulfilled. The important point for our discussion is that the creation of a norm always follows the criteria for its existence, and it can never alter these criteria, which logically precede it. See Raz, supra note 1 at 61.

136. Ibid at 19.

137. Ibid at 20.

138. Ibid at 29.

139. Ibid at 62 [emphasis added].

140. Ibid at 68.

141. Ibid at 62, citing Hans Kelsen, Théorie Pur du Droit (Dalloz, 1962) at 3.

142. See Kelsen, supra note 4 at 71ff. In fact, Kelsen is explicit that an ‘unconstitutional’ statute could only mean one thing: “the constitution aims not only for the validity of the constitutional statute, but also (in some sense) for the validity of the ‘unconstitutional’ statute.” Ibid at 72. He defends the same position in the discussion of the (im)possibility of norm conflicts, surveyed earlier. See our discussion at § II.5, above.

143. Hart, supra note 33 at 610. See also our discussion at § I.3, above.

144. See Waldron, supra note 19 at 32 (discussing the positivist’s inability to account for “doctrinal systematicity”).

145. See Raz, Practical Reason, supra note 6.

146. Ibid at 126.

147. Ibid.

148. Ibid at 127.

149. Ibid at 128.

150. Ibid at 141.

151. Ibid at 129.

152. Ibid.

153. Ibid at 130.

154. Ibid at 112.

155. See Kelsen, supra note 4 at 70 (“[r]elativity of the contrast between creating and applying the law”); Kelsen, supra note 41 at 352-53.

156. Raz, Practical Reason, supra note 6 at 137.

157. Ibid at 134.

158. Ibid at 138 [emphasis added].

159. Ibid .

160. Raz, Between Authority, supra note 8 at 1 [emphasis added].

161. Ibid at 5-6.

162. Ibid at 8 [emphasis added].

163. Ibid at 9 [emphasis added].

164. Ibid at 12.

165. Ibid at 9 [emphasis added].

166. Ibid at 9-10.

167. See Dworkin, supra note 83.

168. See Raz, Practical Reason, supra note 6 at 138.

169. See Raz, Between Authority, supra note 8 at 8.

170. Ibid at 9.

171. See the discussion at § II.1, above.

172. See Kelsen, supra note 4 at 15ff.