I would like to thank Richard Bronaugh for his detailed comments on an earlier version.

1. Raz, Joseph, The Authority of Law: Essays on Law and Morality (Oxford: Clarendon Press, 1979) at 104–05 CrossRefGoogle Scholar [Raz, Authority]. To the same effect Raz, Joseph, “On the Nature of Law” (1996) 82 Archiv für Rechts- und Sozialphilosophie 1 at 2Google Scholar (“The universality of the theses of the general theory of law is a result of the fact that they claim to be necessary truths”) [Raz, “Nature”]. Tur and Marmor both say similar things about “analytical jurisprudence”, which I take to be the same as what Raz calls “legal philosophy”. See Tur, R.H.S., “What is Jurisprudence?” (1978) 28 Phil. Q. 149 at 152, 155CrossRefGoogle Scholar; Marmor, Andrei, Interpretation and Legal Theory, 2nd ed. (Oxford: Hart, 2005) at 27 Google Scholar. I should point out that in quoting this passage I do not want to suggest that this fully captures Raz’s views on the subject. Indeed, as I show below (see infra text to notes 2-3 and 16) his views are more complex. See also Bix, Brian H., “Raz on Necessity” (2003) 22 L. & Phil. 537, esp. at 546-49Google Scholar. However, I believe it is useful to examine this statement, because I believe it fairly represents a prevalent view about the scope of legal philosophy as exemplified in matters such as the inclusive-exclusive debate, legal positivists’ responses to Dworkin, analytic philosophers’ accounts of adjudication, etc.

2. See Raz, Joseph, “About Morality and the Nature of Law” (2003) 48 Am. J. Juris. 1 at 3CrossRefGoogle Scholar (stating that “necessarily rape cannot be committed by the law” and that, “the law cannot be in love”, but saying of such necessity statements that “[m]any of them are of little interest”); Raz, Joseph, “Can There Be a Theory of Law?” in Golding, Martin P. & Edmundson, William A., eds., The Blackwell Guide to the Philosophy of Law and Legal Theory (Oxford: Blackwell, 2005) 324 at 324Google Scholar (an account of the nature of law “consists of propositions about the law which are necessarily true, and … explain what the law is”); Raz, Joseph, Ethics in the Public Domain: Essays on the Morality of Law and Politics, rev. ed. (Oxford: Clarendon Press, 1994) at 235 Google Scholar [Raz, Ethics]. For essentially the same view see Dickson, Julie, Evaluation and Legal Theory (Oxford: Oxford University Press, 2001) at 17 Google Scholar and passim. Dickson argues there that “analytic jurisprudence” is interested in those “necessary truths” that also “adequately explain the nature of law,” and that while “a theory’s ability to explain adequately the nature of law is dependent upon its at least consisting of necessarily true propositions”, but that this is “not [a] sufficient condition of the explanatory adequacy of such an approach”. As she elaborates in the whole book she believes the additional feature for an adequate explanation is that the account will focus on the “important” features of the explained phenomena. Cf. Kramer, Matthew H., Where Law and Morality Meet (Oxford: Oxford University Press, 2004) at 224–26 Google Scholar, where Kramer emphasizes that when talking about “necessary connections” (or lack thereof) between law and morality he is interested in “non-trivial” ones.

3. See, e.g., Raz, Ethics, supra note 2 at 237; Raz, Joseph, “Two Views about the Nature of Law: A Partial Comparison” in Coleman, Jules L., ed., Hart’s Postscript: Essays on the Postscript to The Concept of Law (Oxford: Oxford University Press, 2001) 1 at 35Google Scholar (Dworkin’s mistaken view is that “legal philosophy creates the concept of law, whereas in fact it merely explains the concept that exists independently of it”) [the article hereinafter Raz, “Two Views”; the book hereinafter Hart’s Postscript]. This view is shared by other legal philosophers. See, e.g., Gardner, John, “Legal Positivism: 5½ Myths” (2001) 46 Am. J. Juris. 199 at 213CrossRefGoogle Scholar (stating that legal positivism is “normatively inert”); Dickson, supra note 2 at 83-92; Leiter, Brian, “Legal Realism and Legal Positivism Reconsidered” (2001) 111 Ethics 278 at 285CrossRefGoogle Scholar.

4. See Raz, Ethics, supra note 2 at 237; Raz, Joseph, “Incorporation by Law” (2004) 10 Legal Theory 1 at 15-16CrossRefGoogle Scholar: “So long as we allow that it is possible for a population not to be governed by law, there must be a difference between legal standards and those which are not legal, not part of the law.” And it is the task of legal philosophy to discover this difference.

5. Throughout I talk only of law in the sense of nomos (a prescription as to how humans should act) and never in the sense of physis (a description of how nature is).

6. This point is made in Joseph Raz, “Two Views”, supra note 3 at 7-8; Hart, H.L.A., The Concept of Law, 2nd ed. (Oxford: Clarendon Press, 1994) at 213 Google Scholar.

7. See e.g. Leiter, Brian, “Naturalism and Naturalized Jurisprudence” in Bix, Brian, ed., Analyzing Law: New Essays in Legal Theory (Oxford: Clarendon Press, 1998) 79 Google Scholar, following Quine, W.V., “Two Dogmas of Empiricism” (1951) 60 Phil. Rev. 20 CrossRefGoogle Scholar.

8. For the most explicit defence of this approach see Moore, Michael S., Educating Oneself in Public: Critical Essays in Jurisprudence (Oxford: Oxford University Press, 2000) at 310 Google Scholar, following arguments originally made in Kripke, Saul A., Naming and Necessity (Oxford: Blackwell, 1980)Google Scholar. The problem is not with Kripke’s arguments but with their applicability to law. Some reasons for my doubts can be gleaned from what I say infra at text to notes 17-19.

9. The way the question whether something could be law could be relevant to a society’s self-identity may not only be seen in the context of two or more contemporaneous states. The difference in approach as to what could count as law in the Weimar Republic Germany, Nazi Germany and Post World War II Germany is another striking illustration of my point.

10. Raz, Joseph, Practical Reason and Norms, 2nd ed. (Princeton, NJ: Princeton University Press, 1990) at 158–59 Google Scholar. Raz limits this claim to rules “addressed to ordinary individuals.” Ibid. at 158 (emphasis omitted). Hart, in addition to his argument that some legal norms are power conferring and therefore there is no sanction for their violation, see Hart, supra at 34-35, n. 6, has also argued that we can imagine certain legal systems that have no sanctions at all, and gave international law as an example: the “simple truisms” about sanctions “which hold good for individuals do not hold good for states, and the factual background to international law is so different from that of municipal law, [that] there is neither a similar necessity for sanctions … nor a similar prospect of their safe and efficacious use” in international law. Ibid. at 219.

11. This is Raz’s example. See Raz, supra note 10 at 159. Raz offers some reasons for thinking that even a society of angels would need laws. Ibid. at 159-60. Madison, James, by the way, thought otherwise: “If men were angels, no government would be necessary.” The Federalist Papers No. 51 Google Scholar. The answer, I guess, depends on what counts as an angel.

12. While according to Lon Fuller this is a feature of law that refutes legal positivism, legal positivists happily accepted the importance of guidance to understanding law, only they denied it conflicted with their views. For Fuller’s view see Fuller, Lon L., The Morality of Law, rev. ed. (New Haven, CT: Yale University Press, 1969) esp. at ch. 2Google Scholar. For positivist endorsements see Kramer, supra note 2 at 6: “virtually any legal positivist … would ascribe to law the function of guiding conduct”; Raz, Authority, supra note 1 at 214; Scott J. Shapiro, “On Hart’s Way Out” in Hart’s Postscript, supra note 3 at 149, 169-70. And of course Hart’s notion of the rule of recognition is closely related to the idea of the requirement of guidance by law. Indeed in the Postscript to The Concept of Law he says that “it is quite vain to seek any more specific purpose which the law as such serves beyond providing guides to human conduct and standards of criticism of such conduct.” Hart, supra note 6 at 249.

13. I first learned about this practice in Barnes, Julian’s novel (of sorts) A History of the World in 10½ Chapters (London: Jonathan Cape, 1989) at 61–80 Google Scholar. For academic discussion see Evans, E.P., The Criminal Prosecution and Capital Punishment of Animals (London: William Heinemann, 1906)Google Scholar; Hyde, W.W., “Prosecution and Punishment of Animals and Lifeless Things in the Middle Ages and Modern Times” (1916) 64 U. Pa. L. Rev. 696 Google Scholar.

14. See the judgment reproduced in Berriat-Saint-Prix, , “Rapport et recherches sur le procès et jugements relatives aux animaux” (1827) 8 Mémoires de la société Royales des Antiquaires 403 at 444-45Google Scholar.

15. It has been suggested to me that Kelsen’s view according to which laws are not directed at all individuals (even when they are humans), but only seek to guide the conduct of judges in a conditional way: if certain facts have been proven, then the judge has an obligation to act according to the requirement of certain legal rules. See, e.g., Kelsen, Hans, General Theory of Law and State, trans. by Wedberg, Andres (Cambridge, MA: Harvard University Press, 1945) at 58–63 Google Scholar. But while this view offers a “technical” solution to the problem raised by the animal trial cases, it does so at a cost: Kelsen’s account does not seem to fit our understanding of any existing contemporary (even merely human-oriented) legal system. For this criticism see for example Hart, supra note 6 at 35-37, 286.

16. See Raz, Authority, supra note 1 at 50: “There is nothing wrong in interpreting the institutions of other societies in terms of our typologies. … It only means that at some stage you classify their activities … in terms of a scheme for analysing social institutions of which the participants themselves may have been ignorant”. To the same effect see also Raz, “Nature”, supra note 1 at 3-5. For a defence of this idea with regard to other social concepts see Raz, Joseph, The Practice of Value (Oxford: Clarendon Press, 2003)Google Scholar.

17. This, it should be noted, is different from Dworkin’s substantive claim against legal positivism. Dworkin’s claim is that people within the same society differ in what they think is the content of law, i.e., what the law permits or requires, and that this disagreement extends to central cases as much as it does to marginal cases. This, in essence, is Dworkin’s semantic sting argument. My claim is that there may be disagreement about what counts (or could count) as law. Admittedly, however, in Dworkin’s case the distinction between the two is not very sharp.

18. Raz, supra note 10 at 159.

19. It has been suggested to me that if I can know with relative assurance the boundary of English law and the boundary of American law, then I can also know the boundary of English or American law. And similarly I can have a relatively clear boundary of the disjunction of all laws. Does this not undermine the claim that the boundaries of what counts as law are not clear? It does not, because even if I can give an account of what counts as English law, American law etc., the resulting account will be disjunctive as well, and will not grasp the necessary (and thus according to this interpretation, important) features of law.

20. This is not merely an imaginary hypothetical. See Hallen, Barry & Sodipo, J. Olubi, Knowledge, Belief, and Witchcraft: Analytic Experiments in African Philosophy (Stanford, CA: Stanford University Press, 1997) at ch. 2, esp. at 64–72, 76-81Google Scholar, who report the African Yoruba tribe’s lack of a distinction identical to our distinction between knowledge and belief. Yet, we can say about a member of that tribe that he knew a certain thing, meaning by this that he was in the sort of mental state that corresponds to our concept of knowledge.

21. See supra note 2.

22. See, e.g., Waluchow, W.J., Inclusive Legal Positivism (Oxford: Oxford University Press, 1994) at 19–29 Google Scholar; Dickson, supra note 2 at 32-33, 37-38, 65-67 & passim. Dickson, I should stress, distinguishes there between considerations of importance and other theorizing norms, but the distinction she draws is not clear: can a theory be comprehensive if it does not include everything of importance about our object of inquiry? The difficulty in understanding the distinction between importance and other theoretical values is exacerbated when Dickson says that when Finnis talks about “a judgment of significance and importance” he only talks about theory construction values. See ibid. at 39, quoting Finnis, John, Natural Law and Natural Rights (Oxford: Clarendon Press, 1980) at 17 Google Scholar. I try to offer a clearer distinction in the text.

23. Dworkin, Ronald, “Wasserstrom: The Judicial Decision” (1964) 75 Ethics 47 at 47CrossRefGoogle Scholar. Recently Dworkin made it clear that he thought that many of the questions that interest other legal philosophers are in his view not important or interesting. See Dworkin, Ronald, Justice in Robes (Cambridge, MA: Harvard University Press, 2006) 4, 33–34, 185–86 Google Scholar [Dworkin, Justice]. In contrast, see Raz’s opposing views on the relationship between a theory of law and a theory of adjudication in Raz, “Two Views”, supra note 3 at 27-28, 31-36.

24. In fact, they may both be right, if, as I suspect, the disagreements among philosophers about this question are the result of them trying to solve different puzzles about law: some understand the question of law’s normativity as the question when it is the case that a binding legal norm exists, others understand it as the question when law succeeds in creating a prescription that one should at least prima facie, follow.

25. I use the expression “law’s normativity” only as shorthand to the more particular inquiry we have set out, for as I stressed above, there may not be a single answer to this question.

26. For an account of the debate between so-called moral internalists and externalists see Smith, Michael, The Moral Problem (Oxford: Blackwell, 1994) at 60–91 Google Scholar.

27. The work of Ernest Weinrib might seem to contradict my view. See generally Weinrib, Ernest J., The Idea of Private Law (Cambridge, MA: Harvard University Press, 1995)Google Scholar. But I believe this is the most plausible interpretation of his views. If Weinrib’s views are interpreted as saying that the only conceptually possible notion of law is one which maintains the reciprocity between plaintiff and defendant we could simply ask him, what stops us from abandoning the practice of law and replacing it with a different practice, call it “schmlaw”, which aims, say, only at efficiency and cares little about reciprocity. Since there is no logical constraint on having such a practice, and since it is not law, Weinrib’s arguments (under this interpretation) would have no force against such a practice. Therefore it makes more sense to understand Weinrib’s argument as suggesting that there is something morally important about the value of reciprocity which we would lose if we adopted a system (whatever its name was) that dispensed with it.

28. See Bentham, Jeremy, An Introduction to the Principles of Morals and Legislation, ed. by Burns, J.H. & Hart, H.L.A. (Oxford: Clarendon Press, 1996) at 294–95 (ch. XVII, §24)Google Scholar. As to other aspects of universal jurisprudence he was quite dismissive, saying that such universal jurisprudence “must confine itself to terminology”. Ibid. at 295. All this applied only to the “censor”. Bentham of course conceded that in the “expository” sense differences among nations could be significant.

29. Austin, John, “On the Uses of the Study of Jurisprudence” in The Province of Jurisprudence Determined, ed. by Hart, H.L.A. (London: Weidenfeld & Nicolson, 1954) 363 at 373, 367Google Scholar (emphasis added).

30. Some of what I say here (especially in section VII) bears some resemblance to some arguments developed by Ronald Dworkin. However, even Dworkin claims his account “aims at very great generality”. See Dworkin, Justice, supra note 23 at 185. More importantly, he reaches these conclusions via a route (especially his so-called anti-Archimedeanism, but also his distinction between concepts and conceptions and his notion of interpretive concepts), which I find unhelpful. And as Tyler Burge once remarked it “is usually the case in philosophy [that] the important thing is less the conclusion itself [but] the routes that lead to it.” Burge, Tyler, “Intellectual Norms and Foundations of Mind” (1986) 83 J. Phil. 697 at 697 n. 1CrossRefGoogle Scholar.