1. In George, Robert (ed.), The Autonomy of Law 80 (1996).Google Scholar

2. Id. at 80.

3. Id.

4. It is no doubt possible to devise an artificial formal category to serve as an umbrella task for the law. My doubt applies only to the existence of a natural and informative concept that designates the law's umbrella task.

5. See on this Raz, Joseph, The Morality of Freedom (1986), chs. 3 and 4.Google Scholar

6. Though not every argument that can appear as part of legal reasoning is normally referred to as legal reasoning when thought of on its own.

7. The exception being what I termed “conserving” interpretations whose sole purpose is to establish the meaning the original had at some past point in time.

8. George, , supra note 1, at 87–8.Google Scholar

9. Id. at 94.

10. I am not confident that I always used the terms we discuss here in the same way. But here is what I said about them: “Legal reasoning is reasoning either about what the law is or about how legal disputes should be settled according to law.” Raz, Joseph, Ethics in the Public Domain 327 (rev. ed. 1995).Google Scholar

11. The inference is supported by other observations of his as well.

12. Id. at 83. Citing “Natural Law and the Separation of Law and Morals,” in George, Robert (ed.), Natural Law Theory 130 (1992).Google Scholar

13. In various publications. See, e.g., the chapter on “The Autonomy of Legal Reasoning,” in Raz, , Ethics in the Public Domain, supra note 10Google Scholar, from which the preceding definition is quoted.

14. This simplifies the role of legal rules and standards in legal reasoning. Reasoning is structured, with some reasons determining the relevance of others. Legal norms function in legal reasoning not only as stand-alone considerations for one conclusion or another, but also in a structuring role, helping to determine the relevance of various considerations.

15. George, , supra note 1, at 82.Google Scholar

16. Postema may be particularly interested in one that states that not all moral reasons are legal reasons. This is also trivially true. My debt of gratitude to my neighbor is for me a moral reason, which is not, special circumstances apart, also a legal reason.

17. Possibly one should not make the move the chess reasons suggest if it is immoral to do so. But that is a case of conflicting reasons. The thesis about the law I cited above is different: it says that legal standards have no normative force at all; they are no reasons at all, unless they are morally valid.

18. Remember that they can be morally binding even if they are morally defective and in need of reform. They may still be morally binding, in die sense that until changed they ought to be obeyed.

19. Unfortunately, Hart changed his mind about the conventional character of the rule of recognition when he wrote the Postscript to the book (see 2nd ed. at 255). As this makes his theory more vulnerable, and is inconsistent with some of die insights of the original, it is important to remember that this view was not part of the theory expounded in The Concept of Law.

20. I do not mean to deny that, given additional premises not considered by Postema, one may construct an argument for the sources thesis from the pre-emption thesis. I have suggested such an argument in Ethics in the Public Domain, supra note 10Google Scholar, ch. 10. What is denied above is the direct relation between the autonomy of legal reasoning and the pre-emption thesis asserted by Postema.

21. This is Postema's name for it.

22. George, , supra note 1, at 89–94.Google Scholar

23. Id. at 94.

24. I will use “cooperation” and “coordination” interchangeably, as it seems that Postema does. “Coordination” is the more general, and the more appropriate term. But nothing in the argument depends on the distinction.

25. George, , supra note 1, at 92.Google Scholar

26. Because I find the whole approach encapsulated in the generic argument misguided, I will avoid commenting on the last section of Postema's essay. If its arguments are sound, and I am not convinced that they are, then it would show not that there is anything wrong with the autonomy thesis, but that the law is less often binding than many people think. That is my own view, as I explained in The Morality of Freedom, among other places. But the argument of Postema's essay and of my reply concerns aspects of the explanation of the nature of law, and not the morality of various attitudes towards it.

27. George, , supra note 1, at 88.Google Scholar

28. Id. at 92.

29. Id.

30. Which in the present context does not mean nonrevisability.

31. George, , supra note 1, at 97.Google Scholar

32. Perhaps one can engage in detached reasoning in discharging aspects of this task. That would be to surmise what the judge in the case said by simulating his thought processes. But such simulation is controlled by concrete factual checks: “True,” the simulator will say on occasion, “that reasoning from these premises requires the conclusion that. … But the judge is on record as rejecting it; therefore, we can assume that he did not draw that conclusion. It is likely that he committed the fallacy…,” etc.

33. The sources thesis does not exclude reliance on logical norms.

34. George, , supra note 1, at 97.Google Scholar

35. There is no need to repeat the point made earlier that I do not share Postema's assumption that evaluative reasoning is more likely to be controversial than nonevaluative reasoning. I do, however, believe that were this the case then evaluative reasoning would be more likely to be controversial than the reconstruction of the reasoning of others. It may not be easier to engage in detached reasoning about Catholic morality than in engaged reasoning about Catholic morality, but it would be easier to engage in reasoning about how Bishop X interpreted Catholic morality than to engage in reasoning about Catholic morality.

36. George, , supra note 1, at 99.Google Scholar

37. Id.

38. This is roughly the form in which I advanced the thesis in Practical Reason and Norms (2nd ed. 1990)Google Scholar (1975) and The Authority of Law (1979).Google Scholar

39. Up to a point this question has no general answer. The law itself can indicate which reasons are to be excluded. But as I indicated in later writings there is a default exclusion which the law has unless it indicates otherwise. The default exclusion is of the normative considerations that underlie the requirement, i.e., those considerations against the required action which, should they prevail would lead to rejecting the requirement I will not expand on the reasons for this default understanding of the exclusion here. They are not in question.

40. Raz, Joseph, The Authority of Law, essay 10 (1979).Google Scholar

41. You may say that the pre-emptive force of two kinds of laws cannot apply to the courts in such cases: They may not be bound by the law which applies to the litigants, for it does not (normally) apply to courts. And they are not bound by the law which requires courts to apply the law which applies to the litigants, for in the circumstances of the case, the law allows them to change the law which applies to the litigants.

42. George, , supra note 1, at 100.Google Scholar

43. Note that since as I argued on independent grounds, the law claims to have pre-emptive force, to accept it as tending includes accepting its preemptive force as binding.