1. ‘Postema on Law’s Autonomy and Public Practical Reasons: A Critical Comment’ (1998) 4 Legal Theory 1.

2. ‘Law’s Autonomy and Public Practical Reason’ in Robert George, ed., The Autonomy of Law (Oxford: Clarendon Press, 1996) 79.

3. See, e.g., Joseph Raz, The Authority of Law (Oxford: Clarendon Press, 1979) at 157–59.

4. Raz denies the link between the pre-emption thesis and the autonomy of legal reasoning, but not the link between the pre-emption diesis and the autonomy of law as expressed in the sources thesis, supra note 1 at 8 n. 20.

5. Or coordination, ibid, at 9 n. 24.

6. Legal reasoning as ordinary evaluative reasoning is taken by Raz to cover moral and other evaluative factors, ibid, at 4–6. I take ‘other’ to refer to broad values of living together in society, other than those regarded as moral. In a different context, Raz expresses the view that it is ‘fairly obvious that the law plays an important role in securing some forms of social cooperation’, ibid. at 10.

7. Ibid, at 4.

8. Ibid. at 7. Raz follows the formulation used by Postema, which Postema attributes to Raz—Postema, supra note 2 at 113 n. 11.

9. Raz, supra note 1 at 4.

10. Ibid.

11. Ibid, at 4–6.

12. This Raz himself points out in his discussion of what Postema might mean by legal reasoning, ibid, at 4.

13. Ibid.

14. Ibid.

15. Ibid. at 5.

16. Ibid, at 6.

17. Ibid, at 5–6.

18. Ibid.

19. Ibid, at 6. The precise relationship between the law and this form of ordinary evaluative reasoning that is undertaken ‘according to’ it will be examined further below—text at infra note 29 and following.

20. Raz distinguishes an autonomous form of legal reasoning restricted to reasoning as to the content of legal norms, from a non-autonomous form of legal reasoning, ibid, at 4. Raz also refers to reasoning to and from the content of law, ibid, at 6. The same position is maintained by Raz in Ethics in the Public Domain (Oxford: Clarendon Press, 1994) 316–17. There is ample evidence that some form of legal reasoning thesis has been assumed alongside Raz’s sources thesis in his earlier writing without perhaps receiving so much attention—see supra note 1 at 5 n. 10. Whether Raz has always regarded legal reasoning as embracing the two forms of (autonomous and non-autonomous) legal reasoning is questionable—see supra note 3 at 48–50, which contrasts functions of a judge relying respectively on legal ability and moral character.

21. Raz, supra note 1 at 4.

22. Ibid. Raz’s rejection of this possibility is also found boldly stated in ‘Facing Up’ (1989) 62 S. Cal. L. Rev. 1153 at 1204.

23. RSC para 14/3–4/11 (The Supreme Court Practice 1997 1:161).

24. Allegations such as fraud or undue influence may of course be raised in some cases, so as to render the compliance with legal formalities redundant. However, such cases do not question the rules as to what legal formalities are required.

25. I rely upon Raz’s discussion in supra note 20 at 229–32. Acknowledgment of particular cases falling within a general class is found at 229, and acknowledgment that it is possible to derive a true legal statement from a statement of law found in a legal source together with another true factual premise is found at 232. This is sufficient to provide a clear legal answer to a case whose accepted facts fall within a general category found stated within a legal statement found within a source of law. Raz is unconcerned with this conclusion, being occupied with denying that such a clear case exists where a moral (rather than factual) premise is involved (on that point, see further the discussion infra note 42). For further discussion of the law’s use of general classes to provide clear guidance for particular cases, see John Finnis, Natural Law and Natural Rights (Oxford: Clarendon Press, 1980) at 269.

26. Supra note 1 at 19.

27. Ibid. at 20.

28. Text at supra notes 14–19.

29. Raz, supra note 1 at 5. It is worth pointing out that for Raz the legal requirement to use the reasoning does not mark off a distinctive case of evaluative reasoning that can be labelled legal. Although the law may restrict the scope of such reasoning (see point following in main text), the law is providing an opportunity (however so restricted) to engage in ‘ordinary’ evaluative reasoning. Elsewhere Raz expresses his belief about why it is appropriate to engage in ordinary evaluative reasoning on such occasions. There exists a general background reason: there is a natural need to employ such reasoning when determining issues that affect the well-being of others (supra note 20 at 311–12). There may be a specific reason found in the explicit use of an evaluative term in the legal materials (ibid, at 227; cp. supra note 3 at 75). There is also a simple pragmatic reason: there is nothing else other than ordinary evaluative reasoning that could do the job (supra note 20 at 314–17).

30. Supra note 1 at 6. That both requirements are necessary is made clear by Raz in considering the possibility of a perverse judgment, made by an authorised judge in an unauthorised manner—which would not be a case of reasoning according to law.

31. Strictly speaking it is the judges who are autonomous in selecting the premises, or criteria, to employ in their ordinary evaluative reasoning.

32. Differing views can be found as to whether the extreme case ever exists, and if so how common it is. It is interesting to note that Raz provides an example of reasoning from determinate moral premises as stipulated by the law, supra note 20 at 230, though I shall argue below that he does not allow for the full significance of the possibility.

33. See text at supra note 30.

34. I use ‘authorised position’ here to cover the twin aspects of power to give a legally binding judgment and duty to exercise that power. The detailed content of both aspects of the judicial position must contain some reference to exercising that power or performing that duty as the law requires—a judge passing judgment in favour of a party in return for a bribe would be in breach of that duty and would not be exercising that power, though if unchallenged the judgment may remain valid as fulfilling a legal condition of being delivered by a judge in accordance with the legal formalities. On the distinction between exercising a power and fulfilling a condition, see eh. ni of Andrew Halpin, Rights and Law—Analysis and Theory (Oxford: Hart Publishing, 1997).

35. See text preceding supra note 32.

36. This is recognised by Raz elsewhere—supra note 20 at 232.

37. This is included as a case of narrow_a reasoning, Raz, supra note 1 at 5, and is explained in detail, ibid, at 14–16, as being the basis for discovering the content of the court’s decision that is ‘established as precedent’.

38. Ibid, at 5; and similarly at 18. Cp. Raz’s notion of a ‘directed legal power’ discussed in infra note 39.

39. In an earlier treatment of a similar ‘puzzle’, Raz provided a solution which on the one hand boldly eradicated the legal significance of elements (ii-evaluation) and (iii-constraint) by pronouncing that prior to reaching stage (iv-outcome), there simply was no law, whilst on the other hand maintained a link between stages (i-authority) and (iv-outcome) in terms of the judge’s reasoning to the conclusion in (iv-outcome) in accordance with a directed legal power provided in (i-authority). Significantly, this solution is undergirded by treating the sources thesis as axiomatic. It is interesting to note that although the ‘puzzle’ is attributed to the concerns of other theorists, Raz is willing to provide a solution to it, indicating the general importance of the puzzle. Supra note 20 at 223, 228–32. I discuss Raz’s use of a ‘directed legal power’ further in infra note 50.

40. The link is maintained by Raz both by recognising the broader form of legal reasoning, and by the argument built around directed powers, referred to in the previous footnote. The alternative, of treating only the narrow form of legal reasoning as strictly falling within the concern of the law, was perhaps at one time entertained by Raz in making a distinction between the legal and moral roles of a judge—see supra note 20.

41. See supra notes 25 and 26, and text accompanying.

42. Supra note 20 at 232. This is the occasion for Raz’s pronouncement of no law prior to the court’s decision, remarked on in supra note 39. It occurs in juxtaposition to Raz recognising but not pursuing a different possibility where a factual premise is involved (see supra note 25).

43. Ibid, at 230.

44. Ibid, at 232.

45. Referring to the common standards of society is a frequent judicial practice. The danger is that judges can purport to be doing it in a case where there are no common standards, and hence the moral test found in the law is not determinate. I discuss an example of this in relation to dishonesty in the Theft Acts in ‘The Test for Dishonesty’ (1996) Crim. L. Rev. 283. However that is not

46. Raz stresses that this sort of reasoning can be undertaken by judge or citizen—supra note 20 at 311.

47. We should remind ourselves that the moral test is indeterminate and hence there exists a number of possible answers that ordinary evaluative reasoning could provide (cp. Raz in acknowledging that two people sharing the same moral view and information may exercise their moral judgements with different outcomes, supra note 3 at 199). This opens up consideration of an ambiguity in the legal reasoning thesis. Raz speaks of what the court ‘should decide’ if a case were before it. The ‘should’ is ambiguous as to the probable or normative. In the case where the law contains an indeterminate moral test, my legal reasoning about what the court should decide must involve reading the ‘should’ as a matter of probability—I am otherwise indulging myself in moral reasoning as to what I would decide if I were the judge, which given that I am not may have nothing to do with the law. In the case where the law contains a determinate moral test both readings will lead to the same result—the probable outcome will ride on the back of the determinate moral outcome. The legal reasoning thesis accordingly makes sense as a thesis of legal reasoning if we generally take the ‘should’ in the probability sense, bearing in mind that calculations of such probability may require us to engage in such ordinary evaluative reasoning that we could expect of the court.

Two further points need to be made. If we accept the possibility of a determinate moral test, then we must acknowledge the possibility of a probability of 1, i.e., a certain outcome in applying ordinary evaluative reasoning. A more far-reaching consequence of examining the role of ‘should’ in the legal reasoning thesis is to open up a possibility of tensions between what the judge considers he should (normatively) decide, what I consider the judge should (probability) decide, and what I consider the judge should (normatively) decide—which makes the thesis in its present form inadequate. These tensions will be explored in detail in the following section.

48. Raz shows how a test for inflation might be regarded as factual or evaluative, and speaks of determinate and indeterminate outcomes of evaluative tests without providing any clear guidance on how we can recognise one rather than the other—supra note 20 at 228–31.

49. Cp. Raz, ibid, at 233, 223, 237.

50. It is worth making briefly some further comment on two of the arguments Raz employs in order to suggest the existence of the no law zone (see supra note 39). One is the argument on directed powers, already referred to. The argument depends on establishing an analogy between a directed legislative power and a directed judicial power (ibid, at 228) but this analogy fails. Raz is right to point out that there is no law in the former case until the legislative power has been used to bring it about (so that nobody could purport to be acting lawfully or complain that another is acting unlawfully in accordance with such law until it is enacted: see, in particular, Raz’s discussion of EC directives that do not have direct effect—ibid, at 235). However, in the case of a directed judicial power, law does exist prior to the exercise of the power, which can be employed to guide behaviour and be used to complain about the unlawful behaviour of another, even as a preliminary to the very exercise of that directed power in the case in which the complaint is made. Raz acknowledges the distinction (ibid, at top of 235) but fails to allow for its significance, relying on the uncertainty that sometimes accompanies the exercise of a directed judicial power (ibid, at 235–36)—but this is uncertainty as to what the law is, or whether the existing law will survive a challenge to overturn it, not certainty as to there being no law. (See further at infra note 64.)

The second argument is related to the first, suggesting that there is no law for the citizen until the point judgment is reached. This leads on to an acute problem that Raz faces in his exchange with Postema (supra note 1 at 18–20). How can any norm be regarded as binding on a citizen if it is capable of being challenged in the courts, and upon challenge there exists no norm until judgment is given? Raz’s response to the problem is to rely on the limited motivation and practical opportunity to engage in litigation. This ducks the issue because it fails to address all those cases where citizens do not consider litigation because they share a common perception of what the law is without needing a judge to determine it. It also fails to address what happens in those cases where litigation does occur and arguments are mounted as to the existing law, and remedies sought on the basis of its existence at the time of the wrongs alleged.

51. This variant covers a number of types of situation. One is where the legal material contains an abstract proposition which must be subjected to moral or other evaluative factors before determining whether it should be instantiated in a particular concrete case—I have discussed this in relation to rights in chs. V and VI of my Rights and Law—Analysis and Theory, supra note 34. Another variant is where linguistic imprecision in the legal material permits the court to invoke moral or other evaluative factors in considering whether a legal principle should apply to a particular case—e.g., the statutory provision considered earlier requiring contracts for the sale of land to be in writing explicitly provides that this will be satisfied ‘where contracts are exchanged’. The question whether this phrase covered the final stages of an agreement concluded by written letters fell to be decided (in the negative) by the Court of Appeal in Commission for the New Towns v. Cooper (G.B.) Ltd. [1995] Ch 259—overt evaluative reasoning on this point is to be found at 287. Reasoning by analogy may also be found here (overlapping, if not a separate type of situation). On this see Raz, supra note 3 at 201–06, and the extensive discussion by Scott Brewer in ‘Exemplary Reasoning: Semantics, Pragmatics, and the Rational Force of Legal Argument by Analogy’ (1996) 109 Harv. L. Rev. 923. Both conclude that reasoning by analogy from legal materials is not itself determinative of the issue.

52. An explicit admission that the case before them fell into this category was made by the House of Lords in Airedale NHS Trust v. Bland [1993] AC 789, at 879–80 per Lord Browne-Wilkinson.

53. The paradigm setting in English law is the Practice Statement (Judicial Precedent) [1966] 1 WLR 1234, which permits the House of Lords to depart from their own previous decisions when they consider that ‘it is right to do so’—making overt the ordinary evaluative reasoning involved.

54. For an effective denunciation of the House of Lords for subverting statutory material, see Glanville Williams, ‘The Lords and Impossible Attempts, or Quis Custodiel Ipsos Custodes?” (1986) 45 Cambridge L. J. 33. I have argued that the English Court of Appeal acted in breach of the principles of precedent in altering the basis for assessing damages for defamation, in ‘Law, Libel and the English Court of Appeal’ (1996) 4 Tort L. Rev. 139.

55. Cp. Raz, supra note 1 at 14–16. Raz labels what I refer to in 1(e) ‘reconstructive argument’, but does not fully discuss the 1(g) scenario. This may be attributed to Raz subsequently focusing on a case of the court changing previously clear law (variant (c₃)) rather than clarifying unclear law (variant (c₁) or (c₂)), marginalising even this prospect, and stressing that the law is anyway binding either side of the change (ibid, at 18–20). Similarly, in supra note 20 at 323–24, Raz narrows down the discussion to an either-or between judicial evaluative (moral) reasoning in deviating from legal doctrine, and the application of legal doctrine.

56. In the context of ‘reconstructive reasoning’, Raz concedes that ‘we often rely on our ideas of what is reasonable in reconstructing other people’s thought’, but limits the legitimacy of this to occasions when we have reason to believe that the person whose reasoning we are reconstructing shares our view of what is reasonable (supra note 1 at 15–16). The scope for bringing in our own evaluative reasoning is actually greater than Raz allows for a number of reasons: (i) the full range of speculative reasoning extends beyond Raz’s case of reconstructive reasoning, taking in (1)(g) as well as (1)(e) with variants (g₁)-(g₄) and (e₁)-(e₄)—including dealing with the situations illustrated in supra note 51; (ii) even if we are attempting to reconstruct the evaluative reasoning of a single judgment, the material we are working on may not be consistent or coherent, permitting us to choose those parts we find reasonable; (iii) this opportunity is greatly increased when we have multiple judgments to reconstruct; (iv) a similar point applies when we have a variety of precedents to select from. The practical demonstration of this occurs in just about every appellate court case, where counsel learned in the law are able to argue on opposing sides constructions or reconstructions of the law employing evaluative reasoning favourable to their clients.

57. This occurs in Jones v. Secretary of State for Social Services [1972] AC 944, where the House of Lords considered it inappropriate to exercise their power to depart from a previous decision.

58. This is the converse of point 2(c). It also provides a significant contrast with point 1(b) in which the existence and content are given by formal deductive reasoning from a norm recognised by a test of legal sources. For further discussion of how a norm may be said to exist without a clear content, see supra note 50 and infra note 64. The requirement in 3(d) of a recognition of a legal source means that in the case of speculative legal reasoning the opening for ordinary evaluative reasoning is restricted to that which speculates about what a court might actually decide. Hence such reasoning is not elevated over legal sources, and, as I shall discuss below, such finality as they may provide. In relation to the judicial reasoning in 3(a) or 3(c) this requirement cannot offer a bootstrap lift to provide a legal source to the judiciary of that material which will be recognised as having a legal source by their determining it. However, the reasoning so engaged in is not completely unrestricted ordinary evaluative reasoning, not least because it is undertaken in discharging the requirement to provide a formally recognised source of law, and also due to the general intermeshing of doctrinal argument and evaluative reasoning (see further infra note 63).

59. Cp. Raz, supra note 1 at 5: ‘moral … in the sense that they embody … moral considerations’; and supra note 3 at 40.

60. The question being considered is how one should behave according to the law.

61. A wider possibility than Raz considers—see supra notes 43 and 44. Nor is a reasonable reluctance to become involved in litigation necessarily an impediment (Raz, supra note 1 at 20), since a position may sometimes be taken up as lawful where the burden of litigating to establish otherwise is on the other side.

62. The remark applies equally to an attempt to combine the evaluative with the non-evaluative in a version of soft positivism, so far as that too seeks to rely on a formulaic approach. Some recognition of the possible limitations of soft positivism from a perspective in favour is provided by Matthew Kramer, ‘Coming to Grips With the Law: In Defense of Positive Legal Positivism’ (1999) 5 Legal Theory 171 at 193–200 in recognising that there may be cases where ‘soft positivism is unfeasible’—where general perception of a correct moral answer, or at least institutional unanimity on the answer to be given, does not occur. Criticism of the limitations of soft positivism from a hostile perspective is provided by Eleni Mitrophanous, ‘Soft Positivism’ (1997) 170x. J. Legal Stud. 621, particularly concerned with the inability of soft positivism to draw a line between the morality that can be incorporated into law and the morality that remains outside. Both authors see limitations on the basis of a failure by soft positivism to provide a workable formula by which law can be identified.

The position taken by Mitrophanous is that adopted by Raz, supra note 3 at 47 n. 8. It is interesting to note that in taking up Raz’s challenge W.J. Waluchow, inclusive Legal Positivism (Oxford: Clarendon Press, 1994) at 226–29, suggests a distinction between morality incorporated into law and morality that remains outside operating within a realm of judicial discretion, on the basis of whether the moral test within the law is determinate or not. However, in addressing the problem of resolving whether the test on a particular occasion is determinate Waluchow resorts to relying on how it is seen by the judge who applies the test. This simply leaves open at a point of prior formulation the indeterminacy Waluchow seeks to satisfy Raz in closing: it will now be indeterminate which cases are regarded as determinate by the judge. Waluchow thus reinforces Raz’s request for a formulaic test of law without succeeding in providing one. His approach also conceals the danger mentioned in supra note 45. Recognition of some occasions of determinate moral tests in the law does not mean that we can clearly identify every occasion—which tells against a formulaic approach to law of whatever persuasion.

63. It is worth reminding ourselves of the twofold rationale provided by Raz for his original sources thesis, supra note 3 at 48–52: first, it is supposed to reflect a distinction between different conditions the law can be in (settled/unsettled) as mirrored in the different abilities judges must possess in order to fulfil their roles (doctrinal/moral); secondly, it is supposed to capture a fundamental characteristic of law, viz., that it marks off authoritative rulings which set a limit to the possibility of continuing to challenge the justification of those rulings. It is the ‘explanatory power’ that dominates the first rationale, but this can be rebutted in three ways, by our observations of the artificiality that it entails in dividing one form of legal reasoning from another, in the creation of the no law zone, as well as in the purported distinction between judicial roles (rather than recognising that the one slips in practice into the other—as Raz does himself elsewhere, supra note 20 at 319). This last point can cause further confusion by portraying a realm of judicial activity that is objectively doctrinal, which may well be used to cover activities that are anything but. For an illuminating discussion on this phenomenon in relation to the use of the doctrine of the Rule of Law by the Israeli Supreme Court, see Alon Harel, “The Rule of Law and Judicial Review: Reflections on the Israeli Constitutional Revolution’ in David Dyzenhaus, ed., Recrafting the Rule of Law: The Limits of Legal Order (Oxford: Hart Publishing, 1999). For a further illustration of the use of ‘objective principles of law’, see infra note 76. As to Raz’s second rationale, I seek below (infra text at note 70) to develop this further by distilling out of the sources thesis the quality of finality, whilst suggesting that in its original form it goes beyond the purely descriptive (infra note 74).

64. The distinction between judicial and legislative law making, in relation to filling out the content of legal norms, consists in this. Even though for every matter that may be disputed the courts are required to pass judgment, it is not the case that every such matter must be covered by an act of the legislature. Hence, in every such matter there is a potential legal norm in the judgment of the court, the content of which, though not predetermined by existing legal sources, will be determined by the ordinary evaluative reasoning of the courts. Such a potential legal norm does not necessarily exist in legislative activity. Accordingly, it is possible in relation to a certain matter, in the case of a judicially created norm to state that a norm will be recognised as existing, but not know for sure the content of the norm until judgment is given. On the other hand, we cannot say that a legislatively created norm will come into existence in relation to a certain matter until legislation is passed—which will determine simultaneously both the norm’s existence and its content.

65. Such speculative reasoning may range over a number of factors. It may be largely doctrinal, on the assumption that the law in the area of concern is currently developing a coherent doctrine. It may be more evaluative, on the assumption that evaluative factors embedded in existing law can point the way, or on the assumption that the values established in previous law are likely to be departed from. It may even rely on knowledge of the character and inclinations of the individual judge involved, so as to predict the likely outcome. I consider some empirical evidence of this last possibility in ‘Law, Theory and Practice: Conflicting Perspectives?’ in Avrom Sherr & David Sugarman, eds., Legal Education and Theory (forthcoming, Ashgate). The possibility is noted by Raz, supra note 20 at 232.

66. See text following supra note 22.

67. Raz, supra note 1 at 7—again, following Postema’s formulation, which is attributed to Raz himself: Postema, supra note 2 at 114 n. 21.

68. In reality, practical reason may be used in this context not simply to address the issue of how to act lawfully (as I have assumed here), but also the issue of how to act expediently. In particular, if there is some doubt about how the court may determine a case, we may be thinking as much along the lines of whether we can risk getting our speculation about the outcome wrong, as anything else. Taking it a stage further, even if we are confident about what the eventual outcome will be, we may still be concerned with the expediency of being involved in protracted litigation—Raz shows some sensitivity to these wider considerations in the passage cited supra note 61.

69. Raz regards the pursuit of finality as part of the sources thesis, supra note 1 at 13–14; cp. Raz, The Concept of a Legal System, 2d ed. (Oxford: Clarendon Press, 1980) 215–16. Postema incorrectly takes Raz’s position on finality as being linked to excluding a ‘range of moral and evaluative considerations’ (supra note 2 at 93), whereas it is about ending the debate on such considerations.

70. If both parties are prepared to rely on a common understanding of the law the ultimate finality of the law may be taken for granted. Where one party is prepared to challenge the understanding of the other party, then the ultimate basis for the finality of law will be revealed in the judgment of the court which is given the final say on the matter. An extreme illustration of this point is to be found in a Swedish case, discussed by Dennis Tõllborg in ‘Law as Value’ (forthcoming) Archiv fur Rechts- und Sozialphilosophie, in which a Swedish first instance judge refused to uphold a constitutional right for the public to be given access to evidence that had been used in a trial, despite two rulings by the Court of Appeal that she should do so. The evidence requested constituted child pornography. Since the appeal court could only order the first instance judge to make the order, she thus had the final say over whether the order was made—a finality confirmed ultimately by the Government’s decision not to remove her from office but to change the law.

71. For fuller discussion of Postema’s assumption about law’s ‘defining task’, see Raz, supra note 1 at 2–4. The appeal of Postema’s view is indicated in the prominence given to it in Henrik Palmer Olsen & Stuart Toddington, Law in Its Own Right (Oxford: Hart Publishing, 1999) at 10–12. For present purposes, it suffices to make three simple points. First, a fundamental flaw in Postema’s characterisation of the defining task appears in his initial conjunction: ‘designed to unify public political judgment and co-ordinate social interaction’ (Postema, supra note 2 at 80). It simply is not the case that coordinating social interaction depends on unifying public political judgment. It depends on being able to state with finality whose political judgment counts. Second, the tradition of thinking about law ‘from Cicero to the present day’ (ibid.), which Postema cites as authority for his defining task, is simply not matched by the practical experience of law from beyond Cicero to the present day. The golden age of Roman law did depend on being able to state with finality whose political judgment counted, but had nothing to do with unifying public political judgment. Law flourished while the political judgment of plebeians, women generally, and slaves, was not taken into account. Indeed (and this takes us into the third point), Postema himself acknowledges that modern society has ‘problems of social co-operation caused, in part, by the absence of a common public language of deliberation and justification’ and concedes that it ‘may not be possible to construct a framework for common deliberation and public justification in societies deeply divided about matters of fundamental value and principle’ (ibid, at 111–12). However, Postema does not suggest that modern society suffers from a failure to coordinate social interaction, or an absence of law. Third, if we have a problem of discovering a basis for unifying public political judgment, this is a political (or moral) problem, not primarily a legal one. If the problem is solved, the answer will be reflected in our law. Equally, if the problem is not solved we will still have law that reflects that state of affairs. To suggest that there is an issue as to whether law and political morality should be approached by a strategy of isolation or integration (ibid, at 111) as a means of dealing with the problem of finding a basis for unifying public political judgment misses the point.

72. Note the scepticism expressed by Postema himself, cited in the previous note.

73. Raz, supra note 1 at 12–13.

74. There is an aspirational as opposed to descriptive character to both of these features of the respective theories—see Hart, The Concept of Law (Oxford: Clarendon Press, 1961) at 92, with a view to remedying uncertainty; Raz, supra note 3 at 52, in order to provide ‘publicly ascertainable standards’; and cp. W.J. Waluchow, supra note 62 at 119–23, 184,188-90, on the evaluative rather than descriptive ascription of function.

75. Hart, in his imaginative but far from historically accurate account of the transition from a pre-legal to legal world, supra note 74 at 89–91, draws attention exclusively to rules. Raz, supra note 3 at 51–52, does speak of ‘authoritative rulings’ but these are made synonymous with ‘rules’ or ‘ascertainable standards’. The preoccupation with rules ignores the role of the ruler.

76. The recognition of dual roles for the courts in passing judgment is to be found in ch. 10 of Raz, supra note 3. Judges themselves are often more reluctant to acknowledge them. The impending incorporation of the European Convention on Human Rights into UK law led to an outpouring of judicial theorising, tending to dampen fears that incorporation would somehow increase judicial discretion in shaping UK law. One example is the Ganz Lecture in Public Law delivered at Southampton University in November 1997 by Sir John Laws, “The Limitations of Human Rights’ (subsequently published in (1998) Public Law 254). Sir John develops an argument that regards rights as legal constructs in order to free them from an underlying concept of morality; the aim is to reach the conclusion that judges may be trusted to interpret the rights of the Convention, because in so doing they are only performing their traditional function of dealing with objective principles of law—thus reaching a position that Postema could only portray as mythical. For (sceptical) discussion of earlier examples of judicial theorising, see John Griffith, ‘Judges and the Constitution’ in Richard Rawlings, ed., Law, Society, and Economy (Oxford: Clarendon Press, 1997) at 289; and for general reflection on the transition from Rechtsstaat to Justizstaat, see Neil MacCormick & Robert Summers, Interpreting Precedents (Aldershot: Dartmouth, 1997) at 549–50.