I am grateful to Patrick Emerton, Jeff Goldsworthy, Toby Handfield, Scott Hershovitz, Grant Lamond and especially John Tasioulas for their comments.

1. Dworkin, Ronald, Law’s Empire (Oxford: Hart Publishing, 1986)Google Scholar. I shall refer to this work as “LE”, followed by the relevant page number, in parentheses in the text.

2. Dworkin, Ronald, “Objectivity and Truth: You’d Better Believe It” (1996) 25 Phil. & Pub. Aff. 87 CrossRefGoogle Scholar. I shall refer to this work as “OT”, followed by the relevant page number, in parentheses in the text.

3. In fact, Dworkin, supra note 2 defends both the no-sense and no-relevance theses: see text for infra note 40.

4. This account draws heavily upon Smith, Dale, “The Use of Meta-Ethics in Adjudication” (2003) 23 Oxford J. Legal Stud. 25 at 26-28.Google Scholar

5. I refer to validity, not truth, because objectivists need not accept that moral utterances are truth-apt. For example, an objectivist might regard moral utterances as imperatives, and deny that imperatives are truth-apt. (Conversely, some anti-objectivists claim that moral utterances are truth-apt: see, e.g., Harman, Gilbert & Judith, Thompson, Jarvis, Moral Relativism and Moral Objectivity (Cambridge, MA: Blackwell, 1996) at 158-59Google Scholar.) “Validity” is used as an umbrella term, to cover the various senses in which objectivists could believe that some moral utterances are objectively preferable to others.

6. This thesis applies only to those moral utterances that are claimed to be objective. Objectivists can allow that, while some moral utterances are objectively valid or invalid, others are not.

7. This way of dealing with the blameworthiness example in the previous paragraph was suggested by John Tasioulas.

8. Moore, Michael, “Moral Reality” (1982) Wis. L. Rev. 1061 at 1135-36Google Scholar seems to adopt such a view.

9. See, e.g., Foot, Philippa, “Does Moral Subjectivism Rest on a Mistake?” (1995) 15 Oxford J. Legal Stud. 1 at 12Google Scholar (distinguishing her claim that moral reasons depend on facts about human needs, capabilities, etc. from the “non-cognitivist” claim that moral reasons depend on human feelings or desires).

10. Smith, Michael, “The Definition of ‘Moral’” in Jamieson, Dale, ed., Singer and His Critics (Oxford: Blackwell, 1999) 38 at 58Google Scholar.

11. Hare, R.M., Sorting Out Ethics (Oxford: Clarendon Press, 1997) at 134.Google Scholar Whether Hare and Smith are objectivists is controversial. I classify them as such because they accept the rational convergence thesis.

12. It might seem more appropriate to use Simon Blackburn’s ex Pressivism as my exemplar of anti-objectivism, since it is more popular than PR and is discussed by Dworkin. However, I am sympathetic to Dworkin’s claim that Blackburn’s ex Pressivism is self-refuting (in the sense that, if true, it cannot be stated): see Section IVC. While PR may be mistaken, I believe it can at least be coherently stated.

13. PR can allow for overlap between perspectives—that is, a moral belief may be true relative to more than one perspective. (As a perspective is a set of moral beliefs, there can be overlap between perspectives, just as there can be overlap between other types of sets.) However, PR would claim that any overlap that does exist is due to non-rational factors such as socialisation or non-rational persuasion, and so there cannot be a rational convergence on any particular moral belief.

14. It might be thought that, if PR is correct, there is no room for moral knowledge, and so no room for epistemic errors. However, if PR is correct in claiming to allow for moral truth (concerning which, see note 28 below), then it also allows for moral knowledge (of those moral truths). Also note that perspectival relativism is committed to denying that, as we remove epistemic errors from our belief-sets, we will converge upon a single moral perspective, since this would be to embrace a version of the rational convergence thesis.

15. I have expanded upon this brief account elsewhere, arguing that PR can be coherently stated and does not succumb to some standard objections to moral relativism: Smith, Dale, Making a Difference: The Use of Meta-ethics in Adjudication (Doctoral Thesis, Oxford University, 2005) at 31–63 and 107-28 [unpublished]Google Scholar.

16. This is external scepticism about morality. One could also be externally sceptical about other domains of thought, but it is morality that is relevant to the meta-ethical debate.

17. We shall see that Dworkin wavers as to whether external scepticism denies that moral statements can be true or objectively true. However, the position outlined in the text seems to be his official position, and is (I shall suggest) the better position.

18. Dworkin claims that this presupposition is moral in nature.

19. Dworkin takes a stance on the former issue, claiming that true moral utterances are true regardless of what anyone thinks about them. However, we shall see that he argues that this is a moral claim.

20. See, e.g., Mackie, J.L., Ethics: Inventing Right and Wrong (Harmondsworth: Penguin, 1977) at 36–38 Google Scholar.

21. Nagel, Thomas, The Last Word (New York: Oxford University Press, 1997) at 15 Google Scholar argues that treating relativism as itself only relatively true renders it irrelevant to anyone who is not already a relativist (since it may be true for the relativist, but it is not true for the rest of us, and so we can simply ignore it). However, just because we do not believe that PR is true does not mean that it is not true (for us). We must examine our other beliefs to see whether they entail that PR is true, and check that we have adequate responses to the arguments in favour of PR (such as the argument from diversity). Thus, even if PR is claimed to be only relatively true, we should not simply ignore it.

22. We also saw (supra note 5) that some other forms of anti-objectivism also accept that moral statements can be true or false, while some objectivists might deny this.

23. Dworkin also mischaracterises the meta-ethical debate more generally. This problem is addressed in Sections III. B and IVB.

24. Not all versions of anti-objectivism purport to be neutral even in this revised sense: see infra note 26.

25. See supra note 21.

26. The converse is not true. The neutrality requirement can be breached without breaching the austerity requirement. Indeed, Dworkin portrays Mackie as an external sceptic who rejects the neutrality requirement but accepts the austerity requirement, and argues at length against this form of external scepticism (OT, 112-29). This shows that not all external sceptics accept both the neutrality and austerity requirements; some accept only the latter. However, I shall consider a form of external scepticism (namely, PR) that purports to comply with both requirements, since such a view is harder to defend against Dworkin’s critique.

27. When presenting this argument for the no-relevance thesis, Dworkin typically focuses on adjudication (rather than legal practice and theory more generally), and I shall do likewise.

28. Elsewhere, I have sought to show that PR need not collapse into moral nihilism (that is, the view that no moral statement can be true or false, valid or invalid): Smith, supra note 15 at 111-16.

29. See, e.g., Williams, Bernard, “An Inconsistent Form of Relativism” in Krausz, Michael & Meiland, Jack, eds., Relativism: Cognitive and Moral (Notre Dame, IN: University of Notre Dame Press, 1982) 171 Google Scholar (accusing other writers of arguing this).

30. See, e.g., Hurd, Heidi, “Relativistic Jurisprudence: Skepticism Founded on Confusion” (1988) 61 S. Cal. L. Rev. 1417 at 1467.Google Scholar

31. Hurd may instead have intended to claim that moral relativism collapses into nihilism. However, she offers no argument to support that claim, and I have argued against it: see supra note 28.

32. I suggested in Part II that arguments for PR may draw upon moral beliefs (e.g., about the significance of disputed moral issues). It might be thought that such beliefs must be true whenever PR is true (though they would not, strictly speaking, be implications or entailments of PR, since they are part of the case for PR, rather than vice versa). However, no particular moral belief must appear in every argument for PR. Simplifying for ease of illustration, one argument for PR might treat disputes about abortion as morally significant, whereas another might treat disagreements between Kantians and utilitarians concerning why slavery is wrong as morally significant. Furthermore, it is doubtful whether every defence of PR must include the argument from diversity.

33. This assumes that, if PR is correct, one is rationally committed to accepting the dictates of one’s perspective. If this assumption is wrong, the contrast drawn in the text becomes even starker.

34. I discuss this contrast between objectivism and PR in more detail in Smith, supra note 4 at 45-47.

35. Though even this is true only of forms of external scepticism that adhere to the neutrality requirement: cf. supra note 26.

36. Leiter, Brian, “Objectivity, Morality, and Adjudication” in Leiter, Brian, ed., Objectivity in Law and Morals (Cambridge: Cambridge University Press, 2001) 66 at 77-78Google Scholar.

37. Ibid.

38. See, e.g., Mackie, supra note 20 at 15.

39. The mind-independence thesis denies that moral validity ultimately depends on anyone’s moral beliefs or desires, but allows that moral validity may depend on human needs, capabilities, etc. Therefore, it need not involve claiming that moral properties are part of the “fabric of the universe”.

40. See Tasioulas, John, “Consequences of Ethical Relativism” (1998) 6 Eur. J. Phil. 172 at 184CrossRefGoogle Scholar for a somewhat similar interpretation.

41. The qualification in parentheses suggests how the no-relevance thesis might complement the no-sense thesis.

42. Dworkin never discusses the rational convergence thesis. However, my critique of the no-sense thesis can be ex Pressed in terms of the mind-independence thesis, which he does discuss.

43. This characterisation of moral claims is suggested by Mackie, supra note 20 at 106. For readers who regard it as unduly narrow, Dworkin’s claim also does not appear to instruct people how to live or determine what makes for a successful and worthwhile life (which is the characterisation of ethical claims suggested by Raz, Joseph, The Morality of Freedom (Oxford: Clarendon Press, 1986) at 213 Google Scholar).

44. See Dworkin, supra note 2 at 90 for a passage suggesting this response.

45. Though, again, this overlooks the non-cognitivist possibility.

46. A similar point is made by Tasioulas, supra note 40 at 185.

47. Leiter, supra note 36 at 74-75.

48. Ibid. at 67.

49. Ibid. at 85-91.

50. Blackburn, Simon, Ruling Passions: A Theory of Practical Reasoning (Oxford: Clarendon Press, 1998) at 296-97Google Scholar.

51. Elsewhere, he demonstrates an awareness of it, though his suggested solution is very tentative: see his contribution to the symposium on Dworkin, Ronald, “Objectivity and Truth: You’d Better Believe It” (11 November 1996)Google Scholar, online: Brown Electronic Article Review Service http://www brown.edu/Departments/Philosophy/bears/9611blac.html. (last visited Nov. 29, 2005).

52. PR need not claim that every change in one’s moral beliefs amounts to a change in perspective. For example, a change in beliefs may represent an attempt to remove an inconsistency from one’s belief-set, in order to bring one’s belief-set closer into line with one’s existing perspective.

53. Remember that the issue is whether the perspectival relativist’s claims can be given an intelligible meta-ethical reading, not whether those claims are correct.

54. While PR need not claim that a change in one’s beliefs always results in a change in perspective, I am concerned here with changes that do have that result.

55. Even if the case for PR depends partly on certain moral beliefs, this does not prevent the conditional from being neutral in the first sense endorsed in Part II: see supra note 32.

56. I am grateful to Patrick Emerton for helpful discussions of this issue.

57. See supra note 28. Whether anti-objectivist theories like PR can do justice to the phenomenology of moral discourse and debate is controversial. However, to criticise PR on this basis is not to claim that it is really a (flawed) moral theory.

58. Or possibly an interpretive claim, but one that is epistemic (not moral) in nature: see Section IVD.

59. See, e.g., Darwall, Stephen, Philosophical Ethics (Boulder, CO: Westview Press, 1998) at 12.Google Scholar

60. Tasioulas, supra note 40 at 186.

61. Leiter, supra note 36 at 80 (italics in original, footnote omitted).

62. Tasioulas, John, “The Legal Relevance of Ethical Objectivity” (2002) 47 Am. J. Juris. 211 at 228 Google Scholar.

63. Though see Smith, supra note 4 for a critique of Jeremy Waldron’s defence of a position similar to the no-relevance thesis.

64. See text accompanying supra notes 33 and 34.