The author is grateful to François Tanguay-Renaud and Hamish Stewart for comments on an earlier draft.

1. These now-familiar names for the two views were coined by Waluchow, Wil in his book Inclusive Legal Positivism (Oxford: Oxford University Press, 1994).Google Scholar

2. See my “ Legal Positivism: 5½ Myths” (2001) 46 Am. J. Juris. 199.CrossRefGoogle Scholar

3. The “only if” may seem too stringent, since there are many actions concerning the cruelty or unfairness of which morality is indeterminate, and yet concerning the cruelty or unfairness of which the law must nevertheless make a ruling. But this too reflects an institutional consideration to be factored in afterwards, and so is already allowed for in my formulation. The best explanation I know is Honoré, Tony, “The Dependence of Morality on Law” (1993) 13 Oxford J. Legal Stud. 1, esp. at 16-17.CrossRefGoogle Scholar

4. Of course I would ideally like a different label. “Moralistic” has pejorative overtones (of small-mindedness, intolerance, self-righteousness, pettiness, etc.). Even if stripped of pejorative overtones, it has illiberal overtones which ought to be regarded as pejorative. Personally I hold unreconstructed 1960s-style permissive views about the proper role of the law (and of parents, teachers, “communities,” etc.). These views are moralistic only in that they are held on moral grounds, i.e. because the superficially appealing intervention to prevent or rectify an immorality is, in my view, often more immoral than the immorality it is supposed to prevent or rectify. The word “perfectionist” is sometimes used to capture such a view but it too has misleading overtones (of obsessiveness, fussiness, etc.). I have always rather liked Michael Walzer’s witty suggestion that such a view should be labeled “imperfectionist”: Walzer, “The Imperfectionist,” (review of The Morality of Freedom by Joseph Raz) The New Republic, 7 December 1987 at 30.

5. See my Offences and Defences (Oxford: Oxford University Press, 2007) at chs. 4-7.Google Scholar

6. This paper is by and large a discussion of and reply to Thorburn, Malcolm, “Justifications, Powers, and Authority” (2008) 117 Yale L.J. 1070 [Thorburn].CrossRefGoogle Scholar

7. Ibid. at 1083.

9. Ibid. at 1085.

10. Ibid. at 1086.

11. See the critique that Thorburn borrows, at 1078, from Berman, Mitch, “Justification and Excuse, Law and Morality” (2003) 53 Duke L.J. 1.Google Scholar In this critique I am associated with the view that crimes are legally justified if and only if they are morally justified. Since this view leaves no room for moral error by the law it is clearly anti-positivist, and a fortiori anti-exclusivist. It is in that respect diametrically opposed to my actual view.

12. Thorburn, supra note 7 at 1074 [emphasis omitted].

13. Ibid., e.g., at 1097.

14. Ibid., e.g. at 1107.

15. Ibid. at 1074.

16. Ibid. at 1075.

17. See, e.g., DPP v. Morgan [1976] AC 182 at 206Google ScholarPubMed per Lord Hailsham. For a spirited (but in my view failed) moral critique of the received wisdom on this point, see Dempsey, Michelle and Herring, Jonathan, “Why Sexual Penetration Requires Justification” (2007) 27 Oxford J. Legal Stud. 467.CrossRefGoogle Scholar

18. See, e.g., R. v. Brown [1993] 1 AC 212.Google ScholarPubMed The law of assault is a lot more complex than this summary (or this case) suggests. It is not essential to my argument that consent has the same role in all assaults.

19. Or waive the duty of another. A few among many who associate themselves with this permit-or-waive view: Plamenatz, John, Consent, Freedom and Political Obligation, 2nd ed. (Oxford: Oxford University Press, 1968) at 9–10 Google Scholar; Hurd, Heidi, “The Moral Magic of Consent” (1996) 2 Legal Theory 121 at 123-24CrossRefGoogle Scholar; Owens, David, “Duress, Deception and the Validity of a Promise” (2007) 116 Mind 293 at 303-04.CrossRefGoogle Scholar

20. I am here refining, while preserving in spirit, a point made by John, A. Simmonds in his Moral Principles and Political Obligation (Princeton, NJ: Princeton University Press, 1979) at 76.Google Scholar

21. See Raz, J., The Morality of Freedom (Oxford: Clarendon Press, 1986) at 86–88.Google Scholar

22. Ibid. at 53-57.

23. There is a longstanding debate about the extent to which coercion can be identified non-morally. Much of this debate has focused on whether a coercive proposal by A to B can be identified using a “non-moral baseline,” i.e., without considering the moral acceptability (quite apart from its coercive impact) of what A proposes to do if B does not co-operate. For discussion of possible baselines see Wertheimer, Alan, Coercion (Princeton, NJ: Princeton University Press, 1987) at 204–11.Google Scholar However, it is often overlooked that even those who favour non-moral baselines, or who reject the baseline idea altogether, cannot avoid making the identification of coercion depend on moral judgments in other ways. See notably Frankfurt, Harry, “Coercion and Moral Responsibility” in Honderich, Ted, ed., Essays on Freedom of Action (London: Routledge & Kegan Paul, 1973) at 65.Google Scholar My own view is that one identifies As coercion of B only by identifying (i) aprima facie wrong by A against B which is also (ii) either a justification or excuse for B’s cooperation with or submission to the will of A. To identify manipulation, deception or exploitation one needs to apply the first of these criteria although not the second.

24. Thorburn, supra note 7 at 1078.

25. Raz, Joseph, Practical Reason and Norms (London: Hutchinson, 1975) at 106.Google Scholar

26. For the most detailed statement of my position, see Offences and Defences, supra note 6 at ch. 5.

27. The terminology is Raz’s: Practical Reason and Norms, supra note 25 at 39-40.

28. Offences and Defences, supra note 6 at 106-07.

29. R. v. Flattery (1877) 2 QBD 410 Google Scholar; R. v. Williams [1923] 1 KB 340.Google ScholarPubMed As the court explained these cases in R. v. Linekar [1995] 2 Cr. App. Rep. 49 Google Scholar, “it is the non-consent to sexual intercourse rather than the fraud of the doctor or choir master that makes the offence rape.” For a Canadian implementation see R. v. Cuerrier [1998] 2 SCR 371 Google ScholarPubMed per Gonthier and McLachlin JJ.

30. For example, in a departure from English common law, the Sexual Offences Act 2003, c.42, s. 76(2)(a) (U.K.) makes the “purpose” of an act, and not only its nature, relevant to whether it is covered by the consentor’s consent. I fear that this, like most of the innovation in the 2003 Act, is a serious mistake.

31. Offences and Defences, supra note 6 at 103-04.

32. Thorburn, supra note 7 at 1086, associates this view with Paul Robinson’s “utilitarian account of justification.” However enthusiasm for the same view in moral philosophy is more closely associated with anti-utilitarians, such as Judith Thomson, Frances Kamm, and Thomas Scanlon. For a good survey and defence, see Oberdiek, John, “Culpability and the Definition of Deontological Constraints” (2008) 27 L. & Phil. 105.CrossRefGoogle Scholar (I admit to some hesitation in associating these writers with the Robinson view They claim to be discussing the conditions of “permissibility.” I find it hard to work out how, for them, (king’s being permissible is supposed to relate to (king’s being permitted, never mind how it is supposed to relate to Øing’s being justifiable or justified.)

33. See Crocker, Lawrence, “Justification and Bad Motives” (2008) 6 Ohio St. J. Crim L. 277 at 278.Google Scholar

34. Dillof, Antony calls this the “regard” element in his “Unraveling Unknowing Justification” (2002) 77 Notre Dame L.R 1547.Google Scholar However, he mistakenly regards it as a replacement for, rather than a supplement to, the undefeated reason requirement.

35. A variation on the facts in R. v. Deller (1952) Cr. App. Rep. 184.Google Scholar One might still be an attempted rapist under the rule in R v. Shivpuri [1987] AC 1.Google ScholarPubMed

36. A variation on the facts in Dadson v. R. (1850) 4 Cox CC 358.Google Scholar

37. For a general treatment see Peacocke, Christopher, “Deviant Causal Chains” (1979) 4 Midwest Stud. Phil. 123.CrossRefGoogle Scholar

38. Thorburn, supra note 7 at 1114.

39. Ibid. at 1115.

40. In Offences and Defences, supra note 6 at ch. 1 (co-written with Stephen Shute). Thorburn cites the piece (1115 n. 118) for its resistance to a “consequentialist” account of consent but fails to note that it also challenges his own rival view.

41. Thorburn, supra note 7 at 1080-81.

42. Ibid. at 1080, invoking the example of a rule against “taking someone’s property without her consent.”

43. Ibid. at 1074.

44. Ibid

45. Ibid.

46. Ibid.

47. See R. v. Clegg [1995] AC 482.Google ScholarPubMed The court’s determination of necessity and proportionality makes allowances for the fact that the defendant is not in ideal deliberative conditions, to the extent that this is the case: Palmer v. R. [1971] 1 All ER 1077.Google Scholar As Jeremy Horder points out, this introduces a degree of excusatory latitude into the otherwise justificatory plea of self-defence: Horder, , Excusing Crime (Oxford: Oxford University Press, 2006) at 57.Google Scholar It does not, however, alter the fact that it is for the court, not the defendant, to determine what qualifies as a necessary and proportionate reaction in the circumstances, taking account of any excusatory latitude that may be available. R. v. Scarlett [1993] 4 All ER 629 Google Scholar and Zecevic v. Director of Public Prosecutions, Victoria (1987) 162 CLR 645 Google ScholarPubMed may seem to have casually transferred some of this evaluative authority to the defendant. However, R. v. Owino (1996) 2 Cr. App. Rep. 128 Google Scholar and Osland v. The Queen (1998) 197 CLR 316 Google Scholar reaffirm the common law position as stated in Clegg.

48. Thorburn, supra note 7 at 1074.

49. R v. Thain [1985] NI 457 Google ScholarPubMed, tightening the rule in Dadson, supra note 36.

50. So long as her self-defence remains necessary to repel the attack, bearing in mind that the criterion of necessity builds in an element of efficacy. A futile attempt at repulsion cannot be necessary. See Statman, Daniel, “On the Success Condition for Legitimate Self-Defense” (2008) 118 Ethics 659.CrossRefGoogle Scholar

51. Thorburn, supra note 7 at 1072.

52. Ibid. at 1092; Gardner, Offences and Defences, supra note 6 at 115-16. Thorburn approaches this issue, I think unhelpfully, by asking whether we should think of norms conferring justifications as “conduct rules” or as “decision rules.” He criticizes me for equivocating on this point. But I did not equivocate. Rather, I rejected the distinction: Offences and Defences at 116-17. Indeed I anticipated the point ultimately made by Thorburn himself when he concludes at 1097: “It might be best to avoid this language [of ‘decision rules’ and ‘conduct rules’] altogether and to keep in mind that things are rather more complicated.” I am not sure why Thorburn licenses himself to reject the distinction while he insists on holding me to it against my will!

53. Raz, Joseph, “Promises in Morality and Law” (1982) 95 Harv. L.R. 916.CrossRefGoogle Scholar

54. Hart, H.L.A., The Concept of Law (Oxford: Clarendon Press, 1961) at ch. 3.Google Scholar

55. A well-known illustration is Edwards v. Ddin [1976] 3 All ER 705.Google Scholar Compare von Wright, G.H., Norm and Action (London: Routledge & Kegan Paul, 1963) at 192 Google Scholar, where an attempt was made to explain powers as higher-order permissions. For decisive criticisms consonant with those rehearsed here see, e.g., Bulygin, Eugenio, “On Norms of Competence” (1992) 11 L. & Phil. 201 at 205-06.CrossRefGoogle Scholar

56. It further suggests that there is more than one permissible way to exercise the power. Contrast “mandatory powers,” free of permissive latitude, such as the power of executors to transfer assets to beneficiaries under a will. See Hart, H.L.A., “Bentham on Legal Rights” in Simpson, A.W.B., ed., Oxford Essays in Jurisprudence: Second Series (Oxford: Clarendon Press, 1973) at 196 n. 93.Google Scholar

57. Thorburn, supra note 7 at 1074. The proposition that justification defences “involve the exercise of a legal power” mutates into the proposition that such defences “require at least some case-by-case discretion.” On the same page “discretion” is used apparently interchangeably with “authority”: “the police officer exercised his decision-making authority—his discretion—reasonably.”

58. R. v. Clegg, supra, note 47 at 496 per Lord Lloyd.

59. Lynch v. Fitzgerald [1938] IR 382.Google ScholarPubMed

60. Thorburn, supra note 7 at 1092.

61. Wright v. Sharp [1947] 176 LT 308.Google ScholarPubMed

62. Thorburn, supra note 7 at 1092-93.

63. Hussein v. Chong Fook Kam [1970] AC 942 Google ScholarPubMed; O’Hara v. Chief Constable of Royal Ulster Constabulary [1997] AC 286.Google ScholarPubMed

64. Up to the point at which the arrest is so disproportionate or so unnecessary as to qualify as irrational according to the criteria set out in Associated Provincial Picture Houses v. Wednesbury Corporation [1948] 1 KB 223.Google ScholarPubMed See Mohammed-Holgate v. Duke [1984] AC 437 Google ScholarPubMed per Lord Diplock on the application of the Wednesbury principles in determining the validity of an arrest.

65. Thus excessive force used in effecting an arrest does not invalidate the arrest: Simpson v. Chief Constable of South Yorkshire, The Times, 7 March 1991, [1991] 135 SJ 383.Google Scholar

66. The argument proceeds from Hoye v. Bush (1840) 1 Man & G 775 Google ScholarPubMed, admittedly a case about mistaken identity in the issue of an arrest warrant, and perhaps not extendable to mistakes in warrantless arrests.

67. Walters v. WH Smith & Sons [1914] 1 KB 595 Google ScholarPubMed; R. v. Self [1992] 1 WLR 657.Google Scholar

68. Thorburn, supra note 7 at 1092.

69. See text accompanying note 52.

70. Thorburn, supra note 7 at 1096.

71. In other words, “intentionally” does not entail “knowingly.” I can intend to ф even though, as I know, I have very little prospect of фing: Duff, R.A., Intention, Agency and Criminal Liability (Oxford: Blackwell, 1991) at 55–57.Google Scholar

72. Fletcher’s argument is in “The Nature of Justification” in Shute, Stephen, Gardner, John and Horder, Jeremy, eds., Action and Value in Criminal Law (Oxford: Clarendon Press, 1993)CrossRefGoogle Scholar; mine is in Offences and Defences, supra note 6 at 114-18.

73. Thorburn, supra note 7 at 1074.

74. See, e.g., R. (on the application of Corner House Research) v. Director of the Serious Fraud Office [2008] 4 All ER 927.Google Scholar

75. Thorburn, supra note 7 at 1076.

76. Thorburn, ibid. at 1075.

77. R. v. Lerke [1986] 67 AR 390 at 394-95.Google ScholarPubMed

78. Thorburn, supra note 7 at 1127-28.

79. Hans Kelsen warns of this pitfall of the use of the word “source” in General Theory of Law and State, trans. by Wedberg, Anders (Cambridge, MA: Harvard University Press, 1945) at 131–32.Google Scholar

80. Rice v. Connolly [1966] 2 QB 414.Google Scholar

81. R. v. Chief Constable of the Devon and Cornwall Constabulary ex parte Central Electricity Generating Board [1982] QB 458 Google ScholarPubMed, per Templeman LJ at 479-80.

82. Usually known as the “citizens in uniform” doctrine, this is often traced back to remarks of Lord Mansfield CJ in R. v. Kennett (1781) 5 Car & P 282.Google Scholar A more recent leading case is Albert v. Lavin [1982] AC 546 Google ScholarPubMed, see esp. per Lord Diplock at 565, presenting the special police arrest powers as supplementary.

83. No doubt too much, although he was less rabid a zealot for the doctrine than is sometimes remembered. See Dicey, A. V, Introduction to the Study of the Law of the Constitution 8th ed. (London: MacMillan & Co, 1920) at ch. 4Google Scholar for the zealotry. See ch. 12 for more conciliatory remarks about the Conseil d’État.

84. Entick v. Carrington (1765) 19 St Tr 1029 Google Scholar; Bowles v. Bank of England [1913] 1 ch. 57.Google Scholar This doctrine cuts both ways. Just as public officials are burdened by a need to show special legal authority for everything they do that goes beyond what anyone else may lawfully do, so they are legally permitted unless some law specifically restricts them, to do whatever anyone else may lawfully do: Malone v. Metropolitan Police Commissioner [1979] ch. 344 at 357.Google Scholar

85. I am thinking, for example, of the decision of the Independent Police Complaints Commission not to recommend murder charges against individual police officers involved in the death of Jean Charles de Menezes at Stockwell underground station in London on 22 July 2005. See the IPCC’s report Stockwell One at http://news.bbc.co.uk/2/shared/bsp/hi/pdfs/08_11_07_stockwell1.pdf. Although there is high judicial authority for the proposition that mistakes of fact in self-defence need not be reasonable ones in order to furnish an excuse—for police officers or others—this authority ( Beckford v. R. [1988] AC 130 Google ScholarPubMed, cited by the IPCC at para 19.4 of Stockwell One) strikes many as both morally misguided and per incuriam, and it might helpfully have been re-examined in the courts by the launch of a murder prosecution in this particularly shocking case. Alas the Beckford error has since been put on a statutory footing thanks to s. 76(4) of the Criminal Justice and Immigration Act 2008.

86. Thorburn, supra note 7 at 1128-29.

87. Thorburn, ibid. at 1128.

88. An arguable exception is Andrew Ashworth, who travels part of the same road as Thorburn in arguing that certain kinds of official involvement in crime should be available as defences, albeit not for the official herself but for others who rely on her advice or assurance. Why I say “arguable” is that in such cases Ashworth would favour a nolle prosequi (or similar waiver) over a justification defence, given the choice. Ashworth, , “Testing Fidelity to Legal Values: Official Involvement and Criminal Justice” in Shute, S. & Simester, A.P., eds., Criminal Law Theory: Doctrines of the General Part (Oxford: Oxford University Press, 2002).Google Scholar