1. Hegel’s Philosophy of Right, trans. T. M. Knox (Oxford: Oxford University Press, 1967), para. 110 [hereafter Ph.R.].

2. See H. L. A. Hart, “The Ascription of Responsibility and Rights” Proceedings of the Arist. Soc’y, New Series, XLIX (London: Harrison, 1949) at 188; J.-P. Sartre, Being and Nothingness, trans. Hazel Barnes (New York: Washington Square Press, 1966) at 678.

3. See A. Ashworth, “Taking the Consequences” in S. Shute, J. Gardner, & J. Horder, eds., Action and Value in Criminal Law (Oxford: Clarendon Press, 1993) 107; “Belief, Intent and Criminal Liability” in J. Eekelaar & J. Bell, eds., Oxford Essays in Jurisprudence (Third Series) (Oxford: Clarendon Press, 1987) 13; B. Williams, Moral Luck (Cambridge: Cambridge University Press, 1981) at 20; T. Nagel, Mortal Questions (Cambridge: Cambridge University Press, 1979) at 24.

4. See Smith v. State (1888), 50 Ark. 545, 8 S.W. 941.

5. See The People v. Lewis (1899), 57 Pac. 470.

6. See The Queen v. Jordan (1956), 40 Cr. App. R. 152.

7. In all that follows I am assuming a distinction between criminal liability to punishment simpliciter and criminal liability to a particular measure of punishment. Criminal liability in the former sense is traditionally thought to require some distinctive criminal level of fault such as an intention wrongfully to interfere with another’s liberty or reckless indifference toward the commission of a wrong. However, once criminal liability for a rights-infringement has been settled, the accused’s deserved measure of punishment depends on the nature of die consequential harm and on the degree to which the harm is imputable to his agency. Thus criminal responsibility for a particular harm can be based on the degree of imputation called foreseeability once criminal liability for abstract wrongdoing has been established on more stringent criteria of fault. Manslaughter consequent on an assault is an example. Here criminal liability requires an intentional application of force to another person (or recklessness with regard to non-consent), but responsibility for the death (and so liability to the specific measure of punishment attached to manslaughter) is based on foreseeability. Accordingly, it is perfectly consistent to deny that negligence can ever be a basis for criminal liability and accept that it could ground responsibility for a harm once criminal liability for wrongdoing has been settled on subjective criteria. In this essay I am concerned only with the conditions of legal responsibility for harms assuming that criminal liability for wrongdoing has already been established.

8. For a discussion of this aspect of the novus actus doctrine, see D. Galloway, “Causation in Criminal Law: Interventions, Thin Skulls and Lost Chances” (1989) 14 Queen’s L. J. 71 at 75–76.

9. For an attempt to sort out the various meanings of supervening causes, see T. Christlieb, “Why Superseding Cause Analysis Should be Abandoned” (1993) 72 Texas L. Rev. 161 at 162–169. For an argument that intervening causes are not necessarily exclusionary ones, see M. Zimmerman, “Intervening Agents and Moral Responsibility” (1985) 35 Phil. Quart. 347.

10. (1959), 43 Cr. App. R. 121.

11. Ibid, at 31.

12. [1991] 3 All E.R. 670.

13. Ibid, at 677.

14. This is not to say that the novus actus reasoning alone explains the intuitively correct result. If the policeman’s killing of the robber was justified, then there is no unlawful outcome for which the accomplice can be held criminally responsible.

15. Supra note 5.

16. H.L.A Hart & T. Honoré, Causation in the Law, 2nd ed. (Oxford: Clarendon Press, 1959) (Oxford: Clarendon Press, 1985).

17. Ibid, at 136. See also 326. In the first edition, the rule was stated as follows: “the free, deliberate and informed act or omission of a human being, intended to produce the consequence which is in fact produced, negatives causal connection.” H.L.A Hart & T. Honoré, Causation in the Law (Oxford: Clarendon Press, 1959) at 129. We shall see that, though overbroad, this formulation comes closer to the truth than the revised one. For a critique of the earlier formulation, see J. Feinberg, Doing and Deserving (Princeton, NJ: Princeton University Press, 1970) at 152–86.

18. (1949), 362 Pa. 596, 68 A2d 595.

19. Supra note 16 at 334.

20. We can eliminate any other theory of exculpation by supposing that the policeman’s act was unjustified (though excused) because the deceased had a duress justification for joining the robbery. So there is now an unjustified homicide for which the surviving robber is, according to the Hart-Honoré theory, responsible.

21. (1970), 261 A.2d 550.

22. Supra note 16 at 330–31.

23. Ibid. at 331.

24. Ibid, at 329.

25. [1975] 3 All E.R. 446.

26. Supra note 16 at 361.

27. Perhaps a caveat is in order here. In looking to the property relation for clues about imputation, I am far from claiming that the analogy between the two relations is perfect and that we can therefore simply “apply” the assumptions underlying property to resolve problems of criminal responsibility. Rather, I am suggesting that die analogy drawn at the level of everyday language reflects a common set of ontological assumptions and that one way of bringing into focus the assumptions underlying imputation is by first observing their operation in the property context. Since I am not applying property assumptions to imputation, nothing in my argument logically depends on the exactitude of the analogy between the two relations, and so it is irrelevant that there are numerous disanalogies. In the end, what is important for my purposes here are the assumptions underlying imputation, and I use property only as a heuristic tool for bringing to light those assumptions.

28. See The Tubantia, [1924] All E.R. Rep.; see also Hegel, Ph.R. para. 59; I. Kant, The Metaphysics of Morals, trans. M. Gregor (Cambridge: Cambridge University Press, 1991) at 68–69. The sophisticated understanding views property as any legally protected interest, but this makes the notion of property redundant, since it no longer refers to a distinctive kind of entitlement; see J. L. Scroeder, “The Vestal and the Fasces: Property and the Feminine in Law and Psychoanalysis” (1995) 16 Cardozo L. Rev. 805 at 81 On. 23.

29. See Hegel, Ph.R. para. 36. See also Kant, The Metaphysics of Morals at 63: “Freedom … insofar as it can coexist with the freedom of every other in accordance with a universal law, is the only original right belonging to every man by virtue of his humanity.”

30. Robert Nozick has famously posed this alternative as one between assimilating the object to one’s labour or dissipating one’s labour in the object; see Anarchy, State, and Utopia (New York: Basic Books, 1974) at 174–75.

31. Hegel makes the point succinctly: “Occupancy makes the matter of the thing my property, since matter in itself does not belong to itself.” Ph.R. para. 52; see also paras. 44 and 61. Locke expresses a similar idea in theistic language: “The earth and all that is therein is given to men for the support and comfort of their being.” The Second Treatise of Government (Indianapolis, IN: Bobbs-Merrill, 1952) at 17.

32. See F. H. Bradley, Ethical Studies, 2d ed. (London: Oxford University Press, 1962) at 5–6.

33. I leave aside here as well the question as to whether the dichotomy between “interpretation” and “metaphysical truth” is a necessary one. For an argument that it is not, see A. Brudner, The Unity of the Common Law: Studies in Hegelian Jurisprudence (Berkeley: University of California Press, 1995) at ch. 6.

34. See, in addition to the works cited in n.3, “Sharpening the Subjective Element in Criminal Liability” in R. A. Duff & N. Simmonds, eds., Philosophy and the Criminal Law (Wiesbaden: Steiner, 1984) at 79–89. See also B. Chapman, “Agency and Contingency: The Case of Criminal Attempts” (1988) 38 U.T.L.J. 355.

35. See R. A. Duff, “Acting, Trying, and Criminal Liability” in Shute et al., eds., supra note 3 at 83–88; Nagel, supra note 3 at 35. Those who identify complex acts (e.g. killing) with basic bodily movements (e.g. moving a finger) which then cause other states of affairs (e.g. the triggering of a firing mechanism, the death of the victim) fail to explain why the neurological chain of events between volition and finger-moving, or indeed between primordial volition, the imagining of a specific aim, and the decision to execute it, is not also part of the causal (as distinct from the act) portion of the complex action description; see M. S. Moore, Act and Crime (Oxford: Clarendon Press, 1993) atch. 11.

36. See Hegel’s Science of Logic, trans. W. H. Johnston & L. G. Struthers (London: Allen & Unwin, 1929) II, 375. While it is not essential to my argument to make the case for the subordination of causation to teleology (my argument is simply that the practice of imputation requires this assumption), my doctrinal prescriptions will nonetheless carry more weight if this hierarchy were at least made credible. Accordingly, I offer the following brief comments, which are based on Hegel’s Logic II, 374–80.

In order for teleology to be the truth of causation, there must be a theme common to both, and teleology must be a superior manifestation of that theme. The common theme is die mutual dependence of polarities. A cause is not a cause without an effect and vice-versa; an end is not an end without an embodiment and vice-versa. In the causal relation, however, the mutual presupposition of cause and effect is a truth extrinsic to the actual polarities, which for their own part remain external and indifferent to each other. In the teleological relation, by contrast, the mutual presupposition of end and object is a truth internal to the relation itself, since the unaccomplished agent is self-consciously incomplete and so inwardly impelled to self-realization. Thus teleology is the “truth” of external causation because the mutual dependence of polarities that is latent in external causation becomes explicit and self-conscious in teleology.

37. See The Logic of Hegel, trans. W. Wallace (Oxford: Oxford University Press, 1892) at paras. 206–09.

38. Though I distinguish here between purposes and causes, nothing in my argument turns on whether any hard-and-fast dichotomy is possible here. One may grant that a purpose can be a kind of cause and yet maintain that an interpretation of an outcome in terms of final causation (i.e. its point) subordinates to its primacy an explanation of the same outcome in terms of efficient causation. For an argument that reasons are a species of cause, see D. Davidson, Essays on Actions and Events (Oxford: Clarendon Press, 1980) at 3–19.

39. But see supra note 36.

40. Hegel, Ph.R. paras. 118, 120.

41. One might object that this makes the responsibility of the accused in Commonwealth v. Smith depend on the marksmanship of the policeman (since he is excused from manslaughter if the policeman intentionally shoots his accomplice but not if the policeman, in aiming at the accomplice, accidentally kills a passer-by); see Note, (1958) 71 Harv. L. Rev. 1566. However, this is merely one more example of the ubiquitous role of luck in criminal responsibility. Those who think that unfortunate outcomes cannot fairly be imputed to agents forget that, simply by virtue of acting in the world, free agents take responsibility for luck to an extent consistent with their freedom—which is to say, to the extent that they either foresaw or ought to have foreseen the outcome’s likelihood. Those who think that fortunate outcomes cannot consistently with moral desert be credited to an accused forget that the measure of criminal punishment (as distinct from moral blame) must be linked to specific harms and to the degree of imputation rather than to some general assessment of moral character, or else judicial punishment will not be authorized by the agent’s act and what belongs to it. As far as criminal law is concerned, no anomaly is involved in allowing the policeman’s aim to affect the robber’s responsibility. If the difference between a conviction for murder and one for assault causing bodily harm can depend on the skill of a surgeon, why cannot the difference between a conviction for manslaughter and one for robbery alone depend on the marksmanship of a policeman? For ruminations on the role of luck in assigning responsibility, see Feinberg, Doing and Deserving at 30–37; T. Nagel, Mortal Questions at 24–38; B. Williams, Moral Luck at 20–39; M. Zimmerman, “Luck and Moral Responsibility” (1987) 97 Ethics 374; A. Ripstein, “Equality, Luck, and Responsibility” (1994) 23 Phil. & Publ. Affairs 3.

42. This is what occurred in Pagett v. The Queen (1983), 76 Cr. App. R. 279.

43. Ibid.

44. It is another question whether this analysis also applies in tort law. Where the defendant has breached a private duty by imposing an unreasonable risk on a foreseeable plaintiff, and where a third party’s intentional act is just the sort of eventuality the defendant ought to have avoided for the plaintiff’s sake, then it may be that the intervention does not preclude liability. See Q. v. Minto Management 34 D.L.R. (4th) 767 (Ont. CA. 1986), in which a landlord was held liable to his tenant for negligently allowing employees access to master keys, the tenant having been raped by an employee who had used the key to gain entry into her apartment (this had occurred before). If, however, the landlord had been charged with criminal negligence causing bodily harm, our analysis would require an acquittal. Why this difference? In the criminal case, we are punishing a public wrong rather than compensating a private one. Here it is not relevant that the intentional harm is within the risk by virtue of which the accused’s act is a public wrong, because the public wrong is complete before any harm occurs. That wrong lies in the landlord’s reckless indifference to the safety of others, and so the public right is vindicated by punishing the landlord for reckless endangerment. By contrast, the private wrong of negligence does not materialize until harm is caused, and it would not make sense to treat as negating a wrong the very eventuality whose likelihood made the landlord’s conduct negligent in the first place.

45. See, for example, R. A. Duff, Intention, Agency and Criminal Liability (Oxford: Blackwell, 1990) at 99. For a critique of Duff on this score, see J. Hornsby, “On What’s Intentionally Done” in Shute et al., eds., supra note 3 at 69–73.

46. (1977), 34 C.C.C. (2d) 427 (S.C.C.).

47. [1993] 3 S.C.R. 3.

48. This rule had already become well entrenched in precedent if not in reason; see R. v. Larkin (1942), 29 Cr. App. R. 18 (C.A.); R. v. Church, [1966] 1 Q.B. 59 (C.A.); D.RP. v. Daley (1978), 69 Cr. App. R. 39 (P.C.).

49. I deal later with the objection that differences in blameworthiness for outcomes can be reflected at the sentencing stage.

50. Supra note 25.

51. Though this is what seems to have occurred in R. v. Holland (1841), 2 Mood & R. 351.

52. To the extent that only a reasonably mistaken belief in consent will absolve from sexual assault, the latter is a crime undifferentiated as between intentional, reckless, and negligent assaults.

53. See Vaillancourt v. The Queen, [1987] 2 S.C.R. 636.

54. [1990] 2 S.C.R. 633.

55. See McKinney’s New York Criminal Law Pamphlet (St. Paul, MN: West Publishing, 1991) at 487–826.

56. As McLachlin, J. argued in Creighton; see above note 47 at 48–49. See also Sopinka J.’s dissenting judgment in R. v. Daviault, [1994] 3 S.C.R. 63 at 120.

57. This argument is drawn from Alan Brudner, “Proportionality, Stigma and Discretion” (1996) 38Crim. L. Q. 302 314–17.