I am indebted to Professor Stephen Perry for comments on earlier drafts and for helpful discussions on topics which are dealt with in this paper. I would also like to thank Professor Richard Bronaugh for his comments and the law firm of Tory, Tory, DesLauriers & Binnington, which provided financial support through a J.S.D. Tory Writing Award.

1. Bohlen, Francis H. “The Moral Duty to Aid Others as a Basis of Tort Liability” (1908) 56 U. Pa. L. Rev. 217 at 219f.Google Scholar

2. Horsley et. al. v. MacLaren et. al.[1970] 2 O.R. 487 at 499, 11 D.L.R. (3d) 277 (C.A.) at 289, affd. (1972), 22 D.L.R. (3d) 545 (S.C.C.). Dorset Yacht Club v. Home Office [1970] A.C. 1004 at 1027.

3. For a list of recognized special relationships,seeProsser, William L. and Prosser, W. Page Keeton,andKeeton on Torts,5th ed.(St. Paul: West Publishing Co.,1984)s. 56,376–77;Google Scholar Linden, A.M. Canadian Tort Law, 4th ed.(Toronto: Butterworths,1988)s. 56,268.For a discussion, see infra s.II.Google Scholar

4. Mill, J.S. On Liberty (Middlesex: Penguin Books,1983) at 70–71; Google Scholar “Law and Morals”(1908) 22 HarvL. Rev.97; Bohlen, supra, note 1; H.F. McNiece & J.V. Thornton, “Affirmative Duties in Tort” (1949) 58 Yale L.J. 1272; W.M. Rudolph,“The Duty to Act: A Proposed Rule”(1965) Nebraska L. Rev. 499; J.H. Scheid, “Affirmative Duty to Aid in Emergency Situations“(1969) 3 John Marshall J. of Practice and Procedure 1; J.M. Ratcliffe, ed., The Good Samaritan and the Law (Garden City, New York: Doubleday and Co.,1966); A.M. Linden, “Rescuers andGood Samaritans” (1971) 10 Alta L. Rev. 89; Linden,supra, note 3 at 263ff.

The test formulated by Prosser and endorsed by Linden is that “knowledge of serious peril,threatening death or great bodily harm to another, which an identified defendant might avoid with little inconvenience, creates a sufficient relation to impose a duty of action”. See Prosser and Keeton, supra,note 3 at 377; Linden, supra, note 3 at 267.

Two writers who have taken positions opposing a common law duty to rescue are Benditt and Epstein. See Theodore M. Benditt, “Liability for Failing to Rescue”(1982) 1 Law and Phil. 391;Richard Epstein, “A Theory of Strict Liability”(1973) 2 J. Legal Stud. 151 at 160–89.

5. Levmore, S. “Waiting for Rescue: An Essay on the Evolution and Incentive Structure of the Law of Affirmative Obligations” (1986) 72 Virginia L. Rev. 879;Google Scholar Rodriguez, T. “I am my Brother’s Keeper: A Trend Toward Imposing a General duty on a Bystander to Assist a Person in Danger”(1985)26 Boston L. Rev. 497;Google Scholar Weinrib, E.J. “The Case for a Duty to Rescue” (1980)90 Yale L.J. 247.CrossRefGoogle Scholar

6. Soldano v. Daniels(1983) 190 Cal. Rptr. 310, 141 Cal. App. 3d 443.

7. The court held that business establishments have a duty to let goodSamaritans use the business telephone when informed of an emergency.Ibid,at 317, 141Cal. App. 3d at 453.

8. Crocker v. Sundance Northwest Resorts Ltd.(1988) 44 C.C.L.T. 225, [1988] 1 S.C.R. 1186.

9. Jordan House Ltd. v. Menow & Honsberger(1973), [1974] S.C.R. 239, 38 D.L.R. (3d) 105.

10. It should be noted however that Wilson J. did adopt as statements of the present law quotations from two academic writers, Linden and Fleming, who elsewhere have advocated a general duty to rescue. See Linden, supra note 4; Fleming, J.G. The Law of Torts,7th ed.(Agincourt: The Carswell Co. Ltd., 1987) at 135–36.Google Scholar

11. For discussions of existing good Samaritan legislation, see Feldbrugge, F.J.M. “Good and Bad Samaritans: A Comparative Survey of Criminal Provisions Concerning Failure to Rescue” (1966)Google Scholar14 Am. J. of Comp. Law 630; S. Levmore, supra, note 5; Rudzinski, “The Duty to Rescue: A Comparative Analysis” in J.M. Ratcliffe, supra, note 4 at 91; Hoffman, “Statutes Establishing a

Duty to Report Crimes or Render Assistance to Strangers: Making Apathy Criminal” (1983) 72 Kent. Law J. 827. For a discussion of the appropriateness of good Samaritan legislation, see John Kleinig, “Good Samaritanism” (1976) 5 Phil, and Pub. Aff. 382.

12. See, e.g., Levmore,supra,note 5 at 883–84;H. Fingarette, “Some Moral Aspects of Good Samaritanship”, in J.M. Ratcliffe, supra, nte 4 at 213ff; CO. Gregory, “The Good Samaritan and the Bad: the Anglo-American Law”, ibid, at 23ff.; A.M. Honored “Law, Morals and Rescue”, ibid, at 225ff.; T. Rodriguez supra, note 5 at 526.

13. Aristotle’s concept of corrective justice provides a compelling rationale for liability in tort law.The tortfeasor and the injured party are taken to be equal in all relevant respects prior to their interaction. This equality is destroyed when the wrongful act of the tortfeasor causes a loss. To achieve corrective justice, the court compels a transfer from defendant to plaintiff of resources sufficient to make up the loss suffered by the latter.

For discussions of the concept of corrective justice in tort law, see Jules L. Coleman, “MoralTheories of Torts: Their Scope and Limits: Part II” (1983) 2 Law and Philosophy 5; Richard A. Epstein,supra,note 4; George P. Fletcher, “Fairness and Utility in Tort Theory” (1972) 85 Harv. L. Re.537; Richard Posner, “The Concept of Corrective Justice in Recent Theories of Tort Law”(1981) 10 J. Legal Stud. 187; EJ. Weinrib, “Toward a Moral Theory of Negligence” (1983)2Law and Philosophy 37.

14. Izhak Englard, “The System Builders: A Critical Appraisal of Modern American Tort Theory” (1980) 9 J. of Legal Stud. 27. CrossRefGoogle Scholar

15. Weinrib,supra, note 5. Weinrib’s argument has recently been endorsed by Linden. Seesupra, note 3 at 267.

16. Weinrib,supra, note 5 at 251.

17. See Prosser & Keeton,supra,note 3.

18. Buch v. Amory Mfg. Co.(1898) 69 N.H. 257, 44 A. 809.

19. “To charge a man for not acting, you must show that it was his duty to act.” Oliver Wendall, Holmes,The Common Law(Boston: Little, Brown and Co.,1881),p.183.Google Scholar

20. Bohlen,supra note 1 at 22Iff; Fleming, supra note 10 at 137. For a discussion of the arrival of nonfeasance as a cause of action see S.F.C. Milsom, “Not Doing is no Trespass: A View of the Boundaries of Case” (1954) Cambr. L. J. 105.

21. Bohlen suggests that it was an accident of procedural history thataffirmative duties came to be classified under the law of torts at all. A more logical organization,he suggests, would have reserved tort actions for violations of negative duties and classified all affirmative duties,contractual and non-contractual, under one head. See Bohlen, supra, note 1 at 221–22.

22. [1932] A.C. 562 (H.L.).

23. The expression duty of care is also sometimes used by the courts in the context of genuine affirmative duties, as, for example, in cases involving the possible liability of public authorities.See, e.g.,Anns v. Merlon London Borough Council [1978] A.C.728, [1977] 2 W.L.R. 1024,[1977] 2 All E.R. 492 (H.L.).

24. The Canadian courts have not recognized as wide a range of specialrelationships as have the U.S.courts.

25. Farwell v. Keaton(1976) 39 Mich. 281, 240 N.W. 2d217. (Drinking companions are in a special relationship giving rise to a duty to rescue.)

26. Bohlen,supra,note 1 at 243–44; McNeice & Thornton, “Affirmative Duties in Tort” (1949) 58 Yale L. J. 1272.

27. See Levmore,supra,note 5 at 900.

28. Horsley v. MacLaren, supra, note 2, (1972) 22 D.L.R. (3d) 545 (S.C.C.).

29. Weinrib,supra,note 5.

30. Ibid, at 264. This approach to the issue bears astrong resemblance to that of John Stuart Mill.See J.S. Mill, supra, note 4 at 70–71:

To make anyone answerable for doing evil to others is the rule; to make him answerable for not preventing evil is, comparatively speaking, the exception. Yet there are many cases clear enough and grave enough to justify that exception... There are often good reasons for not holding him to the responsibility; but these reasons must arise from the special expediencies of the case: either because it is akind of case in which he is on the whole likely to act better when left to his own discretion than when controlled in any way in which society have it in their power to control him; or because the attempt to exercisecontrol would produce other evils, greater than those which it would prevent. [Emphasis added].

31. Weinrib,supra,note 5 at 275.

32. Ibid, at 267–68.

33. Dworkin calls this requirement the “gravitational force” of judicial precedent, by which he means thetendency of decided cases to influence the outcomes of later casesarising out of similar but essentially different fact situations. Dworkin argues that the phenomenon of gravitational force results from the judicial practice of deciding cases on the basis of principles designed to treat the parties before the court with fairness, which requires that like cases be treated alike, rather than on the basis of policies aimed at furthering the collective good of the larger community. See Ronald Dworkin, Taking Rights Seriously (Cambridge, Massachusetts: Harvard University Press, 1977) at 11 Iff.

34. Weinrib,supra,note 5 at 271–72.

35. Given Weinrib’s premises, it can be seen that a criticism of his argument by Benditt misses its mark.See Benditt, supra, note 4 at 413. Benditt claims that Weinrib’s argument “constitutes an argument for tort liability only if rescue must either be subject to tort duty or else relegated to the market. This is a false dichotomy, however, for it is surely possible neither to permit contracts nor to impose tort duty”. Ibid, at 413. It is precisely Weinrib’s point to show that this third prima facie possibility is not an option within a consistent common law doctrine. Given a moral duty,Weinrib claims, the only way to avoid the conclusion that there should be a legal duty is to show that enforcement ofsuch a duty oversteps the limits of law’s legitimate coercion. The argument from contractual liberty purports to show that these limits are not overstepped by a legal duty to rescue.

36. The argument also leads to a very odd result concerning the relationship between statutory restrictions on contracts and the common law. Any statutory prohibition on aparticular type of contract would create a pocket for a common law duty coercing a transfer. So, for example,the statutory prohibition on contracts for medical care between doctors and patients would bea sufficient reason to impose an affirmative tort duty on doctors to provide medical assistance for even minor ailments.

37. Weinrib,supra, note 5 at 272–73.

38. Ibid, at 279.

39. Ibid,Ibid, at 268. As his authority for this proposition, Weinrib cites Dworkin, “Liberalism”, in Public and Private Morality (S. Hampshire ed. 1978) 113 at 128. On the page cited, however, Dworkin makes no such claim and is concerned with separate issues altogether, namely, the difference between a “liberal” understandingof the role of government in promoting equality among its citizens, according to which government must be neutral on the question of what constitutes the good life in order to give eachcitizen maximum freedom to pursue his or her own vision of the good life, and a “conservative” understanding of the government’s role, which presumes that government cannot be neutral on the question ofthe good life because it cannot treat its citizens as equal human beings without a theory of whata human being ought to be. In pages following the page cited by Weinrib, Dworkin goes on to argue that the liberal conception of equality leads to the conclusion that an egalitarian distribution of goods is best achieved through some form of a free market economy, i.e., one in which there is widespreadfreedom to contract. If this is the argument Weinrib intended to draw on for support, he is no further ahead. Dworkin makes no claims about any particular conception of liberty that may underlie the common law of private obligations

40. SeeFeinberg, J. “Duties, Rights and Claims” (1966) 3 Am. Phil. Quart. 137 at 141–42.Google Scholar

41. Weinrib speaks of a victim’s right to “conscript” the services of a potential rescuer. Weinrib,supra,note 5 at 268. But tort law operates only after a loss is suffered. The right to recover derives froma right to be rescued, but only the former is enforceable since the latter only arises fleetingly during an emergency situation. The question to be decided in court is not whether the defendant shall be“conscripted” into preventing the plaintiff’s loss, but whether the defendant should be forced to bear this loss.

42. Although Weinrib has never explicitly repudiated the arguments in his essay on the duty to rescue,it appears from his later writings that changes in his understanding of the theoretical foundations of tort law have led to a reversal of his position. See “Liberty, Community and Corrective Justice” (1988) 1 Can. J. of Law and Jur. 3. On Weinrib’s later view of private law, the common law isnot a meajis for enforcing moral obligations up to the point where the evils ofcoercion preclude enforcement, but rather “a regime of right to which considerations of humanity are irrelevant.”Ibid, at 6. The intelligibility of private law dependsupon a Kantian “conception of interaction in terms of the mutual externality of the agents”, which entitles parties

to regard themselves as completely external to — and thus indifferent to — the good of each other. This mutual externality is expressed in the difference between misfeasance and non—feasance that lies at the heart of private law and that precludes the existence of a general obligation to respond to another’s need or promote another’s good. In its central cases, private law is a regime of right in which the willing of the good of another —or the failure to do so — has no indigenous significance. We may celebrate the relationships of love and friendship in literature and mark the richness of our personal lives by their presence, but the intelligibility of private law is not informed by any acknowledgement of their primacy.

Ibid, at 6. See also “The Insurance Justification of Private Law” (1985) 14 J. Legal Stud. 681 at 638.

43. See Weinrib,supra, note 5 at 253ff.

44. This view was accepted in Kelly v. Metropolitan Rwy. [1895] 1 Q.B. 944.

45. Weinrib,supra, note 5 at 254–55. For a discussion of the but–for test, see text below at page 119.

46. I will consider only five of the eight possible combinations of answers to the three questions.

47. See Jeremy Bentham, An Introduction to the Principles of Morals and Legislation (Oxford: Clarendon Press, 1892) at 76: “It has every now and then made a question, what it is in such a case that constitutes one act: where one act has ended, and another act has begun: whether what has happened has been one act or many. These questions, it is now evident, may frequently be answered, with equal propriety,in opposite ways: and if there be any occasion on which they can be answered only in one way, the answer will depend upon the nature of the occasion, and the purpose for which the question is proposed.”

48. This point has generally gone unnoticed in discussions of the relation between a person’s causal rolein the creation of the risk and her duty to aid others imperilled or injured as a result. See, e.g., Prosser, supra, note 3 at 373.

49. Weinrib,supra, note 5 at 256, note 34.

50. Ibid.

51. Weinrib cites the dissenting judgement of Freedman J.A. in Oke v. Weide Transportation Ltd. & Carra, 41 D.L.R. (2d) 53 (Man. C.A.) as evidence of judicial support for the kind of connection between causation and the misfeasance/nonfeasance distinction that Weinrib argues for. See Weinrib, supra, note 5 at 256, note 33. In reply to the argument that the position of the defendant, who caused adangerous road condition and then failed to take steps to alleviate it,was equivalent to that of any other motorist who, driving by and observing the danger, took noaction to remove the risk, Freedman J.A. stated at pp. 62–63:

Concerning this argument I have two observations to make. In the first place, no such othermotorist is before the court as defendant in this case and it is therefore unnecessary to consider what his position might have been. In the second place, even if we assume that such othermotorist would not be liable, it is wrong to think that the defendant’s position is on all fours with his. Indeed it is decidedly different. Our other motorist did not collide with the sign; the defendant did, even if it was without negligence. The former, if observing the brokensign–post at all, could do so only fleetingly, while in the act of driving by; the latterstood at the very spot, where he could see the precise results of the collision and the hazardthey created. The former, having had no part in the destruction of the sign, was never anything more than an innocent passer–by who might not be under a legal duly to take active steps to control the situation; the defendant on the other hand participated in the creation of the hazard, recognized his obligation to do something by way of rectification, and in fact took some stepsin that direction — the removal of debris, the resolve to inform the police — but then failed to go far enough. It is entirely unrealistic, in my view, to try to assimilate his position tothat of some passing motorist.

While it is true that Freedman J.A. mentions the defendant’s causal role in the creation ofthe risk as a factor pointing toward a duty to remove the risk, there is no indicationthat thisalone would have been sufficient to give rise to a duty. Other factors are mentioned aswell, in particular,the fact that the defendant “recognized his obligation to do something by way of rectification, and in fact took some steps in that direction… but thenfailed to gofar enough.” The mentioning of the defendant’s recognition of his obligation to do something suggests that knowledge of his causal role played a part in the creation of the duty. Inany case, the fact that the defendant started to take steps, but failed to go far enough brings his conduct within the rule that a person who voluntarily undertakes a rescue is liable for failure to carry the act through with reasonable care.

Weinrib also cites Rivtow Marine v. Washington Iron Works (1975), 40 D.L.R. 3d 530 (S.C.C.) as further judicial support for the validity of what he takes to be FreedmanJ.A.’s “approach to the nonfeasance issue.” See Weinrib, supra, note 5 at256, note 33. Rivtow was a case in which the defendant manufacturer of a defective item that caused economic loss was held to be beyond the reach of liability for negligence in manufacturing it, but undera duty to abate the danger thus created by warning of the defect once he became aware of it. But here, again, the defendant’s dutyarose only once he became aware of the danger, notmerelyby virtue of his causing the risk of harm.

52. Supranote 8.

53. Supranote 9.

54. For an example of such confusion in the literature, see Alexander, E.R., “One Rescuer’s Obligation to Another: The ‘Ogopogo’ Lands in the Supreme Court of Canada,” (1972) 22 U. of Tor. L. J. 98 at 103–04.Google Scholar

55. M’Mister (or Donoghue) v. Stevenson, supra, note 22 at 580.

56. Horsley v. MacLaren, supra, note 22 at 580. note 2, 22 D.L.R. (3d) 545 at 546, overturning Vanvalkenburg v. Northern Navigation Co. (1913), 30 O.L.R. 142 (S.C.A.D.).

57. See, e.g., Mill,supra, note 4 at 70–71; Epstein, supra, note 4 at 189–204; Kleinig, supra, note 11 at 391ff; Harris, J., “The Marxist Conception of Violence,” (1974) 3 Phil and Pub. Aff. 193 at 200ff.Google Scholar

58. Prosser & Keeton,supra, note 3 at 376.

59. Ibid.

60. Tarasoff v. Regents of University of California (1976) 17 Cal.3d 425 at 436–37, 131 Cal. Rptr. 14 at 23–24. This case significantly extends the common law on rescue, not just by recognizinga new type of special relationship, but also by asserting that a special relationship cangive rise to a duty to a person outside the relationship. I am aware of no Canadian cases thathave taken this step.

61. See above at p. 105 of this essay.

62. Fuller, L. & Wm. Perdue, Jr. “The Reliance Interest in Contract Damages,” (1936) 46 46 Yale L. Rev. 52 at 52–53.CrossRefGoogle Scholar

63. The issue of factual causation must be distinguished from the legal issue of proximate causation. Once a defendant’s conduct is identified as having, as a matter of fact, contributed to the plaintiff’s injury, the court will considerwhether or not the defendant’s conduct was the proximate cause of harm, i.e., whether, for reasons of policy or principle, the defendant’s causal connection should be considered so remote as to rule out legal responsibility. See RichardWright, “Causation in Tort Law” (1985) 73 Cal. L. Rev. 1737. The discussion which follows relies in a number of places upon Wright’s analysis of causation.

64. Bohlen, supra, note 1 at 220-21.

65. Mill, supra, note 4 at 70.

66. (1914), 7 B.W.C.C. 101. The civil law also recognizes that non–intervention can be a causeof injury. See A. Tunc, “The Volunteer and the Good Samaritan” in Ratcliffe, supra, note 4 at 49.

67. [1944] 2 All E.R. 350 at 356. T

68. The but–for test has also been criticized for its failure to deal with other fact patterns. For example, where two simultaneous conditions are independently sufficient to cause the injury, neither is a cause according to the but–for test. Thus, iftwo fires are set at the same time by two different people and each alone was sufficient to dothe damage sued upon, then of neither individual can it be said that, but for his starting a fire, the harm would not have occurred. The test also fails to handle a pre–emptive causation case, i.e., a case where a third party would have inflicted an injury identical to the injury inflicted by the defendant, had the defendant not done so first. For example, if two snipers are aiming at the same person and one shoots first, pre–empting the other’s act, then it cannot be said of the first that, but for his act, the victim would not have been shot. See Wright, supra, note 63 at 1775–76.

69. Hart, H.L.A. & Honoréd, A. Causation in the Law (Oxford: Oxford, 1959)Google Scholar

70. See, e.g., Holmes,supra, note 19 at 54.

71. Hart & Honore,supra, note 69 at 27.

72. Ibid, at 37, 47.

73. A similar point is made by Mack. See Mack, E. “Bad Samaritanism and the Causation of Harm” (1980) 9 Phil, and Pub. Aff. 230 at 235ff.Google Scholar

74. Wright,supra, note 63 at 1748–49.

75. Wright,supra, note 63 at 1788ff. The test was first proposed by Hart and Honor6 in Causation and the Law, supra, note 69 at 105–10, 116–19, 122, 216–19. Wright argues that their exposition of the test was distorted by their emphasis upon nonfactual, normative aspects of the ordinary language concept of causation, aspects which in a legal analysis should be treated as issues of proximate cause and tortious conduct.

76. Wright,supra, note 63 at 1790.

77. Seesupra, note 68. The duplicative causation problem is solved because each of two independently sufficient causes is a necessary element of a set of conditions sufficient to bring about the result. Seeibid, at 1791 ff. Pre–emptive causation cases are also well handled. The sniper who shoots is a factual cause of the victim’s injury since his act was a necessary element of a set of antecedent conditions that was sufficient for the injury. The second sniper, who lost his chance to shoot when the first sniper shot, is of no causal import since his behavior is not a part of any set of actual antecedent conditions that was sufficient for the injury. See ibid, at 1794ff.

78. Wright does not address the issue of whether to admit omissions ascauses.

79. For an alternative argument against counting omissions as causes, see Rizzo, Mario J. “Fundamentals of Causation”foreword to “Symposium on Causation in the Law of Torts” (1987)63 Chicago-Kent Law Review 397.Google Scholar

80. Seesupra, note 5.

81. These three criteria for liability for nonfeasance flow from the test proposed by Prosser, which is atypical proposal. See supra, note 4. The second criterion applies an objective standard to the defendant’s conduct. The issue of whether the standard should be subjective or objective is not addressed in the literature. Below I consider and reject both possibilities.

82. A number of writers have remarked on the administrative problems that could arise in dealing with multiple tortfeasors in a rescue situation. See Linden, supra, note 4 at 90; Benditt, supra,note 5 at 409’10; McNeice & Thornton, supra, note 26 at 1288; Weinrib, supra, note 5 at 262. As Weinrib points out, the problem of choosing a tortfeasor is not insurmountable in principle: the plaintiff would have the option of choosing a defendant and the doctrine of contribution would allow the initial defendant to spread the loss among others who were also responsible for the loss.

83. For a discussion of an occupier’s liability, see Linden, supra, note 3 at 599ff. The doctrine of Novis Actus Interveniens would not undermine this analysis since the injury for which liability is at issue is not the entire injury the plaintiff suffered after approaching the machine, but only the portion of the injury that the defendant could have prevented through timely assistance. Assuming that the machine was not negligently set up, application of the doctrine would mean only that the plaintiff’s act of approaching the machine would negate for legal purposes thecausal connection between the defendant’s creation of a potentially dangerous situation and the plaintiff’s immediate injury upon becoming entangled.

84. McNeice & Thornton,supra, note 26 at 1288; Weinrib,supra,note 5; Epstein,supra,note 4.

85. See Epstein,ibid, at 203;Weinrib,supra,note 5 at 267ff.

86. Isaiah Berlin defines “negative liberty” as “the area within which a man can act unobstructed by others. If I am prevented by others from doing what I could otherwise do, I am to that degree unfree”.SeeBerlin, “Two Concepts of Liberty” in Four Essays on Liberty (London: Oxford University Press,(1969) at122.Google Scholar

87. See above at pp. 106–11.

88. See above at pp. 121–22.

89. See above at pp. 122–23.

90. Surprisingly, agreement is not unanimous among writers on the subject. Epstein, apparently a moral minimalist, relies on the distinction between acts which are morally obligatory and acts which are morally praiseworthy but not obligatory. In an obscure passage in which he argues against a legal duty torescue, he seems to claim that an easily performed rescue of a seriously endangered person falls intothe latter category, at least according to conventional morality, so that failure to give aid in sucha situation is no moral wrong. See Epstein, supra, note 4 at 200–01. Benditt refers to, although he does not endorse, “a moral point of view” held by “some”,which denies a moral duty to rescue on the grounds that the only moral duties are duties which are voluntarily undertaken and duties not to cause harm to others. See Benditt, supra, note 4 at 403–04 Benditt omits to give the names of anyexponentsof this peculiar view.

91. SeeLeakey v. National Trust,[1980] 1 All E.R. 17(C.A.);Goldman v. Hargrove, [1966] 2 All E.R. 989 (H.L.); Smith v. Littlewoods, [1987] 1 All E.R. 710 (H.L.).

92. Seesupra, note 4.

93. As noted earlier, this is a view that Weinrib no longer holds. Seesupra, note 42.