March v E & MH Stramare Pty Ltd

High Court of Australia Mason C.J. Deane, Toohey, Gaudron and McHugh JJ. March v E & MH Stramare Pty Ltd [1991] HCA 12

ORDER Appeal allowed with costs. Set aside the orders of the Full Court of the Supreme Court of South Australia other than the order dismissing the cross-appeal and the order that the plaintiff pay the costs of the cross-appeal and, in lieu thereof, order that the appeal to that Court be dismissed with costs.

Like McHugh J., I would allow this appeal but my reasons for taking this course are rather different from those stated by his Honour as I do not accept that the "but for" (causa sine qua non) test ever was or now should become the exclusive test of causation in negligence cases.

The comprehensive statement of the facts contained in the reasons for judgment prepared by McHugh J. makes it unnecessary for me to do more than sketch the circumstances. The appellant sued to recover damages for personal injuries sustained when, on 15 March 1985, at or about 1.00 a.m., the offside of the front of his motor vehicle struck the nearside rear of the tray of a truck, owned by the first respondent, which was parked along the centre line of Frome Street, Adelaide. The second respondent had parked the truck in that position for the purpose of loading it with large wooden bins containing fruit and vegetables from premises in Frome Street where the first respondent carried on business as a wholesale fruit and vegetable merchant. The appellant was travelling south in the lane closest to the centre line of the road, there being three southbound lanes. The rear of the truck, with its parking and hazard lights illuminated, faced the southbound traffic and about one-half of the width of the truck projected into the lane in which the appellant was travelling.

The primary judge (Perry J.) found that the appellant was intoxicated at the time to such an extent that his ability to judge speed (including his own speed) and distance, his eye functions, his co-ordination and reaction times, and his vision while attending to the controls of the car were impaired, some substantially so. The primary judge found that, although the parking and hazard lights of the truck were illuminated, the second respondent should have appreciated that the parked vehicle might, in some circumstances, constitute a danger to oncoming vehicles. Accordingly, his Honour found that the second respondent was negligent in parking the truck in the middle of Frome Street. His Honour went on to find the appellant guilty of contributory negligence in driving when his faculties were impaired by alcohol, in driving through the preceding intersection of Frome Street and Rundle Street at a speed exceeding 60 kilometres per hour, in failing to see the truck and in failing to veer past the truck. Liability was apportioned 70 per cent against the appellant and 30 per cent against the respondents.

In allowing the appeal, the majority in the Full Court (Bollen and Prior JJ.) held that the second respondent's negligence was not causative of the appellant's injuries, the negligence of the appellant being the "real cause", to use the words of Prior J. It seems that both Bollen and Prior JJ. applied a doctrine of "last opportunity" or "last clear chance" against the appellant. On the other hand, White J., who dissented, concluded that the parking of the vehicle in the middle of the road partly caused the collision.

It has often been said that the legal concept of causation differs from philosophical and scientific notions of causation. That is because "questions of cause and consequence are not the same for law as for philosophy and science", as Windeyer J. pointed out in National Insurance Co. of New Zealand Ltd. v. Espagne [1] . In philosophy and science, the concept of causation has been developed in the context of explaining phenomena by reference to the relationship between conditions and occurrences. In law, on the other hand, problems of causation arise in the context of ascertaining or apportioning legal responsibility for a given occurrence. The law does not accept John Stuart Mill's definition of cause as the sum of the conditions which are jointly sufficient to produce it. Thus, at law, a person may be responsible for damage when his or her wrongful conduct is one of a number of conditions sufficient to produce that damage: see McLean v. Bell [2] , per Lord Wright; Sherman v. Nymboida Collieries Pty. Ltd. [3] , per Windeyer J.

1. (1961) 105 C.L.R. 569, at p. 591. 2. (1932) 147 L.T. 262, at p. 264. 3. (1963) 109 C.L.R. 580, at pp. 590-591.

Some of the confusion surrounding the legal concept of causation has been occasioned by the terminology employed in the various attempts which have been made over the years to express the principles governing the measure of damages recoverable in tort or contract. Not infrequently these endeavours have invoked the language of causation with a view to limiting liability by reference to causal considerations. Thus, it has been said that a wrongdoer is liable for consequences of his or her wrongful conduct which have been variously described as "direct" (In re Polemis and Furness, Withy & Co. [4] ), "natural and probable" (Haynes v. Harwood [5] ; Dorset Yacht Co. v. Home Office [6] ), "direct and natural" (The Edison [7] ; affd. sub nom. Liesbosch, Dredger v. Edison, S.S. (Owners) [8] ), when the wrongful act is "the proximate cause" (Yorkshire Dale Steamship Co. v. Minister of War Transport [9] ) or "the real effective cause" (Leyland Shipping Co. v. Norwich Union Fire Insurance Society [10] ).

1. [1921] 3 K.B. 560. 2. [1935] 1 K.B. 146, at p. 156. 3. [1970] A.C. 1004, at pp. 1028-1030. 4. [1932] P. 52, at pp. 62-64, 74. 5. [1933] A.C. 449. 6. [1942] A.C. 691. 7. [1918] A.C. 350, at p. 370.

Modern commentators take the view that these formulae, or at least some of them, conceal the making of value judgments or reliance on unexpressed policy reasons for refusing to allow liability to extend to the damage sustained in particular cases. The shortcomings of formulae of this kind were exposed in Overseas Tankship (U.K.) Ltd. v. Morts Dock & Engineering Co. Ltd. ( (The Wagon Mound ) [11] , where the "direct consequences" measure of damages in negligence endorsed by Polemis was discarded, and in Overseas Tankship (U.K.) Ltd. v. Miller Steamship Co. Pty. ("The Wagon Mound [No. 2]") [12] , where the criterion of reasonable foresight was applied so as to enable the plaintiffs to recover damages for the kind of loss as the reasonable man should have foreseen.

However, in Chapman v. Hearse [13] , this Court said, "the term "reasonably foreseeable" is not, in itself, a test of "causation"; it marks the limits beyond which a wrongdoer will not be held responsible for damage resulting from his wrongful act". More recently, in Mahony v. J. Kruschich (Demolitions) Pty. Ltd. [14] , the Court said:

A line marking the boundary of the damage for which a tortfeasor is liable in negligence may be drawn either because the relevant injury is not reasonably foreseeable or because the chain of causation is broken by a novus actus interveniens: M'Kew v. Holland & Hannen & Cubitts [15] . But it must be possible to draw such a line clearly before a liability for damage that would not have occurred but for the wrongful act or omission of a tortfeasor and that is reasonably foreseeable by him is treated as the result of a second tortfeasor's negligence alone: see Chapman v. Hearse [16] . Whether such a line can and should be drawn is very much a matter of fact and degree [13] . Just as Chapman v. Hearse rejected reasonable foresight as a test of causation, so M'Kew and Mahony rejected it as an exclusive criterion of responsibility. 1. (1961) 106 C.L.R. 112, at p. 122. 2. (1985) 156 C.L.R. 522, at p. 528. 3. [1970] S.C. (H.L.) 20. 4. (1961) 106 C.L.R., at pp. 124-125. 5. (1961) 106 C.L.R. 112, at p. 122.

Another fertile source of confusion in the development of a coherent legal concept of causation has been the common law defence of contributory negligence. The existence of the defence, as well as the absence of any mechanism for apportionment of liability as between a plaintiff guilty of contributory negligence and a defendant and as between co-defendants who were concurrent tortfeasors, was a potent factor in inducing courts to embrace a view of causation which assigned occurrences to a single cause. So long as contributory negligence remained a defence, the adoption of this approach was more likely to produce just results. The approach was reflected in the question: what was the "effective cause" of the injury, being "the one to which may be variously ascribed the qualities of reality, predominance, efficiency" in the words of Lord Shaw of Dunfermline: Leyland Shipping Co. [17] . Although his Lordship was speaking in the context of a claim under a policy of marine insurance, this notion of effective cause came to be applied in cases of contributory negligence by means of the "last opportunity" or "last clear chance" rule. In ordinary circumstances, the plaintiff was defeated by the defence of contributory negligence when his or her negligence was an effective cause of his or her injury, notwithstanding that the defendant's negligence was also an effective cause of that injury: Boy Andrew (Owners) v. St. Rognvald (Owners) [18] .

1. [1918] A.C., at p. 370. 2. [1948] A.C. 140.

According to the "last opportunity" or "last clear chance" rule, the plaintiff was entitled to recover, despite his or her own negligence, if the defendant had the last opportunity of avoiding the accident but failed to do so due to negligence: see the discussion of the rule in Alford v. Magee [19] . Then the defendant's negligence was the effective cause of the plaintiff's injury. The result achieved by the application of the rule was explained in the language of causation; the defendant's later negligence broke the chain of causation so that the defendant's negligence was left as the effective cause of the plaintiff's injury. And it is in that sense that, in the present case, Prior J. described the appellant's negligence as the "real cause" of his injuries. In this respect his Honour's description echoes the comments of this Court in Alford v. Magee [20] :

[T]here are cases in which there is so substantial a difference between the position of the plaintiff and the position of the defendant at the material time that (although the accident could not have happened if the plaintiff's conduct had not been negligent) it would not be fair or reasonable to regard the plaintiff as in any real sense the author of his own harm. 1. (1952) 85 C.L.R. 437, at pp. 450-464. 2. (1952) 85 C.L.R., at p. 461.

The last opportunity rule served only to confuse even further the legal concept of causation because it did not apply as between co-defendants, so that a failure by one defendant to avail himself or herself of a last opportunity to avoid the accident did not prevent the negligence of the other defendant from being the effective cause of the plaintiff's injury.

The effect of the legislation providing for apportionment of liability

The elimination of the defence of contributory negligence and the introduction by legislation (s. 27a(3) of the Wrongs Act 1936 S.A.) providing for the apportionment between tortfeasors of damages in accordance with the degree of responsibility of the parties for the damage have meant that issues of causation could be approached afresh. True it is that there are to be found, since the enactment of the legislation, statements which indicate that the courts will still identify, in some situations, one of two preconditions to a consequence resulting in injury as the effective (and sole) cause of that injury, treating the other precondition as a causa sine qua non having merely the status of an incident preceding the critical occurrence and, hence, irrelevant: see Sherman [21] , per Windeyer J.; Stapley v. Gypsum Mines Ltd. [22] , per Lord Asquith of Bishopstone (dissenting). These statements echo the remarks of Lord Wright in McLean v. Bell [23] . Obviously, in the nature of things, there will be some cases in which a court concludes that a precondition does not play such a part in the consequence that it deserves to be characterized as a cause.

1. (1963) 109 C.L.R., at pp. 590-591. 2. [1953] A.C. 663, at p. 687. 3. (1932) 147 L.T., at p. 264.

Notwithstanding these statements, the courts are no longer as constrained as they were to find a single cause for a consequence and to adopt the "effective cause" formula. These days courts readily recognize that there are concurrent and successive causes of damage on the footing that liability will be apportioned as between the wrongdoers. The modern approach is exemplified by Lord du Parcq in his speech in Grant v. Sun Shipping Co. Ltd. [24] . His Lordship, speaking with reference to the finding of the Lord Ordinary in that case that the effect of any negligence of the second defenders was broken by the later negligence of the first defenders, said [25] :

This reasoning seems to me to be akin to that which has led to frequent and determined attempts to establish the so-called "rule of the last opportunity," of which less will be heard since the decision of your Lordships' House in Boy Andrew (Owners) v. St. Rognvald (Owners). I refer especially to the opinion of Viscount Simon [26] . With the greatest respect for the Lord Ordinary's opinion I think that his reasoning is fallacious. If the negligence or breach of duty of one person is the cause of injury to another, the wrongdoer cannot in all circumstances escape liability by proving that, though he was to blame, yet but for the negligence of a third person the injured man would not have suffered the damage of which he complains. There is abundant authority for the proposition that the mere fact that a subsequent act of negligence has been the immediate cause of disaster does not exonerate the original offender. 1. [1948] A.C. 549. 2. [1948] A.C., at p. 563. 3. [1948] A.C., at p. 149.

This Court severely criticized the last opportunity rule in Alford v. Magee. In so doing, the Court noted [27] , with apparent approval, Professor Glanville Williams' comment (Joint Torts and Contributory Negligence (1951), p. 223) that the rule was introduced to mitigate the hardship caused by the existence of the defence of contributory negligence and went on to point out that the effect of the rule was to preclude negligence of the plaintiff found to be a cause of the damage from affording a good defence. Subsequently, in Chapman v. Hearse, the Court, after referring to Alford v. Magee, observed [28] :

No doubt, in many cases, the rule has been treated as if it had assumed the role of a test of causation but not, as far as we can see, on any occasion when it was of importance to distinguish between its real and what may, perhaps, be called its apparent character. This conclusion was fatal to the argument that the last opportunity rule survived the enactment of s. 27a(3) of the Wrongs Act which requires an apportionment of damages where a person has suffered damage as the result partly of his or her fault and partly of the fault of any other person or persons. That argument depended upon the proposition that the rule was devised as a test of causation so that, when applied, its effect was to brand the defendant's negligence as the sole cause of the plaintiff's injuries. That proposition, as we have seen, was rejected in Alford v. Magee. See also Teubner v. Humble [29] , per Windeyer J. In England also, the view has been taken that the rule did not survive the enactment of the apportionment legislation. In Davies v. Swan Motor Co. (Swansea) Ltd.; James, Third Party [30] , Denning L.J. offered a different reason for the conclusion: [T]he practical effect of the Act is wider than its legal effect. Previously the courts in practice sought to select, from a number of competing causes, which was the cause - the effective or predominant cause - of the damage and to reject the rest. Now the courts have regard to all the causes and apportion the damages accordingly. So the end result of the apportionment legislation was to abolish not only the defence of contributory negligence but also the last opportunity rule. 1. (1952) 85 C.L.R., at p. 452. 2. (1961) 106 C.L.R., at p. 124. 3. (1963) 108 C.L.R. 491, at p. 502. 4. [1949] 2 K.B. 291, at p. 322.

Although the rule did not in reality go to the issue of causal connexion, its operation was often described in the language of causation. Hence the abolition of the rule enabled the courts to apportion liability as between a plaintiff whose contributory negligence and a defendant whose negligence both were concurrent causes of the plaintiff's injuries, in the same way that the courts can now apportion liability between concurrent tortfeasors whose negligence materially contributes to a plaintiff's injuries. In this respect some of the obstacles which precluded the adoption of a legal approach to causation similar to that taken by philosophy and science have disappeared. But, because legal questions of causation are asked and answered with a view to allocating legal responsibility, very often on the basis of fault, an identity of approach is not possible.

Nonetheless, the law's recognition that concurrent or successive tortious acts may each amount to a cause of the injuries sustained by a plaintiff is reflected in the proposition that it is for the plaintiff to establish that his or her injuries are "caused or materially contributed to" by the defendant's wrongful conduct: Duyvelshaff v. Cathcart & Ritchie Ltd. [31] , per Gibbs J.; Tubemakers of Australia Ltd. v. Fernandez [32] , per Mason J.; Bonnington Castings Ltd. v. Wardlaw [33] , per Lord Reid; McGhee v. National Coal Board [34] . Generally speaking, that causal connexion is established if it appears that the plaintiff would not have sustained his or her injuries had the defendant not been negligent: see I.C.I.A.N.Z. Ltd. v. Murphy [35] . But, as the decision in that case illustrates, it is often extremely difficult to demonstrate what would have happened in the absence of the defendant's negligent conduct.

1. (1973) 47 A.L.J.R. 410, at p. 417; 1 A.L.R. 125, at p. 138. 2. (1976) 50 A.L.J.R. 720, at p. 724; 10 A.L.R. 303, at p. 310. 3. [1956] A.C. 613, at p. 620. 4. [1973] 1 W.L.R. 1, at pp. 4, 6, 8, 12; [1972] 3 All E.R. 1008, at pp. 1010, 1012, 1014, 1017-1018. 5. (1973) 47 A.L.J.R. 122, at pp. 127-128.

The common law tradition is that what was the cause of a particular occurrence is a question of fact which "must be determined by applying common sense to the facts of each particular case", in the words of Lord Reid: Stapley [36] . That proposition is supported by a long line of authority in the United Kingdom: Leyland Shipping Co. [37] ; Admiralty Commissioners v. S.S. Volute [38] ; Yorkshire Dale Steamship Co. [39] ; Alphacell Ltd. v. Woodward [40] ; McGhee v. National Coal Board [41] . It is supported also by this Court's decision in Fitzgerald v. Penn [42] .

1. [1953] A.C., at p. 681. 2. [1918] A.C., at pp. 363, 369-370. 3. [1922] 1 A.C. 129, at p. 144. 4. [1942] A.C., at p. 706. 5. [1972] A.C. 824, at p. 847. 6. [1973] 1 W.L.R., at pp. 5, 11; [1972] 3 All E.R., at pp. 1011, 1017. 7. (1954) 91 C.L.R. 268.

It is beyond question that in many situations the question whether Y is a consequence of X is a question of fact. And, prior to the introduction of the legislation providing for apportionment of liability, the need to identify what was the "effective cause" of the relevant damage reinforced the notion that a question of causation was one of fact and, as such, to be resolved by the application of common sense.

Commentators subdivide the issue of causation in a given case into two questions: the question of causation in fact - to be determined by the application of the "but for" test - and the further question whether a defendant is in law responsible for damage which his or her negligence has played some part in producing: see, e.g., Fleming, Law of Torts, 7th ed. (1987), pp. 172-173; Hart and Honoré, Causation in the Law, 2nd ed. (1985), p. 110. It is said that, in determining this second question, considerations of policy have a prominent part to play, as do accepted value judgments: see Fleming, p. 173. However, this approach to the issue of causation (a) places rather too much weight on the "but for" test to the exclusion of the "common sense" approach which the common law has always favoured; and (b) implies, or seems to imply, that value judgment has, or should have, no part to play in resolving causation as an issue of fact. As Dixon C.J., Fullagar and Kitto JJ. remarked in Fitzgerald v. Penn [43] "it is all ultimately a matter of common sense" and "[i]n truth the conception in question [i.e., causation] is not susceptible of reduction to a satisfactory formula" [44] .

1. (1954) 91 C.L.R., at p. 277. 2. (1954) 91 C.L.R., at p. 278.

That said, the "but for" test, applied as a negative criterion of causation, has an important role to play in the resolution of the question. So much was conceded by Dixon C.J., Fullagar and Kitto JJ. in Fitzgerald v. Penn [45] in their discussion of the unreported decision of this Court in Skewes v. Public Curator (Qld.) (6 September 1954) where A and B were driving their vehicles at excessive speeds in conditions of poor visibility so that their vehicles collided. A was on his correct side of the road, B was not. A's negligence was not causative of injury. Their Honours pointed out that, had the action been tried by a jury, it would have been correct for the judge to instruct the jury "to ask themselves the question whether they were satisfied that the collision would not have taken place with the same results if driver A had been driving at a reasonable speed". See also I.C.I.A.N.Z. Ltd. v. Murphy [46] ; Duyvelshaff v. Cathcart & Ritchie Ltd. [47] .

1. (1954) 91 C.L.R., at pp. 276-277. 2. (1973) 47 A.L.J.R., at pp. 127-128. 3. (1973) 47 A.L.J.R., at pp. 414-417, 419; 1 A.L.R., at pp. 134-135, 138, 142-143.

The commentators acknowledge that the "but for" test must be applied subject to certain qualifications. Thus, a factor which secures the presence of the plaintiff at the place where and at the time when he or she is injured is not causally connected with the injury, unless the risk of the accident occurring at that time was greater: see Hart and Honoré, at p. 122. As Windeyer J. observed in Faulkner v. Keffalinos [48] :

But for the first accident, the [plaintiff] might still have been employed by the [defendants], and therefore not where he was when the second accident happened: but lawyers must eschew this kind of "but for" or sine qua non reasoning about cause and consequence. 1. (1970) 45 A.L.J.R. 80, at p. 86.

The "but for" test gives rise to a well-known difficulty in cases where there are two or more acts or events which would each be sufficient to bring about the plaintiff's injury. The application of the test "gives the result, contrary to common sense, that neither is a cause": Winfield and Jolowicz on Tort, 13th ed. (1989), p. 134. In truth, the application of the test proves to be either inadequate or troublesome in various situations in which there are multiple acts or events leading to the plaintiff's injury: see, e.g., Chapman v. Hearse; Baker v. Willoughby [49] ; McGhee v. National Coal Board; M'Kew (to which I shall shortly refer in some detail). The cases demonstrate the lesson of experience, namely, that the test, applied as an exclusive criterion of causation, yields unacceptable results and that the results which it yields must be tempered by the making of value judgments and the infusion of policy considerations. That in itself is something of an irony because the proponents of the "but for" test have seen it as a criterion which would exclude the making of value judgments and evaluative considerations from causation analysis: see Weinrib, A Step Forward in Factual Causation, Modern Law Review, vol. 38 (1975) 518, at p. 530.

In similar fashion, the "but for" test does not provide a satisfactory answer in those cases in which a superseding cause, described as a novus actus interveniens, is said to break the chain of causation which would otherwise have resulted from an earlier wrongful act. Many examples may be given of a negligent act by A which sets the scene for a deliberate wrongful act by B who, fortuitously and on the spur of the moment, irresponsibly does something which transforms the outcome of A's conduct into something of far greater consequence, a consequence not readily foreseeable by A. In such a situation, A's act is not a cause of that consequence, though it was an essential condition of it. No doubt the explanation is that the voluntary intervention of B is, in the ultimate analysis, the true cause, A's act being no more than an antecedent condition not amounting to a cause. But this explanation is not a vindication of the adequacy of the "but for" test.

The facts of, and the decision in, M'Kew [50] illustrate the same deficiency in the test. The plaintiff would not have sustained his ultimate injury but for the defendant's negligence causing the earlier injury to his left leg. His subsequent action in attempting to descend a steep staircase without a handrail in the normal manner and without adult assistance resulted in a severe fracture of his ankle. This action was adjudged to be unreasonable and to sever the chain of causation. The decision may be explained by reference to a value judgment that it would be unjust to hold the defendant legally responsible for an injury which, though it could be traced back to the defendant's wrongful conduct, was the immediate result of unreasonable action on the part of the plaintiff. But in truth the decision proceeded from a conclusion that the plaintiff's injury was the consequence of his independent and unreasonable action.

The fact that the intervening action is deliberate or voluntary does not necessarily mean that the plaintiff's injuries are not a consequence of the defendant's negligent conduct. In some situations a defendant may come under a duty of care not to expose the plaintiff to a risk of injury arising from deliberate or voluntary conduct or even to guard against that risk: seeChomentowski v. Red Garter Restaurant Ltd. [51] . To deny recovery in these situations because the intervening action is deliberate or voluntary would be to deprive the duty of any content.

It has been said that the fact that the intervening action was foreseeable does not mean that the negligent defendant is liable for damage which results from the intervening action: see Chapman v. Hearse [52] ; M'Kew [53] ; Caterson v. Commissioner of Railways [54] . But it is otherwise if the intervening action was in the ordinary course of things the very kind of thing likely to happen as a result of the defendant's negligence. In Dorset Yacht [55] , Lord Reid observed:

But if the intervening action was likely to happen I do not think that it can matter whether that action was innocent or tortious or criminal. Unfortunately, tortious or criminal action by a third party is often the "very kind of thing" which is likely to happen as a result of the wrongful or careless act of the defendant. Much the same approach was adopted by this Court in Caterson where Gibbs J. [54] (with whom Barwick C.J., Menzies and Stephen JJ. agreed) pointed out that, if the plaintiff's action in jumping from the train was, in the ordinary course of things, the very kind of thing likely to happen as a result of the defendant's negligence and was not unreasonable, the jury was entitled to find that the plaintiff's injuries were caused by the defendant's negligence. The finding that the plaintiff's action was not unreasonable was then essential to that conclusion because contributory negligence was a defence in New South Wales at the relevant time. See also Chapman v. Hearse [56] ; and note the reference in Mahony [57] , to the acceptance by Gibbs J. in Dillingham Constructions Pty. Ltd. v. Steel Mains Pty. Ltd. [58] , of the suggestion that, if a pedestrian were run over by two drivers consecutively and both were negligent, the injuries caused by the second driver would be damage for which both drivers were liable if those injuries were also the foreseeable consequence of the first driver's negligence. 1. (1961) 106 C.L.R., at p. 122. 2. [1970] S.C. (H.L.), at p. 25. 3. (1973) 128 C.L.R. 99, at p. 110. 4. [1970] A.C., at p. 1030. 5. (1973) 128 C.L.R. 99, at p. 110. 6. (1961) 106 C.L.R., at pp. 124-125. 7. (1985) 156 C.L.R., at p. 529. 8. (1975) 132 C.L.R. 323, at pp. 329-330.

As a matter of both logic and common sense, it makes no sense to regard the negligence of the plaintiff or a third party as a superseding cause or novus actus interveniens when the defendant's wrongful conduct has generated the very risk of injury resulting from the negligence of the plaintiff or a third party and that injury occurs in the ordinary course of things. In such a situation, the defendant's negligence satisfies the "but for" test and is properly to be regarded as a cause of the consequence because there is no reason in common sense, logic or policy for refusing to so regard it.

Viewed in this light, the respondents' negligence was a cause of the accident and of the appellant's injuries. The second respondent's wrongful act in parking the truck in the middle of the road created a situation of danger, the risk being that a careless driver would act in the way that the appellant acted. The purpose of imposing the common law duty on the second respondent was to protect motorists from the very risk of injury that befell the appellant. In these circumstances, the respondents' negligence was a continuing cause of the accident. The chain of causation was not broken by a novus actus. Nor was it terminated because the risk of injury was not foreseeable; on the contrary, it was plainly foreseeable.

The detailed facts are set out in the judgment of McHugh J. In summary, the appellant sustained serious injury when the car which he was driving ran into a truck which the second respondent had parked in a position where it was straddling the centre line of a six-lane road in central Adelaide. The accident occurred at night and at a time when traffic on the road was free-moving. At the time, the appellant was very much under the influence of alcohol. Obviously, his own negligence in the driving of his vehicle was the major cause of the accident. It is not disputed that, if the second respondent is liable in damages, the first respondent, which was the owner of the truck and the second respondent's employer, is jointly liable.

At first instance, the trial judge held that the second respondent had owed a duty of care to other users of the road, including inattentive and intoxicated drivers, which he had breached by parking the truck where it was parked. His Honour concluded that responsibility for the accident should be apportioned, pursuant to the Wrongs Act 1936 S.A., as to 70 per cent to the appellant and as to 30 per cent to the respondents. On appeal, a majority of the Full Court of the Supreme Court of South Australia (Bollen and Prior JJ., White J. dissenting) held that, while the second respondent had been negligent in parking and leaving the truck where it was, that negligence had not relevantly caused the accident. The sole effective cause of the accident was, in their Honours' view, the appellant's own negligence. That being so, the judgment and orders of the learned primary judge were set aside and, in lieu thereof, it was ordered that the appellant's action be dismissed. The appellant appeals to this Court from the judgment and orders of the Full Court.

It is clear that the second respondent was in a relationship of proximity with other users of the road on which he left the truck. That relationship gave rise to a duty to take reasonable care to avoid foreseeable injury to such other road users. That relationship and that duty of care were not confined to persons who were careful and sober but extended to all foreseeable users of the road, including bad and inattentive drivers and those whose faculties were impaired either naturally or by reason of the effect of alcohol. In that regard, as was pointed out in Bus v. Sydney County Council [59] , "the law has progressed" in recent years "by placing an increased emphasis upon the relevance of the possibility of negligence or inadvertence on the part of the person to whom a duty of care is owed". The progression of the law in that regard can, to a significant extent, be traced to the effect of apportionment legislation which precludes the approach that causative negligence on the part of a plaintiff forecloses the answer to any question of liability on the part of the defendant. Thus, for example, an inattentive driver of a slowly travelling car has been held to be liable in damages, under apportionment legislation, for breach of a duty of care owed to a pedestrian who, dressed in a dark overcoat on a rainy night, sought to cross the road at a "highly unlikely" spot without any real lookout for oncoming traffic: see Teubner v. Humble [60] .

1. (1989) 167 C.L.R. 78, at p. 90. 2. (1963) 108 C.L.R. 491.

There could be circumstances in which the parking and leaving of a truck at night in a position where it significantly obstructs each of the two centre lanes of a six-lane city road would not constitute a breach of the duty of care owed to other users of the road. An example of such circumstances is a case where the road had badly subsided and a well-lit truck constituted a temporary means of lessening the risk of injury to other road users. It is not suggested in the present case, however, that there was any legitimate reason for the truck to be left where it was. Notwithstanding the lights upon it, it represented a hazard to other motorists. In leaving it (without justifying reason) in what White J. in the Full Court accurately described as an "extraordinary position in the middle" of the roadway where it "obstructed not only one half of the right lane for south-bound traffic in which the [appellant] was travelling but also one half of the right lane for north-bound traffic" (March v. Stramare (E. & M. M.) Pty. Ltd. [61] ), the second respondent was guilty of a breach of the duty of care which he owed to other road users. It is true that it is unlikely that a careful and attentive driver would have failed to see the lights of the parked truck and to have taken steps to avoid it. As has been said, however, the duty of care owed by the second respondent was also to the inattentive (and intoxicated) driver whose presence upon the road was obviously foreseeable.

Once it is accepted that the relevant duty of care in the present case was to all users of the road, including the inattentive and those whose faculties were impaired by alcohol, it seems to me to be apparent that the appeal must succeed. In the context of apportionment legislation, the appellant's own contributory negligence did not preclude him from recovering damages for the respondents' negligence. Nor did the appellant's own negligence take him outside the classes of person to whom the respondents owed a duty of care which the second respondent breached when he parked and left the truck in the centre of the road. To the contrary, the appellant's intoxication and associated carelessness took him within the class of inattentive drivers to whom the truck represented the greatest hazard. It is true that the situation which resulted from the leaving of the truck in the middle of the roadway was a static one and that the accident would not have occurred were it not for the negligent driving of the appellant. The case is not, however, one in which an injured plaintiff became aware of the danger created by a static situation and consciously decided to take the risk: see, e.g., Anglo-Newfoundland Development Co. v. Pacific Steam Navigation Co. [62] ; and cf., per Wells J., Kammerman v. Baster [63] . The case is one in which there was fault on both sides and in which, in the context of apportionment legislation, the accident must be seen as the result not only of the negligence of the appellant in driving his own vehicle but also of the negligence of the second respondent in parking the truck in breach of the duty of care which he owed a class of persons of which the appellant was a member. Expressed in terms of causative fault, the effective causes of the appellant's injuries were the negligence of the second respondent in creating a hazard for a careless and inattentive driver and the negligence of the appellant in being such a driver. In these circumstances, it is not possible to isolate the fault of either the appellant or the second respondent as the "sole" cause of the appellant's injuries. The injuries were caused by the fault of both and, that being so, the case called for apportionment of responsibility and reduction of damages pursuant to the provisions of s. 27a(3) of the Wrongs Act. In the context of such an apportionment provision, it is unnecessary and unhelpful to seek further refinement by reference to catchcries of "last opportunity" or "direct cause": see Chapman v. Hearse [4] ; Teubner v. Humble [5] ; Harvey v. Road Haulage Executive [6] ; Fitzgerald v. Penn [7] .

1. [1924] A.C. 406, at p. 420. 2. (1981) 28 S.A.S.R. 571, at pp. 581-582. 3. (1961) 106 C.L.R. 112, at pp. 122-124. 4. (1963) 108 C.L.R., at p. 502. 5. [1952] 1 K.B. 120, at p. 126. 6. (1954) 91 C.L.R. 268, at p. 284.

There are two further matters which I would mention. The first is that I do not subscribe to the view that, under apportionment legislation, causation is automatically established if a negative answer is given to the question whether the plaintiff's injuries would have been sustained "but for" the negligence of the defendant. Causation in the context of the elements of the tort of negligence is not the same thing as the "scientific term descriptive of sequence in physical phenomena" (Clerk & Lindsell on Torts, 16th ed. (1989), par. 1-103; and see, generally, the various papers in "Symposium on Causation in the Law of Torts", Chicago-Kent Law Review, vol. 63 (1987) esp. Professor Wright, "The Efficiency Theory of Causation and Responsibility: Unscientific Formalism and False Semantics", pp. 553ff.). For the purposes of the law of negligence, the question of causation arises in the context of the attribution of fault or responsibility: whether an identified negligent act or omission of the defendant was so connected with the plaintiff's loss or injury that, as a matter of ordinary common sense and experience, it should be regarded as a cause of it (cf. Barnes v. Hay [8] ). The "but for" (or "causa sine qua non") test may well be a useful aid in determining whether something is properly to be seen as an effective cause of something else in that sense. In particular, the test will commonly exclude causation for the purposes of the law of negligence if the answer to the question it poses is that the accident which caused the injuries would have occurred in the same way and with the same consequences in any event: see, e.g., Duyvelshaff v. Cathcart & Ritchie Ltd. [9] . There are however, in my view, convincing reasons precluding its adoption as a comprehensive definitive test of causation in the law of negligence. First, the clear weight of authority is against the substitution of such a formularized test of causation for a "common sense idea of what is meant by saying that one fact is a cause of another": see, e.g., Fitzgerald v. Penn [10] ; National Insurance Co. of New Zealand Ltd. v. Espagne [11] ; Stapley v. Gypsum Mines Ltd. [12] . Secondly, unqualified acceptance of the "but for" test as even a negative or exclusionary test of causation for the purposes of the law of negligence would lead to the absurd and unjust position that there was no "cause" of an injury in any case where there were present two independent and sufficient causes of the accident in which the injury was sustained. Less importantly, acceptance of the "but for" test as a comprehensive test would carry with it the need to draw somewhat artificial distinctions to avoid the type of confusion between an accident which happened and one which did not which is to be seen in the discussion in Fitzgerald v. Penn [13] of the unreported case of Skewes v. Public Curator (Qld.) (6 September 1954). Thirdly, the mere fact that something constitutes an essential condition (in the "but for" sense) of an occurrence does not mean that, for the purposes of ascribing responsibility or fault, it is properly to be seen as a "cause" of that occurrence as a matter of either ordinary language or common sense. Thus, it could not, as a matter of ordinary language, be said that the fact that a person had a head was a "cause" of his being decapitated by a negligently wielded sword notwithstanding that possession of a head is an essential precondition of decapitation. Again, the mere fact that a person makes a gift of money to another is not, in any real sense, a "cause" of the damage sustained by that other person when his agent negligently loses the money notwithstanding that the loss would not have occurred "but for" the original gift. As Lord Reid pointed out in Stapley [14] :

The question [of "what caused an accident from the point of view of legal liability"] must be determined by applying common sense to the facts of each particular case. One may find that as a matter of history several people have been at fault and that if any one of them had acted properly the accident would not have happened, but that does not mean that the accident must be regarded as having been caused by the faults of all of them. One must discriminate between those faults which must be discarded as being too remote and those which must not. Sometimes it is proper to discard all but one and to regard that one as the sole cause, but in other cases it is proper to regard two or more as having jointly caused the accident. I doubt whether any test can be applied generally. 1. (1988) 12 N.S.W.L.R. 337, at p. 339. 2. (1973) 47 A.L.J.R. 410, at pp. 414-417, 419; 1 A.L.R. 125, at pp. 134-135, 138, 142-143. 3. (1954) 91 C.L.R., at p. 277. 4. (1961) 105 C.L.R. 569, at pp. 590-592. 5. [1953] A.C. 663, at pp. 681-682. 6. (1954) 91 C.L.R., at pp. 276-277. 7. [1953] A.C., at p. 681.

It is true that, in the context of apportionment legislation which gives the latitude necessary to enable the relief to be fairly adjusted to fit the circumstances, the courts will be unlikely to deny causation in any case where the fault of a defendant contributed to an accident. Nonetheless, the question whether conduct is a "cause" of injury remains to be determined by a value judgment involving ordinary notions of language and common sense.

The other further matter is that it should be apparent that nothing in what is written above should be read as indicating a view that a plaintiff is entitled to recover compensation under apportionment legislation in circumstances where his or her own negligence was, as a matter of ordinary common sense, the sole real cause of the accident. Even under apportionment legislation, it is an element of the tort of negligence that the injury sustained by the plaintiff be caused by the defendant's breach of duty. In a case where, as a matter of ordinary common sense, the "sole" cause of the plaintiff's injury was his or her own negligence, that element of the tort will be lacking.

The appeal should be allowed. The orders of the Full Court of the Supreme Court of South Australia should be set aside and in lieu thereof it should be ordered that the appeal to that court be dismissed with costs.

I would allow the appeal and, generally, for the reasons appearing in the judgment of Mason C.J.

In particular, I share the Chief Justice's view that the "but for" or "causa sine qua non" test is not and should not be a definitive test of causation where negligence is alleged. The limitations of the test, particularly where there are two or more acts or events, each of which would be sufficient to bring about the plaintiff's injury, or where a defendant seeks to rely upon a "supervening cause" or "novus actus interveniens", are apparent.

Where negligence is in issue, causation is essentially a question of fact, in the sense explained by the Chief Justice, into which considerations of policy and value judgments necessarily enter.

In the present case, once the majority in the Full Court held, as they did, that the respondents were in breach of a duty of care which they owed to road users, including the appellant, a finding of liability against the respondents was, in the circumstances, inevitable. The only question then remaining is apportionment, having regard to the appellant's "share in the responsibility for the damage": Wrongs Act 1936 S.A., s. 27a(3). There is no justification for disturbing the apportionment determined by the trial judge.

It follows that the appeal should be allowed and the judgment of the trial judge restored.

I agree with the judgments of Mason C.J. and of Deane J. There is nothing that I can usefully add.

The plaintiff in an action for damages for personal injuries appeals against an order of the Full Court of the Supreme Court of South Australia which set aside a verdict in his favour and entered judgment in the action for the defendants. The question in the appeal is whether the Full Court erred in finding that the conduct of the defendants in parking a truck in the middle of a street was not a legally operative cause of the damage suffered by the plaintiff when his motor vehicle collided with the rear of the truck.

The collision occurred at about 1.00 a.m. on Friday, 15 March 1985, in Frome Street, Adelaide, a little south of the intersection of that street with Rundle Street. The plaintiff was driving south in Frome Street at a speed in excess of 60 kilometres per hour. He was in the lane nearest to the centre of the road. A continuous white line ran down the centre of the bitumen carriageway. Broken white lines defined three traffic lanes on each side of the centre line. The lanes were approximately 2.6 to 3 metres in width. The second defendant had parked the truck, which was owned by the first defendant, along the centre line. The rear of the truck faced southbound traffic. About one-half of the truck was in the lane along which the plaintiff was travelling. The point of impact was 37.7 metres south of the southern building alignment of Rundle Street. The offside front of the plaintiff's car struck the nearside rear of the tray of the truck. The force of the collision shifted the truck about 3 metres.

The first defendant carried on business as a wholesale fruit and vegetable merchant in a market building which was situated on the eastern side of Frome Street about 35 metres south of the building alignment of Rundle Street. The market building had three archways which opened onto the footpath of Frome Street. The collision occurred at a point in Frome Street opposite the nearest archway to Rundle Street. The second defendant had parked the truck which had a tray-top in the middle of the street for the purpose of loading it with large wooden bins containing bulk fruit and vegetables. The tray was about 7.1 metres in length and stood 1.3 metres above ground level. It was the practice of the second defendant to load the truck by using a forklift. Bins of produce, awaiting loading, were stacked along the eastern footpath and in the traffic lane nearest to the eastern kerb. The middle of the street was the most convenient place from which to load the truck. The second defendant had parked the truck there at night for that purpose for as long as he could remember.

At the time of the collision the plaintiff was intoxicated. The trial judge, Perry J., found that it was unlikely that the plaintiff's blood alcohol concentration was less than 0.18 grams per 100 millilitres at the time of the accident and that it substantially impaired his ability to judge speed and distance, his eye functions, and his co-ordination and reaction times. His Honour also found that the plaintiff's ability to appreciate his own speed and his ability to see moving objects and to maintain adequate vision while attending to the controls of the car were impaired.

Frome Street was lit by four mercury vapour streetlights, two on each side of the roadway, at the time of the collision. Some light came onto the street from a building, a little south of the scene of the accident on the western side of the street. Illumination also came from lights in the aisle leading from the most northern archway of the market building. They created "a "spill" of light through that archway onto the footpath". In addition, two spotlights mounted on either side of the archway faced onto Frome Street. The trial judge found that Frome Street was "moderately well illuminated".

His Honour also found that the parking and hazard lights of the truck were illuminated. The parking lights consisted of red lights in each of the tail-light assemblies at the rear of the truck, separate single red lights mounted on each rear corner of the tray, and three clearance lights down each side of the truck at the level of the tray, each of which had a red lens to the rear and a yellow lens to the front. The rear lights were visible to southbound traffic. Few, if any, of the lights down the side of the truck, however, would have been visible to a driver approaching the rear of the truck. In addition to the red lights, two orange hazard lights were mounted on either side of the rear of the truck. Yellow hazard lights were also mounted on each door of the cabin of the truck and on the front corners of the cabin. The orange and yellow hazard lights flashed on and off. Those at the rear of the truck were visible to approaching traffic.

The conclusions of the trial judge and the Full Court

The trial judge found that, although the truck was lit with parking and hazard lights, the second defendant should have foreseen that, in certain circumstances, even a well-lit vehicle might provide a danger to other traffic. In determining the content of the duty of care which was owed to road users including the plaintiff, his Honour noted that the duty of care was owed not only to careful drivers but also to careless and drunken drivers. Consequently, his Honour found that the second defendant was negligent in parking the truck where he did. The first defendant was held vicariously responsible for that negligent conduct of the second defendant. The learned judge also found that the plaintiff was guilty of contributory negligence in driving while his faculties were impaired by alcohol, in driving through the intersection of Frome Street and Rundle Street at a speed in excess of 60 kilometres per hour, in failing to see the truck and the lights illuminating it, and in failing to veer so as to pass the truck safely. His Honour apportioned liability against the plaintiff 70 per cent and against the defendants 30 per cent.

All members of the Full Court accepted that the defendants were in breach of the duty of care which they owed to the plaintiff. However, Bollen J. said that the findings of the trial judge in relation to the plaintiff's conduct:

[connote] a failure to co-ordinate speed and lookout. I cannot reconcile that entirely correct finding with the view that the ranking of the truck in the centre of the road with its lights, including flashing hazard lights illuminated, and which must have been visible from afar off, was causative negligence on the part of the second [defendant]. With all respect I must differ from the learned trial judge.

The truck was an obstruction. But it was there clearly to have been seen.

Prior J. said: Legal responsibility attaches to the "real cause" of the plaintiff's injuries as distinct from a cause without which the accident would not have happened. In this case, without the vehicle being left there the accident would not have happened, but that matter is one in this particular case that does not attract legal responsibility. The use of the truck in the middle of the road is "merely an incident which precedes in the history or narrative of events": see National Insurance Co. of New Zealand Ltd. v. Espagne [15] , per Windeyer J. His Honour also said: In my view there was not a sufficiently close or direct relationship between the acts of the defendants and the injuries sustained by the plaintiff to justify the finding made. Causal proximity was not established. 1. (1961) 105 C.L.R. 569, at p. 593.

Thus Bollen J. appears to have found that the plaintiff's negligence was the sole cause of the accident while Prior J. found that, although the defendants' breach of duty was a cause of the accident, it was not the "real cause" of the accident. Both judges seem to have applied a doctrine of "last opportunity" against the plaintiff so as to render the breach of duty by the defendants inoperative.

The confusion and distractions which the defendants created on both sides of [the plaintiff's] path ahead, more particularly to his left, could fairly be said to have caused in part the ultimate occurrence of the collision. White J. would have dismissed the appeal against the trial judge's orders.

At common law, a plaintiff could not recover damages for injury which was the result of the negligent conduct of the plaintiff and the defendant. The plaintiff's contributory negligence was an absolute bar to the recovery of damages. But in South Australia, as in many other jurisdictions in the common law world, legislation now provides that, where a person suffers damage as the result partly of his or her own fault and partly of the fault of another person or persons, a claim in respect of that damage is not to be defeated by reason of the fault of the person suffering the damage. Instead, the damages recoverable are to be reduced to such extent as the court thinks just and equitable having regard to the claimant's share in the responsibility for the damage: Wrongs Act 1936 S.A., s. 27a(3). Notwithstanding the enactment of legislation of this kind, however, the courts have held that, in some situations where the defendant has been negligent and the plaintiff's damage would not have occurred but for that negligence, the plaintiff's negligence will prevent him or her from recovering damages. Thus, the courts have held that a plaintiff's action will fail if his or her negligence can be categorized as the sole cause of the damage: Stapley v. Gypsum Mines Ltd. [16] ; Sherman v. Nymboida Collieries Pty. Ltd. [17] ; Nicol v. Allyacht Spars Pty. Ltd. [18] . In Stapley [19] , Lord Asquith of Bishopstone said:

it is still part of the law of this country that two causes may both be necessary pre-conditions of a particular result - damage to X - yet the one may, if the facts justify that conclusion, be treated as the real, substantial, direct or effective cause, and the other dismissed as at best a causa sine qua non and ignored for purposes of legal liability. In this context: "causa sine qua non" means a cause which does not, in the sense material to the particular case, cause, but is merely an incident which precedes in the history or narrative of events, but as a cause is not in at the death, and hence is irrelevant. (Smith, Hogg & Co. v. Black Sea and Baltic General Insurance Co. [20] , per Lord Wright.) 1. [1953] A.C. 663, at pp. 675-676, 681, 683-684, 687-688. 2. (1963) 109 C.L.R. 580, at p. 591. 3. (1987) 163 C.L.R. 611, at pp. 618, 620. 4. [1953] A.C., at p. 687. 5. [1940] A.C. 997, at p. 1003.

This type of reasoning provoked Windeyer J. in National Insurance Co. of New Zealand Ltd. v. Espagne [21] , to ask:

How can one factor be logically more efficacious than another in producing a result for which both must exist? No doubt Lord Asquith of Bishopstone would answer that rhetorical question by saying that the common law doctrine of causation is able to ignore some factors which are necessary to the production of a result because the rationale of that doctrine is the allocation of legal responsibility rather than the determination of what has happened. 1. (1961) 105 C.L.R., at p. 593.

In Thom (or Simpson) v. Sinclair [22] , Viscount Haldane pointed out that in "strict logic the cause cannot be pronounced to be less than the sum of the entire conditions". This statement was derived from John Stuart Mill's theory that the cause of an event is the sum of the conditions which are jointly sufficient to produce it. But since the act or omission of a person charged with a wrongful act will be one only of the set of conditions or relations sufficient to produce the damage which gives rise to the proceedings, the common law has been forced to reject the application of scientific and philosophical theories of causation. They provide no answer to the question whether the person charged should be held legally responsible for the act or omission which constitutes his or her negligence. Lawyers, and particularly academic lawyers, however, have modified Mill's theory of causation and adapted it for legal purposes. The adaptation of Mill's theory holds that every necessary member of the set of conditions or relations which is sufficient to produce the relevant damage is a cause of that damage: International Encyclopedia of Comparative Law, vol. XI, (1983), Ch. 7, at p. 27; Prosser, Law of Torts, 4th ed. (1971), p. 237; Fleming, Law of Torts, 7th ed. (1987), p. 173; Nader v. Urban Transit Authority of (N.S.W.) [23] . Hence, for the purposes of the common law, a person may be causally responsible for damage even though his or her act or omission was one only of the conditions or relations sufficient to produce the damage. This is the basis of the "but for" test of causation which, apart from some exceptional cases, most writers and judges agree is the threshold test for determining whether a particular act or omission qualifies as a cause of the damage sustained. If the damage would have occurred notwithstanding the negligent act or omission, the act or omission is not a cause of the damage and there is no legal liability for it: Duyvelshaff v. Cathcart & Ritchie Ltd. [24] .

1. [1917] A.C. 127, at p. 135. 2. (1985) 2 N.S.W.L.R. 501, at p. 531. 3. (1973) 47 A.L.J.R. 410; 1 A.L.R. 125.

However, as I have already indicated, a powerful school of opinion asserts that the fact that a person's act or omission was a necessary condition of the occurrence of the damage is not itself sufficient to make that act or omission a legal cause of the damage. This school of opinion asserts that, to be a legal cause of damage, the act or omission charged must not only have been a sine qua non of its occurrence, but it must also have been a cause according to "common sense principles". This school of opinion has always had strong judicial support: see, for example, Leyland Shipping Co. v. Norwich Union Fire Insurance Society [25] ; Fitzgerald v. Penn [26] ; Stapley [27] . The view that the notion of "cause" in everyday speech and for legal purposes means more than a necessary condition or causa sine qua non also has the powerful support of Hart and Honoré in the influential textbook, Causation in the Law, 2nd ed. (1985).

1. [1918] A.C. 350, at p. 362. 2. (1954) 91 C.L.R. 268, at p. 277. 3. [1953] A.C., at p. 681.

But when the damage suffered by a plaintiff would not have occurred but for negligence on the part of both the plaintiff and the defendant, a conclusion that the defendant's negligence was not a cause of the damage cannot be based on logic or be the product of the application of a scientific or philosophical theory of causation. It has to be based upon a rule that enables the tribunal of fact to make a value judgment that in the circumstances legal responsibility did not attach to the defendant even though his or her act or omission was a necessary precondition of the occurrence of the damage.

Whatever label is given to such a rule - "common sense principles", "foreseeabilty", "novus actus interveniens", "effective cause", "real and efficient cause", "direct cause", "proximate cause" and so on - the reality is that such a limiting rule is the product of a policy choice that legal liability is not to attach to an act or omission which is outside the scope of that rule even though the act or omission was a necessary precondition of the occurrence of damage to the plaintiff. That is to say, such a rule is concerned only with the question whether a person should be held responsible for an act or omission which ex hypothesi was necessarily one of the sum of conditions or relations which produced the damage.

That a policy choice is involved in the use of some rules which limit liability for wrongful acts and omissions is obvious. Thus, the rule that a defendant is only legally liable for damage "of such a kind as the reasonable man should have foreseen" (Overseas Tankship (U.K.) Ltd. v. Morts Dock & Engineering Co. Ltd. (The Wagon Mound) [28] ) is clearly a rule of policy. The rule that a defendant is not liable for damage which has been brought about by an overwhelming supervening event ("a novus actus interveniens") is also a rule of policy despite the use of metaphors such as "snapping the causal chain" to explain it. It may be less obvious, but it is still true, that the other labels to which I have referred are also the products of policy choices. In Fitzgerald [29] Taylor J. said that:

expressions such as "effective cause", "direct cause", "real cause", "decisive cause" and "proximate cause", amongst others, have been seized upon, not for the purpose of excluding factors which might be thought to constitute causes in the widest philosophical sense, but for the purpose of placing the defendant's ultimate responsibility upon a practical commonsense basis. Nevertheless, the use of commonsense notions of causation and the use of expressions such as those to which Taylor J. referred require the application of a policy choice, for they allow the tribunal of fact to determine legal liability on broad grounds of moral responsibility for the damage which has occurred. 1. [1961] A.C. 388, at p. 426. 2. (1954) 91 C.L.R., at, pp. 284-285.

Despite the numerous judicial statements to the effect that being a causa sine qua non of the occurrence of damage is not enough to make an act or omission a cause of damage for legal purposes, another school of opinion asserts "that the only genuine causal issue is that of sine qua non or "cause in fact" ": see Hart and Honoré, at p. lxvii, who reject this view. Many, probably most, academic legal writers on the subject assert that, once the "but for" or causa sine qua non test is satisfied, the issue of causation is spent. Other issues such as "proximate cause", "sole cause", "novus actus interveniens", "real cause" and so on are the product of policy choices whose objects are to limit legal responsibility for damage resulting from acts or omissions of the kind in question. Accordingly, on this view, if a negligent act or omission materially contributed to the occurrence of damage, it will still be a cause of the damage for legal purposes even though, for policy reasons, there may be no legal liability for that act or omission. Moreover, a negligent act or omission will be held to have materially contributed to the damage if its contribution was not de minimis: Bonnington Castings Ltd. v. Wardlaw [30] .

In Alexander v. Cambridge Credit Corporation Ltd. [31] , Glass J.A. held that the "but for" test applied in a practical commonsense way was the test for determining whether a causal connexion existed between breach of duty and damage: see also my judgment [32] . Alexander was a case of contract, but in that case both Glass J.A. applied what I had said in Nader [33] , on the subject of causation. Nader was a case of tort. But further reflection on the matter has persuaded me that, if the "but for" test is applied in a "practical commonsense way", it enables the tribunal of fact, consciously or unconsciously, to give effect to value judgments concerning responsibility for the damage. If the "but for" test is applied in that way, it gives the tribunal an unfettered discretion to ignore a condition or relation which was in fact a precondition of the occurrence of the damage.

1. (1987) 9 N.S.W.L.R. 310, at p. 315. 2. (1987) 9 N.S.W.L.R., at p. 358. 3. (1985) 2 N.S.W.L.R., at pp. 530-531.

Moreover, it is doubtful whether there is any consistent commonsense notion of what constitutes a "cause". As Dr Lloyd-Bostock has asserted in "The Ordinary Man, and the Psychology of Attributing Causes and Responsibility", Modern Law Review, vol. 42 (1979) 143, at p. 167, modern psychological research:

illustrates how judgments of causes and responsibility [by an ordinary person] are reached by an active, constructive process which goes beyond the information given and is therefore subject to various forms of error and bias: are structured by as well as expressed in language: and are influenced by the motives, values, experiences, and other characteristics of the judger, the specific context, and the anticipated consequences. These various effects are interwoven and difficult to disentangle both conceptually and empirically, but in general talking about the ordinary man, common sense, and everyday judgments appears somewhat hazardous. Indeed, I suspect that what commonsense would not see as a cause in a non-litigious context will frequently be seen as a cause, according to commonsense notions, in a litigious context. This is particularly so in many cases where expert evidence is called to explain a connexion between an act or omission and the occurrence of damage. In these cases, the educative effect of the expert evidence makes an appeal to commonsense notions of causation largely meaningless or produces findings concerning causation which would often not be made by an ordinary person uninstructed by the expert evidence.

It is understandable that, in the days when any contributory negligence on the part of a plaintiff was sufficient to deprive him or her of a verdict, judges should sanction tests for determining causation which in practice allowed juries to avoid the consequences of a strict application of the doctrine of contributory negligence. In that context, instructions to determine whether a particular act or omission was a cause of damage according to commonsense notions were appeals to extra-legal values to determine "hard cases". Significantly, this Court said in Fitzgerald [1] that:

there will not seldom be cases in which the attention of the jury ought to be called by the judge to the question whether a particular act or omission, which they may regard as negligent, can fairly and properly be considered a cause of the accident. (My emphasis.) But now that contributory negligence is no longer an absolute bar, why should the courts continue to sanction the use of formulas which allow tribunals of fact, under the guise of using commonsense, to determine legal responsibility by applying their own idiosyncratic values? Directions to use commonsense notions of causation to find the "proximate", "real", "efficient" or "substantial" cause of an occurrence are invitations to use subjective, unexpressed and undefined extra-legal values to determine legal liability. To hold a person liable for damage resulting from a set of conditions or relations simply because his or her wrongful act or omission was a necessary condition of the occurrence of that damage would be an unacceptable extension of the boundaries of legal liability in some cases. But this truth does not justify the use of vague rules which permit liability to be determined by subjective, unexpressed and undefined values. 1. (1954) 91 C.L.R., at p. 276.

In my opinion, now that legislation allows liability for damage to be apportioned in accordance with what the court thinks is just and equitable having regard to the comparative responsibility of the parties, the preferable course is to use the causa sine qua non test as the exclusive test of causation. One obvious exception to this rule must be the unusual case where the damage is the result of the simultaneous operation of two or more separate and independent events each of which was sufficient to cause the damage. None of the various tests of causation suggested by courts and writers, however, is satisfactory in dealing with this exceptional case. Perhaps no more can be done in this situation than to treat each wrongful act as an independent cause for legal purposes. The terms of a statute, legal rule or legal instrument may also require a different approach from the "but for" test. In general, however, the "but for" test should be seen as the test of legal causation. Any other rule limiting responsibility for damage caused by a wrongful act or omission should be recognized as a policy-based rule concerned with remoteness of damage and not causation.

For a time after the decision in The Wagon Mound [2] , it seemed that foreseeability had become the exclusive common law test of remoteness of damage. But once the courts continued to follow the rule that a tortfeasor must take his victim as he finds him (Smith v. Leech Brain & Co. Ltd. [3] ), it became apparent that foreseeability could not be the exclusive test of remoteness of damage. Moreover, courts have continued to give effect to the doctrine of novus actus interveniens. Thus, the judgment of this Court in Mahony v. J. Kruschich (Demolitions) Pty. Ltd. [4] contains the dictum that:

A line marking the boundary of the damage for which a tortfeasor is liable in negligence may be drawn either because the relevant injury is not reasonably foreseeable or because the chain of causation is broken by a novus actus interveniens. As I have pointed out, however, questions of novus actus interveniens involve value judgments. Thus, Hart and Honoré who support the novus actus interveniens doctrine argue at p. 133 that the decisions of the courts have been controlled by the principle that the connexion between a wrongful act and harm is negatived "if the factors required, in addition to the wrongful act, for the production of the harm include a voluntary human action or an abnormal occurrence". Yet Hart and Honoré concede (see pp. 142-156) that non-voluntary conduct for this purpose includes conduct which is the result of physical compulsion, concussion, fright, self-preservation, preservation of property, protection of interests, legal or moral obligations, unreflective acts, mistake, accident or negligence. As J. Stapleton has pointed out in "Law, Causation and Common Sense", Oxford Journal of Legal Studies, vol. 8 (1988) 111, at p. 125: Quite apart from the fact that this is a remarkable departure from ordinary usage for authors committed to analysis of the accurate use of plain language, it serves to disguise the fact that the division between those acts which negative causal connection and those that do not appears, as we have seen, to be value-based, that is, to depend on the evaluation of the interest served by the intervening act. 1. [1961] A.C. 388. 2. [1962] 2 Q.B. 405. 3. (1985) 156 C.L.R. 522, at p. 528.

Once it is recognised that foreseeability is not the exclusive test of remoteness and that policy-based rules, disguised as causation principles, are also being used to limit responsibility for occasioning damage, the rationalization of the rules concerning remoteness of damage requires an approach which incorporates the issue of foreseeability but also enables other policy factors to be articulated and examined.

One such approach, and the one I favour, is the "scope of the risk" test which has much support among academic writers as well as the support of Denning L.J. in Roe v. Minister of Health [5] , where his Lordship said:

Starting with the proposition that a negligent person should be liable, within reason, for the consequences of his conduct, the extent of his liability is to be found by asking the one question: Is the consequence fairly to be regarded as within the risk created by the negligence? If so, the negligent person is liable for it: but otherwise not. (My emphasis.) 1. [1954] 2 Q.B. 66, at p. 85.

Damage will be a consequence of the risk if it is the kind of damage which should have been reasonably foreseen. However, the precise damage need not have been foreseen. It is sufficient if damage of the kind which occurred could have been foreseen in a general way: Hughes v. Lord Advocate [6] . But the "scope of the risk" test enables more than foreseeability of damage to be considered. As Fleming points out (Law of Torts, 7th ed. (1987), p. 193), it also enables allowance to:

be made to such other pertinent factors as the purpose of the legal rule violated by the defendant, analogies drawn from accepted patterns of past decisions, general community notions regarding the allocation of "blame" as well as supervening considerations of judicial policy bearing on accident prevention, loss distribution and insurance. Thus, the "scope of the risk" test enables relevant policy factors to be articulated and justified in a way which is not possible when responsibility is limited by reference to commonsense notions of causation or to more specific criteria such as "novus actus interveniens", "sole cause" or "real cause", all of which conceal unexpressed value judgments. 1. [1963] A.C. 837.

But for the breach of the duty which the defendants owed to the plaintiff in the present case, the damage which he suffered would not have occurred. Consequently, the defendants' breach of duty was a cause of the accident. But should the defendants be exempted from responsibility on the ground that the plaintiff's damage was not within the scope of the risk created by the defendants' breach of duty?

When a defendant has a duty to a plaintiff to prevent the occurrence of damage of the kind which occurred and the defendant's breach of duty was a cause of that damage, the damage will be held to be within the scope of the risk which the defendant was required to avoid unless the plaintiff sustained the damage intentionally (or, perhaps, recklessly) or the damage occurred in a manner which could not reasonably be foreseen in a general way. Thus, in Stapley the employer was held liable for its negligence and breach of statutory duty even though the plaintiff would not have been injured but for his own concurrent negligence and breach of statutory duty. Cases involving an intervening negligent act on the part of the plaintiff are no different in principle from cases involving the intervening act of a third party. In the latter class of case, it is well established that a defendant will be held liable for intervening acts which are foreseeable if he or she has a duty to prevent them occurring: Stansbie v. Troman [1] ; Dorset Yacht Co. v. Home Office [2] ; Mahony v. J. Kruschich (Demolitions) Pty. Ltd. [3] ; Smith v. Littlewoods Ltd. [4] .

1. [1948] 2 K.B. 48, at p. 52. 2. [1970] A.C. 1004, at p. 1030. 3. (1985) 156 C.L.R., at p. 529. 4. [1987] A.C. 241, at pp. 272-273.

The defendants conceded that it was reasonably foreseeable that injury to another road user could result from the manner in which the truck was parked. Moreover, no attack was made on the trial judge's finding that the defendants were in breach of the duty of care which they owed to the plaintiff. Nor was any attack made, nor could it be made, on his Honour's finding that that duty of care was owed to careless and drunken drivers as well as careful drivers: see London Passenger Transport Board v. Upson [5] ; and cf. McLean v. Tedman [6] ; Bankstown Foundry Pty. Ltd. v. Braistina [7] ; Bus v. Sydney County Council [8] . That an intoxicated driver, driving at excessive speed and failing to keep a proper lookout, might collide with the parked truck was one of the risks which the defendants were obliged to reasonably foresee and guard against. Thus, the collision which occurred was "the very kind of thing which [was] likely to happen" if there was a want of care on the part of the defendants: Haynes v. Harwood [9] . The damage which the plaintiff suffered was a natural and probable consequence of the defendants' breach of duty. Both the damage which the plaintiff suffered and the manner of its occurrence were fairly within the risk created by the defendants' breach of duty. With great respect to the majority of the Full Court of the Supreme Court, once the defendants were held to be in breach of the duty of care which they owed to the plaintiff, it followed, in the circumstances of this case, that, subject to the question of apportionment, the defendants were liable for the plaintiff's damage.

1. [1949] A.C. 155, at pp. 173, 176. 2. (1984) 155 C.L.R. 306, at pp. 309-311. 3. (1986) 160 C.L.R. 301, at pp. 308-309. 4. (1989) 167 C.L.R. 78, at pp. 90-91. 5. [1935] 1 K.B. 146, at p. 156.

The appeal should be allowed and the judgment entered for the plaintiff by the trial judge should be restored.