Chester v Waverley Municipal Council [1939] HCA 25; (1939) 62 CLR 1 (6 June 1939)

Chester Plaintiff, Appellant; and The Council of the Municipality of Waverley Defendant, Respondent.

The defendant council excavated a trench in a road of which it had the control. The trench was from two feet six inches to seven feet deep and at a week-end it became filled with water. There was a railing round the trench above the earth which had been thrown out on to the road. Children could easily get under the railing. The plaintiff's son, aged seven and a half years, went out to play. There was some evidence that he accidentally fell into the trench. He was drowned. The body was not found until some hours after he had been missed by his parents. The plaintiff, his mother, searched for him and became distressed upon failing to find him. She was present when his dead body was found in and taken from the trench. There was evidence that the plaintiff thereupon received a severe nervous shock—more than a fright—more than temporary mental disturbance and distress. She sued the council for damages for negligence. The learned trial judge nonsuited the plaintiff and the Full Court of the Supreme Court of New South Wales refused a motion for a new trial. The plaintiff has appealed to this court.

The plaintiff must, in order to succeed, establish a breach by the defendant of a duty owed to her to take care. If there is a breach of such a duty then the defendant is liable for the direct consequences of that breach, even though such consequences may have been unexpected. This is the law as at present declared in Polemis' Case[1]. But in determining the extent of any alleged duty to take care it is necessary to take into account the probable consequences of the relevant act or omission. As Warrington L.J. said in Polemis' Case[2], "the presence or absence of reasonable anticipation of damage determines the legal quality of the act as negligent or innocent." Scrutton L.J. said in the same case: "To determine whether an act is negligent, it is relevant to determine whether any reasonable person would foresee that the act would cause damage; if he would not, the act is not negligent"[3]. So also in Donoghue v. Stevenson[4] Lord Atkin, referring to the duty to take care in relation to other persons, says that the persons in respect of whom the duty exists are "persons who are so closely and directly affected by my act that I ought reasonably to have them in contemplation as being so affected" (that is, so injured) "when I am directing my mind to the acts or omissions which are called in question"[5]. In this court, see Bunyan v. Jordan[6].

According to these principles the council owed a duty to persons using the highway to take reasonable care to prevent them receiving injury by reason of the presence of the open trench.

The council contends that there was no breach of this duty because the trench was guarded and protected in the ordinary, which was a sufficient, manner. It could have been made more safe for playing children if it had been surrounded by an impenetrable wall or if a watchman had been placed on duty. But the council contends that it was not bound to take such precautions (Latham v. R. Johnson & Nephew Ltd.[7]; Hastie v. Edinburgh Magistrates[8], quoted in some detail in Purkis v. Walthamstow Borough Council[9]). It is not, however, necessary, upon the view which I take, to consider this aspect of the case. I am prepared to assume that the council was guilty of actionable negligence in relation to the child in leaving the trench in the condition in which in fact it was left. If so, the child, if he had been injured and not drowned, would have had a right of action for damages; if he had been an older person with dependants and had been drowned, his dependants would have had a right of action under the Compensation to Relatives Act 1897 N.S.W. (Lord Campbell's Act).

But in this case the plaintiff must establish a duty owed by the defendant to herself and a breach of that duty. The duty which it is suggested the defendant owed to the plaintiff was a duty not to injure her child so as to cause her a nervous shock when she saw, not the happening of the injury, but the result of the injury, namely, the dead body of the child. It is rather difficult to state the limit of the alleged duty. If a duty of the character suggested exists at all, it is not really said that it should be confined to mothers of children who are injured. It must extend to some wider class—but to what class? There appears to be no reason why it should not extend to other relatives or to all other persons, whether they are relatives or not. If this is the true principle of law, then a person who is guilty of negligence with the result that A is injured will be liable in damages to B, C, D and any other persons who receive a nervous shock (as distinguished from passing fright or distress) at any time upon perceiving the results of the negligence, whether in disfigurement of person, physical injury, or death.

The contention might go even further. Would the council have been liable to the mother as for negligence if the facts had been that she had suffered a severe nervous shock caused by fear that her child had been drowned in the trench, though in fact the child had only wandered away for a time and had returned safe and sound? There could in such circumstances have been no action by or in relation to the child because the alleged negligence had not caused damage to the child. Would mothers and others who actually, though mistakenly, suffered shock by reason of apprehension of injury to others have a remedy against the council?

The plaintiff relies mainly upon the case of Hambrook v. Stokes Brothers[10]. In that case it was held that a negligent defendant was liable for nervous shock caused to a mother by immediate apprehension of injury to her children. Such injury in fact happened, and the mother saw the result when she saw her child in the hospital, but the decision was expressly not based upon the latter circumstance: See per Bankes L.J.[11], where he particularly confines his decision to cases where the facts are indistinguishable in principle from the facts of that case, including the fact "that the shock was due to a reasonable fear of immediate personal injury either to herself or to her children"—See also[12], where Sargant L.J., who dissented, particularly states that plaintiff's counsel expressly disclaimed any suggestion that the shock for which damages were sought was due to seeing the injured child in the hospital.

The plaintiff also relies upon Owens v. Liverpool Corporation[13]. This is also a case of shock, found to be due to negligence in the driving of a tramcar so that the tramcar collided with a hearse and the coffin was displaced so that it was in danger of being thrown into the road. Some of the mourners recovered damages for nervous shock in an action for negligence and the judgment was upheld in the Court of Appeal. One plaintiff was the mother of the person whose body was about to be buried. The other plaintiffs were his uncle, a cousin, and a cousin's husband. The decision was based upon the proposition that the driver of the tram was under a duty to other persons upon the highway to drive the tramcar with care. There was a breach of this duty. The result was that on the principle of the Polemis Case[14] the defendant, the employer of the driver, was liable for all the consequences though these consequences were of an unexpected kind. It was said by MacKinnon L.J.:—"It may be that the plaintiffs are of that class which is peculiarly susceptible to the luxury of woe at a funeral so as to be disastrously disturbed by any untoward accident to the trappings of mourning. But one who is guilty of negligence to another must put up with idiosyncrasies of his victim that increase the likelihood or extent of damage to him: it is no answer to a claim for a fractured skull that its owner had an unusually fragile one"[15]. This passage has been relied upon by the plaintiff to support the contention that, even though nervous shock caused by the sudden death of a child may be unusual, the defendant is nevertheless liable. But it will be seen that the statement of MacKinnon L.J. refers only to the liability of "one who is guilty of negligence to another." The basis of the decision was that there was a duty and a breach of a duty to actual users of the highway not to injure them by driving carelessly. The existence of such a duty is undoubted. In the present case the question to be determined is a question as to the definition, the scope and extent, of any relevant duty. If there is no negligence (no breach of a duty to take care), the happening of a consequence (whether usual or unusual) cannot be relied upon to establish the existence of any duty, though it may have to be taken into account when damages are being assessed if a breach of the duty were established. Thus a person who assaults another is guilty of a wrong and therefore is liable in damages for fracturing the other person's skull as a result of the assault even though the skull was unusually thin. But if the defendant was taking part in an ordinary game in an ordinary way with the plaintiff, the fact that the plaintiff's thin skull was fractured by the defendant would not, in the absence of proof of negligence, establish any liability. Thus in the present case the circumstance that the plaintiff in fact suffered a shock does not establish the existence of any duty in the defendant or any breach of duty by the defendant. The question which must be asked in order to determine whether the defendant was negligent or not is whether the defendant should have foreseen that a mother would suffer from nervous shock amounting to illness if she saw the dead body of her child where the death of the child had been brought about by the negligence of the defendant towards the child. This mode of formulating the question is very favourable to the plaintiff. For reasons which I have indicated, the question should probably be put in a form which substituted the words "person" and "another person" for "mother" and "child."

In my opinion (to apply the phraseology which I have quoted from the Polemis Case[16] and Donoghue v. Stevenson[17]) it cannot be said that such damage (that is, nervous shock) resulting from a mother seeing the dead body of her child should be regarded as "within the reasonable anticipation of the defendant." "A reasonable person would not foresee" that the negligence of the defendant towards the child would "so affect" a mother. A reasonable person would not antecedently expect that such a result would ensue (Bunyan v. Jordan[18], per Rich J.). Death is not an infrequent event, and even violent and distressing deaths are not uncommon. It is, however, not a common experience of mankind that the spectacle, even of the sudden and distressing death of a child, produces any consequence of more than a temporary nature in the case of bystanders or even of close relatives who see the body after death has taken place.

In my opinion there was no evidence to establish the existence of the duty of the defendant to the plaintiff which was a necessary part of the plaintiff's case and the learned trial judge acted rightly in directing a verdict for the defendant.

This appeal arises out of the difficulties attending the law of nervous shock, which may be described as in a state of development. The facts of the present case are fully stated by Jordan C.J., in whose conclusion I agree. The breach of duty towards the deceased child on the part of the defendant is clear enough. It is of little importance whether it be called nuisance or negligence. The question appears to me really to be whether the kind of harm of which the plaintiff complains caused by the sight of her child's body on its recovery is within the ambit of the defendant's duty not to put the road in a dangerous condition. I am prepared to adopt Professor Winfield's view that nervous shock is "a particular instance of damage flowing from the commission of some particular tort," and that "nervous shock sustained by someone who is not reasonably within the contemplation of the defendant falls outside the scope of his duty to take care" (Winfield on the Law of Tort (1937), pp. 85, 87), or, as was said in Bunyan v. Jordan[19], "the harm which in fact ensued is not a consequence which might reasonably have been anticipated or foreseen." In the present instance I think that a mother's shock on the production of the dead body of her child falls outside the duty of the municipality in relation to the care of its roads. She was not using the road nor a witness of the accident. Her subsequent shock is not reasonably within the contemplation of the defendant as a consequence of the condition of the road. A negligent motorist who caused great facial disfigurement to a pedestrian could not be made liable to every person who throughout the pedestrian's life experienced shock or nausea on seeing his disfigurement. The train of events which flow from the injury to A almost always includes consequential suffering on the part of others. The form the suffering takes is rarely shock; more often it is worry and impecuniosity. But the law must fix a point where its remedies stop short of complete reparation for the world at large, which might appear just to a logician who neglected all the social consequences which ought to be weighed on the other side. The attempt on the part of the appellant to extend the law of tort to cover this hitherto unknown cause of action has, perhaps, been encouraged by the tendencies plainly discernible in the development which the law of tort has undergone in its progress towards its present amorphous condition. For the so-called development seems to consist in a departure from the settled standards for the purpose of giving to plaintiffs causes of action unbelievable to a previous generation of lawyers. Defendants appear to have fallen entirely out of favour. In this respect perhaps judges are only following humbly in the footsteps of juries.

A boy some seven or eight years of age went out, after his lunch, to play in a public street within the Municipality of Waverley. He did so with the knowledge of his mother, but he was without any attendant or supervisor. In this street the municipality or its employees had, for some municipal purpose, dug an open trench which had filled with water, some seven feet deep at one end and two feet six inches at the other. The trench was not well guarded. It was easy to climb through or over the barriers placed around it. Apparently the boy was playing on the edge of the trench, fell in, and was drowned. During the afternoon the appellant became anxious for her boy. She began to look for him but could not find him. About six o'clock in the evening the body of the boy was found in and taken from the trench. Efforts were made to revive him but without success; he was dead. The appellant was present when her boy was found, and she suffered severely from nervous and mental shock and from distress.

She brought an action in the Supreme Court of New South Wales. She alleged negligence on the part of the municipality in the excavation and protection of the trench, that her son by reason of such negligence fell into the trench and was drowned, that she, apprehending that her son had fallen into the trench and witnessing the removal of his dead body from the trench, had thereby sustained severe nervous and mental shock and suffered great pain and had incurred expense. The learned judge before whom the action was tried directed a nonsuit and this direction was affirmed in the Supreme Court on appeal. An appeal is now brought to this court.

Some relationship of duty on the part of the municipality towards the appellant must be established. Negligence in the abstract or in the air, as has often been said, is not enough. It is a question of law whether any duty to take care exists and what standard of care is required.

In the present case the respondent, I assume, had the care, control and management of the street already mentioned (Local Government Act 1919 N.S.W., sec. 249). A highway authority is not liable in respect of the highways under its care management and control for what is called non-feasance (Buckle v. Bayswater Road Board[20]). But it is liable for what is called misfeasance. It has a duty, if it does work upon the highway, to take reasonable care for the safety of those using or being upon or about the highway (Buckle v. Bayswater Road Board[21]): "to take reasonable care to avoid inflicting personal injuries, followed by damage, even though the type of damage may be unexpected—namely, shock"[22]: Cf. Hambrook v. Stokes Brothers[23]; Owens v. Liverpool Corporation[24]. "To fufil this duty" the municipality "is not bound to guard against every conceivable eventuality but only against such eventualities as a reasonable man ought to foresee as being within the ordinary range of human experience"[25]. "People must guard against reasonable probabilities, but they are not bound to guard against fantastic possibilities" (Fardon v. Harcourt-Rivington[26]; Bunyan v. Jordan[27]). It is a question of law and not of fact in each case whether there is any evidence of a breach of the duty to take care.

The open trench filled with water could not in itself frighten or terrify any normal or reasonable person. Such a person might think that the open trench was unsafe and that more care should be exercised but no more. In the present case the appellant was not present at the accident nor was she alarmed in the happening of the accident; she only saw its consequences some hours after it occurred. The failure to guard the trench was but indirectly connected with the shock to the appellant and the act or omission of the respondent was not so closely and directly connected with the shock sustained by the appellant that it can be traced to that act or omission. In my opinion the shock to the appellant is not within the ordinary range of human experience; it is so remote from the act or omission of the respondent in opening or guarding the trench that no reasonable person ought to or would foresee or contemplate the injury to the appellant.

Consequently there is no evidence of a breach of any duty that the respondent owed to the appellant and the appeal should be dismissed.

This is a case of considerable general importance. It concerns a rule of the common law of England and the principles involved will greatly affect the development of the law of negligence.

On a Saturday afternoon in August 1937, the plaintiff's child, a boy of seven years, was drowned in a deep trench which the defendant council had caused to be constructed in one of the streets of its municipality. Owing mainly to rain, but partly to underground percolation, the trench had become almost filled with water, and at one point it was seven feet deep. It appears that the trench was not fenced off so as to prevent the approach of children to the edge of the deep water although, as the defendant had caused a considerable quantity of sand to be placed at the edges of the trench, the irresistible combination of sand and water brought the children in the neighbourhood to play at the side of the pool.

I. There was evidence that the place was very attractive to children and that this was known to the defendant's servants.

At the trial the defendant called as a witness the foreman ganger who was in charge of the work. He knew quite well that children came so frequently to the trench that during working hours they had to be "chased away."

II. There was evidence that the place was very dangerous to small children, owing to the defendant's negligence in omitting to guard it securely:—

The foreman of the defendant gave the following evidence in cross-examination:—

There had been a very heavy rain storm on the Thursday afternoon before the accident. This caused work to be stopped. On the next day, Friday, the water in the trench made work impossible. On the Saturday morning of the fatality, children were playing about the edge of the trench. There was evidence that the crude railing placed around the trench, supposedly to guard it, did not even extend around it. Even after the accident, when the railings were placed in a more effective position, it was still possible for children to get through.

On the Saturday afternoon, the trench was some 40 feet long and 4½ feet wide. For the first 20 feet of its length, the approximate depth was about 7 feet, and for the remaining 20 feet, the depth ranged from 7 feet to 2½ feet. Thus the depth of water in the trench was for the most part in excess of the height of the small children playing there. In this way the menace of death was very great and very near.

At the trial the learned acting judge who presided rejected evidence of an expert that the so-called protective methods adopted by the defendant fell short of the ordinary precautions adopted in trench and excavation work. It is very difficult to follow the ruling, because the relevant clause in the existing ordinance under the New South Wales Local Government Act 1919 provided as follows:—"Where any person (including the council) by making a hole, or by placing timber or stone or in any other such way, creates on a road a new condition involving unusual danger to the public, he shall—(a) Cause the place to be sufficiently lit during such hours as lighting shall be necessary for the public safety; and (b) Cause the place to be protected with a sufficient fence or to be otherwise so guarded as to warn the public of the danger" (Ordinance 30, clause 30).

It is plain from the provision in (b) above that the jury was entitled to find that the place was neither so guarded as to warn of the danger nor protected with a sufficient fence. In order to assist in a proper finding, evidence was admissible to show the sufficiency or otherwise of the so-called "fence" placed at the spot by the defendant. For present purposes, however, the rejection of the evidence is not material, because the trial judge ruled and ruled correctly that "the trench was left virtually unguarded, and, in my opinion there was ample evidence from which negligence could properly be inferred by a jury if they saw fit so to do."

III. There was evidence that on the Saturday afternoon the plaintiff's child met his death by drowning through the negligence of the defendant.

Not only was the plaintiff's child recovered from a deep portion of the trench, but the defendant was courageous enough to call as a witness another child who gave evidence from which the jury might have concluded—they were certainly not bound to accept all his evidence having regard to his tender years and the vagueness of many of his statements—that the plaintiff's child fell in the water while playing on the edge of the trench. The child said in evidence:

IV. There was evidence that from the moment when the plaintiff discovered that her child was missing, she searched for him without intermission, that in the course of her search she came to the trench for the purpose of finding or aiding her child, and that while the water in the trench was being explored for the same purpose and until the body was recovered, she suffered severe nervous shock as a result of her own unaided sense impressions.

According to the mother's evidence, the child left his home after lunch at about 2 p.m. The family lived at Allen's Parade, Waverley, in the street where the trench was being excavated. At about 3 p.m. his mother became concerned and commenced to look for her child. Upon her husband's returning from work, both parents called in the aid of nearby relatives, who all helped in the search. The plaintiff had resided in Allen's Parade for only 14 days, and at first was unaware of the special menace of the deep trench. As a result it did not occur to her or her fellow searchers for some time that the child might have fallen into the trench. Late in the afternoon, however, it was suggested by someone, perhaps by the mother of the little boy who was called as a witness, that Maxie, the plaintiff's child, might have fallen in the water. Coming with her husband to the side of the trench the plaintiff was at once beset with fear at the sinister significance of the trench, especially when one of the searchers was unable to plumb the depth of its water.

The plaintiff was a woman of Polish extraction, and found special difficulty in narrating the precise nature of her feelings, her fears, her hopes and her sufferings. But it is quite easy, I think, to perceive the order of events. It is abundantly clear that until the recovery of the body she did not know that her child had been drowned in the trench. Like most mothers placed in a similar situation, she was tortured between the fear that he had been drowned and the hope that either he was not in the trench at all, or that, if he was, a quick recovery of his body and the immediate application of artificial respiration might still save him from death. In this agonized and distracted state of mind and body she remained for about half an hour, when the police arrived and the child's body was discovered and removed.

During this crucial period the plaintiff's condition of mind and nerve can be completely understood only by parents who have been placed in a similar agony of hope and fear with hope gradually decreasing. In the present case the half hour of waiting was the culmination of a long and almost frantic searching which had already reduced her to a state of nerve exhaustion. Even after the finding of the body, an attempt at artificial respiration was made and abandoned only after expert lifesavers had worked on the child's body for some time.

William Blake's imaginative genius has well portrayed suffering and anxiety of this kind:—

The Australian novelist, Tom Collins, in Such is Life, has also described the agony of fearfulness caused by the search for a lost child:—

Not only its poets and novelists, but, at any rate in recent years, those engaged in the administration of the common law of England have recognized that shock of the most grievous character can be sustained in circumstances analogous to those of the present case. Elsewhere in the present opinion I examine the question whether the Privy Council decision in Victorian Railways Commissioners v. Coultas[28] precludes the courts of the British Dominions from allowing damages to be recovered where injury takes the form of illness due to nervous shock.

V. In the circumstances, it is not remarkable that there was evidence of some permanent injury to the plaintiff's nervous system. According to the doctor who attended upon her:

I have dealt with the facts of the case at some length not only because an understanding of them is important from the point of view of liability, but because, in my opinion, they are summarily but insufficiently set out in the Full Court's statement that "the discovery that her son had been drowned caused her a severe shock." This statement takes no account of (a) the plaintiff's long agony of waiting when she feared that her son had been drowned but certainly did not "know" it, or of (b) the effect upon the plaintiff of the actual removal from the water of her child, especially as the circumstances suggested at least to her and her husband that even at that moment life was not quite extinguished.

The question as to what actually caused the plaintiff's illness was entirely for the jury. In his judgment directing a verdict for the defendant the trial judge ventures an interpretation of the facts which differed somewhat from that of the Full Court.

"I think," he said, "the reasonable inference is that the cause of the shock to the plaintiff was her realization of the fact that her child was drowned and the subsequent sight of his body and that the cause of his drowning or the nature of the locality of the fatality had no effect on her mind."

I also think that this interpretation of the facts is inadequate and would probably be so regarded by a jury. For instance, I fail to understand the learned judge's insistence upon excluding the cause of the drowning and the nature of the locality from the factors which operated upon her mind. The suggestion is, I suppose, that the plaintiff's suffering was no different in essentials from that of any other mother to whom someone had delivered a message that her child had been drowned although she had no first hand knowledge of any of the attendant circumstances. But it is impossible to abstract from the totality of events any factor which during the critical waiting period contributed to her distress and shock. In particular it is not possible to ignore the outstanding impression operating on the plaintiff's mind—that within the apparently small area of the trench situated so close to his own home her child might be lying dead or in desperate danger of death; within such close reach in one sense, but in circumstances preventing immediate action by way of rescue or assistance.

The jury might reasonably have found that the cause of her nervous shock and subsequent illness was the very terrifying setting of the tragedy—a setting which was attributable to the defendant in the sense that though the precise situation might not have been foreseen by a reasonably prudent person in the defendant's position, something of the kind should have been foreseen and steps taken to prevent it.

The learned trial judge directed the jury to return a verdict for the defendant and this conclusion was affirmed by the Full Court. But the two judgments proceed upon different lines of reasoning. The trial judge thought that the legal principle to be applied was:—

The learned judge ruled, in effect, that the defendant could not be under any liability to the plaintiff unless she witnessed the actual fall of her child into the trench. But she arrived upon the scene of the fatality at a later period—"some appreciable time" after. Despite the fact that the plaintiff then suffered a nervous shock, such shock was not—in the learned judge's opinion—"due to the negligent act."

The legal position here asserted is that the defendant escapes liability because the injury to the plaintiff through shock, although sustained through the evidence of her own unaided senses, was so sustained at a time when the consequences of the defendant's carelessness so far as the child was concerned had been completed or almost completed. Thus the learned judge does not seem to base his decision upon the absence of any legal duty to the plaintiff. He emphasizes one or two expressions to be found in the judgments of Bankes and Atkin L.JJ. in Hambrook v. Stokes Brothers[29]. There Bankes L.J. referred to the shock having been due to a "reasonable fear of immediate personal injury either to herself or to the children"[30], and Atkin L.J. accepted the principle that in circumstances analogous to those in Hambrook v. Stokes Brothers[31] there arose a duty to take reasonable care "to avoid threatening personal injury to a child in such circumstances as to cause damage by shock to a parent or guardian then present"[32]. But I think it is clear that Lord Atkin's phrase "then present" must at least cover the full period of time during which the consequences of the defendant's "primary" negligence were being made manifest. Thus, in the case of a motor car carelessly left standing on the top of a hill, the period during which the consequences of the defendant's negligence will continue must extend from the first movement of the car down the hill until it finally comes to rest. In my understanding of it, Lord Atkin's judgment would certainly not have excluded liability to the mother if, at the first moment of its careering down the hill, the car had injured her daughter so that the consequences of the defendant's primary negligence had been completed, even although, in that event, the mother's fear for her child might well have been fear of subsequent rather than fear of antecedent injury.

It seems very unreasonable to make liability depend upon too nice a psychological analysis of the nature and time of the first onset of the fear and shock suffered by a mother in circumstances analogous to Hambrook v. Stokes Brothers[33]. The principles would seem to be that as fear and shock of a similar character should be foreseen as a probable or possible consequence of the defendant's primary breach of duty, the liability for the resulting illness, if it exists at all, exists as much in cases where, at the moment of the onset of the shock, the casualty feared has been completed as where it is still in progress or where it has not yet eventuated but is about to do so; or even perhaps where it has not yet eventuated and will not do so.

Here the learned trial judge interpreted Hambrook v. Stokes Brothers[34] as depending upon the fact that the shock to the mother commenced to operate upon her at some moment of time before her child had been injured by the car. A close reading of the case shows that it is not possible to say that the child had not been injured before her anxiety reached the crucial point of nervous shock. Further, her anxiety concerned all three children, and two of them were not injured at all. It would be a very narrow reading of this epoch-making decision to limit the principle of liability which it applies to cases where the onset of nervous shock occasioning illness preceded or coincided with the occurrence of the apprehended casualty.

The judgment of the Full Court is partly based upon the theory that the cause of the plaintiff's shock and illness was merely "the discovery that her son had been drowned." It was said by Jordan C.J.:—

The present case presents legal difficulties of a special kind, but nothing is to be gained by giving a special interpretation or colour to the facts. That is one reason why I have stated them so fully. It seems indisputable that the jury could have found that the onset of the plaintiff's nervous shock took place at a point of time when the plaintiff, although at the side of the trench, did not know or even believe that her child had been drowned. Equally she was not "looking for the body of a child." She was looking for her child. She was terrified lest he should have been drowned, was taking notice of little except what her own senses were telling her, was hoping against hope that her very worst fear would not be realized. It is true that from the point of legal liability these differences in fact may not be decisive, but such differences may make the issue of liability easier to determine. Not only the facts, but all reasonable inferences to be drawn from the facts were for the jury; but they have not yet found, and I imagine that they would not have found, the facts in accordance with the Full Court's interpretation thereof.

The Full Court considered that the plaintiff's case as to the existence of a duty towards her is prejudicially affected by the fact that at the scene of the fatality she was not present "in the character of a wayfarer startled by a distressing sight." A priori it would be surprising if, in relation to the question of breach of duty towards a mother, a defendant is in a stronger position in cases where, by reason of the consequences of his primary negligence towards her child, the mother is trying desperately to find and rescue him than in cases where, knowing nothing of any danger or accident, she merely stumbles across the child's body after the accident has occurred. In the distinction made by the Full Court there seems to lurk the fallacy that the principle of Hambrook v. Stokes Brothers[36] applies only for the benefit of "wayfarers" or "passers-by." I think that the law is at once more civilized and more humane. Behind the illustration provided by Hambrook v. Stokes Brothers[37] lies the broader principle enunciated by Lord Atkin in Donoghue v. Stevenson[38] in order to help in determining whether the common law has established a relationship of duty between a defendant on the one hand and a plaintiff, or the class to which a plaintiff belongs, on the other:—"Who, then, in law is my neighbour? The answer seems to be—persons who are so closely and directly affected by my act that I ought reasonably to have them in contemplation as being so affected when I am directing my mind to the acts or omissions which are called in question"[39].

Let us apply this criterion to a reasonable person in the situation of the defendant council. Such a person would foresee that, by leaving the trench inadequately guarded, it would probably become, especially when filled with water and provided with sand, a very attractive place to children in the neighbourhood of the trench. He would also foresee that, having regard to the unfortunate but notorious fact that children of workpeople are frequently compelled to play in the streets and also to the fact that the water was in the trench, the special menace of the place would be that small children might fall in and be drowned. He would also foresee when "directing his mind" to the dangers that, if a child got into the zone of the special danger, his parents (and others) would resort to the spot either to seek for the child or, upon hearing his cries, to rescue him from danger; and that, in so doing, they might themselves sustain physical injury or illness caused by nervous shock and distress.

If this application of Lord Atkin's test is not forbidden by law, it demonstrates that the principle of Hambrook v. Stokes Brothers[40] cannot be limited to "wayfarer" cases or to ordinary street accidents, and should or may cover the facts of the present case. On this footing, if the present plaintiff while searching for her child near the trench had thrown herself into the water to rescue him from drowning, and in the course of so doing had been herself injured, she would have been entitled to recover from the defendant because she belonged to the class of persons who was "neighbour" to the defendant so that the defendant's primary breach of duty to the child concurred with a secondary breach of duty to the mother. In such a case it would seem that recovery should be had for ordinary physical injury and also for physical illness caused by nervous shock. If the law gives a remedy in the former case it can hardly deny it to the plaintiff who, intent upon rescuing her child, was practically compelled to delegate that task to her husband but who suffered physical illness through the effect upon her nervous system of the half hour of waiting, hoping and fearing. If the present defendant had "directed his mind" at all to the possible consequences of his primary default, would he not have foreseen the likelihood of injury being suffered, not perhaps in the precise way in which it was suffered, but in some such way? If so, he owed a duty to the person who suffered. The law does not assume that all beings can bear a burden too great for many to suffer. "There may well be," as Lord Atkin said, "cases where the sight of suffering will directly and immediately physically shock the most indurate heart; and if the suffering of another be the result of an act wrongful to the spectator, I do not see why the wrongdoer should escape" (Hambrook v. Stokes Brothers[41]).

But it is necessary to deal at once with an argument which seems to have been accepted by the Full Court. It may be put as follows: A few people—"susceptible and emotional mothers" let us say—would have suffered nervous shock and injury after undergoing an experience similar to that of Mrs. Chester. This fact a reasonable person would or might have foreseen. But only persons in such exceptional category would have suffered. Therefore the defendant's secondary duty existed only towards those who did not belong to the exceptional category, i.e., only towards the "ordinary normal human being." The non sequitur is easily discernible. So far as the argument rests upon the contention that no other parents would have suffered shock and illness from the ordeal undergone by Mrs. Chester, I think this is a mere assertion and is contradicted by all human experience. I think that only "the most indurate heart" could have gone through the experience without serious physical consequences.

So far as the argument rests upon law, it may, perhaps, claim some support from Professor Winfield's suggestion that the plaintiff's right to recover for physical illness caused by nervous shock is dependent upon the fact that the plaintiff was a "normally firm and reasonable person" (Winfield on the Law of Tort (1937), p. 85). That learned author says:—"Thus, if a nervous old lady sees B, a stranger, hurt in a trivial accident due to A's negligence in a crowded London street and suffers shock in consequence, she has no cause of action against A, not because the consequence is too remote, but because A owes no legal duty to take care towards unreasonably nervous people and if he owes no such duty he cannot commit a breach of it" (Winfield on the Law of Tort (1937), p. 87).

Professor Winfield rather suggests that this statement is supported by Atkin L.J.'s reference in Hambrook v. Stokes Brothers[42] to the imaginary case of "frightening an old lady at Charing Cross." I respectfully dissent from this suggestion. What Lord Atkin said was:—

Atkin L.J. was pointing out that, inasmuch as liability in ordinary street-accident cases comes into existence only where the accident is due to the negligence of the defendant, the decision in Hambrook v. Stokes Brothers[44] as to nervous shock casts no additional duty upon careful drivers of vehicles. Providing they drive carefully, the "old lady at Charing Cross" who becomes frightened on witnessing an accident can have no cause of action against them. It is not that there is no duty owed to the "frightened old lady," but—the duty being to take reasonable care—that there has been no breach of it. Whether, in the case of negligent driving, "the old lady" would have an action did not fall for decision in Hambrook v. Stokes Brothers[45] although the general reasoning of Atkin L.J. there and in Donoghue v. Stevenson[46]—as well as that of Lord Wright in Grant v. Australian Knitting Mills Ltd.[47]—suggests that she would have a cause of action.

The reason for this is that every reasonable person is aware that all sorts and conditions of men and women may be found among the eye witnesses of an accident. The very phrase suggests that the "old lady at Charing Cross" is merely a human type, not a monstrosity. As Professor Goodhart has said, "Although it may not be probable that any particular man may have an unusually thin skull or a weak heart, nevertheless it is a fact which is easily forseeable. A man who strikes another ought to forsee that his victim may be suffering from some weakness. We all know that the average man in the street is not necessarily the average man" (Essays in Jurisprudence and the Common Law (1931), pp. 126, 127).

This general principle seems reasonably applicable to cases like the present. The defendant could not have been assured in advance that if it negligently failed to guard the trench, the mother of any child injured thereat would conform to one category rather than to another. Therefore the duty was to all members of all categories.

One is reminded of the observations of Kennedy L.J. in Dulieu v. White & Sons[48]:—

If the theory advanced by the Full Court were sound, it might equally have been applied to exonerate drivers from a duty to drive with reasonable care in cases where the person injured was "peculiarly susceptible" to fractures. But there too the duty, though now well defined, embodies the principle stated in Donoghue v. Stevenson[49]. It is always to be foreseen that the person injured by careless driving may turn out to be "peculiarly susceptible" to fractures. So too in the case of susceptibility to nervous shock. Rather than limit the duty to those who are not easily susceptible to shock, it is more logical to contend that if a reasonable person would foresee that only "mothers peculiarly susceptible to nervous shock" would suffer physically as a result of their viewing the injuries of their children, the person who has negligently caused such injuries at least owes an antecedent duty to the class of "mothers peculiarly susceptible to nervous shock."

In the course of the recent decision of the Court of Appeal in Owens v. Liverpool Corporation[50], MacKinnon L.J. said: "But one who is guilty of negligence to another must put up with idiosyncracies of his victim that increase the likelihood or extent of damage to him: it is no answer to a claim for a fractured skull that its owner had an unusually fragile one."

It is not possible to avoid the consequences of this reasoning by contending that as the learned Lord Justice uses the phrase "guilty of negligence" his proposition can have no application to cases where the antecedent duty of the defendant exists only in relation to persons without fragile skulls. In truth, MacKinnon L.J. is denying the possibility of such cases wherever the defendant should foresee that one or more of a class may be injured by his carelessness. A duty to take care to avoid injuring a person who has some peculiar physical or nervous weakness may be brought into existence where the special circumstances are known to the defendant. But where a class or group may be endangered by carelessness, the defendant's sphere of duty can seldom be contracted by denying a duty to weak and nervous members of the group.

It is necessary to examine further the judgment of the Full Court so far as it would confine recovery in relation to nervous shock to those cases where the sufferer "is a normal person of ordinary firmness and mental stability," possessing "the polyglot character" referred to by Phillimore J. in Dulieu v. White & Sons[51] and sufficiently satirized by MacKinnon L.J. in Owens v. Liverpool Corporation[52].

The judgment rather concedes that if a serious accident had occurred through someone stumbling into the trench, the defendant council might be regarded as responsible for nervous shock sustained by the plaintiff's "witnessing the agony or hearing the screams of the victim" but only providing "the person in question was a normal person of ordinary firmness and mental stability, or the sights and sounds were of a kind likely to cause such injuries to such a person." This seems to me to assert a proposition of law which is fundamentally at variance with principle and authority alike. For the principle the Full Court relies on Bunyan v. Jordan[53]. That was an extremely unusual type of case. In my opinion it was a case of malicious or wanton injury rather than a case of breach of duty to take reasonable care[54]. Dixon J. said of the defendant:—"He may have intended to frighten those surrounding him, but, if so, it was only for the purpose of sensationalism. The shock he intended to give or the emotions he intended to arouse could not in a normal person be more than transient"[55]. I expressed the opinion that if the defendant wilfully intended to cause alarm and his efforts turned out to be more successful than he expected, it was utterly immaterial whether the "ordinary, normal human being" would have become terrified and suffered illness through shock. It was enough if the victim of the defendant's exhibitionism became terrified and suffered illness[56].

But some of the judgments of this court discussed the matter upon the footing of mere negligence, even suggesting that no duty could ever come into existence if the plaintiff was "peculiarly susceptible to nervous shock." But it is clear that a conclusion involving so general a principle cannot follow from the denial of the duty to take care in the very peculiar circumstances of Bunyan v. Jordan[57]. There the only person to whom the supposed duty lay was the plaintiff herself. It does not follow that in cases where a reasonable person in the defendant's position (assumedly directing his mind to the consequences of a failure to take a precaution) could not know with certainty the idiosyncrasies of every possible eye-witness of the immediate consequences of such a failure, he is not under a duty to all possible eye-witnesses including those who are susceptible to shock as well as those who are not. In such cases, if it should be foreseen that some eye-witnesses will or may be injured by shock, the antecedent duty of the defendant to take care exists towards all members of the class. If this reasoning is sound, Bunyan v. Jordan[58], whatever may be said of the actual decision, can have no application to the present case.

Since the decision in In re Polemis and Furness, Withy & Co.[59] there has been something of a tendency to avoid its important results by the argument that unless the actual damage a plaintiff has sustained as a direct consequence of the defendant's act or omission could have been foreseen by a reasonable person in the defendant's position, that act or omission cannot be imputed to the defendant as negligence. This argument is in flat defiance of the Polemis decision[60]. In substance it seeks to restore the rejected rule of damage by denying the existence of a duty wherever the consequences to the plaintiff of the defendant's careless act or omission were not the "natural and probable" consequences. It attempts to produce the legal result condemned in the Polemis Case[61] by altering the line of attack. It is not surprising to find a similar argument invoked in cases like the present where (a) the factual situation is rather uncommon and where (b) it may always be asserted that the particular damage could not have been foreseen. Of course these facts do not negative a pre-existing duty provided some damage to the class of persons (of whom the plaintiff was one) was reasonably foreseeable. Then, as the present Lord Porter pointed out in an important analysis of the Polemis Case[62], "the fact that some damage must be foreseeable in order to constitute the tort of negligence is no ground for saying that the damage recoverable is confined to that which should have been foreseen" (Cambridge Law Journal (1934), at p. 183).

There is not the slightest ground for supposing that in Hambrook v. Stokes Brothers[63] the Court of Appeal stopped to inquire whether Mrs. Hambrook was "a normal person of ordinary firmness and mental stability," or gave any support to the curious theory that no duty to take care existed except with respect to such persons. Atkin L.J. was clearly of opinion that a reasonable person should have anticipated nervous shock at least to persons like Mrs. Hambrook, not finding it necessary for the decision to determine finally whether the same anticipation should have been made in respect of all "bystanders." It seems to be curious to argue that unless a reasonable person should have anticipated nervous shock to all and sundry, including the casual passer-by, he would not, and could not, have anticipated damage to the parents of the child. Here, as in Hambrook v. Stokes Brothers[64], the duty cannot in principle be limited to parents or relatives. But here, as in Hambrook v. Stokes Brothers[65], the duty certainly extended to the particular plaintiff, the mother of the injured child.

After a careful consideration of the very able arguments addressed to us on the appeal, I have come to the conclusion that it is of assistance to emphasize certain aspects of Hambrook v. Stokes Brothers[66]. The following facts, stated in the introduction to the report, are of very great significance:—

Although the mother happened to be using the highway at the moment of the onset of shock and was therefore one of a class to whom the defendant owed an antecedent or primary duty to avoid injuring by careless operation of the lorry, she was in fact as much outside the zone of danger as if she had remained throughout on the balcony of a house fronting the street and from that vantage point had followed the progress of her children along the street and the subsequent running away of the lorry.

Atkin L.J. thought that the duty to take care to prevent injury through shock should logically extend to "the bystander in the highway"[68], but the general tenor of his judgment suggests that the existence of such a duty arose from the mother's proximity to the scene of the accident and was not based upon her being "in" the highway rather than "above" or "overlooking" it.

II.

Further, the mother's fear was of her children "having been injured." Indeed some considerable time elapsed before she was finally aware that her little girl had been injured. In the meantime, she had hurried to the school, found her little girl still missing, and gone to the hospital. There she found that some of her fears had been realized. The onset of shock had occurred earlier, but when it occurred her fear was that disaster had already overtaken her children. As I have already suggested, liability can hardly turn upon the question whether the impact of shock preceded, or was coincident with, or immediately followed the sight or hearing of the casualty and its injurious consequences and the fears and alarms thereby occasioned. A far sounder principle is suggested by Professor Magruder when he says that psychic impact must be "fairly contemporaneous" with the casualty (Harvard Law Review, vol. 49, at p. 1039).

III.

The judgments of the majority of the Court of Appeal proceed upon the broad lines that the mother, had she survived, would have been entitled to recover if the casualty regarded as a whole (described by Bankes L.J. as "the running away of the lorry") had produced a reasonable fear of "immediate" (i.e., as a result of "the running away of the lorry") injury to her children, and her nervous shock had arisen from what she herself saw or realized rather than from mere reports to her by a third person: See per Atkin L.J.[69] and per Bankes L.J.[70].

IV.

There was no discussion of liability if the facts had been that the mother, apprehending on reasonable grounds and from the evidence of her own unaided senses that her children had been injured during the runaway, had rushed to rescue, aid or comfort them but was unable to ease her mind until later when she discovered that all of them escaped injury. In such a case, the subsequent news of the escape of the children might not alter the physical consequences of the nervous shock previously sustained. But, as has been noted, two of Mrs. Hambrook's children did escape, although her shock had been sustained through her fears as to all three. In principle it is difficult to see why, especially in relation to a casualty like the present, which is not merely the incident of a moment, a defendant whose carelessness had undoubtedly injured the plaintiff by causing alarm and nervous shock should obtain immunity because the plaintiff's reasonable fears turn out to be mistaken. In such cases the secondary duty should—it would seem—be regarded as imposed by law on the defendant because the possibility or probability of nervous shock to some person might have been foreseen whether or not the particular fears of the plaintiff turned out to be groundless. In the present case the question did not arise for decision, and, in dealing with it, it may be necessary to take account of the caveat entered by Lord Wright in Grant v. Australian Knitting Mills Ltd.[71] and quoted later in this opinion.

A learned article by Professor Magruder in the Harvard Law Review attempts to rationalize Hambrook v. Stokes Brothers[72] in a way which seems to me to be unjustified. He argues that by leaving the lorry insufficiently braked at the top of the hill, the defendant subjected the mother and all other users of the highway to "a bundle of risks."

Even if the Court of Appeal supposed that Mrs. Hambrook had been subjected to these series of risks at the moment when the defendant negligently left the lorry unbraked, no special significance was attached to the point. As we have seen, she was never in the zone of danger except through the seventh risk mentioned above. It is true that other persons using the highway might have become subject to a number, or even all, of these risks. But that fact is hardly relevant to the question whether the defendant was under a duty to those persons who in the circumstances were or would be subjected to the seventh risk but to no other. It seems to me that the Court of Appeal were addressing themselves to the question whether the defendant was under a duty to those persons in the street who, being themselves in no danger at all, should nevertheless have been foreseen by the defendant as likely to suffer nervous shock from the sight of the injurious consequences produced by the defendant's primary breach of duty. I think it is plain that if Mrs. Hambrook had not been in the street at all but had been terrified and shocked in the same way, she would have been entitled to recover.

If my analysis is sound, the case of Waube v. Warrington[73], referred to by Professor Magruder, was wrongly decided. He thus refers to it:—"In Waube v. Warrington1(1935) 258 N.W. 497; 216 Wis. 603. a mother, who was in frail health, looked out of the window while her child was crossing the highway and saw the child killed by the negligence of the defendant motorist. The shock to the mother was so severe that she died shortly thereafter. It was held that the defendant was not liable for wrongful death of the mother" (Harvard Law Review, vol. 49, at p. 1040).

I do not think that this decision can be reconciled with any sound principle of liability. Palsgraf v. Long Island R.R.[75], on which the court relied, had very little to do with the case except to illustrate the truism that an act of which A can complain as negligence may not be negligence of which B can complain although B suffers damage and injury.

Professor Magruder seems to justify the decision upon the ground that, "in Waube v. Warrington1(1935) 258 N.W. 497; 216 Wis. 603., the chance that the defendant's negligent driving will actually result in running into someone may be not a great one; the chance that the defendant will not only run over a child in the street, but also cause illness through shock to a person happening to be looking out of the window is even more unlikely" (Harvard Law Review, vol. 49, at p. 1042).

One cannot measure with mathematical precision the relative probability of the contingencies referred to, but I do not think it matters. With roads closely flanked by houses, there must always be a fair probability or possibility that if a small child is found on a road, his mother or father will be found in the vicinity keeping some sort of lookout on the child's movements. If so, a reasonable motorist would certainly direct his mind to the possibility of shock to a mother in the event of carelessness causing injury to her child on the road. If so, a duty to the mother exists and recovery could be had upon proof of physical illness as a direct result of shock caused by seeing the child killed or injured by the passing motorist. Therefore, in Waube v. Warrington[77] liability should have been imputed to the defendant, not because the motorist was negligent in relation to the child, but because he should have anticipated that one possible result of negligent driving in streets used by children was that their mothers would be shocked and terrified by the sight of injuries sustained by the children whether the accident was witnessed by the mother from the street or from the verandah of her house overlooking it, or while she was passing through her front gate or emerging from her front door.

Professor Magruder also suggests:—"Furthermore, if liability were imposed in favour of the mother who from a safe vantage point happened to see her child run over, and thereby suffered shock, it is difficult to suggest any reason why recovery should be denied to a mother who did not see the accident but suffered a similar shock when the bruised body of her child was brought home" (Harvard Law Review, vol. 49, at p. 1042).

I do not see any necessary relationship between the facts in Waube v. Warrington[78] and those suggested above. Of course the principle of ascertaining liability cannot be different in the two cases. But the act of carrying home the bruised body of the child might be the result of advice or action on the part of a number of persons, and, in the usual case, volition and decision would have been exercised for the very purpose of avoiding unnecessary alarm and shock to the mother. All the circumstances would require consideration, and the observations of Lord Wright in Grant v. Australian Knitting Mills Ltd.[79] may well be deemed applicable. There he said:—"But it is clear that such a state of things would involve many considerations far removed from the simple facts of this case. So many contingencies must have intervened between the lack of care on the part of the makers and the casualty that it may be that the law would apply, as it does in proper cases, not always according to strict logic, the rule that cause and effect must not be too remote: in any case the element of directness would obviously be lacking."

Another type of case is where a defendant has negligently caused the disappearance of a child (e.g., at sea) and, after a long search by a parent, the search is abandoned and the body never recovered. I do not think that, if in such circumstances a mother sustained a physical breakdown through the shock caused by the alarm and the search, the defendant should escape liability.

The recent case of Owens v. Liverpool Corporation[80] reinforces the conclusion to be drawn from Hambrook v. Stokes Brothers[81] itself—that, in order to establish liability for nervous shock and illness, it is not necessary for the plaintiff to establish the fact that he saw the actual occurrence of the casualty as distinct from its immediate and injurious results. In Owens' Case[82], only one of the four plaintiffs, all of whom sustained nervous shock upon witnessing the very distressing results of the negligent driving of the tram driver (i.e., the disturbance of the coffin), actually saw the collision between the tramcar and the hearse. That particular plaintiff was awarded damages of only £15, whereas two other plaintiffs, viz., the mother and the cousin, each of whom saw only the effects of the collision, recovered much heavier damages. It is true that as in most cases of road accidents very little time could have elapsed between the actual collision and the witnessing of its consequences by three of the four plaintiffs. But the case shows what should otherwise be clear, that a secondary liability may exist whether the onset of nervous shock coincides with or follows the plaintiff's observation of the disastrous consequences of the defendant's primary negligence.

It seems to me that Owens v. Liverpool Corporation[83] also suggests that if a child is killed through negligent driving and his mother comes upon the body shortly after the moment of death, not being brought to the spot merely for the purpose of viewing or identifying the body after having been informed by a third person of the tragedy, there is no satisfactory ground upon which recovery can be denied provided physical illness through shock has been caused. In all such cases there may be considerable difficulty in proving the direct causation which is necessary to establish a claim for damages.

Again, MacKinnon L.J. was clearly of opinion that it was not necessary to liability to prove that the shock was caused through fear as to the safety of another. If so, it would follow that, in the case I have visualized, the mother could not be denied the right of recovery merely because her realization of disastrous injury to her child synchronized with her certain knowledge of his death. And MacKinnon L.J. went even further by holding that it was not necessary to prove "apprehension as to human safety" and instanced the case of a favourite dog killed by negligent driving. His further statement is also of importance in all these cases: "Fear that unfounded claims may be put forward, and may result in erroneous conclusions of fact, ought not to influence us to impose limitations as to the nature of the facts that it is permissible to prove"[84].

If in the last example I have suggested liability would arise, the case of the present plaintiff is stronger because there is direct evidence here that the onset of her nervous shock commenced before the fatal consequences of the defendant's negligence to the child were evident.

In order to ascertain what is the general principle of liability in cases like the present, considerable assistance is to be derived from a reference to what may be termed "rescue" cases or "search and rescue" cases. Indeed, it is reasonable that the present case should be grouped with the "search and rescue" cases. Most of these cases were decided in the courts of the United States. Because of its subsequent influence upon the course of decision in England, I will refer at once to the very important case of Wagner v. International Railway Co.[85]. There it appeared that the plaintiff and his cousin were passengers on an electric train and, owing to crowding within, were compelled to stand on an outside platform. That platform was provided with doors but these were carelessly allowed to remain open while the train was rounding a curve. There was a violent lurch, and the cousin was, as a consequence, thrown out of the door at the moment when the train was crossing a bridge over the tracks of two other railroads. The cousin was killed, and his body fell from or through the bridge on to the land beneath. Darkness was coming on and the plaintiff, greatly concerned for his cousin's safety, alighted from the train, which had stopped after travelling about 400 feet after crossing the bridge. He then walked back to the bridge, on which he thought he might find his cousin. Other rescuers searched below the bridge and they came across the cousin's dead body at the very moment when the plaintiff (having seen his cousin's hat on a beam of the bridge and having commenced to cross it) missed his footing and fell to the ground beside the other rescuers. The plaintiff sued the railway company alleging a primary breach of duty to his cousin and a secondary breach of duty to himself.

The unanimous judgment of the New York Court of Appeals, delivered by Cardozo J., thus stated the principle of the rescue cases:—

In such cases the rescuer who is injured in endeavouring to save life or to prevent or mitigate the injurious consequences of the defendant's primary breach of duty is also owed a duty by the defendant. The latter may not have foreseen the possibility of rescue, and may, by insurance or otherwise, have merely guarded himself against all liability whatever. But a reasonable person would have foreseen the possibility of rescue.

The primary duty to take reasonable care to avoid inflicting injury coexists with a secondary duty to take care to avoid injuring rescuers. A breach of the primary duty may occur when, to take the examples of Cardozo J., the train is negligently driven round a curve, the opening is left in the bridge, or the train approaches a crossing without warning. In order to show that such a breach of duty has occurred, it is not necessary to prove the actual damage which is essential to the tort of negligence: Cf. per Lord Wright in Lochgelly Iron and Coal Co. v. M'Mullan[87] and per Atkin L.J. in Hambrook v. Stokes Brothers[88]. Where a breach of the primary duty has occurred, there will also be a breach of the secondary duty, although the tort of negligence will be established only upon proof of injury to a particular plaintiff towards whom either or both duties were owed. The same act or omission will constitute a breach of each duty, which breach will usually be completed whether or not any injurious consequences flow therefrom. As Atkin L.J. pointed out in Hambrook v. Stokes Brothers[89], "further, the breach of duty does not take place necessarily when the vehicle strikes or injures the wayfarer. The negligent act or omission may precede the act of injury. In this case it was completed at the top of Dover Street, when the car was left unattended in such a condition that it would run violently down the steep place. Here, then, was a breach of a duty owed to Mrs. Hambrook."

Further, Wagner's Case[90] shows that the rescuer may recover although he acts from deliberate decision rather than spontaneous or instinctive response. Further still—and this is an important consideration in the present case—the rescuer is not debarred from recovery because in the light of after knowledge it is plain that his attempt at rescue could never have aided the victim of the defendant's primary negligence.

In Wagner's Case[92] the search of the plaintiff was futile not only because the body was not on the bridge which the plaintiff attempted to re-cross, but because the cousin was killed by his fall from the train as a result of the lurch. But the plaintiff could not know in advance that intervention would be useless. He could not be sure where exactly his cousin's body had come to rest. It might have been caught by some projection on the bridge, or held by the rails. His expedition was that of search and rescue, and that was enough to establish a relationship of duty between him and the defendant. For the railroad company should have foreseen that if it carelessly allowed train doors to remain open while the train was rounding a curve, a passenger might be thrown out, and in that event the emergency was likely to evoke some response, especially from the victim's relatives or friends on the train. In cases like Wagner's[93], the plaintiff could not, under the law of the State of New York, have recovered for illness caused by nervous shock alone, but this rule as to shock is peculiar to that State. I am clearly of opinion that by the common law of England in analogous cases of "search and rescue," a mother who attempted to find and aid her child, and who suffered illness through shock alone, would not be precluded from recovery merely because of the fact that, when she discovered the dead body some appreciable time after the accident, life was already extinct. Indeed, in such cases nervous shock is more easily foreseen than ordinary physical lesions of the type which occurred in Wagner's Case[94].

In 1934 the principles of law applied in the United States rescue cases and illustrated by Wagner's Case[95] were discussed by Professor Goodhart, who severely criticized certain obiter dicta of the English Court of Appeal in Cutler v. United Dairies (London) Ltd.[96]. In the course of his article, he said:—

Professor Goodhart's prophecy was soon borne out, for, in 1935, in Haynes v. Harwood[97], the principles as explained and elaborated by Professor Goodhart were, in substance, adopted by the Court of Appeal. In that case, a horse van driven by the defendant's servant was negligently left unattended and the horses bolted down a street frequented by people including many children. At the time the plaintiff, a police officer, was in a police station fronting the street and in that position was in no danger whatever. Apprehending serious injury to women and children who were in the path of the bolting horse, he ran out of the station and crossed the street, pushed out of the way a woman who was in grave danger, and succeeded in pulling up the horses. But he was badly injured in doing so. Finlay J. and the Court of Appeal held that he was entitled to recover against the defendant, who was guilty of primary negligence in leaving the horses unattended. I desire to quote one passage from the judgment of Finlay J. because it illustrates how in some cases it is impossible to resist the inference of a pre-existing duty at least towards parents. In Cutler's Case, Slesser L.J. had said:—"There may be cases where, for example, a man sees his child in great peril in the street and, moved by paternal affection, dashes out and holds a runaway horse's head in order to save his child, and is injured; there there is no novus actus interveniens"[98].

While it is extremely difficult in these cases to restrict the duty to "anxious mothers, fathers, wives and husbands" (Salmond on Torts (Ed. Stallybrass), 8th ed. (1934), p. 371), it is obvious that it is almost impossible for a reasonable person to avoid foreseeing intervention by them in search or rescue cases, and suffering by them in case of casualties threatening to a member of their family.

In the Court of Appeal, Greer L.J. emphasized that the vehicle was left unattended in a street where "there are always many children about"[100]. Maugham L.J. furnished an example which, although used by him for a different purpose, also suggests that in the present case the duty of the defendant council existed not only towards the plaintiff, but towards all other persons who were brought to the trench for the purpose of rescuing or assisting the plaintiff's child. He said.:—

Roche L.J., who agreed with the other members of the Court of Appeal, referred to the general principle of ascertaining duty as stated by Lord Atkin in Donoghue v. Stevenson[102] in a passage which I have already quoted.

Haynes v. Harwood[103] and Wagner's Case[104] show that a defendant who is under a duty of care to a class of people including A and whose breach of duty places A in such a situation of peril, or injures A in such circumstances that B, who is not imperilled at all, is induced to come to A's rescue, may become liable to compensate B if he sustains personal injury in the course of his attempts at rescue. Such right of recovery in B necessarily imports a pre-existing duty towards him, although he does not or may not belong to the class of persons who can be foreseen as likely to sustain injury through the defendant's primary breach of duty. This general conclusion is in strict accordance with the principles which are to be deduced from Hambrook v. Stokes Brothers[105] and other cases which I have discussed.

Haynes v. Harwood[106] also illustrates the proposition that it is not necessary to prove that damage has flowed from a breach of what I have called the primary duty before an action can be successfully maintained in respect of a breach of the secondary duty. A defendant may be guilty of negligent conduct which is a breach of both a primary and a secondary duty, and damage may be sustained either by a person to whom the primary duty is owed, or by a person to whom the secondary duty is owed, or by neither, or by both. In the event of injury being sustained as a result of the breach of one of the duties, damages could be recovered in respect of that breach, and the question whether or not actual injury had been sustained as a result of the breach of the other duty would seem to be immaterial. Countless examples could be given of cases in which damages have been recovered in respect of the breach of a primary duty though inasmuch as no injury has been sustained by any of the persons to whom the secondary duty is owed, there has been no action, and there could be no action. Haynes v. Harwood[107] is a case in which injury was sustained as a result of the breach of the secondary duty and damages were recovered, even though no injury resulted from the breach of the primary duty, and, consequently, no action could be brought. It would seem to follow that damages may be recovered for physical injury resulting from nervous shock at the reasonable fear of injury to another even although the person whose safety is thought to be threatened suffers no actual injury.

A nexus between the Hambrook v. Stokes Brothers[108] type of case and the search and rescue type of case was suggested in 1925 by an able commentator (Law Quarterly Review, at pp. 122-125). In truth, both classes of case are fairly clear applications of Lord Atkin's statement of principle in Donoghue v. Stevenson[109], and since the recent case of Owens v. Liverpool Corporation[110] this seems to have been recognized in academic circles in the United States. Hence the comment in the Harvard Law Review, vol. 52, at pp. 844, 845:—"In cautious retreat, courts have hesitated to apply the test of forseeability of risk, and have adopted arbitrary rules limiting recovery to certain factual situations. ... The present decision is commendable in applying the test of forseeability of risk to this type of case; but this test creates a new and broad field of recovery, and courts should be cautious lest they unduly burden activity."

I appreciate the risk of too wide an extension of liability in cases where proof is beset with special difficulties, but I am of opinion that in the present case the plaintiff is clearly entitled to have her claim considered by the jury and to recover damages upon proof that her shock and illness were caused by emotional distress caused by the circumstances existing from the moment when her search brought her to the trench up to the time when her child's body was removed therefrom. It is evident that a reasonable person in the defendant's position would have foreseen resort to the trench both by persons rescuing a child endangered by the failure to guard the trench and by persons seeking a child with intent to rescue.

If my examination of the cases is sound, the plaintiff is not disentitled to recover merely because she came to the scene of the fatality after her child had fallen into the trench, provided that her shock and suffering were due in the main to what she realized from her own unaided senses during the period I have defined. Nor is the plaintiff disentitled to recover because, as it turned out, the onset of shock was probably at a moment of time subsequent to that of her child's death. This fact should no more defeat her case than if, searching for her child throughout the afternoon, she finally came upon his dead and disfigured body left lying in some lane by a "hit and run" motorist who had negligently caused his death. But the plaintiff's case is stronger than that suggested because here it is unreasonable to divide the transaction so as to separate the removal of the child's body from his original fall into the trench.

The following statement of subordinate principles might now be suggested. I am quite aware that they may require qualification and elaboration, but at present they seem to be in substantial conformity with the cases and the principles asserted or implied therein, and, when examined, they all seem to be reasonable or even necessary applications of cases like Donoghue v. Stevenson[111] and Grant v. Australian Knitting Mills Ltd.[112].

Not only do the above principles as to ascertainment of duty apply to permit of the plaintiff's recovery here, but by fair analogy the duty which was owed to her by the present defendant is at least as clear as that owed to the policeman in Haynes v. Harwood[114], to the wife in Brandon v. Osborne Garrett & Co. Ltd.[115], to the travelling companion of the injured passenger in Wagner's Case[116], and to the mourners in Owens v. Liverpool Corporation[117]. So far as damages are concerned, there is ample evidence that those sustained by the present plaintiff flowed directly from the nervous shock received in circumstances where the defendant's duty was operating.

Two other points should be mentioned. In the first place, the Supreme Court suggested that the plaintiff's present action was brought because under the New South Wales Compensation to Relatives Act (Lord Campbell's Act) no liability could be visited upon the defendant because, owing to the dead child's tender years, the probability of pecuniary loss could not be asserted, and because ever since Blake v. Midland Railway Co.[118] it has been established that under Lord Campbell's Act no solatium is recoverable for deprivation of a beloved relative. No doubt this comment is true so that, unless the plaintiff recovers in the present action, the defendant council would be little worse off for its negligence, in case the jury find negligence. But the considerations are as irrelevant as is the argument rejected alike in Donoghue v. Stevenson[119] and Grant v. Australian Knitting Mills Ltd.[120] to the effect that because there is no privity of contract between the manufacturer and the ultimate consumer, and therefore no remedy for breach of contract, the latter can have no remedy in tort. The Full Court's comment may provoke the further comment that it is high time that the New South Wales Compensation to Relatives Act should be brought into line with the statute law of England by preventing persons who negligently cause the death of young children from escaping liability. However that may be, the existing legislative provision as to the measure of damages should not prevent Australian courts from applying the common law of England in accordance with the principles adopted by the courts of England; principles which are not to be rejected or evaded merely because they have introduced into the law an element of humanity and common sense alike.

The second point is of great importance because it involves reference to a decision of the Privy Council which is still binding on all Dominion courts. It is contended by the defendant that the Privy Council decision in Victorian Railways Commissioners v. Coultas[121] decided over fifty years ago should be regarded as creating an impassable obstacle to the plaintiff's claim. In that case Mrs. Coultas suffered a severe nervous shock caused, according to the statement of the Privy Council, "by seeing the train approaching, and thinking they were going to be killed"[122]. She was held disentitled to recover.

I am of opinion that the Privy Council decision should not be regarded as ruling that in the absence of physical impact to the plaintiff, damages for illness caused by nervous shock can never be recovered. It appears that two main questions of law were reserved, viz.:—1. Whether the damages awarded by the jury to the plaintiffs, or either of them, are too remote to be recovered. 2. Whether proof of "impact" is necessary in order to entitle the plaintiffs to maintain the action.

The decision of the Privy Council answered question 1 only. It was said that "damages arising from mere sudden terror unaccompanied by any actual physical injury, but occasioning a nervous or mental shock, cannot under such circumstances, their Lordships think, be considered a consequence which, in the ordinary course of things, would flow from the negligence of the gatekeeper"[123], and also that "they are of opinion that the first question, whether the damages are too remote, should have been answered in the affirmative, and on that ground, without saying that impact is necessary, that the judgment should have been for the defendants"[124].

It is plain, therefore, that the decision laid down only that damages due to "mere sudden terror unaccompanied by any actual physical injury" were too remote. It must always be a question of fact whether shock to the nerves causes "actual physical injury." To-day it is known that it does. In 1888 it was widely assumed that it did not. Even so, the Privy Council seems to have sensed the possibility of advance in medical knowledge, for otherwise they would have proceeded to answer the second question as to "impact." But they expressly left open the question whether, in the case of physical injury due to sudden alarm (without any impact at all) liability might not ensue.

It is on this basis that Coultas's Case[125] is to be understood, and if so understood it has no application to cases like the present where "shock to the nerves" is another name for actual physical disturbance to the nervous system. Therefore, the decision is no bar to recovery here. In England the case has not been regarded as preventing recovery in analogous cases, and I need only refer to the observations upon it by Atkin L.J. in Hambrook v. Stokes Brothers[126]. As showing the advance since 1888 of medical knowledge as to nervous shock, I would like to quote the following passage from a valuable comment in the Canadian Bar Review, vol. 11, at pp. 516, 517:—

Professor Winfield has stated that the Coultas decision[129] "proceeded upon erroneous ideas about pathology and the other courts of this country and the Irish Exchequer Division have persistently declined to follow it. The fallacy lay in supposing that bodily or physical injury must exclude mental injury as being too remote. It would be difficult to contend that damages ought to be given for the mere sensation of fear, but when fear or any other sensation produces a definite illness why should that illness be any more remote than a broken bone or an open wound" (Winfield on the Law of Tort, at p. 86).

Because of the great importance of the present case I have set out my reasons at length. I think that the plaintiff was entitled to have her claim considered by the jury and that the trial judge erred in nonsuiting her. It follows that in my view the judgment and order of the Supreme Court should be set aside and a new trial of the action should be had. Judgment on the defendant's demurrer (which was filed solely in aid of the legal contention which succeeded on the appeal to the Full Court) should be for the plaintiff. Inasmuch as the defendant is entirely responsible for preventing the issues of fact being tried, I would order it to pay the plaintiff's costs of the first trial in any event as well as those of the appeal and demurrer before the Full Court, and those before this court applicable to an appeal in forma pauperis.

[1] (1921) 3 K.B. 560, at p. 574.

[2] (1921) 3 K.B. 560, at p. 574.

[3] (1921) 3 K.B., at p. 577.

[4] (1932) A.C. 562.

[5] (1932) A.C., at p. 580.

[6] [1937] HCA 5; (1937) 57 C.L.R. 1.

[7] (1913) 1 K.B. 398.

[8] (1907) S.C. 1102.

[9] (1934) 151 L.T. 30, at p. 35.

[10] (1925) 1 K.B. 141.

[11] (1925) 1 K.B., at p. 152.

[12] (1925) 1 K.B., at p. 160.

[13] (1939) 1 K.B. 394; 55 T.L.R. 246.

[14] (1921) 3 K.B. 561.

[15] (1939) 1 K.B., at pp. 400, 401.

[16] (1921) 3 K.B., at pp. 574, 579.

[17] (1932) A.C., at p. 580.

[18] (1937) 57 C.L.R., at p. 15.

[19] (1937) 57 C.L.R., at pp. 15, 17.

[20] [1936] HCA 65; (1936) 57 C.L.R. 259.

[21] [1936] HCA 65; (1936) 57 C.L.R. 259.

[22] (1932) 146 L.T. 391, at p. 392; 48 T.L.R. 215, at p. 217.

[23] (1925) 1 K.B., at p. 158.

[24] (1939) 1 K.B. 394.

[25] (1932) 146 L.T., at p. 392; 48 T.L.R., at p. 217.

[26] (1932) 146 L.T., at p. 392; 48 T.L.R., at p. 217.

[27] [1937] HCA 5; (1937) 57 C.L.R. 1.

[28] (1888) 13 App. Cas. 222.

[29] (1925) 1 K.B. 141.

[30] (1925) 1 K.B., at p. 152.

[31] (1925) 1 K.B. 141.

[32] (1925) 1 K.B., at p. 158.

[33] (1925) 1 K.B. 141.

[34] (1925) 1 K.B. 141.

[35] (1938) 38 S.R. (N.S.W.), at pp. 607, 608; 55 W.N. (N.S.W.), at p. 223.

[36] (1925) 1 K.B. 141.

[37] (1925) 1 K.B. 141.

[38] (1932) A.C. 562.

[39] (1932) A.C., at p. 580.

[40] (1925) 1 K.B. 141.

[41] (1925) 1 K.B., at pp. 157, 158.

[42] (1925) 1 K.B., at pp. 158, 159.

[43] (1925) 1 K.B., at p. 159.

[44] (1925) 1 K.B. 141.

[45] (1925) 1 K.B. 141.

[46] (1932) A.C. 562.

[47] [1935] UKPCHCA 1; (1936) A.C. 85; (1935) 54 C.L.R. 49.

[48] (1901) 2 K.B., at p. 679.

[49] (1932) A.C. 562.

[50] (1939) 1 K.B., at pp. 400, 401.

[51] (1901) 2 K.B. 669, at p. 684.

[52] (1939) 1 K.B. 394.

[53] [1937] HCA 5; (1937) 57 C.L.R. 1.

[54] (1937) 57 C.L.R., at p. 7.

[55] (1937) 57 C.L.R., at p. 17.

[56] (1937) 57 C.L.R., at pp. 17, 18.

[57] [1937] HCA 5; (1937) 57 C.L.R. 1.

[58] [1937] HCA 5; (1937) 57 C.L.R. 1.

[59] (1921) 3 K.B. 560.

[60] (1921) 3 K.B. 560.

[61] (1921) 3 K.B. 560.

[62] (1921) 3 K.B. 560.

[63] (1925) 1 K.B. 141.

[64] (1925) 1 K.B. 141.

[65] (1925) 1 K.B. 141.

[66] (1925) 1 K.B. 141.

[67] (1925) 1 K.B., at p. 142.

[68] (1925) 1 K.B., at p. 157.

[69] (1925) 1 K.B., at p. 159.

[70] (1925) 1 K.B., at p. 152.

[71] (1936) A.C., at p. 107; (1935) 54 C.L.R., at p. 67.

[72] (1925) 1 K.B. 141.

[73] (1935) 258 N.W. 497; 216 Wis. 603.

[74] (1935) 258 N.W. 497; 216 Wis. 603.

[75] (1928) 248 N.Y. 339.

[76] (1935) 258 N.W. 497; 216 Wis. 603.

[77] (1935) 258 N.W. 497; 216 Wis. 603.

[78] (1935) 258 N.W. 497; 216 Wis. 603.

[79] (1936) A.C., at p. 107; (1935) 54 C.L.R. at p. 67.

[80] (1939) 1 K.B. 394.

[81] (1925) 1 K.B. 141.

[82] (1939) 1 K.B. 394.

[83] (1939) 1 K.B. 394.

[84] (1939) 1 K.B., at p. 400.

[85] (1921) 232 N.Y. 176.

[86] (1921) 232 N.Y., at p. 180.

[87] [1933] UKHL 4; (1934) A.C. 1, at p. 23.

[88] (1925) 1 K.B., at p. 156.

[89] (1925) 1 K.B., at pp. 156, 157.

[90] (1921) 232 N.Y. 176.

[91] (1921) 232 N.Y., at p. 181.

[92] (1921) 232 N.Y. 176.

[93] (1921) 232 N.Y. 176.

[94] (1921) 232 N.Y. 176.

[95] (1921) 232 N.Y. 176.

[96] (1933) 2 K.B. 297.

[97] (1934) 2 K.B. 240; (1935) 1 K.B. 146.

[98] (1933) 2 K.B., at p. 306.

[99] (1934) 2 K.B., at p. 250.

[100] (1935) 1 K.B., at p. 151.

[101] (1935) 1 K.B., at p. 164.

[102] (1932) A.C., at p. 580.

[103] (1934) 2 K.B. 240; (1935) 1 K.B. 146.

[104] (1921) 232 N.Y. 176.

[105] (1925) 1 K.B. 141.

[106] (1934) 2 K.B. 240; (1935) 1 K.B. 146.

[107] (1934) 2 K.B. 240; (1935) 1 K.B. 146.

[108] (1925) 1 K.B. 141.

[109] (1932) A.C. 562.

[110] (1939) 1 K.B. 394.

[111] (1932) A.C. 562.

[112] [1935] UKPCHCA 1; (1936) A.C. 85; (1935) 54 C.L.R. 49.

[113] (1939) 1 K.B. 394.

[114] (1934) 2 K.B. 240; (1935) 1 K.B. 146.

[115] (1924) 1 K.B. 548.

[116] (1921) 232 N.Y. 176.

[117] (1939) 1 K.B. 394.

[118] [1852] EngR 10; (1852) 18 Q.B. 93 [118 E.R. 35].

[119] (1932) A.C. 562.

[120] [1935] UKPCHCA 1; (1936) A.C. 85; (1935) 54 C.L.R. 49.

[121] (1888) 13 App. Cas. 222.

[122] (1888) 13 App. Cas., at p. 225.

[123] (1888) 13 App. Cas., at p. 225.

[124] (1888) 13 App. Cas., at p. 226.

[125] (1888) 13 App. Cas. 222.

[126] (1925) 1 K.B. 141.

[127] (1888) 13 App. Cas. 222.

[128] (1888) 13 App. Cas. 222.

[129] (1888) 13 App. Cas. 222.

High Court of Australia

Chester v Waverley Municipal Council [1939] HCA 25; (1939) 62 CLR 1 (6 June 1939)

HIGH COURT OF AUSTRALIA