United Kingdom House of Lords Decisions

You are here: BAILII >> Databases >> United Kingdom House of Lords Decisions >> White and Others v. Chief Constable of South Yorkshire and Others [1998] UKHL 45 (3 December 1998)
URL: http://www.bailii.org/uk/cases/UKHL/1998/45.html
Cite as: (1999) 45 BMLR 1, [1998] 3 WLR 1509, [1999] 2 AC 455, [1998] UKHL 45, [1999] 1 All ER 1, [1999] ICR 216, [1999] IRLR 110

[New search] [Buy ICLR report: [1999] ICR 216] [Buy ICLR report: [1998] 3 WLR 1509] [Buy ICLR report: [1999] 2 AC 455] [Help]

JISCBAILII_CASE_TORT

White and Others v. Chief Constable of South Yorkshire and Others [1998] UKHL 45; [1999] 2 AC 455; [1999] 1 All ER 1 (3rd December, 1998)

HOUSE OF LORDS

Lord Goff of Chieveley Lord Griffiths Lord Browne-Wilkinson
Lord Steyn Lord Hoffmann

OPINIONS OF THE LORDS OF APPEAL FOR JUDGMENT IN THE CAUSE

WHITE AND OTHERS
(RESPONDENTS)

CHIEF CONSTABLE OF SOUTH YORKSHIRE AND OTHERS
(APPELLANTS)

ON 3 DECEMBER 1998

LORD BROWNE-WILKINSON

My Lords,

I have read in draft the speeches of my noble and learned friends, Lord Steyn and Lord Hoffmann. I agree that for the reasons they give these appeals should be allowed and the actions dismissed.

LORD GRIFFITHS

My Lords,

I have had the advantage of reading the speeches of your Lordships before giving my own opinion. In those speeches are cited all the relevant authorities that trace the development of the common law's attitude to psychiatric injury. They show that the common law has regarded claims for psychiatric injury with caution and has not until very recent times been prepared to treat them as on a par with physical injury. I do not think I shall serve any useful purpose by travelling over that historic ground again in this opinion, and I shall take as my starting point the two recent decisions of the House of Lords as stating the present state of the common law.

In Alcock v. Chief Constable of South Yorkshire Police [1992] 1 AC 310, claims were brought by those who had suffered psychiatric injury as a result of the Hillsborough disaster. Two of the plaintiffs were spectators in the ground, but not in the pens where the disaster occurred, the remainder of the plaintiffs learned of the disaster through radio or television broadcasts. All the plaintiffs lost, or feared they might have lost, a relative or fiance in the disaster. Thus it will be seen that two of the plaintiffs were witnesses to the disaster, but not in peril themselves, the remainder were not in the ground when the accident occurred. One of the plaintiffs gave some assistance to the injured but the case was not argued on the basis that he should be treated as a rescuer. The two plaintiffs at the ground were treated as bystanders who witnessed the disaster. All the plaintiffs lost their cases because they did not fulfil one or other of the control mechanisms, all of which the present law requires in cases where damages for psychiatric injury are claimed by plaintiffs who were not directly threatened by the accident but learned of it through sight or hearing of it. These control mechanisms are as Lord Hoffmann sets out in his opinion.

"1. There must be a close tie of love and affection between the plaintiff and the victim. 2. The plaintiff must have been present at the accident or its immediate aftermath. 3. The psychiatric injury must have been caused by direct perception of the accident or its immediate aftermath and not by hearing about it from somebody else."

There is a further requirement in the bystander case and that is that psychiatric injury was reasonably foreseeable as a likely consequence of exposure to the trauma of the accident or its immediate aftermath. The law expects reasonable fortitude and robustness of its citizens and will not impose liability for the exceptional frailty of certain individuals. This is not to be confused with the "eggshell skull" situation, where as a result of a breach of duty the damage inflicted proves to be more serious than expected. It is a threshold test of breach of duty; before a defendant will be held in breach of duty to a bystander he must have exposed them to a situation in which it is reasonably foreseeable that a person of reasonable robustness and fortitude would be likely to suffer psychiatric injury. However, as in the situation we are considering, namely the bystander who is seeing or hearing of the event from a safe distance, the only injury that he could suffer would be psychiatric injury so we can substitute personal injury for psychiatric injury, and this will fit more easily with Page v. Smith [1996] AC 155 to which I now turn.

In Page v. Smith the plaintiff was driving his car at 30 miles an hour when the defendant turned right immediately into his path. The consequence was an accident in which both cars suffered considerable damage but the occupants all escaped physical injury. The Plaintiff, however, had suffered for 20 years from a condition known as chronic fatigue syndrome, which manifested itself from time to time. The judge held that the shock of the accident reactivated this condition which was now in all probability permanent and that it was unlikely that the plaintiff would be able to return to full-time employment, and he awarded damages of £162,153. The Court of Appeal allowed the defendants appeal on the ground that psychiatric injury was not a foreseeable consequence of the accident. The House of Lords by a majority held that in circumstances such as a road accident in which a defendant owes a duty of care not to cause personal injury it mattered not whether the injury suffered as a result of the defendant's negligence was physical injury or psychiatric injury and liability would be established without the necessity to prove as an independent part of the cause of action that psychiatric injury, in the absence of physical injury, was foreseeable.

For my part I regard this as a sensible development of the law and I note from the Law Commission Report on Liability for Psychiatric Illness (1998) (Law Com. No. 249) that it has been supported by the majority of practitioners. If some very minor physical injury is suffered and this triggers a far more serious psychiatric disorder no one questions that damages are recoverable for the psychiatric disorder. If the victim of the negligence escapes minor physical injury but the shock or fear of the peril in which he is placed by the defendant's negligent conduct causes psychiatric injury I can see no sensible reason why he should not recover for that psychiatric damage.

As medical science advances we realise how difficult it is to separate out the physical and psychiatric consequences of trauma, and I believe the law would do better to regard both as personal injury as Page v. Smith requires in the case of primary victims, that is victims who are imperilled or reasonably believe themselves to be imperilled by the defendant's negligence. Insofar as secondary victims are concerned, that is those who are bystanders, we can still ask the question was personal injury reasonably foreseeable if we regard psychiatric damage as personal injury. For as I have pointed out no question of physical injury arises in the case of the bystander.

Having set out my understanding of the present state of the law I turn to consider the position of the plaintiffs in the appeal. They were all police officers and it is agreed that as such they are to be considered for the purposes of the law as though they were employees of the Chief Constable. The case for the police officers is put in two ways. First it is said that they are entitled to recover for a breach of the duty owed to them as employees and secondly that at least some of them are entitled to recover as rescuers.

Their case as employees is put thus: they were all at the ground in the course of their duty, as employees. The Chief Constable owed them a duty to take reasonable care not to expose them to unnecessary risk of injury during the course of their employment. The Chief Constable is vicariously liable for the negligence of the police officer who caused the catastrophe by admitting the crowd in to the pens. It was the impact of the horror of the situation on the minds of the police officers that caused them psychiatric injury. By the negligent creation of the horrific situation the Chief Constable was in breach of his duty not to expose the police to unnecessary risk of injury and is consequently liable for their injuries.

If this approach is right it means that the police will be entitled to recover damages whereas spectators and others on duty in the ground who were exposed to the same horror and risk of psychiatric injury will not be able to do so. I can not believe that this would be a fair or acceptable state of the law. If anything one would expect the police to be at a disadvantage. The police are trained to deal with catastrophic incidents and reasonably well compensated under the terms of their service if they do suffer injury in the course of their duties.

The law of master and servant is not a discrete and separate branch of the law of tort, but is to be considered in relation to actions in tort generally. Here we are considering the tort of negligence and the nature of the duty of care owed by one who negligently creates a catastrophic situation. In order that there shall be some limits to the consequence of the negligence for which the defendant is to be made liable the law imposes the controls I have discussed in Alcock.

In my view these should apply to all those not directly imperilled or who reasonably believe themselves to be imperilled, irrespective of whether they are employees or not. Accordingly, I would allow the appeals insofar as the police rely upon their status as employees.

I turn last to the special category known as "rescuers." If a tort feasor creates a dangerous situation he can foresee that others will attempt to rescue the victims, or potential victims of his negligence. It is well settled that if a rescuer suffers physical injury in the rescue attempt he will be entitled to damages from the tort feasor. If it is foreseeable that the rescuer may suffer personal injury in the form of psychiatric injury rather than physical injury, why should he not recover for that injury. The fear is expressed that if foreseeability of psychiatric injury is sufficient it will open the floodgates to claims, many of an unmeritorious kind, from those who give assistance at any accident. I believe that the courts are well capable of controlling any such flood of claims. Whether or not a person is to be regarded as a rescuer will be a question of fact to be decided on the particular facts of the case. Trivial or peripheral assistance will not be sufficient see McFarlane v. E.E. Caledonian Limited [1994] 2 All E.R.1.

If the rescuer is in no physical danger it will only be in exceptional cases that personal injury in the form of psychiatric injury will be foreseeable for the law must take us to be sufficiently robust to give help at accidents that are a daily occurrence without suffering a psychiatric breakdown. But where the accident is of a particular horrifying kind and the rescuer is involved with the victims in the immediate aftermath it may be reasonably foreseeable that the rescuer will suffer psychiatric injury as Mr. Chadwick did when trying to bring relief and comfort to the victims of the Lewisham train disaster. Mr. Chadwick suffered his injury because of the terrible impact on his mind of the suffering he witnessed in his rescue attempt, and not because of any fear for his own safety, see Chadwick v. British Railways Board [1967] 1 W.L.R. 912. What rescuer ever thinks of his own safety? It seems to me that it would be a very artificial and unnecessary control, to say a rescuer can only recover if he was in fact in physical danger. A danger to which he probably never gave thought, and which in the event might not cause physical injury.

A line has to be drawn in rescue cases between rescue in the sense of immediate help at the scene of the disaster, and treatment of the victims after they are safe. I do not believe that this will be difficult to recognize on the facts of a particular case.

After a careful analysis of the evidence, Rose L.J. identified three of the police officers as rescuers. I would myself dismiss the appeals in respect of those officers, namely White, Bairstow and Bevis. I would add that I do not share the view that the public would find it in some way offensive that those who suffered disabling psychiatric illness as a result of their efforts to rescue the victims should receive compensation, but that those who suffered the grief of bereavement should not. Bereavement and grief are a part of the common condition of mankind which we will all endure at some time in our lives. It can be an appalling experience but it is different in kind from psychiatric illness and the law has never recognized it as a head of damage. We are human and we must accept as a part of the price of our humanity the suffering of bereavement for which no sum of money can provide solace or comfort. I think better of my fellow men than to believe that they would, although bereaved, look like dogs in the manger upon those who went to the rescue at Hillsborough.

I would, however, allow the appeal in respect of P.C. Glave who was not a rescuer, but relied only upon his status as an employee.

LORD GOFF OF CHIEVELEY

My Lords,

These appeals arise from further proceedings following the tragic events which occurred at the Hillsborough Football Stadium in Sheffield on 15 April 1989, when 95 spectators died and hundreds more were injured, one fatally, as a result of crushing sustained in Spectator Pens 3 and 4 at the Leppings Lane end of the Stadium. The immediate cause of the disaster was a senior police officer's decision at 2.32 p.m. to open an outer gate (gate C) without cutting off access to Pens 3 and 4. As a result, spectators in those Pens suffered crushing as more spectators entered the ground through gate C.

The present case is concerned with claims by members of the South Yorkshire Police Force who were on duty at Hillsborough that afternoon, and who claim to have suffered psychiatric damage in consequence. Of the 52 serving police officers who commenced proceedings, 15 plaintiffs appear to have abandoned their actions, and the defendants consented to judgment in the case of 14 plaintiffs who went into Pens 3 and 4 and were actively engaged in the removal of fans who were being crushed. Of the remainder, the cases of six plaintiffs who performed different tasks on the afternoon of the Hillsborough tragedy were selected for trial on the issue of liability, for which purpose it was admitted that these plaintiffs had suffered psychiatric damage. The six plaintiffs, and their ranks at the relevant time, are Inspector Henry White, Police Constable Mark Bairstow, Police Constable Anthony Bevis, Police Constable Geoffrey Glave, Sergeant Janet Smith, and Detective Constable Ronald Hallam. There are three defendants to the proceedings, who have admitted that the deaths and physical injuries suffered by those in Pens 3 and 4 occurred as a result of their negligence. The issue of liability has however been tried with reference only to the first defendant, the Chief Constable of South Yorkshire. There was no dispute as to what the six plaintiffs saw and did at the Hillsborough Stadium on the day of the disaster. This is set out in written statements of the plaintiffs, to which I will have to refer in due course.

The issue of liability came on for trial before Waller J. Evidence was restricted to the agreed statements and medical evidence, and a short section of agreed video film footage shown to the court. The issue of causation was not dealt with, it being agreed that if, by reason of the Court's decision, it arose, the issue would be remitted to a trial judge. Waller J. dismissed the claims of all six plaintiffs. All of them except Mr. Hallam appealed to the Court of Appeal. Although Mr. Hallam did not appeal, the Court of Appeal was invited to rule whether there was a breach of duty to police officers in the position of Mr. Hallam. The Court of Appeal (Rose and Henry L.JJ., Judge L.J. dissenting) allowed the appeals of all except Miss Smith, and held that persons in the position of Mr. Hallam too should be entitled to succeed. Judge L.J. would have dismissed all the appeals. It is against that decision that the defendants now appeal to your Lordships' House, with the leave of this House. Miss Smith has not appealed.

Waller J.

There are two essential strands in Waller J.'s judgment.

(1) He rejected the argument that the plaintiffs could recover damages as primary victims simply on the basis that the Chief Constable was in breach of the duty of care owed by him to the police officers which was analogous to that owed by an employer to his employees. He recognised that there were cases in which an employee could recover damages from his employer in respect of psychiatric injury caused by breach of the latter's duty of care. But in his view the position of a Chief Constable was quite different from that of an ordinary employer. It could not be said to be a Chief Constable's duty not to expose a police officer to injury by nervous shock; indeed there will be many situations in which a Chief Constable will deploy officers at incidents which will be horrific and which will thus carry the risk of nervous shock. Here there was no allegation that there was any breach of duty in deploying the officers at the scene either at the beginning of the match or once the disaster had commenced. The case rested purely on the vicarious liability of the senior officers in relation to causing the incidents which killed and injured victims. In this context, the plaintiffs were secondary victims; and the nature of their relationship with the Chief Constable did not give them an advantage over bystanders whose presence was clearly foreseeable, unless it was by virtue of carrying out an operation such as rescue.

(2) So far as rescue was concerned, police officers must be regarded as professional rescuers. They will not be persons of ordinary phlegm, but of extraordinary phlegm hardened to events which would to ordinary persons cause distress; and, if their activity of rescuing is to ground recovery, it must make it just and reasonable that they should recover when bystanders should not. There should be something akin to the fireman's rule so far as psychiatric damage is concerned. The activity and involvement in the incident or its immediate aftermath must be such as to make it fair and reasonable that the plaintiff should recover when a bystander would not. In particular, "immediate" should be construed narrowly; it was unlikely that it should cover anybody not attending the actual scene.

As a result, he held that all the plaintiffs' claims must fail. None of them could establish that he was a primary victim simply by reference to the relationship between himself and the Chief Constable. None of them qualified as a rescuer, except Inspector White, who joined the end of a line bringing victims out of the Pens. Even so, he was not performing a task which would make it just and reasonable to place him within the area of proximity when a spectator who simply viewed the horrific scene would not be. In addition, it was doubtful whether the psychiatric injury suffered by the plaintiffs could be described as "shock-induced," if that was (as the Judge thought) a necessary requirement.

The Court of Appeal

In the Court of Appeal Rose L.J. first of all singled out rescuers as a special category, pointing out in particular that in Alcock [1992] 1 AC 310 Lord Oliver of Aylmerton placed the rescue cases in his first group of nervous shock cases in which the plaintiff was involved as a participant, rather than in his second group in which the plaintiff was no more than a passive and unwilling witness of injury caused to others: see [1992] 1 AC 310, 407C-408G. Whether a particular person is a rescuer is a question of fact to be decided in the light of all the circumstances of the case. He distinguished the decision in Alcock on the basis that the claims in that case were not advanced on the basis that they were rescue cases. On the facts of the cases before him, he held that three of the claimants were entitled to succeed on the basis that they were rescuers, viz. P.C. Bevis, P.C. Bairstow and Inspector White.

Rose L.J. however also held that in the master and servant context a duty of care exists by reason of that relationship; and that an employee may depending on the circumstances recover against his employer for physical or psychiatric injury caused in the course of his employment by the employer's negligence. On this basis he held that P.C. Glave, and those in the position of D.C. Hallam, were entitled to recover because they were at the ground in the course of duty, within the area of risk of physical or psychiatric injury and were thus exposed, by the first defendant's negligence, to excessively horrific events such as were likely to cause psychiatric illness even in a police officer.

Henry L.J. agreed with Rose L.J. on the issue of rescuers; but he devoted his judgment to deciding that those police officers who were directly involved as active participants were entitled to recover as employees. I cannot do full justice to Henry L.J.'s judgment in a summary. But in brief he concluded that those police officers who were directly involved were primary victims because they were active participants in the incidents caused by their employer's negligence, and that they were direct victims because their employer owed them a duty of care to protect them from personal injury, including psychiatric damage, caused by his negligence. Furthermore there were no public policy reasons why they should not succeed in their claims. In the result, therefore, he agreed with Rose L.J. as to the disposal of the appeals on the issue of employer's liability, as he did on the issue of rescue.

Judge L.J. dissented. He was much influenced by the decision of your Lordships' House in Page v. Smith [1996] AC 155, and in particular by passages in the opinion of Lord Lloyd of Berwick in which he stressed the need to distinguish between primary and secondary victims, and described a primary victim as being within the range of physical injury. His conclusion was that neither those who claim as rescuers, nor those who claim as employees, should necessarily be regarded as primary victims. None of the plaintiffs was at any time present in an area where he or she was exposed to the risk (actual or apprehended) of physical injury arising from the chief constable's negligence. The plaintiffs were therefore all secondary victims to whom the control mechanisms applied. In the case of none was the necessary proximity of relationship established; and, with the arguable exception of Inspector White, in the case of all the necessary proximity of time and place was also absent. Moreover there was no better basis for concluding that psychiatric injury was foreseeable in the case of any of these plaintiffs than it was for the plaintiffs all of whose claims failed in Alcock [1992] 1 AC 310. He would therefore have dismissed all the appeals.

The main principles applicable to claims for damages in tort (i.e., in negligence) in respect of psychiatric injury

In the present case we are concerned with claims for liability in negligence in respect of psychiatric injury suffered by persons who claim that they fall into one or other, or both, of two categories--first, the category of employee, and second the category of rescuer. I shall have in due course to examine each of these two categories in some detail. But it is, of course, impossible to consider them in isolation. In order to understand them properly, we must place them within a framework of legal principle. Only if we do that can we understand why these categories exist and what their function is, and identify what are the principles of law applicable to them.

I shall first outline these principles as generally understood before the decision of your Lordships' House in Page v. Smith [1996] 1 AC 155. In doing so, and indeed in writing the whole of this opinion, I have been much assisted by the Law Commission's Report No.249 dated March 1998 on Liability for Psychiatric Illness. There are two basic principles. These are:

(1) The Plaintiff must have suffered psychiatric injury in the form of a recognised psychiatric illness

The function of this principle is to exclude claims in respect of normal emotions such as grief or distress. Since it is not in issue that the claimants in the cases presently under appeal did indeed suffer from such an illness, viz. post-traumatic stress disorder (P.T.S.D.), I need say no more on this subject.

(2) Damage to the plaintiff in the form of psychiatric injury must have been reasonably foreseeable by the defendant

Here the central question relates to the nature of the foreseeability which is required to render the defendant liable. The development of the law on this subject has been recounted so often that it is unnecessary for me to repeat it yet again in this opinion. Until Page v. Smith, it was generally understood that what is required in all cases of this kind is foreseeability of psychiatric injury, which used to be called injury by shock. How this came about is summarised most clearly by Mullany and Handford in their scholarly and comprehensive treatise on Tort Liability for Psychiatric Damage: The Law of Nervous Shock (1993), to which I wish to express my indebtedness. They state, at pp. 69-70:

"In the early shock cases the need for foreseeability of injury by shock was not made clear. The courts were most hesitant to recognise shock as a kind of damage in its own right, and even after repudiating the need for contemporaneous physical impact retained, for a time, the requirement that the plaintiff must be within the area of possible injury by impact . . . rather than by shock--a theory which has been labelled the 'impact theory. . . .'"

Hambrook v. Stokes Bros.

For this last proposition, over 20 cases (from this country, Canada and Australia) are cited. The same understanding of the legal position was expressed by the Law Commission in their Consultation Paper on Liability for Psychiatric Illness (1995) (Law Com. No. 137) at paras.2.3 and 2.9-11.

The formulation of this principle is attributable to a much-quoted statement by Denning L.J. in King v. Phillips [1953] 1 Q.B. 429, 441, that "there can be no doubt since Bourhill v. Young that the test of liability for shock is foreseeability of injury by shock." This principle has been accepted on numerous occasions, but most prominently by Viscount Simonds when delivering the judgment of the Privy Council in Overseas Tankship (U.K.) Ltd. v. Morts Dock & Engineering Co. Ltd. (The Wagon Mound No. 1) [1961] AC 388, 426. In that case, when differentiating damage by fire from other types of physical damage to property for the purposes of liability in tort, he said (at p. 426):

Bourhill v. Young

[1943] AC 92

King v. Phillips

liability for shock

injury by shock

The principle of foreseeability of damage by shock, or psychiatric injury as it is now more correctly described, has been held to be subject to the qualification that, where the psychiatric injury suffered by the plaintiff is consequential upon physical injury for which the defendant is responsible in law, the defendant will be bound to compensate the plaintiff in respect of the former even if unforeseeable (see Malcolm v. Broadhurst [1970] 3 All E.R. 508). This is an application of the rule that a wrongdoer must take his victim as he finds him--sometimes called the "talem qualem" rule or, more colloquially, the "eggshell skull" rule. This is a principle of compensation, not of liability. As Lord Wright said in Bourhill v. Young [1943] AC 92, 109-110:

"No doubt, it has long ago been stated and often restated that if the wrong is established the wrongdoer must take the victim as he finds him. That, however, is only true . . . on the condition that the wrong has been established or admitted. The question of liability is anterior to the question of the measure of the consequences which go with the liability."

Likewise, in cases where no physical damage has been suffered by the plaintiff, Mullany and Handford state (op. cit. at p. 230), that:

". . . a claim for nervous shock is not actionable until the plaintiff incurs psychiatric damage caused, or contributed to, by the tortfeasor as a result of a breach of a duty or duties owed by him or her to the plaintiff. Only once this has been proved is the defendant bound to take the victim as he or she finds him or her."

See, e.g., Brice v. Brown [1984] 1 All E.R. 997. It is right that I should record the doubts on this point expressed by Mr. Tony Weir in his review of Mullany and Handford in [1993] C.L.J. 520, 521.

At all events, the principle of foreseeability of psychiatric injury has long been held to be subject to two special qualifications. First, in assessing whether psychiatric injury is reasonably foreseeable, it is assumed that the plaintiff is a person of reasonable fortitude. Second, as a concomitant of the first, the question of foreseeability of psychiatric injury is addressed with hindsight; as the Law Commission has put it (Report No. 249 at para.2.8), "foreseeability of the psychiatric illness is considered ex post facto in the light of all that has happened." Although it has been recognised that these qualifications raise their own problems, both have, until Page v. Smith, been understood to be of general application.

Finally, in this context, I must mention the position of people such as policemen or firemen, who might be thought to be less prone to suffer psychiatric injury at the sight of the sufferings of others than members of the general public. In two States of the United States there has developed a principle of policy known as the fireman's rule, under which it has been held that there is no "duty owed to the fireman to exercise care so as not to require the special services for which he is trained and paid:" see Krauth v. Geller (1960) 157 A. 2d. 129 N.J. p. 131, per Weintraub C.J. The fireman's rule was subsequently affirmed by the Supreme Court of California in Walters v. Sloan (1977) 571 P. 2d 609. In Ogwo v. Taylor [1988] A.C. 448, however, it was held by your Lordships' House that the American fireman's rule had no place in English law. That case was concerned with a claim in respect of physical injury, but I can see no reason why the same conclusion should not be reached in the case of a claim for psychiatric injury. As I understand it, however, it is generally accepted that, in considering whether psychiatric injury suffered by a plaintiff is reasonably foreseeable, it is legitimate to take into account the fact that the plaintiff is a person, such as for example a policeman, who may by reason of his training and experience be expected to have more resilience in the face of tragic events in which he is involved, or which he witnesses, than an ordinary member of the public possesses who does not have the same background. This is as far as it goes; and, as I shall explain in due course, it does not, in my opinion, affect the result in the wholly exceptional circumstances of the present case. It follows that, unlike Waller J., I would not, except in the limited manner I have indicated, think it necessary to identify a class of "professional" rescuers to which special rules apply.

In this connection I should also add, in relation to Waller J.'s judgment, that, while I agree with him that a police officer may have to take the risk of injury which arises in the course of his employment, there is no reason why he should be exposed to unnecessary risk of injury, i.e. to injury which could be avoided by the exercise of reasonable care by the Chief Constable, or by those for whose negligence he may be vicariously liable. This is a matter to which I will return at a later stage.

It is right that I should conclude this section of this opinion with the observation that foreseeability of psychiatric injury, while constituting a unifying principle of this branch of the law, cannot be regarded as providing a universal touchstone of liability. In this, as in other areas of tortious liability in which the law is in a state of development, the courts proceed cautiously from one category of case to another. We should be wise to heed the words of Windeyer J. spoken nearly 30 years ago in Mount Isa Mines Ltd. v. Pusey (1970) 125 C.L.R. 383, 396:

"The field is one in which the common law is still in course of development. Courts must therefore act in company and not alone. Analogies in other courts, and persuasive precedents as well as authoritative pronouncements, must be regarded."

I will have these words particularly in mind when I come to consider the case of Page v. Smith [1996] 1 AC 155.

Secondary victims

Having set out the two basic principles, I now turn to the special position of secondary victims.

In his opinion in Alcock v. Chief Constable of Yorkshire [1992] 1 A.C. at pp. 310, 406 et seq., Lord Oliver of Aylmerton (at p. 407) divided cases of liability for what was then called nervous shock:

"[b]roadly . . . into two categories, that is to say, those cases in which the injured plaintiff was involved, either mediately or immediately, as a participant, and those in which the plaintiff was no more than the passive and unwilling witness of injury caused to others."

A plaintiff in the latter category he found it convenient to describe as a "secondary" victim (no doubt having in mind cases such as Best v. Samuel Fox & Co. Ltd. [1952] A.C. 716), while reminding us (at p. 411) that:

"that description must not be permitted to obscure the absolute essentiality of establishing a duty owed by the defendant directly to him--a duty which depends not only upon the reasonable foreseeability of damage of the type which has in fact occurred to the particular plaintiff but also upon the proximity or directness of the relationship between the plaintiff and the defendant."

It has become settled that, to establish the necessary proximity, a secondary victim must show (1) a close tie of love and affection to the immediate victim; (2) closeness in time and space to the incident or its aftermath; and (3) perception by sight or hearing, or its equivalent, of the event or its aftermath. See generally McLoughlin v. O'Brian [1983] AC 410, 422-423, per Lord Wilberforce; Alcock v. Chief Constable of South Yorkshire Police [1992] 1 AC 310, 397-398 per Lord Keith of Kinkel, 402-404 per Lord Ackner, 411-417 per Lord Oliver of Aylmerton, 422-424 per Lord Jauncey of Tullichettle; and the Law Commission's Report No.249, paras.2-18--2.33.

I wish to stress that, although Lord Oliver in his opinion in Alcock referred to victims in his first category as "primary" victims, and (as I have indicated) described them as those who were "involved" as "participants," he did not attempt any definition of this category, but simply referred to a number of examples, including "rescuers"--an example which is of relevance to the present appeals. This is scarcely surprising since into this category fall a number of widely differing cases in which recovery is allowed, other than those falling into the second category which is concerned to segregate the special case of witnesses of injury caused to others to which special rules apply. It is also plain that, in the case of primary victims as in the case of secondary victims, Lord Oliver, in accordance with the generally accepted view, regarded the test of foreseeability to be one of foreseeability of damage of a particular type, viz. injury by what was then called shock: see [1992] 1 AC 310, 408F-G. It follows that, when considering whether the plaintiff does or does not fall into the category of secondary victims, the basic question relates to his involvement. This is essentially a question of fact, which I shall consider at a later stage. I should however add that, in the present appeals, the appellants are said to have been primary victims, either because they can claim as employees--or, more accurately, as "quasi-employees" because, as police officers, they are not strictly speaking employees but are able to rely upon a similar duty of care--or because they can claim as rescuers. I shall however postpone examination of these two categories until after I have considered the impact of Page v. Smith upon the general principles I have briefly described.

I have referred to the category of secondary victims, as identified by Lord Oliver, to whom special limiting principles apply. Since however this part of the law is still in a state of development, we should not exclude the possibility that other categories of claimant may come to be identified whose ability to claim damages for psychiatric injury should also be limited. For example, the Law Commission has canvassed the possibility of limits applying in cases arising from damage to property: see their Report No. 249, paras. 7.24--7.31. These matters need not however concern us in the present case.

The impact of Page v. Smith [1996] AC 155

As I have already foreshadowed, the decision of your Lordships' House in Page v. Smith constituted a remarkable departure from these generally accepted principles. The case was concerned with a traffic accident, in which the defendant's car collided with the plaintiff's car--a collision described as one of "moderate severity." Indeed nobody in either car suffered any physical injury, and the plaintiff (who was not even bruised by his seat belt) was able to drive his damaged car away after the accident. However the trial judge, Otton J., held that, as a result of the shock of the accident, the plaintiff suffered a recurrence of chronic fatigue syndrome from which he had suffered, with differing degrees of severity, for 20 years, and that for this he was entitled to recover damages from the defendant. The Court of Appeal (Ralph Gibson, Farquharson and Hoffmann L.JJ.) reversed the decision of Otton J. on the ground that it was not reasonably foreseeable that psychiatric injury to persons of ordinary fortitude would result from such an accident as this, in which the plaintiff suffered no physical injury. However your Lordships' House, by a majority of three to two (Lord Ackner, Lord Browne-Wilkinson and Lord Lloyd of Berwick; Lord Keith of Kinkel and Lord Jauncey dissenting) allowed the plaintiff's appeal but remitted the case to the Court of Appeal on the issue of causation which had been left open by two members of the court.

In reaching that conclusion Lord Lloyd of Berwick, who delivered the leading opinion with which both Lord Ackner and my noble and learned friend Lord Browne-Wilkinson agreed, departed from the previous understanding of the law in a number of respects. Before I turn to these, however, I wish to make two observations about this case. First, this was not a case concerned with a secondary victim. The plaintiff was obviously involved in the accident, and there was no question of his being affected by injury or death suffered by another. The special control mechanisms applicable in the case of secondary victims did not therefore arise for consideration. On the then accepted principles, the only question for consideration was whether the defendant could reasonably foresee that, in the circumstances which in fact occurred, a person of ordinary fortitude in the position of the plaintiff would suffer psychiatric injury. Second, as a subsidiary ground for their decision, the majority of the Appellate Committee briefly found for the plaintiff on that issue, contrary to the unanimous view of the Court of Appeal: see Lord Ackner [1996] AC 155, 170, and Lord Lloyd of Berwick (with whom Lord Browne-Wilkinson agreed) at p. 197.

I now turn to the respects in which Lord Lloyd, in his leading opinion, departed from the previous understanding of the law.

(1) Foreseeability of psychiatric injury. First and foremost, Lord Lloyd dethroned foreseeability of psychiatric injury from its central position as the unifying feature of this branch of the law. This he did by invoking the distinction between primary and secondary victims. In the case of the latter, he recognised that the law insists on certain "control mechanisms", to limit the number of potential claimants. Among these he included the requirement that the defendant will not be liable unless psychiatric injury is foreseeable in a person of normal fortitude, and he also restricted the use of hindsight to secondary victim cases--points to which I will return later. He continued (at p. 197G):

"Subject to the above qualifications, the approach in all cases should be the same, namely, whether the defendant can reasonably foresee that his conduct will expose the plaintiff to the risk of personal injury, whether physical or psychiatric. If the answer is yes, then the duty of care is established, even though physical injury does not, in fact, occur. There is no justification for regarding physical and psychiatric injury as different 'kinds of damage'."

The last statement in this passage he had previously sought to justify (at p. 188F) on the ground that:

"In an age when medical knowledge is expanding fast, and psychiatric knowledge with it, it would not be sensible to commit the law to a distinction between physical and psychiatric injury, which may already seem somewhat artificial, and may soon be altogether outmoded."

In para. 5.12 of their Report No. 249 the Law Commission record that "on the whole the responses, especially from practitioners, were very favourable to the decision" in Page v. Smith. It appears however that the responses from practitioners were simply expressions of view, unsupported by any analysis. Furthermore, as the Law Commission record in para. 5.14 of their Report, the revolutionary thesis in Page v. Smith has provoked severe criticism by a number of scholars with a special interest in this branch of the law, notably by Mr. Nicholas Mullany in (1995) 3 Journal of Law and Medicine 112, and Dr. Peter Handford in (1996) 4 Tort L. Rev. 5; but also by Professor Tan Keng Feng in [1995] Singapore Journal of Legal Studies 649; Mr. F.A. Trindade in (1996) 112 L.Q.R. 22; and Mr. Alan Sprince in (1995) 11 Professional Negligence 124. Most of them deplore the abandonment of the previously accepted general requirement of foreseeability of psychiatric injury. Mr. Mullany asserts that the distinction thus drawn by Lord Lloyd between primary and secondary victims is contrary to countless common law cases, and that the Privy Council's unambiguous endorsement in The Wagon Mound No. 1 [1961] AC 388, 426 of Denning L.J.'s statement of principle was "clearly seen as an all-purpose test for personal injury actions." In particular, the principle that foresight of shock-induced mental damage is relevant in establishing a duty of care had never been doubted in Australia.

In summary the basic grounds of criticism appear to be threefold.

(a) There has been no previous support for any such approach, and there is authority in England and Australia to the contrary. In England, see Lord Oliver's opinion in Alcock [1992] 1 A.C. at p. 408F-G where he regarded the principle of foreseeability of psychiatric damage as applicable in cases concerned with participants, as in the case of secondary victims. In Australia, Denning L.J.'s general statement of principle appears to have been anticipated by Dixon J. in Bunyan v. Jordan (1937) 57 C.L.R. 1, 16. In Mount Isa Mines Ltd. v. Pusey (1971) 125 C.L.R. 383, both Windeyer J. (at p. 395) and Walsh J. (at p. 402) treated the test of foreseeability of psychiatric injury as generally applicable; and in Jaensch v. Coffey (1985) 155 C.L.R. 549 Brennan J. (at p. 566), Deane J. (at p. 595) and Dawson J. (at p. 611) all did likewise. Indeed Mr. Mullany has stated, citing many cases, that all Australian psychiatric damage decisions have proceeded on this basis: see (1995) 3 Journal of Law and Medicine 112, 115.

(b) The approach favoured by Lord Lloyd appears to be inconsistent not only with the adoption by Viscount Simonds in The Wagon Mound No. 1 [1961] AC 388, 426, of Denning L.J.'s statement of principle, but also with the actual reasoning of the Privy Council in that case. There a particular type of damage to property, viz. damage by fire, was differentiated from other types of damage to property for the purpose of deciding whether the defendant could reasonably have foreseen damage of that particular type, so as to render him liable in damages in tort for such damage. That differentiation was made on purely common sense grounds, as a matter of practical justice. On exactly the same grounds, a particular type of personal injury, viz. psychiatric injury, may, for the like purpose, properly be differentiated from other types of personal injury. It appears to be in no way inconsistent with the making of that common sense judgment, as a matter of practical justice, that scientific advances are revealing that psychiatric illnesses may have a physical base, or that psychiatric injury should be regarded as another form of personal injury. Moreover the absence of any previous challenge to the general application of the principle stated by Denning L.J., and adopted by Viscount Simonds, perhaps provides the strongest endorsement of that common sense judgment.

(c) The majority in Page v. Smith may have misunderstood the so-called eggshell skull rule. In the course of his opinion, Lord Lloyd said (at p. 187A-B):

"We now know that the plaintiff escaped without external injury. Can it be the law that this makes all the difference? Can it be the law that the fortuitous absence of actual physical injury means that a different test has to be applied?"

These rhetorical questions Lord Lloyd answered in the negative. Yet the effect of the "eggshell skull" rule, i.e. the rule that a wrongdoer must take his victim as he finds him, is that the absence (or, more accurately, the presence) of physical injury to the plaintiff, may make all the difference. Lord Lloyd said at p. 193G:

Bourhill v. Young

However, it appears from the passage from Lord Wright's opinion in Bourhill v. Young which I have already quoted, that that is not the ordinary rule. The maxim only applies where liability has been established. The criticism is therefore that Lord Lloyd appears to have taken an exceptional rule relating to compensation and treated it as being of general application, thereby creating a wider principle of liability.

I recognise that the impact of this new statement of principle is likely to be relatively slight, in that it does no more than extend liability for psychiatric damage to those cases where physical damage is reasonably foreseeable (though none is suffered) but psychiatric damage is not. In any event, however, this situation does not arise in the present appeals, since none of the claimants was within the range of foreseeable physical injury; and your Lordships do not therefore have to form a view about the validity of the criticisms which I have summarised above. Your Lordships can therefore proceed on the basis that, for the purposes of the present appeals, the relevant test is, as in the past, the test of foreseeability of psychiatric damage.

(2) I now turn to two aspects of the reformulation of principle in Page v. Smith, which are relevant to the present appeals.

(a) Reasonable fortitude. Before the decision of your Lordships' House in Page v. Smith, the requirement of reasonable fortitude was regarded as being of general application, in cases concerned with primary victims as well as those concerned with secondary victims. See, e.g., the Law Commission's Consultation Paper No. 137, para. 2.10; Mullany and Handford, op. cit., ch. 10; and Mullany, (1995) 3 Journal of Law and Medicine, 112 at p. 117. The debate related not to the applicability of the requirement in cases concerned with primary victims, but to the desirability of the requirement as such: see Mullany and Handford, ubi sup. However in Page v. Smith Lord Lloyd, who treated this requirement as a "control mechanism" (see pp. 189D and 197F), held that it had no place where the plaintiff was a primary victim, in which type of case it was not appropriate to ask whether the victim is a person of "ordinary phlegm." Previously, however, the control mechanisms applicable in cases of secondary victims had been regarded as limited to those identified in the speeches of Lord Wilberforce in McLoughlin v. O'Brian [1983] AC 410, 422 et seq., and Lord Oliver of Aylmerton in Alcock [1992] 1 AC 310, 408-412, and to relate, as I have said, to those referred to in paras. 2.19 et seq. of the Law Commission's Report No. 249, viz. (i) tie of love and affection with the immediate victim; (ii) closeness in time and space to the incident or its aftermath; and (iii) the means of learning of the incident. These did not include the requirement of reasonable fortitude.

No reason is given in Page v. Smith for now including the test of reasonable fortitude among the control mechanisms relating to secondary victims, thereby restricting the test to claims by this class of claimant. In any event since, as I see it, the test of reasonable fortitude constitutes part of the enquiry whether psychiatric injury is reasonably foreseeable, it should logically also arise in cases concerned with primary victims. This is relevant in the present appeals, with reference to the fact that the claimants are police officers who may be said to possess greater fortitude than ordinary citizens. There is certainly debate about the proper role of this test in cases of psychiatric injury, though none of this is reflected in Page v. Smith. At all events, for the purposes of the present appeals, which are concerned with primary victims, I am content to proceed on the basis proposed by the Law Commission (Report No. 249, para. 5.26) that the reasonable fortitude (or "customary phlegm") test is

"best interpreted as meaning nothing more than that, in deciding whether psychiatric illness was reasonably foreseeable . . . one can take into account the robustness of the population at large to psychiatric illness."

(b) Hindsight. Although he did not treat this element as a "control mechanism," Lord Lloyd considered that it too had no part to play where the plaintiff is a primary victim (see [1996] 1 A.C. at p. 197F-G). This too appears to be a departure from the law as previously understood: see Mullany (1995) 3 Journal of Law and Medicine 112, 116. Moreover Lord Lloyd gave no reason for this departure, and it is difficult to understand why this approach should not, together with the reasonable fortitude test, be of general application. However where, as here, the court is concerned with a particular type of damage such as psychiatric injury:

". . . the court has to assess culpability by reference to what has actually happened; if you do not know the outcome of an accident it is impossible to determine whether what occurred should have been foreseen."

See Mullany, (1995) 3 Journal of Law and Medicine, 112, 116. It follows that it is, in my opinion, appropriate that your Lordships in the present appeals should have regard to what happened when considering the issue of foreseeability of psychiatric injury by the defendants.

(3) Primary and secondary victims. This is a matter which has a direct bearing on the outcome of the present appeals. As I have already recorded, we owe the distinction between primary and secondary victims to the opinion of Lord Oliver of Aylmerton in Alcock [1992] 1 AC 310, 407. Although he identified a secondary victim as one who is "no more than the passive and unwilling witness of injury to other," he made no attempt to define a primary victim, describing him simply as one who is "involved, either mediately or immediately as a participant," and giving miscellaneous examples of such persons. In Page v. Smith, however, Lord Lloyd (at p. 184A-B) said of the plaintiff in that case that he:

and well within the range of foreseeable physical injury

As the Law Commission have pointed out in their Report (see Law Com. 249 at paras. 2.52--2.60), the words which I have underlined have led to considerable confusion. So indeed has a further passage in Lord Lloyd's opinion, in which he said (at p. 187E-F):

Bourhill v. Young

Since the defendant was admittedly under a duty of care not to cause the plaintiff foreseeable physical injury, it was unnecessary to ask whether he was under a separate duty of care not to cause foreseeable psychiatric injury

The words which I have underlined in these two passages have led many--the Court of Appeal on a number of occasions (e.g., in the case of Young v. Charles Church (Southern) Ltd. (unreported) 24 April 1997, Court of Appeal Transcript No. 810 of 1997) and in the present case); the Law Commission in their Report No. 249 at para. 5.46; at least one textbook writer (see Munkman on Employers' Liability, 12th ed. at p. 125); and a number of commentators on Page v. Smith--to understand that case to have laid down that presence within the range of foreseeable physical injury is a necessary attribute of a primary victim; see also, in particular, the judgments of Henry L.J. and Judge L.J. in the present case in [1997] 3 W.L.R. 1194 at pp. 1213F-G and pp. 1224F-1226A respectively. In the result this point was, not surprisingly, placed by Mr. Collender Q.C. at the forefront of the appellants' case before your Lordships' House.

I am however satisfied that in neither of these passages did Lord Lloyd intend to reach any such conclusion (which would, in any event, have been no more than an obiter dictum). First, as appears from p. 184D-F of his opinion, Lord Lloyd accepted the distinction between primary and secondary victims drawn by Lord Oliver in Alcock [1992] 1 AC 310, 410-11, where, as Lord Lloyd said, Lord Oliver "referred to those who are involved in an accident as primary victims, and to those who are not directly involved, but suffer from what they have seen and heard, as the secondary victims." Yet the effect of the proposition now under consideration would be that the category of secondary victims is no longer to be restricted to witnesses, or "bystanders" as they are sometimes called, but is to be extended to include all victims other than those who were within the range of foreseeable physical injury. Furthermore it appears from Lord Oliver's speech in Alcock, which Lord Lloyd here invoked, that he did not regard presence within the range of foreseeable physical injury as a necessary attribute of a primary victim. This was made plain by the fact that he included among primary victims those who "come to the aid of others injured or threatened" (see p. 408E), citing Chadwick v. British Railways Board [1967] 1 W.L.R. 912, and plaintiffs in cases such as Dooley v. Cammell Laird & Co. Ltd. [1971] 1 Lloyd's Rep. 271

"where the negligent act of the defendant has put the plaintiff in the position of being, or, thinking that he is about to be or has been, the involuntary cause of another's death or injury"

with the result that he has suffered psychiatric illness (see p. 408E-G). In the latter group of cases there is ordinarily no question of the plaintiff having been within the range of foreseeable physical injury, and in Chadwick that factor was treated as irrelevant by the trial judge, George Waller J. Indeed cases such as Dooley, and rescue cases such as Chadwick and the well-known Australian case of Mount Isa Mines Ltd. v. Pusey (1970) 125 C.L.R. 383 (in which the successful plaintiff was never in any physical danger), are in direct conflict with the conclusion which has been attributed to Lord Lloyd in the passages now in question. In this connection it is significant that no reasons were given in Page v. Smith why any such limitation should be placed on recovery by primary victims; the point was not even discussed. Had it been considered, Lord Lloyd would have had to face up to the well-known decisions already referred to which are inconsistent with the proposition, and to consider whether he should follow them or whether he should distinguish or depart from them and, if the latter, why he should do so. The absence of any reference by Lord Lloyd to those decisions of itself renders it inconceivable that the passages in his judgment now in question should have been intended by him to have the effect attributed to them. The matter is, in my opinion, put beyond all doubt by the summary of his conclusions with which Lord Lloyd ended his opinion (see [1996] AC 155, 197E-H). After stating certain principles which he regarded as applicable in the case of secondary victims, he said:

This proposition, plainly designed to express Lord Lloyd's opinion that foreseeability of physical injury to the plaintiff is a sufficient condition of liability for psychiatric injury, is inconsistent with the proposition that it is also a necessary condition of such liability.

But let it be assumed that the passages in Lord Lloyd's judgment now in question were intended to have that effect: the result would be most remarkable. It would be that on the one hand Page v. Smith expands recovery, by holding that foreseeability of physical injury justifies recovery in respect of unforeseeable psychiatric injury even though no physical injury is suffered, while on the other hand the same case restricts recovery, by precluding recovery in respect of foreseeable psychiatric injury unless physical injury is also foreseeable. This does not make sense. The paradox undermines all credibility in the proposition, which is that what was formerly regarded as neither necessary nor sufficient (see Page v. Smith in the Court of Appeal [1994] 4 All E.R. 522, 549, per Hoffmann L.J.) has become not only sufficient but also, without any explanation, necessary. It is plain, in my opinion, that Lord Lloyd's strategy was to expand recovery by primary victims, not only in the manner I have indicated but also by restricting the applicability of the "reasonable fortitude" and "hindsight" tests to secondary victims; but that he had no strategy to restrict recovery by primary victims, whether by restricting recovery to cases where physical injury was foreseeable or otherwise.

For all these reasons I am satisfied that the passages in Lord Lloyd's opinion, to which I have referred, should be read as merely descriptive of the position of the plaintiff in Page v. Smith, and not as having the effect which has been ascribed to them. It follows, however, that to this extent the appellants' case must be regarded as having been framed on a false premise. I understand however that some of your Lordships are of the opinion that, even if my understanding of these passages in Lord Lloyd's opinion is correct, the House should, as a matter of policy, nevertheless impose a requirement of foreseeability of physical damage as an arbitrary limit upon recovery by primary victims in respect of psychiatric injury suffered by them. I shall consider this proposal at a later stage in this opinion.

I wish, however, to add in this connection that in some cases, in particular those in which the plaintiff is claiming damages in respect of psychiatric injury caused by fear of injury to himself, it may indeed be relevant to enquire whether he was within the range of foreseeable physical injury: see, e.g., McFarlane v. EE Caledonia Ltd. [1994] 1 All ER 1. But, as I have said, it is inconsistent with existing authority that any such requirement should be applicable in all cases concerned with primary victims.

The impact of Page v. Smith on the present appeals. In the light of the foregoing I have to consider the relevance of the decision of Page v. Smith to the present appeals. For the reasons I have already given I have reached the conclusion that point (1) (reasonable foreseeability of psychiatric injury is no longer required where the plaintiff is within the range of foreseeable physical injury) does not arise on the facts of the present case; that point (2)(a) (the reasonable fortitude test is now to be limited to secondary victim cases) can be regarded as immaterial, because I am content for present purposes to proceed on the basis proposed by the Law Commission (Report No. 249, para. 5.26); that point 2(b) (the hindsight test is now also to be limited to secondary victim cases) is an obiter dictum which I am disinclined to follow; and that, for the reasons I have given, point (3) (primary victims can only recover if they are within the area of foreseeable physical danger) does not arise, because I do not read the relevant passages in the judgment as having that effect.

Employees and Rescuers

(1) Employees

An employee (I will for present purposes include in this category a "quasi-employee" such as a police officer who, although he holds an office and is not therefore strictly an employee, is owed the same duty by his "employer"--here the Chief Constable of South Yorkshire) may recover damages from his employer in respect of psychiatric injury suffered by him by reason of his employer's breach of duty to him. The basic obligation of the employer arises from the relationship between him and his employee, under which the employer is under a duty to take reasonable care for the safety of his employee at work (see, e.g., Wilsons and Clyde Coal Co. Ltd. v. English [1938] AC 57, 84-85 per Lord Wright) and in particular not to expose his employees to unnecessary or unreasonable risk. It was this latter duty upon which Mr. Hytner Q.C. for the respondents relied in the present appeals. This duty, as Mr. Hytner Q.C. recognised, is generally regarded as tortious.

However all the employer's duties "are connected in some sense to what happens to the employee while at work" (see Munkman on Employers' Liability, 12th ed., (1995) at p. 33); and it is with cases arising in this context that we are concerned. I put on one side those cases in which an employee is seeking damages from his employer in respect of stress at work, as to which see Munkman, op. cit., at pp. 128-130, and Walker v. Northumberland County Council [1995] 1 All ER 737 (commented on by the Law Commission in its Consultation Paper No. 137, paras. 2.49--2.50). But in the authorities relating to the recovery by an employee from his employer of damages for psychiatric injury, arising from the death or physical injury of another, we find a distinction being drawn between those cases in which the employee has in the course of his employment been involved in the event which resulted in the other's physical injury or death, to which I would add involvement in the aftermath of that event, and other cases in which he has, while at work, incidentally witnessed that event and its outcome.

As to the former, a useful example is to be found in the unreported case of Young v. Charles Church (Southern) Ltd. 24 April 1997. There the victim of the accident, Mr. Cook, was erecting a structure consisting of scaffolding poles. He was assisted by two labourers, one of them being the plaintiff, Mr. Young. As the plaintiff turned away to fetch another pole, Mr. Cook raised a 20 foot pole vertically and it came into contact with an overhead power cable carrying 33,000 volts of electrical current. The plaintiff heard a loud bang and a hissing noise. He looked up and saw that the pole held by Mr. Cook had struck and stuck to the electric wiring. He saw that the ground around Mr. Cook had burst into flames. Mr. Cook fell to the ground; he had been electrocuted. The plaintiff heard the other labourer, Mr. Smith, scream. He too had been struck by electricity. The plaintiff was not injured physically, but he suffered a psychiatric illness (P.T.S.D.) as a result of the accident. He claimed damages from two defendants, one being his employer. Both defendants agreed not to dispute liability for the purposes of the proceedings, but contended that the plaintiff did not come within the class of persons entitled to make a claim for nervous shock, because he had suffered no injury in the accident. The Court of Appeal held that the plaintiff was entitled to recover. As a result of Page v. Smith, the Court of Appeal was concerned with the question whether the plaintiff was within the range of physical injury, and were able on the facts of the case to hold that he was. But for present purposes the important finding of the majority, Evans and Hutchison L.JJ., was that the plaintiff was involved as a participant in the accident. Hutchison L.J. put the matter very clearly at p. 32 of the transcript:

"[The plaintiff] had just given the pole to Cook and was within six feet or so of the latter when Cook received the fatal shock. Though he had turned away to get the shorter support poles to be used to maintain the long pole in a vertical position, he could in my view properly be regarded as still participating in the erection of that long pole at the moment it touched the electric cable . . ."

The circumstances of that case can be compared to those of the two Scottish cases of Robertson and Rough v. Forth Road Bridge Joint Board [1995] SC 364. I take the facts from the opinion of the Lord President, Lord Hope of Craighead, at p. 365:

"The actions arose out of a tragic accident which occurred on the Forth Road Bridge on 29 January 1989. The pursuers and a workmate named George Smith were working together in the course of their employment with the defenders, whose function is to maintain and operate the bridge. The winds were blowing at gale force during their shift that day. The accident occurred while the men were in the course of removing a large thin piece of metal sheeting which had been found lying on the south-bound carriageway. It was being taken off the bridge on the open platform of a Ford Transit pick-up van. Smith was sitting on the top of the metal sheet on the back of the van, Robertson was driving the van and Rough was following a few feet behind in a small patrol van which he was driving. In the course of their journey to the south end of the bridge a sudden and violent gust of wind caused the sheet and Smith to be thrown violently off the back of the van and over the side of the bridge."

Mr. Smith fell only a few feet onto a girder, but was killed by the force of the impact. The two pursuers claimed to have suffered nervous shock as a result of witnessing the accident. The Lord Ordinary held that there was no duty of care owing to them by the defenders, and granted decree of absolvitur. The pursuers reclaimed against that decision, but the First Division of the Inner House refused their reclaiming motions.

The principal opinion was delivered by the Lord President. The argument for the pursuers was essentially that they were so directly involved in the accident as to be within the ambit of their employers' duty of care to them. This argument was however rejected by the Lord President, who regarded the case not as one of active participation in the event, but as one where the pursuers were merely bystanders or witnesses, in which event the ordinary rule stated by Lord Oliver in McLoughlin v. O'Brian must apply and, as the pursuers did not comply with the control mechanisms applicable in the case of claimants who were only witnesses, their claim must fail. The case therefore provides authority that, in a claim by an employee against his employer for damages for psychiatric injury arising from the death of or injury to another, his claim will fail if he is simply a bystander who witnesses the event, and is not an active participant in it (or, I would add, its aftermath). It was perhaps open to the Lord President to take the view that the two pursuers were at the time actively involved with Mr. Smith in the operation of removing the sheet from the bridge, in which event the reclaiming motion would no doubt have been granted; but he took a different view of the facts of the case.

It is, in my opinion, consistent with the tortious basis of the employer's duty in these cases that it should, in cases concerned with a claim by an employee for damages in respect of psychiatric injury, be subject to the limits set out in the opinion of Lord Oliver in Alcock [1992] 1 A.C., 310, 407-411, in the case of a claimant who is a bystander in the sense of being no more than a passive and unwilling witness of injury caused to others; and I consider that the same could be said if the employer's duty of care was expressed as an implied term in the contract of employment. In accordance with this approach Stuart-Smith L.J. (with whom McCowan and Ralph Gibson L.JJ. agreed) said in McFarlane v. E.E. Caledonia Ltd. [1994] 2 All ER 1 (a case concerned with an employee's claim against his employer) at p. 14E:

The importance of this conclusion is that it avoids what otherwise might be regarded as an unacceptable distinction between employees on the one hand, and relatives on the other. This is of particular relevance in the present case, where a number of relatives of victims at Hillsborough failed in claims for damages in respect of psychiatric injury which they advanced in the case of Alcock [1992] 1 AC 310, and it has been suggested that it would be unacceptable if police officers were entitled to a wider basis of recovery as employees. However this is not, in my opinion, the position at law. The difference between the two categories arises not from the applicability of special rules in the case of secondary victims (which, in my opinion, apply to both categories) but from the fact that, whereas police officers who became involved on the ground in the aftermath of the disaster can claim against the Chief Constable as "employees", strangers who intervened will have to justify their intervention, for example by bringing themselves within the broad category of "rescuers", to which I will turn in a moment. In this connection I wish to record that the claims of the plaintiffs in Alcock were not advanced on the basis that they were rescuers, a fact which must be borne in mind when comparisons are drawn between those plaintiffs and the plaintiffs in the present case.

(2) Rescuers

I turn next to the category of rescuers. This category is of particular importance for outsiders who intervene in a situation created by a wrongdoer. The fact that an outsider may intervene in such a situation to rescue a victim of the wrongdoing is reasonably foreseeable by a person in the position of the wrongdoer. The intervention is justified by the necessity of the moment, and so is not unlawful. It does not break the chain of causation between the wrongful act of the defendant and injury suffered by the intervener by reason of his act of rescue, whether the rescue is successful or not. Compensation for such an injury may be recovered by the intervener from the wrongdoer, whether the injury is physical as in the classic rescue cases such as Baker v. T.E. Hopkins & Sons Ltd. [1959] 1 WLR 966, or psychiatric as in Chadwick v. British Railways Board [1967] 1 W.L.R. 912.

Chadwick is important in another respect. It shows that we must not be prisoners of our concepts, here the concept of rescue. Mr. Chadwick was not attempting to rescue anybody. He was a small and agile man, who lived close to the railway line in Lewisham, on which two trains collided with catastrophic results. Many passengers were killed or injured; and many of the injured were trapped in the wreckage for a long time during the night before they could be rescued. Mr. Chadwick worked for many hours during the night, crawling under the wreckage of the train and bringing aid and comfort to the victims, some of them severely injured, who were trapped in the wreckage. He was exposed to some physical danger, but the trial judge (George Waller J.) treated that as irrelevant. It was, he held, the whole horror of the situation which affected Mr. Chadwick, who as a result suffered psychiatric injury in a form which would nowadays probably be classified as P.T.S.D. When we contemplate the full horror of the disaster--the terrible injuries suffered by some of the victims, dead and alive, and the cries of the living for help; the long hours of darkness; the claustrophobic conditions in which Mr. Chadwick worked--it is scarcely surprising that the judge treated the physical danger as irrelevant; and it is scarcely surprising too that, in McLoughlin v. O'Brian [1983] 1 AC 410, 438, Lord Bridge of Harwich stated that, as far as he knew, no one had ever doubted that the case was rightly decided. But it is also plain that the circumstances were wholly exceptional. It must be very rare that a person bringing aid and comfort to a victim or victims will be held to have suffered foreseeable psychiatric injury as a result.

In this connection, I should record that there is controversy on the question whether "searchers" may qualify as rescuers. That they may do so is supported by the most famous of all rescue cases, Wagner v. International Railways Co. (1921) 232 N.Y. 176, in which the judgment of the court was delivered by Cardozo J.; and the same view was expressed by Evatt J. in his dissenting judgment in Chester v. Waverley Council (1939) 62 C.L.R. 1, 14 et seq.--a judgment later to be commended by Lord Wright in Bourhill v. Young [1943] AC 92, 110 and by Lord Wilberforce in McLoughlin v. O'Brian [1983] 1 AC 410, 422, and approved by the High Court of Australia in Jaensch v. Coffey (1985) 155 C.L.R. 549, especially at pp. 590-591 per Deane J. However Mr. Mullany and Dr. Handford have expressed the opinion that this approach is not part of the modern law: see (1997) 113 L.Q.R. at p. 417. The point does not however arise in the present case; the solution may perhaps depend on the facts of the particular case.

I wish also to add that, obviously, a rescuer will normally come on the scene after the disastrous event has occurred. It is most unlikely that he will be involved in that event itself. He is involved in the aftermath of that event, and is concerned with its consequences. That involvement is, however, sufficient to bring him within the category of primary victims, so far as liability for psychiatric injury is concerned: see Alcock [1992] 1 AC 310, 408, per Lord Oliver of Aylmerton.

(3) Employees and rescuers

It is of course perfectly possible for an employee of the tortfeasor to be a rescuer. If so, the basis on which he may claim damages from his employer in respect of any psychiatric injury which he may suffer by reason of his involvement will depend on the circumstances of the case. Where he becomes involved in the course of his employment (see Priestley v. Fowler (1837) 3 M. & W. 1, 6, per Lord Abinger C.B.; Munkman on Employer's Liability, 12th ed., (1995) p. 74) he may be able to claim damages simply on the basis of breach by his employer of his duty of care. If not, however, he can rely on his intervention in the character of a rescuer as a stranger may do. A borderline case, which appears to have fallen on the former side of the line, is the important Australian case of Mount Isa Mines Ltd. v. Pusey (1971) 125 C.L.R. 383. A terrible accident occurred at the defendants' powerhouse, when two employees who were testing a switchboard were severely burned by an intense electric arc. This was held to have occurred because the defendants had negligently failed to give the men proper instructions. The plaintiff, who was a foreman on the defendants' staff and could therefore have reasonably been expected to go to the scene of the accident, did so and found one of the men very severely burned. The plaintiff supported him out of the powerhouse, and helped to carry him to an ambulance. Within nine days, however, the man died of his injuries. The plaintiff subsequently developed a serious mental disturbance, diagnosed as a form of schizophrenia. The High Court of Australia upheld the decision of the trial judge, affirmed by the Full Court of the Supreme Court of Queensland, that the plaintiff was entitled to succeed in a claim against his employers in respect of his psychiatric injury. In the course of the judgments of some members of the High Court, the plaintiff was treated as a rescuer; but I understand the prevailing view of the High Court in that case, and of the High Court which sat in the later case of Jaensch v. Coffey (1985) 155 C.L.R. 549 in which the Mount Isa Mines case was considered, to have been that the defendants' liability arose from breach of their duty as employers of the plaintiff: see, in particular, the judgments of Windeyer J. in the Mount Isa Mines case at p. 400, and of Deane J. in Jaensch v. Coffey at p. 597.

In some cases, however, the circumstances may be such that an employee is involved in the aftermath of the relevant event when acting in the course of his employment with the tortfeasor, and that a part of his involvement may fall within the description of rescue and the remainder not. If as a result of his involvement the employee suffers psychiatric injury, it will be necessary to have regard to his involvement as a whole, including his actions of rescue, when deciding whether or not such psychiatric injury is a reasonably foreseeable consequence of a breach by his employer of his duty to him. That is, in my opinion, the position in the present case. It follows that if, as in the present case, there is a group of employees who were involved in the aftermath of the event, and only some of them were involved in acts of rescue, it does not follow that the latter only will be entitled to recover. It is the involvement of each as a whole which has to be considered; and if the involvement is such that the acts of rescue were no more than incidental parts of a wider involvement which caused the psychiatric injury, there is no reason why those employees who were involved in acts of rescue should be singled out as those who alone are entitled to recover. This is because, in such a case as in the case of Chadwick, it is the whole horror of the situation which is the cause of the psychiatric injury suffered by all of the employees so involved.

A new control mechanism?

As I have already recorded, it was submitted by Mr. Collender Q.C. on behalf of the appellants, relying on certain passages in the opinion of Lord Lloyd in Page v. Smith [1996] 1 AC 155, 184A-B, 187E-F, that it was a prerequisite of the right of recovery by primary victims in respect of psychiatric injury suffered by them that they should have been within the range of foreseeable physical injury. I have already expressed the opinion that no such conclusion can be drawn from Lord Lloyd's opinion in Page v. Smith. I understand however that, even if my view on that point is accepted as correct, some of your Lordships nevertheless consider that a new control mechanism to the same effect should now be introduced and imposed by this House as a matter of policy.

I am compelled to say that I am unable to accept this suggestion because in my opinion (1) the proposal is contrary to well-established authority; (2) the proposed control mechanism would erect an artificial barrier against recovery in respect of foreseeable psychiatric injury and as such is undesirable; and (3) the underlying concern is misconceived. I will consider each of these objections in turn.

(1) The proposal is contrary to well-established authority.

I have here in mind the cases to which I have previously referred, concerned (a) with rescuers, and (b) with those who have, as a result of another's negligence, been put in the position of being, or of thinking that they are, the involuntary cause of another's death or injury. As I have already recorded, the most relevant cases concerned with the first category (rescuers) are Chadwick [1967] 1 W.L.R. 912 (in which the trial judge treated the fact that there was some danger of physical injury as irrelevant), and (on one view) Mount Isa Mines Ltd. v. Pusey (1970) 125 C.L.R. 383 (in which the plaintiff was not in physical danger). In this connection it is important that the decision in Chadwick was approved, without qualification, in your Lordships' House in McLoughlin v. O'Brian [1983] AC 410, 419 per Lord Wilberforce, 424 per Lord Edmund-Davies, and 437-8 per Lord Bridge of Harwich, and again in Alcock [1992] 1 AC 310, 408 per Lord Oliver. As to the second category, the most relevant case is Dooley [1971] 1 Lloyd's Rep. 271 in which, as in other cases of this kind, the plaintiff was never in any personal danger. Furthermore, both categories of case were stated by Lord Oliver in Alcock at p. 408 to be examples of primary victims, in the case of which he plainly did not consider that there was any applicable control mechanism, for example any requirement that the plaintiff should have been within the range of foreseeable physical injury. Having regard in particular to the prominence now given to Lord Oliver's opinion in Alcock in segregating cases of secondary victims as those cases to which special control mechanisms apply, it would be a remarkable departure from existing authority now to create a new control mechanism, viz. that the plaintiff must have been exposed to the risk of physical injury, and to hold that this mechanism is applicable in the case of primary victims. What is here at issue therefore is not whether we should extend liability for psychiatric injury to primary victims who do not come within the range of foreseeable physical injury. The question is whether, having regard to existing authority, we should restrict liability for psychiatric injury to primary victims who are within the range of such injury.

(2) The proposed control mechanism would erect a new artificial barrier against recovery in respect of foreseeable psychiatric injury and as such is undesirable.

The control mechanisms now in force are those established in Alcock to be applicable in the case of secondary victims, viz. (a) a close tie of love and affection to the immediate victim, (b) proximity in time and space to the incident or its aftermath, and (c) perception by sight or hearing, or its equivalent, of the event or its aftermath. These rules, being arbitrary in nature, are widely perceived to create unjust and unacceptable distinctions: see, in particular, the criticisms of Professor Jane Stapleton in In Restraint of Tort (Frontiers of Liability ed. Peter Birks 1994) at pp. 95-96. To introduce the control mechanism now proposed in the case of primary victims would in the same way create distinctions regarded as unjust and unacceptable.

To illustrate the point, let me take the always useful extreme example. Suppose that there was a terrible train crash and that there were two Chadwick brothers living nearby, both of them small and agile window cleaners distinguished by their courage and humanity. Mr. A. Chadwick worked on the front half of the train, and Mr. B. Chadwick on the rear half. It so happened that, although there was some physical danger present in the front half of the train, there was none in the rear. Both worked for 12 hours or so bringing aid and comfort to the victims. Both suffered P.T.S.D. in consequence of the general horror of the situation. On the new control mechanism now proposed, Mr. A would recover but Mr. B would not. To make things worse, the same conclusion must follow even if Mr. A was unaware of the existence of the physical danger present in his half of the train. This is surely unacceptable. May I stress that, although I have taken an extreme example, the contrast I have drawn could well arise in real life; and the new control mechanism now proposed could provoke criticisms of the same kind as those which have been made of the mechanisms recognised in Alcock.

(3) The underlying concern is misconceived.

I sense that the underlying concern, which has prompted a desire to introduce this new control mechanism, is that it is thought that, without it, the policemen who are plaintiffs in the present case would be "better off" than the relatives in Alcock who failed in their claims, and that such a result would be undesirable. To this, there are at least three answers. First, the control mechanisms which excluded recovery by the relatives in Alcock would, in my opinion, have been equally applicable to the policemen in the present case if on the facts they had (like the relatives) been no more than witnesses of the consequences of the tragedy. Second, the question whether any of the relatives might be able to recover because he fell within the broad category of rescuer is still undecided; and, strangely, the control mechanism now proposed to exclude the claims of the policemen in the present case would likewise exclude the claims of relatives if advanced on the basis that they were rescuers. Third, however, it is in any event misleading to think in terms of one class of plaintiffs being "better off" than another. Tort liability is concerned not only with compensating plaintiffs, but with awarding such compensation against a defendant who is responsible in law for the plaintiff's injury. It may well be that one plaintiff will succeed on the basis that he can establish such responsibility, whereas another plaintiff who has suffered the same injury will not succeed because he is unable to do so. In such a case, the first plaintiff will be "better off" than the second, but it does not follow that the result is unjust or that an artificial barrier should be erected to prevent those in the position of the first plaintiff from succeeding in their claims. The true requirement is that the claim of each plaintiff should be judged by reference to the same legal principles.

For all these reasons I am unable to accept the need for, or indeed the desirability of, the new control mechanism now proposed.

The present appeals

(1) The medical evidence

Like Henry L.J., I have read with interest and profit the account of P.T.S.D. in the Law Commission's Report No. 249 at paras. 3.4-3.14, and in Mullany and Handford at pp. 33 et seq.; and like him I have also read Professor Sims' "generic report" relating to a psychiatric examination and assessment carried out upon 70 police officers involved in the Hillsborough tragedy, including the plaintiffs in the present proceedings. This material provides a clear description of the nature of P.T.S.D., its causes and effects. I quote from paras. 3.4 and 3.5 of the Law Commission's Report:

"3.4 The phrase 'post-traumatic stress disorder' was coined in the 1970s and was officially recognised with the publication of DSM-III [the American Diagnostic and Statistical Manual of Mental Disorders] in 1980. Veterans returning from the Vietnam War were found to be suffering from severe stress and in need of treatment, yet there was no diagnosis to fit their syndrome. P.T.S.D. was a concept created to meet that need. However, the acceptance of P.T.S.D. among psychiatrists has not been universal and the diagnosis remains controversial . . .

"3.5 The diagnostic criteria for P.T.S.D. in DSM-IV require that the person develop characteristic symptoms following exposure to a traumatic event (frequently referred to as the 'stressor') in which (i) the person experienced, witnessed, or was confronted with an event or events that involved actual or threatened death or serious injury, or a threat to the physical integrity of self or others, and (ii) the person's response involved intense fear, helplessness, or horror. A diagnosis of P.T.S.D. under ICD-10 [the International Classification of Diseases) requires that the individual have been exposed to a stressful event or situation (either short- or long-lasting) of an exceptionally threatening or catastrophic nature, which is likely to cause pervasive distress in almost anyone."

Mullany and Handford refer (at p. 35) to the American literature currently defining P.T.S.D. as requiring "exposure to a psychologically distressing external event that is outside the range of usual human experience." It comes as no surprise therefore that Professor Sims regarded the Hillsborough tragedy as a P.T.S.D. stressor. Moreover, as Henry L.J. observed ( [1997] 3 W.L.R. 1194, 1207E-H), Professor Sims stressed that the trauma in the present case was prolonged exposure to horrifying and uncontrollable circumstances, and that in general the longer the exposure to the traumatic situation, the greater was the degree of psychological distress subsequently. Recurring themes in the police officers' accounts were inability to take useful action, and so feelings of helplessness and guilt; the sheer number of the deaths and the youth of the victims; hostility and abuse from the crowd, and shame that police decisions had caused or contributed to the disaster. I add in parenthesis that the nature of P.T.S.D. illustrates very clearly the need to abandon the requirement of nervous shock in these cases, and to concentrate on the requirement that the plaintiff should have suffered from a recognised psychiatric illness.

(2) The responsibility in law of the appellants to each of the respondent police officers.

Henry L.J. concluded ([1997] 3 W.L.R. 1194, 1209A) that the risk of psychiatric damage to police officers on duty at the ground as a result of negligent crowd control was plainly foreseeable. In a general sense, this is no doubt true. In particular, the fact that the plaintiffs are police officers and as such might, by reason of their training and experience, be regarded as less likely to suffer psychiatric injury as a result of their involvement in this terrible event and its aftermath is, in my opinion, offset by a combination of the nature and scale of the catastrophe and the hostility and shame resulting from police responsibility for the tragedy, for which these individual plaintiffs were in no way responsible. However, in accordance with the "hindsight" test which, in my opinion, is as applicable to primary victims as it is to secondary victims, the question of foreseeability has to be considered in relation to each respondent, having regard to the nature and circumstances of his involvement.

The involvement of the individual police officers who are respondents to these appeals is described in their written statements. I do not propose, however, to summarise each of their statements. To do so would be to overburden this already long opinion; and in any event it is no use just picking out particular events from these statements--they have to be read as a whole. Reading them as a whole, it is plain to me that each of them was, in the course of his duty as a police officer, involved in the aftermath of the terrible crushing which took place in Pens 3 and 4, which brought death or injury to a large number of spectators. It is true that, as a result of the chaos, relatively few instructions were received by individual officers; to a large extent they were acting on their own initiative, but that does not alter the fact that they were acting in the course of their duty. Sometimes they were involved in specific actions in relation to victims of the disaster--trying to find out if a victim was still alive and, in the belief or hope that he was, applying mouth to mouth resuscitation or cardiac massage; transporting, or helping to transport bodies on makeshift stretchers to the gymnasium; laying out the bodies; standing by an individual body; identifying bodies, which involved looking into their eyes and mouths; dealing with enquiries from distraught relatives, which was described by one officer as appalling; and so on. Some of their actions could be described as acts of rescue, but in my opinion that is not important, having regard to the nature and extent of the involvement of the officers in the present case. In one or two cases the actual activities of this kind by a particular officer were relatively few; but for the rest of the time on the ground he was still involved in the course of his duty, looking for useful tasks to perform. It is also true that, during his involvement, each of them saw, and was much affected by, terrible sights; but that does not mean that they should be regarded as bystanders and so treated as secondary offenders, or that what they witnessed should be put on one side. This is because, in a case such as this, what they saw was part and parcel of their involvement in the aftermath of the event.

Moreover, in judging whether psychiatric illness was foreseeable in any particular case, we have to have regard not only to the nature of each officer's involvement, but also to the context in which that involvement took place. Although we get glimpses of the context from the statements--of the chaos, the hysteria, the breakdown of crowd control, the threatening behaviour of the crowd, which was described as shouting and screaming--each statement is very largely devoted to the actual movements of the officer concerned, what he did and what he saw. The background is mostly taken for granted. Yet it is not difficult for us to grasp that the atmosphere of this wholly exceptional tragedy, in the aftermath of which the officers became involved, and the length of time during which the officers were exposed to the consequences of the tragedy, were potent forces which are highly relevant to the question whether, in each of their cases, psychiatric injury was a consequence of their involvement which was reasonably foreseeable by their "employer" who was responsible for their safety at work. Looking at the picture as a whole, I have come to the conclusion that, in the case of each of the five respondents, that question must be answered in the affirmative, as it must also be in the case of officers in the position of Mr. Hallam, and that the appellant Chief Constable was in breach of his duty to them.

(3) Conclusion.

For these reasons I would dismiss all the appeals with costs, and remit the issue of causation to a trial judge.

LORD STEYN

My Lords,

In my view the claims of the four police officers were rightly dismissed by Waller J. (now Lord Justice Waller) and the majority in the Court of Appeal erred in reversing him: Frost v. Chief Constable of South Yorkshire Police [1997] 3 W.L.R. 1194. Different kinds of harm

The horrific events of 15 April 1989 at the Hillsborough Football Stadium in Sheffield resulted in the death of 96 spectators and physical injuries to more than 700. It also scarred many others for life by emotional harm. It is admitted by the Chief Constable that the events were caused by the negligence of the police in allowing the overcrowding of two spectator pens. In an ideal world all those who have suffered as a result of the negligence ought to be compensated. But we do not live in Utopia: we live in a practical world where the tort system imposes limits to the classes of claims that rank for consideration as well as to the heads of recoverable damages. This results, of course, in imperfect justice but it is by and large the best that the common law can do. The application of the requirement of reasonable foreseeability was sufficient for the disposal of the resulting claims for death and physical injury. But the common law regards reasonable foreseeability as an inadequate tool for the disposal of claims in respect of emotional injury.

The law divides those who were mentally scarred by the events of Hillsborough in different categories. There are those whose mental suffering was a concomitant of physical injury. This type of mental suffering is routinely recovered as "pain and suffering". Next, there are those who did not suffer any physical injuries but sustained mental suffering. For present purposes this category must be subdivided into two groups. First, there are those who suffered from extreme grief. This category may include cases where the condition of the sufferer is debilitating. Secondly, there are those whose suffering amounts to a recognizable psychiatric illness. Diagnosing a case as falling within the first or second category is often difficult. The symptoms can be substantially similar and equally severe. The difference is a matter of aetiology: see the explanation in Munkman, Damages for Personal Injuries and Death, 10th ed., 118, note 6. Yet the law denies redress in the former case: see Hinz v. Berry [1970] 2 Q.B. 40, at 42 but compare the observations of Thorpe L.J. in Vernon v. Bosley [1997] 1 All ER 577, at 610, that grief constituting pathological grief disorder is a recognizable psychiatric illness and is recoverable. Only recognizable psychiatric harm ranks for consideration. Where the line is to be drawn is a matter for expert psychiatric evidence. This distinction serves to demonstrate how the law cannot compensate for all emotional suffering even if it is acute and truly debilitating.

The four police officers were actively helping to deal with the human consequences of the tragedy and as a result suffered from post traumatic stress disorder. The police officers put in the forefront of their case that they suffered harm as a result of a tort and that justice demands that they should be compensated. A constant theme of the argument of counsel for the police officers was that there is no justification for regarding physical and psychiatric injury as different kinds of damage, and in so arguing he was repeating an observation of Lord Lloyd of Berwick in Page v. Smith [1996] AC 155, at 197G. It is of some importance to examine this proposition. Courts of law must act on the best medical insight of the day. Nowadays courts accept that there is no rigid distinction between body and mind. Courts accept that a recognizable psychiatric illness results from an impact on the central nervous system. In this sense therefore there is no qualitative difference between physical harm and psychiatric harm. And psychiatric harm may be far more debilitating than physical harm.

It would, however, be an altogether different proposition to say that no distinction is made or ought to be made between principles governing the recovery of damages in tort for physical injury and psychiatric harm. The contours of tort law are profoundly affected by distinctions between different kinds of damage or harm: see Caparo Industries Plc. v. Dickman [1990] 2 AC 605, at 618E, per Lord Bridge of Harwich. The analogy of the relatively liberal approach to recovery of compensation for physical damage and the more restrictive approach to the recovery for economic loss springs to mind. Policy considerations encapsulated by Justice Cardozo's spectre of liability for economic loss "in an indeterminate amount for an indeterminate time to an indeterminate class" played a role in the emergence of a judicial scepticism since Murphy v. Brentwood District Council [1991] AC 398 about an overarching principle in respect of the recovery of economic loss: see Steele, Scepticism and the Law of Negligence, [1993] C.L.J. 437. The differences between the two kinds of damage have led to the adoption of incremental methods in respect of the boundaries of liability for economic loss.

Similarly, in regard to the distinction between physical injury and psychiatric harm it is clear that there are policy considerations are at work. That can be illustrated by reference to the Criminal Injuries Compensation Scheme. Section 109(2) of the Criminal Justice Act 1988 contains this restrictive rule:

"Harm to a person's mental condition is only a criminal injury if it is attributable - (a) to his having been put in fear of immediate physical injury to himself or another; or (b) to his being present when another sustained a criminal injury other than harm to his mental condition."

The reason for the restriction is that Parliament was fearful that a more liberal rule would impose an intolerable burden on the public purse. Parliament has also decided that the only persons who can claim bereavement damages are parents and spouses: section 1(A)(7) of the Fatal Accidents Act 1976. The spectre of a wide a class of claimants in respect of bereavement led to an arbitrary but not necessarily irrational rule.

Policy Considerations and Psychiatric Harm

Policy considerations have undoubtedly played a role in shaping the law governing recovery for pure psychiatric harm. The common law imposes different rules for the recovery of compensation for physical injury and psychiatric harm. Thus it is settled law that bystanders at tragic events, even if they suffer foreseeable psychiatric harm, are not entitled to recover damages: Alcock v. Chief Constable of South Yorkshire Police [1992] 1 AC 310. The courts have regarded the policy reasons against admitting such claims as compelling.

It seems to me useful to ask why such different rules have been created for the recovery of the two kinds of damage. In his Casebook on Tort, 7th ed., Weir gives the following account (at 88):

". . . . there is equally no doubt that the public . . . draws a distinction between the neurotic and the cripple, between the man who loses his concentration and the man who loses his leg. It is widely felt that being frightened is less than being struck, that trauma to the mind is less than lesion to the body. Many people would consequently say that the duty to avoid injuring strangers is greater than the duty not to upset them. The law has reflected this distinction as one would expect, not only by refusing damages for grief altogether, but by granting recovery for other psychical harm only late and grudgingly, and then only in very clear cases. In tort, clear means close--close to the victim, close to the accident, close to the defendant.".

I do not doubt that public perception has played a substantial role in the development of this branch of the law. But nowadays we must accept the medical reality that psychiatric harm may be more serious than physical harm. It is therefore necessary to consider whether there are other objective policy considerations which may justify different rules for the recovery of compensation for physical injury and psychiatric harm. And in my view it would be insufficient to proceed on the basis that there are unspecified policy considerations at stake. If, as I believe, there are such policy considerations it is necessary to explain what the policy considerations are so that the validity of my assumptions can be critically examined by others.

My impression is that there are at least four distinctive features of claims for psychiatric harm which in combination may account for the differential treatment. Firstly, there is the complexity of drawing the line between acute grief and psychiatric harm: see Hedley, Nervous Shock: Wider Still and Wider, 1997 C.L.J. 254. The symptoms may be the same. But there is greater diagnostic uncertainty in psychiatric injury cases than in physical injury cases. The classification of emotional injury is often controversial. In order to establish psychiatric harm expert evidence is required. That involves the calling of consultant psychiatrists on both sides. It is a costly and time consuming exercise. If claims for psychiatric harm were to be treated as generally on a par with physical injury it would have implications for the administration of justice. On its own this factor may not be entitled to great weight and may not outweigh the considerations of justice supporting genuine claims in respect of pure psychiatric injury. Secondly, there is the effect of the expansion of the availability of compensation on potential claimants who have witnessed gruesome events. I do not have in mind fraudulent or bogus claims. In general it ought to be possible for the administration of justice to expose such claims. But I do have in mind the unconscious effect of the prospect of compensation on potential claimants. Where there is generally no prospect of recovery, such as in the case of injuries sustained in sport, psychiatric harm appears not to obtrude often. On the other hand, in the case of industrial accidents, where there is often a prospect of recovery of compensation, psychiatric harm is repeatedly encountered and often endures until the process of claiming compensation comes to an end: see James v. Woodall Duckham Construction Co. Ltd. [1969] 1 W.L.R. 903 (CA). The litigation is sometimes an unconscious disincentive to rehabilitation. It is true that this factor is already present in cases of physical injuries with concomitant mental suffering. But it may play a larger role in cases of pure psychiatric harm, particularly if the categories of potential recovery are enlarged. For my part this factor cannot be dismissed.

The third factor is important. The abolition or a relaxation of the special rules governing the recovery of damages for psychiatric harm would greatly increase the class of persons who can recover damages in tort. It is true that compensation is routinely awarded for psychiatric harm where the plaintiff has suffered some physical harm. It is also well established that psychiatric harm resulting from the apprehension of physical harm is enough: Page v. Smith [1996] AC 155. These two principles are not surprising. In built in such situations are restrictions on the classes of plaintiff who can sue: the requirement of the infliction of some physical injury or apprehension of it introduces an element of immediacy which restricts the category of potential plaintiffs. But in cases of pure psychiatric harm there is potentially a wide class of plaintiffs involved. Fourthly, the imposition of liability for pure psychiatric harm in a wide range of situations may result in a burden of liability on defendants which may be disproportionate to tortious conduct involving perhaps momentary lapses of concentration, e.g. in a motor car accident.

The wide scope of potential liability for pure psychiatric harm is not only illustrated by the rather unique events of Hillsborough but also by of accidents involving trains, coaches and buses, and the everyday occurrence of serious collisions of vehicles all of which may result in gruesome scenes. In such cases there may be many claims for psychiatric harm by those who have witnessed and in some ways assisted at the scenes of the tragic events. Moreover, protagonists of very wide theories of liability for pure psychiatric loss have suggested that "workplace claims loom large as the next growth area of psychiatric injury law", the paradigm case being no doubt a workman who has witnessed a tragic accident to an employee: Mullany and Handford, Hillsborough Replayed (1997) 113 L.Q.R. 410, at 415.

The police officers claims

In the present case, the police officers were more than mere bystanders. They were all on duty at the stadium. They were all involved in assisting in the course of their duties in the aftermath of the terrible events. And they have suffered debilitating psychiatric harm. The police officers therefore argue, and are entitled to argue, that the law ought to provide compensation for the wrong which caused them harm. This argument cannot be lightly dismissed. But I am persuaded that a recognition of their claims would substantially expand the existing categories in which compensation can be recovered for pure psychiatric harm. Moreover, as the majority in the Court of Appeal was uncomfortably aware, the awarding of damages to these police officers sits uneasily with the denial of the claims of bereaved relatives by the decision of the House of Lords in Alcock. The decision of the Court of Appeal has introduced an imbalance in the law of tort which might perplex the man on the Underground. Since the answer may be that there should be compensation in all these categories I must pursue the matter further.

The case law

In order to understand the law as it stands it is necessary to trace in outline its development. In Dulieu v. White & Sons [1901] 2 KB 669 the Court of Appeal enunciated a narrow and relatively simple rule: psychiatric injury was only actionable if it arose from the plaintiff's reasonably apprehended fear for his safety. But in Hambrook v. Stokes Brothers [1925] l.K.B. 141 the Court of Appeal rejected the limitation laid down in Dulieu v. White & Sons in favour of a mother who suffered psychiatric injury as a result an apprehension of an injury to her child from whom she had just parted. The mother was described as "courageous and devoted to her child" and was allowed to recover. The next development was the decision of the House of Lords in Bourhill v. Young [1943] AC 92. There are dicta in this case which appear to favour the confining of liability for psychiatric injury to those within the area of physical harm. But the status of Hambrook v. Stokes Brothers was left unclear. Then came the decision in McLoughlin v. O'Brian [1983] l.A.C. 410. The plaintiff's husband and children were injured in a car accident. She was informed and saw the serious injuries of her husband and children in hospital. She also was informed that one of her children had been killed. She suffered psychiatric injury. The House of Lords upheld the plaintiff's claim. There are passages in the speeches which tend to support a wide theory of liability for psychiatric injury. Lord Wilberforce countenanced "the real need to the law to place some limitation upon the extent of admissible claims": 442A. For somewhat different reasons Lord Russell of Killowen, Lord Scarman and Lord Bridge of Harwich regarded limitations on the ground of policy considerations as essentially arbitrary: see also Lord Edmund Davies, 425G. This decision was given at the peak of the expansion of tort liability in the wake of Anns v. Merton London Borough Council [1978] AC 728.

In 1982 in McLoughlin the House acted on the reassuring picture that the ". . . scarcity of cases which have occurred in the past, and the modest sums recovered, give some indication that fears of the flood of litigation may be exaggerated . . .": at 421H, per Lord Wilberforce. This assumption has been falsified by the growth of claims for psychiatric damage in the last ten years. In Fear for the Future: Liability for Infliction of Psychiatric Disorder, essay in Torts in the Nineties (1997) ed. Nicholas J Mullany, the editor has attested to the "growing appreciation that the scope for psychiatric suits is much wider than traditionally perceived" and he listed the expansion into claims for workplace stress; suits by members of the armed services in respect of mental suffering; claims for psychiatric damage against medical practitioners and health authorities; and so forth. In addition the same author stated that there has in recent years been a steady growth in Australia in the more common place psychiatric injury proceedings based on the death, injury or imperilment of loved ones or fear of ones own safety: at 112. Moreover, nowadays it would be quite unrealistic to describe awards for psychiatric damage as modest. In any event, since McLoughlin the pendulum has swung and the House of Lords have taken greater account of policy considerations both in regard to economic loss and psychiatric injury.

The leading decision of the House of Lords is Alcock (1992). Before this case the general rule was that only parents and spouses could recover for psychiatric harm suffered as a result of witnessing a traumatic event. In Alcock the group of plaintiffs who sued for psychiatric injury resulting from the events at Hillsborough included relatives who were in the stadium. The House dismissed all the claims including the claim of a plaintiff who was himself witnessed the scenes at the football ground where two of his brothers died: see Lord Ackner's comment, at p. 406A that "the quality of brotherly love is well known to differ widely" This decision established that a person who suffers reasonably foreseeable psychiatric illness as a result of another person's death or injury cannot recover damages unless he can satisfy three requirements, viz: (i) that he had a close tie of love and affection with the person killed, injured or imperilled; (ii) that he was close to the incident in time and space; (iii) that he directly perceived the incident rather than, for example, hearing about it from a third person.

Lord Oliver observed that the law was not entirely satisfactory or logically defensible but he thought that considerations of policy made it explicable: at 418. Professor Jane Stapleton has described the law as stated in Alcock as difficult to justify: see In Restraint of Tort, an essay in Frontiers of Liability, ed., by Birks, 1994. She remarked, at 95:

no more

"In future cases will it not be a grotesque sight to see relatives scrabbling to prove their especial love for the deceased in order to win money damages and for the defendant to have to attack that argument?"

But Alcock is the controlling decision.

The decision of the House of Lords in Page v. Smith [1996] AC 155 was the next important development in this branch of the law. The Plaintiff was directly involved in a motor car accident. He was within the range of potential physical injury. As a result of the accident he suffered from chronic fatigue syndrome. In this context Lord Lloyd of Berwick adopted a distinction between primary and secondary victims: Lord Ackner and Lord Browne-Wilkinson agreed. Lord Lloyd said that a plaintiff who had been within the range of foreseeable injury was a primary victim. Mr Page fulfilled this requirement and could in principle recover compensation for psychiatric loss. In my view it follows that all other victims, who suffer pure psychiatric harm, are secondary victims and must satisfy the control mechanisms laid down in Alcock. There has been criticism of this classification: see Teff, Liability for Negligently inflicted Psychiatric Harm: Justifications and Boundaries, 1998 C.L.J. 91, at 93. But, if the narrow formulation by Lord Lloyd of Berwick of who may be a primary victim is kept in mind, this classification ought not to produce inconsistent results. In any event, the decision of the House of Lords in Page v. Smith was plainly intended, in the context of pure psychiatric harm, to narrow the range of potential secondary victims. The reasoning of Lord Lloyd and the Law Lords who agreed with him was based on concerns about an ever widening circle of plaintiffs.

The proceedings below

Waller J. rejected the claims of the police officers. The majority in the Court of Appeal upheld their claims. The first route followed by the majority was to allow some claims because the police officers were on duty in the stadium when they witnessed the gruesome events. The second route was to allow some claims because the police officers were said to be rescuers.

The employment argument

The majority in the Court of Appeal upheld the argument of counsel for two police officers that they fall into a special category. That argument was again deployed on appeal to the House. The argument was that the present case can be decided on conventional employer's liability principles. And counsel relies on the undoubted duty of an employer to protect employers from harm through work. It is true that there is no contract between police officers and a chief constable. But it would be artificial to rest a judgment on this point: the relationship between the police officers and the chief constable is closely analogous to a contract of employment. And I am content to approach the problem as if there was an ordinary contract of employment between the parties. Approaching the matter in this way it became obvious that there were two separate themes to the argument. The first rested on the duty of an employer to care for the safety of his employees and to take reasonable steps to safeguard them from harm. When analysed this argument breaks down. It is a non sequitur to say that because an employer is under a duty to an employee not to cause him physical injury, the employer should as a necessary consequence of that duty (of which there is no breach) be under a duty not to cause the employee psychiatric injury: see Hilson, Nervous Shock and Categorization of Victims, [1998] Tort L.R. 37, at 42. The rules to be applied when an employee brings an action against his employer for harm suffered at his workplace are the rules of tort. One is therefore thrown back to the ordinary rules of the law of tort which contain restrictions on the recovery of compensation for psychiatric harm. This way of putting the case does not therefore advance the case of the police officers. The duty of an employer to safeguard his employees from harm could also be formulated in contract. In that event, and absent relevant express provisions, a term is implied by law into the contract as an incident of a standardized contract: see Scally v. Southern Health and Social Services Board [1992] 1.A.C. 294. But such a term could not be wider in scope than the duty imposed by the law of tort. Again one is thrown back to the ordinary rules of the law of tort. The first way of formulating the argument based on the duty of an employer does not therefore assist the police officers.

The second theme is on analysis an argument as to where the justice lay on this occasion. One is considering the claims of police officers who sustained serious psychiatric harm in the course of performing and assisting their duties in harrowing circumstances. That is, a weighty moral argument: the police perform their duties for the benefit of us all. The difficulty is, however, twofold. First, the pragmatic rules governing the recovery of damages for pure psychiatric harm do not at present include police officers who sustain such injuries while on duty. If such a category were to be created by judicial decision, the new principle would be available in many different situations, e.g. doctors and hospital workers who are exposed to the sight of grievous injuries and suffering. Secondly, it is common ground that police officers who are traumatized by something they encounter in their work have the benefit of statutory schemes which permit them to retire on pension. In this sense they are already better off than bereaved relatives who were not allowed to recover in Alcock. The claim of the police officers on our sympathy, and the justice of the case, is great but not as great as that of others to whom the law denies redress.

The rescue argument

The majority in the Court of Appeal held that three of the police officers could be classed as rescuers because they actively gave assistance in the aftermath of the tragedy: the majority used the concept of rescuer in an undefined but very wide sense: see Rose L.J., at 1220; Henry L.J. expressly agreed with this passage. This reasoning was supported by counsel for the appellant on the appeal.

The law has long recognized the moral imperative of encouraging citizens to rescue persons in peril. Those who altruistically expose themselves to danger in an emergency to save others are favoured by the law. A rescue attempt to save someone from danger will be regarded as foreseeable. A duty of care to a rescuer may arise even if the defendant owed no duty to the primary victim, for example, because the latter was a trespasser. If a rescuer is injured in a rescue attempt, a plea of volenti non fit injuria will not avail a wrongdoer. A plea of contributory negligence will usually receive short shrift. A rescuer's act in endangering himself will not be treated as a novus actus interveniens. The meaning given to the concept of a rescuer in these situations is of no assistance in solving the concrete case before the House. Here the question is: who may recover in respect of pure psychiatric harm sustained as a rescuer?

Counsel for the appellant is invoking the concept of a rescuer as an exception to the limitations recognized by the House of Lords in Alcock and Page v. Smith. The restrictive rules, and the underlying policy considerations, of the decisions of the House are germane. The specific difficulty counsel faces is that it is common ground that none of the four police officers were at any time exposed to personal danger and none thought that they were so exposed. Counsel submitted that this is not a requirement. He sought comfort in the general observations in Alcock of Lord Oliver about the category of "participants": see 407E. None of the other Law Lords in Alcock discussed this category. Moreover, the issue of rescuers entitlement to recover for psychiatric harm was not before the House on that occasion and Lord Oliver was not considering the competing arguments presently before the House. The explanation of Lord Oliver's observations has been the subject of much debate. It was also vigorously contested at the bar. In my view counsel for the appellant has tried to extract too much from general observations not directed to the issue now before the House: see also the careful analysis of the Lord President in Robertson v. Forth Road Bridge Joint Board [1995] SC 364, at 371G-372B. Counsel was only able to cite one English decision in support of his argument namely the first instance judgment in Chadwick v. British Railways Board [1967] 1.Q.B. 912. Mr Chadwick had entered a wrecked railway carriage to help and work among the injured. There was clearly a risk that the carriage might collapse. Waller J. (later Lord Justice Waller) said (at 918A)

"although there was clearly an element of personal danger in what Mr. Chadwick was doing, I think I must deal with this case on the basis that it was the horror of the whole experience which caused his reaction."

On the judge's findings the rescuer had passed the threshold of being in personal danger but his psychiatric injury was caused by "the full horror of his experience" when he was presumably not always in personal danger. This decision has been cited with approval: see McLoughlin v. O'Brian, at 419B, per Lord Wilberforce; at 424D-G, per Lord Edmund Davies: at 437H--438A, per Lord Bridge of Harwich; and in Alcock at 408B-D per Lord Oliver. I too would accept that Chadwick was correctly decided. But it is not authority for the proposition that a person who never exposed himself to any personal danger and never thought that he was in personal danger can recover pure psychiatric injury as a rescuer. In order to recover compensation for pure psychiatric harm as rescuer it is not necessary to establish that his psychiatric condition was caused by the perception of personal danger. And Waller J. rightly so held. But in order to contain the concept of rescuer in reasonable bounds for the purposes of the recovery of compensation for pure psychiatric harm the plaintiff must at least satisfy the threshold requirement that he objectively exposed himself to danger or reasonably believed that he was doing so. Without such limitation one would have the unedifying spectacle that, while bereaved relatives are not allowed to recover as in Alcock, ghoulishly curious spectators, who assisted in some peripheral way in the aftermath of a disaster, might recover. For my part the limitation of actual or apprehended dangers is what proximity in this special situation means. In my judgment it would be an unwarranted extension of the law to uphold the claims of the police officers. I would dismiss the argument under this heading.

Thus far and no further

My Lords, the law on the recovery of compensation for pure psychiatric harm is a patchwork quilt of distinctions which are difficult to justify. There are two theoretical solutions. The first is to wipe out recovery in tort for pure psychiatric injury. The case for such a course has been argued by Professor Stapleton. But that would be contrary to precedent and, in any event, highly controversial. Only Parliament could take such a step. The second solution is to abolish all the special limiting rules applicable to psychiatric harm. That appears to be the course advocated by Mullany and Handford, Tort Liability for Psychiatric Damage: The Law of Nervous Shock, (1993). They would allow claims for pure psychiatric damage by mere bystanders: see (1997) 113 L.Q.R. 410, at 415. Precedent rules out this course and, in any event, there are cogent policy considerations against such a bold innovation. In my view the only sensible general strategy for the courts is to say thus far and no further. The only prudent course is to treat the pragmatic categories as reflected in authoritative decisions such as Alcock and Page v. Smith as settled for the time being but by and large to leave any expansion or development in this corner of the law to Parliament. In reality there are no refined analytical tools which will enable the courts to draw lines by way of compromise solution in a way which is coherent and morally defensible. It must be left to Parliament to undertake the task of radical law reform.

Conclusion

My Lords, I am in substantial agreement with the reasons given by Waller J. for dismissing the claims of the police officers. In my judgment the Court of Appeal erred in reversing Waller J. in respect of the claims under consideration. For these reasons, as well as the reasons given by Lord Hoffmann, I would allow the appeals.

LORD HOFFMANN

My Lords,

On 15 April 1989 there was a horrifying disaster at the Hillsborough Football Stadium in Sheffield. The pressure of crowds trying to get into the ground crushed 95 people on the terraces to death and injured many more. On that day the plaintiffs (respondents to this appeal) were serving members of the South Yorkshire Police Force on duty at the stadium or elsewhere in Sheffield. Each became in some way involved in the dreadful aftermath. Two helped to carry the dead and dying. Two tried unsuccessfully to give resuscitation to those who had been laid out on the ground. One assisted at the hospital mortuary.

As a result of their experiences, the plaintiffs have suffered from what has been diagnosed as post-traumatic stress disorder, a medically recognised psychiatric illness. The symptoms have affected their ability to work and their private lives. They claim damages in negligence against Chief Constable of South Yorkshire and two other defendants. There were of course many people at the stadium that day who also tried as best they could to help the victims: other policemen, first aid workers, ambulance men and members of the public. Some of them, together with bereaved relatives and friends, have also developed psychiatric illnesses. The claims of some of the relatives were considered by your Lordships' House in Alcock v. Chief Constable of South Yorkshire [1992] 1 AC 310. For reasons which I shall discuss, they were all rejected. But the plaintiffs in this appeal say that the police are in a different position. First, they were in a position analogous to employees of the Chief Constable and they claim that the employment relationship gives rise to duties which are not owed to strangers. Secondly, they were present and assisted at the catastrophe and were not merely passive and helpless bystanders. In order to examine the merits of these arguments, the claims of the five members of the police in this case (as well as some others which are not the subject of appeal) were selected as test cases to be tried together. It is admitted that the disaster was caused by the negligence of persons for whom the defendants were vicariously liable. The only question is whether in such circumstances the law allows the recovery of compensation for the type of injury which the plaintiffs have suffered.

Compensation for personal injury caused by negligence is ordinarily recoverable if the defendant ought reasonably to have foreseen than his conduct might cause such injury. But the common law has been reluctant to equate psychiatric injury with other forms of personal injury. In Victorian Railway Commissioners v. Coultas (1883) 13 App. Cas. 222 the Privy Council held that compensation for such injury was not recoverable at all. The main reason which the Board gave for denying recovery was the evidential difficulty of deciding upon the causes of psychiatric symptoms at a time when very little was known about the workings of the mind. Despite scientific advances, this remains a serious problem. As Lord Wilberforce noted in 1982, "the area of ignorance seems to expand with that of knowledge" (McLoughlin v. O'Brian [1983] 1 AC 410, 418). At any rate, the courts have developed sufficient confidence in medical expertise to be willing to award damages for mental disturbances which manifest themselves in bodily symptoms (such as a miscarriage) or in a "recognised psychiatric illness." The latter is distinguished from shock, fear, anxiety or grief which are regarded as a normal consequences of a distressing event and for which damages are not awarded. Current medical opinion suggests that this may be a somewhat arbitrary distinction; the limits of normal reaction to stressful events are wide and debatable, while feelings of terror and grief may have as devastating an effect upon people's lives as the "pain and suffering" consequent upon physical injury, for which damages are regularly awarded.

For a long time during this century it remained unclear whether the basis of liability for causing a recognised psychiatric illness was simply a question of foreseeability of that type of injury in the same way as in the case of physical injury. The decision of the House of Lords in Bourhill v. Young [1943] AC 92, appeared to many to combine what was in theory a simple foreseeability test with a robust wartime view of the ability of the ordinary person to suffer horror and bereavement without ill effect. Cases soon afterwards, like King v. Phillips [1953] 1 Q.B. 429, followed this approach, treating foreseeability as a question of fact but keeping potential liability within narrow bounds by taking a highly restrictive view of the circumstances in which it was foreseeable that psychiatric injury might be caused. But such decisions were criticised as out of touch with reality. Everyone knew that some people did suffer psychiatric illnesses as a result of witnessing distressing accidents in which other people, particularly close relatives, were involved. Some judges, sympathetic to the plaintiff in the particular case, took the opportunity to find as a fact that psychiatric injury had indeed been foreseeable. This made it difficult to explain why plaintiffs in other cases had failed. It seemed that if the foreseeability test was to be taken literally and applied in the same way as the test for liability for physical injury, it would be hard to know where the limits of liability could be drawn. In all but exceptional cases, the only question would be whether on the medical evidence, the psychiatric condition had been caused by the defendant's negligent conduct.

There was a time when it seemed that English law might arrive at this position. It came within a hair's breadth of doing so in McLoughlin v. O'Brian [1983] 1 AC 410, one of those cases in which one feels that a slight change in the composition of the Appellate Committee would have set the law on a different course. But the moment passed and when the question next came before your Lordships' House in Alcock v. Chief Constable of South Yorkshire [1992] 1 AC 310, judicial attitudes had changed. The view which had for some time been in the ascendancy, that the law of torts should, in principle aspire to provide a comprehensive system of corrective justice, giving legal sanction to a moral obligation on the part of anyone who has caused injury to another without justification to offer restitution or compensation, had been abandoned in favour of a cautious pragmatism. The House decided that liability for psychiatric injury should be restricted by what Lord Lloyd of Berwick (in Page v. Smith [1996] AC 155, 189) afterwards called "control mechanisms", that is to say, more or less arbitrary conditions which a plaintiff had to satisfy and which were intended to keep liability within what was regarded as acceptable bounds.

Alcock was, as I have said, a case which also arose out of the Hillsborough disaster. The plaintiffs were persons who had seen the events from other parts of the stadium or on television, or heard about it from others and then found that their relatives were among the dead. It was assumed for the purposes of the appeal that these experiences had caused them psychiatric injury. The House established certain additional conditions to be satisfied for a successful claim. I state them in a summary form which I think is sufficiently accurate for the purposes of the present discussion but it may for other purposes require qualification and does not purport to be a complete description.

(1) The plaintiff must have close ties of love and affection with the victim. Such ties may be presumed in some cases (e.g. spouses, parent and child) but must otherwise be established by evidence.

(2) The plaintiff must have been present at the accident or its immediate aftermath.

(3) The psychiatric injury must have been caused by direct perception of the accident or its immediate aftermath and not upon hearing about it from someone else.

The result of these various control mechanisms was that none of the Hillsborough relatives was held entitled to recover. Mr. Hicks, whose two teenage daughters died, failed in limine and did not make a personal claim because his grief had not given rise to recognised psychiatric illness. His attempt to make a symbolic claim on behalf of his daughters was dismissed by your Lordships' House on the ground that the fear and terror which they must have felt in the moments before death were normal human emotions for which damages are not awarded: Hicks v. Chief Constable of the South Yorkshire Police [1992] 2 All ER 65. Mr. Harrison, who was present elsewhere at the ground and whose two brothers died, failed condition (1) because the House refused to presume that close ties of love and affection exist between brothers and he had adduced no evidence to prove that they existed in his case. Mr. and Mrs Copoc, whose son died, failed condition (2) because they were not present at the ground but saw the scenes on television. Mr. Alcock, who identified his brother-in-law in the mortuary at midnight, failed condition (3) because he was not in time for the immediate aftermath of the tragedy. The claims of other relatives were dismissed on similar grounds.

My Lords, this story of the ebb and flow of tort liability for psychiatric injury has often been told and I have recounted it again at some length only because I think it must be borne in mind when we come to deal with the authorities. In order to give due weight to the earlier decisions, particularly at first instance, it is necessary to have regard to their historical context. They cannot simply be laid out flat and pieced together to form a timeless mosaic of legal rules. Some contained the embryonic forms of later developments; others are based on theories of liability which had respectable support at the time but have since been left stranded by the shifting tides.

The position which the law has reached as a result of Alcock v. Chief Constable of South Yorkshire [1992] 1 AC 310 has not won universal approval. The control mechanisms have been criticised as drawing distinctions which the ordinary man would find hard to understand. Jane Stapleton has said that a mother who suffers psychiatric injury after finding her child's mangled body in a mortuary "might wonder why the law rules her child's blood too dry to found an action": see The Frontiers of Liability ed. Peter Birks, O.U.P. (1994) Volume 2, p. 84. Equally, the spectacle of a plaintiff, who has, ex hypothesi, suffered psychiatric illness in consequence of his brother's death or injury, being cross-examined on the closeness of their ties of love and affection and then perhaps contradicted by the evidence of a private investigator, might not be to everyone's taste: see the Law Commission Report on Liability for Psychiatric Illness (Law Com. No. 249) at para. 6.24).

Academic writers have made contradictory but equally radical suggestions for reform. Mullany and Handford, in their excellent book The Liability for Psychiatric Damage (Sydney, 1993), advocate getting rid of the control mechanisms and, in the light of advances in psychiatric knowledge, equating psychiatric injury to physical injury. Jane Stapleton, on the other hand, would abolish recovery for psychiatric injury altogether and revert to the law as stated in Victorian Railway Commissioners v. Coultas (1888) 13 App Cas 222: see the article to which I have already referred.

The appeal of these two opposing proposals rather depends upon where one starts from. If one starts from the proposition that in principle the law of torts is there to give legal force to an Aristotelian system of corrective justice, then there is obviously no valid distinction to be drawn between physical and psychiatric injury. On this view, the control mechanisms merely reflect a vulgar scepticism about the reality of psychiatric injury or a belief that it is less worthy of compensation than physical injury: therein the patient must minister to himself. On the other hand, if one starts from the imperfect reality of the way the law of torts actually works, in which the vast majority of cases of injury and disability, both physical and psychiatric, go uncompensated because the persons (if any) who caused the damage were not negligent, or because the plaintiff lacks the evidence or the resources to prove to a court that they were negligent, or because the potential defendants happen to have no money, then questions of distributive justice tend to intrude themselves. Why should X receive generous compensation for his injury when Y receives nothing? Is the administration of so arbitrary and imperfect a system of compensation worth the very considerable cost? On this view, a uniform refusal to provide compensation for psychiatric injury adds little to the existing stock of anomaly in the law of torts and at least provides a rule which is easy to understand and cheap to administer.

The Law Commission in its recent Report (supra) inclines somewhat to the Mullany and Handford point of view by recommending that the condition of close ties of love and affection for secondary victims be retained in a modified form but that the other two be abolished. The reason given for retention of any control mechanism was that:

"a review of the medical literature has led us to believe that the adoption of a simple foreseeability test would or could result in a significant increase in the number of claims which, at least at this point in time, would be unacceptable."

I shall in due course return to this concept of unacceptability and try to analyse what it means. But I shall not enter further into the merits of the various proposals for reform because neither of the radical solutions, or indeed the Law Commission solution, is open to your Lordships. It is too late to go back on the control mechanisms as stated in Alcock. Until there is legislative change, the courts must live with them and any judicial developments must take them into account.

The control mechanisms were plainly never intended to apply to all cases of psychiatric injury. They contemplate that the injury has been caused in consequence of death or injury suffered (or apprehended to have been suffered or as likely to be suffered) by someone else. In Page v. Smith [1996] AC 155, 184 Lord Lloyd of Berwick described such a plaintiff as a "secondary victim" who was "in the position of a spectator or bystander." He described the plaintiff in that case (who had suffered psychiatric injury in consequence of being involved in a minor motor accident) as a "primary victim" who was "directly involved in the accident and well within the range of foreseeable physical injury." The issue in Page v. Smith was whether it is sufficient that a primary victim who, in consequence of a foreseeable accident, has suffered psychiatric injury, should have been within the range of foreseeable physical injury or whether it must have been foreseeable, in the light of the circumstances of the accident as it actually happened, that he would suffer psychiatric illness. A majority of your Lordships held that foreseeability of physical injury was enough to found a claim for any psychiatric injury which the accident caused.

This question does not arise in the present case, but the classification into primary and secondary victims has been debated at length. The plaintiffs say that they were primary victims because they were not "spectators or bystanders." The defendants say that the plaintiffs were secondary victims because they were not "within the range of foreseeable physical injury." Both arguments have some support from the speeches in Page v. Smith [1996] AC 155, which did not have the present question in mind. Essentially, however, as I said at the beginning of this speech, the plaintiffs draw two distinctions between their position and that of spectators or bystanders. The first is that they had a relationship analogous to employment with the Chief Constable. Although constitutionally a constable holds an office rather than being employed, there is no dispute that his Chief Constable owes him the same duty of care which he would to an employee. The plaintiffs say that they were therefore owed a special duty as which required the Chief Constable and those for whom he was vicariously liable to take reasonable care not to expose them to unnecessary risk of injury, whether physical or psychiatric. Secondly, the plaintiffs (and in this respect there is no difference between the police and many others in the crowd that day) did more than stand by and look. They actively rendered assistance and should be equated to "rescuers", who, it was said, always qualify as primary victims.

My Lords, I shall consider first the claim to primary status by virtue of the employment relationship. Mr. Hytner Q.C., for the plaintiffs, said that prima facie an employer's duty required him to take reasonable steps to safeguard his employees from unnecessary risk of harm. The word "unnecessary" must be stressed because obviously a policeman takes the risk of injury which is an unavoidable part of his duty. But there is no reason why he should be exposed to injuries which reasonable care could prevent. Why, in this context, should psychiatric injury should be treated differently from physical injury? He referred to Walker v. Northumberland County Council [1995] 1 All ER 737 an employee recovered damages for a mental breakdown, held to have been foreseeably caused by the stress and pressure of his work as a social services officer. This, he said, showed that no distinction could be made.

I think, my Lords, that this argument really assumes what it needs to prove. The liability of an employer to his employees for negligence, either direct or vicarious, is not a separate tort with its own rules. It is an aspect of the general law of negligence. The relationship of employer and employee establishes the employee as a person to whom the employer owes a duty of care. But this tells one nothing about the circumstances in which he will be liable for a particular type of injury. For this one must look to the general law concerning the type of injury which has been suffered. It would not be suggested that the employment relationship entitles the employee to recover damages in tort (I put aside contractual liability, which obviously raises different questions) for economic loss which would not ordinarily be recoverable in negligence. The employer is not, for example, under a duty in tort to take reasonable care not to do something which would cause the employee purely financial loss, e.g. by reducing his opportunities to earn bonuses. The same must surely be true of psychiatric injury. There must be a reason why, if the employee would otherwise have been regarded as a secondary victim, the employment relationship should require him to be treated as a primary one. The employee in to Walker v. Northumberland County Council [1995] 1 All ER 737 was in no sense a secondary victim. His mental breakdown was caused by the strain of doing the work which his employer had required him to do.

Should the employment relationship be a reason for allowing an employee to recover damages for psychiatric injury in circumstances in which he would otherwise be a secondary victim and not satisfy the Alcock control mechanisms? I think, my Lords, that the question vividly illustrates the dangers inherent in applying the traditional incrementalism of the common law to this part of the law of torts. If one starts from the employer's liability in respect of physical injury, it seems an easy step, even rather forward-looking, to extend liability on the same grounds to psychiatric injury. It makes the law seem more attuned to advanced medical thinking by eliminating (or not introducing) a distinction which rests upon uneasy empirical foundations. It is important, however to have regard, not only to how the proposed extension of liability can be aligned with cases in which liability exists, but also to the situations in which damages are not recoverable. If one then steps back and looks at the rules of liability for psychiatric injury as a whole, in their relationship with each other, the smoothing of the fabric at one point has produced an ugly ruck at another. In their application to other secondary victims, the Alcock control mechanisms stand obstinately in the way of rationalisation and the effect is to produce striking anomalies. Why should the policemen, simply by virtue of the employment analogy and irrespective of what they actually did, be treated different from first aid workers or ambulance men? In the Court of Appeal, where four of the plaintiffs succeeded on this ground, Rose L.J. denied that he was giving preference to "policemen over laymen." He said that the distinction existed because "the court has long recognised a duty of care to guard employees and rescuers against all kinds of injury." For the moment I leave aside the "rescuers", where obviously no distinction based on the employment relationship need be made. But with respect to employees as such, Rose L.J. states a broad proposition as settled law. I think it is debatable whether the authorities have gone so far as to recognise a duty to guard employees against psychiatric injury suffered as a result of injury to others and I shall discuss them in a moment. Apart from authority, however, it seems to me that Rose L.J. is stating the distinction rather than explaining it.

Henry L.J. said that employees were in a different position because their contracts of employment (or the analogous duties of policemen) required them to stay at the ground. Other people could avert their eyes and go but the police had to stay. I do not find this a satisfactory distinction from the cases of other people (such as St John's Ambulance workers) who were presumably also under an obligation to stay but could not sue their employer or even those who had no legal duties (such as doctors who happened to be in the crowd) but stayed out of a sense of moral obligation to see if there was anything they could do to help.

In principle, therefore, I do not think it would be fair to give police officers the right to a larger claim merely because the disaster was caused by the negligence of other policemen. In the circumstances in which the injuries were caused, I do not think that this is a relevant distinction and if it were to be given effect, the law would not be treating like cases alike. I must therefore consider whether the authorities require a contrary conclusion. And in examining them, it is important to bear in mind, as I said earlier, that they are not contemporaneous statements of the law but represent legal thinking at different points in half a century of uneven development.

The plaintiffs rely upon four cases as establishing the right of an employee to recover for psychiatric injury caused by witnessing or apprehending injury which his employer's negligence has caused to others. Three are English and the other is a case in the High Court of Australia. Only one of the English cases (Dooley v. Cammell Laird & Co. Ltd. [1951] 1 Lloyd's L.R. 271 is reported in full; the reasoning in Galt v. British Railways Board [1983] 113 N.L.J. 870 has been condensed to a single sentence and that of Wiggs v. British Railways Board (The Times, 4 February 1986) is also abbreviated. All appear to have been ex tempore first instance judgments given on circuit. I think that on a fair reading, they were each regarded by the judges who decided them as raising one question of fact, namely whether psychiatric injury to the plaintiff was a foreseeable consequence of the defendant's negligent conduct. This was in accordance with the law as it was thought to be at the time. There was no reference to the control mechanisms, which had not yet been invented. In Wigg, Tucker J. expressly said that the only question was that of foreseeability, referring to the speech of Lord Bridge of Harwich in McLoughlin v. O'Brian [1983] 1 AC 410. This was a view which might well have prevailed, but the subsequent retreat from principle in Alcock v. Chief Constable of South Yorkshire [1992] 1 AC 310 meant that it, and the other two cases, had either to be given up as wrongly decided or explained on other grounds. The same is true of the Australian case of Mount Isa Mines v. Pusey [1971] 125 C.L.R. 382, whose interest resides entirely in the judgment of Windeyer J. Only one of the other judges found it necessary to discuss the principles of liability for psychiatric injury and he expressly refrained from considering whether it could be based upon the employee relationship. Windeyer J. thought that it could, but only as part of his wider thesis that foreseeable physical and foreseeable psychiatric injury should not be distinguished. He was at pains to say (at p. 404) that although the plaintiff was owed a duty of care as employee, his position was no different from that of anyone else to whom injury, whether physical or psychiatric, was reasonably foreseeable.

In Alcock itself, Lord Oliver of Aylmerton attempted an ex post facto rationalisation of the three English cases by saying that in each, the plaintiff had been put in a position in which he was, or thought he was about to be or had been, the immediate instrument of death or injury to another. In Wigg, for example, the plaintiff was the driver of a train which had caused the death of a passenger by moving off when he was trying to board. The driver had started because the guard, for whom the employer was vicariously liable, had negligently given the signal. This is an elegant, not to say ingenious, explanation, which owes nothing to the actual reasoning (so far as we have it) in any of the cases. And there may be grounds for treating such a rare category of case as exceptional and exempt from the Alcock control mechanisms. I do not need to express a view because none of the plaintiffs in this case come within it. In Robertson v. Forth Road Bridge Joint Board [1995] SC 364 Lord Hope adopted Lord Oliver's explanation of the English cases and rejected a claim for psychiatric injury by employees who had witnessed the death of a fellow employee in the course of being engaged on the same work. I respectfully agree with the reasoning of my noble and learned friend, which I regard as a rejection of the employment relationship as in itself a sufficient basis for liability.

The second way in which the plaintiffs put their case is that they were not "bystanders or spectators" but participants in the sense that they actually did things to help. They submit that there is an analogy between their position and that of a "rescuer", who, on the basis of the decision of Waller J. in Chadwick v. British Railways Board [1967] 1 W.L.R. 912, is said to be treated as a primary victim, exempt from the control mechanisms.

In Chadwick, the plaintiff suffered psychiatric injury as a result of his experiences in assisting the victims of a railway accident. He spent twelve hours crawling in the wreckage, helping people to extricate themselves and giving pain killing injections to the injured. Waller J. said (at p. 921) that it was foreseeable that "somebody might try to rescue passengers and suffer injury in the process." The defendants therefore owed a duty of care to the plaintiff. He went on to say that it did not matter that the injury suffered was psychiatric rather than physical but in any event "shock was foreseeable and . . . rescue was foreseeable." Thus the judge's reasoning is based purely upon the foreseeability of psychiatric injury in the same way as in other cases of that time. And I think there can be no doubt that if foreseeability was the only question, the judge's conclusion was unexceptionable.

References in the authorities to rescuers sometimes give the impression that they are a category of persons who would not qualify for compensation under the strict rules of the law of negligence but receive special treatment on grounds of humanity and as a reward for altruism. A florid passage by Cardozo J. in Wagner v. International Railway Company 232 N.Y. 176, 180 (1921) is frequently quoted. If rescuers formed a specially privileged category of plaintiff, one would expect that the rule would give rise to a definitional problem about who counted�K��K as a rescuer and so qualified for special treatment. In fact, as one can see from the absence of any such problem in the cases, rescuers can be accommodated without difficulty in the general principles of the law of negligence. There are two questions which may arise. The first is whether injury to the rescuer was foreseeable. There is usually no difficulty in holding that if it was foreseeable that someone would be put in danger, it was also foreseeable that someone would go to look for him or try to rescue him or otherwise help him in his distress. The second question is whether the voluntary act of the rescuer, searcher or helper in putting himself in peril negatives the causal connection between the original negligent conduct and his injury. Again, the courts have had equally little difficulty in holding that such a person, acting out of a sense of moral obligation, does not make the free choice which would be necessary to eliminate the causal effect of the defendant's conduct. In the same way, its causal effect is not negatived by an unsuccessful attempt of the person in peril, whose freedom of choice has been limited by the position into which the defendant has put him, to extricate himself from danger: see The Oropesa [1943] 140.

The cases on rescuers are therefore quite simple illustrations of the application of general principles of foreseeability and causation to particular facts. There is no authority which decides that a rescuer is in any special position in relation to liability for psychiatric injury. And it is no criticism of the excellent judgment of Waller J. in Chadwick v. British Railways Board [1967] 1 W.L.R. 912 to say that such a question obviously never entered his head. Questions of such nicety did not arise until the Alcock control mechanisms had been enunciated.

There does not seem to me to be any logical reason why the normal treatment of rescuers on the issues of foreseeability and causation should lead to the conclusion that, for the purpose of liability for psychiatric injury, they should be given special treatment as primary victims when they were not within the range of foreseeable physical injury and their psychiatric injury was caused by witnessing or participating in the aftermath of accidents which caused death or injury to others. It would of course be possible to create such a rule by an ex post facto rationalisation of Chadwick v. British Railways Board [1967] 1 W.L.R. 912. In both McLoughlin v. O'Brian [1983] 1 AC 410 and in Alcock v. Chief Constable of South Yorkshire [1992] 1 AC 310, members of the House referred to Chadwick with approval. But I do not think that too much should be read into these remarks. In neither case was it argued that the plaintiffs were entitled to succeed as rescuers and anything said about the duty to rescuers was therefore necessarily obiter. If one is looking for an ex post facto rationalisation of Chadwick, I think that the most satisfactory is that offered in the Court of Appeal in McLoughlin v. O'Brian [1981] 1 Q.B. 599, 622 by my noble and learned friend Lord Griffiths, who had been the successful counsel for Mr. Chadwick. He said:

"Mr. Chadwick might have been injured by a wrecked carriage collapsing on him as he worked among the injured. A duty of care is owed to a rescuer in such circumstances . . . ."

If Mr. Chadwick was, as Lord Griffiths said, within the range of foreseeable physical injury, then the case is no more than an illustration of the principle applied by the House in Page v. Smith [1996] AC 155, namely that such a person can recover even if the injury he actually suffers is not physical but psychiatric. And in addition (unlike Page v. Smith [1996] AC 155) Waller J. made a finding that psychiatric injury was also foreseeable.

Should then your Lordships take the incremental step of extending liability for psychiatric injury to "rescuers" (a class which would now require definition) who give assistance at or after some disaster without coming within the range of foreseeable physical injury? It may be said that this would encourage people to offer assistance. The category of secondary victims would be confined to "spectators and bystanders" who take no part in dealing with the incident or its aftermath. On the authorities, as it seems to me, your Lordships are free to take such a step.

In my opinion there are two reasons why your Lordships should not do so. The less important reason is the definitional problem to which I have alluded. The concept of a rescuer as someone who puts himself in danger of physical injury is easy to understand. But once this notion is extended to include others who give assistance, the line between them and bystanders becomes difficult to draw with any precision. For example, one of the plaintiffs in Alcock, a Mr. O'Dell, went to look for his nephew. "He searched among the bodies . . . and assisted those who staggered out from the terraces." ([1992] 1 A.C., 354.) He did not contend that his case was different from those of the other relatives and it was also dismissed. Should he have put himself forward as a rescuer?

But the more important reason for not extending the law is that in my opinion the result would be quite unacceptable. I have used this word on a number of occasions and the time has come to explain what I mean. I do not mean that the burden of claims would be too great for the insurance market or the public funds, the two main sources for the payment of damages in tort. The Law Commission may have had this in mind when they said that removal of all the control mechanism would lead to an "unacceptable" increase in claims, since they described it as a "floodgates" argument. These are questions on which it is difficult to offer any concrete evidence and I am simply not in a position to form a view one way or the other. I am therefore willing to accept that, viewed against the total sums paid as damages for personal injury, the increase resulting from an extension of liability to helpers would be modest. But I think that such an extension would be unacceptable to the ordinary person because (though he might not put it this way) it would offend against his notions of distributive justice. He would think it unfair between one class of claimants and another, at best not treating like cases alike and, at worst, favouring the less deserving against the more deserving. He would think it wrong that policemen, even as part of a general class of persons who rendered assistance, should have the right to compensation for psychiatric injury out of public funds while the bereaved relatives are sent away with nothing.

To some extent this opinion would be based upon notions which the law would not accept. Many people feel that the statutory £7,500 (see s.1A of the Fatal Accidents Act 1976) is an inadequate payment to someone like Mr. Hicks, who lost his two daughters in such horrifying circumstances. �K��KAnd on the other side of the comparison, there is the view that policemen must expect to encounter harrowing experiences in the course of their duties and that their conditions of employment provide for ill-health pensions and injury pensions if they suffer injuries, physical or psychiatric, which result in their having to leave the force before normal retirement age. There may be other reasons also, from which I do not exclude ignorance about the nature of mental illness, but, all in all, I have no doubt that most people would regard it as wrong to award compensation for psychiatric injury to the professionals and deny compensation for similar injury to the relatives.

It may be said that the common law should not pay attention these feelings about the relative merits of different classes of claimants. It should stick to principle and not concern itself with distributive justice. An extension of liability to rescuers and helpers would be a modest incremental development in the common law tradition and, as between these plaintiffs and these defendants, produce a just result. My Lords, I disagree. It seems to me that in this area of the law, the search for principle was called off in Alcock v. Chief Constable of South Yorkshire [1992] 1 AC 310. No one can pretend that the existing law, which your Lordships have to accept, is founded upon principle. I agree with Jane Stapleton's remark that: (see The Frontiers of Liability ed. Peter Birks, O.U.P. (1994) Volume 2, p. 87.

"once the law has taken a wrong turning or otherwise fallen into an unsatisfactory internal state in relation to a particular case of action, incrementalism cannot provide the answer."

Consequently your Lordships are now engaged, not in the bold development of principle, but in a practical attempt, under adverse conditions, to preserve the general perception of the law as system of rules which is fair between one citizen and another.

I should say in passing that I do not suggest that someone should be unable to recover for injury caused by negligence, in circumstances in which he would normally be entitled to sue, merely because his occupation required him to run the risk of such injury. Such a rule, called "the fireman's rule" obtains in some of the United States but was rejected by your Lordships' House in Ogwo v. Taylor [1988] 1 A.C. 431. This would be too great an affront to the idealised model of the law of torts as a system of corrective justice between equals. But the question here is rather different. It is not whether a policeman should be disqualified in circumstances in which he would ordinarily have a right of action, but whether there should be liability to rescuers and helpers as a class. And in considering whether liability for psychiatric injury should be extended to such a class, I think it is legitimate to take into account the fact that, in the nature of things, many of its members will be from occupations in which they are trained and required to run such risks and which provide for appropriate benefits if they should suffer such injuries.

Naturally I feel great sympathy for the plaintiffs, as I do for all those whose lives were blighted by that day at Hillsborough. But I think that fairness demands that your Lordships should reject them. I have also read in draft the speech of my noble and learned friend Lord Steyn and agree with his reasons for taking the same course, which seem to me substantially the same as my own. I would therefore allow these appeals and dismiss the actions.