Gray v Thames Trains Ltd & Anor (Rev 1)

This appeal raises questions about the limits of the doctrine still known as ex turpi causa non oritur actio. It arises out of a decision of Flaux J made on 6 July 2007, in which he held that the claimant’s claim was precluded on the ground of public policy based on that doctrine. He refused permission to appeal but permission was subsequently granted by Tuckey LJ.

We can take the facts largely from the judgment. The claimant is now aged 48. He was one of the victims of the Ladbroke Grove rail crash which occurred on 5 October 1999. He suffered relatively minor physical injuries, but the accident had a major psychological impact upon him, in the form of Post Traumatic Stress Disorder (‘PTSD’). On 19 August 2001 he stabbed a stranger to death in Tilbury, Essex. On 22 April 2002 he pleaded guilty to manslaughter on the grounds of diminished responsibility. He was ordered to be detained in a hospital under section 37 of the Mental Health Act 1983.

In this action the claimant claims damages against both defendants, as the operator of the train and as the entity responsible for the rail infrastructure respectively. They both admit that they owed him a duty of care and that they were in breach of that duty. In short, they admit that his injuries, including his PTSD, were caused by their negligence. They further admit that they are in principle liable in respect of his losses, including loss of earnings, incurred before 19 August 2001. However, they deny liability in respect of losses incurred after that date on the basis that ex turpi causa non oritur actio. The claimant’s claim for those losses was struck out by Master Leslie but was reinstated by Holland J on the ground that the issues raised were better determined at a trial.

After Holland J’s order, the claimant's counsel prepared a re-amended schedule of special damages clarifying the nature of the various heads of claim being pursued. In addition, the claimant’s solicitors served a witness statement together with other statements from his family and friends testifying to the effect of the Ladbroke Grove crash upon him and a number of psychiatric reports. The parties were each given permission to call one expert psychiatrist. The claimant’s solicitors served evidence from Dr Philip Joseph, who is a consultant psychiatrist, and indicated their intention to call him. Although the defendants had the claimant examined by Dr Rosen, they gave no notice of any intention to call him and claimed privilege for his report, as they were entitled to do.

At [4] in his judgment, the judge noted that the trial had been set down for five days but that on the first day of the hearing counsel agreed that he should determine the point of principle as to whether, as the defendants contended, the claim was precluded on grounds of public policy, on the basis that the facts set out in the skeleton argument on behalf of the claimant were essentially accepted by the defendants, without the need for any live evidence. It was we think contemplated that the judge would, so far as necessary, look at the evidence of Dr Joseph and, indeed, the other written evidence.

We consider first the issues of ex turpi causa which were debated before the judge. There was no (or very little) argument before the judge on issues or potential issues relating to reasonable foreseeability, causation or contributory negligence (or fault). However after the argument in this appeal the House of Lords decided the important case of Corr (administratrix of the estate of Thomas Corr (deceased)) v IBC Vehicles Limited [2008] UKHL 13 (‘Corr’). Although Corr is a suicide case, it appeared to us to raise questions which might be of significance in this appeal. We therefore asked the parties if they wished to make submissions arising out of it. They have done so and we return to it below after considering the issues before the judge and the oral argument addressed to us.

The judge set out the facts at [5] to [17]. For present purposes they can be summarised in this way:

The claimant does not claim compensation for the consequences of being detained in a mental hospital under sections 37 and 41 of the Mental Health Act 1983. He recognises that to do so would be to fall foul of the ex turpi causa principle because it would involve his relying upon the fact that he was so detained pursuant to the order of the Crown Court consequent upon his plea of guilty to manslaughter. However, he submits that he is entitled to the earnings he lost as a result of the PTSD. He submits that on the evidence it is plain that, by reason of the PTSD caused by the defendants’ negligence, he was not able to earn as much as he would have done but for the accident. The judge rejected his claim for loss of earnings after the manslaughter. He held that a claimant falls foul of the ex turpi causa principle if his claim is, as the judge put it at [18], closely connected with or inextricably bound up with his own criminal or illegal conduct. He further held that this is such a case.

Mr Anthony Scrivener QC submits on behalf of the claimant that the judge applied the wrong test or, alternatively, that the claimant’s claim here is not connected with or inextricably bound up with the manslaughter in the sense used in the cases. Mr Christopher Purchas QC submits that the judge was right.

The principle derives originally from the judgment of Lord Mansfield in Holman v Johnson (1775) 1 Cowp 341 at 343:

“The objection, that a contract is immoral or illegal as between plaintiff and defendant, sounds at all times very ill in the mouth of the defendant. It is not for his sake, however, that the objection is ever allowed; but it is founded in general principles of policy, which the defendant has the advantage of, contrary to the real justice, as between him and the plaintiff, by accident, if I may so say. The principle of public policy is this; ex dolo malo non oritur actio. No court will lend its aid to a man who founds his cause of action upon an immoral or an illegal act. If, from the plaintiff's own stating or otherwise, the cause of action appears to arise ex turpi causa, or the transgression of a positive law of this country, there the court says he has no right to be assisted. It is upon that ground the court goes; not for the sake of the defendant, but because they will not lend their aid to such a plaintiff. So if the plaintiff and defendant were to change sides, and the defendant was to bring his action against the plaintiff, the latter would then have the advantage of it; for where both are equally in fault, potior est conditio defendentis.

The question therefore is, “Whether, in this case, the plaintiff's demand is founded upon the ground of any immoral act or contract, or upon the ground of his being guilty of any thing which is prohibited by a positive law of this country.”

At [19] the judge set out that passage as underlined above. He noted at [20] that Mr Scrivener relied upon the underlined part in support of his submission that the principle only applies where the claimant is relying upon his own unlawful or immoral act. It is not in dispute that that is the principle in a case in which the claimant’s case is brought in reliance upon a contract or in reliance upon collateral rights acquired under a contract. That is clear from the decision of the House of Lords in Tinsley v Milligan [1994] AC 340, in which the House rejected the doctrine that (as the judge put it at [21]), relief should only be denied a claimant where it would be “an affront to public conscience”. Such a doctrine would involve a balancing exercise, which the House did not think appropriate. In the context of a collateral rights case such as Bowmakers v Barnet Instruments [1945] KB 65, Lord Browne-Wilkinson stated the test as follows at page 377B-C of Tinsley v Milligan:

“In a case where the plaintiff is not seeking to enforce an unlawful contract but founds his case on collateral rights acquired under the contract (such as a right of property) the court is neither bound nor entitled to reject the claim unless the illegality of necessity forms part of the plaintiff's case.”

Mr Scrivener submits that that test in principle applies here but that it does not apply on the facts because the manslaughter does not form part of the claimant’s case and, in any event, because the claimant’s cause of action was complete when he suffered injury at the time of the accident.

We take the last of those submissions first. The judge rejected it and so would we. Tuckey LJ and I comprised the majority in this court (Ward LJ dissenting) in Hewison v Meridian Shipping Pte [2002] EWCA Civ 1821, where it was common ground that the claimant could recover the various heads of loss claimed but the issue was whether he could recover loss of earnings as a crane driver. We held that he could not, on the basis of the same principle of public policy, albeit expressed as ex turpi causa non oritur damnum, on the ground that, in order to recover that loss of earnings, the claimant would have had to continue fraudulently to represent to his employers that he had not had any epileptic fits and fraudulently to conceal from them that he was taking anti-convulsants for epilepsy: see [28]. I added in that paragraph that a classic example of the application of the principle would prevent a claimant who makes his living from burglary from having his damages assessed on the basis of what he would have earned from burglary but for the defendant’s negligence. So the question here is whether the claimant should be defeated by the principle ex turpi causa non oritur damnum, where the damnum is the earnings which the claimant would have made if there had been no accident.

We note in passing that the application of the principle applied in Hewison would not defeat the claimant’s claim here because the claimant is not relying upon the consequences of the manslaughter to establish his loss of earnings. So the first question for decision is whether the principle in Tinsley v Milligan applies to a case of this kind.

There are serious difficulties in the claimant’s way. First, we note, as the judge did at [22], that the editors of the 19^th edition of Clerk & Lindsell on Torts point out at paragraph 3-12 that this approach is less readily applicable in the context of tort cases than it is in the case of contractual or property disputes. The judge quoted this passage from paragraph 3-13:

“A strict application of Tinsley v Milligan across all forms of tort action would limit the applicability of the defence to those (probably comparatively rare) situations in which the claimant has to rely (however that term is interpreted) on the illegality as part of his claim.”

That is true but would it matter? It is by no means clear to us that it would or should matter in a case in which the illegal act was caused by the tort.

However that may be, the judge rejected Mr Scrivener’s submission on the basis that it was inconsistent with the decision of this court in Cross v Kirkby, unreported, 18 February 2000, where the facts were these. The claimant, Mr Cross, was vehemently opposed to hunting. The defendant, Mr Kirby, was a farmer who allowed the hunt to go across his land, as it did on the day in question. During the morning there had been some altercations between the claimant and his partner, Mrs Davis, on the one hand and the defendant on the other. On more than one occasion the defendant had escorted the claimant and his partner off the land. There came a time when Mrs Davis bit the defendant while she was being walked off the land. Shortly afterwards the claimant attacked the defendant. Mrs Davis got an iron bar out of the car and the claimant armed himself with a baseball bat. As Beldam LJ put it, the claimant was angry, almost hysterical. He banged the bat on the ground, splitting it. He shouted : “You’re fucking dead Kirby”. He jabbed him in the chest and in the throat with the bat. The defendant started to walk away but the claimant persisted. To ward off blows with the bat, the defendant turned and grappled with the claimant. Wrestling the bat from him, he hit the claimant with a single blow on the side of the head. Unfortunately the blow caused a linear fracture of the claimant’s head and subdural bleeding and the claimant subsequently suffered epileptic attacks.

The claimant claimed damages for assault. The defence was self defence on the basis that the claimant had armed himself with the bat, had repeatedly assaulted the defendant and had threatened to kill him. He said ex turpi causa non oritur actio. After a long trial, the judge rejected both the defence of self defence and the plea of ex turpi causa. He held that the claimant was 60 per cent responsible and awarded him 40 per cent of the full measure of damages. It is plain from the transcript that from the outset the court, which comprised Beldam, Otton and Judge LJJ, took the view that the judge should have upheld the defence of self defence. The court suggested that the defendant might amend his grounds of appeal to take the point and permitted him to do so. After a very detailed analysis of the evidence Beldam LJ concluded that the judge should have held that the defendant was acting in self defence. Otton LJ agreed with Beldam LJ and Judge LJ did so too, although (unlike Otton LJ) he analysed the self defence issues in some detail.

Beldam LJ said that, if he was wrong on self defence, he would nevertheless find that that the claimant’s case failed on the ground that his injuries arose from his own criminal and unlawful acts. Beldam LJ considered the arguments in some detail. The only point which is of significance in this appeal is that he rejected the submission that the principle in Tinsley v Milligan applied to cases of assault of the kind before the court. He stated the general principle, as applied to such a case, as follows (in a passage quoted by the judge at [27]):

“I do not believe that there is any general principle that the claimant must either plead, give evidence of or rely on his own illegality for the principle to apply. Such a technical approach is entirely absent from Lord Mansfield's exposition of the principle. I would however accept that for the principle to operate the claim made by the claimant must arise out of criminal or illegal conduct on his part. In this context “arise out of” clearly denotes a causal connection with the conduct a view which is implicit in such different cases as Lane v Holloway ([1968] 1 QB 379) and the recent case to which we were referred in this court, Standard Chartered Bank v Pakistan National Shipping Corporation (No 2) ([2000] 1 Lloyd’s Rep 218). In my view the principle applies when the claimant's claim is so closely connected or inextricably bound up with his own criminal or illegal conduct that the court could not permit him to recover without appearing to condone that conduct.”

Judge LJ also analysed this part of the case in some detail in an illuminating judgment. He quoted part of the judgment of Bingham LJ in Saunders v Edwards [1987] 1 WLR 1116. His quotation included this statement:

“Where the claimant’s action in truth arises ex turpi causa he is likely to fail. Where the claimant has suffered a genuine wrong to which the allegedly unlawful conduct is incidental, he is likely to succeed.”

Judge LJ (in our view correctly) regarded that statement as a helpful touchstone and expressed the principle in his own words thus:

“In my judgment, where the claimant is behaving unlawfully, or criminally, on the occasion when his cause of action in tort arises, his claim is not liable to be defeated ex turpi causa unless it is also established that the facts which give rise to it are inextricably linked with his criminal conduct. I have deliberately expressed myself in language which goes well beyond questions of causation in the general sense.”

The court held on the facts that, even if the defence of self defence failed, perhaps on the basis that the defendant went further then he should have done in reasonable self defence, the delivery of his blow was inextricably linked with the assaults to which he had been subjected.

We agree with the view expressed by the judge at [29] that Sir Murray Stuart-Smith was not seeking to lay down a different test in Vellino v Chief Constable of Greater Manchester [2001] EWCA Civ 1249, where (having cited inter alia Tinsley v Milligan and Cross v Kirkby) he said (at [72]):

“From these authorities I derive the following propositions:

1.

The operation of the principle arises where the claimant's claim is founded upon his own criminal or immoral act. The facts which give rise to the claim must be inextricably linked with the criminal activity. It is not sufficient if the criminal activity merely gives occasion for tortious conduct of the Defendant.

…”

In these circumstances we agree with the judge that the test in Tinsley v Milligan is too narrow to apply to a case in tort like the instant case. As applied to a case like this, where it is not suggested that the cause of action arises out of an illegal act, the question seems to us to be whether the relevant loss is inextricably linked with the claimant’s illegal act or, as Beldam LJ put it, so closely connected or inextricably bound up with his criminal or illegal conduct that the court could not permit him to recover without appearing to condone that conduct.

On the facts, the judge answered that question yes. Mr Scrivener submits that he was wrong to do so. The judge said at [31] that, in his view, to argue that the loss of earnings after the manslaughter was somehow independent of it was unreal and that on analysis the claim after 19 April 2001 was dependent on the manslaughter. At [32] the judge noted the argument advanced by counsel for the defendants (not then Mr Purchas) that, if the claimant had committed a criminal offence for which he had been imprisoned or detained in hospital but which was not attributable to the PTSD, the defendant would have had a complete defence in respect of losses after his incarceration, not because of any principle of ex turpi causa, but because the commission of the offence and the incarceration for it would be a break in the chain of causation. That is plainly correct.

The judge added in [32] that he agreed with counsel:

“that in order to avoid that consequence in the present case, the claimant has to rely upon the manslaughter and to contend that it was caused by the PTSD which in turn was caused by the negligence of the defendants.”

We do not agree that that is so. The claimant’s case is simply that he has suffered a loss because, but for the tort, he would have earned money both before and after 19 April 2001 and that he is therefore entitled to recover the whole of his loss of earnings from the defendants. The manslaughter is not inextricably bound up with that claim. Although, as the judge demonstrated in [33], the claimant’s re-amended schedule of special damages set out the facts relating to the manslaughter and expressly alleged that the claimant would not have killed Mr Boultwood but for mental illness caused by the accident, that was not, as we see it, a necessary part of his case.

Although the legal burden of establishing causation in respect of each head of loss remains on the claimant, the evidential burden of showing that the manslaughter and the claimant’s incarceration amounted to a break in the chain of causation is on the defendants. In such circumstances it is a matter for the trial judge to decide whether it broke the chain of causation or not. If it did, the claim would fail for that reason.

However, the court might not so hold because it is of course the claimant’s case, supported by the evidence of Dr Joseph (and not contradicted by any other psychiatric expert) that he would not have committed the manslaughter but for the defendants’ negligence. The claimant says that in these circumstances that there was no break in the chain of causation, the true cause (alternatively a cause) being the PTSD caused by the tort. Moreover, he says that that is so notwithstanding that he intended to kill or cause serious harm to Mr Boultwood. If the court were so to hold, it would do so on the basis that the (alternatively a) proximate or effective cause of the loss of earnings was the tort. If the manslaughter did not break the chain of causation between the tort and the loss of earnings (in the sense described by the House of Lords in Corr discussed below), it does not seem to us that it can fairly be said that the loss of earnings after 19 April 2001 was inextricably linked with the claimant’s illegal act or, as Beldam LJ put it, so closely connected or inextricably bound up with his criminal or illegal conduct that the court could not permit him to recover without appearing to condone that conduct. We return below to the question whether, in the light of Corr, there was a break in the chain of causation on the facts.

The other possibility is that the court might hold that this is a case to which the Law Reform (Contributory Negligence) Act 1945 (‘the 1945 Act’) applied. That would be on the basis that the manslaughter was ‘fault’ within the meaning of section 1(1) of the 1945 Act: see further below in the context of my discussion of Corr. It would be on the basis that the loss of earnings was partly caused by the tort and partly caused by the deliberate act of the claimant in stabbing Mr Boultwood and that both were blameworthy.

If it were so held, again we do not think that it could fairly be said that to allow the claim to succeed in part would be to appear to condone the manslaughter. On the contrary, the apportionment would have the effect of separating the responsibility of the claimant from that of the defendant and of ensuring that the claimant only recovered for the loss of earnings for which the defendants were truly responsible and he was not. In that event, the claimant’s recovery would not be inextricably bound up with the manslaughter but distinct from it. We return below to the question whether there was contributory fault on the facts.

However, if there was contributory fault on the facts, we do not see why it should be regarded as contrary to principle or policy to permit the claimant to recover damages in respect of that part of the damage for which, on this hypothesis, the defendants are responsible. We recognise of course that the rule is one of public policy. As, for example, Lord Goff put in Tinsley v Milligan, (albeit in the course of a dissenting speech) at page 355:

“the principle is not a principle of justice; it is a principle of policy, whose application is indiscriminate and so can lead to unfair consequences as between the parties to litigation.”

We accept that that may be so but it appears to us that, where in a particular case the problem is solved by the application of the 1945 Act, there is no need for public policy to require a different solution. As just stated, at any rate on the facts of this case, where the claim is for loss of earnings, the application of the Act avoids the conclusion that the claimant’s recovery is inextricably bound up with his crime such that, if the court permits partial recovery, it will in some way be condoning the crime.

Unless there is binding authority to the contrary, we would hold that, where the manslaughter does not break the chain of causation or where there is contributory fault on the part of the claimant of less than 100 per cent, so that the claim is not, as Beldam and Judge LJJ put it in Cross v Kirkby, inextricably bound up with or linked with the criminal conduct, public policy does not prohibit recovery. In this regard we recognise that there is (or may be) a distinction between a case like this, where the claim is for loss of earnings and (for the reasons we have given) not inextricably bound up with or linked with the manslaughter, and a case where the claimant was seeking to recover damages as a result of his incarceration for years in a mental hospital or a prison as a result of a sentence imposed by a criminal court. We accept, at any rate as the authorities stand, that such a claim would be inextricably bound up with or linked with the manslaughter and that such damages could not be recovered.

Mr Purchas relies upon two decisions of this court in support of his submission that this claim too is contrary to public policy. They are Clunis v Camden and Islington Health Authority, [1998] QB 978 and Worrall v British Railways Board, unreported, 29 April 1999. Both cases were decided before Cross v Kirkby, in which only Judge LJ referred to Clunis and no-one referred to Worrall. However, it is most unlikely that Beldam LJ did not have both cases in mind because he presided and gave the leading judgment in both. In these circumstances we would be loath to hold that the reasoning in Cross v Kirkby, which we have considered in some detail, was inconsistent with them.

In Clunis the plaintiff, who had a history of mental disorder, was detained in a hospital under section 3 of the Mental Health Act 1983. He was released into the community but some time later, in a sudden and unprovoked attack, killed a man by stabbing him. He pleaded guilty to manslaughter on the grounds of diminished responsibility and was ordered to be detained in a secure hospital. He sued the local health authority for damages for breach of a duty said to be owed to him on the basis that the responsible medical officer should have appreciated that he was dangerous and given him appropriate treatment. It was said that, if she had discharged that duty, the plaintiff would not have committed manslaughter and would not have been subject to the prolonged treatment that he faced as a result of the hospital order made by the Crown Court.

In this court Beldam LJ gave the judgment of the court, which also comprised Potter LJ and Bracewell J. It held that the judge should have struck out the claim on two independent grounds, first that of public policy because ex turpi causa non oritur actio and, secondly that the authority owed no actionable duty to the plaintiff. In this appeal, we are of course only concerned with the first ground. By contrast with this case, there was no suggestion in Clunis that any distinction should be drawn between different heads of damage. The court held, as was plainly the case, that the plaintiff’s claim arose out of and depended upon proof of his commission of a criminal act: see page 987C. See also the analysis at pages 989G to 990D and the court’s conclusion at page 990E, where it said:

“In the present case we consider that the defendant has made out its plea that the plaintiff’s claim is essentially based on his illegal act of manslaughter, he must be taken to have known what he was doing and that it was wrong, notwithstanding that the degree of his culpability was reduced by reason of mental disorder. The court ought not to allow itself to be made an instrument to enforce obligations alleged to arise out of the plaintiff’s own criminal act …”

In Worrall Beldam LJ said that the question in Clunis was whether the plaintiff bore any responsibility for the commission of the criminal offence of manslaughter on which his claim was based. He added that the court held that the plaintiff retained a degree of moral and legal responsibility for the commission of the offence and “accordingly” public policy precluded him from recovering damages in respect of his own criminal act.

As we read it, Worrall is an example of the application of the same principle. It was a somewhat startling case on the facts. The plaintiff’s case was that, as a result of an electric shock caused by his employers’ negligence, he suffered a change of personality which caused him to commit serious sexual offences against two prostitutes. In spite of his defence of alibi, he was convicted of both offences and sentenced to six years’ imprisonment. At the trial no mention was made of the electric shock or its alleged consequences, even in connection with sentence. He was subsequently dismissed from his employment and remained unemployed after his release from prison. His claim included a claim for loss of earnings and loss of a pension. The judge struck out the whole of that part of the claimant’s claim on the basis that the whole claim was consequent upon the criminal offences. This court, comprising Beldam, Roch and Mummery LJJ, dismissed the plaintiff’s appeal. Beldam LJ expressed his conclusion thus:

In my view that part of the statement of claim on which the plaintiff relied to establish the vast majority of his loss was founded upon his commission of serious criminal offences for which he was fully responsible in law. The principle that the court will not lend its aid to the recovery of damages based upon criminal offences applies and the case is indistinguishable save in immaterial respects from Clunis. The plaintiff’s responsibility was undiminished in any respect and I consider it would be contrary to public policy to allow him to recover damages consequent upon the commission of those offences.

Roch LJ agreed with Beldam LJ. So too did Mummery LJ on the public policy point, although he added that, in his opinion the plaintiff’s claims were suffered as a direct result of his crimes and not as a direct or foreseeable result of a breach of a duty of care owed to him by the defendants. In Worrall the court rejected the submission that Clunis was distinguishable. It follows that they are both authority for the proposition that, where a claim or a head of claim depends upon the claimant’s criminal act it cannot succeed on the public policy ground that ex turpi causa non oritur actio or ex turpi causa non oritur damnum, as the case may be.

It follows, as we see it, that, on the authorities as they stand, the claimant here cannot recover the consequences of his compulsory detention in a hospital pursuant to an order of the Crown Court. However, as already explained, that is not the nature of his claim. In this connection, it is important to note that in Worrall the plaintiff sought to advance a new point in this court which was not advanced before the judge. It was almost the same point as is advanced here, namely that the claim for loss of earnings, pension rights and the like were caused by the accident (and thus the defendants’ negligence) and not by the commission of the criminal offences or the period of imprisonment. This court did not permit that point to be taken because it had not been taken below. It follows, as we see it, that Worrall does not assist in resolving the issue before us.

There is nothing in Clunis or Worrall which is inconsistent with the decision or reasoning in Cross v Kirkby. It follows that there is nothing in them which is inconsistent with the proposition that a claim is not defeated by public policy unless the claim or a head of claim is inextricably bound up with or linked with the criminal conduct. We have already given our reasons for concluding that, in so far as the claimant claims loss of earnings, this is not such a case. It follows that, on the basis of the arguments that were addressed to us orally, we would allow the appeal.

It is we think helpful to have regard to the recent decision of the House of Lords in Corr. The facts were that Mr Corr sustained a serious injury at work. His physical injuries were much worse than those of Mr Gray. He became depressed, a condition which worsened as time went on, and ultimately, on 23 May 2002 he committed suicide while suffering from an episode of severe depression. Lord Bingham concisely described his mental condition when he committed suicide at [5]:

“On the one hand, he had the capacity to manage his own affairs. His intellectual abilities were not affected. His appreciation of danger was not lessened. He was aware of the likely consequences of jumping from a high building. He acted deliberately with the intention of killing himself. He had from time to time since the accident thought of taking his own life but had hesitated because of the effect on his family. He understood the difference between right and wrong. He knew the nature and quality of his acts. He did not suffer from hallucinations. It would seem clear, had the question arisen, that his mental condition would not have met the M’Naghten test of insanity. On the other hand, at the time of his death Mr Corr was severely depressed. His depression had caused him to experience feelings of hopelessness. These became increasingly difficult to resist. A critical change took place in the balance of his thinking, when he stopped recognising these feelings of hopelessness as symptoms of his depressive illness, and instead they came to determine his reality. At the time of his suicide Mr Corr was suffering from a disabling mental condition, namely a severe depressive episode which impaired his capacity to make reasoned and informed judgments about his future. It was well known that between one in six and one in ten sufferers from severe depression kill themselves.

The House unanimously held that Mr Corr’s suicide was within the scope of the duty which his employer owed him, that depression was a reasonably foreseeable consequence of its breach of duty and that it was not incumbent upon the claimant to show that suicide itself was foreseeable: see per Lord Bingham at [9-13], Lord Scott at [28-29] and Lord Walker at [38-42]. In reaching that conclusion, Lord Bingham focused (at [13]) on Hughes v Lord Advocate [1963] AC 837, especially per Lord Pearce at page 857, whereas Lord Scott and Lord Walker focused on Page v Smith [1996] AC 155: see [29] and [38-41] respectively. Lord Bingham also placed some reliance on Page v Smith at [7]. Lord Mance and Lord Neuberger agreed but both said that their opinion did not depend on the assumption that Page v Smith was correct: see [45-46] and [53-56] respectively. We should add that Lord Bingham said at [13] that, if it was necessary to show that Mr Corr’s suicide was reasonably foreseeable, the employer would have had difficulty in escaping an adverse finding to that effect.

The House further held unanimously that, if there was a breach of duty owed by the defendant to Mr Corr, the chain of causation between the breach and Mr Corr’s death was not broken by his suicide: see per Lord Bingham at [14-17], Lord Scott at [25-39], Lord Walker at [42-43], Lord Mance at [45-46] and Lord Neuberger at [53]. It was further unanimously held that deliberate suicide was ‘fault’ within the meaning of the 1945 Act and thus capable of being ‘contributory negligence’: see especially per Lord Scott at [31-32], Lord Walker at [44] and Lord Neuberger at [59-60]. Lord Bingham and Lord Walker concluded that there should be no reduction on the facts: see [22-23] and [44] respectively. Lord Mance and Lord Neuberger held that although, given that the issue had not been debated in any detail in the courts below, no reduction should be made on the facts, it might well be appropriate to make a reduction in a case of suicide of the kind the House was considering: see [47-52] and [57-71] respectively. Lord Scott would have reduced the damages by 20 per cent: see [32-33].

It might be argued on the basis of the reasoning in Corr that there is no distinction in principle between that case and this case, so far as breach of duty, causation or remoteness is concerned. In both cases it was reasonably foreseeable that if the defendants did not take reasonable care, the claimant might be injured and that if he was, that injury might cause or be PTSD. The reasoning of the House of Lords might then lead to the conclusion that it was not necessary that it should be reasonable foreseeable that the claimant might attack someone else, just as it held that it was not necessary that Mr Corr’s suicide was reasonably foreseeable. It may also be arguable that it was reasonably foreseeable that the claimant’s PTSD might cause him to injure others or himself. It is not easy to see why, public policy apart, there should be any difference in these respects between a man who kills himself because of depression caused by a tort and a man who kills or seriously injures someone else because of such depression.

Again the issues of causation are similar in the two cases. Thus, having described Mr Corr’s mental state at [5], which we have quoted above, Lord Bingham said this at [16]:

“In the present case Mr Corr’s suicide was not a voluntary, informed decision taken by him as an adult of sound mind making and giving effect to a personal decision about his future. It was the response of a man suffering from a severely depressive illness which impaired his capacity to make reasoned and informed judgments about his future, such illness being, as is accepted, a consequence of the employer’s tort. It is in no way unfair to hold the employer responsible for this dire consequence of its breach of duty, although it could well be thought unfair to the victim to do so.”

Lord Bingham then considered a submission that Mr Corr was not in M’Naghten terms insane to which we return below.

The essence of Lord Bingham’s opinion on causation was his approval at [16] of these paragraphs in the judgment of Sedley LJ (with whom Ward and Wilson LJJ agreed) in this court, which Lord Bingham quoted at [14]:

“82.

To cut the chain of causation here and treat Mr Corr as responsible for his own death would be to make an unjustified exception to contemporary principles of causation. It would take the law back half a century to a time when the legal and moral opprobrium attaching to suicide placed damages for being driven to it on a par with rewarding a person for his own crime. Today we are able to accept that people to whom this happens do not forfeit the regard of society or the ordinary protections of the law.

83.

Once it is accepted that suicide by itself does not place a clinically depressed individual beyond the pale of the law of negligence, the relationship of his eventual suicide to his depression becomes a pure question of fact. It is not a question which falls to be determined, as the deputy judge in significant measure determined it, by analogy with the duty of care resting on a custodian. Once liability has been established for the depression, the question in each case is whether it has been shown that it was the depression which drove the deceased to take his own life. On the evidence in the present case, it clearly was.”

At [15] and [16] Lord Bingham concluded that the Corr case was an example of such a case, in contrast with a case in which, when the victim of a car accident took her life “she made a conscious decision, there being no evidence of disabling mental illness to lead to the conclusion that she had an incapacity in her faculty of volition”: Wright v Davidson (1992) 88 DLR (4^th) 698, 705. The Corr case was different because it was shown on the evidence, as Sedley LJ put it, that it was the depression caused by the defendant’s tort that drove the deceased to take his own life. The decision to commit suicide was not therefore an independent cause of death but one which was itself caused by the tort, so that since, as Lord Bingham put it at [15], the rationale of the principle that a novus actus interveniens breaks the chain of causation is fairness, there was no break in the chain of causation on the facts. It would only be fair to hold that there was a break in the chain of causation if the suicide was independent of the tort.

Lord Scott said much the same. He noted at [25] that, in committing suicide, Mr Corr acted deliberately, aware of the consequences and with the intention of killing himself. At [27] Lord Scott quoted two passages from the 19^th edition of Clerk & Lindsell on Torts including the statement that

“whatever its form, the novus actus must constitute an event of such impact that it ‘obliterates’ the wrongdoing of the defendant”.

Lord Scott ultimately concluded that, in spite of the fact that the suicide was a deliberate act, it did not obliterate the wrongdoing of the defendant. Lord Walker said much the same at [42-43], where he recognised that Mr Corr had the relevant capacity to decide what to do, that he was not deprived of his personal autonomy and that he was not M’Naghten insane but that there was no break in the chain of causation.

It seems to us that, subject to considerations of public policy, there is much to be said for the conclusion that the same is true here. It was the depression suffered by Mr Gray caused by the defendant’s negligence that led him to kill, not himself, but Mr Boultwood. In these circumstances, it is at least strongly arguable that there was no break in the chain of causation.

We recognise that that view is or may be contrary to that of Mummery LJ in Worrall, where he said that the plaintiff’s loss complained of

“relates to loss suffered by him as a direct result of crimes committed by him against others, not as a direct or foreseeable result of a breach of a duty of care owed to him by the defendant.”

It is far from clear that this point was argued before the court in Worrall. In any event the issue of causation was not considered in the detail in which it was considered in Corr. Indeed, as pointed out in [34] above, the point that arises here did not arise in Worrall. As we read what Mummery LJ said, he was focusing on the issue of public policy. However that may be, as to causation, it is not easy to see why the law should hold that there is a break in the chain of causation between a tort and the commission of a crime, where the evidence shows that, as Lord Bingham put it in the context of suicide (in the passage from [16] quoted above), the crime is

“the response of a man suffering from a severely depressive illness which impaired his capacity to make reasoned and informed judgments about his future, such illness being, …, a consequence of the employer’s tort.”

It seems to us that in such a case, if it is appropriate to deprive the claimant of recovery, the court should hold, not that there was a break in the chain of causation, but that public policy requires that the claim should fail.

As to contributory fault, there seems to us to be much to be said for the conclusion that there was contributory fault on the part of the claimant here. Such a conclusion would we think be entirely consistent with the views of Lord Scott, Lord Mance and Lord Neuberger in Corr.

As to public policy, in Corr the House of Lords was not considering the question whether public policy should lead to a different view in a case where a death is caused by suicide, which is of course no longer a crime, from a case in which a death (or other injury) is caused by a crime. At [16] Lord Bingham said this, after the passage quoted above:

“Mr Cousins submitted that on the agreed findings Mr Corr was not, in M'Naghten terms, insane, and it is true that in some of the older authorities a finding of insanity was regarded as necessary if a claimant were to recover for loss attributable to suicide: see, for example, Murdoch v British Israel World Federation (New Zealand) Inc [1942] NZLR 600, following McFarland v Stewart (1900) 19 NZLR 22. I do not for my part find these cases persuasive, for two main reasons. First, so long as suicide remained a crime the courts were naturally reluctant to award damages for the consequences of criminal conduct. Thus a finding of insanity, which exculpated the deceased from criminal responsibility, removed this obstacle. Modern changes in the law overcome the problem: there is now no question of rewarding the consequences of criminal conduct, although it remains true that the more unsound the mind of the victim the less likely it is that his suicide will be seen as a novus actus. The second reason is that whatever the merits or demerits of the M'Naghten rules in the field of crime, and they are much debated, there is perceived in that field to be a need for a clear dividing line between conduct for which a defendant may be held criminally responsible and conduct for which he may not. In the civil field of tort there is no need for so blunt an instrument. "Insane" is not a term of medical art even though, in criminal cases, psychiatrists are obliged to use it. In cases such as this, evidence may be called, as it was, to enable the court to decide on whether the deceased was responsible and, if so, to what extent. I agree with Sedley LJ that it would be retrograde to bar recovery by the claimant because the deceased was not, in M'Naghten terms, insane.”

There seems to us to be something to be said for the view that the traditional harsh view of public policy expressed in, for example, Clunis and Worrall should be revisited in a case in which the crime relied upon (whether relied upon by the claimant or the tortfeasor) was itself caused by the tort. In times gone by, it would perhaps have been seen as inconceivable that the murder or manslaughter of another could have been caused by a tort. However, the facts and evidence in Corr and this case, and perhaps a more developed understanding of clinical depression, show that it is no longer inconceivable. It is far from clear to us why the ends of justice are not sufficiently served by the principles of foreseeability, causation and contributory negligence without the need for a further principle of public policy in such a case. It was no doubt for this reason that Auld LJ said obiter in KR v Bryn Alyn Community [2003] EWCA Civ 85:

“Notwithstanding anything said by this court in Clunis’ case, an argument may survive that damages are recoverable in respect of tortious acts that have resulted in a law abiding citizen becoming criminal.”

We agree. However, the argument is not open in this court because we are bound by the authorities to apply the principles in them and, in particular, bound to apply the principle stated in Cross v Kirkby. As indicated above, the application of that principle leads to the conclusion that the claimant could not recover in respect of his incarceration in a mental hospital pursuant to the order of the Crown Court but, in our opinion, is not barred by public policy from recovering his loss of earnings, if they are otherwise recoverable.

We would add this thought. Suppose a man suffering from clinical depression caused by a tort jumps off a tall building and dies and, just before he does so, he deliberately pushes someone else off, who also dies. Suppose then that both the dependants of the suicide and the dependants of the man who has been pushed off, and thus killed by the suicide, take proceedings against the tortfeasor, it is not clear why, either as a matter of foreseeability or causation on the one hand or public policy on the other, the former should be entitled to recover but not the latter.

As indicated above, at our invitation, the parties have made written submissions on the potential questions arising from the decision and reasoning in Corr. The defendants’ submissions may be summarised in this way.

By contrast, the submissions made on behalf of the claimant may be summarised in this way.

The issues of foreseeability, causation and contributory fault were not considered in any detail or at all by the judge because they were not raised before him and the evidence did not explicitly address them, especially the issue of contributory fault, which was not pleaded. Equally they were not raised in oral argument before us. They have been brought to the fore by the reasoning in Corr. We entirely accept that, for the reasons given by both parties and, indeed, in Corr, we should not determine the issue of contributory fault. If it is to be considered, it must be remitted to the High Court for detailed consideration in the light of the medical and other evidence. On the material available to us at present we would remit it to the High Court. It is true that it was not pleaded but we have no doubt that it would have been if the decision in Corr had been made before the trial. We do not think that the claimant is likely to suffer any prejudice which cannot be compensated in costs. However, we will consider such further submissions as the parties wish to make on this point.

Although we have made some preliminary observations on the issues of foreseeability, causation and contributory fault above, we do not think that we should finally determine any of them now. There is some overlap between contributory fault and the other issues and we have reached the conclusion that, if contributory fault is to be remitted, rather than our deciding the other issues now, they too should be remitted for consideration by a judge, who can of course be Flaux J if he is available.

For these reasons we allow the appeal on the ex turpi causa point. We will consider further submissions on the future conduct of the action.