United Kingdom House of Lords Decisions

Such occurrences and consequent mental injuries or illnesses would already have been a reality in 1929. I accept that the medical explanation for mental injuries, and their physical connection, is today somewhat better informed. No competent psychiatrist would, however, assert that this knowledge is complete. The psychiatric nostrums of today may become the scientific heresies of tomorrow. The textually relevant circumstance is that in 1929 it would already have been appreciated that the imposition of strict liability for mental injury and illness would have opened the door to an avalanche of intangible claims, greatly in excess of the number of claims for physical injuries. For the fledgling aviation industry this would have involved a large exposure to (i) judgments and awards, (ii) the cost of expert evidence to sort out what were cognisable claims, and (iii) the cost of litigation, the latter being irrecoverable in the United States. This might have meant larger liability insurance premiums and a resultant increase in passenger fares. In these circumstances Professor Malcolm Clarke, a specialist in this area of the law, commented that the expansive interpretation of the First Division in King v Bristow Helicopters Ltd as including purely psychological injury "caused shock waves": "Air rage: Businessmen behaving badly: Civil liability for uncivil passengers" (2001) LMCLQ 369. (3) In this context it is reasonable to expect that if it had been intended to cover mental injury or illness, it would have been provided for expressly. In the absence of such an express reference it is reasonable to interpret "bodily injury" and "lésion corporelle" as words of restriction, ie as referring to non fatal injury which is physical rather than mental: contrast the wide term "personal injury" in the Guatemala Protocol which never came into force: see Protocol to amend the Convention for the unification of certain rules relating to International Carriage by Air, signed at Warsaw on 12 October 1929, as amended by the Protocol done at The Hague on 28 September 1955, signed at Guatemala City on 8 March 1971. His interpretation involves reading the phrase "bodily injury" or "lésion corporelle" ejusdem generis with death ("mort") and wounding ("blessure"). It has the merit, unlike the contrary interpretation, of interpreting "lésion corporelle" and "bodily injury" in a meaningful sense. As a matter of first impression as well as common sense there is inherent in these phrases the antithesis between "bodily injury" and mental injury. These phrases are prima facie inapposite to convey a meaning wide enough to include mental illness: see Dr Georgette Miller in her work Liability in International Air Transport, (1977), pp 127-128 which was cited by the Master of the Rolls in Morris v KLM [2002] QB 100, 115-116, paras 45-46. (4) It is common ground that the travaux preparatoires reveal no discussion or mention of liability for mental injury or illness. Given the spectre of enormous exposure to liability for carriers if claims for mental injury or illness were held to be within article 17, the omission of a reference to such claims during working sessions is revealing. Undoubtedly, at Warsaw and before carriers, aircraft insurers and countries whose national systems did not recognise liability for mental injury or illness would have been likely to argue against including such claims. This is not a case of mere silence: if the idea of including claims for mental injury was under consideration it would have demanded discussion. This indicates clearly and convincingly that the idea of covering mental injury or illness was never contemplated: see Fothergill v Monarch Airlines Ltd [1981] AC 251, 278B-G, per Lord Wilberforce. (5) While the domestic law position in France is irrelevant, Professor Malcolm Clarke in a book with the title Carriage of Goods by Air to be published in 2002 has shown that the amendments to the French text of the parallel Convention concerning International Carriage by Rail are instructive. In 1952 the text of article 26, the liability provision corresponding to article 17, read "de la mort, les blessures et toute autre atteinte a l'integrite corporelle". However, in 1961 the text was amended to read "de la mort, les blessures et toute autre atteinte a l'integrite corporelle ou mentale". The expression in both the current text of 1980 and the revision of 1999 is "dommage resultant de la mort, des blessures ou toute autre atteinte a l'integrite physique or psychique du voyageur". It follows that successive French drafters of the Convention over the last 50 years did not consider that, in the text of an international Convention, "corporelle" was wide enough to cover purely psychic injury. Moreover, this factor cogently reinforces the proposition that for the framers of the Warsaw Convention the natural and obvious antithesis would have been between bodily injury and mental injury. And it shows how unrealistic it is to describe mental injury as an unscientific term. In using the phrase "bodily injury" the framers were not selecting a scientific term. Rather they adopted a term which in ordinary signification, experience and acceptance had a restrictive meaning. (6) Lastly, there is also the fact that until some 50 years after the Warsaw Convention no claims for mental injury or illness against carriers were ever brought to judgment anywhere in the world: see Daddon v Air France (1984) 1 S & B Av R V11/141. Given that claims for mental injury and illness are inherently likely to occur more frequently and be more controversial than claims for physical injuries, the suggestion that such claims may for decades have been paid or settled is fanciful. The truth is that until the 1970s the view that such claims were not covered was the orthodox view throughout the commercial world. The Court of Appeal rightly attached importance to the fact that decades elapsed after 1929 before such claims were ever advanced against carriers. (7) In combination these factors indicate that a line was drawn in article 17 which excludes liability where a person suffers no physical injury but only mental injury or illness, such as clinical depression.

IV. Comparative jurisprudence

    18. Leaving aside for the moment the decisions under appeal there are decisions of courts of high standing to be considered. Lord Hope of Craighead has reviewed the case law comprehensively and accurately. I can therefore deal with the major cases quite shortly. In Daddon v Air France (1984) 1 S & B Av R V11/141 the Israeli Supreme Court held that claims for mental anguish suffered by passengers which was caused by the Entebbe high-jacking was covered by the phrase bodily injury in article 17 of the Warsaw Convention. Unfortunately, the Israeli Supreme Court impermissibly relied on changes in the aviation industry since 1929 and the current domestic law view of mental and psychological injury. In the two appeals under consideration the Court of Appeal and the Court of Session rejected this reasoning. Counsel for the passengers in the two appeals before the House placed no reliance on Daddon. Profound as my respect is for the Supreme Court of Israel I am driven to the same position in regard to this decision.

    19. The most important decision is that of the United States Supreme Court in Eastern Airlines Inc v Floyd (1991) 499 US 530. On a flight between the Bahamas and Miami a plane experienced engine failure. The crew announced that "the plane would be ditched". Fortunately, the engine was restarted and the plane landed safely. Passengers sued for emotional injury. The judgment of the Supreme Court addressed in detail the question whether article 17 of the Warsaw Convention permits recovery for mental injury unaccompanied by physical injury. For reasons substantially similar to those I have already given the Supreme Court answered this question in the negative. Justice Marshall, writing for a unanimous court expressed the ratio of the decision, at pp 552-553:

"We conclude that an air carrier cannot be held liable under article 17 when an accident has not caused a passenger to suffer death, physical injury, or physical manifestation of injury. Although article 17 renders air carriers liable for 'damage sustained in the event of' ('domage survenu en cas de') such injuries, . . . we express no view as to whether passengers can recover for mental injuries that are accompanied by physical injuries. That issue is not presented here because respondents do not allege physical injury or physical manifestation of injury."

Since then in El Al Israel Airlines Ltd v Tseng (1999) 525 US 155 the United States Supreme Court by an 8:1 decision followed Floyd and reiterated that without bodily injury there could be no recovery under article 17 for solely psychic or psychosomatic injury. That plainly excludes mental injury or illnesses. In a subsequent decision of a lower court in Weaver v Delta Airlines Inc (1999) 56FSupp 2d 1190 there appears to be a development to outflank Floyd and El Al by alleging that psychiatric injury or illness involves physical changes to the body. It is not strictly relevant on the issues before the House but I will briefly comment on it. But there is another decision of high authority on the principal point. In Kotsambasis v Singapore Airlines Ltd (1997) 42 NSWLR 110 the New South Wales Court of Appeal followed Floyd and held that bodily injury in article 17 does not include purely psychological injury.

    20. For my part the trilogy of decisions of high authority which I have mentioned in the last paragraph, and the reasoning in those decisions, reinforce the view that somebody who suffered no physical injuries but suffered mental injury or illness has no claim under article 17. On the other hand, I would hold that in two respects mental injury and illness may be relevant. First, there is no reason in principle to exclude from consideration pain and suffering caused by physical injury. It may therefore cover mental injury caused by a physical injury. In such cases the threshold requirement of bodily injury under the Convention is satisfied. It is therefore not an exception to the rule as I have stated it. Secondly, I would hold that if a relevant accident causes mental injury or illness which in turn causes adverse physical symptoms, such as strokes, miscarriages or peptic ulcers, the threshold requirement of bodily injury under the Convention is also satisfied. In Rosman v Trans World Airlines Inc (1974) 34 NY 2d 385, 399 the Court of Appeals of New York ruled by a majority of 6:1 that:

". . . as we read article 17, the compensable injuries must be 'bodily' but there may be an intermediate causal link which is 'mental' between the cause - the 'accident' - and the effect - the 'bodily injury'. And once that predicate of - liability - the 'bodily injury' - is established, then the damages sustained as a result of the 'bodily injury' are compensable including mental suffering"

For my part this reasoning in Rosman is faithful to the intent of the Warsaw Convention and I would adopt it. This too is not an exception. But mental injury or illness, such as clinical depression, is excluded.

V. Applying the distinction

    21. The outcome of applying the interpretation which I have preferred is that the claim in Morris v KLM falls outside article 17 because it involves mental injury or illness only and that the claim in King is within article 17 inasmuch as it involves a peptic ulcer together with pain and suffering associated with it.

VI. The new point: Weaver v Delta Airlines Inc

    22. Towards the end of the hearing in the House a new and important point emerged. It was based on Weaver v Delta Airlines Inc, 56 F Supp. 2d 1190. The potential effect of it was that psychiatric injury or illness involved physical changes to the body and is therefore "bodily injury". Given the decisions announced in the speeches today this point does not arise on the appeals under consideration. While the House, of course, has a broad discretion to deal with a point which does not strictly arise, it usually only does so when the point has been considered by courts below and addressed in written and oral argument before the House.

    23. It is necessary to explain the forensic background. In Morris v KLM Royal Dutch Airlines, Weaver was not cited to the first instance judge or considered by him. According to the Law Report it was not cited in argument in the Court of Appeal. It is not referred to in the judgment of the Court of Appeal. It is not referred to in the printed cases for the appellant or the respondent. In King v Bristow Helicopters 2001 SLT 126 there is no reference to Weaver in the first instance judgment. In the First Division judgment there are two passing references by Lord Rodger of Earlsferry (para 3 of the opinion) and Lord Reed (para 64 of his dissenting opinion) but it was not an issue addressed by the First Division. Weaver is not referred to in the printed cases of the appellant or respondent. The submissions to the House on Weaver were meagre. In the result the discussion of Weaver in the substantial speeches today contain many matters on which the House has not had the benefit of adversarial argument. That judges are not mere cyphers and may make their independent investigation into matters of law, I do not doubt. Indeed to some extent I have done so in this case. But it is reassuring if an important new issue, on which the House rules, has been properly investigated and debated by counsel. It has not happened in this case. It will be necessary to revisit the Weaver case when it is directly in issue and with the benefit of a proper exploration of the issues. My discussion of Weaver must be read subject to this caveat.

VII. The merits of the Weaver point

    24. In Weaver v Delta Airlines Inc, the United States District Court for the District of Montana, Billings Division, was faced with a claim by a passenger to recover compensation for post traumatic stress disorder resulting from an emergency landing. The issue was whether the plaintiff suffered bodily injury. The judge observed, at p 1192:

"Weaver's action here is distinguishable from previous cases, because her claim is presented as a physical injury and she relies on recent scientific research explaining that post-traumatic stress disorder evidences actual trauma to brain cell structures. Weaver's post-traumatic stress disorder evidences an injury to her brain, and the only reasonable conclusion is that it is, in fact, a bodily injury.

More particularly, the injury to her brain should be considered a 'bodily injury' as defined under the Warsaw Convention. Granted, Weaver's injury manifests itself in ways that are similar to the 'injuries' previously found not compensatable in similar cases under the Warsaw Convention. However, the central factor here is not legal, but medical. The legal question in this case is simply whether the Warsaw Convention allows recovery for this particular kind of bodily injury, ie a brain injury (even with slight physical effects). The answer must be yes.

The court is cognizant that the Warsaw Convention chose to preclude recovery for purely psychic injuries, and the court respects the Supreme Court's determination in Floyd that such was a legislative choice. Moreover, the present holding has the potential of allowing for more valid actions under the Warsaw Convention, with the increase attributable only to the increased sophistication of medical science. However, no floodgates of litigation will be opened by allowing for claims such as Weaver's, which are based on a definite diagnosis of a disorder that arises from a physical injury that is medically verifiable. Fright alone is not compensatable, but brain injury from fright is. Unlike the plaintiffs in Floyd and its progeny, Weaver's injury is a 'bodily injury' as defined by the Warsaw Convention."

The 9th Circuit is the appellate court which supervises the Weaver court. Last year the 9th Circuit followed Floyd and Tseng in a post-traumatic stress syndrome case where there was no evidence of physical injury but left open the issue whether such a physical injury would satisfy the requirements of article 17: Carey v United Airlines (2001) 28 Avi 15,408, 15,415 (footnote 47). On the other hand, Weaver was followed in In re Air Crash at Little Rock, Arkansas (2000) 118 F Supp 21d 916. There are also lower court decisions the other way: Jack v Trans World Airlines Inc (1994) 854 F Supp 654; Terrafranca v Virgin Atlantic Airways Ltd (1998) 151 F 3d 108; Alvarez v American Airlines Inc (1999) 27 Avi 17,214; (2000) 27 Avi 17,475. As a matter of precedent Weaver hardly rests on secure foundations.

    25. In aid of the argument based on Weaver points have been made which I would not dispute. First, there is the undoubted fact that medical science generally and psychiatry in particular have advanced since 1929. While it is a matter for expert opinion - and there was none before the House - I accept that in cases of recognisable psychiatric illnesses, such as clinical depression (as in the case of Morris v KLM) and post traumatic stress disorder (as in Weaver) there is a physical connection between the illness of the mind and the body inasmuch as the central nervous system which includes the nervous tissue of the brain is involved: see Black's Medical Dictionary, 39th ed, (1999), sv "central nervous system", "brain", "depression" and "post traumatic stress disorder"; and Gelder, Mayou and Cowen, Shorter Oxford Textbook of Psychiatry, 4th ed (2001), passim. Secondly, I accept that courts of law cannot ignore advances in scientific knowledge. In R v Ireland [1998] AC 147, 156, and in Frost v Chief Constable of South Yorkshire Police [1999] 2 AC 455, 492, I observed that courts of law must act on the best medical insight of the day. This is also an uncontroversial point and does not provide the answer to the point of interpretation before the House. Thirdly, statutes are generally always speaking, and ought therefore to be interpreted in light of the contemporary social and scientific world. This is not a rule of law but a principle of construction, which may be displaced by a contrary intent revealed by a particular statutory context: R v Ireland, [1998] AC 147 158D-G. Given that the rationale of the principle is that statutes are generally intended to endure for a long time, one can readily accept that multilateral international trade conventions, which are by statute incorporated in our law, should be approached in a similar way. Indeed one may say that a fortiori they ought to be so interpreted. Again, this principle does not provide an answer to the question before the House.

    26. None of these points, nor a combination of them, undermines the conclusion that at the time of the Warsaw Convention a line was drawn between bodily injury (that is, involving non fatal physical injury) and mental injury or illness. If this view is correct the argument based on Weaver cannot succeed. To accept it would be to ignore the contextual meaning of the Warsaw Convention. It is no answer to say that the distinction between the body and the mind is arbitrary. To some extent it is. Nevertheless, scientifically and in common sense there is a real distinction between physical injuries and mental injury or illness. In Weaver the threshold requirement of a bodily injury within the meaning of the Convention was absent. For reasons already explained the Weaver case (involving only alleged psychiatric injury) is quite different from the two qualifications which I mentioned in para 12 (involving physical manifestations). If cases of mental injuries and illnesses are to be brought within the Convention system, it must be done by amendment of the Convention system and not by judicial creativity. In my view Weaver was wrong in holding that there is no legal issue but only a medical question. Weaver was not faithful to the law as settled in Floyd and Tseng by the United States Supreme Court. And the acceptability of the Weaver line of authority is not saved by restricting it to recognisable psychiatric illnesses. That is a legal concept taken from post 1929 English case law: see McLoughlin v O'Brian [1983] AC 410, 418; R v Chan-Fook [1994] 1 WLR 689; R v Ireland, [1998] AC 147. It was not available at Warsaw to serve as an autonomous concept in article 17. It does not satisfy the criterion of approaching the interpretation of a multilateral trade convention "unconstrained by technical rules of English law, or by English legal precedent, but on broad principles of general acceptation": James Buchanan & Co Ltd v Babco Forwarding & Shipping (UK) Ltd [1978] AC 141, 152.

    27. Moreover, the court in Weaver failed to take into account that to this day the extension of the Warsaw system to include mental injury and illnesses is too controversial to command sufficient international support. In part this must be due to policy factors, namely the expected escalation of claims, far beyond the incidence of physical claims, if mental injuries and illnesses are held to be included in article 17. The potential range of such cases is one factor. The cost of meeting such claims is likely to be huge. Another aspect is the cost of psychiatric evidence by rival experts to determine whether claims are cognisable. Finally, there is the cost of legal proceedings. It may be that large countries, and major airlines, could cope with such additional exposure. For smaller countries, and smaller airlines, it could be a serious matter. In 1929 the aviation industry was fragile as it apparently is today. In 1929 the world was not ready to include mental injuries and illnesses within the scope of article 17. It is not ready to do so in 2002.

    28. It is true, of course, that a judge's instinct must be for full corrective justice. But the language and contextual scene of article 17 rule out the Weaver approach to its interpretation. Courts of law must avoid the reproach that they are becoming the redrafters of the Convention.

    29. I would hold that Weaver should not be followed.

VIII. The judgments below

    30. I am much indebted to the judgments of the Court of Appeal and the First Division for a careful and analytical exploration of the issues in the cases under consideration. It will be clear, however, that I prefer the opinion of the Court of Appeal to the views of the majority of the First Division.

    31. In Morris v KLM, [2002] QB 100, 128, para 103 the Master of the Rolls observed:

"If and when the 1999 Montreal Convention comes into force there may be scope for argument, on the basis of the travaux préparatoires evidencing the consideration that was given to mental injury, that those who drafted the Convention intended the meaning of the phrase 'bodily injury' to turn on the jurisprudence of the individual state applying that Convention. We do not consider that this course is open to those who have to interpret that phrase in the Warsaw Convention. In that Convention the phrase means 'physical injury'".

A statement to Parliament by the Government further elucidated the position. The relevant House of Commons Written Answers for 3 July 2000 (Col 87W and Col 88W) were as follows:

"Montreal Convention

Mr Dismore: To ask the Secretary of State for the Environment, Transport and the Regions what representations he has made in relation to the Montreal Convention to ensure UK passengers will be able to claim compensation for psychiatric injury caused by air accidents; and if he will make a statement. [128317].

Mr Hill: Damages for mental injury caused by air accidents are already recoverable in the UK when associated with physical injury. In preparation for the Diplomatic Conference held in Montreal in May 1999, at which the Convention was signed, the UK supported a proposal by Sweden for a separate head of claim for mental injury. Prior to the Conference, however, that proposal was withdrawn from the draft text of the Convention. Our position was that a separate claim for mental injury could be advocated only if there was sufficient support to gain global agreement. There was not sufficient support so, in the interest of securing the best deal for the UK, it was decided to support the text of the Convention without a separate reference to mental injury. The Conference 'travaux préparatoires', nevertheless, indicate that damages for mental injury can be recovered in certain states and that jurisprudence in this area is developing."

This is how matters stand at present. Limited progress towards the admission of claims for mental injury and illness must await the coming into operation of the Montreal Convention.

X. Disposal

    32. For the reasons I have given I would dismiss the appeal in Morris v KLM. For the same reasons I would allow the appeal in King and restore the decision of the Lord Ordinary. I am in agreement with the speech of my noble and learned friend Lord Hope of Craighead.

LORD HOPE OF CRAIGHEAD

My Lords,

    33. The question in these appeals relates to the meaning of the phrase "bodily injury" in article 17 of the Warsaw Convention of 1929 as amended at The Hague in 1955, which was incorporated into the law of the United Kingdom as Schedule 1 to the Carriage by Air Act 1961. To what extent, if at all, is the carrier liable under this article where the injury which the passenger sustained in the accident was a mental injury and not a physical injury?

    34. Although both cases raise the same question, they arose out of different facts and circumstances and they have arrived here by different routes. I shall give a brief summary of the cases in their historical order before examining the point of law which they have raised.

Mr King's case

    35. On 22 December 1993 Philip King was a passenger on a helicopter which Bristow Helicopters Ltd were operating in poor weather in the North Sea. It took off from the helideck of a floating production platform in the Beryl oilfield, rose to a height of about 35 feet and hovered for a short period. Suddenly its two engines flamed out, there was a loud bang and the helicopter descended rapidly and landed heavily on the helideck. It was engulfed in smoke which prevented the passengers from seeing out. They did not know whether it had landed safely or was on the edge of the helideck and liable to fall off into the sea. There was panic on board and the passengers were shouting. Mr King did not sustain any physical injury while he was on board or when he disembarked from the helicopter. But he was extremely frightened, and he developed several psychiatric conditions including a moderate post-traumatic stress disorder as a result of the accident. The stress led to the onset of peptic ulcer disease, which caused him severe pain and an exacerbation of pre-existing dyspeptic symptoms.

    36. Mr King raised an action of damages in the Court of Session against Bristow Helicopters Ltd ("Bristow"). His flight was non-international carriage by air within the meaning of the Carriage by Air Acts (Application of Provisions) Order 1967: SI 1967/480. Schedule 1 to the 1967 Order applies to such carriage the Warsaw Convention as amended by the Hague Protocol. He claimed damages under article 17 of the Convention both for the psychiatric conditions and for the peptic ulcer disease.

    37. On 13 November 1998 the Lord Ordinary (Lord Philip), having heard a debate on the procedure roll, excluded from probation all the averments about the psychiatric conditions which Mr King sustained as a result of the accident. He also dismissed actions which had been raised by two other passengers which had been heard at the same time: Hammond v Bristow Helicopters Ltd, 1999 SLT 919. But he allowed the averments about Mr King's peptic ulcer disease to go to proof before answer. A reclaiming motion was marked by Mr King in which he sought a proof before answer of all his averments. A cross-reclaiming motion was marked by Bristow in which they asked the court to dismiss the action. On 12 July 2000 the First Division (the Lord President (Rodger) and Lord Cameron of Lochbroom, Lord Reed dissenting), allowed Mr King's reclaiming motion, refused the cross-reclaiming motion and allowed the parties a proof before answer of their averments: 2001 SLT 126. It is against that interlocutor that Bristow have appealed to this House.

    38. During the hearing before the First Division Mr King's counsel said that his averments should be interpreted as disclosing that Mr King was suffering from three psychiatric conditions: post-traumatic stress disorder, chronic depression and fear of flying, and that his peptic ulcer disease was caused, or materially contributed to, by his psychiatric conditions: 2001 SLT 126, 129G-H. He also stressed that he was not offering to prove that Mr King's psychiatric conditions were caused by any physiological changes in his body. That position was maintained in the hearing before your Lordships. The appeal in this case was argued on the basis that Mr King suffered two distinct kinds of injury as a result of the accident when he was on board the helicopter. The first was a mental injury which led to the three psychiatric conditions described in the averments. The second was a physical disorder consisting of the peptic ulcer disease which the psychiatric conditions had caused.

    39. On 6 September 1998 Miss Morris was on an international flight which was being operated by KLM Royal Dutch Airlines from Kuala Lumpur to Amsterdam. She was just under 16 years old and was travelling as an unaccompanied minor. She was seated next to two men who were speaking French to each other. After a meal she fell asleep and woke to discover the hand of the man next to her touching her left thigh from the hip to the knee. He was caressing her between her hip and knee and his fingers dug into her thigh. She got up, walked away and told an air hostess what had occurred. She became very distressed and on her return went to see a doctor. He found that she was suffering from a clinical depression amounting to a single episode of a major depressive illness. She has now made a full recovery.

    40. Miss Morris brought a claim for damages against KLM in Bury County Court. The flight on which she was travelling was international carriage by air for reward within the meaning of article 1 of the Warsaw Convention as amended by the Hague Protocol. She claimed damages under article 17 of the Convention as incorporated into English law by Schedule 1 to the 1961 Act. In the particulars of injuries annexed to her claim form she stated that she relied for particulars of her injury on reports from Dr N J Cooling, a consultant psychiatrist. In his report Dr Cooling said that in his opinion Miss Morris was emotionally shocked by what happened to her, that since returning home she had shown the characteristic features of a clinical depression and that the diagnosis according to DSM-IV criteria was one of a single episode of a major depressive illness (296.2) for which her general practitioner, recognising the very dramatic change in her mental state, had started her on anti-depressant treatment.

    41. By consent the issue of liability was treated in the county court as a preliminary issue. This issue was heard on the basis of agreed facts before Judge Carter QC. On 1 December 2000 he gave judgment for Miss Morris for damages to be assessed, but gave KLM permission to appeal. In the appeal the meaning of "accident" and "bodily injury" in article 17 were both in issue. On 17 May 2001 the Court of Appeal (Lord Phillips of Worth Matravers MR, Peter Gibson and Latham LJJ) allowed the appeal: [2002] QB 100. Miss Morris now appeals against that order to this House.

    42. Before Judge Carter both parties had claimed that the issue between them fell to be determined in a manner that entitled each party to summary judgment. The hearing in the Court of Appeal also proceeded on the agreed premise that there is a distinction between physical injury and mental injury: that is to say, that physical injury involves damage or adverse change to the structure of the body, whereas mental illness adversely affects the well-being of the mind without organic change to the body: [2002] QB 100, 113, para 35. Lord Phillips of Worth Matravers MR said, at p 114, para 40, that the appeal had to be approached on the premise that mental illness and physical injury are distinguishable and that the claimant had accepted that she suffered no physical injury.

    43. It can be seen from this summary that no attempt has been made in either case to show that the mental illness or injury was caused to any extent by physical injury. The only physical injury which is said to have been sustained in either case is what may conveniently be described as the physical manifestation of a mental injury in Mr King's case. This is the peptic ulcer disease which he claims to have developed as a result of the psychiatric conditions described in his averments.

    44. As this brief summary has demonstrated, various expressions may be used to describe an injury which affects the mind but is not accompanied by any physical injury. In Mr King's case, the Lord President observed that among the expressions used in the extensive case law and literature on this subject are "shock", "mental distress", "mental injury", "psychic injury", "psychological injury" and "psychiatric injury". For the sake of consistency he chose to adopt the term "psychological injury": 2001 SLT 126, 129K. In Miss Morris's case, Lord Phillips of Worth Matravers MR used the expressions "mental injury" and "mental illness": [2002] QB 100, 113, 114, paras 35 and 40.

    45. For the purposes of this judgment, except when I am dealing which Mr King's peptic ulcer disease, I propose to adopt the phrase "mental injury". I think that it is preferable to adhere to the noun "injury", which is the word used in article 17. Moreover, I would wish to recognise at the outset that the word "injury" does not extend to the reactions which a passenger may experience during or as a result of an accident such as fright, fear or anxiety. Emotional upsets of that kind seem to me to be clearly outside the scope of the article. Article 17 is concerned only with something that can properly be described as an injury.

    46. As for the choice of adjective, I would prefer to use the word "mental" to "psychological", although I adopted the latter expression in Sidhu v British Airways plc [1997] AC 430, 441B-D. In that case it was suggested that the phrase "bodily injury" in article 17 ought to be construed as including psychological damage, especially if it were shown to have a physiological basis by medical evidence. But the point did not require decision in that case. I recognise at once that the phrase "mental injury" is open to criticism for the reasons which have been so carefully explained by my noble and learned friend, Lord Hobhouse of Woodborough. But, for the purpose of these appeals, all I am looking for is a convenient expression to embrace those conditions affecting the mind which are sufficiently serious to fall within the concept of injury. The words "physical" and "mental" seem to me to provide a sufficient contrast to mark the broad dividing line which has been identified between the two sides of the argument. The question where that line is to be drawn is a difficult one, but is not the question to which the passengers' counsel addressed their principal argument. Their argument was that it was sufficient for them to show that the passengers sustained a mental injury.

    47. Underlying the problem of terminology there is a more fundamental question. It has for a long time been recognised that it is not possible to maintain a rigid distinction between the body and the mind in the law relating to liability in damages for negligence. In Bourhill v Young [1943] AC 92, 103 Lord Macmillan recognised that the crude view that the law should take cognisance only of physical injury resulting from actual impact had been discarded and that it was recognised that an action will lie for injury by shock sustained without direct contact. As he put it, the distinction between mental shock and bodily injury was never a scientific one.

    48. This theme was developed in Page v Smith [1996] AC 155, 181A-D by Lord Browne-Wilkinson, where he observed that medical science has demonstrated that the body can suffer injuries which are not demonstrably attributable to physical injury:

"Injuries of this type may take two forms. First, physical illness or injury not brought about by a chain of demonstrable physical events but by mental or emotional stresses, ie by a psychiatric route. Examples are a heart attack or a miscarriage produced by shock. In this case, the end product is a physical condition although it has been brought about by a process which is not demonstrably a physical one but lies in the mental or nervous system. The second form is psychiatric illness itself which is brought about by mental or emotional stresses, ie by a psychiatric route. Because medical science has so far been less successful in demonstrating the nature of psychiatric illness and the processes whereby it is brought about by the psychiatric route, the courts have been more reluctant to accept the risk of such illness as being foreseeable. But since the decision of this House in McLoughlin v O'Brian [1983] 1 AC 410 it has been established that, in certain circumstances, a defendant can be liable for illness or injury, whether psychiatric or physical, produced in a plaintiff by purely psychiatric processes, without any direct physical impact on, or injury to, the limbs or organs of the plaintiff."

    49. This passage helps to set the scene for these appeals. The branch of medical science which is concerned with psychiatric disorders and disturbances is still in the process of development. It is not yet fully understood. We are able to identify physical conditions which have been brought about by a psychiatric route. Mr King's peptic ulcer disorder is a physical condition of this type. We can also identify various psychiatric illnesses which, according to our present state of knowledge, appear not to have been the product of any kind of physical injury. But the distinction between what I have described as a mental injury and a physical injury is unclear, and the extent to which it can be maintained is debatable. It may now be possible to show objectively, by means of expert medical evidence, that a psychiatric illness is due to a disturbance within the patient's own central nervous system which drugs can control or alleviate. As I shall show later, this has been the basis for some recent decisions in the Federal courts in the United States that a psychiatric illness of that kind is a physical injury. It is of some importance therefore to appreciate the basis on which counsel for the passengers in these cases presented their argument.

    50. The argument in the courts below has assumed that the mental injury which Mr King and Miss Morris sustained is one which cannot be attributed to any kind of physical injury. This point was expressly conceded in Mr King's case. Lord Phillips of Worth Matravers MR acknowledged in Miss Morris's case that the stage may one day be reached when a physical cause can be demonstrated for these kinds of psychiatric illnesses: [2002] QB 100, 114, para 40. But he said her appeal had to be approached on the assumption that that stage had not yet been reached. I understand the point to have been a matter of express concession in her case also. Mr Braslavsky was content to present her appeal to this House on the same basis. He did not attempt to link her mental injury to any physical condition affecting the tissues of her body. He maintained that it was sufficient for her to show that Dr Cooling's diagnosis was that she was suffering from a depressive illness which had a recognised international classification. He said that his case was that she was suffering from a recognised mental disorder, not that her disorder could be linked to any injury to her body.

    51. It may be helpful, before embarking upon an examination of the words "any other bodily injury" in article 17 of the Convention, to examine the present state of our own jurisprudence as to how similar words in domestic legislation are to be interpreted. This is not to say that the approach which we take when we are construing our own legislation is the approach which we should take when we are construing the Convention. The construction of an international convention proceeds upon different principles. But this exercise may help to put the problem in its current context.

    52. The word "bodily" is not unknown in the legislation of the United Kingdom. The phrase "bodily functions" in modern social security legislation has received a broad interpretation: see Cockburn v Chief Adjudication Officer [1997] 1 WLR 799. Lord Slynn of Hadley said, at p 813, that this phrase could extend to the operation of the senses and their communication to the brain. But the concept of function is not the same as that of injury. More in point is the use of the phrase "bodily harm" in English criminal law to describe various categories of offences against the person: see sections 18, 20 and 47 of the Offences against the Person Act 1861. The phrase "grievous bodily harm" is used in sections 18 and 20 of that Act. The phrase "actual bodily harm" is used in section 47.

    53. The meaning of the phrase "actual bodily harm" in that context was considered in R v Chan-Fook [1994] 1 WLR 689. The issue was whether an assault which caused no physical injury but caused an hysterical and nervous condition in the victim was an assault occasioning actual bodily harm. It was held by the Court of Appeal that, while the phrase did not include emotions such as fear or panic, it was capable of including psychiatric injury. Hobhouse LJ, delivering the judgment of the court, said at p 694B-C:

"These are three words of the English language which require no elaboration and in the ordinary course should not receive any. The word 'harm' is a synonym for injury. The word 'actual' indicates that the injury (although there is no need for it to be permanent) should not be so trivial as to be wholly insignificant."

    54. Later, at p 694H, he said that in certain cases an explanation might be required of what is involved in the word "bodily". Rejecting the argument that inclusion of the word in the phrase "actual bodily harm" limits harm to harm to the skin, flesh and bones of the victim, he said, at p 6956:

"The body of the victim includes all parts of his body, including his organs, his nervous system and his brain. Bodily injury therefore may include injury to any of those parts of his body responsible for his mental and other faculties."

He recognised that there was a line to be drawn between mere emotions on the one hand and some psychiatric illness on the other, and that this was a matter for expert evidence. He concluded his discussion of this issue with these words, at p 696C:

"Accordingly the phrase 'actual bodily harm' is capable of including psychiatric injury. But it does not include mere emotions such as fear or distress nor panic nor does it include, as such, states of mind that are not themselves evidence of some identifiable clinical condition."

    55. In R v Ireland [1998] AC 147 the question was whether psychiatric illness was capable of amounting to bodily harm in terms of sections 18, 20 and 47 of the 1861 Act. Two appeals were before the House. In neither case had the victims suffered from structural injuries to the brain such as might require the intervention of a neurologist, nor was it suggested that they had developed psychotic or psychoneurotic conditions. The case was that they had developed anxiety and depressive disorders, which could be distinguished from simple states of fear or problems in coping with everyday life. It was held, in the light of contemporary knowledge covering recognisable psychiatric injuries, that injuries of that kind fell within the phrase "bodily harm" as used in the 1861 Act. Having examined the progress in scientific understanding since that date and the recognition in recent English case law that in the relevant context the distinction between physical and mental injury is by no means clear cut, Lord Steyn said, at pp 158G-159A:

"The proposition that the Victorian legislator when enacting sections 18, 20 and 47 of the 1861 Act, would not have had in mind psychiatric illness is no doubt correct. Psychiatry was in its infancy in 1861. But the subjective intention of the draftsman is immaterial. The only relevant inquiry is as to the sense of the words in the context in which they are used. Moreover the Act of 1861 is a statute of the 'always speaking' type: the statute must be interpreted in the light of the best current scientific appreciation of the link between the body and psychiatric injury."

He concluded, at p 159B, that the word "bodily harm" must be interpreted in the context of the 1861 Act so as to include recognisable psychiatric illness.

    56. In the light of these decisions and the reasoning which was used to support them, I think that there is little doubt that, if same words as those in article 17 were used in a United Kingdom statute to describe the kinds of personal injury caused by an accident that would entitle the victim to recover damages, they would now be held to extend to those kinds of mental injury that could be shown to amount to a recognisable psychiatric illness or injury by expert evidence. The depressive illness which Dr Cooling has diagnosed in Miss Morris's case according to DSM IV criteria (296.2) would seem to fall readily into that category. This brings me to the question whether the words used in the Convention can be construed in the same way.

    57. The history of the Convention and its incorporation into our law which I shall now outline has been fully described elsewhere and is not controversial. But it is necessary to place it on record in this judgment by way of background.

    58. The Warsaw Convention was concluded in French on 12 October 1929. Arrangements were made for it to be received into the law of the United Kingdom by the Carriage by Air Act 1932, and the French text was ratified by the United Kingdom on 14 February 1933. The 1932 Act gave the force of law to a translation of the French text into English. Only the English text was set out in the First Schedule to that Act. Section 4 of the 1932 Act gave power by Order in Council to apply the provisions of the First Schedule to non-international carriage by air, subject to such exceptions, adaptations and modifications, if any, as might be specified. The provisions of the First Schedule were applied, in a modified form, to non-international carriage by air by the Carriage by Air (Non-international Carriage) (United Kingdom) Order 1952 (SI 1952/158).

    59. The Warsaw Convention was amended by a Protocol which was concluded at The Hague in 1955. The Carriage by Air Act 1961 gave effect to the Convention concerning international carriage by air, now known as "The Warsaw Convention as Amended at The Hague, 1955", so that it too might have the force of law in the United Kingdom in relation to any carriage by air to which the Convention applied. Force of law to the Convention in regard to international carriage by air is given by section 1(1) of the 1961 Act, read together with Schedule 1. Part I of the Schedule to that Act sets out a translation of the Convention into English. Part II sets out the text in French. Section 1(2) of the Act provides that, if there is an inconsistency between the text in English in Part I of the Schedule and the text in French in Part II, the text in French shall prevail.

    60. Once again steps were taken to enable the provisions of the Convention, in their amended form, to be applied to carriage by air, not being carriage to which the Convention applies: see section 10(1). This was done by the Carriage by Air Acts (Application of Provisions) Order 1967 (SI 1967/480), for the purposes of which the expression "the amended Convention" is defined by article 2(1) of the Order as meaning the English text of the Warsaw Convention as amended by the Hague Protocol. The Order applies to all carriage by air, not being carriage to which the amended Convention applies: article 3. Article 4(a) provides that Schedule 1 to that Order is to have effect in respect to carriage which is not "international carriage" as defined in Schedule 2 to the Order. Schedule 2 sets out the provisions relating to international carriage in the unamended Warsaw Convention, to which continuing force had to be given as some states party to the Warsaw Convention had not adopted the Hague Convention.

    61. The carriage of Miss Morris on the KLM flight from Kuala Lumpur to Amsterdam was plainly international carriage by air to which the rules in Schedule 1 to the 1961 Act ("the Hague rules") apply. In her case the position is straightforward. The statute provides that the French text is to prevail over the English text if there is any inconsistency: section 1(2).

    62. The carriage of Mr King in the helicopter when it took off from the Beryl platform in the North Sea was not international carriage as defined in Schedule 2 to the 1967 Order. The rules which applied to that carriage were those set out in Schedule 1 to that Order ("the United Kingdom rules"). But the parties in Mr King's case are agreed that, although the Hague rules do not apply in his case, it is proper in his case too to look at the French text of the Convention when construing the articles in Schedule 1.

    63. For the reasons which I gave in Herd v Clyde Helicopters Ltd [1997] AC 534, 552B-E I would endorse this approach. The long title of the 1961 Act states that it is "An Act to give effect to the Convention concerning international travel by air known as 'the Warsaw Convention as amended at The Hague, 1955', to enable the rules contained in that Convention to be applied, with or without modification, in other cases and, in particular, to non-international carriage by air; and for connected purposes." One of the primary objectives of the United Kingdom rules, as in the case of the Convention itself, was to eliminate conflict of laws problems which are just as likely to arise from non-Convention as from Convention carriage. As Lord Reed said in Mr King's case, the practical advantages of construing identical provisions in all three sets of rules, including those in Schedule 2 to the 1961 Order ("the Warsaw rules"), in the same sense are manifest: 2001 SLT 126, 159I.

    64. Chapter III of the Convention contains the provisions which are relevant to these appeals. It is headed "Liability of the Carrier." The articles in this chapter are numbered from 17 to 30. Article 17 is concerned with the carrier's liability for death or injury suffered by a passenger. Article 18 is concerned with the carrier's liability for destruction or loss of or damage to registered baggage or cargo. Article 19 is concerned with damage caused by delay in the carriage by air of passengers, baggage or cargo. Article 20 provides that the carrier is not liable if he proves that he and his servants or agents have taken all necessary measures to avoid the damage or that it was impossible for him or them to take such measures. These provisions must be read together with article 24, which provides that, in the cases covered by these articles, any action for damages, however founded, can only be brought subject to the conditions and limits set out in the Convention.

    65. The general character and purpose of the Hague rules, which were authorised to be applied by Order in Council to non-international carriage by air by section 10(1) of the 1961 Act, were described in Holmes v Bangladesh Biman Corporation [1989] AC 1112, 1129 by Lord Bridge of Harwich; see also Sidhu v British Airways plc [1997] AC 430, 453-454. They impose liability on the carrier without proof of fault in respect of the death of or injury to passengers, damage to or loss of baggage or cargo and delay to passengers, baggage or cargo. They impose limits on the amount recoverable in respect of the death, injury, damage, loss or delay. They nullify contractual provisions tending to relieve the carrier of liability or to lower the limits of liability. Actions for damages to enforce the rights given by the Rules can only be brought subject to the Rules, and they exclude other remedies. What was sought to be achieved was a uniform international code which could be applied by the courts of all the high contracting parties. In those areas with which it deals the rules which it lays down were intended to be uniform and exclusive of resort to domestic law.

    66. From the point of view of the passenger or the owner of baggage or cargo, the imposition of liability without proof of fault on the carrier and the nullification of provisions relieving him of liability or restricting the amount of his liability are very significant advantages. From the point of view of the carrier too however there are significant advantages in the system laid down by the Convention. A principal consequence of that system is the exposure of the carrier to liabilities without the freedom to contract out of them. But it defines those situations in which compensation is to be available, and it sets out the limits of liability and the conditions under which claims to establish liability, if disputed, are to be made. A balance has been struck between these competing interests, in the interests of certainty and uniformity.

    67. As Lord Phillips of Worth Matravers MR said in Miss Morris's case [2002] QB 100, 119, para 67, it is not difficult to see that the Convention was intended to create a uniform set of circumstances in which a carrier by air would be obliged to pay compensation for damage sustained by a passenger. But this raises the question, what damage? To answer this question it is necessary to turn to the wording of the articles.

    68. I must first set out the relevant articles in Chapter III of the Convention, using the English text as set out in Part I of the Schedule to the 1961 Act.

"17. The carrier is liable for damage sustained in the event of the death or wounding of a passenger or any other bodily injury suffered by a passenger, if the accident which caused the damage so sustained took place on board the aircraft or in the course of any of the operations of embarking or disembarking.

18(1). The carrier is liable for damage sustained in the event of the destruction or loss of, or of damage to, any registered baggage or any cargo, if the occurrence which caused the damage so sustained took place during the carriage by air . . .

19. The carrier is liable for damage occasioned by delay in the carriage by air of passengers, baggage or cargo.

20. The carrier is not liable if he proves that he and his servants or agents have taken all necessary measures to avoid the damage or that it was impossible for him or them to take such measures.

21. If the carrier proves that the damage was caused by or contributed to by the negligence of the injured person the court may, in accordance with the provisions of its own law, exonerate the carrier wholly or partly from his liability . . .

24(1). In the cases covered by articles 18 and 19 any action for damages, however founded, can only be brought subject to the conditions and limits set out in this Convention.

(2). In the cases covered by article 17, the provisions of the preceding paragraph also apply, without prejudice to the questions as to who are the persons who have the right to bring suit and what are their respective rights."

    69. The article 17 in the French text as set out in Part II of the Schedule is in these terms:

"Le transporteur est responsable du dommage survenu en cas de mort, de blessure ou de toute autre lésion corporelle subie par un voyageur lorsque l'accident qui a causé le dommage s'est produit à bord de l'aéronef ou au cours de toutes opérations d'embarquement et de débarquement."

    70. The wording of article 17 indicates that three things must be established in order to demonstrate that the carrier is liable. The first is that the passenger must have sustained death or wounding or other bodily injury. The second is that there must have been an accident which took place on board the aircraft or in the course of any of the operations of embarking or disembarking. The third is that the passenger's death, wounding or bodily injury must have been caused by the accident.

    71. In Air France v Saks (1985) 470 US 392, 398-399 the United States Supreme Court said that the text of the Convention implies that, however the word "accident" is defined, it is the cause of the injury that must satisfy that definition rather that the occurrence of the injury alone. At p 405 the court concluded that liability under article 17 arises only if a passenger's injury is caused by an unexpected or unusual event that is external to the passenger. This interpretation of the article was approved by the Court of Appeal in Chaudhari v British Airways plc The Times, 7 May 1997; Court of Appeal (Civil Division) Transcript No 590 of 1997. It was also approved and applied by Lord Phillips of Worth Matravers MR in Miss Morris's case [2002] QB 100, 110, para 24. I agree that it gives the word "accident" a natural and sensible meaning in the context of article 17, and I too would adopt this approach.

    72. In Mr King's case it was accepted that the incident which he describes in his averments was an accident. In Miss Morris's case the defendant argued that the indecent assault which she suffered could not reasonably be described as an accident. The Court of Appeal held that it was not necessary to show that the event had any relationship with the operation of the aircraft or carriage by air, that the incident in which Miss Morris was involved exemplified a special risk inherent in air travel and that she had sustained an accident within the meaning of the article: [2002] QB 100, 112, para 31. So long as it occurred during the time when the passenger was in the charge of the carrier, the passenger was entitled to be compensated for its consequences if the carrier was not able to discharge the burden imposed by article 20 of showing that he and his servants or agents had taken all necessary measures to avoid the damage or that it was impossible for him or them to take such measures. There has been no appeal against this finding by the Court of Appeal.

    73. The remaining question, which is common to both cases, is whether the mental injury which the passengers sustained falls within the scope of the expression "bodily injury". That is the central issue in these appeals.

    74. I propose to approach this question by examining first the principles of interpretation that are relevant to this issue. I shall then discuss the natural meaning of the words used in both the English and the French texts. I shall then look at the travaux préparatoires in order to see what light, if any, they may cast on this issue. Finally I shall examine the jurisprudence of other jurisdictions in order to see whether a settled practice has developed as to the meaning of this phrase and, if not, what guidance can be obtained from the approach which has been taken to the problem in the United States. Before undertaking this exercise I should like to pay tribute to the researches of counsel, to the quality of the judgments both in the Court of Session and in the Court of Appeal and to the diligence and attention to detail with which they have set out and examined the relevant material.

    75. Almost everything that needs to be said on this subject was referred to in Sidhu v British Airways plc [1997] AC 430, and it is necessary here to provide only a brief summary.

    76. We are concerned in this case with the meaning of words used an international convention. The Convention must be considered as a whole, and it should receive a purposive construction: Grein v Imperial Airways Ltd [1937] 1 KB 50, 74-76 per Greene LJ; Fothergill v Monarch Airlines Ltd [1981] AC 251, 279 per Lord Diplock. The ordinary and natural meaning of the words used in the English text in Part I of the Schedule provides the starting point. But these words must also be compared with their equivalents in the French text in Part II of the Schedule, as section 1(2) of the 1961 Act tells us that if there is any inconsistency the text in French shall prevail.

    77. As the language was not chosen by English draftsmen and was not designed to be construed exclusively by English judges, it should not be interpreted according to the idiom of English law. What one is looking for is a meaning which can be taken to be consistent with the common intention of the states which were represented at the international conference. The exercise is not to be controlled by technical rules of English law or domestic precedent. It would not be right to search for the legal meaning of the words used, as the Convention was not based on the legal system of any of the contracting states. It was intended to be applicable in a uniform way across legal boundaries.

    78. In situations of this kind the language used should be construed on broad principles leading to a result that is generally acceptable: see Stag Line Ltd v Foscolo, Mango Co Ltd [1932] AC 328, 350 per Lord Macmillan; James Buchanan & Co Ltd v Babco Forwarding and Shipping (UK) Ltd [1978] AC 141, 152 per Lord Wilberforce. But this does not mean that a broad construction has to be given to the words used in the Convention. As Lord Phillips of Worth Matravers MR said in Miss Morris's case [2002] QB 100, 125, para 90, it is not axiomatic that the broad principle of "general acceptation" described in these cases militates in favour of a broad rather than a narrow interpretation of the phrase "any other bodily injury"

    79. It is legitimate to have regard to the travaux préparatoires in order to resolve ambiguities or obscurities: Forthergill v Monarch Airlines Ltd [1981] AC 251, 278 per Lord Wilberforce. But caution is needed in the use of this material, as the delegates may not have shared a common view. An expression by one of them as to his own view is likely be of little value if it was met simply by silence on the part of the other delegates. It will only be helpful if, after proper analysis, the travaux clearly and indisputably point to a definite intention on the part of the delegates as to how the point at issue should be resolved.

    80. It is also legitimate to have regard to subsequent practice in the application of the Convention, if this shows that the contracting parties were in agreement as to its interpretation when it was entered into. General guidance to this effect is given, albeit only prospectively, in the Vienna Convention on the Law of Treaties (1980) (Cmnd 7964), articles 31(1) and 32.

    81. In an ideal world the Convention should be accorded the same meaning by all who are party to it. So case law provides a further potential source of evidence. Careful consideration needs to be given to the reasoning of courts of other jurisdictions which have been called upon to deal with the point at issue, particularly those which are of high standing. Considerable weight should be given to an interpretation which has received general acceptance in other jurisdictions. On the other hand a discriminating approach is required if the decisions conflict, or if there is no clear agreement between them.

    82. The question was raised whether the Convention must be approached on the basis that it is a document of the "always speaking" type described by Lord Steyn in R v Ireland [1998] AC 147, 158D-159A , whose meaning should be interpreted in the light of the current scientific evidence. I would answer that question in this way. The meaning that is to be given to the words used in the Convention must be the meaning which was to be attributed to them when the Convention was entered into in 1929. But it must always have been intended that the application of that meaning to the facts would depend on the evidence. The proper approach is to make use of the best current medical and scientific knowledge that is available. An important question therefore is whether, if the position has now been reached where a physical basis for a mental injury can be demonstrated objectively by the evidence, the framers of the Convention intended to include a condition of that kind within the scope of article 17 when they used the expression "bodily injury".

    83. It is necessary to start by considering the natural meaning of the words "any other bodily injury". At first sight the words "bodily injury" appear to envisage injury of a physical nature, similar in kind to that which is encompassed by the words "death or wounding." Both phrases may be thought to contemplate some kind of physical injury, some physiological damage to the structure of the body. While the phrase as a whole seems to have been intended to extend the scope of the article beyond "wounding", the adjective "bodily" appears to have been included to restrict or limit the extent of the word "injury". The use of this adjective suggests that not everything that might possibly be described as an injury to the passenger which he sustains on board the aircraft or in the course of embarking or disembarking was intended to be covered by article 17. The possibilities for experiencing emotions such as fear, fright or anxiety due to unexpected events during air travel must have been many in 1929 when the airline industry was in its infancy and the ability of unpressurised aircraft to avoid thunderstorms and other adverse weather conditions was limited. It is reasonable to think that they were familiar to the draftsmen of the Convention and to the signatories. But their use of language suggests not only that they did not intend article 17 to apply to emotional reactions of that kind but that they had in mind injuries which were manifestly physical.

    84. On the other hand the Lord President said in Mr King's case, 2001 SLT 126, 136B, that he saw no reason why the phrase "any other bodily injury" should be interpreted narrowly; rather that it should be interpreted as covering any injury whatever which can properly be regarded as affecting the body. So construed, it seemed to him that the phrase was capable of including psychological injury. Lord Reed recognised, at p 168H, that a wider construction, which would include any impairment to health, was possible. I agree that the words are sufficiently imprecise to permit that interpretation. On that approach it would be sufficient to show that the passenger was suffering from a specific mental illness which had diagnostic features that could be objectively demonstrated. But I think that, in its context, the phrase points to some injury to the body and not to a mental injury.

    85. The equivalent phrase in the French text is "de toute autre lésion corporelle". It follows the word " de mort, de blessure". "Blessure" is defined by the Le Petit Robert (1970 ed) as "lésion faite aux tissus vivants par une cause extérieure (pression, instrument tranchant ou contondant, arme à feu; chaleur)". "Lésion" has two meanings, one legal, the other medical. In the medical sense it is defined as "changement grave dans les caractères anatomiques et histologiques d'un organe sous l'infleunce d'une maladie, d'un accident". This points to an injury which affects the tissues of the body. "Corporelle" is defined in the relevant sense as "relatif au corps". This tends to confirm that the word "lésion" should be interpreted in this context as meaning a physical injury: see Dr Georgette Miller, Liability in International Air Transport (1977), pp 127-128. The phrase "lésion corporelle" in the French text seems to convey the same meaning as that which I would be inclined to draw from the English text. It appears to relate to those changes in the tissues of the body which are not covered by the word "blessure". The word "corporelle" directs attention to physical changes in the body rather than something that affects the mind.

    86. It was suggested that the phrase "bodily injury" was intended to extend to all harm that the passenger might sustain in his person, and to distinguish that kind of injury from an injury to his property or his business. The Lord President saw this explanation as more plausible: 2001 SLT 126, 144K-L. This argument is based on the proposition that article 17 must be read in the context of Chapter III of the Convention as a whole. Article 18 deals with damage to registered baggage. Article 19 deals with delay. The juxtaposition of these articles suggests that the intention was to include in article 17 everything that could be regarded as personal injury, while patrimonial loss consisting of damage to property and economic loss was left to be covered by the following articles.

    87. On the other hand I should have thought that, if this was indeed the intention, the word "lésion" on its own would have done and that the adjective "corporelle" would have been unnecessary. It would have been plain from its association with the word "blessure", and the reference to presence on board the aircraft and to the actions of embarking and disembarking, that the phrase "toute autre lésion" in article 17 was being used in the sense of an injury to the person of the passenger. The fact that the word "corporelle" was included appears to me to indicate that the word "lésion" was intended to describe a physical injury to the passenger as distinct from a mental injury.

    88. The Lord President recognised that the argument that the word "corporelle" was included to exclude mental as opposed to physical injury had an immediate appeal to the reader of the article: 2001 SLT 126, 135F. But he said that it would be superfluous to add the word "corporelle" to the word "lésion" if the latter word were to be given a meaning which contains within itself a reference to injury to the body. In his view therefore the word "lésion" in the relevant phrase must be taken to have been used in the wider legal sense, indicated in the dictionary by the words "dommage, préjudice, tort". The fact remains however that the adjective "corporelle" was chosen to qualify, and thus to restrict, the meaning of the word "lésion". Its ordinary meaning tends to exclude any other form of injury to the body other than a physical injury. And I would hold that the word "lésion" is used in its medical sense in article 17.

    89. I would be inclined therefore to read the phrase in a way that would confine its application to an injury to the skin, bones or other tissues of the body and exclude mental injury. But its meaning is not so clear that that one need look no further. I turn therefore to see what guidance is available in the other material.

    90. It must be said at once that, despite diligent research, no indication has been found that there was any discussion of the meaning to be attached to the phrase "any other bodily injury" by any of the delegates in the course of their negotiations at the Warsaw Conference. But there are some points about these negotiations that are of interest.

    91. As Lord Reed noted 2001 SLT 126, 166H-I, the text of the Convention went through previous drafts before the matter reached the stage of the final negotiations. They originated in the French Government's Draft Protocol which was circulated prior to the Paris Conference of 1925. It dealt with liability in extremely general terms, which focused solely on loss, damage and delay without discussing the nature of the accidents which would give rise to liability. Article 3 of that draft said simply:

"Le transporteur est responsable des pertes, avaries et retards qui résultent de ses fautes personelles et du vice propre de l'appareil."

    92. Following various modifications a more detailed provision was produced by the Comité International Technique d'Experts Juridiques Aériens ("CITEJA"), which introduced the three categories which were later to appear in the Convention in three separate articles:

"Le transporteur est responsable du dommage pendant le transport: a) en cas de mort, de blessures ou de toute autre lésion corporelle subie par un voyaguer; b) en cas de destruction, perte ou avarie de marchandises ou de bagages; c) en cas de retard subi par un voyageur, des marchandises ou des bagages."

As Lord Reed said at p 167A-B, it is perhaps easier in this context to appreciate that "lésion corporelle" might be intended to demarcate category (a) from the other two categories. But discussions which then followed during the Warsaw Conference, which are fully described in Lord Reed's speech, focussed on other matters. There was no discussion of what was meant by the expression "lésion corporelle". Nor was there a discussion of any issue which could be related in any way to the issue whether it was intended to include mental injury as well as physical injury.

    93. The Lord President said in Mr King's case that the court would have to conclude that the travaux préparatoires were of no assistance, as it was accepted by both sides that the question of psychological injury was not discussed in them. This was the primary conclusion that he too had reached and that he would apply: 2001 SLT 126, 137I-J. Lord Cameron of Lochbroom said, at p 147E-F, that it was accepted that during the course of the preparatory work which led to the determination of the final text there was no debate minuted which indicated, one way or the other, which meaning the delegates intended to give to the words "toute autre lésion corporelle" as they appeared in the final text. Lord Reed said, at p 167I-J, that in his opinion the travaux préparatoires did not support any theory that the signatories to the Warsaw Convention had a specific intention either to include or to exclude liability for psychiatric disorders. But he added that they did, on the other hand, confirm that articles 17, 18 and 19 should be read as a whole, concerned respectively with passengers, goods and delay.

    94. In Miss Morris's case Lord Phillips of Worth Matravers MR said the court in her case also agreed that the travaux préparatoires did not afford the type of clear indication that enabled them to be used to resolve the ambiguity: [2002] QB 100, 117, para 50. But he went on to say that the fact that no mention was made of mental injury was not without significance when one came to consider the overall purpose and scheme of the Convention.

    95. What inference is it proper to draw from the silence? It is clear that the importance of these articles was not overlooked. The president of the drafting committee, Mr Giannini, is recorded as having said when he drew attention to article 17, and in particular to the fact that it had been decided to divide what had originally appeared in a single article into three articles setting out the causes for liability for person, goods and baggage and delay:

"As our colleagues certainly recall, these are perhaps the most important articles of the Convention."

Nevertheless the draft article 17 was approved without discussion (II Conference International de Droit Privé Aérien, Minutes, trans Horner & Legraz (1975), pp 204-205).

    96. It seems reasonable to conclude from their silence that the delegates did not feel it necessary to discuss what was meant by the words "lésion corporelle". None of them appears to have anticipated that there would be any difficulty in applying the wording of article 17 to the facts in practice. Nor was it suggested that the formula which they were adopting would cause serious concern to the airlines or to their insurers. This suggests that the meaning which they gave to the words was the simplest and least troublesome meaning that they would ordinarily bear.

    97. As to the conclusion to be drawn from this material, I can at least say that I have not found anything so far which contradicts the impression which I have already formed on reading the Convention that the word "bodily" restricts the scope of the word "injury" to injuries which are manifestly physical and that it excludes emotional reactions and mental injury.

    98. The search is for the meaning which was to be attributed to the words "bodily injury" in 1929 when the Convention was entered into. So I think that we should be cautious in the use of subsequent practice. It would not be right to use subsequent practice to show that the meaning of these words has been changed. The proper way to change the meaning used in an international convention is by amending the convention. But it would be helpful if there was some evidence of a settled practice on the part of all the contracting states to show what they have always understood the words to mean. The point does not appear to have been raised in any reported case until a series of hijackings in the 1970s led to the making of claims in the United States for damages for emotional distress and mental injury. I shall come back later to the case law. At this stage however I wish to refer to some other material to which our attention has been drawn.

    99. The Lord President in Mr King's case drew support for a broader interpretation of the phrase from the fact that from the very beginning in the German-speaking countries the phrase was interpreted in a way that would have permitted recovery for mental injury: 2001 SLT 126, 136C. An official German language translation was prepared as the result of co-operation among officials of Germany, Austria and Switzerland and published in 1933. It interprets the phrase "blessure ou toute autre lésion corporelle" as "körperlich verletzt oder sonst gesundheitlich geschädigt". According to this translation, which uses the adjective "körperlich" to qualify the equivalent to the word "blessure" in the French text rather than the word "lésion", the article applies where the passenger is bodily wounded or otherwise damaged in his health.

    100. For the reasons which the Lord President described at p 136E-G, it is thought that one of those who had a hand in the translation was Dr Otto Riese. He was a member of the German delegation in all the negotiations from the Paris conference in 1925 onwards. He continued to take an interest in and write about aviation matters after the Warsaw conference. He had already indicated in an article which he wrote shortly after it ended that the phrase should be given a wide scope. In subsequent articles he adhered to the view that it applied to damage to health, although he acknowledged in his Luftrecht (1949), p 442 that this was a somewhat free translation of "lésion corporelle". In a French text which appeared in 1951, Riese and Lacour, Précis de droit Aérien, p 264, para 323, he said that the difference between the words "blessure" and "lésion corporelle" was that the Convention applied not only to wounding properly so called but also to all injury or the health of the passenger, "telles que les conséquences d'un choc psychique ou d'un mal de l'air". The Lord President said, at p 137C, that he did not think that either the official German translation or Dr Riese's views, which tended to confirm that the expression "lésion corporelle" was not used with the intention of excluding mental or physical injury, could be dismissed lightly.

    101. Commenting on this chapter of evidence, Lord Reed pointed out, at p 165L, that the official German translation was of no greater weight than the English translation and of less weight than the French text, and he noted Dr Riese's own description of the German version as a somewhat free translation of it. Lord Phillips of Worth Matravers MR made similar comments in Miss Morris's case: [2002] QB 100, 124, para 85. I also agree with him that caution must be applied to the views expressed subsequently by one delegate, for the reason which the Lord President himself gave, at p 137G, that delegates at a conference may not actually all share a common view on the point at issue.

    102. I can see that there is force in the point which the Lord President made, at pp 136L-137A that it would be surprising if Dr Riese had felt able consistently to advance his interpretation while conscious either that his fellow delegates at Warsaw had chosen the phrase with the settled intention of excluding mental injury or that his interpretation was wholly inconsistent with the meaning of these words. But the fact remains that, so far as the records show, the question whether the phrase extended to mental injury was simply not discussed at the Warsaw conference.

    103. CITEJA was dissolved in 1947, and its work was taken over by the International Civil Aviation Organisation ("ICAO"). The legal committee of the IACO met at Madrid in 1951 as part of the programme of work of revision of the Warsaw Convention which led to the Hague Conference in 1955. The French representative, M Garnault, proposed as an amendment to article 17 the substitution of the words "affection corporelle" for "lésion corporelle". His reason, as recorded in the minutes, was that the word "lésion" presupposed a rupture in the tissue and that he thought that there could be cases where, without there being a "lésion" in the tissues or a break in their continuity, there could be a mental illness.

    104. As Lord Reed observed at p 166C, the French representative's explanation adopted the medical, rather than the legal, meaning of the word "lésion" and it indicates that he did not think that the words "lésion corporelle" covered mental illness. The Lord President agreed, at p 140E, that M Garnault's views do indeed support the narrower interpretation of those words and that, as he would be a native speaker of the language, they must carry a certain weight. His view was that, taken as a whole, the legal debate at Madrid showed that by 1951 there was no settled view that the words in article 17 were to be interpreted as excluding psychological injury. I would agree with this assessment. But it does not persuade me that it was thought in 1929 by the signatories to the Warsaw Convention that the phrase "any other bodily injury" would include cases where the only injury sustained was a mental injury.

    105. The Montreal Convention on Air Carrier Liability was signed on 28 May 1999, but it is not yet in force. One of the issues which was discussed was whether mental injury could be included in any new, revised system of liability. A document produced under the aegis of the chairman included the following statement with regard to article 17:

"With reference to article 17, paragraph 1 of the Convention, the expression 'bodily injury' is included on the basis of the fact that in some states damages for mental injuries are recoverable under certain circumstances, that jurisprudence in this area is developing and that it is not intended to interfere with this development, having regard to jurisprudence in areas other than international carriage by air."

    106. The Lord President said, at p 141F, that the delegates' official view showed that they were quite prepared to countenance the situation, which had developed since the Warsaw Convention, where courts of different states deal with damages for psychological harm in different ways. He said that this seemed to him to be consistent with the intention of the delegates at the Warsaw Conference who did not seek to impose a uniform system of assessing damages, being content that the cap on the recoverable damages provided sufficient uniformity for carriers and their insurers. On the other hand, as Lord Reed pointed out at p 166F, the position in 1999 was that national courts, including in particular those in the United States where the point was of particular importance in view of the size of its air market, had interpreted the words "any other bodily injury" as excluding mental injury. This brings me to the relevant case law.

    107. The first reported case in the United States which was drawn our attention arose out of an incident when an aircraft en route from Tel Aviv to New York was hijacked, diverted to the Middle East and forced to land in the desert in Jordan: Rosman v Trans World Airlines Inc (1974) 34 NY 2d 385. The passengers were not struck or personally assaulted by any of the hijackers, but they claimed to have suffered severe psychic trauma during the ordeal. The question was whether they were entitled to compensation for their psychic trauma under the Convention. The case came before the Court of Appeals of New York, which held that the defendant was liable for the plaintiff's palpable, objective bodily injuries, including those caused by the psychic trauma of the hijacking, and for the damages flowing from these bodily injuries, but not for the trauma as such or for the non-bodily or behavioural manifestations of the trauma.

    108. Delivering the judgment of the majority, Rabin J recognised, at p 397, that the relationship between the mind and the body was a difficult issue. But he said that it was not one that the court had to decide:

"Rather, in seeking to apply the treaty's terms to the facts before us, we ask whether the treaty's use of the word 'bodily', in its ordinary meaning, can fairly be said to include 'mental'. We deal with the term as used in an international agreement written almost 50 years ago, a term which even today would have little significance in the treaty as an adjective modifying 'injury' except to import a distinction from 'mental'. In our view , therefore, the ordinary, natural meaning of 'bodily injury' as used in article 17 connotes palpable, conspicuous physical injury, and excludes mental injury with no observable 'bodily', as distinguished from 'behavioural', manifestations." (emphasis added)

At p 399, he dealt with the need for a causal connection between the injury and the accident, and with the question whether the expression "bodily injury" could extend to an injury such as a skin rash which was caused or aggravated by fright experienced during the incident. Disagreeing on this point with Stevens J, who dissented, he said:

"In our view, this connection can be established whether the bodily injury was caused by physical impact, by the physical circumstances of the confinement or by psychic trauma. If the accident - the hijacking- caused severe fright, which in turn manifested itself in some objective 'bodily injury', then we would conclude that the Convention's requirement of the causal connection is satisfied."

    109. I have quoted these passages from the majority judgment in Rosman, and added emphasis to the words "palpable", "conspicuous" and "observable", because they provide important guidance about the scope, on what it described at p 399 as a liberal interpretation, which can be given to the phrase "any other bodily injury". As Rabin J put it in a later passage on the same, the predicate of liability is to establish the "bodily injury". Once this is established, then the damages sustained as a result of the bodily injury can be compensated, including mental suffering. It treats as "bodily injury" physical injuries consequent upon shock or distress such as heart attacks, rashes and strokes and damage to the brain. But it excludes manifestations of fear, anxiety or other mental problems which cannot be described as physical injuries, such as sleeplessness, headaches and loss of weight: see Lord Reed's analysis in Mr King's case, 2001 SLT 126, 169C-D. It also excludes conditions affecting the mind which are not due to a palpable, conspicuous or observable physical injury

    110. The first case to come before the Supreme Court of the United States on this issue was Eastern Airlines Inc v Floyd (1991) 499 US 530. Passengers on an aircraft which narrowly avoided crashing during a flight between Miami and the Bahamas claimed damages solely for mental distress arising out of the accident. A conflict had arisen between the decision of the Court of Appeals in that case that the phrase "lésion corporelle" encompasses purely emotional distress and the decision in Rosman. The Supreme Court held that article 17 does not allow recovery for what it described as mental or psychic injuries unaccompanied by physical injury or physical manifestation of injury.

    111. The court used the French text to guide its analysis. It examined the meaning which was given to the words in the French dictionaries, observed that the phrase was not used in any French legislation in force in 1929, that there were no French court decisions in or before 1929 which explained its meaning or had allowed recovery for injury caused by fright or shock absent an incident in which someone sustained physical injury and that there were no materials in French treatises which indicated that "lésion corporelle" embraced psychic injury. At p 540, the court observed that the cause of action that evidently was possible in principle under French law in 1929 would not have been recognised in many other countries represented at the Warsaw Convention. At pp 544-545, the court said that the unavailability of compensation for purely psychic injury in many common and civil law countries at the time of the Warsaw Convention persuaded it that the signatories had no specific intent to include such a remedy in the Convention. At pp 551-552, it said:

"Even if we were to agree that allowing recovery for purely psychic injury is desirable as a policy goal, we cannot give effect to such policy without convincing evidence that the signatories' intent with respect to article 17 would allow such recovery. As discussed, neither the language, negotiating history, nor postenactment interpretations of article 17 clearly evidences such intent…

Moreover, we believe our construction of article 17 better accords with the Warsaw Convention's stated purpose of achieving uniformity of rules governing claims arising from international air transportation… We have no doubt that subjecting international air carriers to strict liability for purely mental distress would be controversial for most signatory countries. Our construction avoids this potential source of divergence."

    112. At pp 552-553, the court summarised the view that it had reached in terms which specifically left open the question whether compensation might be available where mental injury was accompanied by physical injury:

"We conclude that an air carrier cannot be held liable under article 17 when an accident has not caused a passenger to suffer death, physical injury, or physical manifestation of injury. Although article 17 renders air carriers liable for "damage sustained in the event of" ("dommage survenu en cas de") such injuries, we express no view as to whether passengers can recover for mental injuries that are accompanied by physical injuries. That issue is not presented here because respondents do not allege physical injury or physical manifestation of injury."

    113. The decision in Eastern Airlines Inc v Floyd was followed El Al Israel Airlines v Tseng (1999) 525 US 155 in which the Supreme Court confirmed, at p 166, note 9, the conclusion reached in Eastern Airlines Inc v Floyd that the Convention provides for compensation only when the passenger suffers "death, physical injury, or physical manifestation of injury." The passenger claimed that she had sustained psychic or psychosomatic injuries as a result of an intrusive body search but accepted that there had been no "bodily injury" as that term is used in the Convention. Her claim was that, as the Convention allowed no recovery, it did not preclude her from maintaining a separate action for damages. It was rejected by the Supreme Court. The same result would be reached in this jurisdiction: see Herd v Clyde Helicopters Ltd [1997] AC 534.

    114. Judicial thinking in the United States has been developed further in a series of cases in the Federal courts. In Jack v Trans World Airlines Inc (1994) 854 F Supp 654 there had been a crash and a fire following an aborted take-off. The court distinguished three different kinds of injury, at p 664:

"'Impact injuries' refer to the bodily injuries (such as bruises, lacerations and broken bones) that passengers suffer during an airplane accident (here, the aborted take-off and evacuation of the plane). 'Physical manifestations' refer to those bodily injuries or illnesses (such as skin rashes and heart attacks) that result from the distress one experiences during or after an accident. 'Emotional distress' refers to the psychic trauma that one experiences either during or after the accident."

It was not disputed that "impact injuries" and "physical manifestations" were within the scope of the expression "bodily injury". But the court noted, at p 665, that there were four possible approaches to emotional distress: (1) no recovery allowed for emotional distress; (2) recovery allowed for all distress, as long as bodily injury occurs; (3) emotional distress allowed as damages for bodily injury, but distress may include distress about the accident; and (4) only emotional distress flowing from the bodily injury is recoverable. At p 668, the court adopted the fourth approach. It held that the emotional distress recoverable was limited to the distress about the physical impact or manifestation, ie the bodily injury, but that recovery was not allowed for the distress about the accident itself.

    115. In Alvarez v American Airlines Inc (2000) 27 Avi 17,475 the court held that a passenger's claim for psychological and emotional injuries was properly dismissed because the physical manifestations of his post-traumatic stress disorder, including increased heart rate and elevated blood pressure, were directly related to his emotional trauma and not to a physical injury suffered during the evacuation from the aircraft. At p 17,478 the court said that, if these physical manifestations could support a recovery, a passenger frightened by air turbulence could recover on the basis of his increased heart rate which would convert the rule established in Eastern Airlines Inc v Floyd into a mere pleading formality. This decision supports the view that emotions and other mental disturbances which cannot be related to a palpable, conspicuous or observable physical injury are outside the scope of article 17. But it is not wholly consistent with what was said in Rosman v Trans World Airlines Inc (1974) 34 NY 2d 385 about the physical manifestations of a mental injury.

    116. In Weaver v Delta Airlines Inc (1999) 56 F Supp 2d 1190 a passenger brought an action to recover damages for chronic post-traumatic stress disorder which she attributed to terror which she felt during an emergency landing of the aircraft in which she was a passenger. But in this case there was an affidavit from the passenger's doctor which stated that the impact of the event included bio-chemical reactions which had physical impacts upon her brain and neurological system. In response, the airline did not present sufficient evidence to raise a factual issue on this point. The scientific research relied on by the passenger was criticised, but it was not explained and its expert had not examined the passenger. In the light of this evidence the court held that she had met her burden of showing that the emergency landing physically impacted on her brain and that there was no genuine issue on the facts. At p 1192, the court said:

"Weaver's action here is distinguishable from previous cases, because her claim is presented as a physical injury and she relies on recent scientific research explaining that post-traumatic stress disorder evidences actual trauma to brain cell structures. Weaver's post-traumatic stress disorder evidences an injury to her brain, and the only reasonable conclusion is that it is, in fact, a bodily injury… The legal question in this case is simply whether the Warsaw Convention allows recovery for this particular kind of bodily injury, ie, a brain injury (even with slight physical effects). The answer must be yes."

    117. In In re Air Crash at Little Rock, Arkansas, on June 1, 1999 (2000) 27 Avi 18,428 the passenger suffered from smoke inhalation, knee, leg and ankle injuries in a crash and her subsequent escape from the aircraft. But her primary claim was for emotional and psychological injuries consisting of serious chronic post-traumatic stress disorder and major depression. The judge observed, at p 18,431, that Federal courts had come to different conclusions as to the nexus required between physical and psychic injuries. He referred to Jack, Alvarez and Weaver and to Terrafranca v Virgin Atlantic Airways Ltd (1998) 151 F 3d 108 in which it was held that there could be no recovery unless the passenger demonstrated direct, concrete bodily injury as opposed to manifestations of fear or anxiety. He held, at p 18,433, that, once the threshold of liability had been crossed by proving a physical injury, then all damages available under the passenger's domicile are recoverable. As to whether post-traumatic stress disorder was a physical manifestation of injury, the judge noted that different results had been reached in several recent cases on this point. He held in favour of the passenger, regarding this as a matter of evidence. The passenger's doctor said in his evidence that research had shown brain dysfunction in people with PTSD and that, while it could not be said that everybody would demonstrate the same abnormality, the consistent theme was that PTSD was both a biological and a psychological syndrome. As the judge put it, at p 18,436:

"The Terrafranca and Jack decisions reflect our nascent understanding of the nature of mental illness. It is not clear that those courts were presented with the scientific literature to underpin a finding that PTSD is a biological as well as an emotional and psychological illness... Furthermore, the evidence presented at the trial, both in the form of expert testimony and exhibits, established that PTSD is a biological/physical as well as a psychological injury."

    118. The conclusion which I would draw from these cases is that there is a consistent body of judicial opinion in the United States that the phrase "any other bodily injury" does not permit compensation for an emotional reaction which has not been shown to be a manifestation of physical injury. Emotional distress is excluded, but so also is a psychiatric illness which affects the mind only and is not capable of being described as a physical injury. The decisions in the Weaver and Little Rock cases appear to have broken new ground. They have applied the phrase to a mental injury which was shown by expert medical evidence to have been accompanied by physical changes to the patient's brain cell structures. In those cases, in the light of the evidence, the passengers' post-traumatic stress disorder was held to be a physical as well as a psychological injury. This extends the meaning of the phrase "bodily injury" beyond that which was recognised in Rosman, because changes of that kind are plainly not palpable, conspicuous or observable.

    119. I have grouped these two countries together because they have produced the only judicial decisions outside the United States which were drawn to our attention.

    120. In Daddon v Air France (1984) 1 S & B Av R VII/141 the Supreme Court of Israel was concerned with claims by passengers in respect of the mental anguish which they had suffered while being held captive at Entebbe Airport by hijackers. It held that "bodily injury" in article 17 included mental anguish which was not accompanied by any physical injury. It reached this conclusion after recognising, at p 152, that the parties to the Convention apparently had no intention whatsoever in that regard, either because most states at that time had not recognised mental anguish as a cause of action or because the possibility of mental anguish which was not accompanied by physical injury had nor been contemplated. What the court sought to do was to develop the meaning of the word by judicial policy in the light of subsequent developments. This approach has received no support in the other jurisdictions. It has been criticised on the ground that it is impermissible to construe the Convention in the light of changes since 1929. The question is whether purely mental injury was within the intention of the draftsmen and the signatories: Eastern Airlines Inc v Floyd (1991) 499 US 530, 544.

    121. In Kotsambasis v Singapore Airlines Ltd (1997 ) 42 NSWLR 110, a decision of the Court of Appeal of New South Wales, the passenger claimed that she had suffered psychological injuries when shortly after take off a fire broke out in one of the engines of the aircraft which returned safely to the airport after fuel had been jettisoned. Meagher JA, with whose judgment Powell and Stein JJA agreed, said, at p 114, that the words "bodily injury" both in English and in French were ambiguous and that the ambiguity could only be resolved by looking at the intention of the contracting parties and adopting a purposive approach to the Convention:

"It is immediately apparent that the adjective 'bodily' is a word of qualification or limitation. It is a general principle of statutory interpretation, equally applicable to the interpretation of international agreements, that courts are not at liberty to consider any word as superfluous or insignificant - Commonwealth v Baume (1905) 2 CLR 405, 414 - and, more specifically, that effect is to be given to words of limitation: Randwick Municipal Council v Rutledge (1959) 102 CLR 54, 94. It is clear that the draftsmen of the Convention did not intend to impose absolute liability in respect of all forms of injury."

Following Eastern Airlines Inc v Floyd (1991) 499 US 530 he held, at p 115, that the term "bodily injury" was not intended to, and on a proper interpretation of the Convention does not, include purely psychological injury. He noted that the question whether recovery might be available where psychological injury was accompanied by physical injury was left open in Eastern Airlines and said that, in view of the findings of the trial judge, it was not necessary for him to decide that point. This decision supports the view that article 17 does not provide compensation for mental injury.

    122. I find myself in agreement with the conclusions which Lord Reed reached in Mr King's case and those of the Court of Appeal in Miss Morris's case. The Lord President said in Mr King's case that, on the approach suitable to an international Convention, the phrase "any other bodily injury" was apt to cover all injuries of any kind which affected a person in his body, and that these would include purely psychological injuries which an individual suffers, in a certain sense at least, in his body: 2001 SLT 126, 144L. I regret that I cannot agree with this view. It seems to me to give too little weight to the limiting effect of the word "bodily" and to the difficulties that would have been apparent in 1929 if it had been the intention to extend liability to that kind of injury. It is inconsistent with the decision in Eastern Airlines Inc v Floyd (1996) 499 US 530, which has been followed in New South Wales and which I too would wish to follow. It is also inconsistent with the decision in Rosman, which I regard as the other leading authority in this field.

    123. The evidence indicates that there was no uniform approach among the contracting states in 1929 to recovery in damages for mental injury. As the court noted in Eastern Airlines Inc v Floyd, at p 540, a cause of action for mental injury would not have been recognised in many of them. In this situation clear words would have been needed to ensure that a uniform approach to this problem was adopted in all jurisdictions. The absence of clear words allowing recovery for a purely mental injury is telling. The introduction of the word "corporelle" was not the way to make this point clear. The absence of discussion suggests that the contracting parties were content to use this word as a word of limitation to qualify the otherwise unrestricted scope of the word "lésion", and that they intended to restrict it to physical injury of a kind that was manifest and observable.

    124. Insurance practice has not been the subject of evidence. But it is commonplace for insurers to seek to exclude or restrict liability. One of the central points in the Convention was the surrendering of the opportunity to do this in the case of carriage to which the Convention applies. Part of the process of compromise lay in the choice of language for the articles in Part III, including article 17. This adds further weight to the point that the words "bodily" and "corporelle" were used as words of modification or restriction. It would have been a reasonable compromise to provide strict liability for physical injury of a kind that was manifest and observable, but not to extend this to an emotional reaction or any kind of mental injury. Practice since then suggests that this was what the compromise was understood to be, as there was a marked absence of claims of this character until the 1970s despite the many alarming or distressing events that may be encountered during air travel.

    125. The search is for the meaning which the words were understood to have when the Convention was entered into. It would be wrong to regard article 17 as limited by the state of medical and scientific knowledge that was current in the 1920s. There is no reason to think that the contracting parties intended that no account should be taken of developments in medical science in determining the question whether a passenger has sustained a bodily injury. I agree with my noble and learned friend Lord Mackay of Clashfern that, if the brain could be shown to have been injured and the other conditions for compensation under article 17 are satisfied, it would not be right to refuse compensation under the article on the ground only that in 1929 an injury of that kind would not have been capable of being demonstrated. Whether or not there is such an injury will always depend on the evidence. But the fact remains that a bodily injury is conceptually distinct from any injury which affects the mind: Mullany & Handford, Tort Liability for Psychiatric Damage (1993), p18.

    126. The words "palpable" and "conspicuous" were used by the court in Rosman to describe the kind of injury that they envisaged as falling within the expression "bodily injury". But the use of such strong language risks substituting a new test for that used in the Convention. The question whether there is any better way of explaining what the Convention meant is best left over for another occasion, as my noble and learned friend Lord Steyn has indicated. For the time being I would venture to suggest that one would expect an injury falling within the expression "bodily injury" to be capable of being demonstrated by an examination of the body of the passenger, making the best use of the most sophisticated means that are now available. The Weaver and Air Crash at Little Rock, Arkansas cases, as I understand them, did not proceed on that kind of evidence. There was no evidence in either case that the passengers had suffered an injury to the brain that was capable of being demonstrated by means of an examination of the body of the passenger. The argument was that the PTSD itself constituted a manifestation of a physical injury. In my opinion evidence of the kind that was available in those cases is not enough to satisfy, the test of showing that a psychiatric illness is or includes a "bodily injury" for the purposes of article 17.

    127. I believe that this approach is consistent with the mainstream views in Eastern Airlines v Floyd (1991) 449 US 530 and other leading cases in the United States whose interpretation of article 17, like my noble and learned friend, Lord Nicholls of Birkenhead, I too would wish that this House should follow if possible. I regret that I am unable to agree with the opinion of Lord Hobhouse as to the effect of these authorities.

    128. In any event there has been no attempt in either case to demonstrate that the passengers' depressive illnesses had a physical cause or origin. It would not be right to speculate as to whether, in view of the fact that a clinical diagnosis has been achieved in each case and that treatment has been prescribed for them, their illnesses could be shown to have had a physical origin which could be causally linked to the accident. That would have been a matter for proof. But there has been no offer to lead any such evidence. It is sufficient for the decision in these cases to say that it has not been shown that the mental injury which the passengers sustained falls within the scope of the expression "bodily injury".

    129. I would however follow the indications in Rosman v Trans World Airlines Inc (1974) 34 NY 2d 385 and Jack v Trans World Airlines Inc (1994) 854 F Supp 654 that compensation may be awarded to a passenger under article 17 for the physical manifestations of a mental injury. A peptic ulcer disorder involves the tissues of the body, and it is not difficult to see that it is a kind of bodily injury. The requirement of a causal link to the accident will be satisfied if it can be shown, as Mr King seeks to do, that the disorder was caused by a mental illness which was itself caused by the accident. Thus, while there is no general right to recover damages under article 17 for mental injury sustained by a passenger, damages for the physical manifestations of a mental injury will be recoverable. I would hold that the Lord Ordinary was right to allow Mr King a proof of his averments about the peptic ulcer disease.

    130. I would allow the appeal in Mr King's case and restore the interlocutor of the Lord Ordinary. I have considered whether an opportunity should be given in Miss Morris's case to re-open the question whether her depressive illness was linked to changes in her brain cell structures and could be said on this ground to have amounted to a physical injury. But I have come to the conclusion, in view of the way her case was expressly argued both in the Court of Appeal and in this House and because I do not think that even with the benefit of such evidence it could be shown that she has suffered a "bodily injury", that this would not be appropriate. I would dismiss her appeal.

LORD HOBHOUSE OF WOODBOROUGH

    131. These two appeals, the one from the Court of Session (Lord President Rodger and Lord Cameron of Lochbroom, Lord Reed dissenting, 2001 SLT 126) and the other from the English Court of Appeal (Lord Phillips MR, Peter Gibson and Latham LJJ, [2002] QB 100), raise a question of the correct meaning to be given to Article 17 of the Warsaw Convention 1929, and the 1955 Hague Amendment, as incorporated into the law of the United Kingdom by the Carriage by Air Act 1961. In one case the flight was internal to the United Kingdom and in the other was international. As more fully explained by my noble and learned friend Lord Hope of Craighead, nothing turns on this difference and no argument has been advanced based on making such a distinction. Point is lent to the relevant question by the fact that the two courts reached irreconcilable and opposing decisions as to the meaning of Article 17. The question is one upon which the highest courts of other countries have also expressed opinions using language which has been relied upon by one or more of the parties to these appeals to support their submissions and which will need to be carefully examined. The question is one of importance to the air transport industry world-wide since it affects the division of risk between passengers and carriers and the settlement of claims.

    132. The authoritative text of Article 17 of the Warsaw Convention is that in the French language: Article 36 of the Convention. The UK Act schedules both an English text and the French text. Section 1(2) of the Act provides: "if there is any inconsistency between the text in English ... and the text in French ..., the text in French shall prevail." The English text, which is the same as that used in other English speaking countries and that used for the purposes of later air carriage conventions,reads:

"The carrier is liable for damage sustained in the event of the death or wounding of a passenger or any other bodily injury suffered by a passenger, if the accident which caused the damage so sustained took place on board the aircraft or in the course of any of the operations of embarking or disembarking."

The relevant phrase in French reads "dommage survenu en cas de mort, de blessure ou de toute autre lésion corporelle subie par un voyageur" which has the same connotations as the English. Thus the word lésion has a similar breadth of meaning as the English word injury. Your Lordships were referred to various dictionaries, both French and English, but these did not disclose any inconsistency between the French and English texts and therefore did not raise any question under s.1(2).

    133. Before coming to the rival arguments it is necessary to give an example of a factual situation in order to illustrate the significance of the question raised before us. It is convenient to take the pleaded facts of the Morris case for this purpose. The passenger was a 15 year old girl travelling on an overnight international flight as an unaccompanied minor. She was placed in a seat next to a male passenger of a different nationality. During the night she woke up to find herself being indecently assaulted by her neighbour. Unsurprisingly she was deeply disturbed by this. The stewardess moved her to another part of the aeroplane where the girl completed the flight without further incident. The man had not physically injured the girl in any visible way. She had not been scratched or bruised. The seriousness of the assault lay in its indecency. It was a traumatic experience for the girl which had a lasting effect upon her. As a direct result of the incident she became clinically depressed. She was referred to a consultant psychiatrist, Dr Cooling, and he diagnosed her to be suffering from the illness "major depressive disorder, single episode", 296.2[x] in the "DSM-IV". "DSM-IV" is the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders published by the American Psychiatric Association. This manual contains the authoritative classification of psychiatric illnesses defining the signs and symptoms upon which the diagnosis of the various illnesses and disorders should be based. The Manual explains the terminology used and its purpose. The phrase 'mental disorder' is used for reasons of convenience although it "unfortunately implies a distinction between 'mental' and 'physical' disorders that is a reductionist anachronism of mind/body dualism". (p.xxx) The word 'disorder' is used to describe what is clinically significant and mark the boundary between normality and pathology.

"In DSM-IV, each of the mental disorders is conceptualized as a clinically significant behavioral or psychological syndrome or pattern that occurs in an individual and that is associated with present distress (e.g., a painful symptom) or disability (i.e., impairment in one or more important areas of functioning) or with a significantly increased risk of suffering death, pain, disability, or an important loss of freedom. In addition, this syndrome or pattern must not be merely an expectable and culturally sanctioned response to a particular event, for example, the death of a loved one, Whatever its original cause, it must currently be considered a manifestation of a behavioral, psychological, or biological dysfunction in the individual. Neither deviant behavior (e.g., political, religious or sexual) nor conflicts that are primarily between the individual and society are mental disorders unless the deviance or conflict is a symptom of a dysfunction in the individual, as described above." (p.xxxi)

    134. So, there had been an incident during the course of the flight which qualified as an accident for the purposes of Article 17. (See the judgment of the Court of Appeal [2002] QB 100 at 108-112, following Air France v Saks 470 US 392.) The incident is alleged to have caused the girl to suffer a recognised psychiatric illness displaying observable dysfunctional symptoms as a result of which she suffered damage. Is the psychiatric illness alleged a bodily injury? The passenger submits that it is: the carrier submits that cannot be.

    135. The arguments advanced by each side as to what was required for the passenger to have suffered a bodily injury ranged over a spectrum of alternative submissions. These submissions covered not only the 'injury' itself but also how it may have been caused. It was correctly accepted by the parties that the three elements in the first sentence of Article 17 must be causally linked. The use of the phrase "in the event of" ("en cas de") does not mean that any wound suffered by the passenger at any time between the commencement of embarkation and the completion of disembarkation will suffice to permit any claim for damages. Thus, there must have been an accident which has caused the death, the wounding or the bodily injury suffered by the passenger which caused the damage complained of. Some of the submissions sought to distinguish between different ways in which the relevant death, wounding or bodily injury, as the case might be, had been caused. Such submissions have been consistently rejected by the courts though they have undoubtedly affected the reasoning. The simplest form of the argument has been the 'impact' argument: there must have been an actual impact between some physical object and the passenger. Another form of the argument is that the causal link must have been through the medium of some physical object coming into contact with the passenger as, for example, the inhalation of smoke. These arguments, which were advanced again on these appeals by the carriers, were directed to excluding any injury other than one so caused and many psychiatric illnesses would not have been so caused. Ironically, in the case of Morris, there was such physical contact - the indecent touching. However a traumatic experience can be something the passenger merely witnesses such as passively witnessing a life-threatening emergency or a hijacking. The difficulty with the carriers' causation arguments is that they would seek to make an unwarranted distinction between death and injury. The death of a susceptible individual can be caused by being involved in a sufficiently traumatic incident, such as a very severe fright, and the same applies to non-fatal injuries (though not to woundings). The causal link is a link made through the passenger's senses. Once the 'causation' arguments are rejected, it is necessary to examine carefully the carriers' arguments upon bodily injury since they tend to equate bodily with 'physical' and then apply criteria which would have the effect of including or excluding injuries of the same character on the basis that they were fatal or non-fatal.

    136. Each side preferred the extreme form of its argument and in so doing had received support from one or other of the courts below. Thus, they variously argued as to the meaning of bodily injury (using the term "flight" as including embarkation and disembarkation and including the carriers' causation arguments) -

    137. The majority of the First Division preferred argument (7) largely basing themselves on an examination of the various texts and how the Article had been viewed, especially by a German jurist who had attended the Warsaw conference and some of the later conferences. They considered the history of the Convention and concluded that Article 17 should be given a wide and generous meaning: there is little difference between this and argument (8) as the Israeli case, Daddon v Air France (inf), demonstrates. The dissenting opinion of Lord Reed comes closest to argument (6). The Court of Appeal adopted argument (4), having broadly the same result as argument (2), viewing the phrase bodily injury as pointing a contrast with what they described as "mental injury". This argument was developed by the carriers before your Lordships as an argument that, in 1929, those attending the conference at Warsaw would have contemplated only a crude mind-body dualism and the words used implicitly demonstrate an actual intention to exclude anything which would in popular diction be referred to as of the mind. By contrast argument (6) can be developed by reference to basic medical knowledge of the brain and the way it works as an integral part of the whole body so as to show why psychiatric illnesses may come within the term bodily injury and therefore, provided the other conditions are satisfied, should be understood as falling within Article 17.

    138. The diversity of the judicial statements within the United Kingdom and to some extent elsewhere, all expressed in relation to the same two simple English words bodily injury as used in Article 17, disclose a legal confusion and even, it may be thought, a failure of method. The range of the arguments advanced lead to a similar conclusion. My Lords, I believe that it is possible to identify the explanations for this. I will attempt to classify them under 3 headings: first, the adoption of a subjective approach to the construction of the words and the misuse of national law concepts and rules; second, the assumptions made by the lawyers as to medical science and the change of environment between 1929 and the present day and its relevance; third, the misuse of language and the errors which result.

    139. But I will start by stating briefly what I consider to be the correct reading of these words and how it is supported not contradicted by the United States authorities.

Brief Overview:

    140. The composite expression bodily injury involves a combination of two elements. The word injury in the context of personal injury involves a condition which departs from the normal, which is not a mere transitory discomfort or inconvenience and which, whilst not permanent or incurable, has, in conjunction with its degree of seriousness, a sufficient duration. It includes a loss of function. A person who is concussed or who is in clinical shock or who is made deaf or blind is properly described as injured. (As to deafness, see, for example, Air France v Saks 470 US 392.) A condition which requires treatment to enable the person to return to the normal is typical of an injury though not essential; many injuries heal over time without intervention. Contracting an illness may amount to an injury depending upon the degree to which the illness departs from the normal. One would not normally describe a person who caught a cold as having suffered an injury but, on the other hand, one would certainly describe someone who contracted a serious disease or condition, say, 'aids' or hepatitis, as the result of the deliberate or negligent act of another as having suffered an injury.

    141. The word bodily is simpler. It means pertaining to the body. There must be an injury to the body. It is, as it must be, accepted that the brain, the central nervous system and the glands which secrete the hormones which enable the brain and the rest of the central nervous system to operate are all integral parts of the body just as much as are the toes, heart, stomach and liver. They are all susceptible to injury. The mechanisms by which they can be injured vary. An ingested poison might injure the stomach or liver. A lack of oxygen will injure the brain by causing the death of brain cells. An injury to the heart may be caused by a blow or by a traumatic experience or by over-exertion. In every case there is a cause, external to the organ in question, which produces a change in the structure or ability to function of the organ. If the change, either alone or in conjunction with changes in other organs, is properly described as an injury, it is a bodily injury. Since the body is a complex organism depending for its functioning and survival upon the interaction of a large number of parts, the injury may be subtle and a matter of inference not direct observation. The medical science of diagnosis exists to enable the appropriate inferences to be drawn from the observed evidence. Medicinal treatments (as with drugs) are prescribed on the basis that there is a physical condition which can be reversed or alleviated by physical means.

    142. Bodily injury does not import visibility nor palpability nor externality. Its use in Article 17 in addition to the use of the words "death" and "wounding" ("blessure") and the inclusion of the word "any" confirm this. Take an incident which ruptures a spleen or causes some other internal injury. The doctor infers that the injury has been caused from other signs and symptoms and ultimately confirms it by an invasive (surgery) or non-invasive (ultrascan or x-ray) procedure. An incident may damage someone's optic nerve in a location or manner which cannot be observed; the person may as a result have no sight in that eye; this will be a bodily injury even though there is no thing palpable, conspicuous or visible. The use of the word "palpable" in the discussion of the meaning of Article 17 typifies the dangers of the use of loose terminology which does not aid clarity but, rather, creates ambiguity. "Palpable" is a term which has a precise meaning - capable of being felt by touch - which is its medical meaning. But it also has a metaphorical usage - readily perceived by the senses or the mind. It will be readily understood that the precise meaning is too restrictive and cannot be justified as a gloss upon the simple phrase bodily injury. The metaphorical usage likewise cannot be justified and is impractical and unprincipled. Is the judge or arbitrator to say: "Having heard the evidence I am satisfied that the passenger suffered an injury but I cannot say that I readily perceived it"? The obvious attraction of using words such as "palpable" is that they give an illusion of clarity when in truth they enable the user to avoid clarity and simply serve to detract from the clarity of the primary terminology.

    143. Thus, bodily injury simply and unambiguously means a change in some part or parts of the body of the passenger which is sufficiently serious to be described as an injury. It does not include mere emotional upset such as fear, distress, grief or mental anguish (cf argument 8). A psychiatric illness may often be evidence of a bodily injury or the description of a condition which includes bodily injury. But the passenger must be prepared to prove this, not just prove a psychiatric illness without evidence of its significance for the existence of a bodily injury.

    144. Turning to the United States authority, the leading and subsequently influential case which first brought the relevant points to the fore was Rosman and Herman v TWA 34 NY 2d 385 decided in the Court of Appeals in New York in 1974 with the leading judgment given by Judge Rabin. The plaintiff passengers sought to recover damages from the carrier for "psychic" (sic) trauma suffered on board a hijacked aircraft. None of the passengers had been shot or struck or personally assaulted but they said that they had feared, as was natural in the circumstances, that their lives were in grave danger. They complained of being subjected to emotional stress (and physical deprivation in being kept without adequate food and water in excessively hot and arid conditions). The court rejected the claim for emotional stress but held that it would allow damages for any "palpable, objective bodily injuries, including those caused by the psychic trauma of the hijacking, and for the damages flowing from those bodily injuries, but not for the trauma as such or for the nonbodily or behavioral manifestations of that trauma". (p.400) They explained their reasoning in a passage (at pp. 399-400) which it is worth quoting in full:

"A faithful reading of the terms of the Convention leads, we believe, to the following conclusions. A claim for damages under article 17 arises 'in the event of ... bodily injury'. The claim must therefore be predicated upon some objective identifiable injury to the body. In addition, there must be some causal connection between the bodily injury and the 'accident'. In our view, this connection can be established whether the bodily injury was caused by physical impact, by the physical circumstances of the confinement or by psychic trauma. If the accident - the hijacking - caused severe fright, which in turn manifested itself in some objective 'bodily injury', then we would conclude that the Convention's requirement of the causal connection is satisfied. For example, if plaintiff Herman's skin rash was caused or aggravated by the fright she experienced on board the aircraft, then she should be compensated for the rash and the damages flowing from the rash. It follows that, if proved at trial, she should be compensated for her mental anguish, suffered as a result of the rash, since this anguish would have flowed from the 'bodily injury'."

    " … only the damages flowing from the 'bodily injury', whatever the causal link, are compensable. We are drawn to these conclusions by the clear import of article 17. Those terms, in their ordinary meaning, will not support plaintiffs' claim that psychic trauma alone, or even the psychic trauma which caused the bodily injury, is compensable under the Warsaw Convention." (p.400)

    145. I have underlined certain phrases in this quotation in order to illustrate certain of the points being made. Emotional distress is not compensable as such; it is bodily injury which is compensable. Therefore the passenger must prove that he has suffered a bodily injury. That is what Article 17 says. That is what Judge Rabin is saying when he says there must be some "objective identifiable injury"; the passenger must prove the bodily injury. (He later refers to the need for proof at the trial.) Thus, if an injury to the brain can be proved, we can substitute for the phrase 'skin rash' the words 'brain damage'. He is holding that proved brain damage and its sequelae would be compensable. Given causation, it comes down to a question of proof. Judge Rabin rejects the causation and impact tests. 'Psychic' trauma experienced through the senses can be the route by which the bodily injury can be caused by the 'accident'. The course upon which Judge Rabin is setting the United States courts is one which is fully consistent with the reading of article 17 which I have adopted in paragraph 143 above. I will return to the American authorities (including Rosman) later in this opinion.

    146. The historical context in which the Warsaw Convention of 1929 is set was not only the emergent business of commercial international carriage by air of passengers and goods but also the framework of existing international conventions and agreements covering carriage by sea which was at the time and was, for the following decades, to continue to be the dominant means of international transport. The Warsaw Convention substantially incorporated the same scheme and the same main features as those earlier conventions. Thus there is a recognition of the basic contractual relationship pursuant to which the transport is being performed and the requirement of the compulsory application of the internationally agreed terms to the contracts of carriage and the compulsory inclusion of statements to that effect into the contractual documentation. The substantive provisions provide a division of risk between the carrier on the one hand and the goods or baggage owner or passenger on the other. In relation to carriage by sea, the salient features of this division of risk are provisions which impose evidential burdens on the carrier, give him limited exceptions which he can rely on and which give him either an absolute or qualified right to limit his liability. In the Warsaw Convention effectively the same scheme is followed for the carriage of passengers, their baggage and commercial cargo. In each case an agreed code divides the risk between the carrier and his customer. It provides uniformity and certainty; conflicts of laws problems are avoided as far as possible; the incidents of where any accident or litigation may occur are sought to be removed as far as possible; the negotiation and acceptance of the 'five freedoms' of international air transport were facilitated.

    147. It follows from this that considerations of national or local law should not be allowed to intrude upon, let alone govern, any question of construction that may arise on the provisions for division of risk. As Lord Hope of Craighead said in Sidhu v British Airways Plc [1997] AC 430 at 453, "The code is intended to be uniform and to be exclusive also of any resort to the rules of domestic law." It is not right to attempt to construe the words of the Convention by reference to the rules of any domestic law, English, American, German or even French. We know that those rules were and are not all identical. The purpose of uniformity means that it is the duty of the national court to put to one side its views about its own law and other countries' laws. Quite apart from defeating uniformity, such a course can only lead to the complication of simple issues, the inadequately informed investigation of other legal systems and, most importantly, to uncertainty. In few areas can this be more deleterious than in relation to the historical treatment by various legal systems, including our own, of the topic of so called 'nervous shock'. Nor can it be acceptable, as was urged upon your Lordships by the carriers, to seek to find the 'lowest common denominator' of the delegates' national laws in 1929 and adopt that. Whilst it is important to have regard to the international consensus upon the understanding of the provisions of international conventions and hence to what the courts in other jurisdictions have had to say about the provision in question, the relevant point for decision always remains: what do the actual words used mean? (Stag Line Ltd v Foscolo, Mango & Co Ltd [1932] AC 328, the Hague Rules; James Buchanan & Co Ltd v Babco Forwarding & Shipping (UK) Ltd [1978] AC 141, CMR, Fothergill v Monarch Airlines Ltd [1981] AC 251, Amended Warsaw; Sidhu v BA, sup, Warsaw.)

    148. Further things follow. First, the words used should receive an objective interpretation. It is equally mistaken to try and find out what the individual delegates thought they were agreeing to as it is to investigate the various domestic laws of the signatory countries. The investigation is equally liable to be based on incomplete or imperfect evidence. It may well be that different delegates may have had different beliefs. The views of one delegate, however distinguished, articulate and well-published, may not represent the views of others. The examination must be an objective one. The disciplined liberty to make use of travaux préparatoires regulates this. It is common ground that in the present case, to quote Lord Reed at 2001 SLT 126, 167, para 56,

"The travaux préparatoires do not support any theory that the signatories to the Warsaw Convention had a specific intention either to include or to exclude liability for psychiatric disorders."

Therefore, it is the unadorned language of the Article to which attention must be directed. It is again a descent into unprincipled subjectivism to use, as do the Court of Appeal (§§50 and 96) and others have done before them, the absence of travaux préparatoires as a tool of construction. Thus the Court of Appeal say: "We consider that it is highly significant that no mention was made of liability for mental injury [sic] in the course of the negotiations that resulted in the Warsaw Convention." This is reasoning which speculates about the subjective intentions of the delegates and is not directed to the objective autonomous meaning of the words used. Likewise it is erroneous, in the absence of cogent travaux, to infer that a particular interpretation of a provision is intended from the fact that on a later occasion the Convention was amended without making any change to the provision in question. All it shows is that on the later occasion the parties were content to leave the wording of the provision unaltered. The motives or beliefs of the individual delegates for so doing are irrelevant. Similarly, it is unprincipled to say as the Court of Appeal say also in paragraph 96 that it is "equally significant" that no claim was made for 'mental injury' until the 1970s. One can only ask how can it be relevant to the question of the construction of the Convention. In fact there are simple explanations arising from the development of mass passenger air transport, the outbreak of terrorist hijacking and probably developments in United States domestic law.

    149. Secondly it is also mistaken to interpret a convention such as the Warsaw Convention, or the various amended versions of it, as if they were intended to be historical documents frozen in time. They are intended to provide an enduring uniform code which will govern contractual and, where relevant, delictual relationships not just for a finite time but for the future as the transactions to which they apply are entered into. The contracts into which the code is to be incorporated are to be made and will be performed at dates in the future, maybe long into the future. Notionally to relate them back to a supposed state of affairs existing in 1929 is not only wrong but wholly impractical. It leads to the complication and confusion to which I have already referred. It is also destructive of uniformity since when a convention has later been amended, logic would require that one starts the clock again and asks what was in the minds of the delegates at the later conference. It is not necessary in order to understand this point to have regard to the principles applicable to 'always speaking' constitutional documents. The principle is more simple. Words have a meaning which does not change but the application of those words to the decision of any question depends on the facts and circumstances of the case in which that question arises. It is the facts and circumstances of the cases that change, not the meaning of the contractual words.

    150. Thirdly, the code involves a division of risk. It strikes a balance. It is wrong to construe it as a code designed to advantage one interest or the other, the carrier or the customer. Like any code of this character, it contains familiar types of provision assisting one interest balanced against others assisting the other interest. For the passenger, a simple criterion of causation by an 'accident' is adopted but counterbalanced by strong provisions enabling the carrier to limit his liability. These are not like exemption clauses to be construed against one or other party. The phrase bodily injury simply enters the picture as part of the balanced code regardless of which interest it is thought to assist.

    151. I have elaborated the question of the approach to the construction of Article 17 because, in my view, all the courts below were drawn into an essentially mistaken consideration of material which was nationally based, historical, and subjective. It has resulted in an over-complication of the relevant question. It has necessitated a display of the most impressive scholarship and research but has led to the wrong questions being asked and the wrong answers given.

    152. It is a feature of the brain and the central nervous system that the manner in which it works has inevitably limited the knowledge and understanding of how the organ functions and its pathology. It is an interrelated and interdependent system. Only limited knowledge could be gained from the examination of the anatomy of a corpse. In the living subject, the scope for intrusive investigation was very limited. In the popular perception, this has led to the perpetuation of theories of psychic disembodiment long after they have any scientific justification. There now exist techniques for investigating the functioning of the living brain and the central nervous system together with the roles played by neurotransmitters, hormones and electrical impulses. Physical changes can be scanned and observed using sophisticated instruments and the alterations in the normal chemistry of the brain can now be detected by sophisticated sampling techniques. What was previously invisible can now be made visible. These developments have two relevant results. It can now be shown by valid scientific techniques that certain psychiatric symptoms correspond to physical changes in the brain. Psychiatry (the science of mental illness) has been able to develop a more reliable classification and aetiology enabling better diagnoses to be made and more reliable opinions to be given as to the probable causation of observed disorders. What these developments have changed is not the phenomena nor the meaning of the language used in the Article but the ability to adduce evidence relevant to the factual issues raised by the Article.

    153. The judgments of Lord Philip and Lord Reed in the Court of Session both show an awareness of these developments. The judgment of the Master of the Rolls in the Court of Appeal, §§35-40, in a section headed "The distinction between physical injury and mental injury", from which it is necessary to quote certain passages in full since they are fundamental to the thinking of the Court of Appeal, says:

"35  This appeal has proceeded on the premise that there is a distinction between physical injury and mental injury; that physical injury involves damage or adverse change to the structure of the body, whereas mental illness adversely affects the well-being of the mind without organic change to the body.

This was undoubtedly the general belief in the 1920s,

when the terms of the Warsaw Convention were negotiated. ......."

Then, having quoted from two speeches in the House of Lords in two tort cases and two English textbooks on English tort law, the judgment continues -

"40  These passages lead to the reflection that it is possible that every mental illness may, in time, be shown to be accompanied by and consequent upon some change to the physical structure of the body, so that mental illness can properly be described as a type of physical injury. That stage has not yet been reached, however, and this appeal must be approached on the premise that mental illness and physical injury are distinguishable and that the claimant, as she accepts, suffered no physical injury."

    154. The Master of the Rolls says, referring to the time when "mental illness can properly be described as a type of physical injury", that "that stage has not yet been reached". This is in truth a statement about medical science. It is contentious and needs to be made good by qualified expert evidence. Evidence to the contrary was successfully adduced in the American case of Weaver v Delta Airlines Inc (1999)56 Fed Supp 2d 1190. Before Lord Philip, in the Outer House, where there were three actions before the court, the unsuccessful argument was that the cases should proceed to proof so that the pursuers could call just such evidence: see 1999 SLT 919, 922, 925. The argument apparently failed for want of particularity in the pleading: ib p.926. The argument was not pursued on the solitary appeal by King to the First Division of the Inner House, where any reliance upon Weaver v Delta Airlines was disclaimed: see p.129 of the Lord President's judgment. So far as United Kingdom judges are concerned there has long been an appreciation that given mental functions correspond to and depend upon a given physical state of the body. In the Outer House, Lord Philip gave citations going back to the 19th century: pp.925-926 and there are others which he could have cited: for example, McLoughlin v O'Brian [1983] 1 AC 410 at 418 and R v Burstow [1998] AC 147 These citations and a simple perusal of such basic but authoritative textbooks such as 'The Shorter Oxford Textbook of Psychiatry', 4th ed Oxford 2001, written by Professor Gelder and two other professors of psychiatry at Oxford University, show that there is respectable medical support for the view that, for example, a major depressive disorder is the expression of physical changes in the brain and its hormonal chemistry. Such physical changes are capable of amounting to an injury and, if they do, they are on any ordinary usage of language bodily injuries. The passenger needs to prove by qualified expert evidence that these changes have occurred, that they were caused by the accident and that they have led to the psychiatric condition which constitutes the damage complained of. If he discharges this burden of proof he has satisfied the requirements of the Article and the explanation given of it in Rosman and Weaver. For the court to shut out the passenger from his opportunity to prove his case is both wrong in law and involves assuming the role of prejudging an expert issue which it is not for the court to assume.

    155. I recognise that under the procedural rules which governed the proceedings before him in the Outer House, Lord Philip may have been justified in refusing to allow the cases before him to go to proof and it seems that the subsequent history of the actions bears out that the parties apparently shared this view. As regards the proceedings in the case of Morris, the situation was less clear cut and I will have to revert to this point later in this Opinion.

    156. But two mistakes in the reasoning of the courts below need to be emphasised. The scientific developments have not changed the meaning of the Article. The meaning of the phrase bodily injury has not changed. The criterion or test remains the same. All that has changed is the ability of certain plaintiffs to bring their cases within it. The historical argument of saying one must look to some meaning existing in 1929 is, quite apart from the other objections to the argument, simply misconceived. Air transport has been transformed since 1929, mainly because of the invention of the jet engine and its use in civilian aircraft. Land-based planes are no longer confined to a range of 500 miles or less nor civilian aircraft to daytime flight and an operating height of 3 or 4 thousand feet. A delegate to the Warsaw conference asked what he thought about jet engines, helicopters or pressurised aircraft flying at 30,000 feet or more carrying hundreds of passengers or the risk of hijacking would probably have merely looked bemused as he no doubt would have if asked about monamine neurotransmitters, noradrenaline or changes in the size of the hippocampus. But this ignorance or lack of imagination does not affect the meaning of the words which for better or worse they decided in 1929 to use in the text of the Convention.

The Use of Language:

    157. The third point leads on from the second. Lawyers are not expert scientists nor are they always familiar with the accurate use of medical language. This is no criticism of them unless they have had the assistance of expert evidence. For example, they cannot be expected to know the character of the DSM classification unless they have at one time or another been informed of it or have had the professional opportunity to inform themselves. But, on the other hand, it should be possible to detect the shortcomings of arguments which, citing from the works of Descartes (as did counsel for the carriers in these appeals), invoke primitive and patently unscientific dualist theories. Similarly it is not sound to use such expressions as "mental injury", an expression which forms the corner-stone of the reasoning of the Court of Appeal. The adjective mental means relating to the mind. The mind is a metaphysical concept associated with the self-consciousness of human beings. The word can be used in a descriptive but not a substantive sense. One can have a mental illness or disorder just as one can have a respiratory illness or disorder. But one cannot usefully refer to having a mental injury. Little more helpful is the expression psychic injury (though it may have the meaning psychological injury in the United States where it seems to have its main currency). However none of these expressions, nor the confusion of psychiatry with psychology, assist to answer the question what is a bodily injury. At the best they identify that class of complaints which represent normal responses to external experiences, fright, grief, anguish, feeling sad or unhappy, feeling depressed but not clinically, and so on. These emotional upsets are real and unpleasant but they are not bodily injuries. Indeed they will not cross the threshold into what are properly described as injuries at all. It is this class of unpleasant experiences which, without more, do not suffice to ground a recovery under Article 17 and which, as I read them, the American authorities are seeking to exclude.

    158. The reasoning of the Court of Appeal discloses a striking combination of the various features to which I have referred, most fundamentally, the errors which arise from the use of a phrase which does not occur in the Convention, 'mental injury', and which is devoid of actual meaning. This phrase is then used to create a false antithesis with a phrase which is used, bodily injury, bolstered by a supposed investigation of what delegates to the 1929 conference would have actually had in mind when they agreed to the Convention and the proposition that the absence of travaux préparatoires to support a particular reading should be treated as if there were such travaux préparatoires. The conclusion then arrived at is that although a passenger may be able to adduce evidence to prove that she has suffered an injury which comes within the ordinary meaning of bodily injury, she must be precluded from doing so. The quotations which I have made from the judgment of the Master of the Rolls also illustrate a use of language which inevitably tends to confuse the reasoning. 'Mental injury' is used interchangeably with 'mental illness' and then 'mental illness' is contrasted with 'physical injury' and 'physical' is apparently being used as a substitute for 'bodily'. The picture is not made any clearer by the fact that in intervening passages the phrases "damage to the mind" and "mental disorders" are used.

    159. I therefore disagree with the reasoning and conclusion of the Court of Appeal. Whilst I agree with much of the language and reasoning of Lord Philip and Lord Reed, I do not agree with them that the correct reading of the American cases is that they would exclude more than emotional upset.

The Overseas Authorities:

    160. As stated in the judgments of the courts below, the relevant question under Article 17 has been discussed by jurists in civilian jurisdictions but without disclosing a consistent view independent of domestic law considerations. For present purposes I believe that the most useful assistance can be found in the reported decisions of courts. However here as well there are difficulties since the decision may have been given on the basis of assumed facts, imprecisely defined, or without a clear statement in the report of the actual facts found.

    161. The first cases to which your lordships were referred were ones which arose from the hijacking in 1970 of a TWA flight from Frankfurt to New York. It gave rise to the Rosman case (1974), already referred to, and to a case in the US District Court Burnett v Trans World Airlines Inc (1973) 12 Avi 18405 . It has been suggested that the fact that there do not seem to have been earlier cases under this part of Article 17 shows the artificiality of a construction which had never previously been thought to be arguable. The reports do not bear this out. The new factor appears to have been the occurrence of very unpleasant experiences without any actual impact but causing temporary distress and mental anguish. Until the first spate of terrorist hijackings occurred there was little scope for any such incident or claim. The metamorphosis of the carriage of passengers by air into a mass market serving consumers from a wide range of backgrounds also led to novel situations giving rise to the possibility to argue that damage had been caused by an 'accident'.

    162. The Burnett case arrived at similar conclusions to those of the New York Court of Appeals in Rosman. Thus the claim for 'mental anguish' not caused by any bodily injury was held to be irrecoverable: "Therefore, plaintiffs may recover in this action for any such emotional anxiety that they can demonstrate resulted from a bodily injury suffered as a consequence of the hijacking." (p.18,409) The 'contact' argument was rejected - "that any bodily injury sustained must be the result of physical contact between the body and another object" - described as "a sterile interpretation": "Brief reflection allows one to pose many instances in which a bodily injury may result without any physical contact whatsoever." (ib.) The decision therefore is fully consistent with Rosman though the discussion of the Convention is different, drawing on various French materials.

    163. In Rosman, an argument based upon French domestic law was specifically rejected:  

"It does not follow from the fact that the treaty is written in French that in interpreting it we are forever chained to French law, either as it existed when the treaty was written or in its present state of development." (p.394)

    In the same vein they rejected the subjective approach to construction: "Nor do we see any useful purpose to be served by [an investigation of French law] to search out the intent of the drafters of Warsaw." (p 395) The conclusions in Rosman can be summarised: (a) Mere 'psychic' trauma, mental anguish and emotional distress are not bodily injury and do not on their own give a basis of recovery under Article 17. But, (b) such trauma etc may provide the causal link between the accident and the bodily injury. (c) If a bodily injury is proved, all the consequences of that bodily injury should be taken into account in assessing the passenger's damage, including any consequential mental suffering. This, if I may respectfully say so, is a coherent scheme which reflects the language of the Article and is one with which I agree.

    164. The penultimate paragraph of the judgment of Judge Rabin draws a distinction between the "non-bodily or behavioral manifestations" of the psychic trauma of the accident and the "palpable, objective bodily injuries" which may have been caused by that trauma. The former does not, without more, amount to a bodily injury whereas the latter clearly does. I do not read the word "palpable" (the use of which I have already commented upon in paragraph 142 above) as adding anything more than emphasis and a requirement for proof to the phrase "objective bodily injuries". However, the carriers relied before your Lordships in this connection upon some of the language which Judge Rabin had used earlier on at p.397:

"Rather, in seeking to apply the treaty's terms to the facts before us, we ask whether the treaty's use of the word 'bodily', in its ordinary meaning, can fairly be said to include 'mental'. We deal with the term as used in an international agreement written almost 50 years ago, a term which even today would have little significance in the treaty as an adjective modifying 'injury' except to import a distinction from 'mental'. In our view, therefore, the ordinary, natural meaning of 'bodily injury' as used in article 17 connotes palpable, conspicuous, physical injury and excludes mental injury with no observable bodily, as distinguished from 'behavioral', manifestations."

    This passage is expressly governed by the context of the facts of the Rosman case and the limited question which it raised. It is not evident what is the justification for the inclusion of the words "palpable" and "conspicuous" or what is gained by substituting "physical" for "bodily". The passage is open to the same criticisms as those which I have made of the reasoning of the Court of Appeal in the Morris case. I do not accept that this passage of the judgment is meant to contradict what Judge Rabin says more cogently on pages 399 and 400 which I have quoted in paragraph 144 above.

    165. Next in time came a decision of the Israel Supreme Court, Daddon v Air France, (1984) 1 S&B Av R VII/141. It arose out of the Entebbe hijacking of 1976. The passengers were eventually rescued. The passengers, after nearly six years' delay, sued the airline for "damages in respect of the mental anguish caused to them and for the pain and suffering sustained by them due to what was involved" in the hijacking. The court held (in anticipation of Sidhu v BA and El Al v Tseng) that the Warsaw Convention (with the Guatemala protocol) was exhaustive of the passengers' rights. The court concluded that, in view of the objects of the Convention and its background, Article 17 should be interpreted "in the widest possible way so that it would be possible in pursuance thereof to award compensation also for pure mental anguish". (p.153) This interpretation stands alone internationally and is not one I would agree with. It deprives the words bodily injury of their ordinary meaning and equates them with a legal category, personal injury.

    166. The leading decision of the Supreme Court of the United States, Eastern Airlines Inc v Floyd (1991) 499 US 530, arose from an incident during a flight to the Bahamas from Miami. Shortly after take-off, all 3 of the plane's engines failed. It lost height and a crash landing into the sea seemed inevitable. However at the last moment one of the engines restarted and the plane was able to land safely at Miami. No one was hurt but the passengers claimed "damages solely for mental distress arising out of the incident". The Court held that "Article 17 does not allow recovery for purely mental injuries". (pp.533-4) The decision of the case therefore poses no problems for any save those who would support the extreme Daddon view, as had the 11th Circuit Court of Appeals. The subject matter of the case naturally governed the language used in the judgment delivered by Justice Marshall - "when a passenger has suffered only a mental or psychic injury" - "recovery for purely psychic injuries". (pp.536-7) In a footnote to p.540, he seems to have equated these expressions with "emotional distress" and "grief". No doubt because of the judgment below and the arguments presented to them, the judgment spends some time considering the French text and the jurist's commentaries before coming to the conclusion (p.542) that the French legal sources were ambiguous and did not help to resolve the relevant question. After a historical review, taking in subsequent conferences and conventions and the Daddon case, the Court declined to agree with the Israel Supreme Court and rejected its wide interpretation of Article 17. Reverting to the expression "mental distress", they consider how recovery in respect of it has been treated in America and elsewhere and adds that

"We have no doubt that subjecting international carriers to strict liability for purely mental distress would be controversial for most signatory countries." (p.552)"

    167. Justice Marshall concluded -

"....an air carrier cannot be held liable under Article 17 when an accident has not caused a passenger to suffer death, physical injury or physical manifestation of injury. ..... we express no view as to whether passengers can recover for mental injuries that are accompanied by physical injuries. That issue is not presented here because [the passengers] do not allege physical injury or physical manifestation of injury." (p.552-553)

    He also declined to decide the question whether the Convention was exhaustive of the parties' rights and liabilities. Thus the Supreme Court strictly confined their decision to the facts of the case before them and the question whether to follow Daddon and reverse to that extent Rosman. They rejected Daddon and preferred Rosman although they chose to leave open some of the points Rosman had decided.

    168. Jack v Trans World Airlines Inc (1994) 854 F.Supp.654, District Judge Caulfield: This case arose from an incident at John F Kennedy airport New York when a plane was taking off for a flight to San Francisco. The take-off had to be aborted and the plane caught fire. During the evacuation of the plane some passengers suffered minor injuries and many passengers claimed to have suffered "emotional distress" as a result of the incident which was undoubtedly traumatic. The carrier sought to strike out the claims for "emotional distress", in respect of those who had not been injured at all, on the ground that the claims were barred by Floyd and, in respect of those who had been injured during the evacuation, on the ground that emotional distress damages are not allowed. The carrier also complained that the evidence relied upon by the passengers in connection with their emotional distress claims was self-contradictory and sham, a submission which the judge accepted. (p.661) The judge did not accept the evidence of the psychiatrist relied upon by the passengers describing it as speculative. (pp.663-4) The judge however went on to discuss the right to recover for emotional distress following on an injury. His conclusion was to adopt a similar analysis to that in Rosman: see paragraph 163 above. Damages for emotional distress are not recoverable as such but may be recovered if consequent upon a bodily injury. The judge, notwithstanding his views about the passengers' evidence, declined to strike out the claims by passengers who were alleging "physical manifestations" of their emotional distress or had suffered impact injuries. (p.668) This case therefore provides an example of the application of the decisions in Floyd and Rosman but no more. The judge at p.664 sought to define his terms; his definitions do not assist as they raise as many questions as they purport to answer.

    169. American Airlines v Georgeopoulos (1996 and 1998) [1998] NSWSC 463, Supreme Court of Appeal, New South Wales: This case first came before the Court of Appeal on agreed findings which included "that the [passenger] claims damages for nervous shock or mental suffering and no other basis". The carriage was agreed to be governed by the Warsaw Convention. The appeal therefore raised the bare question whether the Australian courts should follow Daddon or Floyd. The court preferred Floyd. They held the question of liability could not be decided without further findings of fact; absent such findings it was impossible to decide whether the passenger suffered bodily injury within the meaning of the Article. To quote the headnote -

"'Nervous shock' as a condition or a cause of a condition for which a defendant may be liable in negligence describes a non-impact injury which may or may not give rise to a body tissue alteration. Assuming shock occurs, the question is whether that shock caused injury and the nature of that injury. These evidentiary findings are essential to any conclusion whether the injury complained of was a 'bodily injury' for the purposes of Article 17".

This decision is therefore in line with the view I have expressed. The case was remitted.

    170. Before Georgeopoulos returned to the Court of Appeal for the second time, the case of Kotsambasis v Singapore Airlines Ltd (1997) 42 NSWLR 110, had come before it. The plaintiff in this case had been a passenger on a flight from Athens to Sydney. Shortly after take-off one of the engines caught fire and, after dumping fuel, the plane returned safely to Athens. The trial judge accepted that the plaintiff had been severely frightened and anxious during the incident. Her onward flight was diverted via Paris and she strained her back carrying her heavy luggage. Meagher JA said:

"I am of the opinion that the term "bodily injury" was not intended to, and on a proper interpretation of the Convention does not, include purely psychological injury. I note that the decision in Eastern Airlines Inc v Floyd specifically left open the possibility that recovery might be available where psychological injury was accompanied by physical injury." (p.115)

    On the judge's findings she had not suffered any injury which had anything to do with the fright and anxiety of which she complained nor was it in the course of any embarkation or disembarkation. Powell JA agreed. Stein JA, also agreeing, followed the first Georgeopoulos case and expressly preferred Floyd to Daddon. He said that Floyd "held that there could be no recovery for 'psychic' injury unaccompanied by physical injury" and continued -

"However where mental anguish follows and is caused by physical injury, recovery for both injuries is covered. I would agree with this latter statement. Moreover, if the psychological injury is proven to be a species of bodily injury, then it would constitute 'bodily injury' within the article." (p.121)

He later summarised his view as being that bodily injury in Article 17 was not intended to include "purely psychological injury". (p.122, his emphasis.)

    171. On the return of Georgeopoulos for the second time to the Court of Appeal, the court included Sheller JA who had given the leading judgment on the first appeal and Meagher JA who had presided on Kotsambasis. The additional facts found by the magistrate were not helpful to the passengers. He found that all they had suffered was a "mild post traumatic stress disorder" which had not resulted from "any physical or bodily injury" to either passenger nor "any structural alteration to body tissues or alteration in the function of an organ or neurochemical change or any other form of damage to tissues or organs". Sheller JA again delivering the agreed leading judgment expressly adopted and approved what Stein JA had said at p.121 of Kotsambasis and accordingly dismissed the passengers' appeals.

    172. The importance of these Australian cases is that they follow the lead given by the US cases; they reject the proposition that there can be a recovery for emotional distress as such; they accept that an accident can cause a 'psychological' (sic, clearly meaning psychiatric) injury which may be proved to be a bodily injury; they do not criticise the criteria "structural alteration in bodily tissues", "alteration in the function of an organ or neurochemical change" and "any other form of damage to tissues or organs"; they see no inconsistency between this and what was said in Floyd.

    173. It is not necessary to refer to El Al v Tseng (1999) 525 US 576, US Supreme Court, the case of the over-intrusive security search, as it was deliberately argued as a case which fell outside of Article 17. There probably had been no accident. It was doubtful whether the material events occurred within the period covered by Article 17. The plaintiff's evidence was that she was "emotionally traumatised and disturbed", "really sick and very upset" and that is all. (p.587) The decision in Floyd was not questioned. The Plaintiff's argument was that she should be allowed to bring a tort claim. The question for the decision of the Supreme Court was that left open in Saks and Floyd.

"Her case presents a question of the Convention's exclusivity: When the Convention allows no recovery for the episode-in-suit, does it correspondingly preclude the passenger from maintaining an action for damages under another source of law, in this case, New York law?" (Justice Ginsberg at p.585)

The Supreme Court following, among other authorities, Sidhu, held that the Convention was exhaustive. It adds nothing presently material to Rosman and Floyd.

    174. Of the later US cases, only a few need be referred to since most are merely examples of the application of Floyd to facts which were not materially different from those on which Floyd was based. Where those facts have included some injury, the injury had no causative or other relevance to the emotional distress. (eg, Alvarez v American Airlines (1999) 27 Avi 17,214 and (2000) 27 Avi 17,475) In two cases Terrafranca v Virgin Atlantic Airways Ltd (1998) 151 F.Supp 3d 108, Third Circuit Court of Appeals, and Carey v United Airlines (2001) 28 Avi 15,408, Ninth Circuit Court of Appeals, the passenger sought to satisfy the criteria in Floyd by saying that the accident had caused emotional distress and the emotional distress had caused physical symptoms like, in Terrafranca, loss of weight and, in Carey, sleeplessness, nausea, perspiration etc. These consequences it was argued amounted to "physical manifestations" for the purpose of Article 17 so as to bring what would otherwise be mere emotional stress within the terms of that Article. It will be appreciated at once that I myself would not accept that argument. What the passenger has to prove is a bodily injury, not something less but with physical manifestations. The argument was based upon the language used in Floyd: but the phrase used there is "physical injury or physical manifestation of injury". (pp.552-553) If it is simply emotional stress which is causing the person to lose weight, no injury, bodily or otherwise, is proved. For the argument to succeed the plaintiff must prove either that the manifestation proves that there is or has been an underlying bodily injury or that the manifestation itself is a bodily injury. As Rosman shows, provided that causation by the accident can also be proved, in the former instance the plaintiff can recover damages for the underlying bodily injury and its consequences and in the latter for the bodily injury but not what preceded it. What I have said corresponds to the reasoning of the Court of Appeals in Terrafranca: see page 15,777 where Rosman is cited. In Carey (at p.15,414) Circuit Judge Nelson followed Terrafranca, saying,

"The Third Circuit concluded ... that there was no support for the argument that the plaintiff's physical manifestations of her emotional injury satisfied the 'bodily injury' requirement. Because the plaintiff could not 'demonstrate direct, concrete, bodily injury as opposed to mere manifestation of fear or anxiety', the court held that she did not satisfy the conditions for liability under Article 17 and thus could not recover for her emotional distress. For reasons similar to those articulated by the Third Circuit in Terrafranca, we hold that physical manifestations of emotional and mental distress do not satisfy the 'bodily injury' requirement in Article 17."

    Carey came after Weaver (inf) and the Court of Appeals referred to it in footnote 47 to its Opinion without expressing either approval or disapproval. The value of Terrafranca and Carey is that they implicitly approve Rosman and confirm the primacy of the simple bodily injury criterion, not any gloss or paraphrase of it.

    175. There have been two cases where the passenger claimants took upon themselves and discharged the burden of proof contemplated by Rosman and the Australian cases. The first is Weaver v Delta Airlines Inc (1999) 56 F. Supp 2d 1190, Chief Judge Shanstrom. The plaintiff was a passenger on a flight from London to Billings. Mechanical problems necessitated an emergency landing at Dayton Ohio. She claimed that during the landing she had been terrified and had subsequently had to be medically treated for emotional and physical problems attributable to her flight experiences with a diagnosis of post traumatic stress disorder. The carrier sought to strike out her claim, relying on Floyd. The judge refused the application. The plaintiff filed affidavits providing uncontradicted expert evidence that "extreme stress causes actual physical brain damage, ie, physical destruction or atrophy of portions of the hippocampus of the brain". "The impact upon [the plaintiff] of the events which occurred on that flight was extreme and included biochemical reactions which had physical impacts upon her brain and neurological system". She had thus presented evidence of "physical injury" and was entitled to say that "her diagnosed post-traumatic stress disorder arose from the physical changes in her brain brought on during the extreme stress of the emergency landing". (p.1191) The judge therefore distinguished her case from "Floyd and its progeny": she was relying upon "an injury to her brain, and the only reasonable conclusion is that it is, in fact, a bodily injury". (p.1192) It is hard to see any basis for disagreeing with the conclusion that, if the passenger can prove that his or her brain was damaged as a result of the accident, the passenger has suffered a bodily injury.

    176. On 1 June 1999 a plane crashed on the runway at Little Rock Arkansas. One of the passengers, Mrs Lloyd was wounded in the crash suffering multiple though not, in themselves, life-threatening injuries. She was trapped for a while in the burning plane and claimed that she also had suffered debilitating emotional and psychological injuries which led to chronic post traumatic stress disorder and major depression. The judge found that a sufficient causal connection had been established between her physical injuries and her 'mental injuries' (sic). (p.433) This sufficed for her to succeed in full on her claim. However he went on to consider briefly what would have been the case if that causal connection had not been established: could she prove that her post traumatic stress disorder was a physical manifestation of injury? He followed Weaver, considered the expert evidence and found sufficient cause to find that her chronic post traumatic stress disorder caused biological, physical changes in brain function. (p.435) Evidence is necessary to underpin the finding that post traumatic stress disorder is "a biological as well as an emotional and psychological illness". (p.436)

    177. District Judge Knapp reached similar conclusions Turturro v Continental Airlines (2001) 27 Avi 18,414. In a careful review of the earlier decisions, he concluded that the Supreme Court's line drawing in Floyd had worked well. (p.419) He also referred (p.420) to Weaver and Little Rock as cases where plaintiffs had successfully tendered evidence that extreme stress such as that experienced in near death situations or being taken hostage may actually change brain cell structure and cause a specific area of the brain to atrophy. He recognised that it will not be in every case of post traumatic stress disorder that such biological abnormalities will have been caused and it will be necessary to prove by objective evidence that in the given case brain damage has ensued. He therefore endorsed the course which the US cases were following. He decided the case before him in favour of the carrier because the plaintiff had neither pleaded nor shown she could prove a case of bodily injury.

    178. To summarise, there is clear support in the overseas authorities for the view I have expressed. I believe that these authorities show an international acceptability for giving a natural meaning to the words bodily injury without imposing any artificial or restrictive gloss upon them. This is the direction in which the jurisdiction which has to deal with the greatest volume of air passenger litigation, the United States, has moved. It is supported by the decisions of the Courts of New South Wales. If your Lordships adopt the same view as I have of the effect of Article 17, you will be promoting international uniformity rather than creating a risk that the law of the United Kingdom will be at odds with that elsewhere. The decision in Weaver was, given the evidence there before the court, wholly unexceptionable. It is in line with the US law and the natural meaning of the words bodily injury. I accept that the judgment of the Court of Appeal in Morris is inconsistent with it. I prefer the reasoning in Weaver.

The Arguments:

    179. What I have already said has covered most of the points raised in the argument of the appeals before us. The extreme arguments in favour of the passengers cannot be accepted. (Nos.7 and 8) There is insufficient reason for reading the words of the first sentence of Article 17 as embracing everything which might be described as "personal injury" as opposed to damage to property or mere financial loss. The actual language of Article 17 must be given effect to. Floyd is to be preferred to Daddon. I disagree with the majority of the First Division on this point.

    180. Similarly, the various forms of the causation argument relied on by the carriers are likewise not acceptable. (Nos.1, 2 and 3) Argument No.6 which was also relied upon by the carriers conforms most closely to what I consider to be the correct understanding of the use of the words bodily injury but because it represents a paraphrase of them can give rise to error and the adoption of argument No.4.

    181. An argument advanced by the carriers in support of their submissions was that if the Court of Appeal view was not accepted, there would be a flood of claims of non-specific injury which it would be expensive for carriers and their insurers to investigate and would be difficult for them to rebut. This argument would have some force if the Daddon approach was to be adopted and mere emotional stress and upset were to be recoverable. But the argument has no force if the passenger has to prove by the appropriate expert evidence some actual bodily injury in the sense which I have identified and is illustrated by Weaver. The passenger will have to be able to make a specific claim of injury consequent upon the accident and support it by convincing expert evidence. Any forensic difficulties for the carriers should be no greater than with, say, an alleged painful spinal injury.

    182. But there is a further version of the carriers' argument to which I should refer: it accepts that the criterion bodily injury provides a threshold which the passenger must cross (by the adducing the necessary evidence) but then argues that, even though the passenger has crossed the threshold, he is to be precluded from recovering for any 'mental injury'. This argument seeks to use as an excluding criterion an unscientific expression ('mental injury') which lacks definition and fails to provide the requisite certainty. An injury to the brain (or, indeed, some other part of the body) may cause a mental disorder, say some form of dementia or amnesia. Suppose that the injury to the brain is accompanied by a fractured skull or merely was the result of some concussion or a minor stress-induced stroke. In a proportion of such cases the sole external evidence of the injury may be the mental disorder. The reasoning that seeks to exclude such a consequence from the scope of Article 17 encounters problems and it is these which the argument fails adequately to address. What is the distinction between the fractured skull case and the concussion or minor stroke case unless it is to be some causal distinction (the causation argument) which is rightly rejected as unsound? On the other hand, if all mental disorder cases are to be excluded, contrary to the US cases, what is the distinction between a brain injury which causes paralysis of a limb and one which causes amnesia? The reality is that the use of the term 'mental injury' serves no useful defining purpose and is inconsistent with the US cases. It has to be analysed down into some form of the causation argument or it has to become a restriction on the damages that can be recovered even though the threshold has been crossed. Given an injury to the brain which was caused by the accident and which in its turn causes a mental disorder, the only way in which compensation for the mental disorder can be excluded is by introducing an arbitrary limit upon the consequences of the physical injury - the brain injury - which can be taken into account. If of course there has been no injury to the brain, there will have been on these hypotheses no bodily injury and therefore no right to claim. The bodily injury criterion as I have defined it provides a criterion whereas 'mental injury' does not.

    183. On this appeal to the Inner House, the majority adopted the Daddon view with which I have said I disagree. The pursuer specifically disclaimed any reliance upon the Weaver decision and, by necessary implication, declined to make any allegation of having suffered a bodily injury involving his brain. Therefore I agree that the appeal to the Inner House should have failed and the order of the Lord Ordinary allowed to stand.

The Morris Case:

     184. The Court of Appeal applied the wrong test and their reading of Article 17 cannot stand. However, where that leaves the plaintiff's case is another matter. She had adequately pleaded a prima facie case of bodily injury in relation to the psychiatric element in her case. But she had also pleaded and accepted that the assault did not cause her any physical injury. On the view adopted by the Court of Appeal, she was bound to fail. According to the Law Reports, the Court was not referred to Weaver even though that case had been referred to in King to which they had been referred. In any event, it is clear that no argument can have been advanced based upon a physical brain injury. In these circumstances, would it be right to leave Miss Morris without any possibility of recovery? The appeal to this House was brought in order to try to obtain the adoption of the Daddon interpretation of the Article and that was the manner in which her appeal was argued. Her printed case submitted that she should not be left without a remedy for her depressive disorder "merely because she is unable to identify any ... physiological damage to her head". Therefore I do not consider that it would be appropriate to remit her case to the County Court when she has implicitly stated that she will not be able to prove any bodily injury. I accordingly agree that her appeal should be dismissed.