- This litigation has raised once again, albeit in a fairly new context, the question as to the scope that should be given to the term "accident" in article 17 of the Warsaw Convention 1929. The purpose of the Convention, as my noble and learned friend Lord Steyn has observed, was "to bring some order to a fragmented international aviation system by a partial harmonisation of the existing law" (para 27) of his opinion repeating his comment in Morris v KLM Royal Dutch Airlines  AC 628, 635). It is common ground, therefore, that it is important that the courts of the respective signatory states should try to adopt a uniform interpretation of the Convention.
- Article 17 set out the basis on which a carrier would be liable for bodily injury sustained by a passenger during the flight (or while embarking or disembarking from the aircraft). The Convention underwent amendment at the Hague in 1955 but none of the amendments affect the point now in issue. The Convention as amended was incorporated into domestic law by the Carriage of Air Act 1961 and is set out in the 1st Schedule to that Act.
- Article 17 provided as follows:-
"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."
It is to be noticed that the conditions for the imposition of liability on the carrier do not include any element of fault or blameworthiness or failure to observe a proper standard of care on the part of the carrier. The requirements of liability are, first, that the passenger has suffered a bodily injury (a requirement that gives rise to questions about psychiatric injury which, happily, do not need to be addressed in the present case), second, that the bodily injury has been caused by an "accident" and, third, that the accident took place on board the aircraft (or in the process of embarkation or disembarkation). The omission from these conditions of any requirement of negligence on the part of the carrier is double-edged so far as an injured passenger is concerned. It is to the passenger's advantage that negligence on the part of the carrier needs to be neither alleged nor proved. It is to the passenger's disadvantage, however, that even clear causative negligence on the part of the carrier will not entitle the passenger to a remedy if the article 17 conditions cannot be satisfied. It has been authoritatively established that if a remedy for the injury is not available under the Convention, it is not available at all (see Sidhu v British Airways plc  AC 431 and El Al Israel Airlines Ltd v Tsui Yuan Tseng 525 US 155 (1999)).
- Nonetheless, negligence, or the absence of it, on the part of the carrier may play a part under article 20. Article 20 provided 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."
Plainly, a carrier who had been negligent could not qualify for an Article 20 defence.
- Contributory fault of the injured person, too, may afford a defence to the carrier. Article 21 provided that:
"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."
I should mention also article 22 which imposes monetary limits on the extent of the liability of the carrier for article 17 damage. It has often been observed that the provisions were designed to strike a balance between the interests of passengers and the interests of the airlines. (See e.g. Morris v KLM Royal Dutch Airlines  AC 628 per Lord Hope of Craighead at para. 66).
- The use in article 17 of the term "accident" is to be contrasted with the choice of a different term in article 18. Article 18 imposes liability on carriers for damage to baggage or cargo:
"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." (emphasis added)
- The use of the term "accident" in article 17 but the term "occurrence" in article 18 must be significant. Both terms impart the idea that something or other has happened. But "occurrence" is entirely general in its natural meaning. It permits no distinction to be drawn between different types of happening. "Accident" on the other hand must have been intended to denote an occurrence of a particular quality, an occurrence having particular characteristics. In the many decided cases in which the issue was whether the occurrence in question constituted an "accident" for article 17 purposes, the judges have had to ask themselves whether the occurrence possessed the necessary quality or characteristics to qualify as an "accident". It is evident that it was never, or should never have been, enough for there to have been an occurrence that caused the damage. For article 17 liability the occurrence had to have the characteristics of an "accident".
- The "fairly new context" to which I referred in paragraph 1 is the growing belief that sitting in a cramped position for many hours may give rise to the formation of small blood clots in the deep veins of the legs, hence, deep vein thrombosis (DVT). For obvious reasons DVT in relation to air travel is sometimes referred to as "economy class syndrome". If blood clots break away from the wall of the vein to which they are attached and are carried along with the flow of the blood, serious complications may ensue, including a stroke, a heart attack or, in the worst cases, death. As long ago as 1976 a passenger claimed to have suffered from deep vein thrombosis brought about by a long distance flight. But the Supreme Court of New York held that the condition had not been caused by an article 17 "accident" and the passenger's damages claim failed (see Scherer v Pan American World Airways Inc. (1976) 387 N.Y.S.2d 580). Since then DVT as an alleged consequence of economy class long distance flights has had increasing attention from the media, from members of the medical professions and from airlines themselves. The hand baggage of most passengers on long distance flights will these days include a pair of tight stockings which, if worn, are believed to provide some protection against the onset of DVT. Many air travellers, however, have suffered the serious consequences of DVT to which I have referred and they, or persons on their behalf, believe the onset of the DVT to have been attributable to the nature of the seating provided for them on the aircraft.
- Litigation has therefore taken place in a number of jurisdictions raising the question whether the onset of DVT in the course of and caused by air travel can constitute an "accident" for article 17 purposes. Litigation by a number of claimants has been commenced in this jurisdiction. They allege that they, or their deceased relatives, have suffered DVT caused by their flying on an aircraft operated by one or other of the respondents, each of which is a commercial airline. They say that the DVT was caused by an "accident" within the meaning of article 17. A Group Litigation Order in respect of these actions was made on 8 March (and amended on 10 March) 2002 and on 10 May 2002 an order was made by the Senior Master directing that a preliminary issue of law be determined on the basis of an agreed "specimen matrix" of fact.
- The preliminary issue was:
"whether the onset of deep vein thrombosis (DVT) sustained during the course of, or arising out of, international carriage by air, whether as result of an act and/or omission of the carrier or otherwise, is capable, in principle, of being 'an accident' causing bodily injury within the meaning of article 17 of the Warsaw Convention."
The specimen matrix is important. Paragraph 1 describes DVT; paragraph 2 describes characteristics of individuals that may predispose them to the onset of DVT. These details are not important for present purposes. Paragraph 3 simply says that the "Defendant is a commercial air carrier". I should, however, set out in full the remaining paragraphs:
"4. The Claimant was carried by air by the Defendant, for reward, on an international flight to which the provisions of the Warsaw Convention applied. The flight was characterized by the following features:
(1) the layout of the passenger cabin, the seating space available to each passenger and the type of passenger seat installed on the aircraft performing the flight were all in accordance with the Defendant's usual standard for an aircraft of that type flying on the route in question;
(2) the flight was operated in accordance with all of the Defendant's usual procedures and practices;
(3) nothing happened in the course of the flight which adversely affected the performance or flight characteristics of the aircraft;
(4) throughout the flight all of the aircraft's seating and all of its systems affecting the passenger cabin environment were in their normal working order;
(5) the aircraft complied with, and the flight was carried out in accordance with, all applicable aviation regulations; and,
(6) whether or not the above operation of the aircraft minimized and/or eliminated the risk of passengers suffering from DVT, the Defendant took no further or other steps to minimize and/or eliminate such risk.
5. The Claimant asserts that there is a causal link between air travel and the onset of DVT. The defendant denies the existence of any such link. For the purpose of this specimen matrix alone it is assumed that the Claimant suffered from a symptomatic DVT caused by the flight.
6. The Claimant asserts that the Defendant knew, or ought to have known, prior to the flight that by virtue of carriage by air passengers would be at an increased risk of suffering DVT over and above that incurred in everyday life. The Defendant denies these assertions. For the purpose of this specimen matrix alone it is assumed that the Claimant's assertions are correct.
7. The Defendant did not give the Claimant any warning as to the risk assumed in paragraph 6 above, or any advice as to how to minimize any such risk, at any time before or during the flight."
These are the assumed facts on the basis which the preliminary question was to be answered.
- Counsel for the parties were in broad agreement as to the principles of interpretation of article 17 that should be applied. The important principles for present purposes are that:
(1) the starting point is to consider the natural meaning of the language of article 17, with the French text prevailing in case of any inconsistency with the English text (fortunately there is no relevant inconsistency);
(2) the Convention should be considered as a whole and given a purposive interpretation;
(3) the language of the Convention should not be interpreted by reference to domestic law principles or domestic rules of interpretation; and
(4) assistance can and should be sought from relevant decisions of the courts of other Convention countries, but the weight to be given to them will depend upon the standing of the court concerned and the quality of the analysis.
I would add to these that the balance struck by the Convention between the interests of passengers and the interest of airlines ought not to be distorted by a judicial approach to interpretation in a particular case designed to reflect the merits of that case. The point was well put by Justice Scalia in his dissenting opinion in Husain v Olympic Airways (2004) 124 S.Ct. 1221, 1234 (an opinion with which Justice O'Connor concurred):
"A legal construction is not fallacious merely because it has harsh results. The Convention denies a remedy, even when outrageous conduct and grievous injury have occurred, unless there has been an 'accident'. Whatever that term means, it certainly does not equate to 'outrageous conduct that causes grievous injury'. It is a mistake to assume that the Convention must provide relief whenever traditional tort law would do so. To the contrary, a principal object of the Convention was to promote the growth of the fledgling airline industry by limiting the circumstances under which passengers could sue. .. Unless there has been an accident there is no liability, whether the claim is trivial ….. or cries out for redress".
- I think at this point a word of caution about the process of interpretation is in order. It is not the function of any court in any of the Convention countries to try to produce in language different from that used in the Convention a comprehensive formulation of the conditions which will lead to article 17 liability, or of any of those conditions. The language of the Convention itself must always be the starting point. The function of the court is to apply that language to the facts of the case in issue. In order to do so and to explain its decision, and to provide a guide to other courts that may subsequently be faced with similar facts, the court may well need to try to express in its own language the idea inherent in the language used in the Convention. So a judge faced with deciding whether particular facts do or do not constitute an article 17 accident will often describe in his or her own language the characteristics that an event or happening must have in order to qualify as an article 17 accident. But a judicial formulation of the characteristics of an article 17 accident should not, in my opinion, ever be treated as a substitute for the language used in the Convention. It should be treated for what it is, namely, an exposition of the reasons for the decision reached and a guide to the application of the Convention language to facts of a type similar to those of the case in question.
- In Fenton v J.Thorley & Co  AC 443, 453 Lord Lindley said that an "accident" was not a technical legal term with a clearly defined meaning. He went on:
"Speaking generally … an accident means any unintended and unexpected occurrence which produces hurt or loss. But it is often used to denote any unintended or unexpected loss or hurt apart from its cause; and if the cause is not known the loss or hurt itself would certainly be called an accident."
The issue in Fenton v J.Thorley & Co was whether a workman had suffered an "injury by accident" for the purposes of the Workman's Compensation Act 1897. The context was very different from that of the Warsaw Convention. Nonetheless Lord Lindley's understanding of what "accident" meant in the ordinary use of the English language is a useful starting point and was treated as that by Justice O'Connor in giving the highly influential opinon of the U.S. Supreme Court in Air France v Saks 470 US392 (1985) (see at p.398).
- But set in the context of the Convention some adjustments to Lord Lindley's definition of "accident" are clearly necessary. First, for Convention purposes the "loss or hurt" cannot itself be the "accident". Article 17 distinguishes between the bodily injury on the one hand and the "accident" which was the cause of the bodily injury on the other. It is the cause of the injury that must constitute the "accident". Second, it is important to bear in mind that the "unintended and unexpected" quality of the happening in question must mean "unintended and unexpected" from the viewpoint of the victim of the accident. It cannot be to the point that the happening was not unintended or unexpected by the perpetrator of it or by the person sought to be made responsible for its consequences. It is the injured passenger who must suffer the "accident" and it is from his perspective that the quality of the happening must be considered.
- In Air France v Saks 470 US 392 Justice O'Connor, having cited Lord Lindley's definition of an "accident", having surveyed the French case-law and dictionaries and having reviewed the history of the negotiations that had led to the Convention, concluded at p.405:
"… liability under Article 17 of the Warsaw Convention arises only if a passenger's injury is caused by an unexpected or unusual event or happening that is external to the passenger. This definition should be flexibly applied after assessment of all the circumstances surrounding a passenger's injuries."
- Two features of this conclusion, both of which can be regarded as prompted by the facts of the Saks case, warrant some comment. The facts were these. The claimant had suffered damage to and become permanently deaf in an ear as a result of pressurisation changes while the aircraft was descending to land. But the pressurisation system of the aircraft had operated in an entirely normal manner. The airline contended that the normal operation of a normal pressurisation system could not qualify as an article 17 accident. Justice O'Connor agreed. The damage to the claimant's ear could not itself constitute the article 17 accident. The cause of the damage had to be the accident. But the pressurisation system had operated normally. This is the factual background that led Justice O'Conner to formulate the requirement of an "unexpected or unusual event or happening that is external to the passenger" (emphasis added). In doing so she noted that
"European legal scholars have generally construed the word 'accident' in Article 17 to require that the passenger's injury be caused by a sudden or unexpected event other than the normal operation of the plane" (p.404)
"American decisions …. while interpreting the term 'accident' broadly …. nevertheless refuse to extend the term to cover routine travel procedures that produce an injury due to the peculiar internal condition of a passenger" (p.404/5).
- Both the requirement that the causative event be unusual and that it be external to the passenger were prompted by the facts of the Saks case. Both requirements were emphasised by Justice O'Connor in passages at page 406 of her opinion:
"… when the injury indisputably results from the passenger's own internal reaction to the usual, normal and expected operation of the aircraft, it has not been caused by an accident and Article 17 of the Warsaw Convention cannot apply."
"Any injury is the product of a chain of causes, and we require only that the passenger be able to prove that some link in the chain was an unusual or unexpected event external to the passenger."
- O'Connor J's opinion in Saks has been widely followed both in the United States and in the courts of other signatory states. Both the standing of the court and the reasoning of the opinion justify that reliance. Moreover, as I have already observed, it is of importance that if possible a uniform interpretation of the Convention should be applied in all signatory states.
The DVT cases
- There have been attempts by claimants in several signatory states to establish article 17 liability for DVT brought about, or said to be brought about, by air travel. I have already referred to Scherer v Pan American Airways Inc 387 NYS 2d 580 in which a DVT article 17 claim was rejected by the Supreme Court of New York. In October 2001 a similar claim was rejected by the Frankfurt Main Regional Court (see Shawcross & Beaumont's Air Law, loose leaf edition para.702). Article 17 claims based on the airline's failure to warn passengers of the possibility of, and the need to take precautions against, DVT have been rejected in Australia, Canada, Germany and the United States. Some decisions to the contrary can be found but the bulk of the authority is firmly against the acceptance of article 17 DVT claims.
- The most important DVT authority is the recent decision of the High Court of Australia in Povey v Qantas Airways Ltd  HCA33. Gleeson CJ and Gummow, Hayne and Haddon JJ gave a judgment rejecting the argument that the onset of DVT brought about by the conditions of passenger travel on the flight could be said to have been caused by an article 17 accident. The onset of DVT, ie. the formation of the blood clots in the veins of the leg, could not be the requisite accident because that onset was the damage, or the first stage of the damage, complained of. And no other event or happening, no occurrence external to the passenger, could be pointed to. It was common ground that "nothing happened on board the aircraft which was in any respect out of the ordinary or unusual" and "the relevant flight conditions were not said to be unusual or unexpected in any respect (p.13). Kirby J and Callinan J gave concurring judgments.
- I should mention also the recent US Supreme Court decision in Olympic Airways v. Husain 124 S.Ct. 1221 (2004). This was not a DVT case. An asthma sufferer, to whom ambient cigarette smoke represented a considerable health risk, asked for and was allocated seats for himself and his wife in the non-smoking section of the aircraft. Once aboard the aircraft he realised that the allocated seats were only three rows away from the smoking section. He, or his wife, asked a flight attendant if he could move to a seat further away from the cigarette smokers. The flight attendant said no other seats were available. This was untrue. During the flight the asthma sufferer became affected by ambient smoke from the smoking section and suffered a serious asthma attack from which he died. A claim was made by his widow under article 17. The issue was whether his death had been caused by an article 17 accident. But for the intervention, as a link in the causal chain, of the exchanges between the asthma sufferer (or his wife) and the flight attendant the case would have been difficult to distinguish from the DVT cases. The issue was whether that link made all the difference. The majority of the Supreme Court, concurring in an opinion delivered by Justice Thomas, held that it did. They accepted the argument that her failure to move the asthma sufferer to an available seat further away from the smoking section was a cause of the damage external to the passenger and, since it was contrary to normal airline industry standards, was therefore neither expected nor usual. They endorsed O'Connor J's Saks criteria but said that a plaintiff need only be able to prove that "some link in the [causation] chain was an unusual or unexpected event external to the passenger" (p.1227). They rejected the argument that the flight attendant's failure to act could not constitute an "event or happening". Dictionary definitions of "event" or "happening" were cited and relied on. The proposition that an article 17 accident could not be produced by mere inaction was rejected as a "fallacy" (p.1229). Justice Scalia gave a dissenting opinion with which Justice O'Connor, as to its important parts, expressed her agreement. The dissent was as to whether the inaction of the flight attendant could constitute an article 17 accident.
- There is no corresponding problem to be resolved in the present case and it is not necessary for your Lordships in this case to express any preference between the majority and dissenting opinions in Husain. I venture, however, to express my respectful disagreement with an approach to interpretation of the Convention that interprets not the language of the Convention but instead the language of the leading judgment interpreting the Convention. This approach tends, I believe, to distort the essential purpose of the judicial interpretation, namely, to consider what "accident" in article 17 means and whether the facts of the case in hand can constitute an article 17 accident.
- Of more importance for present purposes is that nothing in the Husain case casts doubt upon the two important requirements of an article 17 accident that were established in Saks and have been applied fairly consistently ever since, namely, that an event or happening which is no more than the normal operation of the aircraft in normal conditions cannot constitute an article 17 accident and, second, that the event or happening that has caused the damage of which complaint is made must be something external to the passenger.
- These two requirements appear to me to rule out article 17 recovery in DVT cases where no more can be said than that the cramped seating arrangements in the aircraft were a causative link in the onset of the DVT. The failure by an airline to warn its passengers of the danger of DVT and of the precautions that might be taken to guard against that danger does not, in my opinion, improve the case, at least where there is no established practice of airlines generally or of a defendant airline in particular to issue such warnings. How the case would look if there were such an established practice and if by an oversight the usual warnings were not given does not arise for consideration in the present case. The specimen matrix includes no such assumed facts.
- Accordingly for these reasons, as well as those of my noble and learned friends Lord Steyn and Lord Mance, I would dismiss this appeal.
- The question is whether the onset of deep vein thrombosis ("DVT") sustained during the course of, or arising out of, international carriage of goods by air can amount to an accident within the meaning of article 17 of the Warsaw Convention of 1929 as amended at The Hague in 1955 and incorporated in English law as schedule 1 to the Carriage by Air Act 1961. For the present purposes it is unnecessary to consider separately article 17(1) of the Montreal Convention 1999 (Carriage by Air Act 1961, schedule 1B as inserted by S.I. 2002/263).
- The purpose of the Warsaw Convention, following the precedent of the earlier Hague Rules governing carriage by sea, was to bring some order to a fragmented international aviation system by a partial harmonisation of the applicable law. The Warsaw Convention is an exclusive code of limited liability of carriers to passengers. On the other hand, it enables passengers to recover damages even though, in the absence of the Convention and the Act, they might have no cause of action which would entitle them to succeed. It follows from the scheme of the Convention, and indeed from its very nature as an international trade law convention, that the basic concepts it employs to achieve its purpose are autonomous concepts. For present purposes the compromise agreed at Warsaw involved the imposition of a form of strict liability on carriers in respect of accidents causing death, wounding or bodily injury to passengers in return for the limitations of liability expressed in the Warsaw Convention.
- Specifically, article 17 reflects an important part of the bargain struck at Warsaw:
"The carrier is liable for damage sustained in the event of the death or wounding of a passenger or any 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."
Since in the event of inconsistency between the French and English texts the former prevails (see section 1 of the 1961 Act), the French version cannot be ignored:
"Le transporteur est responsible 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."
In the context of the meaning of "accident", in relation to the issue of DVT, nobody has suggested that there is a relevant difference between the two texts.
- The exclusive remedy under article 17 is dependent on the fulfilment of three indispensable requirements. First, that a passenger sustained death, wounding or other bodily injury. Secondly, that an accident took place on board the aircraft or in the course of any of the operations of embarking or disembarking. Thirdly, that the death, wounding or bodily injury was caused by the accident. This is the immediate context in which the meaning of "accident" must be determined. This setting makes clear that accident is used with reference to the cause rather than the injury itself.
- The issues come before the House on assumed facts. Those assumptions include a normal and unremarkable flight. It specifically must be assumed that there was no unusual or unexpected event external to the passenger which could have caused the DVT. The causal mechanism under consideration was the impact of the flight as a whole on a particular passenger. It was not an event external to the passenger. In the ordinary acceptation of the meaning of accident it does not appear to extend to the onset of DVT by itself.
- In order to achieve a more extensive interpretation of the word "accident" in article 17 counsel for the appellants invoked various policy grounds. He submitted that under article 17 there is a need to impose liability upon the party best able to develop defensive mechanisms, who has physical control of the aircraft, who is most capable of assessing the risk and insuring against it, and who is best able to spread remedial cost over all the passengers. This is a variant of a 'deep pocket' tort law theory which has no place in a trade law treaty such as the Warsaw Convention. On the contrary, article 31 of the Vienna Convention on the Law of Treaties provides that a treaty shall be interpreted "in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose." This is the starting point of treaty interpretation to which other rules are supplementary: see article 31.2; 31.3; 31.4; and 32. The primacy of the treaty language, read in context and purposively, is therefore of critical importance. This approach does not permit the adoption of the economic analysis advocated on behalf of the appellants. Once this approach is rejected the argument for treating DVT by itself as an accident can be seen to be fragile.
- It is, however, necessary to examine the point in the context of the development of comparative jurisprudence. An important case was the Supreme Court decision in Air France v Saks (1985) 470 US 342. The case concerned injuries caused by the normal operation of an aircraft's pressurisation system. The court held that liability under article 17 of the Warsaw Convention arises only if a passenger's injury is caused by an unexpected or unusual event or happening that is external to the passenger. This was an important decision. It is also a strong decision: it covers not only discomfort to a passenger in respect of hearing but also actual hearing loss. Nevertheless, it is not regarded as an accident: Shawcross and Beaumont, Air Law, VII, para 701 (Shawcross issue 101). The appellants do not question the correctness of this decision but argue that if flexibly applied it does not stand in the way of their appeal. Once the appellants wider policy arguments are rejected as I have done it is, however, difficult to support the present DVT claim under the principle in Air France v Saks.
- Let it be assumed that it can be shown that an event affecting a passenger adversely on an aircraft was unexpected and unusual. That is generally, however, not enough to make it an accident. It is an integral part of the test of what amounts to an accident that it must have a cause external to the passenger. In the case of DVT this factor is absent. The component parts of the event cannot therefore amount to an accident. It is closer to a passenger who suffers an asthmatic attack, congenital back pain, a hiatus hernia or simply from the wear and tear of extreme old age. Not surprisingly, such cases have in practice been treated as not amounting to accidents: see Shawcross and Beaumont, Air Law, VII, para 701 (Shawcross Issue 101). Consequentialist arguments militate in favour of treating essentially similar medical cases in the same legal way. After all, DVT is not immune from the general process of medical science.
- On these straightforward grounds it is difficult to treat DVT as falling within the category of an accident under article 17.
- There is, however, another perspective to be taken into account. In interpretation of a trade law treaty, like the Warsaw Convention, it is of great important that the courts of treaty states, and particular their higher courts, should strive for uniform interpretations on debatable issues. Such uniformity is not always attainable but it must be the constant aim. It is necessary to explore how far on this road, the courts of major courts have travelled in respect with the interpretation of DVT.
- In North America there is a clear trend to be observed in the decisions on DVT. Experience shows that a carrier does not incur liability under article 17 for the onset of DVT by itself. Such a claim was dismissed in a lower court in Scherer v Pan American World Airways (1976) 54 AD 2d 636. Similar conclusions were reached in two lower courts in the United States in Blansett v Continental Airlines Inc 379F 3rd 177 (5th Cir 2004) and Rodriques v Ausett Australia Ltd 383F 3rd 914 (2004). (In a case of a congenital asthmatic condition the court came to a similar view: Walker v Eastern Airlines Inc 775 F Supp. 111 (1991)). In the Ontario Court of Appeal a Canadian court came to a similar conclusion: McDonald v Korean Air et al (2003) 171 OAC 368.
- There are also decisions of lower courts in other jurisdictions to which my noble and learned friend Lord Scott of Foscote has made reference. There are, however, two decisions to which I must make particular reference.
- The first is the decision of the Court of Appeal in the present case: In re Deep Vein Thrombosis and Air Travel Group Litigation  QB 234. The court held that for accident to exist there had to be an external event which had an adverse impact on the passenger. Moreover, the court held that inaction was a non event which could not rank as an accident. In careful judgments the Court of Appeal held that a simple question must be asked. Was there, on any recognised meaning of the word, an accident in circumstances where a person suffered DVT merely because of the effect of a flight on an aeroplane without there being any triggering event? The court answered this question in the negative. It regarded it as an error to focus on the component parts of the classic definition in Air France v Saks rather than on the simple concept of accident itself.
- The next development in this important branch of air law was the decision of the Australian High Court in Povey v Qantas Airways Ltd  HCA 33. The issues in Povey were substantially the same as in the English Court of Appeal. In the majority judgment Gleeson LJ, Gummow, Hayne and Heydon JJ followed the reasoning of the English Court of Appeal. They observed (para 44):
". . . In In re Deep Vein Thrombosis Litigation, the Court of Appeal of England and Wales held that the word 'accident' in the Warsaw Convention as modified by the Hague Protocol was to be given a natural and sensible, but flexible and purposive meaning in its context and that for there to be an accident within the meaning of the relevant article, there had to be an event external to the passenger which impacted on the body in a manner which caused death or bodily injury and the event had to be unusual, unexpected or untoward. The Court held that inaction was a non-event which could not properly be described as an accident. Not warning of the risk of DVT and not giving advice on the precautions that would minimise that risk were not events. The conditions in which passengers travelled on flights (with cramped seating and the like) were not capable of amounting to an event that satisfied the first limb of the definition of an accident which 'took place on board the aircraft or in the course of any of the operations of embarking or disembarking."
The majority therefore endorsed the reasoning of the Court of Appeal. Kirby JJ and Callinan J came to similar conclusions in separate judgments. Only McHugh J dissented. Povey is a virtually unanimous decision by the highest court of Australia.
- It is now necessary to consider a recent decision of the United States Supreme Court, which is not directly concerned with a DVT case. It is the case of Olympic Airways v. Husain 540 US 644 (2004). It involved the death of a passenger on board an aircraft as a result of exposure to cigarette smoke. He suffered from a congenital asthmatic condition. The passenger had several times asked to be moved from the smoking section of the plane. A flight attendant refused his request to move him to a non-smoking section. The United States Supreme Court held by a majority of six to two that the flight attendant's conduct did (or could) constitute an accident within the meaning of article 17. I would not cast doubt on the majority decision. It is, however, a very different case from the DVT cases arising from consideration in the present case. And I do not regard it as invalidating the analysis which I have suggested or militating against the virtually uniform interpretation of article 17 to DVT which has so far been adopted.
- Standing back from the detail of the case, my view is that the correct interpretation of article 17 excludes classic DVT cases under consideration from its scope. But if the issue of interpretation had been doubtful I would have wanted to support the strong international judicial consensus which has emerged.
- My Lords, I am in full agreement with the reasons expressed in the House today by my noble and learned friends. I also would dismiss the appeal.
LORD WALKER OF GESTINGTHORPE
- I have had the privilege of reading in draft the opinions of my noble and learned friends Lord Scott of Foscote and Lord Steyn. I am in full agreement with them. I add some brief comments of my own.
- In the Court of Appeal  QB 234 Lord Phillips of Worth Matravers MR stated (p 247, para 25):
"A critical issue in this appeal is whether a failure to act, or an omission, can constitute an accident for the purposes of article 17. Often a failure to act results in an accident, or forms part of a series of acts and omissions which together constitute an accident. In such circumstances it may not be easy to distinguish between acts and omissions. I cannot see, however, how inaction itself can ever properly be described as an accident. It is not an event; it is a non-event. Inaction is the antithesis of an accident."
- Later in his judgment (p 254, para 50) the Master of the Rolls related these general observations to Olympic Airways v. Husain, which had been decided by the United States Court of Appeals, Ninth Circuit 316 F 3d 829 (2002); after the decision of the Court of Appeal in this case Husain went to the United States Supreme Court, 540 US 644 (2004), which was divided. Lord Phillips stated, at para 50:
"The refusal of the flight attendant to move Dr Hanson cannot properly be considered as mere inertia, or a non-event. It was a refusal to provide an alternative seat which formed part of a more complex incident, whereby Dr Hanson was exposed to smoke in circumstances that can properly be described as unusual and unexpected."
- I respectfully agree with that view of Husain, and I have also found helpful a passage in the judgment of Kirby J in High Court of Australia in Povey v Qantas Airways Ltd  HCA 33. After a general discussion of omissions in paras 173 to 176 Kirby J stated at para 187:
"It is unnecessary for this Court to choose between the conflicting opinions expressed in Husain. As was predicted in Saks, and is self-evident, cases will present that are at the borderline of establishing an 'accident' or failing to do so. There were peculiar features of the confrontation between the wife, the passenger and the flight attendant in Husain that arguably lifted that case from classification as a 'non-event' into classification as an unexpected or unusual happening or event and hence an 'accident'."
- The same approach can be found in a very familiar literary source. In Conan Doyle's Silver Blaze Sherlock Holmes draws Dr Watson's attention,
"To the curious incident of the dog in the night-time".
"The dog did nothing in the night-time."
"That was the curious incident."
(cited in the Oxford Dictionary of Quotations 4th ed Revised (1996), p 256, para 14. The dog's failure to bark was part of a more complex incident in which an intruder came into the stable yard in the middle of the night, and was evidently not a stranger to the dog.
- I too would dismiss this appeal.
BARONESS HALE OF RICHMOND
- I share the view that this appeal should be dismissed and for essentially the same reasons. Once it is clear that the accident which causes the injury must be something other than the injury itself, it becomes equally clear that the suffering of an internal reaction to an ordinarily uncomfortable journey by air, during which nothing untoward other than that reaction took place, cannot fall within article 17 of the Warsaw Convention. But I would particularly like to associate myself with the observations of my noble and learned friend, Lord Scott of Foscote, on the dangers of interpreting the words of the decision of a court, which is interpreting the words of the Convention, as if the court's words were those of the Convention. If I fall over during a flight to New York, and break an arm, I suspect that we would all agree that my broken arm was caused by the accident of my fall; and we would do so irrespective of the reason for my fall; if it was my own silly fault, article 21 may relieve the airline of some or all the liability imposed by article 17, but that is another matter. In reaching those conclusions, we should not be agonising too much over whether my fall was an event 'external' to me. We should simply be asking whether it was an 'accident' which led to my injury. My own synonym for 'accident' would be 'untoward event' but that is by the way.
- The preliminary issue before the House is "whether, on the basis of the agreed specimen matrix, the claimants have a claim under article 17 of the Warsaw Convention for the DVT allegedly suffered by them, and in particular whether the specimen matrix discloses an 'accident' for the purposes of article 17 …". Pleadings were exchanged in the issue, and the matrix was agreed "solely for the purpose of [its] resolution at a generic level" and "without prejudice to any other issue which falls to be determined at a generic level and/or the particular facts relating to any individual air carrier and/or any individual case …". Consistently with this, the agreed statement of facts and issues put before the House recite the appellants' assertion and the respondents' denial that "the allegations in the Group Particulars of Claim and Further Information concerning a failure to warn or advise passengers and/or the method of operation of the nominal flight, when taken against the specimen matrix, are capable of being an 'accident' causing the DVT to occur and thus of giving rise to liability under article 17 …".
- The relevance of this point is that in the Particulars of Claim the appellants contend not just that the respondents failed to give any warning or advice in circumstances where (it is said) they knew or ought to have known of the increased risk of DVT (paragraphs 6 and 7 of the agreed matrix), but also that this involved a "failure" by the respondents to carry the appellants "with all the due care and skill to be expected of a reasonable, competent and experienced commercial carrier". Whether allegations of such a nature have any, and if so what, relevance to the existence of an "accident" under article 17 lies at the heart of this appeal. But it is material to note that Mr Scrivener QC for the appellants accepts that, at the dates on which the appellants were carried and suffered DVT (in the late 1990s extending possibly into 2000), the respondent airlines were acting in accordance with general industry practice in carrying them as they did and in not giving any warnings about DVT. There was nothing unusual, abnormal or unexpected about the way in which the relevant aircraft were fitted out or operated. One would expect that to be of considerable - although not decisive - importance if it were relevant to consider whether a carrier acted "with all the due skill and care to be expected of a reasonable, competent and experienced commercial carrier" (cf in English law Bolam v. Friern Hospital Management Committee  1 WLR 582, 586-7; Australian law may differs: cf Rogers v. Whitaker (1992) 175 CLR 479, 484 and Povey v. Qantas Airways Ltd.  HCA 33, para. 177, per Kirby J). On any view the appellants' prospects of success in a case such as the present seem to be confined to a narrow situation in which standard airline practice falls short of some higher standard identified by the court as appropriate at the relevant time(s) between a carrier by air and its passenger.
Approach to interpretation
- With that preliminary comment, I turn to the interpretation of article 17. The argument has taken as the relevant version that appearing in the Warsaw Convention as amended by the Hague Protocol and set out, in English and French, in Schedule 1 to the Carriage by Air Act 1961. The group defence introduces (and the respondents kept open before the House) alternative possibilities as to the version of the Convention actually covering the carriage of any particular claimant passenger, while not suggesting that these involved any presently relevant difference. The wording of article 17 to which submissions were directed reads:
"17. The carrier is liable for damage sustained in the event of the death or wounding of a passenger or 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."
- There was little if any difference between the parties regarding the proper approach to interpretation of the relevant version of the Warsaw Convention. Section 1 of the 1961 Act states that "If there is any inconsistency between the text in English in Part I of Schedule 1 or 1A to this Act and the text in French in Part II of that Schedule, the French text shall prevail". The French text provides:
"Le transporteur est responsible de dommage survenu en cas de mort, de blessure ou de toute autre lesion corporelle subie par un voyageur lorsque l'accident qui a cause le dommage s'est produit à bord de l'aéronef ou au cours de toutes operations d'embarquement et de débarquement."
In Air France v. Saks 470 US 392 (1985), the United States Supreme Court considered the French text and in particular the French legal meaning of the term "accident", and concluded that it differed little from, and in effect offered the same possibilities as, the English word. The contrary was not suggested before us.
- The primary consideration is the natural meaning of the language used, taking into account the text as a whole, and such conclusions as can be drawn regarding its object and purpose: Fothergill v. Monarch Airlines Ltd.  AC 251, 272F-G per Lord Wilberforce, 282D-E per Lord Diplock, 287D per Lord Fraser and 290G per Lord Scarman. The text should be interpreted "in a normal manner, appropriate for the interpretation of an international convention, unconstrained by technical rules of English law, or by English legal precedent, but on broad principles of general acceptation": Stag Line Ltd. v. Foscolo, Mango and Co. Ltd.  AC 328, 350, per Lord Macmillan; Buchanan & Co. Ltd. v. Babco Forwarding and Shipping (UK) Ltd.  AC 141, 152, per Lord Wilberforce; and Fothergill, supra, 272E per Lord Wilberforce, 282A per Lord Diplock, 282C per Lord Fraser and 293C per Lord Scarman. The concepts deployed in the convention are thus autonomous international concepts. The legislative history and travaux préparatoires may be considered to resolve ambiguities or obscurities, when the material is publicly available and points to a definite consensus among delegates. It is also legitimate to have regard to any subsequent practice among the parties which is capable of establishing their agreement regarding interpretation. All these points are, for conventions concluded after 27th January 1980, covered in articles 31(3) and 32 of the Vienna Convention on the Law of Treaties (United Nations, Treaty Series, vol. 1155, 331); and in Fothergill, supra, at pp.282D and 290C Lords Diplock and Scarman treated these articles as codifying previous international legal principles.
- Allied to the legitimacy of having regard to subsequent practice is the significant consideration that a convention like the Warsaw Convention as amended is aimed at achieving conformity of law and practice in its field, and that uniformity in its interpretation is highly desirable. The Convention endeavours, as the United States Supreme Court said in Zicherman v. Korean Air Lines Co. Ltd. 516 US 217, 230 (1996), "to foster uniformity in the law of international air travel" (cf also El Al Israel Airlines Ltd. v. Tseng 525 U.S.155 (1999), 169-170). Thus, my noble and learned friend Lord Hope said in Morris v. KLM Dutch Airlines  2 AC 628, para. 81:
"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."
Lords Nicholls and Mackay made the same point. Lord Nicholls said at para. 5:
"5. It really goes without saying that international uniformity of interpretation of article 17 is highly desirable. Like Lord Mackay of Clashfern, I have been much concerned that the interpretation of article 17 espoused by this House should, if possible, be consistent with the mainstream views expressed in leading overseas authorities."
He went on to attach to a decision of the United States Supreme Court particular importance "given the important position of the United States in carriage by air". Lord Mackay said at para. 7 that:
"… I consider it important that the Warsaw Convention should have a common construction in all the jurisdictions of the countries that have adopted the Convention …"
In that light he attached "crucial importance" to two decisions of the United States Supreme Court, saying that he did so "particularly as the United States is such a large participant in carriage by air" and that, but for these decisions, he would have given more weight to the argument for a contrary construction (of the word "bodily" and its French counterpart "corporelle") than he in fact adopted. Lord Hobhouse at para. 147 also identified the importance of having regard to any international consensus upon the understanding of an international convention "and hence to what the courts in other jurisdictions have had to say about the provision in question". Thus, although "the relevant point for decision always remains: what do the actual words used mean?" (per Lord Hobhouse at para. l47), a consensus of domestic legal authority in higher courts of significant standing can weigh significantly in the determination of that meaning, Similar importance was attached to achieving international uniformity of interpretation in Saks, at p.404, and by the majority judges in the High Court of Australia in Povey at pp.25, 137 and 142.
The common ground
- Article 17 is formulated in terms suggesting an assumption that damage sustained in the event of the death, wounding or bodily injury will have been caused by some form of "accident" (a point made by Kirby J in Povey at para. 111). But that is not a sensible reading. First: "Like pilgrims to th'appointed end we tend; the world's an inn and death the journey's end" (Dryden, Palomon and Arcite, (1700) book 3, II, 883-4). Death from entirely natural or internal causes may be described as accidental, but it is not caused by an accident, which is what matters under article 17; and, if a passenger were to suffer bodily injury as a result of a fall on board which was due to some internal condition (such as partial paralysis or drunkenness) not sensibly attributable to the airline, it seems improbable that his injury should be regarded as caused by a relevant accident: compare Chaudhari v. British Airways plc (16th April 1997, Court of Appeal; CCRTI 96/0229/G) - although in that case there were allegations of negligence on the airline's part, which would, it may be, lead Mr Scrivener to submit that the court should have taken a different view on the facts - and Padilla v. Olympic Airways 765 F. Supp. 835 (1999) (U.S. District Court, SDNY). The Supreme Court noted in Air France v. Saks 470 US 392 (1985), 403-4 that many of the signatories to the Warsaw Convention were parties to the Guatemala City Protocol of 1971 (ICAO Doc. 8932) which would, had it been ratified, have imposed liability for an "event which caused the death or injury" but with an express exemption if the death or injury resulted "solely from the state of health of the passenger". The proposed switch from "accident" to "event" was viewed by delegates as expanding the scope of carrier liability to passengers, but, even so, that expansion was not to include liability for death or injury resulting from the passenger's own state of health.
- Secondly, the drafting history, examined in Saks at pp.401-3, indicates that it was deliberately decided to address separately and in separate terms liability for (a) persons, (b) goods and baggage and (c) delay. The original draft submitted to the second international conference in Warsaw in 1929 would, in contrast, have imposed liability "for damage sustained during carriage" in all these three cases, subject only to the carrier's ability to avoid liability under article 22 by proving it had taken reasonable measures to avoid liability. But the process of negotiation led to a differently conceived scheme, which identifies the necessary cause of damage, in relation to persons, as an "accident" (article 17) and, in relation to goods and baggage, as an "occurrence" (article 18), but speaks simply, in relation to delay, of "damage occasioned by delay in the carriage by air" (article 19). As the Supreme Court observed in Saks (p.403), "an 'accident' must mean something different than an 'occurrence' on the plane". Any accident may be regarded as an occurrence, but the reverse does not hold true.
- Thirdly, the pre-condition to liability under article 17 is not simply that there should have been an accident; the damage sustained in the event of death, wounding or bodily injury must be shown to have been caused by an "accident" which "took place on board the aircraft or in the course of any of the operations of embarking or disembarking". For these reasons, courts throughout the world, when addressing claims under article 17, have had to engage in analysis, and application to varying factual situations, of the term of "accident".
- Mr Scrivener does not challenge any of this. He expressly accepts the conclusion of the Supreme Court in Saks (at pp.400-5) that the death or wounding of or other bodily injury to a passenger cannot of itself be regarded as a relevant "accident" under article 17; and that the death, wounding or bodily injury "must be caused by an accident, and an accident must mean something different than an 'occurrence' on the plane" (p.403). Justice O'Connor, giving the Supreme Court's judgment, supported this conclusion by examination of the drafting history (pp.400-403), by reference to subsequent conduct of the parties and subsequent interpretations of the signatories (pp.400 and 403-4) and by reference to what she described as "the weight of precedent in foreign and American courts" (p.400 and 404-5). Again, no issue has been taken with her approach or conclusions. Further, Mr Scrivener expressly accepts the paraphrase of the term "accident" suggested in Saks (at pp. 400 and 405) - "an unexpected or unusual event or happening that is external to the passenger". But he underlines Justice O'Connor's immediately following words:
"This definition should be flexibly applied after assessment of all the circumstances surrounding a passenger's injuries"
It is in this context that Mr Scrivener lays particular stress upon what he submits are relevant purposive considerations.
- Up to a point, the parties are also on common ground regarding the object and purpose of article 17. Both sides accept that the Convention as amended seeks to strike a balance between competing interests in its articles governing liability towards passengers. It was introduced in the infancy of international air travel at a time when airlines were exposed to unrestricted liability under whatever law was regarded as applicable in whatever forum the plaintiff was able to and did chose. This open legal market situation inhibited financial investment in and growth of the airline industry: cf El Al Israel Airlines Ltd. v. Tseng 525 US 155 (1999), 170 and Zicherman v. Korean Airlines Co. Ltd. 516 US 217 (1996), 230. The Convention brought benefits to both sides. My noble and learned friend Lord Hope described these in Morris v. KLM Royal Dutch Airlines  UKHL7; 2 AC 628, para. 66:
"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."
- Lord Hobhouse put the matter as follows at para. 150:
"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. …"
- It is now established that the remedy provided by article 17 is exclusive, in the sense that no other common or civil law remedy (e.g. for want of reasonably care or negligence) is available to a passenger who sustains bodily injury or dies in the course of international carriage by air, but has no claim against the carrier under article 17 (e.g. because the injury or death was not caused by an accident on board or in the course of embarkation or disembarkation). The House decided this in Sidhu v. British Airways plc  AC 430, where passengers were captured by Iraqi forces in an air terminal in Kuwait where their aircraft had landed just before the Iraqi invasion of August 1990.
- Both sides also reject an originalist approach to the application of article 17, or a limitation of the term "accident" by reference to the type of aviation disaster affecting a whole airplane that was probably all too familiar in 1929. It is, Lord Hobhouse said in Morris v. KLM Royal Dutch Airlines  2 AC 628, para. 149
"… 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."
It is thus fully open to the appellants to argue that the present claims involved "bodily injury" caused by an "accident" within article 7, although that article was originally agreed in an era when no-one conceived that air travel could involve the complex risk of DVT and such international flights as existed involved short refuelling hops which may have meant that it did not.
The rival submissions in outline
- At this point, the respective submissions start to diverge. Mr Scrivener submits that the general conception underlying article 17 is that an airline should bear the risk of, and answer for, matters within its control, in respect of which it can take precautions and insure, and that a court should be reluctant to regard an airline as free from responsibility where a passenger has suffered on board an injury due to conditions on board over which the airline had complete control. This, he submits, is especially so, if and so far as any element of culpability could be attributed to the airline. Rather than concentrate on airline practice at the relevant time, he submits that the court should concentrate on what the ordinary passenger or member of the public would have expected of an airline which had, or ought to have had, the awareness of the risks of DVT which the claimants allege; otherwise, an airline could, by relying supinely on existing practice, avoid liability for known risks and would lack incentive to make changes; the flexible approach to the term "accident" suggested by the Supreme Court in Saks should enable and require the court to give effect to such considerations.
- In contrast, Mr Robert Webb QC appearing with Mr Lawson for British Airways argues, in submissions that were adopted by Mr Lawson appearing for China Airlines, that the balance struck by the Warsaw Convention as amended by the Hague Protocol allows limited scope for deeper examination of any underlying object or purpose behind its wording or, indeed, for examination of the consequences of any particular interpretation. The word "flexibly" used by the Supreme Court does not, he submits, necessarily equate with "broadly". Mr Webb refers to my noble and learned friend Lord Steyn's words in Morris at para. 15, and submits that they are equally applicable to the interpretation of the term "accident":
"15. The Warsaw Convention is an exclusive code of limited liability of carriers to passengers. On the other hand, it enables passengers to recover damages even though, in the absence of the Convention and the Act, they might have no cause of action which would entitle them to succeed: Swiss Bank Corp. v. Brink's MAT Ltd  QB 853, 856G-H, per Bingham J (now Lord Bingham of Cornhill). It is therefore not necessarily right to approach the meaning of the phrase 'bodily injury' in article 17 of the Convention through the spectacles of full corrective justice."
- The Supreme Court's admonition in Saks to apply its suggested definition of accident "flexibly" was illustrated with examples, in particular the interpretation by lower courts of the term "broadly enough to encompass torts committed by terrorists or fellow passengers" (p.405). While this indicates that "flexibly" can mean "broadly", it does not point to a general abandonment of the idea of an "unexpected or unusual event or happening that is external to the passenger". An assault by a terrorist or by another passenger can be regarded as falling within a broad conception of these words. It can, as far as the airline and the passenger are concerned, also be described, without very great difficulty, as an accident of the carriage. But the Supreme Court's definition was suggested with the clear purpose of delineating boundaries to the concept of "accident". The claimant's submission in Saks had been that "accident" embraced any "hazard of air travel", and the Court of Appeals, whose judgment the Supreme Court was reversing, had used a definition derived from another convention (that on International Civil Aviation of Dec. 7 1944), namely "an occurrence associated with the operation of an aircraft which takes place between the time any person boards an aircraft with the intention of flight and all such persons have disembarked". Such an approach would in one respect have introduced a limitation on the concept (to hazards or risk in some way associated with aviation travel) which is certainly not explicit if it exists in any degree in the Supreme Court's paraphrase (which I need not consider - cf Wallace v. Korean Air 214 F.3d 293 (2000); U.S. Court of Appeals, 2nd Circ. for discussion of the point). They would have opened up airlines to liability for bodily injury attributable to any such hazard or risk, whether or not involving anything untoward or unexpected or indeed internal in origin.
- The practical significance of a determination whether or not an accident took place within the meaning of article 17 has altered with changes in the surrounding legal terrain. In Saks the plaintiff was pursuing a state negligence action, which the Supreme Court remitted to the Court of Appeals (pp.394 and 408). Since the Supreme Court's decision in El Al Israel Airlines Ltd. v. Tseng 525 US 155 and the House of Lord's decision in Sidhu v. British Airways plc  AC 430, it has been clear that article 17 is an exclusive remedy in respect of death or bodily injury sustained during air carriage. Prior to those decisions, a claimant might well argue for a restricted interpretation of the term "accident", in order to pursue an unlimited claim for (say) negligence at common law, while carriers might respond by arguing for a wider interpretation (cf Fishman v. Delta Air Lines, Inc. 132 F.3d 138 (1998); U.S. Court of Appeals, 2d Circ. for an example). Since Tseng and Sidhu, these incentives have been reversed, with carriers benefiting by a complete exclusion of liability if they can negative the existence of any "accident" within the meaning of article 17. However, there are complicating cross-currents, capable of inducing different attitudes: under the Montreal Agreement of 1966, carriers for flights originating, terminating or stopping in the United States agreed voluntarily to forego any defence that they had taken all necessary measures (or that it was impossible to take them) under article 20; while the Montreal Convention of 1999 (set out in Schedule 1B to the 1961 Act and in force in the United Kingdom since June 2004) omits altogether any such defence (substituting only limited provisions for exoneration in the case of negligence or other wrongful act or omission of the passenger). At the same time, the Montreal Protocol underlines the requirement of an accident under article 17 which becomes:
"The carrier is liable for damage sustained in case of death or bodily injury of a passenger upon condition only that the accident which caused the death or bodily injury took place on board the aircraft or in the course of any of the operations of embarking or disembarking."
These changes in the surrounding legal terrain may, as a matter of psychological reality, justify a degree of caution about any attempt to reconcile every past legal decision (cf words of Sutherland J in Quinn v. Canadian Airlines International Ltd. (1994) 18 O.R. (3d) 326, 331c-f). But there are enough decisions by different courts to mean that that is in any event an unpromising exercise. In my view, even if the application has been inconsistent, it is unconvincing to suggest that the essential meaning of the term "accident" in the relevant articles has in law changed.
- In Sidhu, where the British courts recognised the exclusivity of article 17, my noble and learned friend Lord Hope said at p.453H-454A that "No doubt the domestic courts will try, as carefully as they may, to apply the wording of article 17 to the facts to enable the passenger to obtain a remedy under the Convention", before going on to point out that it was, in that case, conceded that no such remedy was available. A similar thought was expressed by Kirby J in Povey v. Qantas Airways Ltd.  HCA 33 when he said at para. 137 that, although the travaux préparatoires
"show the compromises that were struck in adopting the text of the Warsaw Convention, they also indicate that that recovery of passengers of acceptable compensation for damage, consequent upon injury, as defined, was an objective of the Warsaw Convention";
and also when he said, with reference to article 25, that
"[if] airlines with the requisite knowledge deliberately or recklessly withheld essential information from passengers (assuming that to be provable) the imposition of liability for consequent serious damage involving injury or death would not be entirely surprising, given the overall objects of the treaty".
Nonetheless, Kirby J joined with the majority in concluding that the incidence of DVT in Povey had not been caused by any accident. In a similar vein Callinan J at para.205 questioned whether the concept of "accident" might not now be giving carriers by air "the benefit of an anachronistic approach to the perils of travel".
- In an impressive review of authority and the factors in play in this difficult area it was suggested by Marrero DJ in Fulop v. Malev Hungarian Airlines 175 F.Supp.2d 651 (S.D.N.Y. 2001), 657 that, both before and after the decision in Saks, courts grappling with the application of article 17 had weighed various factors, focusing in particular on
"the relationship of the claimed accidents to (1) the normal operation of the aircraft or airline; (2) the knowledge and/or complicity of crew members in the events surrounding the alleged accident; (3) the acts of fellow passengers, whether intentional or not; (4) the acts of third persons who are not crew or passengers, e.g. hijackers and terrorists; (5) the location of the occurrence in the continuum of travel; (6) the role, condition and reaction of the complainant in connection with the occurrence at issue; and (7) the kind of risks inherent in air travel."
He went on:
"Insofar as any decisive pattern may be discerned [in the authorities], the fulcrum of these considerations often rests on the extent to which the circumstances giving rise to the claimed accident fall within the causal purview or control of the carrier - or at least within its practical ability to influence -as an aspect of the operations of the aircraft or airline. The larger the role of the airline in the causal chain, and the greater the knowledge and involvement of its personnel and operations in bringing about the harmful event, the more likely it is that liability will be found.
Conversely, as the causal balance shifts towards acts and conditions that are independent of the knowledge or will of the carrier, or not associated with the operation of the aircraft or airline nor arising from risks characteristic of air travel, and instead are more unique to the passenger alleging injury, the lesser the claimant's probability."
Taking the paraphrase suggested in Saks which both sides accepted as a starting point, the issue before us can be broken down into three aspects, namely whether there was (a) an event, which was (b) unexpected or unusual and (c) external to the passenger. These could be viewed as involving, if only broadly, Marrero DJ's suggested factors (1), (2), (5), (6) and (7). But in relation to issue (a), the decision of the Court of the Appeal raises a sub-issue not explicitly identified by Marrero DJ, namely the extent, if any, to which inaction can constitute an event. Further, factor (2), crew "knowledge and/or complicity", arises in an expanded form in this case, having regard to the appellants' positive case that the airline or aircrew were at fault; the allegations of fault, relating to the condition of the aircraft and/or failure to warn of the risks of DVT, are advanced as relevant to all three issues, that is to show that (a) there was an event, (b) it was unexpected or unusual and (c) it was external to the appellant passengers. In this case, if they are sufficient to establish an event and that it was unexpected and unusual, then it would also be external. If there was no such unexpected or unusual event, the injury suffered by the appellants would flow from internal causes. So it is on issues (a) and (b) that it is necessary to focus.
- In the Court of Appeal,  EWCA Civ 1005;  QB 234, Lord Phillips of Worth Matravers MR, in a judgment with which Judge and Kay LJJ agreed while adding some reasoning of their own, disposed of the appellants' claim on the basis that it involved no event, merely alleged inaction on the carriers' part in relation to a standard flight on a standard aircraft operating normally and without any undue incident. He said at para. 25:
"25. A critical issue in this appeal is whether a failure to act, or an omission, can constitute an accident for the purposes of article 17. Often a failure to act results in an accident, or forms part of a series of acts or omissions which together constitute an accident. In such circumstances it may not be easy to distinguish between acts and omissions. I cannot see, however, how inaction itself can ever properly be described as an accident. It is not an event; it is a non-event. Inaction is the antithesis of an accident."
In deciding that the matrix disclosed no event, he said, first, that the permanent features of the aircraft, or the subjecting of the passengers to carriage in aircraft with these features was not capable of amounting to an "accident" within article 17 (para. 28), and, secondly, that the alleged failure to warn of the risk of DVT, or to advise on precautions which would avoid or minimise the risk, could also not be categorised as an accident: it was "simply something that did not happen - a non-event" (para. 29). He went on to say that "The question whether there took place on the flight events which were 'unexpected or unusual' was and is unreal, having regard to the judge's conclusion and my provisional conclusion that there was no relevant event" (para. 3l), adding only at paras. 37 and 38:
"37. I am not persuaded that, when considering whether an action has the unusual, unexpected or untoward quality necessary to enable one to qualify it as an accident under the Air France v. Saks test, the questions of whether the actor has knowledge which makes it culpable to perform the action can never be relevant. I am, however, persuaded that it is simply not possible to apply to a state of affairs, or an omission to act, the test that is relevant to deciding whether an event is an accident.
- Looking at the position overall, which is the only realistic way of looking at it, the matrix, as augmented by the Further Information, discloses nothing which is capable of constituting an "accident" within the meaning of article 17."
- Later, he addressed the decisions of the United States District Court and Court of Appeals in Husain v. Olympic Airways (2000) 116 F Supp 2d 1121 and 316 3d 829 (since affirmed by the Supreme Court by a majority at 540 US 644 (2004), 124 S.Ct. 1221 (2004)). The facts there were that Dr Hanson suffered from asthma and was particularly sensitive to second-hand smoke. He and his wife had requested non-smoking seats, but found that these were only three rows in front of the smoking section, which was not partitioned off. His wife's requests on three occasions (twice while the aircraft was on the ground and once in the air) that he be moved further away were refused by the flight attendant, who asserted on the second and third occasions that there were no spare seats, though she also said on the third occasion that they could ask other passengers to exchange seats, but would not be assisted by the crew. Dr Hanson walked to the front of the plane to get some fresh air, and there collapsed and died. The lower courts found that the airline's "failure to act was a 'blatant disregard of industry standards and airline policies", and that it satisfied the meaning of "accident". The Master of the Rolls did not question the result, but did question the reasoning, suggesting (para. 50) that:
"The refusal of the flight attendant to move Dr Hanson cannot properly be considered as mere inertia, or a non-event. It was a refusal to provide an alternative seat which formed part of a more complex incident, whereby Dr Hanson was exposed to smoke in circumstances that can properly be described as unusual or unexpected. … the direct cause of his death was the unnecessary exposure to the smoke. The refusal of the attendant to move him could be described as insistence that he remain seated in the area exposed to smoke. The exposure to smoke in these circumstances could … properly be described as an unusual or unexpected event."
- The decision in Husain was affirmed in the Supreme Court, where the Master of the Rolls' words were quoted by the majority as indicating that their conclusion was not inconsistent with that of the Court of Appeal in the present case. However, the majority in the Supreme Court also rejected (in footnote 9 on p.1229) the Master of the Rolls' "precise reasoning" to the extent that it differed from their own, illustrating what they regarded as "the fallacy of the petitioner's position that an 'accident' cannot take the form of inaction" with the example (in footnote 10) of a failure in accordance with industry standards to divert a flight when a passenger otherwise faces imminent death as illustrating that
"the failure of an airline crew to take certain necessary vital steps could quite naturally and, in routine usage of language, be an 'event or happening'"
Mr Scrivener places particular reliance on these last footnote words, but they must be read in context. The essence of the majority's reasoning at pp.1229-1330 was that the combination of an expected and usual reaction and a categorical refusal or failure by the crew to react in that way on a particular flight could amount to an "accident".
- The minority judges in Husain (Justices Scalia and O'Connor) took issue with the majority's conclusion that inaction could amount to an "accident". They agreed with the Master of the Rolls' analysis of the legal position, while questioning whether it was on the particular facts right to treat the flight attendant as having "insisted" on Dr Hanson remaining seated where he did (since she had in fact left it to Dr Hanson to try to find a seat elsewhere). They did not consider that anything unusual or unexpected had necessarily occurred at the time of the two requests prior to take-off, and they considered that, subject to one point, the flight attendant had been guilty of no more than inaction at the time of the third request. The one point which in their view required remission to the District Court related to the attendant's repeated representation that the plane was full. If that had led Dr Hanson or his wife not to look for another seat, and was both unusual and unexpected (as to which the minority evidently felt little difficulty) and a cause of his death, then it could be said to be an "accident" under article 17.
- The Court of Appeal and the minority in Husain would thus draw a line between action and inaction, while the majority in Husain distinguished between on the one hand the unusual and unexpected and on the other hand the usual and expected (in each case whether involving action or inaction). But the difference between the two approaches may not be as great as might at first sight appear, for reasons indicated by passages from paras 25 and 50 of the Master of the Rolls's judgment quoted in paras. 70 and 71 above. These passages come close to saying that an event may in some circumstances exist, where there has been crew inaction in a context where action would normally be expected. A similar view was expressed by Kirby J in para. 172 of his separately reasoned judgment concurring in the result with the majority in Povey in the High Court of Australia. If that may be so, then the question whether there has been an event inter-relates with the question whether what happened was unexpected or unusual. I therefore turn to that further question.
- The present case involved carriage by air in an aircraft and in a manner, which were, in terms of industry standards and practice, at the relevant times, normal, usual and expected. Like the Master of the Rolls, I can see no basis on which the permanent features of the aircraft, or the subjecting of the passengers to carriage in aircraft with these features could amount to an "accident" within article 17. That is not of course the same as saying that an unexpected event during the flight must always be instantaneous and immediately noticeable, rather than continuous and unrecognised. Ashley J in his partially dissenting judgment in the Court of Appeal of Victoria in Povey gave (at paras. 211-216) some examples of continuous situations (e.g. continuous malfunction of the pressurisation causing injury) where he thought that there would be an 'accident', and I am in sympathy with him on that point. But the situations he was envisaging were on any view unusual and contrary to expected airline standards and practice.
- With the appellants in this case unable to point to anything unusual or unexpected about the permanent features of the aircraft or its operation, the emphasis has been on the respondent airlines' alleged failure to warn of the risk of DVT, or to advise on precautions which would avoid or minimise the risk, in the light of the knowledge which it is alleged that they had or ought to have had. But it is accepted that it was neither industry nor the respondent airlines' practice at the relevant times to give such warnings or advice. The definition suggested in Saks looks (and later authorities treat it as looking) to what was actually expected and usual at the relevant time. Mr Scrivener submits that the airline had or ought to have had knowledge which, if it had also been known to a reasonable passenger or bystander or a court, would have led him or her or it to regard the normal operation of the airline as inappropriate and to expect more of the airline. But that elides what was actually expected with what would or should, in the light of some further knowledge or higher standard, have been expected. The aircraft and airline were operated and able to operate exactly as expected by all concerned, even if this was because the airline did not, in that light, sufficiently address the need to act in a different manner, and because passengers and the public did not have the knowledge to expect anything different.
- Mr Scrivener submits that, if the appellants were able to prove the fault which they allege, that should compel a different and broader view of the term "accident". First, in so far as this argument invokes article 20, it is illegitimate. It is not possible to short-circuit article 17, by ignoring the requirement of an accident causing the death or injury and moving directly to consideration of the factors introduced in article 20. Article 20 only applies if there has been an accident causing death or injury, an occurrence causing damage to baggage or cargo or damage occasioned by delay within, respectively, articles 17, 18 or 19. As Supreme Court observed in Air France v. Saks 470 US 392 (1985), 407:
"The 'accident' requirement of Article 17 is distinct from the defenses in Article 20(1) both because it is located in a separate article and because it involves an inquiry into the nature of the event which caused the injury rather than the care taken by the airline to avert the injury."
Secondly, though associated with the first point, liability under the terms of article 17 is strict, and the exemption which an air carrier can claim under article 25 in the presently relevant form of the Convention as amended requires more than a simple negativing of fault in the ordinary domestic sense: cf Swiss Bank Corporation v Brink's MAT Ltd  QB 853. The general effect of article 25 is not therefore to make the carrier liable on the basis of fault. Hence, it is understandable that article 17 may include additional substantive pre-conditions to the carrier's liability, such as the requirement of an "accident". This will, as I have pointed out, be a point of even greater significance in cases under the Montreal Protocol of 1999, where there is no equivalent to article 20.
- Thirdly, references to fault (e.g. in failing to warn) can be attractive to the domestic lawyer, familiar with contractual or tortious claims for failure to exercise due skill and care. But it is the nature of the event, rather than the care taken to avoid injury which is the focus of the term 'accident' in article 17 (cf the citation from Air France v. Saks in paragraph 77 above). As the majority pointed out in the High Court of Australia in Povey at para. 41, such references "must proceed from unstated premises about the content or origin of some duty to warn". Unless a proper foundation can be found for these in the concept of "accident", introduced in article 17 and elaborated in Saks in a way which both parties accept, they have no place. However, that concept addresses what was actually expected or usual at the time, not what might or should have been according to some undefined different standard.
- Fourthly, I do not accept that such a conclusion is a recipe for maintaining lax standards. Airlines and airline safety are matters which are, and must always have been, very much in the public eye. They are closely regulated at international and national levels. It seems improbable in the extreme that the or even an important driver for aircraft safety and consideration for passengers is the legal responsibility that may result from liability under article 17. Standard regulatory systems, competition between airlines and the incentives to offer better service are much more credible factors. In the context of the balanced compromise at which article 17 aims, I see nothing incoherent about a standard of liability determined by current expected airline practice, rather than by some higher standard which a court would be asked, in effect retrospectively, to identify.
- Fifthly, where an airline and the crew of its aircraft have acted in accordance with usual and expected practice current at the relevant time, it is at least questionable whether a failure to act in some other way could, without artificiality, be regarded as an accident on board or in the course of embarking or disembarking. The realistic target of criticism would seem to be not the crew, but senior officers of the airline, in the airline's office responsible for safety, for failing to identify the risk of DVT and to introduce further precautions, changes or warnings in relation to flights on its airplanes and flights. Further, if, as in Mr Scrivener's submissions, the emphasis is on the failure to warn, the right time for any warning, in view of the seriousness of DVT, could well be regarded as being not merely prior to boarding, but prior to any booking, so as to give the passenger full opportunity not to commit himself or herself to fly at all.
- There is a substantial body of international authority consistent with the views that I have expressed. In Saks itself, the claim was for internal injury allegedly sustained as a result of the usual, normal and expected operation of the aircraft's pressurisation system (cf pp.395-6 and 406). Common law allegations of negligence were nonetheless, left open (pre-Tseng) and remitted to the court below for further consideration. In Capacchione v. Qantas Airways, Ltd. 25 Avi Cas 17,346 (1996) (US District Court, Central District of Colombia) the claim was for injury allegedly sustained as a result of exposure to an insecticide which the airline was required by Australian regulations to spray inside the aircraft. There was an allegation of negligent failure to warn about the use of such insecticides on board. Nonetheless, the court (again pre-Tseng) held that there had been no accident within article 17:
"The failure to warn Capacchione was not unusual or unexpected. Rather, nothing in Qantas's standard procedure required giving a warning about insecticides" (p.17,351).
- In Fulop v. Malev Hungarian Airlines 175F Supp. 2d 651, to which I have referred in paragraph 69 above, the passenger suffered an heart attack on a flight from Budapest to New York, and claimed that he requested a diversion to London for treatment that the carrier failed to carry out. It was accepted that the carrier had no liability for the initial heart attack, but the claim was that it had been aggravated, leading to "additional complications and graver injuries proximately attributable to the flight crew's conduct which conceivably would not have occurred had the carrier's personnel adhered to established operational standards, rules or policies applicable in the circumstances" (p.664). The District Court said (pp.670-1):
"One course, the one chosen, was 'normal' and uneventful insofar as it led the flight directly to its intended destination.
But the Flight was not 'normal' in that, by not diverting, it led to Fulop's more serious injuries. … One way, the route not chosen, would have complied with airline and industry practices, while the 'normal' scheduled flight in fact allegedly contravened established rules.
Whether in the final analysis the flight crew's acts under these circumstances were culpably negligent is beside the point. What does matter, in this Court's view, is that the Flight and the air carrier's operation were not routine or normal in the sense that they allegedly deviated from compliance with expected procedures and that the departure and its associated delay in bringing emergency relief sooner may have aggravated Fulop's injury."
- In Blansett v. Continental Airlines, Inc. 379 F.3d 177 (5th Cir. 2004) (US Court of Appeals), a passenger on a trans-Atlantic flight in June 2001 suffered DVT which left him disabilitated. By that date many international carriers (it appears about half of them) provided passengers with information and advice about DVT, but many, including Continental Airlines, did not (p.182). The crew of the aircraft on which Mr Blansett travelled remained accordingly "entirely passive", i.e. silent, in that respect. The Court of Appeals, considering Husain, observed that the Supreme Court had "declined to base its analysis on language of reasonableness or unreasonableness" and held that the district court had accordingly erred when it "called for consideration whether Continental's conduct was an accident merely because it was "unreasonable" (p.181). It pointed out that there was nothing in the case equivalent to the crew's alleged failure in Husain to react to an urgent situation arising in flight, saying:
"No such circumstances were thrust on the flight crew in the present case, and their compliance with the regular policy of their airline was hardly unexpected. Rather, the Blansetts allege that the "unexpected" nature of the alleged event arose not from the choices of the flight attendants, but from the Continental policymakers who decided not to mandate DVT warnings on Continental flights.
The Blansetts reason that though this decision occurred at a time and place distant from Blansett's flight, article 17 is to be 'applied flexibly' after 'assessment of all the circumstances surrounding a passenger's injuries …" See Saks, 470 U.S. at 405 …"
The Court of Appeals was prepared to assume that the failure to warn was a departure from an industry standard (p.181). The emphasis here is on the word "an", because the Court went on to make clear that it was undisputed that in 2001 many international carriers did not give DVT warnings, and that Continental's "battery" of other warnings accorded with the Federal Aviation Administration's ("FAA's) then current prescriptions. It decided the appeal on the basis that (p.182):
"It was not an unexpected or unusual decision for Continental merely to cleave to the exclusive list of warnings required of it by the agency that has regulatory authority over its flights.
Ultimately, no jury may be permitted to find that Continental's failure to warn of DVT constituted an 'accident' under article 17. Continental's policy was far from unique in 2001 and was fully in accord with the expectations of the FAA. Its procedures were neither unexpected nor unusual."
- Blansett was followed and applied by the Court of Appeals of the Ninth Circuit in Rodriguez v. Ansett Australia Ltd. 383 F 3d 914 (2004), where the claimant sustained DVT during a flight to New Zealand in September 2000, and complained of an alleged failure to warn, without submitting any evidence to raise any genuine issue as to whether there was at the time either a clear industry standard or an airline policy regarding DVT warnings.
- The same approach has now been adopted by the majority judgment in the High Court of Australia in Povey, which referred to both Blansett and Rodriguez as well as the decision of the English Court of Appeal presently under appeal as being "consistent" with its own conclusion. The claim was again for DVT allegedly caused by the conditions of and procedures relating to passenger travel in flight. But neither the flight conditions nor anything happening on board was said to be out of the ordinary or unusual. The conditions and procedures were standard, and "only by the mechanism of describing the absence of warning as a 'failure to warn' did the appellant seek to suggest that the absence of warning was in any respect unusual or unexpected on the flights concerned" (para. 40). The High Court upheld the majority decision below dismissing the claim, stating (at paras. 41 and 42):
"References to 'failure' to warn in this context are irrelevant and unhelpful. They are irrelevant because they must proceed from unstated premises about the content or origin of some duty to warn.
The references to failure are unhelpful because they suggest that the only point at which some relevant warning could or should have been given is on board the aircraft. But if some warning was necessary or appropriate, it is not apparent why it should not have been given at a much earlier point of making arrangements to travel by air, rather than on board the aircraft. Further, reference to failure is unhelpful because it diverts attention from what it is that happened on board to what might have, could have, or perhaps should have happened there and why that should be so. If, as earlier indicated, it is appropriate to ask 'what happened on board?' the answer in this case is that the appellant alleges that nothing unexpected or unusual happened there."
- I agree with this reasoning and conclusion, and it applies equally in the present case. In these circumstances, it is sufficient to say that there was no unexpected or unusual event on board or during embarkation or disembarkation within the useful paraphrase suggested in Saks, however broadly that may be viewed, and that, viewing the matter in the simple terms of article 17 (which is the ultimate test), the situation does not fall within any ordinary or extended conception of "accident". For these reasons, and for those set out in the speeches of my noble and learned friends, Lord Scott of Foscote and Lord Steyn, I too would therefore dismiss this appeal.