Qantas Ltd & British Airways plc v Povey [2003] VSCA 227

"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."

6. In short the dispute is whether the word "accident" both connotes and requires a need for the plaintiff to prove a specific event, happening or incident, assuming it to be a cause of death, wounding or other bodily injury, in order to hold a carrier liable under the Convention, or whether it is sufficient to establish an omission to act or a failure to take some step or precaution together with other circumstances, which viewed as a whole are unusual or unexpected and are sufficient to amount to a cause of death or injury.

9. More particularly the respondent's claim that an "accident" had occurred took the following form. He alleged that he had suffered from deep venous thrombosis ("DVT") "during the course of or following the flights" conducted by the two appellants, saying that that condition and the resulting injuries were "caused by an accident which took place on board the aircraft upon which the flights were undertaken". Particulars of the accident were given in the proposed amendments to the statement of claim which, in the first place, picked up from existing particulars some five "flight conditions", being a confined and restricted physical environment in which the respondent was immobilised for a long time, impediments to his getting out of his seat, the supply of alcohol, tea and coffee to the respondent during the flights, discouragement from moving around the cabin and encouragement to remain seated and a failure to provide him with information or any warning about the risks of DVT and about measures to reduce those risks (which I shall call "the para.6 particulars")[3]. Secondly, it was alleged that the appellants knew that those flight conditions were capable of causing or increasing the risk of passengers' experiencing DVT in the absence of appropriate preventative measures, so it was unexpected and unusual that the plaintiff was subjected to those conditions. Finally, it was said that the respondent was not provided with any information or warning about the risk of DVT or of any measures to reduce the risk and that the respondent and other passengers expected to be provided with appropriate warnings, so that the failures to provide the warning and information were "unexpected and unusual". Otherwise the facts and proceedings leading to this appeal by both carriers from a refusal to strike out the respondent's claim are comprehensively set out in the judgment of Ashley, A.J.A.

12. In my opinion the word "accident" in Article 17 requires proof of a specific incident or occurrence which can be characterised as fortuitous. In other words it should be given its natural and ordinary meaning, namely an unexpected or unusual event, one which occurs by chance and without design, at least on the part of the person suffering death or injury. In the law the word "accident" has been given a number of meanings according to its context and it has perhaps been misused so as to cover the consequences of an untoward or fortuitous event, although rarely has it been used to comprehend occurrences which do not have the necessary element of chance, at least from the viewpoint of the victim. At one time the word ordinarily meant an event for which no reason could be advanced or blame ascribed, a truly fortuitous occurrence for which no person could be held responsible, as was once connoted by the coronial verdict of accidental death. More recently the word has been used to describe circumstances alleged to give rise to claims in negligence, where the essence of the allegation is not chance, but carelessness or worse. The meaning I prefer, however, is consistent with its meaning in "ordinary language" given by a majority of the High Court (Wilson, Deane and Dawson, JJ.) in Australian Casualty Co. Ltd. v. Federico[4]:

"It is something which happens without intention or design. When used with reference to something which causes injury, it means an unexpected and unintended mishap."

16. More importantly the meaning I would give to the word is consistent with the meaning given to the same word, as it appeared in identical terms in Article 17 of the original Warsaw Convention, in the opinion of the United States Supreme Court, delivered by O'Connor, J., in Air France v. Saks[5]:

"We conclude 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."

Her Honour proceeded immediately to say: "This definition should be flexibly applied after assessment of all the circumstances surrounding a passenger's injuries." But, in order to show what flexibility she had in mind, the judge gave examples both as to the place where the accident occurs[6] and as to events when the cause of the accident is deliberate, such as a terrorist attack or a hijacking where it may properly be said that, although the persons causing the accident may have acted deliberately, it is entirely fortuitous and unexpected from the viewpoint of the passenger. Her Honour then continued [7]:

"But 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."

As will be seen, the latter words have been misunderstood on occasions, but it should be noted that again the critical expression "an unusual or unexpected event" appears there in almost identical terms to those found in the definition already quoted.

25. Although the respondent appeared to challenge the applicability and reliability of the definition in Saks, it has been subsequently followed without criticism (as one might expect) in the United States[9] and it has also been cited with approval by superior courts in the United Kingdom, Canada and Australia. In the House of Lords it has been approved in Morris v. KLM Royal Dutch Airlines[10], as well as in the Court of Appeal in particular in relation to a claim for DVT: see The Deep Vein Thrombosis and Air Travel Group Litigation[11]. In Canada it has been approved in Quinn v. Canadian Airlines International Ltd.[12] and McDonald v. Korean Air[13], both in the Ontario Court of Appeal. In this Court it was likewise cited, albeit in relation to the equivalent provision (s.28) in Part IV of the Act: see Agtrack (N.T.) Pty. Ltd. v. Hatfield [14]. Moreover counsel for the respondent were unable to take us to any decision in any jurisdiction, common law or civil, where the definition given in Saks had been disapproved. Having regard to the obvious desirability that treaties should be read and construed consistently,[15] the Saks definition should be accepted, whatever its merits or supposed deficiencies.

28. The respondent contends that the word "accident" has a meaning which "is wide enough to include any kind of unfortunate or undesirable circumstances (or combination of positive and negative temporal conditions) whether brought about by an act or an omission". Counsel on his behalf insisted that the primary focus in deciding whether something was an accident lay in the element of mischance, mishap or misadventure "the product of or involving the miscalculation of forces, or inadvertence to them". This was not precisely the way in which the judge concluded that it was inappropriate to strike out the relevant paragraph of the statement of claim and the proposed particulars. His Honour clearly enough held that a mere failure to warn could not constitute an accident (see para.[35] of his judgment[16]), but he appeared to conclude that it was open to find an accident, within the definition in Saks, by showing that there was a failure to warn of the risks of DVT known to the appellants and to advise as to precautions which they knew passengers could take to minimise or eliminate those risks. In those circumstances his Honour held that a failure to warn would be but one, "even if the most important, of the many factors which the plaintiff includes in the particulars of the relevant accident as he argues it upon his expanded case": para.[44].

31. Doubtless it is desirable, as O'Connor, J. observed in Saks, to give the term "accident" a flexible meaning and one which is capable of adaptation to a variety of circumstances, but I fail to see how that which is now pleaded by way of particulars could amount to an "accident", for those matters amount to no more than a chain of circumstances, indeed, as argued, a "particular set of conditions", which could never be characterised as the kind of "event or happening" or "mishap" which can constitute an "accident". What one is looking for, in the ordinary acceptation of the term, is a mishap, a chance occurrence, a fortuitous and unexpected event. Moreover, in the present context one may not look individually at any one (or more) of those circumstances which might lead to a conclusion that an outcome is fortuitous or unexpected, unless that circumstance amounts to an act or event which is accidental. Nor can one look to a mere set of circumstances, howsoever fortuitous and unexpected the resulting physical condition or disorder might be, as in the chance incurring, unrelated to a specific act or event, of an injury or other medical condition.

34. In the first place, one must be cautious not to characterise the accident by reference to its alleged outcome, the unfortunate and unexpected incurring of DVT. That is not because in different contexts the word "accident" cannot in some circumstances apply to an injury which a person unexpectedly suffers: indeed, in many contexts of the law "accident" may well mean "accidental injury", as is the conventional shorthand in claims for negligence and in a large range of entitlements to compensation arising out of "injury by accident". What is presently clear is that that was not what was intended to be covered by Article 17, and I am yet to see any authority which suggests that it does, although some American decisions, to which I will have to refer briefly, have tended to slide towards that conclusion by a misunderstanding of what was said in Saks. From time to time one can place too fine a distinction upon the use of words or the order in which they appear, but O'Connor, J. was fully aware of this, concluding nevertheless that the word "cannot be stretched to impose carrier liability for injuries that are not caused by accidents"[17].

37. The respondent placed considerable weight on the meaning given to the word "accident" by their Lordships in the well-known decision of Fenton v. Thorley & Co. Ltd.[18], but that is to misunderstand what O'Connor, J. was saying about the observations which she cited from the speech of Lord Lindley. Although his Lordship defined the term in an unexceptionable way, in general by referring to its meaning "any unintended and unexpected occurrence", her Honour introduced her citation of the passage with these words[19]

"Second, the text of article 17 refers to an accident which caused the passenger's injury, and not to an accident which is the passenger's injury."

She then said that the word "accident" can be used in many senses, and followed by quoting the relevant passage from the speech of Lord Lindley which concluded [20]:

"But [the word accident] is often used to denote any unintended and 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 word `accident' is also often used to denote both the cause and the effect, no attempt being made to discriminate between them."

As to this O'Connor immediately observed [21]:

"In article 17, the drafters of the Warsaw Convention apparently did make an attempt to discriminate between `the cause and the effect'; they specified that air carriers would be liable if an accident caused the passenger's injury."

47. The distinction between cause and effect so far as it applies to accidents and consequential injuries has been recognised for many years, the traditional starting point being the decision in Fenton and in particular the speech of Lord Macnaghten who, distinguishing other collocations of the words "accident" and "injury", said this[22]:

"Now the expression `injury by accident' seems to me to be a compound expression. The words `by accident' are, I think, introduced parenthetically as it were to qualify the word `injury', confining it to a certain class of injuries, and excluding other classes, as, for instance, injuries by disease or injuries self-inflicted by design."

The parenthetical or adjectival use of "accident" in relation to claims for workers' compensation and the like based on the statutory requirement of proof of an "injury by accident" has been accepted for a century in the United Kingdom, Australia and the United States, at the least, and it has been well understood that the use of expressions such as "an injury caused by an accident" may well require proof separately of both accident and injury. It is sufficient to point out that the distinction has been repeated and reinforced time and time again by decisions in this country and in particular in the High Court: see e.g. McGuire v. Union Steamship Co. of New Zealand[23], when, even at that time, Knox, C.J. said that it must be taken as "settled by a long series of decisions" that the statutory expression meant "accidental injury"[24] and Isaacs and Rich, JJ. said that Lord Macnaghten's speech had by then "practically the force of an Act of Parliament"[25]. Again in Hetherington v. Amalgamated Collieries of W.A. Ltd.[26], another case of general importance on the meaning of the workers' compensation legislation, Latham, C.J. said[27], in referring to Fenton, that the phrase "injury by accident" meant accidental injury and that since that decision "it has not been necessary to show, first, that something to be described as an accident happened, and secondly, that something else, namely, an injury, was brought about or caused by that accident". It is clear that the other judgments in that case approached the matter on a similar basis but these are all reasons why it is necessary to be cautious in applying Fenton's case to the present, as O'Connor, J. clearly held in Saks. More recently, when a similar issue arose relating to the expression "bodily injury caused by an accident" in a disability policy, the High Court again recognised the distinction, although the majority took a somewhat different approach from that of the other members of the Court: see Federico, in the passage cited above in para.[4] esp. per Gibbs, C.J.[28] and per Brennan, J.[29]. See also the detailed and informative discussion in Sutton Insurance Law in Australia, 3rd ed., paras.10.56 - 10.87.

52. Even if there may be doubt as to the proper construction of certain statutes and insurance policies, I do not doubt that the use of the word "accident" in Article 17 is not parenthetical or adjectival. It is not a provision where words such as "caused by accident" or "caused by an accident" are inserted as merely descriptive of the word "injury", for there is a separate clause stating the circumstances under which liability arises expressed in terms "if the accident which caused the damage so sustained took place on board the aircraft etc.". In those cases where the word "accident" has been used in an adjectival sense it is usually clear that the conditions for liability depend on the time or place of the suffering of injury or at least it is sufficiently ambiguous for it to be said that it depends on the time and place of the "accidental injury", if so it should be properly construed.

55. In Article 17 there can be no ambiguity, for a condition of recovery is clearly proof that the posited accident has occurred at a prescribed time and place, in that the article requires the accident to take place either "on board the aircraft" or "in the course of any of the operations of embarking or disembarking". Whatever the latter expression may precisely mean, and it has posed difficulties over the years, liability must depend on proof of an accident at one of those prescribed places and during one of the stages of the journey (since a journey may take place over a number of legs). The article requires not that the injury should take place during one of those stages (or at those places), although it may be highly likely in most cases that it would, but that the accident should so occur. It is not analogous to some workers' compensation provisions whereby it was sufficient to prove an accidental injury so long as it "arose out of the employment", for the article is concerned with the activities of the carrier, so that liability must logically be confined to those places (or times) for which the carrier may be seen in broad terms to be responsible. It would be curious, especially having regard to the history of the Convention, if a carrier were responsible for a merely fortuitous injury arising in the course of a journey by aircraft, unless something which took place on or in close proximity to the aircraft led to the injury. I would reach this conclusion notwithstanding that Article 20 permits the carrier to establish that it has "taken all necessary measures to avoid the damage or that it was impossible for [it] to take such measures", for that language seems more apposite to the construction which I would put on the article than that which involves liability for any "accidental injury".[30]

58. Of course the respondent did not argue directly along such lines, for that would have been inconsistent with existing authority, though it was argued that the ratio of Saks was narrow and might be challenged hereafter. Rather counsel said that a broad meaning should be given to the word "accident" so as to create liability in respect of those injuries such that, though the injuries did not flow from a direct or sudden occurrence, the word could be treated as wide enough to comprehend a set of circumstances in which one or more activities of the carrier could be described as unexpected or fortuitous. I note that the words actually used in the respondent's submission to describe the required circumstances are "unfortunate or undesirable", neither of which really have the connotation, required by the definition in Saks, that they be "unexpected or unusual". That which is "undesirable" can be so characterised for a number of reasons, most of which would suggest some blame being ascribed to the carrier, whereas "unfortunate" has a pejorative tone but does not truly import the need to show a chance event as a cause of the injury. One may, however, assume that the onset of the plaintiff's illness in this case was one which was fortuitous and unexpected, and it is alleged that the injury was in fact suffered on the last leg of the flight back to Sydney. But that, as I have said, is not sufficient.

61. The second aspect of the judgment in Saks which the respondent relies on, therefore, was the third passage from Saks set out above[31]

in which O'Connor, J. said that the Court required "only that the passenger be able to prove that some link in the chain [of causes] was an unusual or unexpected event external to the passenger". It was argued that it is sufficient to show only that some element in the chain is unusual or unexpected, although it could not be characterised as a specific unexpected or unusual event or happening. As the opening words of the sentence indicate, however, it was directed to a perennial problem in causation, namely, that it is often difficult to identify (where that is required) precisely what is the "proximate cause" of a particular injury. That difficulty, nevertheless, as O'Connor, J. there stated, might properly be overcome by requiring that the required unexpected or unusual event should be only a cause of the subsequent injury. She was not laying down an alternative means of recovery by showing that a chain of circumstances could in some way be characterised as so unusual or unexpected as to convert them into an "accident": all she was recognising was that there could be a series of events and a set of circumstances which might lead to injury, but that amongst the facts and circumstances leading to the injury there should be an event "external to the passenger which could be characterised as unusual or unexpected". Thus the requirement for an "accident", in the correctly understood meaning of that term, remained as she had originally stated it. If some of the courts in the United States which have considered Saks have misunderstood what her Honour was saying, then I would not, with respect, accept their conclusions or their reasoning.

66. A third argument depended upon drawing a positive inference from another passage quoted above from the judgment of O'Connor, J.[32]

where she observed that, when an injury "indisputably results from the passenger's own internal reaction to the usual, normal and expected operation of the aircraft", Article 17 cannot apply because the injury has not been caused by an accident. From this negative proposition it was sought to infer that, if one could find an injury flowing from an unusual, or abnormal, or unexpected aspect of an aircraft's flight (or the embarking or disembarking of passengers from the flight), then that would be sufficient to amount to an "accident", notwithstanding that no specific incident or event could be isolated as one of the causes of the injury. So the argument proceeded to give examples of aspects of flights which were out of the ordinary which might be said to be a cause of injury. In particular it was said that the failure to warn of DVT and give necessary advice about it was here unusual, because knowledge by the appellants of DVT at the relevant time was alleged and it is usual for carriers to issue warnings and advice about risks of which they had knowledge.

71. I am afraid I cannot accept that what cannot be identified as a specific incident or event in relation to this or any other flight can properly be described as an "accident", within the meaning of Article 17. The allegations in substance do no more than state a failure to do something, and this cannot be characterised as an event or happening, whatever be the concomitant background to that failure to warn or advise. That is not to say that a failure to take a specific required step in the course of flying an aircraft, or in picking up or setting down passengers, cannot lead to an event or happening of the requisite unusual or unexpected kind and thus be an accident for the purpose of the article. A failure by a pilot to use some device in the expected and correct manner, such as a failure to let down the landing wheels or a chance omission to adjust the level of pressurisation, may lead, as has been held, to an accident contemplated by Article 17, but I would venture to suggest that it is not the failure to take the step which is properly to be characterised as an accident but rather its immediate and disastrous consequence whether that be the dangerous landing on the belly of the aircraft or an immediate unexpected and dangerous drop in pressurisation. It is a slide in reasoning to say that every failure to do that which a carrier ought to do necessarily amounts to an accident, although it may frequently lead to such an event. A complete failure to do that which one might otherwise be under a duty to do may well be negligent and so "unexpected" by a passenger or any objective observer, but that will not in itself necessarily amount to an event or occurrence of the required kind. The question in each case still is whether there has been an accident. That is clear from the context in which the passage appears in O'Connor, J.'s judgment and from the tenor of her remarks generally.

74. Moreover it is hard to see how a failure to warn or advise passengers, a "non-event" as it were, can ever constitute an accident within the meaning of the article, notwithstanding the presence of surrounding circumstances which would make the failure unexpected or unusual. That again flows, as I have already pointed out, from the requirement to show that the accident "took place on board the aircraft" or in embarking or disembarking from it. Whatever might be said about failures to take specific actions in the course of flying an aeroplane, which will usually have disastrous consequences which themselves may easily be characterised as events or happenings, a failure to warn or advise is the non-performance of an expected duty which by its nature ordinarily is continuous. Obviously a failure to warn or advise would comprehend a period including the embarkation stage and the flight itself, but, if the respondent's contentions be correct, surely in circumstances such as the present and in relation to a condition such as DVT, one would fairly expect the warning and advice to be given, at least in the first place, well before the embarkation and flight stages. The embarkation stage is ordinarily quite short, though it might be sufficient to cover the waiting lounge, but the practicality and usefulness of a warning given then, and certainly of one given on board the aeroplane, would be relatively slight, unless the advice was of a kind which was practical and would invariably prevent the possibility of suffering DVT as a result of a flight. The need for a warning, if there be one (for public knowledge may make that now otiose), must arise at a much earlier stage, indeed before the ticket is purchased or at least before a passenger starts to embark, so that the passenger is given a choice whether or not to run the risk. Is the failure then to warn subsumed in the continuous failure to warn thereafter, including during the flight? As I have already stated, the object of Article 17 in defining the stage at which an accident occurs is surely to make clear that the carrier is liable only for events which take place at an identifiable time, i.e. "on board the aircraft" or during embarkation or disembarkation.

77. Against this it was argued, as it has been in a number of the reported decisions, that to insist upon some event or action in the sense of an identifiable act or action on the part of the carrier or some person other than the passenger, as opposed to a failure to carry out one or more acts whether requested or not, would run the risk of inserting in the article the despised distinction between action and non-action, which from time to time has obtruded itself into the law of torts. But that is exactly what Article 17 on its face, particularly in the light of its gestation, was intended to do. It picked on the simple French (and English at that time) concept of an accident, something which would be capable of resolving the causative enquiry between the accident and the consequent injury required for liability to attach. Even at the time the Convention was entered into it was the concept of negligence which connoted, if not by then denoted, that what was required to be established in tort was a failure to take reasonable care.

80. The battles fought about action and inaction have nothing to do, essentially, with Article 17, for the posited event there is an "accident". There is no need to incorporate into that undefined word concepts which are alien to the required precondition of an event or an occurrence of the requisite kind. Moreover the word in fact chosen was "accident", with its connotation of "that which befalls one". One may concede that usage in the English language tends to become less precise, dependent on the whims and usages of the day, but the same cannot be said[33] of the French language where a definition of the word "accident" in that language has been translated for the benefit of the Court as meaning a "fortuitous and unfortunate event, causing physical injury or material damage". I cannot see that mere non-action or any failure to act, however egregious the negligence involved, should in itself be characterised as an accident. Sometimes inaction, indeed gross negligence, may lead to an event which may be characterised as an accident, but that is quite a different thing from mere inaction or any negligent failure to act, however serious. Thus significance has been placed in some American cases on one particular example of gross negligence by a flight attendant which resulted in severe scalding of a passenger.[34] But it was not the negligence which amounted to the accident in that case, it was the physical consequence or consequential event, in the example given, of the application of a boiling compress by the attendant on board the plane, which amounted to the accident. Of course, for liability to attach under the Convention, there was no need at that stage to concern oneself with negligence at all, for the Court was not required to concern itself with the reason for the application of the compress, only the fact of the unexpected application, just as in any other case it is the accident, not its cause, which will give rise to liability if that itself causes injury. So enquiries into negligence are not only misconceived, but in my view mischievous, inasmuch as it diverts attention from the real question, except where intent to harm or recklessness under Article 25 is also alleged.

83. Fourthly, the respondent relied on contextual indications from other relevant articles. In particular he points to Article 25 of the Convention which reads:

"In the carriage of passengers and baggage, the limits of liability specified in article 22 shall not apply if it is proved that the damage resulted from an act or omission of the carrier, his servants or agents, done with intent to cause damage or recklessly and with knowledge that damage would probably result; provided that, in the case of such act or omission of a servant or agent, it is also proved that he was acting within the scope of his employment." (Emphasis added.)

Article 22, which need not be set out, is the article which imposes specific monetary limits for recovery of damages but includes limits not merely referable to Article 17, but also under Article 18, which gives a right to damages for loss of or damage to baggage during the whole of the period of "carriage", and under Article 19, which makes the carrier liable for damage "occasioned by delay" to either passengers or baggage. It was argued that the word "omission" in Article 25 shows that an "accident" within Article 17 can comprehend such an omission, but in my opinion that is not established by the language used in Article 25. The article is dealing with particular exceptions to the limits laid down by Article 22 and thus it requires proof that the damage resulted from deliberate or reckless conduct on the part of the carrier or its agent. It does not require any separate enquiry whether an accident has occurred, for that must be assumed for the article to have any application to claims proved under Article 17, but looks at the reasons behind the accident which otherwise might be characterised as fortuitous and unexpected. It is thus concentrating on the cause of the accident, not on the accident itself, nor on its consequences, for it is seeking to impose additional liability where the cause of the disaster can be fairly sheeted home to carrier or agent. Thus a reckless omission may give rise to an accident, but that does not mean that the accident itself may be constituted by an omission. Moreover the limits in Article 22 apply not only to Article 17 but to Articles 18 and 19 and clearly enough "delay" in Article 19 might well come about by reason of an omission, for liability under that article is also intended potentially to be excepted from the limits in Article 22. No inference of the kind asserted can fairly be drawn, therefore, from the language of Article 25.

88. Parenthetically I should observe that one of the few matters upon which I agree with the contentions for the respondent is that the "unexpected or unusual" character of the event or happening should be determined from the viewpoint of the passenger. It must be accidental from the viewpoint of a passenger by reason of the fact that to him or her the event or happening is unexpected or unusual, whatever may be in the mind of the carrier or anybody else who perpetrates the act or event which is said to amount to an accident. That is an aspect of what is intended by the word "accident" which is not unlike that which is assumed in compensation and insurance law, for there would be little point in giving a remedy if the carrier could avoid liability by saying that to it the matter was usual, commonplace or merely within its expectations. Such a conclusion is consistent with the reasoning of O'Connor, J. who gave approval[35] to cases where the acts in question were the deliberate acts of others. Nor should there be some objective test based on what might generally thought to be unusual or unexpected, although the test as to what is "unusual or unexpected" is not to be determined by reference to the individual passenger; rather it should be ascertained from the viewpoint of an ordinary, reasonable passenger.

91. Fifthly, it was said that Saks stood only for what did not amount to an "accident" within the meaning of the article. In other words, the case was directed to circumstances where the injury was suffered by the plaintiff by way of severe pressure and pain in her ear occurring on board the aircraft but which was conceded to be no more than the passenger's own "internal reaction to the usual, normal and expected operation of the aircraft". It was, however, there alleged that that was an occurrence which consisted in the normal cabin pressure changes which the lower court held either individually or collectively could amount to an accident of the relevant kind. But that was why O'Connor, J. said that, where an injury results from the passenger's internal reaction to such usual, normal or expected events, it has not been caused by an accident, because, even from the viewpoint of a reasonable passenger, it could not for that reason be so characterised. It was just one of the natural consequences of a flight. The injury itself may have been unexpected or unusual but that was why her Honour spent so much time pointing out the distinction in the article between "injury" and "accident". So it was necessary to conclude that what must be shown is an unexpected or unusual "event or happening that is external to the passenger". The relevant case law, the history of the Convention and the question of causation were each discussed by her in terms which lead me to the conclusion that her definition was directed[36] just as much to the event or happening which constitutes an accident as to the other negative issue raised before the Court. There is no reason in my opinion to limit the ratio of Saks.

94. Next it was said that this Court should not follow the recent decision of the English Court of Appeal in the DVT Group Litigation, but should prefer the decisions given in a number of U.S. jurisdictions, notably Husain v. Olympic Airways[37] and Fulop v. Malev Hungarian Airlines[38].

97. I do not intend to canvass the wide variety of circumstances considered in the many decisions cited to this Court from the United States or in the further many cases themselves cited in those decisions. On occasions they have led to interpretations of the Warsaw Convention which, with the greatest respect, seem to have forgotten the language used there and the careful analysis by O'Connor, J. Regrettably, as I have already suggested, the reasoning seems in many cases to import notions of fault and negligence into an area where, with two very particular exceptions, it is irrelevant. Those exceptions relate to contributory negligence (Article 21) and to the removal of the limit in amount of damages under Article 25, where in effect gross negligence is proved. Each of those articles assumes that liability already has been found against the carrier.

100. What apparently has led the United States' courts down this sterile and confusing path was O'Connor, J.'s correct but misunderstood reference to the need to show that the activating and causative "event or happening" is "unexpected or unusual". This has been turned, by some courts, especially those deciding Husain, into an enquiry whether the injury has resulted from "unexpected or unusual" behaviour on the part of the defendant airline and more particularly of that of its staff. But "behaviour", whether or not negligent, may not comprehend any event, happening or "accident" at all, more particularly where that behaviour consists merely in a failure to take expected steps or steps acceptable, or which otherwise should be required, in the industry. Although such behaviour is to be condemned, as may be discerned from the colourful language used in a number of the judgments, it does not answer the question whether there has been a defined event or happening which amounts to an accident within the meaning of the convention, however "flexibly" that should be applied. Too often in the cases the test of what is unusual or unexpected has been answered by reference to a failure to comply with the "recognised", "controlling", or "industry" standard of care of an airline, or its flight attendants or other staff: see the various expressions used in Husain[39]. The Court there also relied on what was described as a "failure to adequately respond"[40] and the "blatant disregard of industry standards"[41] But failures to respond do not necessarily result in "accidents", though frequently, as may be seen from the law of torts, they lead to the suffering of injury. I believe some courts have been influenced by cases where failures of that kind have in fact led to accidents, such as where there has been a sudden diving of the aeroplane or the loss of pressure or even, as earlier noted, the application of a scalding compress, but colourful descriptions of the events are unnecessary where the enquiry is as to whether there has been an accident and in particular as to whether the event is unexpected or unusual.

103. In summary, the word "unexpected" (and the word "unusual") was used by O'Connor, J., in my opinion, to reflect the obvious connotation of the word "accident", namely, that the "event or happening" should be a fortuitous incident, that is, one which happens by chance and not by design, at least of the person who suffers the consequences. What is "expected" or "unexpected" should have nothing to do, in this context, with what one might expect in an ideal world, where nobody acts carelessly and where the standards of reasonable care are observed. It should be concerned with what actually happens in the ordinary course of events and whether in the conduct of carriage by air one might fairly contemplate or expect that a particular event or incident might happen, having regard to the way in which flights and related activities are ordinarily conducted. It is a simple factual enquiry, at least it ought to be, and no issue should arise as to what might be expected in an ideal, or at least a non-negligent, world. That is why concepts of negligence are irrelevant. As O'Connor, J. said[42], Article 17 requires "an inquiry into the nature of the event which caused the injury rather than the care taken by the airline to avert the injury". That is one of the reasons why I prefer the clear and simple conclusions of the English Court of Appeal in the DVT Group Litigation case. One may regret that so few changes have been made to a convention negotiated over 70 years ago, but it was worked out as a compromise in light of existing circumstances and the laws of many countries at that time. The benefit of obtaining recovery without proof of fault, albeit partly qualified and up to fixed limits, was doubtless seen by the signatories as a fair trade-off for avoiding potential liability in more obscure cases based on allegations of negligence, faute or the like. It may in due course be renegotiated[43] but that had not been achieved by the time of the Montreal No. 4 Convention in 1975 with which we are here concerned.

Application of Article 17 to Alleged Circumstances

107. One may accept that it is inappropriate to strike out a pleading or dismiss it for failing to disclose a cause of action unless that is clearly required by law on consideration of the alleged facts and circumstances. Often it is said, and it is unnecessary to repeat the authorities, that it is better for a court and in particular an appellate court to know precisely what is the evidence before concluding that that evidence is insufficient to amount to a claimed cause of action. The present case may, on its surface, appear to raise some circumstances which would invoke the application of that rule of practice. On closer analysis, however, I do not think it does.

110. The allegations started out in relatively simple terms largely derived from a description of "the flight conditions" particularised in five different but cumulative ways and a bald allegation that there was "an accident which took place on board the aircraft". Insofar as the description of the flight conditions referred to events or activities, such as the supply of beverages and discouraging the plaintiff from moving around the cabin, they could hardly, on the widest of interpretations, come within the expression "unexpected or unusual event or happening" used by O'Connor, J. in Air France v. Saks. So far as the ordinary conduct of flights are concerned those kinds of activities can only be described as common and mundane, whatever vices they might otherwise have. To this was added in paragraph (e) an allegation that the plaintiff had not been "provided with any information or warning about the risk of DVT" or about measures to reduce that risk. That factor on its own certainly could not amount to an accident for not only did it refer to a non-event and mere inaction but it described a state of affairs which did not change from the start to the end of each flight. That was recognised sufficiently by the learned primary judge for him to conclude that it could not amount to an accident on its own. In any event, the particulars were described in the pleading merely as "the flight conditions", albeit that it was said that those "conditions" and procedures caused the plaintiff's injury. On their own or in combination they seem far removed from the description of an accident given in Air France v. Saks for they are merely the broad description of circumstances attending the flights.

113. Both judge and the respondents on the hearing of this appeal placed greater reliance on the particulars to paragraph 21 which were later supplied, although they do not presently form part of the statement of claim, and which are set out in full in the judgement of Ashley, A.J.A. at para.[61]. But those particulars add nothing to the known circumstances except to the extent that they seek to characterise the appellants' conduct and, in particular, seek to characterise the conditions and the carriers' inactivity as "unexpected and unusual". To my way of thinking those allegations seek merely to add to the existing claim by characterising the behaviour of the airlines as blameworthy and thence to treat what they had done as unexpected because they were "expected" in the light of those circumstances to have done better and to have passed on their knowledge of the connection between the flight conditions and DVT by telling passengers of that fact and by giving warnings and information about the risks that they were exposed to.

116. One is left at the end of the day, notwithstanding the added particulars to paragraph 21, with a description of the conditions and circumstances of the flight rather than anything which could in ordinary language be described as an event, occurrence and more particularly an "accident". Nothing occurred on the flight which changed the conditions in which the passengers were placed when the flight commenced. As I understand it, what is said is that, by using the description "unexpected and unusual", the U.S. Supreme Court was implying that anything about a flight which satisfied that description would be an accident within the meaning of the article. No such thing was said, nor could one fairly believe that the Court intended to say that. As I have pointed out, the words were used, not on their own, but in the phrase "an unexpected or unusual event or happening". Now doubtless the English language has been distorted sufficiently for the words "event" or "happening" to have acquired secondary meanings to refer to occasions which extend over a significant period of time. But that is not what was intended in Article 17, nor by what O'Connor, J. said in Air France v. Saks. One must resist the temptation to read the word and O'Connor, J.'s explanation as comprehending an "unusual state of affairs", as a number of the U.S. authorities appear to do.[44] Indeed, as explained by O'Connor, J. there was good reason for her Honour to describe the matter in terms of an event or happening and to exclude even the word "occurrence", because of the contrast with the use of that word in Article 18.[45]

119. The respondent contends that, even if total inactivity might not come within the article, inactivity together with other factors may produce a set of circumstances which could be described as an "accident". Consistently with what is said in plain terms in the DVT Group Litigation, I consider that no accident can be said to have occurred merely because a set of circumstances or conditions exist which may be so unusual as to call for criticism and which would be compensable outside the Warsaw and Montreal Conventions. One must avoid the temptation of using the interpreted phrase devised by O'Connor, J. as the words of the Convention. It is not that O'Connor, J. was in error or that relevantly her definition does not provide great assistance: it is simply that one cannot take those words and seek to place a further construction upon them without having regard to Article 17 itself. In the first place her Honour did not use the words "unexpected or unusual" on their own, so that we are not concerned with an unexpected or unusual injury, nor with unexpected or unusual behaviour of an airline unrelated to a specific incident, as is implicitly suggested in some of the U.S. authorities and in the approach of the respondent.[46]

122. Secondly O'Connor, J. used those words as part of the expression "an unexpected or unusual event or happening", so that to exclude or overlook the last three words is to distort what her Honour intended. Even then she was merely using synonyms as a means of emphasising the chance element clearly connoted by the word "accident". The "event or happening" still has to be an accident and moreover it must take "place on the board the aircraft or in the course of any of the operations of embarking or disembarking". When one looks at those words together, as I have said previously, one is looking to something which occurred at a specified place or time and not otherwise. To speak of conditions or circumstances "taking place" would, I suggest, be an artificial and curious use of language, especially when they consist of inaction. What those drafting the article were concerned with was proof of a specific incident of a fortuitous kind which itself causes injury.

125. Finally it should be observed that it is the accident which must take place on board the aircraft or during embarkation or disembarkation and not the injury suffered.[47] If I may say so, the concentration on the airline's behaviour and whether it is expected or unexpected, usual or unusual, misses the point. The article is directed to the occurrence of an accident which must take place on board the aircraft etc., but it is irrelevant for the purposes of Article 17 who caused it. The strict or presumptive liability of the article does not seek an answer to that question, so long as it occurs in one of the designated places. That is why the acts of hijackers satisfy the article, although there may be no responsibility in common law terms for their activities on the part of the airline. Moreover one should not be misled by O'Connor, J.'s careful exclusion of those injuries which result from "the passenger's own internal reaction" to the normal operation of aircraft and the requirement that the accident should be "external to the passenger", because that is merely reflecting the need under the article to find both an accident and a consequent injury. The article clearly did not give recovery merely by reason of the suffering of an injury: the precondition for liability is the occurring of an accident, so that injury resulting from the passenger's own health or condition is necessarily excluded in the absence of the required accident on board the plane etc. Likewise the word "accident" clearly connotes that the event or incident cannot be the result of activities by the passenger, for self-harm or the like would not have the necessary element of chance.

128. These broad observations, admittedly repetitive of matters stated earlier, is intended to provide an explanation why "accident" cannot ordinarily consist of circumstances or conditions, whatever may have been said in certain United States courts. I deal with one matter at this stage, merely to put it to one side. It is not unusual for an accident of the required kind to take place in circumstances where that accident is likely to cause injury either generally to passengers or to one or more specific passengers by reason of circumstances relating to them. The U.S. Supreme Court adopted the concepts of causation familiar to common law courts[48], so it is sufficient that the particular accident is merely some link in the chain leading to an injury, in other words that it was a cause of the injury. The surrounding circumstances may thus have had a bearing on the likelihood of suffering injury and in some decisions considerable weight has been placed upon them, but in the end it is sufficient that there be an accident which is but one cause of the injury.

131. What is the more difficult case, and several have come before the Courts and have been frequently the subject of discussion, is where there has been a defective piece of equipment which has either become defective without immediately causing injury or, in the more extreme of cases, where the defect has existed from the outset to the end of the flight, but some injury has been suffered in the course of the flight because of the conditions produced by the defective piece of equipment or the like. There it was said that surely one does not have to look for a particular event or incident. But if an aberrant pilot points the controls of a plane towards the ground, though it is easy to characterise his behaviour as unusual or unexpected, no accident in fact takes place until the plane hits the ground.[49] Terrible though that may be, one is simply not concerned with the reasons for the behaviour at this stage of the enquiry, so that liability would likewise result even if the pilot is totally without blame.[50] Indeed, one could see why, when the article was first drafted, this was a desired outcome, for black boxes and the like did not appear on the scene for many years. The same, though with some more difficult aspects of proof, may be said of a wrongful alteration to the pressurisation controls. It must be the change in pressure to that which is harmful, rather than the movement of the controls themselves, which amounts to the accident. Sometimes the two will be virtually coincidental, so that the cause of the accident and the accident itself will be seen to be relevantly causative of the injury and defining the accident more precisely will be of little consequence. Where the defect has come into existence before the aircraft takes off, there may be greater problems, although it must be remembered that in the case of a crash or similar accident there will be no need under Article 17 itself for the passenger to track down the reason: claimants will have no difficulty in identifying the relevant accident. As to pressurisation itself, the accident may come about only when the pressure reaches a dangerous stage, that is, the pressure may be at a tolerable level for some part of the journey, say immediately after take off, but there will be a point reached when the pressure becomes dangerous and that can fairly be characterised as an accident, whatsoever and whosoever was originally responsible for it. Air France v. Saks was itself a case about pressurisation, but the injury was confined to the plaintiff and no deleterious change in pressure was established.

134. Moreover it is the causative event that is critical, not the unusual or unexpected behaviour, if so it may be described, of those responsible for adjusting the plane's controls. The event which is unusual or unexpected is the crash or the pressure reaching a stage where it is dangerous to human safety. The concomitant circumstances will be largely irrelevant, although one may find a series of events each capable of being a cause of the injury and there would be no reason to deny recovery for that reason, so long as any one or more of them was an incident which was in fact a cause of the passenger's injury.

137. I mention only one other example where unfortunately common law courts have picked on inexcusable behaviour and treated that as amounting to an unexpected or unusual "event". A number of decisions have put forward examples relating to diversion of planes to deal with conditions suffered by passengers. They have posited a passenger collapsing, for no reason connected with the flight of the aircraft, and requiring medical assistance not available on the aircraft.[51] It has been said that a failure to divert in these circumstances amounts to an accident, indeed in one case it was said that "the notion that this is not an unusual event is staggering".[52] But there the injury must have already occurred, the example given involving a person who had "ceased breathing through no initial accident"; so why there should be any further enquiry into the issue whether there was or was not an accident is difficult to follow. Doubtless in other circumstances and applying common law rules (as in intrastate flights involving non-commercial aircraft), there may well be some cause of action. Even if one could say that there was a further cause of the damage sustained in the event of death, it is hard to see why a refusal to divert to another airport amounts to an accident in the common and proper acceptation of that term. I concede that the U.S. cases (or some of them) suggest that "refusal" to divert may amount to an accident. Perhaps, on occasions, the surrounding events may amount to what can accurately be described as an "event or happening", but ordinarily inactivity cannot amount to an "accident", however blameworthy it may be. Hard cases make bad law, the Convention ought perhaps to be amended, but pejorative language directed to the conduct of persons in control or on board does not turn inaction into an accident.

140. In the present case the respondent's case, if it goes at all beyond non-action, must consist in the describing of the alleged circumstances in coloured terms, such that the alleged behaviour becomes unexpected or unusual, but it does not, for the reasons I have endeavoured to state, turn it into an accident, i.e. a fact, event, happening or incident of a fortuitous kind for which Article 17 otherwise provides compensation, essentially without fault or at least on a presumptive basis.

143. I am grateful to Ashley, A.J.A. for his careful and detailed exposition of the arguments and the authorities. Although I agree with his analysis of many of those cases, I cannot take the same approach to them, for I consider that several depart from the language of the Convention. I would therefore prefer, with only minor qualifications, the straightforward approach of the English Court of Appeal in the DVT Group Litigation, as described by Chernov, J.A. I agree with Ashley, A.J.A.'s conclusion as to total inactivity, but, with respect, cannot accept his conclusions about the additional particulars and the potential cause of action therein adumbrated. In the end I am in no doubt as to the proper outcome of the appellants' pleading summonses.

146. Consequently, I would allow the appeals and strike out the respondent's statement of claim and make consequential orders directed to bringing the action to an end. If the result seems harsh, and we have no means of testing how the allegations would stand up at common law, then that is the result of this nation's choosing to enter into specific treaty arrangements as are embodied in the Convention, and inevitably they embody compromises designed to benefit some and deny rights to others.

151. I have had the advantage of reading the draft judgment of Ashley, A.J.A. and agree, for the reasons given by his Honour, that it is not reasonably arguable that the respondents' failure to warn the appellant and the other passengers that they risked suffering deep vein thrombosis during the flights in question, where the passengers were not seized of that information, amounted to an "accident" for the purposes of Article 17 of the Warsaw Convention ("the Convention"). This leaves for resolution the question whether it is reasonably arguable that the expanded flight conditions, in particular, discouraging the appellant from moving about the aircraft when it was in flight and, inferentially, encouraging him to remain in his cramped seat, and offering him beverages, in circumstances where, contrary to the "usual" practice, the airlines failed to pass on their relevant knowledge to passengers who were ignorant of the risk, amounted to an "accident" which caused the appellant's injury.

154. A starting point in seeking to establish the meaning of the word "accident" in Article 17 is to ascertain its usual meaning. In ordinary parlance**,** the word suggests a happening or an event that is unforseen or unexpected.[53] Such a definition, however, may not be determinative of the meaning of the word for the purposes of Article 17 simply because "accident" may have a different connotation because of its context. But if one examines Article 17, it seems apparent that, on its proper construction, the meaning to be ascribed to the term "accident" is that which I have just outlined. Indeed, that "accident" in Article 17 contemplates an event or happening that is unexpected or unusual, is confirmed by appellate courts in other jurisdictions that have interpreted the provision in light of, inter alia, its international

operation and the desirability for uniformity in its construction.[54] In that context, it has been recognised in the United States[55] and in England[56] by, inter alia, the House of Lords in Morris v. KLM Royal Dutch Airlines[57] and by the Court of Appeal in The Deep Vein Thrombosis and Air Travel Group Litigation[58] ("the DVT case") that Air France v. Saks[59] provides a useful guide for determining whether an injury was caused by an "accident" within the meaning of Article 17 notwithstanding that the decision does not seek to replace "accident" in the article with the terms used in the judgment to describe its meaning, or provide an exhaustive definition of "accident".

158. In delivering the Opinion of the Supreme Court in that case, O'Connor, J. emphasised[60] that the essential enquiry is whether the accident caused the injury and not whether the accident is the injury. And it is the cause of the injury that must satisfy the definition of "accident" in Article 17. After examining closely the language of the article in the context of the Convention, the evolution of its drafts before it was finally settled, its character and purpose, the ordinary and dictionary meanings ascribed to the word "accident" and the decisions of, inter alia, appellate courts in the United States and elsewhere as to the interpretation and operation of Article 17, her Honour concluded[61] that "accident" suggests an unexpected or unusual event or happening on board of the aircraft that is external to the passenger which causes the injury or death of the passenger. The judge emphasised that, where the injury or death "results from the passenger's own internal reaction to the usual, or normal and expected operations of the aircraft, it has not been caused by an accident [within the meaning of Article 17]".

161. The reasoning in Saks was approved by the House of Lords in Morris and by the Court of Appeal in the DVT case. Thus, for example, in the DVT case, the Master of the Rolls said[62] that the natural meaning of "accident" suggests an untoward event which impacts on the body in a manner which causes death or injury. His Lordship cited with approval the meaning attributed to the word "event" by Lord Mustill in Axa Reinsurance (UK) Plc v. Field[63], namely, "something which happens at a particular time, at a particular place, in a particular way."

164. The agreed matrix of facts, the basis upon which the Court of Appeal was asked to determine whether it was reasonably arguable that the injuries of the plaintiffs were caused by an "accident", emphasised the cramped seating configuration in the cabin of the aircraft in which the plaintiffs were transported and the relevant knowledge of the airlines, rather than, as here, acts of their crew members in the course of the flight. It seems clear enough that it was accepted by the plaintiffs in that case that, for the purposes of the argument, this seating configuration and the circumstances relating to it were typical and usual features of air travel at the relevant time[64]. In that context, the Master of the Rolls considered[65] that such features of the aircraft, and the subjecting of its passengers to them, were not capable of amounting to an "event" that satisfied the first part of the meaning of accident for the purpose of Article 17 to which I have already referred. It followed, said his Lordship, that the question whether there was an "unexpected or unusual" event did not arise for consideration. But even if such features of the aircraft could be said to constitute an "event", said the Master of the Rolls, the fact that they were reflective of what was typical and accepted in the industry, precluded them from being unexpected or unusual. The Master of the Rolls considered that nothing happened that involved the actions of anyone other than the plaintiff and his reaction to the normal flying conditions - there was no unexpected or unusual event or happening. Moreover, said his Lordship, the failure by the airlines to advise the passengers of the relevant risks, of itself, could not be said to fall within the definition of accident, a mere failure to warn, being mere inaction or inertia, could not amount to an accident.

167. Notwithstanding the analysis of the large number of authorities that have been cited to us on the issue now under consideration, as Judge and Kay, L.JJ. recognised in the DVT case[66], at the end of the day, the question is whether it is reasonably arguable that the appellant can establish that, on any recognised meaning of the word, an accident on board the aircraft caused his injury. In my view, the answer is, no, and I have come to this conclusion for the following reasons. First, I consider that the acts alleged do not constitute an event or happening as those terms were interpreted in the DVT case and AXA. In my view, on a proper reading of the pleadings, the only acts of the respondents that could be said to have amounted to an event or a happening were the discouragement of his moving around the cabin of the aircraft and the offers made to him of beverages for consumption during flight. Such acts, I consider, describe continued circumstances which were part of the standard operation of the airlines and do not amount to an event or happening. Secondly, even if the acts in question could be properly described as an event or happening, it seems to me that, on a proper characterisation of the pleaded facts, the appellant's injury resulted from his own reaction to standard flight conditions and was not caused by an event or happening on board the aeroplane that was external to him. Thirdly, even if I am wrong in both of my above views and the acts of discouragement and offers of beverages could be said to have been an event or happening for relevant purposes that caused the injury, they were plainly not unexpected or unusual. It is not said that such acts were contrary to industry practice or the airlines' own procedures, such as arguably arose in Husain v. Olympic Airways[67], McCaskey v. Continental Airlines[68] and Fulop v. Malev Hungarian Airlines[69] and which enabled the plaintiffs in those cases to cross the prima facie threshold. Here, the alleged acts are said in the pleadings to have been standard travel procedures on such flights. In those circumstances such conduct could not be reasonably characterised as amounting to an unexpected or unusual event or happening, even where it was performed in the context where the airlines failed to inform the passengers of the risk of developing deep vein thrombosis during such a flight.

170. In the circumstances, I would uphold the appeals.

176. These appeals concern the meaning of Art. 17 of the Warsaw Convention 1929 ("the Convention") as amended by the Hague Protocol 1955 and by Protocol No. 4 of Montreal 1975.[70] The Convention as thus amended (Montreal No.4) sets out rules relating to international air travel, particularly for present purposes the circumstances in which an injured passenger may recover damages from an airline.

179. By s. 25K(1) of the Civil Aviation (Carrier's Liability) Act 1989 (Cth) ("the Carriers Act") the provisions of Montreal No. 4 have force of law in Australia in relation to any relevant carriage by air. By s. 25L of the Carriers Act, ss. 35-39 of that Act are made applicable to carriage under Montreal No. 4. One consequence is that s. 36 applies to such carriage. By that section the liability of a carrier "in respect of personal injury suffered by a passenger" is, subject to a presently irrelevant exception, "in substitution for any civil liability of the carrier under any other law in respect of the injury". It is not in debate that in consequence of these sections Art. 17 of Montreal No. 4 defines the circumstances in which the appellants, Qantas Airways Ltd and British Airways Pty Ltd ("Qantas" and "BA", collectively "the airlines") could be liable to the respondent, Mr Brian Povey, for bodily injury which he allegedly suffered by reason of the incidents of a return journey between Sydney and London which he undertook between 15 and 20 February 2000.

182. By his statement of claim dated 21 August 2001, Mr Povey made claim against the airlines and the Civil Aviation Safety Authority ("CASA"). The last-mentioned is not a party to the appeal. The gist of his claim, so far as it concerned the airlines, was that during the course of or following the flights, which he undertook travelling economy class, he suffered from deep vein thrombosis ("DVT"), this leading on to pulmonary embolism and to other serious, long-term consequences. His injury, he alleged, was:

"caused by the conditions of and procedures relating to passenger travel upon the flights."

187. He particularised the flight conditions this way:

"The flight conditions included: (a) a confined and restricted physical environment in which the Plaintiff was immobilised for long periods of time in a seated position; (b) impediments to the Plaintiff getting out of his seat during the flights; (c) the offer and supply of alcoholic beverages, tea and coffee to the Plaintiff during the flights; (d) discouraging the Plaintiff from moving around the cabin of the aircraft and encouraging the Plaintiff to remain seated during the flights; (e) The Plaintiff not being provided with any information or warning about the risk of DVT or information about the measures which the Plaintiff could take to reduce such risk."[71]

191. Having pleaded the applicability of the Warsaw Convention as amended at The Hague - this was an error, but nothing turns on it - he alleged by paragraph 21 of the statement of claim that:

"The DVT and the injuries resulting therefrom constituted bodily injuries caused by an accident which took place on board the aircraft upon which the flights were undertaken within the terms of Art. 17 of the Convention."

195. Qantas sought further and better particulars of paragraph 21. By particulars dated 12 October 2001, Mr Povey alleged that the accident causing the injury comprised the flight conditions; and that it occurred for the duration of the flights to and from the UK. He alleged also that he first experienced symptoms of DVT on the return journey whilst travelling between Kuala Lumpar, a stopover point, and Sydney[72].

198. BA also sought further and better particulars of the statement of claim, in particular of paragraph 11(a) by which Mr Povey pleaded, against CASA, that it knew or ought to have known

"the flight conditions, which were the standard conditions of and procedures relating to passenger travel on prescribed flights."

202. By further and better particulars dated 13 February 2002, Mr Povey said that the "standard conditions on prescribed flights" referred to in paragraph 11(a) of the statement of claim were the same as the "flight conditions" defined in paragraph 6 of the statement of claim.[73]

205. Each of the airlines filed a defence which denied, inter alia, if Mr Povey suffered from DVT and any injuries relating therefrom, that the same were caused by any accident which took place on a flight conducted by it (or in the case of Qantas, in the course of a flight for which it could be liable).[74]

208. There the matter stood when on 26 March 2002 the airlines filed summonses which sought, in the alternative:

* judgment against the plaintiff, or a permanent stay of the proceeding under R 23.01 of Chapter 1 of the Rules;

* the striking out of the pertinent paragraphs of the statement of claim.

213. The argument on the applications was conducted on the footing that a number of factual matters alleged by the respondent, though in fact in dispute on the pleadings, should be taken to be established. The issue debated was whether the respondent had pleaded a case, in substance rather than in form, which could establish that the damage he had allegedly sustained was caused by an accident which took place on board the aircraft. That picks up the critical language of Art. 17, which as it appears in Montreal No 4_._reads as follows:

"The carrier is liable for damages 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."

217. Bongiorno J concluded that the statement of claim did not in substance allege anything that could constitute an accident. But in light of foreshadowed further particularisation of the alleged accident, particularisation which he considered would arguably allege an accident, he dismissed the summonses and gave the plaintiff an opportunity to re-plead. It is not in debate that this was a proper course if the proposed particularisation would meet the airlines' attack.

220. The plaintiff in fact provided further particulars. They read in full as follows:

"The plaintiff provides the following Further and Better Particulars of the allegation in paragraph 21 of his Statement of Claim that the DVT and the injuries resulting therefrom were caused by an "accident" within the terms of Article 17 of the Convention: (a) The "accident" causing the Plaintiff's bodily injuries comprised the flight conditions particularised in the particulars sub-joined to paragraph 6 of the Statement of Claim. (b) At the time of the flight Qantas and/or British Airways knew that the flight conditions were capable of causing or increasing the risk of the passengers on the flight experiencing DVT and knew of preventative measures which may have minimised such risk. The plaintiff and other passengers had no such knowledge. Accordingly, in was unexpected or unusual that the Plaintiff was subjected to such flight conditions. (c) The Plaintiff and other passengers were not provided with any information or warning about the risk of DVT or information about measures which they could take to reduce such risks. The Plaintiff and other passengers expected to be provided with warnings and information about risks of which Qantas and/or British Airways knew to which they may be exposed as a consequence of undertaking the flight, including the risk of DVT. The provision to passengers by Qantas and/or British Airways of warnings and information about such risks was usual, commonplace and expected. The failure by Qantas and/or British Airways to provide such warning and information was, accordingly, unexpected and unusual." [75]

224. Shortly thereafter the airlines obtained leave to appeal from the decision of Bongiorno J.

227. The grounds of appeal relied upon by the individual appellants are not identically expressed; but their gist is clear. So, BA contends that:

"1. His Honour erred in law in finding, at paragraph 43 of the judgment, that it would be open to a tribunal of fact to conclude that a failure by the Appellant to: (a) Warn its passengers of the risks of DVT; and/or (b) Advise passengers of precautions that passengers could take to minimise or eliminate the risks of DVT; in circumstances where the Appellant knew of those risks and passengers expected to be provided with warnings and information about risks of which the Appellant was aware, was an unexpected or unusual event or happening that is external to the passenger and therefore an "accident" within the meaning of Article 17 of the Warsaw Convention as amended as The Hague. 2. His Honour should have concluded that the claim pleaded and particularised by the first Respondent in his statement of claim and further and better particulars, and further and better particularised and explained in argument, did not disclose any claim against the Appellant under Article 17 of the Warsaw Convention as amended at The Hague, by reason of which the claim may (sic) by the first respondent against the Appellant should have been dismissed, stayed or struck out."[76]

The references therein to the Convention as amended at The Hague were not in point. Again, nothing turns on it.

232. The appeal was argued on the footing that the statement of claim as supplemented by further particulars including the particulars set out in [61] did not disclose an accident. It was so argued, conveniently, although the learned trial judge had before him only an intimation of what the latter particulars would be.

236. It is not in debate that the learned judge properly formulated the test by which he should decide whether the appellants should have judgment under Rule 23.01.[77]

239. Neither, it seems to me, could there be any criticism of his Honour's analysis of the correct approach to construction of the Convention, a matter about which I shall say something later.[78]

242. Concerning Air France v Saks[79], the relevance and meaning of which was much disputed, his Honour understandably fastened upon parts of the reasons of O'Connor J to which I shall later refer.[80]

245. His Honour did not finally decide whether the definition of "accident" proposed in Saks should be adopted in this case. For, he concluded, "on any reasonable definition of `accident' the plaintiff's case as presently particularised fails." The set of five circumstances particularised under para 6 of the statement of claim - that is :

* The confined and restricted environment provided for the respondent;

* Impediments to the respondent getting out of his seat;

* The offer of alcohol, tea and coffee to the respondent;

* Discouragement of the respondent from moving about the cabin, and encouragement for him to remain seated; and

* Failure to warn the respondent about the risk of DVT, or of measures to alleviate the risk -

could not whether simply or in combination constitute an "accident". That was so whether the meaning of the word was as proposed by Lord Lindley or Lord Macnaghten in Fenton v J Thorley Co Ltd,[81] or was that assigned it by majority of the High Court in Australian Casualty Co Ltd v Federico[82], or a fortiori, was the meaning given it by Saks.

254. The three possible meanings to which his Honour referred derived from cases of quite distinct types. Fenton involved a claim under the English Workmen's Compensation Act 1897, the critical phrase being "injury by accident". It concerned, then, the construction of a domestic statute. Federico involved the construction of a clause in an insurance policy. Only Saks raised the meaning of "accident" in the context of an international convention, where it has been said that a different approach applies.

257. His Honour then moved on to consider what he treated as being a proposed expansion of the particulars said to constitute the accident. He described the areas of expansion as raising allegations that:

* The airlines had certain knowledge concerning DVT, its causes and its relationship to the cramped conditions in economy class on their aircraft, and of precautions which could be taken to minimise the risk of its occurring to passengers travelling in that class.

* It was so usual and commonplace for airlines to issue warnings and advice to passengers concerning risks of which they had knowledge and to which they thought passengers might be exposed in flying on aircraft that passengers expected such warnings and advice. It was part of the objectively usual, normal and expected operation of the aircraft for such warnings and advice to be given.

That was something less than the full import of the expansion, as analysis of Particulars (b) and (c) of the Further and Better Particulars (conveniently described as particulars (b) and (c) to para 21 of the statement of claim) will show. Of this, more later.

263. His Honour evidently treated aspects of the proposed expansion, as he described it, as involving a failure to act. On that assumption he cited Husain v Olympic Airways[83] as authority for the proposition that a failure to act, viewed objectively, may be an unexpected event and so satisfy the Saks meaning of "accident". He said this:

"Where, objectively viewed, an airline would be expected o act in a particular way (or refrain from doing so) having regard to what is usual or expected in air travel at the time of injury, its failure to so act could constitute an accident for the purposes of Article 17 of the Warsaw Convention."[84]

267. He rejected a submission for the airlines that the failure to warn was not "unusual" because no warnings or advice were usually given about the risk of DVT at the time. The question what warning should be given could not be left dependent on the decision of an airline. That pointed up the need for an objective assessment of what should have been done in a particular case.

"If the plaintiff were to particularise his allegation of an accident to include the matters to which I have referred in paragraph 37 and, at trial, to lead evidence of them as matters of fact it would be open to the tribunal of fact trying this case to reach the conclusion that the defendants' failure to warn of the risks of DVT as they knew them and to advise as to the precautions which they knew a passenger could take to minimise or eliminate those risks constituted in all the circumstances "... and unexpected or unusual event or happening that is external to the passenger"; that is to say an accident within the Saks definition."[85]

274. His Honour rejected a submission for the airlines that a failure to warn, being an omission to act, could not be an event on board an aircraft. He referred to Art. 25 of the Convention. He added that:

"... the failure to warn is only one, even if it is the most important, of the many factors which the plaintiff includes in the particulars of the relevant accident as he argued it upon his expended case."[86]

There his Honour referred, fleetingly, to the fact that the expanded case was not confined to an alleged failure to warn.

279. On this appeal, counsel for the appellants attacked his Honour's reasoning as described in [71]-[74] of these Reasons.

283. Counsel developed extensive arguments as to the approach which must be taken to resolving the meaning of Art. 17; and as to that meaning and its application in the circumstances of this case. I will refer to submissions in the first of those connections when describing the necessary approach. I summarise the principal submissions in the latter connection as follows:

* Art. 17 must be viewed in context. The language of Articles 17, 18, 20, 21, 22 and 30 was relevant in that connection.

* Air France v Saks was the leading case concerning the construction and application of Art. 17. It had been considered, affirmed and applied in the United States, England and Australia. There was no basis for this Court not following and applying it.

* The principal issue in the appeal, that is, whether the respondent's allegations were capable of constituting an "accident," had been considered and resolved in The Deep Vein Thrombosis & Travel Group Litigation (DVT Group Litigation).[87] The English Court of Appeal had held that inaction could never properly be described as accident. There must be an event; and the event must be unusual, unexpected or untoward. A failure to warn could not constitute an accident. Art. 25 did not lead to a different conclusion. The Court of Appeal could not agree with the conclusion reached by the trial judge in the present case that a failure to warn was capable of constituting an accident.

* The Court of Appeal had also held that subjecting passengers to the permanent features of an aircraft was not capable of constituting an accident. The trial judge in the present case had made virtually similar finding, and there was no cross appeal.

* Husain v Olympic Airways, relied upon by the trial judge, was irrelevant. It was a case not of failure to act, but of refusal to act.

* The trial judge had wrongly elevated the concept of expectation - derived from the reference in Saks to an unexpected event or happening - to a position which obscured the true import of the word "accident" and the decision in Saks.

* The trial judge had wrongly treated the Convention as setting up a "no fault scheme of compensation"; and had compounded this error by resorting to concepts appropriate to common law negligence.

* The conclusion of the trial judge that failure to warn constituted an accident meant that all passengers on the flights suffered an accident, regardless whether or not individuals amongst them knew of a link between air travel and DVT.

* DVT Group Litigation was but one of a series of cases, both in Australia and overseas, in which courts had rejected DVT-based claims arising from air travel.

* Recent United States' decisions relied upon by the respondent did not assist him.

300. By oral submissions counsel added the following:

* Particular (a) to para 21 of the statement of claim added nothing to circumstances which the trial judge had held could not constitute an accident. Particulars (b) and (c) were not directed to establishing the event which constituted the accident, but rather that such event was unexpected or unusual.

* When O'Connor J said in Saks that the definition of accident there proposed should be flexibly applied in all the circumstances, that meant only that there should be flexibility with respect to what constituted the event. It did not mean that an event could be constituted by a non-event.

* "Unexpected", where used in Saks, meant something objectively unexpected rather than something unexpected or unintended by the injured person. But it mattered not whether the character of the event was to be objectively or subjectively assessed. Yet it could not be enough that one person should react differently to another.

* Failure to warn might create a situation which enabled an event to cause an injury. But the failure to warn would not then be the event which caused the injury.

* The factual matrix considered in DVT Group Litigation was not relevantly different to the case pleaded and particularised by the respondent. The only difference was the present allegation that the appellants had a practice of warning of risks.

* The issue was whether an airline's failure to warn of a risk known to it but unknown to passengers, in circumstances where passengers expected to be provided with warnings about risks, could constitute an accident. That raised questions whether the definition of "accident" in Saks should be applied, and whether failure to warn was capable of constituting an accident.

* An accident does not include something which does not happen or an occurrence which happens as part of the usual, normal and routine operation of an aircraft.

* It was significant that the particulars did not allege accident at any particular point in time, or describe something said to have happened at a particular place and in a particular way.

* It was significant also that the respondent did not assert that failure to warn involved a departure from the procedures of BA or from industry procedures.

* The definition of accident in Saks should be applied. Not only had it been applied in this country and overseas, it had been applied (at least intendedly) in cases relied upon by the respondent.

* The respondent's case as now particularised depended - as the trial judge held - upon a non-event constituting the accident.

* It was faulty reasoning to inquire whether something which did not happen was an accident by reference to the fault-based notion of an expectation which passengers were entitled to have.

* Some omissions may be treated as positive acts; but not a failure to warn of a risk of injury, which is a "mere omission".

* In a case of failure to warn, it is the fact that the risk eventuates which causes the injury, not the failure to warn - at least unless a passenger can show that the warning would have been heeded and the risk minimised or avoided.

* The decisions in Husain and Fulop v Malev Hungarian Airlines[89] did not assist the respondent. In each of those cases something happened which could be described and fixed in time. Blansett and ors v Continental Airlines Inc[90], however, must be regarded as wrongly decided.

* In Saks, O'Connor J held that an event was unexpected or unusual by reference to what ordinarily or usually happened on board the aircraft. Expectations, so far as they were relevant, were the expectations of those familiar with such ordinary and usual operations, not the expectations of particular or reasonable passengers. The decisions in Day v Trans World Airline Inc[91] and Wallace v Korean Air[92] fitted in with that analysis.

* Why should it be said that failure to warn occurred on board the aircraft in the current case, as distinct from before the flight? If it was truly the latter, then the non-event occurred before the period to which Art. 17 applied.

324. By oral submission counsel added the following:

* After El Al Israel Airlines v Tseng[93] it had been necessary for passengers in the United States to bring themselves within the terms of the Convention. Otherwise they were without relief. Before Tseng, such passengers had eschewed allegation of an accident, so as to have the opportunity of obtaining larger damages. This explained recent jurisprudence in that country directed to the question of "accident".

* The respondent had always alleged only that the flight conditions constituted an accident; and had asserted that these were the standard- that is, usual - conditions of travel.

* An operating procedure on a particular flight may be described as usual if so assessed by reference to the procedures of the airline generally or by procedures in the aviation industry.

* The sequence of events which culminated in the final text of the Convention showed that attention had been paid to the cause of injury capable of attracting damages.

* The construction of Art. 17 urged for the appellants would permit them to escape liability for injuries caused by their failure, even if deliberate, to warn passengers of a risk of which they but not the passengers were aware. This was not supported by the Convention read as a whole, was counter-intuitive, and had little if any support in the travaux preparatoires and earlier relevant documents.

* The interpretation of Art. 17 urged for the respondent was at least reasonably arguable, and the claims against the appellants should not be dismissed.

* The trial judge apart, no Australian superior Court had considered in detail the scope and meaning of Art. 17 (although the Saks definition had been cited with apparent approval in Agtrack (NT) Pty Ltd v Hatfield)[94].

* "Accident" was wide enough to include any kind of unfortunate or undesirable circumstances, whether brought about by an act or omission. The important focus was "the element of mischance, mishap or misadventure", the product of or involving "the miscalculation of forces, or inadvertence to them." Its focus was not upon an "event", or an analysis of what is comprehended by "event".

* Art. 17 being the outcome of a compromise by which, inter alia, limits on recovery were set and the Convention was made the exclusive avenue for recovery, "accident" should not be given a narrow and technical meaning. It should be construed as capable of covering at least the circumstances embraced by common law negligence actions.

* The travaux preparatoires contained nothing to show an intention of limiting (other than by reference to a temporal inhibition) the circumstances in which an injured person could recover damages. They showed an intention of enshrining fault-based liability in the sense only that fault was presumed except if the exculpatory provision applied.

* The test of "accident" set out by the majority in Australian Casualty Co Ltd v Federico should be applied; or otherwise the test in Saks. In the latter case it must be kept in mind that Saks addressed a particular situation, and that the definition there proffered did not replace the language of Art. 17.

* The case advanced for the appellants ignored the flexible operation required by the Saks test, an operation requiring consideration of the evidence in the particular case - this alone making it inappropriate to summarily determine the case.

* Saks placed no limitation on the type or nature of the unexpected or unusual causal event. The trial judge was correct to conclude that it could be constituted by an omission. An omission could readily constitute an event - as witness a failure to illuminate a "fasten seat belt" sign, or a failure to stow luggage properly.

* Saks did not support a conclusion that a relevant event must be unexpected or unusual only from the perspective of the carrier.

* In this case the DVT was due to external conditions - the particular flight conditions - including the absence of necessary warnings. The flight conditions referred to acts and omissions which were not expected.

* Saks may have been affected by the waiver of Art. 20 by the Montreal Agreement 1966, a consideration not applying in Australia. Art. 20 gives an airline some protection. Absent its application, protection might be afforded by confining the operation of Art. 17.

* In this case the failure to warn was unexpected. The question from whose perspective the "unexpected or unusual" quality of an event was to be determined did not arise in Saks; for there the plaintiff accepted the "normality" of the external event. Here, the trial judge rightly rejected the appellants' contention that "unexpected or unusual" was to be judged from their perspective.

* On the respondent's case, his DVT was not due to the usual, normal and expected operation of the aircraft. It was unexpected and unusual not to warn of a known risk. The conception described by the phrase "operation of an aircraft" should not be given a confined meaning.

* Nothing in the Convention required the accident to be limited in time, other than the time frame set by Art. 17. The respondent's case could have been pleaded to identify a multitude of accidents - minute by minute. Sensibly, this had not been done.

* The concept of "accident" should not be confused with (common law) concepts of causation.

* Contrary to the submission for BA, the respondent's case was not a case simply of failure to warn.

* It was not relevant that failure to warn before the respondent boarded the first of his flights could not have attracted a remedy. That would not deny a failure in the relevant period. Issues of causation and damages might arise at trial, but they were separate matters.

* The reference in the statement of claim to the flight conditions being standard conditions of and relating to passenger travel was simply a reference to uniform practice - as was the failure of the appellants to warn of the risk of DVT. It was a non-sequitur that the failure to warn was not unexpected or unusual.

* United States cases subsequent to Saks showed the flexible approach which it commended, and supported the respondent's position.

* The factual matrix in DVT Group Litigation did not articulate an accident as alleged by the respondent.

* The reasoning at first instance and on appeal in DVT Group Litigation was in any event unsound. It equated accident with event. On the reasoning of the Master of the Rolls failure to close a door, to keep a cabin pressurised, etc, could not be an accident. His Lordship's attempt to reconcile United States' decisions revealed the problem of treating "accident" as equivalent to "event".

* Decisions in the Queensland and New South Wales District Courts to which the Master of the Rolls referred with approval were not binding on the trial judge; and, importantly, the reasoning which supported those decisions was removed from his Honour's reasoning.

* In the German case relied upon by the appellants warnings of the risk of DVT had in fact been given.

* In this case the accident was a "set of conditions". What is unexpected or unusual sets limits to the set of conditions. The conditions may be a set of positive and negative conditions.

* The respondent's case was that there was failure to warn in the context of other conditions, including positive acts.

* Insofar as the trial judge referred to reliance and vulnerability[95] it was in the context of "unexpectedness". His Honour had specifically put torts law to one side in construing the Convention.[96]

361. In my opinion, this may be said about paragraph 21 of the statement of claim:

First, particular (a) alleges that the "accident" comprised the flight conditions particularised in paragraph 6 of the statement of claim. If that paragraph was read literally then particulars (b) and (c) would be superfluous. For within particular (a) would lie the "event" - whether or not it could be constituted by an omission; and within that particular would lie, as the respondent would have it, "the element of mischance, mishap or misadventure" which enabled the event to be characterised as an accident. So why should particulars (b) and (c) appear at all?

365. I do not read particular (a), in the circumstances, as being intendedly the sole or entire description of the accident relied upon by the respondent. It follows that I reject the submission made for the appellants, the trial judge having held that the particular (a) allegations could not be said to describe an accident, and no contention having been raised against that finding, that therefore (in effect) there was now no allegation of accident beyond that which had already been rejected.

368. Second, particular (b), read together with particular (a), should be understood to allege that although the flight conditions were the standard conditions of and procedures relating to passengers travel on such flights, their character was nonetheless unexpected or unusual ( and thus constituted an accident) because at the time the appellants knew of relevant risks but the respondent and other passengers did not. That is, the respondent's injury should not be said to be a result of his "own internal reaction to the usual, normal and expected operation of the aircraft."[97] The flight conditions, as alleged, include the environment created by the passenger seating configuration, positive acts (encouragement for passengers to remain seated and discouragement against them moving about; and the offer and supply of alcohol, tea and coffee) and a failure to warn of the DVT risk and of available measures to alleviate the risk. Particular (b) thus understood, failure to warn is no more than one aspect of the unexpected and unusual event.

371. Third, the matters alleged by particular (c) to para 21 should be understood to clothe the failure to warn alleged by particular (e) to paragraph 6 - upon an assumption that failure to act could be an event or a component of an event - with the character of something unexpected or unusual.

374. I have not ignored, in this analysis, the description of the accident given by the respondent's counsel in their written submissions.[98]. The content of that description is narrower than what I consider is comprehended by the particulars. In that formulation the flight conditions, other than the failure to warn, seem to be treated as the circumstances giving rise to the risk which must have called for a warning when the appellants knew of the risk and passengers did not, when it was usual for warnings of risks to be given and when passengers expected such warnings.

377. I consider, however, that the burden of the submissions made on behalf of the respondent demonstrates that a wider case was sought to be raised. So, for example, the written submissions[99], discussing Saks, contend that the pleaded case is that DVT is due "to external conditions (the particular flight conditions...) including the absence of necessary warnings"; and that "the flight conditions as alleged in the present case refer to acts and omissions of the airlines which were not expected."

380. It is unsatisfactory that an appeal relating to summary disposition of such significant litigation as this should be decided in a situation of any uncertainty as to the breadth of the respondent's case as pleaded and particularised. I would not resolve the appeal in favour of the appellants unless, on the broader view of the ambit of the respondent's allegations, the respondent's case was evidently untenable.

384. In resolving this appeal the starting point is the approach which must properly be taken to the interpretation of Montreal No 4, particularly Art. 17. About that matter there was a large but incomplete measure of agreement in counsel's submissions.

390. The authentic text of Montreal No. 4_,_ as in the case of the Convention itself, and the Convention as amended at The Hague - is in the French language. By s. 8(1)(e) of the Carriers Act the text of Montreal No. 4 is taken to be the text set out in Schedule 5 of that Act. But by sub-s. (2), in effect, if there is inconsistency between that text and the French text, the latter is to prevail.

393. Art. 17 has been in the same form since the Convention was opened for signature on 12 October 1929. In Saks, O'Connor J who gave the leading judgment, said this:

"A survey of French cases and dictionaries indicates that the French legal meaning of the term "accident" differs little from the meaning of the term in Great Britain, Germany or the United States."[100]

397. There was considerable debate in this case what meaning should be given the word "accident"; but it was not suggested by any counsel that the answer to that question turned on some difference in meaning between the French "l'accident" and the English "accident". The court was referred to no authority which suggests that there is any relevant difference in meaning. I approach resolution of this appeal on the footing that there is no difference.

400. It is, next, clear enough that considerable store is set upon certainty and uniformity of application of treaties and conventions; for which reason considerable deference is paid to the decisions of the courts of other signatory countries. Not always, however, has it been possible to achieve uniformity of construction of such instruments. Authorities which propose the proper approach in the context of the Convention include Emery Air Freight Corporation v Merck Sharpe and Dohme (Australia) Pty Ltd and anor[101]; and as well Sidhu and ors v British Airways Pty Ltd[102], Morris v KLM Royal Dutch Airlines[103], and the very recent decision of the English Court of Appeal in DVT Group Litigation[104]. What is said in those cases mirrors the general approach to resolution of issues of interpretation in connection with international treaties, as to which see, for example, Riverstone Meat Co Pty Ltd v Lancashire Shipping Co Ltd[105], The Shipping Corporation of India Ltd v Gamlen Chemical Co (Australasia) Pty Ltd[106] and Great China Metal Industries Co Ltd v Malaysian International Shipping Corporation, Berhad[107].

403. It is one thing to say that a court should strive for uniformity of interpretation of an instrument such as Montreal No. 4. It is another question whether relevant uniformity of interpretation is disclosed by pertinent authorities here and overseas. In the present case that question was at the heart of the debate - the appellants and the respondent calling decisions of overseas courts in aid of their respective positions.

406. Upon the question whether authorities here and overseas disclosed a relevantly uniform interpretation, counsel cited not only decisions of the highest courts in the United States and England, and of other superior courts in those judicial systems and in Australia, but also decisions of courts lower in the judicial hierarchy. Into the latter category fell decisions of US District Courts[108] and decisions of District Court judges in this country. Both sides relied upon such decisions. Counsel for Qantas, particularly, criticised the respondent's reliance, deriding some of the language used in decisions cited by respondent's counsel.

409. The task which confronts a court in a case such as this was described by Lord Hope in Morris in part as follows:

"As the language was not chosen by English draftsmen and was not designed to be construed exclusively by English judges, it should not be interpreted according to the idiom of English law. What one is looking for is a meaning which can be taken to be consistent with the common intention of the states which were represented at the international conference. The exercise is not to be controlled by technical rules of English law or domestic precedent. It would not be right to search for the legal meaning of the words used, as the Convention was not based on the legal system of any of the contracting states. It was intended to be applicable in a uniform way across legal boundaries."[109]

413. Set in that context, his Lordship said this about consideration of the case law of other countries:

"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 in 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."[110]

I respectfully agree. It is not inconsistent with this approach that Lord Nicholls should refer in Morris to the desirability of achieving consistency "with the mainstream views expressed in leading overseas authorities[111] or that Judge LJ should say in DVT Group Litigation that:

"We need to be a little cautious not to assume that every decision from every other jurisdiction, at whatever judicial level in the country concerned, carries identical weight."[112]

419. Speaking generally, one would expect to obtain the greatest assistance, in the case of foreign decisions, in decisions of courts of high standing. For that is where one would expect to find the more compelling reasoning. But that should not exclude the possibility, if courts at an equivalent level in different countries have not addressed a particular issue, that the reasoning of a court lower in the judicial hierarchy in one country in respect of that issue may prove more compelling than the reasoning of a court higher in the judicial hierarchy in a second country in respect of a slightly different issue. Further, in analysing the quality of the reasoning in the decision of a foreign court it seems to me that it is the substance of the reasoning and not its style which must be of prime importance. Counsel for Qantas, as I noted earlier, derided some of the language used in United States decisions relied upon by the respondent. That was unhelpful. It is a distinct question whether the reasoning was or was not sound.

422. So far I have focussed upon the internationally expressed desire for uniformity of interpretation of international treaties and conventions; and to the way in which, in consequence, decisions of foreign courts are to be brought to account. There are, I think, two other threads to the question of interpretation which should be kept clearly in mind in the particular context of Art. 17.

425. First, from time to time novel factual situations arise. In deciding how Art. 17 applies in such a case, I think that a court must be careful not to lose sight of the verbiage of the Article, assisted by any authoritative exposition of its meaning. It might do so if it took as its starting point uncritical acceptance of the language of a decision upon the meaning of Art. 17 directed to a different factual situation.

428. Second, it has often enough been said that the Convention represents a balance between benefits and detriments for both passenger and carrier; and that its Articles should not be interpreted as if their intent was to achieve full corrective justice for passengers.[113] In Sidhu those propositions were interrelated with the principle of freedom of contract. That freedom, viewed from the perspective of a prospective international air traveller who must fly pursuant a contract to which the Convention applies or not at all, seems to me to be more apparent than real. Nonetheless, I consider that the propositions are revealed by the text of the Convention - and Montreal No 4 - to be sound.

433. In the beginning there is the language of Art 17. Then there is Saks. It is important to understand what was in issue in that case; and what was not. it involved a claim by a passenger who became deaf in one ear because of the normal operation of the pressurisation system of the airliner in which she had been a passenger. The accident, unless it was constituted by the internal damage to the passenger's hearing apparatus - that is, the injury itself - could only have been constituted by the operation of the pressurisation system. The narrow point which the case decided was that an accident must be something unexpected or unusual, and external to the passenger. An internal response to the usual, normal and expected operation of an aircraft could not satisfy the concept of accident.

436. O'Connor J developed this chain of reasoning:

* "Accident" where used in Art. 17 and "occurrence" where used in Art. 18 must mean something different.[114]

* Art. 17 refers to an accident which caused the passenger's injury; not to an accident which is the passenger's injury.[115] That reflected an attempt by the drafters of the Convention to discriminate between cause and effect.

* The French legal meaning of "accident" differed little from the meaning of the term in Great Britain, Germany or the United States. When used to describe a cause of injury "it [was] usually described as a fortuitous, unexpected, unusual or unintended event."[116] So the text of the Convention suggested that "the passenger's injury must be caused by an unexpected or unusual event.[117]

* Although the word "accident" had been broadly interpreted in American decisions, courts had nonetheless refused to extend the term to cover routine travel procedures that produced an injury due to the peculiar internal condition of a passenger.[118]

"That liability under Art. 17... 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."

"... 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."

"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 and unexpected event external to the passenger."

451. Saks has been applied in the United States, England,[119] and Canada.[120] It has also been cited with apparent approval in this Court[121] and has been applied in the District Courts of NSW and Queensland in cases relied upon by the appellants. I do not doubt that it is a decision to which, in accordance with the required approach, very considerable deference should be paid.

454. That is so despite the observation by Sutherland J in Quinn that Saks was decided in the context of a regime in which, by the Montreal Agreement 1966, the operation of Art 20 had been excluded; which could be seen as a circumstance commanding a narrower definition of "accident" than in jurisdictions in which (as here) Art 20 applies. In Saks itself, O'Connor J referred to the Montreal Agreement, in particular the required waiver of Art 20 (1) of the Convention, when noting that the Court's duty was to enforce the accident requirement of Art 17. That gives no warrant for saying that the meaning accorded "accident" in Saks was influenced by the waiver of Art 20 (1) in consequence of the Montreal Agreement.

457. It should therefore be accepted that a passenger's internal physiological processes constituting an injury cannot at the same time be an accident; for then there would be no differentiation between injury/damage and accident causing the same. It should also be accepted that whatever it is that constitutes an accident must be external to the passenger and be unexpected and unusual. As a corollary, it should further be accepted that when the event ( I do not presently use the word in contradistinction to non-event) relied upon as constituting an accident is some aspect of the usual, normal and expected operation of an aircraft, that event will not have the required characteristic of being unexpected or unusual.

460. It is, however, clear that Saks, in terms, did not have to consider whether an "event or happening", to use the language of O'Connor J,

* could be constituted by a mixture of acts and omissions;

* must be recognisable as an event constituting an accident at the time of its happening.

467. It may be said, I think, that underlying the judgment of O'Connor J in Saks is the assumption that it is in the nature of an accident that it is something which happens; an occurrence which assumes the character of an accident because it is unexpected or unusual. In that connection I refer, though it was said in the particular context of observing that accident may refer to both cause and effect, to her Honour's citation of part of Lord Lindley's speech in Fenton v Thorley.[122] Note also her Honour's reference to the French (legal) meaning of accident when describing a cause of injury,[123] and to her Honour's repeated use of a formula such as "an unexpected or unusual event."[124] It may also be said, I think, that the use of nouns such as "event" or "happening" fits the idea of something which occurs at a fixed point in time. But in my opinion it would be quite wrong to elevate those impressions to the status of a determination by Saks that an accident can only be constituted by a positive act occurring at a precise point in time, and having the characteristic of being unexpected or unusual.

471. An important aspect of the appellants' submissions was the contention that the respondent's expanded case essentially relied upon accident constituted by omission to act - specifically, failure to warn. DVT Group Travel, counsel submitted, doomed that pleading. Omission to act, a fortiori failure to warn, was a non-event. It was the antithesis of "accident."

474. Counsel for the respondent denied that the pleaded case was so confined. He submitted that reliance was placed upon both acts and omissions; of which failure to warn was but one matter. For reasons discussed at paragraphs [81]-[86], I consider that the respondent's submission was correct. On any view, however, inaction is alleged; particularly failure to warn. For that reason, it is necessary to examine a number of cases in which, in the context of Art 17, the plaintiff has alleged an accident constituted, arguably, by inaction. In doing so, I will refer to the way in which each case was pleaded, so far as that is discernible.

477. The cases to which the Court was referred divide, on the pleadings, into two classes: those where the plaintiff relied upon circumstances other than failure to warn; and those where the alleged accident was constituted by a failure to warn. The respondent contended that the failure to warn decisions should be understood against the backdrop of decisions of the first type. In that category were Husain, Fulop, McCaskey v Continental Airlines Inc Et Al[125], Fishman v Delta Airlines Inc[126] and Turturro v Continental Airlines[127]. According to the appellants, however, those cases properly analysed were instances of action - not inaction; and in some instances the reasoning adopted by the court was inappropriate to an Art. 17 inquiry.

480. Turning to the second category of case, the appellants relied upon DVT Group Travel, McDonald v Korean Air, Volander[128], Van Luin v KLM Airlines[129] and Rynne v Lauda Air Luftfahrt Aktiengelsellschaft[130]. The respondent relied upon the decision the subject of this appeal; and in addition Blansett, Louie v British Airways[131] and Miller and anor v Continental Airlines Inc and ors; Wylie v American Airlines Inc and anor[132] (conveniently "Miller").

483. I go to the first group of cases. Husain, now on appeal to the US Supreme Court, involved the failure or refusal of a flight attendant to move an asthmatic passenger from an area of the cabin affected by cigarette smoke despite request and despite being apprised of the passenger's medical condition.

486. The judge in the US District Court concluded that of three occurrences relied upon as accidents, only one met that description - the attendant's repeated refusal to move the passenger to another seat. That failure was an unexpected or unusual event because it occurred in defiance of industry standards and of the defendant's internal policy.

489. The Court of Appeals for the 9th Circuit distinguished Krys v Lufthansa German Airlines[133] and Abramson v Japan Airlines Co[134] on the facts; and could not conclude, based on the record before the District Court, that the judge had clearly erred.

492. In substance, the reasoning in Husain is that found in the judgment at first instance. There the judge said that:

"In sum [the attendant's] behaviour was far from usual or expected. She violated the industry standard of care, she misrepresented to [the passenger's wife] that the flight was full, and she acted in contravention of [the defendant's] accepted ... policy. This aberrant behaviour was both unexpected and unusual[135]."

"When a passenger boards an aeroplane, he or she should be able to expect that the flight crew will comply with accepted procedures and rules. A failure to do so is unexpected."

"The practical effect of cases like Abramson is to dissolve the airlines' duty of care and to `create an incentive to airlines engaged in international travel not only not to exercise the highest degree of care but to completely refuse to treat' or assist passengers with medical problems."[136]

"When a flight attendant's acts create a foreseeable risk of injury to passengers, an `accident' has occurred... In this case, [the attendant's] failure to respond appropriately to... requests and her failure to comply with the applicable standard of care were both `unexpected' and `unusual'".[137]

502. In Husain, as will be seen to be the case in other United States' decisions, no particular distinction was drawn between acts and omissions. The language of the judgment at first instance shifts between the language of refusal to act and failure to act.

505. In DVT Group Travel, Lord Phillips MR said that he had no difficulty with the result in Husain, but that the case could not properly be considered as mere inertia, or a non-event. The refusal to provide an alternative seat formed part of a more complex incident whereby the passenger was exposed to smoke in circumstances that could be described as unusual and unexpected[138].

508. Whether or not Husain is capable of being rationalised in such a way, nothing indicates that this was in fact the basis of the decision; for there is nothing to suggest that either at first instance or on appeal was any point made of the distinction said to exist between acts and omissions.

511. Another aspect of Husain requires comment. The District Court judge held that Saks required an objective inquiry whether an unexpected or unusual event or happening, external to the passenger, had occurred. The focus was not upon the perspective of the person experiencing the injury[139]. In that context, his Honour referred to the passenger's reasonable expectation that his wife's requests would be accommodated, to what a passenger in the particular position should expect, to the "industry standard of care" and to what a passenger "should be able to expect"[140]

514. The judge also said, as I noted a little earlier, that "the practical effect of cases like Abramson is to dissolve the airlines' duty of care"; and that "when a flight attendant's acts create a foreseeable risk of injury to passengers `an accident' has occurred...".

517. It seems to me, with respect, that some of the language used by the judge was the language of common law liability. The language of reasonable expectation falls into that category, as does the reference to an airline's duty of care and to the creation of a foreseeable risk of injury. Importing such domestic concepts into an international treaty does not, as I understand it, accord with principle; which is not to say that in considering whether an event was unexpected or unusual the respective state of knowledge of airline and passenger concerning a particular matter will be irrelevant; or that conduct in breach of an industry or internal practice may not render such conduct unexpected or unusual. Of unexpectedness, more later.

520. Fulop involved a passenger who had a heart attack and who claimed that its extent and effects were made worse because the flight did not divert and land early, thus enabling his earlier treatment. An application by the defendant for summary judgment was granted in part and refused in part.

523. The plaintiff's case was put as the failure or refusal of the defendant to divert, a response alleged not to comply with the defendant's own operations and procedures. It was put, then, as a an instance of action and/or inaction.

526. Judge Marrero of the US District Court attempted to find a pattern in the decisions. He concluded that the larger the causal role of the airline, the greater the knowledge and involvement of personnel and operations in bringing about the harmful event, the more likely it was that liability would be found. His Honour described this as a narrower view of the authorities.

529. His Honour concluded that the unusual and unexpected event or happening might be viewed as the alleged aberrant conduct of the defendant's employees in handling the occurrence, the failures of which aggravated the initial heart attack.

532. His Honour said that what was unusual or unexpected should be gauged by what objectively is usual or reasonably to be expected under particular circumstances[141]. But he also said that it was beside the point whether the crew's acts were negligent[142]. What mattered was whether there was deviation from compliance with expected procedures; an external factor arguably causative of the plaintiff's aggravated injury. Yet his Honour also spoke of the airline failing to perform "to reasonable standards of care", and to the objective expectation of a passenger[143].

535. In all, the Judge undertook, in my respectful opinion, a serious examination of principle. He adopted a relatively narrow view of what could be an unexpected or unusual event or happening. He did refer to common law notions in determining unexpectedness. That is how he analysed the District Court judgment in Husain. Consistently with other United States' decisions his Honour drew no distinction between acts and omissions. Insofar as he imported common law notions in determining unexpectedness, for reasons indicated I could not accept the correctness of his Honour's approach.

538. McCaskey involved a claim by the widow of a man who died some weeks after sustaining a cerebrovascular accident on a flight. The claim was pleaded, so far as accident is concerned, in a number of ways. Only the third way in which the case was put is presently relevant. There, the widow alleged, inter alia, that the failure to divert the flight led to an aggravation of injury and thus to discrete injury.

541. The airline brought motion for summary judgment. It relied upon cases which held that mere failure to act in response to injury was not an accident. Judge Kent of the US District Court, who also determined the applications in Blansett, dismissed the airline's motion. Upon the presently relevant pleading, his Honour accepted that failure to divert was not itself an accident. But it was another thing whether it could never be a jury question. A finder of fact might determine, in the particular circumstances, that the failure to divert, inter alia, was an accident. His Honour pointed out that the decisions in Krys and Abramson antedated Tseng, so that in those cases the passenger was not completely deprived of potential remedy. He pointed out a limiting feature of the Krys' judgment. He noted that Abramson "seems to have been heavily affected by the Court's notion that a pre-requisite to an accident is an unusual `physical' event common to the entire flight. There is no such requirement"[144]. His Honour recognised, however, that Abramson had been cited with apparent approval in Saks.

544. In McCaskey, also, the failure to divert might be characterised as a positive act - that is, a refusal to divert in consequence of ill-informed medical advice given by a doctor on the ground. Again, it was a case in which no distinction in principle was drawn between acts and omissions in the context of "accident". Some of the judge's observations about Krys and Abramson were undoubtedly correct; but that is not to say that each basis upon which he sought to distinguish them was evidently sound.

547. Fishman was not a case of failure to act at all. The infant plaintiff was scalded when a flight attendant applied an over-hot compress to alleviate the child's earache. Water dripped onto the child's neck and shoulder, causing burns. A motion to dismiss on the ground that the plaintiff's suit was out of time was granted; and that decision was affirmed on appeal.

550. The only importance of the case, for present circumstances, lies in the observation of the US Court of Appeals for the Second Circuit that an injury resulting from routine procedures in the operation of an aircraft or airline can be an accident if those procedures are carried out in an unreasonable manner[145]. This bears upon the observation of O'Connor J in Saks that when 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"[146].

553. Turturro was an odd case on the facts. A passenger, because of the onset of acute anxiety, sought to de-plane before the aircraft took off. She had discovered that the pocket book in which she kept medication to take because of fear of flying had been stolen. She made a number of increasingly agitated requests to be permitted to leave the plane. Only after some time were her requests accommodated.

556. The plaintiff made a number of claims. The airline moved for summary judgment, or to dismiss. The motion was granted in part and denied in part. Senior District Court Judge Knapp held, inter alia, that injuries relevant to Art. 17 were not pleaded, because the injuries alleged were psychological, not physical. That decided so much of the case as was concerned with events on board the aircraft, or whilst disembarking.

559. At one point in his judgment, his Honour said that "the `accident' included [the defendant's] act of delaying the return of the plane to the gate and its employee's comments directed to the plaintiff and to fellow passengers"[147]. This observation was probably obiter. But in any event the case involved a refusal rather than a failure to act.

562. Against the background of the United States' cases said for the respondent to show that accident can be constituted by omission to act, I turn next, for convenience, to the American failure to warn cases.

565. In Blansett, it was alleged, inter alia, that the defendant had failed to warn of the risk of development of deep vein thrombosis; and that the passenger had developed a clot and then a stroke. Other allegations of accident, also, were raised. The defendant brought a motion to dismiss in respect of the claims based on failure to warn. Judge Kent denied the motion.

568. His Honour was prepared to assume that sitting in an aircraft during a normal flight was not itself an accident. Here, however, the plaintiff argued that the defendant had violated a custom, or rule, or procedure of airlines to give a DVT warning.

"the [plaintiffs] allege that Continental's failure to warn of the risk of contracting DVT was an accident because such failure was an abnormal and an unreasonable application of routine procedure, and thus it was unusual and unexpected. Taking these well-pleaded allegations as true... this Court holds that the Blansetts have alleged a set of facts that, if proven, could support recovery under the Warsaw Convention."

575. The defendant also moved for partial summary judgment on the claims based on failure to warn. In a second ruling, Judge Kent denied that motion. By the time that his Honour dealt with the second motion it had been made clear that the plaintiffs were arguing that the failure to warn was based on the defendant's failure to follow an alleged industry custom. Having referred to Saks, Fishman, Fulop and Husain, his Honour concluded that an airline's violation of an industry standard of care, alone, could be an accident with the Saks' definition. His Honour added that:

"To hold otherwise would have the ironic result of rewarding individual airlines for adopting an `ostrich' approach of failing to institute routine safety procedures in compliance with generally recognised industry standards of care."[148]

579. In Blansett, as in other American cases, no distinction was drawn between acts and omissions. Nor was there any discussion of a possible distinction, for the purposes of Art. 17, between reliance upon circumstances susceptible of characterisation either as acts or omissions; and reliance upon circumstances which could only be characterised as omissions. The question was simply framed as one whether an airline's violation of an industry standard could constitute an accident within the Saks' definition. No attention was paid to the question whether an accident must occur at an instant in time; or whether the accident must have some immediate effect in inducing injury. Also notable, by contrast with the present case, is that on the one hand the failure to warn was said to be in breach of a specific industry standard; and on the other hand that the allegation of failure to warn was not accompanied by an allegation of positive discouragement against movement in the plane in a context of asserted pertinent knowledge on the part of the defendant and lack of equivalent knowledge by the plaintiff.

582. Louie was a June 2002 decision of US District Court Judge Singleton on a motion to dismiss brought by two airlines. No full report of the decision was provided to the Court. The plaintiff apparently pleaded that he was required to sit in a cramped seat and was given no warning of the risk of DVT. The judge said that the airlines had made out a strong case but had not demonstrated the pleading to be so defective that amendment to specifically allege an accident would not cure the defect. As matters stood, it seems, no accident was pleaded.

585. I think that very little can be gleaned from this decision. The case was one of failure to warn. But so far as it appears, there was no allegation that the risk of DVT was known to the airlines but not to the passengers; nor any allegation that what would otherwise have been the normal operation of the airline was abnormal in light of the defendant's knowledge. Neither again, so far as it appears, was any allegation made that the failure to warn occurred against a background of industry practice to the contrary.

588. In Miller, the defendant airlines brought motions to dismiss for failure to state a claim. The plaintiffs alleged that the defendants failed to warn of the risk of DVT despite customary procedure and policy of the airline defendants, and of the airline industry, to warn of risks and to advise steps passengers could take to minimise risks. The plaintiffs did not allege that defective seats or seating configuration was an unexpected or unusual occurrence.

591. District Court Judge Walker concluded that the plaintiffs' allegations were sufficient to survive a motion to dismiss. Blansett did not provide definitive guidance, but counselled against dismissal of a failure to warn case on motions such as were before the Court. His Honour described cases from other countries as involving different claims and procedural contexts. So, in DVT Group Travel at first instance, Nelson J had said that the flight was operated in accordance with the defendant's usual procedures and practices. In Rynne, the Queensland District Court Judge had said that the flight conformed to proper standards and that there were no incidents during the flight which affected the normal cabin environment. Volander was a decision based on evidence presented to the Court. The Ontario Superior Court of Justice had relied on Scherer v Pan Am World Airways Inc[149] when deciding McDonald v Korean Air.

594. Here, the plaintiffs alleged that the defendant's failure to operate the flight according to customary policies gave rise to the accident that caused their injuries. It was those allegations which were sufficient to survive the motion to dismiss.

597. Like Blansett, Miller was a simple failure to warn case. But like Blansett, breach of industry practice was alleged. There was no discussion in the judgment of a distinction between acts and omissions in the context of "accident;" nor whether, in that context, all circumstances capable of being characterised as omissions were essentially alike. The Court seems to have accepted that failure to act in accordance with normal operations may bespeak accident. Again by contrast with the present case, the plaintiffs did not allege positive discouragement of movement within the aircraft cabin; nor was any contrast pleaded between the pertinent knowledge of the airlines on the one hand and the plaintiffs on the other.

600. I go to the authorities relied upon for the appellant, beginning with the English Court of Appeal decision in DVT Group Travel, upon which appellants' counsel placed considerable reliance.

603. The appeal arose out of the decision of Nelson J that a specimen matrix did not disclose an accident under Art 17. It is important to see what the matrix alleged. Relevantly it was in the following terms:

"4. ... The flight was characterised 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 further or other steps[150]to minimize and/or eliminate risk. ... 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 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."

607. On the appeal, the appellant sought to rely upon further information. It was expressed as follows:

"The Defendant: (1) required the Claimant to sit in a seat which due to its insufficient width, the insufficient distance between it and seats on either side of it and the insufficient distance between it and the seats in front of and behind it discouraged and/or prevented the Claimant from moving out of his seat throughout the flight or most of the flight and/or restricted his movement and/or caused him to sit in a cramped position while seated in it when it knew or ought to have known that discouraging and/or preventing the Claimant from doing these things or any of them and/or causing him to sit in a cramped position could cause and/or increase the risk of the flight causing a DVT to the Claimant; (2) provided or caused there to be i. insufficient space between the seats ii. insufficient air pressure within the cabin and/or iii. insufficient levels of oxygen within the cabin and/or iv. insufficient amounts of fresh air and movement of that air within the cabin and/or v. insufficiently humid air within the cabin and/or vi. a temperature within the cabin that was high or excessively high."

611. The Court of Appeal acted on the further information on the footing that it set out "typical and usual features of travel by air at the material time"[151].

614. Before going further, it should be noted that the specimen matrix contained no assertion that the defendant had positively discouraged movement in the aircraft cabin. Further, whilst the claimant alleged that the defendant knew or ought to have known that air passengers would be at increased risk of suffering DVT, no assertion was made that the claimant (or other passengers on the flight) were not seised of that knowledge. Further again, the additional information asserted that the defendant required the plaintiff to sit in a seat which discouraged or prevented the claimant from moving out of the seat, this being in circumstances where the defendant had or should have had knowledge that discouraging or preventing him from moving, or causing him to sit in a cramped position, could cause or would increase the risk of the flight causing DVT. But that involved no allegation of active verbal discouragement; and the static discouragement was treated as being a typical and usual feature of air travel at the material time.

617. Lord Phillips MR, whilst remaining of the view, which he had expressed in an earlier case, that the definition of injury in Saks gave the word "accident" a natural and sensible meaning in the context of Art. 17, nonetheless took as his starting point the language of the Article itself. He said this:

"It seems to me that the concept of an accident taking place on board an aircraft which causes death, wounding or bodily injury is a simple one. The words naturally suggest an untoward event which impacts on the body in a manner which causes death, wounding or injury. The impact of the event may be physical or it may be an impact on the senses which results in bodily injury, but one would normally expect the untoward event to cause the death or injury directly.[152]"

621. Conformably with his own phrase "an untoward event" and the reference in Saks to "an unusual or unexpected event external to the passenger", his Lordship turned to what is meant by an event. He said this:

"The 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."[153]

625. According to his Lordship, counsel for the plaintiff had submitted that it was possible to identify "two accidents" in the specimen matrix as augmented by the further information. The first could be described as subjecting the passengers to carriage by air that had the features itemised in the further information. His Lordship did not consider the existence of those permanent features, or subjecting the passengers to carriage in aircraft with those features, was capable of amounting to an event satisfying the definition of accident.

628. Then his Lordship turned to the allegation of failure to warn. He said:

"I cannot see how this failure can be categorised as an accident. It was simply something that did not happen - a non-event."[154]

632. Having concluded that no event had occurred, his Lordship understandably described as unreal the question whether events which were "unexpected or unusual" had taken place in both the flights. He noted, nonetheless, the competing submissions in that connection. On the one hand it had been argued that a carrier could reasonably be expected to take reasonable steps to warn or minimise, or exclude known dangers to the health, safety and lives of its passengers. Its failure to do so was unusual or expected. On the other hand it had been argued that because Art. 17 did not involve a fault-based test, it was not legitimate to have regard to questions of fault when considering whether an event was unusual or unexpected.

635. Concerning the competing arguments his Lordship said this:

"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 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 which is relevant to deciding whether an event is an accident.[155]"

639. Having determined the merits of the appeal as a matter of language, his Lordship turned to authorities. He noted, at the outset, that there was no English case law bearing directly on the issue before the Court. Then he turned to authorities in the United States. He referred to Abramson, which he described as having been approved in Saks. That case, he said, accorded with his provisional view that there was nothing in the specimen matrix which was capable of constituting an accident. It is to be remembered that Saks raised a narrow issue. It seems to me that approval of Abramson should be understood in that context.

642. Concerning Fulop, his Lordship remarked that he did not find it impossible to reconcile the result in that case with his provisional conclusion that failure to warn or advise could not constitute an accident. The failure to divert was a corollary of the decision to fly on notwithstanding the passenger's heart attack. If that was contrary to standard practice, or even the airline's internal practice, then it was perhaps arguable that there occurred an unusual or unexpected event.

645. Concerning Husain, his Lordship said he had no difficulty with the result. But he questioned the reasoning. The refusal of the flight attendant to move the passenger, he said, could not properly be considered as mere inertia. It was a refusal to provide an alternative seat. The refusal could be described as insistence that the passenger remain seated in the area exposed to smoke; and so the exposure to smoke could in the circumstances properly be described as an unusual or unexpected event. His Lordship's rationalisation of the decision turned on an issue - that is, the distinction between act and omission - which seems not to have been treated as an issue at all in Husain.

648. McCaskey, according to his Lordship, could be analysed in the same way as Fulop.

651. The reasoning in Blansett, his Lordship considered, did not adequately address the need to identify an event as an essential element of any accident. A failure to act, including a failure to warn, could not be said to constitute an event or an accident without stretching flexibility beyond breaking point. He disagreed with the decision.

654. Then his Lordship came to the decision of Bongiorno J in this case. He described the additional allegations which the respondents sought to raise as first an allegation that the airlines had certain knowledge about DVT, its causes, its relationship to the cramped conditions in economy class, and precautions which could be taken to minimise the risk of it occurring; and second, that it was common place for airlines to issue warnings and advice to passengers in respect of risks of which they had notice. That description of the additional allegations, doubtless because his Lordship did not have before him the actual terms of the proposed amendments, was incomplete in a number of respects. So, it did not make clear that the new allegations were essentially directed to the question of unexpectedness, and that the flight conditions remained pertinent to the respondent's claim. Second, his Lordship made no reference to the alleged absence of knowledge of the respondent and other passengers, set in contrast to the alleged knowledge of the appellants. Third, his description contained no reference to the expectation of passengers that they would be provided with warnings and information about risks of which the appellants knew.

657. His Lordship said that the most critical part of the judgment was that in which Bongiorno J considered whether a failure to warn or advise could constitute an event on board an aircraft. Citing paragraph 44 of his Honour's reasons, his Lordship said there could be no quarrel with so much of that paragraph as said that the failure to warn was only one, even if the most important, of the many factors which the plaintiff included in the particulars of the relevant accident as he argued it upon his expanded case. Then his Lordship turned to so much of the paragraph as dealt with failure to warn. He said this:

"...I cannot accept that the first two sentences deal adequately with the difficulty of bringing a failure to warn or advise within the definition of an accident. It is true that the distinction between acts and omissions can, in some circumstances be dubious. Husain, upon which the judge relied so heavily, is an example of such circumstances, as I have already explained. But there is no dubious distinction in the case of a failure to warn or give advice. Such a failure is mere inaction or inertia. There is no way in which it can be suggested that such inaction is tantamount to action."[156]

Moreover, his Lordship said, Art. 25 did not assist the plaintiff, for it had a wider ambit of operation than merely Art. 17.

662. In the event, his Lordship was unable to concur with Bongiorno J that a failure to warn or advise was capable of constituting an accident. That was the limit of his Lordship's disapproval. He specifically said, as I have noted, that he had no quarrel with the observation of Bongiorno J that failure to warn was only part of the expanded pleading of accident. In the event, I do not understand his Lordship to have held that so much of the respondent's expanded case as did not depend upon failure to warn could not constitute an accident.

665. Then his Lordship referred to the decisions in Van Luin, McDonald and Rynne, in each case with evident approval. I shall deal with those cases discretely a little later in these reasons.

668. Judge LJ agreed with the Master of the Rolls. Whilst saying that the pre-eminence of the judgment of the United States Supreme Court in Saks had been universally acknowledged and should be applied, his Lordship also said that the explanation in Saks had sometimes directed attention away from the single word used in Art. 17. The Article had not been re-written, at greater length, incorporating the explanation provided in Saks. He said:

"Inevitably, the forensic endeavour to bring an individual case within the ambit of Article 17 has led to increasingly refined and subtle arguments which, speaking generally, focus on the broader words of explanation used in the judgment in Saks rather than the words of the Article themselves, and their inapplicability to an injury resulting from the passenger's own `internal reaction' to what is otherwise a normal, untroubled flight.[157]"

672. In the event, his Lordship agreed that, on the basis of the agreed specimen matrix, an accident was not disclosed.

"in my judgment, the court cannot lose sight of the fact that the foundation for liability for personal injury that was chosen and agreed was that there had been an `accident'"[158].

679. Echoing Judge LJ, his Lordship emphasised that Saks did not substitute a new form of words. He said this:

"If one asks oneself the simple question was there, on any recognised meaning of the word, an accident in circumstances where a person suffered deep vein thrombosis merely because of the effect of a flight on an aeroplane without there being any triggering event, the answer in my opinion is quite simply no.[159]"

His Lordship added that if the appellant's arguments were correct, it seemed to him inevitable that the language of Article 17 would have been different.

684. I go to the cases cited with approval by the Master of the Rolls.

687. Van Luin was a defendant's application to dismiss on the ground that no reasonable cause of action was disclosed; or that the statement of claim be struck out as disclosing no reasonable cause of action.

690. The judge noted a submission for the defendant that the statement of claim did not plead that injury was caused by an unexpected or unusual event or happening; and a further submission that a proposed amendment which sought to rely upon failure to advise of the need to move about and failure to advise that the plaintiff should drink extra fluid would only plead the plaintiff's own internal reaction to the usual, normal and expected operation of the aircraft.

693. The judge noted that it had been argued for the plaintiff that "accident" should apply where injury consisted of unexpected internal trauma caused or contributed to by the normal operation of the aircraft; and that Saks should not be followed. He recorded an alternative argument that the proposed amended case arguably pleaded accident.

696. His Honour concluded that the acts or omissions of the crew might constitute an unexpected or unusual event; but that failure to advise was not such an event external to the plaintiff. He said that he followed Saks in so concluding.

699. The judge further said that if the alleged failures did constitute an unexpected or unusual event external to the plaintiff, nonetheless the plaintiff's thrombosis resulted not from such event or happening but from the plaintiff's own internal reaction to the normal operation of the aircraft. He referred to normal operations cases including Abramson, Volander, Scherer and De Marines.

702. In summary, the case as pleaded alleged no accident. The plaintiff sought to amend to plead accident constituted by failures to advise. He sought to raise no allegation about the state of pertinent knowledge of airline and passengers. No allegation was made of positive conduct on the part of the defendant. A hopeless contention was advanced that Art. 17 really meant accidental injury, and that the Court should not follow Saks. The judge held that the acts or omissions of a crew might constitute accident; but that according to Saks failures to advise were not such an event external to the plaintiff. He did not say why Saks led to such a conclusion. He did not contrast acts and omissions, nor characterise failure to advise as a particular kind of omission. He said that if the failures did constitute an accident, the DVT nonetheless resulted from the plaintiff's own internal reaction to normal aircraft operation. That seems to have assumed that there could be but one cause of injury, something at odds with Saks. All in all, Van Luin was a considerably different case to that which is before this Court; and one in which the reasoning was not fully exposed.

705. Rynne, decided in the Queensland District Court, was the resolution of an application for summary judgment.

708. The judgment shows that the plaintiff asked a cabin attendant if she could move about the cabin. She was told she would have to wait until service trolleys passed by. Then she developed calf pain. Once during the first leg of her trip she told a flight attendant that her leg was painful. Nothing was done. Several times she asked for water; but she was given none. Twice during the second leg of her journey she told cabin staff of her painful leg. Nothing was done. She had developed superficial thrombophlebitis in the presence of pre-existing varicose veins.

711. The plaintiff did not squarely plead accident. She did plead failure to adopt procedures to advise passengers of risks of venous thrombosis or to advise or warn of risks, or to advise activity or the need to drink large quantities of fluid.

714. According to counsel's submissions, as noted by the judge, the accident comprised the flight conditions and procedures referred to in the statement of claim and further particulars. "Accident" should be broadly construed, being easier to satisfy than the technical common law rules of negligence and capable of covering at least the same circumstances.

717. The judge accepted the applicability of Saks. He concluded that the requirement that the plaintiff remain in her seat until the trolleys passed by was part of the normal operation of the aircraft. The plaintiff had a medical condition of which the defendant could not have been expected to know. Otherwise, his Honour said, the claim alleged failures of one kind or another. The question was whether "such inaction is incapable of amounting to an unexpected or unusual event or happening which is external to the passenger."[160]

720. In support of a conclusion that such inaction was so incapable, his Honour cited Van Luin, McDonald and DVT Group Travel[161]. He referred also to the present case, by way of disapproval of one aspect of the judgment of Bongiorno J.

723. All in all, one matter aside, the plaintiff's proposed case was based solely on failure to advise or warn. No allegation was raised of pertinent risks known to the defendant but not the plaintiff. No positive conduct was alleged. In my opinion the judge was right to accept Saks and he was right to conclude that the Convention should not be interpreted to reflect "modern risk allocation theory". He treated the decision presently on appeal as affecting such a re-allocation of risk. Plainly, he accepted the decision at first instance in DVT Group Travel, concerning which he said that the facts there alleged and in the matter before him were virtually identical. I have attempted to explain that the pleadings here differ significantly from the factual matrix in DVT Group Travel.

726. McDonald, decided by Hermiston J of the Ontario Supreme Court of Justice at first instance, involved a defendant's strike out application.

729. The judge noted that the plaintiff advanced an argument that DVT acquired on board an aircraft is an accident within the meaning of Art. 17; and a further argument that failure to warn and educate passengers represented an unusual or unexpected operation of the aircraft and constituted a marked departure from the duty of care which carriers owed to their passengers.

732. His Honour held that an airline may be negligent in not advising passengers of the risk they assume, but that such negligence was not in itself an accident in the sense that the DVT was not linked to an unusual or unexpected event external to the plaintiff as a passenger[162].

735. In all, the plaintiff's claim in part rested on failure to warn. In part it involved a hopeless assertion that DVT acquired on board was an accident. In deciding the application favourably to the defendants, Hermiston J did not distinguish between acts and omissions. He relied on Scherer, where it had been held that there was no accident when a passenger suffered thrombophlebitis by sitting on a plane. Scherer did not involve any allegation that failure to warn was an accident; whether discretely or as part of a series of allegations of accident.

738. The decision in McDonald was affirmed in one line on appeal.

741. Obviously enough, the way in which McDonald was pleaded and argued was very far from the pleading and argument in this case.

744. Finally, I should refer to Volander. It was not cited in DVT Group Travel. The precise nature of the proceeding - that is, whether it was a trial or some preliminary application - is not clear to me. Some of the language used, both in the judgment at first instance and on appeal, suggests that it was the former. In any event, two issues seem to have been agitated. First, whether the claim was one falling within Art. 17. Second, whether the claim could be pursued under German national law. The precise basis of claim relied upon by the plaintiff in either connection is not extensively described.

747. Both the court at first instance and the higher regional court held that the plaintiff could not rely on Art. 17 because, as the appeal court said, "there was no accident involved with the event of damage according to [Article 17]". The appeal court said that:

"It was not an incident caused by external forces at a particular time and place, which was unforeseeable and which caused personal damage or damage to property. The present case does not involve a sudden event but rather summation damage, which is not covered by the definition of the term `accident'."

751. So far as I can see, Volander stands for the proposition that the thrombosis apparently suffered by the plaintiff was not an accident. The explanation given by the appeal court as to why that was so did not focus upon the issue of act or omission, or upon unexpectedness. In all, I do not think that the Volander is of much present assistance.

754. In reviewing the various authorities, here and overseas, presently without regard to the place of courts in the judicial hierarchy, I have been concerned to see whether uniformity of approach is disclosed upon the question of event and non-event. In my opinion the following conclusions, not all of which relate to that question, can be drawn from the cases to which I have referred:

* First, whether `accident' be paraphrased as an unexpected or unusual event or happening (Saks) or as an untoward event (DVT Group Travel) a failure to act may, as Phillips MR said in DVT Group Travel, result in an accident or form part of a series of acts and omissions which together constitute an accident. In such circumstances, as his Lordship said, it may not be easy to distinguish between acts and omissions. Although the Courts which decided them did not draw a distinction between acts and omissions, Husain, Turturro, McCaskey and Fulop may be so explained. It is readily possible to think of very many circumstances justifying his Lordship's conclusion.

* Second, a bare failure to warn or advise, unaccompanied by other circumstances, is unequivocally inaction, and not "accident" - either as an unexpected or unusual event or happening, or as an untoward event. That was the opinion of the Court of Appeal - most explicitly, of Phillips MR - in DVT Group Travel. It seems to me, with respect, that neither the frequent difficulty of distinguishing between acts and omissions, or Art. 25, the matters referred to by Bongiorno J, lead to a contrary conclusion. Bare failure to warn or advise is, as I have said, unequivocally inaction; and Art 25 does have a wider field of operation than just with respect to Art. 17. For reasons discussed, I do not think that much can be made of the decisions cited by Phillips MR, or of Volander; but neither can much be made of Blansett, Louie or Miller. All other things apart, in each of the American cases it was alleged that the failure to warn or advise was in breach of the usual, normal and expected operation of the aircraft. Moreover, neither the possible distinction simply between acts and omissions nor a possible distinction between failure to warn and the type of circumstances encountered in Husain, Fulop and McCaskey was the subject of examination in those judgments.[163]

* Third, allegations of the kind considered in Blansett, Louie and Miller were not made in DVT Group Travel. The logic of DVT Group Travel, however, is that the additional circumstances would make no difference. Indeed, although the Master of the Rolls did not illuminate the difference in the specimen matrix which was before him and the pleading in Blansett, his Lordship disagreed with that decision. In the present case, the respondent does not plead that it was industry practice, or the practice of the appellants, to warn of the risks of DVT and advise measures to reduce such risks. The allegation made by particular (c) to paragraph 21 of the statement of claim is less specific. In my opinion the occasion does not arise to consider whether in circumstances of the particular kind considered in Blansett, Louie and Miller, or just possibly as one aspect of a number of circumstances, a failure to warn could ever constitute accident; though my provisional opinion is that it could not do so.

* Fourth, none of the cases cited by counsel was on all fours with the relevant pleading in the present case. In my opinion it is not possible to say that the decision in DVT Group Travel decides the fate of this appeal. Indeed, as I have pointed out, Phillips MR agreed with the observation of Bongiorno J that the allegation of failure to warn was only part of the respondent's expanded particularisation of his claim.

* Fifth, in some cases cited in the present connection there was discussion of unexpectedness. As I earlier observed, insofar as the language used could have imported common law notions of reasonable expectation, reliance and the like, I would reject the applicability of such notions in the context of Art. 17.

* Sixth, in DVT Group Travel there was both acceptance of the definition of "accident" provided in Saks, and yet an expressed concern that the language of Art.17 should not be treated as having been replaced by the language of Saks. It is, I think, necessary to tread a careful line in that connection. It is on the one hand true that the language of Art. 17 has not altered. But on the other hand it is clear that in striving for uniformity of interpretation of an instrument such as Montreal No 4, conformably with the approach to interpretation of treaties and conventions, the meaning given "accident" by Saks has been widely accepted. It has been the source of a very considerable jurisprudence. The acceptance of Saks reflects a general acceptance that it correctly paraphrased the word "accident"; for which reason there should be no different result in a particular case whether regard is had to the critical word in Art. 17 or to the Saks' paraphrase.

764. Saks was decided in 1984. Before and since, cases have dwelt upon the characteristic which renders an event (I use that word simply to convey a circumstance or set of circumstances) an accident.

767. Cases show that there may be an accident within the meaning of Art. 17 despite the event consisting of intentional, even criminal conduct by one passenger towards another passenger or towards all other passengers, See, for example Evangelinos v Trans World Airlines Inc,[164] Krystal v British Overseas Airlines Corp,[165] Oliver v Scandinavian Airlines System [166] and Wallace v Korean Air[167].

770. The cases to which I have just referred make the point that the focus is upon whether an event is unexpected or unusual - and not upon whether it is, for example, inadvertent, careless, intentional or criminal. They also show, I think, that characterisation of an event as unexpected or unusual does not involve consideration particularly from the perspective of the airline or the passenger(s); nor, a fortiori, particularly from the perspective of a reasonably careful airline operator or a reasonable passenger in all the circumstances. The latter formulations, in any event, would introduce notions of domestic torts law; and for that reason, as I said earlier when discussing Husain and Fulop, I would not accept their application. Rather, it seems to me, the question whether an event is to be so characterised is to be answered from the perspective of a disinterested bystander who is apprised of all relevant circumstances. Would such a person describe the event so described, an event causing injury, as one that was unexpected or unusual?

773. That conclusion appears to me to be consistent with authority. In Saks, O'Connor J said that:

"...in cases where there is contradictory evidence, it is for the trier of the fact to decide whether an `accident' as here defined caused the passenger's injury."[168]

777. Her Honour cited De Marines v KLM Royal Dutch Airlines [169] in that connection. In De Marines the Third Circuit Court of Appeals held that a charge to the jury which made no suggestion other than the jury should apply its own perspective to determining whether an event was one which was unexpected or unusual "presented the correct legal standard for determining the occurrence of an accident."[170]

780. Such an approach avoids the risk that what is unexpected or unusual will be determined by reference simply to the different sectional interest of airline or passenger. But it does not mean that in a particular case the circumstances of an individual passenger may not be relevant as part of the factual matrix in which the event said to constitute an accident falls to be considered. Neither does it exclude from consideration the practice of a particular airline defendant or an airline industry practice. Husain may be said to be an example of these last observations, although there language appropriate to common law notions was at times used; and I have already expressed my opinion that such use was inappropriate.

783. Counsel for the appellants submitted, inter alia, that the learned trial judge inappropriately imported notions of the common law into resolving the question whether an event could constitute an accident. They referred to his Honour's use of phrases such as "objectively viewed", "peculiar position of reliance and vulnerability", "passengers' expectations", and "the objective requirement of the Saks test."

786. I have no doubt that his Honour used some language apposite to the common law. I am not convinced, however, that in fact he proposed or relied upon common law notions in concluding that the respondent might establish an accident in this case. It seams to me more probable that his Honour was seeking to convey the point that in deciding whether an airline's conduct causing injury was an event which should be characterised as unexpected or unusual, it was proper to bring to account the circumstances in which the conduct occurred, including consideration of the respective positions of airline and passenger; and that it was to be objectively assessed by the trier of fact in the sense that characterisation was not to be determined by reference to the subjective perspective of the airline or the passenger. I acknowledge, however, that this may not have been the intended import of his Honour's language; and if such was the case then in my respectful opinion his Honour's reasoning fell into error.

789. I should refer also to particulars (a) - (c) of para 21 of the statement of claim. I have described my understanding of their import at [83]-[86]. They refer to matters allegedly known to the airlines and unknown to the respondent and other passengers; to what was alleged to be the usual conduct of the airlines in warning of known risks; and to it being expected by passengers that such warnings would be given. The import of the particulars, I consider, is that the conduct of the airlines alleged to constitute an accident should be determined in the factual matrix thus described. There seems to me to be no resort to common law concepts in the respondent's formulation. The references to knowledge, and use of the verb "expected", must be understood in context.

793. A question which arises in this proceeding is whether it is in the nature of an accident that it should occur at an identifiable instant. The appellants so submitted. It was a consequence of the respondent's case, according to their submissions, that every passenger on each of the flights suffered an accident throughout the course of such flight; an accident which went unrecognised. Such a situation, counsel submitted, showed that it made no sense to describe events, or non-events, occurring over the course of a flight as an accident.

796. The appellants' submission gains some support from DVT Group Travel. Phillips MR referred to "an untoward event which impacts on the body in a manner causing death, wounding or injury."[171] Speaking of Saks, where O'Connor J said that it was only required "that the passenger be able to prove that some link in the chain was an unusual or unexpected event external to the passenger", his Lordship said that he had

"not found it easy... to envisage a remote accident in a chain of causation that does not end in a direct cause of the death or injury that is, itself, an accident within Article 17."

Referring to illustrations given in Shawcross and Beaumont of cases where courts had found accident, he said,

"almost all of these constitute untoward events which impacted directly and physically upon the passenger."[172]

802. A similar notion may also be implicit, having regard to the particular circumstances of the specimen metrix, in the reasons of Kay LJ.[173]

805. No doubt it is the case, as Phillips MR said, that in almost all cases the alleged accident will impact directly and physically upon the passenger. But the critical question is whether accident is so confined.

808. In my opinion an accident for the purposes of Art. 17 might be constituted by something which happened, even if unrecognised by passengers at the time, throughout the course of a flight. An example should make the point.

811. Suppose that there was a continuous malfunction in cabin pressurisation equipment throughout the course of a flight, in consequence of which the level of pressurisation was aberrant, and a passenger developed a vascular abnormality in response. Then suppose a situation in which there was no malfunction, but that a crew member set a wrong pressurisation level at the commencement of a flight, that the level was aberrant throughout the flight, and that the hypothetical passenger suffered the same injury.

814. In each of those cases there would be injury. In neither case, let it be assumed, would the passenger know in the course of the journey that the pressurisation level was continuously aberrant. In each case that aberrant feature could properly be described as an event or happening. In each case it could be properly be described as unexpected or unusual - at the very least in the meaning of departing from the routine operation of the aircraft. In only second case, however, could the aberrant function be traced to an action or happening at a moment in the course of the flight.

817. It seems to me that it would be wrong in principle, as well as being at odds with the flexible application of the Saks definition of injury which Saks commended, to treat only the second case as involving accident causing injury.

820. Further, in neither case, I have assumed, and it seems inherently probable, would any passenger know at the time of the accident that he or she had suffered the same. That should not lead to a conclusion that there had been no accident. Again, the fact that all passengers suffered accident, but only one or some suffered injury, should excite no conceptual surprise.

823. A few reported cases are susceptible of description as involving an accident occurring over a protracted period of time. In Rosman v Trans World Airlines, Inc,[174] for example, passengers claimed to have suffered physical and psychological injuries in response to the hijacking of an aeroplane and to their being held on board the plane for a period of days.[175] The defendant did not claim that a hijacking was not an accident. It might be said that the accident was confined to the moment that the hijackers took over the plane; and that all that followed was injury caused by the hijacking. But that would be to ignore the continuing nature of the hijacking, a matter of significance because some of the alleged injuries arose out of the long-time detention of the passengers on the plane.

827. In my opinion, it is fairly arguable that part but not all of the case which the respondent seeks to pursue against the appellants could constitute "accident". The import of particulars (a) and (b) to para 21 of the statement of claim is that what should otherwise have been characterised as the usual, normal and expected operation of the aircraft was not of that character by reason of the appellants' knowledge of the matters set out in particular (b); and the want of knowledge of those matters by the plaintiff and other passengers. The question what was, in all the circumstances, the usual, normal and expected operation of the aircraft is ultimately a question of fact, to be determined by evidence. Fishman is a case in which a usual operation performed in an aberrant manner, was held to constitute an accident. The present pleading does not raise an identical issue; but I am not prepared to say that the respondent's claim is evidently not viable. At least that is so in respect of the allegations raised by particular (d), probably particular (c), and just possibly particular (b) to para 6 of the statement of claim read in conjunction with particular (b) to para 21. Particulars (d) and (c) to para 6, evidently, allege acts done at different moments in the course of the flights. Particular (b) might also do so. Those particulars may be compared with particular (a), which refers to the static condition of the aircraft; and particular (e), which raises an allegation of failure to inform or warn. As to the last-mentioned, more in a moment.

830. There is, of course, a question whether the respondent could establish that the discouragement against moving pleaded by particular (d) to para 6 was causative of him not moving; and another question again whether that in turn was a cause of him developing DVT. In the case of the matter pleaded by particular (c), again, factual issues of some complexity would seem likely to arise. The matter pleaded by particular (b) is more complicated again; for it is not clear whether the pleading refers to impediments to movement created by the configuration of the seating cabin or to impediments resulting from some positive act or acts. I consider, however, that the issues which would seem likely to arise in connection with those three particulars do not provide a basis for summarily disposing of the proceeding in favour of the appellants; or striking out the allegations under discussion.

833. It is necessary, next, to refer to particular (e) to para 6 of the statement of claim read in conjunction with particular (c) to para 21. It follows from the opinion which I have expressed about failure to warn, particularly when considered in the context of the form in which particular (c) to para 21 is pleaded, that in my opinion the respondent has not there alleged an accident. No possible variation to that pleading was mooted in argument.

836. I would strike out the allegation raised by particular (e) to para 6; and I would not permit amendment to allege the proposed particular (c) to para 21. I would also strike out the allegation raised by particular (a) to paragraph 6. I would allow the appeals only to provide a vehicle for making such orders.

[1] See s.8(1)(e), s.25J and in particular the definition in s.5 of the Act. More fully the treaty is known as the "Warsaw Convention as amended at The Hague, 1955, and by Protocol No. 4 of Montreal, 1975": see also the heading to Schedule 5 of the Act and Article XV of the Montreal Protocol No. 4.

[2] The Montreal No. 4 Convention is given "the force of law in Australia" by s.25K of the Act, although regrettably neither the section nor any other part of the Act, nor any part of the Convention as printed in Schedule 5 of the Act, makes clear to which "carriages" by air the Convention applies. Article 1 refers merely to the place of departure and place of destination of the air carriage as being within the territory of "two High Contracting Parties or within the territory a single High Contracting Party". Notwithstanding that the respondent originally pleaded, and the appellants as defendants originally admitted, that the relevant convention was the Warsaw Convention as amended at The Hague set out in Schedule 2 to the Act, the parties later agreed that it was Schedule 5 which set out the relevant convention and they agreed that the pleadings should be amended to reflect that, if the action were to proceed. It is doubly regrettable that there is no easy means of ascertaining who the relevant High Contracting Parties are to treaties which are made the law of the land.

[3] Because they appeared and still appear in para.6 of the Statement of Claim and are directly incorporated into the new particulars of the accident in para.21 of the proposed amended Statement of Claim.

[4] [1986] HCA 32; (1986) 160 C.L.R. 513 at 527.

[5] [1985] USSC 43; 470 U.S. 392 (1985), at 405. Article 17 appears in identical form in the Montreal No. 4 Convention, but the official U.S. translation of the original Convention into English differs from that in Schedule 5 (and the other Schedules, including Schedule 1) of the Act in that "shall be" is substituted for the word "is" as the third word of the article: see 470 U.S. at 397. No significance has ever been placed on the difference.

[6] Especially as to the "course" of embarkation and disembarkation: cf. Maugnie v. Compagnie Nationale Air France [1977] USCA9 75; (1977) 549 F. 2d 1256.

[9] See e.g. El Al Israel Airlines Ltd. v. Tseng (1999) 525 U.S. 155 at 165fn.9 and at 177 fn.1.

[10] [2002] UKHL 7; [2002] 2 A.C. 628 at 655 per Lord Hope of Craighead para.[71] and at 672 per Lord Hobhouse para.[134].

[11] [2003] EWCA Civ 1562; [2003] 3 W.L.R. 956. It had been earlier followed by that court in Chaudhari v. British Airways P.L.C. [1997] EWCA Civ 1413, as well as in Morris [2001] EWCA Civ 790; [2002] Q.B. 100 at 110.

[14] [2003] VSCA 6 at para.[15].

[15] See Great China Metal Indonesia Co. Ltd. v. Malaysian International Shipping Corporation Berhad [1998] HCA 65; (1998) 196 C.L.R. 161 at 176, 186, 213.

[16] [2002] VSC 580.

[18] [1903] A.C. 443.

[19] At 398. The emphases are those of O'Connor, J. in both this passage and in that quoted below at fn.[.

[23] [1920] HCA 37; (1920) 27 C.L.R. 570.

[25] At 585. See also per Gavan Duffy and Starke, JJ. at 591.

[26] [1939] HCA 36; (1939) 62 C.L.R. 317.

[30] See also the observations on article 20 by O'Connor, J. in Saks at 402. Note that the form of article 20 there discussed was that in the original Warsaw Convention, but its substance was virtually identical.

32] See the second passage quoted in para.[[] fn..

33] See esp. the discussion of the French text by O'Connor, J. in Saks at 398-400, esp. the Larousse definition at 400.

[34] Fishman v. Delta Airlines Inc. [1998] USCA2 3; (1998) 132 F. 3d 138.

[37] [2002] USCA9 902; (2000) 116 F.Supp. 2d 1121; (2002) 316 F. 3d 829. An appeal by the defendant to the U.S. Supreme Court has been heard and judgment reserved on 12 November 2003.

[38] (2001) 175 F.Supp. 2d 651, though it should be noted that the defendant was ultimately successful at trial (2003) 244 F. Supp. 2d 217.

[43] Apparently no change was made to the "accident" provision, as such, by the 1999 Montreal Convention, ratified by the U.S. on 31 July 2003.

[44] Cf. McCaskey v. Continental Airlines Inc. (2001) 159 F.Supp. 2d 562.

[46] For example in Fulop the judge eventually reveals the dependence of his reasoning on notions of negligence by concluding (at 672) that, if the carriers were right, then, in cases of "inaction" or "substandard response" to a medical condition, the carrier "could escape all liability and duty of care" to the affected passenger.

[49] Assuming that other effects, such as changes of pressure or the like, are not themselves accidents.

[50] Subject to Article 20, but the onus rests on the carrier. Perhaps it is better to describe the liability as "presumed liability" rather than "strict liability".

[52] McCaskey v. Continental Airlines Inc. (2001) 159 F.Supp. 2d 562 at 574.

[53] See The Shorter Oxford English Dictionary, 3rd ed. See also the definition of accident in the Oxford English Dictionary, 2nd ed.: "An event that is without apparent cause or unexpected; an unfortunate event, especially one causing injury or damage". The Macquarie Dictionary, 2nd ed. defines accident as "an undesirable happening; casualty; mishap."

[54] In England, for example, in Morris v. KLM Royal Dutch Airlines [2002] UKHL 7; [2002] 2 A.C. 628; Deep Vein Thrombosis and Air Travel Group Litigation [2003] EWCA Civ 1562; [2003] 3 W.L.R. 956 ("the DVT case"). In Canada, in Quinn v. Canadian Airlines International Ltd. 18 OR (3d) 326 and McDonald v. Korean Air 18 February 2003, unreported. In Australia, in Agtrack (N.J.) Pty. Ltd., v. Hatfield [2003] V.S.C.A. 6 at [15] per Ormiston, J.A.

[55] See, for example, Blansett v. Continental Airlines Inc. (2002) 246 F.Supp 2d 596; Wallace v. Korean Air [2000] USCA2 205; (2000) 214 F.3d 293.

[56] See, for example, Morris and the DVT case.

[57] [2002] UKHL 7; [2002] 2 A.C. 628.

[58] [2003] EWCA Civ 1562; [2003] 3 W.L.R. 956.

[59] [1985] USSC 43; (1985) 470 US 392.

[63] [1996] 1 W.L.R. 1026 at 1035.

[64] The Master of the Rolls used the term "typical procedures of the industry" at [33].

[66] At [70] and [82] respectively.

[67] (2000) 116 F.Supp. 2d 1121, US DC.

[68] (2001) 159 F.Supp 2d 562, US DC.

[69] (2001) 175 F.Supp. 2d 651, US DC.

[70] The hearing below was conducted on the basis that it was Art. 17 of the Convention as amended at The Hague in 1955 that was in point. The Convention as so amended is Schedule 2 to the Civil Aviation (Carriers' Liability) Act 1959. The Convention as further amended at Montreal in 1975 is Schedule 5 to that Act. It was eventually common ground on the appeal that the Convention as amended in 1975 -the Montreal No. 4 Convention - applied in the present case. There is no difference in the text of Art. 17.

[72] Appeal Book, A23, para numbered 1.

[73] Appeal Book, A43-44, para numbered 7.

[74] Paragraph 13 of Qantas amended defence dated 25 March 2002, Appeal Book, A 47. Paragraph 9 of BA defence dated 12 October 2001, Appeal Book, A 33.

[75] Appeal Book, A60-61. On the appeal, respondent's counsel handed to the Court a document headed "Proposed Amendments to Statement of Claim". So far as para 21 is concerned, it was in the same language as the Further and Better Particulars.

[77] Reasons, [11]-[12], Appeal Book C 6 - 7.

[78] Reasons, [17]-[24], Appeal Book C 8 - 12. His Honour was referring to the Convention rather than Montreal No 4; but as I have said earlier, nothing turns on that.

[79] [1985] USSC 43; 470 US 392 (1985).

[80] Reasons, [29]-[31], Appeal Book, C 14 - 15.

[81] [1903] AC 443.

[82] [1986] HCA 32; (1985) 160 CLR 513 at 527.

[83] 116 F Supp 2 d (ND Cal 2000).

[84] Reasons, [39], Appeal Book, C19.

[85] Reasons, [43], Appeal Book, C20.

[86] Reasons, [44], Appeal Book, C21.

[87] [2003] EWCA Civ 1005

[88] I omit many submissions which mirrored those made on behalf of Qantas.

[89] 175 F Supp 2 D 651.

[90] [2002] WL 31746730 (SD Tex); and US District Court, Judge Kent, 11 December 2002, unreported

[91] 528 F. 2d 313 (1975).

[92] [2000] USCA2 205; 214 F. 3d 293 (2000).

[94] [2003] VSCA 6 at [15] per Ormiston JA.

[95] Reasons, [40], Appeal Book, C 19.

[96] Reasons, [21], Appeal Book, C 10.

[100] At 399. It is strongly arguable that it is not the French legal meaning of "accident" which could be in point. Authorities suggest that the words of a multipartite treaty should be interpreted according to principles of international law and not domestic law - which must logically extend to the domestic law of all countries and not simply the domestic law of a country the subject of particular litigation.

[101] [1999] NSWCA 415; (1999) 47 NSWLR 696 at [50] and [56] per Mason P.

[102] [1996] UKHL 5; [1997] AC 430 at 443 per Lord Hope.

[103] [2002] UKHL 7; [2002] 2 AC 628 at [5], [7], [81], [127], [147], [178].

[104] At [77] and [79], per Lord Phillips, MR.

[105] [1961] AC 807 at 840 - 841 per Viscount Simonds.

[106] [1980] HCA 51; (1980) 147 CLR 142 at 159 per Mason and Wilson JJ: "... a national Court, in the interests of uniformity, should construe rules formulated by an international convention `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'..."

[107] [1998] HCA 65; (1998) 196 CLR 161 at [38] per Gaudron, Gummow and Hayne JJ, [71] per McHugh J and at [137] per Kirby J.

[108] That is, courts of first instance in the Federal hierarchy.

[111] At [5]. See also per Lord Hope at [127].

[112] See also statements to similar effect as follows: Sidhu, per Lord Hope at 443 D_; Fothergill v Monarch Airlines Ltd_ [1980] UKHL 6; [1981] AC 251 where Lord Wilberforce referred at 286 to doubt where "the preponderance in quantity, or quality, lay"; see also per Lord Diplock at 284 C.

[113] See, recently, Sidhu at 453 C - H per Lord Hope; and Morris at [15] - [16] per Lord Steyn and at [65] - [66] per Lord Hope. See also Eastern Airlines v Floyd [1991] USSC 58; 499 US 530 (1991) and El Al Israel Airlines Ltd v Tseng, citation earlier.

[119] For example, in Chaudhari v British Airways plc (Court of Appeal, unreported, decision 16 April 1997), Morris, DVT Group Litigation.

[120] See Quinn v Canadian Airlines International Ltd 18 OR (3d) 326, affirmed OJ No 1832 (QL) and McDonald v Korean Air, Ontario Superior Court of Justice, 18 September 2002, unreported; affirmed, Court of Appeal for Ontario, 18 February 2003, unreported. Perhaps also in the Republic of Germany. See _Volander v Deutsche Lufthansa (_Frankfurt am Main Regional Court, 29 October 2001; affirmed Frankfurt am Main Higher Regional Court, 6 November 2002).

[121] In Agtrack v Hatfield, citation earlier_._

[124] At 400; 404 - referring to the construction of European legal scholars; and 405 - "an unexpected or unusual event or happening."

[125] 159 F Supp 2d 562 (2001).

[126] [1998] USCA2 3; 132 F 3d 138 (2nd Cir, 1998).

[127] 128 F Supp 2d (Southern District New York, 2001).

[129] District Court of New South Wales, Judge Knight, 11 October 2002, unreported.

[130] District Court of Queensland, Boulton DCJ, 7 February 2003, unreported.

[131] US District Court, Alaska, Judge Singleton, 24 June 2002, unreported.

[132] US District Court, ND Cal, Judge Walker, 2003, unreported.

[133] [1997] USCA11 1951; 119 F. 3d 1515(11th Cir, 1997).

[134] 739 F. 2d. 130 (3d Cir, 1984).

[135] At 116 F Supp 2d 1121 (ND Cal 2000)

[148] Unreported decision, 11 December 2002, at 7.

[149] 387 NYS 2d 580 (App Div 1976).

[150] The unrevised version of the judgment reads "no further or other steps" - which would seem to make more sense.

[164] [1977] USCA3 143; 550 F 2d 152 (CA3 1977).

[165] 403 F Supp 1322 (CD Cal 1975).

[166] 17 CCH Air. Cas. 18,283 (Md 1983).

[167] [2000] USCA2 205; 214 F 3d 293 (2nd Cir. 2000).

[169] [1978] USCA3 624; 580 F 2d 1193 (CA 3 1978).