[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	England and Wales Court of Appeal (Civil Division) Decisions
You are here: BAILII >> Databases >> England and Wales Court of Appeal (Civil Division) Decisions >> Nugent & Anor v Michael Goss Aviation Ltd & Ors [2000] EWCA Civ 130 (14 April 2000) URL: http://www.bailii.org/ew/cases/EWCA/Civ/2000/130.html Cite as: [2000] EWCA Civ 130

[New search] [Printable RTF version] [Help]

Case No: 1999/0501

IN THE SUPREME COURT OF JUDICATURE
IN THE COURT OF APPEAL CIVIL DIVISION
ON APPEAL FROM THE QUEEN'S BENCH DIVISION
Royal Courts of Justice,
Strand, London, WC2A 2LL
Friday 14 April 2000

As is plain from those words, where a claimant relies on recklessness he must also prove knowledge of probable damage.

The main issue in the appeal is as to the meaning in Article 25, and in a corresponding provision for a carrier's servants or agents in Article 25A, of the words "with knowledge that damage would probably result". The original Convention resulted from two international conferences, one in Paris in 1925 and one in Warsaw in 1929, and was given effect in this country by the Carriage by Air Act 1932. It is now set out in Schedule 1 to the Carriage by Air Acts (Application of Provisions) Order 1967, made under section 10 of the Carriage by Air Act 1961, and includes, with some immaterial exceptions, carriage by air wholly within the United Kingdom. The provision was amended to its present form by the Hague Protocol as a result of the ICAO conference at the Hague in 1955/6.
The facts and proceedings

The claim is for damages under the Fatal Accidents Act 1976 arising out of the death on 22^nd October 1996 of Mr. Matthew Harding in the crash of a helicopter in which he was travelling from Bolton to London as a passenger. The helicopter was owned by the third defendant and piloted by Michael Goss, who was also killed in the crash and whose executor is the second defendant. The Convention governed the flight. The first and second defendants conceded liability to the claimants under Article 17 of it, which renders an air carrier liable, irrespective of fault, for the death of a passenger in an accident on board its aircraft or in the course of embarking on or disembarking from it, but relied on Article 22 which limits such liability to a sum equivalent to about £80,000. The claimants value the claim at about £59m. and seek to rely on Articles 25 and 25A of the Convention (to which I shall refer simply as "Article 25").
The Master struck out the paragraphs of the amended statement of claim in which the claimants pleaded the Article 25 case. On their appeal to Burton J. the matter was argued on the basis of a sought re-amendment of the claim. He held that it was unsustainable, refused leave to re-amend and gave judgment against the first and second defendants limited to £80,000. The claimants contend that the Judge, in so doing, wrongly construed the words "with knowledge that damage would probably result" and that, in any event, it was not a case for striking out and he should have allowed the sought re-amendment.
The claimants did not persist with an earlier allegation that the pilot, Michael Goss, had intended to cause damage, the first alternative in Article 25. Their sought re-amended claim (which I shall call their "pleaded case") against the first two defendants was under the second alternative, recklessness in a course of conduct coupled with knowledge of probable resultant damage. As to recklessness, they said that the pilot, had: 1) failed to keep his flying skills up to date; 2) failed to acquaint himself with navigational aids with which the helicopter was equipped; 3) failed to plan the flight properly; and 4) flown when he was tired. As to knowledge of probable damage, they said that the probability of damage was "within his knowledge" in that, even if it was not present in his mind at the material time, "if he had addressed his mind to the matter he would have appreciated, by reason of his knowledge and skill as an experienced pilot, that death or serious injury was probable". Alternatively, they said that he had actually known, in the sense of appreciated at the time of the conduct relied on, that it would probably result in damage, a claim that they now acknowledge would be difficult to prove.
Burton J. rejected the claimants' contention that actual knowledge could include background knowledge, holding that it means actual knowledge, in the sense of appreciation, at the time of the conduct in question that it would probably cause the type of damage which occurred - here of death or of serious injury. In so holding, he considered that he was bound by two decisions of this Court: Goldman v. Thai Airways [1983] 1 WLR 1186 and Gurtner v.
Beaton [1993] 2 Ll. 369. He indicated that, in his view, such a construction does not necessarily require an allegation that a pilot has been suicidal, but could also include a case where he has taken a stupid and unnecessary risk -
"knowing that it carrie[d] a probable risk of death, or perhaps where some act, the airborne equivalent of road rage, [has been] committed by an impatient or angry pilot."
As to the claimants' alternative contention of actual knowledge, in the sense of appreciation at the material time, of probable damage, he concluded that they had no prospect of proving it.
The authorities
Goldman concerned a claim by a passenger in respect of injury resulting from the pilot's failure, when the plane was entering an area of turbulence, to illuminate the seat belt instruction light, with the result that the claimant, whose seat belt was unfastened, was thrown from his seat and seriously injured when the plane encountered turbulence. The trial judge had found that the pilot had been reckless, applying the criminal test of recklessness laid down by the House of Lords in R v. Caldwell [1982] AC 341 and R. v. Lawrence [1982] AC 510, and that he had known of the probability of damage. He held, therefore, that the claimant had satisfied the Article 25 test and awarded him full damages. The Court of Appeal reversed that decision, holding that the test of recklessness was subjective and that it was doubtful whether, on that basis, the pilot had been reckless, and that, in any event, there was no evidence that he had known of the probability of damage resulting from his conduct. Eveleigh LJ, who gave the principal judgment and with whom O'Connor and Purchas LJJ agreed, characterised, at 1192F-G, the trial judge's finding as to the latter as one of knowledge "that damage of some kind would probably result if the risk (i.e. of encountering
... [turbulence]) which he was taking, materialised". In his view, that was not enough; the claimant had also to prove knowledge at the time of the relevant acts or omissions in the sense of actual, not attributed, knowledge that damage would probably result. This is how he put it, at 1194G-H and 1195H-1196A:
" If the pilot did not know that damage would probably result from his omission, I cannot see that we are entitled to attribute to him knowledge which another pilot might have possessed or which he himself should have possessed. I appreciate that, when introducing an English version to coincide with a French text, there is naturally
an inclination to follow the pattern of that text and where possible to avoid a free translation. Even so, I cannot believe that lawyers who intended to convey the meaning of the well-known phrase `when he knew or ought to have known' would have adopted `with knowledge'. ...
An act may be reckless when it involves a risk, even though it cannot be said that the danger envisaged is a probable consequence. It is enough that it is a possible consequence, although of course there comes a point where the risk is so remote that it would not be considered reckless to take it. We look for an element of rashness which is perhaps more clearly indicated in the French text `temerairement'. Article 25 however refers not to possibility, but to the probability of resulting damage. Thus something more than a possibility is required. The word `probable' is a common enough word. I understand it to mean that something is likely to happen. I think that is what is meant in article 25. In other words, one anticipates damage from the act or omission." [my emphasis]
He found support for his interpretation in travaux preparatoires derived from the 1955/6 Hague Conference resulting in the present Article 25.
Purchas LJ said much the same in his concurring judgment. In the following passage, at 1202B-C, he emphasised the importance of the pilot's knowledge in the sense of his appreciation at the time of the relevant acts or omissions that damage would probably result:
" I agree that the true interpretation of article 25 when it is read as a whole involves the proof of actual knowledge in the mind of the pilot at the moment at which the omission occurs, that the omission is taking place and that it does involve probable damage of the sort contemplated in the article. I also agree that this interpretation is supported by the travaux preparatoires and in particular with the selections of the particular version which finally found its way into the article in favour of the alternatives which had been debated." [my emphasis]
Gurtner concerned the crash of an aircraft into a hillside by a pilot who mistakenly thought that he was flying over low ground. Counsel for the claimants, passengers and the estates of passengers killed in the crash, put at the forefront of his submission on this point that the test of probable knowledge of damage was whether a pilot knew of the probability, if the risk materialised, that damage of the type which in fact occurred would occur. The Court rejected that and other submissions, following Goldman. Neill LJ, giving the judgment of the Court, referred with evident approval to all three judgments in that case, including the above passage from that of Purchas LJ, and said at 387:
" It seems to us that the words of arts. 25 and 25A focus attention on the act or omission. The introduction of the phrase `if the risk materialises', which was a central feature of the plaintiffs' argument, is not permissible. We are satisfied that
the articles mean that the pilot must do or omit to do something `with knowledge that damage would probably result' from that act or omission."
Neill LJ added that the Court was bound by the decision in Goldman and expressed the view that its reasoning was consistent with the decision of the Court of Appeal in New South Wales in SS. Pharmaceutical Co. Ltd. Quantas Airways Ltd. [1991] 1 Ll. 288
In SS. Pharmaceutical the claim was in respect of damage to air cargo as a result of the defendant having left it out in the open in the course of transit, conduct that it had acknowledged to be "deplorably bad handling". The claimants' evidence was that the cartons containing the cargo bore a sign indicating that exposure to water would damage it and that the weather forecast for the day in question was of rain. The defendant called no evidence. The trial judge, Rogers J., held that the claimants had satisfied the Article 25 test because, on the evidence, the defendant knew of the likelihood of damage to specially vulnerable cargo in the weather conditions then obtaining. In doing so, he cited with approval, [1989] 1 Ll 319, at 330, the following proposition of the Editors of the then edition of Shawcross & Beaumont on Air Law, at para. 447, which is flatly contrary to the claimants' primary case that background knowledge will do:
"It is undeniable that the actor himself must actually have knowledge that damage would probably result. It is not enough to show that some other person had that knowledge, or that he would have had it if only he had applied his mind to the matter." [my emphasis]
The New South Wales Court of Appeal unanimously upheld Rogers J's conclusion on the law and by a majority his decision on the facts, holding that, on the claimants' evidence and the
inferences which could be drawn from the defendant's failure to call evidence, it had been open to him to find that the damage to the cargo resulted from the defendant's recklessness in the knowledge that it would probably cause the kind of damage in question.
There was no issue before the Court of Appeal - though there had been before Rogers J. - as to the meaning of knowledge of the probability of damage. The majority (Gleeson CJ and Handley JA) expressed, at 290-291, the view that Goldman had settled a number of questions arising out of Article 25, including the subjective meaning of knowledge as Purchas LJ had articulated it. They commented on the agreed stance of the parties on the appeal that the claimants had to establish both recklessness and "actual subjective knowledge of the kind referred to in art. 25":
"That is, without doubt, a stringent requirement. It was described in argument as `a standard of highly reprehensible conduct'. We would not disagree with that, provided the more colourful description is not permitted to replace the language of the article as to the test to be applied. Reference was also made to the juxtaposition in art. 25 of the concept of intent to cause damage and the concept which is of present relevance, the latter being said to take some of its colour from the former. Again, so much may be accepted, provided the argument is not pressed beyond its proper limits. Plainly the second concept is different from the first, otherwise its presence would add nothing to the content of the article. Nevertheless it helps to reinforce the valid points that the state of mind involved goes beyond mere carelessness and that actual, as distinct from merely imputed, knowledge must be shown."
Kirby P, though dissenting on the facts, was of like mind as to the law. After a thorough examination of the construction of Article 25 in the context of its international historical development, including reference to the travaux preparatoires, he said, in clear approval of the Goldman decision, at 301:
"Proof of actual knowledge on the part of the carrier, its servants or agents, must be established. And it must be shown to exist at the moment at which the act or omission complained of occurs. This is the consequence of the application of a subjective test." [my emphasis]
I should not leave the main authorities without referring to the history of the Article and the travaux preparatoires. However, I shall not attempt to emulate the thorough examinations of Eveleigh LJ and Kirby P, and I do so with caution, bearing in mind the warnings to that effect of the Appellate Committee in Fothergill v. Monarch Airlines [1981] AC 251, HL, in particular, that of Lord Wilberforce at 278A-B. The requirement in the 1929 predecessor of the Article for overcoming the limit on recoverable damages was "wilful misconduct" or -
"by such default on ... [the carrier's] part as, in accordance with the law of the Court seised of the case, is considered to be equivalent to wilful misconduct".
The amendment at the 1955/6 Hague Conference of the Article to its present form of recklessness coupled with knowledge of probable damage was prompted by differing views of the courts of signatory states to the Convention as to whether the term "wilful misconduct" imported an objective or subjective standard of mind and behaviour; see SS. Pharmaceutical, per Kirby P. at 299 and 301, and Shawcross & Beaumont, Air Law, 1999, Cap VII, paras. 446 and 447. Many, if not most, gave it a subjective interpretation; others, the French courts in particular, appear to have favoured an objective approach; see Miller, Liability In International Air Transport, 1977, Cap. XI. After much debate between the various national
delegations the new formulation emerged as a result of a narrow majority vote favouring it over two other candidates, recklessness alone and recklessness coupled with imputed knowledge of probable damage. That move to identify actual knowledge and to reject imputed knowledge is an indicator of the narrow scope intended for the provision. It was clearly influential on the Court of Appeal's, particularly Eveleigh LJ's, reasoning in Goldman, and, it seems to me, works against rather than in favour of the claimants' case of background knowledge as sufficient for the purpose.
Even before the tightening effect of the 1955/6 amendment, cases referred to in Shawcross & Beaumont, at para. 667, on the meaning of "wilful misconduct" indicate that English judges had focused on the carrier's appreciation or awareness at the time of the conduct in question that it was wrongful. Thus, in Horabin v. British Overseas Airways Corporation [1952] 2 All ER 1016, Barry J, in summing up to a civil jury, directed them, at 1019F-G, 1020D and 1022E-F:
" Wilful misconduct is misconduct to which the will is a party, and it is wholly different in kind from mere negligence or carelessness, however gross that negligence or carelessness may be. The will must be a party to the misconduct, and not merely a party to the conduct of which complaint is made. ... To establish wilful misconduct on the part of ... [the] pilot, it must be shown, not only that he knowingly (and in that sense wilfully) did the wrongful act, but also that, when he did it, he was aware that it was a wrongful act, i.e, that he was aware that he was committing misconduct. ...
...The element of wilfulness is essential ....
To be guilty of wilful misconduct the person concerned must appreciate that he is acting wrongfully, or is wrongfully omitting to act, and yet persist in so acting or omitting to act regardless of the consequences, or acts or omits to act with reckless indifference as to what the results may be." [my emphases]

Ackner J put the matter in much the same way in his judgment in Rustenburg Platinum Mines Ltd. v. South African Airways [1977] 1 Ll. 564, at 569 (affirmed [1979] 1 Ll. 19, CA):
"... It is common ground that `wilful misconduct' goes far beyond any negligence, even gross or culpable negligence, and involves a person doing or omitting to do that which is not only negligent but which he knows and appreciates is wrong, and is done or omitted regardless of the consequences, not caring what the result of his carelessness may be." [my emphasis]

Before moving on to the submissions in the appeal, I should also pause to consider the effect in the 1955/6 amendment of coupling recklessness with knowledge of probability of
damage. Recklessness is notorious for its different meanings in English law according to the subject matter. And it is to be remembered that we are interpreting an international convention, albeit one incorporated into English law by statute. As Eveleigh LJ said in Goldman, at 1193H, citing Fothergill, English decisions on such statutes, if resorted to at all, should be treated with caution. I take for the purpose of analysis Lord Diplock's Caldwell and Lawrence definition of recklessness, an obvious risk of damage and failure to give any thought to the possibility of it or recognition of the risk and going on to take it. Adding a further ingredient, as in Article 25, of knowledge of the probability of damage may or may not, depending on the obviousness of the risk, add much to the task of inferring that a carrier recognised the risk and went on to take it. The greater the obviousness of the risk the more likely the tribunal is to infer recklessness and that the defendant, in so doing, knew that he would probably cause damage. As a matter of proof the two will often stand or fall together, as happened in Goldman; see in particular per Eveleigh LJ at 1199F and 1200E, and in SS. Pharmaceutical, per Gleeson CJ and Handley JA at 293. As so often, practical considerations of what a tribunal is prepared to infer as to a defendant's state of mind may be more determinative than fine matters of principle of what one legal concept adds to another. Thus, as I have indicated, the majority of the New South Wales Court of Appeal in SS.
Pharmaceutical was prepared, in the circumstances of the defendant's acknowledgement of its "deplorably bad handling" of the cargo and of its failure to call evidence, to hold that Rogers J. had been entitled to infer both recklessness and knowledge of the probability of damage. Here, on the claimants' pleaded case, the Judge was prepared - just - to accept that there was an arguable case of recklessness, but not, on the pleaded facts, of a knowledge of probable damage.
The submissions
Sir Sydney Kentridge, QC, on behalf of the claimants, submitted first, and briefly, that even on Burton J's interpretation of the law, he should not have struck out the case as unsustainable since knowledge is almost always likely to be a matter of inference best determined after a full trial on the facts. He and Mr William Wood, QC, referred to the claimants' wish to call expert evidence of such matters as Michael Goss's knowledge and experience as a pilot, of what he did before taking off from Bolton and of what he, as an experienced pilot, would have learned and must have known.
Sir Sydney submitted secondly, and as the main burden of the appeal, that the Judge's interpretation and application of the law to the issues in the case were wrong. In so submitting, he accepted for the purpose of the proceedings in this Court that Goldman and. Gurtner establish: 1) that Article 25 is concerned with a carrier's actual knowledge of probable damage, not with knowledge that he ought to have had; and 2) that actual knowledge of probable damage is not established by proving knowledge that damage would probably result if a particular risk materialised, the unsuccessful argument in Gurtner. His submission was subtly different from that in Gurtner, namely, that actual knowledge of probable damage for this purpose includes "facts within a pilot's knowledge, even if not present in his mind at the time of the relevant acts or omissions, but which, had he thought about them, would have led him to appreciate the probability of damage". He likened this notion of background knowledge to information stored on the hard disc of a computer. He accepted that Purchas LJ's formulation in Goldman of "knowledge in the mind of the pilot at the moment at which the omission occurs" is plainly against his submission. However, he said that such a precise formulation is not found in the other judgments and maintained that it was not necessary for the decision.
Sir Sydney sought to support his argument by noting that an important function of the Convention is to provide for aircraft crashes and that in the case of pilot error causing crashes and injury to or death of passengers the Judge's interpretation effectively limits the application of Article 25 to its first alternative of "intent to cause damage". Only a suicidal pilot, he said, would behave in a way that he appreciated at the time would probably cause death or serious injury to all in the plane, including himself. He submitted that Burton J wrongly left no scope for the second alternative in Article 25 of reckless pilot error and that the examples of stupidity and road rage that he gave as possible applications for it were in truth examples respectively of inattention or emotional distraction excluding appreciation at the material time of the probable result of such conduct. He put forward as equally apt candidates for the working of the provision, the day-dreaming or tired or drunken pilot who, but for such condition, would have appreciated that what he was doing would probably cause the plane to crash. The critical inquiry, he submitted, is as to what a pilot knew, not as to what was in his mind at the material time, an exercise which, he maintained, the Court did not have to undertake in Goldman.
Sir Sydney's submission was, therefore, that the test of background knowledge satisfies the wording of Article 25, is consistent with authority and makes sense of the inclusion in the Article of recklessness as an alternative to an intent to cause damage, so providing, inter alia, for pilot error resulting in damage. He maintained that the claimants' pleading satisfies that test and that, in any event, it is not a plain and obvious case for striking out, whatever the proper interpretation of "knowledge" for the purpose.
Mr. Philip Shepherd, for the first two defendants, took as his starting point the clear statements in Goldman of Eveleigh and Purchas LJJ, with O'Connor LJ's agreement, that the test is one of actual knowledge at the time of the relevant acts or omissions resulting in damage of the probability that damage of that kind would result. He submitted that Sir Sydney's notion of a pilot's knowledge for this purpose, namely background knowledge, is not actual knowledge at the material time at all. He also submitted that such a notion conflicts with the clear purpose of Article 25 as it has developed to its present form and with the rejection in Gurtner of the similar gloss attempted by the claimants in that case. He also took the Court to the jurisprudence of other jurisdictions, in addition to Australia, supporting the English courts' approach, and to some of the travaux preparatoires in the form of contributions to the debate at the 1955/6 Hague Conference.
Mr. Shepherd contended that, although this is a strike-out application: 1) the principle of law is decisively against the claimants' pleaded case; and/or 2) that, whichever view of the law as to the meaning of "knowledge" is correct, the claimants' pleaded case, taken at its highest, fell far short of it, not even making out a case of recklessness; and/or 3) that, in any event, the pleading is of insufficient particularity under RSC Order 18, rule 12(1)(a) and (b) for an allegation of what amounts to wilful default analogous to a charge of manslaughter. For those reasons, he submitted, this is a plain and obvious case for a strike-out, whether considered under RSC Order 18, rule 19 or the possibly more rigorous regime of CPR Parts 3.4 (2)(a) and 24.2. As to the inferential nature of the exercise in determining the pilot's state of mind, he made detailed submissions to the Court as to the claimants' long-standing failure to particularise their case by alleging facts from which it could be inferred that the pilot knew, or even should have known, at the material time that he was probably going to kill or seriously injure his passengers and himself. In the absence of such material, he questioned what light might be thrown on his state of mind by expert evidence which the
claimants say they wish to call. He submitted that, in any event, such an exercise could only be a search for attribution of knowledge, the very thing that the Court has eschewed in Goldman, in particular Eveleigh LJ in the passage at 1194G-H that I have quoted.
.
Conclusions
Goldman and Gurtner are binding on this Court and SS. Pharmaceutical is of strong persuasive authority. There is nothing in them that would support the subtle gloss on the word "knowledge" relied upon by the claimants. On the contrary, as I have indicated, the general reasoning and articulation of principle in those authorities is against it, as also are the
travaux preparatoires and much of the other foreign jurisprudence to which we have been referred. The essential question is whether it is at least reasonably arguable that the suggested gloss satisfies the test stated and applied by the Court in those decisions. Burton J was of the view that it does not. He said, at pages 5-6 of the transcript of his judgment:
" Subject to any possible successful appeal against the two Court of Appeal decisions, which are binding upon me, it is therefore not open to Mr. Wood QC to argue ... that it is not necessary to show actual knowledge of the probability of death resulting, but only that the probability was within his knowledge, i,e. would have been apparent to him if he had turned his mind to it. This alternative method of formulation ... seem to me only another way of alleging negligence or perhaps gross negligence or even perhaps turning a blind eye. But none of this would amount to actual knowledge as defined and expressed by the Court of Appeal decisions and has been found by them, in construing Article 25, to be required."
In my view, Burton J was correct in that ruling. Sir Sydney's contention is in effect that a carrier, in particular a pilot, should be deemed to know at the time of his conduct resulting in damage that such conduct will probably result in damage even if he does not have that knowledge present in his mind at the time. Such an interpretation would not only be offensive to the plain meaning of the word "knowledge" in this context, which it is conceded means "actual knowledge"; it would also be contrary to the firm statements of principle by the Court of Appeal in Goldman and Gurtner. It is true, as Sir Sydney has pointed out, that this issue as to background knowledge did not call for specific consideration on appeal in either of those cases or in SS. Pharmaceutical. However, I do not accept the distinction that he seeks to draw between imputed knowledge and background knowledge, the latter, but not the former, counting as actual knowledge for the purpose. Putting the two notions into
separate compartments is difficult both logically and in their factual application to the exercise of determining a man's state of mind. It is also hard to justify philosophically or as a matter of common sense in a process clearly aimed at determining fault sufficient to overcome the limits of recoverable damages in Article 22. In my view, it runs contrary to the clear intention, as revealed by the words of Article 25 and of those responsible for the 1955/6 amendment, to qualify the notion which common law jurisdictions know as recklessness with a readily identifiable additional ingredient. Purchas LJ put the matter most precisely in Goldman, but he merely stated what was implicit in Eveleigh and O'Connor LJJs' reasoning. And, as I have shown, Purchas LJ's words have the authoritative support of the New South
Wales Court of Appeal in SS. Pharmaceutical. The key, as Neill LJ said in Gurtner, is the state of mind of the pilot at the time of the material conduct. In my judgment, the additional ingredient is actual knowledge, in the sense of appreciation or awareness at the time of the conduct in question, that it will probably result in the type of damage caused. Nothing less will do.
As to Sir Sydney's concern that Article 25 would be virtually unusable in the case of pilot error unless it is given the meaning for which he contended, a similar argument in Gurtner did not deter the Court from following with approval the Goldman test. Equally, the Canadian Federal Court, in Swiss Bank Corporation v. Air Canada [1981] 129 DLR (3D) 85, did not shrink from the significant limitation on the application of Article 25 in the case of pilot error by the addition to recklessness of the requirement of "knowledge that damage would probably result". It said, at 100:
"In the case of a pilot of a light aircraft, who ex hypothesi is exposed to the same personal risks as his passenger, it must be a very extreme case in which it could be held that the pilot acted recklessly and with the requisite knowledge. If so he would be deliberately hazarding not only the aircraft, but his own life with the knowledge that damage would probably result."
Burton J also squared up to that consequence - or "observation of human behaviour rather than ... statement of law", as Owen J put it at first instance in Gurtner. He said, at page 6 of the transcript of his judgment:
"... it is not sufficient for the plaintiffs to plead and establish that the pilot was reckless, or that he ought to have appreciated a risk, or that he did appreciate that he was taking a risk that he should not have taken. He has to have known that he was more likely than not going to cause death. In a private aircraft or helicopter that
would mean, as was pointed out in Swiss Bank Corporation v. Air Canada ..., his own death."
I respectfully agree with the Canadian Federal Court's and Burton J's acceptance of the limited role in the case of pilot error for the second alternative in Article 25. However, as O'Connor LJ observed in Goldman, at 1200H, the Article is not just concerned with death of or injury to passengers, but also with damage to or loss of baggage and cargo. The Article has to be construed consistently over its whole field of operation, not skewed to meet one particular aspect of it. Even in the case of death or injury, there may be many causes attracting liability other than pilot error and in which the dilemma to which Sir Sydney referred does not arise. These may include, for example, conduct of cabin crew in their care of passengers, errors of ground staff in the positioning of boarding equipment or of maintenance staff in keeping the aircraft airworthy. In any event, as the Canadian Federal Court observed, pilot error of the sort for which Article 25 in part provides - flying in a manner or in circumstances in which the pilot anticipates the probability that he will seriously injure or kill himself and/or his passengers - is by its very nature likely to be an extreme case. The examples ventured by Burton J. of a pilot taking a "stupid risk" or acting in a rage are possible applications of the provision. As Mr. Shepherd submitted, both could involve the conscious and deliberate taking of a risk in which the pilot appreciates that he and/or his passengers will probably suffer serious injury or death. The threshold for recovery under the Article is plainly high and intended to be so, as the United Kingdom's comments on its final draft at the 1955/6 Conference observe:
"7. Intention to cause damage The proposal ... to amend Article 25 of the Convention raises the question whether the amended article is too favourable to the carrier ... The United Kingdom agrees with the view that, as a reasonable limit is set to the operator's liability in return for giving the claimant a right to secure compensation without the burden of proof of negligence, the Convention should limit very narrowly indeed, that is to those where there is an element of criminal intent, the cases where liability is unlimited."
Accordingly, my view is that whether considered under the RSC or the CPR this was a suitable case for a striking out order. It was doomed to fail or in today's terms has no real prospect of success. The Judge applied the correct test as to knowledge of probable damage and on that test he was entitled to hold on the claimants' pleaded case that it was unsustainable. Although determining such matters of knowledge - often, as I have said, an exercise dependent on the drawing of inferences - is normally a fact-sensitive inquiry, the
claimants acknowledge that it would be difficult for them to prove it on the Goldman basis as the Judge interpreted it. In my view, the case is one which the Judge, applying the old or new test, was entitled to hold was "unsustainable". As a matter of pleading, it was not capable of allowing the extreme inference necessary in a pilot error case that the pilot knew that he was probably going to kill or seriously injure his passengers and himself. Certainly there would need to be some indication in the pleading or outside it of the availability of evidence of facts sufficiently compelling to support such an inference, and the claimants have produced none here. There is also the point / that Mr. Shepherd advanced as a revealing weakness of the claimants' case - they have not, as they should have done, particularised adequately so serious an allegation; see Cannock Chase Council v. Kelly [1978] 1 WLR 1, ca, per Megaw LJ at 6F-G. Nor have they set out a credible case on causation which the defendants, by their defence, have put strongly in issue in reliance on the report of an Air Accident Investigation Board's inquiry into the accident.

I would dismiss the appeal.

PILL LJ:

I agree in the result. Save on one aspect of the case, I respectfully agree with the reasoning of Auld LJ.
A claimant who seeks to rely on Article 25 of the Warsaw Convention must prove 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 ... ." I agree that it is only in an extreme case that a claimant will be able to establish these requirements. The requisite knowledge must be proved as well as recklessness. As the
Canadian Federal Court stated in Swiss Bank Corporation v Air Canada [1981] 129 DLR (3D) 85, 100:
"In the case of a pilot of a light aircraft, who ex hypothesi is exposed to the same personal risks as his passenger, it must be a very extreme case in which it could be held that the pilot acted recklessly and with the requisite knowledge. If so he would be deliberately hazarding not only the aircraft, but his own life with the knowledge that damage would probably result."
I agree that recklessness may include a state of mind in which no thought is given to the possibility of there being a risk of serious harmful consequences (R v Lawrence [1982] AC 510 per Lord Diplock at p 526) and that the present case is concerned with what Auld LJ has described as the "further ingredient in Article 25, that is, knowledge that damage would probably result".
It is established that the words "if the risk materialises" cannot be read into the second limb of Article 25 and that "with knowledge" cannot be equated with "when he knew or ought to have known" (Goldman v Thai Airways [1983] 1 WLR 1186 and Gurtner v Beaton [1993] 2 Lloyd's Rep 369).
Sir Sydney Kentridge QC, for the claimants, accepted both those propositions. He accepted that there is a distinction between imputed knowledge and actual knowledge but actual knowledge includes the actor's store of knowledge, he submitted. While the expression "background knowledge" was mentioned and canvassed at the hearing, the expression Sir Sydney used, namely, "store of knowledge", or the expression "fund of knowledge", is more appropriate to reflect his submission as I understood it.
Mr Shepherd, for the first and second defendants, submitted that there must be actual knowledge at the time of the relevant acts or omissions. He relied on the statement of Purchas LJ in Goldman that reliance on Article 25 "involves the proof of actual knowledge in the mind of the pilot at the moment at which the omission occurs (p 1202). In SS Pharmaceutical
Co v Quantas Airways [1991] 1 Lloyd's Rep 288, Kirby P, sitting in the Court of Appeal of New South Wales, stated at 301 that knowledge "must be shown to exist at the moment at which the act or omission complained of occurs".
Those statements do not in my judgment defeat Sir Sydney's submission. In Goldman, the pilot had to make a decision as to whether to illuminate the seat belt instruction light. The Court of Appeal, reversing the trial judge, held that on the information available to him "it seems in my judgment impossible to say that Captain Swang's decision not to illuminate the seat belt fell within the test provided by Article 25" (per Purchas LJ at 1202). The Court had considered the information available to the pilot when he took his decision and the expert evidence as to what he should have done on the basis of that information. When using the expression "actual knowledge", Purchas LJ was referring to the actual information available to the pilot and was not in my view seeking to make a distinction between what was in the front and what was in the back of the pilot's mind or suggesting that a pilot's knowledge by way of training and experience should be ignored in evaluating the decision he took. The same applies to the decision to leave cargo in the open in SS Pharmaceutical. Plainly, information, for example as to approaching weather conditions, which the actor does not have at the moment he makes his decision is excluded from his "knowledge".
The Court must not impute to the actor knowledge he does not have but neither is it entitled to ignore his fund of knowledge and experience in assessing his knowledge at the material time. It is the submission that one can isolate what a person is thinking at a particular moment from his fund of knowledge that, with respect, I find artificial. It unjustifiably devalues the processes of the mind. A pilot does not escape liability merely because, by reason of, for example, drink or tiredness, he forgets for a moment his training and the general knowledge his experience of flying brings him. Obscure information, on the other
hand, once received but readily forgotten, would not usually come within the appropriate fund of knowledge. As a simple land-based example, I mention a driver on a busy road. He would not escape liability on the Article 25 use of the word "knowledge" by claiming that he had forgotten for a moment that vehicles drive on the left hand side of the road (in the United Kingdom) or that it is a red traffic light which means stop. There are no doubt equivalent basic rules in the air. The pilot in Goldman would not have escaped liability on the ground that he had forgotten at the relevant moment that seat belts provide protection against damage.
On the same basis, I disagree with the judge's ruling that "turning a blind eye" is not consistent with knowledge as the word is used in Article 25 though I agree that is the outcome of the reasoning of Auld LJ. The Article does not in my view permit the actor to say that his knowledge is no longer his knowledge because he has made a conscious decision to put it out of his mind.
I agree with Auld LJ that inferences may be drawn in determining the presence of knowledge for the purposes of Article 25. Inferring knowledge is different from imputing it and difficulties of categorisation which may sometimes occur do not destroy the reality of the difference. Courts are familiar with applying a practical approach to the drawing of inferences. Knowledge of basic "rules of the road" will more readily be inferred than will knowledge of more esoteric information. Difficulties at the borderline do not justify a definition of knowledge which involves treating the mind as if incapable of retaining information. In some situations, I would be prepared to draw inferences that an experienced pilot in fact had the general knowledge, for the purposes of Article 25, that an experienced pilot would be expected to have.
I have thought the difference of view as to the meaning of knowledge as used in Article 25 sufficiently important in this context to address it but it does not affect my conclusion as to the outcome of the appeal. I agree with Auld LJ that, whichever test of knowledge is applied, it is not possible on the pleadings and other material before the Court, taking the most favourable view from the claimants' point of view, to draw an inference that the pilot or his employers knew that damage would probably result from their conduct.
I agree that the appeal should be dismissed.

DYSON J:
I agree for the reasons given by Auld LJ that this appeal should be dismissed. In view of the difference of opinion between Auld LJ and Pill LJ on the knowledge issue, I shall add a few words of my own limited to that issue. I agree with Pill LJ that the statement by Purchas LJ in Goldman v Thai Airways [1983] 1 WLR 1186, 1202B-C, and the statements in SS Pharmaceutical Co Ltd v Qantas Airways Ltd [1991] 1 LLR 288 do not defeat the submission of Sir Sydney Kentridge QC as to the meaning of "knowledge" in Article 25. That is because those statements were not necessary for the decisions. They are nevertheless authoritative, and, as Sir Sydney conceded, adverse to his argument. He submits quite simply that they are wrong.
Knowledge is an elusive concept, which has always troubled philosophers and judges. It has different meanings in different contexts. An illustration of its diversity is to be found in the five-fold classification of knowledge relevant for the purposes of constructive trusteeship made by Peter Gibson J in Baden v Societe Generale [1993] 1 WLR 509. The fact that it was later said in Royal Brunei Airlines v Tan [1995] 1 AC 378 that this classification was best forgotten does not detract from the point.
For present purposes, it is sufficient to concentrate on three different types of knowledge. The first, which I shall call "actual conscious knowledge", is knowledge of the type that Purchas LJ described in Goldman at page 1202B-C, viz "actual knowledge in the mind of the pilot at the moment at which the omission occurs". The second, which I shall call "background knowledge", is knowledge which would be present to the mind of a person if he or she thought about it. The third, which I shall call "imputed knowledge", is knowledge which a person ought to have but does not in fact have.
It is clearly established that "knowledge" in Article 25 is not imputed knowledge. It is not sufficient to show that, by reason of his training and experience, the pilot ought to have known that damage would probably result from his act or omission. The test is subjective: see, for example, Goldman. Actual knowledge is required, and the question that has been debated before us is whether that is limited to actual conscious knowledge, or whether it also includes background knowledge. Sir Sydney contends that "knowledge" within the meaning of Article 25 includes both actual conscious knowledge and background knowledge.
The difficulty that I have with Sir Sydney's argument is that I can see no basis for concluding that those who agreed the text of Article 25 intended to include background knowledge, but exclude other knowledge (not being imputed knowledge) which is not present to the mind of the actor at the time of the act or omission. The concept of knowledge is complex and subtle. We use the word "knowledge" in diverse ways. We are all familiar with the situation that arises where we cannot recall something that we believe that we know. If asked the question "what is the capital of Peru?" we might say " I do know the answer, but I just cannot recall it". Or we might say " I cannot remember it at the moment, but it will come back to me". That is not imputed knowledge, although it is not a misuse of language for someone in such a situation to say "I ought to know what the capital of Peru is, but I cannot bring it to mind". Why should the state of mind of a person who knows a fact, but does not apply his mind to it because he has temporarily forgotten it, be treated differently from that of a person who does not apply his mind to a fact because he is distracted or tired? In each case, the fact is not in the forefront of the actor's mind at the relevant time. And yet, I do not understand that Sir Sydney would include within Article 25 knowledge that is not present to the mind of a person as a result of a loss of memory. In my view, there is no obvious reason why a distinction should be drawn between the two cases.
Suppose that the difficulty that I have identified were overcome by holding that knowledge which has been temporarily forgotten is included in the definition of knowledge within the meaning of Article 25. That would merely give rise to a different problem. In that event, the crucial distinction between actual and imputed knowledge would become blurred. Take an example from the pleadings in this case. It is alleged that the crash was caused inter alia by the failure of the pilot to switch on the aircraft standby artificial horizon, and that if he had addressed his mind to that omission, he would have appreciated, by reason of his knowledge and skill as an experienced pilot, that death or serious injury would probably result. It is said that he had the requisite background knowledge, but, because he was suffering from fatigue, he failed to apply his mind to the probable effect of a failure to switch on the artificial horizon. On one view, the correct analysis of these facts, if proved, would be that the pilot had temporarily forgotten that the probable consequence of failing to switch on the standby artificial horizon would be death or serious injury. In other words, he had background knowledge of that fact. But an equally tenable analysis would be that the pilot ought to have known (ie had imputed knowledge) that this would be the probable consequence. On the first analysis, there would be knowledge within the meaning of Article 25, but not on the second analysis.
Does the language of Article 25 itself give any clue that indicates whether background knowledge is included or not. Knowledge must add something to "recklessly". It is a requirement of Article 25 that the act or omission should be done recklessly and with knowledge that damage would probably result. I accept the submission of Sir Sydney that background knowledge cannot be rejected on the grounds that it adds nothing to "recklessly". It is sufficient for recklessness that a person should act regardless of the possible consequences of his acts. What Article 25 requires is that there should be knowledge of the probable consequences. Although I accept this submission of Sir Sydney, I do not consider that an examination of the precise language of Article 25 provides any real clues to the meaning of "knowledge". As was said in SS Pharmaceutical, the concept of recklessness with knowledge that damage would probably result is different from the concept of intent to cause damage. The question is: how different?
I do not believe that those who drafted Article 25 intended that anything less than actual conscious knowledge would suffice. That is a mental state that is clear and simple to
understand. Once one moves away from actual conscious knowledge, uncertainty is
introduced, and difficulties of classification will arise. To use a metaphor that is in common currency, a fact may be just below the surface of a person's mind, or it may be deeply or not so deeply buried in its recesses. It may be just below the surface of his mind because he is distracted, or tired, or because he has forgotten it temporarily. It may be buried more deeply because he learnt it some time ago, and cannot bring it to mind without some outside assisitance.There will be circumstances in which a person has so completely forgotten a fact that it cannot sensibly be said that he has actual knowledge of it at all, although he may well have imputed knowledge of it. If knowledge means more than actual conscious knowledge, then there will be argument as to where on the gradient between actual conscious knowledge and imputed knowledge a particular case comes. I do not believe that it was intended that it would be necessary to embark on subtle and difficult questions of this kind in order to determine whether a claimant could rely on Article 25. There is nothing in the language of Article 25 or the travaux preparatoires to indicate that it was intended to include some, and not all, categories of knowledge not present to the mind at the time of the act or omission. Why should knowledge that has been temporarily forgotten be excluded? If a person fails to apply his mind to a fact because he has temporarily forgotten it, he has no more and no less actual knowledge of that fact at the time of his act or omission than a person who fails to apply his mind to it because he has been temporarily distracted.
I should mention the argument advanced by Sir Sydney which has found favour with Pill LJ. He illustrates it by giving the example of a vehicle driver who is familiar with a route, and knows that there are traffic lights at a particular junction. The driver is tired and distracted and drives through the lights when they show red. To say that he was unaware of the red light because he was distracted at the time is, it is submitted, to put a premium on inattention, and, applying the analogy to a pilot, is not warranted by Article 25. It seems to me that on such facts there would be no difficulty in finding that there was recklessness. But the question remains whether, in such an example, at the time when he drives through the red light, the driver knows that that is what he is doing.
In my view, this illustration does not assist in resolving the problem of what is meant by "knowledge" in Article 25. It merely shows that it is possible to construct examples of "hard" cases. But it is clear from the various authorities to which Auld LJ has drawn attention that it
was intended that Article 25 should be of very restricted application. Sir Sydney submits that, if it applies only to actual conscious knowledge, it will be almost impossible for a claimant ever to invoke Article 25. In practice, it will be necessary to prove that the pilot was suicidal. I do not agree. First, there are the cargo and baggage handling cases. It is worthy of note that in SS Pharmaceutical (which was such a case) the Article 25 test was satisfied. Secondly, the actual conscious knowledge test may also be satisfied in cases of reckless manoeuvres in the sky by non-suicidal pilots, although I accept that such cases will be extremely rare. This is not surprising in view of the admittedly stringent requirements of Article 25. In any event, as is clear from the judgment of Pill LJ, it will also be very difficult for a claimant to rely on Article 25 even if he or she has to discharge the background knowledge test.
For these reasons, as well as those given by Auld LJ, I would dismiss this appeal.