Scott v Copenhagen Reinsurance Company (UK) Ltd.

On 2 August 1990 Iraq invaded Kuwait. By mid-morning Iraq’s forces were in complete control of Kuwait International Airport and all the aircraft there on the ground. Kuwait’s civil air fleet was owned by Kuwait Airways Corporation (KAC). It had 15 aircraft at the airport, and these and the fleet’s spares were a prime target of Iraq’s premeditated policy of plunder. Also at the airport was a British Airways (BA) Boeing 747 (and a few other foreign aircraft). The presence of the BA Boeing was unscheduled: it was in transit, and its late arrival meant that its departure failed by a mere half-hour or so to forestall the 5 am closure of the airport which the invasion occasioned. The KAC aircraft and spares were flown off to Iraq as soon as the logistics of that operation allowed and in due course incorporated into the Iraqi fleet. The BA aircraft however remained at Kuwait. It was still there on the outbreak of war between Iraq and the coalition forces (“Operation Desert Storm”) on 16 January 1991. It was destroyed by allied bombing by, on or around 27 February 1991. This appeal is concerned with the excess of loss reinsurance of the claimant/appellant syndicates’ whole account and raises the question whether the loss of the BA aircraft is to be aggregated with the losses of the KAC aircraft and spares.

Langley J held [2002] EWHC 1348 (Comm) that the KAC aircraft and spares had all been lost on 2 August 1990 and that these were all losses “arising from one event”, namely Iraq’s invasion of Kuwait and capture of the airport. There is no appeal from that conclusion. However, the judge also held that the loss of the BA aircraft did not arise out of the same event. He did not define that event save possibly by reference to the cause of its loss, which he said was “the actual destruction of the aircraft or the war or perhaps the inevitability of war” (para 73).

The syndicates no longer submit, as they did before the judge, that the BA aircraft became an actual total loss immediately on 2 August 1990 in exactly the same way as the KAC aircraft and spares were then lost to Iraq’s policy of plunder. However, their appeal is based on their so-called alternative case, which is that even though a total loss of the BA aircraft could not be established until after a reasonable “wait and see” period, once proved it related back to the original event of the invasion and capture of the airport and was properly to be adjudged as having also arisen from the same one event as the KAC fleet losses. In this connection the written skeleton on behalf of the syndicates describes this as an example of –

“those unusual cases, such as this, in which a single event is claimed to have produced a series of immediate total losses at the same time as producing other losses, not immediately classed as total losses, but initially classed as deprivations of possession, requiring review as to their status at the end of a reasonable “wait and see” period.”

This distinction between an immediate total loss and one which is established only after a period of “wait and see” derives from an analysis of the relevant law which is to a large extent common ground to the parties and to which I shall return below.

The syndicates, on whose behalf the claimant, here the appellant, Mr Murray Scott sues as a named representative, obtained whole account excess of loss reinsurance from a number of reinsurers who are in turn represented in these proceedings by The Copenhagen Reinsurance Company (UK) Limited, defendant and on this appeal respondent (“Copenhagen”). The contract of reinsurance is contained in a slip policy No 90040114. It is sufficient to record the following terms:

“PERIOD: Losses occurring during 12 months at 1^st April, 1990, Local Standard Time at the place where the loss occurs.

LIMITS: £15,500,000 or US or CAN$ 31,000,000 each and every loss, as defined, EXCESS of £15,500,000 or US or CAN$ 31,000,000 each and every loss, as defined, WHICH IN TURN IS EXCESS OF the Combined Limit(s) and Retention(s) of the Reassured’s Specific Excess of Loss Protections, as per Schedule attached.

WORDING: Full wording as Joint Excess Loss Committee Clauses 1.1.90 with additional clauses…

War included.”

The Joint Excess Loss Committee Clauses 1.1.90 (“JELC Clauses”) provided as follows:

“EVENT CLAUSE

“Loss” under this contract means loss, damage, liability or expense arising from any one event or as described in section J of the schedule.”

Section J read:

“J LOSS DESCRIPTION

Each and every loss or series of losses arising from one event.”

Thus the critical provisions for present purposes were the period of the reinsurance and the aggregation provisions which defined loss as all those losses “arising from one event”. As the judge said, it is the application of those words to the facts which is in issue in these proceedings.

The syndicates had in turn entered into various inward contracts of excess of loss reinsurance by which they covered (either directly or indirectly) the insurers/reinsurers of KAC’s primary insurance and reinsurance as well as BA’s insurance contracts. The details of those contracts can be found in the judgment below.

How aggregation will affect any particular reinsured is very difficult to predict. As Copenhagen’s opening statement below said –

“In fact, of course, many individual companies will have interests on both sides of the fence, as retrocedents and also as retrocessionaires under different XL contracts, and where their best interests lie is not always clear.”

This is so even after the event. How an aggregation clause will operate in advance is even more unpredictable. As the judge remarked, aggregation is a concept which may assist a reinsured to pierce the limits of an XL layer but may also operate so as to exhaust a layer and so damage his interests. An illuminating exposition of the underwriting considerations can be found in Lord Mustill’s speech in Axa Reinsurance (UK) Plc v. Field [1996] 1 WLR 1026 at 1034.

There were no submissions as to how, if at all, the function of an aggregation clause might assist this court to resolve the issue before it. I suppose, however, that its function is to police the imposition of a limit by treating a plurality of linked losses as if they were one loss. For this purpose the losses have to be identified by a unifying concept: in this case “one event”, or strictly speaking “arising from one event”.

A résumé of the facts has been made easier both by the parties’ Schedule of Agreed Facts (which is annexed to the judgment below) and by Langley J’s own succinct findings, which I would gratefully adopt and make direct use of here. There is in general no dispute as to those findings (save in one aspect indicated below). In the light of the limited nature of this appeal, which as explained above does not relate to the KAC aircraft and spares nor to the syndicates’ primary case that the BA aircraft was as much an object of Iraq’s policy of plunder as the KAC fleet, I can restrict the judge’s findings to the minimum necessary to determine the present appeal.

Under the heading of “The facts” the judge said this:

“30…I accept Captain Clark’s evidence [he was the BA aircraft’s pilot] that the Iraqis took immediate control and possession of the BA aircraft when they occupied the airport on August 2 and if it was to go anywhere it would have required Iraqi permission to do so.”

Under “Politics” the judge made the following findings:

“34…On August 6 UN Security Resolution 661 collectively introduced sanctions against Iraq and Kuwait effective on August 9…On August 6 it appears that Iraq began to formulate the policy towards foreign nationals trapped in Kuwait which became the human shield policy…

“35.

It was also on August 6 that King Fahd agreed to the deployment of American forces in Saudi Arabia. President Bush said that the role of the American troops was defensive…

“38…On 30 August Iraq announced that all foreign women and children would be released and a few days later they were. Throughout September the coalition military presence was increased but further peace initiatives were also proposed. On 21 September the Royal Air Maroc Boeing 727 was allowed to leave flying out of Kuwait via Baghdad. On 18 October the TMA and MEA aircraft were also allowed to leave again flying via Baghdad.

“39.

A flight plan for the BA aircraft was filed in Baghdad on 16 October and re-filed on 23 October. On 2 November BA was requested to pursue the matter through diplomatic channels. Again it is difficult to conclude that this was more than game-playing by Iraq.

“40…On 29 November the UN Security Council passed Resolution 678 which authorised the use of all necessary means against Iraq unless it withdrew from Kuwait by 15 January 1991…

“41.

On 6 December Iraq announced the release of all hostages. The BA aircraft was not used to fly them out. The aircraft would not have been serviceable without a few days maintenance but in any event it seems to have been Iraqi policy to use Iraqi Airways aircraft for the purpose. By 12 December all BA staff and passengers had returned home.

“42.

In December there were exchanges in which it was suggested that the Iraqis might release the BA aircraft in exchange for components and engines belonging to Iraqi Airways held at Heathrow. That was never a practical proposition.

“43.

On 12 January 1991 the US Congress authorised the use of force if Iraq did not withdraw from Kuwait by 15 January. The bombing phase of Operation Desert Storm began on 16 January. On and after 22 January Mosul airfield in Iraq was attacked by coalition forces resulting in the destruction of two KAC A300 airbuses and the two KAC Boeing 767s. On and after 4 February Iraq moved some of the plundered Kuwait aircraft to Iran for “safe keeping”. The ground war began on 23 February. It was virtually over by 27 February. The BA 747 aircraft was badly damaged, apparently by coalition fire, on some date between 13 and 26 February at Kuwait International Airport. On 3 March Iraq unconditionally accepted all terms required of it.”

Under his next heading of “The insurance position” the judge found:

“44.

On 2 August 1990 Mr Lerwill [BA’s general manager of risk management] notified BA’s insurance brokers that the BA aircraft was at Kuwait…BA had nonetheless decided to defer making a claim because the aircraft remained intact…

“45.

Formal notice of “restraint of the aircraft” was given by BA on 15 August in a letter subsequently scratched by underwriters indicating that they awaited further developments…

“47…On 24 October the brokers reported to Mr Lerwill and Mr Mayhew [BA’s claims manager] following a conversation with the leading underwriter…The letter also stated that if the circumstances in Iraq and Kuwait remained unchanged in the next few months the leading underwriter would expect BA to be able to state “that they had exhausted all means at their disposal to effect recovery of the aircraft” which would be of “great use” in getting the agreement of the following market to a valid claim. Much the same theme was reported by the brokers in a letter dated 13 November. There was no claim for a loss at this time and the inference is that it was not then being asserted that a loss had occurred.

“48.

On 13 February 1991 Mr Mayhew wrote to the brokers stating that “As it now seems to be unlikely that the…aircraft…will be secured from Kuwait I feel that it is perhaps time, six months having elapsed, to seek from underwriters their agreement to accept this letter as a claim…for the insured value of $28m”. The letter also stated that “with the probability of a ground war commencing, the chances of (the aircraft) remaining without damage seems slim”.

“49.

The BA aircraft was in fact destroyed while the claim was in the process of being agreed by the following market. The claim for $28m was paid in full on 4 March 1991.”

What is perhaps the only factual dispute on appeal is as to the precise relative timings of the agreement of BA’s claim on its primary insurers and the destruction of the aircraft. The claim letter of 13 February 1991 concluded with the sentence: “Your confirmation that Underwriters will accept this claim would be appreciated”. The letter is then scratched “Settle” and dated 14 February, probably by the leading underwriter. The scratches of agreement of the following market are each dated over the succeeding days, and the latest scratch is dated 26 February. The schedule of agreed facts contained (at para 29) this item:

“…the BA aircraft…was destroyed, on the ground, at Kuwait Airport on or around 27^th February 1991 during the course of the bombing of Kuwait Airport by allied forces as part of “Operation Desert Storm” or during military activity in the airport area.”

Mr Stewart Boyd QC on behalf of the syndicates submitted that 27 February was the probable date of destruction of the aircraft (by reference to this agreed fact). However, the judge found (para 43, cited above) that the aircraft had been badly damaged “between 13 and 26 February” and that it was destroyed “while the claim was in the process of being agreed by the following market” (para 49). The significance of this goes to whether the aircraft was destroyed before or only after the claim was agreed. Mr Boyd submits that its destruction took place only after the claim had been agreed and that therefore its destruction was irrelevant to the issue of “arising from one event”. He therefore criticises the judge’s conclusion (at para 73) that the cause of the eventual loss of the aircraft was “the actual destruction of the aircraft or the war or the inevitability of war”. On behalf of Copenhagen, however, Mr Dominic Kendrick QC supports the judge’s findings and conclusion.

The judge then turned to the evidence which he heard from two distinguished experts, Professor Efraim Karsh and Professor Lawrence Freedman, co-authors of a leading work (“The Gulf Conflict”, faber and faber, 1992) but on this occasion on opposite sides, the former giving his evidence at the request of the syndicates and the latter at the request of Copenhagen. Professor Karsh is Professor of Mediterranean Studies at King’s College London, and Professor Freedman is Professor of War Studies, also at King’s College. The judge pointed out that there was a substantial measure of agreement between them, but where they differed, in particular in respect of the syndicates’ primary case that the BA aircraft had been lost together with the KAC fleet from the beginning, he preferred the views of Professor Freedman. He quoted these as follows:

“56…“the BA aircraft was never part of the Iraqi policy of plunder. The position of the aircraft was the result first of happenstance and then a by-product of decisions to take the passengers and crew hostage…The BA aircraft was not at risk until coalition military operations began on 16 January 1991…These operations were by no means inevitable at the point of the Iraqi invasion and had only become more likely than not by the time of UN Resolution 678 of 29 November 1990 and the likelihood still fluctuated after that point”.”

The judge then added his own comments:

“57.

I find this assessment more realistic and more in accord with the evidence of actual events. I also think that if there had been no war there would have been a realistic chance of the BA aircraft being restored to BA. To my mind, Professor Freedman’s analysis that Iraq had no purposive policy towards the BA aircraft on 2 August and indeed that it remains unclear that it ever did develop one in any concrete way is probably right. The aircraft was of secondary, if any, importance in the context of events. Certainly I accept Professor Freedman’s view that no policy of retribution or acquisition in respect of the BA aircraft has been established. His opinion that any seizure of such an asset by Iraq would have been “extremely provocative” is I think likely to be accurate. There was nothing to be said for risking turning an issue of secondary importance into a serious provocation. There is no evidence of Iraq seizing or claiming to own any foreign non-Kuwaiti assets found in Kuwait.”

Under the heading of “Summary” the judge then concluded (in para 58) that the evidence established that –

“i)

The assets of KAC, both aircraft and aircraft spares, at Kuwait International Airport, were a specific target of the Iraqi invasion, and Iraq intended both to capture them and treat them as acquired from the moment the airport was captured. In fact it did so…

iii)

Once the airport had been captured all KAC’s assets there were effectively lost to KAC with no real prospect of recovery.

iv)

The BA aircraft was in no sense a target of the invasion nor did it become one later in any relevant acquisitive or retributive sense. It was in common parlance “stranded” when the invasion occurred. I think that had the question been asked on 2 August “is the aircraft lost?” the answer would have been “I don’t know. Wait and see.”

v)

At no time did Iraq form any significant policy intention towards the BA aircraft. It remained at Kuwait throughout the following months including the first weeks of Operation Desert Storm.

vi)

The factual situation is reflected in the Insurance documents. The KAC claim was made almost at once. The BA claim was formally pursued only after Operation Desert Storm had begun.”

The judge first considered two areas of jurisprudence, one relating to the concept of actual total loss, the other to aggregation clauses. As for the former, he reminded himself that the doctrine of constructive total loss does not exist outside marine insurance, and that aviation insurance is not marine insurance: Moore v. Evans [1918] AC 185 at 195, Kuwait Airways Corporation v. Kuwait Insurance Company [1999] 1 Lloyd’s Rep 803 at 809. He was prepared to assume therefore that in order to prove an actual total loss the syndicates had to show that they had been “irretrievably deprived” of the aircraft (adopting the test for actual total loss in section 57 of the Marine Insurance Act 1906); that deprivation of possession could be irretrievable without any further elapse of time (as had been judged to have occurred on the same facts relating to the KAC fleet and spares in KAC v. KIC [1996] 1 Lloyd’s Rep 664, an issue only considered at first instance and not subject to appeal,and as Langley J also ruled had occurred in the present case); and that in the absence of an immediate total loss the test required the actual elapse of a reasonable time for recovery of the property with at least uncertainty of recovery remaining at the end of that time (at para 67).

The second area of jurisprudence took him to Axa Reinsurance v. Field [1996] 1 WLR 1026 from which he cited Lord Mustill’s dictum at 1035 that in ordinary speech an event is “something which happens at a particular time, at a particular place, in a particular way”; and especially to KAC v. KIC [1996] 1 Lloyd’s Rep 664 at 682/689 and to the award of Mr Michael Kerr QC (later Kerr LJ) in the Dawson’s Field Arbitration, itself quoted at length in KAC v. KIC. Those last two cases illustrate the usefulness of testing the question of one event (in KAC v. KIC, “any one occurrence”) by reference to factors, so-called “unities”, such as cause, locality, time and the intentions of any human agents. As Mr Kerr said in Dawson’s Field:

“Whether or not something which produces a plurality of loss or damage can properly be described as one occurrence therefore depends on the position and viewpoint of the observer and involves the question of degree of unity in relation to cause, locality, time, and, if initiated by human action, the circumstances and purposes of the persons responsible.”

Immediately before turning to this jurisprudence the judge had remarked (at para 60) that “In a real sense I think the question is one of impression which does not bear too much analysis”. So it was that at the end of his judgment Langley J moved swiftly to his conclusions:

“71.

If the question is simply asked whether or not the loss of the Kuwaiti aircraft and spares arose from a single event, in my judgment the answer is overwhelmingly “Yes”. The informed observer would I think conclude that they were lost on Iraq’s invasion and capture of the airport, motivated as it was to acquire those very assets…

“72.

Equally if the same question is asked on the analytical basis proposed by Mr Kerr, the “unities” are I think present. There was unity of intent on the part of the Iraqis both to capture the aircraft and spares and to deprive KAC of them permanently. There was unity of time: the objective was achieved when the airport was captured. There was unity of cause: the invasion…

“73.

But if the same questions are then asked in relation to the BA aircraft as well, I think the answers are quite different. The aircraft was not in my judgment lost as a result of the invasion and capture of the airport. The informed observer would obviously have recognised real difficulties for BA in accessing and recovering its aircraft but would not I think have concluded it was an actual loss to BA. The position was analogous to ransom and “wait and see” albeit it lacked the feature of an offer of return if a demand was met and also of any expressed intention to exercise permanent dominion over the aircraft. But I think those factors are balanced. The more analytical approach I think again produces the same answer. The evidence belies an intent to capture or plunder and does not support any formed intention on the part of Iraq at any time permanently to deprive BA of the aircraft. Unity of time is lacking because of the presence of “wait and see”. Unity of cause is lacking as I think in commonsense terms the cause of the eventual loss of the aircraft was the actual destruction of the aircraft or the war or perhaps the inevitability of war. But even the inevitability of war cannot sensibly in my judgment be said to have existed until some months after the invasion. The need for and the passage of time before the loss is established and the intervening events in my judgment make it more and more difficult to characterise the loss as one arising from the invasion and on the facts as I have found them I do not think the Claimant has really come near to making out such a case.”

It follows that the judge approached the issue of whether the loss of the BA aircraft arose from the same event as the losses of the KAC aircraft and spares first from the point of view of what he had earlier described as “one of impression”, and secondly by applying as a cross-check the more analytical tools of the “unities”. In his view each approach supported the other.

On behalf of the syndicates Mr Boyd relied on what were essentially four main submissions. First, he concentrated on the contractual phrase arising from. This is plainly expressive of a causal relationship, but, said Mr Boyd, only required a weak or loose causal connection. In this respect he pointed to Caudle v. Sharp [1995] LRLR 433, where the phrase was the almost identical phrase “arising out of one event”. At 438/9 Evans LJ considered the nature of the causal connection and held that it should not be restricted to that of proximate cause: however, other than saying that within the insurance context it was implicit in the concept of aggregation based on a single event that the causal connection was neither that of proximate cause at the one extreme nor was entirely unrestricted by any doctrine of remoteness at the other extreme, no express definition was there given to the causal relationship involved. On this basis, Mr Boyd submitted that, even though the BA aircraft’s loss was only established after a period of “wait and see”, there was no reason to think that it did not arise from the invasion and capture of the airport as much as the KAC losses had. Therefore there was, contrary to the judge’s view, unity of cause. In any event, the test of the “unities” was only a guideline, and the contractual test was “arising from” which was well met here. Given the weak causal connection required, there was nothing that happened subsequently to the invasion and capture of the airport which could be said to be of sufficient causal relevance to displace the original event which had caused the deprivation of possession and the commencement of the “wait and see” period.

Mr Boyd’s second main submission was to concentrate on the judge’s conclusion referred to above that the common sense cause of the eventual loss of the aircraft was “the actual destruction of the aircraft or the war or perhaps the inevitability of war”. Mr Boyd criticised reference to the destruction of the aircraft on the basis that that occurred only after (a) the end of the period of six months which Copenhagen itself relied on as a reasonable “wait and see” period (and which concluded on 2 February 1991), (b) the loss had been claimed and (c) then agreed, and was therefore entirely irrelevant (see paras 17/18 above). He next criticised reference to the inevitability of war on the basis that it cannot be described as an event at all. Unlike an event, one cannot speak of the “inevitability of war” as having “occurred” or “taken place”. It cannot be placed within the function of time. It is a “state of affairs”, but not an event. However, Mr Boyd was not able to level a similar criticism at the reference to war, by which I understand the judge to refer to the war between Iraq and the coalition which commenced on 16 January 1991. On behalf of Copenhagen, Mr Kendrick took Mr Boyd to task for failing to deal with that obvious causal candidate. Mr Boyd’s riposte was in effect to say that in the circumstances the outbreak of war, even if not inevitable from the first time of the invasion of Kuwait, was not something which could be said to break the chain of causation: nor did the judge refer to it in those terms.

The third main strand in Mr Boyd’s argument referred to the concept and timing of actual loss. He submitted that where the deprivation of possession is such as to initiate a reasonable “wait and see” period, then, as and when the passing of such a period and the taking of all reasonable steps in the interim to recover the property conclude in the realisation and proof that actual loss has been established, the time of such a loss relates back to the original loss of possession. He had no authority to cite in support of that proposition, but he argued that it must stand to reason, otherwise an insured deprived of possession may find himself falling between two stools: he could not bring himself within his policy of insurance extant at the time of loss of possession, before its expiry, and he could not, save for a prohibitive premium, insure his property into a second year. He cited as a practical example in support of his submission the fact that, to forestall argument about the date of loss in the case of the taking of motorcars, the insurance ombudsman has procured agreement from the insurance industry that loss occurs on the date of taking. It follows that there was in fact unity of time of loss, contrary to the judge’s conclusion that “Unity of time is lacking because of the presence of “wait and see””. In any event, he submits that unity of time is not indispensable, and that the judge himself recognised that “losses occurring at different times may yet arise from one event” (at para 66).

Mr Boyd’s fourth main submission was that the test of the “unities” is inappropriate in a case, which he described as unusual, where one event causes the deprivation of possession of various items of property, some of which become immediate total losses, whereas other items only become actual total losses after a further intermediate period. He pointed out that in Dawson’s Field (where four aircraft were hijacked, one was flown to Cairo and blown up there, and the other three were flown to Dawson’s Field in Jordan and there blown up together a few days later, and Mr Kerr held that the losses of the three aircraft at Dawson’s Field could be aggregated, but it was not contended that the loss of the aircraft blown up at Cairo could also be aggregated with the losses of the other three), all four aircraft were subject to a period of “wait and see”: for the context was one of ransom, rather than an intention permanently to deprive. In a development of this submission, as I understood him, Mr Boyd further argued that Mr Kerr’s test of the “unities” was expressly applied not to the phrase “arising out of one event”, but to the different phrase “one occurrence” (at 9). In that arbitration both phrases appeared: but it may be that, whereas the latter phrase was tested by reference to the “unities”, the former phrase had a more traditional causal approach taken to it based expressly upon the words “arising out of” (at 11). So, coming back full circle to his first submission, Mr Boyd argued that the application of the “unities” test was inappropriate and over-analytical and tended to alter and obscure the true question, which was whether the subsequently confirmed loss of the BA aircraft arose out of the same event as the immediate loss of the KAC fleet. The judge had misused the test, had applied too strong a causal connection, and had come to the wrong conclusion.

On behalf of Copenhagen, Mr Kendrick submitted that the judge had come to the right result for the right reasons. There was no dispute as to the primary facts. The issue was essentially one of impression and common sense, and as such was primarily for the trial judge. Langley J had dealt with the issue in just such a way, and only then had gone on to test his initial impression with the help of the “unities”. In doing so, he was using that test only as a guideline, not as a substitute definition of the contractual words. His conclusions were impeccable. Thus, there was no unity of intent: whereas Iraq had intended to capture the KAC fleet, the BA aircraft was only present at the airport as a matter of pure chance and was never targeted. Its fate therefore, unlike the fate of the KAC fleet, depended on subsequent events. There was no unity of time: the BA aircraft was not lost on 2 August 1990, since the elapse of a period of “wait and see” during which BA had to take all prudent steps to recover its aircraft was an essential element in any claim. There was no unity of cause: the aircraft was not lost as a result of the invasion but because of subsequent events, notably the outbreak of war on 16 January 1991, which left it exposed at a strategically important airfield and duly led to its destruction. The outbreak of war was not probable at the time of the invasion. The claim of loss only came forward in February 1991.

As for Mr Boyd’s submissions on appeal, Mr Kendrick responded as follows. First, in relation to the phrase “arising from”, he submitted that, even though it did not look to proximate cause, it required a significant and strong causal connection. It was in this way that Mr Kerr had applied the facts of Dawson’s Field to that phrase’s close cousin (“arising out of”), for he did not go back to the original hijackings of the four aircraft but found his unifying event only in the planned destruction of the three parked in Jordan. Secondly, in relation to the criticism levelled at the judge’s findings of causation, he defended the references both to destruction of the aircraft, as having occurred sometime between 13 and 26 February 1991 while the claim was going around the market, and to war. He was less protective about the inevitability of war. Thirdly, as to the concept and timing of loss, he submitted that on the authorities illuminating the finding of actual loss by deprivation, whether in the non-marine or marine setting, the passage of time and proof that all reasonable steps had been taken were essential elements of the insured’s cause of action: they went not merely to confirm what had already happened, but to create loss. Fourthly, as to the “unities”, the judge had dealt with these appropriately and realistically (see above).

In one sense this appeal is about a trial judge’s assessment of facts which, save in relation to the timing of the destruction of the aircraft, are not in dispute, and about his application of those facts to a brief contractual test (“arising from one event”). In another sense, certain points of principle have emerged which divide the parties. I would not be surprised if those points have received more analysis on appeal than below, seeing that the syndicates’ primary case at trial had been to show that the BA aircraft had been lost “then and there” on 2 August 1990 together with the KAC aircraft and all as part and parcel of the initial events of that day. The points of principle are: whether the link referred to by the phrase “arising from” requires a strong or weak causal connection; whether the date of loss in a “wait and see” situation goes back to the original loss of possession or forward to the outcome of the period for testing the possibilities of recovery; and whether the “unities” test is appropriate to an “arising from” clause, as distinct from a “one event/occurrence” clause.

Although, like the judge, I am diffident about the direct help which previous authority can lend to a unique factual situation, it is the insight of the common law that wisdom can be drawn from previous examinations of similar problems. I will therefore consider the material which has been cited to the court, first in relation to actual total loss, and secondly in relation to aggregation clauses.

The Marine Insurance Act 1906 defines actual total loss in section 57(1):

“Where the subject-matter insured is destroyed, or so damaged as to cease to be a thing of the kind insured, or where the assured is irretrievably deprived thereof, there is an actual total loss” (emphasis added).

Section 60 defines constructive total loss, inter alia (section 60(2)(i)) as –

“Where the assured is deprived of the possession of his ship or goods by a peril insured against, and (a) it is unlikely that he can recover the ship or goods…” (emphasis added).

It is a condition precedent of a claim for a constructive total loss that the assured gives notice of abandonment of the subject-matter insured to the insurer (sections 61/62 of the Act).

It was said in Polurrian Steamship Co Ltd v. Young [1915] 1 KB 922 at 937 that the words “within a reasonable time” must be understood to be implicit in section 60(2)(i)(a). In the same case this court said (at 936/7) in the posthumous judgment of Kennedy LJ that whereas the common law previous to the Act had been framed in terms of whether it was uncertain whether recovery could be made within a reasonable time, the Act had now changed the test into one of unlikeliness of recovery (ie on the balance of probabilities). See also Royal Boskalis Westminster NV v. Mountain [1997] LRLR 523 at 534.

However, the doctrine of constructive total loss does not apply outside marine insurance: Moore v. Evans [1918] AC 185. The House of Lords there declared the decision of the court of appeal absolutely right, and the judgment of Bankes LJ was specifically approved. It has become the leading judgment on the subject of actual total loss in the non-marine setting: [1917] 1 KB 548. The case concerned the consignment of pearls on approval in June and July 1914 to jewellers in Frankfort and Brussels, just before the outbreak of war and the subsequent occupation of Brussels by the Germans. A claim for loss was made on 22 October 1914. The policy year expired on 7 January 1915. The writ was issued on 16 February 1915. The evidence was that the pearls in Frankfort had not been interfered with in any way, and were still with the consignee; and that the Brussels jeweller had, with the plaintiff’s consent, placed his consignment of pearls in a bank for safe-keeping. It was held that there was no loss. The issue had to be tested at the date of the writ, citing Polurrian Steamship at927: “That is the governing date” (at 467/8). It was held that although the plaintiff had lost the commercial adventure of the consignment on approval, there had been no loss (at 473). Indeed, it may be doubtful whether there had even been a dispossession (at 472). Since in that case, however, there was no loss, whereas in the present case it is common ground that there was a loss within the period of the policy (which went up to 31 March 1991), the case is of more interest for statements of principle. For instance at 472 Bankes LJ added:

“It is quite inadmissible to conjecture whether since the policy expired they may have been improperly dealt with. If they have, they may have been lost, but any such loss occurred after the policy had ceased to attach.”

As for a general test of loss in a case of deprivation of possession, Bankes LJ said this:

“The word “loss” in such a policy as this may have a very different meaning when applied to perishable goods, or to goods warehoused at a heavy rent…Mere temporary deprivation would not under ordinary circumstances constitute a loss. On the other hand complete deprivation amounting to a certainty that the goods could never be recovered is not necessary to constitute a loss. It is between these two extremes that the difficult cases lie…” (at 471) “…I cannot attempt a definition of what constitutes a loss, but I find in the language of Blackburn J. in Wilson v. Jones (1867) LR 2 Ex 139, 152 a sentence in which I think I see a clue which may help to solve the problem at hand. It is true that he is speaking of a marine insurance and of a total loss when he says: “If the interest was an interest in the cable being laid at any time, there was still a total loss; for although there was some chance of the cable being recovered, it was a mere chance.” I think these last words are applicable to a case like the present. If the true conclusion from the facts existing at the time of action brought was that the plaintiffs’ chance of recovering their jewels was a mere chance, then the plaintiffs might be entitled to recover on the ground of a loss of their jewels” (at 472/3).

It may be noted that in Moore v Evans the policy year expired prior to the date of writ. It is not entirely clear therefore what Bankes LJ meant when he said that the governing date was that of the writ. He referred for that proposition to Polurrian Steamship: but Kennedy LJ was there speaking of a claim for a constructive total loss in circumstances where a vessel was captured in war on 25 October 1912, notice of abandonment given on 26 October, that same date was taken to be the date of writ, but the vessel was released on 8 December 1912. The significance of those dates was that, if the claim for a constructive total loss had been made good, which it was not, then the recovery of the vessel subsequent to the (deemed) date of the writ would not have affected the claim: the vessel would have belonged to the underwriters. For this reason it is common practice for underwriters to be asked to agree, on receiving a notice of abandonment, to deem a writ to have been issued against them by their insured. That makes it unnecessary for the insured to protect their position by immediate issue of a writ. However, Moore v Evans and the present case are not concerned with marine insurance or constructive total loss but with non-marine insurance and actual total loss. Even in marine insurance it has never been authoritatively determined that the date of writ is relevant to the existence of an actual total loss, and the current view of the learned editors of Arnould’s Law of Marine Insurance and Average, 16^th ed, 1981 at para 1137 is that it is not: once there has been an actual total loss, any subsequent recovery is merely salvage for the underwriters, provided however that the loss continues to be total down to the expiry of the risk. The editors also suggest, however, that in practical terms it may be doubted whether any court would find that confiscated goods had been irretrievably lost where the goods had been released before action brought. See also Arnould, Vol 3, 1997, at para 1137.

In Holmes v. Payne [1930] 2 KB 301 pearls were again the centre of attention, but in this case they were missing rather than detained. The plaintiff reported her pearl necklace missing and made a claim on her insurers for it. She had searched for it and did so again at the request of her insurers. It was not found and the claim was settled. Subsequently the necklace was found and the plaintiff claimed rescission of the settlement on the basis that there had never been a loss. The settlement was held to be binding, but Roche J opined that there had been a loss. He avoided any definition of “loss”, but said (at 310):

“Uncertainty of recovery of the thing insured is, in my opinion, in non-marine matters the main consideration on the question of loss. In this connection it is, of course, true that a thing may be mislaid and yet not lost, but, in my opinion, if a thing has been mislaid and is missing or has disappeared and a reasonable time has elapsed to allow of diligent search and of recovery and such diligent search has been made and has been fruitless, then the thing may properly be said to have been lost. The recovery of the thing is at least uncertain and, I should say, unlikely.”

The relevant dates in that case were as follows: the policy year ran from 7 November 1928 to 6 November 1929; the necklace was missed on 20 November 1928, and the plaintiff made her claim on the next day; in December 1928 the settlement was made; and the necklace was found on 27 February 1929. Those dates present no difficulty, for everything happened during the policy year. They illustrate, however, that at any rate by agreement an actual loss can be established quite quickly.

In Webster v. General Accident Fire and Life Assurance Corporation Ld [1953] 1 QB 520 the plaintiff claimed for the loss of his motorcar, which he had put into the hands of a confidence trickster on the fraudulent representation that the latter had a buyer for it; the rogue sold the car at auction and pocketed the proceeds; the police advised the plaintiff that there was nothing he could do to recover the car, and so he took no further steps in the matter, although he knew in whose hands the car was. An arbitrator (the future Chapman J) held there had been a loss of the car. On a special case Parker J referred to Moore v. Evans and said (at 531/2):

“Every case depends upon its own facts. An assured is not entitled to sit by and do nothing. Equally, he is not bound to launch into legal proceedings or if necessary carry them to the House of Lords. The test, as it seems to me, is whether, after all reasonable steps to recover a chattel have been taken by the assured, recovery is uncertain.”

In Panamanian Oriental Steamship Corporation v. Wright [1970] 2 Lloyd’s Rep 365 (The Anita), (reversed on appeal [1971] 1 WLR 882, but on a different point) a vessel was confiscated for infringement of customs regulations under order of a Vietnamese court made on 25 April 1966, during a six-months policy running up to 15 July 1966. Notice of abandonment was given on 2 May 1966 and a writ was deemed to have been issued on the same day. The claim for constructive total loss based on the notice of abandonment (and deemed writ) of 2 May 1966 failed, because the insured could not show that as of that date recovery was unlikely. However, a further notice of abandonment was given on 12 May 1967 and a writ was actually issued on 29 August 1967, at which times the vessel was still detained, and Mocatta J held that on this basis the claim for a constructive total loss was made good. There was also a claim for an actual total loss in the writ of 29 August 1967, by which time the vessel had been detained for 17 months. Since the claim for a constructive total loss had succeeded, this further claim was academic. However, in an obiter passage, Mocatta J said:

“The question has to be answered as at the date of the writ. It may be true that the order of confiscation divested the owners of the legal ownership of the vessel as is the case after condemnation of a ship by a Prize Court. But the test of irretrievable deprivation is clearly far more severe than the test of unlikelihood of recovery of possession and, despite the gloomy prospects for the future as at Aug. 29, 1967, I feel unable to find that the plaintiffs were at that date irretrievably deprived of their vessel.”

The insured in that case suffered deprivation of possession as from 25 April 1966 (at 381), by reason of a peril covered by the policy (restraint of people, at 378), but no loss until 12 May 1967, which must have been outside the policy year covering the incidence of the peril. As far as I can tell, there is no discussion in the case of the relevance of these dates. Mr Kendrick relied on The Anita in his skeleton argument as illustrating the difficulty of proving an actual loss in a case of mere deprivation of possession. Since, however, an actual loss is accepted in our case, that is in a sense beside the point. Such matters must depend on the facts, and Holmes v. Payne provides a different illustration.

The question arose, in the context of Mr Boyd’s third submission (para 28 above), whether further authority threw any helpful light on the premise that an actual total loss only established as occurring following the expiry of a policy but by reason of a peril whose effect commenced within the policy period either had to relate back to the initial date of the incidence of peril or else was uninsured (saving I suppose any partial loss occurring within the policy period). I asked whether there was any doctrine which might be described as a “grip of the peril” doctrine. I was told there was not.

It may be that my enquiry was viewed as relating to specifically non-marine cases. It does appear to me, however, that at any rate marine insurance recognises both a “death’s-blow” doctrine and what may also perhaps be described as a “grip of the peril” doctrine. The “death’s-blow” doctrine arises where a vessel has suffered some grievous damage which cannot be said to amount all at once to a total loss, or even a constructive total loss, and which only develops into a total loss after the risk expires: see Arnould at para 1138 citing Knight v Faith (1850) 15 QB 649 (where the doctrine was recognised, but not applied on the facts of the case). Arnould goes on to discuss cases of deprivation of possession under the same doctrine, but it may be that “grip of the peril” better describes such examples. Arnould says:

“Bailhache J stated that principle in the following terms [in Fooks v Smith [1924] 2 KB 508 at 514]: “Where by a peril insured against there is a constructive total loss and no notice of abandonment is given, then if in the ordinary course of an unbroken sequence of events following upon the peril insured against the constructive total loss becomes an actual total loss – as, for instance, there is a capture followed by confiscation – the underwriter is liable in respect of the total loss. If, however the ultimate total loss is not the result of a sequence of events following in the ordinary course upon the peril insured against, but is the result of some supervening cause, the underwriter is not liable. That is an illustration of the doctrine proxima causa non remota spectatur.”

“Although the doctrine expounded by Bailhache J is unquestionably right in principle, it should be observed that where the assured has been deprived of possession of his ship or goods by capture and some accident befalls the property when it is in the hands of those who have taken it, the proximate cause of loss has generally been regarded as the capture, not the subsequent casualty. Similar reasoning may apply in relation to other perils involving deprivation of possession, in appropriate circumstances, but capture (which has always been regarded as working an immediate total loss) is in some respects sui generis.”

Again, although the doctrine was recognised in Fooks v. Smith, the facts of the case were decided to fall outside it.

Such a doctrine would explain why there was no problem about dates in The Anita. The doctrine appears to have been recognised in Integrated Container Service Inc v. British Traders Insurance Co Ltd [1984] 1 Lloyd’s Rep 154 at 160, where Eveleigh LJ said:

“Moreover, it did not matter if further loss might occur after the expiration of the policy, for the containers had already been the victims of an insured peril within the policy period. They had received a potential death blow.”

It is said, however, that such a doctrine or doctrines do not apply outside marine insurance: see The Law of Insurance Contracts by Prof Malcolm Clarke, 1999 at para 16-2A1, citing Moore v. Evans; but, subject to express contractual provision, it is not entirely clear why not.

In Dawson’s Field, however, Mr Kerr distinguished (at 6) between cases arising from dispossessions which were intentional (such as capture, enemies, piracy, see Cory v Burr [1883] AC 303 at 398) and true “wait and see” situations, such as hijackings. In the latter, he doubted that there would be any difference in outcome depending on whether the subject matter of the insurance was marine or non-marine (at 5/6):

“In my view, however, these are not considerations which are or ought to be decisive of this case or any similar factual situation. If a lorry is hijacked on the M1 near London and the driver is forced to drive the hijackers to Scotland, then the position under an insurance policy on the lorry should in common sense be the same as if a similar incident happened in relation to a yacht off the South Coast and the crew were ordered to take the hijackers (whether or not they be called pirates) to France. If the persons in control then refused to release the lorry and yacht and their crews unless and until certain demands were met, then again the position should not in common sense be any different. Nor do I believe that there is a difference in law between these two situations. If the owners of the lorry or of the yacht claimed for a total loss in this situation, while the ultimate fate of their property was still uncertain, then the test must in each case be whether or not a Court would be bound to give judgment for them if a writ had been issued at once (a notice of abandonment having been given in relation to the yacht) and the action had come on for trial before the outcome was known. In my view such an action would not succeed.”

I would assume that the qualification “and the action had come on for trial before the outcome was known” is meant to emphasise the thought that the matter would have to be judged as at the date of the writ. This reasoning leaves open, however, the situation where total loss is established after the date of an early writ, assuming new proceedings are possible or the writ is deemed to have been issued anew. In the present case, however, the original deprivation of possession of the BA aircraft on 2 August 1990 is even less like a capture situation than a hijacking. It was described by the judge as rather like a stranding.

In Dawson’s Field, as Mr Boyd observed during his submissions, two different aggregation clauses fell to be applied: one was practically the same as that found in this case, namely “arising out of one event”; the other was “one occurrence”. It was in connection with “one occurrence” that Mr Kerr developed the test of the “unities” (at 9). For these purposes it is common ground that there is no relevant difference between an event and an occurrence. It was in that context that Mr Kerr said (at 10):

“No one contended that each explosion was a separate occurrence. In my view there was one occurrence, one event, one happening; the blowing up of three aircraft in close proximity, more or less simultaneously, within the time span of a few minutes, and as a result of a single decision to do so without any one being able to approach the aircraft between the first explosion and their destruction.”

He then turned to the concept of losses “arising out of one event” (at 10) and said:

“I confess that I had even less difficulty with the words “arising out of one event”…The issues were as to the meaning of the “arising out of” and the question whether or not it could be said that the destruction of the aircraft arose out of one “event”. “Arising out of” may perhaps mean no more than “proximately caused by”: see Corporation of the Royal Exchange Assurance v. Kingsley Navigation (1923) A.C. 235, but I think that in the present context it is probably wider, because the Clauses envisage that one event may cause a plurality of loss and/or damage affecting more than one aircraft…I also reject the contention faintly and more or less formally advanced by the Claimants that the hijackings arose out of one event, viz the P.F.L.P.’s overall plan. I agree that a plan cannot by itself constitute an event. But it was then said on behalf of the Respondents that the destruction of the aircraft at Dawson’s Field could also not be said to have arisen out of one event, because the only unifying event could have been the decision to blow up the aircraft. But in my view this approach is much too narrow, though this view must admittedly be coloured by my view about “occurrence”. The destruction of the aircraft arose from the decision or order to detonate the explosive charges in them which was thereupon carried out in the way described above. If three aircraft become total losses because a decision or order to blow them up together is carried out, why is the carrying out of the decision or order not one event?”

At the conclusion of his award Mr Kerr answered the critical questions posed as follows (at 12):

“(3)

If the Claimants can rely on the destruction of the aircraft by explosion as total losses, and it not being contended that the aircraft destroyed at Cairo can be aggregated with the aircraft destroyed at Dawson’s Field for this purpose, did the destruction of the three aircraft at Dawson’s Field constitute

(a)

one loss or

(b)

one occurrence or

(c)

a series of losses or occurrences arising out of one event or

(d)

none of these

within the meaning of the Clauses in the submission?

My answer is: both (b) and (c).”

It follows that in Mr Kerr’s opinion the blowing up of the three aircraft or their destruction was “one occurrence” and their loss arose out of “one event”, namely the carrying out of the decision to blow them up. It seems to me that in essence Mr Kerr’s reasoning was the same in both cases and the test of the unities applied in both cases. In retrospect, it strikes me that Mr Kerr’s answer to question 3(c) comes close to saying that the loss of the three aircraft arose out of one event namely the loss of the three aircraft. Perhaps “one occurrence” emphasises the passive aspect of the occurrence or event, and “arising out of one event” emphasises the active or causative aspect. Moreover, although Mr Kendrick is entitled to point to the fact that the “one event” identified from which the losses arose was a very immediate cause, namely the blowing up of the aircraft, and not something less immediate, like their hijackings, that was at least in part because in any event the hijackings could not be aggregated into one event or occurrence.

In the present case, however, the “one event” identified by Langley J so far as the KAC fleet is concerned is not their loss but the invasion and capture of the airport, and it was from that that their loss arose; and the question is whether the loss of the BA aircraft arose from the same or another event. In these circumstances, and given that it was never even in contention that the loss of the fourth aircraft might have been aggregated, it is not clear to me that the actual decision in Dawson’s Field throws any particular light on the present case, as distinct from the presentation there of the “unities” test for helping to answer the question of whether a plurality of losses constitutes or arises out of a single event.

In this connection I remind myself that in KAC v. KIC, where I adopted the “unities” test and applied it to the facts of the present case so far as at any rate concerned the KAC aircraft and spares, for I was not dealing there with the BA aircraft, I was concerned with the expression “any one occurrence” rather than “arising from one event”: but I nevertheless said (at 686):

“…the losses’ circumstances must be scrutinized to see whether they involve such a degree of unity as to justify their being described as, or as arising out of, one occurrence” (emphasis added).

I note that in the judgment below Langley J cited that passage (at para 64), apparently without any controversy.

Caudle v. Sharp [1995] LRLR 433 concerned excess of loss whole account reinsurance of liability insurance. The phrase concerned was “series of losses and/or occurrences…arising out of one event”. I have already referred (at para 26 above) to the discussion at 438/9 concerning the nature of the causative relationship involved in the concept of “arising out of”. Evans LJ also said (at 438):

“In my judgment, the three requirements of a relevant event are that there was a common factor which can properly be described as an event, which satisfied the test of causation and which was not too remote for the purposes of the clause.”

The dispute in that case was whether the losses caused by an underwriter’s negligence in writing 32 run-off reinsurance contracts without any investigation into or consideration of the risks involved from asbestosis arose out of one event. The case started in arbitration and the arbitrators’ award was appealed to the court. At first instance Clarke J had held that the losses did arise out of one event, since they all arose from the negligent omission of the underwriter to investigate the problems of asbestosis. On appeal, however, this court disagreed: a negligent omission could not be an “event”; and, even if it could, the losses rather arose from the making of 32 separate contracts. Moreover, even if the making of the 32 contracts could themselves be viewed as sufficiently connected to constitute a single historical event, that did not suffice, since the aggregation clause required the series itself to arise out of a single event (at 440 per Evans LJ, with whom Rose LJ agreed). This last aspect of the decision would seem to support my concern as to the correlative aspect of Mr Kerr’s decision on the wording “arising out of one event”. As for causation and remoteness, Evans LJ put the matter thus:

“First, suppose that Mr Outhwaite’s negligence could be attributed, not to his indeterminate failure to inform himself but to a specific occasion when he was misinformed as to the scale of the asbestosis problem. Even assuming that that previous occasion could be regarded as a causative event, it would not, in my judgment, bring all subsequent contracts and therefore losses within the clause, because it would be too remote even to be regarded as a single event out of which the losses arose, for the purposes of the clause.”

Nourse LJ, who agreed in the result, perhaps expressed the matter most pithily in the following passages (at 443):

“An event must be something out of which a loss or series of losses arises. Here neither Mr Outhwaite’s state of mind nor his failure to instruct himself can be said in any real sense to have been something out of which the losses arose. They arose out of his negligent writing of the 32 policies. There was no “event” before the first of them was written…

“The alternative suggestion that the writing of the 32 policies was one event is equally unsustainable. It seems clear that it did not appeal to Mr Justice Clarke. Moreover, in par. 10.17 of their award the arbitrators said there was no doubt that the act of writing each policy was a separate occurrence in itself. Earlier, in par. 9.6 they had recorded statements by Mr Outhwaite to the effect (among other things) that he had considered each risk offered individually on its merits and had exercised a separate, distinct underwriting judgment in relation to each; and that as many risks were rejected as were accepted and as many were not taken up as eventuated in written contracts. Although the arbitrators did not make findings as to those matters, it cannot be reasonably be suggested that there was any relevant linkage between the 32 policies.”

Caudle v. Sharp was decided before permission had been sought and obtained in KAC v. KIC from the parties involved in the Dawson’s Field Arbitration to place Mr Kerr’s award before the court. It was therefore decided without any express reference to the test of the “unities”. Nevertheless, one can detect a similar process of reasoning in Nourse LJ’s analysis of the alternative submission that the 32 contracts constituted a single event. As for the somewhat different question involved in the phrase “arising out of one event” with its causative emphasis, this court stressed four analytical elements, viz (i) something that can be called an “event”; (ii) the function of that event as being prior to the aggregated losses; (iii) a causative link between losses and event, undefined other than being looser than proximate cause; and (iv) the absence of remoteness. To which can of course be added the underlying concept of aggregation itself, that of a single unifying event.

It follows that if the only suggested unifying factor is something which cannot properly be called an “event”, then aggregation cannot occur. It also follows that if a discrete factor is suggested (as in this case, the inevitability is war is posed as a possible candidate) which cannot properly be called an event, then that may suggest (but I do not think I would put it higher than that) that some other factor should be looked for as the event out of which the relevant loss arose for the purpose of testing whether or not it arose out of the same event as other losses.

As for causation and remoteness, it seems plain that the latter concept was used by this court as a tool to limit the otherwise infinite reach of the workings of causation. That is the function of the concept of remoteness in the law generally. It is in other words a legal tool used to separate out relevant from irrelevant causes. As often happens, however, the use of this tool is somewhat opaque. What the decision in Caudle v. Sharp does seem to me to suggest, however, is that, even though the causative link is looser than that of proximate cause, the courts will look for a nearer and more relevant cause than for a more distant one. Another way of saying this is that the causative link has to be a significant rather than a weak one. In this connection too, the factual analysis of Nourse LJ is interesting, in that it emphasised the evidence for saying that the 32 contracts were entered into as separate exercises of underwriting judgment, while other proffered contracts were declined.

Axa Reinsurance (UK) Plc v. Field [1996] 1 WLR 1026 concerned the contrast between two different aggregation clauses. One, which was found in an underlying E & O policy, limited total liability in respect of all claims “arising from one originating cause, or series of events or occurrences attributable to one originating cause”. The other, which was found in an excess of loss reinsurance policy, was in terms of all losses and occurrences “arising out of one event” (ie as in Caudle v. Sharp). It was in this context that Lord Mustill contrasted the concepts of “event” and “ originating cause” at 1035G:

“In my opinion these expressions are not at all the same, for two reasons. In ordinary speech, an event is something which happens at a particular time, at a particular place, in a particular way…A cause is something altogether less constricted. It can be a continuing state of affairs; it can be the absence of something happening. Equally, the word “originating” was in my view consciously chosen to open up the widest possible search for a unifying factor in the history of the losses which it is sought to aggregate. To my mind the one expression has a much wider connotation than the other.”

What I would take from that passage for present purposes is first, confirmation that an event is to be distinguished from a state of affairs; secondly, that the causative link inherent in the words “arising from”, when coupled with the expression “one event”, should be regarded as a relatively strong and significant link, since it is looking to a causal relationship with something, an event, described in Lord Mustill’s particularist terms; and thirdly, that Lord Mustill’s reference to “time…place…way” as defining elements of an event is reminiscent of Mr Kerr’s “unities”.

Those “unities” were again considered in Mann v. Lexington Insurance Co [2001] 1 Lloyd’s Rep 179. A retrocession contained a limit of $5,000,000 “per occurrence” and applied a deductible “each location any one occurrence”, against the background of a reinsurance which spoke of “each and every loss, each and every location”. The losses were at 67 supermarket stores each in different locations, which had been damaged over the course of two days by rioters in Indonesia. The retrocessionaires’ case, that the total riot damage constituted one occurrence, was based on the allegation that all the rioting was deliberately orchestrated by the government then in power. This court, however, held that in its context an “occurrence” had to occur at a particular location and that therefore the losses could not be aggregated. Of particular importance for the present case, however, was the way in which this court went on to deal, in what is probably an obiter passage, with the “unities”. Since the losses occurred at different locations over a wide area, at different times over two days, therefore there was no unity in time or place. The only unifying factor relied on was that of intent or central orchestration, but that did not suffice: in that connection Waller LJ relied on Mr Kerr’s rejection of the orchestration of the hijackings as being a sufficiently unifying factor (at 190).

I revert to Mr Boyd’s four principal submissions. The first (para 26 above) was that the causal link expressed by the phrase “arising out of” is a weak one, and that therefore and possibly in any event there was nothing that occurred after the initial invasion and capture of the airport that outweighed their significance for purposes of causation.

In my judgment, however, there is nothing in the authorities to support that submission as a matter of principle. On the contrary, it seems to me that in Dawson’s Field, and again in Caudle v. Sharp a significant causal relationship was, albeit implicitly, imposed. In those cases Mr Kerr and this court found the relevant event (or occurrence) in the nearer events, rather than in the more distant. In the latter case, the concept of remoteness was expressly adverted to. I accept that in Dawson’s Field the choice was obscured by the fact that the hijackings could not be regarded as a single event, and for that reason could not even be a candidate. Nevertheless, it seems to me ultimately to be inherent in the concept of aggregation (“arising out of one event”) that a significant causal link is required. In this connection I would refer to Lord Hoffmann’s substantial contribution in recent years to an understanding of what lies behind the courts’ intuitive judgments on issues of causation: see, for instance, Empress Car Co (Abertillery) Ltd v. National River Authority [1999] 2 AC 22 at 29/35. Lord Hoffmann emphasises that it is not possible to give an informed answer to a question of causation when attributing responsibility under some rule without knowing the purpose and scope of the rule. In the present context, the purpose and scope of the rule has to be found in the concept of aggregation inherent in wording such as “arising out of one event”. A plurality of losses is to be regarded as a single aggregated loss if they can be sufficiently linked to a single unifying event by being causally connected with it. The aggregating function of such a clause is antagonistic to a weak or loose causal relationship between losses and the required unifying single event. This is the more easily seen by acknowledging that, once a merely weak causal connection is required, there is in principle no limit to the theoretical possibility of tracing back to the causes of causes. The question therefore in my judgment becomes: Is there one event which should be regarded as the cause of these losses so as to make it appropriate to regard these losses as constituting for the purposes of aggregation under this policy one loss?

If that question is asked in the present case, it seems to me difficult intuitively to say that the loss of the BA aircraft arose from the same event as the loss of the KAC fleet, so as to require the aggregation of those losses, when their respective losses arose in different circumstances and at different times and for different reasons, as explained in the detailed findings of the trial judge. Mr Boyd submitted that Langley J’s intuitive judgment differed from that of the market, which promptly regarded the BA aircraft and KAC fleet as all part of one loss. However, the market did not have the benefit of the detailed factual examination conducted at the trial and of the assistance of the expert witnesses. That detailed factual examination involved, whether intuitively or analytically, a consideration of those aspects of the matter which are referred to as the “unities”. In a sense, however, I am anticipating subsequent discussion.

Before leaving Mr Boyd’s first submission, however, it is interesting at this stage to observe that it was not suggested that the invasion and capture of the airport were the proximate cause of the BA aircraft’s loss, but only that they satisfied the weaker causal test of being the event from which that loss arose. The argument, therefore, was the opposite of, or at any rate not the same as, the submission that might have been made, namely that the invasion and capture of the airport led to the detention of the BA aircraft and that that detention, going back to 2 August, was so significant an event that it amounted to the “death’s blow” or at any rate to the “grip of the peril” which led inexorably or at any rate in some other critical sense, viz. Bailhache J’s test of “in the ordinary course of an unbroken sequence of events” to the aircraft’s loss. At the highest it was said at one point that the chain of causation had not been broken. See also under para 79 below.

Mr Boyd’s second main submission (para 27 above) was specific in criticising the judge’s finding as to the cause of the BA aircraft’s loss, inter alia on the basis that the judge was also making a finding as to the event from which the loss arose. I would accept that criticism so far as the inevitability of war is concerned, to the extent that the judge may here have been speaking about events. I would agree that the inevitability of war does not sound like an “event” in Lord Mustill’s terms, but more like a state of affairs, or even, as I would venture to put it, a judgment about a state of affairs. The nature of Mr Boyd’s criticism, however, demonstrates that he finds in the relevant sentence of para 73 of the judge’s judgment what, although expressed in terms of cause, is effectively a finding as to the “event” from which the loss of the BA aircraft arose. An event is not necessarily the same as a cause, but is a narrower concept, as Lord Mustill reminds us: but in context an event within the phrase must also have a status as a cause. The judge nowhere in terms states what the “event” from which the loss of the BA aircraft arose, only that it did not arise from the same event as the loss of the KAC fleet. That was the question which he had to decide, for the purpose of the aggregation clause. Strictly speaking, he did not have to decide from what other event the loss of the BA aircraft arose: and it did not inevitably have to arise from any event as distinct from some cause. Mr Boyd submitted in the alternative that in the absence of a clear and reasoned finding as to what was this other event and as to its causal relevance to loss, the inference should be drawn that the loss arose from the same event that had caused the original deprivation of possession. However, I disagree. The eventual question remains whether the loss of the BA aircraft arose from the same event as the loss of the KAC aircraft, not whether it arose from the same event as caused the original deprivation of possession. I therefore think that Mr Boyd’s criticism, while accurate as to “event”, fails in terms of the judge’s precise reasoning, which was examining the question of unity of cause. Indeed, one of the useful services which the test of the “unities” performs is that it analyses the constituents of what makes up an event (cf Lord Mustill’s time, place and way) in a manner which has the negative virtue of not begging any questions and also has the positive virtue of being geared to Lord Hoffmann’s insight of asking: Why do you want to know? To which the answer is: Because I want to know whether this loss should be aggregated with that loss.

If, however, the inevitability of war is not an event, the destruction of the BA aircraft and the war or its outbreak were both undoubtedly events. Mr Boyd’s criticism about the judge’s identification of the aircraft’s destruction as a cause or event was that this only occurred after the loss had been agreed on the basis of deprivation of possession. I would acknowledge the element of controversy about the finding of the aircraft’s destruction as a cause of loss. How is that controversy to be resolved? The agreed finding of destruction “on or around 27 February” is not precise and therefore to my mind is not determinative. In the end I have no reason to doubt the judge’s finding that the aircraft was destroyed “on some date between 13 and 26 February”. At that time there was as yet no finally binding settlement. Nor had legal proceedings been commenced, which is perhaps another way to obtain a governing date for resolving the status of the property. Let us suppose, however, that the facts had been that the war had ended (which it did on 3 March) with the aircraft still undamaged, no proceedings yet brought, and no binding settlement: in that case a claim based on loss by deprivation of possession would have failed. That seems to me to indicate that the judge was right to say that the primary cause of loss was the aircraft’s destruction.

If, however, I am wrong about that, there is to my mind no answer to the finding that war, or its outbreak, was the cause of the aircraft’s loss. Of course, in one sense the war grew out of the invasion of Kuwait. But in terms of the loss of the aircraft the invasion was in no way as significant as the deterioration of the situation into a war involving Iraq and this country. The details and uncertain course of that deterioration are set out in the judge’s findings. I would refer in particular to the judge’s findings that the BA aircraft, unlike the KAC fleet, was in no sense a target of the invasion; that if the question had been asked as of 2 August 1990 “Is the aircraft lost?” the answer would have been “I don’t know: wait and see”; that at no time did Iraq form any significant policy intention towards the BA aircraft; that Professor Freedman’s analysis, inter alia that war had only become more likely than not by the time of the UN resolution of 29 November 1990 and that the BA aircraft was not at risk until coalition military operations began on 16 January 1991, was realistic; that had there been no war there would have been a realistic chance of the BA aircraft being restored to BA; and that this factual situation was reflected in the difference between KAC’s immediate claim for the loss of its fleet and BA’s claim which was formally pursued only after war had begun.

In this connection it seems to me that the importance of the war was twofold: the more minor consideration was that the BA aircraft became, in Iraqi terms, enemy property; but to my mind of far greater significance was the fact that it placed the aircraft in grave danger of being in the line of fire. That was one of the points made in BA’s claim letter of 13 February 1991: “With the probability of a ground war commencing, the chances of its remaining without damage seem slim.” Kuwait was the ground to be fought over. What was anticipated was what occurred. The loss was caused by the aircraft’s destruction or by the war and not by the invasion. It follows that the aircraft’s loss undoubtedly arose from the event of its destruction or the war. It could of course also be said that, in a sense, it arose from the event of the invasion of Kuwait, and I bear in mind that the proximate cause is not determinative. Indeed, in the larger context of invasion and war, within a policy which covers war, the actual event of destruction might even seem too close to answer to the wording “arising from”. A choice is therefore there to be made. I will revert to that choice below.

Mr Boyd’s third submission (para 28 above) is that the date of the aircraft’s loss should relate back to the date of BA’s first deprivation of possession on 2 August 1990. Although he stresses, and the judge accepted, that unity of date of loss is not by itself critical to the question before the court, he is keen for obvious reasons to emphasise that the BA aircraft was lost at the same time as the KAC fleet. If the aircraft was lost by destruction, as in my judgment it was, then I do not see this third submission as even getting off the ground. The aircraft was lost between 13 and 26 February 1991.

If on the other hand the aircraft was lost by reason of deprivation of possession, then again, in my view, on the facts of this case it was not so lost until 1991 at the earliest. The present case differs from Moore v. Evans in that the BA aircraft was in the control of the Iraqis from the beginning, whereas the pearls remained in the hands of the jewellers to whom they had been consigned, or in a bank with the plaintiff’s consent. Nevertheless, as in Moore v. Evans, it is impossible on the judge’s findings to say that BA was irretrievably deprived of its aircraft from the first, whatever the content of that test may be. It was a “wait and see” situation. Care must no doubt be taken with that expression, because it is capable of being used in two senses. In its real sense, it refers to a situation which is subject to a process of development and change. Will a ransom be paid and honoured and the property recovered? Will the property be released? That is the sense in which it was used by Mr Kerr in Dawson’s Field and again by the judge in this case. In another sense, it might be used to refer to the emergence of evidence about the initial deprivation. Was the car taken by a joyrider or a thief? If by a joyrider, the probability is that it will soon be found and recovered and there is no total loss. If by a thief, the probability is the other way. In the latter case, where the inference of theft is drawn from the very fact that the car is not found abandoned, it seems to me to be right to say that the total loss occurred at the time of theft. Therefore the agreement with the insurance industry brokered by the insurance ombudsman referred to in the course of submissions seems to me to do no more than reflect the realities: but it tells one nothing about the current problem. The case of missing property (Holmes v. Payne) is again another situation. A total loss cannot be proved without a proper search, or more than one, and perhaps also without the passing of a certain period of time just in case the property “turns up” (like the pearl necklace). Where, however, loss is proved, I can see arguments both ways for saying that the loss occurs when it is first discovered (perhaps even earlier), or only after further search. That is not this case: and in practice the problem does not appear to have caused litigation.

In the present case, however, the “wait and see” concept is used in its real sense. In my judgment, when once the syndicates’ primary submission, that the BA aircraft was lost at the same time and as part and parcel of the KAC fleet’s loss, failed, as Mr Boyd accepts it has, then it becomes impossible to say that the aircraft’s loss, whenever it did occur, occurred on 2 August 1990. Ex hypothesi, it occurred only subsequently. That is why to my mind Mr Boyd’s third submission as I understand it is, or must be, a different one, namely that even though the BA aircraft’s loss occurred subsequently, nevertheless that loss, when once it happened, relates back to the time of original deprivation. (Of course, on a broader level of argument, his submission is that that loss at any rate arises from the original deprivation.) Otherwise, says Mr Boyd, an insured could fall between the two stools of two different years’ insurance.

Since, however, on any view the loss in the present case occurred during the period of the insurance (which ended on 31 March 1991), there is no need to decide that point on this appeal. I would merely comment that Moore v. Evans would seem to be against Mr Boyd’s submission, and in that context that it is to be noted that the period clause in the policy under consideration appears to be emphatic that the loss must occur precisely within the contract period for it to be covered (see para 6 above). It may be (and it certainly has not been examined on the present appeal) that marine policies are written or treated in some other way. In theory, of course, the triggering event for a loss covered by a policy can be defined by reference to the incidence of either a peril, or a loss, or a claim, or any combination of the three. The problem is particularly acute in liability insurance, but that is another story.

I have referred above to the “death’s blow” or “grip of the peril” doctrines. The case was not argued on that basis, and would only have had to have been if the aircraft’s actual loss had occurred after 31 March 1991. If, however, the argument had been presented in that way, and even on the assumption that such doctrines apply outside the maritime sphere, I would have been inclined to say that there was no death’s blow or equivalent grip as of 2 August 1990, and that the aircraft’s deprivation did not lead in the ordinary course of an unbroken sequence of events to its loss. This is in that respect an a fortiori example of the conclusion of Mr Kerr in Dawson’s Field that the hijackings were not the proximate cause of loss in that case; and see also paras 49/50 above.

I would go further and say that even if total loss by deprivation of possession actually occurring outside the period of the policy were to be treated as falling in the policy year of the initial incidence of the risk leading on to that total loss, as Mr Boyd has broadly submitted, and even if on that ground and for that purpose the loss is treated as occurring at the date of the initial deprivation of possession, the use of that date is only a convention or metaphor for the true date of actual total loss. Therefore the most that Mr Boyd could say is that he would be entitled to point to the date of the commencement of the deprivation of possession as being in a certain sense relevant to the date of the aircraft’s actual loss. But he is entitled to say that in any event. He is entitled to say that any loss of the aircraft by deprivation of possession in January or February 1991 originated in a deprivation of possession which started on 2 August 1990. He is entitled to say that the aircraft’s loss arose from that initial deprivation of possession, and therefore to submit that the date of 2 August 1990 is as important or even more important than the actual date of total loss. He is entitled to say on these grounds that, when testing the “unities”, the court should take account not only of the respective actual dates of loss but also of the fact that BA’s initial deprivation of possession of the aircraft occurred on the same day as KAC’s total loss. The question is where those facts and that submission take him: which leads me on to Mr Boyd’s fourth submission, and my conclusion.

The fourth submission (para 29 above) is that the test of the “unities” is inappropriate in a case where one event leads both to immediate total losses and to other total losses more distant in time. I disagree. The submission highlights the issue in this case, but also begs the question. Are the losses to be aggregated as all arising from one event? That question can only be answered by finding and considering all the relevant facts carefully, and then conducting an exercise of judgment. That exercise can be assisted by considering those facts not only globally and intuitively and by reference to the purpose of the clause, but also more analytically, or rather by reference to the various constituent elements of what makes up one single unifying event. It remains an exercise of judgment, not a reformulation of the clause to be construed and applied.

It seems to me that there was nothing wrong with that exercise as conducted by the judge. He said that unity of time was lacking because of the presence of “wait and see”, but I do not for a moment think that he lost sight of the fact that the period of “wait and see” was initiated on 2 August 1990. On the contrary, he went directly on to speak of “The need for and passage of time before the loss is established and the intervening events…”, in a passage which effectively incorporated a reference back to 2 August. In a Dawson’s Field situation lack of unity of time may be critical; it was also stressed in Mann v. Lexington. In the latter case, the lack of unity of time and place was not made good by an assumed orchestration of purpose. In the present case, however, there was unity of place only. There was no unity of time, only an identical starting point for enquiry. That was indeed something, but there was no unity of purpose or intent or of cause. Was the element of an identical starting point for enquiry enough? As the judge said: “The need for and the passage of time before the loss is established” – I would prefer to say “before the loss actually occurred” – “and the intervening events in my judgment make it more and more difficult to characterise the loss as one arising from the invasion…” I agree.

In sum, if, on the facts found, and for the purposes of the issue of aggregation, the question is asked “Did the aircraft’s loss arise from the same event as the loss of the KAC fleet, or some separate event?”, or even (to put the question in a form which contains the specific options) “Did the aircraft’s loss arise from the invasion and capture of the airport, or from the war or the aircraft’s destruction in the war?”: I would agree with the judge’s intuitive response. It did not arise from the invasion. In particular it did not arise out of the same event as the loss of the KAC fleet. It arose from its own particular circumstances, a different event. I would be inclined to say that it arose from the event of the war. If the same question is put with the assistance of the “unities”, it seems to me that the intuitive response is confirmed. The loss of the BA aircraft arose from circumstances which differed markedly from those which governed the loss of the KAC fleet. Only their location, and the timing of the start of the story, were the same. But both those factors, as they concerned the BA aircraft, were a matter of chance; whereas the KAC fleet was at home, where Iraq intended to seize it and make it its own, as part of a policy of Kuwaiti plunder, based on historical claims to sovereignty over Kuwait. The tortuous events of international diplomacy and coalition building which led to war, and the loss of the BA aircraft in and as a result of war, some six months later, had nothing to do with the loss of the KAC fleet.

For these reasons I would dismiss this appeal.

In what this court was told was something of a test case for the insurance market, I have sought to treat Mr Boyd’s interesting submissions in full. Nevertheless, there was another submission made by Mr Kendrick which might have led to a shorter response. This was that as a matter of principle this court should not interfere in a result which was as dependent as here on the trial judge’s findings, based as they were on oral evidence and the opinions of geo-political experts, and on his evaluation of those findings, unless the appellant could clearly point to some error of law or principle which had driven the result below. In this connection he referred to the reflections of this court on its appellate function in Assicurazioni Generali SpA v. Arab Insurance Group [2003] 1 All ER (Comm) 140, especially at paras 14/19.

This was a powerful submission, but one which it is unnecessary to determine. I would merely say that the primary facts were not challenged on this appeal (save in respect of the destruction of the aircraft) and that this court was concerned with a clause which is in common use in the insurance industry and falls to be applied in unusual, but for that very reason, possibly illuminating circumstances. Although the relevant principles of law may have appeared to be uncontroversial at trial, where the emphasis was on the syndicates’ primary case, I do not think there was the same lack of controversy once the spotlight was shone onto their alternative case. Those may also be reasons to counterbalance the considerations Mr Kendrick relied on.

In conclusion I thank both counsel for their helpful, wide-ranging, but succinct submissions.

I agree.

From the findings of the judge four things are clear

The question which had to be resolved under the policy was whether the loss of the BA aircraft arose from the same event as the loss of the KAC aircraft. That question had to be answered in the context of a reinsurance policy which defined loss as “each and every loss or series of losses arising from one event” and which sought to limit the total loss for which the reinsurer was liable.

I agree with Rix LJ that in this context a plurality of losses is to be regarded as a single aggregated loss if the loss can be sufficiently linked to a single unifying event by being causally connected to it and that a significant causal link is required in this connection. The question is whether there is one event which should be regarded as the cause of these losses so as to make it appropriate to regard these losses as constituting one loss for the purposes of aggregation under this policy.

I agree with Rix LJ that it is difficult intuitively to say that the loss of the BA aircraft arose out of the same event as the loss of the KAC fleet so as to require the aggregation of those losses since their respective losses arose in different circumstances and at different times and for different reasons as set out by the judge. I agree with him that so far as the loss of the BA aircraft is concerned the invasion was in no way as significant as the deterioration of the situation into a war involving Iraq and this country.

I therefore agree that this appeal should be dismissed.