If P&C Insurance Ltd v Silversea Cruises Ltd. & Ors

A.ii) Loss of anticipated income and extraordinary expenditure incurred to prevent loss of

anticipated income

B. Cruise Credits/On-Board Credits

Cruise credits and/or on-board credits shall be covered at 100% for the subsequent cruise cancellation, with benefit applied to this Section B if any payments are made by the Assured’s Protection & Indemnity Club(s). Limit US$5,000,000 per event.

THE BELOW SIGNED COMPANY INSURES THE RISK DESCRIBED HEREIN FOR THE AMOUNT SET OPPOSITE ITS NAME, AS PER THE TERMS AND CONDITIONS STATED ABOVE, OR IN THE ADDENDUM ATTACHED HERETO.

ISSUED WITH INTEREST SIGNATURE

Gard Services AS 100%

As agent only for If P&C Insurance Ltd. (publ)

from leaving a harbour or other limited area,

remove damaged cargo.

insurer

from any one casualty, and for the total loss of

amount) shall be the amount of freight per day

Under section Ai there was no claim for the Silver Cloud in respect of the consequences of 11 September. In respect of the mechanical breakdown a per diem claim was made for the period 11 October to 20 December 2001. Again a claim of US$1,662,500 was added for “ extra legal/financial/administrative costs.”

For the Silver Shadow the Ai 11 September claim consisted solely of an additional insurance premium of US$ 525,000 apparently incurred in respect of the transit through the Suez Canal on voyage 3123, which left Athens on 10 October 2001, thereafter calling, happily without incident, at Port Said, Sharm-el-Sheikh, Aqaba, Safaga, Djibouti, Salalah, Muscat and Dubai. To this was added a further charge of US$ 1,662,500 in respect of “ extra legal/financial/

administrative costs.”

Clifford Chance went on to say, in their letter of 3 October:-

“ In support of the enclosed spreadsheet and booking curves we also enclose a serious of further documents containing the data and other information which support the spreadsheet and booking curves.”

The five curves are curves which were in fact drawn up for the purposes of this litigation. Four are vessel specific and one is consolidated across the fleet. Contrary to what was said by Messrs Clifford Chance there was absolutely no data or other information enclosed with their letter which in any way supported or cast any further light on how there had been computed the September projection of the revenue which would be expected to have been booked by 11 March 2002.

It is worthy of note that the same projection included a forecast (already several times revised downwards) of a load factor of 62% for 2001, represented by 23,098 total revenue passengers, so that it is obvious that in order to meet the forecast for 2002 Silversea faced a challenge in increasing their passenger base not simply so as to take account of the increased capacity represented by the recent introduction of the fourth vessel but also so as to achieve in 2002 across four ships a load factor which was 10 percentage points in excess of that projected to be achieved in 2001 by reference to 3 ½ ships. The forecast for 2001 can be regarded as by 5 September 2001 fairly accurate since it is the evidence of Silversea that the nature of its business is such that it would not, as at the beginning of September, expect significant “last minute” bookings for cruises departing during the last three months of the year.

The 5 September 2001 projection to which Mr Bajaj referred was not in fact disclosed to insurers until 24 December 2002, pursuant to an order made on 13 December. Unless I have missed something it contains no information from which one can glean how or why there have been made the assumptions that underpin the forecast revenue for 2002, in particular the two assumptions to which I have referred. Furthermore it contains no curve or information from which one could construct a curve showing how much of the revenue Silversea expected to have been booked by 11 March 2002, or indeed by any date except the end date.

In the light of the foregoing history it is little short of breathtaking that Silversea should have maintained before Thomas J on 13 December 2002 that insurers’ requests for disclosure of material underlying the curve on which the claim was now based were made for tactical reasons, that the material should have been sought long ago and that the documents sought had no bearing on the trial. It is true that the insurers had had disclosed to them in July 2002 the documents which ultimately became the “K” bundles but, prior to reformulation of the claim in October 2002 it cannot have been easy sensibly to correlate that material with the claim as then presented. There was certainly no occasion to compare the documents disclosed with the as yet unpresented claim. In any event the document in the K bundles which might have been thought to be most relevant is what came to be known as the “Garey memorandum” of 4 September 2001 which was headed “ 2002 Booking Curves.” This document, prepared for senior management, had attached to it Booking Curves for 2002 cruises. These were booking curves properly so called, rather than revenue curves, which plotted numbers of passengers booked rather than the revenue which they potentially represented. Not all passengers yield the same revenue. Some book more expensive suites than others. In terms of passenger numbers the Garey Memorandum showed that as at 4 September 2001 bookings for 2002 cruises were behind projections – for the first quarter 1,011 passengers behind the curve, albeit that represented an improvement over the previous month. A further 1954 passengers were required for the first quarter if the target was to be met. The second quarter was below the curve by 390 passengers, the fourth quarter ahead by 64.

Intuitively one would expect that a shortfall in passengers of this magnitude would be likely, albeit not certain, to be reflected in a shortfall in revenue. In fact the documents produced contemporaneously showed that this was indeed the position. These documents were likewise disclosed to insurers in July 2002, in the shape of weekly revenue reports produced by Mr Christian Sierralta. The reports were normally produced on Fridays with relevant information downloaded on that day. Friday 31 August was the last working day before Labor Day. That is no doubt why the report with Download Date 31 August 2001 was in fact dated 5 September 2001. It showed the company tracking US$200,000 below the net revenue curve – in fact the shortfall was actually US$171,000 but was rounded up. The revenue target was US$162,159,000, not the same figure as in the 5 September 2001 forecast presented to West LB, although not that far off. Mr Sierralta reported that Silversea needed to book US$2.3 million net revenue per week during the month of September to track in line with the target curve. US$2.9 million had been booked in the previous week, although net revenue increased by US$5.1 million over the week as a result of a positive adjustment for air costs (US$3.4 million) and a negative adjustment for foreign exchange (US$1.2 million).

A further revenue report was produced on Friday 7 September 2001. This was the last before 11 September. It too reported that Silversea was tracking US$200,000 below the net revenue target curve – in fact the shortfall this time was US$ 192,000. Notwithstanding there had been a gain in revenue of only US$ 1.47 million over the week as opposed to the US$2.9 million which had been said to be necessary in order to track in line with the projection, the shortfall to the curve had not apparently deteriorated over the week. There is no reason why this should not in fact have been correctly reported. The US$ 2.9 million was an average weekly figure. The reporting week included and immediately followed the Labor Day holiday. It is entirely likely that the gradient of the projected curve at this point, computed in the light of analysis of historical results, did not anticipate receipts during this truncated and post holiday week of the same order of those required on an average weekly basis over the month as a whole in order to keep actual bookings on track with the projected curve.

The short point however arising from these revenue reports is that they show Silversea as trading very marginally below its projection for 2002 revenue on the last reporting dates before 11 September. This is to be contrasted with the “Booking Curves” enclosed with Messrs Clifford Chance’s letter of 3 October 2002 and then annexed to the witness statement of Mr Bajaj and referred to by him as being drawn by reference to the 5 September revenue projection. These show that, immediately prior to 11 September in terms of revenue actual bookings exceeded projected bookings, as Mr Bajaj observed at paragraph 68 (b)(i) of his witness statement. In fact the excess shown is US$5,719,000. Mr Bajaj put it thus, as had Messrs Clifford Chance before him: -

“Immediately prior to 9/11 actual bookings exceeded projected bookings

and Silversea were in such circumstances well placed to meet projected

revenues on each of the four vessels.”

The position is therefore that the new claim put forward by Silversea in the shape of Messrs Clifford Chance’s letter of 3 October 2002 and the witness statement of 10 October 2002 of Mr Bajaj proceeds upon a factual basis diametrically opposed to that revealed by documents disclosed to insurers in July 2002. The position was in no way clarified by disclosure given on 24 December 2002 of the 5 September 2001 plan by reference to which the relevant projection had been said to have been drawn. Unsurprisingly this led to further requests for disclosure of documents underlying the claim as now presented. Indeed, in a letter of 18 December 2002 Messrs Clyde had indicated that whilst they assumed that the booking curves attached to the 3 October letter purported to summarise or aggregate much of the material in the July 2002 disclosure (thus the Booking Reports and Revenue Reports) insurers “have been unable to identify anything in Silversea’s disclosure that would assist an expert in coming to a view as to the reliability of Silversea’s assertions concerning its projected revenue. This was the point of seeking disclosure of “all relevant historical data and other data, and all documents relevant to the information, variables, assumptions and methodology used to prepare the 2002 projected booking curve on 5 September 2001, which is referred to at paragraph 3 (b) of Clifford Chance’s letter dated 3 October 2002.” ”

It was only when Mr Bajaj gave evidence at the trial that an explanation was proffered for the very different picture portrayed by the “Bajaj curve” as opposed to that apparently disclosed by the contemporary reports. This was twofold. First, it was not based upon the 5 September projection given to West LB but rather upon the 7 September revenue report. This point emerged on Day 2. Leaving on one side the question why the position should have been so consistently misrepresented to insurers this point would not of itself account for the difference in outcome. As I have already remarked the projected target figures for 2002 revenue in these two documents are different, but there is no reason to believe that the projected revenue curve would differ significantly in its shape and gradient as between the two, always assuming that the 5 September plan in fact contained the material from which such a curve could be constructed, as to which I remain wholly unconvinced. The real reason for the “Bajaj curve” showing that revenue booked as at 11 September (or shortly before) was ahead of the projection at that point was because Mr Bajaj had caused the gradient of the projected revenue curve in the 7 September revenue report to be adjusted.

One of the many oddities of the case was that insurers made no effort to satisfy themselves of the accuracy of the “Bajaj curve” once it had been explained how it had been adjusted and why therefore it differed from contemporary curves. There was no provision for reports by forensic accountants but insurers’ expert on the cruise and leisure industry was Mr Nicholas Pattie, a Director of KPMG. He had the assistance of a team of forensic accountants. He has conducted an extensive exercise which involved projecting revenue forward by reference to booking curves properly so called and average yields, and he had the material which would have enabled him to check the accuracy of the Bajaj revenue curve but for whatever reason he was not asked to do so. Indeed he was not told of the nature of the adjustment to the contemporary revenue curve which had caused the Bajaj curve to depict a different situation, i.e. one in which revenue booked was tracking substantially ahead of projection rather than behind it. Insofar as I understood the reason for insurers behaving in this manner it was because they wished to avoid assisting Silversea to prove its case. In the event I have concluded that the claim based on the Bajaj curve is incorrectly formulated in that it does not, in my judgment, produce an ascertained net loss as is required under section Aii. By agreement of the parties I shall not in this judgment make findings concerning a number of the factual issues which were explored before me, of which the accuracy or reliability of the Bajaj curve is one. I merely express my surprise that, in the light of the history which I have outlined, insurers felt able to assert in closing that the Bajaj curve “is an obviously contrived document deliberately manipulated to present a false picture.” Other than in a small respect to which I shall refer in the context of insurers’ claim for rectification I am not yet and may never be called upon to make findings concerning the allegations of dishonesty which insurers have levelled at Mr Bajaj and Mr Peter. It may never now be necessary for forensic accountants to examine the Bajaj curve but I do not assume that it is an inaccurate depiction.

I turn then to consider whether Silversea’s claim under section Ai (other than in respect of the Silver Cloud mechanical breakdown) is well founded in terms of the policy. I will revert to the claim in respect of the Silver Whisper cruises nos. 4107, 4108,4109 and 4110 arising out of the enforced change from New York to Philadelphia. I do not understand there to be any section Ai claim which is alleged to have been brought about by the other changes in itinerary which are set out in paragraph 22.2 of Silversea’s Consolidated Pleading and which I summarised in paragraph 11 above.

A qualification immediately needs to be made in respect of the Silver Cloud. The whole of the Silver Cloud’s programme for 2002 was cancelled by Silversea, together with certain voyages due to be performed at the end of 2001. I assume therefore (and so far as I can check this seems to be accurate) that the claim under section Ai in respect of the Silver Cloud relates exclusively to passengers who declined to take advantage of the “ Sail and Save” offer and merely accepted return in full of their deposit or other amount paid in respect of their booking. They are thus treated as having effectively cancelled their bookings. I am not at all sure that this is an appropriate analysis. The reality is that Silversea cancelled these voyages. As I shall discuss hereafter, there was no impediment to their performance in terms of the perceived safety of the places due to be visited, most if not all of which were visited by other Silversea ships during the relevant period. Furthermore there is a curious result in terms of double counting. Silversea has an enormous claim under section Aii in respect of the Cloud, albeit it subjects it to what it says is the applicable cap. But in principle Silversea claims US$19.2 million in respect of the Cloud under section Aii, on the basis that it would have expected by 11 March 2002 to have earned or booked revenue in that amount in respect of the Cloud’s 2002 programme. It claims the full amount, because actual bookings were by definition zero as at 11 March 2002, the entire programme having been cancelled. However for the moment I will leave on one side how properly the issue is to be analysed, and will merely resolve the question whether in principle a claim under section Ai is maintainable in respect of the cruises of the Cloud which were scheduled to take place in 2001 and 2002 but did not in the event take place.

It can be therefore be seen at the outset that the claim under section Ai is in respect of either (a) cruises which were performed entirely as scheduled, subject only to a very few minor alterations to schedules which are not alleged to have caused loss or (b) cruises which were not performed because Silversea elected not to perform them, notwithstanding there was no impediment to their performance in terms of safety of the places to be visited or the waterways to be traversed. In the former case the claim is put solely and simply on the basis that a passenger resiling from his announced intention to travel, manifested in the cancellation of booking, amounts to an inference with the scheduled itinerary of the insured vessel thus triggering coverage under either the fifth or the seventh perils enumerated under section Ai. It should be noted that it is not even said, still less sought to be proved, that the level of cancellations on any single cruise rendered it incapable of performance at less than break-even cost. It would on Silversea’s case, subject to deductibles, trigger a claim under section Ai that a single passenger expressed his reluctance to travel in the aftermath of 11 September by cancelling his booking. I have to say that I regard this proposition as quite misconceived. I simply do not regard cancellations of bookings in these circumstances as amounting to interference with the scheduled itinerary of the insured vessel. That there was no interference with the scheduled itinerary of the insured vessel is demonstrated, in all cases other than the Cloud, by the fact that the itinerary of the vessel was performed exactly as scheduled. In the case of the Cloud every scheduled itinerary of the vessel could have been so performed. That is as it seems to me the beginning and the end of the point.

Silversea had two main arguments in support of its suggestion that interference in section Ai must embrace economic interference. The first was that unless that were so, the cover afforded by the policy would be unduly restrictive and of little benefit to Silversea. The second was that the exclusion of “lack of support for any scheduled cruise unless as a direct result of an insured event” indicated, by necessary implication, that lack of support for a scheduled cruise when a direct result of an insured event is intended to give rise to an indemnity under the policy.

I would observe at the outset that neither of these points in my judgment grapples with what I see as the central flaw in Silversea’s approach, which is that what triggers coverage under section Ai is not simply interference but interference “with the scheduled itinerary of the insured vessel.”

That said, the second point is I think the better point, although its main strength derives from an inadmissible consideration. The exclusion makes perfect sense in the context of section Aii, which deals with loss of anticipated income expected to be earned on any future cruise, that is, as I see it, on any cruise subsequent to the one with the scheduled itinerary of which there has been interference, in the shape of some impediment to performance. The cover afforded by section Aii is of some width and generality, speaking in terms of negative impact of bookings or a necessitated change to the scheduled cruise itinerary. Of course there has to be a causal link between the warning and the negative impact or the necessitated change, but it is not surprising to find it spelled out in the face of such apparently unbounded coverage that mere deterioration of market, loss of market or lack of support for a particular cruise does not of itself attract cover – it must be tied directly to the insured event which in the context of section Aii is the warning. It so happens that the self-same exclusion was included in an earlier policy issued by the same insurers through the same brokers to Disney Cruise Line in June 1998, which had no cover equivalent to section Aii. It seems to me that is an inadmissible consideration and that I should construe this policy as a whole as it was agreed by the parties. However I do not consider that the exclusion lacks content in the context of parts Ai and B, as they appear here and as they appear in the Disney policy. Take a case where, as a response to a heightened incidence of terrorist activity in the Middle East, Silversea elects to cancel a voyage which had a preponderance of scheduled calls in the eastern Mediterranean, or in the Red Sea and Gulf of Aden, on account of an altered perception of the safety of the ports. Such a decision might well be taken very shortly before such a cruise was due to sail and might well be preceded by wholesale cancellation of bookings. On either side’s approach to this insurance cover there might be an argument to the effect that there had been a loss of income which could be measured in loss of time, particularly if it proved impossible or impracticable to arrange a substitute itinerary. Cruise credits might well be issued. There might be a claim on the basis that the terrorist activity had in this way interfered with the scheduled itinerary. I believe that the claim would be misconceived, because there would on my example be no actual impediment to performance of the itinerary as scheduled, but I regard it as in no way surprising to find an exclusion which spells out the intended ambit of the cover, making it clear that lack of support for a scheduled cruise does not of itself trigger coverage. It would be otherwise of course if the lack of support stemmed from some inability to perform some aspect of the cruise as a direct result of terrorist activity – as where for example one of the intended ports is closed in consequence of temporary unsafety. That is dealt with by the proviso. Even if I am wrong about that, and wrong to disregard the genesis of the exclusion in a policy with no section Aii cover, still I do not consider that I should be induced by the presence of the exclusion to give to the words of the operative covering clauses a meaning other than that which they naturally bear.

I do not however rest my conclusion on these considerations alone. In my judgment Silversea’s argument proceeds upon a misconception as to the essential structure of the cover. This is encapsulated by its wholly arbitrary treatment of the deductible. Silversea asserts that it is entitled to maintain a claim under section Ai in respect of any cruise which departed more than ten days after 11 September. This is wholly bizarre. It would I accept be only marginally less bizarre if it were suggested that there should be applied to the claim as pleaded a deductible equivalent to ten days at the daily rate, which is what the policy in fact requires. That however would beg the question why a claim which is apparently unconstrained by any agreed daily rate should attract a deductible expressed in terms of so many days at an agreed rate. Perhaps it is in anticipation of that question that Silversea formulated its claim and applied its putative deductible as it did. The combined deductible for sections Ai and B makes clear that the loss which is ranking for indemnity is a loss which can be measured in loss of time, as perhaps Silversea’s first claim in November 2001 recognised. The same concept is endorsed by the fact that there is to be a “Sum Insured,” dealt with on p.1 of the policy separately from the limit of liability, which sum insured is contained in the Schedule which is said to be applicable to section 1.A by which is plainly meant Ai since it expressly does not apply to section I A(ii), where recovery is expressly subject to proof of loss. The daily amount recoverable under section Ai is not expressed to be subject to proof of loss – it is plainly an amount agreed to be recoverable in respect of loss of time attributable to an insured peril. It is not in my view an answer to this point that a cruise ship is unlike a merchant ship and that there is no direct correspondence between time lost and earnings lost. That is precisely why it may be necessary, or overwhelmingly convenient, to have an agreed daily rate. Loss of time on voyage may not immediately translate into loss of revenue from the passengers on board who have paid their fare in advance, although it may result in the issue of cruise credits. But one has only to look at Silversea’s Brochure or Cruise Atlas to see that these vessels are for long periods continuously employed. Loss of time on a voyage may and probably will have a knock-on effect on a subsequent voyage or voyages. In any event time is money in the operation of any vessel. The difficulty in measuring it in the case of a cruise vessel, as opposed to a merchant vessel, simply underscores the convenience of adopting an agreed daily rate.

I find support for this approach in the type of perils enumerated in section Ai, and in the contrast with section Aii. The perils enumerated under the first four and the sixth of the bullet points under part Ai are all plainly perils which are calculated to cause and most likely to manifest themselves in delay or loss of time. The same is true of blockage, closure, capture, seizure and confiscation from which genus the other matters enumerated in the fifth peril must take their flavour. Strikes, riots and civil commotions likewise most often manifest themselves in the shipping context in the form of obstruction or delay, or loss of time generally. The same is probably true of acts of war and armed conflict, and I am inclined to think that the word “directly” is absent in bullet point seven simply because the draftsman had it in mind that acts of war and armed conflict have the capacity, or may traditionally have been thought to have the capacity, to cause interference in a less direct manner than most of the other perils enumerated, in particular the specific perils enumerated in bullet point five, blockage, closure, capture, seizure and confiscation. Thus a vessel may have to sail around a war zone so as to avoid it whereas the operation of most of the other enumerated perils could be expected to be more immediate and direct. Strikes, riots and civil commotions are likely to be fairly localised in their effect, although again it may often be the case that a strike-bound port or one in which there is civil commotion is one which is simply avoided whereas on the whole the specific perils enumerated in the first six bullet points would typically afflict or affect vessels present in a specific area and in their immediate grip. I recognise that this is not a watertight classification, and in particular the more generally expressed perils in bullet point five – any other action or event – have a potentially greater range and ambit, but it is considerations such as these which in my judgment account for the presence of the word “directly” in bullet point five and its absence in bullet point seven. At all events, I do not believe that one can infer from the absence of the adverb “directly” in bullet point seven that the expression “ interfere (s) with the scheduled itinerary of the insured vessel” is in that context to be given a meaning which is different in kind from that which it naturally bears, a fortiori in the earlier context of perils which have so clearly identified a genus as do those in bullet point five.

I would also observe that if Silversea is correct in its construction of section Ai then the policy is in this regard wholly silent as to the manner in which loss should be computed under that part. That would be a surprising conclusion in a policy which provides for a sum insured by reference to a variable daily rate and a relevant deductible expressed in numbers of days. Furthermore since the policy does contain a sum insured which is directly applicable to part Ai, one would need to find another provision which justifies its disapplication in the circumstances of this case. Silversea’s suggestion is that such a provision is to be found in Paragraph 16.6 of the Norwegian Plan. Ordinary canons of construction dictate that the circumstances to which reference is there made must be circumstances of or which attend the incorporation of the Plan into the policy, which circumstances indicate that in certain defined instances the fixed amount per day is not intended to be a dispositive pre-estimate of the recoverable loss. I do not consider that there are any such circumstances – indeed the plain language of the policy seems to me to point in the other direction. On analysis, the circumstances upon which Silversea relies as indicating that the fixed amount per day is inapplicable is that if the fixed amount is the measure of loss then its claim under section Ai will not succeed because no loss of time can be shown. This is a circular or self-supporting argument. A typical example of an incident which would potentially attract indemnity under section Ai would be the need to cancel or curtail a cruise because of a real impediment to its performance. That would lead to the loss of time which otherwise would have been occupied in its performance as scheduled. The fact that the parties have selected a ten day deductible is neither here nor there, but I note in any event that many of Silversea’s scheduled cruises are of more than ten days duration. There is also the potential knock-on effect to which I have referred. Cruise credits might well be issued which would give rise to a further claim under section B. There is no lack of potentially useful cover under section Ai, which incidentally covers perils other than “war risks” in any event, as witness the claim in respect of machinery breakdown on the Silver Cloud. The real oddity of Silversea’s claim under section Ai, as it seems to me, is not that it does not succeed but that it does not relate to any cruise (with the possible exception of the Silver Whisper cruises nos.4107, 4108,4109 and 4110) with which there was any interference in any meaningful sense by reason of the terrorist activity of 11 September.

This brings me to another point which is the relationship between section Ai and section Aii. It seems to me that virtually the entirety of Silversea’s claim under section Ai relates to loss of anticipated income expected to be earned on any future cruise, in other words that it is loss which one would expect to be recoverable, if at all, under section Aii rather than under section Aii. I find an instructive contrast in the language in which sections Ai and Aii of the cover are described on p.3 of the policy. Section Ai is concerned with the loss of income expected to be earned by the operation of the insured vessels. It seems axiomatic, especially when read in the light of the references in the relevant enumerated perils to closure of ports etc., interruption of the vessel’s schedule, damage to a facility at which the vessel is located or intended to be located and interference with the scheduled itinerary of the vessel that the subject matter of section Ai is loss caused by an insured peril giving rise to the inability of the vessel to operate as intended. Section Aii by contrast is concerned with loss of anticipated income expected to be earned on any future cruise. That tends to suggest that this section is concerned with loss suffered on cruises which are future in time to the event which is capable of being a relevant insured peril under section Ai. Of course the insured peril under section Aii is not the same as the relevant (i.e. war or terrorist related) perils in section Ai, but I think the draftsman is likely to have had in mind that any terrorist or similar activity which might give rise to actual interference under section Ai might well be followed up by official warnings of the sort which could have a negative impact on the assured’s bookings for future cruises. I can see no reason why a claim under section Aii should be restricted to bookings which would be expected to be made but which were not in the event made. The language used, “negatively impacts the Assured’s bookings” is as it seems to me sufficiently wide to encompass cancellation of bookings on future cruises as well as expected bookings which do not materialise. I thus do not regard Silversea’s claim under section Ai as relating to its natural subject matter. The natural home for the claim would seem to be section Aii, subject of course to satisfaction of the criteria there laid out.

Mr Flaux QC for Silversea was at times tempted to submit that he derived some assistance from the words “whether actual or threatened” as they appear at the end of the fifth and seventh perils. I am not sure that ultimately he pressed this point but in any event it seems to me that this phrase is plainly adjectival and is descriptive of the type of act, whether of war or of terrorists, whether blockage or closure or seizure or confiscation, which may give rise to interference. I do not understand the submission that the reference to threatened acts is meaningless or otiose on insurers’ construction. A threatened blockage, or threatened terrorist act, is as capable of causing interference with a scheduled itinerary as is actual blockage or terrorist act.

Finally I agree with Mr Swainston QC for insurers that the substitution of the language of section Ai for Paragraph 16.1 of Chapter 16 of the Norwegian Plan which is otherwise incorporated into that part of the insurance is wholly insufficient to displace the basic shape of the cover as is apparent from the other provisions of Chapter 16. Thus under section Ai the insured is entitled to claim in respect of loss of time at the per diem amounts fixed for each vessel. There may be interference properly so called which causes no loss of time, as for example where an itinerary has to be changed without that giving rise to any loss of time, as where a scheduled port of call is omitted without overall loss of time, or where a vessel calls at an alternative port again without overall loss of time. In such a case cruise credits issued to compensate passengers for the loss of the relevant amenity arising out of this interruption to the schedule will be potentially recoverable under section B, but it is hardly surprising that there should be no claim under section Ai.

I should say a further word about the claim in respect of the cancelled Silver Cloud voyages, i.e. those cancelled to enable her to perform the Silver Wind itinerary. Those voyages were not cancelled because of any perception that the places which were to be visited were unsafe. It is true that the Silver Cloud was due to visit some eastern Mediterranean, Red Sea and Gulf of Aden ports in late 2001. The Silver Shadow in fact visited all of these ports only weeks after 11 September. The Silver Cloud was not due to visit the eastern Mediterranean or Africa at all in 2002, although she ended up doing so as a result of assuming Silver Wind’s itinerary. The Silver Cloud was due to spend 2002 in the Far East, Australia and New Zealand. Those places were not perceived to be unsafe on account of terrorist activity. It was the case that passengers joining those voyages from the USA would need to undertake long haul flights which they might be reluctant to do in the aftermath of 11 September. But joining other cruises in other parts of the world could involve flights of comparable or virtually comparable duration. In any event if anything this point merely serves to underscore that there was no interference with the scheduled itinerary of the vessel herself, which is the touchstone. The Silver Cloud schedule was cancelled because, by the time the decision was made, Silversea had concluded that it would have over-capacity in 2002. For present purposes I am asked not to decide and I need not decide whether that over-capacity was something which Silversea had identified prior to 11 September as requiring a reduction to a three ship fleet, whether temporary or permanent or whether it was a consequence of the undoubted downturn in demand following the events of 11 September. However on any view the decision that was made was commercial decision. The decision to select the Cloud as the vessel whose itinerary should be cancelled was informed in part by the following factors: -

scheduled to go into dry dock and, unlike her sister the Cloud, had

not undergone a recent refurbishment;

the Whisper had only just entered service;

difficult itinerary to sell than others and, in the perception of Mr

Leiber, had not been sufficiently marketed in the early part of 2001

although he had hoped to embark upon what he expected to be an

effective marketing programme before 11 September intervened.

That leaves only for consideration Silver Whisper voyages nos. 4107,4108,4109 and 4110. Silversea does not formulate a claim under section Ai in respect of any alleged loss of time on those voyages.

For all these reasons therefore I conclude that Silversea’s claim under section Ai of the policy, other than that in respect of the Silver Cloud mechanical breakdown, is misconceived and should be dismissed, subject only to one point. I have not been addressed on Silversea’s claim for extraordinary expenditure under section Ai which is particularised at p.116 of Trial Bundle A and consists very largely of the additional insurance premium allegedly incurred in respect of the Silver Shadow’s October 2001 Suez Canal transit. Silversea is at liberty to pursue the pleaded claim for extraordinary expenditure but otherwise its claim under section Ai, other than the claim in respect of the Silver Cloud mechanical breakdown, should in my judgment be dismissed.

Part B

In the light of my conclusions thus far it is wholly unnecessary that I should consider the question whether the events of 11 September are properly to be characterised as acts of war or armed conflict. I will merely note that English law would approach this question as one of construction of the policy. In the light of the cover afforded in the relevant section in respect of the consequences of actions by terrorists, and the express reference in section Aii to terrorist activities in a list of perils which includes acts of war and armed conflict, I would be surprised if it were concluded that, for the purposes of deciding whether cover is available under this policy, the events of 11 September fall to be characterised as acts of war or armed conflict when action by terrorists or terrorist activities seems more appropriately to encapsulate what occurred. However that is a matter for another day on which I need express no concluded view and in respect of which,since the facts are not in dispute, I need make no findings.

The claim under Section Aii

In its final submissions Silversea submitted that the words “US$5 million in the annual aggregate and in all” are a reference to insured events not to vessels. I agree with this – it seems to me that the limits and deductibles in this policy all naturally refer to events or accidents or occurrences, save where it is otherwise expressly provided. It is otherwise expressly provided only in the case of section Ai, where the provision on page 5 under the rubric “Limitations and Deductibles” cross refers to the “ attached days per event” and to the daily amount in the sum insured schedule which is attached. In the absence therefore of any wording which suggests that the limit applicable to section Aii is, contrary to the overall shape of the policy, intended to apply separately to each vessel, I would unhesitatingly conclude that the limit is obviously intended to be a single limit irrespective of the number of vessels insured under the policy.

My conclusion that that is the correct approach is borne out by the nature of the cover offered under section Aii. This focuses, as I see it, not upon the operation of the individual vessels as does the cover under section Ai but upon the Assured’s business as a whole. The “Cruise Atlas” to which reference is made in the introduction to this cover on p.3 of the policy is obviously Silversea’s contemporary brochure. I would expect the ascertained net loss suffered by Silversea under section Aii to be computed on a fleet-wide basis – indeed I do not see how it could in principle be done any differently. Suppose that a relevant warning or warnings has negative impact only upon actual or prospective bookings for one of the vessels because it alone was due to travel to the region affected by the warning. Suppose also that all of the passengers who elect not to travel on the vessel in question transfer their bookings or make bookings on other ships in the fleet, causing those vessels to achieve load factors substantially in excess of what was forecast. I do not believe that the individual owning company of the “unpopular” vessel could maintain a claim without there being brought into account the enhanced revenue earned by the other individual ship owning companies, always assuming of course that revenues are in fact separately apportioned within the Group accounts. On any view credit would have to be given for an unexpected booking on vessel A which was made in substitution for a “lost” booking on vessel B. Any other approach would as I see it run counter to the concept of ascertained net loss and is simply and obviously not what is intended. I think that the reference to “the Assured’s bookings” is a reference to Silversea’s bookings across the fleet as a whole. Likewise “a change to the scheduled cruise itinerary” is a reference to the overall planned itinerary of all the insured vessels, not a reference to the itinerary of a single vessel. Furthermore the deductible is a single deductible per occurrence, not a deductible per vessel. In his final address Mr Flaux for Silversea was inclined to accept, although he did not quite commit himself, that the cruise credit limit, expressed as US$5million per event, would preclude Silversea from recovering more than US$5 million overall in respect of cruise credits issued to passengers on different ships in consequence of a single event. In my judgment his inclination was correct. Furthermore, in my judgment the US$250,000 per occurrence language in the deductible to section Aii points in the same direction.

In short I have no doubt that the natural reading of the policy is that in respect of section Aii cover there is a single annual aggregate limit of US$5 million in respect both of all losses arising out of a single event or more accurately occurrence, that being the language of the deductible, and in respect of all losses, whether arising out of one or more events. The losses attributable to a single event or occurrence must be aggregated and a deductible applied to them. No more than US$5 million may be recovered in respect of any one occurrence or in all in respect of howsoever many occurrences there may be during the term of the cover. I also doubt if the draftsman intended to draw any distinction between event and occurrence.

Mr Flaux submitted that the result that there is a separate US$5 million limit applicable to each vessel is achieved by the provision in the General Terms and Conditions “Each Vessel to be a Separate Insurance.” It may be of some significance that it is not said that there shall be a separate policy for each vessel. In any event I see no reason why there should not be three, ultimately four separate contracts of insurance effected under the umbrella of a single policy document pursuant to which it is clear, as it is here, that the maximum recovery under one head of cover shall be US$5 million under howsoever many individual contracts claims may in fact be brought. I do not therefore consider that this point adds anything of substance. This provision does not, as does the Schedule to section Ai, make clear that there is to be a separate limit applicable to each vessel. I note incidentally that the Schedule to section Ai has the effect that the limit for each vessel is different.

However I also consider that Mr Flaux seeks to put far too much weight onto a provision the purpose of which is clear and which has nothing whatsoever to do with the amounts insured, limits or deductibles. The separate insurance provision is obviously intended to address a problem often encountered in practice where different vessels in a fleet have different owners and, more significantly, different mortgagees to whom has been assigned the benefit of the insurance. The clause prevents underwriters using against claimant owner or claimant mortgagee A claiming indemnity in respect of vessel A rights of set off in respect of unpaid premium due in respect of a different vessel B and due therefore from a different owner. It will also often prevent non disclosure, misrepresentation or breach of warranty affecting only one vessel being relied upon by underwriters so as to avoid cover in respect of another vessel to which the non disclosure, misrepresentation or breach of warranty was not relevant. I regard as fanciful the suggestion that by this means a plainly intended overall limit of US$5 million across the fleet as a whole can be transformed into an individual limit of US$5 million per vessel. The treatment of limits in section Ai shows that the draftsman of the policy knew how to use clear language to achieve separate limits for each vessel. I would add that Mr Olsen in his evidence in chief said that the words “Each vessel to be a Separate Insurance” are routinely inserted by brokers into marine policies covering fleets of ships, and that the whole purpose of their inclusion is to prevent underwriters withholding payment of a claim in respect of one vessel where premium is outstanding in relation to another vessel in the fleet, in circumstances where vessels in a fleet may often be separately owned or have different mortgagees. Whilst I am not sure whether I have seen these words in fleet slips before (I may well have done) such is the prevalence of the problem which he identifies that I would not have needed Mr Olsen’s evidence to realise that this is one of the points to which the words are directed. For what it is worth Mr Olsen was not cross examined on this evidence nor on his assertion that these words do not have any relevance to how the limits stated in the policy are to be applied.

In the light of my conclusion that section Aii attracts a single limit I do not need to consider the insurers’ contention that the policy should if necessary be rectified so as to reflect the true and common intention of the parties that there should be a single overall limit applicable to all four vessels combined. In that regard I was reminded of the most recent statement of the principles to be applied, to be found in the judgment of Lord Phillips MR in The Demetra K 2002 EWCA Civ 1070; 2002 2 Lloyd’s Rep 581. What is needed is an antecedent agreement, which may be spelled out of an outward expression of a common intent in a manner which makes plain what is intended to be achieved by the written document which is to be produced in order to encapsulate the agreement. The insurers have failed to demonstrate an outward expression of a common and articulated intent that there be a single limit, but the reason why they have failed is that it is so obvious that that it is what the parties intended, that they did not need to spell it out or to discuss it. The underwriters have succeeded in demonstrating that the parties were ad idem that there should in respect of Aii cover be a single flat rate premium. This is entirely consistent with the agreed extent of the cover purchased thereby being a single global amount, as opposed to an amount available three times over for so long as there were only three ships insured and then four times over once the Silver Whisper joined the cover. Had rectification remained a live issue the critical question would, I think, have been whether the agreement as to a single flat rate premium for Aii cover is consistent only and necessarily with the Aii cover being for a single global amount, so that the parties’ agreement as to a single flat rate premium, and the manner in which they expressed it, the one to the other, necessarily carries with it the expression of an intention that there should be a single combined global limit applicable to all the ships insured under the policy.

The possibility of Vesta replacing Silversea’s then insurers was first discussed in July 1999. Discussion about a cover going beyond that afforded by the Disney policy dates back to September 1999. At first Mr Macmillan-Bell of Berkely, Silversea’s brokers, suggested to Mr Olsen, then Vesta’s underwriter, that what was required was cover for exposure which might arise from an event that might not necessarily result in physical damage to a ship but might lead to future itineraries being revised. Cover was sought in respect of reservations cancelled as a result of lack of interest in the revised itinerary offered in substitution therefor. At some stage, possibly at a meeting which Mr Olsen had with Mr Bajaj in November 1999, Mr Macmillan-Bell suggested a separate section with a separate limit, US$3 million and a separate deductible, US$50,000. Following further discussions, on 14 December 1999 Mr Macmillan-Bell proposed a wording which is in most respects identical to that which became section Aii. The same proposal provided in terms: -

respectively;

Following further discussions, on 17 March 2000 Mr Olsen responded, proposing for the suggested “Anticipated Earnings” cover as per Mr Macmillan-Bell’s last submitted wording; “Limit US$5 million in the Aggregate. Deductible US$500,000 in the Aggregate. Premium US$100,000 for all vessels on risk during a 12 months period.”

There followed on 22 March 2000 a telephone conversation between Mr Macmillan-Bell and Mr Olsen’s assistant, Mr Brian Cheney. Mr Macmillan-Bell wanted to reduce the aggregate deductible under section Aii to US$250,000, to which Mr Cheney responded with the suggestion that the premium for that section should in consequence be doubled to US$200,000.

In the event Silversea renewed with its existing Italian underwriters in 2000. The reasons do not matter although it seems to have been as a result of Mr Macmillan-Bell and Mr Bajaj having insufficient opportunity to discuss Vesta’s proposals rather than any dissatisfaction with them.

At the end of February 2001 Mr Macmillan-Bell revived the negotiations which began precisely where they had left off the previous year. By now, as anticipated in the previous negotiations, the third vessel Silver Shadow had come on risk during the year covered by the expiring insurance and it was in prospect that the fourth vessel, Silver Whisper, would come on risk during the ensuing policy year. This notwithstanding, there was no discussion of any necessity to change the section Aii limit, deductible or premium. The only relevant adjustment which did take place was that in order to assist Mr Macmillan-Bell in securing the order from Silversea Mr Olsen agreed to reduce the premium for Aii which would be payable up front from US$200,000 to US$175,000, with the balance of US$25,000 only payable by Silversea when the Silver Whisper joined the fleet. It would appear that the initiative for the proposal may have come from Silversea. This split in the premium is set out in a Schedule sent by Mr Di Mario of Silversea to Mr Macmillan-Bell on 15 March 2001. The identical Schedule was sent by Mr Macmillan-Bell to Mr Olsen on 30 March 2001 and is the basis upon which on 2 April 2001 Mr Olsen for underwriters agreed to be bound with effect from 1 April 2001.

This Schedule is instructive. It sets out the detailed calculation of separate section Ai premiums for each vessel, by reference to weighted average revenue. The rate for section Ai (and section B) was 1.43%. In relation to section Aii however the Schedule provided for a premium of US$175,000 for the Cloud/Wind/Shadow and US$25,000 for the Whisper. The three ship policy document to which I have already referred which was signed shortly thereafter by Mr Olsen has the provision for a section Aii premium of US$175,000 as set out earlier in my judgment.

Thereafter a series of errors were made which are in my judgment wholly irrelevant to any question of rectification. The wording and the premium structure had been agreed in the manner which I have described. However details of the insurance were entered into Gard’s internal records in a manner which did not in fact reflect the true terms of the agreement. Thus on 23 April 2001 Ms Kirsti Liborg, an underwriting assistant who had had no involvement in the writing of the risk, produced a “ Premium/Rate Schedule – Rating Analysis” which attributed to each of the three vessels under section Aii a sum insured of US$5 million. The document also recorded that for this cover the rate on the Cloud was 0.91%, on the Shadow 0.945% and on the Wind 1.645%, producing an arithmetical aggregate of US$175,000. This was plainly nonsense, and was almost certainly brought about by the requirements and shortcomings of the computer software. Other internal documents were generated by underwriters with equally obvious and in some cases wholly absurd errors, although I need not take time setting them out here. One error of more significance was made in preparing the accounting documentation pursuant to which Silversea could be invoiced. It is not entirely clear to me whether this error should be attributed wholly to underwriters or whether the brokers may have contributed to it. However that they be, for the purpose of calculating the quarterly premium instalment due the overall premium of US$175,000 initially due in respect of section Aii was divided equally between the three ships then on risk, providing an annual premium attributable to each vessel of US$58,333.33. This error was followed through into a Cover Note issued by the brokers on 14 June 2001 to reflect the addition to the cover of the Silver Whisper as from 24 June 2001. The section Aii premium for the Whisper was said to be US$58,333,33 per annum and pro rata. The error was carried through into the four ship policy document which Mr Olsen subsequently signed, possibly on 20 June 2001, without noting the error. The section Aii premium was here expressed to be “US$58,333,33 per annum and pro rata,” which was not of course what was agreed. Silversea was invoiced for Silver Whisper Aii cover on this incorrect basis, no one noting the error at the time. The error came to light only in May 2002. Silversea’s response to its revelation reflects no credit on it nor on Mr Bajaj. On 13 May 2002 Mr Macmillan-Bell telephoned Mr Bajaj and explained what had occurred. He had discovered the error whilst looking at the Silver Whisper documentation in order to deal with the fact that although the original cover was for 12 months, no per diem rates were agreed for the period 16 – 31 March 2002. The point arose also in respect of a requested extension of the cover. According to Mr Macmillan-Bell’s contemporary note of the telephone conversation Mr Bajaj agreed that the premium for section Aii had been US$200,000 flat for all vessels. Mr Macmillan-Bell said that the brokers would credit the incorrectly invoiced amount on the Silver Whisper against the premium for a three month extension of the cover which had recently been negotiated and agreed. Mr Macmillan-Bell followed this up with a letter dated 16 May 2002 which introduced this topic with the word “as discussed” and enclosed an appropriate endorsement and credit note to rectify the accounting position. Mr Macmillan-Bell and Mr Bajaj spoke again on the telephone on 29 May 2002. Evidently Mr Bajaj had by now consulted Silversea’s lawyers. One of the issues already under discussion between underwriters and Silversea was the question whether in respect of section Aii cover there was a US$5 million limit per vessel or a limit of US$5 million for the entire policy. In fact as early as 27 September 2001, in response to Silversea’s 21 September 2001 notification in entirely general terms of a potential claim, Gard’s Claims Director Kitty Kendel sent to Mr Macmillan-Bell a memorandum which made the point, quite unprompted by any suggestion to the contrary, that under section Aii there is an annual aggregate of US$5 million for all incidents all vessels. Evidently Silversea’s lawyers and Silversea now saw the possibility of taking advantage of the hitherto unspotted accounting error in order to bolster the argument that section Aii attracted four separate limits, the approach no doubt being that that argument was supported if could be suggested that the original agreement had in truth been for a separate premium for Aii cover payable in respect of each vessel, that separate premium being, presumably, US$58,333,33. Thus in this second telephone conversation Mr Bajaj, according again to the contemporary note made by Mr Macmillan-Bell, “advised that he was not able to comment and, as such, his instructions were to disregard this endorsement.” The evidence of Mr Bajaj when he was asked about this was I am afraid unsatisfactory and evasive. In my judgment Mr Bajaj knew perfectly well from the documentation available to him, both when speaking to Mr Macmillan-Bell and when giving evidence, that the agreement reached had been for a single flat premium of US$200,000, of which US$ 25,000 became payable only when the Silver Whisper came on risk. His suggestion that, in the light of Silversea having been invoiced US$58,333.33 pro rata, he did not know what ultimately had been agreed between his brokers and the insurers, was opportunistic. I infer that the reason why Mr Macmillan-Bell was not called by Silversea to refute Mr Olsen’s evidence as to the agreement which they had made was that Mr Macmillan-Bell would not only not have refuted it but would also have said that it was inherent in that agreement that there was to be a single limit of cover under section Aii. No doubt he would have said that that is what he thought he was agreeing.

As I have already indicated the question the answer to which might otherwise have been critical is in fact whether a common express intention that there should be a single limit can be spelled out of what was said expressly on the question of a single premium. There is no evidence of any discussion as to the number of limits under section Aii although that is, as I have already indicated, in my judgment likely to be because it never crossed the mind of Mr Olsen, Mr Cheney or Mr Macmillan-Bell that anything more than a single overall policy limit was under consideration. Nonetheless I have on reflection concluded that, by their exchanges on the topic of the single premium for section Aii, which was in such sharp contradistinction to their approach on section Ai, the underwriters and the brokers gave outward expression of a common albeit unstated intention that this should reflect a single policy limit in a manner which made it plain, applying an objective test, that this was what they wished to achieve by the policy. Had it been necessary therefore I would have ordered rectification of the policy to reflect that agreement. No doubt that represents a conclusion at the extreme margin of what would ordinarily be achievable by way of rectification. However insofar as it may seem surprising that rectification could be achieved in respect of a matter that was not expressly discussed, it must of course be remembered that in fact the question does not arise precisely because it is in my judgment clear from the written document that the parties intended there to be a single limit under section Aii. In that regard the written document reflects the antecedent discussions.

Underwriters said in opening and repeated in closing that if the court decides that the limit under section Aii applicable to this claim is US$5 million, no assessment of quantum would be necessary as underwriters would be content to pay that limit. That is a very realistic approach, and in the light thereof it is strictly unnecessary for me to resolve any further issues. However I am unsure whether underwriters intended in any event to maintain their pleaded case on the application of the deductible to the claim under section Aii. This was to the effect that, “ a deductible of US$250,000 applies in respect of each instance of negative impact on bookings for cruise(s) by reason of a particular warning and/or changes to the itinerary of the vessel(s) necessitated by reason of a particular warning (“per notice”). I propose therefore to address this point and two other issues which were argued, on one of which I also heard expert evidence.

The application of the section Aii deductible.

Underwriters’ pleaded case was that any diminution in business after the 11 September attacks was attributable either wholly or in overwhelming part to reaction to the attacks themselves, rather than to any official warnings issued in their aftermath. In support of this case underwriters relied on expert opinion evidence from Dr Brian Gibbs, currently Visiting Associate Professor of Management Science in the Sloan School of Management at the Massachusetts Institute of Technology. His opinion was to the effect that the post 9/11 deterioration in demand for Silversea cruises was caused primarily, as to 80-90% by the terrorist attacks themselves and only to a much lesser degree, 10-20%, by the State Department Advisories and similar warnings.

From the outset this approach struck me as unreal and having heard evidence on the point not just from Dr Gibbs but also from Silversea’s expert Dr Reddy I am confirmed in that view. It is simply impossible to divorce anxiety derived from the attacks themselves from anxiety derived from the stark warnings issued in the immediate aftermath thereof. In relative terms very few people will have had any knowledge of the attacks apart from what they learned of them from media reporting. Of course, images of the aircraft flying into the twin towers will have had a profound impact, but few people will have watched coverage of that sort without also being exposed to the warnings and the media exposition of the warnings which swiftly followed. Part of the media coverage of 11 September was the dissemination to the American public of warnings from the United States Government and other responsible authorities. Dr Gibbs for the purpose of his analysis treated media attention to, by which in context he meant coverage of, the attacks as part and parcel of the attacks themselves. In assessing causal impact he lumped in media coverage of the warnings with the warnings themselves. He accepted that it would have been a difficult assignment to consider 9/11 divorced from the media coverage of 9/11 – “tricky empirically but not [logically] impossible.” However I think that the logic which compelled that conclusion similarly compels the conclusion that it is impossible to divorce the effect of the warnings from the effect of the events which they so swiftly followed. Furthermore I am not sure that I understood Dr Gibbs’ reference to difficulty of an empirical nature since he did not base his conclusions upon the results of experimental studies, other than in the most general sense. Dr Gibbs acknowledged that there was undoubtedly an interactive effect between, on the one hand, the attacks and on the other hand, the warnings. Notwithstanding that interaction he thought it possible to separate out the different factors and to assign to them a different weight in terms of their impact on decision making. Dr Reddy, a chartered clinical and occupational psychologist of unrivalled relevant experience, thought that this was not possible and with genuine respect for his training and his experience I do not consider that I really needed his careful evidence to lead me to the same conclusion. I am also I am afraid unable to regard the attribution of relative causal effect in percentage terms as anything other than arbitrary. Dr Reddy could not understand how those percentages could be derived and nor can I. Dr Reddy pointed out that he and Dr Gibbs work in different parts of the forest. I am afraid that I found Dr Gibbs’ part of the forest wholly impenetrable. Like Dr Reddy I acknowledge that Dr Gibbs’ approach may have an important part to play in assessing human performance in a consumer context, but I am not convinced that it is appropriate when considering reaction to a traumatic event. At all events Dr Reddy’s experience is in that context unrivalled and I prefer his approach.

I also note in passing that since, as I find, and as was common ground between the two experts, the events of 11 September and the warnings were concurrent causes of the downturn in bookings, including cancellations thereof, and since the consequences of the events of September 11 are not for the purposes of section Aii excluded from the ambit of the cover, as opposed to being simply not covered, a claim under the policy must lie – see Wayne Tank and Pump Company Ltd v Employers Liability Assurance Corporation Ltd 1974 1 QB 57. I am not sure that, on this hypothesis, insurers contend to the contrary.

The final point with which I must deal is the question whether Silversea’s approach to the quantification of its ascertained net loss for the purposes of its claim under section Aii is appropriate. I have I hope already sufficiently indicated that in my judgment it is not, not least because I regard loss derived from cancelled bookings as in principle recoverable under section Aii rather than under section Ai under which they are presently claimed. However Silversea’s approach is I believe inappropriate for other reasons also. An ascertained net loss is, whatever else it might be, an actual loss, here to be calculated by reference to a finite period of six months. Silversea’s calculation would claim from underwriters as an actual loss the shortfall in the value of bookings which customers might ordinarily be expected to have made by the end of, say, February 2002 for cruises departing in, say, October 2002, notwithstanding it might be demonstrable by the time of departure of the cruises that the same load factors had been achieved as were reliably forecast before the events of 11 September. I do not suggest that Silversea did in fact achieve equivalent load factors – that remains to be examined should it prove necessary. The example merely serves to demonstrate what I see as a flaw in the method. Nor do I consider that Silversea can derive assistance, in relation to the six month limit, from the words “expected to be earned on any future cruise” used earlier in the policy. It seems to me that the six month limit inserted into the body of section Aii is simply intended to be an arbitrary, but agreed, cut off point. Of more importance is the natural meaning of the expression ascertained net loss. The natural meaning of the two parts of the policy in which these different expressions appear is, when read together, that Silversea may recover the difference between the net income which it could have expected to have earned over a six month period and the net income which it in fact earned over that period, subject of course to proof of the causal link required by section Aii. The accounting convention that income referable to a cruise is deemed earned on the day of sailing overcomes any potential problem of attribution in respect of cruises which span the beginning and end points of the nominated six month period. The calculation takes into account savings consequent upon the carriage of fewer passengers, broadly in the way in which Silversea has in fact done for the purpose of its section Ai claim – I say nothing of course about the detailed figures.

It follows that I consider the basis upon which Silversea’s Aii claim is formulated to be misconceived. Silversea has however as an alternative to its primary case under section Ai sought to recover under section Aii its loss due to cancelled bookings. Insofar as those cancellations relate to bookings on cruises due to depart between 11 September 2001 and 11 March 2002 that alternative case is in my judgment in principle well-founded. Whether Silversea should be permitted to add to that a claim in respect of unmade bookings on cruises departing within the relevant six month period is another question. No such claim has as yet been put forward. Any such claim would be likely to relate very largely if not exclusively to cruises departing between 1 January 2002 or thereabouts and 11 March 2002 since it is acknowledged that as at 11 September 2001 Silversea could have expected only insignificant further bookings in respect of cruises departing over the balance of the year 2001. In view of my conclusion on the applicable US$5 million limit I suspect that any such further claim is academic.

I will hear Counsel on any applications arising out of my judgment, including the question what if any directions need to be made in order to permit the resolution of any relevant outstanding issues including issues of detailed quantum.