Gard Marine and Energy Limited v China National Chartering Company Limited and another

This appeal arises out of the grounding of the OCEAN VICTORY (“the vessel”) in the port of Kashima in Japan on 24 October 2006. She was a Capesize bulk carrier, built in China in 2005. By a demise charterparty dated 8 June 2005, the vessel’s owners, Ocean Victory Maritime Inc (“OVM” or “the owners”), chartered the vessel to Ocean Line Holdings Ltd (“OLH”), which is or was a related company, on the widely used Barecon 89 as amended. On 2 August 2006, OLH time chartered the vessel to China National Chartering Co Ltd (“Sinochart”) and on 13 September 2006, Sinochart in turn sub-chartered her to Daiichi Chuo Kisen Kaisha (“Daiichi” or “the charterers”) for a time charter trip.

The demise charterparty and both time charterparties contained an undertaking (on materially identical terms) to trade the vessel between safe ports. On 12/13 September 2006, Daiichi (and thus Sinochart) gave the vessel instructions to load at Saldanha Bay in South Africa and to discharge at Kashima. Between 19 and 21 September she loaded 170,000 tonnes of iron ore. She arrived off Kashima on 20 October and discharge began that afternoon.

The port of Kashima is entered from the sea through the northern end of a specially constructed channel known as the Kashima Fairway, which runs almost due north south, and is the only route into and out of the port. The Kashima Fairway is bounded on one side (to the east) by the South Breakwater and on the other (to the west) by the land.

On 24 October the vessel sought to leave the port during a storm. However, she allided with the northern end of the South Breakwater and grounded. Shortly thereafter another Capesize vessel, the ELIDA ACE, also grounded in the Kashima Fairway while attempting to leave the port. Salvors were engaged but the OCEAN VICTORY eventually broke in two. Her wreck was subsequently removed in the course of a lengthy wreck removal operation.

Some two years later, on 15 October 2008, Gard Marine & Energy Ltd (“Gard”), one of the vessel’s hull insurers at the time of her loss, took assignments of the rights of OLH and OVM in respect of the grounding and total loss of the vessel. In its capacity as assignee of those rights, Gard subsequently brought a claim against Sinochart (which Sinochart passed on to Daiichi) for damages for breach of the charterers’ undertaking to trade only between safe ports.

On 30 July 2013, Teare J (“the judge”) held that the casualty was caused by the unsafety of the port in breach of the safe port undertaking in the time charters. He awarded Gard substantial damages, namely the agreed value of the vessel (US$88.5m), damages in respect of liability for SCOPIC expenses (US$12m), damages for wreck removal expenses (US$34.5m) and damages for loss of hire (US$2.7m).

Permission to appeal to the Court of Appeal on certain specific issues was granted. On 22 January 2015, the Court of Appeal (Longmore, Gloster and Underhill LJJ) allowed the appeal and set aside the judgment of the judge on the grounds that the conditions which affected Kashima were an abnormal occurrence and that there was no breach of the safe port undertaking on the part of the charterers. The Court of Appeal also held that, in the light of the insurance provisions of the demise charterparty, OVM and OLH (and Gard as their assignee) were not entitled to claim in respect of losses covered by the hull insurers. On 20 May 2015 Gard were granted permission to appeal to this Court.

The parties agreed the issues in this appeal as follows.

The events which led to the grounding and subsequent loss of the vessel are summarised in paras 127 and 128 of the judge’s judgment as follows:

In the Court of Appeal Longmore LJ (giving the judgment of the court) noted in para 14 that it was common ground between the parties that, if the damage sustained by the vessel at Kashima on 24 October 2006 was caused by an “abnormal occurrence”, then the charterers would not have been in breach of the safe port warranty. That common ground was based on the classic dictum of Sellers LJ in The Eastern City [1958] 2 Lloyd’s Rep 127, 131 that:

Longmore LJ added in para 15 that what was in dispute between the parties on the appeal in relation to this issue was (i) what, as a matter of law, was the correct test for an abnormal occurrence; (ii) in particular, was the judge correct to hold (in paras 110, 127-128, 132 and 134 of his judgment) that the combination of two weather conditions on the casualty date (namely the phenomenon of swell from “long waves”, which might have forced the vessel to leave the berth, and a very severe northerly gale which meant that the vessel could not safely exit the port) was not to be characterised as an abnormal occurrence, notwithstanding that the coincidence of the two conditions was “rare”, because both conditions were physical characteristics or attributes of the port; and (iii) on the facts as found by, or undisputed before, the judge, did the weather conditions on the casualty date amount to an abnormal occurrence? It is important to note that it was not submitted that the relevant test could or should be other than that described by Sellers LJ in The Eastern City. In any event that test has stood the test of time. The question is what is meant by “an abnormal occurrence”.

The Court of Appeal summarised the charterers’ case in para 44 of their judgment as follows:

It was not in dispute that the question whether the port was unsafe must be tested as at the moment that the charterers instructed the owners to proceed to it. It was submitted on behalf of the charterers that the appropriate test was whether a reasonable shipowner trading the ship for his own account and knowing the relevant facts would decline to proceed to the nominated port. That is essentially the test set out in the Court of Appeal’s summary of the charterers’ case in sub-para 44(ii) above.

To my mind the key points in this appeal are to be found in sub-paras 44(i), (ii), (iii) and (iv). It is important to note that the test is not whether the events which caused the loss were reasonably foreseeable. Reasonable foreseeability is a well known test in some parts of the law of tort, notably negligence and remoteness of damage. The courts could well have adopted such a test but they have not done so. Instead they have asked whether the relevant event was an abnormal occurrence.

What then is meant by “abnormal occurrence”? The question is whether it has the meaning proposed by the charterers and set out in para 44(iii) quoted above or the meaning proposed by Gard in para 66 of its case as follows:

I would accept the charterers’ submission recorded in sub-para 44(iii) that an ‘abnormal occurrence’ has its ordinary meaning. It is not a term of art. As stated in that sub-paragraph,

We were referred to a number of cases which seem to me to support that conclusion. Sellers LJ’s famous dictum quoted above was taken from the judgment of Morris LJ in The Stork [1955] 2 QB 68, 105, which was a time charterparty case. See also Reardon Smith Line Ltd v Australian Wheat Board (The Houston City) [1956] AC 266, which was a voyage charterparty case to which essentially the same principles were applied. In Kodros Shipping Corpn v Empresa Cubana de Fletes (The Evia) (No 2) [1983] 1 AC 736, which was another time charterparty case, Lord Diplock said at p 749 that he regarded the nature of the contractual promise by the charterer in what he called the safe port clause as having been well settled for a quarter of a century at the very least. He referred specifically to the expression “abnormal occurrence” used by Sellers LJ in The Eastern City which he said reflected the previous statement of Morris LJ in The Stork.

At p 749H, Lord Diplock referred to the distinction between damage sustained by a particular vessel in a particular port on a particular occasion caused by an “abnormal occurrence” and damage resulting from “some normal characteristic of the particular port at the particular time of year”. He added that there were dangers that judges of first instance sometimes omitted important qualifications. He was referring (at p 750A-B) to what he called the heresy that, in the previous decade or so, had been embraced by judges in the commercial court culminating in that of Mustill J in Transoceanic Petroleum Carriers v Cook Industries Inc (The Mary Lou) [1981] 2 Lloyd’s Rep 272, 277.

Lord Diplock agreed with the judgment of Lord Roskill which had been prepared with the assistance of Lord Brandon. The heresy to which Lord Diplock referred was that identified by Lord Roskill at p 757. It arose in the construction of what he described as these eight words in clause 2 of the relevant time charterparty: “The vessel to be employed … between … safe ports …” Those were essentially the same words as in the charterparties in the instant case. The heresy identified by Lord Diplock was the conclusion of Mustill J in The Mary Lou and, indeed, of Robert Goff J in The Evia (No 2) that there was an absolute continuing contractual promise that at no time during her chartered service would the ship find herself in any port which was or had been unsafe for her: see Lord Roskill at p 756G.

On p 757 Lord Roskill gave his reasons for rejecting that approach as a matter of construction of the charterparty. In particular, he said at p 757D that a charterer will exercise his contractual right to require the shipowner to carry out his contractual obligations by giving the shipowner orders to go to a particular port or place of loading or discharge. He added that it was clearly at the point of time when that order is given that the contractual promise to the charterer regarding the safety of that intended port or place must be fulfilled.

Lord Roskill then said this at p 757E:

Having expressed those views as a matter of construction of the charterparty, Lord Roskill analysed the cases and concluded that they strongly supported the views he had expressed. The cases included Ogden v Graham (1861) 1 B & S 773 and GW Grace and Co Ltd v General Steam Navigation Co Ltd (The Sussex Oak) [1950] 2 KB 383 in addition to The Stork, The Houston City and The Eastern City. He noted in particular the dissenting judgment of Sir Owen Dixon CJ in the High Court of Australia in The Houston City which was accepted as correct by the Privy Council.

Lord Roskill paid particular attention at p 760 to the reference to “some abnormal occurrence” by Morris LJ in The Stork, which, as he put it, was the foundation of the similar view of Sellers LJ in The Eastern City as described above. Like Lord Diplock, Lord Roskill emphatically adopted that approach as correct. As I read the remainder of his judgment, in which he refers to a number of other cases, they are all to substantially the same effect.

In all these circumstances, I would accept these three submissions made on behalf of the charterers arising out of the cases referred to above and, in particular The Evia (No 2) in the House of Lords. First, the date for judging breach of the safe port promise is the date of nomination of the port. A safe port promise is not a continuing warranty. Second, the promise is a prediction about safety when the ship arrives in the future. These propositions are not in dispute. As I see it, such a promise necessarily assumes normality; given all of the characteristics, features, systems and states of affairs which are normal at the port at the particular time when the vessel should arrive, the question is whether the port is prospectively safe for this particular ship. If the answer is “yes unless there is an abnormal occurrence”, the promise is fulfilled. As Robert Goff J said at first instance in The Evia (No 2) [1981] 2 Lloyd’s Rep 613, 621, “the formulation of a test whether the port is unsafe must assume normality, and must therefore exclude danger caused by some abnormal occurrence.”

I would further accept the third submission made on behalf of the charterers that on the authorities to which I have referred, safe port disputes should be reasonably straightforward. Was the danger alleged an abnormal occurrence, that is something rare and unexpected, or was it something which was normal for the particular port for the particular ship’s visit at the particular time of the year?

I would also accept the submission that this approach, that is the approach in The Eastern City as elaborated by Lord Roskill and Lord Diplock in The Evia (No 2), provides a coherent allocation of risk between the various interests as follows. The owners are responsible for loss caused by a danger which is avoidable by ordinary good navigation and seamanship by their master and crew. The charterers are responsible for loss caused by a danger which was or should have been predictable as normal for the particular ship at the particular time when the ship would be at the nominated port and was not avoidable by ordinary good seamanship. The owners (and ultimately their hull insurers) are responsible for loss caused by a danger due to “an abnormal occurrence”. As Lord Roskill put it at p 757E quoted above, charterers are not insurers of “unexpected and abnormal risks”. On the contrary, the charterparty terms require owners to take out hull insurance (as they will invariably do) which is their protection against rare and unexpected events. On the charterers’ case on the facts, the characteristics of the port were such that the ship was prospectively safe, but they unexpectedly combined in a critical way such as to create an exceptional, and apparently unprecedented danger. This was within the letter and spirit of Lord Roskill’s description of an unexpected and abnormal risk.

It is to my mind important to note the emphasis in the cases upon the meaning of the expression “abnormal occurrence”. I would accept the charterers’ submission in para 44(iii) of the Court of Appeal’s judgment that “abnormal” is something well removed from the normal. It is out of the ordinary course and unexpected. It is something which the notional charterer or owner would not have in mind.

In short, I would accept the charterers’ submission that the first question is whether a reasonable shipowner in the position of the particular shipowner trading the ship for his own account and knowing the relevant facts would proceed to the nominated port. If the answer is “yes unless there is an abnormal occurrence”, the port is prospectively safe for the particular ship and the promise is fulfilled. In a case where the vessel suffers loss or damage, a second question arises, namely whether there was damage caused by an abnormal occurrence as defined above.

This contrast can be seen in the judgment of the Court of Appeal. Having summarised the essential facts at para 48, in para 49 the Court of Appeal described the core of the judge’s reasoning at paras 127-129 of his judgment thus:

In paras 50ff the Court of Appeal embarked upon a critique of the judge’s approach. In para 50 they set out their conclusion that the judge’s approach was flawed. They then referred in detail to the speeches of Lord Diplock and Lord Roskill in The Evia (No 2). In the speech of Lord Roskill they highlighted in bold the passage at p 757E which I quoted at para 21 above. It is to my mind important that that passage includes the proposition that where the characteristics of a port make the port prospectively safe, Lord Roskill did not think that

Immediately after quoting an extensive passage from the speech of Lord Roskill in The Evia (No 2), most of which is quoted above, in para 52 the Court of Appeal said that its import was clear. They said that charterers do not assume responsibility for unexpected and abnormal events which occur suddenly and which create conditions of unsafety after they have given the order to proceed to the relevant port. They are the responsibility of the ship’s hull insurers (if owners have insured) or of owners themselves. The Court of Appeal further noted in para 52 that the concept of safety is necessarily not an absolute one. They did so by reference to the decision of the Court of Appeal in The Saga Cob [1992] 2 Lloyd’s Rep 545, 551, where, in the context of political risks, Parker LJ, giving the judgment of the court, said this:

In the instant case the Court of Appeal, in my opinion correctly, held (at para 53) that a similarly realistic approach should be adopted to the determination of what it called the essentially factual question whether the event giving rise to the particular casualty is to be characterised as an “abnormal occurrence” or as resulting from some “normal” characteristic of the particular port at the particular time of year. The Court of Appeal, emphasised the word “normal” in the term “normal characteristic”, noting that it was used by Lord Diplock when he observed in The Evia (No 2) at p 749 that:

In para 54 the Court of Appeal placed reliance upon the approach of Mustill J in The Mary Lou at p 278. They noted that in his description of what constitutes an abnormal occurrence, Mustill J implicitly recognised the need to approach the identification of an abnormal occurrence realistically and having regard to whether the event had occurred sufficiently frequently so as to become a characteristic of the port. At the end of the same paragraph, having recognised the difficulty of finding an appropriate turn of phrase, Mustill J said this:

The Court of Appeal identified a number of respects in which they concluded that the judge went wrong. They summarised them in para 55 of their judgment:

The Court of Appeal added in para 56 that, instead of asking the unitary question directed at establishing the correct characterisation of the critical combination (abnormal occurrence or normal characteristic of the port), the judge merely addressed the respective constituent elements of the combination (swell from long waves making it dangerous for a vessel to remain at the Raw Materials Quay and gale force winds from the northerly/north-easterly quadrant making navigation of the Fairway dangerous or impossible for Capesize vessels) separately. He looked at each component and decided that, viewed on its own, neither could be said to be rare and both were attributes or characteristics of the port. The Court of Appeal concluded that that was the wrong approach; what mattered was not the nature of the individual component dangers that gave rise to the events on 24 October, but the nature of the event (namely the critical combination of the two) which gave rise, on the judge’s findings, to the vessel effectively being trapped in port.

The Court of Appeal further held in para 57 that the judge was also wrong to hold that, even if the critical combination was rare, nonetheless it was a characteristic of the port, for two reasons. The first (as stated in his para 127) was because, although it might well be rare for these two events to occur at the same time, nobody at the port could be surprised if they did, and there was no meteorological reason why they should not occur at the same time. The second (as stated in his para 128) was because, even if the concurrent occurrence of those events was a rare event in the history of the port, such an event flowed from the characteristics or features of the port.

The Court of Appeal concluded that both reasons were fallacious. In my opinion they were correct so to hold. As to the first, the Court of Appeal noted in para 58 that the conclusion that nobody at the port could be surprised that both the above events occurred at the same time appears to have been based on the idea that, provided an event is theoretically foreseeable as possibly occurring at the relevant port, because of the port’s location, then that is enough to qualify the event as a “characteristic of the port”.

The Court of Appeal correctly said at para 58 that the judge appears to have derived that test from dicta in the judgment of Mustill J in The Mary Lou at p 278, where (in the passage quoted at para 33 above) he referred to long waves and northerly gale winds as being “at least foreseeable”. However the Court of Appeal, in my opinion correctly, held that satisfaction of the test of mere “foreseeability” was per se clearly not sufficient to turn what the judge himself described as “a rare event in the history of the port” into a normal characteristic or attribute of the port. They held that the error made by the judge was to pick up on the words “at least foreseeable” in his citation from Mustill J’s judgment, and to use minimum foreseeability, without more, as some sort of litmus test for establishing whether an event was a characteristic of a port, without having any regard to significant factors such as the actual evidence relating to the past history of the port, the frequency (if any) of the event, the degree of foreseeability of the critical combination and the very severe nature of the storm on the casualty date. The Court of Appeal further held that in doing so the judge departed from the orthodox and practical approach of Mustill J in his judgment in The Mary Lou at p 278 and of Lords Diplock and Roskill in The Evia (No 2), to the question of whether an event was abnormal. Such an approach necessarily includes an examination of the past history of the port and of whether, in that evidential context, the event was unexpected. I agree. The Court of Appeal also noted in para 58 that he took the phrase “at least foreseeable” as used by Mustill J out of context. I agree with the Court of Appeal that it is clear that, when the passage is read in context, Mustill J was certainly not suggesting that mere, theoretical, foreseeability on its own was sufficient. He was not setting up some sort of alternate test which excluded considerations of questions such as the frequency of past occurrences of the particular event, or the degree of likelihood that the event was to occur in the future.

Moreover, the Court of Appeal noted at para 59 that, as the Court of Appeal emphasised in The Saga Cob in the passage cited at para 32 above, one has to look at the reality of the particular situation in the context of all the evidence, to ascertain whether the particular event was sufficiently likely to occur to have become an attribute of the port. Otherwise the consequences of a mere foreseeability test lead to wholly unreal and impractical results. The Court of Appeal focused on these examples in the instant case: does the mere fact that it is “foreseeable” from the location of San Francisco that earthquakes may occur in its vicinity, or from the location of Syracuse, beneath Mount Etna, that there may be volcanic explosions in its vicinity, predicate that any damage caused to vessels in those ports from such events, were they to occur in the future, would flow from the “normal characteristics or attributes” of those ports, and therefore necessarily involve a breach of any safe port warranty? The answer given by the Court of Appeal was obviously not; whether, in such circumstances, there would be a breach of the safe port warranty, or the event would be a characterised as an abnormal occurrence, would necessarily depend on an evidential evaluation of the particular event giving rise to the damage and the relevant history of the port.

The Court of Appeal was particularly struck (at para 60) by the fact, as they put it, that the judge provides no evidential basis for his apparent factual conclusion that “nobody at the port could, I consider, be surprised” if the crucial combination occurred, or for the conclusion reached earlier in para 110 of the judgment that “there must have been … a clear risk of gale force winds from the northerly quadrant in the Kashima Fairway at the same time as long waves were affecting the Raw Materials Quay”.

The final conclusion of the Court of Appeal’s on the first reason advanced by the judge and referred to by the Court of Appeal in para 57 was set out in para 61 of their judgment (which must be read in the context of para 60) as follows:

As to the second reason advanced by the judge, the Court of Appeal responded in this way in para 62:

By way of postscript, I note that on behalf of the owners significant stress was placed upon the failure of the Kashima port authority to carry out a risk assessment and put in place a proper safety system to deal with the risk of the two types of weather conditions referred to by the judge occurring at the same time. However, while it may be relevant in some cases, the question remains whether the event (or in this case the combination of natural events) which led to this casualty was an abnormal and unexpected occurrence or not. For the reasons I have given I conclude that the Court of Appeal were entitled to reach the decision which they did.

The ultimate conclusion of the Court of Appeal was set out in paras 63 and 64 as follows:

I agree with the Court of Appeal.

In my opinion, the Court of Appeal reached the correct conclusions for the reasons they gave. I initially questioned whether the Court of Appeal should have interfered with the decision of the judge at first instance. However, in the light of the submissions made on both sides, I have concluded that this was one of those rare cases in which the correct conclusion is that the casualty was caused by an abnormal occurrence as that expression is explained in the cases. I accept the reasoning of the Court of Appeal and prefer their approach to that of the judge.

I would accordingly dismiss the appeal on the safe port issue. I would answer the questions raised on that issue by the parties and set out in para 8.1(1) and (2) as follows. The port was not unsafe within the meaning of the safe port undertaking so that the charterers were not in breach of it. The conditions at the port amounted to an abnormal occurrence as that expression is understood in the cases.

Issue 2 summarised in para 8 above assumes (contrary to my view) that there was a breach of the safe port undertaking and asks whether the provisions for insurance in clause 12 of the Barecon 89 form preclude rights of subrogation of hull insurers and the right of owners to recover in respect of losses covered by hull insurers against the demise charterers for breach of the express safe port undertaking. The judge held that the owners and hull insurers were entitled to recover notwithstanding clause 12, whereas the Court of Appeal held that they were not. Lord Sumption agrees with the judge whereas Lord Toulson agrees with the Court of Appeal. I agree with the judge and Lord Sumption, essentially for the reasons they give. Lord Sumption has set out clauses 12 and 13, which I will not repeat.

I have been particularly struck by these considerations. I agree with the judge (at para 185) that the demise charterparty must be given the meaning which, having regard to the background known to both parties, it would reasonably be understood to bear and that, in circumstances where, in clause 29, the demise charterparty contains a clear safe port warranty, one would expect any exemption of the demise charterers from liability in damages for breach of the safe port warranty to be clearly expressed.

In para 190 the judge observed that the charterers relied upon the The Evia (No 2), in which the question arose whether (as Lord Roskill put it at p 766) the war risks clause cast upon the owners and their insurers all war risks and thus freed the charterers from liability for them pursuant to the safe port clause. It was held that the charterers were freed from any liability that they might otherwise have. Lord Roskill identified the relevant question as being whether the war risks clause was a complete code exhaustive of the owners’ rights, which depended upon the construction of the time charterparty as a whole.

As the judge noted at para 191, the charterers adopted that reasoning here and said that it was applicable to the demise charterparty and, indeed, that it was a stronger case because, not only did the demise charterers pay for the cost of hull insurance, but they were also named as joint assureds and, generally speaking, an insurer cannot exercise rights of subrogation to pursue a claim in the name of one co-assured against another.

The judge rejected that submission in these terms:

I entirely agree with the judge. It follows from the fact that clause 12 contains no such express exclusion that any such exemption can only arise by necessary implication. In short, there is nothing in clause 12 which provides that the demise charterers have no liability for breach of clause 29 and I see no basis for such a necessary implication, essentially for the reasons given by the judge and by Lord Sumption. In particular, it seems to me to be striking that, as the judge observed in para 195, clause 13, which contained an alternative insurance and repairs clause which not only provided that hull insurance would be paid for by the registered owner but also expressly stated that the registered owners and/or insurers would not have any rights of recovery or subrogation against the demise charterers in respect of insured losses, was deleted from the printed form. Thus the demise charterers chose not to be bound by clause 13.

After a detailed analysis (between paras 196 and the first part of para 198), the judge said that he did not consider that clause 12 codified the rights and liabilities of the parties with regard to insured risks. He noted that it provides for the provision of insurance and who is to pay for it, for the demise charterers to be responsible for insured repairs and to reimburse themselves from the proceeds of the insurance policy, for the demise charterers to be responsible for other repairs and for the claims on a total loss to be paid to the mortgagee for distribution to the registered owners and demise charterers in accordance with their respective interests. I agree with the judge that this does not in the required sense codify the rights and liabilities of the parties with regard to breach of the safe port warranty where the casualty caused by the breach has given rise to a claim on the insurance.

It is true that, as the judge put it in para 199, what clause 12 has, which neither the clause in The Evia (No 2) nor the clause in The Concordia Fjord (as expressly noted by Mr MacCrindle QC) had, is a provision that the owners and demise charterers were to be co-assureds, thereby, on the face of it, bringing into play the principle that, generally, an insurer cannot exercise rights of subrogation to pursue a claim in the name of one co-assured against another.

After referring (in para 200) to the decision and reasoning of Rix LJ in Tyco Fire and Integrated Solutions (UK) Ltd v Rolls-Royce Motor Cars Ltd [2008] 2 All ER (Comm) 584, the judge concluded in para 201 as follows:

For these reasons and those given by Lord Sumption, I would have answered the question or issue 2 summarised in para 8 above, namely whether the provisions for insurance in clause 12 of the Barecon 89 form preclude rights of subrogation of hull insurers and the right of owners to recover in respect of losses covered by hull insurers against the demise charterers for breach of the express safe port undertaking, in the negative. I am not persuaded by the judgments of Lord Mance and Lord Toulson (with whom Lord Hodge agrees) to reach a different conclusion.

Question or issue 3 summarised in para 8 above assumes (contrary to my view) that there was a breach of the safe port undertaking and asks whether Daiichi is entitled to limit its liability for Gard’s losses or any (and, if so, which) of them as against Sinochart (and Sinochart in turn against Gard) pursuant to section 185 and Schedule 7 article 2(1) of the Merchant Shipping Act 1995, which gave the force of law to the Convention on Limitation of Liability for Maritime Claims 1976 (“the 1976 Convention”).

Like the joint insurance issue, this issue does not arise in the light of our decision on the safe port issue. However it raises a point of some potential importance and was fully argued before us. Neither of the courts below considered it because it was accepted that they were both bound by the decision of the Court of Appeal in The CMA Djakarta [2004] 1 Lloyd’s Rep 460. In that case Longmore LJ gave the only substantive judgment, with which Waller and Neuberger LJJ agreed.

As formulated on behalf of Daiichi, who were time charterers, the question at issue is whether Daiichi (hereinafter “the charterers”) can limit their liability for the loss of the vessel and consequential losses arising out of the loss of the vessel. The answer to that question depends largely upon whether The CMA Djakarta was correctly decided in the Court of Appeal. The limitation issue is a short but important one. It concerns the correct interpretation of the 1976 Convention.

The Convention provides, so far as relevant, as follows:

Although I have included the references to salvors in the above quotation, it is common ground that those references are irrelevant to the issues in this appeal. They were included in the Convention in order to depart from the decision of the House of Lords in The Tojo Maru [1972] AC 242.

The critical question for present purposes is whether The CMA Djakarta was correctly decided in the Court of Appeal. Gard say that it was. They rely upon the fact that it has not been criticised in any case since it was decided. They thus rely upon the reasoning of the Court of Appeal in The CMA Djakarta. They also rely upon some at least of the reasoning of David Steel J at first instance in that case, reported at [2003] 2 Lloyd’s Rep 50 and of Thomas J in The Aegean Sea [1998] 2 Lloyd’s Rep 39. By contrast, the charterers say that both cases were wrongly decided. I have reached the clear conclusion that the Court of Appeal were correct, essentially for the reasons they gave.

The 1976 Convention had of course been preceded by earlier Conventions, which David Steel J referred to as part of his historical analysis of the right to limit liability set out in detail in his judgment at first instance reported in [2003] 2 Lloyd’s Rep 50. I entirely agree with his analysis (at pp 51-53) and will not repeat it here, save to note some key points. He referred to the first relevant limitation statute, which was the Responsibility of Shipowners Act 1733 and then to the Merchant Shipping Act 1854, the Merchant Shipping Acts Amendment Act 1862 and the Merchant Shipping Act 1894, which consolidated the earlier legislation.

As David Steel J put it, section 503 of the 1894 Act furnished a limit to an owner’s liability in respect of certain categories of occurrence in these terms:

The two Conventions to which David Steel J then specifically referred were the two Conventions for the Unification of Certain Rules relating to the Limitation of Liability of Seagoing Vessels in 1924 and 1957 (“the 1924 Convention” and the “1957 Convention” respectively). Although the United Kingdom signed the 1924 Convention, it never became part of English law and was replaced by the 1957 Convention as between states that ratified the 1924 Convention.

In para 20 David Steel J noted that the right of limitation was still afforded to owners for certain occurrences but the categories of occurrence were enlarged by article 1(1) to include “(a) … loss of, or damage to, any property on board the ship;” and

The impact of the 1957 Convention was enacted in the form of the Merchant Shipping (Liability of Shipowners and Others) Act 1958 (“the 1958 Act”) by way of amendment to the 1894 Act. In purported compliance with article 1(1)(b), section 2 prescribed that a new subsection (d) should be substituted in subsection 1 of section 503 of the 1894 Act as follows:

At first instance in The CMA Djakarta David Steel J noted at para 25 that the Convention introduced radical changes as regards to both the size of the fund and the circumstances in which the entitlement to limit might be lost. As he put it, in short, the Convention made available a significantly enhanced fund at what he said was perceived to be the maximum insurable level, but the entitlement to which could only be challenged in quite exceptional circumstances: see The Leerort [2001] 2 Lloyd’s Rep 291. He said in para 26 that it was notable that the Convention left largely untouched the range of persons entitled to limit, although it expressed the category in somewhat different terms.

David Steel J identified the rival contentions of the parties succinctly in paras 28 and 29. It was the charterers’ case, first that, as charterers, they fell squarely within the category of persons enabled to limit their liability as prescribed by article 1, and secondly that the entire claim for damages arising out of the casualty fell equally squarely within the category of qualifying claims under article 2. By contrast, it was the owners’ case that it was clear from the overall context, having regard to the object and purpose of the Convention, that the entitlement to limit was restricted to those persons identified in article 1(2) whose liability for the qualifying claim arose qua owner and not otherwise. On the facts, limitation was not available since it was common ground that no part of the claim arose from the role of the appellant charterers qua owners. David Steel J essentially accepted the submissions made on behalf of the charterers and, in doing so, followed the decision and reasoning of Thomas J in The Aegean Sea.

Longmore LJ set out his general approach to the Convention in paras 9-11. He first identified what he concluded were errors made by David Steel and Thomas JJ. In particular, he did not agree with them that, in order to succeed in limiting their liability, it was necessary for charterers’ claims to arise from their role qua owners. I agree that Longmore LJ’s conclusions in that regard were correct for the reasons he gave and do not need to revisit them.

Longmore LJ then cited a number of cases which support the proposition that, given that the Convention is in its own words incorporated into English law, the task of the court is to construe the Convention as it stands without any English law preconceptions. As he put it, the interpretation of international conventions must not be controlled by domestic principles but by reference to broad and general principles of construction. He cited a number of well-known cases: Stag Line Ltd v Foscolo, Mango & Co Ltd [1932] AC 328, 350, James Buchanan & Co Ltd v Babco Forwarding & Shipping (UK) Ltd [1978] AC 141, 152D-E, Fothergill v Monarch Airlines Ltd [1981] AC 251, 272E, 282A and 293C, and Morris v KLM Royal Dutch Airlines [2002] 2 AC 628, 656 at para 78.

He added in para 10 that, while it may be difficult to know what are broad and acceptable principles, some principles are enshrined in articles 31 and 32 of the Vienna Convention on the Law of Treaties (1969), which was ratified by the United Kingdom in 1971 and came into force in 1980. Those articles provide:

Longmore LJ summarised his conclusions derived from articles 31 and 32 in this way. The duty of a court is to ascertain the ordinary meaning of the words used, not just in their context but also in the light of the evident object and purpose of the Convention. The court may then, in order to confirm that ordinary meaning, have recourse to the travaux préparatoires and the circumstances of the conclusion of the Convention. The 1957 Convention was signed by the United Kingdom.

Like Longmore LJ in para 10, I would regard the existence and terms of a previous international convention (even if not made between all the same parties) as one of the circumstances which are part of a conclusion of a new convention but recourse to such earlier convention can only be made once the ordinary meaning has been ascertained. Such recourse may confirm that ordinary meaning. It may also sometimes determine that meaning but only when the ordinary meaning makes the convention ambiguous or obscure or when such ordinary meaning leads to a manifestly absurd or unreasonable result.

In para 11 Longmore LJ set out the object and purpose of the 1976 Convention as agreed between the parties as follows (omitting the reference to the Tojo Maru). First, the general purpose of owners, charterers, managers and operators being able to limit their liability was to encourage the provision of international trade by way of sea-carriage. Second, the main object and purpose of the Convention was to provide for limits which were higher than those previously available in return for making it more difficult to “break” the limit. Under the 1894 Act an owner was entitled to limit his liability if he showed that the casualty occurred without his “actual fault or privity”. Under the 1976 Convention (and the 1995 Act) the (now higher) limit is to apply unless it can be shown that the loss resulted from the personal act or omission of the party relying on the limit “committed with intent to cause such loss or recklessly with the knowledge that such loss would probably result”. It is thus particularly difficult to break the limit, but the amount available for compensation is higher than it was previously. Longmore LJ added that it was not possible to ascertain with certainty any object or purpose beyond that common ground. He therefore turned to the ordinary meaning of the Convention, beginning at the beginning.

Leaving the position of salvors on one side, at para 13 Longmore LJ noted in the context of article 1 that the word “shipowner” was defined as “the owner, charterer, manager or operator of a seagoing ship”. He then rejected the opinion of David Steel J and Thomas J that a charterer could only limit his liability if he was acting in the management or operation of the vessel. In particular he expressed the view that the mere fact that “charterer” is part of the definition of the word “shipowner” cannot of itself mean that a charterer, which was “an expression otherwise unqualified”, has to be acting as if he were a shipowner (ie qua shipowner) before he can limit his liability. He added:

Longmore LJ then said in para 13 that there were two difficulties in the argument to the contrary. I do not think that it is necessary for me to discuss them in any detail, since they were not relied upon on behalf of the charterers here. In para 18, to my mind correctly, Longmore LJ said that he would not give any gloss to the word “charterer” in article 1(2) and that he would give it what seemed to him to be its ordinary meaning. I agree with that approach.

Longmore LJ noted at para 21 that the issue was not resolved by a consideration of article 1 of the Convention because it was still necessary to ascertain whether a claim for damage to the ship by reference to which a charterer seeks to limit his liability is a claim which falls within article 2. In paras 22-24 he considered loss or damage to the ship under article 2(1)(a), which he correctly held extends the right to limit, inter alia, to claims in respect of “loss of or damage to property occurring on board”, which is not apposite to include loss of or damage to the ship itself since neither the loss of a ship nor damage to a ship can be said to be loss or damage to property on board. Property on board means something on the ship and not the ship itself.

The question then arises whether this is a claim in respect of loss of or damage to property “occurring … in direct connexion with the operation of the ship.” Longmore LJ held in para 23 that the most obvious reason for including this category of claim is to cater for cases of collision with another ship. Loss or damage to that other ship (or its cargo) is not “loss of or damage to property … occurring on board” but is “loss of or damage to property … occurring … in direct connexion with the operation of the ship”.

The critical part of his reasoning is to my mind in the next part of para 23, where he said that that wording was not apt to cater for a case where the very ship, by reference to the tonnage of which limitation is to be calculated, is lost or damaged because the loss envisaged is loss to something other than that ship herself. He added at para 24 that it was not without interest that in order to describe this category of claim the framers had used the phrase “occurring … in direct connexion with the operation of the ship”. That was, he said, virtually the same phrase as that used by David Steel J to define what he meant by “qua owner”. If one were to postulate the case of the vessel being in berth when the dangerous cargo exploded and damaged parts of the harbour, the harbour authority could sue for that damage but one would expect that the shipowners would be able to limit any liability for that claim. In order to do so, however, they would have to assert that the loss or damage occurred in direct connexion with the operation of the ship. The fact that dangerous cargo had with their permission been loaded on the ship would, one thinks, be enough for that purpose. But if it would be sufficient for that purpose, it would be odd that a charterer pursuant to article 1 could not say of his own act in permitting such cargo to be loaded that it was an act “in direct connexion with the operation of the ship”.

Longmore LJ treated these provisions as of some importance in reaching his conclusion. In my opinion he was correct to do so. In para 25 he noted that Thomas J set them out in detail in The Aegean Sea and summarised them broadly in this way. Article 9(1) provides for the claims against (a) the persons mentioned in article 1(2) (viz owner, charterer, manager or operator) to be aggregated if they arose on distinct occasions; likewise for claims against (b) the owner of a ship rendering salvage services and a salvor operating from that ship and (c) a salvor not operating from a ship. Article 9(2) then deals with passenger claims. Article 10 provides that liability can be limited without the creation of a fund. Article 11 then provides for the constitution of a limitation fund when that is, in fact, done; it provides for separate funds for the “shipowner” category of those entitled to limit and the “salvor” categories (and for passenger claims) by providing:

Further, at para 26 Longmore LJ approved Thomas J’s view that, if he was wrong in his general conclusion that the charterers could only limit when the loss (a total loss in that case) was caused by an act normally performed by the shipowner, the claim for the loss of the vessel did not fall within article 2(1)(a) because the loss of the ship was not “loss of property … occurring … in direct connection with the operation of the ship”. This in turn was because, as Thomas J put it at (p 51):

Longmore LJ added that, similarly in The CMA Djakarta, which was a case of extensive repair rather than total loss, David Steel J upheld the shipowners’ argument that the vessel cannot be both the victim and the perpetrator and that the “property” envisaged in the article must be the property of a third party either on board the vessel (eg cargo) or external to the vessel, for example an SBM. David Steel J said (at para 52):

In para 27 Longmore LJ gave a further reason for his earlier conclusion based on the ordinary meaning of article 2(1)(a), with which I also agree. I agree that the effect of giving the words their ordinary meaning is not absurd or unreasonable, nor is there ambiguity or obscurity. As Longmore LJ pointed out, David Steel J considered the wording of the 1957 Convention and held that his conclusion derived support from article 1 of that Convention, which drew an express distinction between “the ship” and “other property”. Longmore LJ held that it was, if anything, even clearer than the 1976 Convention on this point. He held that that served to confirm the proposition since any intention to change the previous agreement so that damage to the ship itself would be subject to limitation would have been made much more explicitly. I agree, although I do not regard this point as of any great significance.

I should add that, in my opinion, in agreement with David Steel J and the Court of Appeal in The CMA Djakarta, there is nothing in the travaux préparatoires which supports any other conclusion. Some reliance was placed upon article 4(1)(iv) of the Liens and Mortgages Convention (1967), which provides:

For the reasons I have given, which are essentially the same as those of the Court of Appeal in The CMA Djakarta, I would hold that, if there were a breach of the safe port warranty, the charterers would not be entitled to limit their liability under the Convention in accordance with the limitation fund calculated by reference to the vessel.

On the safe port issue (question 1), I would hold that the port was not unsafe within the meaning of the safe port undertaking so that the charterers were not in breach of it. The conditions in the port amounted to an abnormal occurrence as that expression is understood in the cases.

On the joint insurance issue, if (contrary to para 88 above) there were a breach of the safe port undertaking, I would have answered question or issue 2 summarised in para 8 above, namely whether the provisions for insurance in clause 12 of the Barecon 89 form preclude rights of subrogation of hull insurers and the right of owners to recover in respect of losses covered by hull insurers against the demise charterers for breach of the express safe port undertaking, in the negative. However I recognise that this is a minority view.

On the limitation issue (question 3), again assuming that there were a breach of the safe port undertaking, I would hold that the charterers were not entitled to limit their liability in accordance with a limitation fund calculated by reference to the vessel.

It may be appropriate for declarations to be made reflecting the above. The parties are invited to make submissions on the form of order and on costs within 21 days of the handing down of the judgments.

I agree that there was no breach of the safe port warranty in this case, for the reasons given by Lord Clarke, which substantially correspond to those of the Court of Appeal.

On that footing, it was strictly speaking unnecessary for the Court of Appeal to decide whether, if there had been a breach, there would have been any liability in damages. For the same reason, it is unnecessary for us to deal with it. Nonetheless, I propose to do so, because the question is of some general importance and I am not persuaded that the Court of Appeal answered it correctly. They held that the demise charterers suffered no loss by the destruction of the ship because, although there was a corresponding safe port warranty in the demise charter of which they were (on this hypothesis) in breach, they would have had no liability to pay damages representing the value of the ship. This is said to be the result of clause 12 of the demise charter, which provided for the demise charterers to procure insurance for the vessel at their own expense against marine, war and protection and indemnity risks, for the joint interests of themselves and the head owners. From this, and from the provisions of clause 12 relating to the distribution of the insurance proceeds upon a total loss, it is said to follow that the head owners were obliged to look exclusively to the insurance proceeds and not to the demise charterer to recover the value of their ship. Therefore the demise charterers had no liability to pass on to the time charterers as damages for breach of the safe port warranty, and the insurers, as their assignees, had no greater right. The same would, on this analysis, have been true if the insurers had brought a subrogated claim against the time charterers in the name of the demise charterers. It is accepted that this argument applies only to that part of the loss which represents the value of the ship. It does not apply to the claim for SCOPIC expenses (essentially salvage), wreck removal costs or loss of hire, which together accounted for rather more than a third of the claim.

It is necessary to draw attention at the outset to the limited basis on which this issue comes before the court. There are three possible bases on which a demise charterer might be in a position to claim damages from a subcharterer for the loss of a ship of which he is the bailee but not the owner:

The demise charter was on the Barecon 89 form. The form was originally drafted in 1974 by the Documentary Committee of the Baltic and International Maritime Council, and revised in 1989. It is said to have become, in one or other of its variants, the most commonly used form of bareboat charter world-wide. Under clause 9 of the form, the demise charterers have the usual obligation to maintain the vessel in good repair and efficient operating condition and to take immediate steps to have any necessary repairs carried out. The form as printed contains no trading limits other than clause 5, which simply requires the vessel to be employed in conformity with the terms of its insurances, including any insurance warranties, unless the agreement of the insurers is obtained. The insurances are governed by either clause 12 or clause 13, one of which must be selected. Clause 12, which was selected in this case, provides (so far as relevant):

In the present case, the Barecon 89 form was amended by deleting the trading limits clause (clause 5) and adding at clause 29 a safe port warranty in the following terms:

The provisions which are said to exclude a right to recover the value of the ship as damages for breach of clause 29 are clauses 12(a) and (c). Clause 12(a) requires insurance to be in place for the parties’ joint account. Clause 12(c) provides that “all insurance payments for such loss shall be paid to the Mortgagee, if any, in the manner described in the Deed(s) of Covenant, who shall distribute the moneys between themselves, the owners and the charterers according to their respective interests.” The argument is that the clause as a whole is a complete code governing financial liability for loss or damage to the ship, and that the words quoted provide for the relief, and the only relief, available as between the head owner and the demise charterer for a total loss. To address this argument, I propose to deal first with the law relating to rights as between co-insured. I shall then consider how, if at all, it applies to this demise charter.

The starting point is the general rule that insurance recoveries are ignored in the assessment of damages arising from a breach of duty: Bradburn v Great Western Railway Co (1874) LR 10 Ex 1; Parry v Cleaver [1970] AC 1. This can conveniently be called the collateral payments exception. It is a departure from the general principle that collateral benefits are brought into account, and is probably best regarded as being based on public policy. Insurance recoveries are a benefit which the injured party has bought in consideration of his premiums, which are intended to inure to his benefit alone, not that of third party wrongdoers. Moreover, the courts have traditionally been concerned to preserve the subrogation rights of insurers against those who are legally responsible for the loss, which are an important part of the economics of insurance. The effect of the collateral payments exception is that as between the insured and the wrongdoer who has caused the loss, they are not treated as making good the former’s loss or as discharging the latter’s liability. The assumption underlying it is that as far as the wrongdoer is concerned, insurance is res inter alios acta, ie, loosely translated, none of his business. The rule thus stated falls to be modified in a case where insurance manifestly is the wrongdoer’s business because, for example, he is a co-insured and/or the insurance is taken out for his benefit. The business context in which this has most commonly arisen is the co-insurance of employer, contractor and subcontractors under standard forms of building contract.

It is well established, and common ground between the present parties, that where it is agreed that the insurance shall inure to the benefit of both parties to the contract, they cannot claim against each other in respect of an insured loss. Co-insurance is the paradigm case. The principle first appears in the United States, but was successively adopted in early editions of MacGillivray on Insurance Law, by the Supreme Court of Canada in Commonwealth Construction Co Ltd v Imperial Oil Ltd [1978] 1 SCR 317 and by the English courts in a line of cases beginning with the decision of Lloyd J in Petrofina (UK) Ltd v Magnaload Ltd [1984] QB 127. What is less clear is its juridical basis. Lloyd J was inclined to think that it was based on the rule against circuity of action, which is difficult to accept given that the insurer will not be a party to any litigation between the co-insureds. The better view, which was endorsed by the House of Lords in Co-operative Retail Services Ltd v Taylor Young Partnership Ltd [2002] 1 WLR 1419, paras 61-65 (Lord Hope), is that it is an implied term of the contract of insurance and/or of the underlying contract between the co-insureds pursuant to which their interests were insured. The implication is necessary because if the co-insureds are both insured against the relevant loss, the possibility of claims between them is financially irrelevant. It would be absurd for the insurer to bring a subrogated claim against a co-insured whom he would be liable to indemnify against having to meet it. It should be noted that this reasoning is relevant only to the position as between the co-insureds. In all of the English cases before this one the question arose between the co-insureds and their insurer. None of them raised the question how the principle about co-insurance affects claims against a third party wrongdoer who is not himself a co-insured and is not party to the arrangements between them. There is no necessity to exclude a claim against him and indeed no reason why either of the co-insureds or their insurer should wish to do so. It is impossible to identify any contract whose business efficacy depends upon that result being achieved.

As between a co-insured (or his insurer) and a third party wrongdoer, a different question arises which none of the existing English authorities purports to answer. The question is this: when we say that one co-insured cannot claim damages against another for an insured loss, is that because the liability to pay damages is excluded by the terms of the contract, or is it because as between the co-insureds the insurer’s payment makes good any loss and thereby satisfies any liability to pay damages? The significance of this question may be illustrated by a hypothetical case. Suppose that A and B are engaged in some contractual venture, involving the use of A’s property. The property is insured in their joint names. It is damaged in breach of some contractual duty owed to A by B, but the cause of the damage is some act of B’s agent, X. If the effect of the co-insurance is that B’s liability to pay damages to A is excluded, then B never had a relevant liability and has suffered no loss which he can claim over against X. But if its effect is that payment by the insurer makes good A’s loss as between A and B and thereby satisfies any liability of B, the result is different. The effect is to exclude the collateral payments exception, so as between A and B the receipt of the insurance proceeds must be taken into account. However, the fact that the insurer’s payment has made good the loss as between A and B does not mean that it has done so as between B and the stranger, X. As between B and X the insurance is res inter alios acta. Indeed, its normal consequence is that the claim will survive to be pursued by the subrogated insurers. Either analysis will achieve the object of the implication, namely to prevent claims between co-insureds. But they have radically different consequences for claims against third parties. Which is the correct analysis must depend on the particular terms of the particular contract. The answer will not necessarily be the same in every case.

I therefore return to the contractual arrangements between the head owners and demise charterers of the “OCEAN VICTORY”. We have not seen the actual policy. What matters, however, is not the actual policy but the policy envisaged in clause 12 of the demise charter. The relevant insurance is the insurance against marine risks which is required by clause 12(a). That is an insurance on property. As far as the demise charterer is concerned, although it is not a liability insurance, he is treated as having an insured interest in the property as such, because his potential liability to the head owner as a bailee and time charterer means that he “stands in [a] legal or equitable relation to the adventure or to any insurable property at risk therein, in consequence of which he may benefit by the safety or due arrival of insurable property, or may be prejudiced by its loss, or by damage thereto, or by the detention thereof, or may incur liability in respect thereof”: see Marine Insurance Act 1906, section 5(2) (emphasis supplied); cf Arnould, Law of Marine Insurance and Average, 18th ed (2013), para 11.37-11.38. If the ship is lost or damaged, the measure of any liability of the demise charterer will be the same as the measure of the owner’s loss, namely the diminution, partial or total, in the value of the ship. There is no other basis on which he could be entitled to insure on the same basis as the owner.

When the “OCEAN VICTORY” was lost, the insurers were bound under clause 12(c) to pay its insured value to the head owner or, rather, to the mortgagee for the account of the head owner. But the natural legal inference from (i) the fact that the demise charterer is insured for his interest in the ship, (ii) the implied prohibition of claims for damages between the co-insured for loss of or damage to the ship, and (iii) the avoidance of double recovery, is that the insurer’s payment to the head owner makes good the head owner’s loss not just as between the insurer and the head owner but as between both of them and the demise charterer. The demise charterer’s liability under the demise charter for the loss of the ship has not been excluded. It has been satisfied. It follows that the demise charterer may claim over against a time charterer who is not party to the insurance or any of the contractual arrangements connected with it.

This may be tested by asking what would happen if the insurer did not pay, for some reason which did not involve a breach of duty by either co-insured, for example because the insurer became insolvent after the casualty. The demise charterer’s obligation to insure ceases upon the termination of the demise charter: see clause 12(a). And the demise charter terminates upon the total loss of the ship: see clause 12(d). The result is that there is no effective insurance, no default of the demise charterer in there being no effective insurance, and no basis on which the head owner can be supposed to look exclusively to its proceeds to make good a loss arising from a breach of the safe port warranty. The demise charterer would be bound to pay damages, not because he was responsible for the lack of insurance but because he was liable for the destruction of the ship in breach of his contract. This can only be because it is the payment of the insurance proceeds which discharges the liability of the demise charterer by making good the head owner’s loss. In the absence of payment, there is no discharge and no bar to a claim between the co-insureds.

This does mean that if the insurer, being solvent, delays in paying the claim, the head owner can require the demise charterer to pay the loss at once. But I do not regard that as undermining what I consider to be the way that clause 12 works. Quite apart from the consideration that the contract must be construed on the assumption that the insurer will perform his obligations, an obligation to pay damages upon a breach of contract is a routine consequence of the chartered service. It is no more unthinkable in the case of the demise charter than it is in the case of the time charter, where no question of co-insurance arises.

This analysis derives strong support from a number of other features of this particular demise charter:

In my judgment the Court of Appeal was wrong to hold that the demise charterers were relieved by the terms of the demise charter of the obligation to pay damages for the consequences of an order to an unsafe port. I would therefore have allowed the appeal on this point if it had arisen.

On the limitation issue, I agree with Lord Clarke.

I agree with the judgment prepared by Lord Clarke on the first and third issues and with the judgment of Lord Toulson on the second issue in this appeal. In what follows, I set out some supplementary reasons of my own for agreeing with Lord Toulson on the second issue.

Clauses 12 and 13 are standard clauses of the “‘Barecon 89’ Standard Bareboat Charter” issued by the Baltic and International Maritime Council (“BIMCO”). It is relevant to consider how the scheme introduced by clause 12 (or, where used, clause 13) operates in circumstances where there is no clause 29. A demise charter involves a bailment on whatever terms may be agreed. Clauses 9 and 12 impose on charterers strict responsibility for having all necessary repairs done. But a demise dharterer may well cause total loss of the demised vessel in circumstances constituting a breach of duty or of an express or implied contract term. Printed clause 5, which was replaced in the present charter by clauses 29 and 30, could itself involve such a breach. Its first and third paragraphs include, for example, various obligations relating to lawful trading and the carriage of suitable lawful merchandise.

Clause 12 (or, where used, 13) contains a scheme designed to address the possibility of the vessel requiring repairs or suffering a total loss. First, and fundamentally, these clauses provide who is to take out marine, war risks and protection and indemnity (“P&I”) insurance. Clause 12 deals with circumstances where the charterers are to do both, clause 13 with circumstances in which owners are to maintain marine and war risks insurance while demise charterers have to maintain P&I insurance. Secondly, they provide for the nature of the insurance which is to be maintained. Under clause 12 all the insurances are to be “in such form as the owners shall in writing approve, which approval shall not be unreasonably withheld”, whereas clause 13 provides for marine and war risks insurance “under the form of policy … attached hereto”, while the P&I insurance to be arranged by charterers has again to be “in such form as the owners shall in writing approve, which approval shall not be unreasonably withheld”. Under both clauses 12 and 13, the value of the vessel for marine and war risks insurance purposes is to be that stated in box 27 (here USD 70m), while boxes 28 and 29 are to specify whether either or both parties may take out additional insurance and if so for how much (each being, under the present charter, prohibited from so doing).

Thirdly, under clause 12 all the insurances are to be arranged to protect “both the Owners and the Charterers and mortgagees (if any)”and to be “in the joint names of the Owners and the Charterers as their interests may appear”. Under clause 13, the marine and war risks insurances are likewise to be “in the joint names of the Owners and the Charterers as their interests may appear”, but there is also an express provision that:

Fourthly, both clauses 12 and 13 make the demise charterers responsible for effecting any repairs, securing reimbursement from underwriters to the extent of the insurance coverage, but remaining responsible for all repairs not covered by the insurance and/or falling within any possible insurance franchise or deductibles.

Fifthly, both clauses 12 and 13 address the possibility of the vessel becoming an actual, constructive, compromised or agreed total loss. Clause 12(c) provides that in this event the marine or war risks insurance payments for such loss “shall be paid to the Mortgagee (if any), … who shall distribute the moneys between themselves, the owners and the charterers according to their respective interests”. Clause 13(h) provides that they “shall be paid to the Owners, who shall distribute the moneys between themselves and the Charterers according to their respective interests”.

The scheme of clause 12 (and 13) is clearly intended to be comprehensive. Whatever the causes, both repairs and total losses fall to be dealt with in accordance with its terms, rather than by litigation to establish who might otherwise be responsible for undertaking them, for bearing the risk of their occurrence or for making them good. This is reinforced by the provisions for marine and war risks insurances to be taken out to protect the interests of owners, charterers and any mortgagees, and to be in the joint names of owners and charterers, as their interests may appear. It is well established, as Lord Sumption and Lord Toulson both acknowledge, that, where it is agreed that insurance shall inure to the benefit of both parties to a venture, the parties cannot claim against each other in respect of an insured loss. This principle is now best viewed as resting on the natural interpretation of or implication from the contractual arrangements giving rise to such co-insurance: Co-operative Retail Services Ltd v Taylor Young Partnership Ltd (“CRS”) [2001] Lloyd’s Ins Law Rep 122 (CA), [2002] 1 WLR 1419, per Lord Bingham, para 7 (favouring the rationale suggested by Brooke LJ in the Court of Appeal at para 72) and Lord Hope, paras 61 to 65. It is merely reinforced where, as here, the principal co-insureds, owners and charterers, are in the same group and ultimate beneficial ownership. Hull insurance covers losses whether or not it is due to the fault of any party, and it is, rightly, not suggested that the principle in CRS is subject to any exception where the loss is due to fault: see also on this point Mark Rowlands Ltd v Berni Inns Ltd [1986] QB 211, 232G-233B, per Kerr LJ.

Repairs were to be undertaken by charterers, regardless whether their cost was fully covered by insurance. But the scheme of both clauses 12 and 13 was for marine and war risks insurances in agreed form to be taken out in a fixed amount, and for no more unless otherwise specifically agreed. At the date of her total loss the OCEAN VICTORY is said to have been worth some USD 15m more than the amount for which she was valued for insurance under the demise charter. It is in my opinion implausible to suggest that, having developed this careful scheme for specific protection of their respective interests, it should have been intended that owners should as against charterers, a company in the same group and beneficial ownership, be able to reopen the scheme by claims of breach, exposing charterers to paying damages for the hull loss based on a different alleged value to that which owners and charterers had agreed between themselves. Just as parties must, for better or worse, accept a valuation agreed under a marine insurance (Marine Insurance Act 1906, sections 27(1) and (2)), so here the parties to this charter must be taken to have accepted the value they agreed for insurance purposes as conclusive as between themselves.

In this respect, the schemes of clauses 12 and 13 are in my opinion effectively mirror images of each other. Clause 13 has an express exclusion of any right of recovery or subrogation on the part of owners and/or insurers against charterers. That is explicable in the context of a clause dealing with insurance taken out by owners. BIMCO in their observations on the Barecon 89 form explain the optional clause 13 on the basis that “It has been felt that it may be useful to cover also the possibility which is believed may arise from time to time, that a vessel is bareboat chartered for a short period, say, four to six months”. BIMCO give the example of passenger vessels. They continue: “It is believed that it is normal practice that the Owners carry on with the insurances for their own account”. BIMCO also confirm that:

There is no suggestion by BIMCO that the responsibilities for repairs and total loss differ in any other respect as between clauses 12 and 13. There is no reason to think that clauses 12 and 13 were devised as anything other than two routes to the same substantive allocation of responsibilities for repairs and total loss, irrespective of fault. I conclude that the express exclusion of a right of recovery or subrogation in clause 13 was simply belt and braces in the context of insurances taken out by owners, and that the reason why no such express term appears in clause 12 was that it never occurred that there could be such claims in the context of insurances arranged by charterers to cover their own as well as owners’ interests. It is inconceivable that the parties intended fundamentally to alter the incidence of risk by permitting or excluding breach-based claims as between themselves in respect of a hull loss, depending upon whether it happened to be convenient to continue to use hull insurances taken out by owners or to rely on fresh insurances taken out by charterers. The judge’s speculation (para 202) that the risk of insurers’ insolvency might have motivated such a distinction is unconvincing, where not only is such risk most unlikely to have been in the forefront of the parties’ minds, but the decision which party should be responsible for hull insurances has no connection with any such risk or with the question whether or not breach-based liability should exist.

I do not consider that the substitution of printed clause 5 by typed clause 29 can have been intended to, or did, alter this basic scheme. In many respects clause 29 simply makes different provision for the same subject matter as clause 5. Clause 29 overlaps with clause 5, in so far as it requires employment in lawful trades. It specifies (not entirely felicitously, in so far as it refers to Zimbabwe) certain trading limits. Clause 5 contemplates that trading limits would be found in Box 19, but adds that the charterers must furthermore employ the vessel in conformity with the terms of her insurances. Both clause 29 and clause 5 address additional insurance premiums, which a vessel’s trading might in some circumstances require to be paid. Clause 29 has the provision for employment “only between good and safe berths, ports or areas where vessel can safely lie always afloat” on which Gard relies. Owners could no doubt object if they learned of intended or actual trading contrary to clause 29, and a breach by charterers in this respect might have consequences outside the scope of hull insurance and so outside the scheme of clauses 12 and 13. But it is, in my opinion, most unlikely that the “safe port” provision in clause 29 can have been meant to give rise to a system of recourse for loss of the hull, by way of damages for breach of contract, separate from and potentially counteracting the no fault scheme of responsibility and insurance recovery for a hull loss introduced by clause 12.

Teare J held that the owners, although indemnified by the insurers, had a subrogated right to claim against the demise charterers damages for breach of clause 29: judgment, paras 203-204. That makes no sense in the context of a co-insurance such as this, as Mr Recorder Jackson QC pointed out in Hopewell Project Management Ltd v Ewbank Preece Ltd [1998] 1 Lloyd’s Rep 448, 458, in a passage approved by Lord Hope in CRS at para 65. It would mean treating the present co-insurance as if it involved two separate and severable insurances, leaving charterers exposed, without liability cover, to claims for breach of charter or duty brought by owners in respect of loss of the hull. That is obviously not what anyone contemplated by clause 12 or clause 13. It is no answer that demise charterers might in turn have a back to back claim for damages against sub-charterers. They might equally well not have, eg if they were trading the vessel on their own account, so there was no sub-charter, or the sub-charter was on different terms or the sub-charterer was not worth powder and shot. I add, though no-one suggested this as a correct analysis, that it is inconceivable that anyone contemplated that the co-insurance to be arranged could give rise to successive payments of the same sum to different parties, with the second of such payments going to reimburse insurers for the first.

In these circumstances, Gard suggests an alternative basis for its claim, that liability exists in the first instance for any total loss attributable to a breach of the safe port warranty, but is then discharged to the extent of any recovery under the marine or war risks insurance taken out under clause 12. This analysis has the twin results, that (i) owners could after a total loss call upon charterers to pay the value of the vessel’s hull, even though an insurance claim was on foot, but had not yet been paid, and (ii) owners could look to charterers in damages for any amount by which the vessel’s actual value at the date of loss exceeded her insured value, both before and after insurers had paid the insured value. Both results are in my view inconsistent with the scheme of clause 12 (or 13), or with any sensible understanding of its evident purpose to cater comprehensively for responsibility for repairs and total loss.

Still more fundamentally, in the context of the present claim by Gard as assignees of demise charterers, there is no basis for treating payment of the vessel’s value under her marine or war risks insurance as “discharging” pro tanto any liability existing between owners and demise charterers. Both the ordinary marine and the war risks insurances are property insurances on the vessel’s hull. Payments made under them go to owners (or their mortgagees) and charterers for their respective interests in the hull. They cannot be treated as satisfying, at one and the same time, any liability (if there were any) which charterers had to owners. The contrary analysis involves the proposition that the parties were prepared to treat the hull insurance moneys as going in the first instance to charterers, not to the owners or their mortgagees, and then being passed up to owners. That is clearly not what happened or would ever happen in fact, and it is only in a counter-factual world and by contradicting the clear intent of clause 12 that it could be treated as if it had happened.

In my opinion, the reason why owners have no claim against charterers for damages for loss of the hull is not that such a claim exists under clause 29 but is at some point discharged. It is that, under a co-insurance scheme like the present, it is understood implicitly that there will be no such claim. This understanding applies, in my opinion, whether or not the insurance moneys have yet been paid. But, even if (contrary to my opinion) one were to treat this understanding as biting only upon payment of the insurance moneys, it still would not arise from or involve the proposition that some liability of the charterers to owners was “discharged” in the sense of paid or satisfied by the insurance moneys. The understanding would simply be that, upon payment of the hull insurance proceeds to those interested in the hull for their respective interests, no further liability would exist inter se. On that basis, charterers would still be unable to show that they had or had discharged any liability to owners, on the basis of which they could pursue a back-to-back claim against sub-charterers. But, as I have made clear, in my opinion the implied understanding arising from the co-insurance scheme is that there would be no liability for the hull value in the event of a total loss, whether or not the insured value had yet been disbursed.

Lord Sumption raises the question what would happen if an insurer became insolvent after a loss. This is in my view a remote eventuality which cannot be a guide to the meaning of clause 12 (or 13). It also raises different considerations of risk and implications to any which require determination on this appeal. Two views might be taken. One is that, under a scheme where both parties have agreed to the particular form of policy, presumably by particular insurers, the risk lies where it falls. It is to be noted that the risk could impact either party in different circumstances. It could impact charterers, who undertake an absolute responsibility for repairs, irrespective of fault, and would be unable to recoup themselves from the hull insurance as contemplated by clause 12 or 13. It could impact owners and charterers for their respective interests, in the event of a total loss occurring without fault. There is no reason to think that the parties intended any different result in the event of a loss which might be attributed to a breach of clause 29 or any other clause. But another view might be that the risk lies by implication on the party responsible for maintaining the insurance during the charter period, even though clauses 12 and 13 provide that the charter terminates as of the date of any total loss. The former seems the more likely position, but it is unnecessary on this appeal to decide which applies, since the point cannot assist to determine the present, quite different issue as to the implications of the insurance scheme when it is effective, which is no doubt what the parties, and BIMCO, had in mind when using and devising it.

Finally, it is submitted that charterers must have some liability towards owners under clause 29, because otherwise there can be no back-to-back claim down the line under the equivalent clause in the sub-charter between demise charterers and Sinochart, who in turn cannot pass on liability under the equivalent clause in the sub-sub-charter between Sinochart and Daiichi, who were and should be ultimately responsible. The difficulty with this submission is that it has not been tested and I, for my part, regard it as entirely open. Gard’s case before the Supreme Court has been put exclusively on the basis that charterers had a liability to owners, which in turn enables charterers to claim damages down the line. Mr Howard QC, who did not appear for Gard below, very frankly acknowledged that it had occurred to him, when he came into the case, that there could have been other bases on which the claim could have been presented, but that it had been concluded that Gard was confined to the way in which the case had been argued on its behalf below.

Those other bases are (i) that the charterers’ possessory title gave them a sufficient interest to be able to maintain a claim for the hull loss: compare the principle in The Winkfield [1902] P 42, whereby a bailee can claim in tort in his own name, without showing that he has any liability to the head owner, but accounting to the head owner for any loss exceeding his own loss, and/or (ii) that there are circumstances in contract where a contracting party can claim substantial damages for loss of or damage to property, when another person has actually borne such loss or damage: Dunlop v Lambert (1839) 6 Cl & F 600, The Albazero [1977] AC 774, 846G-847F. More recently, this latter possibility has been more widely recognised, by giving special treatment to contracts relating to property where loss due to a breach of the contract will be suffered by a holder of the property other than the contracting party: see Linden Gardens Trust Ltd v Lenesta Sludge Disposals Ltd and St Martins Property Corp Ltd v Sir Robert McAlpine Ltd (“St Martins”) [1994] 1 AC 85 where it was in the contemplation of the parties when the contract was made that the property, the subject of the contract and the breach, would be transferred to or occupied by a third party, who would in consequence suffer the loss arising from its breach: see also Darlington Borough Council v Wiltshier Northern Ltd [1995] 1 WLR 68 and the narrow ground of decision expressed by Lord Browne-Wilkinson at p 114G-H in St Martins, in which all members of the House joined. In such a situation, the claimant is seen as suing on behalf of and for the benefit of the injured third party and is bound to account accordingly: see St Martins, per Lord Browne-Wilkinson at p 115A-B and McAlpine Construction Ltd v Panatown Ltd (“Panatown”) [2001] 1 AC 518, per Lord Clyde, at pp 530E-F and 532D-E. (An even broader principle was also suggested by Lord Griffiths in St Martins, at p 96F-97D and reviewed inconclusively by Lord Browne-Wilkinson at pp 111F-112F as well as by the members of the House in Panatown, to the effect that a contracting party might itself have an interest in performance enabling it to claim damages without proving actual loss.)

In the absence of argument, it is not appropriate to reach any conclusion as to whether or how far either of these principles might have assisted Gard, had it been open to Gard to rely on them now. Suffice it to say that, since their application has not been tested, I am not prepared to proceed on the basis that charterers must be recognised as having had liability to owners, in order to be able to claim down the line against Sinochart and so for Sinochart against Daiichi. If the absence of such a liability is fatal to a claim by charterers down the line, it must be because neither alternative basis of claim identified in para 125 is available, and, if neither proved to be available, that would, presumably, be because the law did not regard this situation as one where considerations of justice comparable to those reflected in the reasoning in those cases militated in favour of recognising a right on the part of the charterers to claim down the line.

I also agree with Lord Clarke’s judgment on the first and third issues in this appeal. On the second issue I agree with the judgments of Lord Toulson and Lord Mance.

I agree with the judgments of Lord Clarke and the Court of Appeal on the issue whether there was a breach of the safe port warranty, and I would therefore dismiss the appeal. I agree also with Lord Clarke on the issue of limitation of liability and would affirm the decision of the Court of Appeal in The CMA Djakarta [2004] 1 Lloyd’s Rep 460. On the issue as to the effect of the charter’s joint insurance provisions, I agree with the reasoning and conclusion of the Court of Appeal and disagree with Lord Clarke and Lord Sumption for reasons explained in the rest of this judgment.

It is a curious feature of this appeal that permission to appeal on the joint insurance issue was sought on the basis that the Baltic and International Maritime Council (BIMCO) form of bareboat charter codenamed Barecon 89 is a standard form of contract in common use and the effect of clause 12 is therefore a matter of general importance, but the argument turned on the relationship between that clause and clause 29, which is not part of the standard form.

It is nevertheless sensible to begin by considering the effect of clause 12 in the unamended version of Barecon 89. It is important to understand its basic structure. The standard terms, set out in Part II of the policy, contain nothing about safe ports. Clause 5, headed “Trading Limits”, stipulates that the vessel is to be employed in lawful trades for the carriage of suitable lawful merchandise within trading limits which may be specified in the schedule which forms Part I of the policy, and that the vessel is not to be employed otherwise than in accordance with the terms of the insurance which is required to be maintained.

The insurance requirements are set out in clause 12 or its alternative, clause 13. Clause 12 is headed “Insurance and Repairs”. It provides:

Clause 13 applies in place of clause 12 if the parties so choose in part I of the policy. In relation to P and I risks during the charter, clause 13 follows the provisions of clause 12, but in relation to marine and war risks clause 13(a) puts the responsibility for maintaining cover on the owner. It provides:

The significant feature of bareboat chartering, or chartering by demise, is that during the period of the charter (in the words of BIMCO’s explanatory notes to Barecon 89) “the vessel comes in the full possession, at the absolute disposal, and under the complete control of the bareboat charterers.” The notes add that bareboat chartering is therefore entirely different from ordinary time chartering when it comes to the allocation of costs, liabilities and responsibilities.

Clause 12 in the unamended form deals comprehensively with the risks of loss or damage to the vessel and what is to happen in such an event. In summary, the demise charterer is responsible for arranging and maintaining insurance, in a form approved by the owner, in the names of both parties for an agreed value; the charterer is responsible for effecting all insured repairs; the charterer is responsible for repairs not covered by the insurance, for example, due to deductibles under the terms of the insurance (or, for that matter, due to use of the vessel outside the terms of the insurance); and in the case of a total loss covered by the insurance, the clause provides for the processing of the insurance moneys.

BIMCO’s explanation for the optional alternative of clause 13 was that sometimes a vessel is bareboat chartered for only a short period and it may make sense for the owners to carry on with the insurances which they are likely to have in place. Clause 13 therefore provides that the vessel is to be kept insured by the owners against marine and war risks, and that the owners and their insurers are to have no right of recovery or subrogation against the charterers on account of loss or damage covered by such insurance. It would be unnecessary to include equivalent words in clause 12. It cannot have been the parties’ intention that the charterer’s exposure to liability should be greater under clause 13, where cover against marine and war risks was to be maintained at the owner’s expense than under clause 12, where it was to be maintained at the charterer’s expense. Longmore LJ put the point pithily when he described the exclusion of rights of recovery or subrogation in clause 13 as “a confirmation rather than a negation of such exclusion in the more usually adopted clause 12 for the longer term charters when it is the charterers who pay the premium” (para 88).

The critical question then arises as to the effect, in relation to the operation of clause 12, of the substitution of clause 5 by clause 29, which provides that the vessel is to be employed in lawful trades for the carriage of lawful merchandises “only between good and safe berths, ports or areas where the [vessel] can safely lie always afloat”, etc.

On the hypothesis that Kashima was not prospectively a safe port for the OCEAN VICTORY when Daiichi gave instructions for her to discharge there, Daiichi was thereby in breach of the safe port undertaking in the time sub-charter between itself and Sinochart, which in turn was in breach of the equivalent undertaking in the time charter between itself and the demise charterer, OLH, which in turn was in breach of clause 29 of the demise charter between itself and the owners, OVM. The consequence of that breach was that the vessel was lost, but the demise charterer and the owners were co-insured (as required by clause 12) and the insurers paid its insured value. One of the insurers, Gard, claims to be entitled to recover that sum from the time charterers as assignee of the rights of the demise charterer. The claim therefore depends on the demise charterer being liable to the owners for that sum by way of damages for breach of clause 29, and thus entitled to recover the same sum from the time charterer as loss suffered by the demise charterer.

Gard’s case is that the breach of clause 29 caused the loss of the vessel; therefore the demise charterer was liable to the owners for the vessel’s value, and the fact that the owners were paid that amount by the insurers is res inter alios acta as between the demise charterer and the time charterer.

The critical question is whether the contractual scheme between the owners and the demise charterer precluded any claim by the former against the latter for the insured loss of the vessel. This is a matter of construction. It has become a common practice in various industries for the parties to provide for specified loss or damage to be covered by insurance for their mutual benefit, whether caused by one party’s fault or not, thus avoiding potential litigation between them. The question in each case is whether the parties are to be taken to have intended to create an insurance fund which would be the sole avenue for making good the relevant loss or damage, or whether the existence of the fund co-exists with an independent right of action for breach of a term of the contract which has caused that loss. Like all questions of construction, it depends on the provisions of the particular contract: see, for example, Co-operative Retail Services Ltd v Taylor Young Partnership Ltd [2002] 1 WLR 1419.

In that case a building owner entered into a standard form of building contract for the construction of office premises. Under its terms the contractor was required to take out and maintain a policy in the names of the owner, the contractor and specialist electrical subcontractors, Hall, for all risks insurance covering loss or damage to the works from specified perils including fire. Hall entered into a collateral contract with the owner warranting that it had exercised and would exercise all reasonable care and skill in the design and execution of the sub-contract works. A fire occurred causing extensive damage. The owners sued their architects and mechanical and engineering consultants, who brought third party proceedings against Hall. This raised the question whether Hall was liable to the owners in respect of the fire damage, alleged by the third party claimants to have been caused by Hall’s negligence and breach of warranty. The House of Lords, upholding a decision by the first instance judge and the Court of Appeal, held that it cannot have been the parties’ intention that parties who were jointly insured under a contractors’ all risks policy could make claims against one another in respect of damage covered by the insurance, or that the insurers could make a subrogated claim in the name of the owners against Hall, and that the court would if necessary hold that there was an implied term to such effect (which I infer in relation to Hall must logically have taken effect as an implied term of the collateral contract between itself and the owners). In so holding the House of Lords approved and applied the reasoning of Mr Recorder Jackson QC, as he then was, in Hopewell Project Management Ltd v Ewbank Preece Ltd [1998] 1 Lloyd’s Rep 448, 458, where he described it as nonsensical if those parties who were jointly insured under a contractors’ all risks policy would make claims against one another in respect of damage to the contract works. The implied term presupposes, of course, that the party relying on it has not by his own conduct prevented recovery of the loss under the policy - a point made by Jackson J (as he had by then become) in Board of Trustees of the Tate Gallery v Duffy Construction Ltd [2007] BLR 216.

In the present case the Court of Appeal followed the same reasoning in holding that the proper construction of clause 12 was that there was to be “an insurance funded result in the event of loss or damage to the vessel by marine risks” and that, if the demise charterers had been in breach of the safe port clause, they would have been under no liability to the owners for the amount of the insured loss because they had made provision for looking to the insurance proceeds for compensation. It did not consider that the introduction of clause 29 was intended to alter the way in which clause 12 was to operate.

I agree with the Court of Appeal. The demise charter allowed for a sub-demise with the owners’ consent, which was not to be unreasonably withheld. The risk existed that the vessel might be directed to an unsafe port, not necessarily by negligence on anyone’s part, so causing peril to the vessel, but the risk of consequential damage to the vessel was catered for by the insurance required to be maintained by the demise charterer in the joint names of itself and the owners. The commercial purpose of maintaining joint insurance in such circumstances is not only to provide a fund to make good the loss but to avoid litigation between them, or the bringing of a subrogation claim in the name of one against the other. I do not accept that by substituting clause 29 for clause 5 the parties intended to subvert that purpose. If anything, the present case is stronger in this regard than Co-operative Retail Services Ltd v Taylor Young Partnership Ltd, because in that case Hall’s obligation to use skill and care was in a separate contract from the contract between the owner and the contractor which contained the provisions about insurance whereas in the present case clauses 12 and 29 are part of a single contract. Gard’s submission that clause 29 becomes pointless if clause 12 has the effect held by the Court of Appeal is fallacious. It sets limits on the use of the vessel, breach of which may give rise to loss, but clause 12 deals with the consequences of loss or damage to the vessel, regardless of whether it resulted from negligence or other fault of the demise charterer (or a sub-charterer).

Mr Mark Howard QC argued that this interpretation misapprehends the purpose of clause 12, which in a case of loss caused by a breach of contract by the charterer does no more than to ensure that the owner’s right to recover damages is backed by an available fund. This in substance was the argument advanced unsuccessfully against Hall in Co-operative Retail Services Ltd v Taylor Young Partnership Ltd. Lord Hope (paras 39 and 40) distinguished between a provision for insurance which curtails the means of recovering loss whether or not it was caused by a contracting co-insured’s default, and a provision which backs the other party’s other obligations with an insuring obligation but leaves the other obligations enforceable against the other party by other means. He agreed with the judge’s conclusion that the contractual insurance arrangements meant that if a fire occurred, the owners were to look to the joint insurance policy to provide the fund for the cost of restoring and repairing the fire damage rather than, in the judge’s words, indulge in litigation with each other.

In the present case, if one were to ask whether it would have accorded with the parties’ intentions that on the morning after the loss the owners would have been entitled to demand immediate payment from the demise charterers, rather than make a claim on the insurers and wait for it to be settled, my answer would be that they intended no such thing. The insurance arrangements under clause 12 provided not only a fund but the avoidance of commercially unnecessary and undesirable disputes between the co-insured.

It does not follow that the demise charterers (or their insurers in their shoes) necessarily had no available remedy against the time charterers. The court was told that at one stage the insurers intimated a claim analogous to the claim which a bailee may bring under The Winkfield [1902] P 42. However, the matter was not pursued, and it would be inappropriate to express a positive view about the likelihood of success of such a claim (or any alternative), about which the court has heard no argument.

For those reasons I would have upheld the decision of the Court of Appeal on the recoverability issue, if the demise charterers had been in breach of the safe port clause. I have had the benefit of reading Lord Mance’s additional reasons, with which I agree. Like him, I do not think it is necessary to reach a final conclusion about the position in the case of an insolvent insurer, which it is not reasonable to suppose was in the minds of the BIMCO drafters of Barecon 89 and ought not to affect its core interpretation.