Hercultio Maritime Limited & Ors. v Gunvor International BV & Ors.

This is a claim by the owner of the mv POLAR to recover cargo’s proportion of general average expenditure, the expenditure in question consisting of a ransom payment to pirates who had detained the vessel in the Gulf of Aden. The claim is defended by the cargo owners on the ground that the shipowner’s only remedy in the event of having to pay a ransom to pirates was to recover under the terms of insurance policies, the premium for which had been paid by the voyage charterer. Whether that is a good defence depends on the construction of the contract contained in or evidenced by the bill of lading, which incorporated the terms of the charterparty.

An arbitration tribunal (Mr Timothy Young QC, Mr Dominic Kendrick QC and Mr Simon Gault) held that the cargo owners were not liable to contribute in general average. On an appeal to the Commercial Court under section 69 of the Arbitration Act 1996 Sir Nigel Teare reached the opposite conclusion but gave leave to appeal to this court.

By a charterparty dated 20^th September 2010 the shipowner chartered the vessel to Clearlake Shipping Pte Ltd (“the charterer”) for a voyage from one or two safe port(s) Tallin/St Petersburg range to one safe port Fujairah or, in charterer’s option, one or two safe port(s) or STS transfers in the Singapore area with a cargo of maximum two grades of fuel oil. No charterparty was drawn up, the terms of the contract being contained in a fixture recap which incorporated with some amendments the BPVOY 4 standard form together with some further additional claims. The fixture recap provided: “All above via Suez with Suez costs to be for Owners’ account”.

Clause 30.2 of the BPVOY 4 form provided for all bills of lading issued under the charter to be deemed to contain War Risks, Both-to-Blame Collision and New Jason clauses.

Clause 39 of the BPVOY 4 form was a detailed and lengthy war risks clause. It defined “War Risks” as including, among other things, “acts of piracy”. The clause was set out by the judge as an appendix to his judgment, but it was common ground that he summarised it fairly and accurately at [9] as follows:

Pursuant to clause 39.2 the owner was entitled to cancel the charter if, at any time before the vessel commenced loading, it was considered that performance of the contract of carriage might expose the vessel to war risks.

Pursuant to clause 39.3, the owner was not required to continue to load or to sign bills of lading or to proceed or continue on a voyage where it appeared that the vessel might be exposed to war risks. If it should so appear, the owner was entitled to request the charterer to nominate a safe port for the discharge of the cargo. If within 48 hours the charterer failed to nominate such a port, the owner was entitled to discharge the cargo at any safe port of its choice in complete fulfilment of its obligations under the charter. The extra expenses of such discharge were payable by the charterer.

Pursuant to clause 39.4, if, at any stage of the voyage, it appeared that the vessel might be exposed to war risks on any part of the route and there was another longer route to the discharge port, the owner was entitled to give notice to the charterer that this route should be taken. The extra expenses of such route, if the extra distance exceeded 100 miles, were payable by the charterer. An amendment to the standard form stated that these extra expenses included extra time taken and additional bunkers consumed.

Pursuant to clause 39.5, the owner was at liberty to comply with the orders of identified third parties, including governmental authorities and war risks underwriters.

Pursuant to clause 39.6, anything done or not done in compliance with the clause was not to be a deviation.

There were several additional clauses, including a Gulf of Aden clause (expressed to be “for this CP only”) and a further war risks clause. The Gulf of Aden clause provided that half of any time awaiting an escort or protection team or other protective measures would count against used laytime or (if applicable) as time on demurrage and that any additional costs of such measures (including time and bunkers) would be shared equally between the owner and the charterer. It continued:

The additional war risks clause provided that any additional premiums payable by the owner in respect of war risks were for the charterer’s account. Such premiums were payable by the charterer together with the freight against the owner’s invoice supported by appropriate documents.

It was common ground that the effect of these clauses was that the charterer would pay for the additional war risks and K&R cover up to a maximum of US $40,000. If the cover cost more than that, the shipowner would pay the balance.

A cargo of 69,493.28 mt of fuel oil was loaded at St Petersburg between 29^th September and 2^nd October 2010. Six bills of lading were issued. The shipper in each case was Warley International Ltd, part of the Rosneft group, and the consignee was “to the order of BNP Paribas (Suisse) SA”. The arbitrators found that the lawful holder of all six bills of lading throughout the voyage was Gunvor International BV, the receiver of the cargo in Singapore. As the arbitrators pointed out, however, that need not have been so, given that the six bills of lading were separately negotiable. It appears also that Clearlake and Gunvor were connected companies although, again, that need not have been so and cannot affect the construction of the bills of lading.

The port/place of discharge was stated as “Singapore for orders”. Five of the bills were on the INTANKBILL 78 form. The sixth bill had the face of that form, but the reverse of the Congenbill form. Bills of lading 1 to 5 contained these words of incorporation on their face:

Bill of lading 6 provided, on its reverse side:

It was not suggested that there was any material distinction for the purpose of this appeal between these two formulations.

In fact none of the bills of lading identified which charterparty was intended to be incorporated, but it was common ground that the material charterparty was that between the shipowner and Clearlake referred to above.

Each of the bills contained express general average clauses providing for general average to be settled in accordance with the York-Antwerp Rules 1974 (or 1994 in the case of bill of lading 6).

Bills of lading 1 to 5, but not bill of lading 6, had a vertical marginal note on the reverse reading:

Each of the bills contained on its face a statement that “By taking delivery of the cargo the Consignee shall make himself liable for unpaid freight, deadfreight, demurrage and other charges”.

The arbitrators found that the cargo was insured by cargo underwriters on what they described as “familiar terms”. They did not explain what they meant by this, but said that these terms were irrelevant for their purposes. However, it was common ground that the cargo was insured on the Institute Cargo Clauses (A) terms. These cover all risks of loss or damage to the cargo, with certain exclusions, one of which is war risks. However, the exclusion of war risks is itself subject to an exception for “piracy”. Accordingly, although in general war risks are excluded from the cover, loss or damage caused by piracy and its consequences was covered. It was common ground also that this is a standard term of marine cargo insurance.

The shipowner had annual hull & machinery and war risks insurance but, as with all such insurance, there were certain excluded or Additional Premium areas, one of which was the Gulf of Aden. By an endorsement to their war risks cover dated 26^th October 2010 the shipowner purchased insurance cover for the vessel’s proposed transit through this otherwise excluded area. In addition, it took out kidnap and ransom (“K&R”) insurance for a single voyage through the Gulf of Aden. This provided cover against the payment of ransom as a result of piracy up to a limit of US $5 million. It was not in dispute that the effect of this additional cover was that any ransom up to the limit of US $5 million would be reimbursed by the K&R underwriters. How the war risks insurance would respond to any ransom in excess of that amount was not the subject of any argument before us.

As it happened, the combined premiums for this K&R and additional war risks cover fell just short of the figure of US $40,000 contained in the Gulf of Aden clause in the charterparty. The premiums were paid by the shipowner but reimbursed by the charterer in accordance with the charterparty terms. (The shipowner also took out loss of hire insurance which took the total premium above US $40,000, but this was for its own account).

While transiting the Gulf of Aden on 30^th October 2010, the vessel was seized by Somali pirates. She was held captive for ten months before being released on 26^th August 2011. In order to obtain this release, a ransom payment of US $7.7 million was paid, funded by the K&R and war risks underwriters.

Some of the cargo was abstracted during the period of seizure, but most of it remained intact and was carried to its destination. After diverting for repairs, supplies and re-crewing, the vessel continued the voyage and finally delivered the cargo at Singapore.

The shipowner declared general average and, in order to obtain delivery of the cargo, the cargo underwriters provided a general average guarantee dated 16^th September 2011, while Gunvor as the cargo owner provided a general average bond dated 28^th September 2011. The guarantee and the bond provided for arbitration in London. In due course a general average adjustment was issued, concluding that approximately US $4.8 million (in very broad terms, approximately 60% of the ransom payment) was due to the shipowner from the cargo interests. This reflected the respective values of the vessel and the cargo.

The shipowner and its various underwriters, and the mortgage bank, claimed this sum from Gunvor under the average bond and from the cargo underwriters under the average guarantee. Two arbitrations were commenced, one under the bond and the other under the guarantee, but the same arbitrators were appointed in each and they have been dealt with together.

The arbitrators ordered that two preliminary issues should be determined. These were, in essence:

Were the terms of the voyage charter, including in particular the war risks and Gulf of Aden clauses, incorporated into the bills of lading?

If so, did the shipowner, on the true construction of the bills of lading and/or by implication, agree to look solely to its insurance cover under the war risks and/or K&R insurance in the event of a loss covered by that insurance?

The arbitrators held that the answer to both questions was yes. In outline, an outline which may not do full justice to a careful and thorough award, their reasoning proceeded as follows:

The war risks and Gulf of Aden clauses fell within the widely expressed incorporating language (“all terms and conditions, liberties and exceptions”) of the bills of lading.

Clause 30.2 of the charterparty demonstrated an intention that war risks clauses should be incorporated into the bills of lading, although that did not answer the question of which war risks clauses should be so incorporated; however, what really mattered was the Gulf of Aden clause.

Leaving aside (in the light of The Miramar[1984] AC 676) those aspects of the Gulf of Aden clause concerned with demurrage, the Gulf of Aden and additional war risks clauses were “germane to the carriage” of the cargo (an expression derived from the speech of Lord Atkinson in T.W. Thomas & Co Ltd v Portsea Steamship Co Ltd[1912] AC 1, 6 and hallowed by long usage). They affected the route to be taken which was of central importance to the adventure and the legal relations between carrier and cargo. If these clauses were not incorporated into the bills, there would be a disconnect between the charterparty and the bill of lading contracts: the shipowner might exercise liberties contained in the charterparty which would be wrongful under the bills of lading.

Accordingly these clauses were incorporated into the bills, irrespective of whether the effect of doing so was to impose an obligation on the bill of lading holders to pay the insurance premiums.

Turning to the effect of the incorporation of these terms, the arbitrators began by considering their effect as between the shipowner and the charterer. With some hesitation, they concluded that the parties had made specific allocation of the well-known risk of piracy in the Gulf of Aden, by establishing a fund for the payment of ransom in the form of insurance for which the charterer was to pay, and (applying the reasoning of Lord Roskill and Lord Brandon in The Evia (No. 2)[1983] AC 736) that it would be a “remarkable” and wrong result if the charterer had to pay the premium but was then liable to the insurers (exercising rights of subrogation), effectively reimbursing the insurers for the loss against which they had insured. In this regard, it made no difference that the claim in The Evia (No. 2) had been a claim for breach of contract (a breach of the safe port warranty) while the present case was a claim for general average.

Accordingly, under the charterparty, the shipowner had agreed not to claim contribution from the charterer for losses resulting from the operation of piracy risks in the Gulf of Aden. That was in effect a “code” whereby the shipowner would look exclusively to the insurance cover for which the charterer had paid.

The next question was whether that “code” was incorporated into the bills of lading. The arbitrators’ view here was that it did not matter whether the effect of incorporating the war risks and Gulf of Aden clauses into the bills of lading was to impose an obligation on the cargo owners to pay the relevant premiums. What mattered was that “the charter code” involved an agreement by the shipowner not to seek contribution for piracy losses. It was not a case of incorporating a positive obligation on cargo, but an agreement by the shipowner excusing cargo from liability. That agreement was incorporated into the bills of lading.

However, although they did not think it was necessary, if necessary the arbitrators would have been prepared to “manipulate” the word “charterer” to mean “lawful holders of the bills of lading” so as to impose an obligation on the bill of lading holders to pay the premium, although they recognised that in reality the occasions when the shipowner would actually look to the bill of lading holders for such payment would be “vanishingly small”.

The arbitrators found as a fact that at the material time the risk of piracy in the Gulf of Aden and the growth of K&R insurance were well known. The risk of seizure by pirates in order to obtain a ransom payment was a distinct risk, different from ordinary navigational risks such as grounding or other seafaring casualty. When concluding the charterparty, the parties had focused specifically on that risk and how, in principle, it was to be paid for, namely by the relevant insurers.

The shipowner obtained permission to appeal under section 69 of the Arbitration Act 1996 on two questions of law. These were:

Are agreements between a shipowner and a charterer in a charterparty which delineate the responsibility for the payment of additional war risk or K&R premia between the shipowner and the charterer germane to the carriage of the vessel’s cargo in the context of the bill of lading?

Does an agreement between a shipowner and bill of lading holder concerning the allocation of responsibility for the payment of war risks, machinery and K&R insurance premia give rise to an exclusive insurance fund precluding the shipowner from recovering in GA a contribution from cargo interests in respect of any losses suffered as a result of perils falling within the insurances?

It seems to me, with respect, that these are rather general questions to which the answer is likely to be “it all depends”. The real question, as Mr Justice Andrew Baker observed when granting permission to appeal, is whether, properly construed, the bills of lading excluded liability on the part of the bill of lading holders in respect of cargo’s contribution in general average in the event the vessel encountered a peril insured under any of the insurances referred to in clause 39, the Additional War Risks clause or the additional Gulf of Aden clause of the voyage charter.

The judge noted that disputes about the incorporation of charterparty terms into bills of lading have a long history, although the present case is unusual in that it is the bill of lading holders who contend that the terms in question were incorporated and the shipowner who resists this. However, whether the relevant terms were incorporated depends upon an objective process of construction of the bills. Having summarised the arbitrators’ approach to this process, which was not suggested to be in error, the judge (in an equally careful and thorough judgment) proceeded as follows:

It was not disputed that the words of incorporation in the bills of lading are wide enough to incorporate the war risks and Gulf of Aden clauses.

It was necessary to deal with the component parts of the clauses separately. Although they comprised a “package”, the cargo owners accepted that some elements of the package (e.g. relating to laytime and demurrage) were not incorporated.

The liberties given to the shipowner not to continue with the voyage or to deviate from the usual route were germane to the loading, carriage and discharge of the cargo and were therefore incorporated into the bills of lading. However, the charterer’s obligation to bear the expenses caused by the exercise of such liberties was not incorporated. That was because to impose such an obligation on the bill of lading holders would be inconsistent with their agreement to pay freight “as per Charter Party” for the performance of the contract of carriage by the shipowner: they had agreed to pay freight, but not to pay additional expenses, which did not fall within the reference to “other charges” on the face of the bills.

Turning to the Gulf of Aden clause, the judge recorded that it was common ground that the provisions relating to time used awaiting an escort or protection team or the implementation of protective measures were not incorporated into the bills of lading. Nor were the provisions for sharing of additional costs, time or bunkers incurred following a fixed route, entering a convoy, deviating to pick up or drop off a protection team or implementing protective measures or deviating from the usual route.

So far as the liability for additional premium was concerned, the judge’s view was that it was important to decide whether, when reading these clauses into the bills of lading, it was appropriate to substitute “bills of lading holders” for “charterer” so as to impose an obligation on the bill of lading holders to pay this premium. He held that it was not appropriate to do so. While the obligation to pay for additional war risk and for K&R insurance was directly germane to the carriage and delivery of the cargo, there were a number of reasons why it was not appropriate to read the bills of lading as imposing an obligation on the bill of lading holders to pay the premium. One reason was that to do so would be inconsistent with the provisions in the bills for payment of freight. Another was the uncertainty as to how responsibility for the premium would be allocated between different bill of lading holders or (if each holder was jointly and severally liable for the full amount) what rights a holder who had paid in full would have to recover a proportion of the premium from other bill of lading holders.

Accordingly the judge concluded that the part of the Gulf of Aden clause which referred to payment of additional premium, although incorporated into the bills of lading, did not impose any liability on the bill of lading holders.

On that basis, he turned to the question whether the bills of lading gave rise to an exclusive insurance fund precluding the shipowner from recovering a general average contribution from cargo interests in respect of losses suffered as a result of perils falling within the insurances. He dealt with this question in two stages.

The first stage was to consider the position as between shipowner and charterer under the charterparty. Here he held, in agreement with the arbitrators, that the parties had agreed that the shipowner would look to the insurers for indemnification in respect of such losses and not to the charterer. Accordingly the shipowner was precluded by that agreement from seeking to recover the loss by way of a contribution in general average from the charterer.

That being the position under the charterparty, the next question was to consider whether those parts of the charterparty which were incorporated into the bills of lading (which did not include the obligation to pay premium) had the effect of precluding the shipowner from recovering a contribution from the bill of lading holders. The judge held that they did not. In contrast with the position in The Evia (No. 2) and The Ocean Victory[2017] UKSC 35, [2017] 1 WLR 1793, both charterparty cases, the defendants seeking to take the benefit of the insurance as precluding a claim against them (here the bill of lading holders) had not agreed to pay the premium. Therefore the “remarkable result” described by Lord Roskill in The Evia (No. 2) whereby the party paying the premium would be liable to the insurers exercising subrogation rights did not arise here. That being so, there was no proper basis for concluding that, in the event of general average arising from payment of a ransom to pirates, the shipowner had agreed not to look to the cargo owners for a contribution. All that the shipowner had agreed was that it would not seek a contribution from the charterer.

For the cargo owner Mr Stephen Hofmeyr QC made two principal submissions:

First, that regardless of whether the bill of lading holders were under any obligation to pay the premium, the shipowner’s agreement in the charterparty to look exclusively to its insurance in the event of detention by pirates was incorporated in the bills of lading for the benefit of the cargo owners. This was the obvious commercial purpose of this agreement. It was the cargo owners who had the greatest interest in the successful carriage of the cargo and the greatest exposure to liability in general average for ransom payments, a risk which was well known and for which the charterparty code was intended to provide. However, the judge’s analysis meant that the bill of lading holders took the burden of the charterparty regime (e.g. the various liberties given to the owner to deviate from the proposed voyage through the Suez Canal or to deliver the cargo at an alternative port) but without the benefit of the insurance cover.

Second, that (if it mattered) the terms of the bills of lading should be “manipulated” so as to impose on the holders an obligation to pay for the additional premium up to a maximum of US $40,000. The bill of lading holders would then be in the same position as the charterers to resist any contribution in general average on the ground that, being liable to pay the premium, they should not then be liable to the insurers (exercising rights of subrogation) for the loss against which the insurers had insured. Such a liability was not inconsistent with the terms of the bills of lading and gave rise to no difficulty: each bill of lading holder was jointly and severally liable for the full amount of the premium up to US $40,000.

For the shipowner Mr Guy Blackwood QC submitted as follows:

Without manipulation of the reference to the charterer to include the bill of lading holder, the obligation to pay premium was an obligation of the charterer and not the bill of lading holders and amounted to nothing more than an agreement by the shipowner not to seek a contribution in general average from the charterer. The position of the bill of lading holders was therefore not comparable to that of the charterers in The Evia (No. 2) and The Ocean Victory. Moreover, it was unlikely that the shipowner would have intended to give up its right to a contribution in general average from the cargo interests, particularly in circumstances where it would have been understood that, in accordance with market practice, the cargo interests would have their own insurance against such liability. Clear words would be required for such a conclusion. Moreover, The Evia (No. 2) and The Ocean Victory were distinguishable because they were cases of breach of contract as distinct from general average.

Manipulation of the bill of lading to impose an obligation to pay the premium on the bill of lading holders should be rejected as commercially unreal. There might be multiple bill of lading holders, but there was no way to determine how liability for the premium should be allocated between them. If each bill of lading holder was jointly and severally liable for the full amount, there would be legal and practical difficulties in obtaining reimbursement inter se. In any event, five of the bills of lading contained an express provision defining “Charterer” as “the person entering a Charter Party”.

Although these were the shipowner’s principal submissions in resisting the appeal, it also contended, by a Respondent’s Notice, that the judge was wrong to have concluded (1) that the part of the Gulf of Aden clause dealing with additional premium liability was incorporated into the bill of lading and (2) that the effect of the charterparty was to prevent the shipowner from seeking a contribution in general average from the charterer in the event of a ransom payment.

The principles applicable to the incorporation of charterparty terms into a bill of lading contract have been developed in a series of cases, many of which have been concerned with arbitration clauses. The approach developed in these cases is described in Scrutton on Charterparties, 24^th Edition (2020) at paras 6-016 to 6-018 (omitting footnote citations of well-known cases such as Thomas v Portsea[1912] AC 1, The Annefield [1971] P 168, The Miramar[1984] AC 676 and The Channel Ranger[2014] EWCA Civ 1366, [2015] QB 366):

It is important to keep firmly in mind that any question about the incorporation of charterparty terms into a bill of lading is a question of construction of the bill of lading contract. Thus the aim of the exercise on which the court is engaged is to ascertain the objective meaning of the terms which the parties to the bill of lading contract have used to express their agreement, taking account where appropriate of the commercial background as it would have been understood by both parties (cf. cases such as Wood v Capita Insurance Services Ltd[2017] UKSC 24, [2017] AC 1173). The process of construing a bill of lading is in principle no different from that of any other contract, although it needs to take account of the particular features of such contracts. As the Supreme Court has explained, construction of contracts is a single “iterative” process. That means, in this context, that while it is convenient to approach the issue by reference to the sequence of steps described in Scrutton, these should not be too rigidly applied. Moreover, it is important at each subsequent stage to be prepared to revisit a conclusion reached at an earlier stage, and to stand back at the end of the process to test the conclusion reached against the terms of the contract and business common sense. In this regard Scrutton’s reference to “prima facie incorporation” is helpful as indicating that, at each preliminary stage, the conclusion reached can be no more than provisional.

Particular features of a bill of lading contract which are relevant to issues of incorporation include the following. First, although the contract is concluded between the carrier (typically the shipowner) and the shipper, contractual rights and obligations are intended to be transferred in accordance with the Carriage of Goods Act 1992 as the bill is negotiated (often down a chain of contracts) to the ultimate receiver who presents it at the discharge port and takes delivery of the cargo. Second, whereas the charterparty is a single indivisible contract between the shipowner and the charterer covering the whole adventure from (at latest) the vessel’s arrival at the (first) load port until completion of discharge at the (final) discharge port, there may be numerous bills of lading issued, each of which only comes into existence as cargo is loaded on board and relates to the particular cargo covered by the bill. Bills of lading on a single voyage may be for different cargoes and may be negotiated to a variety of receivers, who may be located in different discharge ports or even different countries. When the bill of lading contract is concluded, the identity of the receivers and the proposed discharge port(s) may not be known and, even if they are, plans may have changed by the time the vessel arrives at its destination. Although in the present case all six bills of lading were negotiated to a single receiver at a single discharge port, this need not necessarily have been so, as the arbitrators pointed out, and cannot be assumed. Third, although in one sense English law allows considerable slackness on the part of those who complete bills of lading, so that a charterparty may be incorporated even though it is not properly identified in the bill, it must be remembered that the cargo interests may have no knowledge of the charterparty terms and no ready means of finding out what they are when they have to decide whether to accept negotiation of the documents under their purchase contracts. That may make it easier to conclude, as in The Miramar, that the parties to the bill of lading contract did not intend to incorporate an onerous or unusual clause even if that clause is directly relevant to the carriage of the cargo.

I should add that Mr Hofmeyr emphasised (more than once) that the arbitrators in this case were “an experienced tribunal”. I accept, of course, that that is so, but the same can be said of the judge. However, this is not a case which turns on issues of market practice or understanding on which the views of commercial arbitrators might be entitled to particular weight. The question is one of construction of the bill of lading contracts in which the background, although important, is clear and uncontroversial.

A logical first step is to consider whether, in the charterparty, the shipowner agreed not to seek a general average contribution from the charterer in the event of a ransom payment to pirates seizing the vessel in the Gulf of Aden. (Footnote: ²) Unless that question is answered affirmatively, the issue of incorporation into the bill of lading does not arise: there is nothing of relevance to be incorporated. The arbitrators and the judge did answer the question affirmatively, but the shipowner challenges this conclusion by its Respondent’s Notice.

It should be noted that the issue here is solely concerned with allocation of the risk of detention of the vessel by pirates while passing through the Gulf of Aden. It is not suggested that the shipowner agreed not to seek general average contributions as a result of any other kind of general average event. Further, although the war risks clauses in the charterparty were modified by the shipowner’s agreement to proceed through the Suez Canal on the basis that insurance would be paid for by the charterer up to a maximum of US $40,000, those clauses continued to apply in the event of other war risks arising.

In The Evia (No. 2) the House of Lords held that a clause in a time charterparty providing for the charterer to pay for insurance against war risks had the effect of releasing the charterer from liability (in that case, for potential breach of a safe port warranty) for loss and damage covered by that insurance. Any other conclusion would be the “remarkable result” to which Lord Roskill referred whereby “the charterers would have paid the premiums not only for no benefit for themselves but without shedding any of the liabilities which [the safe port warranty] would, apart from clause 21, impose on them”. Similarly in The Ocean Victory, another safe port case, a clause in a bareboat charterparty provided for joint insurance of the vessel in the name of both owner and charterer. A majority of the Supreme Court (Lord Mance and Lord Toulson, with whom Lord Hodge agreed on this issue) held that there could be no claim in respect of losses covered by the insurance.

The judge considered that the majority in The Ocean Victory had approved the approach of Lord Justice Longmore in this court:

In my view it is questionable whether the majority of the Supreme Court went so far as to approve these passages from Lord Justice Longmore’s judgment. Lord Toulson said that he agreed with the Court of Appeal, but did not expressly refer to these passages. Lord Mance said that he agreed with Lord Toulson, as well as adding some observations of his own. Lord Toulson made clear at [139] that the question was one of construction, the question in each case being whether the parties should 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-existed with an independent right of action for breach of a term of the contract which had caused that loss. He added that, like all questions of construction, the answer depends on the provisions of the particular contract. He did not start from a prima facie position that in these circumstances the parties will have looked to the insurers for indemnification. The existence of two dissenting judgments, by Lord Sumption and Lord Clarke, demonstrates that this question may be difficult to answer. For the majority it was an important feature of The Ocean Victory that both parties were to be named as insureds. Accordingly, what Lord Mance described as the “well established” principle “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” applied (see Co-operative Retail Services Ltd v Taylor Young Partnership Ltd[2002] UKHL 17, [2002] 1 WLR 1419).

That feature was not present in The Evia No. 2, where the charterer had an obligation to pay for the insurance, but was not to be a named insured. As demonstrated by Mr Oliver Caplin, who argued this point on behalf of the shipowner, the conclusion that clause 21 of the charterparty in The Evia No. 2 constituted a complete code was not based solely on the charterer’s obligation to pay for the insurance. Other features of the clause were also important, including that the vessel was to remain on hire (notwithstanding the off-hire clause elsewhere in the charter) during any time lost owing to the master, officers or crew refusing to proceed to a war zone or to be exposed to war risks.

In the present case there was no provision for the charterer (let alone the bill of lading holders) to be named as joint insured with the shipowner. Moreover, the charterer’s obligation was to make a contribution to the cost of additional war risks and K&R insurance up to a maximum of US $40,000 which might or might not be sufficient to cover the full cost. It is, therefore, a weaker case than either The Evia (No. 2) or The Ocean Victory for concluding that the shipowner agreed not to seek a general average contribution from the charterer.

It is, however, unnecessary to decide this question. I can proceed on the basis that the charterparty does include an agreement by the shipowner not to seek a general average contribution from the charterer in the event of a ransom payment to pirates seizing the vessel in the Gulf of Aden. If such an agreement exists, that is because, although not expressed, it is implicit in what the parties have agreed. As Lord Mance put it in The Ocean Victory at [114] where there was co-insurance, the principle that the parties cannot claim against each other in respect of an insured loss rests “on the natural interpretation of or implication from the contractual arrangements giving rise to such co-insurance”. That is equally so when the same conclusion is reached even without the parties being named as co-insured.

On the assumption that the charterparty does include an agreement by the shipowner not to seek a general average contribution from the charterer in the circumstances which occurred, the question is whether that agreement was incorporated into the bills of lading in such a way that the shipowner agreed also not to seek a contribution from the bill of lading holders. I approach that question in stages.

First, I agree with the arbitrators and the judge that the incorporating words in the bills of lading (“all terms and conditions, liberties and exceptions”) are extremely wide. They are sufficiently wide to encompass the war risks and Gulf of Aden clauses in the charterparty. It is doubtful, however, whether they are sufficiently wide to encompass what is merely implicit in the charterparty’s express terms considered as a whole. Accordingly, if we are to find in the bills of lading an agreement by the shipowner not to seek a general average contribution from the cargo owners, that must be because the express terms of the charterparty which are incorporated into the bills demonstrate that the same (or an equivalent) agreement was intended to apply also under the bills of lading.

It is plain, however, that not all of the charterparty terms dealing with war risks were intended to be incorporated into the bills. Those parts of clause 39 of the BPVOY4 form and the additional war risk clause which govern the position before completion of loading have no place in a bill of lading contract. Further, the cargo owner did not suggest that the parts of the Gulf of Aden clause which deal with time counting for laytime and demurrage and with the sharing of expenses between the shipowner and the charterer were incorporated. Accordingly, to the extent that the charterparty constitutes a complete code dealing with the risk of piracy during transit through the Gulf of Aden, any code incorporated into the bills of lading is different, at least in some respects, from the charterparty code.

Nevertheless I agree with the arbitrators and the judge that the provision for the charterer to pay for the additional war risks and K&R insurance was directly relevant to the carriage and discharge of the cargo. The route which the vessel was to take, that is to say through the Suez Canal and therefore through the Gulf of Aden rather than round the Cape of Good Hope, was plainly of direct relevance. The shipowner’s agreement to proceed by that route, and not to exercise the liberty to deviate around the Cape which clause 39 of the BPVOY4 form would otherwise have allowed, was conditional on and inextricably linked with the availability of insurance and the charterer’s agreement to pay the premium. Prima facie, therefore, that part of the additional war risks and Gulf of Aden clauses was incorporated into the bill of lading contracts.

It is convenient, and I think logical, to consider next whether, when prima facie incorporated into the bill of lading contracts, the requirement on the charterer to pay the premium should be “manipulated” so as to impose that obligation on the bill of lading holders. I am not persuaded that to do so would be inconsistent with the provisions of the bills of lading concerning freight, but I agree with the judge that it is not appropriate to engage in such manipulation. As the judge explained, there is nothing in the bills (or the charterparty) to say how liability for the premium would be apportioned between different bill of lading holders. But an individual bill of lading holder would want to know for how much of the premium it was liable. It is by no means obvious either that each bill of lading holder should be jointly and severally liable for the full amount or that each bill of lading holder should be liable for a proportionate amount of the premium, whether calculated by reference to value or volume (which, where more than one grade of cargo was allowed, would not necessarily be the same thing). The fact that neither the bills nor the charterparty addresses this question tends to suggest that the bill of lading holders were not intended to be liable for the premium.

Moreover, if (as the cargo owners contend) each bill of lading holder was to be jointly and severally liable for the full amount, a holder called upon to pay would want to know what rights of reimbursement it had against other bill of lading holders. However, the exercise of such rights would give rise to considerable practical difficulties where bill of lading holders could be located at different ports or even in different countries and might not even know each others’ identities. That such matters are not addressed in the bills (or the charterparty) is a further indication that the bill of lading holders were not intended to be liable for the premium.

It is no answer for the cargo owners to say that the amount of the premium was modest in the overall scheme of things, that the shipowner would only look to bill of lading holders for payment of the premium in rare cases, or that such difficulties could be worked through if and when they arose. If anything, the fact that “the occasions when the Owners would actually look to the bill of lading holders would be vanishingly small” (as the arbitrators put it) tends against any suggestion that the parties intended to impose such a liability on them. There is little point in manipulating the language of the charterparty in order to impose a liability on bill of lading holders which will only be invoked in a vanishingly small number of cases.

I conclude, therefore, that manipulation is not appropriate. Like the judge, I do so without needing to consider the effect, if any, of the vertical marginal note contained in bills 1 to 5 but not bill 6.

I have so far reached the provisional conclusion that the charterparty terms requiring the charterer to pay for the additional war risks and K&R insurance were prima facie incorporated into the bills of lading, but that they should not be manipulated so as to impose a liability on bill of lading holders to pay the premium. Since those clauses therefore impose no liability on the bill of lading holders, it must be asked what they are doing in the bills of lading and whether their incorporation serves any useful purpose. If not, the issue of incorporation would need to be revisited as there may be no point in incorporating a term which has no effect.

Mr Hofmeyr submitted that the purpose of incorporating the charterparty clauses into the bills of lading was to ensure that the shipowner’s rights and obligations vis-à vis the bill of lading holders were not materially different to their rights and obligations vis-à vis the charterer. It would therefore be nonsensical to incorporate the charterparty code into the bills of lading on the ground that it is germane to the carriage of the cargo, but nevertheless to treat it merely as a record of an agreement between the shipowner and the charterer. I do not agree. In my judgment the incorporation of these terms does serve a useful purpose. It makes clear, even if only as a record of the terms agreed between the shipowner and the charterer, the basis on which the shipowner has agreed in the bill of lading contract that the voyage will be via Suez and the Gulf of Aden. That basis is that the shipowner will have insurance against the risk of piracy, albeit paid for by the charterer and not the bill of lading holders. Without the incorporation of these terms the important issue of the vessel’s route under the bill of lading contract would at best be highly uncertain.

I have now identified the terms of the bills of lading which have to be construed and can come to the ultimate question in this appeal, which is whether the bills excluded liability on the part of the bill of lading holders to pay cargo’s contribution in general average in the event the vessel encountered a peril insured under any of the insurances.

Mr Hofmeyr advanced a powerful argument that they did have this effect. In particular, he submitted that it was the cargo owners rather than the charterer which would be directly concerned with any general average contribution in the event of a ransom payment and that the shipowner would inevitably look to the cargo owners, as happened in this case, by insisting on an average bond from the cargo owners and an average guarantee from their insurers, as a condition for delivering the cargo at the discharge port. Accordingly the premium paid by the charterer under the war risks and Gulf of Aden clauses could properly be regarded as paid for the benefit of the bill of lading holders. There is no reason, he submitted, why one party should not agree to pay premium for the benefit of another on the basis that its counterparty will not seek a contribution from the party for whose benefit the premium is paid in the event of an insured loss. This would give effect to the obvious commercial purpose of the requirement on the charterer to pay the premium, which was the quid pro quo for the various liberties given to the shipowner in these clauses.

Despite the force of this argument, in the end I do not accept it. The risk of piracy and the potential need to pay a ransom were not only foreseeable but were foreseen by the parties to the bill of lading contract and dealt with expressly by them. They knew that, if the shipowner was to allow its vessel to transit the Gulf of Aden, insurance against this specific risk would need to be taken out. They knew or can be taken to have known that payment of such a ransom would give rise to general average. It would have been straightforward in the circumstances, if that is what they intended, to say in terms that cargo was not to contribute in general average in the event of such a payment. But they did not do so, instead leaving an implicit understanding to this effect to be inferred (on the cargo owners’ case) by reference to complex arguments concerning the incorporation of charterparty terms into a bill of lading contract. That seems an unnecessarily convoluted way to express a simple concept, which at least calls into question whether this is what the parties intended.

In the circumstances it seems to me that the “presumption” referred to in Gilbert-Ash (Northern) Ltd v Modern Engineering (Bristol) Ltd[1974] AC 689, 717 applies, that is to say that “in construing such a contract one starts with the presumption that neither party intends to abandon any remedies for its breach arising by operation of law, and clear express words must be used in order to rebut this presumption”. That may state the position too strongly, but in the present case there are no clear express words to rebut the presumption that the shipowner did not intend to abandon its right to a contribution from the cargo owners in general average and any “implicit understanding” (to borrow Lord Mance’s term in The Ocean Victory) is not so clear that we can be confident that it is what the parties intended, not least where the charterer was not necessarily paying the whole of the additional premium which would be necessary to obtain the cover required.

Nor do I accept that there is any commercial imperative for the cargo owners’ construction. Mr Hofmeyr began his submissions by pointing out that the shipowner, although the nominal claimant, has no financial interest in the claim, which is brought for the benefit of its insurers exercising rights of subrogation, and that the shipowner had not even paid for this insurance, the premium for which was paid by the charterer. He added (although of course this is inherent in any subrogated claim) that the insurers had received premium to cover precisely the event which had occurred, namely the payment of a ransom to pirates. He submitted that this demonstrated the remarkable and uncommercial nature of the shipowner’s claim. However, this is a case in which both parties were insured. The arbitrators did not think that the cargo owners’ insurance was relevant but, with respect, I disagree. The cargo owners likewise have no financial interest in this claim, as their liability to contribute in general average, including in the event of piracy, was expressly insured under their own insurance, which was in standard terms. The points made by Mr Hofmeyr therefore apply with equal force to the cargo owners.

In reality this is a case where both parties were insured against the risk of piracy and where allowing the shipowner to claim will mean that each set of insurers will bear its proper share of the risk which it has agreed to cover. In contrast, the effect of construing the bills of lading to exclude a claim by the shipowner will mean that the loss is borne entirely by the shipowner’s insurers and that the cargo owners’ insurers escape liability for a risk which they agreed to cover.

Standing back, therefore, in my judgment the judge’s conclusion accords with both legal principle and commercial sense.

I would dismiss the appeal.

I agree.

I also agree.