Glencore Energy UK Ltd & Anor v Freeport Holdings Ltd (The 'Lady M')

The primary issue raised on this appeal is whether article IV rule 2(b) of the HagueVisby Rules is capable of exempting the carrier from liability to the cargo owner for damage caused by fire if that fire were caused deliberately or barratrously. The appeal also raised a potential issue as to the mental element for an act of barratry¹.

In the early hours of 14 May 2015, while the ‘Lady M’ (‘the vessel’) was in the course of a voyage from Taman in Russia to Houston in the USA, a fire started in the engine room. As a result, the owners of the vessel (‘the Owners’) engaged salvors and the vessel was towed to Las Palmas, where general average was declared.

The appellants (‘Glencore’) brought proceedings in the Commercial Court claiming (as owners of a cargo of approximately 62,250 m.t. of fuel oil carried on board the vessel) such sums as it had incurred to the salvors, as well as the costs of defending the salvage arbitration proceedings. Its claim was founded on alleged breaches of contracts of carriage contained in or evidenced by four bills of lading dated 28 April 2015, alternatively in bailment.

The contracts of carriage were subject to the Hague-Visby Rules which, so far as material, provide as follows:

The term ‘barratry’ includes every wrongful act wilfully committed by the master or crew to the prejudice of the owner, or, as the case may be, the charterer.

Glencore pleaded its claim against the Owners in conventional form, relying on article III rules 1 and 2. It also contended, without prejudice to the burden of proof, that:

Its pleading went on to set out facts supportive of the case that the fire had been deliberately started, although Glencore was unable to say which individuals among master and crew had set the fire. At §20(4) the pleading contained the following:

By their Defence the Owners took issue with the Particulars of Claim in a number of respects but pleaded a positive case on the cause of the fire:

There were further averments in support of the contentions set out in §13c, but for present purposes one can pick up the Defence at §16.

It is unnecessary to say anything further about the pleadings other than to note that in its Amended Reply Glencore reiterated that a defence under article IV.2(b) was not available to the Owners because it did not apply where a fire was caused by barratry.

On 16 June 2017, Sara Cockerill QC (as then she was), sitting as a Deputy High Court Judge of the Commercial Court, ordered the hearing of two preliminary issues on the basis of agreed and assumed facts. So far as relevant to this appeal, these agreed and assumed facts were as follows:

On the basis of these facts, and against the general background of the casualty, two preliminary issues were identified: (1) whether, on the basis of the agreed and assumed facts, the conduct of the Chief Engineer constituted barratry; and (2) if so, whether the Owners were precluded from relying upon article IV.2(b) and/or 2(q) of the Hague-Visby Rules.

In the course of argument at the trial of the preliminary issues, Mr Justice Popplewell (‘the Judge’) agreed to a slight expansion and refinement of the issues, which he set out by reference to each party’s case.

Glencore contended, on the basis of the agreed and assumed facts, that:

On the same basis, the Owners contended that:

In the course of a full and careful analysis, the Judge concluded on the first question that whether the conduct of the Chief Engineer constituted barratry depended on further facts that would need to be found as to his state of mind; but that the issue was not determinative of whether the Owners were exempt from liability for the fire under article IV.2(b) or (q). So far as the second question was concerned, he found that article IV.2(b) was capable of exempting the Owners from liability if the fire were caused deliberately or barratrously. On the third question, he held that the Owners were not exempt from liability for the fire under article IV.2(q).

Glencore appealed against the Judge’s decision on the basis that: (1) on the agreed and assumed facts, the conduct of the Chief Engineer in starting the fire constituted barratry and that this conclusion did not depend on a close analysis of his state of mind at the time; and (2) the article IV.2(b) defence was not available where the fire was caused by the barratrous act of the Master or crew.

It is convenient to start with the second issue.

The issue involves the interpretation of the phrase: ‘Fire, unless caused by the actual fault or privity of the carrier’ in the Hague-Visby Rules (Footnote: ¹); and the first question that arises is the correct approach to the interpretation of the Hague-Visby Rules, and in particular these 12 words, in order to see whether they operate to exclude the liability of carriers for fire caused deliberately or barratrously by a crew member.

The parties cited a large number of authorities and other material in support of what they said was the proper approach to construction.

In broad summary, Mr Thomas QC submitted on behalf of Glencore that the defences in article IV.2 were based on standard forms of exclusion clauses which had been used in contracts of carriage prior to the establishment of the Hague Rules; and it followed that as a matter of English law the meaning and effect of words used in such standard clauses should inform the operation and effect of the article IV.2 defences.

In answer, Mr Hill QC argued that the Judge was correct in his interpretation of article IV.2(b). The words are clear and emphatic, and set out an exception for all loss or damage arising or resulting from fire, subject to the proviso: where the fire is caused with the actual fault or privity of the carrier. Glencore’s interpretation would require a further implied proviso to be added, ‘or the barratry of master or crew’. There is no proper basis for implying such words, not least because ‘barratry’ is not a relevant concept in the Hague Rules. The relevant interpretative rules require that it is only if the words of the Hague Rules are unclear, that it is permissible to look at their background; and Glencore’s wide-ranging search for a prior meaning of words which are clear was plainly impermissible.

The history of the Hague Rules begins with the International Law Association conference in Gray’s Inn between 17 and 20 May 1921, which produced an early draft. A few months later the International Law Association Conference took place at The Hague between 30 August and 3 September 1921 (‘The Hague Conference’). This involved negotiations between representatives of different commercial interests (primarily cargo interests and carriers); and redrafting by the Maritime Law Committee; and formed the travaux préparatoires, whose admissibility was in dispute on the appeal. The negotiations culminated in an agreed text which became known as the 1921 Hague Rules.

In October 1922 there was a further conference of the Comité Maritime International in London, at which further amendments were negotiated and agreed, in what became known as the 1922 Hague Rules or London Rules. Shortly thereafter, a diplomatic conference in Brussels appointed a sous-commission to consider the Rules further. It was after meetings of the sous-commission in Brussels in 1922 and 1923 (Footnote: ³) that the final version of the Hague Rules was adopted at the Brussels Conference on 25 August 1924 as the International Convention for the Unification of Certain Rules of Law relating to Bills of Lading.

At [27] of his judgment, the Judge referred to a number of authorities on the correct approach to the interpretation of the Hague Rules by the English Courts. These included: Stag Line v. Foscolo, Mango & Co Ltd [1932] AC 328, Lord Atkin at 342-3 and Lord MacMillan at 350; Aktieelskabet de Danske Sukkerfabriker v. Bajamar Compania Naviera S.A. (The Torenia) [1983] 2 Lloyd’s Rep 210, Hobhouse J at 219;

The Judge’s summary succinctly summarises what is clear and binding authority; and I shall confine myself to some short passages which support specific aspects of the Judge’s synthesis.

First, two passages from the speeches in Stag Line Ltd v. Foscolo, Mango & Co [1932] AC 428, in relation to the circumstances in which it is permissible to look at earlier meanings of words or phrases used in the Rules. Lord Atkin addressed the point at 432-4:

The speech of Lord Macmillan at 350 was to similar effect:

Second, in Gosse Millerd Ltd v. Canadian Government Merchant Marine [1929] AC

It was to this type of case that the Judge was referring when he spoke of ‘words or expressions which have judicial interpretation as terms of art’ and the presumption that they would have been used ‘in the sense already judicially imputed to them.’ Such an approach avoids what would otherwise be a tension between the cases set out above.

Third, in a passage in his speech in Effort Shipping Co. Ltd v. Linden Management SA (The Giannis NK) [1998] AC 605 at 621H, Lord Steyn emphasised the importance of ascertaining meaning from the language of words:

It is unnecessary to add further reference to authority. Some of the cases relied on by Mr Thomas showed a willingness by the Courts to have regard to earlier decisions, to a greater or lesser extent, so as to confirm a particular meaning in the Hague Rules or

This leaves one further matter for consideration at this stage: the effect of the Vienna Convention on the Law of Treaties 1969 (‘the Vienna Convention’) as a guide to interpretation.

The Vienna Convention was ratified by the United Kingdom in 1971 and came into force in 1980.

In Gard Marine & Energy Ltd v. China National Chartering Company Ltd (The Ocean Victory) [2017] UKSC 35, [2017] 1 WLR 1793 at [74] Lord Clarke of Stonecum-Ebony JSC summarised the approach to the interpretation of the Convention on Limitation of Liability for Maritime Claims (1976) (Footnote: ⁵). He referred, among other cases, to the decision in Stag Line v. Foscolo (above) and to the importance of not interpreting international conventions by reference to domestic principles, but rather by reference to ‘broad and acceptable principles.’ He recognised at [73] that it may be difficult to identify broad and acceptable principles, but identified some such principles in articles 31 and 32 of the Vienna Convention.

Article 31 of the Vienna Convention is headed, ‘General Rule of Interpretation.’

Lord Clarke concluded at [74]:

The need to focus on the ordinary meaning of the words used in their context and in the light of their object and purpose is consistent with the approach to interpretation established before the Vienna Convention took domestic effect; and it is therefore unnecessary to consider whether the earlier approach was inconsistent and, if so, how any such inconsistency might need to be resolved.

The essential characteristic of the Hague Rules was the pragmatic compromise (Footnote: ⁶) described by Lord Steyn in The Giannis NK (above) at 621H, quoted above.

The Imperial Shipping Committee report issued in 1921 had recognised that the renewed pressure on shipowners to relax their exclusions meant that, although they generally continued to insert broad exclusion clauses in their contracts, many of them, perhaps a majority, did not fully rely upon them. As the Judge noted, this was a reason for caution in construing the Hague Rules by reference to prior decision.

At [30], the Judge summarised his view of the assistance to be gained from the context, object and purpose of the Hague Rules:

In my view that was an accurate statement of the correct legal approach to the construction of the Hague Rules.

I would accept that discussions and resolutions in travaux préparatoires may illustrate in broad terms the context, object and purpose of an international convention, but in the case of the Hague Rules this can be ascertained without recourse to travaux préparatoires. I therefore turn to the meaning of the words in issue.

The words, ‘fire, unless caused by the actual fault or privity of the carrier’, is a phrase whose natural and ordinary meaning is clear. The words exclude the carrier from liability for fire however caused, provided it is not caused with the actual fault or privity of the carrier or in breach of its obligations set out in article III.1. The word ‘fire’ contains no implicit qualification as to how the fire is started, whether accidentally or deliberately, negligently or otherwise. Nor is there any implicit qualification depending on who may be responsible for the fire. The only express qualification is the proviso in the second part of the clause, ‘unless caused by the actual fault or privity of the carrier.’ Unless the cause of the fire falls within the proviso, fire is an excepted peril. I deal with the implicit qualification where there is a breach of article III.1 below.

The Judge drew additional support for the ordinary meaning of the word ‘fire’ where it is an insured peril under a policy of marine insurance. It is clear that arson to which the assured is not a party is within the scope of the fire peril, see Arnould, Law of Marine Insurance and Average (18th edition) at 23-29), Busk v. Royal Exchange Assurance Company (1818) 2 B & Ald 73 at 82-83, Trinder v. Thames and Mersey Insurance Company [1898] 2 QB 114 at 124, and the cases at footnote 195 of Arnould, to which the Judge referred at [34] of his judgment: Slattery v. Mance [1962] 1 QB 676 at 680-681; Continental Illinois National Bank & Trust Co of Chicago and Xenofon Maritime S.A. v. Alliance Assurance Co. Ltd (The Captain Panagos D.P.) [1986] 2 Lloyd’s Rep 470 at 510-511; [1988] 1 Lloyd’s Rep and

Mr Thomas’s observation that barratry and fire are separately identified marine perils does not advance his argument. ‘Barratry of master, officer and crew’ is a peril now covered in the second part of the Perils Clause of the Institute Time Clauses, Hulls 1.11.83 at clause 6.2.5, and is therefore subject to the proviso at clause 6.2: ‘provided such loss or damage has not resulted from want of due diligence by the assured, owners or managers’; see also Arnould (above) at 23-25.

This view of the ordinary meaning of article IV.2(b) gains support from a number of authoritative textbooks.

In Aikens, Lord & Bools on Bills of Lading (2nd edition) §10.231, there is this:

The editors of Scrutton on Charterparties and Bills of Lading (23rd edition) at §14074 adopt a similar analysis of the article IV.2(b) exception:

According to the editors of Carver on Bills of Lading (4th edition) at §9-215:

Finally, the editors of Voyage Charters (4th edition) at 85.261:

It is plain that, if the carrier has failed to comply with its obligations under article III.1 to exercise due diligence before and at the commencement of the voyage (a) to make the ship seaworthy, or (b) to properly man the ship, it will not be able to rely on the fire exception if a negligent or deliberate act of the crew have caused the fire, see for example Maxine Footwear Co. Ltd v. Canadian Government Merchant Marine Ltd [1959] AC 589 (JCPC) 589 at 602-3 (Footnote: ⁷). However, where there has been no prior causative breach of the carrier’s obligations under article III.1, its liability for loss by fire is excluded by article IV.2(b) unless caused by the actual fault or privity of the owner.

I would add that there is no sound policy reason for reading the word ‘fire’, both in isolation and in context, in a way that excludes fire where deliberately caused by the crew, from the carrier’s defence under Article IV.2(b). In cases of barratry the carrier’s agents are acting contrary to the carrier’s interests and in breach of the trust reposed in them (Footnote: ⁸). As the authors of Aikens Lord & Bools on Bills of Lading observe in a footnote at p.360:

Although, I have reached the clear conclusion as to the ordinary meaning of the words in their context, I must address Mr Thomas’s further arguments relied on in support of his construction of Article IV.2(b): first, a developed argument by reference to the judgment of Lord Sumption in the Volcafe case (above); second an argument by reference to English law prior to the Hague Rules; third, a reliance other non-domestic authorities; and fourth, reliance on the travaux préparatoires.

The case was decided by the Supreme Court on 8 December 2018, very shortly before the hearing of the appeal; and was concerned with what was an unusual situation where the burden of proof in a cargo claim brought under the Hague Rules was material, see [1].

The bills of lading, which were subject to English law and jurisdiction, incorporated the Hague Rules. The carriers were responsible for loading a cargo of coffee beans (which was hygroscopic in nature) into unventilated containers. The carriers used absorbent corrugated paper to line the containers so as to protect the cargo from condensation damage as was usual commercial practice at the time. Despite these precautions, part of the cargo was found to be damaged by condensation; and a claim was made against the carriers.

The cargo owners pleaded their case in conventional terms alleging a breach of bailment in failing to deliver the cargo in the same good order and condition as when shipped, alternatively relying on a breach of the carrier’s obligation under Article III.2 properly and carefully to load, stow, carry, keep, care for and discharge the cargo. The carrier relied on the defence in article IV.2(m): damage caused by inherent defect or quality of the cargo.

One of the issues at trial was upon whom lay the burden of proving whether the cargo damage was caused by (i) negligent preparation of the containers, as asserted by the claimants, or (ii) inherent vice, as asserted by the defendants, see [6]. Lord Sumption JSC (giving a judgment with which Lord Reed JSC, Lord Wilson JSC, Lord Hodge JSC and Lord Kitchin JSC agreed) concluded, at [43], that the carrier had the legal burden of showing that it had taken due care to protect the cargo from damage, including taking due care to protect it from damage arising from its inherent characteristics such as its hygroscopic character; and on the facts found at trial, or perhaps more accurately the absence of material findings, the carrier had failed to discharge that burden.

In the course of a comprehensive judgment, Lord Sumption analysed the law of bailment so far as it applied to the issue, and the relevant provisions of the Hague Rules. He also addressed a number of matters which led to his conclusion on the appeal. First, although the Hague Rules provide a complete code in relation to those matters which they cover, they do not address issues of evidence, or modes of proving a breach of a prescribed standard or the application of an exception, see [15]. Second, the well-established principle that the Hague Rules should not be construed by reference to principles of purely domestic application (Footnote: ⁹) did not bear materially on the issue for decision, see [16].

It was on the third issue, the operation of the burden of proof, that much of Lord Sumption’s judgment focused, and the parties to the present appeal addressed their argument.

For present purposes it is convenient to identify what was said to be the ‘true rule’, at [25]:

It is only necessary to refer to two further passages. The first is at [28]:

It is important not to lose sight of Lord Sumption’s observation that there is ‘no unifying legal principle’ behind the list of exceptions in article IV.2. It follows that the correct approach is to construe the exceptions in their own terms, while bearing in mind that they fall under a general heading and have to be construed as part of the overall scheme of obligations, liabilities and exceptions set out in articles III and IV.

The second passage is at [33]:

Mr Thomas relied on the judgment in Volcafe to establish two main propositions. First, Lord Sumption’s observation at [7]:

I do not regard this passage as either changing the well-established approach to construction of the Hague Rules, see [16] of Lord Sumption’s judgment, nor amounting to encouragement to embark on a wide-ranging examination of the common law position other than where it is necessary and likely to throw light on the particular point in issue. In Volcafe it was necessary to do so on the issue of burden of proof which was not a matter dealt with in the Hague Rules at all, see [15]:

Secondly, Mr Thomas submitted that the judgment in Volcafe demonstrated that it was wrong to approach the word ‘fire’ in article IV.2(b) in isolation. Instead, it was necessary to analyse the exception in the light of the contractual obligations undertaken by the carrier: for example, under articles III.1 and 2. He laid emphasis on the phrase, ‘Neither the carrier nor the ship shall be responsible for loss and damage arising or resulting from’, which introduces the list of exceptions. This highlights the importance of both the excepted circumstances and its causative effect, see also [32] of the judgment in Volcafe. Since it is not every fire that causes loss, Mr Thomas submitted that the carrier must show both that the fire was an excepted peril, and that it was the effective cause of the loss.

I do not consider that this argument materially advances Glencore’s case on the appeal. On the assumed facts, the fire that caused the loss was the deliberate act of the Chief Engineer. Provided the Owners were not in breach of their obligations under article III.1 (which is not an issue which arises on the preliminary issues) they are entitled to rely on article IV.2(b) unless loss or damage resulting from the fire was caused by their actual fault or privity. I would add that Lord Sumption’s observations (at [33]) that the carriers have the legal burden of disproving negligence for the purposes of invoking an exception under article IV.2 did not address any argument in relation to article IV.2(b); and does not greatly assist on the assumed facts where there has been a deliberate act by a crew member to the prejudice of the carrier and without the carrier’s actual fault or privity.

There were two limbs to this part of Mr Thomas’s argument. First, he relied on In re Polemis and anor v. Furness, Withy and Co Ltd [1921] 3 KB 560 in support of the contention that, at the time of the Hague Rules, English law had established that as between cargo interests and carriers, a clause excluding liability for fire was insufficient to exclude liability for negligently caused fires. Secondly, he submitted that the position at common law prior to the Hague Rules was that, in the absence of clear words, exclusion clauses in a contract of carriage were not construed as applying to intentional acts of wrongdoing by the carrier’s servants or agents against the ship or cargo, i.e. barratry.

The first difficulty is that both of these arguments run contrary to the approach to interpretation of the Hague Rules set out in the speeches of Lord Atkin and Lord Macmillan in Stag Line v. Foscolo (see above): that the meaning of the provisions are not to be ‘rigidly controlled by domestic precedents of antecedent date’ (Lord Macmillan) or coloured ‘by considering whether a meaning otherwise plain should be avoided if it alters the previous law,’ (Lord Atkin). On the contrary, they suggest a ‘predilection for former law’ of which Lord Atkin disapproved.

This is the answer to the extensive exercise in forensic archaeology on which Mr Thomas embarked.

Nor am I persuaded by these arguments even if they had reflected a permissible approach. So far as the first argument is concerned, the Judge noted at [48]:

In The Polemis there was an exception in a time Charterparty for ‘fire … always mutually accepted.’ The Court of Appeal held that these words were not sufficient to exclude damage caused by a fire due to the negligent act of stevedores (the charterers’ agents) in the course of loading, since there was no express stipulation to that effect. I am doubtful whether a decision on the construction of a Time Charterparty clause is of great assistance in interpreting article IV.2(b); but in any event the word ‘fire’ is not the only word that must be construed in article IV.2(b). It is a word that must be read in context.

So far as the second argument is concerned, Mr Thomas relied on three cases on dissimilar facts: Taylor v. Liverpool & Great Western Steam Co (1873-74) L.R. 9

In Taylor v. Liverpool & GWS, on a case stated, the Court (Lush and Archibald JJ) held that an exception for loss caused by ‘thieves’ did not cover theft by the crew. At p.550 Lush J said:

In Steinman v. Angier, the plaintiff sought to recover from the carrier the value of the goods which had been stolen by the stevedores during the course of stowage. The carrier sought to rely on a term which excluded liability for losses caused by ‘thieves of whatever kind, whether on board or not, or by land or sea.’ The Court held that the term did not exclude the carrier’s liability, with Bowen LJ saying at p.624:

Mr Thomas acknowledged the dissimilarities with the present case, which does not involve theft by the carrier’s agent; but he submitted that the principle limiting exclusion of liability applied to damage which was intentionally caused by the crew.

In The Chasca (above) LR 4 A. & E. 446, holes were discovered to have been bored with augers below the waterline in the sides of the vessel, causing water damage to cargo. The crew confessed that they bored the holes (Footnote: ¹⁰); and the owners of the vessel argued against the cargo owner that the crew’s conduct fell within the exception in the bill of lading for dangers of the seas (Footnote: ¹¹). Unsurprisingly, the owners’ argument did not find favour with the court. Sir Robert Phillimore held that the authorities showed that:

Mr Thomas submitted that the decision in The Chasca showed that, even where there were wide exceptions, those exceptions were not to be construed as extending to barratry in the absence of clear words.

In my view none of these cases assist Glencore’s argument on the construction of article IV.2(b). Neither Taylor v. Liverpool & GWS nor Steinman v. Angier concerned fire or barratry, and The Chasca had nothing to do with fire and was concerned with the perils of the sea exception. The cases may provide historic support for the broad proposition that exclusion of liability for damage caused by deliberate wrongs committed by the crew will require clear words; and, in the case of The Chasca, clearer words than those which might exclude negligence. The reason why the ‘perils of the sea’ exception did not cover the barratrous acts was that the wrongful boring of holes in the vessel’s hull was not a fortuity.

It seems to me that Mr Thomas’s extensive researches have revealed that there was no pre-Hague Rules judicial interpretation of ‘fire’ as a term which had a clearly assigned meaning that excluded fire caused by the crew, so that it must be presumed that it was used in article IV.2(b) in the same way.

Mr Thomas relied on two further transpontine authorities.

The first was the decision of the US Court of Appeals for the Fourth Circuit in In the matter of Intercontinental Properties Management S.A. as owner of the Motor vessel ‘MIMI’ (1979) 604 F 2d 254. In that case, there was a claim by cargo interests for loss caused by the deliberate casting away of the vessel by a member of the crew. The issue in the case was identified as being simply whether the shipowners could bring themselves within the exception in rule 4.2(q) of the US Carriage of Goods by Sea Act in a case of barratry by a crew member. The barratrous conduct was the opening of the vessel’s sea-valves, which had been preceded by acts of mortal violence against fellow crew members (see p.257 of the report). The court concluded that the acts were within the scope of Supardi’s employment and as a matter of construction fell within the excluding proviso in rule 4.2(q) so that the shipowners could not bring themselves within that exception.

The potential relevance of the decision on this appeal comes from the supportive reasoning in the opinion of the Court given by Circuit Judge Philips at 265-266:

In my view this extract does not assist on the interpretation of Article IV.2(b). The reasoning was specific to Article IV.2(q) with its particular excluding proviso where

The second case was the New Zealand decision in Tasman Orient Line CV v. New Zealand China Clays and ors (The Tasman Pioneer) [2010] 2 Lloyd's Rep. 13. In that case, the master of the vessel took a risky short cut through a narrow passage to save time. The vessel grounded, and the master failed to alert either the owners or the coastguard. Instead he instructed the crew to lie and try to cover up what had happened. The delay in seeking assistance caused significant cargo damage, but despite the master’s deceptions the owners were held entitled to rely on Article IV.2(a) (‘Act, neglect or default of the master, mariner, pilot, or the servants of the carrier in the navigation or in the management of the ship’) on the basis that the exception was designed to protect carriers from the acts of their employees, provided the conduct did not amount to barratry. This qualification for barratry was the result of a concession and became common ground, see paragraph 10. The court concluded that the concession was rightly made. However, the reasoning in support of that conclusion is slight, perhaps unsurprisingly since it was not in issue. Furthermore, to the extent that it was based on the travaux préparatoires, the conclusion appears to be unfounded.

Having noted that the Tasman Pioneer case was not concerned with Article IV.2(b) and had been subjected to ‘penetrating academic criticism’ (Footnote: ¹²), the Judge observed:

In my view neither of these cases assists on the preliminary issue.

There are two relevant questions in relation to the travaux préparatoires: first, what is the test for recourse to them as a means of interpretation; and second, how do they assist in the interpretation of article IV.2(b) in the present case?

In Effort Shipping v. Linden Management (The Giannis NK), Lord Steyn addressed the first question at p.623D:

In Fothergill v. Monarch Airlines Lid [1981] AC 251, Lord Wilberforce had considered the travaux préparatoires contained in the minutes of the 1955 Hague Conference (leading up to the 1955 Hague Protocol, amending the Warsaw Convention 1929). At 278C he had said:

I would add a third potential objection, which may or may not apply in a particular case, namely: that it is possible that parties to an international convention may choose (or at least acquiesce in) imprecise language.

The answer to the second question requires consideration of the travaux préparatoires.

The wording of the draft of the Rules which first came before the 1921 Hague Conference included ‘fire’ as an excepted peril. At some stage before the second day, the draft was amended so that Rule 2(b) exempted ‘barratry’ and Rule 2(c) exempted ‘fire’. Both these changes were discussed and negotiated in passages which I have set out as an appendix to this judgment. In the event, the wording of both exceptions was retained.

The discussions during the second day appear to have concluded that ‘fire’ in what was then draft Rule 2(c) was understood to mean fire however caused including, in particular, fires started by servants or agents of the carrier either deliberately or negligently. There was a discussion about whether to add wording which excluded fires caused with the fault of servants or agents in the specific context of a proposed amendment covering fires wilfully started. However, that amendment was rejected. The understanding was that fire caused with the privity of the owners could not be exempted even if the language were left simply as ‘fire’.

At some stage before the end of the Conference the barratry exception was removed, and the Rule 2(b) exception in the 1921 Hague Rules became ‘fire’.

A year later, at the Brussels conference in October 1922, the words ‘unless caused by the actual fault or privity of the carrier’ were added, following a proposal by the US delegation, such wording already being the basis of shipowners’ exception under s.502 Merchant Shipping Act 1894 (Footnote: ¹³). Thus, in The Diamond [1906] P 282, it was held that s.502(i) protected owner unless it was ‘in fault or privy to [the] misconduct or carelessness on the part of the crew,’ in starting the fire.

Having considered the travaux préparatoires, the Judge concluded that they showed that the participants proceeded on the basis that ‘fire’ meant fire even if deliberately caused by the shipowner’s servants or agents, or resulting from their negligence; and not that they only contemplated fires which were caused accidentally or without negligence. It followed that the travaux préparatoires supported the plain meaning of the text of article IV.2(b).

Mr Thomas submitted that the fact that the delegates to the Hague Conference considered that the carrier would have no liability for a barratrously started fire did not show that the word ‘fire’ in the draft was understood to mean fire even where caused by barratry. On a proper reading, the travaux préparatoires show that the draftsmen made a deliberate decision not to exclude losses caused by barratry. It would, he submitted, be strange if the defence had been reintroduced for certain forms of barratry through the fire defence.

Mr Hill submitted that the Judge was right to dismiss this argument as a non sequitur. The fact that the discussion did not favour a general exception for all forms of barratry did not imply, either necessarily or otherwise, that the fire exception did not encompass barratry. The point simply begged the question of what those drafting the Hague Rules did intend to exclude, which depended on the true interpretation of article IV.2(b).

It seems to me that the Judge was right in his analysis on the material he was invited to consider. However, I am very doubtful whether the threshold for consideration of the travaux préparatoires came close to being met. This was not a provision in respect of which there were ‘truly feasible alternative interpretations’ of the words, see Lord Steyn in the Giannis NK (above). Nor was it one of those ‘rare’ cases where

It was common ground that an act of barratry occurs without the actual fault or privity of the carrier. However, Glencore’s argument necessarily implies an additional qualification to the words, ‘Fire, unless caused by the actual fault or privity of the carrier, or the fault or neglect of the crew,’ [emphasis added]. I can see no proper basis for implying such words either as a matter of ordinary meaning nor on any of the supplementary arguments advanced by Glencore, and I see principled reasons for not doing so.

In my view the Judge was right in his conclusion on issue 2.

Since I have concluded that the Judge correctly decided the issue 2, it is unnecessary to deal at the same length with this issue. If the fire were set by the deliberate act of the Chief Engineer, provided it was caused without their actual fault or privity, the Owners can rely on the article IV.2(b) defence. It follows that it is unnecessary to decide whether or not the Chief Engineer’s assumed conduct would properly fall within the definition of barratry.

Before the Judge, Glencore defined barratry as a wilful act of wrongdoing committed by the master or crew against the ship or goods without the privity of the shipowner; or alternatively an act or omission of the master or crew with intent to cause damage, or recklessly with knowledge that damage would probably result. The Owners defined barratry as any wilful or intentional act of wrongdoing by the master or mariners to the prejudice of the owner or charterer, without the privity of that owner or charterer, where the intention is criminal or fraudulent.

The significant difference for the purposes of the hearing was that Owners argued that there must be the commission of a crime with the necessary intent; and that, on the agreed and assumed facts, the necessary intent to commit a crime would be absent were the mental state of the Chief Engineer such as to afford him a defence of insanity

Mr Hill argued that as a matter of English law, a defence of insanity lies where a person is labouring from such a defect of reason, from disease of the mind, as not to know the nature and quality of the act he was doing, or, if he did know, that he did not know he was doing what was wrong, see McNaughton’s Case (1843) 10 Cl & F 200 at pp. 210-211. The facts that were assumed for the purposes of the preliminary issue were that the Chief Engineer intended to cause damage; but they left open the question of whether by reason of a mental illness he did not understand that what he was doing was wrong.

Mr Thomas submitted that barratry does not require the mens rea of intention to commit a crime; the relevant ingredient in barratry is simply wrongdoing, which can be committed by any serious breach of duty towards owners; and that however defined, the act of setting fire to the vessel intending to cause damage was so obviously contrary to the duty owed to the Owners and so clearly within the concept of wrongdoing that it amounts to barratry.

Importantly for present purpose, Glencore took a threshold point that it was not open to the Owners to submit that the definition of barratry was not conclusive of the result on the basis of the existing and agreed assumed facts. Mr Thomas submitted that the preliminary issues had been ordered to be tried on the understanding that those facts would be sufficient to resolve, one way or the other, whether or not there had been an act of barratry. He also complained that the Owners’ Defence did not plead insanity or the necessary factual ingredients to support such a plea.

The Judge rejected Glencore’s threshold objection at [11]:

The Judge referred to a number of decisions which assisted him in reaching a definition of barratry: Knight v. Cambridge (1724) 1 Stra 58 (Raymond J); Earle v. Rowcroft (1806) 8 East 126 (Lord Ellenborough CJ); John Cory & Sons v. Burr (1888) 8 App. Cs. 393 (Lord Blackburn); Briscoe & Co v. Powell & Co (1905) 21 TLR 128 (Channell J); Mentz Drecker & Co Ltd v. Maritime Insurance Co (1909) 15

However, in the event, he came to no firm conclusion on the preliminary issue:

The reasoning that led to this inconclusive result can be found in three earlier passages in the judgment:

Glencore is critical of these passages and in particular the conclusion that the Chief

In my view, the Judge was wrong to allow the Owners to raise the issue of insanity in the way they did. There was no factual basis upon which he could consider the question. By inviting him to address the issue of insanity, the Owners were acting as if they were conducting a tutorial group, asking, ‘Would your answer be different if the barrator were insane?’ and then, ‘by what standards would you judge the issue of insanity?’ These are no doubt interesting questions and the Judge did his best to answer them; but he should never have been asked to address them at all. Unsurprisingly, he came to a conclusion that it would depend on the facts, which had been the basis of Glencore’s initial objection to the matter being dealt with in this way. This case provides a good illustration of the importance of closely defining the ambit of preliminary issues before the trial of those issues, and not adding further hypotheses during the course of them.

While I understand why the Judge approached the matter in this way, since his approach was entirely consonant with the practical approach that characterises case management decisions in the Commercial Court, the result was unsatisfactory. As noted above, §15 of the agreed or assumed facts was that Chief Engineer was either (a) under extreme emotional stress and/or anxiety due to the illness of his mother; or (b) alternatively, suffering from an unknown and undiagnosed personality disorder and/or mental illness; or (c) neither (a) nor (b) above. This reflected §13.b of the Owners’ Defence. They had not pleaded, and still have not pleaded, that he was insane; and there was no proper basis for assuming that he was. Even if the matter were to be analysed as a matter of domestic law (and in my view the Judge was right to consider that it should not), there is a clear analytical distinction to be drawn between mental illness (which was pleaded) and insanity (which was not) (Footnote: ¹⁶). As the Chancellor observed during the course of argument, it is not generally the function of the Courts to answer hypothetical questions. To allow such a departure from the agreed and assumed facts so as to introduce a matter that was not, and could not be, pleaded was undesirable and wasteful of resources.

When he came to hand down his judgment, the Judge made clear that the Owners would have to amend their pleaded case so as to reflect the argument they had advanced before him. At the date of the hearing of this appeal, they had not done so. Tellingly, Mr Hill informed us that they did not have the evidence that would permit anyone to sign a Statement of Truth.

Accordingly, and subject to any argument of the parties as to the form of an order, I would allow Glencore’s appeal on issue 1.

Accordingly, I would dismiss Glencore’s appeal on issue 2 (the fire exception) and allow its appeal on issue 1 (whether the conduct of the Chief Engineer necessarily constituted barratry on the assumed facts).

I would propose that the order made on the appeal should be as follows: On the agreed and assumed facts:

I agree that, for the reasons given by Lord Justice Simon, the appeal on the principal issue (issue 2) should be dismissed.

In my view, the appellants’ case on article IV.2(b) suffered from two insurmountable difficulties. First, Mr Thomas QC was unable to show that there was any ambiguity in the words ‘fire, unless caused by the actual fault or privity of the carrier’, and that, in particular, there was no reason not to read the word ‘fire’ as including all fires, howsoever started (unless due to the actual fault or privity of the carrier). In consequence, there was no need to have regard to the pre-Hague Rules cases, or the deliberations of the Committee in 1921, for any clarification or guidance.

Secondly, however, even assuming that there was a need or requirement to look in detail at the earlier authorities, they are, on a proper analysis, of no assistance to the appellants. At one point during his oral submissions on behalf of the respondents, Mr Hill QC accepted that, if there had been an authority which decided that ‘fire’ excluded a fire which had been barratrously started, then Glencore’s case might at least have a starting-point. But he submitted that there was no such authority and he was right to do so.

On analysis, and as Mr Thomas QC accepted, there were only three cases prior to The Hague Rules on which the appellants could rely in any event: Taylor v Liverpool & Great Western Steam Co (1873-74) L.R. 9 Q.B. 546; Steinman & Co v Angier Line Ltd [1891] 1 QB 619; and The Chasca (1875) LR 4 A & E 446. None of those authorities was concerned with fire at all: the first two involved ‘thieves’ and did not consider barratry; and the third concerned the perils of the sea exception. Even on their face, therefore, they offered no guidance whatsoever as to how Article IV.2(b) was to be interpreted, much less providing any justification for departing from the ordinary meaning of the words used.

In essence, the appellants’ interpretation of Article IV.2(b) would have it read:

I also agree that, for the reasons given by Simon LJ, the appeal on issue 1 should be allowed. In my view, given the absence of any clear pleaded case, let alone evidence, as to the Chief Engineer’s actual mental state, the judge should not have been asked to determine this question at all.

I agree with both judgments and with the disposal proposed by Lord Justice Simon. I add a few words on issue 1 only because, in my judgment, the litigation process took a wrong turning.

It must have been apparent at the hearing before the Judge that the question of whether or not the Chief Engineer was suffering from a mental illness that meant that he did not understand the nature and quality of his actions could be relevant, if not crucial, to the determination of the first preliminary issue, namely ‘whether on the basis of the agreed and assumed facts … the conduct of the chief engineer constitute[d] barratry.’ Nonetheless, the preliminary issues trial progressed without the Owners pleading insanity (a term I will use as a useful shorthand). Indeed, it appears that they did not know then, and do not know now, whether they will ever be in a position to plead insanity.

When the Judge said in paragraph 11 that ‘[a]lthough the Defence does not specifically allege insanity, it does plead reliance on mental illness and the submissions to me made clear that the allegation of insanity is made’, he fell into error, because in fact no such allegation could properly have been pleaded at that stage. Whilst I would accept what he said subsequently, namely that ‘[i]t would not be right to shut [the Owners] out on a pleading point which could at this stage be cured

Paragraph 24 of the Judge’s judgment makes clear that the nature of the alleged mental illness underlay the decision on issue 1 (which was, as I say, whether, on the basis of the agreed and assumed facts, the conduct of the Chief Engineer constituted barratry). The Judge mentioned in that paragraph that the Chief Engineer ‘would not be guilty of the kind of knowing wrongdoing which would be sufficient’ or be committing a crime or knowingly acting in breach of his duty to the owners, if ‘he were suffering from a clinical mental disorder, such that he could not distinguish between right and wrong’. He gave as an example of a possible assumed fact that the Chief Engineer might have been in ‘a psychotic state in which he heard voices telling him that the owners wished him to act as he did’. Thus, the Judge knew that the nature of the mental illness could be important, and ought to have concluded that he had no real facts (whether agreed or assumed) on which to base the legal decision he was being asked to make.

Accordingly, I agree with Simon LJ that it was undesirable and a waste of resources to allow a departure from assumed facts that introduced an issue that was not, and could not be, pleaded. The determination of whether a particular crew member was acting barratrously, when suffering from a particular kind of mental illness, will only need to be decided if and when there are precisely pleaded facts as opposed to assumed possibilities that go beyond what was agreed for the purposes of the preliminary issues directed to be tried.

Glencore appealed on the basis that there was no need for a close analysis of the Chief

It is sufficient, in these circumstances, for us to say that, on the agreed and assumed facts, the fire was deliberately caused with intent to cause damage, and the Owners were, therefore, exempt from liability under article IV.2(b).