Volcafe Ltd and others v Compania Sud Americana De Vapores SA

This appeal is about the burden of proof in actions against a shipowner for loss of or damage to cargo. It may seem strange that a species of litigation which has generated reported decisions over four centuries should not yet have returned a definitive answer to this question. The reason is probably to be found in the fact that the courts very rarely decide issues of fact on the burden of proof. The trial judge is usually able to find some persuasive evidence, however exiguous, to break the impasse. This case is, or may be, different.

The six claimants were the owners and bill of lading holders for nine separate consignments of bagged Colombian green coffee beans shipped at Buenaventura in Colombia between 14 January and 6 April 2012 on various vessels owned by the defendant shipowners for carriage to Bremen. They were stowed in a total of 20 unventilated 20-foot containers. These were transhipped at Balboa in Panama and discharged at Rotterdam, Hamburg or Bremerhaven for on-carriage to Bremen. Each consignment was covered by a bill of lading covering the entire carriage to Bremen.

The bills of lading, which were subject to English law and jurisdiction and incorporated the Hague Rules, were on LCL/FCL (less than full container load/full container load) terms. It is common ground that this means that the carriers were contractually responsible for preparing the containers for carriage and stuffing the bags of coffee into them. They employed two firms of stevedores to perform this function. Coffee is a hygroscopic cargo. It absorbs, stores and emits moisture. It can be carried in ventilated or unventilated containers. In 2012 both types of container were in widespread use for the carriage of bagged coffee, and the shippers had specified unventilated containers for these consignments. The use of unventilated containers is cheaper, but if they are used to carry coffee beans from a warm to a cooler climate, as they were in this case, the beans will inevitably emit moisture which will cause condensation to form on the walls and roof of the container. This makes it necessary to protect the coffee from water damage by dressing the containers, ie lining the roof and walls with an absorbent material such as cardboard, corrugated paper or “Kraft” paper. The use of Kraft paper was a common commercial practice in 2012, and it was employed in this case. When the containers were opened at Bremen, however, the bags in 18 of them were found to have suffered water damage from condensation.

The cargo owners pleaded their case in what has for many years been standard form. Their primary case was that in breach of their duties as bailees the carriers failed to deliver the cargoes in the same good order and condition as that recorded on the bill of lading on shipment. Alternatively, they pleaded that in breach of article III, rule 2 of the Hague Rules they had failed properly and carefully to load, handle, stow, carry, keep, care for and discharge the cargoes. A number of particulars of negligence was pleaded. For present purposes, the only relevant one is that the carriers failed to use adequate or sufficient Kraft paper to protect the cargoes from condensation. The carriers joined issue on all of these points, and pleaded inherent vice on the ground that the coffee beans were unable to withstand the ordinary levels of condensation forming in containers during passages from warm to cool climates. The cargo owners pleaded in reply that any inherent characteristic of the cargo which resulted in damage, did so only because of the carrier’s negligent failure to take proper measures for its protection.

The case was tried in the London Mercantile Court by David Donaldson QC, sitting as a deputy High Court judge. He held that there was no legal burden on the carrier to prove that the damage to the cargo was caused without negligence or by an excepted peril. There was only a factual presumption that damage ascertained on discharge was due to negligence. The critical issues of fact, as they emerged at the trial, concerned the weight of paper and the number of layers that (i) were, and (ii) should have been used. The deputy judge’s conclusions were as follows:

These conclusions were criticised on a number of grounds by the Court of Appeal, which proceeded to make its own findings. I shall return to the Court of Appeal’s treatment of the facts later in this judgment. But for the moment I shall proceed on the basis of the deputy judge’s conclusions, for it is those which give rise to the major issue of law on this appeal. On whom was the burden of proving whether the cargoes were damaged by (i) negligent preparation of the containers, or (ii) inherent vice?

The bills of lading in this case incorporated the Hague Rules. It is, however, necessary to examine the common law position apart from the Rules, first, because it is an essential part of the legal background against which they were drafted; and, secondly, because the common law position had been considered in a number of authorities decided before the Rules were promulgated, which have remained influential since and indeed were relied upon on this appeal.

The delivery of goods for carriage by sea is a bailment for reward on the terms of the bill of lading. Bailment is a transfer of possession giving rise to a legal relationship between the bailor and the bailee which is independent of contract, although in practice it is commonly contractual and the terms of the contract will commonly modify its incidents. Two principles of the common law of bailment are fundamental. The first is that a bailee of goods is not an insurer. His duty is limited to taking reasonable care of the goods. This has been true of bailees generally for as long as bailment has existed as a recognised source of legal responsibility at common law: see S Stoljar, “The early history of bailment”, American Journal of Legal History, vol i (1957), p 5, and the landmark decision of Chief Justice Holt in Coggs v Bernard (1703) 2 Ld Raym 909, 917-918. In the 19th century some shipowners, especially in the liner and tariff trades, were common carriers, bearing a more onerous responsibility at common law. The characteristic feature of a common carrier was that he held himself out as accepting for carriage the goods of all comers on a given route, subject to capacity limits. As such, he was strictly liable at common law for loss of or damage to the cargo subject only to exceptions for acts of God and the Queen’s enemies. The absence of negligence was irrelevant. But although the position of common carriers is commonly referred to by way of background in the case law, as it was in the judgments below, it is no longer a useful paradigm for the common law liability of a shipowner. Common carriers have for many years been an almost extinct category. For all practical legal purposes, the common law liability of a carrier, unless modified by contract, is the same as that of bailees for reward generally.

The second principle, which is equally well established, is that although the obligation of the bailee is thus a qualified obligation to take reasonable care, at common law he bears the legal burden of proving the absence of negligence. He need not show exactly how the injury occurred, but he must show either that he took reasonable care of the goods or that any want of reasonable care did not cause the loss or damage sustained. As Cockburn CJ put it in Reeve v Palmer(1858) 5 CBNS 84, 90:

Three points should be made by way of amplification of these statements. First, it is clear that the burden of proof with which these decisions were concerned was a legal burden. It is quite different from the evidential burden which may arise where the facts give rise to a rebuttable inference of negligence or under the principle res ipsa loquitur. Secondly, while the rule about the burden of proof in English law developed long before any pragmatic justification was advanced for it, its continued importance in the law of bailment has consistently been supported on the ground that because the bailee is in possession of the goods it may be difficult or impossible for anyone else to account for the loss or damage sustained by them: see The “RUAPEHU” (1925) 2 Ll L Rep 310, 315 (Atkin LJ); British Road Services Ltd v Arthur V Crutchley & Co Ltd (No 1) [1968] 1 All ER 811, 822 (Sachs LJ). Modern scientific techniques of investigation have eased this particular problem to some degree, but have not removed it. Thirdly, although the principle regarding the burden of proof was independently developed by the common law, it is not a peculiarity of the common law. The duty of a depositary to justify his inability to deliver the goods in the condition in which he received them is a basic feature of the civil law. So far as carriers are concerned, it originates in the Roman praetorian edict Nautae, Caupones Stabularii, which in modified form remains the basis of the law of deposit in French and Scots law and other civil law systems: see Pothier, Traité du Contrat de Louage 6th ed (1821), para 199; R Zimmermann, The Law of Obligations: Roman Foundations of the Civilian Tradition (1996), pp 514-526. For the current position in France, see Code Civil, Book III, Title VIII, Chapter III, article 1784 governing carriers by land and water (“Ils sont responsables de la perte et des avaries des choses qui leur sont confiées, à moins qu'ils ne prouvent qu'elles ont été perdues et avariées par cas fortuit ou force majeure”); and for Scotland, Gloag on Contract 2nd ed (1929), p 721; McBryde, The Law of Contract in Scotland 3rd ed (2007), paras 9.53-9.57 and The Laws of Scotland, Stair Memorial Encyclopaedia, vol 8 (Deposit), at para 13.

There is no significant dispute about these principles on this appeal. The real issue is whether the incidence of the burden of proof is different in a modern contract for carriage by sea incorporating the Hague Rules. I therefore turn to the Rules.

The relevant provisions of the Rules are as follows:

The burden of proof arises on this appeal at two stages of the analysis. The first is concerned with article III.2. Does the cargo-owner bear the legal burden of proving breach of that article, or is it for the carrier, once loss or damage to the cargo has been ascertained, to prove compliance? The second relates to article IV.2, and particularly to exception (m). The carrier accepts that he must bear the burden of proving facts which bring the case within an exception, but submits that once he has done so it is for the cargo-owner to prove that it was the negligence of the carrier which caused the excepted peril (in this case, inherent vice) to operate on the cargo.

Article III.2 imposes on the carrier a general duty to take reasonable care of the cargo during carriage. Mr Rainey QC, who appeared for the carrier, submitted that the burden of proving a breach of it lay upon the cargo-owner. His argument proceeded as follows: (i) the Hague Rules constitute a complete code governing the care of the cargo; (ii) an international convention such as the Hague Rules should not be construed in the light of particular features of English law or any other domestic system of law; and (iii) article III.2 of the Hague Rules, by imposing an obligation to take reasonable care of the cargo, displaces the English law rule about the burden of proof, because as a general rule he who asserts must prove. In my judgment, each of these steps in the argument is fallacious.

As to proposition (i), the Hague Rules never had statutory application, save to the carriage of goods shipped from a port in the United Kingdom: Carriage of Goods by Sea Act 1924, section 1. That Act has now been repealed by the Carriage of Goods by Sea Act 1971 which gives the force of law to the Hague-Visby Rules but does not apply to the shipments in issue on this appeal. Accordingly, the Hague Rules have effect only by virtue of their contractual incorporation into the bill of lading. Subject to the other terms of the contract, the Rules are a complete code on those matters which they cover. But they are not exhaustive of all matters relating to the legal responsibility of carriers for the cargo. As is well known, the background against which they were drafted was the attempt of (mainly British) shipowners in the late 19th century to limit their legal responsibility for cargo, and the attempt of other countries, notably the United States, Canada and Australia, to impose a minimum standard of performance by law. The purpose of the Rules was to standardise the obligations of the carrier and to limit the exceptions on which he should be entitled to rely. They are accordingly concerned almost exclusively with the standard of performance. Apart from certain articles, such as IV.1 and IV.2(q), which deal in terms with the burden of proof for specific purposes, the Rules do not deal with questions of evidence or the mode of proving a breach of the prescribed standard or the application of an exception. These are matters which in accordance with generally recognised principles of private international law are for the law of the forum. They are part of the law of evidence and the rules of procedure, which are liable to vary from one jurisdiction to another.

Turning to proposition (ii), it is well established that even in a contract governed by English law, provisions derived from an international convention are intended to have an internationally uniform effect and should not be construed by national courts by reference to principles of purely domestic application: Stag Line Ltd v Foscolo, Mango & Co Ltd[1932] AC 328, 350 (Lord MacMillan); James Buchanan & Co Ltd v Babco Forwarding and Shipping (UK) Ltd[1978] AC 141, 152 (Lord Wilberforce). But this principle has no bearing on the present issue. In the first place, as I have explained in the preceding paragraph, the Rules are not concerned with the incidence of the burden of proof save in limited respects. Secondly, it has not been shown that the common law principles regarding the burden of proving negligence or the lack of it in the carriage of goods are principles of purely domestic application, except in the limited sense that their historical origins lie in the common law. As I have pointed out above, the principle that the custodian of goods has a legal responsibility to justify their loss or redelivery in damaged condition is common to civil law jurisdictions as well. One of those jurisdictions is Scotland, to which the Hague Rules were applied by the Carriage of Goods by Sea Act 1924 on the same basis as England. The common law of bailment has no application in Scotland, but the law of Scotland regarding the burden of proof lying on a warehouseman or a carrier is substantially the same. Civil law jurisdictions party to the Hague Convention vary widely in the way that they implement the Rules. In many of them, the occurrence of damage during carriage is treated not just as casting the burden of proving absence of fault on the carrier, but as conclusive of breach of article III.2, unless the carrier can prove that one of the article IV.2 exceptions applies. This is, for example, the effect of the French Code des Transports L 5422-12 and of successive decisions of the Cour de Cassation: see Chambre Commerciale, 27 mai 1975, 74-10388, 10 juillet 2001, 99-12258, 13 décembre 2016, 14-28332. The researches of counsel have shown that a corresponding principle applies in Belgium, the Netherlands, Italy, Germany, Norway and Spain.

The carriers’ proposition (iii) is based, in my view, on a misconception. Mr Rainey argued that the reason why at common law the bailee had the burden of disproving negligence was that at common law a bailee had a strict obligation to redeliver the goods in the same condition as when received. The position, he submitted, was different where the obligation was a qualified obligation to take reasonable care, as it is in article III.2. However, as I have pointed out, the common law obligation of a bailee is not strict, save in the somewhat theoretical case of common carriers. His obligation is to take reasonable care. The common law has always treated that as consistent with a rule imposing on him the burden of disproving negligence. In the same way, the imposition of a corresponding duty of care on the carrier by article III.2 is consistent with his bearing the burden of disproving negligence.

When one examines the scheme of the Hague Rules, it is apparent that they assume that the carrier does indeed have the burden of disproving negligence albeit without imposing that burden on him in terms. This is because of the relationship between articles III and IV. Article III.2 is expressly subject to article IV. A number of the exceptions in article IV cover negligent acts or omissions of the carrier which would otherwise constitute breaches of article III.2: for example articles IV.1 and IV.2(a). It is common ground, and well established, that the carrier has the burden of proving facts which bring him within an exception in article IV, and in the case of articles IV.1 and IV.2(q) this is expressly provided. It would be incoherent for the law to impose the burden of proving the same fact on the carrier for the purposes of article IV but on the cargo owner for the purposes of article III.2. As will be seen below, a rather similar problem arises in relation to the exception for inherent vice.

Nothing in the Hague Rules alters the status of a contract of carriage by sea as a species of bailment for reward on terms. As Hobhouse J pointed out in Aktieselskabet de Danske Sukkerfabrikker v Bajamar Compania Naviera SA (The “TORENIA”) [1983] 2 Lloyd’s Rep 210, 216:

For these reasons I consider that in principle where cargo was shipped in apparent good order and condition but is discharged damaged the carrier bears the burden of proving that that was not due to its breach of the obligation in article III.2 to take reasonable care. I say “in principle” because it is next necessary to consider whether the authorities in cases governed by the Hague Rules point to a different rule. I turn, therefore, to the authorities.

The first case to address directly the burden of proof in relation to article III.2 was Gosse Millard v Canadian Government Merchant Marine Ltd[1927] 2 KB 432, a decision of Wright J shortly after the Carriage of Goods by Sea Act came into force. On the footing that the damage ascertained on outturn had not been explained, he held the carrier liable because the burden of disproving negligence lay on him. He gave two reasons. The first was that the words “properly … discharge” in article III.2 of the Hague Rules meant deliver in the same apparent order and condition as on shipment, so that if damage was ascertained on discharge there was a “prima facie breach”. Wright J did not mean by this that the damage was prima facie evidence of breach of the article. He meant that the carrier was in breach unless he could excuse himself under article IV: see pp 435-436. This reason was bad, as the House of Lords later held in G H Renton & Co Ltd v Palmyra Trading Corpn of Panama[1957] AC 149. “Properly” in article III.2 did not impose an obligation to achieve a particular outcome, but to load, carry and discharge “in accordance with a sound system”: see pp 166 (Viscount Kilmuir LC), 169-170 (Lord Morton of Henryton), 174 (Lord Somervell of Harrow), approving the statement of Lord Devlin in Pyrene Co Ltd v Scindia Navigation Co Ltd[1954] 2 QB 402, 417-418. That error did not, however, affect Wright J’s second reason, which was that the carrier was a bailee. Citing the pre-Hague Rules cases on the liability of carriers as bailees, he put the point in this way, at pp 435-436:

Wright J’s decision was followed by Scrutton LJ, delivering the leading judgment in the Court of Appeal in Silver v Ocean Steamship Co Ltd[1930] 1 KB 416, 424-425. He held, as Wright J had done, that the Hague Rules had made no difference to the incidence of the burden of proof in cases of bailment for carriage.

I have already referred to the statement of Hobhouse J in The “TORENIA” [1983] 2 Lloyd’s Rep 210, 216 about the importance of the legal characterisation of a contract of carriage as a bailment. Hobhouse J went on to state the law as follows at pp 216-217:

Scrutton LJ and Lords Wright and Hobhouse (as they later became) were notable authorities in this area of law. Apart from Wright J’s error as to the meaning of “properly … to discharge” in article III.2, an error which was not made by either Scrutton LJ or Hobhouse J, their analyses of the burden of proof under contracts of carriage incorporating the Hague Rules are in my view entirely in accordance with principle.

This proposition has sometimes been expressed by saying that once it is shown that the cargo was loaded in good condition and discharged in bad, the carrier bears the burden of proving that this was caused by one of the excepted perils in article IV: see, for example, the statement of Viscount Sumner in F C Bradley & Sons Ltd v Federal Steam Navigation Co Ltd (1927) 27 Ll L Rep 395, 396, and of Lord Wright, delivering the advice of the Privy Council in Paterson Steamships Ltd v Canadian Co-operative Wheat Producers Ltd[1934] AC 538, 545-546. This formulation is common in codes giving effect to the Hague Rules in civil law jurisdictions, as I have observed. But it is not entirely satisfactory to an English lawyer, because it misses out a stage of the analysis. The true rule is that the carrier must show either that the damage occurred without fault in the various respects covered by article III.2, or that it was caused by an excepted peril. If the carrier can show that the loss or damage to the cargo occurred without a breach of the carrier’s duty of care under article III.2, he will not need to rely on an exception.

So far as that analysis has been doubted, it is because of dicta in the House of Lords in the Scottish case of Albacora SRL v Westcott & Laurence Line Ltd 1966 SC(HL) 19; and in the High Court of Australia in Great China Metal Industries Co Ltd v Malaysian International Shipping Corpn Bhd (The “BUNGA SEROJA”) [1999] 1 Lloyd’s Rep 512.

The “Albacora” is important mainly for its analysis of the meaning of inherent vice, and I shall return to it in that context. However, Lord Pearce expressed doubt in that case about the correctness of Wright J’s view that the burden of disproving negligence lay upon the carrier, without giving reasons. And Lord Pearson rejected Wright J’s view, giving as his reason that Wright J had been wrong about the meaning of “properly … to discharge”. In The “BUNGA SEROJA” McHugh J dealt briefly with the burden of proof under article III.2 of the Hague Rules in the following terms, at para 98:

Article IV.2 of the Hague Rules is a notoriously unsatisfactory provision, because there is no unifying legal principle behind the highly miscellaneous list of excepted causes of loss. Some of them refer to matters which by their nature would otherwise constitute breaches of the carrier’s duty to care for the cargo. Some refer to matters which may or may not be caused by such a breach. In other cases, such as act of God, the carrier would not be liable even in the absence of an exception. The explanation for this intellectual disorder is historical. The exceptions are generally those which were allowed by the draftsmen of the Rules because their inclusion in bills of lading was sanctioned by long-standing practice, or because they were common law exceptions to the liability of a common carrier, or because they were excepted in existing national legislation such as the US Harter Act and corresponding legislation in Canada and Australia. Only one of the article IV.2 exceptions expressly imposes the burden of proof on the carrier, namely (q). It is, however, well established that the carrier bears the burden of bringing himself within any of the exceptions. Mr Rainey does not challenge this. His case is that once he has proved that the cargo suffered from an inherent vice, the cargo-owner must positively prove that it was only because of the carrier’s negligence that the cargo’s vicious propensities resulted in damage.

The starting point in the authorities is the decision of the Exchequer Chamber in Notara v Henderson(1872) LR 7 QB 225. The facts were that a cargo of beans was delivered damaged by seawater. The beans had been wetted when the vessel was involved in a collision. She put into Liverpool for repairs, and it was proved that it would have been reasonable for the master temporarily to discharge the beans there, so that they could be spread out and dried in a warehouse, and then reloaded before the vessel proceeded on her voyage. If that had been done, part of the damage would have been avoided. The bill of lading excepted “loss or damage arising from … accidents of the seas”. The court held that the exception did not protect the carrier from liability for that part of the damage which could have been avoided by the exercise of due care. Willes J, delivering the judgment of the court, stated the law as follows at pp 235-236:

The decisions of the House of Lords in Thomas Wilson, Sons & Co v Owners of the Cargo per The “XANTHO”(1887) 12 App Cas 503 and Hamilton, Fraser & Co v Pandorf & Co(1887) 12 App Cas 518, were handed down on the same day by differently constituted appellate committees. In both cases, it was held that a bill of lading exception for “dangers and accidents of the seas” excused the carrier only if the relevant danger or accident happened without his fault. The law on this point can conveniently be taken from The “XANTHO”. Lord Herschell, delivering the leading speech, put the matter in this way at pp 510-511:

The sheet-anchor of the carriers’ case on this appeal is the decision of the Court of Appeal in The “GLENDARROCH” [1894] P 226, where the question left open by Lord Herschell was decided. The “GLENDARROCH” was decided 30 years before the Hague Rules were adopted. It concerned a specific exception assumed to be included in a bill of lading. Like most of the case law on the burden of proving an exception in the contract of carriage, it was about the exception for perils of the sea, which was commonly included in bills of lading before the Hague Rules and was reproduced in article IV.2 of the Rules as exception (c). The facts were that the cargo suffered water damage because the vessel ran aground on St Patrick’s causeway in Cardigan Bay. During the trial, the judge ruled that the carrier had the burden of proving that the grounding had occurred without negligence in the navigation of the vessel. Thereupon, the carrier’s counsel declined to call any evidence to explain how the vessel came to be in a place where the action of the wind and waves was liable to put her aground. Accordingly, the judge entered judgment for the cargo-owner. The Court of Appeal allowed the appeal, holding that the burden of proving that an excepted peril had been occasioned by the carrier’s negligence lay on the cargo owner. Today, the result on the same facts would be the same because, assuming that the vessel was initially seaworthy, the carrier would be exempted from liability for negligent navigation under article IV.2(a) as well as perils of the sea under article IV.2(c). But that is by the way. What matters for present purposes is the reasoning of the Court of Appeal on the footing that the only relevant exception was for perils of the sea. Lord Esher MR delivered the leading judgment. He appears to have regarded the shipowner as a common carrier, for he observed (p 230) that but for the contractual exception for perils of the seas, the carrier would have been strictly liable and negligence “utterly immaterial”. On that footing, any negligent navigation of the vessel could be relevant to the exception for perils of the seas only if one read it as meaning “Except the loss is by perils of the sea, unless or except that loss is the result of the negligence of the servants of the owner.” Lord Esher proceeded on the assumption that that construction was correct. But he held that on that footing negligent navigation was an exception to an exception. It followed that once the carrier had proved that the wind and waves had deposited the vessel on St Patrick’s causeway, he had done all that the exception required of him, and the burden of proving that this had happened only because of its negligent navigation by the crew lay upon the cargo-owner. He referred in support of this conclusion to the practice of the court regarding the pleading of exceptions upon exceptions. The reasoning of Lopes LJ was the same. As he pointed out (p 235), the result was that

The decision in The “GLENDARROCH” derives some support from obiter dicta by Lord Sumner in Owners of Steamship “MATHEOS” v Louis Dreyfus & Co[1925] AC 654, 666, Lord Wright in Joseph Constantine Steamship Line v Imperial Smelting Corpn Ltd[1942] AC 154, 194, Lord Pearson in The “Albacora”,supra, at p 31, and Mason and Wilson JJ in Shipping Corpn of India Ltd v Gamlen Chemical Co (A/Asian) Pty Ltd(1980) 147 CLR 142, 167-168, although in none of these cases were the authorities examined. None of them, moreover, confronted the practical and conceptual problems about the analysis in The “GLENDARROCH”. Fundamental to that analysis was a distinction between an exception and an exception to an exception. This distinction is in my view unsatisfactory. This is partly because the attempt to distinguish between the case where absence of fault is part of the test for the exception and the case where it is an exception to the exception seems to me to import a refinement of some subtlety, unrelated to any commercial purpose which the parties can sensibly be thought to have had in mind. But it is mainly because if an exception is subject to an exception for cases where it was avoidable by the exercise of due care, then the issue must ultimately be one of causation. Thus, in The “GLENDARROCH” itself, the question was whether the effective cause of the loss was the action of the wind and waves, or the conduct of the crew in allowing the action of the wind and waves to damage the cargo when with reasonable diligence on their part both ship and cargo could have withstood the storm. If, as the Court of Appeal rightly accepted in that case, the burden of proving facts bringing the carrier within the exception lay on him, that must extend not just to the question whether the sea conditions were perilous, but also to the question whether that was the effective cause of the damage. That is unquestionably the position of a bailee at common law: Coldman v Hill[1919] 1 KB 443; British Road Services Ltd v Arthur V Crutchley & Co Ltd (No 1) [1968] 1 All ER 811, 822, 824 (Sachs LJ, with whom Danckwerts LJ agreed). It is also, under the express terms of article IV.2 of the Hague Rules an integral part of what must be proved for any of the exceptions to apply (“Neither the carrier nor the ship shall be responsible for loss or damage arising or resulting from …” (emphasis supplied).) Yet, as Lopes LJ recognised, the reasoning of the Court of Appeal involved a distinction between the existence of the excepted circumstance on the one hand and its causative effect on the other.

The “GLENDARROCH” has stood for a long time. But it has rarely featured in the reasoning of subsequent case law, and the basis on which it was decided is technical, confusing, immaterial to the commercial purpose of the exception and out of place in the context of the Hague Rules. The decision may have been justifiable in the more formal conditions of pleading and trial practice in the 1890s, or as applied to the notional bill of lading terms which the Court of Appeal was considering. But as the source of a general rule governing the burden of proof, it should no longer, in my view, be regarded as good law. I consider that the carrier has the legal burden of disproving negligence for the purpose of invoking an exception under article IV.2, just as he has for the purpose of article III.2.

Even if I had thought that The “GLENDARROCH” was correct as applied to the exception for perils of the sea, I would not have regarded it as applicable to the exception for inherent vice. This is because the distinction between the existence of the peril and the standard of care required of the carrier is impossible to make in that context. A cargo does not suffer from inherent vice in the abstract, but only in relation to some assumed standard of knowledge and diligence on the part of the carrier. Thus the mere fact that coffee beans are hygroscopic and emit moisture as the ambient temperature falls may constitute inherent vice if the effects cannot be countered by reasonable care in the provision of the service contracted for, but not if they can and should be. At the time that The “GLENDARROCH” was decided, this had already been established by the decision of the Court of Appeal in Nugent v Smith(1876) 1 CPD 423. In that case, a mare carried in the hold died as a result of a combination of more than usually bad weather and the fright of the animal herself which caused her to struggle and injure herself. Cockburn CJ described inherent vice (p 438) as the rule that

The leading modern case is The “Albacora”, which I have already mentioned in another context. It is authority for the proposition that the standard of care by reference to which the exception for inherent vice is to be assessed may depend on the nature of the service contracted for. The issue was whether a cargo of fish was capable of withstanding carriage in unrefrigerated spaces, that being the service stipulated in that case. The ratio can conveniently be taken from the speech of Lord Reid, at p 23:

The effect of these statements, and others to the same effect, are accurately summarised in Scrutton on Charterparties and Bills of Lading 23rd ed (2015), para 11.055:

It follows that if the carrier could and should have taken precautions which would have prevented some inherent characteristic of the cargo from resulting in damage, that characteristic is not inherent vice. Accordingly, in order to be able to rely on the exception for inherent vice, the carrier must show either that he took reasonable care of the cargo but the damage occurred nonetheless; or else that whatever reasonable steps might have been taken to protect the cargo from damage would have failed in the face of its inherent propensities.

The judgment of the Court of Appeal was given by Flaux J, with whom Gloster and King LJJ agreed. Flaux J’s analysis of the burden of proof is not entirely clear. He appears to have considered that if damage was ascertained on outturn, the cargo owner had the legal burden of proving that this was due to a breach of article III.2, but that there was an evidential inference from the mere fact of damage that such a breach had occurred. The carrier would have to point to evidence to rebut the inference or fail on that issue. On that footing, there appears to have been little practical difference between a legal and an evidential burden. Once a breach of article III.2 had been established (whether by evidence or inference), Flaux J considered that the carrier had the legal burden of establishing a “prima facie case” for the application of one of the exceptions in article IV.2. Flaux J cannot have been using the expression “prima facie case” in its ordinary sense, ie a case on the facts which unless rebutted by further evidence would entitle the carrier to succeed. He must, I think, have meant that the carrier had the legal burden of proving that one of the circumstances listed in article IV.2 existed at the relevant time. It was then, he thought, for the cargo owner to bear the legal burden of showing that it only resulted in damage to the cargo because the carrier had failed to take reasonable care of it.

I have already explained why I do not accept this analysis in principle. I also have great difficulty with the way in which Flaux J applied it to the exception for inherent vice. What is a “prima facie case” for the application of that exception? Flaux J seems to have thought that the carrier need only prove that the cargo had an inherent propensity to deteriorate, but not that he took reasonable care to prevent that propensity from manifesting itself. He criticised the deputy judge for “conflating the issues of whether there was some inherent defect, quality or vice in the cargo and whether the carrier properly and carefully cared for and carried the cargo.” At the same time, he recognised that there was “a degree of overlap, if not of circularity” between these two things, “in the sense that one is focussing on the ability of the cargo to withstand the ordinary incidents of carriage, pursuant to obligations of the carrier under the contract of carriage.” The reason why, in spite of the overlap or circularity, he nevertheless felt it necessary to treat them as discrete questions was that he thought that the burden of proof lay with the carrier on the former issue and the cargo owner on the latter. The problem with this analysis is that, as I have observed, and as Flaux J came close to acknowledging, it is conceptually impossible to define inherent vice except by reference to some assumed standard of care for the cargo. A cargo may have inherent characteristics that make its deterioration inevitable whatever care is taken of it. In that case negligence is irrelevant and inherent vice is proved without more. But inevitable deterioration is rare in cargoes that are habitually carried by sea. In the great majority of cases, the cargo’s inherent propensity to deteriorate may or may not manifest itself in damage, depending on the ambient conditions of stowage and the way it is handled. If, within the limits of the kind of carriage contracted for, reasonable care would have prevented the cargo’s inherent propensity from causing damage, then the cargo is fit to withstand the ordinary incidents of the carriage contracted for and there is no inherent vice. This makes it difficult to support Flaux J’s analysis of the incidence of the burden of proof. If the existence of inherent vice depends on the appropriate standard of care, the law cannot coherently apply a different burden of proof to one of them from that which applies to the other.

Having held that the cargo owner had the burden of proving that the carrier’s absence of care had caused the exception to operate, the Court of Appeal went on to find that he had failed to discharge it. This was not because of the absence of evidence of matters which it was for the cargo owner to prove. Starting from the proposition that the carrier’s obligation under article III.2 was to care for and carry the goods “in accordance with a sound system”, they thought that the deputy judge had misdirected himself that this meant in accordance with a system that would prevent damage, and that inherent vice could be demonstrated only if damage was inevitable. I am not persuaded that the deputy judge made either of these mistakes. He simply observed that at the relevant time bagged coffee was commonly carried in unventilated containers from warm climates to cooler ones without mishap, and there was nothing out of the ordinary about this particular cargo. It therefore seemed probable that with reasonable care the cargo was perfectly capable of withstanding the risks reasonably to be expected during unventilated carriage. But it is unnecessary to examine this point further, because while the Court of Appeal’s criticisms encouraged them to reopen the facts, the decisive reason why they overruled him was that they disagreed with two of his critical conclusions about the evidence, and made positive findings of their own which he had felt unable to make. In the first place, the deputy judge had found that there was no evidence of any generally accepted industry practice to which the carrier could claim to have conformed. The Court of Appeal found that there was an accepted industry practice in 2012 for lining unventilated containers for the carriage of bagged coffee. It was, they said, to use two layers of paper of at least 80 gsm or one layer of at least 125 gsm. Secondly, the deputy judge had found that apart from a tentative conclusion to be drawn from the photographs, there was no evidence of the number of layers of paper in place at the time of shipment, and no evidence of its weight except that it was more than 80 gsm. The Court of Appeal found by an examination of the photographs and of the pre-loading documentation that two layers of paper had been used. It therefore followed from the deputy judge’s finding that the paper used weighed at least 80 gsm that the containers had been lined in accordance with accepted industry practice.

This court has on a number of occasions pointed out that while an appeal to the Court of Appeal is by way of rehearing, a trial judge’s findings of fact should not be overturned simply because the Court of Appeal would have found them differently. It must be shown that the trial judge was wrong: ie that he fundamentally misunderstood the issue or the evidence, or that he plainly failed to take the evidence into account, or that he arrived at a conclusion which the evidence could not on any view support. Within these broad limits, the weight of the evidence is a matter for the trial judge. There is a world of difference between the impression which evidence makes on a judge who has followed it as it was deployed and the impression that an appellate court derives from cold transcripts. The judgment of Flaux J in the Court of Appeal attaches no intrinsic weight to the deputy judge’s analysis. It simply substitutes his own.

In my judgment, the Court of Appeal was not justified in overturning the deputy judge’s findings on either of the two critical points which I have identified. On the question of industry practice, the judge had before him a joint memorandum by the expert witnesses, referring to a number of recommendations, none of them more recent than 2004. The issue was addressed by the experts in their oral evidence in terms which were not wholly consistent. Some of it suggested that in 2012 practice in the trade was variable, fluid and developing. The assessment of this evidence was very much a matter for the deputy judge. He found that it fell short of proving any sufficiently uniform or accepted practice to serve as a benchmark against which to measure the carrier’s care for the cargo. To my mind this was a conclusion which was open to him on the material deployed at trial. Turning to the paper actually used, the Court of Appeal thought that the deputy judge’s conclusions were “against the weight of evidence”. There was an issue at trial about what could be deduced from the reports of inspectors who examined the containers on outturn, on which the evidence was inconclusive and the deputy judge made no finding. There was an issue about the weight to be attached to documentary evidence suggesting that before shipment it had been intended to use two layers of paper. The deputy judge discounted this, mainly because of the absence of documentation such as invoices which one would have expected to find if this intention had been carried into execution. That left the inferences to be drawn from photographs taken at the port of loading. The deputy judge thought that they tended to show that single layers of paper had been used, the Court of Appeal that it was doubled. The difference depends on whether certain of the photographs showed double layers or an overlap of sheets laid as single layers. For my part, I think that there is some force in the criticism that the deputy judge too readily discounted the documentary evidence generated before shipment. But this was only part of a larger and more complex body of evidence which, taken as a whole, could have supported either of the competing analyses. This falls well short of what was necessary to show that the deputy judge was wrong.

I would hold that the carrier had the legal burden of proving that he took due care to protect the goods from damage, including due care to protect the cargo from damage arising from inherent characteristics such as its hygroscopic character. I would reinstate the deputy judge’s conclusions about the practice of the trade in the lining of unventilated containers for the carriage of bagged coffee and the absence of evidence that the containers were dressed with more than one layer of lining paper. In the absence of evidence about the weight of the paper employed, it must follow that the carrier has failed to prove that the containers were properly dressed.

For these reasons, I would allow the appeal and restore the order of the deputy judge.