England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE QUEEN'S BENCH DIVISION, COMMERCIAL COURT
Sir Jeremy Cooke sitting as a Judge of the High Court
CL/2016/000070; [2016] EWHC 2514 (Comm)

		Royal Courts of Justice Strand, London, WC2A 2LL
		22/02/2018

Lord Justice Flaux:

Introduction

1. The appellant appeals with the leave of Gross LJ against the Order of Sir Jeremy Cooke (sitting as a High Court Judge in the Commercial Court) dated 3 November 2016, declaring that the appellant is not entitled to limit its liability to the respondents for damage to cargo carried on its vessel to the sum of £54,730.90. The appeal raises for decision the long-standing debate as to the meaning of "unit" in Article IV rule 5 of the Hague Rules and, specifically, whether "unit" refers to a physical item of cargo or shipping unit as the respondents contend or is a reference to a unit of measurement as used by the parties to denominate or quantify the cargo in the contract of carriage and is thus capable of applying to bulk or liquid cargo, as the appellant contends.

Factual background

2. The facts are and were not in dispute and can be summarised as follows. The dispute arises out of damage to a cargo of fish oil in bulk carried on board the tanker AQASIA pursuant to a charterparty between the appellant as owner and the first respondent as charterer contained in or evidenced by a Fixing Note dated 23 August 2013. The charterparty provided for the carriage of 2,000 tons of fish oil in bulk (5% more or less in charterer's option) from Iceland to Norway on board the vessel WEST STREAM for a lumpsum freight of Nok 817,500. The appellant subsequently nominated the AQASIA, of which it was the disponent owner, as the substitute performing vessel.
3. The Fixing Note provided that the charterparty was to be on the "London Form", an old tanker voyage charter form, replaced in common usage by the Intertankvoy 76 form. The London Form provides inter alia as follows:

"26. – The Owners in all matters arising under this Contract shall also be entitled to the like privileges and rights and immunities as are contained in Sections 2 and 5 of the Carriage of Goods by Sea Act 1924 and in Article IV of the Schedule thereto …"

4. The charterparty thus incorporated Article IV of the Schedule to the 1924 Act which contains the Hague Rules. Article IV rule 5 provides:

"… Neither the carrier nor the ship shall in any event be or become liable for any loss or damage to or in connection with goods in an amount exceeding 100l per package or unit, or the equivalent of that sum in other currency, unless the nature and value of such goods have been declared by the shipper before shipment and inserted in the bill of lading ...."

5. The Fixing Note also incorporated various rider clauses including terms providing for the charterparty to be governed by English law and for disputes to be resolved by London arbitration.
6. On 6 September 2013, the vessel loaded a cargo of 2,056,926 kg of the first respondent's fish oil in bulk at two Icelandic ports. About 550,000 kg of the cargo ("the subject cargo") was loaded in tanks 1P, 2P and 5S. The Master (or his agent) signed a Congen bill of lading acknowledging shipment of the cargo in apparent good order and condition. The first respondent was named as the shipper of the cargo. The bill recorded the shipper's description of the cargo as "Icelandic Fishoil in bulk-2,056,926kgs". The appellant, as disponent owner of the vessel, was not party to the contract of carriage contained in or evidenced by the bill of lading, which was with the head owners. It was common ground that it was the charterparty which contained or evidenced the contract of carriage between the parties to the present proceedings.
7. After loading, the vessel proceeded to Lovund in Norway and loaded another cargo of fish oil, some of which was loaded in tanks 1P, 2P and 5S and thus became comingled with the subject cargo. On arrival at the discharge port(s), 547.309 metric tons (547,309 kg) of the subject cargo was found to have suffered damage.
8. The first respondent claimed damages from the appellant for the loss it has suffered as owner of the subject cargo, in the sum of US $367,836. The second respondent was the insurer of the cargo, joined to the proceedings out of an abundance of caution, lest title to sue had somehow passed from the first to the second respondent. The appellant accepted liability in principle for the damage to the subject cargo but contended that it was entitled to limit its liability to £54,730.90 (i.e. £100 per metric ton of cargo damaged) under article IV rule 5 of the Hague Rules. This was contested by the respondents.
9. The parties agreed that, notwithstanding the arbitration clause in the charterparty, the Commercial Court should have jurisdiction to determine the agreed preliminary issue as to whether the appellant was entitled to limit its liability to £54,730.90.

The judgment below

10. The judge dealt first at [5] to [8] with what was then the appellant's primary case, that when read as a whole as a contractual term of the charterparty, Article IV was clearly intended to apply to a bulk cargo, because the London Form was only for use with bulk or liquid cargoes. All of Article IV had to apply in order to give meaning to the parties' agreement that the appellant was to have the privileges, rights and immunities afforded by Article IV. The judge rejected the submission that the effect of clause 26 was that the words of Article IV were written into the charterparty so that every provision in the Article must be given meaning and effect in the context of the carriage of the bulk cargo contemplated.
11. The judge held at [5] that: the effect of the words: "the like privileges and rights and immunities as are contained in…Article IV" in clause 26: "must be that the Defendant is entitled to rely on the package or unit limitation only in the same circumstances as it would be entitled to do, if there had been a full incorporation of the Hague Rules". He went on to hold at [7] that:

"The effect of the clause is to allow the Defendant the like privileges, rights and immunities as are contained in Article IV, where they apply. Although it is true to say that the Charterparty was expressly a charter for the carriage of a bulk cargo of fishoil in a tanker, the Defendant is entitled to rely on no more than what is provided by the limitation in Article IV, so that if the word unit, as used in Article IV, does not apply to bulk cargoes as a matter of construction of the Rules, it cannot change its meaning because of the nature of the contract of carriage. There may be parts of Article IV which are applicable and other parts which are inapplicable. It is accepted that Article IV r.1, 2 and 4 provide protection to the owner but it is self-evident that Article IV r.2(n) which exempts the owner from liability for "insufficiency of packing" cannot apply to a cargo which is not packed. Such a provision, although part of the charter, is inapplicable to the factual situation which obtains in the carriage of a bulk cargo. Similarly, it is accepted that the words "per package" in Article IV r. 5 cannot apply to a bulk cargo, so the determinative issue is whether or not the word "unit" in that Article can do so."

12. One of the appellant's grounds of appeal is that the judge was wrong in not accepting the argument as to the correct construction of the charterparty, that it was clearly the intention of the parties that the appellant's liability under the charterparty be limited, although it is fair to say that Mr Debattista for the appellant placed a great deal more emphasis in his appeal skeleton and his oral submissions before this Court on his other ground of appeal as to the true meaning of "unit" in the Hague Rules themselves.
13. The judge then set out at [10] that it was accepted that, as a matter of ordinary language, the word "unit" is capable of referring to either an individual physical item or a unit of measurement. He referred to the appellant's argument by reference to various provisions in the Hague Rules, that "unit" meant a unit of measurement. I do not set out all those arguments here as they are in large measure the same arguments as Mr Debattista advanced before this Court, which I consider later in this judgment.
14. At [13], the judge then set out the argument of Mr Lionel Persey QC for the respondents that the context in which the word was used in the phrase "per package or unit" in Article IV rule 5 referred to a physical item or composite of items rather than a unit of measurement. Because a "package" was undoubtedly a physical item, the use of the words together pointed to their both being concerned with physical items rather than abstract units of measurement, on the principle of noscitur a sociis. It was argued that "package" was also used in Article III rule 3(b) which refers to "packages or pieces or the quantity or weight" where "packages or pieces" were referring to physical items of cargo and quantity or weight were seen as different, hence their being specified.
15. The judge accepted these arguments, saying at [14] to [16]:

"[14] I find these arguments compelling, as have others, since, despite the Defendant's best efforts, as set out above, I can see nothing in the Hague Rules which lends any support for the argument that the word unit connotes a unit of measurement, in circumstances where the Rules specifically refer to quantity or weight when measurable units are in mind.

[15] The Defendant did not submit that the word "unit" in the Rule meant a unit of measurement for all purposes, because it was accepted that it covered unpackaged items for shipment. The Defendant argued, however, that the word was apt to cover unpackaged physical items as units of shipment but was also apt to cover a unit of measurement in the case of bulk cargoes. This creates an obvious issue, where the word is given different meanings for different types of cargo. It also can be seen as creating a problem in the case of a package where a weight or volume also appears on the Bill of Lading. If there is one package and a weight, which give rise to different limitation amounts, which is to be taken? Unlike the Hague- Visby Rules which provide for the application of the higher of the limits assessed in accordance with their terms, there is no such provision here. Moreover, if the word is apt to cover both a shipping "unit" in the sense of an unpackaged item, such as a car, and a unit of measurement expressed on the Bill of lading, such as the weight of such a car, which of those two measures is to be taken, in the absence of any guidance in the Rules themselves? These appear to me to be powerful points against the argument that "unit" can mean both a shipping unit in the sense of a physical unpacked object and a unit of measurement, whether for freight purposes or otherwise. The choice, as appears from many commentaries, lies between a shipping unit in the sense described and a unit of measurement, and most invariably, if not always, seen as that utilised for freight purposes in the light of the "customary freight unit" which is the expression employed in the US Carriage of Goods by Sea Act 1936, instead of the word "unit" used in r.5.

[16] As appears later in this judgment, the construction of "unit" which I adopt as a physical shipment unit is the one favoured by the majority of the commentaries and text books on the point and is the construction accepted by courts in other common law jurisdictions."

16. The judge then went on to consider the extent to which the Hague-Visby Rules or the United States Carriage of Goods by Sea Act 1936 were aids to construction of Article IV rule 5 of the Hague Rules, which he held they were not. He then set out in detail the travaux préparatoires for the Hague Rules, the authorities, English and Commonwealth, and the views of textbooks and academic commentators. Since I propose to engage in essentially the same exercise below, it is not necessary to set out those sections of his judgment. I refer to the relevant passages of the judgment in my Analysis and Conclusions.
17. The judge concluded at [59]:

"When regard is had to the English authorities, the Commonwealth authorities, travaux préparatoires and the textbooks and commentaries, I have no hesitation in coming to the conclusion that the word "unit" in the Hague Rules can only mean a physical unit for shipment and cannot mean a unit of measurement or customary freight unit as is the case in the United States."

18. He said that, in those circumstances, it was unnecessary to determine how a limit based on a unit of measurement would apply on the facts of the case, but since the point had been argued, he would decide it. The judge recorded the argument for the appellant at [60] in these terms:

"The Defendant argued for a limit per ton on the basis that the Charter party referred to a cargo of 2,000 tons, albeit that the Charterers had an option to declare 5% more or less. The Defendant did not contend that the court should apply the customary freight unit for which the US Carriage of Goods by Sea Act provides. The freight under the Charterparty was a lump sum freight and there is a decision of the United States District Court for the Southern District of New York, Ulrich Amman Building Equipment Ltd v M/V Monsun 609 F. Supp. 87 (S.D.N.Y. 1985) to the effect that a lump sum freight is simply one freight unit which would here give rise to a pitiable limit. The expression "customary freight unit" is defined in that case as meaning 'the unit of cargo customarily used as the basis for the calculation of the freight rate to be charged'.

19. The judge then referred to the rival alternatives as to the meaning of "unit" described by the editors of Scrutton in the 18^th edition at pp 442-3 (with which I deal in more detail below), that is, "shipping unit" or "freight unit". He noted at [62] that Mr Debattista's argument did not follow that analysis, but simply wished to take the unit of measurement for the cargo expressed in the charterparty and apply the limitation to that as if a metric ton were the relevant unit, but the judge noted that, as the respondent pointed out, the bill of lading referred to kilograms and limitation by reference to kilograms would lead to a figure greater than the claim. Having considered the difficulties with the appellant's argument, the judge concluded at [63] and [65] that the only realistic alternative to the shipping unit construction which he had accepted was the customary freight unit measure of limitation. As the judge said, there was no benefit to the appellant adopting that measure, because of the lump sum nature of the freight.

Grounds of appeal

20. In summary, the appellant pursues two grounds of appeal:

(1) That the judgment failed to give effect to the clearly expressed intention of the parties to the charterparty that the appellant should be entitled to limit its liability in respect of this bulk cargo pursuant to Article IV rule 5.

(2) That the judge erred in concluding that the limitation of liability in Article IV rule 5 of the Hague Rules did not apply to bulk cargo in a number of respects.

Summary of the parties' submissions

21. Given that my analysis of the issues raised by this appeal set out below identifies the principal submissions advanced by the parties, it is not necessary to set them all out in detail. However, in summary, on behalf of the appellant, Mr Debattista's principal submissions were as follows:

(1) Since it was common ground that as a matter of ordinary language, "unit" can mean either a physical item of cargo or a unit of measurement, the judge erred in not concluding that as a matter of construction of Article IV rule 5 of the Hague Rules, the word did not have that dual meaning but concluding that the word in the rule meant a shipping unit or physical item of cargo.

(2) Resort should only be had to the travaux préparatoires of the Hague Rules if the respondent could establish that the meaning of "unit" was ambiguous or obscure which it could not and, in any event, the respondent could only rely upon the travaux préparatoires in support of its construction if it could demonstrate a "bull's eye" which it could not.

(3) The judge gave insufficient weight to the definition of "goods" in Article I.

(4) The judge erroneously assumed that the duality of meaning for which the appellant contends would give rise to problems for example where both the quantity of units of cargo and their weight were specified in the bill of lading.

(5) The judge misinterpreted the decision of Evans J in The Aramis [1987] 2 Lloyd's Rep 58 where, so Mr Debattista submitted, that judge had assumed that the limitation provision in Article IV rule 5 could apply to bulk cargo.

(6) The judge had placed too much weight on the Commonwealth authorities which he had described as "directly on point". They could not be so described because they were not concerned with bulk cargoes.

(7) The judge erred in concluding that the preponderance of the textbooks and commentaries favoured the construction he had reached.

(8) That the judge failed to have regard to the market expectations and practice of those in the bulk shipping market who continued to use charterparty forms such as the London Form with provisions like clause 26, expecting that they would be able to limit liability under Article IV rule 5.

(9) On the appellant's alternative case, the judge erred in failing to give effect to the parties' intentions under this charterparty that the appellant should be entitled to limit its liability in respect of this bulk cargo.

(10) The judge should have concluded that the appellant was entitled to limit its liability on the basis that the metric ton enumerator in the charterparty constituted the relevant "unit".

22. The submissions on behalf of the respondent by Mr Lionel Persey QC can be summarised as follows:

(1) The judge's analysis of the meaning of "unit" in Article IV rule 5 as a physical item of cargo or shipping unit was entirely correct.

(2) The travaux préparatoires clearly confirmed that meaning. If, which he disputed, it was necessary to demonstrate a "bull's eye" in the travaux préparatoires, the respondent could demonstrate several.

(3) The judge had interpreted The Aramis correctly. There was no English or Commonwealth authority where the construction for which the appellant contended had been accepted. It had been expressly rejected in Canada and Australia.

(4) The judge had also been correct in his assessment of the textbooks and the academic commentaries. The preponderance of opinion was in favour of the construction of "unit" as meaning physical item of goods or shipping unit.

(5) There was no basis for reliance on alleged market expectations and practice of which there was no evidence and which could not, in any event, alter the clear meaning of the word "unit".

(6) The alternative case was misconceived. Clause 26 only gave the appellant the privileges, rights and immunities of Article IV to the extent they were applicable and there was no warrant for concluding that rule 5 had a different meaning under the charterparty than in the Hague Rules themselves.

(7) The judge had been right to conclude that, even if the appellant was right as to the meaning of "unit", its attempt to limit liability failed on the facts.

Analysis and conclusions

The meaning of "package or unit" in Article IV rule 5

23. Whilst, as the respondents accept, as a matter of ordinary language the word "unit" is capable of meaning both a physical item of cargo, a shipping unit, and a unit of measurement, such as weight or volume, the critical question on this appeal is what the word means in the particular context in which it is used in Article IV rule 5. Despite Mr Debattista's submissions to the contrary, I have reached the firm conclusion that in the context of the Hague Rules, "unit" means a physical item of cargo, not a unit of measurement, for the following reasons.
24. First, the word "package" is clearly referring to a physical item and the use of the words "package" and "unit" together and in the same context points strongly to both words being concerned with physical items rather than units of measurement, on the principle noscitur a sociis, as the judge held.
25. Second, that "package or unit" both relate to physical items of cargo is borne out by Article III rule 3(b) of the Hague Rules, which provides:

"After receiving the goods into his charge the carrier or the master or agent of the carrier shall, on demand of the shipper, issue to the shipper a bill of lading showing among other things:

(b) Either the number of packages or pieces, or the quantity, or weight, as the case may be, as furnished in writing by the shipper."

26. Whilst it is true that the phrase used there is "packages or pieces" rather than "package or unit" (indeed, "unit" is not used elsewhere in the Rules than in Article IV rule 5), it is clearly referring to individual physical items of cargo, in contradistinction to the weight or quantity of cargo which were, as the judge said at [13]: "seen as different and specified accordingly". I consider that, in the context of the Rules, a "unit" can be regarded as synonymous with a "piece", they are both descriptive of a physical item of cargo which is not a "package", because, for example, it is incapable of being packaged or is not in fact packaged. Contrary to Mr Debattista's submissions, I do not consider that "pieces, quantity or weight" all fit into a "portmanteau" definition: "unit". Rather, where the Rules wish to refer to the weight of cargo, they do so specifically as in Article III rule 3 or later in that Article at rule 5.
27. Third, contrary to Mr Debattista's submission, I do not consider that the definition of "goods" in Article I of the Hague Rules is of any particular assistance in construing Article IV rule 5. The definition is:

"'Goods' includes goods, wares, merchandise and articles of every kind whatsoever except live animals and cargo which by the contract of carriage in stated as being carried on deck and is so carried."

Mr Debattista submitted that this was an inclusive definition, which clearly included and was always intended to include goods carried in bulk, from which it followed that the limitation in Article IV rule 5 also included bulk cargoes. He submitted that the judge had been wrong to dismiss this submission at [11] of the judgment, on the basis that Article IV rule 5 was directed to a different point and uses different language. In my judgment, the judge was right to conclude that the definition of "goods" could not assist in determining the meaning of the completely different word "unit" in a different provision of the Rules. Mr Debattista's argument essentially assumes what it seeks to prove. Merely because there is a wide definition of "goods", it does not follow that every provision in the Rules applies to every type of goods. Some provisions are clearly inapposite to particular types of goods. To take an obvious example: the exception in Article IV rule 2(n) for "insufficiency of packing" clearly cannot apply to bulk cargoes or other goods which are not packaged.

28. Fourth, given that Mr Debattista accepts that the word "unit" in Article IV rule 5 does not just refer to a unit of measurement for all purposes, but does also refer to unpackaged physical items for shipment, I agree with the judge that this creates an obvious problem, that the word is given different meanings for different types of cargo. As the judge said, that problem would arise with unpackaged items such as cars, where the bill of lading not only specifies the number of "units" in the sense of the number of cars, but their weight. Which of the two is to be taken to be the "unit" for limitation purposes, in the absence of any guidance in the Rules themselves? Furthermore, as Mr Persey QC pointed out, even in the context of bulk cargoes, the appellant's construction would cause problems, where for example more than one unit of measurement is used to describe the cargo, both a volume and a weight or, as in the present case, different units of weight, metric tons in the charterparty and kilograms in the bill of lading.
29. Mr Debattista sought to answer that obvious problem by saying it could be resolved by "sensible construction". For example, if the cargo consisted of 5 cars, each weighing 2,500 kg and the bill of lading stated both the number of cars and their weight, he asserted that obviously, the "unit" would be the car, the physical item, rather than its weight. I do not consider that the issue of what is the "unit" for limitation purposes in article IV rule 5 can be determined by reference to how the parties have chosen to describe the cargo in the bill of lading and it is far from obvious to me that, on the appellant's case that "unit" can refer to both a physical item and its weight, limitation must necessarily be by reference to the former in the example given. Like the judge, I consider that the problem which the appellant's argument creates is a powerful point against the correctness of the argument.
30. The judge noted at [12] of his judgment that at the time the Hague Rules were adopted, the price of bulk cargoes which were being shipped was such that the limitation provisions would not have been seen as relevant, as I consider is borne out by the travaux préparatoires, which I consider below. The judge noted that it was the increase in the price of commodities since that time which led to shipowners now contending that the limitation provisions in Article IV rule 5 ought to apply to bulk cargoes. The judge said that if, as a matter of construction of the Rules "unit" does not mean a unit of measurement, then there is no basis upon which bulk cargoes could be subject to limitation, however desirable that might be and it would be wrong to resort to fictions to achieve that end.
31. Mr Debattista was critical of the judge, contending that this construction froze the Hague Rules in the economics of the 1920s and that this time-bound approach was not that adopted by the Courts, citing The River Gurara [1998] QB 610 where the Court of Appeal applied the Rules to containerised transport even though that was unknown when the Rules were enacted. This criticism is unwarranted and misunderstands what the judge was saying. He was not saying that the construction of the Rules had to be assessed by reference to the 1920s. He was simply making the point that the value of bulk cargoes in the 1920s explained why no-one considered it necessary to insert any limitation provision in Article IV rule 5 applicable to bulk cargo. If, on its true construction, Article IV rule 5 did not apply to bulk cargo, it was not permissible to strain the language of the provision to make it apply to bulk cargo, however desirable that might be. In my judgment, that analysis is correct.
32. In concluding that "unit" means a physical item of cargo or a shipping unit, I have borne well in mind that the Hague Rules had and (where they are still applicable) have an international currency. As Lord Macmillan said in Stag Line v Foscolo, Mango & Co [1932] AC 328 at 350:

"It is important to remember that the Act of 1924 was the outcome of an International Conference and that the rules in the Schedule have an international currency. As these rules must come under the consideration of foreign Courts it is desirable in the interests of uniformity that their interpretation should not be rigidly controlled by domestic precedents of antecedent date, but rather that the language of the rules should be construed on broad principles of general acceptation."

33. If Mr Debattista were able to point to a substantial body of authority in a number of foreign jurisdictions or other material demonstrating that the construction which the appellants seek to place on "unit" in Article IV rule 5 is the generally accepted construction, then obviously that would give me pause for thought. However, he cannot. Rather, as the judge said, the construction of "unit" as a physical item or shipping unit is the one accepted by courts in other common law jurisdictions and favoured by the majority of academic commentators and textbooks. This construction is also clearly confirmed by the travaux préparatoires for the Hague Rules, which I will consider, before turning to the authorities and the academic commentaries.

The travaux préparatoires

34. The decision of the Supreme Court in Gard Marine and Energy Ltd v China National Chartering Co Ltd ("The Ocean Victory") [2017] UKSC 35; [2017] 1 WLR 1793, decided since Sir Jeremy Cooke gave judgment in the present case, has clarified the circumstances in which it is permissible to have recourse to the travaux préparatoires in ascertaining the meaning of words used in an international Convention. Lord Clarke of Stone-cum-Ebony JSC (with whose judgment all their Lordships agreed on the issue to which this point related) endorsed the approach of Longmore LJ in CMA CGM S.A. v Classica Shipping Co Ltd ("The CMA Djakarta") [2004] EWCA Civ 114; [2004] 1 Lloyd's Rep 460. Lord Clarke referred, as Longmore LJ had, to Articles 31 and 32 of the Vienna Convention on the Law of Treaties (1969) ratified by the United Kingdom in 1971 and which came into force in 1980. Those provide:

"Article 31

General rule of interpretation

1.         A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.

2.         The context for the purpose of the interpretation of a treaty shall comprise, in addition to the text, including its preamble and annexes:

(a)       any agreement relating to the treaty which was made between all the parties in connexion with the conclusion of the treaty;

(b)       any instrument which was made by one or more parties in connexion with the conclusion of the treaty and accepted by the other parties as an instrument related to the treaty. …

Article 32

Supplementary means of interpretation

Recourse may be had to supplementary means of interpretation, including the preparatory work of the treaty and the circumstances of its conclusion, in order to confirm the meaning resulting from the application of article 31, or to determine the meaning when the interpretation according to article 31:

(a)       leaves the meaning ambiguous or obscure; or

(b)       leads to a result which is manifestly absurd or unreasonable."

35. Lord Clarke agreed with the approach of Longmore LJ in relation to these Articles at [74] and [75] of his judgment:

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

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

36. Mr Debattista also placed particular reliance on what Lord Steyn said in Effort Shipping Co Ltd v Linden Management SA ("The Giannis NK") [1998] AC 605 at 623E-G:

"Following Fothergill v Monarch Airlines Ltd., [1980] 2 Lloyd's Rep 295; [1981] AC 251, I would be quite prepared, in an appropriate case involving truly feasible alternative interpretations of a convention, to allow the evidence contained in the travaux préparatoires to be determinative of the question of construction. But that is only possible where the Court is satisfied that the travaux préparatoires clearly and indisputably point to a definite legal intention: see Fothergill v Monarch Airlines Ltd., per Lord Wilberforce, at p. 202 col. 1; p.278C. Only a bull's eye counts. Nothing less will do."

37. He submitted that, since it was common ground that "unit" could as a matter of language mean either a physical item or a unit of measurement, it was the respondents who were arguing for a meaning other than that ordinary meaning, and therefore it was for the respondent to persuade this Court that either the word "unit" was ambiguous or obscure or that the ordinary meaning would lead to a manifestly absurd or unreasonable result. Even if the respondents could do so, they would still have to persuade the Court that it could score a bull's-eye on the travaux préparatoires, which they could not do.
38. It will be apparent from my conclusion that, in the context of the Hague Rules, "unit" clearly means a physical item of cargo, a shipping unit, and not either a unit of measurement or both a shipping unit and a unit of measurement (as the appellant contends), that I regard this submission of Mr Debattista as misconceived. In my judgment, the primary purpose for looking at the travaux préparatoires is to confirm the meaning of "unit" in article IV rule 5 which I have concluded is the clear meaning of the word.
39. The travaux préparatoires were examined and analysed in great detail by Allsop J (now the Federal Chief Justice) in the Federal Court of Australia in El Greco v Mediterranean Shipping [2004] 2 Lloyd's Rep 537, particularly at [169]-[193]. Like the judge, I agree with the conclusions he reached. I see no reason to set out in extenso in this judgment all the discussions, at the various conferences in 1921 to 1923, of what became the Hague Rules, before they were finalised and adopted. However, I would emphasise the following points derived from my own consideration of the travaux.
40. As appears from [169] of Allsop J's judgment the original draft of what became Article IV rule 5 contained a weight/volume limitation in these terms:

"neither the carrier nor the ship shall be responsible in any event for loss or damage to or in connection with goods for an amount greater than £ per package or £ per cubic foot or £ per cwt (as declared by the shipper and inserted in the contract of carriage, whichever shall be the least) of the goods carried, unless the nature and value of such goods have been declared by the shipper and inserted in the bill of lading …" (my emphasis)

41. As the judge noted in [25] of the judgment, when this was discussed at the conference of the International Law Association in The Hague on 31 August 1921, M Dor, the French jurist who represented cargo interests, was anxious to remove the weight/volume limitation, on the basis that with a valuable item in a small package, a limitation on the basis of the least of package, weight or volume would enable shipowners to limit liability to a very small sum, in circumstances where it mattered most. During a lively debate, M Dor described this as a "trick", giving the example of a case of silk which does not weigh much and is of limited volume. He wanted to have only a limitation as he said: "[of] £100 per package and per package only, leaving out the cubic foot and the cwt." In response, Sir Norman Hill, President of the Liverpool Shipowners Association and head of the drafting committee, said, in support of the weight limitation: "…if you have a hold full of wheat, is that a package?" to which M Dor's riposte was: "You cannot limit your liability for wheat. There is no limitation of liability for wheat. It is for parcels."
42. In the debate on the same day, there was also some discussion about limitation by reference to freight. M de Rousiers, a French shipowner, suggested limiting to, say, ten times the freight, which would give some relation between the profit of the shipowner and his responsibility. M Dor was not enamoured of limitation by reference to freight either, again giving the example of a case of silk which might be worth 1,000 or 3,000 francs, but the freight from Marseille to Algiers was only 37 francs. He said: "the freight system of M de Rousiers may be all right if you multiply by a very high figure. If you say twenty times the freight, that is all right; but if you say once, twice or three times the freight, that is absolutely no good."
43. Overnight, Sir Norman Hill drafted a new proposed limitation provision which limited liability to a sum equal to so many times the freight or the value of the goods, whichever was the least. At the outset of the session on 1 September 1921, the Chairman, Sir Henry Duke (the President of the Probate, Divorce and Admiralty Division) introduced that proposal and stated that M Dor had given notice of a proposal that limitation should be £100 sterling, gold per package, with omission of any alternative calculation by cubic foot or cwt.
44. During the discussion that day, M Dor indicated that limitation by a multiplier of freight, because it was proposed by shipowners, would be viewed with suspicion by the Chambers of Commerce in France. Only a clear limitation of £100 per package would be acceptable to them. He returned slightly later to the question of cubic feet or hundredweight, which he referred to as a trap. In relation to bags of wheat, where Sir Norman Hill had asked how a shipowner was going to limit unless it was by weight, M Dor said: "for bags of wheat you do not need to limit liability at all; where you need to limit it is for packages of extraordinary value", his point being that a bag of wheat was worth less than £100 at that time. That is also an indication that, where wheat was shipped in bulk (as it sometimes was at that time) a £100 limit would not be relevant, as its value was much less, whether measured in bushels, kilograms, long or metric tons.
45. At that session on 1 September 1921, the shipowners remained concerned about the adequacy of a limitation only by reference to "package". Mr L.C. Harris, another British shipowner, gave the example from his own experience of a boiler which was being loaded into a hold and dropped from above the hatch damaging the ship, although it might have gone through the ship's plating. He said: "Now you see, we might have lost our ship but if we damaged that boiler we should pay £100 for it. It is not satisfactory in either direction." The concern there was evidently as to limitation for goods which could not be described as a package. M Dor had also referred earlier to a court case in France of a car carried on deck.
46. As the judge said at [26] of his judgment, by the end of the session on 1 September 1921, there was a general consensus on the wording: "neither the carrier nor the ship shall be responsible in any event for loss or damage to or in connection with goods to an amount beyond £100 per package unless the nature or value of such goods has been declared …".
47. After that session closed, the drafting committee (chaired by Sir Norman Hill and including M Dor) agreed unanimously upon the language of the Rules with which it was satisfied. At the beginning of the next day's session of the full committee on 2 September 1921, Sir Henry Duke read out the draft wording of what were described as "the Hague Rules 1921". When he got to what is now Article IV rule 5 (then Article IV rule 4) he said:

"Now there is a slight alteration made to which I call your attention –"£100 per package or unit" - As you know, there are goods to which the Code will apply which are not described as per package, and the matter was raised yesterday, and upon consideration the Committee thought that by adding the words "or unit" the intent would be made clear".

48. As Allsop J said in El Greco at [177], and the judge agreed in the present case at [27] and [28], the addition of "unit" was to cover items shipped as single units and not packaged in any way, such as the car and boiler discussed the previous day, which, as Sir Henry Duke said: "are not described as per package". They were obviously not intended to reintroduce the weight or volume limitation, which had been abandoned by that stage. As the judge said at [28]:

"It has, I think, never been suggested that the intention behind the drafting committee's addition of the words "or unit" was somehow to introduce (or reintroduce) limits of liability based on weight or volume and it was accepted by Mr Debattista for the Defendant that the objective in mind was to cover unpackaged items for shipment. It is clear that although the concepts of weight and volume were discussed initially, they were abandoned and were never resurrected, whether by the introduction of the word unit or otherwise."

49. Furthermore, there was no suggestion at that time that "unit" had been introduced to cater for bulk cargoes. In so far as bulk cargoes were discussed at all at The Hague, it appears to have been assumed that they would not be the subject of limitation. After Sir Henry Duke had read out the draft Rules they were unanimously adopted by the committee. There was no further discussion of the word "unit".
50. The issue of limitation was discussed again at the Diplomatic Conference in Brussels in October 1922. In the sub-committee chaired by a U.S. admiralty judge, Judge Hough, Mr Bagge the Swedish delegate, recalled that at The Hague, Sir Norman Hill had stated that the clause [i.e. what is now Article IV rule 5] should not apply to bulk cargoes and Judge Hough conceded that a bulk cargo was not a package. Although there was no separate discussion of the meaning of "unit", the discussion generally proceeded on the basis that the clause did not apply to bulk cargo, no doubt because, as was also pointed out again, limitation was unnecessary in the case of bulk cargoes because of their low value. At a later point in the discussion, Judge Hough referred to the proviso [to Article IV rule 5] concerning declaration of value as only applying "when the declared value is greater than £100 per piece or item", a further clear indication that the provision is concerned with physical items of cargo and also support for the conclusion that "piece" and "unit" were considered as synonymous.
51. Finally, at the meeting of the sub-committee on 7 October 1923 in Brussels, there was the following exchange between Mr Bagge and Sir Leslie Scott (former Solicitor-General and subsequently Scott LJ):

"Mr Bagge declared that it seemed clear to him that article 4(5) did not apply to bulk cargoes.

Sir Leslie Scott said that it was clear that in that case it was not a question of package or unit."

52. In my judgment, the travaux préparatoires demonstrate clearly a number of matters which confirm that the meaning of "unit" is, as I have found, a physical item of cargo or shipping unit, not a unit of measurement: (i) any limitation by reference to weight or volume was abandoned by the end of the session on 31 August 1921, as was any limitation by reference to a multiplier of freight by the end of the session on 1 September 1921; (ii) the word "unit" was clearly introduced into Article IV rule 5 not to reintroduce limitation by reference to weight, volume or freight but to cater for items of cargo which are carried without packaging, such as cars or boilers; (iii) the "package or unit" limitation in Article IV rule 5 was never intended to apply to bulk cargoes. I agree with the judge that the position is accurately summarised by Allsop J in El Greco at [278]:

"The terms of art. IV, r. 5 of the Hague Rules were negotiated and agreed upon as a package limitation […] The addition of the words "or unit" can be seen to have been intended to clarify the rule by making unnecessary any debate in individual cases about the extent and nature of wrapping and the like, so that individual articles capable of being carried without packaging - boilers, cars and the like, and which could be seen as units of cargo as shipped - would be covered. This approach involves a rejection of the notion that "or unit" was inserted to cover bulk cargo by reference to freight unit, as in U.S. COGSA. The weight of judicial and other views that I have earlier referred to makes this a safe conclusion …"

53. Accordingly, the travaux préparatoires do confirm the clear meaning which I consider is to be attributed to "unit" in Article IV rule 5. Even if Mr Debattista were right that it is necessary to identify a "bull's eye", I consider that the three matters I identified in the previous paragraph as emerging clearly from the travaux préparatoires, do amount to "bull's eyes" in the sense meant by Lord Steyn or, in the more measured language of Lord Wilberforce in Fothergill v Monarch Airlines [1981] AC 251 at 278, the travaux préparatoires "clearly and indisputably point to a definite legislative intention". Finally in relation to the travaux préparatoires, I should mention the attempt by Mr Debattista to dilute their impact by submitting that they were in the nature of minutes of meetings, so that the Court should be cautious lest they are inaccurate. I was unimpressed with this submission. Even if there were inaccuracies (for which there is no evidence) the three matters I have mentioned emerge clearly from the travaux préparatoires.
54. In his written submissions, Mr Debattista criticised the judge for having failed to deal with the important point that, in the debate in the House of Lords on the Bill which became the Carriage of Goods by Sea Act 1924 and enacted the Hague Rules, one of the problems identified by the opponents of the Rules was that they applied to bulk cargoes. This point was not developed orally, which is perhaps unsurprising, since it is without merit. Having read the passages from Hansard relied upon, whilst they contain an eloquent defence of the freedom of contract of British shipowners by Lord Sumner and Lord Phillimore in opposing the Rules, I do not consider that there is anything which is relevant or admissible in construing the Hague Rules.

The U.S. Carriage of Goods by Sea Act 1936 and the Hague Visby Rules

55. Before turning to consider the various authorities which have considered the meaning of "unit", I should deal briefly with two other matters which, like the judge, I consider cannot affect the meaning of "unit" in article IV rule 5. First, the equivalent provision in the United States Carriage of Goods by Sea Act 1936 ("USCOGSA"). That Act followed the wording of the Hague Rules closely but is not identical. One of the respects in which it differs is in the limitation provision in section 4(5):

"Neither the carrier nor the ship shall in any event be or become liable for any loss or damage to or in connection with the transportation of goods in an amount exceeding $500 per package lawful money of the United States, or in case of goods not shipped in packages, per customary freight unit, or the equivalent of that sum in other currency, unless the nature and value of such goods have been declared by the shipper before shipment and inserted in the bill of lading…" (my emphasis)

56. Mr Debattista placed considerable reliance on the fact that the United States Department of Trade, in a memorandum dated 5 June 1937 described the differences from the Hague Rules in USCOGSA as "intended primarily to clarify provisions in the Convention which may be of uncertain meaning thereby avoiding expensive litigation in the United States for purposes of interpretation" (see Tetley: Marine Cargo Claims 4^th edition (2008) at p 2175). However, that view of the Department of Trade cannot possibly lead to a construction of the original Article IV rule 5 which interprets "unit" as a freight unit, let alone a customary freight unit. In The Bill 55 F. Supp. 780 (1944) in the District Court of Maryland, District Judge Chesnut referred to "per customary freight unit" as having "expanded or changed" the phrase "per unit" and said that: "it seems reasonably clear that the phraseology finally adopted was intended to be more definite than the shorter phrase 'per unit'…" In Falconbridge Nickel Mines v Chimo Shipping [1973] 2 Lloyd's Rep 469, a decision of the Supreme Court of Canada (to which I will return later in this judgment) Ritchie J, having cited what was said by Judge Chesnut in The Bill, said at 476 col. 1:

"It is thus plain to me that it was only after considerable debate that the United States adopted the present form of their statute and I am satisfied that the words 'per package or, in case of goods not shipped in packages, per customary freight unit' do constitute a change from the Hague Rules as adopted in Great Britain and in Canada, and I do not think that they afforded any substantial guidance in the solution of the problem as to the meaning of the phrase 'per package or unit' as it occurs in art. IV r. 5".

57. Like the judge, I agree with that analysis and have concluded that USCOGSA is of no assistance as to the meaning of "unit" in Article IV rule 5. In the circumstances, it is not necessary to consider further the meaning of "customary freight unit" or the discussion in Cooke on Voyage Charters to which the judge refers in [32] of his judgment.
58. Before the judge, Mr Debattista also argued that support for his construction of "package or unit" as including bulk cargoes was to be derived from the terms of Article IV rule 5(a) of the Hague-Visby Rules which contains a weight limitation multiplier introduced in the Visby Protocol of 1968:

"Unless the nature and value of such goods have been declared by the shipper before shipment and inserted in the bill of lading, neither the carrier nor the ship shall in any event be or become liable for any loss or damage to or in connection with the goods in an amount exceeding 666.67 units of account per package or unit or 2 units of account per kilogramme of gross weight of the goods lost or damaged, whichever is the higher" (my emphasis).

59. Mr Debattista argued that if the weight limitation had been introduced in order to include bulk goods within the limitation regime where they had previously been excluded under the Hague Rules, the underlined words would have stopped at "lost or damaged". This argument by reference to the Hague-Visby Rules limitation provision was not really pursued on appeal, but to the extent that it remains live, it is misconceived for the reasons the judge gave at [19] and [20] of the judgment, that: (i) the terms of the Hague-Visby Rules cannot conceivably affect the construction of the Hague Rules adopted 45 years earlier and (ii) in any event, the wording of Article IV rule 5(c) strongly suggests that the draftsmen of the Hague-Visby Rules considered that a "unit" constituted a physical item of cargo rather than a freight unit.

The English authorities

60. There was no English authority directly on the point in the context of the Hague Rules prior to the decision of the judge in the present case. However, such dicta as there were proceeded on the basis that a "unit" in Article IV rule 5 meant a physical item of cargo rather than a unit of measurement. Thus, Studebaker Distributors Ltd v Charlton Steam Shipping Co Ltd [1938] 1 KB 459 concerned a bill of lading clause which limited liability to $250 for "each package" where unboxed cars had been carried from the USA to London. The judge, Goddard J, considered the word "package" there in contradistinction to "package or unit" in the Hague Rules in terms from which it was clear that he considered "package or unit" referred to an individual piece of cargo, not a unit of measurement. At 467 he said:

"… The goods are expressly stated to be unboxed, and the case was argued before me by both parties, who doubtless want a decision on what are known to be the actual facts, on the footing that the cars were put on board without any covering, or, to state it in another way, just as they came from the works. I confess I do not see how I can hold that there is any package to which the clause can refer. "Package" must indicate something packed. It is obvious that this clause cannot refer to all cargoes that may be shipped under the bill of lading; for instance, on a shipment of grain it could apply to grain shipped in sacks, but could not, in my opinion, possibly apply to a shipment in bulk. If the shipowners desire that it should refer to any individual piece of cargo, it would not be difficult to use appropriate words, as, for instance, "package or unit," to use the language of the Hague Rules …"

61. In The Aramis [1987] 2 Lloyd's Rep 58, a case concerning damage to a bulk cargo of linseed expellers, one of the issues which arose was the extent to which the defendant shipowners could limit their liability under article IV rule 5. At 67 col. 1. Evans J referred to the "long-standing debate as to the proper meaning of 'unit'. He continued:

"The view put forward by the defendants, which receives some support from Scrutton on Charterparties (18th ed.) pp. 441-443, is that for a bulk cargo the "freight unit" or "customary freight unit" should be adopted. This contention only avails the defendants if that unit can be identified in the present case as one tonne. The bills of lading do not do so. The weight is expressed in kilos, both in print and in type. There is no evidence that the customary freight unit is one metric tonne. I reject this contention, therefore, as being unsupported by, or contrary to, the evidence before me."

62. Mr Debattista developed a somewhat convoluted argument that Evans J was expressing the opinion that the Scrutton analysis that "unit" was a freight unit or customary freight unit was correct or to be preferred, albeit that, in that case, there was no evidence that the customary freight unit was a metric ton and the bills of lading stated the quantity of cargo in kilograms, so that any limit by reference to kilograms would be in excess of £100 per unit, meaning the shipowner could not limit liability. In my judgment, that is a complete misreading of what Evans J was saying in this passage. He was noting the long-standing debate and recording on which side of it the editors of Scrutton stood. However, the sentence beginning: "This contention only avails the defendants…" is saying that, even if the defendants were right that the Scrutton analysis was correct, they would lose on the facts. The judge was careful not to say that the Scrutton analysis (to which I will return below, but which amongst textbooks and commentators was clearly a minority view) was correct, because he did not have to, given his conclusion on the facts. I have little doubt that if he had been intending to express an opinion, albeit obiter, that the Scrutton analysis was correct, he would have said so expressly, even if he had then gone on to decide that the defendant failed on the facts.
63. In Bekol BV v Terracina Shipping Corporation et al., The Jamie (unreported 13 July 1988) timber shipped in nine bundles created by fastening individual timber pieces together had been wrongfully shipped on deck and was damaged. Leggatt J decided that the relevant "packages or units" for the purposes of Article IV rule 5 were the bundles. He said that each individual length of timber:

"measuring typically two or three inches by four or five inches in cross-section and many feet in length, viewed by itself is a single item and therefore capable when considered in isolation of being called a unit. If pieces of this kind were carried loose, each of them might be said to constitute a unit; but when, as here, a number of pieces are fastened together with steel straps they become a composite shipping unit."

64. At [42], the judge said of this case and another unreported decision of the Commercial Court, that of Hobhouse J in The Troll Maple (1990), that neither case gave "much assistance save that each proceeded on the assumption that a "unit" was a physical item of cargo rather than a unit of measurement". However, I agree with the assessment of Andrew Baker J in Kyokuyo Co Lt v AP Møller-Maersk A/S ("The Maersk Tangier") [2017] EWHC 654 (Comm); [2017] 1 Lloyd's Rep 580 at [83] that The Jamie: "confirms that under English law, when considering 'units' under the Hague Rules, the search is for the identifiably separate items of cargo, as in fact shipped." To that extent it is of some assistance in relation to the principal issue on the appeal.
65. Mr Debattista placed some emphasis in his submissions on the decision of the Court of Appeal in River Gurara v Nigerian National Shipping Line Ltd [1998] QB 610. In that case, the ship, carrying a cargo of containers, stranded and was a total loss. The bills of lading described the cargo said by the shippers to be within the containers as constituting a specified number of "bales" or "parcels" or the like. The issue was whether the "packages" on which the limit in Article IV rule 5 was to be calculated were the containers or the individual items within them.
66. Like Colman J at first instance, the Court of Appeal was not attracted by the argument that the container was the appropriate package for limitation purposes. At 617F-618A, Phillips LJ said:

"Mr. Russell [counsel for the cargo claimants] submitted that, when the Convention was concluded in 1924, a figure of £100 represented a fair figure for the average value of a package shipped. To apply the same figure to a huge container stuffed with many packages would defeat the object of preventing shipowners from limiting their liability to sums that were absurdly low by reference to the average values of cargoes shipped. I consider that there is force in this submission. If Mr. Kay is correct, the change in the method of stowing and carrying cargo that occurred when containerisation was introduced effected a radical change in the limitation regime. I would not readily reach such a conclusion.

Mr. Russell further submitted that to describe a container as a package was to strain the natural meaning of that word. With this also I agree. In Bekol B.V. v. Terracina Shipping Corporation (unreported), 13 July 1988, which seems to be the only recorded case in which the English court has considered the meaning of "package" in the Hague Rules, Leggatt J. referred to the definition of that word in the Oxford English Dictionary: "a bundle of things packed up, whether in a box or other receptacle, or merely compactly tied up." A huge metal container stuffed with goods which will normally themselves be made up in individual packages is not naturally described as a package

These two considerations alone would lead me, in the absence of authority, to conclude that, where the Hague Rules limit falls to be computed in relation to parcels of cargo which are loaded in containers, it is the parcels, and not the containers, which constitute the relevant packages."

67. Phillips LJ was reinforced in that conclusion by decisions in other jurisdictions, citing Friendly CJ in the United States Court of Appeals Second Circuit in The Mormaclynx [1971] 2 Lloyd's Rep 476 at 486:

"Still we cannot escape the belief that the purpose of section 4(5) of C.O.G.S.A. was to set a reasonable figure below which the carrier should not be permitted to limit his liability and that 'package' is thus more sensibly related to the unit in which the shipper packed the goods and described them than to a large metal object, functionally a part of the ship, in which the carrier caused them to be 'contained.'"

Phillips LJ noted that this preference for the packages rather than the containers in which they were stuffed, at least where the bill of lading states the number of packages, had been shown in other jurisdictions as well: Canada, Australia, France, Holland, Italy and Sweden.

68. He went on to decide, declining to follow U.S. authorities, that the description of the cargo in the bill of lading (for example, so many containers) was not an agreement which was determinative of the limit of liability under Article IV rule 5. What determined that limit of liability was the number of packages proved to have been loaded in the containers rather than the number of containers. Hirst LJ dissented on this point, although he agreed with the overall result of the appeal.
69. Mr Debattista suggested that Phillips LJ had applied commercial creativity and invited us to do the same. In my judgment, there was no particular commercial creativity in that case. The Court of Appeal reached a conclusion as to the meaning of "package" in Article IV rule 5 which had also been reached in a number of other jurisdictions. That case is of no assistance as to the meaning of "unit" in the provision, which is the issue for this Court.
70. Since the decision of Sir Jeremy Cooke in the present case, that issue has also arisen before Andrew Baker J in The Maersk Tangier, to which I have already referred. That case concerned a number of preliminary issues in relation to damage to three container loads of frozen tuna loins and bags of frozen tuna shipped on the defendant's vessel. One of the issues was whether, if liability was limited pursuant to Article IV rule 5 of the Hague Rules, the relevant packages or units were the containers or the individual pieces of tuna. Since the first issue the judge had to decide was whether the Hague Rules or the Hague-Visby Rules applied to the relevant contracts of carriage and he decided the Hague-Visby Rules applied with the force of law, his decision in relation to package or unit under Article IV rule 5 was strictly obiter.
71. The judge first decided, at [71]-[72] of his judgment, that each of the bags of tuna was a "package" and then went on to consider whether each of the frozen loins was a "unit". Having cited and discussed many of the same authorities and textbooks as the judge did in the present case (including approval of the analysis of the travaux préparatoires by Allsop J in El Greco) together with The River Gurara, Andrew Baker J concluded that "units" are physical items of cargo. At [86] he said:

"Drawing the threads together, in my judgment, for the Hague Rules under English law:

(i) The possible reading of Article IV rule 5 for containerised cargo that it is, by definition, packaged cargo, the containers being the only relevant packages, was authoritatively rejected by The River Gurara.

(ii) In providing for a limit of liability 'per package or unit', the sense of 'or' is 'whichever (if either) be relevant to the cargo in question', and the focus is upon the cargo as in fact transported. Any given item of cargo cannot be both packaged and unitised cargo, although the entire cargo can be neither (e.g. bulk cargoes are neither: see The Aqasia). The cargo can of course be a mix, so (e.g.) there is no difficulty in principle here over the possibility that the frozen loins might be 'units' but the bagged tuna was packaged cargo, each bag being a 'package'.

(iii) It follows that if cargo as in fact transported is packaged, the limit of liability for that cargo applies per package, even if what has been packaged would have been suitable for transportation without that packaging: see The Jamie. The further question, for packaged goods, of packages within packages, dealt with in The River Gurara and discussed in El Greco, does not arise in the present case and I need make no particular decision about it.

(iv) If cargo as in fact transported is not packaged, but is made up of identifiably separate items of transportable cargo, those items are 'units'. For break-bulk shipments, the identification of any 'units' will be by reference to the cargo as in fact shipped. For containers, what is in fact shipped (in the strict sense) is the containers; but following The River Gurara (see i) above) the container walls are transparent under the gaze of Article IV rule 5; or to put it another way, irrespective of any allocation of responsibility for the stuffing of the container or of where (if at all prior to shipment in the strict sense) any responsibility for the care and carriage of the container attaches to the carrier, from the perspective of the cargo and how it is made up (if at all) for transportation, the journey begins at the door of the container not at the ship's rail. That is of the essence of the efficiency of modern container transport, to the mutual benefit of cargo interests and carriers.

(v) There is no reason of language or purpose why 'units' should not be identified, for a container load, by reference to the characteristics of the cargo as it was stuffed into the container. To the contrary, in the light of iv) above in particular, that is the natural way of assessing any question of the characteristics of a containerised cargo, if (always) it is not for the relevant purpose to be determinatively characterised by the container itself. There is no source in the language or purpose of Article IV rule 5 for a special, added, rule calling for a focus not upon the cargo as shipped, but upon how (if at all) the cargo could have been shipped if not containerised.

(vi) In this case, looking through the notionally transparent walls of the three Containers to examine the cargo as shipped (or, if this be the preferred way of looking at it, watching the Containers being stuffed to see what the cargo was, as stuffed), one sees: individual frozen tuna loins, transportable and shipped/stuffed 'as is'; bags. The natural, and correct, conclusion if asked whether, and if so how, the cargo as shipped comprised 'packages or units', is that the cargo was a mixed cargo of 'packages' (the bags, each bag one package) and 'units' (the unpackaged tuna loins, each loin being one unit since each was identifiable as a separate article for transportation as such, within the container)."

72. Although other issues decided by Andrew Baker J (not relevant to the present issue) are the subject of a forthcoming appeal to the Court of Appeal later this year, Andrew Baker J's analysis is not challenged in sofar as he agreed with Sir Jeremy Cooke that a "unit" in Article IV rule 5 of the Hague Rules is a physical item.

The Commonwealth authorities and other cases

73. Although at the time of the judgment of Sir Jeremy Cooke there was no English authority directly on the point, there are decisions in Commonwealth jurisdictions which are directly on the point and which reach the conclusion that "unit" means a physical item of cargo, not a unit of measurement or a freight unit.
74. Perhaps the clearest exposition to that effect is that of the Supreme Court of Canada in Falconbridge to which I have already made reference. Ritchie J, giving the judgment of the Court decided in terms at 475 that "unit" in Article IV rule 5 means a shipping unit, of physical goods:

"The meaning of the word "unit" as it occurs in the phrase 'package or unit' in Rule 5 has given me very great difficulty but I am now satisfied that no substantial assistance can be obtained from the U.S. cases because of the clear difference in the wording of the Rules and such authorities as exist in this country and in England appear to me to bear out the statement of Mr Justice Rand that the word in this context means a shipping unit, that is a unit of goods."

75. Later in the judgment, after he had referred to USCOGSA and concluded that it did not give any guidance (see the passage which I cited at [56] above), Ritchie J referred at 476 col. 1 to the analysis in Temperley & Vaughan: The Carriage of Goods by Sea Act 1924 4^th edition (1932) at pp 81-82:

"The word unit connotes one of a number of things rather than a thing standing by itself, and with reference to goods carried by ship, it does not seem appropriate to describe the whole of a cargo or parcel of cargo in bulk. Further, the natural interpretation of the word "unit" in the phrase "package or unit" appears to be that it has been added in order to cover parts of a cargo similar in a general way to a package, but not strictly included in that term which properly implies something packed up or made up for portability and would therefore not include such a thing as a log of wood or a bar of metal. The word "unit" has, it is suggested, been added in order to embrace such things and not to extend the scope of the Rule to bulk cargoes or parts thereof. Moreover, the whole purpose of Rule 5, which is directed against excessive claims for things of undisclosed abnormal value, supports this limited interpretation of the word."

76. Ritchie J said: "The learned authors of this work then refer to 'an alternative view for which there is much to be said' and which they describe as follows":

"… inasmuch as the term "unit" is commonly used to mean a standard of measure or enumeration, or one of a series of things split up either physically or notionally for the purpose of enumeration or measurement, the phrase "package or unit" here used must refer back to the particulars of enumeration or measurement which must be shown on the bill of lading as provided by Article III Rule 3 …"

77. As Ritchie J then commented:

"It is clear, however, that the authors prefer the former view. This interpretation is further borne out by the note to be found in Halsbury's Laws of England, 3rd ed., vol. 35 at p. 535 where the learned editors observed in a cryptic note speaking of the word "unit" as used in the Rule: "… which latter term is no doubt apt to indicate an unboxed vehicle."

78. In New Zealand, a judge at first instance, Tompkins J in New Zealand Railways v Progressive Engineering Company Ltd [1968] NZLR 1053 had to consider the meaning of "package or unit" in a similar limitation provision to Article IV rule 5 in the Government Railways Act 1949. Having referred to Goddard J's judgment in Studebaker, he concluded: "…a package imports the notion of articles packed together as these were. A unit on the other hand, imports something which is a separate thing, such as a single manufactured article, though of course any single article, if accepted for transport as a separate article, would be a unit."
79. I have already referred to the judgment of Allsop J (with whom Black CJ agreed) in the Federal Court of Australia in El Greco. Having considered the travaux préparatoires he analysed many of the authorities to which I have referred, including Studebaker and Falconbridge (evidently the "weight of judicial…views that I have earlier referred to" in [278]). He reached the conclusion in that paragraph that "unit" covered items of cargo not packaged, and did not cover bulk cargo by reference to freight unit.
80. Before this Court (but not before the judge) Mr Debattista relied on decisions of the French courts and legal commentary from which it appears that in France the courts have taken the view that "unit" can connote a unit of measurement applicable to bulk cargo. I agree with Mr Persey QC that the decision of the Cour de Cassation in 1947 on which the appellant relies is devoid of any reasoning and seems to have been subject to academic criticism.
81. Furthermore, as Mr Persey QC submitted, all this demonstrates is that "unit" has been interpreted differently in some contracting states, as is clear from the travaux préparatoires for the Hague-Visby Rules which the judge cited at [33] of his judgment. None of this material detracts from the fact that the preponderance of English and Commonwealth decisions and obiter dicta are to the effect that "unit" means a physical item of cargo or shipping unit, not a freight unit or unit of measurement, so that Article IV rule 5 does not apply to bulk cargo. Mr Debattista sought to downplay the impact of these decisions and dicta by pointing out that none of the cases (with the exception of The Aramis where as I have held, Evans J was careful not to express a view) concerned a bulk cargo, so that the judge had been wrong to say they were "directly on point". I do not consider that criticism to be justified. Whilst it may be correct that the facts of the cases do not concern bulk cargoes, the nature of the cargo carried cannot diminish the strength of the opinion as to the meaning of "unit" in Article IV rule 5 which emerges from these authorities.

The textbooks and commentaries

82. The judge dealt comprehensively at [50] to [58] with the various textbooks and academic commentaries. I do not propose to reiterate everything he said, but to summarise what seem to me to be the salient points. First, the authors of Carver on Bills of Lading 3^rd edition (which was the current edition at the time of the judgment), Sir Guenter Treitel QC and Professor Francis Reynolds QC, stated that the view that "unit" meant a freight unit had never been accepted in England and specifically rejected in Canada. Their view was that "unit" is "an identifiable article or piece of goods that cannot be called a package (usually because it is not packed)". The 4^th edition which was published after the judgment, retains that passage and observes that the judgment in the present case: "confirmed that there is no limit under the Hague Rules in respect of bulk cargo". As the judge noted at [52], Professor Reynolds had expressed the view that Article IV rule 5 does not apply to bulk cargoes in two articles published in 1990 and 2005 (the latter referring to the judgment of Allsop J in El Greco).
83. As the judge said at [53], the editors of Aikens on Bills of Lading (2^nd edition 2015) say that the prevalent view is that "unit" means shipping unit and that the better view is that there is no limit under the Hague Rules for bulk cargo. The same view was reached by Griggs, Limitation of Liability for Maritime Claims (4^th edition 2005). Professor Francesco Berlingieri in International Maritime Conventions Vol. 1: The Carriage of Gods and Passengers by Sea (1^st edition 2014) concludes that a "unit" is "a physical unit that cannot be described as a package."
84. In my judgment, the analysis in the leading Canadian textbook Tetley: Marine Cargo Claims 4^th edition (2008) does not support Mr Debattista's construction. Having set out the rival arguments for "unit" in the Hague Rules meaning a freight unit on the one hand and unpacked objects on the other, the author says at 2178-9:

"(c) The word 'unit' in the English and Canadian case law has come to mean shipping units - generally large, unboxed and unpacked objects, such as cars, generators and tractors – rather than freight units as in the United States [citing Studebaker, Falconbridge and other Canadian cases and Professor Reynolds' 2005 article]. The English interpretation of 'unit' in the Hague Rules has also been adopted by the Federal Court of Australia, albeit in a case governed by the Australian enactment of the Hague-Visby Rules [citing El Greco].

(d) This understanding of 'unit' is more consistent with the approach taken under the Hague-Visby Rules with respect to the word 'unit' in art. 4(5)(c) [again citing El Greco]."

85. The editors of Cooke on Voyage Charters (4^th edition 2014) express the view that the juxtaposition of "unit" and "package" tends to show that the rule was aimed at a physical unit, with the distinction being between units which are packed and those which are not. They refer to the fact that this is the construction adopted in Canada (Falconbridge) and Australia (El Greco) and appears to have been the view of Goddard J in Studebaker, although they note that this would provide no effective limit of liability in cases of bulk cargo.
86. As the judge says at [58], the current edition of Scrutton on Charterparties (23^rd edition 2015) is largely unhelpful, simply referring to the fact that the alternative limitation based on weight in Article IV Rule 5(a) of the Hague-Visby Rules has removed the controversy which existed under the Hague Rules as to the application of "unit" to bulk cargo. There is then a footnote reference to the passage in the 18^th edition (1974) at pp 442-443 referred to by Evans J in The Aramis. That stated:

"What is a unit? The alternatives are (a) the 'freight unit', i.e. the unit of measurement applied to calculate the freight, or (b) the 'shipping unit', i.e. the physical unit as received by the carrier from the shipper. The 'freight unit' has been authoritatively rejected in Canada in favour of the 'shipping unit' at least so far as concerns individual articles such as automobiles not shipped in packages, and there is some authority in England for adopting this approach. But if the 'shipping unit' solution is adopted, it is not easy to see why the Rule treats 'package' as an alternative to 'unit', since 'shipping unit' would include a package. Furthermore the concept of the 'shipping unit', unlike the 'freight unit', is not at all appropriate when applied to bulk cargo: a possible solution is to apply the 'shipping unit' to individual articles not in packages and the 'freight unit' to bulk cargo."

87. As Evans J said, that passage supports the view that for a bulk cargo the "freight unit" or "customary freight unit" should be adopted. That passage in the 18^th edition did not appear in any of the earlier editions of Scrutton published after the Hague Rules were enacted, nor did it appear in any subsequent edition, presumably because, by the time the next edition was published in 1984, the Hague-Visby Rules had been enacted. All editions from the 12^th (1925) to the 17^th (1964) simply contained a footnote against the word "unit" that it: "probably means the unit of enumeration or measure shown in the bill of lading as provided by Article III rule 3" i.e. the number [in the case of packages or pieces] or quantity or weight. In Falconbridge Ritchie J considered that footnote, at 476 col. 2 and said: "…it is my view that the tractor here in question is a piece of cargo within the meaning of that rule". This seems to me to confirm the view I expressed at [26] above that "piece" and "unit" are synonymous.
88. The views expressed in the passage in the 18^th edition reflect, to an extent, the views expressed by Mr Michael Mustill QC in an article written in 1971, on which Mr Debattista also relied (to which the judge referred at [50] of the judgment), which may not be surprising since he was one of the editors of Scrutton at the time. As the judge said, both the article by Mr Mustill QC and the passage in the 18^th edition of Scrutton were written at a time before the decision of the House of Lords in Fothergill v Monarch Airlines [1981] AC 251, when it was considered impermissible to have regard to the travaux préparatoires. Now that it is permissible to have regard to the travaux préparatoires, for the reasons I have given, they demonstrate that limitation by reference to weight or multipliers of freight was abandoned, it was recognised that it was not possible to limit liability for bulk cargo and "unit" was clearly introduced to cover unpackaged individual items of cargo such as cars and boilers. In those circumstances, the views expressed by Mr Mustill QC and by that edition of Scrutton are misconceived, as are those in the earlier editions to the extent that it is suggested that one meaning of "unit" is a unit of measurement. In any event, the views in Scrutton are very much in a minority, the preponderance of the textbooks and commentaries favouring the construction of "unit" as a shipping unit, a physical item of cargo, not a unit of measurement or a freight unit.
89. In the First Supplement to the 23^rd and current edition of Scrutton, the editors have added a new footnote in Appendix III, which sets out the Hague Rules, against the word "unit" in Article IV rule 5:

"As pointed out in The Aramis [1987] 2 Lloyd's Rep 59, the 18^th edition of this work (at pp 441-443) offered some support for the suggestion that in cases concerning bulk cargoes "unit" should be taken to mean "freight unit" or "customary freight unit". In Vinnlustodin HF v Sea Tank Shipping AS [2016] EWHC 2514 (Comm) Sir Jeremy Cooke disagreed and held that the word "unit" was not apt to apply to bulk cargoes, meaning that the package limitation provision of Art. IV of the Hague Rules did not apply to bulk cargoes."

90. In my judgment, Mr Debattista derives no assistance from Scrutton either in the 18^th edition or the current edition. In any event, I agree with Mr Persey QC that, although the textbooks and commentaries recognise the debate as to whether Article IV rule 5 applies to bulk cargo, the broad consensus (with the exception of Scrutton) is that it does not.

Alleged market expectation

91. In his submissions, Mr Debattista asserted that there was a long standing expectation in the market that shipowners can limit in relation to bulk cargoes by incorporation of clause 26 of the London Form or a similar clause to be found in other forms of bulk charterparties, to which he drew our attention. This had been the expectation for 90 years since the Hague Rules were enacted and the conclusion that Article IV rule 5 did not apply to bulk cargoes would surprise the market and confound those expectations. I agree with Mr Persey QC that this submission is misconceived. There is no evidence whatsoever of market expectation or practice and even if there were, it is difficult to see how that could affect the meaning of Article IV rule 5, given that whatever shipowners may or may not have thought it meant, the debate as to the meaning of "unit" is, as Evans LJ said, a long-standing one.

Conclusion on the meaning of "unit"

92. My conclusion that the clear meaning of "unit" is a physical item of cargo or shipping unit and not a unit of measurement or a freight unit and, hence, Article IV rule 5 does not apply to bulk cargo, is confirmed by the travaux préparatoires, by the preponderance of the authorities and textbook and academic commentaries. Like the judge I have no hesitation in rejecting the appellant's construction of Article IV rule 5.

The appellant's alternative case

93. The appellant's alternative case is that, even if Article IV rule 5 would not ordinarily apply to bulk cargoes (as I have found), the provision should be construed as applying to bulk cargoes in this case because of clause 26 of the London Form. Mr Debattista contends that these parties in this charterparty clearly intended that the appellant's liability would be limited in respect of this bulk cargo. He complains that the judgment fails to give effect to the bargain struck by the parties. It is fair to say that, although at the permission to appeal stage this was the appellant's primary case (as appears from the Grounds of Appeal), it has assumed much less prominence at the appeal hearing, perhaps because of the discouragement from Gross LJ when granting permission to appeal.
94. The effect of the submission is that all of Article IV had to apply to the cargo carried under the charterparty in order to give effect to the parties agreement in clause 26 that the appellant would: "be entitled to the like privileges and rights and immunities as are contained in Sections 2 and 5 of the Carriage of Goods by Sea Act 1924 and in Article IV of the Schedule thereto …" As the judge said, this provision does not mean that the words of Article IV are written into the charterparty with the consequence that every provision in the Article must be given meaning and effect in the context of the carriage of the bulk cargo contemplated by the charterparty. The reference to: "the like privileges and rights and immunities" means that the shipowner will have the same or equivalent privileges, rights and immunities as if the Hague Rules applied to the carriage.
95. Accordingly, on the correct construction of the charterparty, the appellant is entitled to rely upon no more than what Article IV provides. If, as I have found, "unit" does not apply to bulk cargoes as a matter of the construction of the Hague Rules, then the appellant does not have the protection of the limit of liability in Article IV rule 5 under the charterparty. I agree with the judge that the meaning of the rule and of the word "unit" in it cannot change because of the nature of the contract of carriage.
96. Mr Debattista placed great emphasis on the assertion that the market or commercial expectations of those engaged in the carriage of bulk cargo who traded on this form of charterparty or other forms with a similar provision to clause 26 were that they would be able to limit liability in respect of those bulk cargoes under Article IV rule 5. I have already indicated above that I am unimpressed by that submission. There is absolutely no evidence of the knowledge or expectations of those who trade in this market. In any event, whatever those expectations were, they could not possibly alter the meaning of Article IV rule 5.
97. Furthermore, as Mr Persey QC pointed out, the appellant could have protected itself by seeking to incorporate in the charterparty some form of deeming provision giving Article IV rule 5 and "unit" a different meaning from that they would have had in the absence of the deeming provision: see per Lord Bingham of Cornhill in The Tasman Discoverer [2004] UKPC 22; [2004] 2 Lloyd's Rep 647 at 651 col. 1, also an Article IV rule 5 case. See also the provision used in the 1930s in coal bills of lading as referred to by Temperley and Vaughan at 82 which stated: "the unit under Article IV rule 5 being the ton".

Limitation of liability on the facts

98. The judge found that even if "unit" did not mean a shipping unit as he found, the only viable alternative was a freight unit or customary freight unit and the appellant would not have been able on the facts to limit its liability, because under the charterparty, the freight was lump sum. Furthermore, there was no evidence that any particular freight unit was "customary" for the carriage of fish oil. As Mr Persey QC pointed out, that was the basis on which the carrier's defence failed before Evans J in The Aramis. The appellant seeks to avoid the difficulty which it would face if it relied on a limit per freight unit or customary freight unit, by arguing that the limit should be by reference to the weight of the cargo as described in the contract of carriage. Specifically, Mr Debattista argues that the limit should be per metric ton under the charterparty.
99. Even if the appellant were right that "unit" should be construed as a unit of measurement, there is an insuperable difficulty with this argument. There is no description in the charterparty of the cargo as shipped. It does not describe the cargo as 2,000 metric tons, let alone the 2,056.926 metric tons actually shipped. The Fixing Note simply stated the intended cargo as 2,000 tons of fishoil in bulk, 5% more or less in the charterers' [not the appellant's] option. The bill of lading is the only document issued in respect of the cargo actually shipped and it describes the cargo as weighing 2,056,926 kilograms. The problem for the appellant in relying upon that document is twofold. First the appellant was not a party to the contract of carriage contained in or evidenced by the bill and second, as the judge noted, limitation by reference to kilograms of cargo shipped would produce a sum higher than the amount claimed.
100. In the circumstances, it seems to me that the judge was correct to conclude that, even if the appellant were right that "unit" was capable of meaning something other than a shipping unit, its attempt to limit its liability under Article IV rule 5 must fail on the facts.

Conclusion

101. In all the circumstances, for the reasons I have given, I consider that the appeal should be dismissed.

Lord Justice Richards

102. I agree.

Lady Justice Gloster

103. I also agree.