This was an application by the plaintiffs pursuant to O 14 r 12 of the Rules of Court for a determination on a point of law, namely, whether the weight limitation under art 4.5.a of the Hague-Visby Rules should be calculated based on the actual weight of the goods lost or damaged, as opposed to the total weight of such goods. The facts are not in dispute. The plaintiffs are the owners of a cargo of 7 drums of oil-filled electric cables. Each drum contained a single cable measuring 526 metres in length and weighed an average of 22,916 kgs. The first defendants are the operators of the terminal at the port of Singapore. The cargo was shipped on board the vessel La Loire for carriage from Southampton to Singapore, for transhipment to Bangkok thereafter. A bill of lading for '7 x 20 foot flat rack containers said to contain 7 drums electric cable' was issued by the second defendants. The cargo was discharged in Singapore on 27 April 1999 in good order and condition, and stored by the first defendants whilst awaiting transhipment to Bangkok. Prior to loading onto the connecting vessel, three out of the seven drums of cables were found to be damaged. Of these three drums, one was completely damaged, while the other two drums had 103 and 114 metres of damaged cable respectively. The matter was eventually settled. The first defendants agreed to pay the plaintiffs the damages up to the limitation amount under the Hague-Visby Rules, subject to a referral to the Court for determination of the limitation quantum.

Held: The limitation quantum applicable to the three damaged drums is that proposed by the defendants, ie, based on the actual weight damaged, yielding a limitation quantum of SGD 151,814.97.

Article 4.5.a of the Hague-Visby Rules (applicable by virtue of s 3 of the Carriage of Goods by Sea Act (Cap 33)), reads as follows: '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 the equivalent of 10,000 francs per package or unit or 30 francs per kilo of gross weight of the goods lost or damaged, whichever is the higher.' The local currency equivalents are specified in the Carriage of Goods by Sea (Singapore Currency Equivalents) Order 1982, which states that 10,000 francs is equivalent to SGD 1,563.65, and that 30 francs is equivalent to SGD 4.69.

The interplay between the two principles of limitation is succinctly spelt out in The Hague and Hague-Visby Rules, Lloyd’s Practical Shipping Guides, John Richardson, 1998 4th ed, at 45: 'The limit in the Hague-Visby Rules incorporates a package/weight alternative, with whichever produces the higher limit applicable. From the figures, it is possible to calculate that, where a package weighs less than 333.33 kilos, a package limit will be applied, but where it exceeds this weight a weight based limit is applicable.'

In the present case, as the weight of each drum far exceeded 333.3 kg, no issue arose in respect of limitation by reference to a package or unit, which for the three drums would have amounted to only SGD 4,690.95. Instead, the dispute related to the method of calculating the 'gross weight of the goods lost or damaged'. The plaintiffs' case was that where a single article is partly damaged, the term 'gross weight of the goods lost or damaged' refers to the total weight of that entire article. The limitation quantum, based on the total weight of the three drums, would then be SGD 332,428.12.

The defendants, however, argued that the correct calculation should be based on the actual weight of the damaged cable, ie the weight of the totally damaged drum together with the weight of the damaged portions of the other two drums. This yielded a reduced limitation quantum of SGD 151,814.97.

It is pertinent at this juncture to recall the context in which the weight limitation under art 4.5 was drafted. The concept of limitation of liability has been present in various forms since the sixteenth century. Its primary function is to protect the carrier from liability for cargoes of high undisclosed value. By standardising the level of liability, the carrier is then able to offer uniform and hopefully cheaper freight rates to shippers. The limitation to liability is also instrumental to the parties in the determination of insurance premiums to be charged by insurers. To prevent carriers from abusing limitation clauses, the Hague Rules sought to provide a standard basis for limitation in art 4.5, calculated by reference to a fixed sum per package or unit. The shipper is of course at liberty to evade the limitation clause by declaring the full value of the cargo on the bill of lading (the exception in art 4.5.a). However, a carrier would only waive the protection of a limitation clause at a price. Such ad valorem bills of lading therefore invariably attract increased freight rates from carriers.

The weight limitation in art 4.5.a was an alternative formula of limitation introduced by the Hague-Visby Rules, which came into force in 1977. The amended art 4.5.a included a limitation quantum determined by the weight of the cargo damaged or lost. The shipper would then be at liberty to invoke whichever limb produced the higher quantum. This alternative formulation was intended primarily to deal with bulk cargoes, for which there was no sensible package description, and for which the limit would be much too low if the entire shipment was considered as one unit. Although proposed primarily with bulk cargo in mind, the weight limitation could of course be invoked where there were large units of heavy cargo (for example, cars and heavy machinery), for which the package and unit limitation would similarly have been woefully inadequate.

Counsel for the plaintiffs essentially relied on four main arguments in support of his case. First, he submitted that on a plain reading of the phrase 'gross weight of the goods lost or damaged', it was the clear intention of the drafters to refer to the entire weight of the relevant article or goods. He reasoned that should the drafters have intended the weight limitation to apply only to the damaged portion, they would have said so specifically, and the phrase would have read as 'gross weight of the goods or part thereof that are lost or damaged'. Furthermore, the plaintiffs' view was that given that the new weight limitation in the Hague-Visby Rules was a concession to bulk cargo owners, the drafters must have intended an interpretation that was favourable to cargo owners.

Secondly, counsel for the plaintiffs reasoned that penalising the cargo owner by calculating limitation on the actual weight damaged or lost would have militated against the duty to mitigate. A cargo owner who truly and successfully mitigated its losses by recovering whatever portions of cargo were salvageable would find itself doubly disadvantaged when its mitigatory efforts resulted in a lower limitation ceiling.

Thirdly, the plaintiffs contended that raising the issue of actual weight would create intractable difficulties in practice in terms of ascertaining the exact weight of the damaged portion. To illustrate, counsel cited an example of a television set damaged during shipment due to ingress of seawater. He argued that if only part of the circuitry had come into contact with seawater, there would be needless dispute as to what was the exact weight of the damaged portion. This would lead to an increase in litigation, and a corresponding increase in costs in the shipping business.

The fourth argument raised by the plaintiffs is one based on commercial certainty. Counsel for the plaintiffs submitted that the limitation provisions in art 4.5.a were intended to allow both ship and cargo interests to know at the start of the shipment the exposure of the shipowner to liability. This in turn would allow parties to better determine the appropriate freight and insurance rates. By referring to the weight enumerated on the bill of lading and calculating the weight limit based on that entire weight, parties could work with a fixed limitation quantum. Additionally, the plaintiffs submitted that such an exercise would be consistent with the interpretation of the package and unit limitation in instances where art 4.5.c applies, since the meaning of package or unit is likewise ascertained from the bill of lading where a container, pallet or similar article of transport is used to consolidate goods.

The main plank of the defendants’ case arises from a paper presented by Anthony Diamond QC on 8 December 1977, entitled The Hague-Visby Rules, written soon after the Hague-Visby Rules came into force on 23 June 1977. Counsel for the first defendants claimed that Anthony Diamond QC, in assessing the relationship between the two principles of limitation, implicitly treated 'gross weight' as referring to the actual weight of goods lost or damaged. At 240, the author stated: 'Accordingly, the general rule is that if a package or unit weighs 333.3 kilos or less, then the limit is 10,000 francs irrespective of whether all the goods were lost or damaged or only some of them. If a package or unit weighs more than 333.3 kilos and all the goods within it are lost or damaged, then the weight alternative will provide a higher limit. Finally, if a package or unit weighs more than 333.3 kilos and only some of the goods within it are lost or damaged, the limit will be 10,000 francs unless those goods weigh more than 333.3 kilos.'

Counsel for the first defendants submitted that in assessing whether the relevant weight exceeded 333.3 kgs, Anthony Diamond QC clearly referred only to the goods lost or damaged, and not all the goods in the package or unit.

The author makes the same assertion later on in the article, when he raised the question of what the limitation principle would be when goods were shipped in a container, where there was no enumeration of packages on the bill of lading. He took the view that the weight limitation was the applicable principle as opposed to the package or unit limitation. Implicit in his reasoning is his clear assumption once again that the ‘gross weight’ limitation principle is based on actual weight of goods lost or damaged, at 244: 'Fifth question: Suppose there is no enumeration of packages, how is the limitation figure to be arrived at? The claimant can never now be in a worse position than by being faced with a limit based on the weight of the goods lost or damaged. If he owns all the goods in a container and they are all lost, then a limit based on the weight of the goods is likely to produce a higher limit than a single figure of 10,000 francs. If only some of the articles stowed in a container are lost or damaged, those lost or damaged may weigh less than 333.3 kilos, in which event the limit will be based on the concept that the container is the package or unit. I can see not reason for taking the weight of all the goods stowed in a container unless all those goods are lost or damaged.'

The defendant pointed out that contemporary texts still regard Anthony Diamond QC as correct on this point: see Treitel and Reynolds, Carver on Bills of Lading, 2001 1st ed, at 530: 'Difficulties can also arise where the goods are partly damaged: is the limit calculated by reference to the weight of the damaged goods or by reference to the weight of the whole consignment? Related problems could occur if a detachable part of a machine was damaged. It seems that reference should be made to the weight of what is damaged.'

The same assumption has been made by at least one more writer. Nicholas Gaskell, in Bills of Lading: Law and Contracts, states at 519 (this text is reproduced in David Yates, Contracts for the Carriage of Goods, at 1-576/6): 'The important consequence of the weight or package alternative is that there is now a definite limit for bulk cargoes. On the above figures, a limit of £1,690 per tonne would only begin to operate for cargoes such as nickel and tin, virtually all other major bulk cargoes having a value beneath this figure. It seems reasonable to conclude that the limit applies to the weight of the goods lost and not that of the whole consignment.'

Finally, in reply to the plaintiffs’ argument, counsel for the first defendants pointed out that a plain reading favoured their interpretation of the phrase 'goods lost and damaged'. Indeed, he argued that if the drafters had intended to give effect to the plaintiffs' interpretation, they would have specified 'total' or 'entire' weight in order to negate the clear implication arising from the qualifier 'lost and damaged'.

It must be recalled that the primary canon of interpretation of the Rules is to achieve uniformity with the views taken in other jurisdictions. This was the case for the Hague Rules: see Stag Line Ltd v Foscolo, Mango & Co Ltd [1932] AC 328, per Lord Atkin at 343: 'For the purpose of uniformity it is, therefore, important that the Courts should apply themselves to the consideration only of the words used without any predilection for the former law, always preserving the right to say that words used in the English language which have already in the particular context received judicial interpretation may be presumed to be used in the same sense already judicially imputed to them.' Lord Macmillan made the same observation at 350: '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.'

The same canon of interpretation applies to the Hague-Visby Rules. In The Morviken [1983] 1 Lloyd’s Rep 1 (CMI597), Lord Diplock said at 5: '[The Hague Visby-Rules] should be given a purposive rather than a narrow literalistic construction, particularly wherever the adoption of a literalistic construction would enable the stated purpose of the international Convention viz. the unification of domestic laws of the contracting States relating to bills of lading to be evaded by the use of colourable devices that, not being expressly referred to in the Rules, are not specifically prohibited.'

While the academic texts above are not conclusive on the point, such unchallenged opinion should nevertheless be departed from only with the greatest caution. As mentioned, the plaintiffs had argued that if the drafters had intended the weight limitation to apply only to the damaged portion of the goods, they would and should have said so specifically. This argument is convincing. It inexplicably places the onus on the drafter to explain what seems perfectly unambiguous. The phrase 'goods lost and damaged' means just what it says.It cannot be taken by any stretch of the imagination to refer to undamaged goods as well.

The only instance where ambiguity can arise is where the damaged portion of the goods is somewhat inseparable, either physically or commercially, from the undamaged portions of the goods. The physical difficulty of apportionment has already been touched on by the plaintiffs' example of television sets, where it may be practically difficult to assess how many kilogrammes of circuitry were damaged by ingress of seawater. The next difficulty is a commercial one and arises from the 'affected' value as a result of a damaged component. For example, should the glass screen of the television set, though physically undamaged, be rendered commercially useless as a result of the damaged circuitry, it is not clear whether the weight of the screen should be considered under the 'actual' weight for the purposes of limitation.

These thorny issues, however, do not arise in every case, and indeed, do not arise in the present one. They do not detract from a finding that the plain and literal meaning of the phrase 'goods lost and damaged' refers to actual weight. The definition is clear even though its application may give rise to evidential disputes in difficult cases.

Conversely, the plaintiffs' interpretation of the phrase 'goods lost and damaged' is not as plain and literal as it seems on first blush. Counsel for the plaintiffs submitted that the relevant 'article' for which the total weight must be taken into consideration is that enumerated on the bill of lading. He gave an example of a bill of lading stating its contents as '1 container containing 10 boxes of handphones', each box in turn holding 10 handphones. If 3 boxes were damaged such that only half the number of handphones in each box were spoilt, the weight limitation would be based on the total weight of the three boxes. However, since '1 container' was enumerated as well, it could very well be that the weight limitation could be based on the total weight of the entire container. Hence the logical conclusion of the plaintiffs' postulation must be that the relevant 'article' from which the total weight is taken is that of the smallest of the items or modes of packaging enumerated. This seems to be the missing link in the plaintiffs' definition of the phrase. Seen in this light, the plaintiffs' definition of 'goods lost and damaged' takes on a somewhat contrived meaning. It is a tall order to say that the phrase plainly and literally means 'the total weight of the smallest article or articles of items or modes of packaging enumerated in the bill of lading'. Needless to say, the problem is exacerbated when no items are enumerated at all on the bill of lading, for there would then be disputes as what constitutes the relevant 'article' on which the weight limitation is based. On the whole, it seems that the plaintiffs' reading of the phrase 'goods lost and damaged' is by far more convoluted than that proposed by the defendant, and that on balance, a plain reading of the provision favours the defendants' case.

The plaintiffs' strongest contention is that calculating the limitation by reference to the weight of the entire article promotes commercial certainty. This manifests at two stages - first, at the stage of the formation of the contract of carriage, and secondly, at the time of litigation. At the formation of the contract, the plaintiffs say that by examining the weight of the item enumerated on the bill of lading, a fixed limitation quantum could be derived, which in turn allows the parties to make better risk assessments for the purposes of freight and insurance. The bill of lading, in short, would be the 'be all and end all' in respect of the parties' rights and liabilities. Liability would no longer be contingent on the actual damage suffered. This argument, while superficially attractive, betrays blemishes on closer inspection. First, as mentioned earlier, if no specific items are enumerated on the bill of lading, disputes may arise as to which is the relevant article from which to calculate 'total' weight. This would detract from rather than promote commercial certainty. Next, where multiple items are enumerated, all that is certain is the maximum ceiling of liability. The actual ceiling in such cases would vary according to the number of items affected (although they could be affected only in part). Quite clearly, a calculation based on 'total' weight as opposed to 'actual weight' does not, in such instances, promote the cause of certainty. Indeed, the only instance where a fixed limitation could be achieved is where there is only one item is enumerated on the bill of lading. Only in such a case would calculating limitation based on 'total' weight produce an absolute as opposed to a variable figure.

This of course begs the question whether such absolute certainty is desirable, or indeed intended by the drafters. The problem with absolute certainty is that it runs against the tenor of the package and unit limitation, which by definition varies in accordance with the number of packages and units damaged or lost. The notion of absolute certainty also defies common sense when taken to its logical extremes. This is best exemplified in large bulk cargo shipments of valuable cargo, say of nickel or tin. The limitation quantum would then be based on the weight of the entire shipment, whether or not it was 1 kg or 1 tonne that was actually damaged. This limitation quantum could thus be alarmingly disproportionate to the quantum of damage. Worse, in effect, the limitation clause would in reality be rendered redundant since it would only kick in where the damaged cargo was of sufficient weight and value to breach the high limitation ceiling in the first place. It is indeed questionable whether the drafters of the Hague-Visby Rules intended certainty to be purchased at so high a price to shipping interests.

As for the question of certainty at the stage of litigation, the plaintiffs contend that the need to ascertain the extent of the actual damage leads to increased costs of litigation. Expert assessors need to be called, creating issues rife for litigation, thereby increasing legal costs and hampering settlement. Contrarily, counsel for the first defendants argued that actual loss or damage had to be proved by the plaintiffs in any case. To this end, the fact that disputes on assessment arise could not fairly be attributed to the limitation issue. While assessment of actual damage may be more hotly disputed given its additional significance on limitation, the plaintiffs' argument is unpersuasive. The need to assess actual damage does not militates against the duty to mitigate. It is never in the plaintiff’s interest to mitigate its damages. The fact that its mitigation would additionally reduce the 'actual' damage and thereby face a lower limitation quantum does not alter the plaintiff's propensity to perform its obligations under the general law of contract.

It must be recalled that the Hague Rules and the Hague-Visby Rules are the result of a careful balance of interests between cargo owner and carrier. In exchange for duties and obligations which cannot be contracted out of, the carrier receives the benefit of the limitation clause in art 4.5.a. The amendment to art 4.5.a under the Hague-Visby Rules was targeted towards filling a lacuna relating to bulk cargoes (which could not fit into the package or unit description), as well as for large units of cargo with greater weight (for which the package and unit limit would be disproportionately low).

The plaintiffs' proposal for calculating the weight limitation based on 'total' weight, based on what is enumerated by the shipper on the bill of lading, may disturb the careful balance of interests achieved thus far under the Rules. The plaintiffs' method of calculation conceives of the weight limitation replacing the package or unit limitation the moment the total weight of that particular article exceeds 333.3 kg. Thus, if a bill of lading stipulates a container as the relevant package, pursuant to art 4.5.c, once the total weight of the container exceeds 333.3 kg (which it almost certainly will), the limitation quantum would rise above SGD 1,563.65. Such adventurous interpretation may understandably have serious financial repercussions on the costs of shipping. Moreover, this is a route of interpretation along which no other judge or academic in a relevant jurisdiction has trodden. Accordingly, it is not wise to adopt the plaintiffs interpretation of art 4.5.a until such time as a clear academic and judicial consensus has crystallised on the issue.