Reproduced with permission of 27 Virginia Journal of International Law (1986-1987) 729-802

     [Introduction]

1. The Problems of Conflicting Interpretations
2. International Conflicts on the Validity of Himalaya Clauses under the Hague Rules
   1. Himalaya Clauses in the United States
   2. Himalaya Clauses in England
   3. Himalaya Clauses in Australia
   4. Himalaya Clauses in Canada
   5. Himalaya Clauses in West Germany
   6. Summary
3. International Conflicts on the Validity of Choice of Forum Clauses under the Hague Rules
   1. Choice of Forum Clauses in Countries with Statutory Bans
   2. Choice of Forum Clauses in the United States
   3. Choice of Forum Clauses in England
   4. Choice of Forum Clauses in Belgium
   5. Choice of Forum Clauses in Pakistan
   6. Choice of Forum Clauses in the E.E.C.
   7. Summary
4. Conclusion

[INTRODUCTION]

Uniform laws are each year governing more and more aspects of private activity, both domestically and internationally. They have not yet achieved their full promise, however, because uniformity of interpretation has not always followed uniformity of enactment. Different courts have construed identical provisions in different ways, and these conflicting interpretations have undermined the uniformity of the uniform laws. Although the resulting partial uniformity is preferable to total diversity, there is still considerable room for improvement.

A solution to the problem of conflicting interpretations requires an understanding of how conflicts arise. Existing scholarship recognizes the problem, and some writers have tried to explain it, but current explanations are inadequate. This Article takes the first step toward a solution. It reexamines the issue in the international context, and offers a new hypothesis to explain how conflicts develop in the interpretation of uniform laws.

Domestic uniform laws are familiar in the United States and other federal systems.[1] The Commissioners on Uniform State Laws [page 729] have proposed over a hundred such statutes,[2] the Uniform Commercial Code being the best-known example.[3] A rapidly expanding body of international uniform legislation governs individual activity on substantially the same terms in different countries.[4] These international uniform laws obtain force in national legal systems either through enactment as domestic statutes or through ratification of international conventions.[5] Worldwide organizations, such [page 730] as the United Nations Commission on International Trade Law (UNCITRAL), the International Institute for the Unification of Private Law (Unidroit), and the Hague Conference on Private International Law, and regional organizations, such as the Organization of American States and the European Economic Community, are all actively promulgating new uniform laws.[6]

Significant benefits exist in having the same rule apply in each jurisdiction whenever a transaction involves more than one legal system. Citing advantages such as certainty,[7] predictability,[8] convenience,[9] and stability,[10] the legal community has recognized the benefits for generations.[11] Indeed the point is so well established that some commentators simply assert that uniformity is beneficial without further discussion [12] The advantages of uniformity, however, require more than a uniform text.[13] The various courts that [page 731] apply the text must do so uniformly before all of the promised benefits can be fully realized.

The opening sentence of the Uniform Commercial Code's official commentary recognizes the importance of uniform interpretation in the domestic context: "Uniformity throughout American jurisdictions is one of the main objectives of this Code; and that objective cannot be obtained without substantial uniformity of construction."[14] The problem of non-uniform construction is even greater in the international context, where national courts interpret international uniform law that has become part of the domestic legal system. Viscount Simonds expressed the danger clearly in the British House of Lords: "It would be deplorable if the nations should, after protracted negotiations, reach agreement … and that their several courts should then disagree as to the meaning of what they appeared to agree upon."[15]

Courts [16] and commentators [17] have long recognized the existence of conflicting interpretations of uniform laws. A few commentators have attempted to explain why such conflicts arise.[18] The existing [page 732] literature, however, tends to treat the problem as merely an instance of judicial centrifugal force. The number of courts independently construing a uniform law, it suggests, makes diverging interpretations inevitable.

This Article argues that international [19] legal conflicts are not simply random occurrences. On the contrary, independent domestic legal concerns push national courts into differing interpretations of supposedly uniform laws. Each court considers itself bound to interpret and apply international uniform law in a manner that will avoid inconsistency or tension with its own domestic law. Constrained by substantively different domestic laws, national courts allow their desire to minimize the disruptive effects of international law to overwhelm their mandate to maintain uniformity. Diverging interpretations of the international law are the result.

The Brussels Convention for the Unification of Certain Rules of Law Relating to Bills of Lading,[20] commonly known as the Hague Rules, provides an excellent example of this process[21] The United [page 733] States ratified the convention and enacted its operative provisions as the Carriage of Goods by Sea Act (COGSA).[22] Virtually every country involved in maritime commerce has ratified the Hague Rules [23] or enacted a statute in substantially the same terms.[24] Study of systematic differences among national court interpretations of the Rules strongly suggests that conflicts are inevitable as long as the ultimate authority to interpret them is entrusted to national courts that feel bound by differing domestic law constraints.

Part I describes the problem of conflicting interpretations and its significance in the context of a particular international uniform law (the Hague Rules), reviews the contributions that others have made in explaining the problem, and summarizes the Article's central hypothesis. The hypothesis explains these conflicts as a predictable product of variations in the substance of independent aspects of domestic law. Part II tests the hypothesis through a detailed examination of a leading controversy in the interpretation of the Hague Rules. It considers decisions governing the ability of an independent contractor to claim the benefit of limitations on liability that the Hague Rules grant the carrier.[25] Part III offers a second illustration, examining conflicts among a number of countries on the effect of a choice of forum clause under the Hague Rules. The purpose in each case is not simply to demonstrate that an international conflict exists, but to use the specific example as a [page734] case study to see how the conflict arises. Finally, Part IV summarizes the Article's conclusions.

I. THE PROBLEM OF CONFLICTING INTERPRETATIONS

The Hague Rules are a well-established example of international legislation intended to produce uniform law throughout the maritime world. In essence, they create a set of standard terms and incorporate them into bills of lading to allocate responsibility for loss or damage to cargo carried in international commerce. Under general maritime law, a carrier was absolutely liable for any cargo loss unless it could show that one of a short list of exceptions applied.[26] To extend these very narrow exceptions, ship owners included limitation and exemption clauses in their bills of lading to excuse themselves from liability in additional circumstances. The Hague Rules set strict limits on the protection that a carrier could claim, thus determining which party is responsible for specified losses and the extent of the responsibility.[27]

A compromise between carriers and cargo interests determined the precise terms of the risk allocation,[28] but that is of relatively minor importance here. This Article addresses neither the merits of the compromise nor the proper construction of specific provisions.[29] The important point is that the Hague Rules seek to create a clear and uniform allocation of responsibility for cargo losses on an international scale.[30] [page 735]

Conflicts in the interpretation of the Hague Rules do not simply destroy some aesthetic symmetry in the international legal order. The lack of uniformity imposes real costs on the commercial system that the Rules govern. As a practical matter, insurance covers virtually every loss in the international carriage of goods by sea.[31] The Hague Rules' allocation of risk, therefore, determines which insurer (cargo's direct insurer or carrier's liability insurer) will bear the loss, and thus which party will bear the direct cost of the insurance premium.[32]

In an ideal system, the law would allocate risk to minimize overall costs.[33] In the real world of shipping, however, it is often difficult to predict what effect, if any, liability rules will have on costs.[34] The next best solution is a clear and uniform rule. Even if arbitrary, it would enable each party to insure against only those risks for which it will be held responsible. When conflicts in interpretation create uncertainty, however, neither party can know whether it bears a given risk. Thus both parties must insure against the risk, creating a needless double insurance expense.[35]

Furthermore, uncertainties in the governing law produce litigation, which is always wasteful.[36] Each side spends money to allocate [page 736] losses that have already occurred and to clarify the allocation of future losses. Litigation does not reduce overall losses at best it does what the Rules themselves were intended to do.

Unfortunately, conflicting interpretations are not an abstract problem. Many Hague Rules issues have been litigated in more than one country, and the results have often varied. Clear conflicts exist on such topics as the right of an independent contractor to claim the carrier's limitations of liability,[37] the enforceability of a choice of forum clause in a bill of lading,[38] the meaning of the term "package,"[39] the definition of a "peril of the sea,"[40] the question whether deck carriage is a "deviation,"[41] the burden on the carrier in fire damage cases,[42] and the application of the time limit for [page 737] bringing suit.[43] Although the existence of such conflicts is well-recognized, no one has satisfactorily explained how and why they arise.

Conflicts in the interpretation of the Hague Rules [44] and other international uniform laws [45] have been the subject of academic attention, but the discussion has generally focused on the existence of particular conflicts with little thought for their origin. One of the more interesting contributions in the maritime field, for example, is a short piece in which Charles Black advocates an "international court of appeals" with discretionary jurisdiction over national court decisions construing the Hague Rules,[46] Black, however, assumes the existence of international conflicts as his starting position:

Some commentators suggest more directly that international conflicts in interpretation are random occurrences.[48] The idea is simply that, when so many national courts construe uniform laws, a form of judicial centrifugal force makes diverging interpretations [page 738] inevitable. The logic of this theory is apparent, for a simple numerical example can illustrate the potential impact of random distribution. Suppose a uniform law provision has five equally plausible interpretation.[49] The first court to construe it will adopt one of them. If a foreign court independently [50] construes the same provision, there is only a 20% chance it will adopt the same interpretation. Thus there is an 80% chance that a conflict will develop. If ten courts faces the issue, the odds against their agreeing are almost two million to one. Even with far weaker assumptions, the improbability of concurrence quickly becomes apparent,[51] so the random distribution hypothesis has considerable appeal.

Some writers support this view implicitly. O.C. Giles, for example, argues that conflicts develop when judges are unwilling or unable to study foreign legal systems.[52] The failure to study foreign law is significant because a court then construes international conventions independently, without regard for foreign views. But that only leads to a random distribution among parochial courts. It does not increase the likelihood that their interpretations will systematically differ; it is simply a matter of chance which results they happen to reach.[53]

Similarly, several authors suggest that conflicts arise because of the varying means by which nations incorporate a convention into [page 739] domestic law.[54] A court is more likely to follow the international understanding when it has the actual convention before it, the theory suggests, and is more likely to diverge if the substance of the convention is embedded in a national code in which judges lose sight of its international nature. Once again, the theory says no more than that interpretations will tend to be random when courts lack relevant material to guide them. There is no reason to suppose that poorly informed courts would systematically disagree with better informed courts. They may, simply by chance, happen to reach the same results, or they may diverge in either direction.[55] When and where conflicts occur is a matter of random distribution.

Some British observers argue that conflicts arise through differences in the methodology of treaty interpretation.[56] English courts were traditionally unable to consult the travaux préparatoires in the interpretation of a convention,[57] while courts in other countries suffered no such disability.[58] Yet again, the underlying theory is simply that interpretations will tend to be random when courts lack relevant material to interpret a convention.

A few commentators have gone beyond random distribution thinking. They offer theories that not only predict the existence of [page 740] conflicts, but explain why specific conflicts occur in the way they do. Jacob Sundberg, for example, observes that "the technical terms" of a uniform law "often have different connotations in the different legal systems," and that this leads to diverging interpretations.[59] This theory suggests that conflicts occur systematically, rather than randomly. A French court will interpret a convention by applying the French connotation of a technical legal term while a German court will apply the German connotation. This is not random, for the result is predictable.[60] Implicit in this theory is the notion that conflicts in interpretation occur primarily through imprecise drafting. If only the choice of language had been more circumspect, the connotations would not have differed. Although it is true that better drafting could eliminate some conflicts,[61] there are still too many cases where this simple explanation is inadequate.[62] Furthermore, this risk is well-recognized and courts take affirmative steps to avoid it.[63]

A similar idea on a. larger scale attributes conflicts to differences in legal tradition.[64] A common law court approaches a problem differently than does a civil law court -- and both traditions are different than one would find in a socialist court. An English court may [page 741] give more weight to prior judicial decisions when construing an international convention, for example, while a French court may pay closer attention to academic literature. This theory may be a useful start, but it is still inadequate. These broad differences in legal tradition do not explain why there are conflicts between United States and Canadian,[65] or between Belgian and German,[66] interpretations of a convention. Moreover, the argument loses much of its force in cases where the courts are fully aware of the international status of the convention and how foreign courts have interpreted it. It may offer a systematic explanation for some conflicts, but it does not go far enough.[67]

Another possibility is that national policies affect judicial interpretation.[68] Perceptions of national interest influence the diplomatic conferences that negotiate conventions [69] and the political bodies that ratify them.[70] If these national policies are a significant cause of conflicts in the interpretation of international uniform laws, the Hague Rules should provide an ideal illustration of the phenomenon. The Hague Rules themselves were a response to [page 742] conflicting national laws promoting distinct national interests.[71] Even today, countries are commonly identified as "carrier" nations or "cargo" nations.[72] For the most part,[73] however, this seems to be an irrelevant consideration. England is a leading carrier nation, but the House of Lords' initial response to the "Himalaya clause" problem [74] was strongly pro-cargo.[75] The United States is a cargo nation, but its Himalaya clause jurisprudence favors the carrier.[76] The primary judicial interest is in reconciling a technical legal convention with national law, not with national policy.[77]

The principal weakness in prior attempts to explain international conflicts of interpretation has been the failure to consider the impact of substantive domestic law on the process of interpretation.[78] Interpreting a uniform law is not simply a matter of looking to a specific provision and construing the language used. National courts have considerable leeway in reaching their conclusions, and a court operating in a vacuum might be justified in adopting any of a wide range of permissible interpretations.[79] Although random selection may play some role in the process of choosing the interpretation to adopt,[80] the problem goes deeper [page 743] than that. Substantive domestic law greatly influences the national court's choice of interpretations and, unless checked by stronger considerations of international uniformity or comity, ultimately makes conflicts inevitable.

When a national court must interpret a uniform law, it faces not only a range of permissible interpretations of the international legislation but also an entire body of domestic law.[81] A court will try to reconcile the uniform law with the domestic law. Some permissible interpretations may be totally inconsistent with domestic law; others may be to some extent inconsistent or create tensions with domestic law; and still others may be entirely consistent with domestic law.[82] A court generally chooses from among the consistent interpretations.

The process of interpreting one law to maintain consistency with [page 744] another is familiar to United States constitutional lawyers. If a court can construe a statute in several ways, and one construction would violate the Constitution while another would not, the court must construe the statute consistently with the Constitution.[83] The Supreme Court has candidly admitted that it " 'will often strain to construe legislation so as to save it against constitutional attack'."[84] Although domestic law in the present context generally lacks the force of the Constitution, the process is analogous. If it can, a court will construe an international uniform law to be consistent with domestic law, even when it must strain to do so.

Closer analogies [85] exist under well-established canons of statutory interpretation. The presumption against repeals by implication, for example, simply requires a court to construe two statutes consistently with each other if possible.[86] Similarly, a court will endeavor to construe a statute to avoid a conflict with international law [87] or a treaty.[88] In each case, a court faces a range of permissible interpretations and chooses one that will be consistent with the [page 745] external constraints that exist in its legal system. For a court construing an international uniform law, the most significant constraint in fact, if not in theory,[89] is substantive domestic law.

Once a court has interpreted a uniform law to have one meaning, a conflict arises when a national court in a second country faces a different body of domestic law that constrains it in different ways. The range of permissible interpretations of the uniform law will be the same, but the domestic constraints are not. Options that were consistent with the national law of the first country will violate the law of the second, and options that failed in the first will be perfectly acceptable in the second. The second court will generally adopt an interpretation that is consistent with its own domestic law, thus creating a conflict between the two interpretations. As this scenario is repeated in a number of different courts, an entire range of interpretations arises.

National Courts' opinions in cases applying the Hague Rules reveal the major role played by domestic law constraints. Time and again, the perceived need to reconcile the Rules with national law has produced interpretations that vary significantly from country to country. A detailed examination of relevant decisions on two specific issues leads to the conclusion that these variations will continue as long as the ultimate interpretation of the Hague Rules is left to national courts subject to different domestic law constraints.

II. INTERNATIONAL CONFLICTS ON THE VALIDITY OF HIMALAYA CLAUSES UNDER THE HAGUE RULES

As part of the essential compromise between carrier and cargo interests,[90] the Hague Rules limit a carrier's liability in specified ways. Article 4(5), for example, limits the carrier's liability to £100 [page 746] "per package or unit" unless the shipper declared a higher value before shipment.[91] Article 3(6) discharges all liability unless the claimant brings suit within a year of delivery.[92] Article 4(2) exempts the carrier from liability for loss or damage from certain designated causes, such as errors in the navigation or management of the ship,[93] perils of the sea,[94] or acts of God.[95]

The Hague Rules specify that each of these limitations and exemptions benefits "the carrier" and "the ship."[96] The term "carrier," however, is not limited. The Rules say only that it "includes the owner of the vessel or the charterer who enters into a contract [page 747] of carriage with a shipper."[97] The Rules are ambiguous, therefore, as to how far the carrier's protections extend or how far the carrier can extend them. Most carriers are corporate entities that can perform a contract of carriage only through agents, but the Hague Rules do not explicitly protect agents. Do the limitations in favor of the carrier extend to the master and crew? Or can a cargo owner sue them for cargo damages resulting from errors in the navigation or management of the ship, notwithstanding the carrier's Article 4(2)(a) exemption? Independent contractors pose another complicated question of interpretation. Although some carriers load and unload their own ships,[98] most contract with an independent stevedore to perform this service.[99] Thus a third party often performs the carrier's duties during loading or unloading, which is when a high proportion of cargo damage occurs. The carrier is ordinarily liable for these losses, but that liability is subject to the Hague Rules limitations. Can a cargo owner avoid the effect of these limitations by bringing an action against the stevedore directly?

In some countries, the courts automatically extend the carrier's exemptions and limitations to those performing its obligations under the contract of carriage. In others, including the major common law countries, the courts interpret the Hague Rules not to protect third parties. Here another question arises: What is the effect of a bill of lading clause that purports to extend the carrier's exemptions and limitations to third parties? Suppose, for example, that a bill of lading includes the following provision:

This is a simple example of a "Himalaya clause," which is named for an early case in the field.[101] Carriers commonly include such a clause in their bills of lading. Cargo owners, or their insurers, have challenged its legal validity many times, in several countries, with diverse results. Under present law, for example, a United States [102] or German [103] court would almost certainly allow a stevedore to rely on the clause. An English, Australian, or Canadian court would probably reach the opposite conclusion.[104]

The remainder of Part II examines the development of the law on Himalaya clauses in five countries. The reported cases, for the most part, involve essentially the same factual situation: The stevedore (or occasionally another independent contractor) negligently damages or loses the goods while loading, unloading, or storing them. The cargo owner cannot recover for the full extent of the damages from the carrier (either because the Article 4(5) package limitation reduces the carrier's liability below the actual loss or because the Article 3(6) time bar prohibits suit completely) and therefore sues the stevedore. The stevedore then claims the benefit of the carrier's limitations and exemptions.

A. Himalaya Clauses in the United States

The stevedore's entitlement to the carrier's limitations of liability [page 749] under the Hague Rules has been the subject of considerable litigation in the United States, even before carriers routinely included Himalaya clauses in bills of lading, beginning with the seminal case of A.M. Collins & Co. v. Panama Railroad Co.[105] The defendant stevedore damaged a package during unloading, the consignee sued for the full extent of the damages, and the stevedore claimed the benefit of the carrier's $500 per package limitation. The bill of lading did not include a Himalaya clause. The Fifth Circuit Court of Appeals acknowledged that the stevedore was not a party to the bill of lading,[106] but decided that the identity of the formal parties was not controlling. It instead ruled that the bill of lading governs every step of the transportation, including unloading the goods, whether the carrier or its agent performed the service.[107] The court found no reason to distinguish this case, in which the carrier had contracted with an independent stevedore, from the case in which the carrier had hired its own employees to do the unloading. In the latter situation the COGSA limitations undoubtedly would have applied.[108] The stevedore thus succeeded on the Hague Rules alone, without an explicit Himalaya clause.

The reasoning in Collins illustrates the influence of independent domestic law. Despite a passing reference to "the need for uniformity in ocean bills of lading,"[109] the court devoted its attention to reconciling its conclusion with other aspects of domestic law. It concluded that COGSA superseded an earlier local statute,[110] that a Supreme Court case decided before the promulgation of the Hague Rules was distinguishable,[111] and that a federal statute governing railroad bills of lading offered a useful analogy.[112] The court's primary authority for its ultimate conclusion was section 347 of the Restatement of Agency, which provided that an agent acting pursuant to its authority enjoys the non-personal exemptions [page 750] and limitations of its principal.[113] Although the Restatement rule is not unique to American law,[114] the principle it embodies is not universally recognized.[115] Even United States courts have declined to apply it in similar circumstances.[116]

The Collins dissent does not take issue with the majority's focus on domestic law concerns. On the contrary, it adopted the same approach and argued that the opposite result was more consistent with United States law. The dissent relied primarily on a pre-Hague Rules decision of the Supreme Court,[117] the principle that statutes in derogation of the common law must be strictly construed,[118] and the rule that an agent is liable for its own negligence.[119] The goal was not simply to apply the Hague Rules, but to apply them in the way that was most consistent with domestic law.

Several courts followed Collins during the mid-fifties,[120] but in 1958 the Fourth Circuit Court of Appeals reached the opposite conclusion [121] and the Supreme Court granted certiorari to resolve [page 751] the conflict. The shipper in Robert C. Herd & Co. v. Krawill Machinery Corp.[122] sued the stevedore for the full extent of the cargo damages, challenging the defendant's reliance on the $500 package limitation. The stevedore first contended that COGSA section 4(5) protected it directly, but the Court determined that Congress had not intended the limitation to extend beyond the carrier itself.[123] (The Court did not consider what the diplomatic conference that adopted the Hague Rules may have intended.) The stevedore next argued that it was entitled to the limitation under the bill of lading. The bill of lading also referred only to the carrier, and the Court would not allow this extension of the limitation clause without clear language to that effect.[124] Finally, in a subtle variation on its first argument, the stevedore contended that it was entitled to succeed indirectly under the Collins theory, even if not entitled to any limitation in its own right. The Court, however, found Collins contrary to the "long-settled" rule that an agent is liable for its own negligence, and disapproved of the decision.[125]

The Herd Court was exemplary in its recognition of the international aspect of the case. It saw that COGSA "was lifted almost bodily from the Hague Rules,"[126] noted that its purpose "was to establish uniform ocean bills of lading,"[127] and considered English and Australian decisions on the issues before it.[128] It nevertheless decided how it would apply the Hague Rules on the basis of consistency with other aspects of domestic law. It relied, for example, on the principle that a rule of law in derogation of the common law must be strictly construed.[129]

The centerpiece of the Court's reasoning was the rule that an agent is liable for its own negligence. Although the rule exists in some form in other legal systems, neither the rule nor its application is universal.[130] The English courts, for example, did not recognize [page 752] it in this context in the nineteenth century, and arguably would not have done so when the Hague Rules were first adopted.[131] German courts apply an exception in these precise circumstances [132] In any event, the Court treated the rule strictly as a matter of domestic law. It cited three cases to establish the rule's scope; all were uniquely American. (The first held that a private operator of a public vessel did not enjoy the protection available to the government under the federal Suits in Admiralty Act [133] The second held that a corporation acting as agent for the United States was not entitled to sovereign immunity or the right to be sued only in the Court of Claims.[134] And the third held, in relevant part, that a state's eleventh amendment privilege is not communicated to its agents.[135]) The Court simply construed the Hague Rules in the manner it found most consistent with domestic common law.

Although the Herd Court decided every issue against the defendant, the case ultimately favored stevedores. Relying implicitly on the United States third-party beneficiary doctrine [136] (a doctrine that is not recognized in England or the Commonwealth [137]), the Court strongly suggested that a properly drawn bill of lading could limit an agent's liability.[138] Later courts cited Herd for the proposition that COGSA limitations for the benefit of the carrier do not extend to third parties such as stevedores,[139] but the case ensured [page 753] that future bills of lading would include Himalaya clauses and almost ensured that they would be effective to protect stevedores.[140]

Subsequent litigation in the United States addressed a different question: Given that COGSA itself does not protect stevedores, what language in the bill of lading will entitle a stevedore to rely on the carrier's limitations?[141] In the leading case, the bill of lading extended "all limitations of and exonerations from liability" to "all agents and all stevedores and other independent contractors whatsoever."[142] The court cited Herd for the proposition that the parties to a bill of lading may extend a benefit under the contract to a third party by clearly expressing their intention to do so, and held that the clause was sufficient. Later courts upheld more widely-drawn limitation clauses, even when the bill of lading did not specify "stevedores."[143] Clauses covering "all agents and independent contractors,"[144] or even "bailees,"[145] have been sufficient. Not all courts have been so willing to uphold an exemption, however, and several Himalaya clauses have failed despite their similarity to those that have succeeded. A bill of lading clause extending the carrier's limitations to "all persons rendering services in connection with performance of this contract," for example, failed on the [page 754] ground that its language did not "exhibit the clarity required" to protect the stevedore.[146]

United States law denies stevedores an automatic entitlement to a carrier's limitations on liability, but as a practical matter the carrier's protection is readily available. Despite the problems in determining exactly what language will be effective, a carrier can extend its limitations through a bill of lading clause clearly expressing the intention to make the extension. The law has changed substantially since the issue first arose in the early 'fifties, but the course of the development -- and every step along the way -- has been firmly guided by domestic law constraints independent of the Hague Rules.

B. Himalaya Clauses in England [147]

England has also been the site of considerable litigation on this issue, but despite broad dicta about the need for international uniformity,[148] the results have been fundamentally different from the results in the United States or on the Continent. The leading [page 755] English decision is Scruttons v. Midland Silicones.[149] The defendant stevedore negligently damaged a package shipped from New York to London on an American carrier under a bill of lading incorporating the United States COGSA. There was no Himalaya clause. The consignee brought suit for £593, the full extent of the damages, but the stevedore claimed the benefit of the carrier's $500 per package limitation. The House of Lords, following the Supreme Court's holding in Herd,[150] found that the term "carrier" in COGSA and the bill of lading did not include stevedores,[151] and that the consignee was accordingly entitled to recover its full damages from the stevedore.

The House continued beyond this straight-forward application of United States law, however, and declared that even an express term in the bill of lading would have been inadequate because the stevedore was not a party to the bill of lading. This conclusion rested on the English rule that contracts for the benefit of a third party cannot be enforced by the beneficiary,[152] a doctrine unique to England and the British Commonwealth.[153] The Law Revision Committee had recommended the abolition of the restrictive third-party beneficiary rule over twenty years earlier,[154] and since Scruttons a number of law lords have expressed their support for the recommendation.[155] Indeed a majority in Scruttons expressed reservations about the rule.[156] Moreover, it would have been easy to avoid reaffirming the third-party beneficiary rule without violating the doctrine of precedent.[157] But the House of Lords, as an English [page 756] court, was more concerned with reconciling the Hague Rules with English law (just as the Supreme Court had reconciled them with United States law) than with finding the proper international interpretation or furthering international uniformity.

The principal dispute in Scruttons concerned the proper application of Elder, Dempster & Co. v. Paterson. Zochonis & Co.,[158] a pre-Hague Rules decision that had suggested limitations on the third-party beneficiary rule in the context of the carriage of goods by sea. The charterer in Elder, Dempster had issued bills of lading excluding liability for damages caused by stowage. The consignee contended that the exclusion clause did not protect the ship owner, who was not a party to the bill of lading contract between the charterer and consignee. Each of the law lords rejected this argument in a single paragraph near the end of his judgment.[159] Although there was no clear rationale in Elder, Dempster,[160] the narrow holding itself was clear, and each member of the Scruttons court discussed in some detail how it might apply.[161] The majority [page 757] decided that the decision was distinguishable, but the amount of attention it received is nevertheless instructive. Elder, Dempster had been decided five months before the Hague Rules were signed, and it would have had a different outcome if they had applied.[162] Yet the House of Lords was apparently prepared to base its interpretation of the Hague Rules in Scruttons on this earlier case -- a striking confirmation of the influence of independent domestic law.

Lord Denning's dissent in Scruttons highlights the extent to which substantive domestic law influences the interpretation of the Hague Rules. He examined the development of English contract and tort law, and concluded that the consignee would have had no claim against the stevedore if the case had arisen in the nineteenth century.[163] The stevedore would have owed no duty of care to the owner of the goods. Its duty would have been defined by its contract with the carrier -- a contract to which the consignee was not a party.[164] Only after the recognition of negligence as an independent tort in 1932,[165] eight years after the adoption of the Hague Rules, could the consignee have sued the stevedore directly, without regard for contractual limitations [166] That change in British law effectively changed the scope of the protection available under the Hague Rules in England.

Lord Reid agreed with the majority, but in dicta he proposed a complex arrangement that could give the stevedore the benefit of the carrier's Hague Rule limitations and exemptions consistently with English law. He suggested creating a contractual relationship between the stevedore and the shipper. Elaborate measures were required to evade the third-party beneficiary rule: [page 758]

This suggestion may circumvent the problem in Scruttons, but it does not escape the constraints of domestic law. The procedure not only requires careful execution to succeed under the third-party beneficiary rule, it is also subject to other requirements of contract and agency law. Even the acceptance of the proposal would leave English law with far stricter requirements than under American law as to what a bill of lading must contain to limit a stevedore's liability.[168]

Thirteen years after Scruttons, the Privy Council [169] accepted Lord Reid's suggestion. In The Eurymedon,[170] the defendant stevedore negligently damaged cargo shipped from Britain under a bill of lading that arguably satisfied Lord Reid's suggestions.[171] The consignee brought suit in the New Zealand courts for £880, the full extent of the damages, more than a year after the accident. The stevedore claimed the benefit of the carrier's limitations -- including the requirement that all claims be brought within [page 759] one year and the £100 limitation of liability. The trial court accepted Lord Reid's four requirements for validity and decided that the bill of lading had fulfilled them [172] The New Zealand Court of Appeal, in a unanimous judgment, also accepted Lord Reid's requirements, but decided that the stevedore had not satisfied the consideration requirement.[173]

The Privy Council, by a three-to-two vote, reversed and held for the stevedore. The critical issue was again consideration.[174] Lord Wilberforce, speaking for the majority, admitted that "English law [had] committed itself to a rather technical and schematic doctrine of contract," but he "forc[ed] the facts to fit uneasily into the marked slots of offer, acceptance and consideration."[175] The bill of lading, he argued, created "a bargain initially unilateral but capable of becoming mutual, between the shipper and the [stevedore], made through the carrier as agent."[176] The stevedore's discharging the goods for the shipper's benefit was the consideration supporting the agreement. The exemptions and limitations in the bill of lading therefore protected the stevedore.[177]

Lord Wilberforce's analysis may have conformed the contract to [page 760] "commercial reality,"[178] but it is fair to say that it "g[a]ve an unnatural and artificial meaning to the clause."[179] And this artificiality was due entirely to the desire to reconcile the operation of the Hague Rules with domestic law. Furthermore, such a strained approach was not universally acceptable, even in that part of the common law world that recognizes the English third-party beneficiary rule. In The Eurymedon itself, for example, five of the eight appellate judges who considered the matter were unpersuaded.[180] While the Eurymedon decision allowed the stevedore to limit liability, it did little to clarify the law. Privy Council decisions have virtually no binding precedential value,[181] and this case proved to be easily distinguishable on its facts.[182] Moreover, its persuasive value was limited because the decision depended on the resolution of arcane and technical points of English contract law.

In 1980, the Privy Council returned to the issue in The New York Star [183] -- its final appeal from the High Court of Australia.[184] [page 761]

Lord Wilberforce, speaking for a unanimous court this time, reaffirmed the Eurymedon principles in no uncertain terms.[185] A Himalaya clause must still satisfy Lord Reid's requirements,[186] but a stevedore's limitation of liability appears to be the rule rather than the exception under English law.

One way or another,[187] English law now permits a stevedore to obtain the benefits of a bill of lading limitation clause, but only if the clause is drafted to satisfy highly technical domestic rules. A clause that merely meets the minimum requirements of American law will not succeed in England.[188] This may be surprising to the [page 762] vast majority of lawyers who believe that the law in England and the United States is essentially the same on commercial maritime matters, particularly when an international uniform law governs in both countries. But fundamental differences in the legal status of contracts for the benefit of a third party result in a different interpretation of the governing "uniform" law.

C. Himalaya Clauses in Australia

The New South Wales courts decided two of the earliest cases dealing specifically with a stevedore's right to the benefit of a carrier's limitations of liability under the Hague Rules.[189] In Gilbert, Stokes & Kerr Pty. v. Dalgety & Co.,[190] a stevedore claimed the benefit of the bill of lading's £100 limitation clause. The consignee, relying on the English third-party beneficiary rule, objected that the stevedore was not a party to the bill of lading and thus was not entitled to any benefit under it. The court, relying heavily on Elder, Dempster & Co. v. Paterson, Zochonis & Co.[191] and other cases decided prior to the adoption of the Hague Rules, rejected this objection and allowed the defense on common law bailment principles.[192] Three years later, Waters Trading Co. v. Dalgety & Co.[193] reaffirmed this reasoning, Courts [194] and commentators [195] [page 763] cited the two cases together to support the proposition that a carrier's limitations on liability also protect the stevedore.

Five years after Waters, the High Court of Australia, by a three-to-two vote, overruled the New South Wales decisions and held the stevedore fully liable. The facts in Wilson v. Darling Island Stevedoring & Lighterage Co.[196] were considered indistinguishable from Waters,[197] so the controversy focused almost exclusively on a stevedore's entitlement to the protection of the limitation clauses in the bill of lading. The author of the principal judgment distinguished Elder, Dempster and argued, on the basis of the third-party beneficiary rule, "that the [stevedore] is not a party to the contract evidenced by the bill of lading, that it can neither sue nor be sued on that contract, and that nothing in a contract between two other persons can relieve it from the consequences of a tortuous act committed by it against the plaintiff."[198] He also considered the "curious, and seemingly irresistible, anxiety to save grossly negligent people from the normal consequences of their negligence"[199] to be a development entirely inconsistent with normal common law trends.[200] He preferred "the modern tendency ... to expand the field of liability in tort."[20l]

The dissenters also considered the case to be governed by the proper interpretation of Elder, Dempster [202] and fundamental common law principles.[203] They concluded that Elder, Dempster protected the stevedore on a bailment rationale.[204]

Wilson v. Darling Island Stevedoring left Australian law in an uneasy state. A three-member majority of the High Court ruled against the stevedores in the strongest possible terms. The leading judgment, in particular, treated the bill of lading as irrelevant and [page 764] left no hope for protection under a suitably drafted clause.[205] Two judges, on the other hand, dissented in equally strong terms. They were willing to extend the carrier's limitations to the stevedores automatically. The extreme differences between the two sides arose from different views of Elder, Dempster (a case in which the Hague Rules did not apply), common law principles of agency and bailment, and fundamental contract technicalities (including the third-party beneficiary rule), with little regard for the Hague Rules or their international character.

After the Privy Council's Eurymedon decision,[206] the High Court of Australia reconsidered the issue in The New York Star.[207] The disagreement was carried forward to a new generation of judges who were as deeply divided as their predecessors. The stevedore claimed the benefit of the carrier's time limitation in a suit by the consignee of goods that had been stolen through the stevedore's negligence. Every judge in the case accepted Lord Reid's Scruttons suggestions, at least in theory, but the interpretations varied widely.

The chief concern at trial had been whether the carrier had the stevedore's authority to contract as its agent; the consignee conceded that the consideration requirement was satisfied.[208] The trial court found the necessary agency, applied the Eurymedon doctrine, and held for the stevedore. On appeal, the consignee [page 765] challenged the adequacy of the consideration.[209] The New South Wales Court of Appeal, distinguishing The Eurymedon, held unanimously that there had not been adequate consideration because the stevedore's performance had not been in response to any offer from the shipper.[210] Thus the stevedore was fully liable.

The High Court of Australia affirmed the decision on different grounds,[211] but discussed in detail whether the stevedore could rely on the limitation clause. Three members of the court, although disagreeing among themselves on the precise analysis,[212] agreed that the carrier's limitations protected the stevedore. The dissenting judges on this issue did not simply differ on the application of the Eurymedon principles, but argued that those principles were fundamentally wrong. Justice Stephen explicitly agreed with "those of their Lordships who dissented [in The Eurymedon] and with the members of the New Zealand Court of Appeal."[213] Justice Murphy declared his willingness to accept the "undesirable technicalities"[214] required by The Eurymedon if "the adoption of such a theory ... would serve Australia's interests,"[215] but in the belief that Australia was better served by a rule protecting cargo interests he concluded "that a contract should not be conjured up out of the [page 766] circumstances in order to extend the exemptions and immunities under the bill of lading to the stevedore."[216]

It is difficult to predict with confidence what path Australian law will take next. The High Court has upheld the stevedore's position, but only in dicta, and only by a three-to-two vote in which those favoring The Eurymedon disagreed among themselves on the proper analysis. The New South Wales Court of Appeal has conscientiously applied the Privy Council's New York Star decision to protect stevedores.[217] Given the High Court's greater freedom to depart from Privy Council decisions,[218] and the abolition of appeals to the Privy Council,[219] there remains a real possibility that cargo interests may yet prevail in Australia.[220] The close similarity between the domestic laws of England and Australia has meant that the applications of the Hague Rules have been similar in the two countries. But to the extent that Australia applies common law doctrines, such as the third-party beneficiary rule, differently than does the mother country, there are also differences in the interpretation of the Hague Rules.

D. Himalaya Clauses in Canada

The impact of domestic law on the interpretation of the Hague Rules is particularly striking in Canada,[221] where there are two systems of domestic law. The common law courts followed the English third-party beneficiary rule and the English decisions applying it. [page 767] In Quebec, on the other hand, the rule does not apply. The Quebec courts extended the protection of the Hague Rules to stevedores, but in a manner that reconciled them with the Quebec Civil Code and civil law contract principles.

The status of the Himalaya clause under Canadian common law changed with developments in English law. Thus when English courts were unwilling to consider the stevedore's claims, the common law courts in Canada applied the third-party beneficiary rule rigorously. The earliest Canadian Supreme Court decision on the subject rejected the stevedore's claim solely on the authority of the House of Lords decision in Scruttons.[222]

After the Privy Council's indecisive decision in The Eurymedon,[223] the common law courts in Canada accepted the Himalaya clause in principle, but continued to apply a strict view of the third-party beneficiary rule.[224] In a British Columbia case,[225] for example, the stevedore claimed the benefit of a very broad Himalaya clause that appeared to cover all of the common problem areas.[226] The court, however, distinguished The Eurymedon on the ground that the carrier there had been a subsidiary of the stevedore with admitted authority to contract as agent for the stevedore; the court ruled that in this case the carrier lacked such authority.[227] It therefore followed Scruttons and held that the stevedore could not rely on the limitation clause. [page 768]

After the strong endorsement of the stevedore's claim in The New York Star, the Canadian common law courts again followed the English lead. The Supreme Court's most recent decision essentially adopts the Privy Council's reasoning to uphold the stevedore's claim.[228]

The Quebec courts have consistently recognized the stevedore's right to claim the carrier's limitations -- but the cases have turned on the application of Quebec law. Shortly after The Eurymedon, for example, the Quebec Court of Appeal announced that it would recognize the Himalaya clause,[229] but held that the stevedore before it had been guilty of gross negligence, and thus was unable to claim the protection of an exoneration clause under Quebec law.[230] The following year the Quebec Superior Court relied heavily on domestic law to rule for the stevedore.[231] The court based its decision primarily on Article 1029 of the Quebec Civil Code,[232] which explicitly permits contracts for the benefit of third parties.[233] Indeed, the court distinguished a Canadian Supreme Court decision on the ground that the earlier case had arisen in Nova Scotia, where common law principles apply.[234]

The influence of domestic law on the application of the Hague Rules in Canada was dramatically revealed in the recent Miida Electronics litigation. The trial court, relying on Quebec law, dismissed the action against the stevedore.[235] It concluded that the [page 769] Himalaya clause was sufficient under Quebec law to bind the consignee to the limitation clause in the stevedore's contract with the carrier.[236] The Court of Appeal, deciding the case under Canadian common law, reversed the trial court on this issue and found the stevedore fully liable.[237] Although the court admitted the validity of a Himalaya clause in principle,[238] its restrictive interpretation of common law doctrine would have made the clause worthless to a stevedore in practice.[239] Five years later, after English law had clearly shifted in the stevedore's favor, the Supreme Court reversed the appellate court on this issue and restored the original judgment.[240] The principal controversy was whether Quebec law or Canadian maritime law (i.e., common law) governed; once the majority decided to apply common law, it resolved the Himalaya clause issue by adopting the English jurisprudence on the subject.

The Canadian courts eventually concluded that a stevedore may generally rely on a Himalaya clause, but there were two very different bases for this conclusion. The common law courts followed the English decisions through a complicated body of technical doctrine. The Quebec courts relied on the Civil Code and civil law methods of contract analysis. The explanation for these systematic differences within a single country can be found in the differences between the substantive laws of Quebec and those of the rest of Canada. Not only do the Canadian results differ from those in [page 770] other countries, but the impact of domestic law constraints within Canada itself is obvious.

E. Himalaya Clauses in West Germany [241]

Under West German law,[242] the validity of a Himalaya clause is not an issue. A. stevedore can undoubtedly rely on a bill of lading clause extending the carrier's limitations of liability to third parties.[243] Section 328 of the German Civil Code explicitly permits contracts for the benefit of a third party,[244] so the obstacle that the English third-party beneficiary rule creates in Commonwealth countries simply does not exist.[245] The shipper and carrier are free to extend the benefits of the contractual regime to the stevedore without regard for common law doctrines such as consideration.[246]

The more interesting question under West German law is whether a third party can claim the benefit of the carrier's limitations on liability even when there is not a Himalaya clause.[247] The prevailing opinion is that this is indeed possible.[248]

The early cases in which a third party sought the benefit of a [page 771] carrier's limitations involved fact patterns analogous to The Himalaya [249] itself: a cargo owner, unable to collect full damages from the carrier, sued the master of the vessel.[250] In the leading case, decided in 1960, the Bundesgerichtshof (West Germany's highest court) held that the parties to a contract could exclude the future delictual [251] liability of a third party by implication.[252] In concluding that the shipper had implicitly intended to extend the exclusion to the master, the court considered several factors: The shipper knew that people such as the master would perform the contract. The shipper's interest, in any event, was the same no matter who performed the contract.[253] Furthermore, the purpose of the exclusionary clause was to establish the carrier's risk, which was a factor in setting the freight rate. A risk of higher liability would result in a higher freight rate. Because the purpose of the clause rather than its wording was the material factor (at least under German law), the court would not allow the shipper to evade the exclusion after paying the lower rate.[254] In a similar situation a decade later, the Bundesgerichtshof treated the master's entitlement to the carrier's limitation of liability as an established principle requiring no elaboration.[255]

More recent decisions extend the principle beyond the carrier's employees. In a 1977 case, for example, the original carrier subcontracted with a second carrier. The shipper then sued the second carrier, who claimed the right to limit its liability. The Bundesgerichtshof, citing its 1960 judgment, held that a third party's protection against a contracting party can be either explicit or implicit, and can include reliance on limitations of liability.[256] Such protection [page 772] is implicit where appropriate and reasonable.[257] The court explained that a carrier has an interest in limiting its liability and should be able to offer the shipper lower freight rates in return. Furthermore, shippers prefer to obtain their insurance elsewhere so that they can have faster and safer claims. When a carrier uses a subcontractor, the court argued, the new rate should be based on the original rate. The same economic analysis would still apply. The subcontractor, therefore, may limit its liability under the terms of the original contract.[258]

Although theoretical issues remain,[259] it is clear that the West German courts will enforce an explicit Himalaya clause. Moreover, they are very likely to grant the benefits of a Himalaya clause even if the bill of lading does not contain one.[260] German courts are thus willing to grant stevedores rights that United States and Commonwealth courts have explicitly denied them. The explanation for this stark contrast in the application of the Hague Rules can be traced to differences in domestic law. While German courts apply section 328 of the Civil Code,[261] the Commonwealth Courts are bound by the English third-party beneficiary rule.[262] While the Bundesgerichtshof applies the principle that an exclusion clause's purpose, rather than its terms, is material,[263] United States courts repeat the common law rule that exclusion clauses are strictly construed against the party claiming under them.[264]

F. Summary

In each case discussed here, the national court applied substantially the same version of the Hague Rules. There were some variations [page 773] in detail concerning the precise extent of protection available to the carrier,[265] but there were no textual variations that even arguably concerned the decision to extend the carrier's protections to an independent contractor. There is nevertheless a wide range of interpretations on the issue. The problem is due in large measure to national courts' attempts to reconcile the Hague Rules with domestic law that varies from country to country.

In the United States, the courts have reconciled the Hague Rules with domestic agency principles and the doctrine permitting contracts for the benefit of third parties. In England there is a very different third-party beneficiary rule. Applying their domestic agency and contract principles, the English courts developed methods to uphold the Himalaya clause, but only via a tortuous trail. Australia has tended to apply a more restrictive third-party beneficiary rule; thus its courts have been less willing to recognize the Himalaya clause. The Canadian courts have tended to follow the English decisions -- except in Quebec, where the common law does not apply and the Himalaya clause jurisprudence is completely different. Like Quebec, Germany recognizes contracts for the benefit of third parties and applies more liberal rules of contract construction. Because of these differences in domestic law, the German courts have inferred enforceable Himalaya clauses even when the bills of lading did not contain them.

Each national court faced a different set of domestic law constraints and applied the Hague Rules to conform with the constraints. The result has been varying interpretations of an international uniform law according to variations in the substance of domestic law.

III. INTERNATIONAL CONFLICTS ON THE VALIDITY OF CHOICE OF FORUM CLAUSES UNDER THE HAGUE RULES

Perceiving the conflicts among national courts in the interpretation of the Hague Rules, many carriers have sought to remedy the problem by inserting choice of forum clauses [266] in their bills of lading. [page 774] A Dutch carrier, for example, might provide that "[a]ll actions under the present contract of carriage shall be brought before the Court of Amsterdam and no other court shall have jurisdiction with regard to any such action."[267] In theory this provision helps the parties to predict the outcome of litigation because they know that a Dutch court will interpret the Rules. It also gives the carrier the benefit of a convenient local forum. And to the extent that Amsterdam is inconvenient for potential plaintiffs, the carrier also avoids claims that are not worth pursuing if they must be filed in the Netherlands.

The inclusion of a choice of forum clause in a bill of lading, however, raises a difficult question: do the Hague Rules permit the carrier to stipulate the forum? A national court must resolve this issue when a claimant, notwithstanding the choice of forum clause, brings an action outside the specified forum, and the carrier claims the clause's protection.[268] Because the Hague Rules contain no explicit provision governing the issue,[269] there is an obvious potential for conflicts among national courts.

Some courts have found Article 3(8), which exists to prevent carriers from undermining the essential compromise of the Hague Rules,[270] to be relevant in determining the validity of a choice of forum clause. It provides:

As the second sentence indicates, Article 3(8) invalidates not only direct but also indirect attempts to lessen the carrier's liability. The issue, therefore, is whether a choice of forum clause creates a sufficient obstacle to recovery that it should be regarded as reducing the carrier's liability in violation of this provision.

American courts hold that any choice of forum clause is an inherent violation of Article 3(8).[272] In England, Article 3(8) applies in extreme cases, but it does not create a per se rule.[273] Other countries determine the validity of a choice of forum clause without reference to Article 3(8). The result is a wide range of interpretations of a supposedly uniform law. The only constant factor has been the influence of sources independent of the Hague Rules on the process of interpretation.

A. Choice of Forum Clauses in Countries with Statutory Bans

The strongest reaction against choice of forum clauses is a complete and automatic prohibition. Although the Hague Rules do not take this course explicitly, some legislatures, in enacting the Hague Rules, have prohibited attempts to deny jurisdiction to the local courts.[274] The Australian COGSA, for example, provides that "[a]ny stipulation or agreement ... purporting to oust or lessen the jurisdiction of the Courts of the Commonwealth [of Australia] or of a State in respect of any bill of lading ... shall be illegal, null and void, and of no effect."[275] The High Court has held that this provision invalidates a choice of forum clause even when a case is governed by a foreign country's version of the Hague Rules.[276] Similar prohibitions exist in a number of other [page 776] countries,[277] including Lebanon,[278] New Zealand,[279] South Africa,[280] and Syria.[281] Thus in some countries an explicit provision in a national statute determines the validity of a choice of forum clause without reference to the Hague Rules. The domestic law constraint is absolute, and the courts do not even address the proper interpretation of Article 3(8).

B. Choice of Forum Clauses in the United States

Historically, the United States courts were among the least willing to recognize choice of forum clauses in any context.[282] Without even considering the possible applicability of the Hague Rules, the courts refused on general principle to recognize the forum chosen in a bill of lading.[283] By the 1940s, however, a more receptive attitude began to develop.[284] In 1972 the Supreme Court, in The Bremen v. Zapata Off-Shore Co.,[285] finally reversed the long tradition and declared that choice of forum clauses are presumptively valid.[286] Although the Bremen Court recognized that there were situations in which the presumption could be defeated [287] it firmly [page 777] rejected the "parochial concept that all disputes must be resolved ... in our courts."[288]

With the growing recognition that choice of forum clauses might succeed under general principles, American courts began to consider the impact of COGSA section 3(8), the domestic enactment of Hague Rules Article 3(8).[289] In Wm. H. Muller & Co. v. Swedish American Line,[290] a New York consignee brought suit in the Southern District of New York for damages to goods on board a Swedish ship that had been lost at sea. The defendant claimed the benefit of a bill of lading provision requiring "[a]ny claim against the carrier arising under this bill of lading [to] be decided ... in the Swedish courts."[291] The plaintiff responded that this clause lessened the carrier's liability within the meaning of section 3(8) because it forced a claimant "to undergo a substantial expense" to obtain a trial in Sweden.[292] With very little discussion of the point, the Second Circuit rejected the plaintiff's argument and concluded "that such possible expense, which is only incidental to the process of litigation," was insufficient "to bring this jurisdictional agreement within the ban of [section 3(8)]."[293] The court instead held that "the enforceability of such an agreement depends upon its reasonableness."[294] Here the clause "was not unreasonable" and would be enforced.[295] Thus the court addressed the Hague Rules issue, but ultimately decided that independent domestic principles governed the case.[296]

A number of district courts followed Muller,[297] but other courts [page 778] of appeals did not address the Article 3(8) issue. In Carbon Black Export. Inc. v. The S.S. Monrosa,[298] an American shipper brought in rem and in personam proceeding:) in Texas, despite a bill of lading clause calling for suit in Genoa." The district court, following Muller, declined jurisdiction.[299] On appeal, the Fifth Circuit distinguished Muller and reversed the district court. It concluded that the choice of forum clause, by its terms, did not apply to an in rem action.[300] It then decided that the district court should also have taken jurisdiction over the in personam aspect of the case under "the universally accepted rule that agreements in advance of controversy whose object is to oust the jurisdiction of the courts are contrary to public policy and will not be enforced."[301] The Carbon Black plaintiff argued that the choice of forum clause effectively lessened the carrier's liability and thus violated COGSA section 3(8),[302] but the court did not even mention the issue in its summary of the grounds for appeal.[303] In other words, the court decided the case on the general domestic principles recognized at the time without reference to section 3(8).[304]

The Second Circuit reconsidered the issue and overruled Muller [page 779] in 1967. Indussa Corp. v. S.S. Ranborg [305] involved a shipment from Antwerp to San Francisco on a Norwegian vessel under a bill of lading providing for disputes to "be decided in the country where the Carrier has his principal place of business."[306] When the consignee arrested the vessel and brought an in rem action in New York, the carrier claimed the protection of this provision. The district court, applying Muller, declined jurisdiction in order to permit a Norwegian court to decide the case.[307] The panel hearing the appeal, however, concluded that Muller was wrongly decided and sua sponte requested consideration of the case en banc.[308] The full court voted eight to one to overrule Muller on the ground that it was inconsistent with COGSA.[309]

The Indussa court found choice of forum clauses invalid under section 3(8) on several grounds. First, permitting a claimant to bring an action only in a distant forum has the practical effect of substantially lessening liability. "Such a clause puts 'a high hurdle' in the way of enforcing liability ... and thus is an effective means for carriers to secure settlements lower than if cargo could sue in a convenient forum."[310] Second, a foreign court might not apply COGSA or an equivalent enactment of the Hague Rules. Even when it would, the risk of inconsistent interpretations remains. The carrier's liability might be reduced if a foreign court applied different law than that of an American court.[311] Finally, the Muller rule had proved unworkable. Because section 3(8) forbids any lessening of the carrier's liability, Muller required district courts to predict how foreign courts would decide cases to ensure that there would be no reduction in a plaintiff's rights. Such predictions were difficult and uncertain, and the results were inconsistent.[312] The court concluded that section 3(8) protects plaintiffs from these risks. "[A]ny contractual provision ... that would prevent cargo able to obtain jurisdiction over a carrier in an American court from having that court entertain the suit" is void.[313] [page 780]

The Indussa court was not content to base its decision solely on section 3(8). COGSA's enacting clause [314] and section 13 [315] -- provisions with no equivalent in the Hague Rules -- both declare that COGSA shall govern every contract for the carriage of goods by sea. to or from ports of the United States in foreign trade. There is no way to ensure that a foreign court will apply the United States COGSA so broadly.[316] Indussa is a good illustration of the problem. The Norwegian courts could plausibly have applied American, Belgian, or Norwegian law.[317] The Second Circuit found deference to a court that might apply foreign law almost as objectionable under these two domestic COGSA provisions as applying foreign law directly.[318]

Since Indussa, American courts have consistently denied carriers the benefit of choice of forum clauses [319] in bills of lading that are subject by law [320] to the United States COGSA.[321] But in other [page 781] cases -- where a foreign enactment of the Hague Rules governs, for example, or where the carrier voluntarily incorporates the United States COGSA into the bill of lading so that it governs as a matter of contract -- courts have applied The Bremen to enforce choice of forum clauses without reference to Article 3(8) or corresponding foreign COGSA provisions.[322] If the Hague Rules were truly uniform international law, Article 3(8) and the identical section 3(8) of COGSA would operate in the same way. The courts' failure to recognize this suggests that they find only the domestic rationale [323] persuasive. The broader section 3(8) rationale, which should apply under any version of the Hague Rules, is insufficient to overcome independent domestic legal principles.

In sum, a thoroughly American doctrine on this issue has developed over the years. At first there was no distinction based on the applicability of COGSA section 3(8), so decisions were made on general principles without regard for the international concerns inherent in the subject. Historically this meant that choice of forum clauses were always ineffective, but as general attitudes changed the courts began to uphold clauses in bills of lading they found reasonable. Although since Indussa the courts have distinguished bill of lading cases, they continue to view the issue as a peculiarly American concern. They apply section 3(8) when the United States COGSA governs as a matter of law, but do not give the same reading to the substantially identical Article 3(8) of the Hague Rules.

C. Choice of Forum Clauses in England

English law is more sympathetic to a carrier's desire to impose [page 782] its own forum on the cargo interests through a clause in the bill of lading,[324] but even in England the carrier's choice is not enforced automatically. English courts have consistently rejected the Indussa argument.[325] They nevertheless decline to enforce choice of forum clauses in at least two situations. Under Article 3(8) of the Hague Rules,[326] English courts will strike down a clause when the carrier's liability would be reduced in the specific forum chosen under the circumstances of the particular case.[327] Under general principles of English law, without reference to the Hague Rules, they will also refuse to give effect to a clause when the chosen forum is considered less appropriate in the circumstances.[328] And even when a carrier successfully argues that a choice of forum clause should apply, its success is often limited to a conditional stay of the pending English action.[329]

The English courts first considered the validity of a choice of forum clause in a bill of lading under the Hague Rules in Maharani Woollen Mills Co. v. Anchor Line [330] shortly after the enactment of the British COGSA.[331] The consignee of goods damaged in shipment from Liverpool to Bombay brought suit in England despite a bill of lading clause requiring that "all claims ... shall be determined at the port of destination according to British laws."[332] The trial court dismissed the case on the ground that jurisdiction existed only in India. On appeal the plaintiffs argued that Article [page 783] 3(8) of the Hague Rules invalidated the clause because it reduced the carrier's liability. Lord Justice Scrutton, the great commercial judge of his generation, rejected the argument in three sentences:

The court affirmed the dismissal of the action, effectively denying the plaintiff any remedy.[334]

Maharani may be explained in part by the fact that in 1927 the Bombay courts were British,[335] in part by the belief that those courts would have decided the case under British law, and in part by the view that Bombay was the more appropriate forum under the circumstances. The rationale for the summary rejection of the argument that the Second Circuit later adopted in Indussa, however, was based on a formalistic view distinguishing substance from procedure. Although such a distinction is recognized to some extent in many countries,[336] English courts were much readier to apply it in a manner having significant legal consequences.[337] In [page 784] recent years the House of Lords has occasionally taken a less formal approach to the substance/procedure distinction,[338] but no English court has reconsidered the argument that a choice of forum clause is an inherent violation of Article 3(8).[339]

The English rejection of the Indussa argument has not resulted in the automatic enforcement of all choice of forum clauses, or even in the conclusion that Article 3(8) of the Hague Rules is irrelevant to the issue. The decision whether to recognize a choice of forum clause is at least subject to general domestic law principles. Moreover, when the carrier seeks to remove the action to a jurisdiction where the courts will not apply the British version of the Hague Rules and the carrier's liability will be lessened as a result, the choice of forum clause is unenforceable under Article 3(8).

Two recent cases, The Hollandia [340] and The Benarty,[341] illustrate both the scope of Article 3(8) in England and the major influence of domestic law. In The Hollandia a British company shipped cargo on a Dutch vessel for ultimate delivery in the Netherlands Antilles. The bill of lading gave exclusive jurisdiction to the Court of Amsterdam.[342] When the cargo was damaged, the shipper brought suit in England, where the COGSA limitation on the carrier's liability was approximately £11,1500,[343] instead of in Amsterdam, where the limitation would have been about £250.[344] The House of Lords ruled that Article 3(8) [345] rendered the choice of [page 785] forum clause null and void as applied in this dispute, although it might have been valid in a case where it did not have the effect of reducing the carrier's liability in a manner inconsistent with the British COGSA.[346]

In The Benarty various cargo owners filed suit in England despite bill of lading clauses requiring all actions to be brought in Indonesia. The Indonesian charterer" relying on the choice of forum clauses, moved for a stay. The court accepted that Article 4(5) of the Hague-Visby Rules, which all English court would apply, permitted the carrier to limit its liability on the principal claim to £217,800; that Article 4(5) of the Hague Rules, which an Indonesian court would apply, permitted a £4,000 limitation; and that Article 474 of the Indonesian Commercial Code [347] allowed a limitation (based on the tonnage of the vessel) of $1,200 for all claims.[348] The charterer waived its right to rely on the Hague Rules limitation and agreed to accept the much higher limitation of the Hague-Visby Rules,[349] thus bringing itself outside the precise holding of The Hollandia. The charterer nevertheless retained its more valuable right to rely on the Indonesian tonnage limitation. The court ruled that Article 8 of the Hague Rules [350] permitted this, and thus Article 3(8) did not prohibit it [351] The ultimate result, therefore, was the enforcement of a choice of forum clause that effectively reduced the plaintiff's potential recovery by over 99% and made further proceedings "not worth the effort."[352]

Once a choice of forum clause survives scrutiny under the British COGSA, the decision whether to recognize the chosen forum and grant a stay remains within the court's discretion under general principles independent of the Hague Rules.[353] There is a [page 786] presumption favoring a stay unless the plaintiff can show "strong cause" to deny it. Although English plaintiffs have not been able to meet this burden automatically,[354] in over half the reported cases the court has refused to enforce the choice of forum clause.[355]

In sum, an English court ruling on the validity of a choice of forum clause in a bill of lading will apply Article 3(8) only in extreme cases. Generally a court's decision will be based on domestic principles of English law [356] that have nothing to do with the Hague Rules or considerations of international uniformity.

D. Choice of Forum Clauses in Belgium [357]

The Belgian courts,[358] applying a broad view of public policy [page 787] quite unlike that recognized under the common law,[359] construe Article 3(8) of the Belgian enactment of the Hague Rules [360] to protect Belgium's national interest in the application of its own law rather than to protect any specific rights of cargo owners.[361] Thus the courts defer to a foreign jurisdiction specified in the bill of lading when it is sufficiently certain that the foreign court will apply Belgian law, even if the result is to require suit in a distant country.[362] But they will not defer to a court in a neighboring country that might apply a different version of the Hague Rules, even if there is no suggestion that the interpretation applied would differ from that of a Belgian court.[363]

It is clear that the Belgian courts find this national interest only in their domestic law, and that they do not view the Hague Rules as creating any paramount interests in favor of the countries involved in a particular shipment. In The Grecian Med,[364] for example, the court upheld a Belgian choice of law and choice of forum clause even though the case concerned a shipment from the Netherlands to Greece.

The principal legal issue is how certain the Belgian court must be that the chosen forum will apply "Belgian law before it will give [page 788] effect to a choice of forum clause. Despite some early inconsistency,[365] the basic standard is well established. The burden is on the carrier [366] to prove with reasonable certainty [367] that the chosen forum will apply the Belgian version of the Hague Rules as construed by the Belgian courts.[368]

The application of the standard has changed over the years as Belgian courts have become more suspicious of foreign decisions. In the 1950s and 19605, they would generally uphold a choice of forum clause if the bill of lading also contained a paramount clause requiring the application of Belgian law.[369] In the early 1970s, however, courts in West Germany [370] and Holland [371] declined to apply [page 789] the Belgian version of the Hague Rules despite such clauses paramount. Thereafter, the Belgian courts regularly denied effect to forum clauses without regard to the existence of a clause paramount,[372] often citing the offending German [373] and Dutch [374] cases to justify the action.

The practical result is that a Belgian court is unlikely to enforce a choice of forum clause,[375] although in theory it will judge each case on its own facts. In any event, the Belgian interpretation of Article 3(8) is based on Belgian law, and not on a generally applicable construction of the Hague Rules.

E. Choice of Forum Clauses in Pakistan

Pakistani law on the interpretation of the Hague Rules is, for [page 790] the most part, based on English doctrines.[376] Indeed, the Pakistani COGSA [377] was enacted when the country was still part of British India. The enforceability of choice of forum clauses in bills of lading, however, has turned on the interpretation of an independent provision in the Pakistani Contract Act.[378]

By the late 1960s there was a long and consistent line of cases in East Pakistan [379] holding that a bill of lading could validly specify a competent forum to the exclusion of the local courts.[380] Narayanganj Iron Works Ltd. v. Pakistan [381] is typical. The plaintiff brought suit in Dacca for the short delivery of goods shipped on an English vessel from Liverpool to Chittagong under a bill of lading that required disputes to be "determined in England ... to the exclusion of the jurisdiction of the Courts of any other country."[382] When the carrier claimed the benefit of the choice of forum clause, the plaintiff responded that the clause violated section 28 of the Contract Act.[383] Section 28, subject to two exceptions for arbitration cases, invalidates agreements "by which any party thereto is restricted absolutely from enforcing his rights under or in respect [page 791] of any contract by the usual legal proceedings in the ordinary tribunals."[384] The High Court concluded, in keeping with the line of similar cases,[385] that the clause at issue was consistent with the statute because it did not prohibit either party from enforcing its rights in an ordinary tribunal; it merely specified which one of two competent courts must be used.[386] There was no suggestion that Article 3(8) of the Hague Rules might apply.

The courts in West Pakistan, meanwhile, had adopted a different view of section 28,[387] and in Chowdhury v. Mitsui O.S.K. Lines,[388] the Pakistani Supreme Court resolved the conflict. The court began its analysis with a review of the English authorities.[389] It also discussed the Hague Rules, noting that they do not mention choice of forum clauses and that the practices of countries that have adopted the Hague Rules vary.[390] The court only alluded to Article 3(8),[391] and apparently gave no thought to the Indussa argument.[392]

The Chowdhury court ultimately focused on the impact of the local Contract Act on general English principles.[393] It decided that [page 792] choice of forum clauses in bills of lading violate the general rule of section 28, but that the two statutory exceptions [394] save them because "foreign jurisdiction clauses ... are really in the nature of arbitration clauses."[395] A court thus has the discretion to decide whether to stay an action to allow their case to be tried in the chosen forum. In exercising this discretion, Pakistani courts should be guided by general English principles with one exception: as section 28 expresses a general public policy against choice of forum clauses, the burden is on the carrier to satisfy the court that a stay is appropriate.[396]

The law in Pakistan is substantially the same as the law in England, save that the basic presumption on the validity of the clause is reversed. The difference is frequently one of form rather than substance; the noteworthy point is that the difference results from national legislation independent of the Hague Rules.

F. Choice of Forum Clauses in the E.E.C.[397]

Article 17 of the Brussels Judgments Convention [398] may preempt national law analysis of the validity of choice of forum clauses for courts in the member states of the European Economic Community.[399] As originally enacted, Article 17 required enforcement [page 793] of a clause if (1) the parties to a transaction had chosen a forum in an E.E.C. member state "by agreement in writing or by an oral agreement evidenced in writing" and (2) one of the parties was domiciled in a member state.[400] The amended version is even broader.

The Court of Justice of the European Communities recently construed the original Article 17 in The Tilly Russ,[40l] a case involving a bill of lading subject to the Hague Rules. An American firm shipped a consignment of timber from Toronto to Antwerp on a German vessel under a standard-form bill of lading providing that "[a]ny dispute ... shall be decided, by the Hamburg court."[402] The consignee nevertheless filed suit in the Commercial Court of Antwerp, which, in keeping with established Belgian principles,[403] accepted jurisdiction.[404] On appeal, however, the Belgian Court of Cassation asked the European Court of Justice for a preliminary ruling [405] on whether a bill of lading could be considered an "agreement in writing" or an "agreement evidenced in writing" within the meaning of the Convention.

The Court of Justice was uncertain whether Article 17 applied on the facts before it,[406] but it ruled that a choice of forum clause in the standard terms and conditions of a bill of lading could satisfy Article 17's requirements in three situations: (1) if both parties [page 794] had expressed their written agreement to the conditions containing the clause; (2) if the clause had been the subject of a previous oral agreement expressly referring to the clause; or (3) if the bill of lading were within the framework of a continuous business relationship governed by the general conditions containing the clause.[407] As a practical matter, the first two possibilities seem less significant. The shipper often does not sign the bill of lading, and discussion of the boilerplate provisions is rare indeed. The third situation seems fairly common, however, and the effect is likely to be the application of Article 17 whenever there is a regular course of dealing between shipper and carrier. The court did not mention Article 3(8), despite the fact that it is in force in every E.E.C. member state except land-locked Luxembourg.[408]

The amended version of Article 17 [409] is even wider, for coverage is now extended to cases where

The final clause presumably will swallow most of the field. The vast majority of Hague Rules cases arise in "international trade or commerce," a disputed choice of forum clause is almost invariably in a form that "accords with practices in that trade or commerce," and it is hard to imagine an acceptable excuse for the parties not to have been aware of those practices.

Article 17 will not completely displace national law and the Hague Rules in E.E.C. member states. There will still be cases in which neither of the parties is domicile in a member state or the [page 795] chosen forum is not in a member state.[411] But it appears that in a large proportion of the cases, courts will be forced to reconcile Article 3(8) of the Hague Rules with Article 17 of the Brussels Judgments Convention. Thus law independent of the Hague Rules will determine how national courts apply the Hague Rules.

G. Summary

Although each country discussed here has enacted Article 3(8) of the Hague Rules in substantially identical terms, there is a Wide range of interpretations on the validity of a choice of forum clause. As with conflicts on the validity of the Himalaya clause, the problem arises when national courts attempt to reconcile the Hague Rules with domestic law that varies from country to country.

In several countries, explicit provisions added to the domestic statute implementing the Hague Rules resolve the issue against choice of forum clauses. In the United States, the courts interpret two provisions that Congress added to COGSA to have the same effect. The Belgian courts apply Article 3(8) and a broad concept of public policy to achieve the same practical result. On the opposite end of the spectrum, an E.E.C. convention may often resolve the issue in favor of choice of forum clauses without considering Article 3(8). English courts rarely consider Article 3(8), but decide the issue under general principles -- including a presumption favoring enforceability. Pakistan also relies on general principles, but because of an independent statute, applies a presumption against enforceability.

Each national court faces a different set of domestic law constraints, which may assume any of a variety of forms. But in each case the national courts have relied heavily on the substance of domestic law to determine how they will apply the Hague Rules.

IV. CONCLUSION

A detailed examination of national court decisions applying the Hague Rules in cases involving Himalaya clauses and choice of [page 796] forum clauses illustrates the central role that domestic law constraints play in the interpretation of international uniform laws. National courts resolving ambiguous provisions have invariably attempted to reconcile the uniform law with their own domestic legal doctrines. In the process, they have interpreted the uniform law to make it consistent with independent domestic law. Because domestic law varies from country to country, conflicts in interpretation have arisen.

Consider, for example, the position of a national court deciding whether a stevedore may claim the benefit of a Himalaya clause. The court faces a wide range of permissible interpretations of the Hague Rules:

In the abstract, any of these interpretations might be plausible. The national court, however, faces not only this range of permissible interpretations of the Hague Rules, but also an entire body of domestic law. Some potential interpretations of the Hague Rules might conflict with domestic law, other potential interpretations could be to some extent inconsistent or create tensions with domestic law, and still other interpretations would be entirely consistent with domestic law.[420] A German court, for example, might find that the second and third alternatives are inconsistent or create some tension with the principle that an exclusion clause's purpose, rather than its terms, is material.[421] And the fourth and fifth alternatives are inconsistent with section 328 of the German Civil Code.[422] It therefore chooses the first alternative.

Conflicts arise when other national courts face different bodies of domestic law that constrain them in different ways. A United States court might find the first alternative inconsistent with the rule that an agent is liable for its own negligence [423] and the fifth alternative inconsistent with the doctrine permitting contracts for the benefit of a third party.[424] The second alternative might create some tension with tort and agency rules, while the fourth alternative creates some tension with domestic contract doctrine. It therefore [page 798] chooses the third alternative. A Commonwealth court might find the first, second, and third alternatives inconsistent with the English third-party beneficiary rule. The court therefore chooses the fourth or fifth alternative, depending on its domestic agency rules.

This example, a simplified summary of the analysis in Part II of the Article, shows how international conflicts arise because of the universal judicial attitude toward domestic law. Each national court faces the same theoretical range of permissible interpretations, but domestic law constraints push the German court to the first alternative, the United States court to the third alternative, and the Commonwealth court to the fourth or firth alternative.

An examination of the decisions on choice of forum clauses yields similar results. Consider the position of a national court deciding whether a carrier may claim the benefit of a choice of forum clause. Once again, the court faces a wide range of permissible interpretations of the Hague Rules:

Here, too, national courts face a wide variety of domestic law constraints that affect their choice of permissible interpretations of the Hague Rules. Some national courts face explicit statutory provisions that effectively compel the fifth alternative.[432] United States courts face statutory provisions that create tension with all but the fifth alternative.[433] For Belgian courts, their domestic doctrine of public policy creates tension with all but the fourth and fifth alternatives.[434] In England, general common law principles on choice of forum clauses are inconsistent with the first and fifth alternatives, and create tension with the fourth.[435] These same principles apply in Pakistan, but an independent statute imposes an additional constraint.[436] Similarly, the Brussels Judgments Convention imposes an additional constraint in E.E.C. member states.[437] Once again, differing domestic law constraints drive national courts into conflicting interpretations of international uniform law.

The influence of substantive domestic legal doctrine does not completely explain every conflict in the interpretation of international uniform law in national courts. Random distribution,[438] parochial ignorance of international concerns, varying methods of enacting a uniform law convention or interpreting a treaty, differing connotations of technical legal terms, and broad differences in legal tradition [439] undoubtedly play their parts. But domestic legal doctrine plays a major role in explaining why systematic conflicts have arisen among national courts, and prior scholarship has failed to recognize its significance.

Recognizing the impact of domestic law constraints is only the first step toward a solution to the problem of conflicts in the interpretation of international uniform law. The next step is to consider [page 800] measures to reduce the divisive influence of domestic law in this context, for as long as the influence continues, international uniform law will not achieve its full potential. The next step is for national courts to become aware of the dangers of interpreting international uniform law according to domestic legal doctrine. Many conflicts could undoubtedly be avoided if courts were more sensitive to the problem, and made the extra effort required to discover the international meaning of a convention. Counsel arguing a case governed by an international uniform law should likewise investigate the international understanding of the law, and take responsibility for calling this information to the court's attention.

Perhaps the identification and explanation of the problem and the role of domestic law in causing it will begin a process of increasing judicial awareness so that no more conflicts in interpretation will arise. A more realistic prediction is that the world's judges will continue to allow domestic legal doctrine to shape their interpretations of international, uniform law. Even when judges are sensitive to the problem, the difficulty of deciding cases without being unduly influenced by well-developed professional assumptions is probably too great to overcome.

If this latter prediction is correct, the next step toward a solution should be the consideration of institutional measures to conflicts in the interpretation of international uniform law. The range of possibilities is wide. Some measures, such as better developed choice of law rules, do little to avoid conflicts, but enable parties to deal with conflicts that do arise. Other measures, such as increasing the availability of international materials in the subject areas of international uniform laws, will better enable courts to avoid the problem outlined here, but they still must trust domestic judges to avoid the influences of domestic law. More radical measures, such as Black's proposal to create an "international court of appeals" with discretionary jurisdiction over national court decisions construing the Hague Rules,[440] go directly to the heart of the problem, but may create greater difficulties in the process.

A full discussion of possible solutions is beyond the scope of the present Article. A subsequent Article will consider these suggestions, and others like them, and propose a new mechanism to resolve conflicting interpretations of international uniform law. In [page 801] the meantime, it is essential to understand the major influence of domestic legal doctrine in causing these conflicts of interpretation among national courts. Any proposal to correct the problem must take account of its true causes. [page 802]

FOOTNOTES

* Assistant Professor of Law and Fellow to John Mark McLaughlin Centennial Research Professorship, University of Texas Law School; B.A., J.D., Vale; B.A. (Jurisprudence), M.A., Oxford. A Summer Reach Award from the University Research Institute, University of Texas at Austin. supported part of the work on this Article. Hans W. Baade, Julius J. Getman, John O. Honnold, Douglas Laycock, David W. Robertson, David J. Seipp, Louisa Weinberg, and Jay L. Westbrook read earlier drafts of this Article, and provided useful advice and criticism. Karen Patton Bogle, Carolyn Hurt, and Pamela Denney Nielson provided useful research assistance.

1. See, e.g., Goldring, "Unification and Harmonisation" of the Rules of Law, 9 Fed. L. Rev. 284, 309-22 (1978) (discussing Australian experience); Leach, The Uniform Law Movement in Australia, 12 Am. J. Comp. L. 206 (1963) (same); MacTavish, Uniformity of Legislation in Canada -- An Outline, 25 Can. B. Rev. 36 (1947) (discussing Canadian experience); Palmer, Federalism and Uniformity of Laws: The Canadian Experience, 30 Law & Contemp. Probs. 250 (1965) (same).

2. The Conference of Commissioners recommends over 90 "uniform" acts for adoption by an jurisdictions, 1983 Handbook of the National Conference of Commissioners on Uniform State Laws and Proceedings of the Annual Conference 282-85 [hereinafter Handbook]. In addition, the Conference has promulgated almost 20 "model" acts. Id. at 285-86. The Conference has withdrawn over 125 acts that are now obsolete or that have been superseded. Id. at 287-91.

3. Other generally enacted uniform laws cover a wide range of subjects. E.g., Unif. Anatomical Gift Act (1968); Unif. Child Custody Jurisdiction Act (1968); Unif. Reciprocal Enforcement of Support Act (1968); Unif. Gifts to Minors Act (1966); Unif. Testamentary Additions to Trusts Act (1960); Unif. Arbitration Act (1956); Unif. Declaratory Judgments Act (1922); Unif. Partnership Act (1914); see Handbook, supra note 2, at 282-85.

4. International uniform law is particularly common for international transportation. E.g., Convention Concerning International Carriage by Rail (COTIF), May 9, 1980, 1987 Gr. Brit. T.S. No. 1 (Cm. 41); Convention on the Contract for the International Carriage of Goods by Road (CMR), May 19, 1956, 399 U.N.T.S. 189 [hereinafter CMR]; Warsaw Convention for the Unification of Certain Rules Relating to International Transportation by Air, Oct. 12, 1929, 49 Stat. 3000, T.S. No. 876, 137 L.N.T.S. 11, reprinted in 49 U.S.C. § 1502 note (1982) [hereinafter Warsaw Convention]; Brussels Convention for the Unification of Certain Rules of Law Relating to Bills of Lading, Aug. 25, 1924, 51 Stat. 233, T.S. No. 931, 120 L.N.T.S. 155 [hereinafter Hague Rules]. International uniform law, however, covers a wide range of fields. E.g., Paris Convention for the Protection of Industrial Property, Mar. 20, 1883, as revised, July 14, 1967, 21 U.S.T. 1583, T.I.A.S. No. 6923, 828 U.N.T.S. 305: Convention on the Recognition and Enforcement of Foreign Arbitral Awards, June 10, 1958, 21 U.S.T. 2517, T.I.A.S. No. 6997, 330 U.N.T.S. 38; Universal Copyright Convention, Sept. 6, 1952, 6 U.S.T. 2731, T.I.A.S. No. 3324, 216 U.N.T.S. 132; Geneva Convention Providing a Uniform Law for Cheques, Mar. 19, 1931, 143 L.N.T.S. 357. The Vienna Sales Convention will govern international sales contracts beginning January 1, 1988. United Nations Convention on Contracts for the International Sale of Goods, Apr. 11, 1980, 52 Fed. Reg. 6264, reprinted in 19 I.L.M. 671.

5. Although in the United States a self-executing treaty may seem to be the prototypical international uniform law, this is not always the case. Some uniform laws are intended to be "model laws" that nations may enact, but that are not part of an international convention. See Droz, The Hague Conference on Private International Law -- Ninth Session: Report of the Work of the Small Committee in the Matter of Uniform Laws, 9 Am. J. Comp. L. 592, 592 (1960). Some nations (such as England and the commonwealth countries) do not recognize self-executing treaties, but require implementing legislation. E.g., Attorney-Gen. for Can. v. Attorney-Gen. for Ont., 1937 A.C. 326, 347-48 (P.C.) (Can.) (Lord Atkin). And a country may treat an international convention as a model law by enacting its substance as domestic legislation without ratifying the convention itself. See, e.g., infra note 24 and accompanying text.

6. The work of these organizations is well-documented. E.g., David, The International Institute of Rome for the Unification of Private Law, 8 Tul. L. Rev. 406 (1934); Farnsworth, UNCITRAL -- Why? What? How? When?, 20 Am. J. Comp. L. 314 (1972); Hornsey, The Unification and Harmonisation of Law in Europe, 22 N. Ir. L.Q. 301 (1971); Nadelmann, Conflicts between Regional and International Work on Unification of Rules of Choice of Law, 15 Harv. Int'l L.J. 213 (1974); Nadelmann, The United States Joins the Hague conference on Private International Law: A "History" with Comments, 30 Low & Contemp. Probs. 291 (1965); Pfund, International Unification of Private Law: A Report on United States Participation, 1985-86, 20 Int'l Law, 623 (1986).

7. E.g., Honnold, A Uniform Law for International Sales, 107 U. Pa. L. Rev. 299, 299 (1959); Yntema, Unification of the Laws Respecting Negotiable Instruments, 4 Int'l L.Q. 178, 178 (1951).

8. E.g., Honnold, supra note 7, at 302; Lando, The Contribution of Comparative Law to Law Reform by International Organizations, 25 Am. J. Comp. L. 641, 643 (1977).

9. E.g., Dunham, New Reasons For Uniformity, 16 U. Chi. L., Sch. Rec. 3, 26-27 (1968); Lawther, Uniform State Laws, 18 Tex L. Rev. 436, 437 (1940).

10. E.g., Yntema, supra note 7, at 178.

11. E.g., Handbook, supra note 2, at 220-21; Pound, Unification of Law, 20 A.B.A. J. 695 (1934); Wigmore, The International Assimilation of Law -- Its Needs and Its Possibilities From an American Standpoint, 10 Ill. L. Rev. 385 (1916).

12. E.g., Boal, Efforts to Achieve International Uniformity of Laws Relating to the Limitation of Shipowners' Liability, 53 Tul. L. Rev. 1277, 1295, 1297 (1979); Lorenz, Some Comparative Aspects of the European Unification of the Law of Products Liability, 60 Cornell L. Rev. 1005, 1006 (1975); Note, Uniform Rules for a Combined Transport Document in Light of the Proposed Revision of the Hague Rules, 20 Ariz. L. Rev. 953, 953 (1978).

13. This Article does not address textual differences, although they undeniably impede uniformity. When a jurisdiction enacts a "uniform" law with changes in language, the law is only partially uniform. This represents a political problem in reaching agreement rather than a judicial problem of interpretation. Several commentators have addressed the problem of inconsistent texts, both in the international and domestic contexts. E.g., Henson, The Problem of Uniformity, 20 Bus. Law. 689 (1965); Mann, Uniform Statutes in English Law, 99 Law Q. Rev. 376, 389-90 (1983); Note, The Uniform Commercial Code: Major Differences Between Massachusetts and Pennsylvania, 71 Harv. L. Rev. 674 (1958); Note, Admiralty -- International Uniformity and the Carriage of Goods by Sea, 60 Tul. L. Rev. 165, 174 -- 82 (1985); cf. infra note 82.

14. U.C.C. general comment (1978).

15. Scruttons Ltd. v. Midland Silicones Ltd., 1962 A.C. 446, 471 (1961).

16. E.g., TWA v. Franklin Mint Corp., 466 U.S. 243, 257-58 & n.31 (1984) (recognizing conflicts in interpretation of Warsaw Convention, supra note 4); General Motors Overseas Operation v. S.S. Goettingen, 225 F. Supp. 902, 904-05 (S.D.N.Y. 1964) (recognizing conflicts in interpretation of Hague Rules, supra note 4); The Thameshaven, Judgment of Apr. 15, 1975, Tribunal de commerce, Antwerp, 11 Eur. Transport L. 92, 100 (1976) (same); Chowdhury v. Mitsui O.S.K. Lines, 1970 Pak. Legal Dec. (Sup. Ct.) 373, 392-93, [1970] 2 Lloyd's Rep. 272, 282 (Pak.) (same); Ulster-Swift Ltd. v. Taunton Meat Haulage Ltd., [1977] I W.L.R. 625, 631 (C.A.) (recognizing conflicts in interpretation of CMR, supra note 4).

17. E.g., O.C. Giles, Uniform Commercial Law 21 (1970); H.C. Gutteridge, Comparative Law 101 (2d ed. 1949); Beutel, The Necessity of a New Technique of Interpreting the N.I.L. -- The Civil Law Analogy, 6 Tul. L. Rev. 1, 1-3 (1931); Hargest, Keeping the Uniform State Laws Uniform, 76 U. Pa. L. Rev. 178, 179-81 (1927); Mann, The Interpretation of Uniform Statutes, 62 Law Q. Rev. 278, 278 (1946); Munday, The Uniform Interpretation of International Conventions, 27 Int'l & Comp. L.Q. 450, 450, 454 (1978); Nadelmann, Uniform Interpretation of "Uniform" Law, 1959 Unidroit Y.B. 383, 385 (1960); Rossman, Uniformity of Law: An Elusive Goal, 36 A.B.A. J. 175, 178, 258-59 (1950); Sundberg, A Uniform Interpretation of Uniform Law, 10 Scandinavian Stud. L. 219, 221 (1966); Note, Uniform Construction of the Uniform Acts, 29 Harv. L. Rev. 541, 541-42 (1916); see also infra notes 44 & , 45.

18. See infra notes 52-67 and accompanying text.

19. Although this Article focuses on international conflicts, them is no reason why its analysis should not apply in a domestic context. The impact of local law is more obvious in the international context because substantive law differs more among nations than among states or provinces ,within a single nation. To the extent there are relevant international differences, however, it is predictable that they will tend to produce conflicts in the interpretation of domestic uniform law. The difficult task or testing this hypothesis must be left for future research.

20. Hague Rules, supra note 4. Two protocols (which the United States has not yet ratified) have amended the Hague Rules. Twenty countries have ratified or acceded to the Protocol to Amend the Hague Rules, Feb. 23, 1968, 1977 Gr. Brit. T.S. No. 83 (Cmnd. 6944) (entered into force June 23, 1977) [hereinafter Visby Protocol]. Ten countries have ratified or acceded to the Protocol Amending the Hague Rules, Dec. 21, 1979, 1984 Gr. Brit. T.S. No. 28 (Cmnd. 9197) (entered into force Feb. 14, 1984) [hereinafter Brussels Protocol]. Both protocols amend the, Hague Rules in matters of detail (albeit some important detail), not in any fundamental way. The Hague Rules as amended by the Visby Protocol are known as the Hague-Visby Rules. The Hague-Visby Rules are now in force in most of Western Europe. See, e.g., Carriage of Goods by Sea Act, 1971, ch. 19 (U.K.) [hereinafter British COGSA] (enacting Hague-Visby Rules), as amended by Merchant Shipping Act 1981, ch. 10, § 2 (U.K.) (implementing Brussels Protocol) (superseding Carriage or Goods by Sea Act, 1924, 14 & 15 Geo. 5, ch. 22 (U.K.) [hereinafter British 1924 COGSA] (enacting Hague Rules).

21. In some ways the Hague Rules are an extreme case, for a great many conflicts have arisen under them. They are nevertheless an appropriate object of study because they offer a wide range or possibilities. They have been the subject of judicial interpretation for over 60 years in almost every maritime country of the world. National courts have issued opinions explaining their application in literally thousands of cases on a wide variety of issues. Few uniform laws offer such a rich source of material for academic inquiry.

22. Carriage of Goods by Sea Act, ch. 229, 49 Stat. 1207 (1936) (codified at 46 U.S.C. App. §§ 1300-15 (Supp. III 1985) ("COGSA" or "U.S. COGSA").

23. 1984 U.S. Dep't of State, Treaties in Force 257 (listing 71 parties to Hague Rules).

24. The Soviet Union, for example, has not ratified or acceded to the Hague Rules, but Chapter 8 of its Merchant Shipping Code enacts substantial portions of the Hague Rules. See, e.g., Merchant Shipping Code art. 165 (U.S.S.R.) (enacting Hague Rules art. 4(5) with limitation of 250 rubles per parcel or customary freight unit), translated in The Soviet Codes of Law 1186-87 (Law in Eastern Europe No. 23, W. Simons ed. 1980); see also infra note 408 (noting Greek situation).

25. Just as the Hague Rules might be considered an extreme case, see supra note 21, the examples discussed in Parts II and III, infra, might appear to be extreme cases, for each concerns an issue on which the text of the Hague Rules offers no explicit guidance. Although there are many conflicts where the text is explicit, see, e.g., infra notes 39-43, the choice hero is deliberate. This Article is not concerned with cases where national courts simply misconstrue the plain meaning of a uniform law. It seeks to explain how conflicts arise in the interpretation of ambiguous provisions. It thus examines two "extreme" cases on the theory that the influences present in every case will be most obvious there.

26. The general maritime law exceptions (which protected the carrier only when it was not negligent or otherwise at fault) were Act of God, act of a public enemy, inherent vice of the goods, or fault of the shipper. These exceptions are continued in the Hague Rules. supra note 4, art. 4(2)(d), (f), (i), & (m).

27. Representatives of the interests of vessel owners and cargo owners negotiated the terms of the Hague Rules. For a history of their development. tee G. Gilmore & C. Black, The Law of Admiralty 139-44 (2d ed. 1975), and sources cited therein.

28. In essence, the carrier is obligated to exercise due diligence to make the ship seaworthy and to use due care in handling the cargo. Hague Rules, supra note 4, art. 3(1)(a) & (2). In return, the carrier is not liable for losses due to negligence in the navigation or management of the ship, see infra note 93, or for any of a list of specific causes, or for any other cause arising without the actual fault or privity of the carrier. Hague Rules, supra, art. 4(2).

29. Part II of this Article, for example, discusses the enforceability of Himalaya clauses. It does not join the debate on whether Himalaya clauses should be enforceable. A subsequent Article will consider the proper construction of a specific provision of the Hague Rules. It will use one example to illustrate how the U.S. courts have approached COGSA and to explain how they should approach it. See Sturley, The Fair Opportunity Requirement Under COGSA Section 4(5): A Case Study in the Misinterpretation of the Carriage of Goods by Sea Act, 19 J. Mar. L. & Com. ____ (1988) (forthcoming).

30. The need for international uniformity in the interpretation of the Hague Rules has been expressed time and time again. E.g., Robert C. Herd & Co. v. Krawill Mach. Corp., 359 U.S. 297, 301 (1959); Riverstone Meat Co. v. Lancashire Shipping Co., 1961 A.C. 807, 840 (Viscount Simonds) (quoting R.F. Brown & Co. v. T. & J. Harrison, 137 L.T.R. 549, 556 (C.A. 1927) (Atkin, L.J.); G. Gilmore & C. Black, supra note 27, at 191-92; A. Mocatta, M. Mustill, & S. Boyd, Scrutton on Charterparties and Bills of Lading 421-22 (19th ed. 1984); 1 R. Colinvaux, Carver's Carriage by Sea 250-51 (13th ed. 1982).

31. G. Gilmore & C. Black, supra note 27, at 53.

32. In the long run, the cargo interests will bear the insurance premiums, for the carrier must ultimately consider the cost of insurance in setting its freight rates. E.g., Egger, The Unworkable Per-Package Limitation of the Carrier's Liability under the Hague (or Hamburg) Rules, 24 McGill L.J. 459,473 (1978); Moore, The Hamburg Rules, 10 J. Mar. L. & Com. 1, 5, 11 (1978,); Zamora, Carrier Liability For Damage or Loss to Cargo In International Transport, 23 Am. J. Comp. L. 391, 393 (1975).

33. See generally G. Calabresi, The Costs of Accidents 26-31 (1970) (summarizing cost reduction goals of accident law).

34. Compare Tetley, The Himalaya Clause -- Heresy or Genius?, 9 J. Mar. L. & Com. 111, 112 (1977) (asserting that imposition of liability reduces negligence) with Rossmere, Cargo Insurance and Carriers' Liability: A New Approach, 6 J. Mar. L. & Com. 425, 429 (1975) (arguing that other incentives to reduce negligence make liability rules irrelevant).

35. In theory, the premiums should drop for both parties to allow for each insurer's reduction in risk due to the possibility that the other insurer might be liable. Even if this theory were confirmed in practice, however, there would still be double transaction costs (including the costs required for each insurer to evaluate and monitor the same risk) and an increased risk of litigation costs (in cases where the insurers could not agree which of the two was liable). See generally Zamora, supra note 32, at 393-95.

36. It is only when the parties' perceptions of the likely outcome of a dispute vary by an amount large enough to cover the expenses of litigation that it is rational for both parties to pursue a lawsuit through to decision. If the outcome had been clear before the litigation, the party who was certain to lose should have been unwilling to bear the costs. When the Hague Rules are uncertain enough to permit such varying perceptions, therefore, litigation under them is more likely to occur. See generally Priest & Klein, The Selection of Disputes for Litigation, 13 J. Legal Stud. 1 (1984).

37. See infra Part II.

38. See infra Part III; see also Mann, supra note 13, at 401-04 & nn. 14-28.

39. Article 4(5) permits the carrier to limit its liability to a fixed sum "per package," see infra note 91, but the Hague Rules do not define the term "package." The resulting litigation on such issues as whether a container is a "package" has been extensive and inconsistent. See Vance & Sassoon, The Container as a Package Controversy; A Comparative Jurisprudential Review, 16 Eur. Transport L. 315 (1981).

40. Article 4(2)(c) excuses the carrier from liability for damages caused by a "peril of the sea," but provides no further definition of the term. West German and U.S. courts have adopted different interpretations. See General Motors Overseas Operation v. S.S. Goettingen, 225 F. Supp. 902, 904-05 (S.D.N.Y. 1964).

41. At common law, "deviation" (which mayor may not involve physically deviating from the carrier's route) deprived the carrier of the benefit of its exemptions. The scope and effect of deviation under the Hague Rules is uncertain, although Article 4(4) protects the carrier in cases of "reasonable" deviation. See G. Gilmore & C. Black, supra note 27, at 176-83. U.S. and Canadian courts have reached different conclusions on whether deck carriage constitutes unreasonable deviation. Compare Encyclopaedia Britannica, Inc. v. S.S. Hong Kong Producer, 422 F.2d 7 (2d Cir. 1969) (deviation), cert. denied, 397 U.S. 964 (1970) with Consumers Glass Co. v. Farrell Lines, 1986 A.M.C. 443 (Ont. Sup. Ct. 1985) (not deviation).

42. Article 4(2)(b) excuses the carrier from liability for damages due to fire unless the fire is "caused by the actual fault or privity of the carrier." Hague Rules, supra note 4, art. 4(2)(b). Article 3(1), however, imposes an obligation on the carrier to exercise duo diligence to provide a seaworthy ship. Id. art. 3(1). Under Belgian law, the carrier can claim the benefit of the Article 4(2)(b) exclusion without proving that it exercised due diligence to make the ship seaworthy. The Patricio Murphy, Judgment of Mar. 3, 1972, Cour d'appel, Brussels, 1972 Jurisprudence du Port d'Anvers [Jur. Port Anv.] 21, reprinted in [1974] 1 Uniform L. Rev. 250. The Privy Council has ruled that Article 3(1) imposes an overriding obligation that the carrier must fulfill before it can rely on Article 4(2). Maxine Footwear Co. v. Canadian Gov't Merchant Marine Ltd., 1959 A.C. 589, 602-03 (P.C.) (Can.).

43. Article 3(6) discharges the carrier's liability if suit is not brought within one year of the delivery of the goods. See infra note 92. Under West German law this one-year time bar is a rigid requirement, while under Belgian law it can be interrupted. See The Travestein, Judgment of June 18, 1970, Landgericht, Bremen, W. Ger., 7 Eur. Transport L. 330 (1972) (declining to apply Belgian law that differed from German law); The Regine, Judgment of Feb. 23, 1972, Tribunal de commerce, Antwerp, 7 Eur. Transport L. 345 (1972) (refusing to enforce choice of forum clause calling for suit in Germany on ground that German courts would not apply Belgian law) (citing The Travestein); see also Mann, supra note 13, at 390 & nn. 63-66 (discussing other conflicts under art. 3(6).

44. E.g., E. Selvig, Unit Limitation of Carrier's Liability 3-4 (1961); A. Yiannopoulos, Negligence Clauses in Ocean Bills of Lading 8 & n. 24 (1962); Graveson, Bills of Lading and the Unification of Maritime Law in the English Courts, in Lectures on the Conflict of Laws and International Contracts 57, 61 (1951); Vance & Sassoon, supra note 39.

45. E.g., Mankiewicz, Conflicting Interpretations of the Warsaw Air Transport Treaty, 18 Am. J. Comp. L. 177 (1970); Wijffels, Legal Interpretations of C.M.R.: the Continental Viewpoint, 11 Eur. Transport L. 208, 211-29 (1976).

46. Black, The Bremen, COGSA and the Problem of Conflicting Interpretation, 6 Vand. J. Transnat'l L. 365, 370 (1973); cf. G. Gilmore & C. Black, supra note 27, at 191-92. This Article does not address proposals, such as Black's, to resolve conflicting interpretations. A subsequent Article will propose a solution building on the U.S. experience with certified questions and the European experience with preliminary references to the Court of Justice of the European Communities.

47. Black, supra note 46, at 369.

48. E.g., Mann, supra note 13, at 388-89, 390-92; Nadelmann, supra note 17, at 386.

49. The discussion in the text assumes that a court is equally likely to choose any of the five interpretations. Without this assumption, the mathematics would be more complicated but the essential point would be the same: a conflict is likely even if only a few courts consider the issue. Suppose that one interpretation is significantly more plausible than the others, and there is a 40% chance that a court would select it. The odds against ten courts reaching the same result would be over nine thousand to one.

50. For the second decision to be truly random, it must be independent of the first decision. The greater the first decision's influence on the second court, the less random the second decision will be. To illustrate the concept the discussion in the text makes the extreme assumption that the first decision will have no influence on the second court In the all too common situation where a national court is unaware of what foreign courts have decided, this assumption is fully justified.

51. With two equally likely interpretations, there is a 50% chance that two courts will disagree and a 75% chance of a conflict when three have considered the issue. If ten courts face the issue, the odds against their agreeing are over five hundred to one. If there are two plausible interpretations but one is far more likely than the other, conflicts are still surprisingly likely. Suppose there is a 90% chance that a court would select the more plausible view. In a universe of ten courts, there is still a 65% chance that a conflict will develop.

52. O.C. Giles, supra note 17, at 193-94.

53. Giles offers no additional explanation for the conflicts. At one point, he rebuts suggestions that conflicts arise due to differences between common law and civil law countries, or due to different methods of interpretation. Id. at 75.

54. E.g., id. at 58; E. Selvig, supra note 44, at 4; A. Yiannopoulos, supra note 44, at 181-82; Mann, supra note 17, at 289; Sundberg, supra note 17, at 224; cf. Sinclair, The Principles or Treaty Interpretation and their Application by the English Courts, 12 Int'l & Comp. L.Q. 508, 530-34 (1963) (English methods of interpretation vary according to method by which convention adopted); Schreuer, The Interpretation of Treaties by Domestic Courts, 45 Brit. Y.B. Int'l L. 255, 257 (1971) (same).

55. There is no reason to suppose, for example, that poorly informed courts will be more restrictive or less restrictive in their interpretations, or that they will be more liberal or more conservative, or that their conclusions will be more favorable to carriers or to cargo claimants. One could argue that their interpretations are more likely to be wrong, but the problem here is with provisions that plausibly bear more than one meaning. See supra note 25.

56. E.g., P. Devlin, The English Rules on Choice of Law With Reference to Maritime Contracts 18 (Handelshogskolans I Goteburg Skriftserie [Gothenburg School of Economics Publications] No. 1, 1952); H.C. Gutteridge, supra note 17, at 109; Mann, supra note 17 at 278-79; Minutes or the Third Meeting of Organisations Concerned With the Unification of Law, 1963 Unidroit Y.B. 149 (1964) (statement of N. Marsh). Contra O.C. Giles, supra note 17, at 75.

57. The House of Lords has recently relaxed this rule. Gatoil Int'l Inc. v. Arkwright-Boston Mfrs. Mut. Ins. Co., 1985 A.C. 255, 263-65 (1984); Fothergill v. Monarch Airlines, 1981 A.C. 251 (1980).

58. E.g., Air France v. Saks, 470 U.S. 392, 400 (1985) (When "interpreting a treaty it is proper, or course, to refer to the records of its drafting and negotiation."); Vienna Convention on the Law or Treaties, May 23, 1969, art. 32, 1980 Gr. Brit. T.S. No. 58, at 19 (Cmnd. 7964), reprinted in 8 I.L.M. 679, 692 ("[S]upplementary means of interpretation includ[e] the preparatory work or the treaty.").

59. Sundberg, supra note 17, at 221; cf. Schreuer, supra note 54, at 265.

60. It is possible to apply this theory in combination with the "random distribution" theory. Suppose that a French court applying the French connotation of a technical term bas a 60% probability of construing a convention one way and a 40% probability of construing it a second way. For a German court the probabilities are reversed. There is only a 52% chance that a conflict will arise (2% higher than if the French and German connotations had not altered the probabilities), but if one does, the chances are almost 70% that the French court will have adopted the first construction and the German court the second. Thus there is both a systematic and a random element to the conflict.

61. See J. Honnold, Uniform Law for International Sales Under the 1980 United Nations Convention 114 (1982) (describing efforts to draft convention in "language that refers to things and events for which there are words of common content in the various languages").

62. The conflicts described in Parts II and III of this Article, for example, did not arise through any simple disagreement on the meaning of specific technical legal terms in the Hague Rules.

63. Kg., Fothergill v. Monarch Airlines, 1981 A.C. 251, 272-75, 286-87, 301 (comparing French connotation of "avarie" and English connotation of "damage" in Warsaw Convention, supra note 4, art. 26(2)). See generally German, Interpretation of Plurilingual Treaties: A Study of Article 33 of the Vienna Convention on the Law of Treaties, 11 Harv. Int'l L.J. 400 (1970); Hardy, The Interpretation of Plurilingual Treaties by International Courts and Tribunals, 37 Brit. Y.B. Int'l L. 72 (1961); Lipstein, Some Practical Comparative Law: The Interpretation of Multi-Lingual Treaties With Special Regard to the EEC Treaties, 48 Tul. L. Rev. 907 (1974).

64. E.g., Beutel, supra note 17, at 19-22; Blomquist, The Proposed Uniform Law on International Bills of Exchange and Promissory Notes, 9 Cal. W. Int'l L.J. 30, 67-68 (1979); Mann, supra note 17, at 278-91. Contra O.C. Giles, supra note 17, at 75.

65. See supra note 41 (U.S. and Canadian interpretations of "deviation"); compare infra notes 122-46 and accompanying text (U.S. law on Himalaya clauses) with infra notes 221-40 and accompanying text (Canadian law on Himalaya clauses); compare infra notes 305-22 (U.S. law on choice of forum clauses) with infra note 277 (Canadian law on choice of forum clauses).

66. See supra note 43 (Belgian and West German interpretations of Hague Rules art. 3(6) time bar); cf. Wijffells, supra note 45, at 211-29 (describing conflicts among six civil law countries in interpre1etion of CMR, supra note 4).

67. The suggestion that conflicts arise through differences in legal tradition is most useful when there are specific differences in substantive law associated with each legal tradition. To the extent that civil law countries permit contracts for the benefit of a third party, for example, while common law countries do not, conflicts are likely on the validity of the Himalaya clause. U.S. law demonstrates that this problem arises through differences in substantive law rather than legal tradition. Although a common law country, the United States permits contracts for the benefit of a third party. And its law on the Himalaya clause is more like that of other countries that permit such contracts than that of other common law countries. Part II develops this analysis in detail.

68. There is some support for this theory in The New York Star, Port Jackson Stevedoring Pty. v. Salmond & Spraggon (Austl.) Pty., 139 C.L.R. 231, 285 (Austl. 1978) (Murphy, J.), rev'd, [1981] 1 W.L.R. 138 (P.C. 1980) (Austl.). See infra note 215 and accompanying text.

69. See, e.g., U.N. Conference on the Carriage of Goods by Sea, Comments and proposals by Governments and international organizations on the draft Convention on the carriage of goods by sea, U.N. Doc. A/CONF.89 (1977); Zamora, supra note 32, at 395.

70. Cheng, Centrifugal Tendencies in Air Law, 10 Current Legal Probs. 200, 200-07 (1957); Tetley, Canadian Comments on the Proposed Uncitral Rules, 9 J. Mar. L. & Com. 251, 252, 256-57, 268 (1978).

71. See generally G. Gilmore & C. Black, supra note 27, at 142-43.

72. E.g., Tetley, supra note 70, at 252 (Canada is cargo nation).

73. But see infra note 215 and accompanying text.

74. See infra Part II.

75. See infra notes 149-68 and accompanying text (discussing Scruttons Ltd. v. Midland Silicones Ltd., 1962 A.C. 446, 471 (1961).

76. See infra notes 122-46 and accompanying text.

77. To the extent that substantive domestic law reflects national polices, a court will still apply them indirectly.

78. A general work on treaty interpretation addresses the situation in which a domestic constitution limits the executive's treaty-making power, and the court feels that it must choose between a narrow treaty or an invalid treaty. Schreuer, supra note 54, at 265. Scbreuer quotes a decision or the German Federal Constitutional Court to illustrate the point: "[w]here several interpretations are feasible, preference must be given to an interpretation which permits the treaty to exist, having regard to the requirements of the Constitution'." Id. at 266 (quoting Judgment of May 4, 1955, Bundesverfassungegericht. [BVerfGE], 4 Entscheidungen BVerfGE 157, 168, translated in 22 I.L.R. 630, 631); cf. infra note 83 and accompanying text. This appears to be the closest any prior treatment has come to addressing the impact of substantive domestic law on the process of interpretation.

79. A "permissible" interpretation, as the term is used here, is legally plausible in the abstract, without consideration for its consistency with other national or international legal norms.

80. A random distribution hypothesis is most convincing in situations where the courts of a single country disagree on the interpretation of a provision. Such conflicts are not unusual The U.S. Courts of Appeals for the Second and Ninth Circuits, for example, explicitly disagree on the proper allocation of the burden of proof under COGSA section 4(2)(b). Compare In re Ta Chi Navigation (Pan.) Corp., 677 F.2d 225, 228-29 (2d Cir. 1982) (burden on shipper) with Sunkist Growers, Inc. v. Adelaide Shipping Lines, 603 F.2d 1327, 1335-36 (9th Cir. 1979) (burden on carrier), cert. denied, 444 U.S. 1012 (1980). Similarly, the Courts of Appeals for the Fifth and Ninth Circuits disagree on the opportunity that a shipper must have to declare the value of a shipment before the carrier can rely on the COGSA section 4(5) package limitation. Compare Komatsu, Ltd. v. States S.S. Co., 674 F.2d 806 (9th Cir. 1982) (denying carrier benefit of limitation) with Brown & Root, Inc. v. M/V Peisander, 648 F.2d 415, 419-25 (5th Cir. 1982) (granting carrier benefit of limitation). Other countries have also experienced such conflicts. See, e.g., infra note 365 and accompanying text (Belgian experience).

81. The situation is more complicated than the text suggests, for the relevant domestic law is also subject to a range of permissible interpretations.

82. In the rare case where an international convention and national law necessarily conflict, i.e., where no permissible interpretation of the convention is consistent with a permissible interpretation of national law, the court must choose one and reject the other. The court might resolve the conflict by giving effect to the later enactment. E.g., Edye v. Robertson, 112 U.S. 580, 597-99 (1884) (later statute takes precedence over treaty); United States v. Schooner Peggy, 5 U.S. (1 Cranch) 103, 109-10 (1801) (treaty supersedes prior law); see Restatement of Foreign Relations Law of the United States (Revised) § 135 (Tent. Draft No. 6, 1985). Or it might have a per se rule favoring one or the other. Several European countries give precedence to a treaty over a domestic statute. E.g., Judgment of May 27, 1971, Cour de Cassation, Belg., 1971 Pasicrisie Belge I 886, 919, translated in 1972 Common Mkt. L.R. 330, 373; Judgment of May 24, 1975, Cour de Cassation, Fr., 1975 Dalloz-Sirey, Jurisprudence 497, 506, translated in [1975] 2 Common Mkt. L.R. 336, 368-69; Const. art. 55 (Fr.); Grondwet art. 94 (Neth.); Brinkhorst & Lammers, The Impact of International Law, Including European Community Law on the Netherlands Legal Order, in Introduction to Dutch Law for Foreign Lawyers 561, 571-72 (1978) (discussing Grondwet art. 65 (1953) (superseded by Grondwet art. 94 (1983)). In England and the Commonwealth, an act of Parliament -- whenever passed -- is superior to an unenacted treaty. Attorney-Gen. for Can. v. Attorney-Gen. for Ont., 1937 A.C. 326, 347-48 (P.C.) (Can.) (Lord Atkin). Whichever choice is made, there is no conflict in interpretation. If the court refuses to apply the convention, uniformity of application has certainly suffered. But the court has ignored the convention, not construed it. Whatever practical impact this might have on the aims of the convention, it is not the issue here.

83. E.g., NLRB v. Catholic Bishop, 440 U.S. 490, 500 (1979). The Supreme Court not. only construes statutes to avoid constitutional violations, it construes statutes to avoid even a constitutional question. E.g., United States v. Security Indus. Bank, 459 U.S. 70, 78 (1952); St. Martin Evangelical Lutheran Church v. South Dakota, 451 U.S. 772, 780 (1951); United States v. Clark, 445 U.S. 23, 27 (1980). See generally 2A J. Sutherland, Statutes and Statutory Construction § 45.11 (N. Singer 4th ed. 1984 rev.).

84. Aptheker v. Secretary of State, 378 U.S. 500, 515 (1964) (quoting Scales v. United States, 367 U.S. 203, 211 (1961).

85. A closer analogy in terms of substance rather than process can be seen in the U.S. implementation of the Hague Rules. When Congress enacted COGSA, it made several changes in the internationally agreed text. See COGSA §§ 3(2), 3(4), 3(6), 3(7), 4(2)(j), 4(2)(q), 4(4), 4(5), 8; see also infra note 91 (discussing section 4(5). The State Department. explained that. these changes were "intended primarily (1) to clarify provisions in the [Hague Rules] ...and (2) to coordinate [COGSA] with other legislation of the United States." Memorandum of the Department. of State, June 5, 1937, 51 Stat. 269, 274 (1937). The courts do implicitly what the political branches did explicitly.

86. E.g., American Bank & Trust Co. v. Dallas County, 463 U.S. 855, 868 (1983). See generally 1A J. Sutherland, supra note 83, § 23.10 (1985 rev.) (U.S. law); P. Langan, Maxwell on the Interpretation of Statutes 191-93 (12th ed. 1969) [hereinafter Maxwell] (English law); D.C. Pearce, Statutory Interpretation in Australia 94 (1974) (Australian law); A.P. Chatterjee, Interpretation of Statutes 425-31 (2d ed. 1983) (Indian law).

87. Murray v. Schooner Charming Betsy, 6 U.S. (2 Cranch) 64,118 (1804); Maxwell, supra note 86, at 183-85.

88. Whitney v. Robertson, 124 U.S. 190, 194 (1888). Since a treaty is equivalent to a federal statute under U.S. law, id.; Foster v. Neilson, 27 U.S. (2 Pet.) 253, 314 (1829), this rule can be seen as an example of the presumption against repeals by implication. Cf. TWA v. Franklin Mint Corp., 466 U.S. 243, 252 (1984) (applying "canon of construction against. finding implicit repeal of a treaty in ambiguous congressional action"). See generally 1A J. Sutherland, supra note 83, § 32.06 (1985 rev.); Maxwell, supra note 86, at 185.

89. U.S. courts theoretically construe treaties more liberally than statutes. See, e.g., Nielsen v. Johnson, 279 U.S. 47, 51-52 (1929). The Supreme Court has also recognized, at least in the domestic context, that courts should not construe uniform laws in accordance with prior local doctrine. Commercial Nat'l Bank v. Canal-Louisiana Bank & Trust Co., 239 U.S. 520, 528-29 (1916) (construing Uniform Warehouse Receipts Act). The Court ignored both of these principles in Robert C. Herd & Co. v. Krawill Mach. Corp., 359 U.S. 297 (1959) (construing COGSA). English courts also declare that they will construe treaties liberally. E.g., Imperial Japanese Gov't v. Peninsular & Oriental Steam Navigation Co., 1895 A.C. 644, 657 (P.C.) (Sup. Ct. for China & Japan). But the House of Lords did not consider this possibility in Scruttons Ltd. v. Midland Silicones Ltd., 1962 A.C. 446 (1961). See infra note 157.

90. See supra note 28 and accompanying text.

91. Article 4(5) provides, in relevant part:

92. Article 3(6) provides, in relevant part, that "the carrier and the ship shall be discharged from all liability in respect of loss or damage unless suit is brought within one year after delivery of the goods or the date when the goods should have been delivered." Hague Rules, supra note 4, art. 3(6).

93. Id. art. 4(2)(a). This "most striking" exemption, G. Gilmore & C. Black, supra note 27, at 155, is based on the assumption that the vessel's safety is a sufficient incentive for the carrier to select a captain and crew who will exercise due care in the navigation and management of the vessel. Id. at 143.

94. Hague Rules, supra note 4, art. 4(2)(c).

95. Id. art. 4(2)(d).

96. Article l(d) defines "ship" as "any vessel used for the carriage of goods by sea. " Id. art. 1(d). The Rules protect the ship to take account of jurisdictions, such as the United States, that permit in rem actions against the vessel.

97. Id. art. l(a) (emphasis added).

98. In practice, of course, this means that some carriers employ their own longshore and harbor workers to load and unload their ships. See, e.g., Jackson v. Lykes Bros. S.S. Co., 386 U.S. 731, 735 (1967) ("[L]ongshoreman was hired directly by the [vessel] owner."); Incorvala v. Hellenic Lines, 668 F.2d 650, 650 (2d Cir.) ("Hellenic Lines provides its own stevedoring services."), cert. denied, 459 U.S. 967 (1982).

99. L. Kendall, The Business of Shipping 101 (5th ed. 1986); G. Gilmore & C. Black, supra note 27, at 278; see, e.g., Edmonds v. Compagnie Generale Transatlantique, 443 U.S. 256, 263-64 (1979) (contrasting "recurring situation" in which longshoreman is "employed by a stevedoring concern" with the "less familiar arrangement where the ... longshoreman loading or unloading the ship is employed by the vessel itself").

100. Secrest Machine Corp. v. S.S. Tiber, 450 F.2d 285, 286 (5th Cir. 1971) (per curiam). The precise wording of the Himalaya clause is often important in determining its effect under different national laws.

101. The Himalaya clause received its name from an English case popularly known as The Himalaya, and reported as Adler v. Dickson, [1955] 1 Q.B. 158 (C.A. 1954). Some well-known bill of lading clauses are named for the cases that first upheld them. See, e.g., The Jason, 225 U.S. 32 (1912) (upholding "Jason clause"); G. Gilmore & C. Black. supra note 27, at 266-68 (describing "Jason clause"). Adler v. Dickson, however, demonstrated to Carriers a need for a "Himalaya clause." The plaintiff was injured while a passenger on the cruise ship The Himalaya. She sued the master and the boatswain for their negligence bemuse the carrier was contractually exempt from all liability. Because the contract did not have a "Himalaya clause," she succeeded. The carrier, having indemnified its employees, ultimately paid the damages. It thus lost its contractual exemption indirectly.

102. See infra notes 141-45 and accompanying text.

103. See infra notes 243-46 and accompanying text.

104. The quoted clause does not provide explicitly that the carrier is contracting as agent for the stevedore. See infra note 167 and accompanying text. Thus it probably would not succeed in England, see infra notes 147-86 and accompanying text. Australia, see infra notes 189-220 and accompanying text, or Canada, see infra notes 221-40 and accompanying text.

105. A.M. Collins & Co. v. Panama R.R., 197 F.2d 893 (5th Cir.), cert. denied, 344 U.S. 875 (1952), overruled, Robert C. Herd & Co. v. Krawill Mach. Corp., 359 U.S. 297 (1959).

106. Id. at 895.

107. Id. at 896-97.

108. Id. at 897.

109. Id. at 896.

110. Id. at 895 (C.Z. Code tit. 3, § 933 (1934) (superseded 1963) "must give way to the provisions of the later enacted [COGSA]"). There is a clear suggestion that the local statute would have applied if Congress had enacted COGSA first.

111. Id. at 896 (distinguishing Reid v. Fargo, 241 U.S. 544 (1916)).

112. Id. at 897 (citing Galveston Wharf Co. v. Galveston, H. & S.A. Ry., 285 U.S. 127 (1932) (case arising under Carmack Amendment, 49 U.S.C. § 20(11) (1926) (superseded 1978))).

113. Id. (citing Restatement of Agency § 347 (1933). The current Restatement provides:

114. At one point there was strong support for the rule in England. E.g., Elder, Dempster & Co. v. Paterson, Zochonis & Co., 1924 A.C. 522, 534 (Viscount Cave) ("[A]gents ... can claim the same protection as their principals."); Mersey Shipping & Transp. Co. v. Rea, Ltd., [1925] 21 Lloyd's Rep. 375, 378 (K.B. Div'l Ct.) (Scrutton, L.J.) ("[W]here there is a contract which contains an exemption clause, the servants or agents who act under that contract have the benefit of the exemption clause.").

115. E.g., Wilson v. Darling Island Stevedoring & Lighterage Co., 95 C.L.R. 43, 69 (Austl. 1956) (Fullagar, J.) ("no foundation whatever for suggesting that there is any such general rule of law"); Gilbert, Stokes & Kerr Pty. v. Dalgety & Co., 48 N.S.W. St. R. 435, 443 (N.S.W. Sup. Ct. 1948) (believes "no such general principle ... to be round in the law of agency"), overruled on other grounds, Wilson v. Darling Island Stevedoring & Lighterage Co., 95 C.L.R. 43 (Austl. 1956).

116. E.g., Robert C. Herd & Co. v. Krawill Mach. Corp., 359 U.S. 2.97, 303.05 (1959).

117. A.M. Collins & Co. v. Panama R.R., 197 F.2d at 898 (Holmes, J. dissenting) (citing Reid v. Fargo, 241 U.S. 544 (1916).

118. Id.; see infra note 129.

119. A.M. Collins & Co. v. Panama R.R., 197 F.2d at 898; see infra notes 131-35 and accompanying text.

120. E.g., United States v. The South Star, 210 F.2d 44 (2d Cir. 1954); Autobuses Modernos, S.A. v. The Federal Mariner, 125 F. Supp. 780 (E.D. Pa. 1954); Ford Motor Co. v. Jarka Corp., 134 N.Y.S.2d 52, 1954 A.M.C. 1095 (Mun. Ct. N.Y.C.).

121. Robert C. Herd & Co. v. Krawill Mach. Corp., 256 F.2d 946 (4th Cir. 1958), aff'd, 359 U.S. 2.97 (1959).

122. Robert C. Herd & Co. v. Krawill Mach. Corp., 359 U.S. 297 (1959).

123. Id. at 301-02.

124. Id. at 302-03.

125. Id. at 303-05.

126. Id. at 301.

127. Id.

128. Id. at 306-08.

129. Id. at 304; see 3 J. Sutherland, supra note 83, § 61.01 (C.D. Sands 4th ed. 1974). Commentators have criticized the principle that statutes in derogation of the common law must be strictly construed, see, e.g., R. Dickerson, The Interpretation and Application of Statutes 206-08 (1975), and a number of states have abolished it by statute, 3 J. Sutherland, supra, § 61.05 & n.1.

130. Even in the United States there are limitations on the rule, such as the doctrine of Restatement section 347, see supra note 113, on which the Collins court bud relied. The Herd Court did not mention section 347, but cited several cases involving a principal's personal immunities. Robert C. Herd & Co. v. Krawill Mach. Corp., 359 U.S. at 302-04 (citing Brady v. Roosevelt S.S. Co., 317 U.S. 575 (1943); Sloan Shipyards Corp. v. United States Shipping Bd. Emergency Fleet Corp., 258 U.S. 549 (1922); and Osborn v. President. Directors, & Co. of the Bank of the United States, 22 U.S. (9 Wheat.) 738 (1824). By its terms, section 347 does not apply to a principal's personal immunities.

131. See infra notes 163-66 and accompanying text.

132. See infra notes 247-58 and accompanying text. West Germany recognizes the general rule in its Civil Code. Bürgerliches Gesetzbuch [BGB] § 840(2) (W. Ger.).

133. Brady v. Roosevelt S.S. Co., 317 U.S. 575, 577-85 (1943).

134. Sloan Shipyards Corp. v. United States Shipping Bd. Emergency Fleet Corp., 258 U.S. 549, 566-70 (1922).

135. Osborn v. President, Directors, & Co. of the Bank of the United States, 22 U.S. (9 Wheat.) 738, 842-44 (1824). The point was in fact admitted. Id. at 843.

136. U.S. courts had permitted contracts for the benefit of a third party since the mid-nineteenth century. See Lawrence v. Fox, 20 N.Y. 268 (1859).

137. See infra notes 152-57 and accompanying text (England); infra note 198 and accompanying text (Australia); infra notes 222-28 and accompanying text (Canada).

138. Robert C. Herd & Co. v. Krawill Mach. Corp., 359 U.S. 297, 302 (1959).

139. E.g., Secrest Mach. Corp. v. S.S. Tiber, 450 F.2d 285, 286 (5th Cir. 1971) (per curiam); Mediterranean Marine Lines v. John T. Clark & Son, 485 F. Supp. 1330, 1333 (D. Md. 1980); Scruttons Ltd. v. Midland Silicones Ltd., 1962 A.C. 446, 460, 471 (1961).

140. See, e.g., Assicurazioni Generali v. D'Amico, 766 F.2d 485, 487-88 (11th Cir. 1985) (citing Herd for proposition that bill of lading may extend limitations to third parties); B. Elliott (Can.) Ltd. v. John T. Clark & Son, 704 F.2d 1305, 1308 (4th Cir. 1983) (same); Toyomenka, Inc. v. S.S. Tosaharu Maru, 523 F.2d 518, 520-21 (2d Cir. 1975) (same); De Laval Turbine, Inc. v. West India Indus., 502 F.2d 259, 264 (3d Cir. 1974) (same).

141. For a recent survey of U.S. case law in this area, see Zawitoski, Limitation of Liability for Stevedores and Terminal Operators Under the Carrier's Bill of Lading and COGSA, 16 J. Mar. L. & Com. 337, 345-49 (1985).

142. Carle & Montanari, Inc. v. American Export Isbrandtsen Lines, 275 F. Supp. 76, 78 (S.D.N.Y.), aff'd, 386 F.2d 839 (2d Cir. 1967), cert. denied, 390 U.S. 1013 (1968).

143. Courts have, a fortiori, upheld clauses explicitly mentioning "stevedores." See, e.g., Brown & Root, Inc. v. M/V Peisander, 648 F.2d 415, 418 n.6 (5th Cir. 1981); Timco Eng'g, Inc. v. Rex & Co., 603 F. Supp. 925, 928 n.3 (E.D. Pa. 1985); Koppers Co. v. S/S Defiance, 542 F. Supp. 1356, 1360 (D. Md. 1982), aff'd, 704 F.2d 1309 (4th Cir. 1983).

144. Bernard Screen Printing Corp. v. Meyer Line, 328 F. Supp. 288, 289 (S.D.N.Y. 1971), aff'd, 464 F.2d 934 (2d Cir. 1972), cert. denied, 410 U.S. 910 (1973); cf. Tessler Bros. (B.C.) v. Italpacific Line, 494 F.2d 438, 446-47 (9th Cir. 1974) ("independent contractors" includes stevedores); Secrest Machine Corp. v. S.S. Tiber, 450 F.2d 285, 287 (5th Cir. 1971) (per curiam) ("clear that the term 'independent contractor' includes stevedores").

145. Assicurazioni Generali v. D'Amico, 766 F.2d 485, 489-90 (11th Cir. 1985) ("bailee" covers terminal operator). But see De Laval Turbine, Inc. v. West India Indus., 502 F.2d 259, 265-68 (3d Cir. 1974) (clause including "bailee" inadequate).

146. Cabot Corp. v. S.S. Mormacscan, 441 F.2d 476, 478 (2d Cir.), celt. denied, 404 U.S. 855 (1971); accord Rupp v. International Terminal Operating Co., 479 F.2d 674, 677 (2d Cir. 1973). The Cabot court also relied on the principle that "limitations of liability are to be construed most strongly against the parties who introduced them into the contract." 441 F.2d at 478; cf. De Laval Turbine, Inc. v. West India Indus., 502 F.2d 259, 254-65 (3d Cir. 1974) ("[S]uch extension ... will be strictly construed against the party seeking protection.") (applying Robert C. Herd & Co. v. Krawill Mach. Corp., 359 U.S. 297, 305 (1959)).

147. This section includes Privy Council appeals from Commonwealth countries. Although not technically an "English" court, the Judicial Committee of the Privy Council is composed of essentially the same judges as the Appellate Committee of the House of Lords. See P.F. Smith & S.H. Bailey, The Modern English Legal System 84 & nn. 5-7 (1984) (describing composition of Judicial Committee). Privy Council decisions are thus a good indication of the state of English law. Indeed English courts and commentators occasionally treat them as if they were House of Lords decisions. The Wagon Mound, Overseas Tankship (U.K.) Ltd. v. Morts Dock & Eng'g Co., 1961 A.C. 388 (P.C.) (N.S.W.), for example, is often regarded as though it had overruled Polemis v. Furness, Withy & Co., [1921] 3 K.B. 560 (C.A.). E.g., Doughty v. Turner Mfg. Co., [1964] 1 Q.B. 518, 525 (C.A. 1963) (Lord Pearce); id. at 528 (Harman, L.J.); id. at 532 (Diplock, L.J.); Goodhart, Obituary: Re Polemis, 77 Law Q. Rev. 175 (1961). Furthermore, Commonwealth commentators have long regarded the Privy Council as an essentially English court. E.g., H. Hughes, National Sovereignty and Judicial Autonomy in the British Commonwealth of Nations 49-51 (1931); Blackshield, The Last of England -- Farewell to Their Lordships Forever, 56 Law Inst. J. 779, 779-80 (1982); see also R. Stokes, New Imperial Ideals 86 (1930).

148. E.g., Scruttons Ltd. v. Midland Silicones Ltd., 1962 A.C. 446, 471 (1961) (Viscount Simonds); Riverstone Meat Co. v. Lancashire Shipping Co., 1961 A.C. 807, 840 (Viscount Simonds) (quoting R.F. Brown & Co. v. T. & J. Harrison, 137 L.T.R. 549, 556 (C.A. 1927) (Atkin, L.J.); Stag Line v. Foscolo, Mango & Co., 1932 A.C. 328, 350 (1931) (Lord Macmillan); The Eurymedon, 1938 P. 41, 61 (C.A. 1937) (Scott, L.J.).

149. Scruttons Ltd. v. Midland Silicones Ltd., 1962 A.C. 446 (1961).

150. See supra notes 122.38 and accompanying text.

151. The Scruttons bill of lading was issued before the Herd decision. Scruttons Ltd. v. Midland Silicones Ltd., 1962 A.C. at 449. That is presumably why it did not specify stevedores.

152. There is a long line of authority in support of the third-party beneficiary rule. E.g., Dunlop Pneumatic Tyre Co. v. Selfridge & Co., 1915 A.C. 847, 853.

153. U.S. courts had begun recognizing third-party beneficiary contracts over a century before. Lawrence v. Fox, 20 N. Y. 268 (1859). Civil law codes explicitly permit third-party beneficiary contracts. See, e.g., Code civil art. 1121 (Fr.); Bürgerliches Gesetzbuch (BGB] § 328(1) (W. Ger.); Que. Civ. Code art. 1029.

154. Law Revision Committee, Sixth Interim Report (Statute of Frauds and the Doctrine of Consideration), Cmd. No. 5449, paras. 41-49 (1937).

155. E.g., Woodar Inv. Dev. Ltd. v. Wimpey Constr. U.K. Ltd., [1980] 1 W.L.R. 277,300 (H.L.) (Lord Scarman); id. at 297-98 (Lord Keith); id. at 291 (Lord Salmon, dissenting); Beswick v. Beswick, 1968 A.C. 58, 72 (1967) (Lord Reid); id. at 93 (Lord Pearce).

156. Scruttons Ltd. v. Midland Silicones Ltd., 1962 A.C. at 473, 478 (Lord Reid); id. at 492 (Lord Denning, dissenting); id. at 494 (Lord Morris).

157. From 1898 until 1966, the House of Lords considered itself bound by its prior decisions. London Tramways Co. v. London County Council 1898 A.C. 375: Practice Statement (Judicial Precedent), [1966] 1 W.L.R. 1234 (H.L.). In 1962, therefore, Dunlop, see supra note 152, was binding precedent. The Scruttons court might nevertheless have limited itself to a simple application of U.S. law. The parties, Scruttons Ltd. v. Midland Silicones Ltd., 1962 A.C. at 454, 460, and the law lords, id. at 466, 495, saw that the U.S. COGSA governed. Or the House might have decided that its Elder, Dempster decision, see infra notes 158-60 and accompanying text, created a relevant exception to the third-party beneficiary rule. Alternatively, it might have recognized the Hague Rules as an international uniform law that should be interpreted "unconstrained by technical rules of English law, or by English legal precedent." James Buchanan & Co. v. Babco Forwarding & Shipping (U.K.) Ltd., 1978 A.C. 141, 152 (1977) (Lord Wilberforce).

158. Elder, Dempster & Co. v. Paterson, Zochonis & Co., 1924 A.C. 522.

159. Whether the exclusion Clause protected the shipowner was a secondary issue in Elder, Dempster. The primary issue was whether the vessel's unseaworthiness (to which the exclusion Clause did not apply) or improper stowage (to which the Clause did apply) was responsible for the damage. The House of Lords ultimately held that negligent stowage had been responsible.

160. Viscount Cave, joined by Lord Carson, said simply that the shipowner "took possession of the goods ...on behalf of and as the agents of the charterers, and so can Claim the same protection." Elder, Dempster & Co. v. Paterson, Zochonis & Co., 1924 A.C. at 534. Viscount Finlay declared that "when the [tortious] act is done in the course of rendering the very services provided for in the bill of lading, the limitation on liability therein contained must attach, whatever the form of the action and whether owner or charterer be sued." Id. at 548 (Viscount Finlay, dissenting). Only Lord Sumner, with whom Lord Dunedin joined, seemed troubled by extending protection to the shipowner. Despite his uncertainty with the rationale, however, he had no doubt that the shipowner was entitled to the bill of lading's protection. Id. at 564-65.

161. Scruttons Ltd. v. Midland Silicones Ltd., 1962 A.C. at 468-71 (Viscount Simonds); id. at 475-79 (Lord Reid): id. at 479-81 (Lord Keith); id. at 485-89 (Lord Denning, dissenting); id. at 494 (Lord Morris).

162. If the Hague Rules had applied in Elder, Dempster, the Article l(a) "carrier" definition would have covered the charterer and the shipowner. See supra note 97 and accompanying text. Thus the bill of lading would have protected them both. Article 3(2), however, would have imposed liability for improper stowage on the carrier. Article 3(8), see infra note 271 and accompanying text, would have invalidated the bill of lading clause to the contrary, see supra note 159.

163. Scruttons Ltd. v. Midland Silicones Ltd., 1962 A.C. at 483-88 (Lord Donning, dissenting).

164. E.g., Winterbottom v. Wright, 10 M. & W. 109, 152 Eng. Rep. 402 (Ex. 1842).

165. Donoghue v. Stevenson, 1932 A.C. 562. This decision is analogous to the decision in MacPherson v. Buick Motor Co., 217 N.Y. 382, 111 N.E. 1050 (1916).

166. See also Reynolds, Tort Actions in Contractual Situations, 11 N.Z.U.L. Rev. 215, 220 (1985) ("[R]esult of Donoghue v. Stevenson was that employees, agents or sub-contractors, who had previously been protected by the privity of contract rule, would now be liable to the other contracting party in cases of damage.") (footnote omitted).

167. Scruttons Ltd. v. Midland Silicones Ltd., 1962 A.C. at 474.

168. In the nineteenth century, British courts were far more willing than U.S. courts to allow exemption clauses eliminating liability for negligence entirely. See G. Gilmore & C. Black, supra note 27, at 142. The complete reversal of this situation may seem ironic, but it shows the impact of domestic law on the application of the Hague Rules.

169. This section treats the Privy Council as an English court. See supra note 147. That is particularly appropriate in The Eurymedon, New Zealand Shipping Co. v. A.M. Satterthwaite & Co., 1975 A.C. 154 (P.C. 1974) (N.Z.), which was governed by English law.

170. New Zealand Shipping Co. v. A.M Satterthwaite & Co., 1975 A.C. 154 (P.C. 1974) (N.Z.).

171. The bill of lading (drafted before the Scruttons decision, id. at 160, and hence without explicit attention to Lord Reid's suggestions) purported to extend the carrier's limitations to "every independent contractor from time to time employed by the carrier," and stated that the carrier acted as agent for all independent contractors for the purpose of this clause. Id. at 165.

172. A.M. Satterthwaite & Co. v. New Zealand Shipping Co., 1972 N.Z.L.R. 385 (N.Z. Sup. Ct. 1971), rev'd, [1973] 1 N.Z.L.R. 174 (N.Z. Ct. App. 1972), rev'd, 1975 A.C. 154 (P.C. 1974) (N.Z.).

173. A.M. Satterthwaite & Co. v. New Zealand Shipping Co., [1973] 1 N.Z.L.R. 174 (N.Z. Ct. App. 1972), rev'd, 1975 A.C. 154 (P.C. 1974) (N.Z.).

174. The Judicial Committee had little trouble including the stevedore within the phrase "every independent contractor from time to time employed by the carrier," see supra note 171, although it seemed to attach some weight to the fact that the defendant "habitually acts as stevedore in New Zealand" for the carrier. It also had little trouble finding the agency clause in the bill of lading adequate. Finally, it had no trouble finding authority for the agency, since the stevedore was the parent company of the carrier. The Eurymedon, New Zealand Shipping Co. v. A.M. Satterthwaite & Co., 1975 A.C. at 166-67.

175. Id. at 167.

176. Id. at 167-68.

177. Id. at 168. Under the "unilateral contract" analysis, the shipper might be deemed to have said to the carrier's independent contractors, "I promise that when you perform services for me, I will grant you the benefit of the carrier's exemptions and limitations." The bargain is not yet a full contract because it lacks mutuality.

178. The Eurymedon, New Zealand Shipping Co. v. A.M. Satterthwaite & Co., 1975 A.C. at 167.

179. Id. at 173 (Viscount Dilhorne, dissenting).

180. Viscount Dilhorne's dissent was typical. He did not believe that the Himalaya clause contained an offer by the shipper, or an agreement between the shipper and the stevedore, whereby the stevedore obtained the benefit of the exemption and limitation clause in the bill of lading if it unloaded the shipper's goods. He would have recognized such an offer only if the clause contained "clear and unequivocal language" to that effect. Id. at 170.

181. Privy Council decisions generally have only persuasive value. Even when the House of Lords considered itself bound by its own decisions, see supra note 157, the Judicial Committee of the Privy Council did not. R. Cross, Precedent in English Law 20-21 (3d ed. 1917). The English courts allow great respect for Privy Council decisions, see supra note 147. but are not bound to follow them, e.g., Port Line Ltd. v. Ben Line Steamers Ltd., [1958] 2 Q.B. 146, 168 (declining to follow Lord Strathcona S.S. Co. v. Dominion Cole Co., 1926 A.C. 108 (P.C. 1925) (N.S.). In recent years the major Commonwealth countries have also claimed their freedom to depart from Privy Council decisions. E.g., Viro v. The Queen, 141 C.L.R. 88 (Austl. 1978); Calkins & Burke Ltd. v. Far E. S.S. Co., 72 D.L.R.3d 625. 635 (B.C. Sup. Ct. 1976). Indeed few Commonwealth countries still permit appeals to the Privy Council See Barwick, Some Observations on the Privy Council, [1985] 2 Malayan L.J. cxix, cxxviii-cxxix (listing appellate jurisdiction as of May 1, 1985); see also infra notes 184 & 218.

182. The Privy Council found the corporate relationship between the carrier and stevedore significant. See supra note 174. This relationship proved important in later decisions applying The Eurymedon. E.g., Calkins & Burke Ltd. v. Far E. S.S. Co., 72 D.L.R.3d 625 (B.C. Sup. Ct. 1976); Lummus Co. v. East Afr. Harbours Corp., [1978] 1 Lloyd's Rep. 317, 322-23 (Kenya High Ct. 1977).

183. Port Jackson Stevedoring Pty. v. Salmond & Spraggon (Austl.) Pty., (1981] 1 W.L.R. 138 (P.C. 1980) (Austl.).

184. Australia abolished appeals from the High Court to the Privy Council in 1975. Privy Council (Appeals from the High Court) Act 1975, No. 33, § 3 (Austl.). The Act did not. apply to pending cases. The New York Star was the last Privy Council appeal from the High Court. Blackshield, supra note 147, at 783. For a discussion of the litigation in the courts below, see infra notes 206-16 and accompanying text.

185. The New York Star, Port Jackson Stevedoring Pty. v. Salmond & Spraggon (Austl.) Pty., [1981] 1 W.L.R. at 143-44.

186. Id. at 143. One commentator argues that a Himalaya clause should be able to succeed without satisfying Lord Reid's requirements. The argument is based on technical English contract, tort, 8Ild bailment principles. Palmer, The Eurymedon in Australia -- Three Recent Cases, 135 New L.J. 919, 919-21 (1985). The most recent English decision involving a Himalaya clause, however, turned on whether the stevedore had concluded a separate contract with the cargo owner. Raymond Burke Motors Ltd. v. Mersey Docks & Harbour Co., [1986] 1 Lloyd's Rep. 155 (Q.B. (Com. Ct.) 1985).

187. The Elbe Maru, Nippon Yusen Kaisha v. International Import & Export Co., [1978] 1 Lloyd's Rep. 206 (Q.B. (Com. Ct.) 1977), illustrates a different method for a stevedore to obtain the carrier's Hague Rule exemptions under English law. The bill of lading provided that the consignee would make no claim against any sub-contractor of the carrier, and would indemnify the carrier against all consequences if a claim were made. Id. at 207. The carrier, by a separate contract, agreed to indemnify the sub-contractor against any claims of the consignee. When the consignee sued the sub-contractor (in breach of the bill of lading agreement), the carrier brought a second action against the consignee to stay the original suit. The court, exercising its discretion, granted the stay because the carrier had a real, financial interest in seeing the contract enforced. Id. at 210.

188. The results discussed in the text may be different under the Hague-Visby Rules, which are now in force in Great Britain. A new provision protects "a servant or agent of the carrier (such servant or agent not being an independent contractor)." British COGSA, supra note 20, sched., art. IV bis (2). The English courts have not yet discussed this provision, and the literature is divided on its meaning. Some commentators argue that a stevedore may, in appropriate circumstances, rely on the new rule. E.g., W. Tetley, Marine Cargo Claims 387 (2d ed. 1978); Gronfors, Why not Independent Contractors?, 1964 J. Bus. L. 25, 27; Zaphiriou, Amending the Hague Rules, 1971 J. Bus. L. 12, 14. Others argue that the Hague-Visby Rules do not protect stevedores. E.g., Chandler, A Comparison of COGSA, the Hague-Visby Rules, and the Hamburg Rules, 15 J. Mar. L. & Com. 233, 265 (1984); Diamond, The Hague-Visby Rules, 5 Lloyd's Mar. & Com. L.Q, 225, 249-51 (1978); Richardson, The Hague-Visby Rules -- A Carrier's View, in The Hague-Visby Rules and The Carriage of Goods by Sea Act, 1971, at 8 (Lloyd's of London Press Seminar 1977) [hereinafter Lloyd's Seminar]. Yet others find the new provision ambiguous, and are unable to predict how courts will resolve the issue. E.g., Maskell, The Influence of The New Rules On Contracts Of Carriage, in Lloyd's Seminar, supra, at 6; Mustill, Carriage of Goods by Sea Act. 1971, 11 Arkiv for Sjórett 684, 709 (1972).

189. Australia was one of the first countries to enact the Hague Rules, Sea-Carriage of Goods Act 1924, No. 22 (Austl.).

190. Gilbert, Stokes & Kerr Pty. v. Dalgety & Co., 48 N.S.W. St. R. 435 (N.S.W. Sup. Ct. 1948), overruled, Wilson v. Darling Island Stevedoring & Lighterage Co., 95 C.L.R. 43 (Austl. 1956).

191. Elder, Dempster & Co. v. Paterson, Zochonis & Co., 1924 A.C. 522. See supra notes 158-60 and accompanying text.

192. The court held that a servant or agent employed by the carrier to perform part of the contract or carriage takes possession of the goods for the purpose of carrying out. that contract as a bailee, for the cargo owner on the bill of lading's terms. Gilbert, Stokes & Kerr Pty. v. Dalgety & Co., 48 N.S.W. St. R. at 437.

193. Waters Trading Co. v. Dalgety & Co., 52 N.S.W. St. R. 4 (N.S.W. Sup. Ct. 1951) (in banco), overruled, Wilson v. Darling Island Stevedoring & Lighterage Co., 95 C.L.R. 43 (Austl. 1956).

194. E.g., Adler v. Dickinson, [1955] 1 Q.B. 158, 181 (C.A.); Pyrene Co. v. Scindia Navigation Co., [1954] 2 Q.B. 402, 421-22.

195. E.g., W.L. McNair & A. Mocetta, Scrutton on Charterparties 286-87 (16th ed. 1955).

196. Wilson v. Darling Island Stevedoring & Lighterage Co., 95 C.L.R. 43 (Austl. 1956).

197. Id. at 73 (Fullagar, J.); id. at 54-55 (Williams, J., dissenting).

198. Id. at 67 (Fullagar, J.); see also id. at 78-79.

199. Id. at 71.

200. Id. at 70-71; cf. id. at 73 (Approach is "alien to the whole spirit of the common law.").

201. Id. at 70.

202. Id. at 56-60 (Williams, J., dissenting); id. at 88.93 (Taylor, J., dissenting).

203. Justice Williams saw that the British and Australian COGSAs were "passed to give effect to the International Conference on Maritime Law which met in Brussels in 1922," and that they "made important alterations in the common law rights and liabilities of shipowners." Id. at 55 (Williams, J, dissenting). He nevertheless felt that "the rules of the common law remain applicable except in so far as they are expressly modified." Id.

204. Id. at 60; id. at 92-93 (Taylor, J., dissenting).

205. Justice Kitto, the third member of the majority, did suggest that the shipper could consent in the bill of lading to assume the risk that a stevedore might damage the goods. The stevedore would then have a volenti non fit injuria defense if the shipper brought suit in tort. Id. at 81-83 (Kitto, J.). This, in his view, was the basis of the Elder, Dempster decision. Id. at 83-85. He did not believe that such consent existed in Wilson, and (more significantly) even if it had, he did not believe such consent would bar an action by the consignee. Id. at 85-86. The suggestion, therefore, was of limited value. On the principal line of analysis discussed in the text, Justice Kitto agreed without elaboration. Id. at 80-81.

206. See supra note 169-82 and accompanying text.

207. Port Jackson Stevedoring Pty. v. Salmond & Spraggon (Austl.) Pty., 139 C.L.R 231 (Austl. 1978), rev'd, [1981] 1 W.L.R. 138 (P.C. 1980) (Austl.). Although the High Courts decision was reversed by the Privy Council, it is still a good indication of the Australian attitude toward the Himalaya clause. The High Court has already announced that it is not bound by Privy Council decisions. Viro v. The Queen, 141 C.L.R. 88, 93 (Austl. 1978) (Barwick, C.J.); id. at 118-22 (Gibbs, J.); id. at 129-30 (Stephen, J.); id. at 135-36 (Mason, J.); id. at 150-51 (Jacobs, J.); id. at 158-67 (Murphy, J.); id. at 172-74 (Aickin, J.); cf. supra note 181.

208. Salmond & Spraggon (Austl.) Pty. v. Joint Cargo Services Pty., [1977] 1 Lloyd's Rep. 445, 448 (N.S.W. Ct. App. 1976) (describing case at trial), aff'd on other grounds sub nom. Port Jackson Stevedoring Pty. v. Salmond & Spraggon (Austl.) Pty., 139 C.L.R. 231 (Austl. 1978), rev'd, [1981] 1 W.L.R. 138 (P.C. 1980) (Austl.).

209. Id. at 448.

210. Id. at 449 (Glass, J.); id. at 452 (Hutley, J.); id. at 453 (Mahoney, J.).

211. The stevedore's negligence had permitted the theft of the goods after they had been unloaded from the ship. Because unloading was already complete, the High Court concluded that the limitation clauses in the bill of lading no longer applied and that the consignee was thus free to bring suit against the stevedore more than a year after the accident. The New York Star, Port Jackson Stevedoring Pty. v. Salmond & Spraggon (Austl.) Pty., 139 C.L.R at 260-66 (Stephen, J.); id. at 280-83 (Mason & Jacobs, JJ.); id. at 285-86 (Murphy, J.).

212. Justices Mason and Jacobs argued that the bill of lading contained an offer by the shipper to extend the limitation clauses, and that the stevedore accepted this offer by discharging the goods. Id. at 271-74. Chief Justice Barwick, on the other hand, felt that the relevant bill of lading provision was an "arrangement" between the shipper and the stevedore for which the stevedore provided consideration by discharging the goods. Id. at 242-47. Although Barwick specifically rejected the offer and acceptance analysis of Mason and Jacobs's opinion, id. at 243, he admitted that under such an analysis the stevedore had accepted the offer and provided adequate consideration, id. at 247. In The Eurymedon, the Privy Council treated the "unilateral contract" analysis favored by Barwick as essentially the same as the "offer and acceptance" analysis favored by Mason and Jacobs. See supra note 177.

213. The New York Star, Port Jackson Stevedoring Pty. v. Salmond & Spraggon (Austl.) Pty., 139 C.L.R at 257-58 (Stephen, J.).

214. Id. at 284 (Murphy, J.).

215. Id. at 285. It is surprisingly unusual to find such blatant nationalism in judicial opinions construing the Hague Rules. See supra notes 6877 and accompanying text.

216. The New York Star, Port Jackson Stevedoring Pty. v. Salmond & Spraggon (Austl.) Pty., 139 C.L,R. at 285 (Murphy, J.).

217. Godina v. Patrick Operations Pty., [1984] 1 Lloyd's Rep. 333 (N.S.W. Ct. App. 1983).

218. When the New South Wales Court of Appeal followed the Privy Council's New York Star decision, a state court (unlike the High Court, see supra note 207) was bound by a Privy Council decision unless it was inconsistent with a High Court decision. H. Renfree. The Federal Judicial System of Australia 143 (1984); Geddes, The Authority of Privy Council Decisions in Australian Courts, 9 Fed. L. Rev. 427, 442-55 (1978). This rule may change, now that Australia has eliminated appeals from state courts to the Privy Council Australia Act 1986, No. 142, § 11 (Austl. 1985).

219. See supra notes 184 & 218.

220. J.P.M. de Koning, writing in the same year that the High Court decided The New York Star, predicted that

221. Canada enacted the Hague Rules two months after the United States did. Water Carriage of Goods Act. 1936, 1 Edw. 8, ch. 49, currently codified as Carriage of Goods by Water Act. Can. Rev. Stat. ch. C-15 (1970) [hereinafter Canadian COGSA].

222. Canadian Gen. Elec. Co. v. Pickford & Black Ltd., 14 D.L.R.3d 372, 374 (Can. 1970).

223. See supra notes 169-82 and accompanying text.

224. The most relevant decision of the Canadian Supreme Court during this period was a non-maritime case declining to apply The Eurymedon. Greenwood Shopping Plaza Ltd. v. Beat tie, 111 D.L.R.3d 257, 263-64 (Can. 1980). Although Greenwood reaffirmed the English third-party beneficiary rule, later courts treated the decision to distinguish The Eurymedon, rather than to reject it outright, as an endorsement or the Eurymedon principles. See, e.g., L. & B. Constr. Ltd. v. Northern Can. Power Comm'n, [1984] 6 W.W.R. 598, 616 (N.W.T. Sup. Ct.); Dyck v. Manitoba Snowmobile Ass'n Inc., 136 D.L.R.3d 11, 24-25 (Man. Ct. App. 1982).

225. Calkins & Burke Ltd. v. Far E. S.S. Co., 72 D.L.R.3d 625 (B.C. Sup. Ct. 1976).

226. The bill of lading extended the carrier's limitations to "all servants, agents and independent contractors including in particular, but not by way of limitation, any stevedores) used or employed by the carrier." Id. at 631. The bill of lading also specified the consideration and provided that, for purposes of extending the limitations, the carrier acted as agent on behalf of the intended beneficiaries and that they were to this extent parties to the bill of lading contract. Id.

227. Id. at 633-35. See supra note 174. The court also commented on the lack of binding force of a Privy Council decision, see supra note 181, and observed that all three of the judges of the New Zealand Court of Appeal and two judges in the Privy Council had disagreed with the result in The Eurymedon.

228. International Terminal Operators Ltd. v. Miida Elecs. Inc., 28 D.L.R. 4th 641 (Can. 1986); see infra note 240 and accompanying text.

229. Ceres Stevedoring Co. v. Eisen und Metall A.G., 72 D.L.R.3d 660, 659 (Que. Ct. App. 1976) (Owen, J.); id. at 661 (Tremblay, C.J.); id. at 674 (Lajoie, J.).

230. Id. at 669.70 (Owen, J.).

231. Miles Int'l Corp. v. Federal Commerce & Navigation Co., [1978] 1 Lloyd's Rep. 28.5 (Que. Super. Ct. 1977).

232. Article 1029 provides:

233. Miles Int'l Corp. v. Federal Commerce & Navigation Co., (1978) 1 Lloyd's Rep. at 290-92.

234. Id. at 290 (distinguishing Canadian Gen. Elec. Co. v. Pickford & Black Ltd., 14 D.L.R.3d 372 (Can. 1970).

235. Marubeni America Corp. v. Mitsui O.S.K. Lines, 96 D.L.R.3d 518 (Fed. Ct. Trial Div. 1979), aff'd in part & rev'd in part sub nom. Miida Elecs., Inc. v. Mitsui O.S.K. Lines 124 D.L.R. 3d 33 (Fed. Ct. App. 1981), aff'd in part & rev'd in part sub nom. International Terminal Operators Ltd. v. Miida Elecs. Inc., 28 D.L.R. 4th 641 (Can. 1986).

236. Id. at 532-34.

237. Miida Elecs., Inc. v. Mitsui O.S.K. Lines, 124 D.L.R.3d 33 (Fed. Ct. App. 1981), aff'd in part & rev'd in part sub nom. International Terminal Operators Ltd. v. Miida Elecs. Inc., 28 D.L.R.4th 641 (Can. 1986).

238. Id. at 45-46 (Le Dain, J.); id. at 51-52 (Lalande, D.J.).

239. Judge Le Dain concluded that the relevant clauses, which were substantially the same as the limitation clauses in, for example, The New York Star, did not apply to liability for negligence. Id. at 48. Judge Lalande decided that the relevant clauses gave the stevedore only "the benefit of the rights and immunities ... afforded by the Hague Rules ... . There is nothing in those rules that exonerates the carrier from liability for loss of goods by theft because of negligence in the care and custody of the goods after their discharge." Id. at 52. The Hague Rules, however, permit the carrier to exclude all liability after discharge, and the carrier had done so here. Both judges appear to have been influenced by the common law doctrine disfavoring exclusion of liability for negligence. Cf. supra note 146. See generally G. Treitel, The Law of Contract 172-74 (6th ed. 1983). The dissenter simply followed The Eurymedon and The New York Star, making no real attempt to justify this course despite his recognition that "the validity of [a Himalaya] clause is still a subject of discussion in Canadian law." Miida Electronics, 124 D.L.R.3d at 39 (Pratte, J., dissenting in port).

240. International Terminal Operators Ltd. v. Miida Elecs. Inc., 28 D.L.R. 4th 641 (Can. 1986).

241. Hans W. Baade provided useful advice and criticism, particularly on this section of the Article. Karsten Otte provided useful research assistance on this section.

242. West Germany incorporated the Hague Rules into its Commercial Code. See, e.g., Handelsgesetzbuch [HGB] § 660 (W. Ger.) (enacting Hague Rules art. 4(5) with limitation of 1250 marks per package or unit).

243. Schmidt, The Himalaya Clause under the Law of the Federal Republic of Germany, 19 Bur. Transport I,. 675, 677 & n. 12 (1984); id. at 679 & n. 19.

244. Section 328 provides, in relevant part: "A contract may stipulate performance for the benefit of a third party, so that the third party acquires the right directly to demand performance." Bürgerliches Gesetzbuch [BGB] § 328(1) (W. Ger.) (I. Forrester, S. Goren & H. Ilgen trans. 1975).

245. Similar results can be seen in other civil law countries that permit third-party beneficiary contracts. See, e.g., Sandstrom, The Limitation of the Stevedore's Liability. 1962 J. Bus. L. 340, 342-43 (Scandinavian courts would uphold Himalaya clause).

246. Some commentators view this as a simple example of an agreement in favor of a third party under section 328(1). See Schmidt, supra note 243, at 680 & n. 30 (citing authorities). This analysis is similar to that in Miles Int'l Corp. v. Federal Commerce & Navigation Co., [1978] 1 Lloyd's Rep. 285, 290-92 (Que. Super. Ct. 1977) (applying Que. Civ. Code art. 1029). See supra notes 231-34 and accompanying text. Such an approach may create theoretical problems under German law, however, so most commentators vie VI the extension of benefits 00 a third party as an application of section 328 by analogy. Schmidt. supra note 243, at 680 & nn. 31-33.

247. The issue is essentially whether a third party can claim the automatic protection that the Fifth Circuit upheld in Collins, see supra notes 105-19 and accompanying text, or that the New South Wales courts recognized in Gilbert, Stokes and Waters, see supra note 189-95 and accompanying text.

248. Schmidt, supra note 243, at 678 & n. 18. Schmidt cites three authors who disagree with his conclusion, id. at 678 n. 17, but their work predates the cases cited infra notes 256 & 257.

249. Adler v. Dickson, [1955] 1 Q.B. 158 (C.A. 1954). See supra note 101.

250. E.g., Judgment of Jan. 21, 1971, Bundesgerichtshof, 1971 Monatsschrift für Deutsches Recht [M.D.R.] 462 (W. Ger.); Judgment of July 7, 1960, Bundesgerichtshof, 1960 M.D.R. 907 (W. Ger.).

251. The basic delict provision of the West German Civil Code provides that "[a] person who ... negligently ... injures the ... property ... of another is bound to compensate him for any damage arising therefrom." BGB § 823(1) (I. Forrester, S. Goren & H. Ilgen trans. 1975).

252. Judgment of July 7, 1960, Bundesgerichtshof, 1960 M.D.R. 907 (W. Ger.).

253. Id. at 907.

254. Id. at 907-08. The court also held that the clause, being reasonable, was not contrary to public policy. Id. at 908.

255. See Judgment, of Jan. 21, 1971, Bundesgerichtshof, 1971 M.D.R. 462, 462.63 (W. Ger.) (citing Judgment of July 7, 1960, Bundesgerichtshof, 1960 M.D.R. 907 (W. Ger.).

256. Judgment of Apr. 28, 1977, Bundesgerichtshof, 1977 M.D.R. 819, 819 (W. Ger.).

257. The Bundesgerichtshof has continued to apply this standard. In a case involving on action against the master of the vessel, the court (citing the judgment discussed in the text) declared that it was a general rule of West German law that clauses protecting the carrier also protect third Parties if the purpose of the contract and the interests of the parties involved make the third party's non-liability appropriate and reasonable. Judgment of Nov. 26, 1979, Bundesgerichtshof, 1980 M.D.R. 648, 648 (W. Ger.) (citing Judgment of Apr. 28, 1977, Bundesgerichtshof, 1977 M.D.R. 819 (W. Ger.).

258. Judgment of Apr. 28, 1977, Bundesgerichtshof, 1977 M.D.R. at 819.

259. See supra note 246.

260. Cf. Sandstrom, supra note 245, at 343-49 (arguing that Scandinavian courts could and should imply benefits of Himalaya clause even if bill of lading does not contain one).

261. See supra note 244.

262. See supra notes 152-57 and accompanying text.

263. Judgment of July 7, 1960, Bundesgerichtshof, 1960 MD.R. 907, 907 (W. Ger.). See supra note 254 and accompanying text.

264. See supra note 146.

265. Under U.S. COGSA section 4(5), for example, the package limitation is 500 U.S. dollars. Under the corresponding Canadian provision, it is 500 Canadian dollars. See Canadian COGSA, supra note 221, sched., art. 4(5). And under HGB section 660, the limitation is 1250 West German marks. Cf. supra notes 91 & 242.

266. Choice of forum clauses, also known as jurisdiction or forum selection clauses, specify the court in which a case shall be heard -- generally to the exclusion of any other court. See generally R. Weintraub, Commentary on the Conflict of Laws 223-26 (3d ed. 1986). It is the "exclusive" jurisdiction clause that is of interest here.

267. The Hollandia, 1983 A.C. 565, 570 (1982). See infra notes 340-46 and accompanying text.

268. The concern here is with the normal situation in which the currier, as the defendant in a law suit in a forum other than that specified in the bill of lading, seeks the benefit of the choice of forum clause that it drafted. If the shipper raises the choice of forum clause. e.g., Roach v. Hapag-Lloyd, A.G. 358 F. Supp. 481 (N.D. Cat 1973). Article 3(8) concerns do not apply, and there is no reason to invalidate the choice of forum clause.

269. See supra note 25.

270. See supra note 28 and accompanying text.

271. Hague Rules, supra note 4, art. 3(8).

272. See infra notes 305-21 and accompanying text. In the United States, COGSA section 3(8), 46 U.S.C. App. § 1303(8) (Supp. III 1985), is substantially identical to Article 3(8).

273. See infra notes 324-56 and accompanying text.

274. It is common for statutes to specify the compulsory application of the Hague Rules in certain circumstances. The enacting clause of the U.S. COGSA, for example, provides "[t]hat every bill of lading ... for the carriage of goods by sea to or from ports of the United States, in foreign trade, shall have effect subject to the provisions of this Act. " 46 U.S.C. App. § 1300 (Supp. III 1985); cf. COGSA § 13, 46 U.S.C. App. § 1312 (Supp. III 1985). In England, Parliament provided that the Hague-Visby Rules "shall have the force of law," British COGSA, supra note 20, § 1(2), and shall apply to all shipments from British ports, id. § 1(3). These statutes, unlike those discussed in the text, are compulsory choice of law provisions.

275. Sea-Carriage of Goods Act 1924, No. 22, § 9(2) (Austl.). Section 9(2) applies "to the carriage of goods from any place outside Australia to any place in Australia." Id. Section 9(1) applies in similar terms to shipments from Australia to foreign countries.

276. Wilson v. Compagnie des Messageries Maritimes, [1954] 2 Lloyd's Rep. 544 (Austl.).

277. Before the Hague Rules, when there was no semblance of international uniformity, the original Canadian COGSA had a similar prohibition Water Carriage of Goods Act 1910, 9-10 Edw. 7, ch. 61, § 5, superseded by Canadian COGSA, supra note 221, § 4. Today Canada follows English law on choice of forum clauses. E.g., Nichimen Corp. v. Gearbulk Ltd., 1986 A.M.C. 593 (Can. Fed. Ct. Trial Div. 1985) (enforcing clause under English principles); Burrard-Yarrows Corp. v. The Hoegh Merchant, [1982] 1 F.C. 248 (Trial Div. 1981) (same); A.S. May & Co. v. Robert Redford Co., 6 D.L.R.3d 288 (Ont. H. CL 1969) (denying enforcement under English principles); Polito v. Gestioni Esercizio Navi Sicilia Gens, 1960 Can. Exch. 233 (Que. Adm. Dist.) (same).

278. Code de Commerce Maritime art. 212 (Lebanon 1947), translated in Lebanese Law of Merchant Shipping (S.E.A. Hakim trans. 1965).

279. Sea Carriage of Goods Amendment Act 1968. No. 17. § 3 (N.Z.) (adding section 11A to Sea Carriage of Goods Act 1940, 4 Geo. 6, No. 31 (N.Z.)).

280. Merchant Shipping Act, 1951, No. 57, § 3101(6) (S. Afr.).

281. Code de Commerce Maritime art. 212 (Syria 1950).

282. See The Bremen v. Zapata Off-Shore Co., 407 U.S. 1, 9-10 & no. 10 (1972); Note, Validity of Contractual Stipulation Giving Exclusive Jurisdiction to the Courts of One State, 45 Yale L.J. 1150, 1150 & n. 2 (1936).

283. See A. Knauth, The American Law of Ocean Bills of Lading 238 (4th ed. 1953).

284. E.g., Krenger v. Pennsylvania R.R., 174 F.2d 556, 560-61 (2d Cir.) (L. Hand. C.J., concurring), cert. denied, 338 U.S. 866 (1949).

285. The Bremen v. Zapata Off-Shore Co., 407 U.S. 1 (1972).

286. The Court explicitly limited its holding to federal courts sitting in admiralty, id. at 10, but later courts have applied The Bremen more expansively. See, e.g., Benson v. Interstate Battery Sys. of America, Inc., 683 F.2d 718, 720-22 (2d Cir. 1982), and cases cited therein.

287. The Court said it would not enforce a clause when enforcement would be "unreasonable and unjust," or when the clause was invalid through "fraud or overreaching," or when enforcement "would contravene a strong public policy ... declared by statute or by judicial decision." The Bremen v. Zapata Off-Shore Co., 407 U.S. at 15.

288. Id. at 9.

289. The Bremen Court explicitly left open the possibility that choice of forum clauses might fail under COGSA. See id. at 10 n. 11 (noting that The Bremen was not subject to COGSA; distinguishing Indussa Corp. v. S.S. Ratlborg, 377 F.2d 200 (2d Cir. 1967) (en banc)).

290. Wm. H. Muller & Co. v. Swedish Am. Line, 224 F.2d 806 (2d Cir.), cert. denied, 360 U.S. 903 (1955), overruled, Indussa Corp. v. S.S. Ranborg, 377 F.2d 200 (2d Cir. 1967) (en banc).

291. Id. at 807.

292. Id.

293. Id.

294. Id. at 808.

295. Id.

296. The en banc Second Circuit later criticized the Muller court for "lean[ing] too heavily on general principles of contract law" rather than on COGSA. Indussa Corp. v. S.S. Ranborg, 377 F.2d 200, 202 (2d Cir. 1967) (en banc).

297. E.g., Amicale Indus., Inc. v. S.S. Rantum, 259 F. Supp. 534 (D. S.C. 1966); Pakhuismeesteren, S.A. v. S.S. Goettingen, 225 F. Supp. 888 (S.D.N.Y. 1963); Takemura & Co. v. The S.S. Tsuneshima Maru, 197 F. Supp. 909 (S.D.N.Y. 1961); Schuster Naval Stores Co. v. Ozean/Stinnis Lines, 1962 A.M.C. 999 (D. Ga. 1961); Aetna Ins. Co. v. The Satrustegui, 171 F. Supp. 33, modified, 174 F. Supp. 934 (D. P.R. 1e159); see also infra note 299 and accompanying text. But see Sociedade Brasileira de Intercambio Comercial e Industrial, Ltda. v. S.S. Punta Del Este, 135 F. Supp. 394 (D. N.J. 1951) (distinguishing Muller).

298. Carbon Black Export, Inc. v. The S.S. Monrosa, 254 F.2d 297 (5th Cir. 1958). celt. dismissed, 359 U.S. 180 (1959).

299. Id. at 299 & n. 2 (quoting district court opinion).

300. The clause provided "that no legal proceedings may be brought against the Captain or ship owners or their agents in respect to any loss of or damage to any Goods herein specified, except in Genoa, it being understood and agreed that every other Tribunal in the place or places where the goods were shipped or landed is; incompetent, notwithstanding that the ship may be legally represented there." Id. at 299.

301. Id. at 300-01.

302. Brier of Appellant at 13-17. Carbon Black Export Inc. (No. 16,667). The carrier responded to the section 3(8) argument in detail Brier of Appellees at 15-18.

303. Carbon Black Export, Inc. v. The S.S. Monrosa, 254 F.2d at 299 n. 3.

304. The Supreme Court granted certiorari to resolve the apparent conflict between the Fifth Circuit's Carbon Black decision and the Second Circuit's Muller decision. but after oral argument the Court dismissed the writ. The Monrosa v. Carbon Black Export, Inc., 359 U.S. 180 (1959). The parties had briefed the section 3(8) question for the Court. Brief for the Respondent at 21-25, The Monrosa (No. 178); Reply Brier for the Petitioners at 10-14; cf. Brier of Amicus Curiae American Institute of Marine Underwriters at 8-9 (arguing that choice of forum clause violates COGSA enacting clause, 46 U.S.C. § 1300); id. at 11-12 (arguing that choice of forum clause violates COGSA section 8); Reply for Petitioners to Brier of Amicus Curiae at 1-4.

305. Indussa Corp. v. S.S. Ranborg, 377 F.2d 200 (2d Cir. 1967) (en banc).

306. Id. at 201.

307. Id.

308. Id. at 200.

309. Judge Moore, arguing that Muller was distinguishable, concurred in the result. Id. at 204-05.

310. Id. at 203 (quoting G. Gilmore & C. Black, supra note 27, at 125 n. 23 (1st ed. 1957)).

311. Id. at 203-04.

312. Id. at 202.

313. Id. at 204.

314. 46 U.S.C. App. § 1300 (Supp. III 1985).

315. Id. § 1312.

316. In many cases a foreign court will clearly not; apply the U.S. COGSA so broadly, for foreign COGSAs provide for their own compulsory application in situations that may conflict with the U.S. COGSA. See, e.g., British COGSA, supra note 20, § 1(3). An English court would be required to apply the British COGSA to a shipment from London to New York, while a U.S. court would be required to apply the U.S. COGSA.

317. The "U.S. Trade" clause essentially provided that if the bill of lading were subject to COGSA, then COGSA would govern. Indussa Corp. S.S. Ranborg, 377 F.2d at 201. Under COGSA, of course, the bill of lading would be subject to COGSA. But under Belgian law the bill of lading would be subject to the Belgian COGSA (because the Goods were shipped from Belgium) and the U.S. Trade clause would be invalid. See infra notes 357.75 and accompanying text. As the contract was made in Belgium, it is logical that a court might apply Belgian law. Furthermore, the "Paramount Clause" declared that the Hague Rules as enacted in the country of shipment (Belgium) applied. Indussa, 377 F.2d at 200-01. Indeed the plaintiff thought Belgian law relevant enough to file an affidavit of a Belgian lawyer on Belgian law with the district court. Id, at 201. Finally, the "Jurisdiction Clause" effectively provided for Norwegian law "except as provided elsewhere herein." Id. The carrier thought Norwegian law relevant enough to file an affidavit of a Norwegian lawyer on Norwegian law with the district court. Id.

318. Indussa Corp. v. S.S. Ranborg, 377 F.2d at 203. The court also relied on Knott v. Botany Mills, 179 U.S. 69 (1900), a decision under the U.S. Harter Act.

319. Indussa left open the possibility that a court; could decline jurisdiction in a COGSA case under the doctrine of forum non conueniens. Indussa Corp. v. S.S. Ranborg, 377 F.2d at 204. See also Union Ins. Soc'y of Canton, Ltd. v. S.S. Elikon, 642 F.2d 721, 725 (4th Cir. 1981) (refusing to enforce choice of forum clause but remanding for consideration under doctrine of forum non conueniens); cf. Philippine Packing Corp. v. Maritime Co. of Phil, 519 F.2d 811 (9th Cir. 1975) (per curiam) (applying doctrine of forum non conueniens in case not subject to COGSA).

320. Courts have upheld choice of forum clauses in cases where COGSA did not apply as a matter of law, but had been incorporated by contract. See e.g., North River Ins. Co. v. Fed Sea/Fed Pac Line, 647 F.2d 985, 988-89 (9th Cir. 1981) (citing cases), cert. denied, 455 U.S, 948 (1982).

321. See. e.g., Union Ins. Soc'y of Canton, Ltd. v. S.S. Elikon, 642 F.2d 721 (4th Cir. 1981); Pacific Lumber & Shipping Co. v. Star Shipping A/S, 464 F. Supp. 1314 (W.D. Wash. 1979); Mitsui & Co. v. M/V Glory River, 464 F. Supp. 1004 (W.D. Wash. 1978); Northern Assurance Co. v. M/V Caspian Career, 1977 A.M.C. 421 (N.D. Cal.); cf. Roach v. Hapag-Lloyd, A.G., 358 F. Supp. 481 (N.D. Cal. 1973) (choice of forum clause upheld when shipper sought to rely on it).

322. See, e.g., Dukane Fabrics Int'l, Inc. v. M.V. Hreljin, 600 F. Supp. 202 (S.D.N.Y. 1985) (shipment from Italy to Nigeria, chosen forum in Italy); Ampac Trading Co. v. MN Ming Summer, 566 F. Supp. 104 (W.D. Wash. 1983) (shipment from Canada to Saudi Arabia, chosen forum in Japan); Galaxy Export Carp. v. M.V. Hektor, 1983 A.M.C. 2637 (S.D.N. Y.) (shipment from Jamaica to Nigeria, chosen forum in Italy); Paterson, Zochonis (U.K.) Ltd. v. Compania United Arrows, 493 F. Supp. 626 (S.D.N.Y. 1980) (shipment from Hong Kong to Nigeria chosen forum in Japan). Italy, Nigeria, Canada, Japan, Jamaica, and Hong Kong had all ratified or acceded to the Hague Rules.

323. See supra notes 314-18 and accompanying text.

324. Stephen Denning, a vigorous advocate of the carrier's position on this issue, cites the English approach to the enforcement of choice of forum clauses as a model for the United States to follow. See Denning, Choice of Forum Clauses in Bills of Lading, 2 J. Mar. L & Com. 17, 17-23, 38-39 (1970).

325. The "Indussa argument" is used here to refer to the broadest aspect of Indussa Corp. v. S.S. Ranborg, 377 F.2d 200 (2d Cir. 1967) (en banc), i.e., the contention that a choice of forum clause is an inherent violation of Article 3(8) of the Hague Rules, even if the chosen forum would apply a version of the Hague Rules. See supra notes 310-13 and accompanying text.

326. The British 1924 COGSA, supra note 20, incorporated the Hague Rules. The current British COGSA, supra note 20, enacts the Hague-Visby Rules. Article 3(8) is identical in the two versions.

327. E.g., The Hollandia, 1983 A.C. 565 (1982); see infra notes 340-46 and accompanying text.

328. E.g., Aratra Potato Co. v. Egyptian Navigation. Co., (1981) 2 Lloyd's Rep. 119 (C.A.); see infra notes 353-55 and accompanying text.

329. E.g., The Eleftheria, 1970 P. 94, 100, 105 (1969) (granting stay subject. to provision for security to answer judgment in chosen forum).

330. Maharani Woollen Mills Co. v. Anchor Line, [1927] 29 Lloyd's Rep. 169 (C.A.).

331. British 1924 COGSA, supra note 20.

332. Maharani Woollen Mills Co. v. Anchor Line, (1927) 29 Lloyd's Rep. at. 169.

333. Id.

334. The Indian plaintiff's English underwriter had brought the action in England because it had insufficient time to file an action in Bombay before the expiration of the one year permitted for suit under Hague Rules Article 3(6).

335. India remained a part of the British Empire until 1947, and the Privy Council retained appellate jurisdiction over the Indian courts until 1949. See M.P. Jain, Outlines of Indian Legal History 440-86 (2d ed. 1966); Eddy, India and the Privy Council: The Last Appeal. 66 Law Q. Rev. 206 (1950).

336. E.g., Lindblom, On the Distinction Between Procedural and Substantive Law, 18 Scandinavian Stud. L. 109, 113-16 (1974). But see In re K.S. Line Corp., 596 F. Supp. 1268, 1272 (D. Alaska 1984) (substance/procedure definition not commonly used in Korean legal system).

337. See. e.g., Foster, Some Defects in the English Rules of Conflict of Laws, 16 Brit. Y.B. Int'l L. 84, 101-02 (1935) (criticizing English law for overly broad "procedure" classification). The problems with attempting to establish a clear distinction between procedural and substantive rules, or with assigning legal consequences solely on the basis of such a distinction, are well recognized in the United States. E.g., Restatement (Second) of Conflict of Laws § 122 comments b & c (1969).

338. E.g., Miliangos v. George Frank (Textiles) Ltd., 1976 A.C. 443, 465 (1975) (Lord Wilberforce) (declining to apply English procedure rules to impair creditor's substantial rights under Swiss law). But cf. Black-Clawson Int'l Ltd. v. Papienwerke Waldhof-Asehaffenburg AG, 1975 A.C. 591 (treating statute of limitations as procedural rule to permit plaintiff to enforce claim under West German law that German court had held time-barred).

339. Counsel raised the Article 3(8) issue in The Media, [1931] 41 Lloyd's Rep. 80, 81 (K.B.), but the court did not even mention the point in its opinion.

340. The Hollandia, 1983 A.C. 565 (1982).

341. The Benarty, 1985 Q.B. 325 (C.A. 1984).

342. The Hollandia, 1983 A.C. at 570. The relevant Clause is quoted above. See supra note 267 and accompanying text.

343. Article 4(5) of the Hague-Visby Rules, which would have set liability on the basis of weight in this case, was the applicable provision. See supra note 91.

344. The Netherlands had not yet ratified the Visby Protocol, supra note 20, so the court would have applied the package limitation of the unamended Article 4(5) of the Hague Rules. See supra note 91.

345. The House also gave some weight to British COGSA, supra note 20, § 1(2) & (3). See The Hollandia, 1983 A.C. at 571, 577. This provision requires English courts to apply the Hague-Visby Rules with "the force of law" to all shipments from British ports. It is thus analogous to COGSA's enacting clause and section 13. See supra note 274; cf. supra notes 314-18 and accompanying text (discussing Indussa court's reliance on COGSA's enacting Clause and section 13).

346. The Hollandia, 1983 A.C. at 574-75.

347. Article 474 of the Indonesian Commercial Code is a limitation statute analogous to the English Merchant Shipping Act, 1894, 57 & 58 Vict., ch. 60, § 503. The U.S. Limitation Act does not consider the tonnage of the vessel in cargo damage cases. See 46 U.S.C. App. §§ 181-89 (Supp. III 1985).

348. The Benarty, 1985 Q.B. at 334, 340.

349. Id. at 338.

350. Article 8 of the Hague Rules provides:

351. The Benarty, 1985 Q.B. at 342-43.

352. Id. at 344.

353. The principles guiding the court's discretion were established in The Eleftheria, 1970 P. 94, 99-100 (1969), and endorsed by the Court of Apple in Arntra Potato Co. v. Egyptian Navigation Co., [1981] 2 Lloyd's Rep. 119, 127, 129 (C.A.). Under The Eleftheria, an English court considers essentially the same factors as a U.S. court would consider in ruling on a forum non conveniens motion, including such things as the availability of witnesses and evidence, the governing law, and the likelihood of a fair trial in the chosen forum. Cf. Edelman, Forum Non Conveniens: Its Application in Admiralty Law, 15 J. Mar. L. & Com. 517, 522-23 (1984). English forum non conveniens law is more restrictive. See The Abidin Daver, 1984 A.C. 398.

354. See, e.g., The Benarty, 1985 Q.B. 325, 343, 349 (finding trial in Indonesia appropriate in view of limited nature of issue to be tried): The Makefjell, [1976] 2 Lloyd's Rep. 29 (C.A.) (enforcing clause requiring claims against the carrier to be decided in Oalo, Norway): The Eleftheria, 1970 P. 94, 100,105 (1969) (enforcing Clause requiring disputes to be decided in Greece): cf. The Sennar, [1985] 1 W .L.R. 490 (H.L.) (enforcing clause giving exclusive jurisdiction to Sudanese courts when Dutch courts had upheld clause in earlier action between same parties).

355. See, e.g., Aratra Potato Co. v. Egyptian Navigation Co., [1981] 2 Lloyd's Rep. 119 (C.A.) (declining to enforce Clause choosing Egypt as forum); The Adolf Warski, [1976] 2 Lloyd's Rep. 241 (C.A.) (declining to enforce clause choosing Poland as forum); The Fehmarn, [1958] 1 All E.R. 333 (C.A. 1957) (declining to enforce clause choosing Soviet Union as forum); The Athenee, [1922] 11 Lloyd's Rep. 6 (C.A.) (declining to enforce clause choosing the Tribunal of Commerce at Marseilles, France, as forum): The Vishva Prabha, [1979] 2 Lloyd's Rep. 286 (Q.B.) (declining to enforce clause choosing India as forum); The Vestris, [1932] 43 Lloyd's Rep. 86 (Adm.) (declining to enforce clause choosing New York as forum).

356. Under the Civil Jurisdiction and Judgments Act, 1982, ch. 27, § 2 & sched. 1, the Brussels Judgments Convention now has the force of law in England. See infra note 398. It may compel the English courts to enforce a choice of forum clause specifying jurisdiction in the courts of another E.E.C. member state if the requirements of E.E.C. law are satisfied. See infra notes 397-411 and accompanying text. The convention has only been in force since January 1, however, so it is still too early to say how the English courts will apply the Convention. See Civil Jurisdiction and Judgments Act 1982 (Commencement No. 3) Order 1986, S.I. 1986, No. 2044.

357. Bernard Rudden provided useful advice and criticism on this section of the Article.

358. The bulk of the litigation on this issue has been in the Commercial Court ("Tribunal de commerce" or "Rechtbank van Koophandel") of Antwerp or on appeal in the Court of Appeals ("Cour d'appel" or "Hof van Beroep") in Brussels. For the sake of consistency and to avoid confusion, citations here will use the French names of the courts regardless of the language of the cited report.

359. See generally Husserl, Public Policy and Ordre Public, 25 Va. L. Rev. 37 (1938) (contrasting Anglo-American public policy and Continental ordre public); id. at 61-65 (discussing ordre public in Belgian and French law); Forde, The "Ordre Public". Exception and Adjudicative Jurisdiction Conventions, 29 Int'l & Comp. L.Q. 259, 259 (1980) (discussing differing interpretations of ordre public under civil and common law),

360. Code de Commerce liv. II, tit. II, art. 91, § III, 80º.

361. See, e.g., The Monte Cinco, Judgment of Jan. 23, 1969, Tribunal de commerce, Antwerp, 1969 ,Jurisprudence du Port d'Anvers [Jur. Port Anv.] 170, 172, reprinted in 1969 Uniform L. Cases 184, 186 (holding choice of forum. clause void as contrary to Belgian public policy ("contraire a l'ordre public")).

362. See, e.g., The Frieda, Judgment of Dec. 10, 1970, Tribunal de commerce, Antwerp, 1970 Jur. Port Anv. 390, reprinted in 1971 Uniform L. Cases 206 (recognizing Canadian forum); The Costa Rica, Judgment of Mar. 29, 1960, Tribunal de commerce, Antwerp, 1961 Jur. Port Anv. 55, reprinted in 1961 Uniform L. Cases 264 (recognizing Nicaraguan forum in action involving less than $400).

363. See, e.g., The Oklahoma, Judgment of Mar. 1, 1962, Cour d'appel, Brussels, 1962 Jur. Port Anv. 232, reprinted in 1963 Uniform L. Cases 16 (refusing to enforce clause indicating Copenhagen as forum when Danish court would have been free to apply Belgian, Italian, or Scandinavian versions of Hague Rules); Judgment of Mar. 8, 1956, Cour d'appel, Ghent, 1957 Pasicrisie Belge II 64 (refusing to enforce Bordeaux as forum when prior French case had applied Hague Rules in lieu of Belgian law).

364. Judgment of June 3, 1970, Tribunal de commerce, Antwerp, 1970 Jur. Port Anv. 377, reprinted in 1971 Uniform L. Cases 200.

365. Compare The Lichtenfels, Judgment of Mar. 14, 1960, Tribunal de commerce, Antwerp, 1960 Jur. Port Anv. 276, 280, 282, reprinted in 1961 Uniform L. Cases 29, 32. 34 (choice of forum clause valid because reasonable certainty ["certitude raisonnable"] German court would apply Belgian law) with The Rhodos, Judgment of Mar. 8, 1960, Tribunal de commerce, Antwerp, 1960 Jur. Port Anv. 323, 325, reprinted in 1961 Uniform L. Cases 25, 26 (choice of forum clause null and void because no absolute certainty ["l'absolue certitude"] Danish court would apply Belgian interpretation of Hague Rules).

366. The Atlanta, Judgment of Sept. 4, 1973, Tribunal de commerce, Antwerp, 9 Eur. Transport L. 269 (1974); The Brandenburg, Judgment of Oct. 20, 1966, Tribunal de commerce, Antwerp, 4 Eur. Transport L. 131 (1969).

367. The Marillia, Judgment of Oct. 31, 1974, Cour d'appel, Brussels, 10 Eur. Transport L. 50 (1975), reprinted in [1975] 1 Uniform L. Rev. 203; The Germania, Judgment of Oct. 5, 1963, Cour d'appel, Brussels, 1964 Jur. Port Anv. 129, reprinted in 1964 Uniform L. Cases 203; The Puerto Somoza, Judgment of May 19, 1961, Cour d'appel, Brussels, 1961 Jur. Port Anv. 313, reprinted in 1962 Uniform L. Cases 12; cf. supra note 365.

368. The Silvergate, Judgment of Jan. 22, 1965, Cour d'appel, Brussels, 1965 Jur. Port Anv. 146, reprinted in 1966 Uniform L. Cases 157; The Delia, Judgment of Feb. 8, 1962. Cour d'appel, Brussels, 1962 Jur. Port Anv. 143, reprinted in 1963 Uniform L. Cases 14; The Griesheim, Judgment of Nov. 30, 1961, Cour d'appel, Brussels, 1962 Jur. Port Anv. 117, reprinted in 1962 Uniform L. Cases 236.

369. E.g., The Silvergate, Judgment of Jan. 22, l965, Cour d'appel, Brussels, 1965 Jur. Port Anv. 146, reprinted in 1966 Uniform L. Cases 157; The Loodsrecht, Judgment of Mar. 2, 1961, Cour d'appel, Brussels, 1961 Jur. Port Anv. 296, reprinted in 1962 Uniform L. Cases 5; The Aludro, Judgment of Mar. 14, 1968, Tribunal de commerce, Antwerp, 1969 Jur. Port Anv. 43, reprinted in 1969 Uniform L. Cases 178; The Songhkla, Judgment of July 27, 1956, Tribunal de commerce, Antwerp, 1958 Jur. Port Anv. 72, reprinted in 1959 Uniform L. Cases 16; see also The Lekhaven, Judgment of Dec. 24. 1965, Cour d'appel, Brussels, 1966 Jur. Port Anv. 118, reprinted in 1967 Uniform L. Cases 35 (choice of forum clause void in, absence of clause paramount requiring application of Belgian law); The Oklahoma, Judgment of Mar. 1, 1962, Cour d'appel, Brussels, 1962 Jur. Port Anv. 232, reprinted in 1963 Uniform L. Cases 16 (same).

370. The Travestein, Judgment of June 18, 1970, Landgericht, Bremen, 7 Eur. Transport L. 330 (1972). A Belgian court had previously declined jurisdiction in the case on the basis of a choice of forum clause. The Travestein, Judgment. of Jan. 22, 1969, Tribunal de commerce, Antwerp, 7 Eur. Transport L. 342 (1972).

371. The Ganymedes, Judgment of Dec. 21, 1971, Alrondissementsrechtbank [Rb.], Rotterdam, 1974 Weekblad voor Privaatrecht, Notariaat en Registratie [WPNR] 852, 7 Eur. Transport L. 356 (1972); cf. The Alnati, Judgment of May 13, 1966, Hoge Raad der Nederlanden, Neth., 1967 Nederlandse Jurisprudentie 16, 2 Eur. Transport L. 707 (1967) (applying Dutch law in preference to Belgian law in case between Dutch parties first brought in Holland).

372. E.g., The Themis, Judgment of Apr. 4, 1974, Tribunal de commerce, Antwerp, 9 Eur. Transport L. 433 (1974); La Concorde, Judgment of May 8, 1973, Tribunal de commerce, Antwerp, 8 Eur. Transport L. 497 (1973); The Atlanta, Judgment of Mar. 21, 1972, Tribunal de commerce, Antwerp, 7 Eur. Transpore L. 831 (1972); The Regine, Judgment of Feb. 23, 1972, Tribunal de commerce, Antwerp, 7 Eur. Transport L. 345 (1972).

373. E.g., The Ruhrstein, Judgment of Dec. 7, 1972, Cour d'appel, Brussels, 8 Eur. Transport L. 281, 285 (1973) (citing The Trauestein, Judgment of June 18, 1970, Landgericht, Bremen, 7 Eur. Transport L. 330 (1972); Judgment of May 16, 1983, Tribunal de commerce, Antwerp, 20 Eur. Transport L. 82, 87 (1985) (same); The Tokyo Bay, Judgment of Sept. 10, 1979, Tribunal de commerce, Antwerp, 14 Eur. Transport L. 790, 798 (1979) (same); The Thameshauen, Judgment of Apr. 15, 1975, Tribunal de commerce, Antwerp, 11 Eur. Transport L. 92, 100 (1976) (same); The Atlanta, Judgment of Sept. 4, 1973, Tribunal de commerce, Antwerp, 9 Eur. Transport L. 269, 27,1 (1974) (same); La Concorde, Judgment of May 8, 1973, Tribunal de commerce, Antwerp, 8 Eur. Transport L. 497, 499 (1973) (same); The Regine, Judgment of Feb. 23, 1972, Tribunal de commerce, Antwerp, 7 Eur. Transport L. 345, 347 (1972) (same).

374. E.g., The Thameshauen, Judgment of Apr. 15, 1975, Tribunal de commerce, Antwerp, 11 Eur. Transport L. 92, 100 (1976) (citing The Ganymedes, Judgment of Dec. 21, 1971, Rb., Rotterdam, 1974 WPNR 862, 7 Eur. Transport L. 356 (1972); The Themis, Judgment of Apr. 4, 1974, Tribunal de commerce, Antwerp" 9 Eur. Transport L. 433, 438 (1974) (same); The Atlanta, Judgment of Sept. 4, 1973, Tribunal de commerce, Antwerp, 9 Eur. Transport L. 269, 274 (1974) (same); La Concorde, Judgment of May 8, 1973, Tribunal de commerce, Antwerp, 8 Eur. Transport L. 497, 499 (Jl97:3) (same); The Kennemerland, Judgment of July 27, 1972, Tribunal de commerce, Antwerp, 7 Eur. Transport L. 1017, 1026-27 (1972) (same).

375. Belgium is a member of the European Economic Community, so the Belgian court may be compelled to enforce a choice of forum clause specifying jurisdiction in the courts of another member state if the requirements of E.E.C:. law are satisfied. See infra notes 397-411 and accompanying text.

376. See, e.g., Chowdhury v. Mitsui O.S.K. Lines, 1970 Pak. Legal Dec. (Sup. Ct.) 373, 384-89, [1970] 2 Lloyd's Rep. 272, 277-80 (Pak.) (discussing English cases); Abdul Razak v. East Asiatic Co., 5 Dacca L.R. 394, 398-403 (H. Ct. E. Pak. 1953) (same).

377. Carriage of Goods by Sea Act, 1925, No. XXVI (Pak.).

378. Contract Act, 1872, Act IX, § 28 (Pak.).

379. In 1971 East Pakistan became Bangladesh, an independent country. Since Bangladesh continues to recognize the law in force prior to independence, Bangladesh Const. §§ 149, 152, the discussion here applies to both Parisian and Bangladesh. Indeed the Supreme Court of Bangladesh has explicitly recognized the validity of "pre-liberation" decisions of the Supreme Court of Pakistan. Jamal Jute Baling & Co. M. Sarkis & Sons, 2 Bangladesh Sup. Ct. Rep. 27 (1977).

380. Swedish E. Asia Co. v. Masud Reza & Co., 21 Dacca L.R. 343 (H. Ct. E. Pak. 1969) ("Norwegian, Swedish or Danish Courts"); British India Steam Navigation Co. v. ALA. Wadud & Co., 1968 Pak. Legal Dec. (Dacca) 860, 20 Dacca L.R. 367 (H. Ct. E. Pak. 1968) (England); British India Steam Navigation Co. v. Chodhury, 19 Dacca L.R. 54 (H. Ct. E. Pak. 1966) (England); Osaka Shosen Kaisha O.S.K. Line, Japan v. East Pakistan, 19S5 Pak. Legal Dec. (Dacca) 481, 17 Dacca L.R. 659 (H. Ct. E. Pak. 1964) (Japan); Narayanganj Iron Works Ltd. v. Pakistan, 15 Dacca L.R. 434 (H. Ct. E. Pak. 1963) (England); Tar Muhammad & Co. v. Federation of Pakistan, 9 Dacca L.R. 197 (H. Ct. E. Pak. 1957) (Karachi); Abdul Razak v. East Asiatic Co., 5 Dacca L.R. 394 (H. Ct. E. Pak. 1953) (Denmark). All of these cases were overruled by Chowdhury v. Mitsui O.S.K. Lines, 1970 Pak. Legal Dec. (Sup. Ct.) 373, [1970] 2 Lloyd's Rep. 272 (Pak.). See, infra notes 388-96 and accompanying text.

381. Narayanganj Iron Works Ltd. v. Pakistan, 15 Dacca L.R. 434 (H. Ct. E. Pak. 1963), overruled, Chowdhury v. Mitsui O.S.K. Lines, 1970 Pak. Legal Dec. (Sup. Ct.) 373, [1970] 2 Lloyd's Rep. 272 (Pak.).

382. Id. at 437.

383. Id. at 436-37.

384. Contract Act, 1872, Act IX, § 28 (Pak.). The Narayanganj Iron Works court did not consider the two arbitration exceptions relevant.

385. See cases cited supra note 380.

386. Narayanganj Iron Works Ltd. v. Pakistan, 15 Dacca L.R. at 437.

387. In Ali Akbar v. M.G.M. India Ltd., 1952 Pak. Legal Dec. (Lahore) 249 (H. Ct. W. Pak.), a non-maritime case, the court held that a choice of forum clause was equivalent to an agreement to submit a case to arbitration, and thus within section 28's exceptions. This meant that the enforceability of the clause was a matter of discretion. Under the circumstances of Ali Akbar, the court declined to enforce a contract provision requiring that any legal proceedings be instituted in Bombay.

388. Chowdhury v. Mitsui O.S.K. Lines, 1970 Pak. Legal Dec. (Sup. Ct.) 373, [1970] 2 Lloyd's Rep. 272 (Pak.). Chowdhury was a consolidation of three pending cases from East Pakistan, each involving shipment on a foreign vessel from a foreign port to Chittagong under a bill of lading requiring disputes to be decided in a foreign forum. In each case the consignee brought an action in Chittagong, the trial court dismissed the action on the basis of the choice of forum clause, and the High Court summarily affirmed the dismissal. Id. at 378-81, [1970] 2 Lloyd's Rep. at 273-75.

389. Id. at 385-89, [1970] 2 Lloyd's Rep. at 277-80.

390. Id. at 392-93, [1970] 2 Lloyd's Rep. at 282.

391. Presumably unaware of Indussa, the Chowdhury court declared that U.S. courts would enforce a choice of forum clause "only if [the defendant] can satisfy the Court that it ... in no way lessens the liability of the carrier beyond that which is permitted under the Hague Rules." Id. at 392, [1970] 2 Lloyd's Rep. at 282.

392. See supra note 325. The court cited The Media, [1931] 41 Lloyd's Rep. 80 (K.B.), where counsel raised the Indussa argument. See supra note 339. There is no indication that the court considered that aspect of the case. See Chowdhury v. Mitsui O.S.K. Lines, 1970 Pak. Legal Dec. (Sup. Ct.) at 393, [1970] 2 Lloyd's Rep. at 283.

393. Chowdhury v. Mitsui O.S.K. Lines, 1970 Pak. Legal Dec. (Sup. Ct.) at 390-96, [1970] 2 Lloyd's Rep. at 281-84.

394. See supra note 384 and accompanying text.

395. Chowdhury v. Mitsui O.S.K. Lines, 1970 Pak. Legal Dec. (Sup. Ct.) at. 394, (1970) 2 Lloyd's Rep. at 283.

396. Id. at 395-96, [1970] 2 Lloyd's Rep. at 284.

397. Although the E.E.C. is itself an international organization, its position with respect to Article 3(8) of the Hague Rules (like its position with respect to a number of other matters) is much more analogous to that of a federal nation. Like a national court, the Court of Justice has determined the application of Article 31(8) solely by reference to its own legal system without regard for the wider international order.

398. Convention on Jurisdiction and the Enforcement of Judgments in Civil and Commercial Matters, Sept. 27, 1968, art. 17, 15 J.O. Comm. Eur. (No. L 299) 32, 35.36 (1972) [hereinafter Judgments Convention], translated in 21. O.J. Eur. Comm. (No. L 304) 36, 39-40 (1978), amended by Convention on the Accession of Denmark, Ireland, and the United Kingdom to the Judgments Convention, Oct. 9, 1978, art. 11, 21 O.J. Bur. Comm. (No. L 304) 1, 5-6 (1978). The Greek accession required further amendments, but Article 17 was unaffected. See Convention on the Accession of Greece to the Judgments Convention, Oct. 25, 1982, 25 O.J. Eur. Comm. (No. L 388) I (1982). The Judgments Convention as amended is reprinted at 26 O.J.Eur. Comm. (No. C 97) 2 (1983) [hereinafter Amended Judgments Convention].

399. The unamended Brussels Judgments Convention applied only in the six original E.E.C. member states: Belgium, France, West Germany, Italy, Luxembourg, and the Netherlands. Denmark, Greece, Ireland, and the United Kingdom acceded to the amended version. See supra note 398. Portugal and Spain have not yet ratified the Convention.

400. Article 17 (as originally enacted) provided, in relevant part, as follows:

401. Partenreederei ms Tilly Russ v. NV Haven & Vervoerbedrijf Nova. 1984 E. Comm. Ct. J. Rep. 2417. [1984] 3 Common Mkt. L.R. 499.

402. Id. at 2419. [1984] 3 Common Mkt. L.R. at 502.

403. See supra notes 358-74 and accompanying text.

404. The Tiny Russ. Partenreederei ms Tilly Russ v. NV Haven & Vervoerbedrijf Novo. 1984 E. Comm. Ct. J. Rep. at 2419. [1984] 3 Common Mkt. L.R. at 502.

405. See generally L.N. Brown & F. Jacobs, The Court of Justice of the European Communities 151-78 (2d ed. 1983) (discussing preliminary rulings). In The Tiny Russ, the Court of Cassation requested the ruling under the Protocol to the Brussels Judgments Convention, which establishes a procedure similar to that under the more familiar Article 177 of the E.E.C. Treaty.

406. It was unclear, for example, if the party issuing the bill of lading satisfied the domicile requirement. The Tiny Russ, Partenreederei ms Tilly Russ v. NV Have- & Vervoerbedrijf Nova, 1984 E. Comm. Ct. J. Rep. at 2431-32, [1984] 3 Common Mkt. L.R. at 505 (opinion of Slynn, Adv. Gen.).

407. Id. at 2432.34. [1984] 3 Common Mkt. L.R. at 513-14.

408. Every E.E.C. member state except Greece and Luxembourg has ratified or acceded to the Hague or Hague-Visby Rules. Greece has substantially enacted the Hague Rules as part of its Code of Private Maritime Law. See, e.g., Code of Private Maritime Law art. 142 (Greece) (substantially enacting Hague Rules art. 3(8)), translated in The Greek Code of Private Maritime Law 39 (B. Karatzas & N. Ready trans. 1982).

409. See supra note 398.

410. Amended Judgments Convention, supra note 398, art. 17. reprinted at 26 O.J. Eur. Comm. (No. C 97) at 7.

411. See, e.g., The Sennar, [1985] 1 W.L.R. 490 (H.L.) (Sudanese forum); The Benarty, 1985 Q.B. 325 (C.A. 1984) (Indonesian forum); The Makefjell, [1976] 2 Lloyd's Rep. 29 (C.A.) (Norwegian forum); The Frieda, Judgment of Dec. 10, 1970, Tribunal de commerce, Antwerp, 1970 Jur. Port Anv. 390, reprinted in 1971 Uniform L. Cases 206 (Canadian forum); The Costa Rica, Judgment of Mar. 29, 1960, Tribunal de commerce, Antwerp, 1961 Jur. Port Anv. 55, reprinted in 1961 Uniform L. Cases 264 (Nicaraguan forum).

412. This was essentially the approach of the Fifth Circuit in Collins, see supra notes 10519 and accompanying text, and of the New South Wales courts in Gilbert, Stokes and Waters, see supra notes 189-95 and accompanying text. It is the Approach of the West German courts today. See supra notes 247-60 and accompanying text.

413. This was essentially the approach taken under Quebec low by the court in Marubeni America Corp. v. Mitsui O.S.K. Lines, 96 D.L.R.3d 518, 532-34 (Fed. Ct., Trial Div. 1979), aff'd in part & rev'd in part sub nom. Miida Elecs., Inc. v. Mitsui O.S.K. Lines, 124 D.L.R.3d 33 (Fed. Ct. App. 1981). aff'd in part & rev'd in part sub nom. International Terminal Operators Ltd. v. Miida Elecs. Inc., 28 D.L.R.4th 641 (Can. 1986). See supra note 235-36 and accompanying text.

414. This was the position adopted by the Supreme Court in Herd. See supra notes 122-38 and accompanying text.

415. This has been the issue in the U.S. litigation of the last 25 years. See supra note 141-46 and accompanying text.

416. This is essentially the approach taken by the recent English cases. See supra notes 169-86 and accompanying text.

417. This has been the focus of the Australian litigation since The Eurymedon. See supra notes 206-16 and accompanying text.

418. There are, of course, more than five possibilities open to the court. Discussion of these five, however, is sufficient to illustrate the point.

419. This is not too far from the House of Lords approach in Scruttons, see supra notes 149-68 and accompanying text, or the Australian High Court approach in Wilson, see supra notes 196-205 and accompanying text.

420. In rare cases, the Hague Rules and national law may explicitly conflict, i.e., there may be no permissible interpretation of the Hague Rules that ill consistent with any permissible interpretation of national law. That situation is not the issue here. See supra note 82.

421. See supra note 254 and accompanying text.

422. See supra note 244 and accompanying text.

423. See supra notes 130-32 and accompanying text.

424. See supra note 136 and accompanying text.

425. The House of Lords implicitly considered and rejected this proposal in The Hollandia, 1983 A.C. 565, 574 (1982). European Community law may permit the bill of lading to name any forum in a member state. See supra notes 397-411 and accompanying text.

426. Denning, supra note 324, argues for this position.

427. When Article 3(8) concerns are not implicated, English law puts the burden on the plaintiff. See supra text accompanying notes 353-54. Pakistani law, in contrast, puts the burden on the carrier. See supra note 396 and accompanying text.

428. This was the position adopted by the House of Lords in The Hollandia, 1983 A.C. 565 (1982). See supra notes 340-46 and accompanying text.

429. This is theoretically the position of the Belgium courts when the Brussels Judgments Convention does not apply. See supra notes 357-74 and accompanying text.

430. Once again, discussion of five possibilities is sufficient to illustrate the point. Cf. supra note 418.

431. This is the U.S. position with respect to cases governed by COGSA. See supra notes 305-21 and accompanying text.

432. See supra notes 274-81 and accompanying text.

433. See supra notes 314-18 and accompanying text.

434. See supra notes 358-68 and accompanying text.

435. See supra notes 353-55 and accompanying text.

436. See supra notes 376-96 and accompanying text.

437. See supra notes 397-408 and accompanying text.

438. As the distinctions among permissible interpretations become finer, it is logical to suppose that domestic law constraints will have relatively less impact and random distribution will have relatively more impact. Cf. supra note 60.

439. See supra notes 48-67 and accompanying text.

440. See Supra note 46.

---

Pace Law School Institute of International Commercial Law - Last updated March 27, 2006

---