Go to Database Directory || Go to Bibliography

Reproduced with permission of 20 Journal of Law and Commerce (Spring 2001) 155-208

[INTRODUCTION]

The United Nations Convention on Contracts for the International Sale of Goods (CISG) [1] has been effective in the United States since January 1, 1988. It should substantially reduce the need for choice of law by American courts. The CISG contains a body of law that applies to sale of goods contracts between an American party and a party located in another country that has ratified the CISG.[2] There will soon come a time when virtually all of the leading trading nations have ratified the CISG; although a choice-of-law process will be needed to determine the governing law for transactions involving a party located in a non-ratifying country, such transactions will be few in number. Even if both parties are located in ratifying countries, the CISG does not apply to certain excluded transactions such as consumer contracts and contracts for ships or aircraft,[3] but the excluded transactions are not frequent in international trade. The CISG does not apply if the parties have opted out of the CISG altogether.[4] Parties often do opt out, but this should not create difficult choice-of-law problems for American courts. When the parties opt out, they usually agree on the law that is to replace the CISG, and American courts will be required to uphold this agreement.[5] [page 155]

In the context of international sales, the frequent and difficult choice-of-law problems will arise when the CISG applies to a transaction but does not resolve all the legal issues before the tribunal. The CISG expressly excludes a number of issues from its scope. Furthermore, like most rule regimes, the CISG leaves numerous gaps; some issues within the scope of the CISG are not specifically addressed by its rules. This article deals with the question: What choice-of-law methods should American courts use to select the applicable law in such situations? (We will assume that the parties have not made a legally effective agreement as to the law applicable to the issue(s) not resolved by the CISG.).

I. CISG GAPS AND EXCLUSIONS

In identifying issues for which choice of law may be required, we should note an important distinction between gaps and exclusions. The distinction is important because gaps can often be filled without any choice of law process, whereas excluded issues can be resolved only by means of choice of law.

A. Filling Interstitial Gaps Pursuant to Article 7(2)

Some types of issues are expressly excluded from the scope of the CISG. But if an issue is not so excluded, it is "governed" by the CISG. An issue that is governed by the CISG but not resolved by any express rule of the CISG is said to fall in a gap.[6] For example, article 78 provides that if a party fails to pay a sum of money that is due, the other party is entitled to interest on that sum.[7] Interest at what rate? No article of the CISG excludes the interest rate issue from the scope of the CISG, but neither article 78 nor any other article specifies how the interest rate is to be determined. The issue is governed by the CISG but not resolved by any express rule of the CISG. It falls into a gap, a gap that lies within the scope of the CISG regime.

Such interstitial gaps are to be filled pursuant to article 7(2): [page 156]

"Questions concerning matters governed by this Convention which are not expressly settled are to be settled in conformity with the general principles on which it is based or, in the absence of such principles, in conformity with the law applicable by virtue of the rules of private international law."[8]

Note that a court may resort to choice-of-law rules (rules of private international law) and apply the domestic law of some state only if the issue cannot be resolved by the general principles underlying the CISG.

Commentators have identified a number of broad principles that underlie the CISG. One such principle is that the parties should act in good faith.[9] This principle is expressed in article 7(1), which provides that the CISG should be interpreted so as to "promote ... the observance of good faith in international trade."[10] Related norms (which could be viewed as additional principles or as aspects of good faith) require the parties to act for their mutual benefit [11] and to act reasonably.[12] Another reasonableness principle is that a party's statements are to be interpreted as they would be interpreted by a reasonable person.[13] Additional principles related to good faith and reasonable conduct include: (1) the principle that each party should communicate information needed by the other party;[14] (2) the estoppel principle that a party cannot effectively contradict his statement on which the other party has relied;[15] (3) the principle disfavoring premature termination of the contract;[16] and, (4) the principle requiring a party to mitigate loss from the other party's breach.[17] [page 157]

Unlike the principles that are related to good faith and set standards of conduct for the parties, some of the most important principles underlying the CISG are adjudicative standards that encourage tribunals to decide cases in ways that facilitate international trade. Commentators have identified a number of such principles: (1) freedom of contract and party autonomy as to the rules that will govern the contract;[18] (2) international sales contracts should not be subject to writing requirements or other formal requirements;[19] and (3) a general presumption that the parties have formed a binding contract.[20] Arbitration tribunals have found, in CISG article 74, a general principle of full compensation for breach,[21] and have used this principle to fill the interest rate gap in article 78.[22]

In light of the broad and pregnant principles listed above, most interstitial gaps in the CISG can be filled by using general principles. Contract formation issues can be resolved by the principle that negotiation statements be interpreted according to an objective reasonable person standard, or by one of the other principles related to good faith, or by the principle favoring a finding that a binding contract has been formed. Most questions whether a party has violated a contractual obligation can be resolved by the principle requiring reasonable conduct or by one of the other principles related to good faith. Most remedy gaps can be filled by applying the full compensation principle or the mitigation of loss principle. Only rarely should tribunals have to use choice of law and apply rules external to the CISG regime in order to fill CISG gaps.[23] [page 158]

B. The Article 4 Third-Party Exclusion

If an issue is expressly excluded from the scope of the CISG, it is not "governed" by the CISG, article 7(2) does not apply, CISG general principles do not come into play, and the court must apply its choice-of-law rules leading to substantive rules that are external to the CISG regime.

One such exclusion is found in the first sentence of article 4: "This Convention governs only the formation of the contract of sale and the rights and obligations of the seller and the buyer arising from such a contract."[24] The rights of third parties are thus excluded from the scope of the CISG.

Assume that Seller sells defective goods to Buyer under an international sales contract controlled by the CISG. Buyer in turn sells the goods to Remote Purchaser, who suffers some loss or injury because of the defective nature of the goods. Remote Purchaser sues Seller for "breach of warranty," claiming standing as a third-party beneficiary of the warranty from Seller to Buyer. The question whether Seller breached his obligation concerning the quality of the goods will be governed by CISG article 35 and other CISG provisions.[25] But the question whether Remote Purchaser, as a third-party beneficiary, has enforceable rights against Seller is not governed by the CISG and must be resolved by way of a choice-of-law process.[26] Similarly, third-party claims asserted by persons who used the goods as employees of Buyer, or were mere bystanders who never used the goods, would be subject to non-CISG law concerning enforceability by third-party beneficiaries. And non-contractual claims of negligence or strict liability in tort (even if asserted by Buyer) would seem to be excluded from the scope of the CISG by article 4 because they do not arise from the contract between Seller and Buyer. [page 159]

Because of the article 4 third-party exclusion, the CISG does not govern issues that arise when one party assigns one or more of her contractual rights.[27] A choice-of-law process must be employed to select the law resolving such issues as: the assignability of the right, the effect of the assignment on the obligor, and priority between successive assignees of the same right.

C. The Article 5 Death or Personal Injury Exclusion

CISG article 5 provides that "[t]his Convention does not apply to the liability of the seller for death or personal injury caused by the goods to any person."[28] Claims brought by any person (including the buyer) for death or personal injury to anyone are thus excluded from the scope of the CISG and governed by the law that is applicable under the forum's choice-of-law rules.

We can now see that articles 4 and 5 exclude from CISG coverage a large number of product liability claims: article 4 excludes all claims of third parties; article 5 excludes all claims for death or personal injury. These exclusions will often necessitate a choice-of-law process, even for international sales transactions otherwise controlled by the CISG.

On the other hand, the CISG rules govern the buyer's contractual claim for economic loss caused by nonconformity of the goods, and also govern the buyer's contractual claim for damage to other tangible property caused by nonconformity of the goods. Here the CISG rules displace the contractual remedies afforded by domestic laws, and there is no place for a choice-of-law process.

The extent to which the CISG displaces products liability claims sounding in tort (or "delict") is not entirely clear.[29] Presumably, the CISG does not displace domestic law tort claims when the liability issue is expressly excluded from the scope of the CISG. Thus, third parties can probably recover damages from the seller in tort, and buyers can probably recover tort damages from the seller for personal injury, when such remedies are provided by the applicable law selected by choice-of-law rules.

What is not so clear is whether the buyer can maintain a domestic law tort claim concurrently with, or in lieu of, his CISG contractual claim when he has [page 160] suffered economic or property loss. (The buyer might assert a tort claim because he may not be able to recover on his CISG claim, due to the CISG article 39 notice requirement [30] or some other impediment created by the CISG rules.). It can be argued that because a tort claim does not arise from the contract and is thus excluded from the scope of the CISG under article 4, room is left outside the CISG regime for choice of law and the application of some domestic tort rule. The prevailing view, however, seems to be that the CISG contractual remedies pre-empt a buyer's tort claim for losses with which the CISG rules deal (buyer's economic or property losses).[31]

Because many products liability cases involve statute of limitations issues, this may be a good place to note that the CISG contains no statute of limitations provisions. However, the 1974 United Nations Convention on the Limitation Period in the International Sale of Goods, as amended by a 1980 Protocol,[32] is in force and was ratified by the United States. It applies, for the most part, to the same claims to which the CISG applies.[33] For claims not governed by the CISG and the UN Limitations Convention, a choice-of-law process must be employed to resolve conflicts between statutes of limitations.

D. The Article 4(a) Validity Exclusion

CISG article 4(a) states that "except as otherwise expressly provided in this Convention," the CISG "is nor concerned with ... the validity of the contract or of any of its provisions or of any usage."[34] Validity issues are thus excluded from the scope of the CISG regime and governed by rules selected in a choice-of-law process, except when the CISG is expressly concerned with [page 161] a validity issue. Two problematic questions are presented. Which issues are "validity" issues? And which CISG rules are "expressly" concerned with these issues?

With respect to a number of issues, it is generally agreed that they are validity issues and are not expressly addressed by the CISG rules. A list of such issues includes: (1) incapacity (due to insanity or infancy, for example);[35] (2) lack of agency authority;[36] (3) fraud and misrepresentation;[37] (4) duress;[38] (5) unconscionability (at least when "warranty" disclaimers and limited remedy clauses are not involved) ;[39] (6) illegality or violation of public policy;[40] and (7) mistake.[41]

The status of certain other issues is uncertain. It is not clear whether they are validity issues excluded from the scope of the CISG by article 4(a) and thus subject to a choice-of-law process, or are issues expressly addressed in the CISG and thus controlled by the CISG. Much depends on how the issue is characterized. Perhaps the most important problems involve disclaimers of implied warranties and clauses limiting a buyer's remedies for breach.

Using warranty disclaimers as an example, one can argue that domestic rules like those in the Uniform Commercial Code (UCC) section 2-316 are validity rules. Section 2-316 subsections (2) and (3) clearly provide that disclaimers of implied warranties are ineffective unless certain words or types of words are used.[42] Furthermore, section 2-316, like section 2-302, is [page 162] primarily concerned with preventing unfair surprise,[43] and it is "generally agreed that section 2-302's unconscionability rule concerns validity. The CISG contains no express rule dealing with this validity issue. (Although the CISG notion of good faith could be relevant, it is merely an underlying principle and not an express rule.). Therefore, the effectiveness of the disclaimer is a validity issue excluded from the scope of the CISG by article 4(a). And if some state's UCC is the applicable law identified by the forum's choice-of-law rules, section 2-316 governs the effectiveness of the disclaimer.

On the other hand, it can be argued that certain CISG rules expressly deal with the issue, and there is no place for a choice-of-law process. CISG article 35(2) provides that its implied obligations of the seller (similar to the UCC implied warranties of merchantability and fitness for particular purpose) do not come into play when "the parties have agreed otherwise."[44] CISG article 6 says the same thing in more general terms.[45] Thus the issue is whether the parties have agreed to the warranty' disclaimer. This issue is largely governed by CISG article 8(2), which requires that the seller's disclaimer language be interpreted as a reasonable person in the buyer's circumstances would interpret it.[46] As for domestic rules like those in UCC section 2-316, they are essentially interpretation rules, rather than validity rules, and must therefore give way to CISG article 8(2).[47]

We need not resolve the debate here. We are concerned with what choice-of-law method should be used when an issue is deemed to be a validity issue excluded from CISG governance by article 4(a).

E. The Article 4(b) Property Rights Exclusion

CISG article 4(b) states that "except as otherwise expressly provided in this Convention," the CISG "is not concerned with ... the effect which the contract may have on the property in the goods sold."[48] Although the CISG contains some rules that may be relevant to property rights as between the seller and the buyer,[49] the CISG does not deal with property claims of third parties. Thus, for example, the question whether the buyer takes the goods [page 163] free of some third-party's claim is left to the applicable domestic law selected in a choice-of-law process.[50]

F. Article 6 Opt-out Exclusions

As previously noted,[51] CISG article 6 allows the parties to contractually opt-out of the CISG altogether or to opt-out of, or modify, a specific CISG rule.[52]

We are assuming that the parties have not effectively opted-out of the CISG altogether and that the transaction is generally controlled by the CISG.[53] But the parties may have effectively excluded the CISG rule that addresses an issue before the court. And the parties might not have included in their contract a rule to replace the excluded CISG rule and not even made a legally effective agreement as to what body of non-CISG law is to govern issues not resolved by the CISG.

In such a situation, what should the court do? Should the court rule that the disputed issue has been "excluded" from CISG coverage by virtue of article 6 and immediately invoke its choice-of-law rules to select the law governing the issue? Or should the court say that the parties have created a "gap" in the CISG, a gap to be filled pursuant to article 7(2), which directs courts to resort to choice of law only when the issue cannot be resolved by the general principles underlying the CISG?

If the parties had made an unsuccessful attempt to designate the domestic law of some state as the governing law for issues not resolved by the CISG rules, we would probably say that the court should immediately resort to its choice-of-law process. The parties would have indicated that with respect to the issue addressed by the CISG rule they have rejected, they want to play by domestic law rules and not the international CISG rules or principles. This [page 164] intention of the parties should be respected, because the CISG itself, in article 6, gives priority to party autonomy.[54]

It has been suggested, however, that if the parties have not attempted to designate some domestic law as the governing law for issues not resolved by CISG rules, the court should apply article 7(2) and look for CISG general principles before resorting to a choice of law.[55] This suggestion is not persuasive. Article 6 says that the parties may "derogate from" a CISG provision. If such a derogation amounts to an exclusion, so that the issue addressed by the CISG rule from which the parties have derogated is not "governed by" the CISG, article 7(2) does not apply. A good reason for regarding derogations as exclusions and bypassing article 7(2) is that any rule constructed from the CISG general principles would probably be similar to the CISG rule the parties have tried to knock out. Applying article 7(2) is thus likely to frustrate the intention of the parties and the CISG policy of party autonomy. Because the parties have not indicated what domestic law they want applied, the court should resort to its own choice-of-law process.

G. Article 92 Reservations

Under CISG article 92(1), a state may ratify the CISG with a declaration (reservation) that it will not be bound by Part II of the CISG (dealing with contract formation) or that it will not be bound by Part III (dealing with obligations of the parties and remedies).[56] Denmark, Finland, Norway, and Sweden have ratified the CISG with reservations rejecting the Part III contract formation rules.[57]

Article 92(2) provides that a state that has ratified the CISG with a declaration that it will not be bound by Part II "is not to be considered a Contracting State within paragraph (1) of article 1" with respect to contract formation issues.[58] Assume that an American court tries a case involving an American party and a party located in Denmark (which has ratified the CISG [page 165] with an article 92 reservation rejecting Part II). Even though the United States has not made an article 92 reservation, the American court must not apply the CISG Part II rules to any contract formation issues that arise. The article 1(1)(a) CISG applicability rule is not satisfied, because Denmark is not a Contracting State with respect to Part II.[59] The American court cannot apply the Part II rules pursuant to article 1(1)(b), because of the United States reservation (under article 95) that it will not be bound by article 1(1)(b).[60] The court will thus have to employ a choice-of-law process and select some non-CISG rule to resolve contract formation issues.[61]

H. The Article 96 Statute of Frauds Reservation

A number of CISG provisions instantiate the general principle that international sales contracts should not be subject to writing requirements or other formal requirements.[62] Article 11 states that the contract "need not be concluded in or evidenced by writing and is not subject to any other requirements as to form."[63] Article 29(1) indicates that contract modifications and terminations need not be in writing.[64] The contract formation provisions in Part II expressly allow oral offers and oral or nonverbal acceptances.[65]

However, article 96 provides:

"A Contracting State whose legislation requires contracts of sale to be concluded in or evidenced by writing may at any time make a declaration in accordance with article 12 that any provision of article 11, article 29, or Part II of this Convention, that allows a contract of sale or its modification or termination by agreement or any offer, acceptance, or other indication of intention to be made in any form other than in writing, does not apply where any party has his place of business in that State."[66]

Despite the American Uniform Commercial Code sale of goods statute of frauds in section 2-201, the United States has not made an article 96 reservation. [page 166] But article 96 reservations have been made by Argentina, Belarus, Chile, China, Estonia, Hungary, Latvia, Lithuania, the Russian Federation, and Ukraine. And CISG article 12 provides that if either party has his place of business in a state that has made an article 96 declaration, the "no writing required" rules of article 11, article 29, and Part II do not apply.[67]

Thus, if an American court hears a case involving an American party and a party located in a state that has made an article 96 reservation, the court cannot apply the CISG "no writing required" rules referred to above. The prevailing view is that this does not mean that the reservation state's statute of frauds necessarily applies. The court must use its choice-of-law process to determine which state's law concerning writing requirements will apply.[68] If the choice-of-law process leads to the law of the reservation state, that state's statute of frauds applies. If the choice-of-law process leads to the law of an American jurisdiction, such as New York, UCC Article 2 rules apply.[69]

II. AMERICAN CHOICE-OF-LAW METHODS

We have seen that even in international sales transactions generally governed by the CISG, some issues may require resolution by way of a choice-of-law process. We will now survey the various choice-of-law rules and approaches currently used by American courts for contract issues. We will see that none of these rules or approaches provide sound or helpful guidance for choice of law in the international sales context. (Although most of my criticisms of American choice-of-law methods also apply in the context of domestic sales, we are here concerned only with international sales.) [page 167]

A. UCC Section 1-105

Our survey must begin with the choice-of-law rules contained in UCC section 1-105. In most of the cases with which we are concerned, these statutory rules are mandatory for American courts.[70] Subsection (1) of section 1-105 provides that when the parties have not effectively agreed on the applicable law, the local UCC applies to transactions bearing an "appropriate relation" to the state in which the court sits.[71] The phrase "appropriate relation" is ambiguous and has been interpreted in two disparate ways.

According to the first interpretation, forum UCC law must be applied if the forum state has at least a minimal or reasonable relation to the transaction, such that application of forum law would comply with constitutional due process requirements. Some courts have adopted this interpretation.[72]

A greater number of courts, however, have interpreted "appropriate relation" to mean a relation that renders forum UCC law applicable under the forum's judicial choice-of-law rules.[73] The preference for this second interpretation is supported by a number of good reasons. First, comment 3 to section 1-105 states that when a transaction "has significant contacts with a [page 168] state which has enacted the Act [the UCC] and also with other jurisdictions, the question what relation is 'appropriate' is left to judicial decision."[74] Second, although comment 3 also indicates a preference for UCC law over non-UCC law, this is due to the original drafters' desire to give the UCC the widest possible application at a time when it was not clear how many states would enact the UCC. Now that every state has enacted some form of the UCC, this original purpose is obsolete.[75] Third, in the context of international sales, the second interpretation accords more respect to foreign law than does the first interpretation and thus better serves the goal of international comity. Fourth, because the first interpretation virtually assures the application of forum law, it encourages forum shopping.[76]

A proposed revision of UCC Article 1 would replace the present section 1-105 with a new section 1-301. Subsection (a) of this proposed section provides that in the absence of an effective agreement of the parties as to applicable law, their rights and obligations will be determined by the law that would be selected by the forum state's conflict-of-laws principles.[77] If this proposed revision (which is, in effect, a statutory adoption of the second interpretation of the present section 1-105(1)) is enacted by the various states in the United States, it will be clear that state courts are free to develop and apply their own choice-of-law rules for issues not resolved by the CISG. We will now consider the choice-of-law methods American courts have been using for contracts issues.

B. Applying the Law of the Place of Contracting

For a long period extending well into the 20th century, one of the most popular American choice-of-law rules for many contract issues was lex loci contractus: apply the law of the place where the contract was formed. This rule was adopted in the 1878 leading case of Milliken v. Pratt [78] and enshrined [page 169] in the Restatement (First) of Conflict of Laws section 332.[79] The rule is still the authoritative rule in a number of states.[80]

For the rule to work, it must be carefully structured so that the place of contracting can be identified without identifying the state whose law would govern the ultimate issue (otherwise a vicious circularity would ensue). The Restatement (First) solution was to identify as the place of contracting, the state where the last event necessary to form a contract occurred under the contract law rules of offer and acceptance specified in the Restatement itself.[81] Thus, if the offeree mailed her acceptance from State A, a court in State B would identify State A as the place of contracting and would apply State A law to the ultimate contract law issue.[82]

With respect to an international sale of goods generally governed by the CISG, however, an American court must normally apply the offer and acceptance rules of the CISG. CISG article 18(2) indicates that the contract is formed when the offeror receives the offeree's acceptance; thus the place of contracting would be in the offeror's state or country.[83]

A choice-of-law rule requiring the court to apply the law of the place of contracting suffers from two fatal defects. First, if the parties negotiated their contract by means of several communications, it may be difficult to identify the acceptance, and thus difficult to identify the place where the acceptance was received. Second, even if the court can confidently identify the acceptance, there is no good reason to make the law governing a measure of damages issue, for example, depend on the accidental and substantively irrelevant facts concerning which party was the offeror and which party the offeree. These facts are relevant to few of the issues left unresolved by the CISG. As a choice-of-law rule for CISG transactions, the place of contracting rule should therefore be rejected because it arbitrarily selects the law of the offeror's state or country, simply because he was the offeror and not the offeree. [page 170]

C. Applying the Law of the Place of Performance

Under the traditional approaches to choice of law, courts that did not apply the law of the place of contracting to a particular issue might apply the law of the place where a contractual obligation was to be performed. Some courts applied the law of the place of performance to most contract issues, including issues unrelated to performance.[84] Under the bifurcated approach of the Restatement (First), however, issues related to contract performance (such as the nature of the performance required, the sufficiency of a party's performance, excuse for non-performance, and damages for non-performance) were to be governed by the law of the place of performance, while validity issues and many other issues were to be resolved by the law of the place of contracting.[85]

But where is the place of performance? It would be fanciful to suggest that all issues be governed by the law of the place where "the contract" was to be performed. The contract is an agreement involving a number of promises by each party and thus a number of obligations to be performed by each party. In an interstate or international sales contract, there is no one state or country where all the obligations are to be performed.

It would not be irrational to suggest that questions concerning a party's performance of a particular obligation be governed by the law of the place where that particular obligation was to be performed. (This is what the Restatement (First) prescribed.).[86] But problems remain. Performance of an obligation may involve a number of acts in different places. Focusing on just one of these acts and selecting the law of the place where that act was to occur may be as arbitrary as selecting the law of the place where the acceptance was received by the offeror. We will see an example when we consider Restatement (Second) of Conflict of Laws § 191, which can be understood only after a review of the general approach of that Restatement.

D. The Restatement (Second) Approach

The choice-of-law approach prescribed in the Restatement (Second) of Conflict of Laws is the outgrowth of the "most significant relationship" (or "center of gravity" or "dominant contacts" or "grouping of contacts") [page 171] approach that emerged in W.H. Barber Co. v. Hughes [87] and Auten v. Auten [88] and which was later adopted in many states.[89] For contract issues, the Restatement (Second) approach, or something like it, is now followed in a majority of American jurisdictions.[90]

At the heart of the Second Restatement's complex scheme lies section 188. Subsection (l) provides that in the absence of an effective choice of law by the parties, the rights and duties of the parties are to be determined by the local law of the state that, with respect to the particular issue to be resolved, has the "most significant relationship" to the contractual transaction and the parties.[91] Note that a separate choice-of-law analysis is to be performed for each issue; the court need not decide all issues under the local law of a single state.[92]

The state with the most significant relationship is to be identified by applying the principles stated in section 6.[93] Section 6(2) (which applies to choice of law generally, not just contract issues) lists seven factors or policy goals to be considered and balanced:[94] (1) the needs of the interstate and international systems;[95] (2) the relevant policies of the forum; (3) the relevant policies of other interested states and their interests in the outcome of the issue;[96] (4) protection of the parties' justified expectations; (5) the basic policies underlying the particular field of law; (6) predictability and uniformity of result; and (7) ease in determining the law to be applied and applying it.

In addition to the section 6 factors, the court must also take into account certain geographical contacts the transaction has with various states:[97] (1) the place of contracting;[98] (2) the place where the contract was negotiated; [page 172] (3) the place of performance; (4) the location of the subject matter of the contract; and (5) the domicile, residence, nationality, place of incorporation, and place of business of the parties. The contacts to be given the most weight are those that are most relevant to the particular issue being decided.[99]

In international sales cases, the Restatement (Second) approach outlined above may do more harm than good. The possibility of injustice is great, because the large number of factors and contacts to be balanced makes it all too easy for a non-neutral court to find that its forum state has the most significant relationship.[100] And if the court tries to be neutral, it is likely to discover that the Restatement approach does not clearly point to the law of one particular state or country and is thus not very helpful.[101]

Consider a hypothetical sale of goods transaction between a French seller plaintiff and an Illinois buyer defendant. An Illinois court is deciding a validity issue excluded from the CISG, fraud or mistake, for example, and is trying to be neutral and impartial. With respect to the section 6(2) policy factors, the policies and interests of the forum (Illinois) are probably offset by the policies and interests of the other interested state (France). Although the policy of protecting the justified expectations of the parties may seem to suggest applying whichever law upholds the validity of the contract, typical parties not only expect their contract to be binding, but also expect not to be induced into a contract by fraud or mistake.[102] Party expectations thus do not clearly point to either French law or Illinois law. Predictability and uniformity of result can be achieved by any of a number of judicial practices consistently followed, for example, always applying the law of the plaintiff's State, always applying the law of the defendant's state, always applying the law of the seller's state, or always applying the law of the buyer's state. Acknowledging the value of predictability and uniformity will not help the court decide which practice to adopt and thus will not help the court decide whether to apply French law or Illinois law. Ease in determining what law to apply and determining the content of that law may indicate that the Illinois court should apply Illinois law, rather than the relatively unfamiliar French law. [page 173] But this factor, by itself, provides an insufficient reason for applying Illinois law, especially if the French law can be adequately explained by competent experts.

With regard to the section 188(2) contacts, the place of contracting should have little weight. Whether the court follows section 188 Comment (e) and uses the offer and acceptance rules of the forum, or uses the CISG rules of offer and acceptance to determine when, and thus where, the contract was formed, the place of contracting will usually be the result of accidental and irrelevant facts as to which party was the offeree and which the offeror. There will be no one place of negotiation, assuming that the seller negotiated from an office in France while the buyer negotiated from Illinois, in which case the French contact and the Illinois contact cancel out each other. When the issue is the possible invalidity of the contract due to fraud or mistake, the places of performance seem irrelevant. The location of the goods (the subject matter of the contract) is likely to vary. When the deal was being negotiated, the goods may have been in Switzerland. Just prior to the seller's delivery, the goods were probably located in France. If the buyer has received the goods, they may now be in Illinois, unless the buyer has sent the goods into another state. It is difficult to see how any of these locations could be relevant to the issue of whether the contract should be enforced. The "domicile, residence, nationality, place of incorporation, and place of business of the parties" is probably evenly split between France and Illinois (or between France and the United States if the buyer is incorporated in Delaware, for example).

Consideration of the seven section 6 policy factors and the five section 188 contacts does not yield any strong reason for concluding that France has the most significant relationship and does not yield any strong reason for deciding that Illinois has the most significant relationship. Because of the nature of international sales, each factor or contact is likely to either be insignificant or point equally to the law of the seller's state and the law of the buyer's state.

The Restatement offers a tiebreaker, a way out of the impasse. Section 191 provides that the parties' rights under a contract for the sale of goods are governed by the local law of the state where, under the contract terms, the seller is to deliver the goods, unless some other state has a more significant relationship, in which case the local law of that other state will be applied.[103] Thus, if neither France nor Illinois has the more significant relationship under section 188, the Illinois court in our hypothetical example is to resolve the [page 174] validity issue by applying the law of the place where the seller was to deliver the goods. The place of delivery is the place where the seller would complete his performance regarding physical delivery of the goods.[104]

Normally, the place where the seller completes his delivery performance is determined by a contract term containing a three-letter symbol such as "FOB," "CIF," or "DES." Assuming that the parties have agreed (either expressly or by way of usage of trade or course of dealing) that such symbols are to be interpreted according to the International Chamber of Commerce Incoterms 2000, a contract term "FOB Bordeaux" means, among other things, that the French seller completes his delivery when the goods are delivered at Bordeaux to a ship hired by the Illinois buyer, and the buyer must bear the risk of loss from that point on.[105] But if the contract term is "DES Chicago," the seller does not complete his delivery performance until the goods arrive in Chicago; until that point, the seller must bear transportation costs and the risk of loss.[106]

Thus, under Restatement (Second) section 191, French law governs the validity issue if the contractual delivery term is "FOB Bordeaux," but Illinois law applies if the term is "DES Chicago." This does not make sense. A three-letter delivery term is merely a convenient way for the parties to allocate certain responsibilities and risks related to transportation of the goods. It does not give either the seller's state or the buyer's state a closer connection with, or greater interest in, the contractual transaction. It is quite irrelevant to validity issues, issues concerning the quality of the goods, and most measure of damages issues. The rule of section 191, like other place of performance rules, may be appropriate when the question is whether a party has performed a particular obligation, but such questions can usually be resolved by interpretation of contract terms, usage of trade or course of dealing, the very comprehensive CISG rules specifying the obligations of each party, or the general principles underlying the CISG; rarely will choice of law even be necessary. In short, the Restatement (Second) approach presented in sections 6, 188, and 191 gives courts little that is helpful to choice-of-law analysis for international sales contracts.[107] [page 175]

E. Interest Analysis

The "interest analysis" approach to choice-of-law problems requires an attempt to discern the governmental policy behind each state's legal rule and determine whether each state could reasonably assert an interest in having its own rule applied to the particular issue before the court. Although no American jurisdiction relies exclusively on interest analysis for contract cases, a number of states use some form of interest analysis in combination with other approaches.[108]

The form of interest analysis proposed by Brainerd Currie usually results in the application of forum law.[109] Consider the typical situation where only two states are involved, one of which is the forum. If only one of these states can be regarded as having a real interest in the application of its law, there is a false conflict, and the law of the interested state should be applied. If both states have a real interest and these interests conflict, even after a moderate and restrained interpretation of state policy, there is a "true conflict." In true conflict cases, forum law should be applied, said Currie. Because a court will only infrequently find that its own state has no real interest in the litigation, Currie's analysis tends to result in choosing forum law.

Despite Currie's insistence that courts should not try to determine which state has the stronger or more worthy interest, the courts that actually use interest analysis tend to resolve true conflicts by doing that very thing (or by applying the law of the state whose interests would be most "impaired" if its law were not applied, a choice-of-law method that is similar, if not functionally equivalent, to applying the law of the state with the greater interest).[110] [page 176]

As a choice-of-law approach for international sales cases, interest analysis has serious defects. If a court follows Currie and resolves true conflicts by automatically applying forum law, this lack of respect for foreign law will impair international comity and invite retaliation. (We should keep in mind that international comity is a more sensitive concern than interstate comity.)[111] If, on the other hand, the court weighs conflicting governmental interests or assesses comparative impairment, it is engaging in an enterprise so speculative that final outcomes become very difficult to predict.

And whichever form of interest analysis is employed, the court faces the often impossible task of identifying and explicating the governmental policy underlying a foreign country's legal rule. (Note that even Currie's approach requires this.) An Illinois judge can probably understand the policy behind a New Jersey rule, because Illinois and New Jersey share a (largely) common legal culture. But will an Illinois judge ever understand the policy behind a French rule? We should conclude that although interest analysis may be helpful in cases where false conflicts are clearly present, it is not suitable as a general approach to choice of law for international transactions.

III. ALTERNATIVES TO AMERICAN CHOICE-OF-LAW METHODS

Fortunately, there are alternatives to the choice-of-law methods now used by American courts. We will examine three such alternatives.

A. Applying International Sources of Sales Law

Although American choice of law typically points to the domestic law of some state (as Restatement (Second) section 188 explicitly does), there is no [page 177] good reason why judicial choice-of-law methods cannot point to some international source of law.[112]

1. The Medieval Lex Mercatoria

Throughout much of the European Middle Ages, international trade was largely governed by transnational commercial law.[113] Beginning in the 11th and 12th centuries, Europeans enjoyed a revival of international trade. This trade continued to expand well into the 14th century. Although the increase in international trade was facilitated by population growth and urbanization, it was also made possible by greater security of vital trade routes. By the early 12th century, the Europeans had wrested control of the Mediterranean from the Moslems, and Viking raids on the coasts of northwestern Europe had pretty much ceased.

In order to maintain the growth of international trade, merchants needed a new commercial law. It had to be fairly simple. It had to be a uniform commercial law, an international body of law that could protect merchants from the vicissitudes of local law. It had to regulate new commercial devices such as negotiable paper, letters of credit, marine insurance, and the commenda and societas maris forms of limited-liability investment partnership; for these new commercial devices, the old Roman law and the new Roman-based law taught in the law schools were inadequate.

These needs were met by a new lex mercatoria (law merchant) developed in the 11th through 15th centuries, primarily by the merchants themselves. [page 178] The new law was based largely on the maritime customs of port cities and the customs of the inland fairs and markets.

The medieval law merchant had a number of characteristics that made it easy to apply and highly successful as a lubricant for commerce, international commerce in particular. First, the law merchant was a distinct body of law, designed for commercial transactions and clearly separated from royal law, feudal law, local manorial or urban law, and canon law. Second, the merchants had their own special courts, usually presided over by judges who were themselves merchants, and always employing a speedy and informal procedure. Third, the law merchant was international law. Although it was localized and diverse in the 11th century, it gradually became rather uniform throughout Europe. The Amalfitan Table of 1095, for example, was produced in the Italian city of Amalfi as a codification of maritime practices and was later adopted in many other Italian cities. The maritime judgment Rolls of the court of Oleron (an island off the west coast of France) were published in about 1150 and were widely accepted in England and other Atlantic and North Sea countries. The 14th century maritime Laws of Wisby (on the island of Gotland in the Baltic) were accepted throughout the Baltic areas. The maritime customs recognized by the Consular Court of Barcelona were collected in the Cansolato del Mare of about 1340 and were then accepted in numerous Mediterranean ports. Fourth, the law merchant was based on mercantile custom and notions of good faith and fair dealing, rather than strict legal rules. Fifth, the law merchant emphasized freedom of contract. Sixth, the law merchant enforced oral contracts for the sale of goods; any requirement of a writing was regarded as a hindrance to trade.

Although the medieval lex mercatoria worked quite well as an international system, it disintegrated in the modem era as commercial law became nationalized. With the rise of powerful nation-states, the law merchant was blended into distinctive national legal systems. The law of sales was no longer international, and it was no longer created by merchants. In England, the demise of the lex mercatoria can be dated to 1606, when Edward Coke became Chief Justice of the Common Pleas and began to submerge mercantile law into the common law; sales law was now developed in common law courts rather than merchant courts, and it became heavily influenced by property law concepts. On the continent, some aspects of commercial law were preserved in a distinctive body of law (even in the 19th century codifications), but all commercial law was nationalized. [page 179]

2. Contemporary International Sources of Sales Law

Today, we once again have a body of international law that can be applied to transnational sales of goods. This body includes the CISG, merchant customs, the UNIDROIT Principles of International Commercial Contracts, and international case law. The CISG obviously plays an important role, but we are now concerned with using choice of law to resolve issues that the CISG leaves unresolved.

Merchant customs regarding international trade have been created by merchants and by trade associations and other private organizations such as the International Chamber of Commerce. (The ICC's Incoterms and its Uniform Customs and Practice for Documentary Credits are now widely accepted as customs governing certain aspects of international sales transactions.) Such merchant customs are, however, unlikely to be applied by way of a judicial choice-of-law process. Under CISG article 9(2), unless the parties have otherwise agreed, they are already bound by a mercantile custom if it: (1) is a usage of trade the parties knew about or should have known about, and (2) is regularly observed in the particular international trade in which the transaction is involved.[114] And if the usage does not meet these requirements, it would probably be improper for a court to apply the usage as international law selected by a choice-of-law rule. CISG article 9(2) seems to contain a negative implication that usages not meeting its requirements (for example, a usage one party had no reason to know about) are not to be applied by a court. Furthermore, an attempt by one party to invoke a usage that is not regularly observed, or not reasonably accessible to the other party, is likely to be regarded as a violation of the basic CISG principles of good faith and reasonable conduct.

This leaves us with the UNIDROIT Principles and international case law as possible international sources of law to be selected in a choice-of-law process. [page 180] The UNIDROIT Principles of International Commercial Contracts [115] was drafted by the International Institute for the Unification of Private Law (UNIDROIT) and provides an unofficial "Restatement" of rules for international commercial contracts, including contracts for the sale of goods. A number of features make the UNIDROIT Principles an appropriate source of rules that can be applied to issues not resolved by the CISG. First, the major criterion in deciding which rule to espouse for a given issue was very practical: Which rule best serves the needs of international trade?[116] In addition, the UNIDROIT rules are based on the same general principles that underlie the CISG rules,[117] and there is thus an overall harmony between the two sets of rules. This is an important consideration for a court selecting rules with which to supplement the CISG. Furthermore, the UNIDROIT rules deal with many issues excluded from the CISG. Finally, the UNIDROIT rules are much easier to research and prove than are the rules of domestic legal systems.

Courts and arbitration tribunals throughout the world are developing a growing body of international case law on issues not resolved by the CISG, though arising from transactions to which the CISG applies. This case law is now accessible by means of UNILEX and other international reporting services that focus on CISG cases. Compared to the UNIDROIT Principles, however, international case law has some disadvantages. Because it may not be uniform from one tribunal to another, international case law affords less predictability of outcome than the UNIDROIT rules provide. And researching international case law is more difficult and costly than researching the UNIDROIT rules.

Nevertheless, both international case law and the UNIDROIT Principles are in two respects preferable to domestic law as sources of rules for international sales. First, the rules of domestic sales or contract law are often inappropriate for international transactions.[118] For international trade to [page 181] flourish, merchants must be flexible and innovative and thus need extensive freedom of contract. International trade also requires contracts formed quickly and relatively cheaply between parties separated by long distances; merchants must therefore be freed from formal requirements concerning prescribed language or signatures. Like the medieval lex mercatoria, the CISG and UNIDROIT Principles favor freedom of contract and informal methods of contracting. So does much of the international case law, especially that developed by arbitration tribunals. The domestic laws of some nations, however, greatly restrict freedom of contract or impose burdensome formal requirements.

Second, choice-of-law methods that apply only domestic rules can impose an unfair burden on one of the parties. Such choice-of-law methods usually result in the application of the law of one party's state. Even when courts use a choice-of-law rule that makes it clear which state's law will be applied (for example, a rule that always selects the law of the seller's state), the predictability problem is reduced, but the fairness problem persists. The party located in the state whose law will not be applied will be at a disadvantage. Assuming that he uses counsel in his own country, he will incur legal fees that far exceed the legal fees paid by the other party (his lawyers must research a foreign body of law, while the other party's lawyers deal with law with which they are already familiar). To make matters worse, he cannot rely on his lawyers' advice as confidently as the other party can rely on her lawyers' advice. If the courts were to apply the UNIDROIT rules or international case law, rather than domestic law, the two parties would face roughly equal legal research burdens and roughly equal risks.[119]

B. Applying the Law of the Seller's State

A number of international treaties deal with choice of law for international commercial contracts. In international sales cases, three of these treaties generally require a court to apply the law of the seller's state. This general rule is subject to some exceptions, but they will not apply in the majority of sales cases. Although the United States is not a party to any of these treaties, American courts may find their general rule attractive. [page 182]

1. The 1955 Hague Convention

The 1955 Hague Convention on the Law Applicable to International Sales of Goods [120] was sponsored by the Hague Conference on Private International Law. It was ratified by Belgium, Denmark, Finland, France, Italy, Niger, Norway, Sweden, and Switzerland and came into force in 1964.

Article 3 of the Convention provides that when the parties have not contractually designated the governing law, the sale:

"shall be governed by the domestic law of the country in which the vendor has his habitual residence at the time when he receives the order. If the order is received by an establishment of the vendor, the sale shall be governed by the domestic law of the country in which the establishment is situated."[121]

The next paragraph provides that the sale:

"shall be governed by the domestic law of the country in which the purchaser has his habitual residence, or in which he has the establishment that has given the order, if the order has been received in such country, whether by the vendor or by his representative, agent or commercial traveller."[122]

The law to be applied thus depends on where the buyer's order is received. If the order was received by the seller's subsidiary, branch office, or agent in the buyer's country, the law of buyer's country governs the transaction. In the majority of international sales cases, however, the buyer's order is received in the seller's country, and the law of the seller's country therefore governs the transaction.

2. The 1986 Hague Convention

In 1985, the Hague Conference on Private International Law produced a new Convention on the Law Applicable to Contracts for the International Sale of Goods.[123] Because it was not signed until 1986, it is referred to by that date. It was designed to replace the 1955 Hague Convention, but has not yet been blessed with enough ratifications to come into force. [page 183]

Article 8(1) of this Convention provides that if the parties have not chosen the applicable law, "the contract is governed by the law of the State where the seller has his place of business at the time of conclusion of the contract."[124] This basic rule is subject to exceptions, but none of these exceptions is likely to apply to the typical international sale.[125]

3. The 1980 Rome Convention

In 1980, the members of the European Economic Community adopted a Convention on the Law Applicable to Contractual Obligations.[126] This Convention came into force in 1991 and has been ratified by all fifteen members of the present European Union.

Article 4(1) of the Convention provides that if the parties have not chosen the applicable law, "the contract shall be governed by the law of the country with which it is most closely connected."[127] This provision does not lead to an open-ended "most significant relationship" analysis, however, because article 4(2) provides a presumption "that the contract is most closely connected with the country where the party who is to effect the performance which is characteristic of the contract has, at the time of conclusion of the contract," its principal place of business or other place of business from which its performance is to be effected.[128] [page 184] The official Report on the Convention indicates that in a sale of goods, the characteristic performance is the seller's delivery of the goods; therefore, the law of the seller's place of business is usually the governing law.[129] The commentators generally agree with the official Report.[130]

With respect to contracts for the sale of goods, the basic rule is thus quite simple. The contract will be governed by the law of the seller's state.[131]

4. Advantages of the Treaty Rule

We have seen that three international treaties use a basic choice-of-law rule requiring application of the law of the seller's state.[132] This rule determines the applicable law for most legal issues and for the majority of international sales contracts in which the parties have not selected the applicable law. [page 185]

This choice-of-law method deserves serious consideration. In the first place, it is simple and easy to administer (especially if the exceptions found in the treaties are eliminated). By simply applying the law of the seller's state to all issues not resolved by the CISG, a court would avoid the difficulties of determining where the contract was formed, where the contract was to be performed, which state has the most significant relationship to the contract and the particular issue, or which state has the strongest interest in having its law applied.

Also (and for similar reasons), a rule requiring application of the law of the seller's state makes the choice-of-law process more predictable for the parties. If they know they will be litigating in a court that uses this choice-of-law rule, an American buyer and an Italian seller will know that Italian law and not the UCC will be applied. (The rule does not necessarily achieve predictability of outcome, however; an American buyer might have difficulty in identifying and understanding the applicable Italian rule.)

Furthermore, selecting the law of the seller's state may be the wave of the future. It has already been accepted in a majority of European states. We have noted that the fifteen members of the European Union have ratified the 1980 Rome Convention. And the former Czechoslovakia, as well as Hungary, Poland, and Switzerland enacted choice-of-law statutes pointing to the law of the seller's state.[133] At some point in the future, it may be desirable for American courts to swim with the tide in order to promote uniformity and predictability in choice of law.

C. Rule-Dépeçage

The term "dépeçage" is often defined as a choice-of-law method that cuts up a case and applies the law of different legal systems to different issues.[134] Thus, a court employs dépeçage when it applies the law of State A to one issue and applies the law of State B to another issue.

The leading American choice-of-law approaches encourage extensive dépeçage. The Restatement (Second) requires that its "most significant relationship" test be applied separately to each contract issue.[135] [page 186] This can result in the law of State A being applied to one issue while the law of State B is applied to another issue. Similarly, interest analysis approaches permit a court to find that State A has the only (or greater) interest in one issue and that State B has the only (or greater) interest in some other issue. Note, however, that whether the court uses the Restatement or interest analysis, it always applies one approach, one analytical method with one set of factors to consider, to all issues.

In using choice of law to resolve issues not resolved by the CISG, courts should consider another kind of dépeçage, one in which different choice-of-law rules are employed for different issues. Instead of following one broad open-ended approach, a court would use precise choice-of-law rules. And a different rule would be used for each issue (or small set of related issues).[136] A choice-of-law rule requiring application of the law of the seller's state might be used for one issue, while a rule requiring application of the law of the buyer's state is used for a different issue. Let us call this choice-of-law method "rule-dépeçage."

We are considering choice-of-law methods for cases in which the CISG applies but does not resolve a particular issue. Obviously, dépeçage is inevitable in such cases; the CISG will be applied to some issues, and non-CISG law will be applied (by way of a choice-of-law selection) to issues not resolved by the CISG. When dépeçage is inevitable, the question is what kind of dépeçage should be used.

A strong argument in favor of rule-dépeçage focuses on the weaknesses of the choice-of-law methods we have examined. Choice-of-law rules that resolve all or most issues by applying the law of the place of contract formation, or the law of the place of contract performance, or the law of the seller's state resolve too many different issues by concentrating on one connecting factor that is likely to be irrelevant to most of these issues. On the other hand, broad approaches such as the Restatement (Second) "most significant relationship" approach or interest analysis lead to unpredictability. Rule-dépeçage offers an attractive solution of the dilemma. It avoids the arbitrary inflexibility of a choice-of-law rule that covers too many issues, and its use of precise rules avoids the unpredictability of the broad approaches. [page 187]

IV. PROPOSED CHOICE-OF-LAW RULES

A choice-of-law method for international sales issues not resolved by the CISG should be based on certain goals. It should facilitate international trade, preserve international comity, protect party autonomy, and enhance predictability of outcomes.

In many cases, these goals will be promoted by the application of international sources of law (the UNIDROIT Principles or international case law). For some issues, however, international sources will be unavailable or inappropriate, and the court will have to apply the domestic law of some state.

A choice-of-law rule requiring the application of the domestic law of the seller's state has the great virtue of simplicity. But it would not make sense to apply the law of the seller's state to all kinds of issues if this is an inferior method with respect to most kinds of issues. We will now examine the issues left unresolved by the CISG and try to determine, for each type of issue, what choice-of-law rule would be best. If applying the law of the seller's state appears best for most kinds of issues, we will conclude that this method should be used for all issues. Otherwise, we will conclude that American courts should employ rule-dépeçage.

A. Filling Gaps Pursuant to CISG Article 7(2)

In filling gaps pursuant to CISG article 7(2), a court may resort to a choice-of-law process only in those rare situations in which the issue cannot be resolved by the general principles underlying the CISG. Here, the best choice-of-law rule would point to law in international sources. Such law is more apt to be consistent with the CISG scheme of rules, and more likely to be suitable for international sales, than is domestic law.

If the UNIDROIT Principles contains a relevant rule, that rule should be preferred. The UNIDROIT rules are easy to find, and their application will reduce legal research costs and enhance predictability of outcomes. If no UNIDROIT rule is available, any prevailing rule of international case law should be applied. If the international case law is in serious conflict, the court should apply the case law rule that best facilitates international trade. If there is no existing international case law on the issue, the court should begin the process of developing some.[137] [page 188]

B. Products Liability Issues Excluded by CISG Articles 4 and 5

With respect to any products liability issue excluded from the scope of the CISG by article 4 or 5, it is unlikely that international sources of substantive law will be available and appropriate.[138] The court's choice-of-law rule must thus select some state's domestic law.

Although American courts have been using a variety of choice-of-law methods for products liability issues,[139] interest analysis seems to have become the most popular method.[140] Interest analysis leads to unpredictable results, however. If a court makes an unbiased appraisal of all possible state interests, it will usually be unable to conclude that one interested state has stronger interests than another interested state.

Any state will want to pursue some, if not all, of the following policy goals related to products liability: (1) providing compensatory damages to residents who are injured by defective products (the "narrow compensation" goal); (2) providing compensatory damages to nonresidents who are injured by defective products brought into the state (the "expansive compensation" goal); (3) imposing liability to discourage domestic sellers from selling defective products (the "narrow regulation" goal); (4) imposing liability to discourage foreign sellers from sending defective products into the state's stream of commerce (the "expansive regulation" goal); (5) protecting domestic sellers from excessive liability for product defects (the "narrow protection" goal); (6) protecting foreign sellers from excessive liability for product defects (the "expansive protection" goal). The three "expansive" policy goals are not at all fanciful and in a global economy may be at least as important as the three "narrow" goals. A state might pursue an expansive compensation goal in order to attract tourists. It might pursue an expansive regulation goal in order to keep defective products out of its territory, regardless of their state of origin. [page 189] It might pursue an expansive protection goal because it wants to encourage foreign firms to export their products to the state.

In most international products liability cases, more than one state would want at least one of its policy goals promoted. Because protection goals are apt to conflict with compensation goals and regulation goals, the interests of these states are likely to conflict. If the law of the defendant seller's state favors the seller while the law of the plaintiff's state favors the plaintiff, the narrow protection goal of the seller's state conflicts with the narrow compensation goal (and perhaps an expansive regulation goal) of the plaintiff's state.[141] There is no reliable way for an American court to decide which state (one of which is a foreign country) has the stronger interest.

Interest analysis yields a determinate outcome only in the false conflict cases that can arise when the law of the defendant seller's state favors the plaintiff while the law of the plaintiff's state favors the seller. If the seller's state clearly has a narrow regulation goal and the plaintiff's state clearly has no expansive protection goal, the seller's state has an interest in having its law applied while the plaintiff's state has no interest in having its law applied. If the plaintiff's state clearly has an expansive protection goal and the seller's state clearly has no narrow regulation goal, the plaintiff's state has an interest in having its law applied while the seller's state has no interest in having its law applied. In all other cases, comparative interest analysis fails to yield determinate outcomes. Here, courts should not simply apply forum law - doing so could seriously impair international comity.

Instead of asking which state has the stronger interest, courts should ask which state's lawmakers are in the best position to prescribe and balance products liability goals in international cases. If all six of the possible policy goals are considered, this question can be answered. In most cases, the state in the best position to balance the policy goals and make law for the plaintiff's claim is the state where the goods have been introduced into commerce (the "market state"). If the plaintiff acquired possession of the goods as a buyer or lessee in a commercial transaction, the market state is the state in which the plaintiff took possession. If the plaintiff was a mere user or bystander who never acquired possession of the goods in a commercial transaction, the market state is the state in which the goods were last acquired (by someone [page 190] taking possession under a commercial sale or lease) prior to the plaintiff's injury.

Lawmakers in the market state can best estimate the social costs imposed by defective products in that state and can best assess the need to discourage domestic and foreign sellers from injecting defective products into the commerce of that state. They are thus in the best position to establish narrow and expansive regulation goals. Market state lawmakers are also in the best position to set narrow and expansive protection goals. They are familiar with the state's economy and the need to limit liability so as to encourage domestic and foreign sellers to supply the state with goods. If the plaintiff resides in the market state, that state's lawmakers are in the best position to establish the relevant narrow compensation goal.

If the plaintiff does not reside in the market state, market state lawmakers will be well qualified to set an expansive compensation goal, but lawmakers in the state of plaintiff's residence are probably best qualified to establish the narrow compensation goal. (Resident plaintiffs' needs for liability compensation may depend on the insufficiency of local alternative compensation schemes such as insurance and government assistance.) Nevertheless, lawmakers in the state of plaintiff's residence are not in a good position to balance a compensation goal against protection goals and regulation goals, which should be based on local conditions in the market state. Lawmakers in the market state are in a better position to balance compensation goals with protection goals and regulation goals.

If the state where the injury occurred is not the market state and not the state of the plaintiff's residence, it is unlikely that lawmakers in the state of injury would be qualified to assess and balance any of the relevant policy goals. The place of injury, by itself, does not seem relevant to any of these six goals.

Lawmakers in the defendant seller's state are in the best position to prescribe a narrow protection goal and a narrow regulation goal. But they are not in a good position to balance these two goals with any of the four other policy goals. They are not qualified to assess the market state's need for an expansive regulation goal or expansive protection goal. Nor are lawmakers in the seller's state qualified to prescribe and balance the compensation goals related to the plaintiff.

American courts should apply the law of the market state to substantive products liability issues arising from international sales but excluded from the CISG by article 4 or 5.[142] [page 191] The market state is the only state whose lawmakers are in a good position to establish and balance regulation, protection, and compensation goals. Its lawmakers will always be qualified to prescribe an expansive regulation goal, an expansive protection goal, and either a narrow compensation goal or an expansive compensation goal. Furthermore, the market state is the only state in which the activities of the seller in the international sale, the buyer in the international sale, and the plaintiff are likely to come into contact in cases involving a plaintiff who is not the buyer in the international sale. A choice-of-law rule selecting the substantive law of the market state also seems to be a good way to promote predictability in international products liability cases.

For statute of limitations issues excluded from the scope of the UN Limitations Convention, the court should not use the market state rule, but should simply apply the relevant limitations law of the forum. A statute of limitations is largely justified by considerations involving civil procedure and evidence and does not essentially reflect a resolution of regulation, compensation, and protection goals.[143]

C. Assignment Issues Excluded by CISG Article 4

In an American court, some assignment issues excluded from the scope of the CISG by article 4 will be governed by mandatory choice-of-law rules in the UCC. For other assignment issues, however, judicial choice-of-law rules must be employed.

1. Seller's Assignment of Letter of Credit Proceeds

In many cases, the seller assigns its right to the proceeds of an international letter of credit issued by the buyer's bank and confirmed by a bank in the seller's state. The letter of credit will probably incorporate by reference the provisions of the current International Chamber of Commerce Uniform Customs and Practice for Documentary Credits (UCP). [page 192] The UCP has very little to say, however, about assignments of proceeds.[144] Resolution of an assignment issue is thus likely to require a choice-of-law process.

If the issue is the legal effectiveness of the assignment or the effect of the assignment on the obligor confirmer, the choice-of-law rules in UCC section 5-116 require that in the absence of the parties' agreement as to governing law, the issue will be governed by the law of the jurisdiction in which the confirmer bank's confirming branch is located.[145] If the seller made the assignment to secure a loan and thus created a security interest, issues concerning perfection and priority of the assignee's security interest are, under UCC section 9-306, governed by the local law of the jurisdiction whose law governs the confirmer's liability under section 5-116.[146]

If the seller assigned its rights to the letter of credit proceeds in an outright sale of those rights and not to secure any obligation, the assignment did not create a security interest within the scope of UCC Article 9.[147] Therefore, the choice-of-law rules in Article 9 would not govern issues concerning the assignee's priority. Unless the court finds that UCC section 5-116 was intended to provide the choice-of-law rule, it will have to develop [page 193] its own judicial choice-of-law rule. For the sake of simplicity and predictability, however, the judicial choice-of-law rule for priority issues should be borrowed from the section 5-116 rule, which governs other assignment issues. The court should thus apply the priority rules of the jurisdiction where the confirming branch of the confirmer bank was actually located at the time of the confirmation.[148]

2. Seller's Assignment of Account

In some cases, the goods are sold on open account and no letter of credit is involved. Here, the seller may assign its right to payment under the sales contract.

If an issue arises concerning the assignee's priority vis-à-vis other assignees or creditors of the seller, UCC Article 9 provides mandatory choice-of-law rules. The seller has assigned an "account."[149] Whether the assignment provides security for a loan obligation or takes the form of an outright sale, it is within the scope of Article 9.[150] The choice-of-law rule in section 9-301(1) provides that perfection and priority of the assignee's security interest is determined by the local law of the jurisdiction in which the debtor (the assignor seller) is located.[151] If the debtor is an American corporation, it is located in its state of incorporation.[152] If the debtor is a foreign corporation, it is located in the country where it has its chief executive office, unless that country fails to maintain a security interest filing system comparable to the UCC filing system, in which case the debtor is located in the District of Columbia![153]

What if the issue concerns the assignability of the seller's account or the assignment's effect on the buyer (the buyer's rights and duties vis-à-vis the assignee)? [page 194] The statutory choice-of-law rules in UCC sections 9-301 through 9-307 deal only with the perfection and priority of the assignee's security interest, and Article 2 does not contain any relevant choice-of-law rule. The court will thus be free to use its own judicial choice-of-law rule.[154] This choice-of-law rule should point to the local law of the jurisdiction where the assignor maintained the office from which it made the assignment. Such a rule is similar to the baseline rule of UCC section 9-301 (1) (which will govern perfection and priority issues), but it omits analogs of the section 9-307 locating rules. Although those locating rules may make sense when the question is where a financing statement should be filed, we are now concerned with issues to which filing is irrelevant. The suggested choice-of-law rule would make it easy for the assignee and assignor to predict the applicable law. The obligor (buyer) would have to make inquiries in order to ascertain the office from which the assignor made the assignment. But this should be less difficult than ascertaining the assignor's chief executive office or place of incorporation. (Because the assigned rights arise from an international sales contract, the obligor buyer and assignor seller will be in different countries.)

3. Assignment by Buyer

The buyer in an international sales transaction might assign one of its rights under the sales contract or assign a claim, for example, a claim against the seller for breach of warranty. It is unlikely that the buyer will grant a security interest in its rights or claims under the sales contract (such rights or claims will not often be acceptable collateral for a loan). It is therefore unlikely that the UCC Article 9 choice-of-law rules will apply and likely that the court will be free to use a judicial choice-of-law rule for all issues arising from a buyer's assignment. This choice-of-law rule should be the one recommended in the previous paragraph: apply the law of the jurisdiction where the assignor maintained the office from which it made the assignment. The reasons for such a rule when the seller assigns an account apply with equal force when the buyer is the assignor.

D. Validity Issues Excluded by CISG Article 4(a)

The UNIDROIT Principles of International Commercial Contracts provides substantive rules for most validity issues. [page 195] These rules deal with consideration or cause,[155] initial impossibility,[156] mistake,[157] fraud,[158] duress,[159] gross disparity (unconscionability),[160] limitations on the ability to avoid the contract and remedies accompanying avoidance,[161] the validity of exemption (exculpatory) clauses limiting or excluding liability for nonperformance,[162] and the validity of liquidated damages clauses.[163] The effectiveness of a warranty disclaimer could be viewed as a validity issue.[164] Although the UNIDROIT Principles lacks a rule directly addressing the validity of warranty disclaimers, this issue could be resolved by article 1.7 (requiring good faith and fair dealing), article 2.20 (regarding surprising standard terms), article 2.22 (a battle of the forms rule adopting the "mutual knockout" approach), and article 3.10 (invalidating a term for gross disparity).

For any validity issue excluded from the scope of the CISG by article 4(a) but susceptible to resolution by the UNIDROIT Principles, an American court should use a choice-of-law rule requiring the application of the Principles. Predictability is best enhanced by such a choice-of-law method. The rules in the UNIDROIT Principles are very accessible, and both parties would at least be able to identity the substantive rule that would govern a particular validity issue. The choice-of-law rule should not point to a domestic legal system. If the outcome might depend on the unfamiliar law of the other party's country, a party in an international sale will be able to predict that outcome only by incurring considerable attorney fees.

For some validity issues, the UNIDROIT Principles provides no rules. It does not deal with contractual incapacity or lack of agency authority.[165] Nor is it concerned with illegality.[166] [page 196]

In the absence of international law providing a substantive rule for an incapacity issue, a court must use a choice-of-law rule pointing to some domestic body of law. In the interests of simplicity and predictability, a court might well adopt a rule requiring application of the local law of the state where the person whose capacity is in question had his principal place of business at the time of the sales agreement.

With respect to agency issues, the 1983 Geneva Convention on Agency in the International Sale of Goods[167] provides substantive rules for many issues concerning the effect that an agent's or purported agent's act has on a third party.[168] Although the United States has not ratified this treaty, American courts should apply its rules for the sake of uniformity, predictability, and the facilitation of international trade. For agency issues not resolved by the 1983 Geneva Convention, the court should use the choice-of-law rules in the 1978 Hague Convention on the Law Applicable to Agency.[169] This treaty provides choice-of-law rules that usually select the agency law of the state in which the agent had his business establishment, or in which the agent performed the relevant acts, depending on other connecting factors.[170] Although the United States has not ratified this treaty, American courts should use its choice-of-law rules as models. The focus on the location of the agent makes sense: the agent plays a pivotal role between the buyer and the seller, and the agent is the only one of the three affected persons (principal, agent, and third-party) who can readily be identified and located by the other two.

Whether an international sales contract is invalid due to illegality is an issue closely related to the effects of "mandatory rules" and the public policy exception to choice-of-law rules. We will defer our examination of illegality issues to section IV(I) below, where all of these matters will be discussed together. [page 197]

E. Property Issues Excluded by CISG Article 4(b)

Is the buyer a good-faith purchaser whose purchase cuts off the property interest of a third party who has a claim to the goods? This issue and other property issues involving third-party claims are excluded from the scope of the CISG by article 4(b).

If an American court adjudicates the dispute, it must apply any relevant choice-of-law rule in the UCC. If the third-party claimant obtained a security interest in the goods, UCC section 9-301 provides a mandatory choice-of-law rule requiring that priority between the buyer and the secured party be determined by the local law of the state in which the goods are located.[171] If the third-party claimant is an unsecured creditor of the seller, the choice-of-law rule in UCC section 2-402 may apply. This rule provides that a creditor of the seller may treat the sale as void if, as against her, "a retention of possession of the seller is fraudulent under any rule of law of the state where the goods are situated."[172]

In cases for which the UCC provides no choice-of-law rule (other than the one in Article 1), an American court is free to fashion its own choice-of-law rule. For the sake of simplicity, uniformity, and predictability , the court should follow the lead of UCC sections 2-402 and 9-301 and apply the local law of the jurisdiction in which the goods are located. In an international sales transaction, this is likely to be the buyer's state.

F. The CISG Article 6 Opt-out Exclusion

We are assuming that the parties to the international sales contract have not opted out of the CISG altogether. Pursuant to CISG article 6, however, they may have opted out of the CISG rule that addresses an issue before the court. If they also failed to include a replacement rule or choice-of-law clause in their contract, the court should resort to a judicial choice-of-law rule.

Respect for party autonomy is one of the basic principles underlying the CISG and should be one of the values guiding the design of choice-of-law methods for international sale issues. In opting out of a specific CISG rule, the parties have explicitly agreed not to be bound by the CISG rule and have [198] implicitly indicated that they do not want to be bound by any similar rule that might be found in other international legal sources. The rules in the UNIDROIT Principles or international case law will often be similar to the CISG rules. In order to protect party autonomy, the court should therefore eschew a choice-of-law rule pointing to international sources.

In the interests of simplicity and predictability, the court should always apply the local law of the seller's state or, alternatively, always apply the local law of the buyer's state. Because there is no strong policy reason for preferring either alternative to the other, this seems to be an appropriate situation in which to simply follow the tendency of the recent choice-of-law treaties and always apply the law of the seller's state.

G. CISG Article 92 Contract Formation Reservations

If one of the parties is located in a nation that ratified the CISG with an article 92(1) declaration that it will not be bound by CISG Part II (dealing with contract formation), the usual result is that an American court must not apply the CISG Part II rules to a contract formation issue.[173] The court must thus use a choice-of-law process.

The court should not adopt a choice-of-law rule pointing to the UNIDROIT Principles or other international sources of law. The nation making the article 92 reservation has clearly indicated that it does not accept the CISG rules on contract formation and has implicitly rejected any international rule that is similar to a CISG rule on contract formation. (Otherwise, a court could sneak in the CISG rule in the guise of some other international rule.) The UNIDROIT rules on contract formation are very similar to the CISG rules, and we can expect the same of international case law. An American court's application of such international rules could thus impair international comity .

As in the CISG article 6 opt-out situation, it is difficult to discern any strong reason for always selecting the law of the seller's state or for always selecting the law of the buyer's state. Here too, the court should simply emulate the choice-of-law treaties and always apply the non-CISG domestic law of the seller's state.[174] [page 199]

H. CISG Article 96 Statute of Frauds Reservations

Assume that an American court hears an international sales case involving a statute of frauds issue. One of the parties is located in the United States. The other party is located in a foreign state that has ratified the CISG with an article 96 reservation declaring that the "no writing required" rules of the CISG will not apply when any party to the sales contract has its place of business in that state. The court must not apply the CISG rules, but it need not apply the reservation state's statute of frauds. It is free to employ a choice-of-law process in order to resolve the statute of frauds issue. (The issue could be whether a writing is required, what kind of writing is required, or whether a noncomplying agreement can be enforced under an exception to a statute of frauds.)

The court's choice-of-law rule should not point to international sources of law. The UNIDROIT Principles provides that a contract need not be evidenced by a writing.[175] Applying this rule would be tantamount to letting the CISG "no writing required" rule in through the back door and could impair international comity between the United States and the reservation state. It would probably be inappropriate to apply any international case law dealing with writing requirements.

The choice-of-law rule should therefore point to either the domestic law of the reservation state or the domestic law of the state in which the American party is located. The best choice-of-law rule would be a rule of validation that chooses whichever state's law validates the contract and makes it enforceable. Such a rule would facilitate international trade by protecting both buyers and sellers against the often surprising formal requirements of domestic law. It would thus promote one of the most important goals of choice of law for international sales cases.[176] [page 200]

American courts should find a rule of validation quite congenial. It seems to have been the (usually unacknowledged) motivation for many, perhaps most, of the American decisions dealing with statute of frauds issues or substantive validity issues (incapacity, fraud, etc.).[177] We should also note that the recent choice-of-law treaties that point to the law of the seller's state for most issues make an exception for statute of frauds issues; here they provide choice-of-law rules that look very much like rules of validation.[178]

Using a rule of validation is not equivalent to imposing the CISG rules. An American court using a rule of validation will not be applying a substantive rule that states, as the CISG does, that no writing is required. The court will be applying a statute of frauds. In most cases, this will be the statute of frauds in UCC section 2-201. It is even possible that the validating law would be the statute of frauds of the reservation state. Although the reservation state and its citizens must be protected against application of the CISG (or similar) rules, they must take their chances under the forum's choice-of-law process. [page 201] Even if the court always applied the statute of frauds law of the seller's state (or the buyer's state), the result would often be validation of the contract.

I. Illegality, Public Policy, and Mandatory Rules

At the end of section IV(D), we postponed discussion of invalidity due to illegality. This issue often requires courts to consider the "public policy" exception to choice-of-law rules or determine the impact of "mandatory rules." Because public policy and mandatory rules may override any choice-of-law rule, whether it deals with illegality or some other issue excluded from the scope of the CISG, we will examine all of these matters in this portion of the article.

When a choice-of-law rule leads to foreign law that offends a strong public policy of the forum, the public policy exception allows the court to apply forum law rather than the foreign law.[179] Although the public policy exception is sometimes phrased as merely a jurisdictional preclusion of claims based on obnoxious foreign law, most courts use the exception to reject both claims and defenses based on foreign law, and then apply their own forum law.[180]

A domestic rule of substantive law is "mandatory" in the weak sense of the term if it cannot be displaced by a rule agreed to by the parties.[181] It is "mandatory" in the strong sense of the term if it is not only mandatory in the weak sense, but also intended to be applied in interstate and international transactions, despite any choice-of-law rule that would select a rule of a foreign legal system.[182] We will use the term "internationally mandatory" to refer to rules that are mandatory in this strong sense. [page 202]

Another important distinction involves "private law" mandatory rules and "public law" mandatory rules. Although it is not always easy to draw the line between private law and public law, we might say that private law mandatory rules include, inter alia, rules of contract law that are designed to protect one party against overreaching by the other party, rules that seek to avoid private harms; public law mandatory rules, on the other hand, are essentially matters of economic regulation designed to protect the public from negative externalities that would harm the public if the parties were to violate these rules.[183] Public law mandatory rules are found in antitrust statutes, import-export regulations, and currency controls, for example.

1. Illegality Under Private Law Mandatory Rules

An international sales contract (or a particular contract term) might be illegal under the private law mandatory rules of at least one party's state. Private law mandatory rules are usually found not to be internationally mandatory.[184] Such a finding is especially appropriate for contracts subject to the CISG. When neither party is a consumer, neither party is likely to need special protection, and private law mandatory rules may unduly impede international commerce. Assuming that no internationally mandatory private law rule is involved, the court is free to apply a choice-of-law rule.

The best choice-of-law rule would be a rule of validation that selects either the law of the seller's state or the law of the buyer's state, whichever deems the contract or clause to be legally valid. If the court regards the validating law of a foreign country as particularly unjust, it could resort to its public policy exception and apply the invalidating law of the forum. If the contract or clause is illegal under the law of both seller's state and buyer's state, it must, of course, be deemed illegal.[185] [page 203] A rule of validation seems to be the best choice-of-law rule because it would respect party autonomy and facilitate international trade. It is also consistent with what American courts actually do.[186]

2. Illegality Under Public Law Mandatory Rules

A defendant who has failed to perform a contractual obligation might argue that her performance was prohibited by a public law mandatory rule, that her obligation is unenforceable because illegal.[187] When public law is involved, there will probably be no place in an American court for a choice-of-law process.

In many cases, an American court will find that American public law is the only law purporting to regulate the defendant's conduct. For example, the defendant might be an American seller whose delivery of the goods to a buyer in Country B is prohibited by export regulations issued by the executive branch of the United States Government. It is unlikely that Country B has enacted any law purporting to govern the export of such goods from the United States. If American law purports to regulate the defendant's conduct, and the law of Country B does not, there is no conflict of laws and thus no occasion for use of a choice-of-law rule. [page 204]

In other cases, an American court will find that the only law purporting to regulate the defendant's conduct is a public law of a foreign country. For example, the defendant might be a buyer located in Country C; although she promised to pay the American seller in United States dollars, Country C's currency exchange regulations prohibit her from sending dollars out of her country. (Assume that the American court somehow has personal jurisdiction over the defendant.) It is unlikely that any American law purports to prescribe which currencies may leave Country C. If the law of Country C purports to regulate the defendant's conduct, and American law does not, there is no conflict of laws and no place for a choice-of-law rule.[188]

In a few cases, an American court will find that the defendant's conduct is regulated by both American and foreign public laws, one of which prohibits defendant's performance while the other permits it. Here there is a conflict of laws. But there is still no place for a choice-of-law rule. So long as the American public law is a federal law construed to be applicable to the international sale and internationally mandatory, the American court has no choice; it must apply the American law.[189] [page 205] Indeed, application of American federal law seems to be required, even if that law cannot be construed to conform to international law standards for prescriptive jurisdiction.[190]

3. The Effect of Mandatory Rules on Issues Other Than Illegality

When an issue other than illegality is excluded from the scope of the CISG, should the court's judicial choice-of-law rule give way to the mandatory rule of a state whose law would not be selected by that choice-of-law rule? It is unlikely that any of the issues we have considered (other than illegality) will be subject to public law. Issues concerning products liability, assignment of contractual rights, statute of frauds, and the like are matters of private rights and private law.

In international sales cases, mandatory rules of private law should not be interpreted as internationally mandatory if that can be avoided. If a mandatory rule is not internationally mandatory, it has no privileged status in an international transaction and is subject to the court's normal choice-of-law rule. If a mandatory rule of private law expressly states that it must be applied in international transactions, the rule must, of course, be classified as internationally mandatory in intent. If such a rule is contained in the statutory or administrative law of the American forum, it has the effect of a statutory choice-of-law rule and must be applied unless application would be unconstitutional. But an American court is not bound to apply the private law mandatory rule of a foreign country simply because the foreign legislature intended the rule to be internationally mandatory. Here the court should use its judicial choice-of-law rule for the issue involved. Presumably, this choice-of-law rule reflects the court's considered judgment as to what choice-of-law method is most appropriate for international sales cases and the particular issue. [page 206] As a matter of judicial integrity, a court should not abandon this judgment if it does not have to.

4. The Forum Public Policy Exception

American courts using choice-of-law rules rather than a broad approach will probably want to retain a public policy exception that allows them to apply forum law when the normal choice-of-law rule points to a foreign law that offends a strong public policy of the forum. Whether the issue is illegality or some other issue excluded from the scope of the CISG, this public policy exception should rarely be invoked.

When forum public policy is embodied in public law, there is likely to be no place for choice of law and thus no need for the public policy exception. When forum public policy is embodied in private law (contract law, commercial law, tort law, or property law), it will probably not be strong enough to override choice-of-law rules designed for international sales between merchants. The court should not invoke the public policy exception unless the test suggested by Judge Cardozo in Loucks v. Standard Oil Co. of New York is met: the foreign law must menace public welfare in the forum or violate a forum principle of justice that is fundamental and deep-rooted in tradition.[191] Indeed, because international, and not merely interstate, transactions are involved, American courts should go a step beyond Loucks and employ an international standard of justice: the foreign law must violate a fundamental principle of justice that the forum shares with most civilized nations.[192]

CONCLUSION

When the CISG applies to an international sale of goods, it might not resolve all the legal issues that arise. This article has examined the issues not resolved by the CISG, and for each type of issue, has identified the judicial choice-of-law rule that seems most appropriate.

1. In filling gaps pursuant to CISG article 7(2), the court should apply law from international sources to issues that cannot be resolved by the CISG's general principles. [page 207]

2. For products liability issues excluded from the CISG by article 4 or 5, the court should apply the substantive law of the "market state" and the statute of limitations law of the forum.

3. In the absence of a statutory choice-of-law rule, assignment issues excluded from the CISG by article 4 should be resolved as follows: when the seller has assigned letter-of-credit rights, the court should apply the law of the jurisdiction where the confirming bank's (or issuing bank's) branch was located; when no letter of credit is involved, the court should apply the law of the jurisdiction where the assignor maintained the office from which the assignment was made.

4. For validity issues excluded from the CISG by article 4(a), the court should apply the UNIDROIT Principles when its rules resolve the issue. Incapacity issues should be governed by the law of the state where the person whose capacity is in question had her principal place of business. For agency issues, the court should apply the substantive law of the 1983 Geneva Convention and fill its gaps by using the choice-of-law rules of the 1978 Hague Convention. For illegality issues involving mandatory rules of "private law," the court should use a rule of validation that selects whichever state's law validates the contract. (For illegality issues involving mandatory rules of "public law," no choice-of-law rule is necessary.)

5. Property issues excluded from the CISG by article 4(b) and not subject to statutory choice-of-law rules should be governed by the law of the jurisdiction in which the goods are located.

6. If the parties have opted-out of a CISG rule pursuant to article 6 and failed to agree on a replacement rule or governing law, the court should apply the law of the seller's state.

7. If the non-American party's government has ratified the CISG with an article 92 reservation, the court should apply the (non-CISG) domestic law of the seller's state to contract formation issues.

8. If the non-American party's government has ratified the CISG with an article 96 reservation, the court's choice-of-law rule should be a rule of validation that selects whichever state's statute of frauds law validates the contract.

* Professor of Law, University of South Carolina Law School.

1. United Nations Convention on Contracts for the International Sale of Goods, April 11, 1980, S. TREATY DOC. No. 98-9 (1983), 19 I.L.M. 668 (1980), reprinted at 15 U.S.C. app. 52 (1997) (entered into force Jan. 1, 1988) [hereinafter CISG].

2. Id. art. 1(1)(a). Article 1(1)(b) provides that the Convention also applies to a sale of goods contract between parties whose places of business are in different states if the rules of private international law (conflict of laws) lead to the application of the law of a state that has adhered to the Convention. Id. art. 1 (1)(b). This provision is not effective in American courts, however, because the United States ratified the CISG with a declaration (pursuant to CISG article 95) that it would not be bound by article 1(1)(b).

3. For a list of excluded transactions, see CISG, supra note 1, arts. 2 and 3.

4. CISG, supra note 1, art. 6 (stating that the parties may "exclude the application of this Convention or, subject to article 12, derogate from or vary the effect of any of its provisions").

5. See infra note 70 and accompanying text (explaining that American courts hearing a sale of goods case are bound by the statutory choice-of-law rules in Uniform Commercial Code Art. 1). The proposed revision of Article 1 contains a general rule that the parties' contractual designation of applicable law will be effective, whether or not the transaction bears a relation to the state or country whose law is designated. UNIFORM COMMERCIAL CODE REVISED ARTICLE 1. GENERAL PROVISIONS § 1-301(a) (proposed Discussion Draft Apr. 14, 2000).

6. For the distinction between issues excluded from the scope of the CISG and gaps within the matters governed by the CISG, see M.J. Bonell, Article 7, in COMMENTARY ON THE INTERNATIONAL SALES LAW: THE 1980 VIENNA SALES CONVENTION 65, 75-76 (C.M. Bianca & M.J. Bonell eds., 1987); Rolf Herber, Article 7, in COMMENTARY ON THE UN CONVENTION ON THE INTERNATIONAL SALE OF GOODS (CISG) 59, 65 (Peter Schlechtriem ed., Geoffrey Thomas trans., 2d ed. 1998).

7. CISG, supra note r, art. 78.

8. CISG, supra note 1, art. 7(2).

9. Bonell, supra note 6, at 80,84..85; Herber, supra note 6, at 65.

10. CISG, supra note 1, art. 7(1). The only way to interpret the CISG so as to promote good faith in international trade is for tribunals to interpret CISG rules as requiring good faith conduct by the parties. Good faith is required of the parties, not the tribunals. (The tribunals are not engaged in international trade.)

11. Bonell, supra note 6, at 88. The Preamble to the CISG states that "the development of international trade on the basis of equality and mutual benefit is an important element in promoting friendly relations among States." CISG, supra note 1, pmbl.

12. ALBERT H. KRITZER, GUIDE TO PRACTICAL APPLICATIONS OF THE UNITED NATIONS CONVENTION ON CONTRACTS FOR THE INTERNATIONAL SALE OF GOODS 116 (1989); Bonell, supra note 6, at 80-81; Herber, supra note 6, at 67. The general principle requiring that conduct meet an objective reasonable person standard is instantiated in, inter alia, CISG arts. 60(a), 72(2), 75, 77, 85, 86(1), and 88(2).

13. Herber, supra note 6, at 67. This general principle is expressed in CISG art. 8(2).

14. JOHN O. HONNOLD, UNIFORM LAW FOR INTERNATIONAL SALES UNDER THE 1980 UNITED NATIONS CONVENTION, para. 100 (3d ed. 1999); Herber, supra note 6, at 67. This principle underlies CISG arts. 19(2), 21(2), 26, 39(1), 43(1), 48(2), 65(2), 68, 71(3), 72(2), 79(4), and 88(1), (2).

15. Bonell, supra note 6, at 81.

16. Id. This principle is reflected in CISG arts. 25, 34, 37, 48, 49, 51(2), 64, and 71-73.

17. HONNOLD, supra note 14, para. 101; Bonell, supra note 6, at 81; Herber. supra note 6, at 67. This principle is instanced in CISG arts. 74, 77, 85, and 88.

18. M.J. Bonell, Article 6, in COMMENTARY ON THE INTERNATIONAL SALES LAW: THE 1980 VIENNA SALES CONVENTION 51, 51 (C.M. Bianca, & M.J. Bonell eds., 1987); Bonell, supra note 6, at 80; Herber, supra note 6, at 67. This principle is expressed in CISG art. 6, quoted supra note 4.

19. Bonell, supra note 6, at 80; Herber, supra note 6, at 67. This principle is expressed in CISG arts. 11 and 29(1).

20. Bonell, supra note 6, at 81. This principle is reflected in CISG arts. 16, 18(1), 19(2), and 21.

21. The first sentence of article 74 prescribes that "[d]amages for breach of contract by one party consist of a sum equal to the loss, including loss of profit, suffered by the other party as a consequence of the breach." CISG, supra note 1, art. 74.

22. Full compensation requires interest at the average prime lending rate in the aggrieved party's country and with respect to the contractual currency of payment. See, e.g., Case SCH-4366, 1995 R.I.W. 590, UNILEX D.1994-14, E. 1994-14 (Internationales Schiedsgericht der Bundeskammer der gewerblichen Wirtschaft-Wien 1994) (Austria); Case SCH-4318, 1995 R.I.W. 591, UNILEX D.1994-13, E.I994-13 (Internationales Schiedsgericht der Bundeskammer der gewerblichen Wirtschaft-Wien 1994) (Austria).

23. KRITZER, supra note 12, at 118 (noting that CISG art. 7(1) requires that 7(2) be interpreted so as to promote international uniformity, and predicting that tribunals will make every effort to find relevant general principles in the CISG before applying choice-of-law rules); Peter Winship, Private International Law and the U.N. Sales Convention, 2l CORNELL INT'L L.J. 487, 529-30 (1988) (suggesting that if tribunals take a properly generous approach to the CISG text, there should be little need to resort to choice-of-law rules).

24. CISG, supra note I, art. 4.

25. Although article 35 does not use the term "warranty," it imposes on the Seller obligations very similar to the express warranties, implied warranty of merchantability, and implied warranty of fitness for a particular purpose imposed by the American Uniform Commercial Code. Compare CISG, supra note I, art. 35 with U.C.C. §§ 2-313, 2-314, 2-315 (2000).

26. HONNOLD, supra note 14, para. 71; Rolf Herber, Article 4, in COMMENTARY ON THE UN CONVENTION ON THE INTERNATIONAL SALE OF GOODS (CISG) 42, 47 (Peter Schlechtriem ed., Geoffrey Thomas trans., 2d ed. 1998).

Honnold notes, however, that in situations in which Seller asks Buyer to provide Remote Purchaser with Seller's warranty, a tribunal might find that Seller participated as a seller in the sales contract with Remote Purchaser. See HONNOLD, supra, para. 63.

27. Case I ZR 5/96, UNILEX D.1998-5, E.1998-5 (Bundesgerichtshof 1998) (Germany); Case 11 U 206/93, UNILEX D.1995-2, E.1995-2 (Oberlandesgericht Hamm 1995) (Germany); Case BG 9341/94, UNILEX D.1994-29.1, E.1994-29.1 (Bezirksgericht Arbon 1994) (Switz.); Herber, supra note 26, at 47.

28. CISG, supra note 1, art. 5.

29. KRITZER, supra note 12, at 96-97.

30. The buyer cannot rely on nonconformity of the goods unless he notified the seller of the nonconformity within a reasonable time after he discovered it or should have discovered it. CISG, supra note 1, art. 39(1). An outside limit is imposed on the reasonable time for notice (unless this is inconsistent with a contractual period of guarantee): buyer must notify seller "within a period of two years from the date on which the goods were actually handed over to the buyer ..." CISG, supra note 1, art. 39(2).

31. HONNOLD, supra note 14, para. 73 (suggesting that CISG rules displace domestic rules that turn on the same facts, regardless of "contract" or "tort" labels); Rolf Herber, Article 5, in COMMENTARY ON THE UN CONVENTION ON THE INTERNATIONAL SALE OF GOODS (CISG) 49, 50 (Peter Schlechtriem ed., Geoffrey Thomas trans., 2d ed. 1998); W. Khoo, Article 5, in COMMENTARY ON THE INTERNATIONAL SALES LAW: THE 1980 VIENNA SALES CONVENTION 49, 50 (C.M. Bianca & M.J. Bonell eds., 1987).

32. United Nations Convention on the Limitation Period in the International Sale of Goods (as amended by 1980 Protocol), opened for signature June 14, 1974, S. TREATY Doc. No. 103-10 (1993),13 I.L.M. 952 (1974) (entered into force in the United States Dec. 1, 1994) [hereinafter UN Limitations Convention].

33. Compare UN Limitations Convention, supra note 32, arts. 1-6, with CISG, supra note 1, arts. 1-6.

34. CISG, supra note 1, art. 4(a).

35. HONNOLD, supra note 14, para. 66; KRITZER, supra note 12, at 86; Helen Elizabeth Hartnell, Rousing the Sleeping Dog: The Validity Exception to the Convention on Contracts for the International Sale of Goods, 18 YALE J. INT'L L. 1, 64 (1993) (citations omitted); Herber, supra note 26, at 44.

36. See HONNOLD, supra note 14, para. 66; KRITZER, supra note 12, at 86; Hartnell, supra note 35, at 64-65 (citations omitted); Herber, supra note 26, at 44.

37. HONNOLD, supra note 14, § 65; KRITZER, supra note 12, at 86; Hartnell, supra note 35, at 70-72.

However, the proper treatment of seller's nonfraudulent misrepresentation of the quality of goods already identified to the contract is controversial (because there is a breach of express warranty, a matter expressly addressed by CISG rules). See Hartnell, supra note 35, at 76-77.

38. KRITZER, supra note 12, at 86; Hartnell, supra note 35, at 70.

39. JOHN HONNOLD, UNIFORM LAW FOR INTERNATIONAL SALES UNDER THE 1980 UNITED NATIONS CONVENTION paras. 232-35 (2d ed. 1991); KRITZER, supra note 12, at 82; Hartnell, supra note 35, at 83-84.

40. See Case VIII ZR 134/96, 1997 NJW 3309, UNILEX D. I997-13, E. I997-13 (Bundesgerichtshof 1997) (Germany); HONNOLD, supra note 14, § 64; Hartnell, supra note 35, at 79, 80; Herber, supra note 26, at 44.

41. See Case HG 48/1994, UNILEX D.1995-23, E.1995-23 (Handelsgericht St. Gallen 1995) (Switz.); Herber, supra note 26, at 45. However, the proper treatment of mutual mistake that causes seller to misdescribe goods already identified to the contract is controversial (because there is a breach of an express warranty, a matter expressly addressed by CISG provisions). See KRITZER, supra note 12, at 90-91; Hartnell, supra note 35, at 72-77; Herber, supra note 26, at 45.

42. U.C.C. § 2-316(2), (3) (2000).

43. Hartnell, supra note 35, at 86 (citing U.C.C. § 2-316 cmt. 1 and § 2-302 cmt. 1).

44. CISG, supra note I, art. 35(2).

45. Id. art. 6 (quoted supra note 4).

46. Id. art, 8(2).

47. For similar arguments, see HONNOLD, supra note 39, paras. 230-34; KRITZER, supra note 12, at 27,81-82.

48. CISG, supra note 1, art. 4(b).

49. KRITZER, supra note 12, at 93.

50. HONNOLD, supra note 14, para. 70.

51. See supra note 4.

52. It is generally agreed that opting-out may be accomplished either by express agreement or by clearly implied agreement. See Decision of October 18, 1995, 1995-96 No. 40 Rechtskundig Weekblad 1378, UNILEX D.1995-28.1.0, E.1995-28.1.0 (Rechtbank van Koophandel, Hasselt, 1995) (Belgium); Case 15 U 29/92, 1993 NJW-RR 1316, UNILEX D. 1992-28, E.1992-28 (Oberlandesgericht Karlsruhe 1992) (Germany); Case 16 S 40/91, UNILEX D.1991-6, E.1991-6 (Landgericht Stuttgart 1991) (Germany); Societa X v. Societa Y, 1994 (8.3-4) Diritto del Commercio Intemazionale 861, UNILEX D.1994-9, E. 1994-9 (Ad hoc Arbitral Tribunal-Firenze 1994); HONNOLD, supra note 14, para. 77; Bonell, supra note 18, at 54; Rolf Herber, Article 6, in COMMENTARY ON THE UN CONVENTION ON THE INTERNATIONAL SALE OF GOODS (CISG) 52, 53, 54-55 (Peter Schlechtriem ed., Geoffrey Thomas trans., 2d ed. 1998).

53. See supra text accompanying notes 5-6.

54. For a similar argument, see Bonell, supra note 18, at 61.

55. Id. at 58-59; Herber, supra note 52, at 54, 57.

56. CISG, supra note 1, art. 92(1). The English-language text refers to "ratification, acceptance, approval or accession," but I am using the word "ratify" to signify any of these technically different modes of adopting a treaty.

57. No state has yet ratified the treaty with an article 92 reservation rejecting the Part III rules, and it is unlikely that any state will do so. There would be little reason to ratify the CISG without its rules concerning the obligations of the parties. In our discussion of article 92 reservations, we will therefore focus on reservations rejecting only the Part II contract formation rules.

58. CISG, supra note 1, art. 92(2).

59. Id. art. l(l)(a) (referring to parties located in different "Contracting States").

60. See supra note 2.

61. In some unusual situations, it is conceivable that an American forum's choice-of-law rules would eventually lead to the application of CISG Part II rules, despite an article 92 reservation. HONNOLD, supra note 14, paras. 47.4, 47.5, 467; KRITZER, supra note 12, at 77-78; M. Evans, Article 95, in COMMENTARY ON THE INTERNATIONAL SALES LAW: THE 1980 VIENNA SALES CONVENTION 654, 656-57 (C.M. Bianca & M.J. Bonell eds., 1987).

62. See supra note 19 and accompanying text.

63. CISG, supra note 1, art. 11.

64. See id. art. 29(1)

65. Id. arts. 18(1), 18(2),20(1),24.

66. Id. art. 96.

67. Id. art. 12.

68. Case AZ 12.G.41.471/1991, UNILEX D.1992-8, E. 1992-8 (Metropolitan Ct. of Budapest 1992) (Hungary); HONNOLD, supra note 39, para. 129; KRITZER, supra note 12, at 143; Herber, supra note 26, at 43-44; J. Rajski, Article 96, in COMMENTARY ON THE INTERNATIONAL SALES LAW: THE 1980 VIENNA SALES CONVENTION 658, 659 (C.M. Bianca & M.J. Bonell eds., 1987); Peter Schlechtriem, Article 12, in COMMENTARY ON THE UN CONVENTION ON THE INTERNATIONAL SALE OF GOODS (CISG) 91, 91-92 (Peter Schlechtriem ed., Geoffrey Thomas trans., 2d ed. 1998). But see Vital Berry Marketing BV v. Dira-Frost F.F.I. BV, UNILEX D.1995-15.1.2, E.1995-15.1.2 (Rechtbank van Koophandel, Hasselt 1995) (Belgium); HONNOLD, supra note 14, § 129 (taking a position contrary to that of the previous edition).

69. The Uniform Commercial Code rules require writings for some but not all manifestations of assent. U.C.C. §§ 2-201, 2-205, 2-206, 2-209, 2-316 (2000).

Most commentators agree that if the forum's choice-of-law rules point to the law of a state that has not made an article 96 reservation, the applicable law concerning writing requirements is that state's domestic law (e.g., New York's U.C.C.) and not that state's CISG law. KRITZER, supra note 12, at 143; Herber, supra note 26, at 43-44; Rajski, supra note 68, at 659. But see Schlechtriem, supra note 68, at 92 (noting disagreement and suggesting that CISG art. 11 should be applied).

70. If the sales contract is within the scope of U.C.C. Article 2, a state court dealing with an issue not resolved by the CISG must follow section 1-105 in deciding whether to apply the U.C.C. rules of its state or to apply law from a source outside its state. A federal court exercising diversity jurisdiction over a case involving an American party and a non-American party must use the choice-of-law rules of the state in which it sits, see Klaxon Co. v. Stentor Elec. Mfg. Co., 313 U.S. 487 (1941), and must therefore use section 1-105.

Because CISG treaty law is federal law , a federal court might also base its subject-matter jurisdiction on federal questions. When a federal court has both diversity jurisdiction and federal question jurisdiction, it is not clear whether the court is to use federal common law choice-of-law rules or the choice-of-law rules of the state in which it sits. See EUGENE F. SCOLES ET AL., CONFLICT OF LAWS § 3.36, at 176 n.1 (3d ed. 2000).

71. U.C.C. § 1-105(1) (2000).

72. See, e.g., Allmond Assocs., Inc. v. Hercules Inc., 960 F. Supp. 1216, 1228 n.9, 1236 n.23 (E.D. Mich. 1997) (interpreting Michigan UCC § 1-105); Gen. Instrument Corp., F. W. Sickles Div. v. Pa. Pressed Metals, Inc., 366 F. Supp. 139, 146 (M.D. Pa. 1973) (interpreting Pennsylvania UCC § 1-105).

73. See, e.g., Thornton v. Cessna Aircraft Co., 886 F.2d 85,90 (4th Cir. 1989) (interpreting South Carolina UCC § 1-105); Westerman v. Sears, Roebuck & Co., 577 F.2d 873, 879 and n.7 (5th Cir. 1978) (interpreting Florida UCC § 1-105); P&E Elec., Inc. v. Util. Supply of Am, Inc., 655 F. Supp. 89,93-94 (M.D. Tenn. 1986) (interpreting Tennessee UCC § 1-105); Madaus v. November Hill Farm, 630 F. Supp. 1246, 1248 (W.D. Va. 1986) (interpreting Virginia UCC § 1-105); Golden Plains Feedlot, Inc. v. Great Western Sugar Co., 588 F. Supp. 985, 989-91 (D.S.D. 1984) (interpreting South Dakota UCC § 1-105); United Overseas Bank v. Veneers, Inc., 375 F. Supp. 596, 600-01 (D. Md. 1974) (interpreting Maryland UCC § 1-105); Travenol Lab., Inc. v. Zotal, Ltd., 474 N.E.2d 1070, 1073 (Mass. 1985); Boudreau v. Baughman, 368 S.E.2d 849,854-55 (N.C. 1988).

74. U.C.C. § 1-105 cmt. 3.

75. Boudreau, 368 S.E.2d at 855; JOHN A. SPANOGLE & PETER WINSHIP, INTERNATIONAL SALES LAW: A PROBLEM-ORIENTED COURSEBOOK 52 (2000); Peter Winship, Private International Law and the U.N. Sales Convention, 21 CORNELL INT'L L.J. 487, 526 n.183 (1988).

76. See, e.g., Golden Plains Feedlot, 588 F. Supp. at 990; United Overseas Bank, 375 F. Supp. at 601; Boudreau, 368 S.E.2d at 855.

77. U.C.C. Revised Article 1 § 1-301(a) (proposed Discussion Draft of April 14, 2000). A reporter's note points out that the policy of gaining widespread application of the UCC among the states now has little vitality and suggests that international comity militates against a pro-forum bias in international transactions. Id; reporter's note e.

78. 125 Mass. 374. (1878).

79. RESTATEMENT (FIRST) OF CONFLICT OF LAWS § 332 (1934).

80. Symeon C. Symeonides, Choice of Law in the American Courts in 1998: Twelfth Annual Survey, 47 AM. J. COMP. L. 327, 334-45 (1999) (indicating that although support for the rule in some of these states is weak or subject to exception, ten states have not yet abandoned the rule: Alabama, Florida, Georgia, Kansas, Maryland, New Mexico, Rhode Island, South Carolina, Tennessee, Florida Virginia). The same observations are presented in SCOLES ET AL., supra note 70, § 2.20, at 86 table 5, § 2.21, at 91-95.

81. RESTATEMENT (FIRST) OF CONFLICT OF LAWS § 311 cmt. d, §§ 312-31 (1934).

82. Id. § 325, § 326 cmt. a (employing the generally accepted American contract law rule that the contract is formed when the offeree dispatches her acceptance).

83. CISG, supra note 1, art. 18(2).

84. See generally ROBERT A. LEFLAR ET AL., AMERICAN CONFLICTS LAW § 146 (4th ed. 1986).

85. RESTATEMENT (FIRST) OF CONFLICT OF LAWS §§ 353-72 (place of performance), §§ 311-52 (place of contracting) (1934).

86. Id. § 355 and cmt. a, § 358.

87. 63 N.E.2d 417 (Ind. 1945).

88. 124 N.E.2d 99 (N.Y. 1954).

89. LEFLAR ET AL., supra note 84, §§ 90, 149.

90. Id. § 149. It has been estimated that twenty-five states use the Restatement (Second) approach for contract issues, that five states use a similar significant contacts approach, and that at least six states combine the Restatement (Second) approach with other approaches. SCOLES ET AL., supra note 70, §§ 2.20-2.25.

91. RESTATEMENT (SECOND) OF CONFLICT OF LAWS § 188(1) (1971).

92. Id. § 188 cmt. d.

93. Id. § 188(1).

94. Id. § 6(2).

95. The goal here is to further harmonious relations between states and facilitate commercial intercourse between them. Id. § 6 cmt. d.

96. Generally, the state whose interests are most deeply affected should have its local law applied. Id. § 6 cmt. f.

97. Id. § 188(2).

98. "[T]he place of contracting is the place where occurred the last act necessary, under the forum's rules of offer and acceptance, to give the contract binding effect. Id. § 188 cmt. e (emphasis added).

99. Id. § 188(2).

100. SCOLES ET AL., supra note 70, § 2.14, at 66-67 (noting that a non-neutral court could magnify the forum-oriented policies of § 6 into a general preference for forum law).

101. The Restatement (Second) has often been criticized for its failure to produce predictable and uniform outcomes. See, e.g., LEFLAR ET AL., supra note 84, § 149, at 224-26; SCOLES ET AL., supra note 70. § 2.14, at 67.

102. RESTATEMENT (SECOND) OF CONFLICT OF LAWS § 201 cmt. d (1971) (suggesting that the protection of party expectations has a relatively small role to play with respect to such issues).

103. Id. § 191.

104. Id. § 191 cmt. d.

105. INCOTERMS 2000, 49-55 (International Chamber of Commerce 1999). Under the Incoterms, all "F" terms (FCA, FAS, FOB) and an "C" terms (CFR, CIF, CPT, CIP) create shipment contracts, with the seller completing his delivery performance in the country of shipment. Id. at 12-14. All "D" terms (DAF, DES, DEQ, DDU, DDP) create destination (arrival) contracts, with the seller's delivery performance completed in the destination country. Id. at 18.

106. Id. at 97-103.

107. In the international sales context, the Restatement (Second) does make one valuable contribution: the ready acceptance of dépeçage. See infra Part III.C.

108. SCOLES ET AL., supra note 70, § 2.20, at 86 table 5, § 2.24, at 100, § 2.25. The authors indicate that at least nine American jurisdictions use a combination of approaches that includes some form of interest analysis: California, District of Columbia, Hawaii, Massachusetts, New Jersey, New York, North Dakota, Oregon, and Pennsylvania. Id. § 2.25.

109. For concise summaries of Currie's theory, see LEFLAR ET AL., supra note 84, § 91; LUTHER L. McDOUGAL III ET AL., AMERICAN CONFLICTS LAW: CASES AND MATERIALS 28-29 (3d ed. 1998); SCOLES ET AL., supra note 70, § 2.9, at 25-33.

110. See LEFLAR ET AL., supra note 84, § 91; SCOLES ET AL., supra note 70, § 2.9, at 31-32 (noting that the "comparative impairment" method seems to involve a weighing of interests), §§ 2.24, 2.25 (indicating that states using interest analysis for contract or tort cases either weigh interests or assess comparative impairment).

111. Like Currie's interest analysis, Robert A. Leflar's proposed approach has a forum-bound tendency. Leflar's approach has been adopted for contracts conflicts in Minnesota and Wisconsin. See SCOLES ET AL., supra note 70, § 2.25, at 101. Leflar proposed that choice of law be influenced by five primary considerations: (1) predictability of results; (2) maintenance of interstate and international order; (3) simplification of the judicial task; (4) advancement of the forum's governmental interests; and (5) application of the better rule of law. For summaries of Leflar's approach, see LEFLAR ET AL., supra note 84, § 95; SCOLES ET AL., supra note 70, § 2.13. We can agree that the first three considerations are important goals. In the interests of international comity, however, foreign interests should not be subordinated to forum interests. The search for the "better rule of law" all too easily results in the application of forum law. Id. at 54-56 (citing illustrative Minnesota cases). The goal of substantive justice is better served by choice-of-law rules that direct a court to apply whichever state's rule validates the contract, or whichever state's products liability rule favors the victim, or whichever state's rule promotes some other policy designated in the choice-of-law rule itself.

112. Note that the present Uniform Commercial Code section 1-105 does not preclude application of international law. U.C.C. § 1-105(1) (2000). Although the first sentence of subsection (1) limits party choice of law to the law of a state or nation, no such limitation is imposed in the second sentence, which applies when the parties have not effectively chosen applicable law.

Nor would section 1-301 in the proposed revision of Uniform Commercial Code Article 1 preclude judicial selection of international sources of law. U.C.C. Revised Article 1 § 1-301 (a) (proposed Discussion Draft of April 14, 2000). The second sentence in subsection (a) refers to "the law" that would be selected by the forum's conflict-of-laws principles. Reporter's note e states that this second sentence in subsection (a) determines which "jurisdiction's" law governs. Id. reporter's note e. But use of the word "jurisdiction's" seems inadvertent, and we may hope that it will either be avoided in the final comments or ignored by the courts.

113. The following discussion of the medieval lex mercatoria is largely based on HAROLD I. BERMAN, LAW AND REVOLUTION 333-56 (1983); HENRI PIRENNE, MEDIEVAL CITIES (Frank D. Halsey trans., 1952); THEODORE F.T. PLUCKNETT, A CONCISE HISTORY OF THE COMMON LAW 657-70 (5th ed. 1956); RUDOLF B. SCHLESINGER ET AL., COMPARATIVE LAW: CASES-TEXT-MATERIALS 301-04 (5th ed. 1988); RICHARD E. SPEIDEL ET AL., TEACHING MATERIALS ON COMMERCIAL AND CONSUMER LAW 624-34 (3d ed. 1981); LEONE. TRAKMAN, THE LAW MERCHANT 1-44 (1983).

114. CISG, supra note 1, art. 9(2). If a trade usage applicable under article 9(2) conflicts with a substantive rule of the CISG, the usage prevails. See Bermatex S.r.l. v. Valentin Rius Clapers S.A., UNILEX D. 1994-24.2, E.I994-24.2 (Juzgado Nacional de Primera Instancia en lo Commercial No. 10 1994) (Argentina); Case 20b 191/98X, UNILEX D.1998-17.2, E.1998-17.2 (Oberster Gerichtschof 1998) (Austria); Case I U 69/92, UNILEX D.1993-2.1, E.1993-2.1 (Oberlandesgericht Saarbrücken 1993) (Germany); JOHN O. HONNOLD, UNIFORM LAW FOR INTERNATIONAL SALES UNDER. THE 1980 UNITED NATIONS CONVENTION, para. 122 (3d ed. 1999); Werner Junge, Article 9, in COMMENTARY ON THE UN CONVENTION ON THE INTERNATIONAL SALE OF GOODS (CISG) 75, 76 (Peter Schlechtriem ed., Geoffrey Thomas trans., 2d ed. 1998).

115. UNIDROIT (INTERNATIONAL INSTITUTE FOR THE UNIFICATION OF PRIVATE LAW), PRINCIPLES OF INTERNATIONAL COMMERCIAL CONTRACTS (1994) [hereinafter UNIDROIT PRINCIPLES].

116. See generally MICHAEL JOACHIM BONELL, AN INTERNATIONAL RESTATEMENT OF CONTRACT LAW 16-17, 31, 32 (2d enlarged ed. 1997) (espousing the benefit of applying UNIDROIT Principles).

117. Alejandro M. Garro, The Gap-filling Role of the UNIDROIT Principles in International Sales Law: Some Comments on the Interplay Between the Principles and the CISG, 69 TUL. L. REV. 1149, 1164-66 (1995).

118. See generally BONELL, supra note 116, at 9; JOHN A. SPANOGLE & PETER WINSHIP, INTERNATIONAL SALES LAW: A PROBLEM-ORIENTED COURSEBOOK 41 (2000); Garro, supra note 117, at 1153; Luther L. McDougal III, "Private" International Law: Ius Gentium Versus Choice of Law Rules or Approaches, 38 AM. J. COMP. L. 521, 533 (1990) (observing that domestic rules reflect domestic policies and ignore transnational policies; even when domestic policies and transnational policies are the same, a particular resolution of conflicting policies may be appropriate for domestic transactions but inappropriate for international transactions).

119. UNIDROIT PRINCIPLES, supra note 115, pmbl. cmt. 5 (observing that application of UNIDROIT rules avoids applying domestic law which will usually be more familiar to one party than to the other); Garro, supra note 117, at 1153-54 (making the same observation).

120. Convention on the Law Applicable to International Sales of Goods, June 15, 1955, 510 U.N.T.S. 147 [hereinafter 1955 Hague Convention].

121. Id. art. 3 para. 1.

122. Id. art. 3 para. 2.

123. Convention on the Law Applicable to Contracts for the International Sale of Goods, Dec. 22, 1986, 24 I.L.M. 1575 [hereinafter 1986 Hague Convention].

124. Id. art. 8(1). If the seller has more than one place of business, "the relevant place of business is that which has the closest relationship to the contract and its performance ..." Id. art. 14(1).

125. The contract is governed by the law of the buyer's state if the "negotiations were conducted, and the contract concluded by and in the presence of the parties, in that State." Id. art. 8(2)(a). But this exception will seldom apply, because it seems to require that all negotiations be conducted in the buyer's state and that the contract be signed in a face-to-face meeting of the parties in that state. Francis A. Gabor, Emerging Unification of Conflict of Laws Rules Applicable to the International Sale of Goods: UNCITRAL and the New Hague Conference on Private International Law, 7 NW. J. INT'L L. & Bus. 696, 713-15 (1986).

The contract is governed by the law of the buyer's state if the contract is expressly a destination contract rather than a shipment contract. See 1986 Hague Convention, supra note 123, art. 8(2)(b). But most international sales contracts are shipment contracts.

The contract is governed by the law of the buyer's state if the contract terms were largely dictated in the buyer's invitation for bids from prospective sellers. See id. art, 8(2)(c). Obviously, this exception will not apply often.

The contract is governed by the law of a state other than the seller's state if the contract is "manifestly more closely connected" with that other state. Id. art. 8(3). In most international sales, however, the connecting factors will be rather evenly split between the buyer's state and the seller's state, and it would be difficult to find that the buyer's state (or a third state) was "manifestly" more closely connected than the seller's state.

126. Convention on the Law Applicable to Contractual Obligations, opened for signature June 19, 1980, 1980 0.1. (L 266) 1, 19 I.L.M. 1492 (1980) [hereinafter 1980 Rome Convention].

127. Id. art. 4(1).

128. Id. art. 4(2).

129. Mario Giuliano & Paul Lagarde, Report on the Convention on the Law Applicable to Contractual Obligations, 1980 O.J. (C 282) I, 20, 21.

Note that although seller's delivery is the characteristic performance, the governing law is the law of the country where the seller is located, not the law of the country in which the seller completes performance of his delivery obligation. The basic rule of the 1980 Rome Convention thus differs from the rule in the American Restatement (Second) of Conflict of Laws section 191.

130. See, e.g., PETER KAYE, THE NEW PRIVATE INTERNATIONAL LAW OF CONTRACT OF THE EUROPEAN COMMUNITY 57 (1993); Campbell McLachlan, The New Hague Sales Convention and the Limits of the Choice of Law Process, 102 L.Q. REV. 591, 605 (1986).

131. Article 4(5) provides an exception. The article 4(2) presumption does not apply (and thus the law of seller's state does not apply) if the contract is "more closely connected" with some other country. See 1980 Rome Convention, supra note 126, art. 4(5). Although article 4(5) does not use the "manifestly more closely connected" language of 1986 Hague Convention article 8(3), it would seem that the typical international sales contract has equally close connections with seller's state and buyer's state and that the article 4(5) exception should be used sparingly.

132. The fourth and most recent choice-of-law treaty is the 1994 Inter-American Convention on the Law Applicable to International Contracts, opened for signature Mar. 17, 1994, OAS Doc. OEA/Ser. K/XXI.5, 33 I.L.M. 732 (1994) [hereinafter 1994 OAS Convention]. The Convention was approved at a Mexico City conference sponsored by the Organization of American States. Although the Convention has not yet been ratified by the United States, it has been ratified by Mexico and Venezuela and entered into force in 1997.

Article 9 of this Convention provides that if the parties have not effectively chosen the applicable law, "the contract shall be governed by the law of the State with which it has the closest ties." 1994 OAS Convention, supra, art. 9 para. 1. We should not assume that the quoted language necessarily or primarily refers, in a sale of goods case, to the law of the seller's state. Although many portions of the Convention were based on the 1980 Rome Convention, there is not even a hint of a "characteristic performance" presumption. Instead, courts are merely directed to determine which state has the "closest ties" by taking into account "all objective and subjective elements of the contract." Id. art. 9 para. 2. The Convention thus seems to leave courts free to employ a "most significant relationship" approach.

133. Gabor, supra note 125, at 703; Kurt Lipstein, Characteristic Performance - A New Concept in the Conflict of Laws in Matters of Contract for the EEC, 3 NW. J. INT'L L. & Bus. 402, 411-13 (1981).

134. See, e.g., LEFLAR ET AL., supra note 84, § 96; McDOUGAL III ET AL., supra note 109, at 118; SCOLES ET AL., supra note 70, § 2.14, at 60; Willis L.M. Reese, Dépeçage: A Common Phenomenon in Choice of Law, 73 COLUM. L. REV. 58, 58, 75 (1973).

135. RESTATEMENT (SECOND) OF CONFLICT OF LAWS § 188(1), (2), cmt. d.

136. This method is favorably considered in Reese, supra note 134, at 74.

137. Luther L. McDougal III. "Private" International Law: Ius Gentium Versus Choice of Law Rules or Approaches, 38 AM. J. COMP. L. 521, 535-37 (1990) (suggesting that judges develop international case law rather than apply domestic rules).

138. Economic, social, and legal conditions vary greatly from one country to another. In the absence of international treaties providing substantive products liability law, each state should be allowed to decide how to allocate losses caused by defective products that enter its territory. An attempt to impose international case law rules would ignore variations in local conditions and could impair international comity.

139. SCOLES ET AL., supra note 70, § 17.65, at 824, 827.

140. Id. §§ 17.66-17.73 (surveying recent cases). Of the ten court decisions named in the text, five were based on interest analysis or a "most significant relationship" approach in which interest analysis played a major role. No other choice-of-law method was used more than once.

141. I assume that each state has at least a narrow protection goal and a narrow compensation goal. In purely domestic cases, these goals are apt to conflict and must be compromised in the state's products liability law. In international cases, a state can have an interest in promoting only one of these goals (either protection for a domestic seller or compensation for a domestic plaintiff).

142. The Convention on the Law Applicable to Products Liability, adopted at a Hague Conference session in 1972, is a small step in this direction. Convention on the Law Applicable to Products Liability, opened for signature Oct. 1, 1973, arts. 4-7, 11 I.L.M. 1283. This international treaty on choice of law for products liability issues has entered into force among a number of European states. Under this treaty, the law of the state where the product was acquired by the victim may be applied, but only if that state is also the state where the victim resides or the state where the injury occurred. Id.

143. This is probably true even if the time limitation is contained in a statute that created the substantive right and could be characterized as a "statute of repose" which bars the right and not merely the remedy.

144. Article 49 merely provides that the beneficiary may assign proceeds "in accordance with the provisions of the applicable law." ICC UNIFORM CUSTOMS AND PRACTICE FOR DOCUMENTARY CREDITS, ICC Publication No. 500 art. 49 (1993).

We will ignore the possibility that the beneficiary has "transferred" the credit pursuant to UCP article 48.

145. U.C.C. § 5-116(a), (b) (2000). Section 5-116 deals with the liability of an issuer or "nominated person." A confirmer is a nominated person. Id. § 5-102(a)(4).

146. Id. Revised § 9-306(a), (b). However, section 9-306 does not apply if the security interest is only perfected automatically under section 9-308(d) (automatic perfection of a security interest in letter-of-credit rights by perfecting a security interest in seller/beneficiary's account arising from the sales contract). Id. Revised § 9-102(a)(77) (defining "supporting obligation"), § 9-306(c), § 9-308(d).

Furthermore, if section 5-116 points to the law of a foreign country, section 9-306 does not apply, and the choice-of-law rule in section 9-301(1) applies. Id. Revised § 9-102(76) (defining "State" to exclude foreign countries), § 9-301(1) (providing a baseline rule that applies when sections 9-303 through 9-306 do not apply), § 9-306(a) (making applicability of its rule depend on the section 5-116 jurisdiction being a State). Section 9-301(1) provides that perfection and priority of the security interest are governed by the local law of the jurisdiction where the debtor (our seller) is located. Id. Revised § 9-301(1). If the debtor is an American corporation, it is located in its state of incorporation. Id. Revised § 9-102(a)(70) (defining "registered organization"), § 9-307(e). If the debtor is a foreign corporation, it is located either in the country where it has its chief executive office or in the District of Columbia, depending on whether its country maintains a security interest filing system comparable to the UCC filing system. Id. Revised § 9-307(b), (c).

147. See U.C.C. Revised Art. 1 § 1-201(38) (proposed Discussion Draft of April 14, 2000) (defining "security interest"); U.C.C. Revised § 9-109(a) (2000) (bringing within the scope of Article 9 outright sales of only accounts, chattel paper, payment intangibles, or promissory notes).

148. If there is no confirmer, the court should, of course, apply the law of the jurisdiction where the issuer's branch was located.

149. U.C.C. Revised § 9-102(a)(2) (2000).

150. Id. Revised § 9-109(a)(1), (3).

151. Id. Revised § 9-301(1).

152. Id. Revised § 9-307(e).

153. Id. Revised § 9-307(b), (c). Hans Kuhn indicates that most foreign legal systems will not pass the Section 9-307(c) comparability test. Hans Kuhn, Multistate and International Secured Transactions Under Revised Article 9 of the Uniform Commercial Code, 40 VA. J. INT'L L. 1009, 1048 (2000). Kuhn also suggests that use of the Article 9 choice-of-law rules requiring application of UCC filing requirements in transactions having little connection with the United States would violate the due process standard developed in Allstate Insurance Co. v. Hague, 449 U.S. 302 (1981) and subsequent decisions. See Kuhn, supra, at 1081-82, 1084, 1101.

154. U.C.C. Revised Art. 1 § 1-301(a (proposed Discussion Draft of April 14, 2000).

155. UNIDROIT PRINCIPLES, supra note 115, art. 3.2 and cmts. 1, 2 (providing that neither consideration nor cause is required for contract validity).

156. Id. art. 3.3. In some legal systems, a sales contract is invalid if at the time of contract formation, the specified goods no longer exist or the seller cannot convey legal title; article 3.3 provides that such facts do not invalidate the contract.

157. Id. arts. 3.4-3.7.

158. Id. arts. 3.8, 3.11.

159. Id. arts. 3.9, 3.11.

160. Id. arts. 3.10, 3.11.

161. Id. arts. 3.12-3.18.

162. Id. art. 7.1.6.

163. Id. art. 7.4.13.

164. For an explanation of how this issue could be characterized as either a validity issue or an interpretation issue, see supra text accompanying notes 42-47.

165. UNIDROIT PRINCIPLES, supra note 115, art. 3.1(a), (b).

166. Id. art. 3.1(c).

167. Convention on Agency in the International Sale of Goods, opened for signature Feb. 17, 1983. 22 I.L.M. 249.

168. For a discussion of the Convention's provisions, see Michael Joachim Bonell, The 1983 Geneva Convention on Agency in the International Sale of Goods, 32 AM. I. COMP. L. 717 (1984).

169. Convention on the Law Applicable to Agency, June 16, 1977, Hague Conference on Private International Law, 1 Actes et documents de la Treizieme Session 42-47 (1978), reprinted in 26 AM. I. COMP. L. 438 (1978).

170. For a discussion of the choice-of-law rules provided in this Convention, see SCOLES ET AL., supra note 70, § 18.36.

171. U.C.C. Revised §9-301(3)(C) (2000). Under section 9-301(1), perfection of the security interest would be governed by the law of the jurisdiction in which the debtor is located.

172. Id. § 2-402(2).

173. See CISG, supra note 1. art. 92(2); see also supra notes 58-61 and accompanying text. Because it is unlikely that a nation would make an article 92(1) declaration rejecting the CISG Part III rules, we are ignoring this possibility. See supra note 57.

174. The court should not use a rule of validation that selects whichever state's law affirms effective contract formation. Such a rule would not resolve all contract formation issues. In a battle of the forms, for example, it may be clear that a contract has been formed, and the issue may be which form's term is included in that contract. This question frequently arises in international sales cases and is addressed in Part II of the CISG. CISG, supra note 1, art. 19. But the CISG rule cannot be applied, and the court needs a choice-of-law rule that selects a domestic rule for this issue.

175. UNIDROIT PRINCIPLES, supra note 115, art. 1.2.

176. A rule of validation would usually lead to American UCC law. The reservation state is likely to have a rather rigorous statute of frauds. The American party's state, on the other hand, will have enacted the UCC section 2-201 statute of frauds (unless that state happens to be Louisiana). Compliance with section 2-201 is not difficult. U.C.C. § 2-201 (2000). The present version of this statute of frauds applies if the contract price is $500 or more. But under subsection (2), the statute can often be satisfied by a writing signed by the plaintiff. And under subsection (3), no writing is required with respect to goods that the buyer has received and accepted, or for which the buyer has made payment. Essentially the only sales contracts that would be unenforceable under this statute of frauds are purely oral contracts for goods that have neither been delivered nor paid for.

If the April 14, 2000 version of the proposed revision of UCC Article 2 is enacted, formal invalidity will be even more unlikely. The revised section 2-201 would raise the $500 threshold to $5000, would replace the present "signed writing" requirement with an "authenticated record" requirement that can be satisfied by the typical fax or e-mail message, and would retain the provisions found in the present subsections (2) and (3). U.C.C. Revised Article 2 § 2-103(a)(1) (defining "authenticate"), (a)(34) (defining "record"), § 2-201 (proposed Discussion Draft of April 14, 2000).

It should also be noted that under the federal Electronic Signatures in Global and National Commerce Act and under the Uniform Electronic Transactions Act, a contract record or signature cannot be denied effect solely because it is in electronic form. See Electronic Signatures in Global and National Commerce Act, Pub. L. No. 106-229, §§ 101(a), 103(a)(3), 106, 114 Stat. 464, 464, 468, 472 (2000); Uniform Electronic Transactions Act §§ 2, 3(b )(2) and cmts. 4, 7, § 7 (2000).

177. See LEFLAR ET AL., supra note 84, § 148 (citing empirical studies and numerous leading cases and concluding that American courts have usually chosen the law that upholds the contract, unless the contract was regarded as bad and vicious).

178. The 1980 Rome Convention provides that a "contract concluded between persons who are in different countries is formally valid if it satisfies the formal requirements of the law which governs it under this Convention [generally, the law of the seller's country] or of the law of one of those countries." 1980 Rome Convention, supra note 126, art. 9(2). The operative provision in the 1986 Hague Convention is almost identical to that of the Rome Convention. 1986 Hague Convention, supra note 123, art. 11(2). The 1955 Hague Convention does not apply to statute of frauds issues. 1955 Hague Convention, supra note 120, art. 5(2).

The general choice-of-law rule in the 1994 OAS Convention does not expressly point to the law of the seller's state. See supra note 132. This Convention contains a special rule for formal validity issues, and the rule has a strong validating tendency. 1994 OAS Convention, supra note 132, art. 13 para. 2 (providing that if the contract is formed by persons in different states, it is formally valid if it meets the requirements of the law that generally governs, or the law of either state from which the contract was formed, or the law of the place where the contract is to be performed).

179. SCOLES ET AL., supra note 70, §§ 3.15, 3.16. Courts using broad choice-of-law approaches, rather than choice-of-law rules, are likely to use forum policy as part of the choice-of-law process and thus unlikely to need a public policy exception. See id. § 3.16. I am proposing choice-of-law rules, however, and courts using choice-of-law rules will probably want to preserve a public policy exception.

180. SYMEON C. SYMEONIDES ET AL., CONFLICT OF LAWS: AMERICAN, COMPARATIVE, INTERNATIONAL: CASES AND MATERIALS 96 (1998).

181. RUDOLPH B. SCHLESINGER ET AL., COMPARATIVE LAW: CASES-TEXT-MATERIALS 587 and n.3 (5th ed. 1988) (noting that civil law systems make a similar distinction between ius dispositivum (not mandatory, subject to agreement otherwise) and ius cogens (mandatory)).

For example, the risk of loss rules in UCC section 2-509 are nonmandatory because they are subject to the contrary agreement of the parties. U.C.C. § 2-509(4) (2000). But the good faith rule of section 1-203 is mandatory (in this weak sense) because it may not be disclaimed by agreement of the parties. Id. § 1-102(3).

182. See BONELL, supra note 116, at 214-15; Hartnell, supra note 35, at 57-58.

183. Philip J. McConnaughay, Reviving the "Public Law Taboo" in International Conflict of Laws, 35 STAN. J. INT'L L. 255, 305-06 (1999).

184. Id. at 304-05 (noting that usury statutes and other private law mandatory rules are traditionally subject to contractual choice-of-law elections and judicial choice-of-law methods).

185. If the court deems the contract to be illegal, it must then determine the effects of illegality. Is the contract void and thus unenforceable by either party? Or is it merely voidable, so that one of the parties could elect to enforce it? Or can either party enforce it despite its illegality? Even if a party cannot enforce the contract, may she be awarded restitution for her performance?

In some cases, the statute that was violated prescribes the effects of illegality. In determining the effects of illegality when they are not statutorily prescribed, American and European courts tend to consider not only the comparative equities of the parties, but also the purpose of the statute that was violated and whether denying damages or restitution would substantially further that purpose. JOHN D. CALAMARI & JOSEPH M. PERILLO, THE LAW OF CONTRACTS §§ 22.1, 22.2, 22.3, 22.7 (4th ed. 1998) (summarizing American case law); KONRAD ZWEIGERT & HEIN KÖTZ, INTRODUCTION TO COMPARATIVE LAW 386-87 , 575-81 (Tony Weir trans., 3d rev. ed. 1998) (summarizing European case law). The best opinions concerning the purpose of the statute and what type of remedial relief might be compatible with that purpose will be found in the case law of the state whose statute was violated.

Therefore, when an American court invokes its public policy exception and finds the contract illegal under forum law, the effects of illegality should also be governed by forum law. But what if the contract is deemed illegal because it is illegal under the laws of both the seller's state and the buyer's state? For reasons similar to the reasons for employing a rule of validation for the illegality issue, the effects of illegality should be governed by whichever state's law allows the damages or restitution being sought. In order to facilitate international trade and respect the autonomy of the parties, American courts should seek to minimize the effects of illegality and carry out the intentions of the parties to the extent possible. But if neither state's law permits the relief being sought, the court must leave the parties where it finds them.

186. LEFLAR ET AL., supra note 84, § 152 (noting that although American courts usually do not announce that they are doing this, they tend to apply whichever state's law sustains an allegedly usurious contract).

187. We will assume that if the relevant contractual provision is illegal, it was already illegal at the time of contract formation. If the defendant argues that her performance became illegal under a new law enacted after the contract was formed, her defense would probably be governed by CISG article 79, which deals with supervening post-contract impediments to performance. See CISG, supra note 1, art. 79. Whether she has a good defense would depend on how the court interprets article 79 pursuant to the interpretation rule in article 7(1). Article 7(1) does not make room for a choice-of-law process. Compare CISG art. 7(1) with CISG art. 7(2).

188. The American court's decision on whether to apply Country C's law may depend on the outcome of other issues: obligations under the Articles of Agreement of the International Monetary Fund, Country C's prescriptive jurisdiction under international law standards, the court's duties under the federal Act of State doctrine, and so forth. But the decision should not depend on any choice-of-law rule prescribing how the court should choose between Country C regulatory public law and American regulatory public Jaw.

189. Hartford Fire Ins. Co. v. California, 509 U.S. 764, 794-99 (1993) (noting that the Sherman Act applies to foreign conduct intended to produce and actually producing some substantial effect in the United States, even if such conduct is also regulated by foreign law, and indicating that in such a case, the Sherman Act should be applied unless it prohibits conduct required by foreign law); United States v. Aluminum Co. of America, 148 F.2d 416, 443-44 (2d Cir. 1945) (holding that Congress can impose liability on foreign persons for conduct outside the United States if the conduct has intended effects in the United States; a court should be concerned only with whether Congress intended to impose such liability and whether such imposition was constitutional, not with whether liability would be imposed by another country); McConnaughay, supra note 183, at 268, 291, 293, 298-99 (suggesting that Hartford Fire reaffirms the Alcoa principle that when the internationally mandatory public law of the American forum applies, it should not be displaced by choice-of-law analysis).

The authorities cited above deal with situations in which the question is whether the defendant should be subject to civil or criminal liability for conduct that violates American public law. These authorities do not deal with our hypothetical situation in which a defendant who has been sued for breach of an international sales contract argues that her contractual obligation is unenforceable because performance would have been illegal. I see no reason, however, why it should matter whether the defendant performed her contract and violated American public law, or complied with that public law and broke her contractual promise. Either way, the question is whether the promised performance was illegal, and the answer should be that it was illegal if it was prohibited by the internationally mandatory public law of the American forum.

190. Aluminum Co. of America, 148 F.2d at 443 (indicating that when a federal statute is construed to apply to extraterritorial conduct, American courts do not ask whether the statute violates international law; they merely ask whether it is constitutional); United States v. Georgescu, 723 F. Supp. 912. 921 (E.D .N .Y. 1989) (indicating that a federal statute is controlling, even if Congress's exercise of prescriptive jurisdiction was contrary to preexisting international law); RESTATEMENT (THIRD) OF FOREIGN RELATIONS LAW OF THE UNITED STATES § 402 cmt. i (noting that "when the intent of Congress to apply United States law in particular circumstances is clear, courts will apply that law notwithstanding any inconsistency" with international law), §403 cmt. g (noting that if a federal statute cannot be construed to be consistent with international law, the statute is nevertheless valid, but its application may subject the United States to international responsibility) ( 1986).

If the American public law is not federal but state law, an American court may hold the statute invalid for contravening international law limitations on prescriptive jurisdiction; international law is federal law and is thus supreme over state law. See id. § 402 cmt. k.

191. Loucks v. Standard Oil Co. of New York, 120 N.E. 198, 201-02 (N.Y. 1918).

192. In considering whether a foreign law is contra bonos mores, German courts have often looked for an international standard of justice. SCHLESINGER ET AL., supra note 181, at 857-63, 863 n.3.

Pace Law School Institute of International Commercial Law> - Last updated January 17, 2003

---