Rewarding Ignorance of the CISG: A Response to John Coyle
June 13, 2022
In a recent post, Professor John Coyle considers the interpretation of the following choice of law (“COL”) clause in an international contract for sale of goods where both parties are located in Contracting States to the U.N. Convention on Contracts for the International Sale of Goods (CISG): “This Agreement shall be governed by the laws of the State of New York.” The specific issue is whether the clause should be read to exclude application of the CISG, as permitted by Article 6. Coyle argues that, employing what might be called a “U.S. majority expectations” approach to contract interpretation, such a clause should be interpreted to exclude the CISG because that is the intention of “most U.S. companies” that employ such generic COL clauses in their international sales contracts.
Coyle accomplishes three rather significant tasks, not all of which he may have intended. First, he sets out a “U.S. majority expectations” approach to contract interpretation under the CISG (an approach with which we disagree). Second, he demonstrates the problems with presuming that drafters (at least U.S. drafters) of international sales contracts should be rewarded for unprofessional ignorance of the law. Third, he proposes “solutions” that are not likely possible and, in any event, would not be a good idea and run counter to U.S. treaty obligations.
The “U.S. Majority Expectations” Approach
The rationale for a majority expectations approach is obvious: in interpreting ambiguous contract provisions, a reading that reflects the result intended by the majority is most likely to capture the intention of the parties to the contract.
However, we have several reservations about a majority expectations approach generally. First, such an approach undercuts another principle of contract interpretation – one that is internationally accepted according to Article 4.6 of the UNIDROIT Principles of International Commercial Contracts – the contra proferentem principle. That principle indicates that ambiguous contract language should be interpreted against the drafter of the language, thus creating an incentive to avoid ambiguities. By discovering and favoring what the drafter likely intended to say, instead of choosing an interpretation favoring the other side, the majoritarian approach eliminates that penalty and the incentive for clear contract drafting that the contra proferentem rule is designed to create.
Second, majoritarian intentions regarding ambiguous language may change over time, although we do not claim this has occurred with respect to generic COL clauses and the CISG. Of course, empirical studies could keep track of this changeability, but this would be difficult and expensive.
Third, and more fundamentally, application of the majoritarian approach itself alters the evolution of majoritarian expectations over time. For example, application of the majoritarian approach to generic COL clauses during the early years of the CISG could have interrupted the process of lawyers becoming familiar with (and thus more willing to apply) the CISG. After all, if any ambiguous contract language would be interpreted to preclude the Convention’s application, why bother learning about the Convention (or even cleaning up obviously problematic COL language)?
These problems apply to majoritarian approaches in general. But Coyle’s argument suffers from an additional problem: he considers only the expectations of U.S. parties, which is why we have termed his approach the “U.S. majority expectations” approach.
Rewarding Unprofessional Ignorance
Coyle’s basic premise is that tribunals in the United States – and presumably the rest of the world – should adopt a rule presuming that a choice of law clause selecting the law of a U.S. state in an international sales contract represents a choice of the domestic sales law of the chosen state (i.e., except in Louisiana, the Uniform Commercial Code as adopted in the selected state), and not the CISG. But this premise requires a number of rather destructive contortions of private international law itself, and of basic treaty law.
Coyle’s interpretive rule is based on his empirical study of contracts drafted by U.S. parties, including interviews with U.S. contract drafters. The initial problem here is that the inquiry was limited to parties from only one Contracting State. When the default law is the CISG, the contract interpretation rules of the CISG must apply to the interpretation of the COL clause. After all, until the COL clause is interpreted, how can a judge or arbitrator know that rules different from the default law have been chosen by the parties? Moreover, CISG Article 8 (just like U.S. domestic law) requires determining the intent of both parties, not just the party from the United States. Coyle’s argument necessarily assumes that the non-U.S. party to the contract intends a choice of “the law of New York” to be a choice of the New York UCC, and not the CISG, but he provides no evidence (empirical or otherwise) supporting that conclusion.
It might be helpful to get a little international perspective on this issue by looking at the COL clause from the standpoint of the non-U.S. party to an international sales transaction. In many other countries the domestic law governing sales is national, whereas in the United States it is governed by “sub-national” (state) law – specifically, Article 2 of the UCC except in Louisiana. Internationally, a clause that merely designates the law of a Contracting State without language excluding the CISG is held not to be an effective exclusion of the Convention. [Links for the following citations are in the parentheses at the end of each citation] Consider, for example, the decision of the Bundesgerichtshof, Germany, 11 May 2010; the decision of the Oberster Gerichtshof, Austria, 2 April 2009; the decision of the Court of Arbitration of the International Chamber of Commerce, France, 2002 (Arbitral Award No. 11333); and the many other decisions cited in footnotes 43 and 44 in the discussion of CISG Article 6 in the UNCITRAL Digest of Case Law on the United Nations Convention on Contracts for the International Sale of Goods (2016 edition).
That international consensus certainly makes sense, given that the state designated in the COL clause has ratified the CISG and agreed to apply the Convention when the applicable requirements of Article 1 are met. In short, the law of France for international sales is the CISG (unless the CISG is affirmatively excluded in favor of municipal French sales rule), and a COL clause designating French law does not change that. In this context, the COL clause designating French law is not meaningless: it operates to provide that non-CISG French law will apply to questions beyond the scope of the CISG, such as the “validity” of the contract for sale.
The situation in the United States is somewhat less straightforward, because our domestic sales law is state rather than federal. The COL clause we are considering designates New York law, and of course (unlike France) New York State itself did not directly ratify the Convention. As Coyle accurately declares, however, “under the Supremacy Clause, the law in every state is that the CISG is applicable to contracts where the contracting parties are from different countries that have adopted the CISG.” Thus, absent clear language excluding the CISG, there is no obvious difference between a COL clause designating the law of a particular U.S. state and one designating the law of a country that has ratified the CISG.
A COL provision designating the law of a U.S. state without explicit exclusion of the CISG is overwhelmingly likely to have been proposed by the U.S. party rather than non-U.S. party in an international sales contract. But the empirical evidence Coyle cites in support of the idea that exclusion of the CISG is the effect intended by the majority appears to be based entirely on data from U.S. lawyers. This skews the evidence to favor the intention of U.S. lawyers, and completely ignores the perspective of the non-U.S. party to such clauses.
A non-U.S. party, faced with contract language choosing the law of a particular U.S. state and likely less familiar with the intricacies of the interaction between state and federal law in the United States, may well be misled about the effects of such language. Such a party may be startled to learn that the intended effect of a reference to the law of one particular state was to exclude the operation of a federal treaty, even though the clause never even mentions the CISG.
The clear, even overwhelming, evidence Coyle cites that U.S. lawyers prefer to exclude the CISG, appears to assume that the choice of law is a contest between the CISG and the familiar UCC. But in an international contract, of course, that is not necessarily the case. The choice facing a client represented by a U.S. attorney may be a choice between the CISG and domestic Argentinian, Italian, or Chinese sales law, in which case the enthusiasm of U.S. lawyers for exclusion of the CISG might ebb. How about this for an empirical question: “Your client has negotiated commercial terms with a French party for a very favorable international sales contract. Your client is fully committed to the transaction, but the other party says it will not agree to application of U.S. law and will insist upon application of French law. Would you advise exclusion of the CISG in favor of municipal French sales law, or would you advise agreeing to apply the CISG?”
The Proposed Solution
Coyle proposes that individual U.S. states fix the problem he identifies by adopting an interpretive rule with respect to opting out of the CISG. The Supremacy Clause makes application of the CISG a matter of federal law, binding in every state; a state cannot change the language of a treaty. Any effort to go that route would be disastrous for U.S. treaty relationships and make future negotiation of similar treaties nearly impossible. The Canadian example of provincial statutes mentioned in Coyle’s post simply does not apply. The U.S. and Canadian constitutional frameworks create very different approaches to private international law treaties, with Canada requiring that each province ratify individually. That simply is not possible under the U.S. Constitution. This is why such treaties have very specific general provisions that relate to Canadian ratification (although they do not specifically single out Canada in the language).
Moreover, Coyle’s proposed solution likely violates U.S. treaty obligations. The obligation imposed by CISG Article 7(1) to have regard for uniform interpretation and the international character of the CISG is found in many other treaties affecting private international law. A domestic law rule (state or federal) that specifically rejects the uniform interpretation of CISG COL clauses would not only violate the U.S. obligation under Article 7(1), but would do harm to that obligation in other treaties as well.
But, before such an effort should even be considered, it is necessary to ask whether Coyle’s proposition is itself a good idea; should there be a rule reversing the existing rules of application of the CISG as that law of every state in the United States? The answer is that there should not. Like any law, the CISG creates both advantages and disadvantages for the parties it governs: buyers and sellers in international trade. Those rules can and do sometimes differ from those found in the comparable domestic law on sales. This means that automatic opting out of the CISG promotes lawyer malpractice. There are situations in which the CISG is preferable for a client when compared to the UCC. A simple rule that rubber stamps a lack of understanding of the law, and encourages lawyers to avoid determining which law is better for their client in a given relationship, is a system that promotes the automatic adoption of rules that may be disadvantageous to a lawyer’s own clients. In other situations, that is considered malpractice. Not only does CISG Article 7 obligate the United States and its courts (state and federal) to respect the “international character” of the CISG through the promotion of uniformity in its application, but Model Rule of Professional Conduct 1.1’s obligation of basic competence requires that lawyers know just what they are doing when opting into or opting out of a given law on behalf of a client.
Basic contract drafting requires a consistent and reliable process, a process that places certain obligations on lawyers engaged in that drafting. That process should always (a) begin with an understanding of the default law that would apply absent any contract COL provisions; (b) follow with an understanding of when the default contract law contains mandatory rules that cannot be changed; (c) take into account any other limitations on party autonomy in the choice of law, like public policy; (d) consider and understand the law that applies to alteration of the default law; and (e) rely ultimately on the careful use of language when drafting contract terms to change the default law.
As the post indicates, even after the CISG has been U.S. law for over thirty years, many lawyers, including those who draft international contracts, seem to be oblivious to its existence. But it is difficult to find an example of a comparable situation outside the CISG where lawyers drafting contracts would reasonably believe that a choice of the law of state X should not include rules that clearly are the law of state X. Nor is it easy to find examples of decisions regarding any other rule of positive law in which the rule is ignored because one of the parties simply was unaware of the rule or inadvertently assumed the rule should not be applied.
U.S. lawyers’ abject ignorance of law revealed in Professor Coyle’s empirical work is not something to be encouraged and rewarded by the interpretation he advances. For example, Coyle’s post highlights the following statement of one lawyer: “We would never select the law of Indiana, say, as a means of getting the [CISG]. We are just not that Machiavellian.” The statement is, frankly, an indictment of U.S. legal education in the commercial law area. A choice of Indiana law could never be a “Machiavellian” manipulation to have the CISG apply in the circumstances examined in the post: where both parties are located in Contracting States to the CISG, the CISG is the default law that would apply absent any choice of law clause at all. The truly “Machiavellian” manipulation would be to use a highly ambiguous COL clause choosing Indiana law as a means of excluding the CISG and imposing the UCC on a foreign party unlikely to understand the strange U.S. domestic law theory upon which that reading is based. For the reasons stated here, we strongly oppose that result.