Go to Database Directory || Go to Bibliography || Go to Case Search Form

Reproduced with permission of 17 Journal of Law and Commerce (1998) 187-217

excerpt from

The Several Texts of the CISG in a Decentralized System: Observations on Translations, Reservations and other Challenges to the Uniformity Principle in Article 7(1)

Harry M. Flechtner [*]

(. . .)

In an article written for the 1987 symposium on the CISG sponsored by the Journal of Law and Commerce, I addressed the standards found in Articles 71 and 72 dealing with the effect of a party's prospective inability to perform. Article 71(1) permits a party to suspend temporarily its performance if "it becomes apparent that the other party will not perform a substantial part of his obligations."[10] Article 72(1), on the other hand, allows a party to avoid the contract, thus putting a permanent end to its obligation to perform, if "it is clear" that the other side "will commit a fundamental breach of contract."[11] In my earlier article I noted (as have others) that the Convention appears to require a higher degree of certainty that a future breach we'll occur in order to justify permanent avoidance of contract as compared to temporary suspension of performance. In other words, Article 72 states that the prospect of future breach must be "clear" to justify avoidance, whereas suspension under Article 71 requires only that the threat of breach "becomes apparent."[12] But I went further and made an original (albeit minor) argument that the Convention also required the threat of a more serious breach -- one with more significant consequences -- to trigger avoidance under Article 72 as compared with suspension of performance under Article 71.[13]

My argument -- that avoidance of contract under Article 72 required the prospect of a more serious breach than did mere suspension of performance under Article 71 -- was in large part a textual one. This textual argument focused on the fact that the English version of Article 72 required the threat of a "fundamental breach of contract," whereas the English text of Article 71 required only the possibility that a party would not perform "a substantial part of his obligations." The thrust of the argument was that the drafters would not have used two different phrases ("fundamental breach" as opposed to non-performance of "a substantial part of his obligations"), and in particular, two different adjectives describing the seriousness of the breach ("fundamental" as opposed to "substantial"), had they not intended to distinguish the seriousness of the threatened breach that would satisfy the standards of the respective articles.

Some time after publishing this argument I had occasion to look at the official French version of Articles 71 and 72, and was surprised to discover that my argument was, at the least, much harder to make under the French text. In the French version, both Article 71 and Article 72 use the same adjective to describe the seriousness of a threatened breach that would trigger their provisions. In both, the standard is a breach or non-performance that is "essentielle," i.e., Article 71 states that, to justify suspension, a party must threaten non-performance of "une partie essentielle de ses obligations," and Article 72 requires a threat of "une contravention essentielle au contrat" to warrant avoiding the contract.[14] One need not be an accomplished linguist to recognize that the French text's use of the same adjective to describe the severity of the threatened breach in both Articles 71 and 72 undercuts my argument, particularly when a different adjective equivalent to (indeed, a cognate of) the adjective used in the English text of Article 71 -- "substantielle" -- was available to the drafters of the French text.

I will not here try to sort out whether the French text of Articles 71 and 72 conclusively rebuts my argument.[15] I merely wish to give an example of how different official language versions of the same CISG provisions constitute substantially different texts, a point that seems intuitive enough that it need not be further belabored. It is worth noting, however, that the textual non-uniformity created by the six official language versions of the CISG is only part of the problem in this area. Naturally, translations of the CISG are being made for those whose language does not coincide with any of the six official versions. I have already alluded to the German translation commissioned by the governments of Austria, Germany and Switzerland,[16] and I am also aware of an Italian version.[17] Such translations may be "unofficial,"[18] but that does not mean they are unimportant. As a practical matter, they will undoubtedly constitute the primary source of Convention provisions for courts, arbitral panels and practitioners that work in a language lacking an official version. It is, of course, inevitable that discrepancies and inaccuracies will crop up in such unofficial translations as they would in any translation. The fact that there is no "official" vetting of such translations by UNCITRAL presumably increases the risk of deviations.

(. . .)

Where a particular failure of uniformity is unintended and would violate Article 7(1), the choice from among the competing resolutions can be extraordinarily difficult. This is true even in the relatively easy case where the drafters clearly intended a particular uniform result. For example, suppose non-uniform rules arise from discrepancies in the official language versions of the CISG. As suggested above, this represents unintentional and undesirable non-uniformity: the drafters presumably intended the results in one language version as opposed to another, a single uniform rule. The uniformity principle in Article 7(1) appears to require an attempt to promote the uniform rule intended by the drafters and (presumably) embodied in one of the language versions. The Final Act of the CISG, however, declares that the six official versions of the Convention constitute "a single copy in the Arabic, Chinese, English, French, Russian, and Spanish languages, each text being equally authentic."[66] How does one reconcile different rules expressed in different official language versions?

One commentator has suggested that "the English and French texts of the CISG best represent the intentions of the representatives at the 1980 Diplomatic Conference in Vienna as to the exact wording of the Convention's text," and that in cases of irreconcilable differences among official texts, the English and French versions should form the basis for interpreting the CISG.[67] Such an approach, however, flatly contradicts the Convention language declaring that all six official language texts of the CISG are "equally authentic." The People's Republic of China, for instance, might be surprised to learn that the official Chinese text of the CISG (on which I presume it relied in ratifying the Convention) could be deemed subordinate to the English and French texts in cases of conflict. And what if the English and French diverge? In the end, the idea of elevating one official language text over another in resolving conflicts appears both unwise and unworkable. An alternative methodology, however, is not easy to devise.[68] My point is that, even when the drafters intended a uniform rule in CISG, honoring the uniformity principle in Article 7(1) is often an extraordinarily difficult task. Moreover, a tribunal will sometimes have to choose from among conflicting interpretations of the CISG in situations where the text itself does not express a single unified intention. As was noted above,[69] the drafters themselves could and did fall prey to unconscious background assumptions and imperfect communications that would lead them to believe that there was agreement when in fact they had not formed a common intent for a particular result.[70]

(. . .)

Go to entire text of Flechtner commentary


* Professor, University of Pittsburg School of Law, A.B. 1973, Harvard College; A.M. 1975, Harvard University; J.D. 1981, Harvard University School of Law.

(. . .)

10. CISG, supra note 1, art. 71(1).

11. Id. at 72(1).

12. Compare Harry M. Flechtner, Remedies Under the New International Sales Convention: The Perspective From Article 2 of the U.C.C., 8 J.L. & Com. 53, 94 (1988) (arguing that the CISG requires a higher probability that a threatened future breach will occur in order to justify avoidance of contract under Article 72 as opposed to suspension of performance under Article 71) with John Honnold, Uniform Law for International Sales under the 1980 United Nations Convention 388 at 487 & 396 at 495 (2d ed. 1991) (the same) [hereinafter Honnold Treatise].

13. See Flechtner, supra note 12, at 94.

14. CISG, supra note 1, arts. 71, 72 (French version), reprinted in Documentary History, supra note 3, at 825 (emphasis added).

15. For one approach to reconciling divergent official language versions of the CISG (an approach that would be quite unhelpful in the particular case described in the text), See the text accompanying note 67 infra. The official Spanish text of Articles 71 and 72 echoes the English text in using two different adjectives ("sustancial" in Article 71; "esencial" in Article 72) to describe the seriousness of the breach required to trigger the two provisions. I do not possess the requisite language skills to comment on the Chinese, Russian or Arabic texts.

16. See supra text accompanying note 9. For another description of the German translation, see Diedrich, supra note 8, at 317-18.

17. The German and the Italian translations are reproduced in C.M. Bianca et al., Commentary on the International Sales Law: The 1980 Vienna Sales Convention 807-40 (1987).

18. They are "unofficial" in that they are not sanctioned by the sponsoring body of the Convention, UNCITRAL. That does not mean that they are not sponsored by governments or that they do not have official (or at least quasi official) status within countries in which the language of the "unofficial" translation is spoken. The German translation, for example, was produced by the governments of the German-speaking countries, presumably for the purpose of being used by courts and other public bodies. See Volken, supra note 9, at 39-42. Indeed, the German version has been labeled "official but non-authentic." Diedrich, supra note 8, at 317.

(. . .)

66. Documentary History, supra note 3, at 765 (Final Act of the U.N. Conference on Contracts for the International Sale of Goods).

67. Diedrich, supra note 8, at 317. The reasoning behind this conclusion is that English and French "were the languages in which the deliberations and legal negotiations among the representatives of the Contracting States took place." Id.

68. Paul Volken suggest, rather vaguely, that "sincere efforts toward achieving uniform application of the Vienna Convention may require consulting its texts not only in one but in several official languages," and that achieving uniform application of the Convention requires "taking other linguistic versions of the same provision into account. . . ." Volken, supra note 9, at 41. He does not, however, elaborate on what is meant by "taking into account" different language versions, nor does he specify how to proceed if "consulting" the various official language texts of the CISG reveals a discrepancy.

69. See supra text accompanying note 60.

70. Parol evidence issues, once again, can serve as an example. Professor Honnold, who played a central role in the drafting of the Convention, has written that Article 8(3) "override[s]" the U.S. domestic parol evidence rule, but he asserts that "[j]urists interpreting agreements subject to the Convention can be expected to continue to give special and, in most cases, controlling effect to detailed written agreements." Honnold Treatise, supra note 12, 110, at 171. As was previously noted, however, a German court has asserted that under the Convention an oral agreement can contradict a written one. See OLG Hamm, UNILEX No. 19 U 97/91 (Sept. 22, 1992). Thus, it is not clear that Professor Honnold's view is shared by the civilians who participated in the drafting of the CISG.

(. . .)

Pace Law School Institute of International Commercial Law - Last updated August 16, 1999

Go to Database Directory || Go to Bibliography