On Multi-Lingual Interpretation
(Former Permanent Representative of Israel to the United Nations Office at Geneva.
Member of the United Nations International Law Commission.
The views expressed in this note are personal to the writer.)
Published in Israel Law Review, Vol. 6, 1971; reprinted in The Arab-Israeli Conflict, Vol. II: Readings, ed. John Norton Moore (Princeton University Press, 1974).
Security Council resolution 242 (1967), adopted on November 22, 1967, contains the following phrase:
"Withdrawal of Israeli armed forces from territories occupied in the recent conflict."
In the other languages used by the Security Council (except Chinese), that phrase is framed as follows:
"Retrait des forces armées israéliennes des territoires occupés lors du récent conflit."
"Vyvod izrailskikh voruzhennykh sit s territorii, okkupirovannykh vo vremya n'edavn'ego konflikta."
"Retiro de las fuerzas armadas israelís de los territorios que ocuparon durante el reciente conflicto."
That phrase has produced considerable controversy inside Israel, but within that controversy a secondary issue has arisen, of some juridical interest, since some of the protagonists of one point of view or another have purported to see a fundamental difference between one or other of these language versions of this phrase. We have no intention of taking sides in that particular controversy. The aim of this note is more limited, namely, to indicate certain factors relevant to the interpretation of a multi-lingual resolution of an organ of the United Nations.
The above quoted phrase in resolution 242 (1967) is identical with the draft submitted by the United Kingdom on November 16, 1967, in Security Council document S/8247: similarly the French and Russian versions of the resolution are identical with the translations into those languages, prepared by the United Nations Secretariat, of the draft resolution S/8247. On the other hand, there is a not insignificant difference between the Spanish translation of the draft resolution and the official Spanish version of the resolution itself, the draft reading: "Retiro de todas las fuerzas armadas israelís de territorios que occuparon durante el reciente conflicto".1
Of these four languages, two, the English and the French, had, under the Provisional Rules of Procedure of the Security Council in force in November, 1967, the status of working languages, and the others the status of official languages. The regulatory texts-in this case the Charter of the United Nations and the Provisional Rules of Procedure of the Security Council do nothing to refine the concept of the status of the working languages as compared with the official languages, and the distinction is even more blurred in fact, since on the one hand the procedure of simultaneous interpretation (as distinguished from consecutive interpretation, and translation) is now employed in the meetings of the Security Council, and on the other hand the Secretariat is expeditious in issuing the documents simultaneously in all the languages. By rule 45 of the Provisional Rules of Procedure, verbatim records of the meetings of the Security Council shall be drawn tip in the working languages; but by rule 46: "All resolutions and other important documents shall forthwith be made available in the official languages...".2 To some extent, it may be assumed for practical purposes that the working languages have a standing roughly equivalent to that of the authentic texts of an international treaty, although the analogy must be treated with considerable caution, bearing in mind that in the law of treaties the status of "authentic text" derives from the agreement of the parties, and is not imposed by mere procedure. That standing itself will when necessary relate back to the language in which the negotiation and drafting took place, there being all the difference in the world between a negotiated language version and one produced mechanically by some translation service, however competent. Here it must also be recalled that the function performed by interpretation, when the object of the exercise is a treaty being interpreted by its parties for the purposes of its own application by them, in the nature of things differs from that of interpretation of a resolution of an international organ, especially when the interpretation is being undertaken not by a party for the purposes of application by it, but fly a State which was not even a member of the organ when the resolution in question was adopted.
A close study of the rules for the interpretation of a multi-lingual treaty embodied in articles 31 to 33 of the Vienna Convention on the Law of Treaties of May 23, 19693 suggests that what the International Law Commission wanted to stress, when it put forward the draft of these rules, was, in case of doubt, the importance of determining the history of the multi-lingual texts concerned in order to establish their interrelationship as a matter of fact, as the point of departure for an operation designed to establish the intention of the parties to the treaty in question. Already in 1964 the Commission indicated that it would not be content to rely oil purely doctrinal studies, and requested the Secretariat to furnish further information regarding the practice of the United Nations in drawing up the texts of multi-lingual instruments.4 The practical considerations which prompted that attitude in relation to the interpretation of a multi-lingual treat) obviously apply with even greater force when what is being interpreted is a resolution which itself exists in a number of language versions, the precise status of which differ.
As stated, the draft of resolution 242 (1967) was submitted by the United Kingdom and naturally the original text is English. It is an historical fact, which nobody has ever attempted to deny, that the negotiations between the members of the Security Council, and with the other interested parties, which preceded the adoption of that resolution, were conducted on the basis of English texts, ultimately consolidated in Security Council document S/8247. Investigations which have since been made establish that at some stage the question was raised whether the translations prepared by the Secretariat (and especially the French translation) were adequate and accurate renderings of that original. This question was answered in the affirmative. The translations were prepared in the usual way by the appropriate language services of the Secretariat. It appears that these translations were checked by the substantive Secretariat officials in the Department of Political and Security Council Affairs of the Secretariat assigned to the Security Council, and later by the members of the Security Council themselves in informal meetings, those checks supplying the necessary political controls over the technical work. There is also some evidence to indicate that contemporary independent checks were made in different Foreign Ministries.5 The upshot was at the time a general understanding that, in the same way that the English and Russian languages can. get by without use of the definite (or indefinite) article, the genius of the French and Spanish languages requires use of definite articles to a degree which a non-Latinist may find excessive and misleading, or at least confusing. A suggestion that the translation may have been faulty (although in what respect is not clear) has only been advanced recently: but even so the question is, and will remain, an open one. Many experts in the French language, including academics with no political axe to grind, have advised that the French translation is an accurate and idiomatic rendering of the original English text, and possibly even the only acceptable rendering into French. As an independent scholar of the law has recently written: "the expression 'des territoires' in [the French] translation may be viewed merely as an idiomatic rendering into French, not intended to depart ... from the English...".6
This contention is further supported by the proceedings in the Security Council itself. In its composition on the date in question, English was the language used by ten members (Brazil, Canada, China, Denmark, Ethiopia, India, Japan, Nigeria, United Kingdom and the United States); French was used by three members, (Bulgaria, France and Mali); and Russian and Spanish by one each, the USSR and Argentina (it will be noted that all the French-speaking delegations were then favourable to the Arab thesis). Furthermore, of the non-members invited to take part in those meetings - UAR, Israel, Jordan and Syria - all were habitually using the English language both in the meetings of the Security Council and in the private negotiations.
The draft resolution was introduced by the United Kingdom at the 1379th meeting of the Security Council on November 16, 1967. In subsequent meetings there was sporadic mention - without particular stress on linguistic problems of the meaning to be given to the phrase under consideration here. On the question of concordance, the French representative was explicit in stating that the French text was "identical" with the English text.7 The Israel representative intervened at the end of the debate to state that he was communicating to his Government nothing else except the original English text of the draft resolution as presented by the original sponsor on November 16.8 It is known from an outside source that the sponsors resisted all attempts to insert words such as "all" or "the" in the text of this phrase in the English text of the resolution,9 and it will not be overlooked that when that very word "all" erroneously crept into the Spanish translation of the draft, it was subsequently removed.
It is submitted that when resolution 242 is closely examined in the light of the practice and procedure of the Security Council and the Secretariat and of its own legislative history, the supposition of any real or assumed lack of concordance between some of the language versions of the official text of that resolution is not helpful in resolving any interpretation problems which that resolution may present. One reason for this is that in order to remove such lack of concordance, one or other of the language versions would have to be rewritten. But that process would in the nature of things introduce arbitrariness and subjectivities into the matter. For instance, it is said that the indefinite quality of the English and Russian versions - which was a matter of political determinism-ought to be met by the introduction of a word such as "certains" into the French version (and its equivalent in the Spanish). But in such a context, certains would need some equivalent in English, for instance some, a word which does not appear in the English text and which, moreover, it is unlikely that a draftsman with any command of the English language, from either side of the Atlantic, would have willingly or wittingly inserted. If on this score there is any ambiguity in resolution 242 as it stands (which we do not think to be the case), it is rendered neither greater nor less by comparison of the different language versions, but is inherent in the text as adopted, in all its language versions. In this connection it may be observed that categorical assertions that the resolution obliges Israel completely to withdraw all its armed forces from all the occupied territories are not based on preference for one or other of the language versions of the resolution, but on the resolution in its integrity, in each one of its language versions. That was made clear, by the pro-Arab spokesmen, using the English, French and Russian languages, in the Security Council debate in November 1967. However, the real problem of what the resolution means on this cardinal question, or to put it differently, what the Security Council intended, arises whatever the language in which the resolution be read or a given contention expressed.
Curiously enough, there is remarkably little international jurisprudence on the interpretation of multi-lingual resolutions of international organs. In the South-West Africa-Voting Procedure case the International Court of Justice noted that there was a slight difference between the English and French texts of a resolution of the General Assembly, and reached the conclusion, from an examination of the debates in the General Assembly, that the French version seemed to express more precisely the intention of the General Assembly.10 Applying that test to the Security Council resolution under discussion, it would follow that in order to reach viable conclusions as to the intention of the Security Council (assuming that to be the aim of the process of interpretation), all the antecedent discussions in the Security Council (from May, 1967) and in the Fifth Emergency Special Session of the General Assembly, must be taken into account.11 That includes the rejection of all resolutions calling for the complete withdrawal of all Israeli armed forces from all the occupied territories. It also includes the predominant position in fact occupied by the English language in all the formal and informal meetings and negotiations which preceded the adoption of the Security Council's resolution under consideration here.
It is commonplace to say that interpretation is an art, not an exact science. Likewise, translation is an art, not an exact science. The most the law call do in such circumstances is to indicate in general terms, as we have done here, the nature of the rules governing the process by which this art is applied in a concrete case, the kind of intellectual discipline with which the interpreter must gird himself. In this connection, the wise counsel furnished by the greatest of Jewish jurists to the translator may appropriately be recalled:
Any translator who intends to render a work from one language to another merely by rendering word for word, and slavishly following the order of the chapters and sentences in the original, will come to grief. The product of his labor will be unintelligible and ludicrous. That method is utterly incorrect. The first step is to read the original until the translator is fully at home in it and has complete understanding of what the author has written. Then he should render the contents clearly and idiomatically. However, this can be done only if he grasps syntax by the neck and vigorously shakes it, changing the order of paragraphs and words, substituting many words for one when necessary and vice versa, and altering punctuation until the translation reads clearly, gracefully and meaningfully.12
The critic of the different language versions of the Security Council resolution would do well to keep that in his mind.
1. The change in the Spanish text was probably the result of renewed scrutiny following the intervention of the only Spanish-speaking member of the Security Council at that time, the representative of Argentina. S/PV. 1382 at 78 (Provisional English version).
2. Document S/96/Rev. 4. Following Security Council resolution 263, adopted on January 24, 1969, Russian and Spanish now have the status of working languages consequential amendments being made in the Rules of Procedure in force since then, doc. S/96/Rev. 5.
3. The Law of Treaties: A Guide to the Legislative History of the Vienna Convention, by the present writer (1970). For the draft articles on the law of treaties submitted by the International Law Commission, see Reports of the International Law Commission on the work of the second part of its seventeenth session and on it eighteenth session, in U.N. Official Records of the General Assembly, twenty-first session, supplement No. 9 (A/6309/Rev. 1), Part II, Chapter II. Yearbook of the International Law Commission (1966), vol. II at 173.
4. Report of the International Law Commission on the work of its sixteenth session, footnote 170. Official Records of the General Assembly, nineteenth session, supplement No. 9 (A/5809), Yearbook of the International Law Commission (1964), vol. II at 206. See also the discussion at the 767th meeting of the Commission, Yearbook of the International Law Commission (1964), vol. I at 298. For the memorandum of the Secretariat (in the preparation of which the present writer assisted), see document A/CN.4/187, Yearbook of the International Law Commission (1966), vol. II at 104.
5. Eugene Rostow, "The United Nations and Legal Aspects of the Search for Peace in the Middle East" (1970), Proceedings of the American Society of International Law, 69. In 1967, Dean Rostow occupied a responsible position in the State Department during the administration of President Johnson, and his remarks were made with personal knowledge and authority.
6. Julius Stone, "The 'November Resolution' and Middle East Peace: Pitfall of Guidance" in A Collection of Essays in honor of Josef L. Kunz (reprinted from (1970) Toledo Law Review), footnote 7.
7. S/PV.1382 at 58 (Provisional English version). Note also the remark of the representative of France at the 1895th plenary meeting of the General Assembly on November 3, 1970, insisting "in order to avoid reviving an old quarrel" that this part of resolution 242 (1967) must be quoted in exactly the same terms as those that were adopted, the English text in the original English, the French text in the original French version, the Russian text in the original Russian version, and so on. A/PV.1895 at 53 (Provisional English version). This is probably the most authoritative confirmation one could have that the French text was intended to convey exactly the same meaning as the English, and not vice versa.
8. S/PV.1382 at 96 (Provisional English version). Note also on this point Mr. Eban's Press Conference of March 7, 1971.
9. Arthur Lall, The U.N. and the Middle East Crisis (1968) at 253-4. Ambassador Lall had earlier been Deputy Permanent Representative of India to the United Nations, and although in 1967 he held a teaching post at Columbia University, in the City of New York, he is widely regarded as reflecting the views of the Indian delegation, which at that time was a member of the Security Council.
10. I.C.J. Reports, 1955, 67 at 72.
11. See, for instance, document S/8235, circulating as a Security Council document the text of a draft resolution (which was not adopted) submitted by a group of delegations at the emergency meetings of the General Assembly. That circulation was undertaken at the request of India, made at the 1373rd meeting of the Council on November 9, 1967.
12. Maimonides to Shmuel ibn Tibbon in 1199. Translation from Leo W. Schwarz, Memoirs of My People (1943) at x.