AmbienteDiritto.it - Rivista giuridica - Electronic Law Review - Copyright © AmbienteDiritto.it
Testata registrata presso il Tribunale di Patti Reg. n. 197 del 19/07/2006
The Interpretation Problems of Multilingual Treaties
EDOARDO BINDA ZANE*
Interpretation in international law1
“Interpretation” is, at the same time, a definite and vague term. Simplistically, it could be thought as “a way of explaining what is unclear”, a definition applicable in every possible field or topic. However, when it comes to identify in detail what this term refers to in the specific field of international law, things are not as easy.
This problem has been dealt with by an incredible number of experts and the different definitions that have been given are countless, given the vastness of the matter and all the slight differences and points of view. Considering these facts, I will not attempt to introduce my definition, nor to refer some other authors’ one, in fact, by doing so, I would implicitly deny the validity of all the ones that are not taken into consideration. Instead, I will try to give a quick, general classification of the different sides of the question, so to clarify on what levels the debate has been taking place.
Namely, there are three different approaches or classifications relevant for this paper:
1. Textual, subjective, teleological interpretation;
2. Restrictive, extensive, analogical interpretation;
3. Autonomous, uniform, international interpretation.
Textual, subjective, teleological interpretation
These three different types of interpretation are centred on a distinction of the basis chosen to rely on, they are therefore to be considered as three different methods, or ways, of facing the interpreting process. This distinction might seem somewhat weak and irrelevant, but it is of great importance in a realistic international environment, also considering its link with the inner structure of the 1969 Vienna Convention on the law of treaties.
Textual interpretation is also called objective. Its rationale is to focus on the text of the treaty, considering it the only source for interpretation as it is thought to be the authentic expression of the intention of the parties; the declared and effective will of the parties are therefore considered identical and embodied in the text.
Subjective interpretation is not as strict, it does rely on the text, but it tends to consider the intention of the parties as an element separate from the text. The reference is hence to the effective will of the parties, which does not appear from the text as it has been written, whereas from the logical construction of the treaty itself as well as from the object and purpose of the treaty and the contingent situation in which the treaty has been drafted.
Last but not least, teleological interpretation takes another step forward: it is not based on the intention of the parties, nor on the text of the treaty, but on its declared or intended purpose. In this third meaning, the most relevant subject is not the parties anymore: the text of the treaty itself is awarded a much higher position. This consideration is particularly worth of attention in situations in which there are problems of authenticity of a text, or when it comes to distinguish between a text and a version of a treaty.
None of these three methods is necessarily right or wrong, nor obviously is its application in a particular context. The general trend, however, has recently been of having the subjective interpretation slowly fade in favour of the textual one, which is often jointly considered with the teleological. This “unwritten rule” also appears in the Vienna Convention of 1969, that embodies all the interpreting methods outlined above, implicitly giving a sort of hierarchical order on the one to use.
Restrictive, extensive, analogical interpretation
This second classification is somewhat different from the previous one, in fact these three other meanings refer to the outcome of the interpretative process, and not to an interpreting method. On a logical level this is proved by the fact that, by admitting one, the possibility of considering the other two is inevitably rejected.
Restrictive interpretation has been for a long time predominant in international law, although now the focus is shifting to the extensive and analogical one. Its concept is apposing a sort of limitation to the “broadness” originally given to the sense of the treaty, just as if its text were saying more than what the legislators really intended it to do.
On the other hand, extensive interpretation uses precisely the opposite concept, which is to give a meaning to a treaty that is larger than what has been written. In a nutshell, the treaty is considered to mean less than what the legislators were trying to say, and therefore the interpretation is based not on what is written, whereas on what the real intention of the legislators is thought to be.
When does each one of the two apply then? There is no established rule for applying one or the other, all is left to the work of the interpreter in relation to the situation and to the linkages of the considered treaty or law. In the past, restrictive interpretation of international treaties was thought to be essential for guaranteeing State liberty and sovereignty, however, the current trend is to favour the extensive one. This shift finds its origin in the developing collaboration that has recently been taking place between Governments, or, in other words, on the proof that States can deal, participate in other ones’ foreign policy and accept recommendations without giving up their sovereignty and liberty.
Thirdly, analogical interpretation belongs to the same dimension of the outlined two, but differs for the method it follows and is generally used when a case needs solved and other outcomes are proved useless. What is sought for here, is not understanding the real intention of the legislator, whereas comparing international treaties or laws ruling similar matters and move from that basis to look for a solution in order to clear the practical situation.
Autonomous, uniform, international interpretation
This last classification touches yet another dimension of meaning, in fact it relates especially to a linguistic and terminological field. Translation is alone a very delicate topic, but it becomes even more delicate when the object of translation is an official document, a treaty for example. Terms are almost never perfectly translated in another language, because the correspondent one has never the exact meaning, due to linguistic or cultural differences. The implications of choosing a term for a translation instead of another one are therefore virtually infinite and so are the consequences when considering international law. Hence, the different applicable methods are not discussed here, this division only tries to clarify what the generally followed trends are.
These three are, as well as the previous ones, different results of a process and not interpreting methods. Moreover, they are not mutually exclusive, in fact “autonomous interpretation is always international and international interpretation is always uniform, but uniform interpretation is not necessarily international or autonomous”2.
Uniform interpretation is a larger “container” that includes the autonomous and the international ones; it is an obligation taken by the States that participate in a treaty for which they all agree that a particular term should have one single meaning in the considered context. The will of the States is then the basis for its existence: when a common will disappears, the basis for a uniform interpretation falls as well.
Autonomous interpretations takes place in the situations in which an expression or a term of a treaty is given a meaning that is coherent with the text and the conditions considered, but does not necessarily respond to the meaning the same term has in the parties’ internal legislations. There is a slight but important difference between these two cases: in the former, everything relies on the effective will of the States to carry on the situation, whereas the latter only identifies an existent situation in which the States and their will do not play any direct role.
International interpretation works in a similar way to the autonomous: there is a recognized difference between the meaning of a term in a national environment and in an international one. In this second case, there might be conditions under which some expressions, not linked to any treaty, convention or law, are awarded a specific meaning; this unique connotation corresponds to its international interpretation.
Multilingualism in international treaties
The previous chapter was focused on three views of what is meant by the term “interpretation”. Specifically there are three relevant classifications I identified: the first one dealing with the choice of an interpreting method, the second one considering the result of the interpreting process, the third one relative to the different meanings that foreign words can assume in an international environment.
In this chapter I will draw the attention on the importance of multilingualism in international law. First I will give a brief description of the International Law Commission, a very relevant actor in international law, secondly I will focus on the drafting and on some parts of the Vienna Convention on the Law of Treaties of 1969. Specifically, I will consider Article 33 on interpretation of multilingual treaties, Articles 31 and 32 on rules of interpretation, Article 79 on the correction of errors.
The International Law Commission and the drafting of the 1969 Vienna Convention on the Law of Treaties
The International Law Commission (ILC henceforth) is, in the frame of international law, an unprecedented institution for its importance and tasks. It was created after the end of World War II through the United Nations by virtue of General Assembly Resolution 174 (II) of November 21st, 1947, which provides that the “General Assembly resolves to establish an “International Law Commission”…which shall exercise its functions in accordance with the provisions of the annexed Statute”3. This resolution – and therefore the creation of the Commission itself – was inspired directly by the UN Charter, which provides inter alia that “the General Assembly shall initiate studies and make recommendations for the purpose of… encouraging the progressive development of international law and its codification”4; it is worth noting, with referral to the text of G.A. Resolution 174 (II), that the ILC Statute’s wording in defining the Commission’s duties, is a perfect recalling of this article5.
At its first session, in 1949, the Commission reviewed, on the basis of a Secretariat memorandum entitled "Survey of international law in relation to the work of codification of the International Law Commission", twenty-five topics for possible inclusion in a list of topics for study. Following its consideration of the matter, the Commission drew up a provisional list of fourteen topics selected for codification, as follows:
√ Recognition of States and Governments
√ Succession of States and Governments
√ Jurisdictional immunities of States and their property
√ Jurisdiction with regard to crimes committed outside national territory
√ Regime of the high seas
√ Regime of territorial waters
√ Nationality, including statelessness
√ Treatment of aliens
√ Right of Asylum
√ Law of treaties
√ Diplomatic intercourse and immunities
√ Consular intercourse and immunities
√ State responsibility
√ Arbitral procedure
Although it was understood that the above list of fourteen topics was only provisional and that additions or deletions might be made after further study by the Commission or in compliance with the wishes of the General Assembly, the list has continued to constitute the Commission's basic long-term programme of work.
Since 1949 the Commission has submitted final drafts or reports with respect to eleven of the above-mentioned topics or sub-topics6.
The Law of Treaties, particularly, has been dealt in several occasions and in referral to different aspects of the topic, as it was one of the three topics the Commission gave priority to7: in 1951 the ILC studied the Question of Reservations in Multilateral Conventions, in 1963 it reported on the Question of Extended Participation in General Multilateral Treaties Concluded under the Auspices of the League of Nations, in 1978 the Commission summarized years of work in the Vienna Convention on the Succession of States in respect of Treaties8. However, it is with the 1969 Vienna Convention on the Law of Treaties that the Commission reached its greatest success in this area.
The ILC decided in 1962 to appoint a Drafting Committee in charge of preparing the draft articles on the law of treaties. The aim of the ILC was originally to produce an expository code of a general character, primarily descriptive or exhortative, however, in 1961, it decided that the work should have been focusing in another direction, as reported by the Commission itself:
First, an expository code, however well formulated, cannot in the nature of things be so effective as a convention for consolidating the law; and the consolidation of the law of treaties is of particular importance at the present time when so many new States have recently become members of the international community. Secondly, the codification of the law of treaties through a multilateral convention would give all the new States the opportunity to participate directly in the formulation of the law if they so wished; and their participation in the work of codification appears to the Commission to e extremely desirable in order that the law of treaties may be placed upon the widest and most secure foundations.9
The work was originally carried out in English and French as working languages, however, in 1964, Spanish was added as a third one. The basic texts for the Drafting Committee were produced in English by Sir Humphrey Waldock, Special Rapporteur, on the basis of the debates, but the drafting process itself was then trilingual; subsequently the Secretariat prepared the Russian and Chinese translations of the ILC draft articles. The five different language versions were all given equal status, given the fact that the language process had been thoroughly carried out as well as the work on the concordance between the different versions after the drafting process had ended in 1966.
The Vienna Convention on the Law of Treaties was signed at Vienna, on 23 May 1969, in a single copy in the Chinese, English, French, Russian and Spanish languages, each text being equally authentic10.
The 1969 Vienna Convention on the Law of Treaties and the interpretation process
The 1969 Vienna Convention on the Law of Treaties deals each and every aspect of the topic, however, for the aim of this paper, only a few articles are relevant, specifically: 31, 32, 33 (regarding interpretation) and 79 (on the correction of errors).
Questions arisen in the drafting of Article 33
Article 33 is in particular dedicated to the interpretation of treaties in different languages:
Interpretation of treaties authenticated in two or more languages
1. When a treaty has been authenticated in two or more languages, the text is equally authoritative in each language, unless the treaty provides or the parties agree that, in case of divergence, a particular text shall prevail.
2. A version of the treaty in a language other than one of those in which the text was authenticated shall be considered an authentic text only if the treaty so provides or the parties so agree.
3. The terms of the treaty are presumed to have the same meaning in each authentic text.
4. Except where a particular text prevails in accordance with paragraph 1, when a comparison of the authentic texts discloses a difference of meaning which the application of articles 31 and 32 does not remove, the meaning which best reconciles the texts, having regard to the object and purpose of the treaty, shall be adopted.
There are two important terminological questions which arose during the drafting session of the ILC regarding this article. The first was the problematic distinction between a text and a version of a multilingual treaty, on which opinion was divided in two positions. On one side there was the conviction that there is only one text of a treaty, of which there might be versions in several languages, the main point used to defend this thesis was the fact that considering a treaty as existing in several different texts, would have derogated to the principle of unity of the treaty as a single document. However, in the opposite terms, it had been observed that international practice spoke quite often of authentic texts rather than authentic versions of a treaty11, moreover it was doubtful whether the principle of unity of a treaty was derogated more by the term “text” than by the term “version”. The decision was to have the first interpretation prevail, in order to give a distinction between the language versions which have been explicitly given the status of authentic text, from the ones which have not. This result reflected in the wording of Article 33 of the Vienna Convention of 1969, in which “text” should be referred to as any rendition in a language in which the treaty was authenticated, whereas “version” refers also to languages other than those in which the text was authenticated12.
The second question to be considered refers to treaties drawn up, authentic, authenticated and authoritative in different languages. This group of terms has lead to some terminological problems, as the distinction between them might seem somewhat unclear. Leaving out the details of the discussion, the result has been that, in a general way, an authentic text, which is authoritative, is one which may be relied upon for purposes of interpretation. Moreover, “sometimes a text of a treaty is authentic between some parties, and a different text between other parties. The practice was mentioned of bilateral treaties ‘drawn up’ in the national languages of both parties, which were regarded as ‘authentic’, and to which a translation in a third, widely known diplomatic language considered to be “authoritative” was annexed. Other bilateral treaties ‘drawn up’ in two languages were ‘authentic’ only in one of them”13.
In the frame work of the Vienna Convention, it is important to make a distinction between two articles, both dealing with authenticity, which are slightly different: Article 10 considers the cases in which the status of authenticity is given to a text of a treaty, article 33, on the other hand, deals with the status of authenticity of different language versions of a treaty. Here again, the distinction between “text” and “version” plays an important role.
Articles 31, 32, 33 and the process of multilingual interpretation14
General Rule of Interpretation
1) A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.
2) The context for the purpose of the interpretation of a treaty shall comprise, in addition to the text, including its preamble and annexes:
(a) any agreement relating to the treaty which was made between all the parties in connection with the conclusion of the treaty;
(b) any instrument which was made by one or more parties in connection with the conclusion of the treaty and accepted by the other parties as an instrument related to the treaty.
3) There shall be taken into account, together with the context:
(a) any subsequent agreement between the parties regarding the interpretation of the treaty or the application of its provisions;
(b) any subsequent practice in the application of the treaty which establishes the agreement of the parties regarding its interpretation;
(c) any relevant rules of international law applicable in the relations between the parties.
4) A special meaning shall be given to a term if it is established that the parties so intended.
Supplementary Means of Interpretation
Recourse may be had to supplementary means of interpretation, including the preparatory work of the treaty and the circumstances of its conclusion, in order to confirm the meaning resulting from the application of article 31, or to determine the meaning when the interpretation according to article 31:
(a) leaves the meaning ambiguous or obscure; or
(b) leads to a result which is manifestly absurd or unreasonable.
Article 31 is the one to refer to when dealing with “plain” interpretation of treaties, however it is article 33 the most important for interpreting multilingual treaties and it should therefore be the first to be relied on in such cases.
The starting point of the process is paragraph 3 of article 33: “The terms of the treaty are presumed to have the same meaning in each authentic text”. This presumption, however, is not always possible, quite often, in fact, terms that are considered to be equivalent in two languages have in fact a somehow different meaning in their respective languages. This might lead to a problem of lack of clarity, for which it is necessary to refer to paragraph 4 of the same article, where it is suggested a comparison of the authentic texts of the considered treaty, in an effort to find their common meaning.
In the case this method proves unsuccessful as well, that is when there is in fact a difference in meaning between the texts, the general rule of interpretation provided by article 31 is applied. If even this would not be sufficient, article 32 provides some supplementary means of interpretation that could be used15. According to paragraph 4 of article 33, if all these efforts do not solve the arisen problem, “the meaning which best reconciles the texts, having regard to the object and purpose of the treaty, shall be adopted”.
As a last possibility, as pointed out by Sir Humphrey Waldock, “if no reconciliation of the texts was possible, the interpretation should be left to be determined in the light of all the circumstances”16.
The correction of errors in multilingual treaties: article 79 of the Vienna Convention
Articles 31 to 33 of the Vienna Convention of 1969 deal with interpretation, article 79 deals specifically with the correction of mistakes in international treaties. Apparently, these articles belong to two different fields, however, in certain cases it is problematic to determine whether a situation should be treated under an article rather than under the others.
The key to the solution of this question is interpretation itself: article 79 (considering only paragraph 3, which specifically relates to multilingual treaties) provides that the article itself and its provisions apply to multilingual treaties when “it appears that there is a lack of concordance which the signatory States and the contracting States agree should be corrected”. Two are hence the conditions under which article 79 should be applied: there is a lack of concordance between two language versions of a treaty and all parties do not give rise to disputes or problems.
If the problem is not left undisputed (and therefore requires interpretation), the situation ceases to be relevant for article 79 and falls under the rules of interpretation provided by articles 31 to 33, or, in another case, if one or more parties consider the error to be essential, article 48 applies:
A State may invoke an error in a treaty as invalidating its consent to be bound by the treaty if the error relates to a fact or situation which was assumed by that State to exist at the time when the treaty was concluded and formed an essential basis of its consent to be bound by the treaty17 .
Problems of multilingual interpretation18
The previous chapter was centred on the phenomenon of multilingualism, which has become usual in the practice of international law. The International Law Commission is the highest authority in international law, with the duty of promoting the progressive development of international law and its codification. Taking into account its object, the ILC has drafted several conventions, among which is the Vienna Convention on the Law of Treaties of 1969; this convention provides the fundamental procedures of interpretation of international treaties in its articles 31, 32, 33.
However, interpretation, and specifically the multilingual one, is not a smooth process, in fact several obstacles can occur. First of all, translation is never precise, as terms in different languages assume slightly different meanings that can originate infinite political implications. Also, translation is a long-lasting process and is rarely coordinated with the political process of negotiations and acceptance, which may cause discrepancies. These considerations lead to the problem of defining what text or version should be considered authentic when these two factors (time and terminology) occur.
Partially untied from the other ones, but still in the same framework, some disputes have arisen from the structure of the Vienna Convention: One is linked to the concept of unity of the treaty and to how it applies during the interpretation process, before or after a divergence between two texts is discovered. The second one analyses the rules provided by the Convention with regard to a possible hierarchy on the rules to rely on.
Terminological problems and solutions19
It is generally difficult to find an exact correspondence between a term in a language and an equivalent term in another language, let alone in the framework of international law, where sometimes terms in one legal system have even no equivalent in another one. It is interesting that this problem occurs not only within systems whose official languages are different, but also within countries that share the same language but attribute slightly different meanings to the same words.
There are four main kinds of terms that can be identified with respect to this topic:
Terms that sound alike in two languages but have different meanings
It is crucial that the translator put as much attention into properly choosing terms which are to be used in his job, but, since even translators are human beings, sometimes it might occur that a term that sounds like the one to translate will be used for that purpose.
However, “words which sound almost alike in two different languages but have different meaning are a common source of confusion and misunderstanding. For example, while the French verb demander means ‘to ask, request’, the English like-sounding demand is much more forceful; the French verb contrôler is often mistakenly rendered as ‘to control’”20.
Terms that have more than one literal translation
This is another situation that might occur, “for example, should the Russian expression obschii poriadok be translated into German as allgemeines Recht (general law) or gemeinsames Recht (common law)? Is the term nierushimost’ (inviolability) more accurately rendered as Unantastbarkeit, or the weaker Unverletzichkeit?”21
Terms that have no translation in another legal system
It is not uncommon to face the impossibility of translation, because the other legal system does not even conceive a similar concept (for example when anglosaxon and roman systems are involved). For example, “art.1.1 letter d of the Geneva Convention of 1948 on the International Recognition of Rights in Aircraft indicates “l’hypothèque, le ‘mortgage’ et tout droit similaire” in the French text, “the ‘hypothèque’, the mortgage, and all similar rights” in the English text and “la hipoteca, ‘mortgage’ y derechos similares” in the Spanish text” .22
Translated words whose literal equivalent discloses a different meaning
It might occur that some words, when literally translated, result in an existent term in another language which nevertheless has a different meaning. “For example, alsbald in German may mean ‘as soon as possible’, while in French aussitôt, the equivalent rendition of the two parts of the word, when used alone always has the stronger meaning of ‘fortwith’”23.
These obstacles have been partially alleviated by new instruments such as the compilation of multilingual glossaries (for example the UN terminology bulletins25): something comparable to a normal dictionary, but with a special focus on the problems outlined above, on the framework of the object of translation or on its topic. However, even with the help of this instrument, the problem could not be completely solved because glossaries, being very specific and thorough in explanation, give access to a very high number of meanings and implications. This in the end might lead to a solution, but would result in devoting too much time to the process (see infra). Nevertheless, their usefulness has been proven by practice.
Several other attempts have been done in order to ease off translations, such as implementing the results achieved by the International Institute for the Unification of Private Law (UNIDROIT), an independent intergovernmental organisation, whose purpose is to study needs and methods for modernising, harmonising and co-ordinating private and in particular commercial law as between States and groups of States26. Although its field of interest is private or commercial law, its achievements are fully adaptable to the needs of international law.
A last resource for translation needs mentioned: technology. Communication technologies are particularly important for instant, long-distance translations for the impact they have on the time factor (see infra), which is in this case almost non-existent. However, for international treaties things change. Although time plays an important role, accuracy cannot be avoided. Referral to technology is hence to translating machines, or, in a more general way, computers. Theoretically the idea of inserting a text written in English and witnessing an almost-instantaneous translation is correct. However,
the feasibility of a fully automatic and accurate translating machine… requiring… literary components and judgement value, seems doubtful. These areas will continue to require human attention to avoid results such as the translation of the English phrase “out of sight, out of mind” into the Russian as “invisible or insane”, or the expression “the spirit is willing, but the flesh is weak” as “the whiskey is all right, but the meat has gone bad”.
(…) [However], although still in a research stage, the use of electronic means for linguistic ends is progressing in the framework of the European Communities27.
Terminology as a tool of diplomacy
At times, however, terminological problems can be helpful in order to solve problems or ease off tensions in delicate situation.
A historic example is the role that UN Security Council’s Resolution 242 (1967) played in the 1967 Israel-Palestine crisis, specifically, translation itself has been capable of easing off all diplomatic tensions. The connection is on the clause regarding the withdrawal of troops from occupied territories: “the notable omissions in regard to withdrawal are the words ‘the’ or ‘all’ and ‘the June 5, 1967 lines’... the resolution speaks of withdrawal from occupied territories, without defining the extent of withdrawal"28. On top, the word “the” is absent in the Russian language and not always necessary in English29.
The time factor
This problem is not linked with doctrine or theory of international law: it is the result of the practice that has been developing in the drafting and signing of international multilingual treaties in history.
The difference between a text and a version of a treaty has been analysed earlier in chapter 2; recalling that difference, what the parties sign after having agreed on a topic, is a text of a treaty that may consist of several language versions. In the current practice, it happens that some parties avoid or neglect to verify the correspondence between the translated texts before signing; this is the moment in which the time factor gains importance. The focus is here brought on the period elapsing from the moment when the draft of the text is finished in the working language or languages to the moment when the treaty is translated in the other languages that will be considered official. Taking it to the extreme, there are even cases when the parties sign a text whose language versions have not yet been completed.
“Generally in current practice, the very concept of multilingual authenticity often contains a fictitious legal element, in the sense that sometimes the parties are in fact authenticating a nonexistent text. Likewise, a somewhat similar problem, although less disturbing from a visual and temporal viewpoint, exists when an authentic version which has not been negotiated by the parties is produced by the technical service of a secretariat and is not actually reviewed by the signatory”.
In reality, time is only a variable in a wider problem, which deals with authenticity. The question is on whether the translated text of a treaty should receive the status of being authentic or it should be considered in a minor position with respect to the originally drafted version.
The following paragraph is devoted to the analysis of this topic.
The status of different multilingual authentic texts in the interpretation process
“The texts of an authentic multilingual document, which are of legal validity to the parties for purposes of interpretation, may be subdivided according to:
a. Working o official language versions,
b. Drafted or translated versions, and
c. Versions originally authenticated at the time of signature, or those authenticated at a later date”30.
The problems arising from this subdivision is whether all these authentic versions should be considered at an equal level or it exists a sort of hierarchy between them, as the two factors of time and terminology concur to determine the three groups here.
There was, originally, a distinction between working and official languages, as in the case of the UN Charter, for which English and French, being the languages used for drawing up and drafting the text of the Charter, are considered to carry more weight with respect to Chinese, Russian and Spanish. However, in the current practice, the number of working languages has dramatically increased, reducing and somehow even blurring the distinction between them and official languages. It is the reason why they can be considered in a single group now, as the legal weight of a working language has very much decreased due to this process.
With regard to drafted or translated versions, the 1969 Vienna Convention might be taken as an example. Its Article 85, in the final provisions states that the “Chinese, English, French, Russian and Spanish texts are equally authentic”31, however, reviewing the history of the drafting process, English and French were the two original working languages, Spanish was subsequently added as a third and the Chinese and Russian versions of the final draft articles were prepared by the ILC Secretariat. Given this situation, it appears that there is a contrast between the provision of Article 85 and the facts relating to the drafting process. Is it really correct to consider equal a version drafted in an official language, and hence reflecting the intentions of the parties, and a version translated by language experts, technically competent but not involved in actual negotiations?
As for the third group, if the status of authenticity has been granted to several language versions, they are all considered to be equal, regardless of whether some have been produced after an interval of years or even if they have been accepted before having been drafted (see above, “the time factor”). However, at least in case of divergence or contrast, the versions belonging to this third group are granted a lesser interpretative value.
The unity of multilingual documents32
The principle of unity of a multilingual document can be understood by considering the fact that all language versions are thought to represent a single document (or text) and that, being such, they all embody a single meaning. In fact, “in law there is only one treaty – one set of terms accepted by the parties and one common intention with respect to those terms – even when two authentic texts appear to diverge”33.
Plurilingual in expression, the treaty remains a single treaty with a single set of terms the interpretation of which is governed be the rules set out in articles 27 and 28 [31 and 32 of the Vienna Convention]. The unity of the treaty and of each of its terms is of fundamental importance in the interpretation of plurilingual treaties and it is safeguarded by combining with the principle of the equal authority of authentic texts the presumption that the terms are intended to have the same meaning in each text.34
What does this imply? When interpretation is needed, should the right to rely on only one version (according to the principle of unity) be invoked, or should there better be an obligation to consult several versions?
There are two different cases in which this question is ought to be considered: before a divergence emerges and after a divergence has emerged.
Before the discovery of a divergence
In this case, the starting point should be Article 33(3) if the Vienna Convention:
The terms of the treaty are presumed to have the same meaning in each authentic text.
This presumption implies that it is theoretically irrelevant to consult one or more versions, although consulting several would result in greater accuracy of the interpretation. Of course the positions in doctrine are not uniform, with some authors maintaining a side and vice versa, however, nor the Vienna Convention nor other ones provide any obligation for consulting several versions instead of one. The rationale of this lack of obligation is easily understood: simply there are very, very few experts or groups of experts who are able to manage more languages in the framework of international law.
After the discovery of a divergence
In such a situation, the principle that holds this entire digression is put at stake: should the presumption of unity of a treaty still continue to hold or should one language version prevail on the others when two language versions suggest two different provisions? In other words, do articles 31, 32 and 33 of the Vienna Convention indicate that a more correct version, to which preference shall be given, could be identified among all?
In its 1966 commentary to the Final Draft articles, the ILC states that “every effort should be made to find a common meaning for the texts before preferring one to another” , and therefore reaffirms that the principle of unity should be maintained. However, in the same sentence it is provided that there could be a possibility to prefer a text to another, if all other possible methods have proved unsuccessful, as stated in Article 33 of the Convention.
The structure of this Article has created a rather harsh dispute, as one hand it provides the principle of equality in texts and on the other it provides an exception to the principle itself. Some authors regard this paradox to the article and to the attitude of the ILC, which “shows discrepancies and is unsatisfactory” (Bernhardt), others believe the ILC to be free of guilt and regard this “exception” to be “a simple consequence of the very nature of the presumption” (Germer).
In a nutshell, there is no solution to this question, the disputes have been going on for years and they will probably keep on going as, lacking a general rule “the answer probably differs for each individual linguistic discrepancy”36.
A Hierarchical Order in the Rules of Interpretation?
There are three points that need elucidation regarding this topic: whether the paragraphs of article 31 are put in a sort of order or hierarchy, whether articles 31 and 32 are to be considered at the same level, whether the procedure and the linkages between the articles should change when interpreting multilingual treaties.
Regarding the first point, an answer can be found in the way article 31 is provided: a “general rule of interpretation”. The ILC, by choosing to define this article as a single rule instead of several ones, has chosen to indicate that the four paragraphs are intensively connected within each other, and that, having been drafted as a whole and on logical, rather than on hierarchical basis, their application involves a single operation. This vision is generally accepted and largely undisputed in international law, in fact, when considering all the possible context that could occur, having a pattern to follow would be too constrictive, whereas a certain amount of freedom in the art of interpretation is needed.
The second dispute takes place on a comparison level: the provisions of article 32 are labelled as “supplementary means of interpretation”, does this imply that article 31 should enjoy a higher position? Moving to a broader approach, the question is really on whether precedence should be given to the text of the treaty over the intentions of the parties (Bernhardt), or if the two are to be considered at the same time for a starting point (Mc Dougal). It doesn’t seem however that the ILC has had many doubts on stating the correct viewpoint from which the situation should be analysed. Although it pointed out that it did not intend to draw a rigid line between the two, the provisions of article 31 are considered as enjoying a superior authentic character in relation to the elements in article 3237.
Since many more factors are implied, multilingual interpretation is, in many terms, different than “normal” interpretation. This draws attention to the thesis, according to which, the same set of rules is not as proper in such cases, and therefore also all the existing links and relations within articles and paragraphs should be reviewed under a different light. However, such a thought does not take into account two elements:
1. since there is not a strict pattern to be followed, the considered rules result to be pretty plastic and can be adapted in different ways.
2. the general (and non-written) procedure outlined (see supra, “Articles 31, 32, 33 and the interpretation process”), which is generally accepted, is just an enlargement of the plain procedure used for unilingual interpretation.
In conclusion “the system of interpretation for multilingual treaties is the same as for those which are unilingual, featuring only the added element of the comparison of the texts”38
A “transversal” problem: the provisions of article 32 – “the supplementary means of interpretation”39
Article 32 states that “recourse may be had to supplementary means of interpretation, including the preparatory work of the treaty and the circumstances of its conclusion”40.
There are two points of major concern that this article leaves unsolved: first, the problem regarding the opposability of the preparatory works to States that had not participated in the negotiations, secondly, article 32 does not indicate any criteria for individuating other supplementary means, different from the two quoted ones.
Regarding the first point,
The problem of the opposability of the preparatory works of a treaty to States that had not participated in the negotiations has been solved differently […]. Some maintained that a State that had not participated to negotiations is not entitled to know the positions maintained by the States and that it is only bound by the text of the treaty, especially when the working papers have not been published.
Against this position, some others argued that the States agree on a treaty, not on a “text” and, especially, when considering the perspective of the agreement, they don’t just quickly read the text, but prepare accurate studies [on the matter].
Moreover […] the exclusion of relying on the working papers in the relationships with some States would lead to use two different criteria in the interpretation of the same treaty, which would then assume two different meanings.41
Another linkage to this point derives from the interpretation of single words, or terms, there are three cases that ought to be considered:
1. When the ordinary meaning of a term results absurd or unreasonable;
2. When the general rule does not allow to distinguish, within all the special and ordinary meanings of the considered term, which one is the one to be taken into account in the contingent situation;
3. When, although it exists an ordinary meaning of a term, it is to another one that the parties have chosen to consider, manifesting their intention in the preparatory works.
Point 1 does not imply particular difficulties for the interpreter, as the main point to rely on is considered to be the “object and purpose of the treaty”. A different situation exists in points 2 and 3, where the discretion of the interpreter assumes a much broader importance and where, consequently, the risks of misunderstandings and conflicts on interpretation are much higher42.
Regarding the second point, it is to exclude that the two quoted methods are to be considered the only accepted ones. The use of the term “including” in the English version is enough to give credit to this interpretation. However, in the case other “supplementary means” are accepted, problems of contradiction within them might occur. Specific referral is made to the technical rules or logics of the interpretative ratio, embodied in sentences such as ut res magis valeat quam pereat, expressio unius exclusio alterius, genus per specialia derogatur, in dubio mitius etc. The principles here expressed are common in most juridical environments and are applicable in very different situations and with very different goals. Exactly for their nature so common and, also, so vague, the interpreters must be very careful not only in recalling them, but also, and most of all, in selecting them. The interpreter is, in such cases, awarded with a broad range of discretion and freedom for his job; however, this vastness of choice might also result in a greater difficulty for fulfilling the task as, for their lack of specificity, these general principles might act as a farther obstacle for they might be applied in several different ways and, when many of them are used for the interpretation process, might also build a rather complicated network of contradictions.
> STEFANIA BARIATTI, L’interpretazione delle convenzioni internazionali di diritto uniforme, in Studi e pubblicazioni della rivista di diritto internazionale privato e processuale, 1986;
> JEFFREY S. MORTON, The International Law Commission of the United Nations, 2000, Columbia, SC, United States of America;
> MALA TABORY, Multilingualism in international law and institutions, Alphen aan den Rijn, The Netherlands, 1980;
> SIR ARTHUR WATTS, The International Law Commission 1949 – 1998, 1999, New York, NY, United States of America;
> Charter of the United Nations;
> Statute of the International Law Commission;
> Yearbook of the International Law Commission 1962, 1966;
> Convention on the Law of Treaties of 1969;
> Convention on the Law of Treaties between States and International Organizations of 1978;
> International Law Commission Commentary to the Final Draft Articles of the 1969 Vienna Convention on the Law of Treaties.
1 For more infomation on interpretation and the implications regarding its practical application see: STEFANIA BARIATTI, L’interpretazione delle convenzioni internazionali di diritto uniforme, in Studi e pubblicazioni della rivista di diritto internazionale privato e processuale, 1986.
2 STEFANIA BARIATTI, L’interpretazione, p. 264, 1986.
3 JEFFREY S. MORTON, The International Law Commission of the United Nations, 2000, Columbia, SC, United States of America, p. 4.
4 Charter of the United Nations, article 13, paragraph 1, l.a.
5 Statute of the International Law Commission, article 1, paragraph 1: “The Commission shall have for its object the promotion of the progressive development of international law and its codification”.
6 http://www.un.org/law/ilc/ - for more information refer to the Yearbook of the International Law Commission.
7 SIR ARTHUR WATTS, The International Law Commission 1949 – 1998, 1999, New York, NY, United States of America, p. 610.
8 Ibid., p. 610.
9 YEARBOOK OF THE INTERNATIONAL LAW COMMISSION (14th session, 1962), Vol. II, p. 160, para. 17.
10 MALA TABORY, Multilingualism in international law and institutions, Alphen aan den Rijn, The Netherlands, 1980, p. 100 to 101.
11 An example of this practice can be found in the Charter of the United Nations itself at Article 111: “The present Charter, of which the Chinese, French, Russian, English, and Spanish texts are equally authentic, shall remain deposited in the archives of the Government of the United States of America. Duly certified copies thereof shall be transmitted by that Government to the Governments of the other signatory states”. This is also considered to be “consuetudinary law”.
12 MALA TABORY, Multilingualism, p. 170 to 171.
13 MALA TABORY, Multilingualism, p. 172.
14 The 1969 Vienna Convention on the Law of Treaties refers specifically to treaties within States, Treaties within other actors are ruled by the Convention on the Law of Treaties between States and International Organizations of 1978. Rules on interpretation can be found at articles 31, 32, 33, which recall exactly the Vienna Convention of 1969.
15 It is necessary to point out that, although the provisions of article 32 are regarded as “supplementary”, there is no distinction between them and the general rule because the ermeneutical process mantains its sistematical unity (S. Barlatti).
16 YBILC, (1966-I) 874th meeting, para. 13.
17 Vienna Convention on the Law of Treaties of 1969, Article 48, paragraph. 1.
18 Please note that the quoted articles are the ones of the 1969 Vienna Convention on the Law of Treaties, but the concepts here explained can be applied to treaties whose parties are not necessarily only States, but also – for example – international organisations.
19 For a practical example, see MALA TABORY, Multilingualism, Appendix – “Use of Languages in the UN System”, p. 237.
20 MALA TABORY, Multilingualism, p.133.
21 Ibid., p.132.
22 STEFANIA BARIATTI, L’Interpretazione, p.140.
23 MALA TABORY, Multilingualism, p.133.
24 See MALA TABORY, Multilingualism, Conclusion, page 227 to 234.
25 The UN terminology bulletins currently in use are a very large number. Their application started in the early 1970s and it now covers the most diverse fields, from “Tea Terminology” (1979) to “UNCTAD: General terminology - provisional glossary of titles” (1983).
27 MALA TABORY, Multilingualism, p. 141.
28 Arthur J. Goldberg. U.S. Ambassador to the United Nations (1965-1967), June 10, 1977.
29 MALA TABORY, Multilingualism, p.135.
30 MALA TABORY, Multilingualism, p. 193.
31 Vienna Convention on the Law of Treaties of 1969, Article 85.
32 Specifically in this case, see note 18.
33 ILC Commentary to Article 29 of the Final Draft Articles of the 1969 Vienna Convention on the Law of Treaties, para. 6.
34 Ibid. para. 7.
35 Ibid. para. 7.
36 MALA TABORY, Multilingualism, p. 202.
37 ILC Commentary on articles 27, 28 of the Final Draft Articles of the 1969 Vienna Convention on the Law of Treaties, para. 9.
38 MALA TABORY, Multilingualism, p. 195.
39 See note 18.
40 Vienna Convention, Art. 32.
41 STEFANIA BARIATTI, L’Interpretazione, p.211.
42 STEFANIA BARIATTI, L’Interpretazione, p.205 to 207.
Pubblicato su www.AmbienteDiritto.it il 14/01/2008