There is one and only one internationally accepted document which supplies a legal frame for the interpretation of treaties by the means of provisions that may be taken as compulsory rules. The document in question is the Vienna Convention on the Law of Treaties (VCLT), Articles 31 and 32, adopted unanimously at the 1968/9 Vienna Conference.

The two articles read as follows:^[3]

31 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 connexion with the conclusion of the treaty;
b) any instrument which was made by one or more parties in connexion 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.

32 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.'

If one decides to call the two articles 'the international law of interpretation', and I think that there are no compelling reasons why we should not call them so, then, first of all, we need to have in mind that the law of interpretation is an outcome of a long process of negotiation (stretching from the early 1964 till May 1969) involving many individuals and institutions. Three key institutions were directly engaged: the International Law Commission (ILC);^[4] the Committee of the Whole of the Vienna Convention Conference (a representative of the UN General Assembly) whose task was to agree on the final version of the Convention; and the General Assembly itself. The General Assembly, of course, made the ultimate decisions concerning Articles 31 and 32 by a majority vote. Secondly, the ILC was definitely the key pillar of the whole process because it fed the other actors with the draft articles and, after receiving their feedback, worked on the penultimate and ultimate versions of the law concluded at the Vienna Conference on 23 May 1969.^[5] Humphrey Waldock, the Special Rapporteur of the ILC, was the key individual in the process as he prepared the first drafts and exerted an enormous philosophical influence on the other members of the ILC.

Arts. 31 and 32 embody something one can call 'a legal response to the problem of ambiguity' which can be presented roughly in the following words:

"Imagine that you have recognised an ambiguity^[6] in a legal document/treaty. The ambiguity implies that the meaning of the document is not fully transparent, or given; hence a legal response to the document requires that we restore its meaning, or recover the transparency of its meaning, through the process of interpretation that takes into account all the elements needed to restore the meaning. Provided that we can spell out those elements, and rank and couch them in a relatively brief and practical form, there would not be any reason why we should not view such elements as the rules of interpretation to which the status of compulsory rules, the force of law, should be attributed. It is the last step that makes this response strictly legal one."

There is no doubt that the parties to an agreement have the duty to interpret it because the very adoption of the agreement implies such duty. But, does the accepting of such duty entail the requirement to formulate and apply a set of compulsory rules? Given that we are already endowed with the ability to interpret a text, or a pattern of language, - the fact that we learned how to construct and understand sentences in our mother tongue implies such ability, can we, however, formulate a universal algorithm that can help us to remove all ambiguities in an objective/impartial and unambiguous manner? Looking into the VCLT, and measuring overall achievement of the international legal scholars, and governmental representatives, in this field by the achievements of the Convention, this paper answers the question in the negative.

In other words, here I attempt to demonstrate that Articles 31 and 32, formulated with the aim to remove ambiguities from the treaties, are ambiguous too; that, instead of reducing ambiguities to an unambiguous language, they in fact reproduce ambiguity. In a sense, this is a huge problem because the said articles deliver the opposite of their promise - instead of removing the problem they double it. In another sense, however, this is good news because it can inspire one to search for a more plausible view of the problem of ambiguity in international-political, legal, or generally diplomatic contexts.

The method I apply to substantiate and support my key claim is of both genealogical and textual nature. Simply, the documents and records of the main stages of the process of both drafting and negotiating the 31 and 32, taken as both a context and travaux preparatoires (i.e. preparatory work) for the two articles, are looked into to form a plausible view of the meaning of the articles. For analytical purposes, I divide the negotiating process in three stages:

1. 1964 - the year when the ILC prepared the first draft articles and passed them to the General Assembly (Section I.1);
2. 1965-1966 - the period in which individual governments prepared their responses to the first drafts, after which the ILC amended its drafts accordingly (Section I.2);
3. 1968-1969 - the very Vienna Conference on the Law of Treaties, which concluded with adoption of the VCLT (Section I.3).

In the course of my discussion I increasingly focus on Waldock's responses to the US critique of the ILC drafts to make an increasingly convincing case for the thesis that, in order to gain something, Waldock deliberately made the draft Articles interpretable and thus sacrificed their 'single voice'. This is the essence of the argument proposed in Section I. Section II offers briefly some ideas and conjectures about the worth and the consequences of the argument.

I Negotiating the international law of interpretation

I.1 Stage One - from the ILC to the governments

Negotiations on Articles 31 and 32 commenced on 14 July 1964 at the ILC. At a meeting under the chairmanship of Roberto Ago, an introductory word was given to Humphrey Waldock who pointed out that the ILC should answer the question of whether the Convention should contain any provisions dealing with interpretation; interestingly, he also added that it was impossible to find enough state-owned documents relating to the states' interpretative practice.^[7] He decided to propose four articles dealing with interpretation (70-73 in the enumeration at the time), of which the 72 stipulated that the interpretation which gives a better effect to a treaty should prevail, whilst the 73 addressed a retroactive effect that the rules of law might have on the matters of interpretation; the 70 and 71 were very similar to the ultimately adopted 31 and 32.

At the same session the ILC members provided various comments some of which were somewhat sceptical. For instance, member Ruda cast doubt on the meaning of the process in which the ILC engaged at the time and insisted that at the stage of development of the international law one could not yet speak about the rules of interpretation that were obligatory for the states; the only secure rule at the time was 'Vattel's rule' ('in claris non fit interpretatio' meaning 'when matters are clear, there could be no question of interpretation').^[8] Verdross added that the articles would not be binding for the states as they could agree on any means of interpretation they like; in a similar spirit, member Yassen added that, 'excessive detail should be avoided.'^[9] Summarising the spirit of the session, chairman Ago said that no member of the ILC explicitly proposed to remove the interpretation articles from the Draft Convention, and that those articles were 'of an even more provisional nature than the rest of the draft [of the VCLT].'^[10]

On 15 July the ILC held another session at which a majority of its members agreed that Article 72 should be removed on the ground that, normally, any acceptable interpretation would give effect to the interpreted treaty. As a result of the revision, the ILC agreed about the draft of three articles - the 69, 70, and 71. Article 71 reads almost the same as paragraph 4 of Article 31, whilst Article 70 reads the same as the 32. Article 69 reads as follows:

1. A treaty shall be interpreted in good faith in accordance with the ordinary ^[11] meaning to be given to each term: a) In the context of the treaty and in the light of its objects and purposes; and
b) In the light of the rules of general international law in force at the time of its conclusion.

2. The context of the treaty, for the purposes of its interpretation, shall be understood as comprising in addition to the treaty, including its preamble and annexes, any agreement or instrument related to the treaty and reached or drawn up in connexion with its conclusion.

3. There shall also be taken into account, together with the context:
a) any agreement between the parties regarding the interpretation of the treaty;
b) any subsequent practice in the application of the treaty which clearly establishes the understanding of all the parties regarding its interpretation.

In order to enable the GA members to get a clear picture of the direction the ILC was taking, Waldock attached a commentary to the three articles, which formed a crucial part of the ILC report to the GA.^[12]

The commentary addressed many issues pertaining both to the topic and the wording of the articles and, more importantly, provided a key part of travaux preparatoires for any future interpretation of Articles 31 and 32. It starts with the American approach as formulated in the Harvard Law School. Crucially, the School claimed that, The process of interpretation, rightly conceived, cannot be regarded as a mere mechanical one of drawing inevitable meanings from the words in a text…interpretation involves giving a meaning to a text…This is obviously a task which calls for investigation, weighing of evidence, judgment, foresight, and a nice appreciation of a number of factors varying from case to case. No canons of interpretation can be of absolute and universal utility in performing such a task, and it seems desirable that any idea that they can be should be dispelled.^[13]

Waldock, however, hoped that some general rules of interpretation could be formulated. In paragraph 3 of his commentary he focuses on the issue of relative weight that should be given to the text of a treaty, on the one hand, and to the intentions and purposes of the treaty parties, on the other; in his view, a majority of scholars consider the treaty text as a primary basis of interpretation.^[14] Furthermore, in paragraph 5, he explains the ILC attitude to the rules of interpretation - those rules represent the rules of a common sense and logic, but their application also depends on a variety of considerations which have to be appreciated by the interpreter of the document... even when a possible occasion for their application may appear to exist, their application is not automatic… In other words, recourse to many of these principles is discretionary rather than obligatory and the interpretation of documents is to some extent an art, not an exact science. ^[15]

But, if interpretation is more an art than a science, how can one frame it in the language of compulsory rules, as Waldock also suggests in one of the crucial parts of his commentary?^[16] We will see that this question will haunt ILC's attempts to formulate a viable international law of interpretation till the very end.

Throughout 1965 the UN GA members were responding to this report. On behalf of the ILC, Waldock presented those responses and replied to them in the ILC 1966 Sixth Report on the Law of Treaties. ^[17]

I. 2 Stage Two - from the governments to the ILC

Some governments, including Finland, Turkey and Kenya, briefly welcomed the ILC draft. ^[18] However, some have raised more substantial objections. For instance, the Greek and the Hungarian opposed both the empirical undertones of the draft and its insistence on 'text' as the primary element of interpretation.^[19]

However, the commentary by the US government carried a special weight. It included four crucial points as follows: 1. The US representatives proposed that the rules of interpretation should be called 'guidelines' instead of 'rules'. 2. Secondly, and in many ways this was crucial, they did not agree with the primacy assigned to 'ordinary meaning'. 3. They also considered Article 70 concerning 'preparatory work' as 'unduly restrictive'. 4. Finally, the US emphasised that, should the concept of context be defined, it should be done in clearer terms than those the ILC proposed. ^[20]

Hence the US suggested that the 'business' of drafting of the interpretation articles was not complete and that substantial changes needed to be introduced. ^[21]

Waldock tried to respond constructively to all comments. As to the US proposal to use the term 'guidelines', he pointed out that governments seem to 'endorse the attempt of the Commission to isolate and codify the general principles which constitute general rules for the interpretation of treaties', and, as the ILC intended to avoid rigidity, the rules proposed could already be taken as 'guidelines'. ^[22]

Secondly, in his opinion, a majority of governments endorsed the textual approach to interpretation, not an approach in which interpretation implies an ab initio inquiry into the intentions of the treaty parties. However, in his response to the Czechoslovakian proposal to add a provision endorsing explicitly the textual approach, Waldock makes an enigmatic remark:

The presumption suggested by the Czechoslovak Government is closely in line with the concept of interpretation expressed in the article. But the statement of the presumption may tend to raise the question how far the presumption is rebuttable and what precisely is the relation between the presumption and other elements of interpretation mentioned in articles 69-71.^[23]

Curiously again, responding to the US proposal to enlist the rules of interpretation seriatim without a hierarchy implied, Waldock emphasised that the ILC has not intended to suggest a hierarchy. The ILC title for Article 69 is 'General Rule of Interpretation' (in singular), and the rule conceives of the process of interpretation as if 'all the various elements, so far as they are present in any given case, would be thrown into the crucible and their interaction would then give the legally relevant interpretation.' ^[24] In other words, Waldock here compares the process of interpretation with a creative act that does not imply a hierarchy.

He also explained that the 'hierarchy' that the ILC had in view was not a matter of legal methodology, but of the most fundamental logic: logic requires an exposition of a set of rules in a certain order; this had nothing to do with the legal aspects of the process of interpretation. For example, if we start with 'ordinary meaning' then logic requires that the meaning is situated within its context and related to the object and the purpose of a treaty - this implies that the ILC view of 'context' was built on the most fundamental logic, not on some controversial legal theories.^[25]

After having received the governmental comments, the ILC revised Articles 69-71 and proposed their revisions as Articles 27 and 28,^[26] under new enumeration; obviously, some of the US critical observations had been incorporated in the 27 and 28. For instance, the ILC accepted the US proposal that 'general' should be deleted from 'general international law'. Article 71 concerning 'special meaning' was moved to Article 27 [previously 69] as its last paragraph. In sum, the ILC 1966 draft Articles 27 and 28 ^[27] were identical to today's 31 and 32, except that the wording 'which establishes the understanding…' (27 3b) was later replaced with 'which establishes the agreement…' (31 3b). The 28 reads the same as the 32.

The new draft articles were published as a part of the ILC Report on the work of its 18th session (Geneva, 4 May -19 July 1966)^[28] and passed to the GA.

There is no doubt that the reduction of the initial four to two articles should be taken as one of the key aspects of this whole process. In that respect, a commentary from the 18th session of the ILC, too, should be given a due consideration: …the Commission was of the opinion that the distinction made in articles 27 and 28 between authentic and supplementary means of interpretation is both justified and desirable. At the same time, it pointed out that the provisions of article 28 by no means have the effect of drawing a rigid line between the 'supplementary' means of interpretation and the means included in article 27. The fact that article 28 admits recourse to the supplementary means for the purpose of 'confirming' the meaning resulting from the application of article 27 establishes a general link between the two articles and maintains the unity of the process of interpretation. ^[29]

This means that the whole negotiating process can also be interpreted plausibly as resulting in a single article. Hence, on the one hand, the number of rules (articles) was reduced but, on the other, something else was gaining in complexity as a result of the pressure the US put on the ILC. The next sub-section will try to illuminate that aspect of 'growing complexity'.

I. 3 Stage Three - the UN Conference

The Vienna/UN Conference on the Law of Treaties commenced on 26 March 1968. At such conferences a body is ad hoc established to prepare the final drafts that the state representatives vote through or down. For the Conference, the Committee of the Whole (CoW) represented such a body, and Humphrey Waldock was given another role, of an Expert Consultant, for both the Committee and the Conference. The first session was held from 26 March till 14 May 1968.

The 31st meeting of the CoW, held on 19 April,^[30] commenced with the US critique of the proposed articles and its demand that the proposals be replaced with an US amendment. On behalf of the US, Myres McDougal argued that the 27 and 28 were excessively rigid.^[31] Furthermore, he stated that the canons of interpretation have rarely been considered as mandatory rules of law that would prevent investigation into the relevant circumstance,^[32] and called on an ILO observer who, at the 7th meeting of the CoW, argued that 'the ILO practice on interpretation had involved greater recourse to preparatory work than was envisaged in article 28.' Even worse, in McDougal's opinion, Waldock's inclusion of travaux preparatoires into supplementary means was based on a misreading of the judgements of the International Court of Justice. Crucially, The rigid system of articles 27 and 28 was thus not an expression of existing rules of international law. Furthermore, if an attempt were made to introduce it, it would prove totally unworkable. It was based on the assumption that a text had a meaning apart from the circumstances of its framing, and that it could be interpreted without reference to any extraneous factor. In reality, words had no fixed or natural meaning… ^[33]

Drawing on a maximum of his rhetorical skill, at the end he offered two additional arguments against the ILC approach. First, the very presence of Waldock in the role of 'expert consultant' and frequent reliance of the CoW on his interpretations suggest that, for the purposes of interpretation, one's familiarity with the circumstances of the conclusion of a treaty is more important than the treaty text itself.^[34] Secondly, the rigid system of Articles 27 and 28 'should not be made international law because it could be employed by interpreters to impose upon the parties to a treaty agreements that they have never made.'^[35]

That is why the US proposed a single article/amendment which reads as follows:

A treaty shall be interpreted in good faith in order to determine the meaning to be given to its terms in the light of all relevant factors, including in particular:
a. the context of the treaty;
b. its objects and purposes;
c. any agreement between the parties regarding the interpretation of the treaty;
d. any instrument made by one or more parties in connexion with the conclusion of the treaty and accepted by the other parties as an instrument related to the treaty;
e. any subsequent practice in the application of the treaty which establishes the common understanding of the meaning of the terms as between the parties generally;
f. the preparatory work of the treaty;
g. the circumstances of its conclusion;
h. any relevant rules of international law applicable in the relations between the parties;
i. the special meaning to be given to a term if the parties intended such term to have a special meaning. ^[36]

Expectedly, the US proposal intensified the debate within the CoW and prompted further disagreements within the international community.^[37]

The crucial 33rd meeting was held on 22 April. After a long session it resulted in a decision to vote down the US amendment. Before the vote, Waldock delivered the closing words that stand out as the session's most interesting part.

They are the most interesting part, and the part of vital importance for the argument proposed in this paper, because they enable one to realise that Waldock accentuated certain contextual factors from within the history of drafting of the 'interpretation' articles to bring their meaning closer to the spirit of the US amendment. In other words, perhaps intentionally, Waldock gave the 'flavour' of ambiguousness to the means the ILC proposed to be the means of disambiguation of treaties. His 'ambiguation' of those articles becomes especially visible once we focus on the way his closing words tackle three key issues (in the following text marked with A, B and C).

A Waldock says that the adjective 'special' in the phrase 'special meaning', and perhaps the whole paragraph 4 of Article 27, may be superfluous (!) - a special meaning is simply an ordinary meaning relative to a particular context - and that, 'as far as article 27 was concerned, the intention had been to place on the same footing all the elements of interpretation therein mentioned.' ^[38] However, let us note that such a claim brings the concept of 'ordinary meaning' closer to the spirit of the US amendment - if the distinction between an ordinary and a special meaning depends fully on the relationship between a meaning and a context in which it is situated, then it makes no sense to speak about 'ordinary meaning' in its isolation from the context. In other words, to arrive at a proper meaning of a word, or a sentence, one has to place unexceptionally all meanings within their context and then the context will determine if the word, or the sentence, carries a special or an ordinary meaning. Hence Waldock suggests both indirectly and vaguely to McDougal that he has got an open room to interpret the 27 (i.e. 31) as if the assumption of the primacy of 'ordinary meaning' is as provisional as the assumption of the primacy of 'special meaning'.

B In his closing word Waldock blurs the division line between the general rule of interpretation (27/31) and the supplementary means of interpretation (28/32); or, more precisely, he says a number of very unclear things about the division line, which is why one can get but an ambiguous answer to the question of whether the interpreter should take the 27 as compulsory rule and the 28 as a permissive one, or s/he should take the two as a single article which consists of not neatly distinguished both compulsory and permissive elements that provide only loose guidelines for interpretation. To remind, the 'supplementary means of interpretation' are not explicitly or exhaustively defined. We know that they include travaux preparatoires^[39] as well as the circumstance of the conclusion of a treaty, but the two factors are not the exclusive ones.^[40] Secondly, the supplementary rule was formulated permissively, as 'recourse may be had…' By contrast, initially the general rule of interpretation had the status of a compulsory rule. The US request to combine the 27 and the 28 obviously aimed to increase the flexibility of interpretation and eliminate the hierarchy of the means of interpretation. How Waldock replies to such a request in his closing word?

First, he claims that, according to his personal experience, legal interpreters do not rely on a preparatory work unless a problematic situation arises. This is a curious reply because the requirement to interpret already implies an interpretative obstacle, hence a problematic situation. In other words, his first remark can be taken as being in line with both the US request to allow for an automatic recourse to travaux preparatoires prior to any reliance on the interpretative means enumerated in the 31, and the understanding that the 27 and the 28 have a different legal status. Note also that in his closing word Waldocks emphasises that, 'there had certainly been no intention of discouraging automatic recourse to preparatory work for the general understanding of a treaty.' ^[41]

Secondly, it is in 1966 that Waldock points out that the formulation of Article 70 (i.e. the 28 in the 1968 enumeration, or the 32 in today's) seemed to the Commission to be about as near as it is possible to get to reconciling the principle of the primacy of the text…with the frequent and quite normal recourse to travaux preparatoires without any too nice regard for the question whether the text itself is clear. Moreover, the rule formulated in article 70 is inherently flexible, since the question whether the text can be said to be 'clear' is in some degree subjective.

But, if the application of Article 28 (i.e. 32) is triggered by the realisation that the application of Article 27 (i.e. 31) leads to 'an ambiguous or obscure result', the realisation that the application of Article 27 in its final result ought to trigger Article 28 is, if we follow Waldock's logic, subjective too. This implies that nothing can prevent one to view Articles 27 and 28 as a single, combined and flexible rule that should be applied in one go as the US proposed. But, this may also imply that nothing can prevent one to view the result of the application of the 31 as not requiring a further recourse to the 32 - the opinion that the meaning resulting from the application of the 31 is clear, or obscure, is always 'in some degree subjective'. ^[42]

To sum up, in 1964 Waldock suggests that Article 27 (i.e. 31), whenever it can deliver a clear and unambiguous result, is the only 'must' of legal interpretation. In 1968/9, and to a lesser degree in 1966, he complicates and 'ambiguates' such a suggestion by emphasising that a) the opinion that the result of the application of Article 27 (31) needs to be confirmed by the application of Article 28 (32) is to an extent subjective; b) combination of Articles 27 and 28 maintains the unity of the process of interpretation; c) Article 28 adds an additional firmness and reliability to the results of the application of Article 27, and can be used in forming the general understanding of a treaty.

C Waldock opened his closing words by pointing out that, He wished to dispel any impression that the ILC had approached the problem of interpretation from the point of view of settling a doctrinal controversy. The Commission had of course taken into account the various theories…but the rules, which it had framed, had been conceived as reflecting what happened in State practice. ^[43]

His words indicate that the premises on which the ILC founded its work should not be taken as strictly and straightforwardly as they were initially conceived. To remind, in his 1964 report Waldock stated that, with regard to the doctrinal differences, the ILC should take a clear view as to the role of text in the treaty interpretation. Paragraph 3 of the same report should be read as an endorsement of the textual approach to the treaty interpretation. Now, in 1968, Waldock substantially qualifies such an endorsement either to quiet the US concerns or to suggest to the other governments that the US opposition to the draft articles, on the alleged ground of their doctrinal bias, was unwarranted. However, whatever his motive, by this move Waldock complicates the interpretation of the 31 and 32 because, in the light of some of his claims, the ILC gave primacy to 'text (ordinary meaning)', whereas, in the light of the last citation, such primacy was not given at all (In this context, recall also his 1966 metaphor of 'the crucible'!).

Secondly, his statement describing the work of the ILC as a reflection of 'State practice' does not cohere with his 1964 statement (made at one of the first meetings of the ILC) that, 'evidence of it [State practice] was difficult to obtain as not much was to be found in publications of State practice.' ^[44] The question one should pose is, 'How is it possible that now the ILC articles reflect the State practice, though not much evidence of that practice was to be found'? Perhaps Waldock raised the point concerning the work of the ILC as a reflection of the State practice to quiet the US concerns as he obviously implied that the ILC had no intention to impose the rules of interpretation on any state - the ILC drew only on the (existing) State practice.

To conclude then: Waldock's closing comments are of enormous significance because they reveal that the international law of interpretation is a host to dramatic, puzzling, and vertigo-inducing forms of ambiguity. Perhaps such an outcome is predictable. To an argumentative attack by a party A against a party B the latter can respond by arguing that the former misinterpreted her case or that the substance of A's objections is already incorporated in her case. However, to launch her counter-argument successfully B has to twist, to reinterpret, and to inflate and blur, the meaning of her initial claims. This gives rise to ambiguity. ^[45] The same applies to Waldock's response to the US objections. Obviously he was very eager to get the US support, and hoped that he would demonstrate to the US that the substance of its objections was already incorporated in 'his' draft articles; 'ambiguation' of the articles was a logical consequence of such argumentative strategy.^[46]

II The worth and the consequences of the argument

The argument proposed above has a sufficient explanatory worth vis-à-vis certain phenomena in the field of international legal interpretation.

First, imagine that a court needs to apply Articles 31 and 32 to a particular international dispute. Due to the ambiguities of the international law of interpretation, one should expect a deeply polarised outcome of such an application. In other words, the ambiguities of the 31 and 32 imply that an outbreak of fresh disputes, not a dispute-settlement, is the most likely outcome of the application of the two articles. And, if one, for instance, looks into the International Court of Justice (ICJ) handling of the 1991 'Qatar vs. Bahrain' dispute, ^[47]one will realise that the ICJ's internal dispute over the meaning of the 31 and 32 can best be explained by the hypothesis that the two articles are indeed ambiguous.

The ICJ's handling of the dispute is not the only example of an interpretative disagreement prompted by various equally plausible readings of the 31 and 32. As, in 1984, Sir Ian Sinclair admitted: 'The Convention [interpretation] rules are expressed in very general terms and much in the way of discretion and appreciation is left to the tribunal called upon to interpret a particular treaty provision. Our review of recent international case law on treaty interpretation reveals only too clearly that widely differing results can still be achieved even if conscious effort is being made to apply the Convention [interpretation] rules.' ^[48]

Note here that Sinclair explains the widely differing results by the generality of the 31 and 32; in my opinion, the Section I argument makes 'the widely differing results' even more logical and expected.

Secondly, the argument proposed in Section I is explanatorily worthy in another way. It enables one to understand apparent contradictions in the attitudes of several key legal scholars involved in the process of negotiating of the 31 and 32 to the two articles. For instance, in 1968 Ian Sinclair opposed the US amendment on the grounds that it would encourage unnecessary disputes and that the 31 and 32, as drafted in 1968, possessed the capacity to put an end to debate at international conferences.^[49] However, in the light of his 1984 admission, the 31 and 32, too, leave too much a room open to discretion and interpretation, and thus have the capacity to encourage unnecessary disputes; this simply means that in 1968 Sinclair's focus remained only on one of the two equally possible interpretations of the 31 and 32, whilst in 1984 he became more aware of the unsettled meaning of the said articles. Hence the thesis, that the ambiguities of the 31 and 32 are indeed ambiguities, explains why Sinclair's successive attitudes to the international law of interpretation are only apparently contradictory.

In my presentation of the CoW debate I deliberately omitted one very curious part of McDougal's debate with Waldock. A part of McDougal's argument against Waldock's proposal was expressed in the following way: The criterion of ordinary meaning, because of its ambiguity, opened the door to arbitrary interpretations of the text and would create greater uncertainties than an insistence upon a comprehensive, contextual examination of all factors potentially relevant to common intent. ^[50]

But, now, let us also recall that McDougal's first objection against the system of articles 31 and 32 was that the system is excessively rigid. How can the same system be ambiguous and rigid at the same time? Does not this imply that McDougal was very uncertain about the way one should interpret the two 'interpretation' articles? And, more importantly, does not the qualification of the articles as ambiguous ones make the US opposition seem suspect?^[51] Whatever the answer to the third question, the argument proposed in Section I enables one to understand easily the unease and uncertainty so clearly manifested in McDougal's response to the international law of interpretation.

Finally, how has Waldock responded to McDougal's claim that 'the criterion of ordinary meaning' is ambiguous? Actually, that claim is the only part of McDougal's case against the 31 and 32 to which Waldock has not responded. In other words, Waldock has not denied it nor confirmed it, which, when seen in the light of his argumentative strategy of ambiguity to which I referred at the end of I.3, seems fully expected and perfectly logical.

Now, what are the most important consequences of the argument proposed in Section I? First of all, Articles 31 and 32 cannot be applied algorithmically to the case at hand. The two articles are ambiguous and, should we decide to apply them, we need to agree about a proper interpretation of the two; as they are equally open to two equally plausible and yet incompatible interpretations, such an agreement about a proper interpretation cannot be deduced from them. Hence something else should be added to the 31 and 32 to provide the ground for the agreement on the proper interpretation of the 31 and 32, from which it follows that the two articles should be considered as inadequate in the sense of their being essentially incomplete. ^[52] However, this invites another question. Namely, why should one first agree about the meaning of the rules of the treaty interpretation and only thereafter proceed actually to interpreting a treaty? Is not this indirect route of interpretation less plausible than the route of proceeding directly to negotiating an agreement about the meaning of the treaty? It seems to me that this makes the international law of interpretation also redundant in a sense. That much is clear. However, it remains unclear whether this 'failure' of the VCLT interpretation provisions can be explained by a deeper and more general reason. Can one, for instance, reasonably claim that all attempts to formulate an algorithmic method of legal interpretation are bound to fail?

The thought that one could uncover and construct the meaning, or settle a dispute concerning the meaning, algorithmically and through a straightforward application of a number of rules is an odd one. It implies that, for those who possess such an algorithm, ambiguity would never arise. This explains why the notion of algorithmic formulation of the rules of (legal) interpretation has always had to face a strong and persistent opposition that argued approximately along the following lines. ^[53] First, language can be viewed from within two distinct perspectives - as a tool of socially controlled communication and a convention-bound medium; ^[54] or as a tool of individual creation, metaphors and poetry, and a guardian of individualism^[55] bound by conventions only at its periphery. ^[56] Despite the fact that both perspectives must be accepted generally and in principle, they are a source of different and sometimes opposed guidelines that, in specific cases, may lead to irreconcilable interpretations of a pattern of language. Therefore all canons of interpretation must, on the one hand, assume both perspectives and, on the other, acknowledge a potentially uneasy nature of their relationship, which explains why such canons are often couched in vague and ambiguous terms. ^[57] Secondly, language seems to be particularly resistant to legislation.^[58] For instance, the international attempt to pass the legislation for so-called 'hate-speech' ended with adoption of an ambiguous convention.^[59] Prima facie any legal and hence compulsory definition of a required use, and interpretation, of language seems to be too restrictive.

The argument presented here, however, does not depend on such general claims. It starts with the textually and genealogically confirmed thesis that Articles 31 and 32 provide an ambiguous response to ambiguity. This means that we could attempt to disambiguate the articles, but we should not describe our attempt as an attempt to find or uncover their true meaning, or to illuminate their inherent meaning through accumulation of additional meaning-packages; we should describe it as an attempt to reach an creative agreement about a new pattern of language.^[60] This, obviously and automatically, implies that, in our dealing with the problem of diplomatic/legal/political ambiguity, we need to move beyond the logic that animates the 'rules of interpretation' approach - ultimately, the language of 'the rules of interpretation' needs to be replaced with a different language.

Due to the limitations of space, here I cannot offer an outline of such 'different language', but can give a few hints concerning the direction in which we ought to seek it. Note that the VCLT approach to interpretation focuses only and narrowly on meanings - it treats a to-be-disambiguated meaning as a puzzle that can be solved by gradual addition of certain atomic packages of meaning, or small pieces of information from which separate factors of meaning can be first distilled and then combined to generate a transparent and rock solid meaning. But, does not such an approach miss something important for understanding and interpreting the language? Does not it miss an important insight by Wittgenstein put in the following words: 'Communication through language implies not only an agreement in definitions, but also (queer as this may sound) an agreement in substantial claims'?^[61] In other words, Wittgenstein reminds us that the factor of meaning cannot be in any sensible way divorced from the factor of belief; the two factors are inextricably intertwined and each can be clarified, settled, and illuminated, only in connection with the other.^[62] Interpretation is not only a search for meanings, but also a contest of beliefs, and this is what makes it both challenging and rewarding. This, however, also implies that we need to bring the concept of ambiguity in the proximity of the concept of belief, and then make an attempt to couch the former not only in terms of meaning, but also in terms of belief, by redrawing it as a conflict, or potentiality, of both meanings and beliefs. In spite of the fact that, upon the first encounter, such a theoretical strategy sounds unpromising, it can nonetheless be tried and plausibly shown to be coherent, illuminating, relevant, and productive.

References:

Abel, R. (1994), Speech and Respect, London: Stevens & Sons, Sweet & Maxwell.
Aristotle, (1877), The Rhetoric of Aristotle, Book III (with a commentary by Edward M. Cope, revised and edited by John E. Sandys), Cambridge at the University Press
Aust, A. (2000), Modern Treaty Law and Practice, Cambridge: Cambridge University Press.
Barwise, J. (1988), 'On the Circumstantial Relation between Meaning and Content', in U.Eco, M. Santambrogio and P.Violi (eds.), Meaning and Mental Representation, (Bloomington and Indianapolis: Indiana University Press), pp. 23-39.
Beardsley, M. (1961), 'On Contexts and Vagueness', in K. W. Hunt, P. Stoakes (eds.) Our Living Language (Boston: Houghton Mifflin Company) pp. 37-47.
Byers, M. (2002), 'The Shifting Foundations of International Law: A Decade of Forceful Measures against Iraq', European Journal of International Law, vol. 13 no. 1, pp. 21-41.
Chomsky, N. (1989), Necessary Illusions. Thought Control in Democratic Societies, London: Pluto Press.
Davidson, D. (1982), Essays on Actions and Events, Oxford: Clarendon Press.
_____ (1984), Inquiries into Truth and Interpretation, Oxford: Clarendon Press.
_____ (1986), 'Nice Derangement of Epitaphs', in E. LePore (ed.), Truth and Interpretation. Perspectives on the Philosophy of Donald Davidson (Oxford, New York: Basil Blackwell), pp. 433-446.
_____ (1991), 'James Joyce and Humpty Dumpty', in P.A. French, T.E.Uehling, Jr., Howard K. Wettstein (eds.), Midwest Studies in Philosophy, vol. 16, (University of Notre Dame Press), pp. 1-12.
_____ (1999), 'Interpretation: Hard in Theory, Easy in Practice', in M. De Caro (ed.) Interpretations and Causes. New Perspectives on Donald Davidson's Philosophy (Dordrecht, Boston, London: Kluwer Academic Publishers), pp. 31-44.
Dickson, J. (2001), 'Interpretation and Coherence in Legal Reasoning', in E. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Fall 2001 Edition) (URL= http://plato.stanford.edu/archives/fall2001/entries/legal-reas-interpret, accessed on 6 February 2006).
Dworkin, R. (1982), 'Law as Interpretation', Critical Inquiry, vol. 9 no. 1, pp. 179-200.
Falk, R. A. (1970), 'On Treaty Interpretation and the New Haven Approach: Achievements and Prospects', in R.A.Falk, The Status of Law in International Society (Princeton, New Jersey: Princeton University Press), pp. 342-377.
Hirsch, E.D. (1967), Validity in Interpretation, New Haven and London: Yale University Press.
Johnstone, B. (1996), The Linguistic Individual, New York, Oxford: Oxford University Press.
Kearny, R.D., Dalton, R.E. (1970), 'The Treaty on Treaties', The American Journal of International Law, vol. 64 no. 3, pp. 495-561.
McDougal, M. S. (1967), 'The International Law Commission's Draft Articles Upon Interpretation: Textuality Redivivus', The American Journal of International Law, vol. 61 no. 4, pp. 992-1000.
Partsch, K.J. (1992), 'Racial Speech and Human Rights: Article 4 of the Convention on the Elimination of All Forms of Racial Discrimination', in S. Coliver (ed.) Striking a Balance: Hate Speech, Freedom of Expression and Non-discrimination (London and University of Essex: Article 19 and Human Rights Centre, University of Essex), pp. 21-28.
Pehar, D. (2001), 'Use of Ambiguities in Peace Agreements', in J.Kurbalija, H. Slavik (eds.) Language and Diplomacy (Malta: DiploProjects), pp. 163-200
Raz, J. (1996), 'Why Interpret?', Ratio Juris, vol. 9 no. 4, pp. 349-363.
Rosenne, S. (1970), The Law of Treaties- A Guide to the Legislative History of the Vienna Convention, Dobbs Ferry, New York: A. W. Sijthoff-Leyden, Oceana Publications, Inc.
Sinclair, I. (1984), The Vienna Convention on the Law of Treaties, Manchester: Manchester University Press.
Sur, S. (1974), L'Interpretation en droit international public, Paris: Libraire General de Droit et de Jurisprudence.
UNCLT (1969), United Nations Conference on the Law of Treaties, 1968, Official Records, New York: United Nations.
UNCLT (1970), United Nations Conference on the Law of Treaties, 2nd Session - Vienna 9 April - 22 May 1969, Official Records, New York: United Nations.
Wittgenstein, L. (1953), Philosophical Investigations, Oxford: Basil Blackwell.
Yearbook (1965a), Yearbook of the International Law Commission for 1964, Vol. I, 11 May-24 July, New York: United Nations.
Yearbook (1965b), Yearbook of the International Law Commission for 1964, Vol. II. Documents of the 16th session including the report of the Commission to the General Assembly, New York: United Nations.
Yearbook (1967a), Yearbook of the International Law Commission for 1966, Vol. I Part II, New York: United Nations.
Yearbook (1967b), Yearbook of the International Law Commission for 1966, Vol. II, New York: United Nations.

Endnotes