At its most recent (70th) session, the International Law Commission adopted two important sets of “restatements” on two important sources of international law on second reading, namely the Draft Conclusions on the Identification of Customary International Law and the Draft Conclusions on Subsequent Agreements and Subsequent Practice in Relation to the Interpretation of Treaties (see the ILC’s 2018 Report (UN Doc A/73/10) here). This post concerns the second of these restatements, subsequent agreements and subsequent practice (see Chapter IV of the Report). In particular, this post expresses a concern about an apparent, almost surreptitious, attempt by the Commission to elevate subsequent agreements and subsequent practice as tools of interpretation to the same level as the more objective tools outlined in article 31(1) of the Vienna Convention of the Law of Treaties. The concern may seem like a storm in a cup – and I hope that is the case. However, there is a real possibility – a possibility which could risk the stability of treaties – that the ordinary meaning of the words of a treaty, in their context and in light of the object and purpose of the treaty could give way to ever-changing moods of States expressed through subsequent agreements and subsequent practice. If states don’t like the terms of the treaties they have adopted, they should amend it through the means provided for in the treaty or in the customary rules on amendments of treaties. Amendment through interpretation, a real likelihood if subsequent agreements and subsequent practice were elevated to an independent status of equal value – perhaps some day even greater – to ordinary meaning, in context and in light of the object and purpose, would be a dangerous course. It is hoped that this implicit suggestion in the work of the Commission is not taken up the practice of courts in the application of article 31.
I should begin by two caveats. First, this post, like the draft conclusions themselves, concerns only subsequent agreements and subsequent practice in relation to treaty interpretation. Thus, what is said here does not affect the role that subsequent agreements or subsequent practice might have, say for modification of treaties in general. Second, there is, admittedly, nothing in the draft conclusions themselves that can be interpreted as the elevation of subsequent agreements and subsequent practice. The (attempted) elevation comes in the commentaries to a number of provisions in the set of draft conclusions. I should note, in connection with the last-mentioned caveat, that the commentaries themselves seem to have been elevated to a higher position than before – not quite on par with the draft conclusions but certainly approaching that level. While in the past, it has been understood that the draft texts adopted by the Commission were to be read with commentaries, during the 70thsession, the Commission inserted language as the first paragraph in the general commentary of both second reading topics to emphasise this point, which had not been emphasised in this manner before.
There is No Order of Importance between Means of Interpretation
The first example of the elevation of subsequent agreements and subsequent practice is the over-emphasis of the idea, reflected in the title of article 31 of the Vienna Convention, that article 31 as a whole is the general rule. Draft Conclusion 2 is titled “General rule and means of treaty interpretation.” According to the first paragraph of the commentary, this title was chosen to signal two points:
“First, article 31, as a whole, is the ‘general rule’ of treaty interpretation. Second, articles 31 and 32 together list a number of ‘means of interpretation’, which shall (article 31) or may (article 32) be taken into account in the interpretation of treaties.’
The notion that the whole of article 31 of the Vienna Convention reflects a general rule is repeated at various places in the commentary to draft conclusion 2. Paragraph 3 of the commentary, for example, having recalled that the whole of article 31 is the general rule, refers to “the primary means of interpretation according to article 31, all of which are to be taken into account in the process of interpretation”.
The coup de grace, at least in the commentary to draft conclusion 2, can be found in paragraph 6 of the commentary. There, the Commission extols the importance of article 31 paragraph 1 and then proceeds to undermine its primacy:
“Article 31, paragraph 1, is the point of departure for any treaty interpretation according to the general rule contained in article 31 as a whole. The reference to it is intended to ensure the balance in the process of interpretation between an assessment of the terms of the treaty in their context and in the light of its object and purpose, on the one hand, and the considerations regarding subsequent agreements and subsequent practice in the present draft conclusions, on the other. The reiteration of article 31, paragraph 1, as a separate paragraph, is not, however, meant to suggest that this paragraph, and the means of interpretation mentioned therein, possess a primacy in substancewithin the context of article 31 itself.” (Emphasis added).
Thus, the title of article 31, “General Rule”, is used to justify a subtle elevation of subsequent agreement and subsequent practice, in the interpretation of treaties, so that article 31(1), which provides for the main rule of interpretation, namely that 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 to purpose”, no longer “possess a primacy” in interpretation. This elevation in subsequent agreements and subsequent practice is also advanced in the blurring of the line between modification and interpretation, which hints at the possibility (or at least makes it possible to conclude) that through the process of interpretation, subsequent agreements and subsequent practice can modify a treaty. It is this blurring of the line to which I now turn.
Blurring the Line between Interpretation and Modification
The elevation of subsequent agreements and subsequent practiceis also reflected in the Commission’s equivocation in relation to the effects of subsequent agreements and subsequent practicein the context of interpretation. Draft Conclusion 7 set out the possible effects of subsequent agreements and subsequent practice. In general terms, they “in interaction with other means of interpretation, contribute to the clarification of the meaning of the treaty.” (Draft Conclusion 7 paragraph 1). It explains that this may lead to the “narrowing, widening, or otherwise determining the range of possible interpretations”. (Id). The third paragraph concerns modification and not interpretation – so it falls outside the scope of the draft conclusions – and provides that it “is presumed that the parties to a treaty, by an agreement or a practice in the application of the treaty, intend to interpret the treaty”. There is nothing in these provisions that suggest an elevation of the subsequent agreements and subsequent practice in the interpretation of process.
The commentaries, however, might suggest something different. Although paragraph 3 of draft conclusion 7 foresees the possibility of modification through subsequent agreements, the commentary makes it clear that this is not in the context of interpretation (article 31), but rather in the application of article 39 of the Vienna Convention:
“The paragraph reminds the interpreter that agreements may serve to amend or modify a treaty, but that such subsequent agreements are subject to article 39 of the 1969 Vienna Convention and should be distinguished from subsequent agreements under article 31, paragraph 3 (a) (paragraph 21 of the commentary to draft conclusion 7).
Having made clear that the modification by interpretationis not permitted, the commentaries then engage in a slight backtracking. Paragraphs 23 to 31 of the commentaries contain very complex (and at times hard to follow) description of subsequent agreements and practice. At some places, the line between subsequent agreements and practice as modifiers of treaties and subsequent agreements and practice as tools for interpretation become blurred. For example, at paragraph 23, the commentaries state:
“Articles 31, paragraph 3 (a), and 39, if read together, demonstrate that agreements that the parties reach subsequently to the conclusion of a treaty can interpret and amend or modify the treaty.”
However, it is not clear why article 31(3) and article 39 of the Vienna Convention should be read together because they have nothing to do with each other. But more importantly, this obscure sentence makes it difficult to tell whether the commentary foresees an agreement modifying a treaty through interpretation (article 31(3)(a)). This uncertainty is reproduced elsewhere in the commentary. Indeed, reflecting the blurring of the lines, the commentary declares that “[i]t may sometimes be difficult to draw a distinction between agreements of the parties under a specific treaty provision that attributes binding force tosubsequent agreements, simple subsequent agreements under article 31, paragraph 3 (a), which are not binding as such, and, finally, agreements on the amendment or modification of a treaty under articles 39 to 41.” (paragraph 24 of the commentary to draft conclusion). The blurring of the lines is perhaps clearer in respect of subsequent practice. Paragraph 27, for example, makes the following conclusion:
“Indeed, the dividing line between the interpretation and the amendment or modification of a treaty is in practice sometimes ‘difficult, if not impossible, to fix’”.
As noted above, the blurring of the line between subsequent agreements and subsequent practice for the purposes of interpretation and as simple modifiers results in the possibility that, through interpretation, subsequent agreements and subsequent practice could be seen as modifying a treaty. Admittedly, the commentaries never make the point expressly because some members of the Commission had expressed the view that this would be incorrect. However, anyone seeking to justify the conclusion that modification of treaties through the use of subsequent agreements and subsequent practice in the interpretation of treatieswould certainly be able to profitably rely on these commentaries to make their case. In this sense, at least with respect to this point, the commentaries, rather than clarify the draft conclusions, simply add uncertainty.
A Comment on the Use of the Commentaries to Elevate Subsequent Agreements and Subsequent Practice in the Interpretation Process
First of all, it is true that article 31 is entitled “general rule”. This does not, however, justify the conclusion that paragraph 1 of article 31 does not “possess a primacy” in the interpretation process. A couple of things are worth point out about article 31. First, while, like the first paragraph, paragraph 3 of article 31 is phrased in the obligatory “shall”, the command is only to take into account subsequent agreements and subsequent practice. Good faith, ordinary meaning, context and object and purpose are, in contradistinction, not merely to be taken into account but are to be applied – that is the import of the phrase a “treaty shall be interpreted in good faith in accordance with the ordinary meaning of the terms in their context and in the light of its object and purpose.” It should not be forgotten that under article 31(3), it is not only subsequent agreements and subsequent practice that have to be taken to into account in the interpretation of treaties. Article 31(3) also provides that “relevant rules of international law” are also to be taken into account. Yet it can hardly be suggested that in the interpretation of treaties, other relevant (jus dispositivum ) rules of international law – that parties may have wished to modify – are to be given the same weight as ordinary meaning, context and object and purpose (article 31(1)). Elsewhere, I have made this following point:
“Art 31(3) constitutes part of the general rule in the sense that it contributes to, or facilitates, the determination of the meaning of the terms of the treaty, in good faith, in their context and taking into account the object and purpose of the treaty. It is for this reason that the injunction in Art 31(3) is for the interpreter to take into account subsequent agreements, subsequent practice, and relevant rules of international law, while there is an absolute obligation in Art 31(1) to give effect to the ordinary meaning of the terms, in their context and taking into account the object and purpose of the treaty.” [D Tladi “Interpretation of Treaties in an International Law-Friendly Framework: The Case of South Africa” in H Aust and G Nolte (eds.) The Interpretation of International Law by Domestic Courts: Uniformity, Diversity, Convergence (Oxford, 2016), at 145, footnote 48]
This view has been expressed in other reputable texts on the interpretation of treaties. Dörr, for example, while discussing article 31(3) under the general title “General Rule”, divides the elements of Art 31 into ‘general rule’, and ‘interpretive means additional to context’, including subsequent agreements and subsequent practice in the latter category i.e. reserving “general rule” for article 31(1) [O Dörr, ‘Article 31: General Rule of Interpretation’ in O Dörr and K Schmalenbach (eds), Vienna Convention on the Law of Treaties: A Commentary(Springer 2012) 521 et seq.]. In his view, interpretive means additional to context are meant “to establish the true meaning of the relevant terms of the treaty”[p. 553]. Even clearer, Sorel andBoré Eveno, having analysed decisions of the International Court of Justice and regional courts, have concluded that Art 31(1) provides the primary method of interpretation while Art 31(3) serves as a confirmatory purpose[JM Sorel and V Boré Eveno, ‘1969 Vienna Convention: Article 31: General Rule of Interpretation’ in O Corten and P Klein (eds), The Vienna Conventions on the Law of Treaties: A Commentary. Volume I (Oxford 2011), at 817-829].
The statement in the second paragraph of the commentary to draft conclusion 2 quoted above referring to “the primary means of interpretation according to article 31, all of which are to be taken into accountin the process of interpretation”, is thus incorrect. It is only the tools of interpretation in article 31(3) that are to “taken into account”. The injunction in article 31(1) is not merely to take into account but to give effect or to apply.
In a judgment delivered on 6 June 2018, Immunities and Criminal Proceedings (Equitorial Guinea v France), the International Court of Justice made observations consistent with the view that article 31(1) does have prominence:
“Pursuant to customary international law, as reflected in Articles 31 and 32 of the Vienna Convention on the Law of Treaties, the provisions of the Palermo Convention must be interpreted in good faith and in light of the object of purpose of the Convention. To confirm the meaning resulting from that process, or to remove ambiguity or obscurity, or to avoid a manifestly absurd or unreasonable result, recourse may be had to supplementary means.” [para. 91.]
In this extract, the Court sets out the general ruleof interpretation, namely article 31(1), and then the supplementary means of interpretation in article 32. That the Court does not mention article 31(3) is, in my view, a recognition that article 31(3) form part of the enquiry into article 31(1) in that its elements serve to facilitate a good faith interpretation of the ordinary meaning of the words of the treaty, in their context and in the light and object of the purpose of the treaty. This is what the Commission meant in the commentary to its 1966 Draft Articles on the Law of Treaties when (in the commentary to draft article 27) it spoke of a “single combined operation”. In fact the Commission is explicit about this:
“Equally, the opening phrase “There shall be taken into account together with the context” is designed to incorporate in paragraph 1 the elements of interpretation of set out in paragraph 3.” [Para. 8 of the commentary]
This description by the Commission in 1966 seems logical. It is confirmed by the approach of the International Court of Justice in perhaps one of the most important cases to address subsequent agreements and subsequent practice, namely the Kasikili/Sedudu case(Botswana/Nambia), ICJ Reports 1999, p 1045. In that case, the Court first sets out the rule of interpretation in article 31(1) [para. 18]. Although it refers also to article 31(2), it should be remembered that article 31(2) does not set forth a rule, but merely describes what “context”, used in article 31(1), refers to. The Court then, in an in-depth analysis of various factors, proceeds to establish the ordinary meaning of the terms in their context and in the light of the object and purpose of the treaty [para. 19-45]. It is interesting that the Court considers the travaux (article 32 of the Vienna Convention), before article 31(3). When the Court does (eventually) address article 31(3) it seems clear that this is done to confirm the meaning arrived at from the application of article 31(1) of the Vienna Convention [paras. 48-75].
Yet the Commission, in 2018, rejected the primacy of article 31(1) opting to suggest, even if implicitly, that article 31(3) had a value on its own, which value was equivalent to that of article 31(1). It may be said (and I am sure some in response to this post will suggest) that this not what the Commission meant by its emphasis of the title “General Rule” and that in fact what the Commission meant is what has been described in this post. That this would be incorrect can be illustrated by the vehement objection of some members of the Commission – Sir Michael Wood, Mr Georg Nolte (the Special Rapporteur) and Mr Sean D Murphy – to the suggestion to clarify this See (A/CN.4/SR.3444). Indeed the objection from these members was so vociferous that they would not even countenance a reference, in a footnote, to literature expressing a different view. In response to the arguments, for example, Wood said he “would be opposed to trying to accentuate what seemed to be an important point that should be stated clearly.” In response to the objection to even state that there were different views, I stated that the proposal to put forward a different view was
“merely an attempt to explain what was meant by the title of article 31 and to explain that there were others that held views that were different from the position of the Commission. From [my] perspective it would be a shame for the Commission to pretend that the view that it had expressed in the commentary was the only one. In seeking to hide the fact that there were alternative views on the matter, the Commission would give the impression that it was unsure of its arguments and did not want objective readers to be able to see different views and to make determinations for themselves.”