4 Interpretation and modi fi cation
4.1 Treaty interpretation – the conventional debate
4.1.1 The dichotomy of terms and intent
In textbooks discussing treaty interpretation, one of the first things mentioned is the split between at least two basic views of what treaty interpretation is ‘all about’. It is ‘the question of precedence of the text of the treaty with its objective meaning as against true party intent, or of precedence of true party intent as against the text’.4 It is the age-old question of declared and true intent, between the objective and the subjective theories. Owing to the decentralised nature of
3 E.g.: Hans Kelsen, Reine Rechtslehre (2nd ed. 1960) 346–354 (Chs 45–47); Karl Larenz, Methoden- lehre der Rechtswissenschaft (6th ed. 1991). For international law: Maarten Bos, A methodology of international law (1984); Maarten Bos, Theory and practice of treaty interpretation, 27 Netherlands International Law Review (1980) 3–38, 135–170.
4 ‘die Frage nach dem Vorrang entweder des Vertragstextes in seiner objektiven Bedeutung vor dem wirklichen Parteiwillen oder aber des wahren Parteiwillens vor dem Text.’ Rudolf Bernhardt, Die Auslegung vửlkerrechtlicher Vertrọge insbesondere in der neueren Rechtsprechung internationaler Gerichte (1963) 15.
Uncertainty in International Law 88
international treaties, which sometimes closely resemble contracts, private law or even Roman law (ius civile) is the model which has traditionally been applied to treaty interpretation.5 The extent of the traditional debate on that point will be discussed now; we will leave a critique of that tradition to later sections.
In a sense, the view that the intentions of the parties are the ideal of treaty interpretation is congenial to international law.6 The traditional view is that all international law derives from the will of states.7 If a treaty is a meeting of wills and states will a treaty into ‘existence’, then the resultant text is a manifestation of that will. The will of the parties is on this view not only a logical moment. The will has content; therefore, that juncture of wills is an intent or intention. Here, the
‘party intention’ is the treaty an sich. Assuming all that, should not our cognition then aim for the ‘true’ treaty, i.e. party intent?8 This approach has been described as the ‘juridically natural view’.9 And some scholars believe that this view is generally agreed; Gerald Fitzmaurice argues that ‘no one seriously denies that the aim of treaty interpretation is to give effect to the intentions of the parties.’10
Granted that party intent is the aim of treaty interpretation – as was majority opinion before the Vienna Convention on the Law of Treaties 1969 – the argu- ment brought to bear by the opposing camp (the textualists) is that the text of the treaty can be the only reliable source for the joint intentions of the parties. Only the treaty text is signed, only its terms are agreed upon and only its terms the parties promise to abide by.11 The treaty text is specifically designed to express the intentions of the parties.12 Yet this traditional (moderate) textualism is not incon- sistent with a focus on intentions, because a categorical shift takes place behind the rhetoric against intentions. The moderate textualists switch their arguments to the epistemological plane, the means to reliably find out what the ontology looks
5 Bos (1980) supra note 3 at 15; Hersch Lauterpacht, Private law sources and analogies of international law (with special reference to international arbitration) (1927) 155–188.
6 Francis G. Jacobs, Varieties of approach to treaty interpretation: With special reference to the draft convention on the law of treaties before the Vienna diplomatic conference, 18 International and Comparative Law Quarterly (1969) 318–346.
7 The Case of the S.S. ‘Lotus’ (France v. Turkey), Judgment of 7 September 1927, PCIJ Series A No. 10 (1927) at 18: ‘The rules of law binding upon States therefore emanate from their own free will as expressed in conventions or by usages generally accepted as expressing principles of law.’
8 Gyửrgy Haraszti, Some fundamental problems of the law of treaties (1973) 28; Hersch Lauterpacht, Restrictive interpretation and the principle of effectiveness in the interpretation of treaties, 26 British Year Book of International Law (1949) 48–85 at 83.
9 Gerald G. Fitzmaurice, The law and procedure of the International Court of Justice: Treaty interpretation and certain other treaty points, 28 British Year Book of International Law 1951 (1952) 1–28 at 3.
10 Gerald G. Fitzmaurice, The law and procedure of the International Court of Justice 1951–4:
Treaty interpretation and other treaty points, 33 British Year Book of International Law 1957 (1958) 203–293 at 204; Heribert Franz Kửck, Vertragsinterpretation und Vertragsrechtskonven- tion. Zur Bedeutung der Artikel 31 und 32 der Wiener Vertragsrechtskonvention 1969 (1976) 26.
11 Bernhardt (1963) supra note 4 at 31.
12 Fitzmaurice (1958b) supra note 10 at 205; Arnold Duncan McNair, The law of treaties (1961) 365;
Ian Sinclair, The Vienna Convention on the Law of Treaties (2nd ed. 1984) 115, 141.
4.1.1 Interpretation and modification 89
like, yet both see ‘intentions’ as the ontology that forms the law and as the goal of interpretation. While both want to cognise party intentions, the textualists claim to have found ‘where those intentions are to be found, . . . where they are (primarily) to be looked for’.13
However, it is also said that the methods of interpretation are not at issue here.
We are told that we are being confused by detractors wishing to argue that it is all about whether recourse to travaux préparatoires is a ‘licit’ method of treaty interpretation.14 If one accepts that it is the goal or aim of treaty interpretation to find something beyond the text of the treaty and if one argues that that goal can only validly be reached by reference to the text, rather than extra-textual references such as travaux préparatoires – as the traditional textualists were forced to argue – then the issue is indeed about the ‘legitimacy’ of recourse to such other methods.
Strict subjectivists counter by assaulting the traditional formulation of the textualist position. The hypothesis is that a text has a meaning, not merely one given specifically to the term by the drafters, but a ‘plain’15 or ‘ordinary meaning’.
This is understood as a meaning customarily used by most speakers of a particu- lar language and as can be found in a dictionary. The corollary of that hypothesis is the assumption that each word has one meaning, or, at least, one meaning that can be fixed contextually. Also, if the drafters of a treaty used a word, they are presumed to have intended its ‘ordinary’ meaning, unless they specified a ‘special meaning’, such as termini technici by way of a purpose-built definition. Upon these assumptions rests the textualist theory that an interpreter needs to find the meaning of a text. The Vienna Convention is classically texualist. Article 31 sets the standard of ‘ordinary meaning’ as well as context, while keeping open the option that a term has a ‘special meaning’. The critique proceeds upon two main arguments which are interlinked. First, focusing on the ‘plainness’ of the meaning is putting the cart before the horse:
This rule [of ‘plain meaning’] seems pre-eminently reasonable. Its obviousness explains the frequency with which it is invoked. Its only – but, upon analysis, decisive – drawback is that it assumes as a fact what has still to be proved and that it proceeds not from the starting point of the inquiry but from what is normally the result of it.16
Thus, interpretation is seen as always necessary, however clear the words may sound:
If Article 3, according to the natural meaning of its terms, were really perfectly clear, it
13 Fitzmaurice (1952) supra note 9 at 4.
14 Bernhardt (1963) supra note 4 at 20; Fitzmaurice (1958b) supra note 10 at 206; Sinclair (1984) supra note 12 at 116.
15 McNair (1961) supra note 12 at 364–382.
16 Hersch Lauterpacht, The doctrine of plain meaning, in: Elihu Lauterpacht (ed.), International law. Being the collected papers of Hersch Lauterpacht (1978) Volume 4, 393–403 at 396.
Uncertainty in International Law 90
would be hardly admissible to endeavour to find an interpretation other than that which flows from the natural meaning of its terms. But I do not see how it is possible to say that an article of a convention is clear until the subject and aim of the convention have been ascertained, for the article only assumes its true import in this convention and in relation thereto.17
The opponents of the textual view deny that there can be such a thing as ‘ordin- ary meaning’. For them, words have no fixed meaning that need only be extracted;
words can portray the intentions of the parties only imperfectly.18 Indeed, ordin- ary meaning is disqualified as an impossible concept. ‘Language constantly varies and meanings may fluctuate from one age to the next.’19 Language contains
‘subjective elements’20 and there is doubt as to the objective reality of the ‘mean- ing’ or of the possibility of its reliable ascertainment.21 If followed, this line of argument would lead to the conclusion that ascribing a meaning to a text is impossible, because, in the end, no such thing can exist. This, however, would
‘lead to a denial of the very possibility of verbal communication’22 – a radical sceptic’s view which negates cognition.
However, a general term like ‘party intentions’ cannot fare better, like any concept that needs a definition. Using ‘intentions’ assumes what has yet to be established, namely that the text does not adequately reflect intent.23 Further, one may question what ‘intention’ is. Is it the intention of a party or a joint intention?
Surely the latter, subjectivists interject, because treaties are made by a meeting of wills, not by unilateral will. If common intentions are sought, what do they look like? Do the original intentions have to ‘cover’ every possibility that may arise – as otherwise there is no law on the issue? Hersch Lauterpacht points out that there may in fact be no common intention of the parties on a certain issue, inter alia because they simply did not intend the same result or were merely seeking a dilatory compromise formula.24 ‘[T]reaties . . . are often a political substitute for
17 Interpretation of the 1919 Convention Concerning Employment of Women During the Night, Advisory Opinion of 15 November 1932, PCIJ Series A/B No. 50 (1932), dissenting opinon Anzilotti at 383. It needs to be stressed that in this opinion Judge Anzilotti primarily constructs the meaning from the context of the article in question, despite his allusions to the subjective and teleological view.
His references to travaux préparatoires come only after he has found sufficient reasons for supporting his dissenting view in the context, and only to support the conclusion previously reached. Only his rhetoric is subjectivist, not his argument.
18 Bernhardt (1963) supra note 4 at 16.
19 Bos (1980) supra note 3 at 149.
20 Fitzmaurice (1952) supra note 9 at 2 (FN 1).
21 Such epistemological doubts are hedged by: Illmar Tammelo, Treaty interpretation and practical reason. Towards a general theory of legal interpretation (1967) 6; Myres S. McDougal, The International Law Commision’s draft articles upon interpretation: Textuality redivivus, 61 American Journal of International Law (1967) 992–1000 at 997.
22 Jacobs (1969) supra note 6 at 340.
23 Fitzmaurice (1958b) supra note 10 at 205.
24 Lauterpacht (1949) supra note 8 at 76–78. The phrase ‘dilatorischer Formelkompromiò’
was apparently coined by Carl Schmitt, e.g. Carl Schmitt, Legalitọt und Legitimitọt (1932) 91.
4.1.1 Interpretation and modification 91
rather than a legal expression of the agreement of the parties.’25 Also, how is common will, the will of a majority of persons, possible? This is a problem in municipal legislation where hundreds of members of parliament form some sort of will to enact a statute (or do not, but merely vote on command), but it becomes exacerbated in treaty-making, where there is a multitude of factions within the state party (the actual drafters from the foreign ministry, the members of parlia- ment agreeing to a signed treaty, the head of state ratifying it) and a multitude of state parties seeking to form a common intention.26 Could one say that this conglomerate of ‘states of mind’ is either capable of forming a will to enact or – much more dangerous to the subjectivist cause – that this multitude forms intentions as to the content of the treaty law independently of the text? This can be doubted. To seek to cut the Gordian knot and to ‘presume – to imply – intention is to predicate that intention does not matter’.27