The Vienna Convention on the Law of Treaties (VCLT) was perhaps the most important product of the International Law Commission’s 75 years of existence to date. It has been thought finally and definitively to have settled the struggle between Sir Gerald Fitzmaurice and Sir Hersch Lauterpacht, both of them having served as Judges of the International Court of Justice, over how treaties should be interpreted. Lauterpacht believed that in order properly to interpret the substantive provisions of a treaty one should search for the treaty Parties’ intentions. Fitzmaurice, on the other hand, was convinced that the treaty Parties’ intentions should be understood as having been expressed in the treaty language those Parties had adopted, and therefore one should interpret the actual words of the treaty. It was Fitzmaurice’s view that was hailed as having prevailed over Lauterpacht’s when the VCLT was opened for signature in 1969 in Vienna.[1]
Indeed, Section 3 of the Convention, titled “Interpretation of treaties,” embraced Articles 31-33, which in fact were phrased in a hierarchical order of steps for interpreting the terms of a treaty. Article 31, titled “General rule of interpretation,” provides at 31(1) 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 and purpose.
Article 31(2) clarifies that “context…shall comprise, in addition to the text, its preamble and annexes”: the ensuing two items (Article 31(2)(a) and (b)):
(a) Any agreement relating to the treaty which has been 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.
Article 31 then added Sub-section 3: “There shall be taken into account, together with the context:” three further considerations, as additional means of arriving at an “ordinary meaning:”
(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.
Finally, Section 3 of the VCLT concluded with a Sub-section 4, which is not relevant to the subject of this blog, but for the sake of completeness should be mentioned, consisting of one short sentence: “A special meaning shall be given to a term if it is established that the parties so intended.”
The VCLT, having thus set forth in Article 31 the “General rule of interpretation” of treaties,” entirely geared to determining the “ordinary meaning” of the terms of treaties, then is followed by Article 32, titled “Supplementary means of interpretation,” which thus clearly is intended to be applied when an “ordinary meaning” cannot be determined by application of Article 31, including its descending order of the sources to be considered.
Article 32 first is notable for its absence of compulsion [contrary to Article 31(1)’s “A treaty shall be interpreted”] as it begins with “Recourse may [not shall] be had to supplementary means of interpretation,” which are somewhat loosely described as “including [while not necessarily excluding anything else] the preparatory work of [?]the treaty and the circumstances of its conclusion [thus including surrounding political, historical and sociological circumstances?],” and largely only “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.”
So far, so good! But Article 32 contains one more “Supplementary means of interpretation,” namely the purposive “in order to confirm the meaning resulting resulting from the application of Article 31.” If, however, the required “ordinary meaning” is established by application of Article 31, hence the result of applying Article 31 is not “ambiguous,” “obscure,” or “Leads to a result which is manifestly absurd or unreasonable,” why should there be any room to “confirm” that meaning? To that extent the hierarchy of the VCLT is disturbed! It seems that somehow the Lauterpachtian camel of “seek the intention of the parties” managed somehow to sneak its nose under the edge of Fitzmaurice’s “interpret the text” tent when all were convened in Vienna!
The intended hierarchical application of the VCLT in fact was potentially impaired by the procedure adopted essentially by all international courts and tribunals, as well as international arbitral tribunals, pursuant to which a claimant or applicant first serves its Memorial, followed by the respondent submitting its Counter-Memorial, then the Reply Memorial of the initiator of the proceeding, and so on. This has led to claimants and applicants putting forward in their opening Memorials all of their alternative arguments under both Articles 31 and 32. Why wait for the Reply Memorial to wheel Article 32 up to the firing line? As a result, adjudicators are confronted regularly, from the start, with a kind of “everything INCLUDING the kitchen sink” mélange of arguments employing both Articles 31 and 32, and all the individual provisions thereunder, and it is left up to them as to how to sort things out.
The result is that Investor-State arbitral tribunals have split between two approaches to application of those two Articles. More often than not, tribunals have correctly adhered to the hierarchical approach, stopping their interpretive analysis at Article 31 if they have established an “ordinary meaning” following its application, and eschewing any involvement of Article 32. They do as the International Court of Justice did in Case concerning the Arbitral Award of July 31, 1989 (Guinea-Bissau v.Senegal), Judgment of November 12, 1991, ICJ Rep. 1991, ¶ 48(emphasis added):
[T]he first duty of a tribunal which is called upon to interpret and apply the provisions of a treaty, is to endeavour to give effect to them in their natural and ordinary meaning in the context in which they occur. If the relevant words in their natural and ordinary meaning make sense in their context, that is an end of the matter. If, on the other hand, the words in their natural and ordinary meaning are ambiguous or lead to an unreasonable result, then, and then only, must the Court, by resort to other methods of interpretation, seek to ascertain what the parties really did mean when they these words. (quoting Competence of the General Assembly for the Admission of a State to the United Nations, Advisory Opinion, I.C.J. Reports 1950, p. 8.)
This strictly hierarchical approach has been followed by tribunals in Investor-State cases such as Asian Agricultural Products LTD (AAPL) v. Republic of Sri Lanka, ICSID Case No. ARB/87/3, Final Award, June 27, 1990, ¶40; Canfor Corporation and Others v. United States of America, UNCITRAL, Decision of Preliminary Question, June 6, 2006, ¶ 324; and Enron Creditors Recovery Corporation and Ponderosa Assets, L.P. v. Argentine Republic, ICSID Case No. ARB/01/3, Decision on Jurisdiction (Ancillary Claim), August 2, 2004, ¶ 32.
On the other hand, a number of arbitral tribunals, instead of following the hierarchical approach that was intended to be the essence of the VCLT, faced with the Parties before them having offered them the entire “stew” of Articles 31 and 32, have deviated into what is known as the “crucible approach” to those two articles, as the International Law Commission described it in 2 Yearbook of the International Law Commission 218 (1960):
All the various elements, as they were in any given case, would be thrown into the crucible, and their interaction would give the legally relevant interpretation.
Indeed, that approach has been encouraged by, for example, Richard K. Gardiner, Treaty Interpretation 10, 51 (Oxford University Press 2nd ed. 2015).
Most unfortunately, a significant, and prestigious, minority of Investor-State tribunals have adopted it. The author of this blog may be pardoned for noting that those cases included the following three cases in which he had been appointed by the investor to the tribunal and to the award in which he dissented, due to the failure of his colleagues to apply the VCLT hierarchically. They are Daimler Financial Services AG v. Argentine Republic, ICSID Case No. ARB/05/1, Award, August 11, 2012 (Dissenting Opinion of Judge Charles N. Brower, August 15, 2012, ¶¶ 18, 20-22); Austrian Airlines v. Slovak Republic, UNCITRAL, Ad Hoc, Final Award, October 9, 2009 (Separate Opinion of Judge Charles N. Brower to the Final Award, October 9, 2009, ¶ 4); and HICEE B.V. v. Slovak Republic, PCA Case No. 2009-11, Partial Award on Jurisdiction, May 23, 2011 (Dissenting Opinion of Charles N. Brower to the Partial Award on Jurisdiction, May 23, 2011), ¶¶ 5-42.
The fight goes on between those who apply the VCLT as it was designed and intended to be applied, and those, albeit fewer, Investor-State arbitrators who insist on the “crucible approach” in order to “justify” results that they prefer.
The promo code 20VCLT22 offers a 20% discount on the retail price of The Vienna Convention on the Law of Treaties in Investor-State Disputes: History, Evolution and Future in the Wolters Kluwer eStore. The code is valid through to 28 February 2023.
Judge Charles N. Brower, an Arbitrator Member of Twenty Essex Chambers, has sat as Judge ad hoc of the International Court of Justice in three active contentious proceedings (2014-2022), sits as Judge of the Iran-United States Claims Tribunal (1983-present), and has sat as Judge ad hoc of the Inter-American Court of Human Rights (1999-2002).
[1] This blog is based on a chapter entitled “Competing Theories of Treaty Interpretation and the Divided Application by Investor-State Tribunals of Articles 31 and 32 of the VCLT,” published in “The Vienna Convention on the Law of Treaties in Investor-State Disputes,” which chapter was co-authored by Judge Brower, Devin Bray and Pem Chhoden Tshering, all of whom would like to acknowledge and thank Asa Johannson and Ariana Talaie for their research assistance and comments on earlier drafts of this contribution.