Running in Circles: A Comment on Bjorge’s Evolutionary Interpretation of Treaties

It was a pleasure to read Eirik Bjorge’s The Evolutionary Interpretation of Treaties. The book is well written and exceptionally well researched. Eirik demonstrates nothing less than an encyclopedic knowledge of the relevant case law and scholarship, and has seemingly read every single bit of text that the International Law Commission and its rapporteurs have produced on the question of interpretation. Eirik’s book is beyond question the most comprehensive examination to date of the issue of evolutionary treaty interpretation, and it fills an important gap in the literature.

While the virtues of the book are many, I cannot help but feel that, had Eirik chosen a different methodological path, the book could have been significantly more illuminating with regard to the nature of the phenomenon of evolutionary interpretation. This is not because I take issue with the main thrust of Eirik’s argument, namely that evolutionary interpretation is perfectly compatible with the rules of interpretation set out in Articles 31-33 of the Vienna Convention on the Law of Treaties. Yes, it is – at least partly because the Vienna ‘rules’ are so broad and flexible that one can do (almost) whatever one wants with them.

I do, however, take issue with Eirik’s argument that evolutionary interpretation is entirely explainable by reference to the concept of the common intention of the parties, and at that a particular objective conception of that concept, which is not about the actual mental state of any individual or group of individuals representing their respective states when a given treaty was concluded. Nor does Eirik by such an objective conception mean the objective indicia for proving an otherwise unprovable subjective mental state, as e.g. found in the travaux préparatoires. Rather, as Eirik explains in Chapter 3 of the book (which is its argumentative core and is more or less self-contained), by ‘intention of the parties’ he means ‘the result which one reaches if the general rule of interpretation is applied correctly.’ The intention is, in other words, the outcome of the interpretative process, rather than a factor to be taken into account in that process (p. 63). Once the ‘intention of the parties’ is defined in this particular way, then the problem of evolutionary interpretation is, like any interpretative problem, definitionally one of establishing the ‘intention of the parties,’ i.e. of correctly applying the admissible means of interpretation to get to the right result.

This is not, in my view, a methodologically satisfactory approach, not necessarily because it is incorrect as such, but because it has little explanatory power of its own. First, it is obviously circular – the task of (evolutionary) interpretation is to establish the intention of the parties, defined as the correct result one reaches in applying all admissible tools of interpretation. This circularity is present not just in this definition, but in the structure of the whole argument of Eirik’s book – whenever a particular interpretative problem is raised, the book ultimately ends with the intention of the parties (as it must), often followed by a barrage of quotations from ICJ judgments or ILC reports invoking the intention of the parties.

Second, the use of the intention of the parties concept in this particular way (i.e. as the correct outcome of the whole interpretative process) is potentially very confusing, because other people (including judges, ILC members, and scholars being quoted and cited) may use the same words, but not in this particular way. In other words, that everybody talks about the ‘intention of the parties’ does not mean that they are actually talking about the same thing.

In my admittedly anecdotal experience, lawyers as a profession can be incredibly sloppy in how they talk and think about intention, often seamlessly moving from one conception of intention to another without even realizing it. Even Eirik himself at least at one point in the book switches from one conception of intention to another, at pp. 61-62, where he discusses Dworkin’s thesis (as adopted by Letsas) that intentions can exist at different levels of abstraction. (This is by the way just another articulation of an important insight from analytical philosophy, pioneered by Elizabeth Anscombe in the 1950s, that actions are intended under some description, but not other descriptions – see here for a short primer). Thus, for example, in voting for an equality law the same legislator can intend to generally prohibit discrimination on grounds of sexual orientation, while not intending that this ban require access to marriage for same sex couples. Crucially, intention in this sense is in fact a subjective psychological state, not the objective outcome of some interpretative process, as Eirik otherwise defines it.

Third, while he invokes many, often widely different, individual instances, Eirik does not spend much time on defining what ‘evolutionary interpretation’ actually is. He only does so in a single paragraph on p. 59, where he says that ‘evolutionary interpretation’ includes ‘situations in which an international court or Tribunal concludes that a treaty term is capable of evolving, that it is not fixed once and for all … a situation where account is taken of the meaning acquired by the treaty terms when the treaty is applied.’ The key concepts deployed here (terms, evolution, fixation, meaning, application) are not themselves defined and are capable of multiple definitions. I am not trying to be facetious here, but one cannot meaningfully talk about ‘meaning’ without clearly explaining what one means by ‘meaning’ (for instance, semantic meaning in conventional language).

Similarly, some kind of typology of evolutionary interpretation would have been useful. For example, we frequently encounter situations (which normally pass without much objection), in which the interpreter applies an old rule to a completely novel problem that was manifestly beyond the grasp of the rule’s drafters: e.g. is the interception of email communication or the seizure by police agents of data on the hard-drive of one’s computer an ‘interference’ with the private life of that person within the meaning of Art. 8 ECHR or a ‘search and seizure’ within the meaning of the Fourth Amendment to the US Constitution. Then we have more controversial situations in which the interpreter examines an issue at time point A, but re-examines the decision at later time point B, and says that A was wrongly decided: for instance, racial segregation and the US Constitution (Plessy v. Ferguson and Brown v. Board). Finally, and perhaps most interestingly, we have situations in which, just like in the example above, the interpreter reaches two different decisions on the same issue at times A and B. However, the interpreter does so by saying that while the decision A was correct at the time, it is no longer correct today: think, for example, of the rights of transsexuals under the ECHR (Cossey and Goodwin). Are all of these examples ‘evolutionary interpretation’ properly so called? Are they conceptually the same or are they significantly different? What exactly ‘evolves’ here, and on what basis? And when and why should this evolution happen? Answering these questions requires more, I think, than references to the intention of the parties.

Fourth, somewhat vexingly, Eirik is himself completely aware that his definition of the ‘intention of the parties’ is circular. He thus defends his adoption of this particular conception of intention, on which the entire argument then turns, by saying (at p. 63) that while ‘one may criticize this circular definition, it should nonetheless be retained, if for no other reason that it is also the best definition which the ILC has been able to produce.’

What I fail to understand is why we as scholars trying to better appreciate a problem should be at all content with whatever the ILC has been able to produce, be it in the 1960s or today. I fully accept Eirik’s point (at p. 20) that he wants to stick to the authoritative, conventional grammar of international law, as expressed in Arts. 31-33 VCLT. But that grammar might simply not be enough to enable us to think truly deeply about problems as complex as is evolutionary interpretation. There is a whole conceptual universe out there that could potentially help us understand interpretation, be it dynamic or static. Any kind of intellectual straitjacketing to the small world of the VCLT and the ILC’s work on the topic is, at best, counterproductive. At worst, it is conducive of an insufficiently critical appraisal of the work product of the ILC or the ICJ, as somehow always managing to confirm the broader ‘intention of the parties’ narrative, even when that work product is conceptually manifestly rudimentary or confused (see, e.g., here for my early take on the ICJ’s Costa Rica v. Nicaragua judgment).

Finally, in that regard, the topic of evolutionary interpretation is a particularly fertile ground for interdisciplinary engagement. Our understanding of that process could be enriched by insights from the philosophy of language, legal philosophy, and linguistics – for example, with regard to the role of intention or perceived intention of the speaker in language and human communication more generally. We could also draw inspiration from municipal legal theorists, who have expended an enormous amount of time and sheer brainpower in trying to answer the same or very similar questions. Think, for instance, of UK scholars grappling with the elusive divide between interpretation and modification in the context of s. 3 of the Human Rights Act 1998, which requires ECHR-compatible interpretations to be given to statutes as far as it is possible to do so (what, then, are the limits of the ‘possible’? – see cases like Ghaidan v. Godin-Mendoza or the scholarship of Aileen Kavanagh).

Or think, most relevantly, of the mountains of scholarship produced by American lawyers over decades of debates between many different varieties of ‘originalism’ and ‘living constitutionalism,’ in a system whose foundational document have been subjected to two centuries of (different forms of) evolutionary interpretation. Consider, for instance, conceptual distinctions drawn between interpretation and construction, between the original meaning and the expected original applications of a provision, and between (fixed) original semantic meaning and (evolving) constructed and applied legal meaning, drawn by eminent scholars such as Larry Solum and Jack Balkin. There is much that we international lawyers can learn from others; there are many ways in which their experiences can enrich our own when it comes to a topic as deep as interpretation.

In conclusion, while I again take no issue with Eirik’s argument that evolutionary interpretation (however exactly defined) can be compatible with Arts. 31-33 VCLT, the explanatory power of his book is significantly diminished by its strict focus on the VCLT and the ‘intention of the parties’ as its idée fixe. Whether, for example, ten or twenty years from now the European Court decides to overturn Schalk and Kopf v. Austria and holds that the legal recognition of gay marriage is required by the Convention, depends only to a very limited extent on a proper application of the VCLT ‘rules.’ The European Court’s discretion is constrained by the VCLT no more than the US Supreme Court’s in deciding the same question under the US Constitution (which it will likely do soon). The constraints, such as they are, come from somewhere else.

I hope that this critique does not come across as overly harsh. Again, the book is quite excellent in doing what it sets out to do, and it would be unfair to criticize Eirik for writing the book that he wanted to write, rather than a book that I would have wanted him to write. A strict focus is also perfectly understandable and indeed desirable in a PhD thesis. Nonetheless, this book really cries out for a sequel, and I do hope that Eirik will write one.