Painting “Hampton Court Green” by Camille Pissarro via Wikimedia Commons. Cover courtesy of Cambridge University Press.

Back to Symposium

Reflections on Customary International Law and Interpretation


It is not often that one reads about Schrödinger’s cat, the particle and wave qualities of light, and Latour’s idea that modern discourses are always driven by their foundational contradictions in a single book, let alone one on law. Indeed, if one thing emerges with certainty from this edited volume, it is that customary law continues to capture the imagination of international lawyers. One is even tempted to say that customary law itself often serves as a platform for reflecting on broader problems of legal theory and philosophy.

How does one then engage with a book of such immense breadth and depth? When I was kindly asked to contribute to this symposium, I initially contemplated that I would converse with one author or speak to the volume’s overall qualities. Yet after doing my reading, I found it more appropriate to escape both classic formats of either drilling in on a single chapter or offering overly general comments on the quality of the book as a whole (spoiler: it’s very good). Instead, I thought I would briefly engage with certain ideas and problems that have given me some pause throughout the years, issues that I, perhaps arbitrarily, consider to lie at the heart of what still makes custom hard to theorize, and at the same time worth-while exploring. My comments are thus themselves tentative and in the spirit of scholarly dialogue, rather than featuring definitive answers or perfectly crystallized viewpoints.

I hereby offer three reflections on specific book chapters, plus two broader ones that cut across more than one chapters. I start with my three chapter-specific reflections:

An ‘Empowerment Norm’? (Kammerhofer)

In his contribution, Jörg Kammerhofer (Ch. 1, pp. 3-28) seems to suggest that customary international law needs an ‘empowerment norm’ before it can actually lead to the creation of legal norms (an ‘empowerment norm’ being a separate positive norm that explicitly grants law-making power to custom). I am not sure this is the most accurate way to conceptualize things from an analytical point of view. In my mind, customary law is not mere custom (meaning habit or usage) that happens to meet the requirements of an empowerment norm. Rather, normativity, and even an intention of law-making (or at least law-abidance), should be understood as an inherent element of the very nature customary law, perhaps even a definitional one. Other norms may recognize this quality of customary law and incorporate it into the list of applicable laws within a given legal order, (as one could argue is the case with international law), but this is not the same as saying that custom is law because of an empowerment norm, as in that an empowerment norm that exists outside of and separately from custom is constitutive of its normativity for all intents and purposes. At this point, I wonder whether we can obtain a better analytical picture by drawing some inspiration from how social ontologists, such as John Searle, and more recently Brian Epstein, have conceptualized dependency relationships and institutional facts. In this regard, Epstein’s distinction between ontological ‘grounding’ and ‘anchoring’ may help disambiguate at least part of the puzzle – a normative intention could be seen as a grounding condition for the making of and existence of customary law, whereas what makes customary law visible or applicable in a particular legal setting may indeed have to be traced back to a separate set of institutional facts, involving some sort of ruled that is backed by collective intentionality.

Custom’s ‘Birth-certificate’ (d’Aspremont)

For his part, Jean d’ Aspremont (Ch. 2, pp. 29-39) critiques the assumption that customary law is based on social reality by zeroing in on the apparent impossibility of pinning down the custom-making moment to a specific time in the past. As he argues, ‘CIL could not be upheld as being upheld in social reality if it could not be presumed as being made at a certain moment in the past’. I admit that I found this part of d’ Aspremont’s critique unconvincing. Surely, social (as well as psychological) reality exists regardless of our very human need of seeing the world through a clean before/after divide. For example, just because you cannot pinpoint a specific moment in the past when disco music became popular, this does not mean that it was never popular. And even if by convention we pin down a culminating point for the rise of disco music –such as the release and success of a popular album– this would say more about our intention to impose order and boundaries on social reality than about that reality itself. Similarly, those arguing about the existence of a customary norm should not have to produce its birth-certificate, not necessarily because that would be impossible, but because this is not how we judge the validity of claims about the existence of social reality in general.

The Role of Empirical Data in Analytical Theories about Custom (Merkouris)

For Panos Merkouris I have a methodological question. Much of Panos’ chapter (Ch. 16, pp. 347-369) is devoted to producing evidence from international and national courts to show that the notion that customary international law is accepted in practice. It is indeed very insightful to witness how these often seemingly abstract debates about custom and interpretation have been translated in the work of judges and lawyers. Yet I wonder, is it necessary to produce all this empirical evidence from an analytical point of view? Isn’t the possibility of interpreting custom something that has to be established at a level of theoretical analysis, or does it need to be accompanied with practice to that effect? If the latter is the case, I wonder what the value of such practice would be if it pointed to the opposite direction. In other words, what would the verdict be in a scenario where the theory suggested that we can interpret custom, but there was a well-established practice that we cannot, or, at least, will not? Conversely, it would be interesting to also hear from the opposite side of this debate. What sense do those that vehemently deny the possibility that we can interpret customary norms make of this practice of interpretation –is this practice misguided and should thus be disregarded in toto, or could it have an effect on how scholarship should understand and present this matter?

Next to these chapter-specific comments, I have two observations that cut across different contributions.

Beware of Analogizing Custom!

The first and perhaps most straightforward observation concerns a reliance by many authors, either explicitly or implicitly, on an analogy between customary law and treaties or legislation, regarding how we visualize the process of their creation as well as the form of what has been created. The often-used analogy between customary international law and oral agreements is one such example. The expectation that there must be ‘a’ moment in which customary law is created, mirroring the moment when an agreement or a piece of legislation enters into force, is another such case. While such analogies have certain explanatory power, they can also lead to doctrinal conundrums when taken as revealing deeper analytical truths about custom. Perhaps the better known such conundrum concerns our paradoxical theories surrounding custom’s birth. An equally problematic, but much less studied aspect of this problem relates to our issues with individuating the various norms of custom. Implicitly or explicitly analogizing norms of customary law to articles of a treaty or a piece of legislation tends to (a) give immense power to those in position to articulate the level of generality at which putative rules are pitched (it is harder to prove a customary norm on head of state immunity before international criminal courts as opposed to a respective norm before criminal courts in general; same with a norm prohibiting nuclear weapons specifically as opposed to a more general class of weapons etc); and (b) obfuscate analytical thinking about the very real question of what is indeed the most appropriate level of generality. Analogizing custom to other sources of law may be helpful for pedagogical purposes, including for explaining things to ourselves, but should be seen, in my mind at least, as a provisional theoretical construct rather than an accurate or complete reflection of customary law.

On Interpretability, Philosophically Speaking

The second general point that I would like to raise is more tentative and bears upon the very idea that customary law is ‘interpretable’. This is a term that I have also employed in my own work, yet I have come to question its accuracy from a purely philosophical perspective. Not to be too Aristotelian, but I wonder whether interpretability is an inherent feature of customary norms themselves, as it is often talked about, or whether it is human beings that have an innate capacity to interpret the world around them, including their own behavior and the norms that may come out of it. This might seem a trivial point of semantics, but it may be worth noting given the increasing relevance of AI for applying all sorts of rules. As ChatGPT – everyone’s go-to AI chatbot these days– readily explains, it is not actually interpreting anything when answering questions that are posed to it; rather its responses are ‘always based on statistical patterns rather than true understanding or subjective interpretation’. Indeed, this way of providing answers through analyzing statistical patterns seems to be quite distinct from how human beings produce and engage with abstract (not just open-textured, but abstract), norm-generative concepts (such as friendship, love, or sovereign equality) that lie at the foundation of many types of social interaction and attempts to regulate it, including custom itself. Thus, while it is technically possible to answer all sorts of questions about rules and their application without engaging in interpretation, this should arguably lead us to a discussion of what types of thinking and processes we would like our norm application to involve rather than an investigation of ‘who does it better’. In that sense, the debate surrounding the interpretability of norms, including customary ones, could as well be a discussion about the added value that human nature can bring to the application of these norms via interpretation rather than simply whether they are inherently interpretable or not.

Orfeas Chasapis Tassinis

Dr Orfeas Chasapis Tassinis is Research Fellow at Gonville & Caius College at the University of Cambridge and Fellow at the Lauterpacht Centre for International Law

View profile
Print article

Leave a Reply

We very much welcome your engagement with posts via the comment function but you do so as a guest on our platform. Please note that comments are not published instantly but are reviewed by the Editorial Team to help keep our blog a safe place of constructive engagement for everybody. We expect comments to engage with the arguments of the corresponding blog post and to be free of ad hominem remarks. We reserve the right to withhold the publication of abusive or defamatory comments or comments that constitute hate speech, as well as spam and comments without connection to the respective post.

Submit your Contribution
We welcome contributions on all topics relating to international law and international legal thought. Please take our Directions for Authors and/or Guidelines for Reviews into account.You can send us your text, or get in touch with a preliminary inquiry at:
Subscribe to the Blog
Subscribe to stay informed via e-mail about new posts published on Völkerrechtsblog and enter your e-mail address below.