07 Oct On the Interpretability of Customary International Law: A Response to Nina Mileva and Marina Fortuna
[Odile Ammann is a postdoctoral researcher at the University of Zurich, Switzerland.]
“Customary international law cannot be interpreted because it’s not written.” I have heard this objection many times, including from the most seasoned international lawyers. While the interpretability of customary international law (CIL) may seem less obvious than that of written laws, I do not think that the written or unwritten character of a legal act impacts its interpretability. Nina Mileva and Marina Fortuna’s recent post, which forcefully highlights the interpretability of CIL, is thus more needed than one might initially think.
I have looked into the interpretation of CIL in my own doctoral research, which focused on international law in domestic courts and which was supervised by Samantha Besson. The monograph resulting from this reflection will be published at the end of this year. Given that Mileva and Fortuna’s contribution speaks to and echoes my own work, I would like to take this opportunity to comment on some of the important aspects they have raised.
What is interpretation in international law?
Mileva and Fortuna write that “interpretation is the process through which the interpreter attempts to determine the true meaning of the rule that is being interpreted,” and that “most cases brought before international courts and tribunals” involve interpretative questions. I would take this even further and argue that interpretation is always at play in international (and domestic) dispute settlement. Law application is not possible without interpretation, no matter how modest the interpretative act may seem in some cases. As Andrei Marmor notes, interpretation is the “ascription of meaning to an object.” Moreover, interpretation refers both to a process and to the outcome of this process. Whether there is a “true” meaning (or, to use Ronald Dworkin’s famous expression, a “one right answer”) is a different matter. Interpretation is, after all, a social construction, and different interpreters can reach different and incompatible, yet equally compelling interpretative outcomes.
Mileva and Fortuna are right to call for greater rigor in the interpretative process. Interpretation is not a “dance floor” where literally every move is permitted, as Jean d’Aspremont humorously observes. Indeed, interpretation in international law (just like interpretation in any other legal and non-legal field, including in medicine, music, and literature) is subject to specific constraints, no matter how loose or demanding these constraints are. In other words, interpretation takes place (recte: must take place) within a set of methods that are defined by the respective interpretative field.
All too often, international lawyers and courts treat interpretative methods, and especially the methods of art. 31–32 of the Vienna Convention on the Law of Treaties (VCLT), as a convenient set of guidelines that can be twisted at will. Some authors even argue that these methods are not mandatory. Yet, first, following the interpretative methods of international law is a corollary of States’ international legal obligations. Second, in the case of art. 31–32 VCLT, there is widespread agreement that these methods are customary. Third, why would States have adopted provisions on the methods of treaty interpretation if they thought that these methods were not important?
Following Timothy Endicott, a method is “a way of doing something”; thus, methods of legal interpretation are ways of interpreting laws. In international law, interpretative methods are prescribed by specific norms. The constraints established by these norms are precisely that: constraints. These basic methods are not just “nice to have,” as many (including States, of course) would like to believe. If States do not comply with them, the world will of course not collapse. But in such a case, States need to face that they are doing something that does not qualify as interpreting international law.
The fact that these constraints exist and that they must be observed does not mean that all interpreters of international law will (and must) arrive at the same interpretative conclusion by using these methods. Interpretative methods create, to use Frederick Schauer’s (and, originally, Hans Kelsen’s) telling metaphor, “a frame without a picture.” While they constrain the interpretative process and, therefore, its outcome, they do not predetermine the interpretative result. Interpretative methods often point in different directions and need to be weighed against each other, so that an interpretative conclusion can be extracted from their interplay. Importantly, they create a frame inside of which meaningful communication between various interpreters becomes possible. If we do not even use the same set of tools to interpret international law, then it is unlikely that we will ever find common ground. By contrast, if at least our basic tools are the same, a conversation (and, potentially, an agreement) becomes possible.
Interpretative norms and principles
Instead of talking about “rules of interpretation,” I would use the more generic term of “interpretative norms” to describe normative provisions that require interpreters to use specific methods. Indeed, although rules, like every legal act, are open-textured and can thus become vague in particular cases, they allow for deductive reasoning once they have been interpreted. The notion of “rules” seems apposite when we need to resolve conflicts, for instance, by using conflict rules like lex posterior and lex specialis, or when we rely on logic through rules of logic like ejusdem generis.
By contrast, norms that prescribe the use of interpretative methods can seldom (if at all) be applied deductively. Instead, norms requiring the use of specific methods point to relevant features of the interpretandum that decision-makers must consider, and they require inductive reasoning (e.g. to determine the context or history of a legal act). Interpretative methods show in which direction we should be going, but they do not constrain us completely.
I would also use this terminology (“norms”) when talking about the interpretandum (e.g., a paragraph in a treaty or a practice that States have been carrying out for decades), as this makes it possible to capture the full set of interpretative material that courts and other interpreters of international law encounter. Strictly speaking and from a jurisprudential perspective, however, the norm is the result of the interpretation, and not its object. Following H.L.A. Hart, the interpretative object is not a norm, but a social fact. The normative component is found in the conclusion that the court eventually reaches. This distinction is acknowledged by Tullio Treves: “Customary rules are the result of a process (…) through which elements of fact, empirically verifiable, acquire a legal character thus creating rights and obligations for the subjects of international law.” While it would be cumbersome to be perfectly consistent about this terminology in the context of scholarship, I believe that it is worth making this clarificatory point.
The interpretability and interpretation of CIL
As Mileva and Fortuna highlight, Treves argues that custom cannot be interpreted. “Even though language is necessary to communicate their content,” he writes, “expression through language is not an indispensable element of customary international law rules. The irrelevance of linguistic expression excludes interpretation as a necessary operation in order to apply them.” Like Treves, many scholars are reluctant to refer to the “interpretation” of CIL, presumably because custom is vaguer than written law. Moreover, for unwritten law, establishing the existence of the law plays a more important role than for written law, the sheer existence of which usually seems evident because written laws are fixed on a physical medium. Thus, one may think that unwritten law needs to be identified, but not interpreted. Another reason for this reluctance to talk about “interpretation” in relation to custom is that this activity is (erroneously) associated with textual material.
Yet interpretation, qua inquiry into the meaning of a legal act, allows us to make sense of written and unwritten laws alike: when we interpret the law, we interpret social facts, and these facts may be verbal or non-verbal. Moreover, all laws, written or unwritten, leave room for indeterminacy and, therefore, for interpretation.
What Treves rightly acknowledges is that CIL is often closely related (or, as he writes, stands in a “strict relationship”) with texts. Indeed, it suffices to look at the ILC’s draft conclusions on custom to realize that most of the elements of determination of State practice and opinio juris (including State conduct) will eventually be fixed on a written medium (see e.g. draft conclusions 6 and 10). This means that to identify CIL, interpreters will almost always resort to textual interpretation. To grasp the meaning of the interpretative object, they will also need to use historical, contextual, and teleological interpretation.
To respond to Maarten Bos’ claim that the interpretation of CIL can be reduced to (and exhausts itself in) its identification, it is helpful to rely on d’Aspremont’s exploration of the distinction between law ascertainment and content determination. Law ascertainment is the act by which a legal norm is identified; content determination consists in defining the meaning of this norm. Even if they are often intermingled and complementary in practice, these two operations are conceptually distinct. Importantly, the entanglement of law ascertainment and content determination in practice does not allow us to conclude that CIL cannot be interpreted. Law ascertainment and content determination go hand in hand, yes, but interpretation is always necessary. Two interpreters may agree about the existence of a norm of CIL on a specific matter, but they may disagree about the meaning of this norm in a specific case.
A lack of interpretative guidance on CIL?
Mileva and Fortuna write that while treaty interpretation is governed by the VCLT, “there is currently no such guidance” for CIL. To argue that the interpretation of CIL (and of unwritten international law more generally) is understudied is correct. Nonetheless, I believe that there is such guidance on the interpretation of CIL: last year, the ILC issued its draft conclusions on identification of CIL (for an early analysis of the work of the Special Rapporteur, Sir Michael Wood, see Besson and Ammann 2016).
Of course, one could argue that the ILC’s draft conclusions cannot (yet) be deemed customary and, therefore, mandatory. This objection is legitimate, as it remains to be seen whether the draft conclusions will be accepted by States on the long run. Still, the ILC’s work is the most recent and comprehensive statement of the principles guiding the interpretation of CIL. The draft conclusions were adopted in a universal and (relatively) inclusive forum, even if this forum is still imperfect, e.g. from the perspective of gender equality (see United Nations Office of Legal Affairs, Seven Women in Seventy Years, New York, 24 May 2018) and ethnic diversity (taken together, European States – including Russia and Turkey – and the United States occupy 12 of the 34 seats of the ILC, while Asia has 5 seats, the Middle East 2, Africa 8, and Latin America 7).
What needs to be stressed, however, is that the Special Rapporteur on custom, Sir Michael Wood, explicitly requested States to provide comments on previous versions of the draft conclusions and on their own domestic practice on CIL. Many States followed up on this request (the Swiss practice, for instance, has been set out in detail in Besson and Ammann 2016). Of course, States with more resources will have an advantage in terms of their ability to contribute to this process. Moreover, as already mentioned, some uncertainty inevitably remains at this stage as far as the long-term acceptance of the draft conclusions is concerned. Still, the odds are high that the ILC’s draft conclusions will become well-established standards, and that these standards will provide not “rules,” but “authoritative guidance for those called upon to identify customary international law, including national and international judges,” as the ILC intended (see here, para 4).
I would argue that the ILC’s draft conclusion on custom encapsulate the same four basic methods as those that govern treaty interpretation (and, one can add, the interpretation of domestic law): the textual, contextual, teleological, and historical method. Some passages in Mileva and Fortuna’s text suggest that this is indeed the case. It is important to mention in passing that my use of the term of “interpretative methods” may not coincide with the understanding that other international lawyers have of this term. This is what can make research on interpretation so frustrating: while we think that we are discussing the same concept, some of us are talking about rules of logic, others about maxims, yet others about normative interpretative theories, etc. Admittedly, the ILC does not spell out these four methods in an explicit fashion. As Mileva and Fortuna note, the draft conclusions are centered on the constitutive elements of CIL, that is, State practice and opinio juris. As the two authors point out, these constitutive elements are not methods (or, as they put it, “rules of interpretation”) as such: they only represent the basic material that we (and courts) need to rely on in order to identify a CIL norm.
I would not attribute too much importance to the terminology courts use in relation to interpretation. The fact that they do (not) use the word “interpretation” is, in my opinion, not decisive, including when it comes to determining whether they are doing a stringent job from the perspective of interpretative methods and reasoning. Much more important, I think, are the interpretative tools they refer to and actually put to use. This is where international legal scholars come into play: they can (and should) hold courts accountable by requiring more precision and rigor in these courts’ interpretative reasoning. This stringency is essential for the rule of international law to be preserved.
State practice and opinio juris and, therefore, the interpretation of CIL are always in flux. Therefore, I agree with those who say that the identification of a CIL norm is not a one-time occasion, but a continuous process. Even if the existence of such a norm has been identified in the past, we should not take this existence (nor the meaning of this norm) for granted. This process of continuous change also applies to other sources of international law, such as treaties. It explains why treaty interpreters must pay attention to the treaty parties’ subsequent practice as per art. 31(3)(b) VCLT.
Taking interpretative methods seriously
Contemporary international legal practice and scholarship show that we need to take interpretative methods more seriously in international law. One excellent contribution that addresses this issue is Stefan Talmon’s analysis of the ICJ’s interpretation of CIL. Among other things, Talmon shows that all too often, the ICJ asserts interpretative conclusions without providing a careful justification, and that it identifies State practice and opinio juris based on convenient (yet sometimes unproper) proxies and shortcuts.
This lack of stringency also applies, mutatis mutandis, to domestic courts’ interpretations of CIL, but also to their interpretations of norms stemming from other sources of international law. Like the practice of international adjudicatory bodies, the case law of domestic courts offers many examples of the interpretability of custom, yet these examples are rarely as rich and detailed as they could be. Indeed, domestic courts tend to shy away from unwritten law for a variety of reasons that cannot be fully explored here. Moreover, interpretative methods are, as already mentioned, often neglected, especially in relation to CIL, but not only.
I am excited by Nina Mileva and Marina Fortuna’s project to bring more clarity to the interpretation of customary international law. I am convinced that such clarity is much needed. I hope that this sketchy post will provide some food for thought in their endeavor, and I look forward to continuing the conversation with them and other international lawyers.