07 Jul Customary International Law Symposium: Making Sense of Law as Practice (Or: Why Custom Doesn’t Crystallize)
[Jutta Brunnée is University Professor and Metcalf Chair in Environmental Law at the Faculty of Law of the University of Toronto.]
With her provocative new article Making Sense of Customary Law, Monica Hakimi challenges doctrinalists as well as theorists of international law to engage in a sophisticated conversation about a classical problem: how do we know when customary international law (CIL) exists as “a general practice accepted as law”? Monica argues that “orthodox” accounts of custom are overly concerned with identifying stable primary rules (i.e. rules regulating behavior) and, to that end, unduly reliant on secondary rules (for Monica’s purposes, rules that establish the criteria for the identification of primary rules) (1490). The central problem with this “rulebook conception” (1497), argues Monica, is that the secondary rules upon which it rests do not in fact exist (1506). In practice, CIL does not derive from “reliable and generally applicable secondary rules,” but “emerges more organically” (1515-16). CIL should be understood as a “real-world sociological phenomenon” (1505), not a rulebook.
In this brief comment, I want to focus on three key strands of Monica’s argument. First, international lawyers should not be distracted by purported secondary rules but focus their methodological attention on what actors “use and receive as CIL in the everyday practice of law” (1491). Second, while the fact that CIL is at bottom a practice “does not mean that CIL never functions like rules” (1491), it does mean that its norms are far more fluid than the orthodox account would allow (1505, 1526). Third, even in the absence of fully settled rules, CIL structures an argumentative practice that promotes the rule of law (1532-36).
I share Monica’s skepticism vis-à-vis the “rulebook” conception of CIL and her understanding of CIL as a practice. To the extent that Monica’s observation that CIL at times “functions like rules” is meant to express a practice-based understanding of rules, I am very much on board with that too. I would go further, however. CIL does encompass rules. As I will explain, there is no inherent tension between a practice-based conception of law and the existence of rules, be they rules of conduct or parameters for the identification of legal norms and arguments. In fact, I would argue that such rules and parameters are crucial to the existence and operation of a cognizable practice of “law.”
To be sure, CIL rules – or any legal rules, for that matter – are not static. But that does not mean that CIL can’t be relatively stable even as it retains fluidity. After all, a practice-based understanding of law reveals that both stability and change are dynamic. Both the parameters of the practice itself and the rules it instantiates remain in place only when they are maintained by the participants in the legal system, and they shift or decay when practices change or fade. It turns out that one of international law’s favourite metaphors – that of the crystallization of custom – is actually wrong. Custom, being generated and maintained by practice, does not crystallize; it gels. Monica’s compelling challenge to the rulebook conception helps us appreciate this central feature of CIL.
So, let me take these points in turn. First off, I entirely agree with the thrust of Monica’s challenge to the rulebook account of custom. CIL is not a set of rules that emerge fully formed from a “source.” Instead, it is generated, maintained, changed and, potentially, destroyed through international practice. In a practice conception of law the notion of a source, then, is really a shorthand for a lawmaking process that operates according to certain parameters, parameters that are also generated and maintained by social practice.
It will not come as a surprise, given what I have said so far, that I also agree with Monica’s observation that assessing whether CIL has emerged, or whether a given proposition accords with CIL, cannot be a quantitative exercise that measures the putative rule against the “two elements” of widespread practice and opinio juris (1497). Rather, a contextual, qualitative assessment of the salient practice is required, something that most adherents to the rulebook conception would acknowledge, as Monica notes in driving home her point that the orthodox view does not actually have secondary rules to show for (1506-10).
So, how then do we assess whether CIL has emerged and what its content is? Monica tells us that a practice conception of CIL does not mean that “anything goes;” it just means that “what goes is not determined by secondary rules” (1510). I agree that there is no on/off switch for CIL. Monica notes that, because the CIL process is “unstructured, and authority within it is diffusely held, normative positions must be accepted and treated as CIL in order to have the force of CIL. Those who want a position to be CIL must work to earn support for it from other participants in the process” (1525). Again, I agree. Monica’s account aligns with the orthodox view to the extent that for a rule to be CIL, it must have “widespread support” as law (1525). Yet again, I concur. But without any agreed parameters for this assessment, how do we break through the circularity of the proposition that CIL exists when something is (widely) accepted and treated as CIL?
If the assessment is not quantitative, and if the sorts of criteria assembled by the ILC in its Conclusions on the identification of custom fail as secondary rules, how do we know whether a norm is accepted as CIL? Monica notes that we may well look to the same “raw data” that the rulebook conception employs, i.e. “operational practice, judicial decisions, verbal pronouncements, and so on” (1520-21), not to determine whether a given position satisfies the two-element test, but to determine whether it is ”legally salient.” What amounts to legal salience, however, remains somewhat opaque. That is perhaps the inescapable consequence of Monica’s premise that there are no secondary rules. Monica does seem to draw a distinction between CIL and other modes of interaction. But the boundary is fluid, and this fluidity is central to her account of CIL (1505, 1526). For her, the argumentative practice that is CIL is “quintessentially legal in nature so long as it centers on the authority to make particular governance decisions and places this authority outside the hands of any one player” (1536).
I agree with Monica that the key to unlocking the mystery of CIL lies in the kind of practice actors engage in. But I differ with her when it comes to what kind of practice is needed. To my mind, the question of demarcation is more important than Monica acknowledges. That is because a “practice” does not arise simply when actors engage in material conduct or argumentation. Rather, according to an influential social science definition, practices are “socially meaningful patterns of action which, in being performed more or less competently, simultaneously embody, act out, and possibly reify background knowledge and discourse in and on the material world.” Hence, certain parameters must be developed to help constitute and maintain the practice called “law,” and to allow for argumentative patterns to be recognized as legal and guided by law. In other words, parameters for demarcation are not only compatible with but indispensable for a practice-based account of law.
And so, I concur with the many observers who have maintained that, although it is difficult to demonstrate that states hold an opinio juris, it remains important to assess why they engage in a given practice. There must be some kind of evidence that states consider “themselves to be applying a mandatory rule of customary international law” (North Sea Continental Shelf, para. 76). The ICJ, and the ILC, insist that this requirement needs to be satisfied by searching for separate evidence of opinio juris, although both acknowledge that the requisite evidence may well be found in the relevant practice. Regardless, rather than fixate on evidence for the “existence of a subjective element” (North Sea Continental Shelf, para. 77), I suggest that it is more productive to examine the practice with a view to determining whether it amounts to what I have called elsewhere a “practice of legality.”
In the “interactional” account of international law that I have developed with Stephen Toope, legal norms emerge from social interaction and when norms assume certain traits of legality. Legal norms must be general, promulgated, reasonably clear, consistent with other legal requirements, prospective, and refrain from asking the impossible. What is more, there must be a degree of congruence between the existing law and international practice. But congruence is more than conformity of behavior with the norm. Actors must respect the requirements of legality in their application, interpretation, justification of actions on the basis of the relevant norm – that is, they must engage in a practice of legality. Focusing on the traits and practices of legality, I suggest, allows a concrete assessment of whether a norm has evolved into CIL, and continues to be treated as such.
While the requirements of legality have been highlighted by legal theorists as elements of the rule of law, they are not abstract, philosophical notions. They too were generated through and have been maintained by practice, providing resilient background understandings concerning the parameters for legal interaction and legal argumentation. They are a “real-world sociological phenomenon,” to borrow Monica’s term. As such, they enable not only the identification of legal norms, in this case CIL. They also provide a concrete explanation of why it is not the case that “anything goes.” In the context of the “unwilling or unable” standard for self-defence against non-state actors, for example, norms that fall short of the requirements of legality are likely to struggle to emerge as CIL. By the same token, interpretations, arguments or justifications that strain the requirements of legality are unlikely to be persuasive and likely to meet with resistance.
I return to the proposition that shared parameters for what constitutes law, and what amounts to a legal argument, are indispensable for a practice-based conception of law. As I noted, without such parameters, a cognizable practice can neither arise nor be maintained. That is why it is of concern when actors’ actions and arguments over-stretch CIL and strain the criteria of legality, or when actors don’t bother with a legal justification at all. Such patterns, if they persist and fail to meet with push-back, not only risk undermining particular rules, but may eventually erode the very parameters that account for the practice of international law in the first place.
All told, Monica appears to be more sanguine about the practice of CIL than I am. I entirely agree with her observation that CIL can be “law,” without necessarily “authoritatively resolving what ought to be done” (1536). But while CIL’s “argumentative practice” does not require a “rulebook,” I would suggest that it does require shared parameters along the lines I sketched above. Further, I agree with Monica that conduct or positions that exceed existing CIL are “legally cognizable” (1521) and have the potential to affect the content of CIL (1494). However, although such positions are very much part of the process that shapes CIL, they are not CIL, as Monica appears to suggest (1511). In my view, it is important to maintain a distinction, and to be able to assert that such conduct or argumentation is in violation of CIL. Until a new rule, or an existing rule’s new content, has garnered widespread acceptance as law, the old rule remains in effect. In this context, the aforementioned requirements of legality may on occasion push towards change. More often than not, however, requirements such as generality, constancy over time, and non-contradiction will serve to stabilize existing rules. In my view, the requirement of generality, which demands that law provide “rules” of some kind, rather than standards for some actors or settings (1505, 1511) but not others, is central to any account of the rule of law, including one (rightly) grounded in argumentative practice (1533).
Monica’s thought-provoking article pushes us not only to make sense of custom, but also to make sense of law as practice. I much look forward to following the debate that it is bound to inspire.