In Determining Customary International Law: The ICJ’s Methodology between Induction, Deduction and Assertion, Stefan Talmon revisits the old debate over inductive and deductive methods for finding customary international law (CIL) to see whether we can now, fifty years after the original debates, learn any lessons about whether, when, and how the International Court of Justice uses each. What Talmon finds is that there are no clear patterns in how the ICJ uses each method, and that in fact, it is a third method, assertion, that best exemplifies the Court’s approach. What Talmon only hints at though are much broader lessons—that the ICJ’s failure to adopt a clear methodology for finding customary international law is only a symptom of a much broader problem, that the ICJ has never articulated a clear, coherent explanation of its authority to interpret customary international law or for whom. The ICJ’s efforts to find customary international law may simply be incoherent, a mirage, or even impossible.
The questions that Talmon’s study begs are why. Why isn’t the ICJ more interested in developing a clear methodology and why are there no patterns in the ICJ’s use of deductive or inductive methods? And why aren’t states more concerned? Why haven’t states demanded a clear methodology before treating ICJ decisions on custom as authoritative?
Methodology as Justification
Starting with the first question, the lack of clear methodology hints that the ICJ’s choice of induction, deduction, or assertion has little to do with methodology and everything to do with justification. When the Court invokes each one, it is attempting to justify its authority to interpret or find rules of CIL. Assertion makes this clearest—in the absence of any real evidence of a customary rule, the Court justifies it rules with appeals to “obviousness.” As Talmon wisely observes, both inductive and deductive methods are claims of derivative authority—the Court is not “making” or “choosing” a rule, but merely “finding” the rule made by states themselves in their interaction with one another. This is true whether the Court counts practice and weighs opinio juris or attempts to deduce a specific rule from recognized more general ones. It also tracks the requirement of article 38 of the ICJ Statute that the Court apply “international custom, as evidence of general practice accepted as law.” In essence, the Court‘s claim is that it is simply stepping into the shoes of states and making the same judgment they would make about specific rules and actions.
Negotiated Law and Adjudicated Law
The problem reflected in the ICJ’s methodological muddle is that this task may be impossible and the justification something of a fib. A court cannot step into a state’s shoes. A court forced to find a rule to decide a case is engaged in a fundamentally different activity than a state discerning a rule to guide its actions or jockeying for its favoured interpretation in relations with other states. As I explain in greater depth in International Law’s Erie Moment, states and courts are engaged in two entirely different forms of lawmaking that operate differently, instantiate different values, and derive legitimacy from different sources. Customary international law imagines a specific model of lawmaking, what I refer to as “Negotiated law,” in which states develop the law through their regular interactions. It imagines a constant back-and-forth, in which the meaning of the rules are discussed, debated, and fought over. Ambiguity and flexibility are features of Negotiated law rather than bugs; they encourage renegotiation and learning through practical experience. And the legitimacy of the rules that emerge from Negotiated law is supported by values of consent, compromise, and pragmatism. Courts face a very different task. The goal of “Adjudicated law” is to resolve ambiguity, to find with certainty for one party or the other. Justice rather than pragmatism is the goal. And whether using deductive or inductive logic, it is logic and reasoned elaboration, not jawboning, that is the valued method of interpretation. Adjudicated law accordingly derives authority from different sources of legitimacy—delegation, expertise, neutrality, reasoning, and finality.
These differences between adjudicated law and negotiated law complicate the transposition of rules from one context to other. The development of “custom” via adjudication is not just a continuation of that rule’s development, but its transformation into something new.
The Unraveling of the ICJ Statute’s Compromise
But the differences between Negotiated and Adjudicated law could largely be ignored when judgments of international courts were rare and ad hoc. The realization that the two forms of lawmaking are somewhat at odds may have inspired the drafters of the statutes of the Permanent Court of International Justice and ICJ to make those courts’ decisions binding only in specific cases and to relegate precedents to a secondary role in others. Essentially, states demanded that courts, as best as possible, decide cases based on what states had done or decided; the Court is asked to approximate what states would themselves perceive the law to be in their interactions with one another. In turn, the legitimacy of the ICJ’s decisions would be based on its ability to accurately reflect the results of state practice and will. And because of the impossibility of capturing those perfectly, states were only willing to accept the ICJ’s authority to translate Negotiated law into Adjudicated law for specific cases. They were not willing to give up the flexibility built into Negotiated law entirely. States would take the ICJ’s view into account in their continuing dialectic on custom, but the ICJ’s view had no authority to supercede that dialectic.
Two developments upended that understanding. First, as Talmon explains, almost inevitably, the ICJ was asked to decide questions that state practice could not definitely answer; standing in the shoes of states would be, at best, a guess. Shifting from inductive to deductive methods, and eventually to assertion, reflects a continued desire to use the same derivative legitimacy—the Court wasn’t doing anything that needed to be legitimated; it was only discerning the will of states embodied in existing Negotiated law. Whether using inductive or deductive logics though, this charade was somewhat hard to maintain. The Court was engaged in making Adjudicated law, which needed to be legitimated in its own way. Second, international court decisions became more common, both at the ICJ and elsewhere. The increasing density of adjudication had system effects for the relevance of ICJ decisions. An ICJ opinion on CIL might be cabined to a dispute when adjudication was relatively rare, but in a denser ecosystem of courts in which adjudication is frequent, it becomes impossible. Courts looking for rules of CIL now have the ICJ’s “clear” opinion alongside less than clear state practice and opinio juris. The ICJ’s and other courts’ opinions become new baselines against which future analyses of CIL take place. Whether desired or not (and there were backers of the PCIJ and ICJ from the beginning who did so desire), a system of precedent emerged.
The development of precedent upends the compromise implied in the ICJ statute (Adjudicated law threatens to replace Negotiated law entirely) and puts much greater pressure on the ICJ to legitimate its decisions on the content of CIL. But while some states have pushed back against this emerging system of precedent in some contexts (see more examples and discussion in Part II.B here), the ICJ has refused to legitimate it or to develop its own account of its authority to develop international law. Instead, it has hewed to methods—induction, deduction, or assertion—that continually reassert the Court’s authority as merely derivative. Where the use of those methods is unconvincing, it merely evidences the lie that Adjudicated law and Negotiated law are the same and that the Court solely applies the rules developed by states.
Do States Prefer Methodological Mayhem?
This then returns us to my second original question: why haven’t states demanded that the ICJ develop a clear methodology for finding rules of CIL? States could make a clear methodology a sin qua non of bringing cases and accepting jurisdiction. The fact that they haven’t suggests that they don’t necessarily want one. Why not? In any given case, there is a strong chance that deductive and inductive methods will support different answers, and states will likely prefer one method to another in a particular case. Across issues and time though, states have a multitude of positions on customary international law, some of which will be strengthened by appeals to state practice and others by appeals to logic. While allowing the ICJ to fudge a bit on methodology might worry states, particularly if ICJ opinions are likely to have precedential effect, that methodological fudge also guarantees that the room for states to continue debating custom will remain. And states may realize that an ICJ methodology risks becoming a CIL methodology more broadly. The ICJ may have an outsize ability to freeze a fluid debate over a particular custom; it might be worse, for states, if it could freeze the methodology for finding custom generally, which would threaten to disrupt the entire Negotiated law project of customary international lawmaking and its back-and-forth parrying. In turn, the possibility that states prefer the flexibility of methodological uncertainty over tying the ICJ’s hands suggests lessons not only for the ICJ, but for projects, like the International Law Commission’s, to codify the meaning and processes of CIL more broadly.