International Law as Behavior Symposium: An Agenda

[Harlan Cohen is an Associate Professor of Law at the University of Georgia School of Law]

What is the study of “International Law as Behavior”?  At the workshop in November, Elena Baylis, Tomer Broude, Galit Sarfaty, Jean Galbraith, and Tim Meyer (whose chapters/presentations were described earlier) were joined by Kathryn Sikkink, who presented on the role of agency in constructivism, Ron Levi and Sungjoon Cho, who drew upon sociology to study the “fields” of international criminal law and international human rights practice and the social structure of the WTO, respectively, Adam Chilton, who presented on the potential of experimental methods for studying human rights, and Anne van Aaken, who explored behavioral law and economics’ implications for international legal theory.  What, if anything, binds these ten projects together?  Are there lessons to be learned about how these projects and methods can fit together into some greater whole?  These will be topics discussed in the book arising out of this project, but for now, a few thoughts on ways forward.

It can be tempting to see these projects as puzzle pieces, which when assembled in the correct order, reveal a larger picture of the international order. Each brings its own insights: Rational choice sets up testable, generalizable hypotheses about how states might interact given express assumptions about state behavior. Sociology, anthropology, and behavioral law and economics can test those hypotheses against real world scenarios, explaining why specific situations diverge from those expectations, whether as a result of social structures, culture, or human psychology. Experimental methods can help identify the actual preferences of international actors. Constructivist accounts can build upon sociology, anthropology, and psychology to explain where state preferences come from and how they change. Focused primarily on different, overlapping units of analysis—individual actors, the communities in which they practice, the culture in which their embedded, the states on behalf of whom they act, and the larger structures in which those states are embedded, these approaches might seem like natural complements—snapshots taken from one angle, which when spliced together might provide a panoramic view of the international system. Together, these accounts might provide a more complex account of the different processes, preferences, beliefs, and incentives that might drive the vast array of actors who operate in international law, whether grass-root activists, transnational norm advocates, technocratic experts, politicians, bureaucratic careerists, or diplomats. Where these levers converge or diverge may help explain both the emergence of consensus over rules and continued contestation. Successful strategies for achieving particular international goals will flip all the right switches.

But imagining all of these accounts as different harmonies converging in one glorious tune is too simplistic and overly optimistic; a grand theory of international legal behavior is an unrealistic goal. For one thing, these methodologies overlap considerably more than the generalizations above might suggest, sometimes competing to answer the same questions. But even where they are complementary, any attempt to eliminate the boundaries between methodologies faces significant hurdles. Some of these may be discursive hurdles: Different methodologies are steeped in their own languages and terminologies that structure quality work in those fields, but that might seem impenetrable to those outside it, like discussions of “field” and “practice” within sociology. Terms of art may also sound similar but mean different things in different fields, creating confusion. As Martha Finnemore noted, “sociologists use the term ‘institution’ in very different ways than do rational-choice scholars or historical institutionalists, emphasizing the social and cognitive features of institutions rather than structural and constraining features.” Other hurdles may be set by conflicting methodological bias: Rational choice favors parsimony, prediction, and generalizability; anthropology and sociology favor deep description. Anthropological research, Galit Sarfaty explains, “is not based on prior assumptions or models. Rather, hypotheses and theories emerge from the data, and are constantly evaluated and adjusted as the research progresses.” This contrasts it and methods like it with others designed specifically to test hypotheses. Beginning from such different starting points may make such contrasting approaches difficult to pursue in tandem. For example, “interviews [in anthropology] are usually unstructured or semi-structured with open-ended questions developed in response to observations and ongoing analysis,” a stark contrast to the structured, carefully worded questions used in some experiments inspired by other methodologies. Training in any of these methodologies can also create blinders, framing international law questions for researchers as ones their expertise can answer and obscuring the need for other tools, as Elena Baylis suggests. Finally, different approaches may also create research hurdles; they may involve methods difficult to pursue at the same time or by the same scholars. Rationalist and experimental methods may require mathematical expertise not widely shared outside of certain fields. “Ethnographic research involves case-oriented study, including long-term fieldwork and in-depth interviews,” as Galit Sarfaty explains. The burdens of doing such work well may make it difficult for all but the most committed to undertake.

What might be more fruitful is to think about different ways they methods might relate to one another in a study of international law as behavior. The chapters in this book and discussion at the workshop suggest four basic models of interaction: Collaboration, Reflective Dialogue, Burden Sharing, and Competition.

In a model of collaboration, multiple methods might be used as part of a single effort (either by a single researcher of by collaborators with different expertise) to answer a specific question. Such multi-method approaches may be fruitful ways to answer or explore certain international law questions or phenemona, as Tim Meyer suggests regarding epistemic cooperation. A particularly good example of such a project would be Katerina Linos’ The Democratic Foundations of Policy Diffusion, discussed on Opinio Juris, which uses public opinion experiments, cross-national statistical analysis, and specific country case studies to explore where state health and family policies come from. When a multi-method approach works, as it does there, the whole ends up greater than sum of its parties, with the convergence between methods itself acting as a powerful argument in favor of the thesis. Some projects, particularly ones that require understanding the actions of multiple actors facing a range of different incentives (like the emergence of precedent within international law), or those where the unit of focus overlaps, may lend themselves to this kind of study.

A different model might be reflective dialogue. In this model, the results of parallel studies by different scholars following different approaches are used to inform or tweak each other in an effort to improve their respective results. As Jean Galbraith and Tomer Broude suggest, insights from behavioral law and economics might be used to revise rationalist assumptions about either compliance or rational treaty design. Insights from rational design might provide starting points for sociological or anthropological research, to study as Galit Sarfaty suggests, “disjunctures between how laws are written and how they are implemented on the ground.” Insights from anthropology or sociology might inform the design of experiments. Ethnographic and historical case studies may help refine constructivist theories of norm diffusion and normative change.

A third model might be burden-sharing. Some of these approaches are better equipped to answer some international law questions than others. Rationalist accounts may be better equipped to make general predictions about “states” than to explain specific individual decisions. Anthropology, on the other hand, may be able to explain those individual decisions, but with its emphasis on deep description, may not yield many generalizable hypotheses. Controlled experiments will be easier to carry out in some contexts than others: it is easier, for example, to test the opinions of the general public than that of experts in negotiation or on the battlefield. Public opinion though may not readily answer the questions that concern us; public opinion might or might not have any bearing on the behavior of diplomats, or politicians, or soldiers, and different approaches may be necessary to explain the relationship between public opinion and official action and the mechanisms—elections, training, culture, the market—that might relate one to the other. As Tomer Broude has explained, behavioral psychology may be on firmer ground describing areas of international law where individual decisionmaking dominates, for example, individual targeting decisions in International Humanitarian Law or judicial decisions. It may need to be more tentative in describing group decisions, where individual psychological biases might be muted, or the decision-making of agents, like diplomats, who may negotiate at some remove from those directly affected by the treaties they might produce. We might ask scholars using different methods to focus on specific areas where their methods might be most powerful and suggestive. Studying international courts and tribunals, for example, we might look to rational choice to explain their design, sociology to study the practice that emerges around them (as Ron Levi, Sungjoon Cho, Mikael Rask Madsen, and I have suggested), and behavioral psychology to better understand judicial votes and panel effects. We could then daisy-chain these explanations together to explain how courts evolve from their original mandate to their current practice

Finally, we might simply encourage competition. With regard to some questions, the assumptions or biases of different methodologies may simply be too hard to reconcile. We might instead encourage the strongest, most methodologically pure responses to those research questions, allowing their possibly contradictory explanations or predictions to live side-by-side (something akin to Elena Baylis’ methodological pluralism ). The ideal here might be something along the line of the classic Kurosawa film Rashomon, which, depicting the conflicting testimony of multiple characters without commentary, forces the viewer to reconcile the stories for herself, who ideally gains greater insight into the actual events from the overlapping accounts than might be possible from one supposedly neutral version.

Key in considering these models is shifting away from thinking about what these methods can do and towards what particular international law question might require. These different bridging mechanisms will be better or worse suited to different international law projects and questions. What the conversation at the November symposium demonstrates though is that figuring out how to relate these projects is something best achieved through active dialogue between the scholars pursuing them. I’m grateful to have been able to start this conversation with the scholars in this book project, all of whom are at the cutting edge using these methodologies and look forward to continuing it with both them and the readers of Opinio Juris.