Introduction to the Symposium on Limitations of the Behavioral Turn in International Law

In recent years, scholars have increasingly applied empirical insights from cognitive research and behavioral economics to the analysis of international law and governance, a trend known as the behavioral turn in international law. These empirical insights are especially welcome in international legal scholarship where, for several decades, legal theory has been dominated by normative assumptions about individual and state behavior grounded in constructivist and rationalist approaches. However, the methodological and normative limitations of the behavioral turn have not yet been sufficiently explored. This symposium aims to examine these limitations and to provide avenues for further research. This framing essay sets the scene for the symposium, starting with a description of the behavioral assumptions of rationalist and constructivist approaches to international law. This is followed by a description of how insights into actual human behavior might inform rationalist and constructivist approaches. Subsequently, the methodological and normative limitations of these studies are elaborated upon, laying out the contributions to this symposium. The final section brings the insights together and provides avenues for further research.

behavior. 4 This fundamental social structure is, then, assumed to shape the identities, interests, and behavior of actors and vice versa. 5 These rationalist and constructivist approaches operate on a theoretical level and, hence, do not emphasize the collection of empirical evidence to support their theoretical assumptions about human behavior. 6 Insights from cognitive research and behavioral economics could provide empirical evidence to support, supplement, or undermine those theoretical assumptions.

How Insights from Cognitive Research and Behavioral Economics Can Inform Rationalist and Constructivist Approaches to International Law
Insights from cognitive research and behavioral economics challenge the notion of instrumental human rationality implied by rationalist and constructivist approaches to international law. These empirical studies show that actual human behavior sometimes systematically and predictably deviates from the normative behavior expected from a rational actor (that is, choosing the available option that maximizes expected utility). 7 These deviations are known as cognitive biases. In international legal scholarship insights on cognitive biases have been used to explain how and why international actors make certain decisions. 8 For example, cognitive biases, and specifically the availability bias, have been used to explain why policy-makers only started to consider the risks of Bilateral Investment Treaties (BITs) after having been subject to a BIT claim themselves. 9 The availability bias refers to individuals' tendency to believe that the probability of events that come to mind easily are more representative than what is actually the case. 10 Insights like this one provide a deeper understanding of how governments process information about the implications of their economic policies when negotiating and signing BITs.
International legal scholarship has also used insights into cognitive biases to explain how international law could best be designed. 11 These insights provide support for constructivists' assumptions that go beyond instrumental rationality. For example, constructivists have long held that actors make decisions through a certain lens; this is known as framing. Empirical research confirms that, contrary to what rational choice theory would assume, framing may affect actual behavior because actors' aversion to perceived loss is greater than their pleasure from 4 Adler, supra note 2. 5 Identities are representations of an actor's understanding of who they are, including preferences. perceived gains. 12 For example, it has been demonstrated that states behave differently depending on how a choice to resolve disputes at the ICJ is framed in a multilateral treaty. When states were required to explicitly opt out of adjudication by the ICJ, 80 percent accepted ICJ jurisdiction, while when states were required to explicitly opt into adjudication by the ICJ, only 5 percent accepted ICJ jurisdiction. 13 Such studies illustrate that how a treaty option is framed can affect the likelihood that states will accept it-leaving low-hanging fruit for treaty negotiators interested in increasing state participation. Furthermore, such findings go against instrumental rationalist assumptions that states would always choose the option that would maximize their own self-interest. Instead, the findings may be explained by the status quo bias-people's tendency to adjust their behavior to what they perceive to be the status quo, in this case the default option. 14

Methodological and Normative Limitations of Applying Cognitive Research and Behavioral Economics to International Legal Scholarship
The methodological and normative limitations of using cognitive research and behavioral economics in international legal scholarship have, however, not yet been sufficiently explored. 15 This symposium aims to explore questions related to the limitations of applying empirical insights from cognitive research and behavioral economics to international law. The contributions in this symposium reflect on when and how such insights can be useful in analyzing international law, and where the limits of that usefulness lie.
An important methodological limitation stems from the unit of analysis: whose behavior should be studied and how? In international law, the unit of analysis could vary from individuals and small groups (e.g., judges, trade negotiators, military commanders, political leaders) to large institutional groups (e.g., states, parliament, or administrative agencies). 16 Most of the empirical insights from cognitive and behavioral science stem from experiments with individuals or small groups in different contexts. Can such insights be applied to large institutional groups, such as the aggregate black-box state? One could argue that they can, since much of international legal decisionmaking is actually by individuals or small groups, even when operating under the authority of the state. 17 Such an approach would be similar to the approach adopted in rational choice theory, where the same behavioral assumptions are applied to individuals and states. And it would be particularly valuable in the international legal context where conducting experiments involving elite decision-makers, such as heads of state, judges on international tribunals, and other policy-makers, is rarely feasible. But what are the methodological limitations of transposing empirical insights from the individual level to the aggregate level, and of applying these insights in the context of states, bureaucracies, and institutions? And how can insights from an experimental study in a lab setting inform real-life situations?
Emilie Hafner-Burton, of the University of California, San Diego, addresses these methodological limitations in her contribution to this symposium. 18 She explains the potential and limitations of using convenience samples of students or other segments of the general public to provide insights into the behavior of elite decision-makers. She warns that studying how convenience samples of non-expert populations make decisions regarding tasks that in the real world are elite-driven may produce incorrect conclusions in a wide array of fields. Hafner-Burton provides several potential remedies to help minimize the dangers, including focusing inquiry on the types of decisions that the general public might generally make, studying people with specific behavioral traits similar to those presupposed in the underlying theoretical model, studying the elite decision-makers directly, or complementing lab experiments with other methods.
Sungjoon Cho, of Chicago-Kent College of Law, addresses the inevitable blind spots of behavioral approaches in order to improve the underlying methodology of this approach. 19 He explores the possible effect on BAIL of the double hermeneutic, where actors may conform their behavior to research findings. Cho raises several possible negative effects of the double hermeneutic on international law. To overcome these negative effects, he argues that normative judgments must remain central in international law, emphasizing the advisory rather than instructive role of behavorial approaches to international law.
Other methodological limitations of behavioral approaches to international law include concerns about the complexity of social science approaches, the fact that they require expertise not generally held by lawyers, and translation challenges between fields. Lauge Poulsen, of University College London, and Michael Waibel, of the University of Vienna, attempt to overcome these translation challenges by highlighting how the drivers and functions of nearly identical terms that reflect settled legal language in treaty provisions, known as boilerplate treaty provisions, could be informed by different social science approaches, most notably neoclassical economics, social psychology, and behavioral economics. 20 Anne van Aaken, of Hamburg University, also attempts to overcome these translation challenges by applying insights from cognitive science and behavioral economics to shed light on rules of interpretation in international law that have, so far, been dominated by doctrinal research approaches. 21 Normative limitations concern the legitimacy and desirability of using behavioral insights in international law and governance. Doron Teichman and Eyal Zamir, of the Hebrew University of Jerusalem, focus on the normative aspects of using nudges in international governance. 22 A nudge is a concept derived from cognitive science, proposing indirect suggestions and positive reinforcement to influence human behavior and the decision-making of groups or individuals. These include, for example, default rules, deadlines, goal-setting, and ranking. The authors first review how nudges are used in international governance, before moving to explore the normative limitations of nudging-both in general and in the concrete context of international law. Along with concerns that nudges may undermine actors' control over their choices by exploiting their cognitive biases, the authors discuss the whether it is desirable to offer actors with limited political accountability a powerful tool to steer the human behavior of individuals. The authors also discuss concerns that nudges may gain political acceptability at the expense of effectiveness.

Conclusion and Avenues for Further Research
The contributions provide food for thought on the limitations of the behavioral turn in international law. These limitations are not unique to international law's behavioral turn; it is well-known that research findings are limited. 23 For example, threats to internal validity (i.e., that the research design will accurately measure what it intends to measure) and external validity (i.e., that the sample accurately represents what you would get if you were able to assess the entire target population) are generally acknowledged in scientific papers of good quality. 24 Furthermore, what spurred the behavioral turn in international law is the major limitation of rationalist and constructivist approaches, that is, the lack of supporting empirical evidence for the normative assumptions of these theories.
In general, researchers should be aware of the limitations of their studies so they can accurately place their scientific findings in the right context. This allows for a better understanding of the added value of their research findings to the research literature. Furthermore, identifying limitations is an important step towards improving research methodology and formulating new research questions, as limitations expose the weakness of current research findings. Identifying limitations is especially important with respect to the behavioral turn of international law, which is still in its infancy. Some studies have simply never been conducted. Other studies need replication to provide a more substantial basis for the external validity of the original study findings.
The importance of exploring limitations to behavioral approaches was demonstrated within this symposium by the multiple suggestions for future research by contributors. Hafner-Burton, for example, argues for further empirical research regarding public opinion in international law, as well as studying elite decision-makers directly. She also proposes complementing lab experiments with observational data or historical case studies based on archival work. Teichman and Zamir argue that future research should identify additional nudges that operate in the international arena, develop robust models that can account for how nudges affect state behavior, and consider efforts by international organizations to use nudges to directly influence individual behavior. Van Aaken calls for further research on potential cognitive biases in legal interpretation by international courts and other interpreters in order to improve legal decision-making. Poulsen and Waibel's contribution opens a whole new research agenda to explore the implications of rationalist and behavioral economic insights on the design and adoption of boilerplate treaty provisions. Cho calls for further research on the interrelation between effortful non-heuristic thinking and effortless heuristic thinking in BAIL.
Failure to expose the weaknesses of the behavioral turn may cost it credibility when exaggerated promises fail to deliver. Accurately acknowledging the limitations of the behavioral turn in international law could increase the knowledge of lawyers and legal scholars not only with regard to its promises but also its perils. A proper understanding of the strengths and weaknesses of behavioral approaches in international law will contribute to a more robust body of international legal scholarship that is grounded in both theory and empirical investigation.