Search
Browse By Day
Browse By Time
Browse By Person
Browse By Mini-Conference
Browse By Division
Browse By Session or Event Type
Browse Sessions by Fields of Interest
Browse Papers by Fields of Interest
Search Tips
Conference
Location
About APSA
Personal Schedule
Change Preferences / Time Zone
Sign In
X (Twitter)
Some of the most salient questions of international cooperation relate to treaty formation and ratification. Why do states join binding agreements? How much do treaties' substantive terms affect states' expected utility from the treaty? Does that expected utility also predict ratification or behavioral change? Or is membership in some treaties primarily a form of international or domestic signaling, such that the terms make little difference to state participation? Given the immense variation between agreements--their substantive objectives, participants, and operative mechanisms--there is no single solution to these puzzles. Rather, the answers vary markedly between classes of treaties, and even between individual agreements. These questions matter because, for decades, the role of multilateral treaties in world politics has been steadily growing.
To gain greater insight into these questions, we develop a Bayesian item response theory (IRT) model, variants of which have previously been applied to roll calls of domestic legislative (Carey 2003; Morgenstern 2003; Myagkov and Kiewiet 1996; Rosenthal and Voeten 2004) and judicial political (Martin and Quinn 2002) actors. Our model carries forward the key assumptions of the formal theoretical model developed by Morrow and Cope (2021), which explains the strategic logic of multilateral treaty negotiations generally. Our empirical model is therefore sufficiently flexible to apply across a broad range of time, places, and subject matters. It can be used to test theories of multilateral legal bargaining and explore a host of related questions of international cooperation.
As with legislative and judicial bodies' session roll call votes, the published records of treaty negotiations contain a wealth of information on actor preferences. Unlike those actors, however, states are less likely to express their preferences so clearly or succinctly. Rather, these positions come from groups of states engaging in protracted working sessions over the course of months or years. State representatives use these meetings to discuss, defend, and lobby for their positions on various treaty provisions and amendments. They sometimes enter roll-call votes or otherwise state their positions plainly. But other times they couch those positions in the nuanced language of diplomacy, greatly complicating quantification. Fortunately, these negotiations are usually recorded or summarized in detail, allowing us to use text analysis and/or hand coding by international legal experts to analyze and quantify the statements.
This paper makes several contributions to the literature. Methodologically, it illustrates how researchers can use formal models to motivate and inform their IRT models; demonstrates how scholars can use text data to estimate policy positions; and shows how others can use IRT models in international relations research. Substantively, it opens the door to a line of research on how states and IGOs can improve the gains from international cooperation by studying the positions that states take during negotiations.