Kantian fractionalization predicts the conflict propensity of the international system

Many studies in international relations have investigated relationships between pairs of countries and the likelihood of conflict, yet none have connected the overall structure of the network of relationships between countries with the total level of international conflict. Here, we blaze a new path in the study of international conflict by introducing a measure of the overall fractionalization in the network of international relationships which we call Kantian fractionalization and demonstrating that this measure has been closely correlated with the number of new international conflicts in the following year. Moreover, we show that jointly democratic pairs of countries contribute negligibly to Kantian fractionalization, casting doubt on one of the most prominent concepts in international relations and policy prescriptions in Washington