Kantian fractionalization predicts the conflict propensity of the international system.

Network science has spurred a reexamination of relational phenomena in political science, including the study of international conflict. We introduce a new direction to the study of conflict by showing that the multiplex fractionalization of the international system along three key dimensions is a powerful predictor of the propensity for violent interstate conflict. Even after controlling for well-established conflict indicators, our new measure contributes more to model fit for interstate conflict than all of the previously established measures combined. Moreover, joint democracy plays little, if any, role in predicting system stability, thus challenging perhaps the major empirical finding of the international relations literature. Lastly, the temporal variability of our measure with conflict is consistent with a causal relationship. Our results have real-world policy implications as changes in our fractionalization measure substantially aid the prediction of conflict up to 10 years into the future, allowing it to serve as an early warning sign of international instability.