Résumés
Abstract
Concentrating on "the how of International Relations", this article deals in its first half with three major issues in scientific methodology : 1) importance of the scientific mode in acquiring and transmitting knowledge, and also rendering it more policy-relevant; 2) methods of forecasting international events and their evaluation; 3) different routes to theory-building : induction, deduction, analogy, gaming and computer simulation. The second half of the paper substantiates this methodological survey by presenting some findings from the Correlates of War Project that has been going on at the University of Michigan for two decades. The emphasis is on the rigorous identification of the phenomenon (War) and on the presentation of data on different factors at different levels of analysis correlated with its incidence.