"In this book," wrote the late Quincy Wright, "Richard Barringer has made a contribution to the study of international conflict by devising a new method of classifying empirical data to characterize conflict and its stages of development, with results of considerable predictive value."Historians have long applied experienced judgment to distill the most significant factors in the development of large-scale conflict, and modern social scientists have attempted to reduce its acknowledged multidimensionality to manageable proportions by the application of correlation-based factorial techniques. The present study represents a new departure in this oldest and most persistent of civilized man's intellectual preoccupations. It establishes the beginnings of an objective, systematic, and automated program of research into the origins, development, and termination of war, and into the means of its control.An eclectic theory of conflict, a descriptive model of its significant stages, a novel technique of data collection, and an original method of data manipulation, analysis, and presentation are developed. A conflict codebook of 300 social, political, economic, and military indicators is presented as a comprehensive system within which all conflicts develop. Agreement analysis -- a noncorrelational technique for determining the dominant empirical patterns in a data base -- is developed in its complete form.Application of the method to eighteen wars of the twentieth century reveals the various combinations of factors that precipitate each significant stage of conflict. Finally, the contribution of this method to policy making, through computer simulation of conflict and early detection of conflict potential, is illustrated by example of the war in Vietnam.

A number of original techniques with broad application in the social sciences are developed in War: Patterns of Conflict. This companion volume contains the supporting material and computer programs for those techniques and the instruction for their use. It includes simple and compound agreement analysis, a logarithmic scaling technique for compound data, and a revised edition of the conflict data codebook.