TOWARD A SCIENTIFIC THEORY OF WAR
Daniel S. Geller
The scholars most frequently credited with development of the quanti-
tative empirical study of war include Frederick Adams Woods, Alexan-
der Baltzly, Pitirim A. Sorokin, Quincy Wright, Lewis Fry Richardson,
Karl Deutsch, and J. David Singer. Of this set, the contributions of
David Singer to the scientiac study of international conbict will be
judged as paramount. Although Woods and Baltzly (1915), Sorokin
(1937), Wright (1942), and Richardson (1960) all compiled data on
wars, Singer’s efforts with the Correlates of War (COW) Project stand
alone. Over the last four decades, the vast majority of systematic scien-
tiac analyses conducted on the subject of international conbict have
employed some component of the expansive COW Project database.
Today, our empirical knowledge of the factors associated with patterns
of war and peace is attributable largely to the vision of David Singer.
1
The Correlates of War Project had its genesis in 1963 with a grant
from the Carnegie Corporation to the Center for Research on Conbict
Resolution at the University of Michigan. A portion of this grant went to
David Singer for the study of war. As did Sorokin, Wright, and Richard-
son, Singer and his associate Melvin Small culled historical materials for
information on war—in this case the frequency, participants, duration,
and battle deaths of all wars since 1816 (Singer and Small 1972; Small
and Singer 1982). Additional data sets were generated dealing with mil-
itarized interstate disputes, alliance membership, diplomatic ties, geo-
graphic proximity, territorial changes, intergovernmental organizations,
civil wars, and national material capabilities (inclusive of the military,
economic, and demographic dimensions of power).
2
Singer believed that with few exceptions most previous analyses on
the causes of war were insufaciently systematic and rigorous. Even the
work conducted by Sorokin (1937), Wright (1942), and Richardson
222
