12 CHAPTER OnE ■ ApproAches to InternAtIonAl relAtIons
history? Are states as power hungry as some phi los o phers would have us believe? How
can we explain empirical findings? Can we use those findings to predict the future?
Behavioralism proposes that individuals, both alone and in groups, act in pat-
terned ways. The task of the behavioral scientist is to suggest plausible hypotheses
regarding those patterned actions and to systematically and empirically test those
hypotheses. Using the tools of the scientific method to describe and explain human
be hav ior, these scholars hope to predict future be hav ior. Many will be satisfied, how-
ever, with being able to explain patterns, because prediction in the social sciences
remains an uncertain enterprise.
The Correlates of War proj ect permits us to see the application of behavioralism.
Beginning in 1963 at the University of Michigan, the po liti cal scientist J. David
Singer and his historian colleague Melvin Small investigated one of the fundamental
questions in international relations: Why is there war?^13 Motivated by the normative
philosophical concern with how peace can be achieved, the two scholars chose an
empirical methodological approach. Rather than focusing on one “big” war that
changed the tide of history, as Thucydides did, they sought to find patterns among
a number of dif er ent wars. Believing that generalizable patterns may be found across
all wars, Singer and Small turned to statistical data to discover the patterns.
The initial task of the Correlates of War proj ect was to collect data on international
wars between 1865 and 1965 in which 1,000 or more deaths had been reported in a
12- month period. For each of the 93 wars that fit these criteria, the researchers found
data on its magnitude, severity, and intensity, as well as the frequency of war over time.
This data- collection pro cess proved a much larger task than Singer and Small had
anticipated, employing a bevy of researchers and gradu ate students.
Once the wars were codified, the second task was to generate specific, testable
hypotheses that might explain the outbreak of war. Is there a relationship between the
number of alliance commitments in the international system and the number of wars
that are fought? Is there a relationship between the number of great powers in the
international system and the number of wars? Is there a relationship between the
number of wars over time and the severity of the conflicts? Which factors are most cor-
related over time with the outbreak of war? And how are these factors related to each
other? What is the correlation between international system– level factors— such as
the existence of international organ izations— and the outbreak of war? Although
answering these questions will never prove that a par tic u lar factor is the cause of war,
the answers could suggest some high- level correlations that merit theoretical expla-
nation. That is the goal of this research proj ect and many others following in the
behavioralist scientific tradition.
Another example of research in the behavioral tradition can be found in human
rights lit er a ture. The question many scholars probe is why countries violate human
rights treaties. Is it because states never intended to follow the provisions? Is signing