Escalation processes are found in many types of international conflict
. However, a great deal of the theoretical and empirical literature on
escalation is context specific and concentrates on explaining the out
comes of an escalation process. This approach has generated numerous i
nsights; however, our understanding of escalation processes, in genera
l, remains partial and incomplete. In this article, the author develop
s a two-sided incomplete information model to identify the kinds of es
calation strategies states are likely to adopt in conflict. The model
produces several hypotheses, one of which is tested empirically in the
context of militarized interstate disputes. The hypothesis states tha
t as the disparity between the players' cost tolerances increases, the
lower cost tolerant actor is more likely to escalate to the maximum o
f his or her ability on the first move in the conflict. The results of
the test confirm the theory's expectations of an inverse relationship
between cost tolerance and an actor's escalation behavior. The articl
e concludes by noting implications for future research on escalation p
rocesses.