19
another and even when they are not, they possess a shared history. This means that the territorial dispute
salience may not drop off as sharply over distance, making it profitable for the leader to act. Furthermore,
the rump state exercised influence over the entire pre-secession territory. Combined with the in-depth
knowledge of the secessionist state’s terrain, this may make it easier for the rump state to act militarily.
In turn, the secessionist state leader’s threshold for using force to demonstrate the state’s resolve may be
reduced – much like Maoz (1989) argues. After considering the theoretical linkages between the key
variables and territorial conflict, we turn to the research design issues.
Research Design
Operational Definitions
Secessions. A secession occurs when a homeland, non-colonial region of a country becomes an
independent, new state. For the definition of state, see Small and Singer (1982). De facto control of
territory by a rebel group or internal government restructuring is not considered a secession because it
does not alter recognized international borders. The Tir et al. (1998) list of territorial changes reports the
cases of 20
th
century secessions; the related country-pairs are listed in the Appendix.
23
The Dependent Variable. The main dependent variable is the onset of militarized conflict over territory
that takes place between two SISs. To operationalize this concept, we consult Ghosn and Palmer (2003),
who provide a listing of Militarized Interstate Disputes (MIDs)
24
through 2001. To evaluate hypothesis
H-1, we use all MIDs listed, but for the other hypotheses we use only those instances in which the MID
concerned territorial control for at least one of the disputants. The data are arranged on a yearly basis, so
23
Note, first, that German and Korean divisions at the end of World War II are not included because they do not
represent cases of internally-motivated secessions. Second, the divisions of India and Palestine are included despite
their colonial background. The cases are nevertheless included because they are common in prior research, so the
omissions would render this work less capable of speaking to the secession literature. Moreover, the Muslim
population in India and the Jewish population in Palestine were pursuing states separate from their neighbors as the
colonial rule was nearing its end, thus indicating an internal motivation for the division. Arguably, these are not so
much the cases of separation from the colonial ruler but cases of separation from the common state the groups would
have formed after decolonization.
24
A MID is defined as “a set of interactions between or among states involving threats to use military force, displays
of military force, or actual uses of military force . . . . these acts must be explicit, overt, nonaccidental, and
government sanctioned” (Gochman and Maoz, 1984: 587).
Convention