Research on the `democratic peace' has received considerable attention in the last few years. Democracies, though, are often thrown together when examining the propensity of different regime types to become engaged in international conflict. Yet, democratic governments vary dramatically across nation-states. Whether it be presidential versus parliamentary, or multi-party versus single-party, democratic states clearly differ in the structure of their governing institutions. This article examines the relationship between government type, domestic political opposition, and the threat, show or use of military force. The analysis finds that Western parliamentary governments are rarely involved in militarized interstate disputes, but when they are they tend to be the targets of aggression by non-democratic states. Furthermore, the evidence suggests that these democratic governments are much more likely to reciprocate disputes when their opponent is a non-democracy. Reciprocation, though, also tends to be influenced by the type of government in power during a dispute. Coalition democratic governments, rather than single-party governments, are much more likely to reciprocate militarized disputes. In fact, the findings suggest that coalition governments are more likely to reciprocate disputes in general, and particularly more likely to reciprocate with the actual use of military force. The results also suggest that the level of polarization of a parliamentary government tends to decrease the probability of dispute reciprocation.

Economic sanctions are frequently used as a tool of foreign policy, described by some as falling between diplomacy and military force. An important question regarding the use of sanctions is whether they can function as an alternative to military force by demonstrating the sender's resolve and making military force unnecessary, or if their use tends to result in an increased probability that military force will be used. Based on a theory of sanctions as costly signals, the authors develop and test hypotheses regarding the relationship between sanctions and military force. The results show that after a sanction occurs, there is a significantly increased probability of a use of military force. Democracies, because of their propensity to tie their hands with audience costs, while at the same time facing domestic pressure to devise sanctions to be costless to the sender, are highly likely to be involved in a militarized dispute after using sanctions.

As a manifestation of its self-reliance doctrine, the state of Israel has acted unilaterally in the military realm throughout its history. Several scholars have also linked Israeli actions to elections and the economy in the context of diversionary theory. Hypotheses derived from earlier empirical work on internalization-externalization linkages and additional causal hypotheses are tested to enhance diversionary theory. Vector autoregression (VAR) is used to relax causal assumptions and test basic relationships in the Israeli case from 1948 to 1998. Findings suggest that Israel responds militarily to Arab military actions and domestic political protests. However, Israeli actions also lead to more domestic protests. The potential for this reciprocal relationship is typically ignored in the current diversionary research program. Arab military actions appear to be driven by Israeli military actions and seem to decrease in response to Israeli actions.

Poliheuristic (PH) theory models foreign policy decisions using a two-stage process. The first step eliminates alternatives on the basis of a simplifying heuristic. The second step involves a selection from among the remaining alternatives and can employ a more rational and compensatory means of processing information. The PH model posits that strategic/realist factors are more important in the second step of the process. The model is tested for the years 1918 to 1994, using crisis actors from the International Crisis Behavior data set. Results show that domestic political loss has a negative impact on the use of violence in response to a crisis trigger. Contiguity, joint democracy, and trigger are also significant in the expected directions. Relative capabilities have a positive impact, and enduring rivals do not appear more likely to use violence against each other as a first response in a crisis.A key divergence in the empirical study of international relations has been that between cognitive and systemic/rational approaches (Hagan 2001). All too often, each school has considered the other as overly deterministic or ad hoc. Jerel Rosati (2001) takes this discussion a step further because he calls for a merger between rational and cognitive approaches. Like Hagan (2001), Rosati asserts that scholars should not treat the state as a "black box." In other words, we need to explore state actions as a function of the human cognition of their leaders. By dismissing the human cognition component of decisions, we may be discarding important information. Focusing on the rational model alone might give only information about outcomes and preferences. Looking at cognitive issues can tell us about processes and beliefs as well as where preferences come from and how they are established. Rosati suggests that the