Regime Types and International Conflict, 1816-1976

Abstract

This study replicates and extends previous inquiries on the relations between regime type and conflict involvement of states. It examines the robustness of previous findings with respect to various regime attributes, various conflict involvement measures, and units of analysis. Using two comprehensive datasets on polity characteristics and militarized interstate disputes, the empirical analyses reveal: (1) There are no relations between regime type and conflict involvement measures when the unit of analysis is the individual polity (i.e., a state characterized by a certain regime type over a given time span); this finding is robust in that it holds over most regime characteristics and conflict involvement measures. (2) There is a significant relationship between the regime characteristics of a dyad and the probability of conflict involvement of that dyad: Democracies rarely clash with one another, and never fight one another in war. (3) Both the proportion of democratic dyads and the proportion of autocratic dyads in the international system significantly affect the number of disputes begun and underway. But the proportion of democratic dyads in the system has a negative effect on the number of wars begun and on the proportion of disputes that escalate to war.

An Institutional Explanation of the Democratic Peace

“…One such generalization, sometimes even asserted to be an empirical law of international relations (Levy 1988), is that democracies do not¯ght wars with one another. The empirical evidence for this claim is, in fact, quite strong (Maoz and Abdolali 1989;Bremer 1992;Oneal and Russett 1997;Ray 1995). Recent efforts to cast this empirical observation in doubt notwithstanding (Layne 1994;Spiro 1994;Farber and Gowa 1995;Schwartz and Skinner 1997), extensive, rigorous statistical tests all show a signi¯cant propensity for democracies to have been virtually immune from wars with one another (Russett 1995;Maoz 1998).…”

Navigating the Range of Statistical Tools for Inferential Network Analysis

“…Perhaps the most persuasive example of this involves the "democratic peace," with the idea being that democracies tend not to fight one another (for canonical works, see Maoz and Abdolali 1989;Maoz and Russett 1993). The effect of the democratic peace has been largely established via logistic regression models, which assume conditional independence of observations.…”

Constructing Multivariate Analyses (of Dangerous Dyads)

“…(He compared the results of that analysis with another focusing on contiguous states only.) 1 Maoz and Abdolali (1989) analyzed pairs of states over a much longer time period than previous analysts, from 1816 to 1976. They reported on the relationship between the regime types of pairs of states and the probability of interstate disputes, as well as the wars in which they became involved.…”