2
We often conceive of democratic governments as being conducive to peace and stability.
This conception is often justified, particularly when compared to non-democratic
regimes. However, within the family of democracies there is certainly evidence that
suggests even stable democracies suffer from their fair share of political unrest. One only
need look to the United States during the nineteen sixties to realize that democracies are
not always able to fully channel dissent through democratic institutions. Occasionally
violence erupts on the streets of a democracy despite the ability of citizens to fully
participate in legitimate political processes. The goal of the current paper is to sort out,
within the family of established democracies, what factors lead to more or less political
violence.
Surprisingly, there have been relatively few empirical examinations into the
question of democracy and violence. This may be due, in part, to the notion that
democracies are generally less prone to violence. Nevertheless, I argue there is great
value in understanding the levels of violence that do occur within democratic systems.
Much of the work that has examined levels of violence within democratic systems
has focused on two broad concepts: majoritarian and proportional representational
systems of democracies (Powell 1982; Krain 1998). Powell (1982) finds that
proportional systems seem better able to contain riots and deaths than majoritarian
systems. Krain (1998) in a replication of Powell’s work find similar results.
2
Powell (1982) seems less concerned with the connection between democracy and
violence than he is with the broader concept of democratic performance. An analysis of
political violence is included only to the extent that how well a democracy can mitigate
2
Krain (1998) uses Powell’s data to demonstrate how the poisson regression model is the appropriate
model to test event count data such as Powell’s.
Convention