Scholars and laypeople alike often consider nationalism to be a primarily destructive phenomenon because it supposedly has the potential to promote extremely violent international conflict and other atrocities. By glorifying the history, myths, and attributes of a national group while marginalizing or dehumanizing excluded groups, nationalism creates unbridgeable differences, irresolvable animosities, and perceived existential threats conducive to the outbreak of violence.

Despite this argument’s widespread acceptance, the extant literature does not adequately theorize about or assess the relationship between nationalism and violent international conflict. Theorization has focused almost exclusively on the association between nationalism and the incidence of interstate violence, preventing the formulation of a comprehensive theory regarding nationalism’s impact on international conflict behavior. Furthermore, any insights derived from existing theories are dubious because they are supported mainly on the basis of a nonrandom sample of case studies.

My dissertation addresses these problems in the literature. I first develop a comprehensive theory of the relationship between nationalism and international conflict behavior, specifically in terms of war initiation, severity, and duration. I argue that, in order to obtain mass cooperation, political elites embed their goals and the related policies in nationalist rhetoric and images. However, these “nationalist persuasion campaigns” can result in several mechanisms that promote long and deadly international conflict. Nationalism can provoke “national enemies” and their foreign allies, generate biased strategic assumptions, create domestic interest groups that favor war, permit the suppression of opposition groups, and promote “nationalist bidding wars” that bring about unexpectedly more aggressive military policies. It is possible that these mechanisms affect international conflict behavior to varying degrees depending on the form of nationalism present in a particular state. Therefore, I construct a series of hypotheses regarding the effect of four types of nationalism on three facets of international conflict behavior.

In order to assess the accuracy of my argument and move beyond the qualitative techniques that have limited generalizability, I use original data on the existence and type of state-level nationalism from 1816 to 1991 and quantitative analysis (namely, summary statistics, cross-tabulations, and specialized regression models). I find that nationalism contributes to the initiation of more deadly conflicts, but not necessarily to longer wars. Nationalism significantly increases both the probability that a state will initiate an interstate war and the number of battle deaths the war will cause. There is greater uncertainty surrounding the effect of nationalism on interstate war duration. Nationalism in the initiator significantly decreases the length of war, while nationalism in the target significantly increases it, though only during the twentieth century. None of nationalism’s subtypes has an unambiguous effect on international conflict behavior that is consistent with the expected pattern of violent conflict.

Given that nationalism encouraged the initiation of more lethal conflicts from 1816 to 1991, one may suspect that the international community might not have to concern itself as much in the future with the violent ramifications of nationalism due to increasing globalization. Yet, the empirical evidence described in the last part of my dissertation shows that this is not the case; national attachment and pride are still ubiquitous among individuals in more developed countries, due to perceived unsatisfactory economic conditions. Political leaders and the global community in general must then be cognizant of nationalism’s potentially violent consequences and support efforts to combat its divisiveness and prevent the accession of extreme-right parties. Otherwise, the possibility of extremely violent nationalist conflict and the ensuing human suffering remains.