In this article, Christopher Barrie uses a natural experiment setup to study how war affects group identities. The article builds on a survey that was in the field during the fall of Mosul to the Islamic State of Iraq and al-Sham (ISIS) in June, 2014.
The fall of Mosul occurred in the middle of the fieldwork for the survey, meaning it was possible to compare the responses of those surveyed before and after the outbreak of large-scale war. The analysis focuses on the identity responses of Sunni and Shi’i Muslims to see if both groups rally behind a national identity in response (i.e., whether war breeds unity) or whether they identify with divergent identities (i.e., was has disintegrative effects). It found the latter—that only Shi’i Iraqis rally more behind the nation.
The findings are explained with reference to what the author refers to as prior attachment – when groups derive a sense of security from the nation, they will rally behind it in face of threat. This was not the case, Barrie argues, for Sunni Iraqis who have confronted an ethnically exclusionary Shi’i-dominated state.
In the article, Barrie also finds that in areas where Shi’i and Sunni Muslims co-mobilized in defensive partnership against ISIS, Sunnis were much more likely to rally behind a national identity. This he explains with reference to achieved cooperation . When groups can be convinced of the cooperation of their ethnic opposites, this manifests in renewed attachment to the nation.
In summary: war can generate national unity—but only when groups mobilize in common.