Communalizing colonial policies (CCPs) include a variety of practices that recognize and institutionalize communal difference among colonized populations, and several qualitative analyses find that they promoted postcolonial ethnic conflict. In contrast, the few quantitative analyses that explore this issue focus on several mechanisms, make conflicting claims, and provide mixed results, thereby suggesting that CCPs do not have general effects. Yet the quantitative findings might be inaccurate for several reasons: Some use the identity of the colonizer as a proxy for CCPs, others measure a CCP but have small samples with limited variation in the focal independent variable, and all of these analyses are unable to explore whether CCPs affect ethnic conflict through different and competing mechanisms. To address these limitations, we create four ideal types of CCPs, gather data on a CCP that conforms to each ideal type, and test the relationships between CCPs and ethnic civil warfare onset using the set of former British and French colonies. We find that a discriminatory CCP is associated with high odds of ethnic civil war onset, especially shortly after independence. Alternatively, differentiating and accommodating CCPs lack general relationships with ethnic civil war onset, and an empowering CCP is negatively related to ethnic warfare in most models.

In this article, we reorient the literature on colonialism and ethnic violence by exploring how different types of communalizing colonial policy (CCP) affected postcolonial patterns of ethnic warfare. We hypothesize that CCPs have limited or mixed effects when they simply recognize or empower communities but that they promote ethnic warfare when explicitly favoring some communities over others, especially when this discrimination affects the power of communities. To test these hypotheses, we combine a statistical analysis of the British Empire with a focused case study of Myanmar. We find that two relatively non-discriminatory CCPs—the use of communal census categories and high levels of indirect rule—had limited or mixed effects on postcolonial ethnic warfare. Unequal communal representation in the legislature and security forces and a mixed use of indirect rule, on the other hand, are three highly discriminatory CCPs, and we provide evidence that they increased the odds of postcolonial ethnic warfare.

A significant body of literature argues that American evangelical missionaries working in the Ottoman Empire in the 19th century directly contributed to the rise of Armenian nationalism. While acknowledging that missionaries had an effect on Armenian nationalism, this article finds that the impact of missionaries was much more unintended than is commonly assumed and resulted primarily from Armenian reactions to growing missionary influence. Employing new data on the biographies of Armenian nationalist leaders as well as comparative-historical methods, the article offers evidence that missionary influence spurred a backlash among the Armenian community that intensified preexisting local initiatives, increased investment in mass education in the provinces, and modernized its schooling system, all of which popularized and strengthened Armenian nationalism.

A significant literature demonstrates that the presence of historic missionary societies—especially Protestant societies—during the colonial period is significantly and positively associated with increased educational attainment and economic outcomes. However, we know less about the mechanisms underlying the long-run consequences of institutions, as it is commonly very hard to disentangle direct effects from indirect effects. One clear way to do so, however, is to explore the long-term impact of missionary influence in places in which the direct beneficiaries of missionary education are no longer present. The present article considers one such region, the Anatolian region of the Ottoman Empire. Due to the ethnic violence and population movements at the start of the twentieth century, the newfound Turkish nation-state was largely religiously homogenous. This provides us with a unique situation to empirically assess the long-run indirect effects of Christian missionary societies on local human capital. For this purpose, I present an original dataset that provides the locations of Protestant mission stations and schools, Ottoman state-run schools, and Armenian community schools contained within Ottoman Anatolia between 1820 and 1914. Contrary to the common association found in the literature, this study does not find missionary presence to be correlated with modern-day schooling. Rather, I find that regions with a heightened missionary presence and an active Christian educational market perform better on the gender parity index for pretertiary schooling during both the Ottoman and Turkish periods.

No abstract