Throughout history, states have sought to reshape the culture and identities of their citizens through assimilation policies and indoctrination efforts. In this paper, I show that coercive assimilation policy can cause substantial cultural change among ethnic minorities, but that these effects do not necessarily persist into later generations, and may even reverse. I focus on a historical policy in the United States under which authorities removed Native American children to distant boarding schools. I exploit the staggered recruitment patterns of schools and variation in cohort exposure to facilitate causal identification. I show that exposure to boarding schools offered few economic benefits, but did lead to rapid cultural assimilation. Treated cohorts were more likely to speak English, more likely to give their children western names, and more likely to be counted as `White’ in later censuses. However, I find that these effects reversed in the next generation, with the adult children of treated cohorts being more likely to live in a rural area and less likely to intermarry. I argue that stronger ethnic identification, associated with exposure to boarding schools and transmitted across generations, is a plausible channel for these effects. Ultimately, the schools seem to have strengthened the identities they sought to erase.

Protest movements are drivers of political, economic and social change. A fundamental goal of such movements is mobilisation on the basis of shared identities and common interests. This project examines whether protest - and in particular, ethnic activism - can affect the racial identities of non-protestors. I propose a new method to measure changes in racial identity using self-identified race in US social security card applications. Using a sample of individuals with Native American ancestry linked between historical censuses and social security records, I document an increase in the share identifying as ’nonwhite’ following high-profile ‘Red Power’ protests in the late 1960s. I propose to study the tangible effects of changes in racial identity on a range of short- and long-run outcomes, including the occurrence of subsequent protests, tribal enrollment, legal action against the US government, and the extent of cultural renewal later in the 20th century.

Individualism has been shown to have important economic, social and political consequences. This project examines whether individualism can be fostered by government policy, the degree to which it persists across generations, and its long-run implications for local economic development. We study these questions in the context of an ambitious land allotment programme targeting Native Americans in the early-20th century, using a range of historical and contemporary data sources. At the individual-level, we examine the effects of allotment on naming practices, intermarriage, participation in Native American civil rights associations, and location choice among descendants up to 100 years later. In order to document the long-term political and social consequences of allotment at the reservation-level, we construct new datasets on public goods provision, the occurence of and issues raised in local public meetings, and the content of modern tribal constitutions.

Terrorist groups often aim to extract concessions (e.g., religious or regional autonomy) from governments. While there is evidence that out-group attacks increase support for hawkish policies, less is known about the effects of terrorism on the attitudes and political preferences of those that are (supposedly) represented. In this project, I examine the effects of local terrorist attacks perpetrated by Basque nationalists on nationalist sentiment within the Basque country, and in other Spanish provinces with strong regional identities (e.g., Catalonia and Galicia). I use a combination of attitudes surveys that intersected terrorist attacks and highly-granular electoral data to study these questions.