The Effects of Language Policy on Identity: A “Natural Experiment” in Estonia

• In Event: Language Policy, State Building, and Social Cohesion

Abstract

How does a state-mandated language policy affect ethnic self-identification, national identity, and social trust and cohesion in the domestic political system? Extensive fieldwork and qualitative interviews with policymakers, specialists responsible for the policy’s implementation, and school personnel have underscored the deep and long-standing divisions in the Estonian education system along ethnolinguistic lines. From September 2024, Estonia will begin a nationwide education system overhaul which will require Estonian as the language of instruction for all public schools. For a nation in which roughly a third of the population is Russian-speaking and 15% of schools have operated predominantly in Russian at the time of transition, this language mandate will have significant socio-political as well as individual, psychological ramifications. Existing research evinces that language and socio-political identity are closely, though complexly, related (Laitin, 1998; Pérez and Tavits 2019). However, the causal effects of “civic integration,” including educational, policies are rarely studied, in part due to the rarity of as-if random policy variation (Fouka 2023, 3). While there is reason to anticipate some “material” gains from this language mandate such as greater higher educational uptake (which is only available for free in Estonian), prior research on compulsory language integration suggests that “symbolic” backfire effects are possible, can be long-lasting, and usually arise from “parental compensatory responses” (Bisin & Verdier 2001, Fouka 2020). The results of a triple difference identification strategy coupled with deep qualitative fieldwork and an analysis of socialization and spillover effects will yield broadly relevant findings about the effects of “civic integration” policies on identity and social cohesion among majority and minority groups in democratic societies.

To measure the effects of language mandates, this paper leverages the fact that the policy will go into effect for 4th grade students in the 2024–2025 academic year but not students enrolled in grades 5 and above. This policy intervention provides a unique opportunity for a “natural experiment” in which language of education is an as-if randomly assigned “treatment” set by policymakers. Causal effects can be measured using a difference-in-difference-in-differences estimator with testable identification assumptions (Olden and Møen 2022). With the official support of the Estonian Ministry of Education as well as municipal education boards, I will conduct a baseline survey in Spring 2024 of all students — in both transitioning and regular schools — who will be in the 4th and 5th grades in the 2024–2025 academic year as well as their parents and teachers. A series of endline surveys at yearly intervals thereafter will yield a fuller picture of the causal effects of a state-mandated language policy in terms of ethnic self-identification, national identity, social trust and cohesion, and other integration outcomes for members of both the ethnolinguistic majority and minority in society.

A separate analysis of parent and teacher identities and attitudes will shed light onto the socialization effects on their children and students, respectively, as well as spillover effects of the language transition on these adults. These data, coupled with qualitative interviews of parents and teachers, will provide an opportunity to investigate the effects of family and educational transmission as it relates to identity and various integration outcomes.

While this is a unique opportunity to study the causal effects of a significant language education change on various socio-political outcomes, the implementation of policies to promote the state or titular language, especially for the sake of social cohesion, is not. Integration of migrants and ethnic minority groups is a topic of great concern across Europe. Moreover, nation-building policies to promote the titular national language and culture in the long wake of the Soviet collapse are prevalent. Kindred educational policies can be seen in Ukraine and Latvia as well as the successful or still ongoing de-Cyrillization of alphabets in Azerbaijan (1991), Moldova (1989), Turkmenistan (1993), Uzbekistan (2021), and Kazakhstan (2025). This paper presents a research design that leverages a rare opportunity to learn about changes in identity, inter-ethnic relations, and general integration of minority language-speakers into society in response to an educational policy. Complemented by careful qualitative research, this paper illuminates the heterogeneous drivers of particular effects in different subsets of the population as well as socialization influences from — and spillover effects on — parents and teachers.

Author

• Laura Howells, Princeton University