ABSTRACT
Post-imperial ethnic identities and divides are often constructed and construed through direct and indirect references to imperial legacies. In this study, we use nationally representative survey data to examine proficiency, use, and valuation of the Russian language – a major such legacy of the Soviet empire – in Kyrgyzstan, a multiethnic Central Asian nation with a long and complex history of ethnic and regional cleavages. The multivariable analyses produce instructive net variations in Russian proficiency and use across regional subgroups of ethnic Kyrgyz, the titular ethnicity, and between Kyrgyz and Uzbeks, a marginalized ethnic minority. The analyses also show that the command and use of Russian increase with community ethnic heterogeneity. Yet, no variations along these axes are found in the perceived importance of Russian language knowledge for success in the domestic labour market. These findings are situated within the interconnected contexts of historical ethnolinguistic legacies, dynamics of nation-building, and geopolitics.
Acknowledgements
The first author acknowledges support from UCLA’s California Center for Population Research (NICHD Population Research Infrastructure Grant #P2C-HD041022).
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
De-identified survey data and coding programs are available upon request.
Notes
1 Notably, although respondents could choose more than one ethnicity, all, except one respondent in Round I, chose only one.
2 The wording of the questions on the language of reading and writing was slightly different in the two rounds. Whereas the Round I questionnaire asked about the language in which respondent last time read a newspaper, in Round II, the corresponding question also covered other printed and digital media such as a book or a text on the Internet. Also, to reflect the growing access to new technologies, in Round II the question about writing specifically mentioned digital messages. We do not believe, however, that these differences greatly impact the comparability of the two datasets.