¹ Three important exceptions are the studies by Armstrong, John A., “The Ethnic Scene in the Soviet Union: The View of the Dictatorship,” in Ethnic Minorities in the Soviet Union, ed. Goldhagen, Erich (New York: Praeger, 1968), pp. 3–49 Google Scholar; Aspaturian, Vernon V., “The Non-Russian Nationalities,” in Prospects for Soviet Society, ed. Kassof, Allen (New York: Praeger, 1968), pp. 143–198 Google Scholar; and Vardys, V. Stanley, “Communism and Nationalities: Soviet Nation Building,” Unpublished paper presented to the 65th Annual Meeting of the American Political Science Association, New York, September 1–6, 1969 Google Scholar.

² For reviews of official doctrine and policy, in addition to the works cited above see: Pipes, Richard, The Formation of the Soviet Union: Communism and Nationalism, 1917–1923, rev. ed. (New York: Atheneum, 1968)Google Scholar; Conquest, Robert, Soviet Nationalities Policy in Practice (New York: Praeger, 1967)Google Scholar; Hodnett, Grey, “The Debate over Soviet Federalism,” Soviet Studies, 18 (April, 1967), 458–481 CrossRefGoogle Scholar; and Rakowska-Harmstone, Teresa, “The Dilemma of Nationalism in the Soviet Union,” in The Soviet Union under Brezhnev and Kosygin: The Transition Years, Strong, John W., ed. (New York: Van Nostrand, 1971), pp. 115–134 Google Scholar.

³ For a review of Soviet language policy, see Conquest, Soviet Nationalities Policy. For a discussion of educational policies with respect to the nationalities, see Pennar, Jaan, Bakalo, Ivan I., and Bereday, George Z. F., Modernization and Diversity in Soviet Education: with Special Reference to Nationality .Groups (New York: Praeger, 1971)Google Scholar; Bilinsky, Yaroslav, “Education of the Non-Russian Peoples in the USSR, 1917–1967: An Essay,” Slavic Review, 27 (September, 1968), 411–437 CrossRefGoogle Scholar; and Lipset, Harry, “The Status of National Minority Languages in Soviet Education,” Soviet Studies, 19 (October, 1967), 181–189 CrossRefGoogle Scholar.

⁴ Aspaturian writes: “Sovietization is … the process of modernization and industrialization within the Marxist-Leninist norms of social, economic and political behavior. Russianization is defined as the process of internationalizing Russian language and culture within the Soviet Union. … Finally, Russification … is defined as the process whereby non-Russians are transformed objectively and psychologically into Russians, and is more an individual process than a collective one.” See “The Non-Russian Nationalities,” pp. 159–160.

⁵ I shall use the terms “ethnic identity” and “national identity” interchangeably. Although there is a more specialized Russian terminology referring to ethnic groups according to their stage of historical development, Russian writers usually refer to the Russian people and to the other ethnic groups generically as either “nationalities” (natsional'nosti) or “peoples” (narody).

It may be of some interest to add that the Soviet Union and the Soviet people are never referred to collectively as a nation or nationality. One may speak of the Soviet government (pravitel'stvo), state (gosudarstvo), or country (strana). Similarly, one may refer to the Soviet people (narod) or peoples (narody). But never to the Soviet “nation” (natsiia), a term which is reserved for some thirty Soviet ethnic groups. For an interesting and critical discussion of Russian ethnic terminology, see Kozlov, V. I.. Dinamika chislennosti narodov: Metodologiia issledovaniia i osnovnye faktory (Moskva: “Nauka,” 1969), pp. 20 and 55–69 Google Scholar.

⁶ Shibutani, Tamotsu and Kwan, Kian M., Ethnic Stratification: A Comparative Approach (New York: Macmillan, 1965), p. 47 Google Scholar.

⁷ Note that two common usages of the term “national identity” are avoided here. First, one often finds the term employed synonymously with “national integration,” especially in discussions of a nation's or new polity's “search for identity.” By this is variously meant a quest for a set of common values (transcending, perhaps, more “parochial” loyalties) or for agreement on the norms governing political action. While one might wish to label such a consensus on the ends and forms of the political community “national identity,” such a usage tends to confuse the concepts of “state” and “nation” and seems to imply that the achievement of “national integration” (or of a viable political system) is by definition inimical to the preservation of narrower “ethnic” loyalties. For a discussion of the tendency to confuse the concepts of nation and state, see Connor, Walker, “Nation-Building or Nation-Destroying?” World Politics, 24 (April, 1972), 332–336 CrossRefGoogle Scholar.

A second approach equates the term “national identity” with consensus on the set of beliefs expressed in answer to the following questions: What sort of people are we? What makes me a (Canadian)? How am I different from other people? This approach more or less equates national identity with a collective self-description of a people. Such self-description seems to be identical to what Daniel Glaser has called “ethnic ideology” (to be discussed further below). See, for example, such uses of the term “national identity” in Eisenstadt, S. N., “Israeli Identity: Problems in the Development of a Collective Identity of an Ideological Society,” The Annals of the American Academy of Political and Social Science, 370 (March, 1967), 116–123 CrossRefGoogle Scholar; and Schwartz, Mildred A., Public Opinion and Canadian Identity (Berkeley: University of California Press, 1967)Google Scholar.

⁸ Glaser, Daniel, “Dynamics of Ethnic Identification,” American Sociological Review, 23 (February, 1958), 32 CrossRefGoogle Scholar.

⁹ See especially Rosenthal, Erich, “Acculturation Without Assimilation? The Jewish Community of Chicago, Illinois,” American Journal of Sociology, 66 (November, 1960), 275–288 CrossRefGoogle Scholar; Gordon, Milton M., Assimilation in American Life: The Role of Race, Religion, and National Origins (New York: Oxford University Press, 1964)Google Scholar; Glazer, Nathan and Moynihan, Daniel, Beyond the Melting Pot: The Negroes, Puerto Ricans, Jews, Italians, and Irish of New York City (Cambridge, Mass.: MIT Press, 1963)Google Scholar; Nahirny, Vladimir C. and Fishman, Joshua A., “American Immigrant Groups: Ethnic Identification and the Problem of Generations,” The Sociological Review, 13 (November, 1965), 311–326 CrossRefGoogle Scholar; and Parenti, Michael, “Ethnic Politics and the Persistence of Ethnic Identification,” American Political Science Review, 61 (September, 1967), 717–726 CrossRefGoogle Scholar.

¹⁰ Wallerstein, Immanuel, “Ethnicity and National Integration in West Africa,” Cahiers d'études africaines, 3 (October, 1960), 133 Google Scholar.

¹¹ Cohen, Abner, Custom and Politics in Urban Africa: A Study of Hausa Migrants in Yoruba Towns (Berkeley: University of California Press, 1969)Google Scholar; and Smock, Audrey C., Ibo Politics: The Role of Ethnic Unions in Eastern Nigeria (Cambridge, Mass.; Harvard University Press, 1971)CrossRefGoogle Scholar.

¹² For a similar argument, see Rubel, Paula G., “Ethnic Identity Among the Soviet Nationalities,” in Soviet Nationalities Problems, ed. Allworth, Edward (New York: Columbia University Press, 1971), pp. 217–218 and 222–223 Google Scholar.

¹³ Kozlov, , Dinamika chislennosti narodov, p. 48 Google Scholar. M. S. Dzhunusov also emphasizes the importance of ethnic self-consciousness, although he imbues the term with somewhat broader meaning than does Kozlov. While Kozlov tends to restrict the usage to “attachment to the group,” Dzhunusov interprets ethnic consciousness more or less as ethnic ideology. See “Teoriia i praktika razvitiia sotsialisticheskikh natsional'nykh otnoshenii,” Voprosy filosofii, 1967, No. 9, pp. 26–36 Google Scholar. See also Gurvich, I. S., “Nekotorye problemy etnicheskogo razvitiia narodov SSSR,” Sovetskaia etnografiia, 1967, No. 5, pp. 62–77 Google Scholar.

¹⁴ In the short run, however, different “fates” are envisioned for the languages of the numerically small peoples than for the languages of the larger ones. Languages of Union Republic nationalities are considered to be destined for further functional development as their users advance socially and economically. Languages of nationalities of Autonomous Republics or lesser status national regions, on the contrary, are finding their further functional development limited. Many spheres of activity there (science, technology, government, education) are increasingly being serviced by the Russian language as the “internationality language.” On this see especially Desheriev, Iu. D. and Protchenko, I. F., Razvitie iazykov narodov SSSR v sovetskuiu epokhu (Moskva: “Prosveshchenie,” 1968), pp. 121–125 Google Scholar; and Burmistrova, T. Iu., Teoriia sotsialisticheskoi nalsii (Leningrad: Izdatei'stvo Leningradskogo universiteta, 1970), pp. 46 and 65 Google Scholar. Desheriev and his colleagues have developed the model of functional development of Soviet languages in a number of works. The volume cited here presents the model in its fullest form.

¹⁵ Isaev, M. I., “Natsiia i iazyk,” Voprosy istorii, 1968, No. 2, p. 106 Google Scholar.

¹⁶ Kholmogorov, A. I., Internatsional'nye cherty sovetskikh nalsii (Na materiialakh konkretno-sotsiologicheskikh issledovanii v Pribaltike) (Moskva: “Mysl',” 1970), p. 141 Google Scholar. For similar arguments or assumptions, see Guboglo, M. N., “Vzaimodeistvie iazykov i mezhnatsional'nye otnosheniia v sovetskom obshchestve,” Istoriia SSSR, 1970, No. 6, p. 24 Google Scholar; Kozlov, V. I., Dinamika chislennosti narodov, p. 26 Google Scholar; and Bertagaev, T. A., Desheriev, Iu. D. (supervisor), Isaev, M. I. et al., “Rol' russkogo iazyka v razvitii slovarnogo sostava iazykov narodov SSSR,” in Voprosy terminologii (Materiialy vsesoiuznogo terminologicheskogo soveshchaniia) (Moskva: “Nauka,” 1961), p. 34 Google Scholar.

¹⁷ Rough approximations might be made of losses due to complete identity change in recent years on the basis of a comparison of actual 1970 census totals with estimates based on 1959 census figures and fertility and mortality rates for the intervening years. Such estimates would be very crude, however, because there are no published figures on mortality and fertility rates for specific nationalities but only rates for regions (provinces and republics), which are usually ethnically mixed. Even more severe problems are presented if one wishes to project the extent of assimilation for earlier years, because of a similar lack of fertility rates, compounded by the severe disruptions due to population and border shifts and war losses between the censuses of 1926 and 1959.

¹⁸ When examining language identity preservation, this analysis ignores changes that may be taking place in the languages themselves. It may be misleading to speak of “preservation” of a national language when the language itself may be only recently standardized or created as a self-standing language. Moreover, many spheres of non-Russian languages are heavily infused with Russian loan words. It has been estimated, for example, that in many non-Russian languages, 70 to 80 percent of the new technical-scientific, sociopolitical, educational, and other specialized terms consist of borrowings from the Russian language or through it. ( Bertagaev, et al., “Rol' russkogo iazyka,” p. 14.)Google Scholar Nevertheless, Soviet linguists do not envision fusion of Russian and non-Russian languages but rather eventual displacement of national languages by a single internationality language. It is therefore appropriate to treat identification with a particular language and its literary heritage as a meaningful indicator distinguishing speakers of one language from those of another.

¹⁹ For a description of the contents of the passport and for an explanation of how the entry on nationality is determined, see Kudriavtsev, P. I. Chief ed., Iuridicheskii slovar', 2nd ed. (Moskva: Gosudarstvennoe izdalel'stvo iuridicheskoi literatury, 1956), II, 102–103 Google Scholar.

It is worth noting that recent studies of the choice of nationality by children of ethnically mixed parents do not reveal an overwhelming attraction to the Russian nationality. Indeed, some evidence points to a process of reverse assimilation of Russians by the local. nationalities. For example, L. N. Terent'eva showed that in the capital cities of the Baltic republics, children of mixed Russian-native parents selected the local nationality in a majority of cases—though the frequency also depended on whether the father or mother was Russian, since children tend to favor the father's nationality. (See Terent'eva, L. N., “Opredelenie svoei natsional'noi prinadlezhnosti podrostkami v natsional'no-smeshannykh sem'iakh,” Sovetskaia etnografiia, 1969, No. 3, pp. 20–30.)Google Scholar Similar patterns have also been noted in the Tatar ASSR and the North Caucasus. (See Shkaratan, O. I., “Etnosotsial'naia struktura gorodskogo naseleniia Tatarskoi ASSR,” Sovetskaia etnografiia, 1970, No. 3, p. 15 Google Scholar; and Sergeeva, G. A. and Smirnova, Ia. S., “K voprosu o natsional'nom samosoznanii gorodskoi molodezhi,” Sovetskaia etnografiia, 1971, No. 4, pp. 86–92.)Google Scholar Such reverse assimilation is probably largely confined to the native republics and provinces, however.

²⁰ For a discussion of the validity and reliability of the census questions on nationality and language, see Bruk, S. I. and Kozlov, V. I., “Etnograficheskaia nauka i perepis' naseleniia 1970 goda,” Sovetskaia etnografiia, 1967, No. 6, pp. 3–20 Google Scholar.

²¹ For a discussion of the contents of the 1970 census, see Zinchenko, I., “Voprosy natsional'nosti i iazyka pri perepisi naseleniia SSSR,” Vestnik statistiki, 1969, No. 11, pp. 53–58 Google Scholar.

²² Deutsch, Karl W., “Social Mobilization and Political Development,” American Political Science Review, 60 (September, 1961), 494 Google Scholar.

²³ See Deutsch, Karl W., Nationalism and Social Communication: An Inquiry into the Foundations of Nationality, 2nd ed. (Cambridge, Mass.: MIT Press, 1966), pp. 162–163 Google Scholar. On this point see also Kozlov, V. I., Dinamika chislennosti narodov, pp. 332–334 Google Scholar; and Gumperz, John J., “Language and Communication,” The Annals of the American Academy of Political and Social Science, 373 (September, 1967), 228 CrossRefGoogle Scholar.

²⁴ For discussions of the motivational and aspirational levels of Soviet urban and rural residents (though not specifically of the non-Russian nationalities) see Mickiewicz, Ellen, “Uses and Strategies in Data Analysis of the Soviet Union: Cleavages in Industrialized Society,” in Handbook of Soviet Social Science Data, ed. Mickiewicz, Ellen (New York: Free Press, 1973), pp. 4–16 Google Scholar; Yanowitch, Murray and Dodge, Norton T., “The Social Evaluation of Occupations in the Soviet Union,” Slavic Review, 28 (December, 1969), 619–643 CrossRefGoogle Scholar; and Rutkevich, M. N. and Filippov, F. P., Sotsial'nye peremeshcheniia (Moskva: “Mysl’,” 1970), pp. 237–238 Google Scholar.

²⁵ Once again it should be emphasized that when measuring Russification this analysis measures only linguistic Russification, not complete loss of original national identity.

²⁶ Although this study will continue to hypothesize that social mobilization encourages Russification, Karl Deutsch has observed that as a rule the rate of mobilization outruns that of assimilation. Rapid mobilization tends to destroy the unity of multinational states because with increasing education and exposure to mass communications in their native tongues, members of each nationality become more aware of their differences from other groups and of their affinities with persons of kindred language or heritage. Thus the effect of social mobilization may be to exacerbate ethnic conflict rather than to lessen it. See Deutsch, , Nationalism and Social Communication, p. 126 Google Scholar. On this same point see also Fishman, Joshua A., “Language Maintenance and Language Shift as a Field of Inquiry,” Linguistics, No. 9 (November, 1964), 52–53 Google Scholar. For a similar argument applied to the Soviet case, see Pipes, Richard, “The Forces of Nationalism,” Problems of Communism, 13 (Jan.-Feb., 1964), 1–6 Google Scholar.

Furthermore, one cannot rule out the possibility that both assimilation and strengthening of national identities may occur simultaneously among different segments of the mobilized population. One Soviet geographer has noted the dual role of cities as centers of both ethnic rapprochement (sblizhenie) and ethnic consolidation. ( Pokshishevskii, V. V., “Etnicheskie protsessy v gorodakh SSSR i nekotorye problemy ikh izucheniia,” Sovetskaia etnografiia, 1969, No. 5, pp. 4–5.)Google Scholar Moreover, resistance to assimilation may in fact be a reaction by certain social groups, such as the national intelligentsias or political leaders, to assimilatory pressures or to evidence of assimilation. We should remain aware, therefore, that mobilization processes may retard as well as hasten the rate of assimilation, and that the aggregate measures of mobilization and assimilation employed here are not fully sensitive to these complex social processes.

²⁷ On some additional drawbacks to such a measure, see Fishman, Joshua A., “Sociolinguistic Perspective on the Study of Bilingualism,” Linguistics, No. 39 (May, 1968), 31–34 Google Scholar.

²⁸ See, for example, Bondarskaia, G. A., “Rol' etnicheskogo faktora v formirovanii territoriial'nykh razlichii rozhdaemosti,” in Voprosy demografii (Issledovaniia, problemy, metody), ed. Volkov, A. G. et al. (Moskva: “Statistika,” 1970), p. 171 Google Scholar; Karakhanov, M. K., “Dinamika narodonaseleniia srednei azii,” pp. 23–4Google Scholar; and Melamed, M. P., “Vosproizvodstvo naseleniia v Turkmenskoi SSR i vliianie na nego faktory,” pp 95–8Google Scholar, both in Problemy narodonaseleniia, Karakhanov, M. K. chief ed. (Moskva: Izdatel'stvo Moskovskogo universiteta, 1970)Google Scholar. On the preservation of other traditional customs relating to family life, see also Pipes, Richard, “Assimilation and the Muslims: A Case Study,” in Soviet Society: A Book of Readings, ed. Inkeles, Alex and Geiger, Kent (Boston: Houghton Mifflin, 1961), pp. 593–595 Google Scholar.

²⁹ It is difficult to find truly comparable data for several groups. Figures for intermarriage between Tatars and Russians in Kazan and figures for marriage between Ukrainians and Russians in Kharkov reveal that the former rate is far lower than the latter despite similar overall proportions of Russians and natives in the two cities. See (on Kazan) Khanazarov, K. Kh., “Ob odnom aspekte dal'neishego sblizheniia natsii v SSSR,” Obshchestvennye nauki v Uzbekistane, 1968, No. 6, pp. 23–26 Google Scholar; and (on Kharkov) Kurman, M. V. and Lebedinskii, V., Naselenie bol'shogo sotsialisticheskogo goroda (Moskva: “Statistika.” 1968), p. 126 Google Scholar.

³⁰ Of the Muslim intellectual in the Soviet Union today, Bennigsen and Lemercier-Quelquejay have written: “He may be indifferent to religious dogma and practice, but he remains attached to the whole body of customs and traditions which make up the special character of the Muslim way of life.” This continuing attachment to the community of believers is symbolized by the universally practiced rite of circumcision; more-over, intermarriage is strongly frowned upon. See Bennigsen, Alexandre and Lemercier-Quelquejay, Chantal, Islam in the Soviet Union, trans. Wheeler, Geoffrey E. and Evans, Hubert (New York: Praeger, 1967), p. 213 Google Scholar.

³¹ Fishman, Joshua A., “Language Maintenance and Language Shift,” 62 Google Scholar.

³² Tsentral'noe statisticheskoe upravlenie, Ilogi vsesoiuznoi perepisi naseleniia 1959 goda, 16 vols. (Moskva: “Gosstatizdat,” 1962–1963)Google Scholar. There is one volume for the USSR as a whole and one for each Union Republic.

³³ Specifically, in the North Caucasus one finds the bi-national Chechen-Ingush ASSR, Kabardino-Balkar ASSR, and Karachaevo-Cherkess AO, as well as the multi-ethnic Dagestani ASSR, for which six nationalities have been included in the analysis: Avarians, Darginians, Laks, Lezghians, Kumyks and Nogais.

³⁴ The Germans and Crimean Tatars did have national territories until the Second World War, at which time the territories were abolished and the groups deported to Siberia and Central Asia. The Jews nominally have their own Jewish Autonomous Oblast in the Far East as a cultural homeland, but the population of this province, which was formed in the late 1920's (and formally designated the Jewish AO in 1934) has never been dominated numerically by Jews and can hardly be considered a homeland of Soviet Jews. For a description of the founding of the Jewish AO and its population makeup, see Solomon Schwarz, , “Birobidzhan: An Experiment in Jewish Colonization,” in Russian Jewry: 1917–1967, ed. Aronson, Gregor et al. (New York: Thomas Yoseloff, 1969), pp. 342–395 Google Scholar.

By excluding groups having no official homeland one is in effect controlling for a significant variable in the Soviet context: the amount of official recognition granted the groups, which is associated with the granting of native-language schools, publications, and representation within the state and Party hierarchies. Simultaneously one is also focusing attention on population groups that are still relatively compactly settled within a traditional homeland, a factor that also hinders Russification.

³⁵ Moreover, each nationality is treated as if it were equal in size and importance to others, and each subcategory of the population is treated as if it, too, were equal in size with others defined by the independent variables. The groups are not weighted according to population size because of the inordinate population sizes of a few groups such as the Ukrainians (37 million in 1959), Belorussians (8 million), and Uzbeks (6 million), which together account for 60 per cent of the non-Russian population. The proliferation of small nationalities in the analysis might have been avoided if one wanted to focus on the larger or politically more important groups. However, since this is the first systematic statistical analysis of these data, and since the census data were coded as part of a larger project examining various aspects of ethnic change in the Soviet Union, it was thought desirable to include all ethnic groups for which substantial statistical data could be gathered in both census and non-census sources. This meant talcing all groups whose official national areas have Autonomous Oblast status or higher, as well as all major groups officially recognized as native to these regions.

³⁶ For a description of the use of dummy variables, see Cohen, Jacob, “Multiple Regression as a General Data-Analytical System,” Psychological Bulletin, 70 (1968), 426–443 CrossRefGoogle Scholar; and Rao, Potluri and Miller, Roger L., Applied Econometrics (Belmont, Cal.: Wadsworth, 1971), pp. 88–99 Google Scholar.

³⁷ On the justification for using unstandardized rather than standardized coefficients see Blalock, Hubert M. Jr., “Causal Inferences, Closed Populations, and Measures of Association,” American Political Science Review, 61 (March, 1967), 130–136 CrossRefGoogle Scholar.

³⁸ There are substantial data available pertaining to levels of education and the distribution of professional manpower among the nationalities, but these data are not directly cross-tabulated with language usage, and employment of them could lead to faulty “ecological inferences.”

³⁹ All “averages” given in this paper are computed on the basis of group-level data. In addition, unless stated otherwise, all such figures will be based on the total N of 46 nationalities.

⁴⁰ According to the 1926 census report, 16.9 per cent of the rural Altais were Russified, while 7.9 per cent of the Khakass ruralites were Russified. These figures are calculated from the 1926 census report, Vsesoiuznaia perepis” naseleniia 1926 goda, 56 vols. (Moskva: Ts. S.U., 1928–1933)Google Scholar. For figures extracted or derived from the 1926 and 1959 census reports page numbers will not be cited because it would have been far too cumbersome to record the pages from which figures (or bases of figures) for each nationality were derived.

⁴¹ Weinreich, Uriel, Languages in Contact: Findings and Problems (The Hague: Mouton & Co., 1953), p. 94 Google Scholar. I have not undertaken a detailed examination of generational differences in Russification because of certain imperfections in the reported data. All age-specific figures on language usage pertain to the somewhat loosely defined “areas of primary settlement” of the ethnic groups. As a rule, these areas are congruent with neither the official national areas nor the entire USSR. As a result, the age-specific figures tend to confound the effects of geographical mobility with other factors, and it is impossible to employ the age-specific figures in conjunction with the “exposure to Russians” variable (except for a few groups for whom the recorded “area of primary settlement” happens to coincide with the official national area).

⁴² Since the addition of any new variable to a regression model is likely to increase the amount of variance accounted for (the R ²), the statistic is generally more appropriate than R ² for assessing the incremental change in the variance accounted for by the addition of a new variable to an equation. is derived from R ² but includes a correction or adjustment to R ² depending upon the number of cases and variables in the equation. In effect, the adjustment penalizes the researcher who is working with a small N or a large number of regressors (relative to the N). On the derivation and justification for , see Rao, and Miller, , Applied Econometrics, pp. 20–21 Google Scholar.

⁴³ See Gordon, Robert A., “Issues in Multiple Regression,” American Journal of Sociology, 73 (March, 1968), 592–616 CrossRefGoogle Scholar.

⁴⁴ This “interaction model” might be constructed by creating a new interaction term equal to the product of “exposure” (X ₃) and “urban-rural” (X ₁) and adding it to Equation 2. It is also appropriate, however, to split the sample and to run separate regression analyses on the relevant population sub-groups (here, urban and rural). See Rao, and Miller, , Applied Econometrics, pp. 148—150 Google Scholar; and Melichar, Emanuel, “Least-Squares Analysis of Economic Survey Data,” American Statistical Association, Proceedings of the Business and Economic Statistics Section, 1965, p. 379 Google Scholar.

⁴⁵ The product-moment correlation between Russian exposure and linguistic Russification is .37 for the urban and .57 for the rural groups.

⁴⁶ Note that the lower impact of Russian exposure on Russification in rural areas is gauged by the relative sizes of the unstandardized regression coefficients, not the standardized coefficients. Indeed, the correlation between Russian exposure and Russification is higher in the rural than in the urban areas. Consequently, Russian exposure accounts for a greater proportion of the variance in Russification of rural populations (32.4 per cent) than of urban populations (13.7 per cent).

⁴⁷ Rao, and Miller, , Applied Econometrics, pp. 35–38 Google Scholar. For a similar argument see also Wonnacott, Ronald J. and Wonnacott, Thomas H., Econometrics (New York: Wiley, 1970), pp. 65–66 Google Scholar. We should add that the contribution of the male-female variable to the over-all R ² is extremely small. Nevertheless, the differences highly consistently support the hypothesis: among rural populations, men are more Russified than women in 36 of the 46 cases, while in the other 10 cases the male-female differences are negligible (less than one-tenth of a percentage point); among urban populations, men are more Russified than women in 34 of the 46 cases, while (contrary to expectations) women are more Russified than men in 9 cases, and in the 3 remaining cases the differences are negligible. These totals are subject to a certain amount of rounding error in computing of percentages.

⁴⁸ The special susceptibility of (former) Orthodox groups to Russification (upon becoming mobilized) has been noted by Armstrong, , “The Ethnic Scene,” p. 16 Google Scholar.

⁴⁹ The product-moment correlation between “Muslim-non-Muslim” and “exposure to Russians” is .06 for urban residents and —.21 for rural.

⁵⁰ John A. Armstrong has labeled the Baltic nations, along with the Georgians, Moldavians, Finns and Poles, as “state nations”: “nationalities with strong traditions of national identity, including distinctive languages, well-developed cultures, and distinctive historical traditions” ( Armstrong, , “The Ethnic Scene,” p. 21 Google Scholar). Thus, the Baits are likely to maintain an ethnic ideology strongly distinguishing themselves from Russians.

In fact, a more general case may be made that groups with Union Republic status should be better able than groups of lesser status to preserve their ethnic identities because of (a) their relatively greater access to native-language schools and other cultural facilities; (b) their (usually) relatively rich and long-established literary traditions and histories of independence from Russian rule; (c) their relatively large population and territorial size, their location on the periphery of the USSR, and their social and economic diversity, which give them more of a claim to nationhood; and finally, (d) the prestige associated with being termed a “nation.” This greater ability of Union Republic nationalities to preserve their identities has been proposed by Vardys, , “Communism and Nationalities,” 13 Google Scholar, and has been suggested by Armstrong for the Ukrainian case. See Armstrong, John A., Ukrainian Nationalism, 2nd ed. (New York: Columbia University Press, 1963), p. 306 Google Scholar. Avtorkhanov also distinguishes the Union Republics, where native cadres tend to predominate numerically in leadership positions, from the Autonomous Republics and Oblasts, where non-indigenous personnel predominate. The former are termed “self-governing” republics; the latter, “governed.” See Avtorkhanov, Abdurakhman, The Communist Party Apparatus (Chicago: Henry Regnery, 1966), p. 173 Google Scholar. Although the argument here would suggest the importance of examining the impact of “official status” on ethnic identity change, this factor is not treated here because, from a statistical standpoint, the test of the “impact of status” is extremely complex.

⁵¹ The cases of the three Buddhist groups may each have different explanations. The Tuvinians entered the USSR only after World War II and therefore have not had a long time for Russianization to have an impact. The Buriats, on the other hand, do show a relatively high level of Russification (12 per cent of the urban population) and fall very nearly at the predicted level of Russification. Certainly in contrast to the Muslims with equally high levels of Russians exposure, they are rather highly Russified. The Kalmyks are one of the nationalities banished by Stalin for alleged collaboration with the German invaders in World War II. Although I shall not attempt to explain their high resistance to Russification, it should be noted that they are barely Russified despite the fact that of all the groups examined here, as of 1970 the Kalmyks showed the highest level of knowledge of Russian as a second language. See “Census Data: Age, Education, Nationality,” The Current Digest of the Soviet Press, 23, No. 16 (May 18, 1971), 16–17 Google Scholar.

⁵² Even these rough comparisons can only be suggestive, however, since one might still pose the argument that it is the combination (interaction) of duration and extent of contact that accounts for the results.

⁵³ There have been a number of such reports concerning attitudes of Tatars and Russians towards mutual collaboration at the workplace, intermarriage, and inter-ethnic friendships. See, for example, Arutiunian, Iu V., “Opyt sotsial'no-etnicheskogo issledovaniia (po materiialam Tatarskoi ASSR),” Sovetskaia etnografiia, 1968, No. 4, pp. 3–13 Google Scholar; and Shkaratan, O. I., “Etno-sotsial'naia struktura,” pp. 3–16 Google Scholar. Brief reports of similar studies among other nationalities are to be found in Drobizheva, L. M., “O sotsial'noi odnorodnosti respublik i razvitii natsional'nykh otnoshenii v SSSR,” Istoriia SSSR, 1967, No. 1, pp. 67–82 Google Scholar; and Drobizheva, L. M., “O sblizhenii urovnei kul'turnogo razvitiia soiuznykh respublik,” Istoriia SSSR, 1969, No. 3, pp. 61–79 Google Scholar. An extraordinary wealth of relevant data is presented also in Kholmogorov, Internatsional'nye cherty.

⁵⁴ The figures and generalizations presented here are not meant to be definitive. In individual cases, in fact, the increment in Russification since 1926 has been negligible and perhaps even negative. Groups such as the Ukrainians and Belorussians, for example, appear to be less Russified in 1959 than they were in 1926 within the urban and rural populations taken separately. The methodology for making longitudinal comparisons of ethnic assimilation is very complex and needs more attention. Particular problems are presented by such factors as the changing borders of the Soviet Union, the disbandment of certain ethnic territories between census periods, changing census definitions of nationality and native language, and changing criteria for classifying distinctive ethnic groups in the census reports. For a brief discussion of the last problem, see n. 59, below. In the calculations presented in Table 9, only that portion of the nationality officially designated with the given national label in the census report has been included in the tabulation. Thus, for example, in calculating figures for the Volga Tatars in 1926, the Mishars and Kriashens are excluded; but in the 1959 totals, Soviet statisticians have combined the Mishars and Kriashens with the Tatars, and Table 9 therefore relies upon the combined totals.

⁵⁵ Of course, even if the percentage of natives Russified remains constant, the absolute numbers may increase enormously because of population growth. However, the current aim is not to examine such numbers but to attempt to understand the processes of Russification by measuring the probability that a given non-Russian will be Russified under specified conditions (e.g., residence in an urban or rural setting).

⁵⁶ The present measures of the Russification effect of urban residence are still imperfect, however, because the levels of Russification among urbanites of different nationalities are dependent partly upon the rates of recent migration of natives to the cities. Where recent urban immigration rates (of largely unrussified rural natives) have been high, one would expect the level of Russification displayed by urban residents to be low; where recent immigration rates have been low, the level of Russification of the urban residents should be high. Without information on the migration rates of specific nationalities, one cannot attempt to compensate for this factor. The 1970 census results will offer some information on this when they become available.

⁵⁷ Kozlov, V. I., Dinamika chislennosti narodov, p. 29 Google Scholar. The quotation marks are in the original.

⁵⁸ See Conquest, Robert, The Nation Killers: The Soviet Deportation of Nationalities (London: Macmillan, 1970)Google Scholar.

⁵⁹ Some of this assimilation may be more apparent than real. For example, the 1959 census report does not list the Adzhars, Mingrelians, Svanians, and Laz, all of which were listed in the 1926 report. However, even if an individual named one of these titles as his subjective nationality in 1959 he was recorded in the census report as a Georgian. Similarly, self-designated Mishars and Kriashens, listed separately in 1926, were counted as Tatars in 1959. These cases of “assimilation” are probably more apparent than real because they result at least in part from administrative decisions to combine certain groups. In fact, census enumerators were provided in 1959 with a “Glossary of Nationalities and Languages” listing 733 ethnic titles, which were then combined into 126 titles in the census report. Unfortunately, figures on the populations of the 733 self-designated groups have not been revealed. On this matter see Isupov, A. A., Natsional'nyi sostav naseleniia SSSR (po itogam perepisi 1959 g.) (Moskva: “Statistika,” 1964), pp. 11–13 Google Scholar.