8 Why the Peripheral Peoples Did Not Become Russians David D. Laitin In the nineteenth century Alexis de Tocqueville accurately foresaw the clashes of civilization that would mark Russian society for a century and a half: There are, at the present time, two great nations in the world, which seem to tend towards the same end.... I allude to the Russians and the Americans. Both of them have grown up unnoticed; and whilst the attention of mankind was directed elsewhere, they have suddenly assumed a most prominent place amongst the nations. ... All other nations seem to have nearly reached their natural limits, and only to be charged with the maintenance of their power; but these are still in the act of growth. ... The American struggles against the natural obstacles which oppose him; the adversaries of the Russian are men; the former combats the wilderness and savage life; the latter, civilization with all its weapons and its arts: the conquests of the one are therefore gained by the ploughshare; those of the other, by the sword. The Anglo-American relies upon personal interest to accomplish his ends, and gives free scope to the unguided exertions and common sense of the citizens; the Russian centres all the authority of society in a single arm: the principal instrument of the former is freedom; of the latter, servitude. Their starting-point is different... yet each of them seems to be marked out by the will of Heaven to sway the destinies of half the globe.[1] Manifest destiny, Russian style, did not foster the construction of a new continent-wide identity (such as “American”). To be sure, attempts to construct an inclusive “Russian” identity in the nineteenth century, and a “Soviet” identity in the twentieth, had some success. But within the boundaries of the Russian empire (and the Soviet Union), linguistic diversity in particular and national diversity more generally remained and even prospered. The incomplete rationalization of language and culture within the boundaries of the Russian empire provides the historical context for this book. But this chapter – in going over well-trodden historical fields – seeks as well to demonstrate the contingency of this outcome. I seek to show how Russian (and Soviet) rationalization was possible but unsuccessful. The subsequent rationalization projects of four of the former Union republics resulted (or have as yet failed to result in) a strategic turning of the cultural tide rather than the inevitable (in the Hegelian sense) fulfillment of four national dreams. We cannot assume that all states will become nation- states by historical necessity, a myth that underlies many of the post-Soviet national projects – but with historical perspective it will be possible to analyze the likelihood that these four will. Russian Expansion to the Continental Peripheries From the end of the fifteenth century, with Ivan the Terrible’s conquest of Kazan and Astrakhan, through the end of the nineteenth, with the conquest of the khanates of Kokand, Bokhara, and Khiva and the annexation of the Transcaspian region, the Russian empire expanded at a rate of some fifty square miles per day.[2]^ Eastern Ukraine (the left bank of the Dnieper) came under tsarist protection in 1654. In the early eighteenth century after victory over Sweden, Peter the Great took the territory that is today’s Estonia and Latvia. Throughout the eighteenth century, Russian trade and state control moved into what is most of today’s northern Kazakhstan and annexed it. Like the Habsburgs in Spain and the Bourbons in France, the Romanovs did not make formal distinctions between their governance in what Pipes calls “Russia proper” and in its “imperial hinterlands.”[3] To be sure, the western end of empire was marked by “boundaries” (rubezhy, or granitsy) while the east was marked by a frontier (mezha). These terms were used conventionally and represented a clear distinction in imperial expansion. In the east, expansion was through peace treaties (shert’, a Turkic word), which were conceived of by Moscow as pledges of allegiance of “eternal slavery to the grand tsar.” The main mechanisms of rule were through hostages. Tribute in furs was expected; the tsar made exchange through gifts (gosudareevo zhalovan’e, or sovereign’s compensation).[4] Despite these different vocabularies of rule, tsarist lands east and west were divided into provinces (gubernii), which were ruled by governors general or viceregents. All rules were equally valid throughout the empire, were written in Russian, and were administered in a unified bureaucratic system of classified offices. Furthermore, nobles from non-Russian lines were given rights similar to those of the Russian nobility, and russified foreigners were quickly recruited into state service.[5] Russian-speakers emigrated to the new tsarist territories in waves. From the earliest period of territorial incorporation, the state recruited a quasi-independent military caste, the Cossacks, and entrusted them with the task of protecting Russian settlements in the borderlands, and with protecting the boundaries of the empire from foreign predators. For this, after twenty years of service, they were given land to till, and they became moderately wealthy landowners. By the beginning of the twentieth century, 4.4 million people living outside Russia proper traced their origins to Cossack settlement. Peasant migrations from Russia and Ukraine to the south and west were a second major source of settlement into imperial territories. The lust for the black earth, especially by freed serfs who were unable to cover their redemption payments, led at first to temporary migrations to the cities to earn cash, and after 1906 (when redemption payments were canceled, railroad transportation was available, and formal travel restrictions had been removed), to the rapid prerevolutionary colonization of Central Asia and the eastern steppes.[6] Traders, artisans, and skilled workers moved inexorably to the frontier through much of the second half of the nineteenth and the early twentieth centuries. By the 1897 census, those characterized as Russians constituted 8.5 percent of the population (and over a quarter of the urbanites) outside the boundaries of what was to become the Russian Federation.[7] State and Nation-Building in Russian History This tale of expansion appears no different from that of France into Languedoc or England into Wales.[8] Those expansions are without question thought of as examples of early state building. Yet today (with the knowledge of the Soviet collapse in 1991), we think not of France (vis-à-vis Languedoc) or England (vis-à-vis Wales) as the proper framework for understanding Russian expansion, but rather of the Ottomans and the Austrian Habsburgs (where the states shed their separate national components). It is true that in comparison with the expansions of England and France, Russia’s cultural impact on the indigenous populations was quite limited. Linguistically, Welsh and Languedocians by the end of the nineteenth century were assimilated into the dominant state language; meanwhile Baits, Kazakhs, and even Ukrainians relied principally on the language of their forefathers; very few had developed fluency in Russian. What explains the incomplete incorporation of peripheral subjects into the Russian state-building project? Russian state building was not all that unlike its Western counterparts. In fact, Russian tsars since Catherine II saw rationalization of the Russian language – that is, its standardized use in all official and quasi-official domains – throughout their empire as an important ingredient of state building.[9] This is not simply a function of “Great Russian chauvinism.” Catherine II, after all, was a German princess. And for generations, the Russian nobility communicated with one another in French. Rather it was a part of a program “to extend [the legal] administrative system into the countryside.”[10] Catherine II wrote (in 1764) that Ukraine, the Baltic provinces, and Smolensk should russify “and cease to look like wolves in the forest,” implying that russification would lower the chances of political subversion. Surely this is why she promoted Russian most actively in the Polish provinces, where loyalty was of the greatest concern. The nineteenth century is a story of toleration for language diversity (under Alexander I) mixed with periods of promotion of an “official nationality” (under Nicholas I). Yet the logic of rationalization – especially in response to foreign threat – regularly appeared on the tsarist agendas. Nicholas’s response to the Polish rebellion (1830-31) was to demand a fusion of the languages spoken in the Polish- influenced areas of the empire – a mix of dialects that were much later formalized as Byelorussian and Ukrainian – with Russian.[11] Alexander II put down a rebellion in Poland and Lithuania in 1863 and subsequently sought to limit the use of Polish and Lithuanian. In the face of Polish aristocrats’ courting of Ukrainians peasants with an eye to possible incorporation of western Ukraine into a restored Poland, Alexander II issued the Ems Ukaz (1876) prohibiting inter alia the import of Ukrainian-language books and the teaching of the Ukrainian language. Rationalization continued under Alexander III and Nicholas II. Alexander III reversed earlier efforts by Catherine II to rule the eastern provinces through a unified Turkic tongue. In the eighteenth century, Catherine induced the Tatars to settle in the steppe area of today’s Kazakhstan, and Tatar became the official language of imperial administration. When a Kazakh became a clerk, he had to write in Tatar, which was of the same family but a distinct language nonetheless. By the mid-nineteenth century, egged on by Nikolai Il’minskii, a Russian Orthodox lay missionary (and linguist), who feared that the Tatars represented a threat to Orthodox rule of Central Asia, Alexander III ordered instruction in the schools to be conducted in Russian, effectively banning the Tatars from teaching in the school system.[12] Later, Nicholas II promoted Russian in the administration of Finland, fearing that the continued use of Finnish would make their administrative incorporation into Russia less secure.[13] To be sure, the tsars were not entirely successful in their russification program. But the point here is that they perceived an administrative and security advantage in having a single official language and sought to change the language repertoires of officials in incorporated territories so that Russian would become predominant for official uses. While they often faced counter-pressures, there is little doubt that except for Alexander I, the tsars tried to enhance the role of Russian whenever they had the chance. Perhaps (as Tocqueville implies) the elites in the Russian periphery were more resistant to russification than the lords ruling over the incorporated regions of Western European states? One reason might be that the lords in the periphery of the Russian empire could seek the protection of the rulers of other states. Georgian elites, for example, wavered between the tsars and the rulers of Iran and Turkey.[14]^ But the lords of Catalonia, Toulouse, and Alsace had options to negotiate with more than one central leader, and did so. Once a state establishes effective administrative rule, as was the case in France and Spain, peripheral elites might hope for systemic breakdown, but they cannot bargain their way out of the central state. The situation for lords in Russia’s periphery was therefore similar to that faced by lords in other states’ peripheries. Historical evidence suggests that Russian rulers paid a higher cost for compliance with language-rationalization legislation than the rulers of other multinational societies did. But the outcomes were not wholly different. Evidence from the Baltics, Georgia, and Kazakhstan demonstrates the considerable pressure that regional elites faced to pay the transactions costs in communicating with the center by learning Russian. In the Baltics, despite the attractions of German for ambitious Estonians and Latvians, the rising classes (in the mid-nineteenth century), as well as the German nobles (in the late nineteenth century), began making concerted efforts to learn Russian. Edward Thaden reports that the Baltic representatives in the first two dumas (mostly Baltic peoples) and the second two dumas (mostly from the Ger man nobility) all spoke Russian fluently.[15] Ronald Suny has similar findings in his work on the Georgian nationalist intelligentsia in the mid-nineteenth century. Dmitrii Kipiani, one of its luminaries, wrote primarily in Russian, including his memoirs.[16] By the end of the century, 91 percent of the schools in the Caucasus relied on Russian as the sole medium of instruction. Both the intelligentsia and those co-opted by the Russian state apparatus had become fluent in Russian.[17] Martha Olcott’s history of the Kazakhs tells a similar story. After 1870, aristocrats and elders sent their children to Russian schools “in order better to represent their people upon assuming their fathers’ positions.” That generation wrote exclusively in Russian.[18] The regional elites in the Russian empire, this evidence demonstrates, had a strong incentive to invest in Russian-language competence. If language rationalization occurred on the peripheries of the Russian empire as it did in the continental peripheries of Western European states, a major difference between the cases is that the broader societal ramifications of rationalization were different. Although France, Spain, and England would face language-revival movements in their peripheries in the 1970s, it is fair to say that a dominant language had achieved quasi-hegemony by World War I. But Russia could only be described as an aggregation of nationalities where co-opted elites and a small intelligentsia in various provinces had facility in Russian. Why was the expansion of Russian so limited, and why did the ultimate success of rationalization among the elites not motivate successively lower strata of the populations to learn Russian?[19] Scholars point to two crucial factors to account for these differences in outcome. First is the geography of the open steppe.[20] The Russian countryside was always open to marauding bands of conquerors. One consequence, it is sometimes pointed out, is a far greater cultural heterogeneity than in the West. A second consequence was that the Muslim hordes brought such fear to the Russian settlers that cultural intermingling rarely took place. Muslims were considered to live on the other side of a divide that was unbridgeable for Europeans. A third consequence of the open steppe – and this adumbrated by Anthony Smith[21] – is that unlike the state builders of Western Europe, whose expansionary appetites were constrained by natural barriers, the Russian tsars could continue to expand without limit. Their ability to assimilate such a gargantuan space would be beyond even a Napoleon. Although the expanse and heterogeneity of the empire cannot be denied, I am reluctant to rely too heavily on this explanation. First, Eastern expansionary appetites are probably exaggerated in Western historiography. Second, cultural differences among peoples in today’s successfully consolidated nation-states would have been emphasized more strongly, had their national projects failed. A second reason why rationalization in language did not quickly penetrate to the lower strata is that state rationalization occurred later than it did in Western Europe, which put new constraints on leadership.[22] The early state rationalization laws in France were in place by the early sixteenth century; in Tudor England rationalization of language was in full development in the late sixteenth century. In Spain, the essential decrees were passed in the early eighteenth century, but the pressure on lords to learn Spanish occurred much earlier.[23] Comparable legislation did not occur in Russia until the mid-nineteenth century, in a world-historical period of mass literacy, which was itself a factor inducing language-revival movements among incorporated nationalities throughout the world. Under these circumstances, newly instituted mass education policies relying on the language of state rationalization more easily engendered popular resistance. The spread of a state language, under novel historical circumstances, faced new and powerful popular obstacles. We get a clear sense of language politics for late developers by examining the career of Il’minskii, whose activities in mid-nineteenth-century Kazan coincided ideologically with Nicholas I’s promotion in 1833 of “official nationality.” Just like his French and English “colleagues” in Africa, Il’minskii was faced with the problem of teaching “natives” basic literacy. He found that it is easier to fulfill educational goals (and to reach students’ souls) in the mother tongues of the students, and therefore helped develop written forms for a variety of Asian tongues. To be sure, as already mentioned, Il’minskii’s work had russifying elements. He relied on the Cyrillic alphabet as a means of promoting Orthodox values, and to wean Asians away from the Arabic script with its links to Islam. Also, in his schools, while the local languages were taught as subjects, the language of instruction after the second year of primary education was Russian.[24] Yet we see in Il’minskii’s career a problem for nineteenth-century state builders. Since mass literacy was becoming an essential aspect of the “state function” by virtue of the needs of increasingly bureaucratized business firms and the state itself (soldiers, clerks, and others needed to send and receive written messages), virtually all states had Ministries of Education with mandates to provide trained personnel to fulfill these new functions. As these ministries sought to widen the scope of educational activities, the issue of the language of instruction in the mass public school became relevant for the first time.[25] There remained, however, pressures and counterpressures in the educational establishment. On the one hand, the lessons of the missionaries had considerable influence in state educational circles. As early as 1879, the Russian Ministry of Education issued regulations authorizing “elementary schools with Volga languages as media of instruction.”[26] On the other hand, in the 1880s, I. D. Delianov, the minister of education, strongly favored russification; and his successor, D. A. Tolstoi (who had earlier russified the Polish educational system), fought to get the Baltic educational system out of Lutheran (with German as the medium of instruction) hands through the promotion of Russian. Yet Tolstoi also supported Il’minskii’s efforts to promote native-language education in the east, suggesting that the Ministry remained cross-pressured on this issue. Economic planners also began to speak, albeit haltingly, for the recognition of peripheral languages. N. Bunge, minister of finance in the 1880s, supported tolerance of the Baltic languages in order to permit economic growth without provoking nationalist disturbances; and Count Witte in the 1890s supported German education in the Baltics in order to get competent and loyal economic managers.[27] When mass education and government economic management arrived in states that had consolidated in earlier eras, there was already an elite stratum among nearly all language groups – children of state functionaries or commercial bourgeois families seeking “national” markets – that was capable of teaching in and interested in spreading the language of the central state. In Russia, the historical fact of starting late changed the context of language rationalization. Built into the state apparatus was a core institution (the Ministry of Education) that was in the business of standardization and development of languages of the periphery. Also, with the state interested in the management of industrial firms, its functionaries perceived (especially in the regions designated as growth nodes) language rationalization as a threat to development. In light of these two variables – the special geographical context of Russian state expansion and the fact that Russia did not begin building a consolidated state until so late in its history that state interest in linguistic rationalization conflicted with state interest in mass literacy and economic growth – we see that at the time of the Revolution, despite successful language rationalization among peripheral elites, the Russian language was not a core part of the language repertoires of many social strata in the periphery. Whereas in the final third of the nineteenth century, peasants had already become Frenchmen,[28] in Russia, although peripheral nationality groups had many elites who were capable Russian-speakers (fulfilling the rationalization program), their peasants (and members of other strata as well) had not become Russian. Russification in the Soviet Period The legacy of limited language rationalization was not substantially altered during the seventy-four years of Soviet rule, in spite of unremitting state centralization. The startling fact among the nationalities of the Soviet Union was that “unassimilated bilingualism” remained the widespread and stable language repertoire.[29] To be sure, there were exceptions. In eastern Ukraine, Belarus’, and in a number of industrial cities outside of Russia, there were unambiguous trends toward full assimilation. Meanwhile, in rural Central Asia, there are large rural pockets where parochialism remains the linguistic norm. Data collected by Martha Olcott and William Fierman suggest that in the Asian republics, vast numbers of youth have no functional knowledge of Russian. Finally, in some areas such as in Armenia and in the Baltics there has been a deliberate deemphasis of Russian, even to the extent of (based on census figures) intergenerational loss of proficiency. From a Russian or Soviet specialist’s point of view, these exceptions carry great weight. From a comparative perspective, however, the variations are small compared to the general out- come of stable “unassimilated bilingualism.” Brian Silver’s analyses of Soviet census data suggest that the level of intergroup contact was sufficient to explain variation in the move from parochialism to unassimilated bilingualism.[30] The Muslim/Orthodox variable and the degree of urbanization, however, had little explanatory power. Silver suggests from this that the acquisition of Russian as a second language is almost entirely a matter of economic and practical consideration, with the ethnic significance of this language step playing only a minor role.[31] As for the switch from unassimilated bilingualism to assimilation, Silver’s data show a clear difference by cultural group. With high levels of contact with Russians, non-Russian but Orthodox nationalities (which include the ambiguous case of the Ukrainians) move in a monotonic way toward assimilation. For the Muslim, Armenians, and Baltic nationalities, both contact and high levels of urbanization are necessary for a full switch, and even in those cases the level of switch is very low. More precisely, Silver found that the switch from original language to Russian was higher than predictions based solely on contact and urbanization in 14 of the 17 Orthodox ethnic groups, lower than expected in 22 of 23 Muslim groups, and lower for the three Baltic groups.[32] Marginal differences in region aside, the overwhelmingly clear outcome (based on data from the 1970 and 1979 Soviet censuses and ignoring the more nationally charged environment of the 1989 census; see Table 2.1) is that of stable unassimilated bilingualism, especially among the “titular nationalities,” those language groups that gained control over Soviet republics.[33] As E. Glyn Lewis puts it, perhaps too forcefully, the data on retention of national languages, “make nonsense of the claim that... it is possible in the foreseeable future to envisage a merging of languages or the creation of a common language.”[34] An explanation for the maintenance of unassimilated bilingualism in the Soviet peripheries needs to account for how the titular national elites successfully consolidated local power through a linguistic regime under their control.[35] To be sure, nationality policy in the early Soviet years focused on the linguistic rights of the individual, not the territorial republic of settlement. In many areas, this policy continued to define the Soviet educational mission. By 1938-39 I. T. Kreindler points out that Uzbekistan offered instruction in twenty-two languages; Ukraine in seventeen, and Dagestan in twenty.[36] Table 2.1. Unassimilated and assimilated bilingualism of titular nationalities. 1970-1979 – see next page Percent claiming Russian Percent switching native as second language language to Russian . 1970 1979 1970 1979 Turkmen 15.4 25.4 1.1 1.3 Kirghiz 19.1 29.4 1.2 2.1 Uzbek 14.5 49.3 1.4 1.5 Tajik 15.4 29.6 1.5 2.2 Georgian 21.3 26.7 1.6 1.7 Azerbaijan 16.6 29.5 1.8 2.1 Kazakh 41.8 52.3 2.0 2.5 Lithuanian 35.9 52.1 2.1 2.1 Estonian 29.0 24.2 4.5 4.7 Latvian 45.2 56.7 4.8 5.0 Moldovan 36.1 47.4 5.0 6.8 Armenian 30.1 38.6 8.6 9.3 Ukrainian 36.3 49.8 14.3 17.2 Byelorussian 49.0 57.0 19.4 25.8 Note: These data include members of each nationality living republic and therefore overstate the switch to Russian for titulars living outside their republic and therefore overstate the switch to Russian for titulars living in “their” republics. Source: Tsentral’noye Statisticheskoye Upravleniye pri Sovete Ministrov, SSSR, Itogi vsesoyusnoy perepisi naseleniia 1970 goda (Moscow: Statistika, 1973), 4:20-319, and Chislennost’ i sostav naseleniia SSSR: Po dannym vsesoiuznoi perepisi 1979 goda. (Moscow: Finansy i Statistika, 1984), pp. 71-137. But a very different policy arose from seeds planted in the People’s Commissariat for Nationalities, the Treaty with Union Soviet Republics (December 1922), and the first constitution of the USSR (1924), where the political relations between nations were organized along strictly territorial principles, and within each republic, political advantages accrued to the titular nationalities at the expense of minorities. These decisions had far-reaching effects. As Lewis puts it:[37] In consequence, whereas before this decision 66.5% of the total Uzbek population of Central Asia lived within the Turkestan ASSR, 22.2% and 11.3% within the Republics of Bukhara and Khorezm respectively, after the delimitation of 1925 over 82% of all Uzbeks in Central Asia were concentrated in Uzbekistan. The Turkmen population were originally even more dispersed, only 43% living within the Turkestan Republic, 27% and 29.8% in the Bukhara and Khorezm Republics respectively. After delimitation of territories over 94% of all Turkmen were brought together. . . . Naturally the concentration helped to ensure ... the greater linguistic homogeneity of the various republics. The Uzbeks came to constitute nearly 75% of the population of the Republic, Turkmen over 70% of the population of the Turkmen SSR, and in the case of Tajikstan and Kirgistan the national group in each constituted over 74% and 66% respectively. Such unification and increased homogeneity made the task of providing vernacular education and literature much easier, and so promoted the non-Russian national languages. Thus began a system, originally sold as a policy of korenizatsiia (nativization). Its origins were in the Twelfth Party Congress in 1923, when Great Russian chauvinism was declared a greater danger than local nationalism. The campaign, lasting more than a decade, not only gave special rights to minorities within both Russian and non-Russian regions through the creation of “national Soviets” (a policy that had no long-term legacy), but also (and with a profound legacy) gave considerable leeway for national elites controlling titular republics to promote their national cultures. In one interpretation, korenizatsiia “provided opportunities for nationalities representing over 93% of the non-Russian population to create ethnically distinct stratification subsystems within Union or autonomous republics.”[38] Within the distribution politics of each republic, based in part on access to linguistic capital, the titular nationals used their positions to assure their ethnic brethren the more visible jobs of power and patronage;[39] they were (in the Asian republics especially) the beneficiaries of affirmative action programs for educational placement and technical jobs;[40] they had subsidized publications in their languages;[41] and they used the lack of language competence to deny minorities within their republics access to educational and job opportunities.[42] In light of these policies, children of mixed Russian/titular marriages, when living in the titular republic, often declared themselves as members of the titular nationality.[43] And, after decades of russianization (that is, the migration of Russians into the titular republics), the titular elites were by the 1980s able to reverse the tide. In the late Soviet period, therefore, there was significant migration of nontitulars out of the republics, and of titulars from outside back into their “home” republics.[44] The titular elites took responsibility for managing ethnic relations within their republics in order to retain control of a vast “neotraditional” patronage system.[45] To be sure, not all titular nationals were pleased; there is some evidence that many titular nationality parents wanted more Russian education than “their” elites were willing to provide.[46] Yet as long as the titular leaders could assure Moscow that there would be peace in the republic, they were, for decades, able to sustain neotraditional political structures. This system permitted the Soviet elites to promote “primordialism” but to severely limit its mobilization into direct confrontations with Soviet power.[47] Primordialism, in the Soviet context, could be promoted by allowing nationalities to use their “own” languages, just so long as they did not make political demands on the basis of nationality on the central state. But with titular languages used not only for cultural expression but for republic-level administration as well, titular nationals had an incentive to remain unassimilated bilinguals. Their languages represented capital for jobs and opportunities. The question that remains puzzling is why would a regime that was so centralized support and even protect titular languages? Some might suggest that the Leninist ideology of national self-determination became enshrined in the organization of language zones controlled by titular nationals. But Lenin’s writings on language and nationality give equal and contradictory regard to the goals of national self-determination and proletarian internationalism. If either of these themes is emphasized at the expense of the other, Lenin’s pronouncements can be invoked in support of a wide continuum of language and nationality policies. Lenin is widely known for his support for the national self-determination of peoples. Yet he can also be cited for his ultimate goal of the “complete Russification of non-Russian nationalities.”[48] In fact, Lenin’s followers used his name to give greater autonomy to the titular nationalities and to make special opportunities available for native-language primary education to the members of nontitular nationalities. Lenin’s writings on nationality could have been used to support both the rationalization of Russian and the promotion of regional languages. There is a macrohistorical as well as a micro-incentive alternative to the explanation based on Leninist ideology. The macro dimension concerns the pacification efforts of the new Soviet state in the period immediately following the civil war, both to preempt pan-Turkism and pan-Islamicism and to expand the revolution westward. Both of these strategies had the consequence of giving republican elites far more linguistic autonomy than an overall strategy of russification would have permitted. The micro dimension brings us to the tipping game, introduced in Chapter I, to explain why there was no cascade toward Russian after accommodating elites attained power helped by their knowledge of Russian. Overcoming Pan-Turkism The awesome power of nationalism shocked Lenin in 1917, and again during the civil war.[49] The territorial integrity of Russia was immediately threatened. In Central Asia, the revolutionaries faced a credible pan-Turkic threat. Pan-Turkism had been crystallized in 1882 when Ismail bey Gaspraly developed a Turkic koine for his newspaper, Terjuman (The Interpreter), that was easily understood through much of Russian-controlled Central Asia. In 1904-5, a political organization, Ittifaq al- Muslimin, advocated pan-Turkic social and linguistic but not political goals.[50] By 1917, at the Pan-Russian Conference of Muslims, delegates assumed – surely ignoring the existence of quite separate Tatar, Azeri, and Kazakh literary languages – that there was in Russia a single Muslim nation with one tongue. The Soviet regime was compelled, for lack of other allies, to deal with these Pan-Turkic ideologues. For example, Najmuddin Efendiev-Samurskii, leader of the Communist Party of Daghestan, published a book in 1924 advocating education in a common Turkic language. Sultan Galiev, the highest-ranking Muslim official in the communist hierarchy, was also a strong advocate of pan-Turkism. These “national communists” found themselves in a good bargaining position, mediating between local cultures and the weak Soviet state, whose leaders thoroughly distrusted their Asian allies. The Bolsheviks probably overestimated the possibility of a mobilized secession from a united Turkic movement. Nonetheless, this fear explains their primary goal to weaken these elites’ claims for a common Turkic culture. The Communist Party’s “divide and rule” policy “involved breaking up the large mass of Muslim and Turkic populace into fragments and then putting the pieces together into the required number of units, each of them having an exact territorial demarcation.”[51] This was not a difficult task, given the dialect, tribal, and social differences that already existed, and the fears of some Central Asian peoples that they might face Volga Tatar domination in the name of pan-Turkic unity.[52] The regime named and subsidized publications in new languages and worked to distinguish them from their dialectical cousins. Examples of dividing close speech forms and calling them separate languages, as was done with Tatar and Bashkir, with Kazakh and Kirgiz, and with Balkar and Karachai, have been well documented. Lenin, whose father was a colleague of Il’minskii, and (perhaps because of that) was very sensitive to the nationality issue, felt that it was important not to alienate the Turkic peoples. Turning them into active supporters of the regime would need to be delayed. He therefore pressed Stalin and others to respond positively to nationalist demands and to challenge “Great Russian chauvinism.”[53] In Central Asia the development of loyalty was especially crucial, given the nature of the Muslim communist elite. In the period of communist consolidation, this elite was still composed of “unreliable” class elements heavily tainted with nationalism; they were tolerated only because of the weakness of the Soviet state. The central party foresaw that long-term stability would depend on replacing this old elite with a new set of cadres from the general population, especially workers. Forcing the Russian language on this population would no doubt alienate the future base of recruitment to an extent that the Bolsheviks could not afford.[54] The politics of orthography in Central Asia nicely illustrates the desire first to block pan-Turkism, and only later to seek loyalty to the Soviet regime. In the 1920s, the Latin alphabet became the new script for the Turkic languages to replace the Arabic script, which might have helped foster a pan-Muslim identity. The Latin script, richer in vowel representation than Arabic, also helped to differentiate dialects that looked the same with Arabic spellings.[55] It was not until the late 1930s, after the First Five-Year Plan had utterly destroyed the so-called Turkish feudal elite, when loyalty to the Soviet regime was no longer a pressing problem, that the conversion to Cyrillic (done through a central dictate, without any linguistic preparation) was made. The Cyrillic script, along with the introduction of Russian root words for key concepts dealing with political and technical areas, helped the project of blending the languages, a precursor, in Soviet thinking, to linguistic rationalization.[56] Thus the primary goal to prevent pan-Turkism was met first by eliminating the Arabic script; the secondary goal of procuring loyalty (or at least not engendering anti-Russian feelings) was met by abjuring the Cyrillic script; the tertiary goal of language rationalization was delayed a decade. By the late 1930s, circumstances had changed; full rationalization was both feasible and desirable. With the pacification of Central Asia, cyrillicization would be possible. At this time, international security issues might have been decisive. Perhaps reflecting the need for Russian-speaking conscripts in the face of the growing security threat from Germany, the Soviet state in 1938 required the teaching of Russian in all non-Russian schools.[57] In the following year, again with the probable linking of state security to rationalization, the Latin script for Central Asian languages was officially replaced by Cyrillic. This would facilitate bilingualism in Russian among Turkic speakers. Turkic elites had their own agenda. At the Pan-Russian Congress of Muslims in May 1917, “the delegates [including traditionalists and radical communists such as Sultan Galiev] expressed near unanimity on the fundamental concern of all factions – that the destiny of the Muslim peoples must be made separate and distinct from that of Russians.”[58] This desire for a separate destiny was held equally strongly by the Muslim National Communists in the period following the October Revolution. They had replaced the traditionalists who dominated the 1917 congress but shared their goals in regard to language.[59] The primary goal was indeed to nurture a Pan-Turkic language. But if this goal could not be reached, they saw a necessity to prevent Russian from overrunning their language(s), and would have accepted separate development of their local languages rather than russification. Given the Soviet fear of pan-Turkism, the Turkic elites saw room for a compromise that would serve their interests. If the representatives of Central Asia were willing to accept the promotion of local languages, they speculated, Soviet authorities would grab the opportunity. Indeed this compromise depicts what transpired in language politics during the period following 1923: the center opted for promotion of local languages while the Muslim elites held on to the pan-Turkic alternative. Alexandre Bennigsen and S. Enders Wimbush describe the historical dynamics in the following terms:[60] The Muslim national communists [‘]... opposition to the linguistic division of the Soviet Turkic world after 1923 was especially pronounced. Seeking at first to have Kazan Tatar declared as the lingua franca of all Turkic territories of the Soviet Union ... in 1926 [they] fell back to a three-region linguistic strategy. Under this plan Kazan Tatar would become the language of the European part of the Soviet Union, Azeri would be the language of the Caucasus, and Chagatay would serve all Central Asia. They also opposed the introduction of Latin and later of Cyrillic alphabets, arguing instead for the universal use of the Arabic alphabet. From 1923 on, the regime sought to reduce the power of the pan-Turkic elite. By 1928, as Stalin consolidated his power, a massive purge severely reduced the leadership ranks of the already weakened pan-Turks. Sultan Galiev was imprisoned and later executed, and anyone suggesting a pan-Turkic agenda was accused of “Sultangalievism’’ and suffered a similar fate.[61] The purge effectively eliminated the pan-Turkic (or even the three-region) option, but there was still a local alternative to russification. The Muslim elite, recognizing the fact that the Party could now insist, without opposition, on the local variants of Turkic, accepted this as certainly better than russification. Galimdzhan Ibragimov, an influential Muslim communist, argued along these lines. His widely publicized essay in 1927 Which Way Will Tatar Culture Go? advocated the spread of his local Tatar language, which in his statistical semantics, was growing in importance and use. Unconnected to the “Sultangalievist” circle, his controversial statements gave legitimacy to the local option.[62] Stalin’s korenizatsiia campaign, which lasted in Central Asia through 1933, and in many places up till 1938, reflects this agreement. The Piedmont Principle Soviet pacification of Ukraine had a very different plot but an outcome similar to that of Central Asia – the promotion of the republican language at the expense of russification.[63] When the leaders of the Ukrainian Communist Party felt pressured to russify after the civil war, an article in the newspaper of the Central Committee observed: “There was a time when Galicia served as the ‘Piedmont’ for Ukrainian culture. Now, when Ukrainian culture is suffocating in ‘cultured,’ ‘European’ Poland, its center has naturally shifted to the Ukrainian SSR.” The image of Piedmont – the magnet to draw in all of Italy – suggested the idea, in Terry Martin’s account, “that Soviet Ukraine would likewise first culturally and then politically unify the divided Ukrainian populations of Poland, Czechoslovakia and Romania . . . [since] cross-border ethnic ties could be exploited to foment national discontent in neighboring states.” This Piedmont perspective helps explain a good deal of the nativization that took place in the western republics. It is certainly true, as was the case in Central Asia, that the Bolsheviks lacked support in the countryside and did not want further to provoke the populace with a russification drive. Yet in the west, the Bolsheviks had a grander vision supporting nativization policies. Ukraine’s communists pressed for minority national Soviets within each republic, as this would humiliate Poland, which was being criticized by the League of Nations for its national minority policies. Similarly, the Soviets created the Moldavian ASSR, in the hope of using that territory as a Piedmont to attract Bessarabia, which Romania had annexed to the chagrin of the Soviets. Again, in late 1923, the Politburo of the Soviet Union agreed to extend the boundaries of Byelorussia by transferring sixteen uezdy from the RFSFR (the Russian Federation), in order to make it a strong and attractive republic. This would help, in the eyes of Soviet authorities, to foment rebellion by White Russians against Polish rule in Polish Belorussia. The Soviet chairman of the border dispute commission insisted that for this policy to succeed, it was necessary to derussify the population. The apotheosis of the Piedmont principle took place in the late 1920s in Kuban, where Cossacks were able to ukrainize all districts in which they were the majority in this area of the RFSFR. A bureaucratic cascade ensued, in which even the fully russified Ukrainian peasants in Kazakhstan were compelled to ukrainize. In accordance with Piedmont strategy, Ukraine sent teachers, books, theatrical productions, and radio programs to all areas in which Ukrainians lived within the Soviet boundaries. By 1932, after a brutal collectivization effort had nearly caused a counterrevolution in the Kuban, and after Moscow had adopted a less missionary foreign policy orientation, the Piedmont principle was drastically modified. Ukraine would have no more cultural influence outside its republican boundaries. Although Ukrainian schools would continue to educate virtually all Ukrainians, there was to be full russification of Slavs within the RSFSR. Stalin continued his assault on the titular languages through the decade. In 1933 a decree abolished the right of the constituent republics to grant orders of distinction. In this era, Frederick Barghoorn writes, Writers such as [Mykola] Khvylovi, who proclaimed an “Asiatic Renaissance,” in which a Western-oriented Ukraine, not Moscow, was to be the leader of socialism, statesmen such as education Commissar [O.] Shumski, economists like [N.] Volobuev, who in 1928 denounced Moscow’s “colonialism,” leaders like [G. I.] Petrovski, who as early as 1926 attacked the habitual use of the Russian language at Ukrainian Party meetings, were imprisoned or shot, committed suicide, or simply disappeared. By World War II, Stalin was unrestrained in his substitution of Russian nationalism for Soviet patriotism. In November 1941, in an oft-quoted speech, Stalin declared that the fascists “have the impudence to demand the destruction of the Great Russian nation, the nation of Plekhavov and Lenin, of Belinski and Chernyshevski, of Pushkin and Tolstoi, of Glinka and Chaikovski, of Gorki and Chekhov, of Sechenov and Pavlov, of Repin and Surikov, or [and now, Stalin lists two tsarist generals] Suvorov and Kutuzov.”[64] After Stalin’s death in 1953, the Soviet government continued what was tantamount to a policy of russification. Khrushchev’s educational reforms of 1958, for example, promoted parental free choice for educational medium of instruction in schools throughout the Union. Many national elites saw this as a code word for russification. Talk about the “merging” of nations (that is, the russification of the Soviet Union) as a historically inevitable process was a leitmotif in the Khrushchev years. But under Brezhnev, the notion of “merging” was quietly dropped. The historical legacies of the Piedmont principle and the effort to destroy pan-Turkism through the promotion of distinct Turkic languages had an enduring legacy in the Soviet Union. Both policies, although repealed, helped to perpetuate the unassimilated bilingual outcome throughout the Soviet Union through the greater part of the twentieth century. The Failure to Tip to Russian Macrohistorical factors, having to do with the state-building efforts of Russian tsars and commissars, explain why language rationalization did not spread through all social strata as it did in Western European states. Yet a microanalyst will ask: what prevented ordinary people, despite interests by local elites to prevent them, from assimilating into the dominant culture (and becoming fluent in the dominant language) of the ruling state? To answer this question, the microfoundations of Russian language rationalization must be examined. Before I provide those foundations, I shall specify more fully than I did in Chapter I how payoffs can be assigned in a tipping model. Consider a state-building tipping game, as portrayed in Figure 2.1 but identical in structure to the national-revival game portrayed in Figure 1.1. In this game, the principal players are people living in the periphery of a heterogeneous state who must decide whether to adopt, or equip their children to adopt, the state language. An examination of Figure 2.1 shows that the overall payoff for speaking the central state language (Lc) as opposed to the regional language (Lr) is largely determined by whether other members of their community (speakers of Lr) are adopting Lc. Thus people, by the logic of the model, are pushed into an intergenerational coordination game with their conationals. As with all coordination efforts, strategic problems arise. What incentive was there for the first Kazakh to learn Russian, or to send his child to a Russian-language school? Was he excoriated by his own community for having done so? Did the Russians praise assimilators in general but discriminate against them in particular because of other cultural differences? Would it have been wiser to wait for a significant Russian-speaking community of Kazakhs to develop? If so, suppose every speaker of Lr decided to wait until 40 percent of the Kazakhs had learned Russian, or a significant number of Kazakh children had completed a Russian-language school? The macroresult of this microprocess would have been the nonassimilation of Russian by the Kazakh speech community. It follows from this strategic situation that it is individually irrational for a speaker of Lr to switch to Lc (or prepare her children to do so) if virtually no one in the speech community has made the switch. If this is the case, and we accept ratio nality assumptions, how could intergenerational assimilation ever occur? There are a number of possibilities. For one, an individual’s payoff for switching may be higher than the average, for example if he learns language easily or if he falls in love with a speaker of Lc. Also, states often coerce individuals into assimilation – by putting them in labor camps or in the army, for example. In the Soviet army, there were units with Russian majorities and units with mixed minorities, but no units of a majority of any one minority. The mixed minority units were usually stationed in Russian-speaking zones. This strategy evidently worked well in catalyzing the widespread understanding of Russian in all social strata.[65] When soldiers returned home, they automatically raised the percentage of Lr members who spoke Lc. Figure 2.1. State-building tipping game: Percentage of population that speaks central state language Several intricacies of the model might be raised now. First, if it is irrational at first to assimilate, at the tipping point k, it becomes irrational to refuse to switch. The tipping model should therefore enable us to calculate the rates of language shift depending on the number of Lr speakers who have already made the shift. Second, on both payoff functions (Lc-Lc and Lr-Lr) the curves reverse direction at the extremes. I explain the phenomenon that this represents – the high returns for being one of the very few members of your group with an unusual language repertoire – in Chapter 9. Here it is best to focus on the central aspect of tipping – that great social shifts seem impossible at one point but inevitable at another. If rational-choice theory is to be applied to issues of cultural identity, as I indicated in Chapter I, it must go beyond material calculations. If it were the case that the only rationality is that of material benefit, then rational models would quickly find themselves ill-equipped to deal with issues of culture. But I do not believe that the only form of rational calculation is that of material benefit. In fact, a good first cut into the utility functions Lr-Lr and Lc-Lc is to think about three separate calculations. First, and this concerns material benefits, a potential assimilant needs to calculate the expected economic returns for adding a language to her child’s repertoire (less the opportunity costs for learning it). But after this calculation, we enter the world of honor and status. Second, a potential assimilant will want to assess whether members of the in-group will punish potential assimilants as cultural apostates. This value I call in-group scorn. Third, the potential assimilant will want to consider the degree to which members of the out-group (who speak the language her child might be assimilating) will accept an assimilant as one of their own, for example, as a potential marriage partner for a member of their family, or as a member in a private club. This factor I call out-group acceptance. Applying the tipping model to the Soviet Union makes it possible to analyze the largely successful tip from parochialism to unassimilated bilingualism.[66] The economic benefits for bilingualism were moderately high, and the first learners of Russian did not face any significant in-group scorn. Though out-group acceptance was not great outside the Slavic republics in the west, in the move toward unassimilated bilingualism, out-group acceptance (outside of gaining rewards for linguistic mediation) was not consequential. Most regions of the Soviet Union, as I showed, passed the tipping point toward unassimilated bilingualism. As long as the Union was holding, an increasing number of non-Russian Soviet citizens were developing competence in Russian as a second language. The Soviets, at least by the 1930s, were not satisfied with unassimilated bilingualism. Their hope was basically to attain assimilated bilingualism.[67] But this was in most republics not very successful. Consider first economic returns. Given the patronage power of titular elites who favored speakers of Lr, the added economic returns for assimilation into Russian were low through much of the country. Perhaps only in regions economically behind central Russia, and especially in ASSRs and lesser units where the patronage for speakers of the titular language was paltry, were these payoffs (less learning and in-group scorn costs) for assimilated bilingualism positive. A Tatar – in Tatarstan, an ASSR – who knew Russian better than Tatar, one study found, had over a 50 percent chance to improve his or her job training while a Tatar who still knew the Tatar language better than Russian had only a 10 percent chance of improvement.[68] But in regions that were economically more advanced or not dependent on Russia, such as the Baltic republics, Georgia, and Armenia, people throughout the Soviet era saw little economic advantage in making Russian their primary language, though learning Russian as a second language was considered useful. Concerning in-group status, there was some variation across republics, and over time. In regions that are culturally similar to Russia (the Slavic republics of Ukraine and Byelorussia), early assimilators faced little in-group scorn. In Ukraine of the 1970s, for example, with arrests still being made of Ukrainian cultural figures, it was imprudent for self-appointed patriots to sanction their brethren for not upholding national traditions.[69] In other republics, assimilators were held in much deeper suspicion by their fellow nationals. Many sold out indeed, but they paid an in-group cost for so doing. As time went on, however, in-group status took on greater importance in titular calculations. This was due to the perception, beginning in the Brezhnev period but accentuated in the Gorbachev years, of imperial decline. If, for example, Georgians believe that Georgia is likely to become an independent country in the next generation, they will worry far more about ingroup status and will begin to discount heavily the expected economic returns for learning Russian. To the extent that people believe that one’s children will be living in a sovereign state of Armenia, Georgia, Estonia, Kirghizia, they will worry about what fellow nationals will think about their having assimilated into the culture of the former center. Especially for those groups that had tasted sovereignty in the recent historical past (Baltics, Georgia, west Ukraine), and those whose linguistic brethren have sovereignty across an international border (Moldovans and Tajiks), the in-group status costs of Russian dominance in one’s linguistic repertoire increase. Concerning out-group status, despite the propaganda supporting the “merging” of all peoples, the rewards for assimilation were hardly impressive. Ability to penetrate all-Union party circles was clearly related to whether the non-Russians were Orthodox in religion and Slavic in culture, and even then the widespread belief that all power positions were reserved for Great Russians lowered the potential status rewards for assimilation. A Ukrainian who spoke Russian may not have been considered a complete outsider; but a Kazakh who spoke perfect Russian continued to experience residual prejudice and suspicion as a possible fifth columnist. So religious and cultural similarity raised the probability of out-group acceptance, but they never raised it very high. Despite the seventy-seven years of “Moscow Center” in Soviet rule, the expected repertoires in the Soviet period. In: Laitin, D. Identity in Formation: The Russian-Speaking Populations in the Near Abroad. Cornell University Press, 1998, pp. 37-58. ________________________________ [1] Alexis de Tocqueville, Democracy in America, trans. Henry Reeve (1835; New York: Schocken Books, 1961), 1:521-22. [2] Richard Pipes, The Formation of the Soviet Union: Communism and Nationalism, 1917-1923 (Cambridge: Harvard University Press, 1957), p. 1, calculation attributed to A. Bruckner, Die Europäisierung Russlands (Gotha, 1888). [3] Pipes, Formation of the Soviet Union, pp. 6-7. [4] Michael Khodarkovsky, “From Frontier to Empire: The Concept of the Frontier in Russia, 16th-18th Centuries” Russian History 19 (1992): 115-28. [5] Roman Levita and Mikhail Loiberg, “The Empire and the Russians: Historical Aspects” in Vladimir Shlapentokh, Munir Sendich, and Emil Payin, eds., The New Russian Diaspora (Armonk, N.Y.: M. E. Sharpe, 1994), p. 5. [6] Robert Kaiser, the Geography of Nationalism in Russia and the USSR (Princeton: Princeton University Press, 1994), pp. 47-50. [7] Robert A. Lewis et al., Nationality and Population Change in Russia and the USSR (New York: Praeger, 1976), p. 149. [8] See James Given, State and Society in Medieval Europe (Ithaca: Cornell University Press, 1990). [9] This notion of rationalization relies on Max Weber’s “formal rationality” that is, the use of standardized procedure rather than “substantive rationality” which concerns the content of the law. See Weber, Economy and Society (Berkeley: University of California Press, 1968), pp. 809-38. [10] George L. Yaney, the Systemization of Russian Government: Social Evolution in the Domestic Administration of Imperial Russia, 1711-1905 (Urbana: University of Illinois Press, 1973), pp. 75-76. [11] Hugh Seton-Watson, The Russian Empire, 1801-1917 (Oxford: Oxford University Press, 1967), p. 269. [12] Martha Brill Olcott, “The Politics of Language Reform in Kazakhstan,” in Isabelle T. Kreindler, ed., Sociolinguistic Perspectives on Soviet National Languages: Their Past, Present, Future (Berlin: Mouton de Gruyter, 1985), p. 188. [13] Edward C. Thaden, “The Russian Government,” in Edward C. Thaden, ed., Russification in the Baltic Provinces and Finland, 1855-1914- (Princeton: Princeton University Press, 1981), p. 82; Baron von der Osten-Sacken, The Legal Position of the Grand-Duchy of Finland in the Russian Empire (London: Lamely, 1912), p. 154. [14] Ronald Suny, The Making of the Georgian Nation (Bloomington: Indiana University Press, 1988), chap. 3. [15] Thaden, Russification, p. 75. [16] Ronald Suny, personal communication. [17] Suny, Making of the Georgian Nation, pp. 70, 351. [18] Martha Brill Olcott, The Kazakhs (Stanford, Calif.: Hoover Institution Press, 1987), pp. 104-5. [19] This dichotomy between a bilingual elite and monolingual lower strata is exaggerated. Itinerant merchants and artisans (such as the Chuvash) picked up Russian; Muslim elites, however, tended to ignore the imperial language. Yet compared with other states where language rationalization occurred, in Russia there was far less of a cascade through all levels of society toward monolingualism in the state language. This is the outcome that I seek to explain. [20] John A. Armstrong, “Toward a Framework for Understanding Nationalism in Eastern Europe,” East European Politics and Societies 2 (1988): 280-305. Reinhard Bendix, Kings or People (Berkeley: University of California Press, 1978), as I discuss in Chapter II, emphasizes this same Phenomenon to explain the persistence of Russian patrimonialism. [21] Anthony D. Smith, “Ethnic Identity and Territorial Nationalism in Comparative Perspective” in Alexander J. Motyl, ed., Thinking Theoretically about Soviet Nationalities (New York: Columbia University Press, 1992), p. 54. [22] David Laitin, Roger Petersen, and John Slocum, “Language and the State,” in Motyl, Thinking Theoretically about Soviet Nationalities, pp. 129-68. [23] I show this quantitatively in David D. Laitin, Stathis Kalyvas, and Carlota Sol_, “Language and the Construction of States: The Case of Catalonia in Spain, “Politics and Society 22 (1994): 5-29. [24] Isabelle T. Kreindler, “Nikolai Il’minski and Language Planning in Nineteenth-Century Russia” International Journal of the Sociology of Language 22 (1979): 5-26; see also her article “The Non-Russian Languages and the Challenge of Russian” in Kreindler, ed., Sociolinguistic Perspectives on Soviet National Languages (Berlin: Mouton, 1985), pp. 345-67. [25] Abram de Swaan, In Care of the State (Oxford: Oxford University Press, 1989). [26] Isabelle T. Kreindler, “The Mordvinian Languages: A Survival Saga,” in Kreindler, SociolinSuistic Perspectives, p. 241. [27] Thaden, “Russian Government,” pp. 48-49,54-55,70-71. [28] Eugen Weber, Peasants into Frenchmen (Stanford: Stanford University Press, 1976). [29] This term is Brian Silver’s. He distinguishes parochialism (knowing only the language of the locality), unassimilated bilingualism (knowing as well the language of the center, but using it in limited domains, and with great difficulty), assimilated bilingualism (relying principally on the central language, but maintaining some facility in the local language), and assimilation (becoming monolingual in the central language). Brian Silver, “Methods of Deriving Data on Bilingualism from the 1970 Soviet Census” Soviet Studies 27 (October 1975): 574-97. [30] Soviet census data are not fully to be trusted. Respondents gave subjective accounts of their language abilities, but the accuracy of these self-assessments was not tested. Also, respondents often denied competence in Russian to voice opposition to the regime. There is also evidence of political tampering with the data. William Fierman reports, for example, that the first secretary of the Uzbekistan Communist Party, a promoter of the Russian language, delivered a census report in 1979 that showed a 34.8 percent rise in the use of Russian over a period of nine years. See William 1 Fierman, “Language Development in Soviet Uzbekistan,” in Kreindler, Sociolinguistic Perspectives, pp. 220-21. This is highly unlikely from a sociolinguistic point of view. The 1989 census data show ;| that knowledge of Russian among Uzbeks dropped from approximately 53 percent in 1979 to about 22 percent in 1989, confirming that the reported 54.8 percent rise of the previous decade was most likely fraudulent. On these 1989 data, see Barbara A. Anderson and Brian D. Silver, “Demographic Sources of the Changing Ethnic Composition of the Soviet Union” Population and Development Review 15, no. 4 (1989): 644-46. For an excellent study of verbal self-identification in the Soviet census, see Rasma Karklins, “A Note on “Nationality’ and ‘Native Tongue’ as Census Categories in 1979,” Soviet Studies 32, no. 3 (1980): 415-22. [31] Brian Silver, “Language Policy and the Linguistic Russification of Soviet Nationalities,” in Jeremy Azrael, ed., Soviet Nationality Policies and Practices (New York: Praeger, 1978), pp. 282-86, and “Bilingualism and Maintenance of the Mother Tongue in Soviet Central Asia” Slavic Review 35, no. 3 (1976): 414. [32] See Brian Silver, “Social Mobilization and the Russification of Soviet Nationalities,” American Political Science Review 68, no. 1 (1974): 59; similar findings are reported in Ronald Wixman, Language Aspects of Ethnic Patterns and Processes in the North Caucasus, Research paper no. 191, University of Chicago, 1980, for the North Caucasus and Central Asia; Olcott, “Politics of Language Reform” for Kazakhstan; and George B. Hewitt, “Georgian: A Noble Past, a Secure Future” in Kreindler, Sociolinguistic Perspectives, pp. 163-79, for Georgia. [33] These data are from titular nationalities of only the highest level of republic, known as union republics. Below that level are the autonomous soviet socialist republics (ASSRs), which received, as I shall analyze later, far fewer resources for cultural autonomy. In Table 2.1, I ignore 1989 census data, since it is highly likely that in 1989 respondents were declaring ignorance of Russian to make ^a political point, at least in several republics. [34] E. Glyn Lewis, Multilingualism in the Soviet Union: Aspects of Language Policy and its Implementation (The Hague: Mouton, 1972), pp. 134-35. [35] B. A. Anderson and B. D. Silver, “Some Factors in the Linguistic and Ethnic Russification of Sovwt Nationalities: Is Everyone Becoming Russian?” in Lubomyr Hajda and Mark Beissinger, eds., The Nationalities factor in Soviet Politics and Society (Boulder, Colo.: Westview Press, 1990), pp. 95-130. [36] I. T. Kreindler, “The Changing Status of Russian in the Soviet Union,” Research paper no. 37, Soviet and East European Center, Hebrew University of Jerusalem, 1979, pp. 4-5. [37] Lewis, “Multilingualism,” pp. 58-59. But, as Edward A. Allworth, The Modern Uzbeks: from the fourteenth Century to the Present, A Cultural History (Stanford, Calif.: Hoover Institution Press, 1990), perceptively insists, it is anachronistic to think of this as the movement of peoples to “their” republics. More probably, people identified themselves (or were compelled to identify themselves) with the titulars and made the (often slight) cultural adjustments to fit in with their newly adopted group. [38] On korenizatsiia see William Fierman, Language Planning and National Development: The Uzbek Experience (Berlin: Mouton de Gruyter, 1991), and George Liber, The Urban Harvest: Ethnic Policy, Legitimization, and the Unintended Consequences of Social Change in the Ukrainian SSR, 1923-1933 (London: Cambridge University Press, 1992). The quotation is from Philip G. Roeder, “Soviet Federalism and Ethnic Mobilization,” World Politics 43 (1991): 204. He exaggerates the level of opportunity for non-Russians, because he counts members of nationality groups living outside their titular republics as being advantaged by the system when they were not. According to Kaiser, Geography of Nationalism, p. 155, in 1926 only 79.5 percent of the members of the eleven titular nationalities lived within their titular republic. [39] Rasma Karklins, Ethnic Relations in the USSR (Boston: Alien & Unwin, 1986), pp. 80-81; Suny, Making of the Georgian Nation, p. 290. [40] Nancy Lubin, “Assimilation and Retention of Ethnic Identity in Uzbekistan,” Asian Affairs 12 (1981): 227-85. [41] Lewis, Multilingualism, p. 177. [42] Soviet ethnographers confirm the linguistic advantages that have accrued to members of titular nationalities. See M. Guboglo et al., “Etnolingvisticheskie protsessy,” in I. V Bromlei, ed., Sovrewennye etnicheskie protsessy v SSSR (Moscow: Nauka, 1975), pp. 259-313; table 12, p. 301. For example, between 1926 and 1970, the correspondence between nationality and native language increased for ten out of fifteen titular nationalities, but decreased for seventeen out of nineteen nationalities having a lower form of autonomy. See also Roeder, “Soviet Federalism and Ethnic Mobilization” p. 204, and Karklins, Ethnic Relations. [43] Karklins, Ethnic Relations, p. 38. [44] G. I. Litvinova and B. Ts. Urlanis, “Demograficheskaia politika Sovetskogo Soiuza,” Sovetskoe gosudarvstvo i pravo 3 (1982): 45, quoted in Karklins, Ethnic Relations, p. 94; see also Anderson and Silver, “Demographic Sources.” [45] Andrew G. Walder, Communist Neo-Traditionalism: Work and Authority in Chinese Industry (Berkeley: University of California Press, 1986). [46] Karklins, Ethnic Relations, p. 105, interprets her data to emphasize the considerable majorities of the titular nationals who wanted mare education in their language. But the data can also be read to show that there were significant minorities that wanted more Russian. The Soviet state attempted at various times to respond to the felt demand for Russian as a medium of instruction in the union republics. Khrushchev’s policy of 1958 to allow for parental choice in regard to medium of I instruction was severely criticized by the titulars, since they feared that their own people would 1 choose Russian. This policy did not have a major impact on educational reality within most re- publics. See Lewis Multilingualism, pp. 67-80. In the 1970s, the Ministry of Education proposed that in Georgia, technical material in higher education should be in Russian. Concomitant attempts to delete Georgian from its official status in Georgia induced the titular elites to argue vociferously against these efforts. See Suny, Making of the Georgian Nation, p. 301. [47] Roeder, “Soviet Federalism and Ethnic Mobilization,” pp. 196-231. [48] This quotation is Anatoly Khazanov’s gloss. See Khazanov, After the USSR (Madison: University of Wisconsin Press, 1995), p. 12. [49] This section is borrowed from Laitin, Petersen, and Slocum,” Language and the State.” In that article, we portray this conflict as a formal game in which the equilibrium outcome is the promotion of the local variants of Turkic. Here I summarize the argument without the game model. In that article, we portrayed the outcome of this conflict as the motor for korenizatsiia throughout the Soviet Union. Since the publication of that article, Terry Martin has convinced me, as reflected in the next section, that there was a more powerful motor driving korenizatsiia in the west. [50] Alexandre Bennigsen, “Panturkism and Panislamism in History and Today,” Central Asian Survey 3, no. 3 (1985): 41-43. [51] Alexandre Bennigsen and Chantal Lemercier-Quelquejay, Islam in the Soviet Union (New York: Praeger, 1967), p. 126. [52] Allworth, Modern Uzbeks. [53] I. T. Kreindler, “A Neglected Source of Lenin’s Nationality Policy” Slavic Review 36 (1977): 86-100, and “The Non-Russian Languages and the Challenge of Russian: The Eastern Versus the Western Tradition,” in Sociolinguistic Perspectives, p. 348. [54] See Olcott, Kazakhs, p. 206. [55] Michael Kirkwood, “Language Planning: Some Methodological Preliminaries” in M. Kirkwood, ed., Language Planning in the Soviet Union (New York: St. Martin’s Press, 1990), p. 32. [56] Simon Crisp, “Soviet Language Planning, 1917-53,” in Kirkwood, Language Planning, pp. 29-30. [57] Terry Martin, “The Soviet Nationalities Policy, 1923-1938” (Ph.D. diss., University of Chicago, 1996), p. 919, reports that Stalin’s speeches in 1935 offer a military justification for linguistic rationalization (knowing enough Russian to obey orders), without any implications of assimilation. [58] Alexandre A. Bennigsen and S. Enders Wimbush, Muslim National Communism in the Soviet Union: A Revolutionary Strategy for the Colonial World (Chicago: University of Chicago Press, 1979), p. 20. [59] Although there were some among this new communist elite who believed the best way to avoid Russian domination was to more fully develop local cultures, the pan-Turks who stressed Central Asian regional unity in politics and culture were the more significant group, especially in the maneuverings concerning language development of the 1920s. [60] Bennigsen and Wimbush, Muslim National Communism, p. 88. [61] The question of Sultan Galiev磗 loyalty to the Soviet regime has been reopened in “Who Is Sultangaliev,” published in 1989 in the Tatar journal Kazan Ultari. See Azade-Ayse Rorlich, “The Disappearance of an Old Taboo: Is Sultangaliev Becoming Persona Grata?” Report on the USSR, September 29, 1989, pp. 16-17. [62] Azade-Ayse Rorlich, “Which Way Will Tatar Culture Go? A Controversial Essay by Galimdzhan Ibragimov,” Cahiers du monde russe et sovi閠ique 15, no. 2 (1974): 363-71. [63] This section relies heavily on Martin, “Soviet Nationalities Policy.” Although references to “Piedmont” abound in contemporary accounts, the term “Piedmont principle” in this context is Martin’s invention. [64] Frederick C. Barghoorn, Soviet Russian Nationalism (New York: Oxford University Press, 1956), pp. 16, 37, 39. Jerry F. Hough, in chapter 6 of his Democratization and Revolution in the USSR: 1985-91 (Washington, D.C.: The Brookings Institution, 1997) takes the counter view. By his recruitment of non-Russians into the Party, and by reifying nationality on the passport in 1934, Stalin, because “he remained a Georgian at heart,” actually promoted non-Russian identities. [65] Robert Cullen reports that the Soviet comedian Evgenii Petrosian earned hilarious laughter in Rostov-on-Don when a punch line mentioned that in an army base of thousands of soldiers, “none of them speak Russian.” New Yorker, June 12, 1989. M. B. Olcott and William Fierman, “Soviet Youth and the Military” (U.S. Department of State, contract no. 1724-620124, Washington, D.C.), report that this image of incompetence in Russian among non-Russian army recruits is reflected in many stories in the regional newspapers. But once in mixed minority units, we can surmise, with Karklins, Ethnic Relations, p. 101, that most soldiers achieve some facility in Russian. In my discussions with Estonians who had served in the Soviet army, I found that they developed a rudimentary understanding of Russian and a rich and eloquent mastery of Russian mat_ (profanity). [66] Readers might still object—as have scores of seminar participants who have heard me present this model—that this is not a realistic portrayal of how people really think about cultural matters, even if I abandon a pure materialist choice perspective. I urge the reader to suspend disbelief, until Chapter 5, where I use extensive ethnographic data to demonstrate the real-world calculus of cultural identity. [67] There is no consensus on this point. Frederick Barghoorn, “Russian Nationalism and Soviet Politics: Official and Unofficial Perspectives” in Robert Conquest, ed., The Last Empire: Nationality and the Soviet Future (Stanford, Calif.: Hoover Institution Press, 1986), p. 32, refers to “Stalin’s policy of maximum feasible Russification.” Most Soviet commentators, e.g. Iu. Bromlei, in Natsion- al’nye protsessy v SSSR: V poiskakh novykh podkhodov (Moscow: Nauka, 1988), emphasized the goal of the flowering of all nations. Although the Soviet policy was not graced by consistency in this matter, it is fair to say that in Soviet eyes, an ideal patriot was, in an earlier statement by Frederic Barghoorn, in Soviet Russian Nationalism, p. 11, “a complex of the highest values and loyalties of Soviet citizens, with loyalty to the particular culture of one’s own nation in the second order of priority and loyalty to international communism on the third level.” This sense of priority is consistent with the goal of assimilated bilingualism. [68] Peter Shearman, “Language, Sovietization and Ethnic Integration in the USSR,” Journal of Social, Political, and Economic Studies 8, no. 3 (1983): 243-44. [69] Alexander Motyl, Will the Non-Russians Rebel? (Ithaca: Cornell University Press, 1987), chap. 6.