THE GREAT DIVIDE
Literacy, Nationalism, and the
Communist Collapse
By KEITH DARDEN and ANNA GRZYMALA-BUSSE*
I. INTRODUCTION
WHY do some governing parties, closely associated with a collapsed
authoritarian regime, nonetheless retain power and continue
to govern? This paradoxical outcome occurred in 45 percent of
countries of the former Soviet Union and its satellites. In some of these
countries the ﬁrst free elections returned the Communist Party to rule,
while in others unreconstructed communists retained power and free
elections were never held. In the remaining 55 percent, however, communist
parties lost the ﬁrst free elections and exited power completely.
In this article, we seek to explain these patterns of communist exit,
which has spawned both empirical and theoretical controversies. The
communist exit in the ﬁrst free elections has been strongly correlated
with subsequent democratic consolidation, successful economic reforms,
and patterns of political party competition.1
Communist persistence,
by contrast, resulted in “democracy with adjectives”: (1) quasidemocratic
systems that hold elections but do not foster competition
or representation and (2) the rise of antireform coalitions that extract
private beneﬁts from the state and sabotage reforms.2
*The authors are listed alphabetically and share responsibility. This paper was ﬁrst presented at the
annual meeting of the American Political Science Association, Washington, D.C., 2005. We are grateful
to Stephen Hanson, Jeff Kopstein, Grigore Pop-Eleches, the participants of faculty workshops at
Duke University and the University of Michigan, and the three anonymous reviewers, for their very
helpful comments. We would like to thank Shale Horowitz and David Reilly for sharing their data.
1
M. Steven Fish, “The Determinants of Economic Reform in the Post-Communist World,” East
European Politics and Societies 12 (Winter 1998); Valerie Bunce, “The Political Economy of Postsocialism,”
Slavic Review 58 (Winter 1999); Herbert Kitschelt, “Accounting for Postcommunist Regime
Diversity: What Counts as a Good Cause?” in Grzegorz Ekiert and Stephen Hanson, eds., Capitalism
and Democracy in Central and Eastern Europe (Cambridge: Cambridge University Press, 2003).
2
Steven Levitsky and Lucan Way, “The Rise of Competitive Authoritarianism,” Journal of Democracy
13 (April 2002); Joel S. Hellman, “Winner Take All: The Politics of Partial Reform in Postcommunist
Transitions,” World Politics 50 (January 1998).
World Politics 59 (October 2006), 83–115
84 WORLD POLITICS
At the same time, considerable criticism of the communist exit as
an explanation for subsequent trajectories has emerged. First, Herbert
Kitschelt, in particular, has charged that accounts focusing on the communist
exit as a main explanatory variable suffer from excessive causal
proximity, leading to an explanation that “yields little insight into the
causal genealogy of a phenomenon.”3
His critique begs a signiﬁcant
question: if the communist exit is so highly correlated with favorable
outcomes, how do we account for the exit? Second, the mechanisms of
this inﬂuence on economic and democratic outcomes have been difﬁcult
to pinpoint. Does the communist exit act as an independent causal
factor that eliminates a source of antireformist sentiment and thus promotes
greater elite consensus? Is it a simple symptom of deeper readiness
for democracy and the free market or a necessary but not sufﬁcient
condition for subsequent reforms? Third, there has been little explanation
of the reasons behind the communist exit, or how those reasons
might themselves relate to postcollapse outcomes. This shortcoming
has led to the criticism that, much as with the communist collapse itself,
political science has focused on the wrong set of explanatory factors
and mechanisms. One striking omission, for example, is a theory
of one-party rule and the factors that could sustain its efﬁcacy and even
its legitimacy.4
The critical question underlying all these controversies is why communist
rule ended in such divergent outcomes. Why, that is, was there
communist exit from power in some countries but not in others? We
argue that the ultimate roots of the explanation lie in precommunist
schooling, which fomented and fostered nationalist ideas that led to the
delegitimation of communist rule. The exit itself was the culmination
of decades of nursed nationalist grievances, invidious comparisons, and
carefully sustained mass hostility to the communist project as a foreign
and inferior imposition. Section II reviews and tests the competing explanations.
Section III examines the patterns of schooling. Section IV
presents an alternative model that establishes a causal chain linking the
introduction of mass schooling, subsequent ideas about the nation and
its legitimate authority, the rise of anticommunist opposition, and the
communist exit. As a result, some countries were much less hospitable
to communism, more likely to kindle an opposition, and more likely
3
Kitschelt (fn. 1).
4
Stathis Kalyvas, “The Decay and Breakdown of Communist One-Party Regimes,” Annual Review
of Political Science 2 (1999); Timur Kuran, “Now out of Never: The Element of Surprise in the
East European Revolution of 1989,” World Politics 44 (October 1991).
to promote the kind of communist party that could and would leave.
Section V concludes.
II. EXISTING EXPLANATIONS OF THE COMMUNIST EXIT
The literature on both the collapse of the communist regimes and the
transitions that followed provides us with several competing explanations
for the variation in the communist exit. These locate the forces
behind the communist exit along a temporal continuum that begins
with long-standing structural forces, such as geography, moves through
the legacies of the precommunist era, and ends with the immediate
causes of the communist collapse, such as the strength of the anticommunist
opposition.
One prominent explanation for postcommunist trajectories focuses
on structural factors, the favorable geopolitical settings that placed some
of these countries in the Western Christian orbit, with its Enlightenment
tradition, potential for trade, and diffusion of democratic ideas.
One manifestation of this inﬂuence is the remarkable correlation that
exists between proximity to the West (deﬁned as the geographic distance
between state capitals and either Vienna or Berlin) and the favorable
conﬁguration of communist exit, democratic reforms, and market
liberalization, as Kopstein and Reilly 2000 demonstrate.5
Such settings
allowed for the favorable inﬂuence of international organizations, “not
so much of actual EU or NATO membership as of anticipated membership.
These divergent, externally induced incentives are part of what
accounts for differences in institutional reform, state behavior, and
popular discourse in the countries of postcommunist Europe.”6
After
all, the nearest neighbors were also the objects of the most intense focus
of the EU and other West European initiatives, and communism may
have appeared less desirable if one’s neighboring noncommunist points
of reference were Germany and Austria rather than Afghanistan and
Iran. Geographic proximity had also earlier fostered a sense of “belonging”
to Europe.
A second inﬂuential approach examines the legacies of precommunist
development and the ways in which the political experiences of the in-
5
Jeffrey S. Kopstein and David A. Reilly, “Geographic Diffusion and the Transformation of the
Postcommunist World,” World Politics 53 (October 2000); Anna Grzymala-Busse and Pauline Jones
Luong, “Reconceptualizing the State: Lessons from Post-Communism,” Politics and Society 30 (December
2002).
6
Kopstein and Reilly (fn. 5), 25.
THE GREAT DIVIDE 85
86 WORLD POLITICS
terwar era shaped the politics of the communist years and beyond.7
In the classic “modernization” account, wealth, industrialization prior
to statehood, and overall levels of economic development are critical
to the development of democracy and to the maintenance of regime
stability.8
Considerable disparities persisted here in the communist era.
Even though East Central Europe has been characterized as a backward
periphery relative to Western Europe,9
its development levels
were always higher than in Central Asia, for example. The communist
exit, therefore, could be a function of precommunist economic development
and, more broadly, of modernization.
In a similar account, the length of the prewar democratic experience
and the political conﬁgurations that dominated it translate into
distinct communist regime types.The resulting typology of communist
regimes comprises bureaucratic-authoritarian, national-accommodative,
and patrimonial systems.10
The ﬁrst type, built on interwar working-class
parties and a preexisting professional bureaucracy, results in a
conﬁguration of weak communist insiders, unable to forestall their own
replacement during the communist collapse. At the opposite end of
the spectrum, patrimonial communism, built on authoritarian regimes
and nonprofessional bureaucracies, privileges the communist insiders
and allows them to hold on to political and economic power. One prediction
is that the longer the democratic and legal-rational experience
prior to the onset of communism, the greater the likelihood of a communist
exit.
Finally, the most temporally immediate explanation for the communist
exit is the strength of the anticommunist opposition in the waning
years of the communist regimes. As several scholars of regime transi-
7
George Schöpﬂin, Politics in Eastern Europe (Oxford: Blackwell, 1993); Grzegorz Ekiert,“Democratic
Processes in East Central Europe: A Theoretical Reconsideration,” British Journal of Political
Science 21, no. 3 (1991); Ken Jowitt, A New World Disorder (Berkeley: University of California Press,
1991); Ekiert and Hanson (fn. 1); Beverly Crawford and Arend Lijphart, “Enduring Political and
Economic Change in Post-Communist Eastern Europe: Old Legacies, New Institutions, Hegemons,
Norms and International Pressures,” Comparative Political Studies 28, no. 2 (1995).
8
Seymour Lipset, Political Man: The Social Basis of Politics (New York: Doubleday and Company,
1960); Daniel Lerner, The Passing of Traditional Society: Modernizing the Middle East (Glencoe, Ill.:
Free Press, 1958); Alexander Inkeles, Becoming Modern: Individual Change in Six Developing Countries
(Cambridge: Harvard University Press, 1969); Karl W. Deutsch, “Social Mobilization and Political
Development,” American Political Science Review 55 (September 1961); see also Adam Przeworski,
Michael E. Alvarez, Jose Antonio Cheibub, and Fernando Limongi, Democracy and Development: Political
Institutions and Well-being in the World, 1950–1990 (New York: Cambridge University Press,
2000).
9
Andrew C. Janos, East Central Europe in the Modern World:The Politics of the Borderlands from Pre
to Post-Communism (Stanford, Calif.: Stanford University Press, 2000).
10
Herbert Kitschelt, Zdenka Mansfeldová, Radosl⁄aw Markowski, and Gábor Tóka, Post-Communist
Party Systems (Cambridge: Cambridge University Press, 1999).
tions have noted, the strength of the opposition (and its constituent
radicals and moderates) affects not only the likelihood of negotiations
with an authoritarian government but also its outcome.11
The more
powerful the opposition, the more likely is communist exit. Conversely,
as Andrew Janos argued, “because it was rooted in the communalism
and paternalism of Byzantine Orthodoxy, communism resonated far
more positively in the Orthodox societies of the southeast, than in the
legalistic, contract societies of the northwest tier.”12
The implication
here is that cultural receptiveness to communism underlies the patterns
we observe and either promotes the rise of the opposition or sustains
communist rule.13
While all of these explanations offer compelling accounts, they face
two challenges. First, many of the accounts tend to rely on powerful
correlations, rather than on clearly speciﬁed mechanisms by which the
legacies of the past, for example, translate into outcomes decades later.
Second, these accounts tend to address national-level variation: therefore,
they cannot as easily explain the differences among the countries
emerging from the former Soviet Union or the intranational diversity
in support of the communist party. Yet these are as intriguing and as
potentially important for theory building as are their national-level
counterparts. As several scholars have noted, pockets of subnational
authoritarianism can coexist with democratic national governments,14
producing very divergent regimes within the same nation-state.
More speciﬁc problems arise with the individual explanations. One
complication with an emphasis on geographic factors is that some coun-
11
Samuel Huntington, The Third Wave (Norman: University of Oklahoma Press, 1991); Laszlo
Bruszt and David Stark, “Remaking the Political Field in Hungary,” Journal of International Affairs 46
(Summer 1992). For the negotiations, see Jon Elster, ed., Round Table Talks and the Breakdown of Communism
(Chicago: University of Chicago Press, 1997). See also Pauline Jones Luong, Institutional Change
and Political Continuity in Post-Soviet Central Asia (Cambridge: Cambridge University Press, 2002); John
Ishiyama, “Transitional Electoral Systems in Post-communist Eastern Europe,” Political Science Quarterly
112 (Spring 1997); Thomas Remington and Steven Smith, “Institutional Design, Uncertainty, and
Path Dependency during Transition,” American Journal of Political Science 40 (October 1996); Patrick
O’Neil, “Presidential Power in Post-communist Europe: The Hungarian Case in Comparative Perspective,”
Journal of Communist Studies 9 (September 1993);Timothy Frye,“A Politics of Institutional Choice:
Post-communist Presidencies,” Comparative Political Studies 30 (October 1997).
12
Janos (fn. 9), 326.
13
An excellent analysis of these arguments is in Grigore Pop-Eleches, “Which Past Matters?
Communist and Pre-Communist Legacies and Post-Communist Regime Change” (Paper presented
at the annual meeting of the American Political Science Association, Washington, D.C., September
1–4, 2005).
14
Richard Snyder, “After the State Withdraws: Neoliberalism and Subnational Authoritarian
Regimes in Mexico,” in W. Cornelius, T. Eisenstadt, and J. Hindley, eds., Subnational Politics and
Democratization in Mexico (La Jolla: Center for U.S.-Mexican Studies, University of California, San
Diego, 1999); Levitsky and Way (fn. 2); Robert Mickey, “Paths out of Dixie: The Democratization
of Authoritarian Enclaves in America’s Deep South, 1944–1972” (Book manuscript, Department of
Political Science, University of Michigan, 2006).
THE GREAT DIVIDE 87
88 WORLD POLITICS
tries were considerably more pro-West than their geographic location
would seem to suggest: Estonia, Latvia, and Lithuania are all farther
from Western capitals than Belarus, for example. Nor can geography
explain the differences in achieving a “return to the West.” How, that
is, did geographic or cultural afﬁnities translate into domestic political
action. Moreover, the mechanisms of the afﬁnity for Europe remain
underspeciﬁed. Integration into speciﬁc regional and international
organizations is an unlikely candidate: neither the EU nor NATO offered
any serious prospects for membership until many years after the
communist exit and the initial democratic and market reforms.15
In
East Central Europe, when even the most ardent reformists spoke of
a “return to Europe,” they were referring to a normalcy deﬁned by the
lack of Soviet imperial interference and not to anticipated membership
in speciﬁc international structures. Further, these sentiments did not
break cleanly along geographic lines. Thus, European Serbians kept
their communist party in power, while non-European Georgians rapidly
dispensed with theirs. The desire to “join the West” or “return to
Europe” was a signiﬁcant motive and clustered geographically, but its
roots do not lie in location alone. We thus need a better account of the
mechanism underlying the desire to “return to Europe.”
The mechanisms by which precommunist bureaucratic development
and regime types translate into the communist exit are similarly
underspeciﬁed. Temporal distance should not be conﬂated with causal
depth. What, for example, are the causal links between a professional
interwar bureaucracy and the weakness of communist insiders at the
time of the communist collapse? There is little question that the precommunist
development of a rational bureaucracy and direct rule distinguishes
the communist and postcommunist development of many
Central European countries from that of Central Asian states, where
the Soviet Union was the ﬁrst to bring any bureaucracy to nomadic societies.
However, it is not clear why these differences would lead to the
rejection of communist parties. Indeed, one could easily argue that the
superior bureaucracy made communism more rational, efﬁcient, and
competent and that its bureaucratic legacy should have bolstered rather
than undermined the legitimacy of communism. Moreover, having the
right bureaucratic legacy was not a necessary condition; otherwise patrimonial
Moldova, Armenia, and Georgia would have kept their communist
parties in power.
15
Thus, the EU was content to conduct bilateral agreements with postcommunist countries; membership
as a feasible option did not even enter the discussion until 1995 and the Bosnia conﬂict.
The strength of the anticommunist opposition movement as an explanatory
variable explains both too much and too little. On the one
hand, it is so strongly correlated with the communist exit that the possibility
the two are part of the same phenomenon cannot be discounted.
Once again, if we take warnings of excessive causal proximity seriously,
we ought to beware of positing such powerful causal relationships between
events that are so temporally close together. On the other hand, this
explanation requires that we identify the determinants of a strong opposition:
what are they, and do they explain the communist exit as well?16
A better, alternative explanation of the communist exit is therefore
needed, one that accounts both for the immediate impetus for the exit
and for its facilitating conditions. It should also provide a mechanism
through which the communist exit becomes feasible and likely. If we
turn to the precommunist past, we need to specify precisely which factors
and processes made some societies less hospitable to communism
on both the individual and the collective level, more likely to foment
opposition, and more likely to foster the kind of ruling party that could
and would leave power. A country’s location and “neighborhood” by
itself is unlikely to inﬂuence either the popular opposition or the legitimacy
of communist rule. Prewar democratic statehood is a more likely
candidate, since it may very well engender memories of noncommunist
governance and the subsequent identiﬁcation of communism as an
“abnormal” form of governance. Finally, the development of anticommunist
opposition itself needs explaining, since it is analytically and
empirically so close to the communist exit.
One of the most striking aspects of the anticommunist opposition
was its fusion with nationalism. Opposition movements claimed to be
rescuing the nation from the grasp of an alien, imposed, and illegitimate
communist regime. As Valerie Bunce argues, “The diffusion of
the national idea served as the mechanism for opposition elites to confront
imperial domination by seeking states and regimes of their own
making.”17
Yet if nationalism drives the opposition to communism, this
begs the questions of how and why these mass sentiments varied across
the region and of which mechanisms link those beliefs—largely historical
in origin—to contemporary events.
Although the development of nationalism has been linked to many
tools of the modern state, the clearest mechanism for the transmission
16
Bunce (fn. 1) and Janos (fn. 12) identiﬁed the rise of the opposition with prior statehood and the
linking of the state project with liberalism.
17
Bunce, “The National Idea: Imperial Legacies and Post-Communist Pathways in Eastern Europe,”
East European Politics and Societies 19 (Summer 2005).
THE GREAT DIVIDE 89
90 WORLD POLITICS
of a shared national identity and history is education.18
Although education
is typically seen as part of a bundle of developments—urbanization,
industrialization, income growth—due to the legacy of modernization
theory,19
the role of education is causally and empirically
distinct. Schooling provides the one clear channel for the deliberate and
systematic inculcation of a set of values.20
And the critical aspect of mass
literacy is its timing: the national ideas instilled in a population during
the ﬁrst round of mass schooling—when a community ﬁrst shifts from an
oral to a literate mass culture—are durable, and the ﬁrst schooled generation
will transmit those values in ways that previous or subsequent
cohorts do not, as we will see.21
We therefore focus on a historical legacy that predates not only communism
but also, in several cases, nation-states; the legacy includes the
timing of mass schooling and the nature of its national content. The
communist exit could not have occurred without mass opposition to
the regime—and that opposition in turn rested on notions of statehood
and legitimate governance ﬁrst inculcated by mass schooling. Attaining
literacy under a noncommunist regime led to the transmission of a
national identity separate from, and often directly opposed to, the communist
regimes. Precommunist schooling thus lowers the magnitude
of support for the communist party and increases the likelihood that
widespread opposition to the communist party will arise. We therefore
hypothesize that the communist exit is more likely to occur where literacy
preceded the onset of communism.
The extent and content of mass literacy for the region are shown in Appendix
1.To establish the basic correlations and to test the rival hypotheses,
we turn to straightforward statistical analyses that compare the impact of
precommunist schooling on the communist exit with the inﬂuence of economic
development, years of precommunist democratic statehood, and regional
effects, including distance to the West. (The full operationalization
and coding of these variables is included in Appendix 2.)
The ﬁndings support the claim that where precommunist schooling
was ﬁrmly established and literacy was widespread, the populations
were more likely to vote the communists out of power at the ﬁrst avail-
18
Ernest Gellner, Nations and Nationalism (Ithaca, N.Y.: Cornell University Press, 1983); E. J.
Hobsbawm, Nations and Nationalism since 1780 (New York: Cambridge University Press, 1990), chap.
3; Barry Posen, “Nationalism, the Mass Army and Military Power,” International Security 18 (Autumn
1993).
19
Deutsch (fn. 8); Lipset (fn. 8); Inkeles (fn. 8); Lerner (fn. 8).
20
Sylvia Scribner and Michael Cole, “Cognitive Consequences of Formal and Informal Education,”
Science 182 (November 9, 1973).
21
Keith Darden, “Mass Schooling and the Formation of Enduring National Loyalties” (Book
manuscript, Department of Political Science, Yale University, 2007).
able opportunity. As shown in the simple scatterplot of Figure 1, there
is a clear linear relationship between the percentage of the population
that was literate at the onset of communism and the defeat of the communists
in the ﬁrst free elections.22
In all countries with high levels of
precommunist literacy, the communists were soundly defeated in the
ﬁrst free elections.
To conduct some basic statistical tests of our hypothesis, we chose a
simple cross-sectional OLS regression as our estimation procedure and
applied it to the universe of postcommunist cases.23
The ﬁndings in
22
Only Macedonia and Bosnia-Herzegovina, both with low levels of education prior to 1945,
diverge from this general pattern. These two cases will require further research; given the high level
of decentralization in the Yugoslav school system, we currently lack sufﬁcient data on the content of
schooling in these provinces under communism.
23
With one notable exception, Ukraine, our units were the current countries and the historical data
are for the population that lived within the current boundaries. Given the signiﬁcant regional differences
in the content and extent of schooling prior to communism, we created two units out of Ukraine,
one for the three provinces that made up Austrian Galicia and another for the remaining provinces.
THE GREAT DIVIDE 91
FIGURE 1
PRECOMMUNIST SCHOOLING AND SHARE OF SEATS TO NONCOMMUNIST PARTIES
IN THE FIRST FREE POSTCOMMUNIST ELECTIONS
0 20 40 60 80 100
Levels of Precommunist Literacy
100
80
60
40
20
0
%SeatstoNoncommunists1stFreeElection
Kyrgyz Republic
Albania
Serbia
Romania
Bulgaria
Moldova
Macedonia
Bosnia-Herzegovina
Armenia
Ukraine W
Croatia
Georgia
Lithuania
Poland
Slovenia
Slovakia
Hungary
Czech Rep.
EstoniaLatvia
Ukraine E
Azerbaijan
Belarus
Kazakhstan
Tajikistan
Turkmenistan
Uzbekistan
Russia
92 WORLD POLITICS
Table 1 conﬁrm what is readily apparent from the scatterplot—that
precommunist schooling alone accounts for over 80 percent of the variance
(model 1).
Once we include precommunist literacy, the effects of urbanization
wash out (model 2). In contrast to modernization theory, then, we ﬁnd
that “all good things” do not go together: the impact of schooling is
independent both of prewar democratic statehood and of urbanization
levels (one proxy for industrialization in precommunist countries).24
The possibility that literacy captures the effects of precommunist statehood
and the advantages of national unity that it might confer are disconﬁrmed
by model 3, which includes a control for years of prior statehood.
The inclusion of these variables, either individually or together,
added nothing to improve the ﬁt of the model; nor did it change the explanatory
primacy of precommunist literacy. Finally, in contrast to the
structural accounts, neither a dummy for East Central Europe (which
would capture the effects of being outside of the Soviet Union) nor the
distance to Western capitals captures any of the variation once included
in the equation with precommunist literacy (models 4 and 5).
24
Janos (fn. 12).
TABLE 1
SHARE OF SEATS TO NONCOMMUNIST PARTIES IN THE
FIRST FREE POSTCOMMUNIST ELECTIONS
(OLS REGRESSION, BOOTSTRAPPED STANDARD ERRORS; 50 REPLICATIONS)
Model 1 Model 2 Model 3 Model 4 Model 5
Precommunist literacy rate .849 *** .960 *** .826 *** .777*** .669**
(.087) (.107) (.156) (.211) (.243)
Precommunist urbanization –.761 –.614 –.439 –.457
(.432) (.458) (.527) (.398)
Years of prior democratic .882 .769 .721
statehood (.980) (.834) (.996)
East Central European 6.29
dummy (9.27)
Distance to West –.007
(.0066)
Constant 5.65 13.91 13.92 10.89 27.14
(7.10) (9.64) (10.95) (9.26) (18.40)
N 28 28 28 28 28
Adj. r-squared .60 .62 .62 .63 .64
*** (P>|z|) = .000; ** (P>|z|) ≤ .01
All these results are robust to speciﬁcations of the communist exit
either as votes for the opposition or as seats captured by noncommunists.
Since there are no reliable, consistent, and comparable measures
of historical wealth for the region, we did not use estimates of historical
GDP as a proxy for precommunist modernization.25
Instead, we used
several other proxies: percentage of population employed in agriculture,
doctors per capita, and railroad density (kilometers of rail per square
kilometer of territory). None of these was substantively or statistically
signiﬁcant, either alone or in combination with the other variables.The
results conﬁrm our initial ﬁnding, that indicators of economic development
and modernization do not undermine the dominant impact of
precommunist literacy.26
In short, structural/geopolitical factors and precommunist legacies
(including those of modernization) explain less of the patterns of communist
exit than mass literacy. This suggests that we may be capturing
the unmeasured cultural differences that many scholars have attributed
to “regional effects” or the “Soviet legacy.” The identiﬁcation of prior
national schooling as the source of supposed “regional effects” is further
borne out with more microlevel comparisons. The regional differences
within the Soviet and Yugoslav cases bear out this conclusion. The
differences between, for example, Ukrainian nationalist Lvov and its
procommunist neighbor Zakarpatska, or between communist Western
Belarus and nationalist Lithuania show the importance of precommunist
cultural differences for postcommunist political trajectories.
While we recognize and acknowledge the potential problems associated
with cross-sectional country-level statistical analyses with a
relatively small number of observations, these ﬁndings are consistent
with our hypotheses about the link between precommunist schooling
and anticommunism. Further, this analysis suggests that nationalism,
which modernization ought to have supplanted in favor of more secular
understandings of societal interests, remained a driving force behind
the rejection of communism.
Moreover, concerns about endogeneity (a problem endemic in crosssectional
analyses of postcommunist data) are alleviated because the
25
The widely used historical GDP data set by Angus Maddison is unfortunately incomplete and
does not break down the data for the republics in the Soviet, Yugoslav, and Czechoslovak federations;
Maddison, Monitoring the World Economy, l820–l992 (Paris: OECD Development Centre, 1995). Much
of these data are simply not available, especially in the areas under Soviet control, such as Central
Asia. See David Good, “The Economic Lag of Central and Eastern Europe: Income Estimates for
the Habsburg Successor States, 1870–1910,” Journal of Economic History 54 (December 1994); and
Michael Pammer, “Proxy Data and Income Estimates: The Economic Lag of Central and Eastern
Europe,” Journal of Economic History 57 (June 1997).
26
Full results and the data set are available from the authors.
THE GREAT DIVIDE 93
94 WORLD POLITICS
primary variables of interest—precommunist schooling and postcommunist
electoral outcomes—are so far apart in time. Given the radical
changes that took place in this region between the onset of communism
and its collapse, the fact that a precommunist variable could account for so
much of the variation in a postcommunist outcome demands explanation.
III. CORRELATING MASS SCHOOLING AND THE COMMUNIST EXIT
What, then, were the patterns of mass schooling? To measure the extent
to which the population has been schooled, our primary measure is
the literacy rate, because by the turn of the twentieth century, virtually
all education was taking place in schools. Typically, a literate citizen
would have spent at least four years in school and been exposed to the
standard history, literature, geography, and music curricula that shaped
common national identities and political loyalties. The illiterate peasantry
or herdsmen, by contrast, were typically fragmented into “islands
of local customs and relationships, festivity and folklore, which were set
apart from the rest of society by dialect and tradition, as well as by the
limits on scale which oral communication inevitably imposes.”27
The
rate of literacy thus marks the extent of schooling and the degree to
which a standardized “high culture” and written history had become a
common culture that was shared on a mass scale.
With respect to precommunist literacy and culture, postcommunist
countries fall into three basic categories. In the ﬁrst group are those
with highly schooled populations and substantial national content in
the schooling curriculum at the onset of the communist period: the
Czech Republic, Estonia, Hungary, Latvia, Slovakia, Slovenia, Poland,
Lithuania, Croatia, Romania, and the Western Ukrainian region of
Galicia. All of these countries/regions had achieved over 70 percent
literacy before communism. Most of these areas had already sustained
universal primary education for several generations prior to communism
and typically also had high rates of secondary education and wellestablished
institutions of higher education.
The precommunist schooling in these countries was infused with
national content.28
In the Habsburg provinces that became Hungary,
27
James Sheehan, “What Is German History? Reﬂections on the Role of the Nation in German
History and Historiography,” Journal of Modern History 53 (March 1981), 8.
28
For a useful contemporary overview of these cases published under the auspices of the League
of Nations, see Jonathan French Scott, The Menace of Nationalism in Education (London: G. Allen
and Unwin, 1926). On the general development of history and social studies in education, which at
the time was highly infused with national and racial categories, see Aaron Benavot, Yun-Kying Cha,
David Kamens, John W. Meyer, and Suk-Ying Yong, “Knowledge for the Masses: World Models and
National Curricula, 1920–1986,” American Sociological Review 56 (February 1991).
the Czech Republic, Poland, Croatia, Slovakia, Slovenia, Romania, and
Ukrainian Galicia this nationalist content was, in part, the legacy of a
long-standing Habsburg policy of divide and rule—national antagonisms
were cultivated in an effort to prevent challenges to the monarchy
from coalescing.29
In the nineteenth century Vienna used school
curricula to cultivate a new Ukrainian identity in Galicia to counter
the active nationalism of the Poles and potential irredentism of the
Russians, and similarly assisted the national development of the Romanians,
Croats, and Slovaks to undermine the growing inﬂuence of
Hungarian nobles (the Hungarians, in response, cultivated Serbian
identity to divide the Serbo-Croats).30
Subsequently, the progressive
decentralization of Habsburg control over educational content also
created opportunities for new nationalist elites, particularly in Hungary,
the Czech lands, and Poland, to build support for their claims
to independent statehood. Even Croatia, the least educated of these
countries, had crossed a 50 percent threshold of literacy by 1910, and
its educational system had substantial national content. Thus, even under
imperial rule, the curriculum had national content and stressed the
cultivation of distinctive identities.
During the interwar period, these governments and the three Baltic
states were using nationalist school curricula to build loyalty to the
new states and to legitimate their territorial claims. Hungary, Romania,
Bulgaria, the three Baltic states, and (by the 1930s) Poland created
centralized, standardized school curricula and teacher training with a
strong focus on nationalizing and homogenizing their populations. In
the Czech and Slovak territories, the vast majority of schools cultivated
the identity of the titular nationality of their respective provinces:
schooling was decentralized, so the curriculum was often decided at the
level of the school.31
Elsewhere, in Transcarpathia, there was no consistent policy to introduce
national content in education, and a mix of Russophile Russian
émigrés, Rusynophile intellectuals, and Ukrainian nationalists from
neighboring Galicia provided the limited education that the province
29
In the words of Francis II: “My people are strangers to one another, and so it is best.They do not
take the same malady at the same time. In France, when the fever comes, it attacks you all the same
day. I put the Hungarians in Italy and the Italians in Hungary. Each watches his neighbor. They do
not understand each other and they detest each other. From their antipathy order is born, and from
their reciprocal hatred, the general peace.” Sarah E. Simons, “Social Assimilation I,” American Journal
of Sociology 6 (May 1901), 814.
30
Robert A. Kann, The Multinational Empire: Nationalism and Reform in the Habsburg Monarchy,
1848–1918 (New York: Octagon Books, 1970); Charles Jelavich, South Slav Nationalism: Textbooks
and Yugoslav Union before 1914 (Columbus: Ohio State University Press, 1980), 18–20; Paul Robert
Magocsi, A History of Ukraine (Seattle: University of Washington Press, 1996), 407–15.
31
Magocsi (fn. 30), 168.
THE GREAT DIVIDE 95
96 WORLD POLITICS
had achieved by 1938, when it was occupied by Hungary.32
Yugoslavia’s
educational apparatus was so poorly funded that prior teachers, curricula,
and many texts remained in place through the interwar years.33
This meant that regions with strong school systems and a history of
education (Slovenia, Croatia, and Vojvodina) typically retained their
literacy levels, whereas the less educated regions showed only marginal
advancement.34
In all these cases, the explicit purpose of the mass education
of the peasantry was to infuse them with new forms of national
loyalty believed to be necessary to secure loyal bureaucrats, loyal soldiers,
and loyal citizens.35
At the other end of the spectrum falls the second cluster of countries,
those with little or no exposure to literate culture prior to communism.
Albania, Azerbaijan, the ﬁve countries of Soviet Central Asia, Bosnia-Herzegovina,
and Macedonia each had no more than 30 percent
literacy when the communists took power. Prior to communism, these
regions had little or no schooling and were marked by high levels of
cultural heterogeneity and fragmented, localized identities and political
loyalties attributed to traditional societies.36
The few literate residents were either national minorities or imperial
administrators. In Azerbaijan and Central Asia those who could read
and write were typically Russians or other nontitular nationalities. Of
the non-Slavic populations of these regions, none but the Uzbeks had
achieved even 5 percent literacy by 1920, when the Soviet Union began
its literacy campaigns and conducted its ﬁrst census in the region.
In the tsarist empire, the limited schooling in Azerbaijan and Central
Asia was almost exclusively in Russian and the cheap, popular reading
materials of the time (lubok) were entirely Russocentric.37
Aside from a
handful of Jadidist (pan-Turkic nationalist) intellectuals, the schooling
and educational life of the region were entirely lacking in “national”
content that would link the local cultures of the region to a claim to
legitimate rule or sovereign rights. As late as 1909 Albania did not have
32
Ibid., 177.
33
Jelavich (fn. 30).
34
In fact the slight improvement in literacy rates for Yugoslavia as a whole is probably attributable
to the death of an older generation of illiterates rather than any signiﬁcant improvements in educa-
tion.
35
Hobsbawm (fn. 18), chap. 3; Posen (fn. 18).
36
Robert J. Kaiser, The Geography of Nationalism in Russia and USSR (Princeton: Princeton University
Press, 1994), 44; Nicholas Vakar, Belorussia: The Making of a Nation (Cambridge: Harvard
University Press, 1956); Loring Danforth, The Macedonian Conﬂict (Princeton: Princeton University
Press, 1995).
37
Kaiser (fn. 36), 67; Ben Eklof, Russian Peasant Schools: Ofﬁcialdom, Village Culture, and Popular
Pedagogy, 1861–1914 (Berkeley: University of California Press, 1986).
even a single teacher training school and the Albanian language had no
standard alphabet.38
Even by the end of the interwar period, Albania
had achieved only a meager 20 percent literacy rate and this training
was almost entirely of a religious nature. The schooling that these regions
eventually received would be under communist control.
In the remaining cases mass education was in the process of being
established or varied considerably within the country, having been
completed in some regions but hardly begun in others. In Romania,
for example, an overall literacy rate of 57.1 percent in 1930 masked
differences between Transylvania, with nearly full literacy and strong
national content as in other Habsburg communities, and Bessarabia,
where only 38 percent of the population could read and write. Serbia,
similarly, had nearly full literacy in Vojvodina and in areas around Belgrade
but exceptionally low rates of literacy in Kosovo, conﬁrming “that
the Serbian national program was not complete.”39
Russia, Ukraine, Belarus, and the interwar Moldovan ASSR (current
Transdniestria) had all achieved moderate rates of literacy, and the
youngest generation would have had nearly universal coverage in most
of these provinces.40
These populations were schooled to believe, however,
that they were part of a broader Russian nation, with a common
past in Kievan Rus, comprising the Great Russians, the Lesser Russians
(Ukrainians), and the White Russians (Belorussians).41
Hence, while
the Slavic populations of Belarus and Ukraine had achieved moderate
rates of schooling prior to communism, the schooling was Russian in
language and focused on broader Russian and Orthodox unity.
Of the cases where mass schooling was initiated but full literacy was
not yet attained prior to the communist period, the most interesting
are Armenia and Georgia, where the content of schooling was nationalist,
despite the fact that the main educational push was carried out
when these territories were under Soviet control. Following the Bolshevik
Revolution, Georgia and Armenia formed a separate independent
Transcaucasian Confederation with Azerbaijan that lasted until
1920–21. During this brief time, both republics set up national primary
school systems, with nationalist texts and teacher training.42
The Soviet
38
Barbara Jelavich, History of the Balkans (Cambridge: Cambridge University Press, 1983), 88.
39
Jelavich (fn. 30), 16.
40
Jews, who were nearly all annihilated on the territory of Ukraine and Belarus during the Second
World War and thus could have little effect on postcommunist trajectories, would have inﬂated the
precommunist literacy statistics of these regions somewhat.
41
Eklof (fn. 37)
42
On Georgia, see Wladimir Woytinsky, La Démocratie géorgienne (Paris: Librairie Alcan Lévy,
1921), 266–70; and Karl Kautsky, Georgia: A Social-Democratic Peasant Republic, trans. H. J. Stenning
(London: International Bookshops, 1921).
THE GREAT DIVIDE 97
government, according to Martin, following the annexation “boasted
that Soviet power had deepened the national work begun”43
by precommunist
nation-building governments. Teachers and curricula were retained,
and by the time that the “Great Retreat” from this policy began
in 1933 and especially by the “Great Purge” of 1937,44
the populations
of these territories were already schooled in an alternative pre-Soviet
national curriculum. Thus, even according to the 1939 census, ethnic
Georgians made up 72.2 percent of the teachers in the republic and
73.6 percent of the professors and researchers even though they were
only 61.4 percent of the population of the republic.45
Similarly, ethnic
Armenians constituted 84.2 percent of teachers and 88.5 percent of
professors and researchers in the Armenian SSR, both in excess of their
82.8 percent share of the population as a whole.46
More important than the raw percentage of titulars, however, is the
fact that the titular nationality had been educated in a fundamentally
different way in Georgia and Armenia than it had been elsewhere in
the USSR prior to World War II. As a result, even though Georgia and
Armenia had literacy of less than 50 percent upon their incorporation
into the USSR, these two countries preserved their pre-Soviet nationalist
school curricula and teaching cadre into the early 1930s, when the
population as a whole achieved nearly full literacy.
SECTION IV. FROM SCHOOLING TO ANTICOMMUNISM
The processes linking mass schooling to broad-based nationalism unfolded
via three mechanisms. First, schools brought mass literacy. Literacy
meant that face-to-face communication was no longer required
for the easy dissemination of ideas across time and space. It also dramatically
increased a society’s capacity to record and convey history,
literature, and myth; the amount that could reliably be stored in books
and accessed from them was much greater than what could be retained
in memory.47
Mass literacy allowed the social communication among
strangers that Deutsch, Anderson, and others have identiﬁed as a pre-
43
Terry Martin, The Afﬁrmative Action Empire: Nations and Nationalism in the Soviet Union, 1923–
1939 (Ithaca, N.Y.: London: Cornell University Press, 2001), 15.
44
It was a purge that, in Georgia, largely targeted teachers and the intelligentsia.Tamara Dragadze,
Rural Families in Soviet Georgia (New York: Routledge, 1988), 185.
45
USSR Census of 1939: Aggregate Statistics from the USSR Census of 1939 (Computer ﬁle, Toronto,
University of Toronto, Centre for Research in East European Studies), Robert E. Johnson, producer
(Toronto: University of Toronto, Data Library Service [distributor], 1995), table 63b.
46
Ibid., table 66b.
47
Jack Goody, The Power of the Written Tradition (Washington, D.C.: Smithsonian Institution
Press, 2000), 27.
98 WORLD POLITICS
condition of nationalism.48
Education thus increased society’s “carrying
capacity” for transmitting, replicating, and sustaining nationalist ideas.
Second, and more important, mass schooling used the curriculum to
convey nationalist ideas. Most schools in the late nineteenth and early
twentieth century purveyed nationalist content in the basic subjects of
literature, history, geography, and music and thereby inculcated a sense
of patriotism in the children. As a result, the rise of nationalism in
most European cases can be traced directly to changes in the education
and socialization of the young and to the simultaneous introduction
of nationalist content into school curricula. Prior to the development
of mass schooling, education and socialization consisted of children
imitating the rituals, mores, habits, skills, and beliefs of their parents or
elders.49
This form of informal education was inherently conservative.
Culture was predominantly learned—or replicated—through observation
and experience. Nationalism, however, was a novel idea that could
not be conveyed well through such traditional channels. It was not a set
of ideas prevalent among the previous generation; and as an abstract,
imaginary bond among a community of strangers, nationalism could
not be learned through experience. It could, however, be learned in
school. As the reading of school texts replaced the observation of experience
as the primary form of instruction, children came to think and
categorize the world more in terms of the general and the abstract than
in terms of the practical categories based on experience.50
The nation
and their place in it was one of the abstract categories that children
acquired in school.51
48
Karl W. Deutsch, Nationalism and Social Communication (Cambridge: MIT Press, 1953), 87, 101;
Benedict Anderson, Imagined Communities (New York: Verso, 1991), 44–46.
49
Scribner and Cole (fn. 20), 554–55.
50
Considerable experimental evidence supports the notion of a direct causal relationship between
schooling and abstract thought and reliance on general concepts not drawn from experience. See the
discussion of these experiments in A. R. Luria, “Towards the Problem of the Historical Nature of Psychological
Processes,” International Journal of Psychology 6, no. 4 (1971); idem, Cognitive Development:
Its Cultural and Social Foundations (Cambridge: Harvard University Press, 1976); Scribner and Cole
(fn. 20); and especially idem, The Psychology of Literacy (Cambridge: Harvard University Press, 1981).
As demonstrated most persuasively by the experiments conducted by Scribner and Cole (pp. 130–33),
with the Vai tribe of Liberia, the shift toward abstraction is linked only to Western-style schooling
(that is, curricular content), not to the development of literacy or written languages of the type of rote
memorization typical of Islamic madrassas.
51
In one of the few efforts to test the hypothesis linking schooling and nationalism systematically,
Howard Schuman, Alex Inkeles, and David H. Smith, in their study of East Pakistan in the 1960s,
found that even controlling for factors such as urbanization and occupation, literacy had a massive
substantive impact on the likelihood that respondents would identify ﬁrst and foremost as Pakistanis
(rather than with their Bengali, regional, or village identities); Schuman, Inkeles, and Smith, “Some
Social Psychological Effects and Noneffects of Literacy in a New Nation,” Economic Development and
Cultural Change 16 (October 1967).
THE GREAT DIVIDE 99
52
The model, emulated by most countries in both Eastern and Western Europe, was the Prussian
school system adopted in 1808. The ideology justifying the nationwide system was the need to build a
new set of shared ideas to legitimate state authority in the face of the breakdown of the old traditional
order.
53
On the importance of this modern “pyramid” of schooling for the development of nationalism,
see Gellner (fn. 18), 34.
54
The postwar demographic shifts had little impact on nationalist sentiments. Germans returned
to Germany, Poles moved from Ukraine and Lithuania into the new borders of Poland, and so on.
The forced migrations did not insert new populations with distinct linguistic or national identities
into established national communities. The tragedy of the Holocaust further homogenized the ethnic
composition of Poland, Hungary, and other countries.
Third, this content was monitored and purveyed by a broad, national-institutional
apparatus devoted to indoctrinating the younger
generation in a common set of ideas.52
In each community a bounded,
controlled setting was established. It was in this environment—the
school—that a majority of children spent much of their waking life
over a period of many years. Authority over the dissemination of ideas
in this setting was strictly controlled by a teaching staff. The staff, in
turn, was under the control of the state, educated in its central institutions,
and teaching from a generally standardized curriculum.53
This
curriculum and the broad nationwide educational apparatus that disseminated
it conveyed a consistent national culture over an enormous
territorial domain, typically within a single generation.
In short, mass schooling explains a critical element in the supply of nationalist
ideas. A considerable institutional apparatus is required to convey
an abstract, shared idea like common nationhood to masses of people
who have little or no face-to-face interaction with one another. The rise
of formal schooling accomplished this (1) by bringing a shift from oral,
informal socialization to general training based on the written word and
school texts, (2) by introducing nationalist content into school curricula,
and (3) by disseminating this nationalist cultural message through
titular-language teachers and standardized statewide institutions.
The mass schooling received by this ﬁrst literate generation was
critical, in that it “immunized” that generation against subsequent attempts
to inculcate the populace with different notions of nation, such
as those found under communist education. Once established, these
ideas were sustained despite border shifts, the decimation of wars, and
the migrations of populations. Thus, Germans living in Poland or in
Czechoslovakia saw themselves as distinctly German. And Poles forced
to migrate to the Soviet Union or the “Volga Germans” maintained
their respective identities.54
As a result, ideas initially introduced through schooling came to be
instilled in the popular culture and eventually accepted as common-
100 WORLD POLITICS
place truths. Unlike previous modes of ideational transmission such as
church organizations, which rested on the authority of trained elites
with cloistered knowledge, mass schooling embedded ideas in the community
at large, making all members “authorities.”55
Schooling carried
ideas directly into the household, and any schoolchild could replicate
the history of the nation, could sing the national songs, recite the national
poets, and explain the signiﬁcance of dates and symbols of national
importance. A society transformed by literacy was particularly
suited to the rigors of underground subversion, solidarity, and perseverance.
These identities were also robust to the enormous upheavals
of war, genocide, and population transfers that surrounded World War
II: the national identities were instilled in families and thus survived
with family units as they were transferred and resettled hundreds of
miles away. (Ironically, the increased population homogeneity in Poland,
Czechoslovakia, and Hungary that followed World War II also
meant greater homogeneity of anticommunist sentiment.)
Two caveats are important here. First, mass schooling and literacy
were not necessary conditions for the spread of nationalism—nationalist
ideas could also be conveyed through church pulpits or other forms
of mass oral communication, if not as effectively or irreversibly. Second,
schooling alone was not sufﬁcient to foment nationalism; it had to
be schooling with nationalist content. Nineteenth-century schooling
included new subjects such as geography and history that served to
inculcate ideas of nation and nationalism. Earlier, Protestant Reformers
and Jesuit Counter-Reformers had employed schools in a battle for
the minds of children in the seventeenth century, but such schooling
resulted in strong attachments to a sectarian catechism rather than to
different national identities.56
Nationalism and schooling become causally
linked only when a nationalist curriculum was introduced in the
nineteenth century. It was at that time that literacy began to spread
to the populations of eastern Europe and parts of what would become
the USSR. Hence, in the precommunist cases that we examine here,
nationalism accompanied the rise of the school.
FROM NATIONS TO VALUES
Notions of a shared national identity then provided standards for
what would constitute legitimate rule—governments consonant with
55
Gellner (fn. 18), 32.
56
C. John Sommerville, The Discovery of Childhood in Puritan England (Athens: University of
Georgia Press, 1992), chap. 9.
THE GREAT DIVIDE 101
102 WORLD POLITICS
national values that advance shared understandings of economic and
political development. Both potential counterelites and citizens used
three criteria in evaluating the legitimacy of communism. First, was
communism a domestic development or was it the imposition of an
occupying force? Second, was its sponsor, the Soviet Union, a historical
enemy or a more friendly power? Third, and most important, did communism
represent a political, cultural, and economic advancement for
the nation, or was it an antimodern step backward? Put differently, was
communism in keeping with religious, cultural, and economic national
expectations?
Where national schooling existed prior to communist rule, these
questions were answered in the negative. The communist regime was
perceived to be an alien, inferior imposition by a suspect regional superpower.This
is not to say that some local elites were not fervent communists
who played a key role in establishing communist rule or that
the values of socialism, such as guaranteed employment, social security,
and egalitarianism, were rejected. But as both historical accounts and
communist-era public opinion polls show, the communist regimes were
not seen as legitimate. Instead, communist parties ﬁrst battled popular
resistance in establishing their rule and then battled apathy once it was
established, while the plurality of society looked back with nostalgia to
the precommunist era.57
Thus, after World War II, the Communist Party experienced success
in free elections only in Czechoslovakia, where a strong local tradition
of social democracy had existed prior to the war.58
This was largely the
result both of a moderation in communist appeals and of the uncertain
status of the newly regained Sudetenland.59
As soon as the communists
began the move to consolidate their power, however, they met with
protest and criticism. In Slovakia the communists lost the 1946 elections
by a wide margin. In Poland and in Hungary communism arrived
with Soviet tanks, was imposed with the clear support of the Soviet
Union and its occupying forces, and was never approved electorally.
57
See in particular the chapters by Bela Zhelitski, John Micgiel, and Inessa Iazhborovoskaia, in
Norman Naimark and Leonid Gibianskii, eds., The Establishment of Communist Regimes in Eastern
Europe, 1944–1949 (Boulder, Colo.: Westview Press, 1997). The few reliable communist-era public
opinion polls tended to show a support for socialist values without support for the regime itself (in
Poland, for example, roughly 70 percent of the students polled supported socialist values but rejected
Marxism). In both the Czech Republic and Poland, these polls also show interest and gloriﬁcation of
the prewar era. Walter D. Connor and Zvi Gitelman, Public Opinion in European Socialist Systems (New
York: Praeger, 1977), 45–178.
58
The Communist Party won 38 percent in the Czech lands but far less in Slovakia.
59
The Red Army and Soviet presence were seen as the main guarantee of border integrity in the
Sudetenland.
Even in Yugoslavia, where communism was “homegrown,” it support was
not ensured.60
Russia and then the Soviet Union were seen as enemies of both the
nation and the national idea. This notion stemmed both from the wartime
experiences of Soviet occupation (and in the case of Poland, the
refusal to aid the anti-Nazi uprisings) and from the prewar teaching of
history. The partial exceptions here are Bulgaria61
and to some extent
Serbia:62
where nationalist schooling also included a heavily pro-Russian
and then a pro-Soviet component, largely due to Russia’s role in
the liberation of these areas from Ottoman rule.
For countries that had tasted independent economic and cultural development
in the interwar period, communism represented a step back
from modernity and from national values. It was secular, whereas many
of the national ideas of Poland, Hungary, Croatia, and Slovakia were
closely aligned with the Catholic church.63
It was nominally antinationalist
and cosmopolitan at a time when local national sentiments were
intense. Finally, it entailed an industrial production proﬁle that ignored
local comparative advantages and pushed both standards of living and
gross national products below prewar levels. For the petite bourgeoisie
of Czechoslovakia, Slovenia, and Hungary, nationalization of property
and the perception that economic policy now favored Soviet interests
represented an economic disaster.64
Thesestandardsforwhatwouldconstitutelegitimaterulearosefroma
set of shared expectations about who would best serve national interests
and about what constituted the bundle of cultural and economic goods,
civic rights, and political responsibilities such a government ought to
provide for its citizens.The existence of these standards then led to two
sets of comparisons made throughout the communist era: with noncommunist
neighbors and with the counterfactual for the communist
countries—the imagined trajectories of development had communism
not been imposed. As a result, where Central Asian republics saw de-
60
The victorious Yugoslav partisans under Tito relied on the presence of the Soviet Army and its
entrance into Belgrade in 1945, to ensure communist rule.
61
The lower enmity to Russia in Bulgaria stems from the 1905 liberation of Bulgaria from Ottoman
rule by the Russians.
62
The Serbian Orthodox church claimed a kinship with its Russian Orthodox brethren.
63
The most explicit example of this conﬂation is the Polish Black Madonna, both an object of
religious veneration (pilgrimages, prayers) and the symbol of divine protection of Poland (credited
with inspiring the resistance to the Swedish invasion in the seventeenth century and with the success
of the Bolshevik War of 1920).
64
For example, Hungarian postwar inﬂation was the worst the country had experienced. Similarly,
industrialization plans in the Czech lands failed largely because the country was already industrialized,
had few natural resources, and had been dependent on foreign trade, largely with Germany.
THE GREAT DIVIDE 103
104 WORLD POLITICS
velopment and enormous advances in the rights of women, economic
growth, industrialization, and the building of national infrastructure,
many of their East Central European counterparts saw a massive slide
backward into oppression,economic stagnation,nationalization of markets,
and national underdevelopment. In Poland the constant standard
of reference was przedwojenny, or “prewar,” an adjective used to describe
lost excellence in everything from buildings to scouting associations to
fruit preserves. In Hungary, Czechoslovakia, and Poland, scholars covertly
calculated the economic costs of belonging to the Soviet sphere
of inﬂuence and of having to engage in forced trade with the Soviet
bloc on unfavorable terms, while their Slovenian and Croatian counterparts
estimated the costs of subsidizing other republics in the Yugoslav
federation.The comparisons were not favorable. Even as these societies
became more industrialized and urbanized under communism, for example,
relative incomes did not increase. While East Central European
incomes per capita averaged 37 percent of West European per capita
incomes in the prewar 1930s, this proportion dropped slightly to an
average of 35 percent by the 1970s and 1980s.65
Even where material
conditions improved under communism, as in Slovakia and Western
Ukraine, the hostility to communism continued unabated.
At the same time, as Bunce and Kopstein and Reilly point out, geographic
proximity meant constant and invidious comparisons to neighboring
countries. For Czechs, Slovaks, Slovenians, and Hungarians, the
proximity of Austria (and their shared historical past) was one benchmark
against which their countries could be measured—and their postwar
development was found to be wanting. The availability of Austrian
and German media broadcasts in the border regions only accentuated
the shortcomings of the socialist system. For Estonians and Latvians,
Finland provided the comparison. Romanian intellectuals looked to
France, whereas for Polish dissidents, Scandinavia provided an alternative
model.66
As a result, even though communism produced a massive
economic and societal transformation with some positive effects, particularly
in developing rural areas, such achievements were discounted
in comparison with the counterfactual of noncommunist statehood and
the ready comparisons with the West. As one observer noted, the prevailing
sentiment was that “if not for communism, we would have been
like the West.”67
65
Figures calculated from Janos (fn. 12), 349.
66
A 1970s joke plaintively asked, “Why did the Black Madonna ﬁght back the Swedish onslaught?
We could have been Sweden . . . ”
67
Adam Przeworski, Sustainable Democracy (New York: Cambridge University Press, 1995), 3.
The same dynamics appear in those nations of the former Soviet
Union that had received mass schooling prior to the advent of communism.
In Georgia communism was seen as alien and violently imposed
during the Bolshevik annexation of the Transcarpathian republic in
1921 and the repression of the Georgian uprising in 1924. In the 1920s
and 1930s senior Soviet ofﬁcials (including Stalin himself) recognized
that Georgians refused to accept the legitimacy of Soviet communism,
especially in Western Georgia.68
A detailed ethnographic account from
the 1970s from the Abari and Likheti regions found that even rural
residents romanticized their precommunist past, schooled their children
at home in precommunist national myths, and viewed Russian
and Soviet culture as inferior.69
Sentiments in Armenia tended to be
more pro-Russian, because in precommunist Armenian myth, Russians
were depicted as providing Armenians with a haven from their historical
oppressors, the Turks.
In Estonia, Latvia, Lithuania, and the areas of Ukraine that had national
schooling prior to their incorporation into the USSR, anti-Russian
sentiments were—and remain—pervasive. Even though the USSR
brought industrialization and electriﬁcation to rural areas, communism
never achieved broad popular support; indeed, it was seen as undermining
the more progressive European character of these regions. And nationalist
views were expressed in underground publications despite the
ruthless suppression of any public expression of anticommunist sentiment.
In Ukraine these regions had considerably higher levels of arrests
of nationalist dissidents in the 1960s and 1970s.70
And as in Georgia,
a parallel underground culture existed alongside ofﬁcial Soviet life that
consisted of anticommunist histories, literature, and songs. Signiﬁcant
religious or national holidays were celebrated privately and the underground
culture was passed on in private gatherings of family and friends.
Communism was seen as forcibly imposed by Russian (or “Muscovite”
in the Ukrainian case) occupation, and the horrors of Soviet annexation
in 1939 and then 1944 were recounted to show the nonindigenous,
ethnically Russian (or Jewish) nature of communist rule. Local communists
were viewed as collaborators with an occupying power. Only
anticommunist activities were considered patriotic and loyal.
By contrast, in Central Asia and Azerbaijan the comparisons with
both the pre-Soviet past and the noncommunist neighboring states were
68
Tamara Dragadze, Rural Families in Soviet Georgia (New York: Routledge, 1988).
69
Ibid., 183–84
70
Kenneth C. Farmer, Ukrainian Nationalism in the Post-Stalin Era: Myth, Symbols and Ideology in
Soviet Nationalities Policy (Boston: Martinus Nijhoff, 1980), 176–79.
THE GREAT DIVIDE 105
highly favorable. Communism, closely linked to the Soviet Union, was
identiﬁed with enlightenment and progress, rather than with forced
imposition. Communism was credited with bringing electriﬁcation,
industry, schooling and high culture, and all forms of modern life to
the region. It was associated in particular with the liberation of women
from the oppression of traditional life. In Central Asia primary identiﬁcation
appears to have been with the communist-era administrative
units (oblasts), rather than with clans, tribes, or any real or imagined
precommunist community.71
The precommunist past was seen in the
late 1980s and the 1990s largely as it had been depicted in Soviet accounts,
as a world of ignorance, superstition, and poverty.72
Even as
speciﬁc policies from Moscow could be subject to criticism, the general
assessment was that Soviet rule had brought dramatic improvement to
the lives of Central Asians. Russians and their cultural impact on the
region were viewed in a very positive light.73
For the Central Asian states, comparisons with their neighbors conﬁrmed
their favorable view of the impact of communism. Today, Afghanistan
and Pakistan still represent examples of life without communism.
Afghanistan under the Taliban ﬁt precisely the Soviet stereotype
of precommunist life. Only neighboring China is viewed in a favorable
light, but as a communist country the Chinese example simply reinforces
the general view equating communism with the forces of enlightenment
and modernity in the region. Indeed, it is typically viewed
as the counterfactual “path not taken,” that is, what life might be like
had communism not ended with the USSR.
In areas of Ukraine, Belarus, and Moldova schooled by the Soviets
—much like Central Asia—pro-Russian sentiments predominated and
continue to predominate. Consistent with Soviet ideology, communism is
associated with technology, industry, and progress. Even in the areas of
Ukraine hit hardest by the famine during collectivization, communism
is credited with bringing the mechanization of agriculture, literacy, and
progress.Indeed,surveys indicate that these areas are the most staunchly
71
See Jones Luong (fn. 11), chap. 3.
72
The same depictions of the positive impact of communist rule depicted in the 1920s propaganda
ﬁlm, Three Songs for Lenin (Dziga Vertov), were repeated in many conversations with Central Asian
respondents in the mid-1990s.
73
According to State Department surveys conducted through the 1990s, the share of respondents
with a “favorable” opinion of Russia in 1994 was 74 percent in Azerbaijan, 78 percent in Kazakhstan,
86 percent in Uzbekistan, 83 percent in Kyrgyzstan, 90 percent in Armenia (up from 43 percent in
1992), but only 36 percent in Georgia. See Regina Faranda, “Ties That Bind, Opinions That Divide”
(Manuscript, U.S. State Department Opinion Surveys, 2001); see also David D. Laitin, Identity in
Formation: The Russian-Speaking Nationality in Estonia and Bashkortostan (Glasgow: Centre for the
Study of Public Policy, University of Strathclyde, 1995).
106 WORLD POLITICS
pro-Russian and most nostalgic for the Soviet system.74
Communism
was also viewed as an indigenous movement in these regions, with local
communist heroes lionized and considered to be true patriots. While
Ukrainians and Belarussians see themselves as having a distinct identity
and culture, this culture is viewed as closely related and equal to
or inferior to the literature and art of the broader Russian community.
The collective effort to defend the Soviet Union from Nazi aggression
is considered the major patriotic achievement of the local populations
and the Communist Party, especially in Belarus. Communist partisans
and Red Army soldiers were viewed as defenders of the motherland,
not as an occupying force.75
FROM COMPARISONS TO OPPOSITION
These widespread comparisons meant that the appeals of nationalist
opposition elites would ﬁnd different levels of resonance among the
populace. Across the communist world, elite groups were opposed to
communist rule, criticizing it on nationalist and cultural grounds. However,
whether or not such critiques would ﬁnd broad popular support
depended on extant national sentiment and identity. Where nationalist
schooling existed, anticommunist opposition resonated with popularly
held notions of communism as an alien and antimodern imposition.
Such opposition took the form of highly organized mass movements,
such as Solidarity in Poland in 1980 (with over ten million members, or
one-third of the adult population) or the smaller but similarly recognizable
opposition groups in Hungary. In Slovenia a nationalist communist
leader, Milan Kucˇan, assumed power in 1986 and led the republic
to greater independence from the Yugoslav federation, culminating in
the 1990 elections of DEMOS, a six-party opposition coalition that had
arisen in the late 1980s.
The anticommunist opposition, whether organized or still inchoate,
made explicit references to the prewar past. Prewar national holidays,
such as August 17 (the anniversary of the Battle of Vistula during the
Bolshevik War) or May 3 (the 1792 constitution) were occasions for
Poles to ﬂy the Polish ﬂag on their houses and balconies.76
March 15,
74
In 1994, 81 percent of respondents in Ukraine and 88 percent in Belarus took a favorable view
of Russia, and these ﬁgures were consistent throughout the 1990s. Only 27 percent of respondents in
Western Ukraine viewed Russia as an ally; Faranda (fn. 73), 44.
75
Andrew Wilson, Ukrainian Nationalism in the 1990s: A Minority Faith (New York: Cambridge
University Press, 1997); Karel C. Berkhoff, Harvest of Despair: Life and Death in Ukraine under Nazi
Rule (Cambridge: Harvard University Press, 2004), 206–13.
76
See Jan Kubik, The Power of Symbols against the Symbols of Power (Pittsburgh: Pennsylvania State University
Press, 1994); Janine Wedel, The Private Poland (New York: Facts on File Publications, 1986).
THE GREAT DIVIDE 107
the anniversary of the 1848 revolution in Hungary, was similarly celebrated
with marches and slogans, many of which took on an anticommunist
cast. In 1989, when the opposition’s demonstration was ﬁve
times larger than the Hungarian communists’ ofﬁcial observance, the
holiday conﬁrmed the public rejection of the communist regime.77
Where the communist regime had been more repressive but schooling
had inculcated nationalist identities, the opposition consisted of ad
hoc coalitions of mass movements uniting behind elites in the last days
of the communist regime. Thus, in Czechoslovakia, mass opposition
groups, Public against Violence and Civic Forum, arose only in the
days and weeks before the communist collapse. The Czechoslovak case
also shows that the communists may be forced to exit in face of popular
hostility, even where no entrenched organization of dissent exists. The
ultimate cause for the communist exit is the popular sentiment engendered
by nationalist schooling and the comparisons it breeds, rather
than the power of elite organization itself. Conversely, Communist
Party strategies may shape the form of the protests, but they do not
determine the party’s legitimacy or the extent of its delegitimation.
Once communism began to collapse, these forces revived prewar
democratic parties (the Polish PPS, Czech Social Democrats, Hungarian
Smallholders’), resuscitated prewar party acronyms (the Polish BBWR),
and adopted historical institutional solutions (parliamentary systems
or prewar constitutions, which were initially adopted wholesale by the
new postcommunist governments in the Baltic Republics).The essence
of the opposition project was encapsulated in the slogan of a “Return to
Europe,” as the motivation for the enormous project of democratization,
market reforms, and social transformation that followed.78
This is
not to say the prewar period was idolized without reservations or without
awareness of its obvious ﬂaws: but much of the opposition saw the
communist collapse and the possibility of establishing new domestic
forms of rule as a return to “normalcy,” the natural state of these na-
tions.79
In the former Soviet Union internal variation corresponds precisely
to precommunist nationalist schooling. Popular anticommunist opposition
movements appeared in the three Baltic states, Moldova, Georgia,
and Western Ukraine and amassed demonstrations of over one hundred
77
Rudolf Toʹʹkes, Hungary’s Negotiated Revolution (Cambridge: Cambridge University Press, 1986),
318.
78
Even the 1989 slogan for Kucˇan’s extremely liberal Slovene communists was “Europe Now!”
79
See Padraic Kenney, A Carnival of Revolution (Princeton: Princeton University Press, 2002).
108 WORLD POLITICS
thousand people, in many cases already as early as 1988.80
In each of
these cases the movement organizers laid out an explicit nationalist
agenda that drew on banned precommunist ﬂags and symbols (the ﬂags
of the interwar period in the Baltics, the blue-and-yellow ﬂag and trident
in Western Ukraine, the Georgian ﬂag from 1918, and the ﬂag
and symbols of interwar Romania in Moldova). The largest protests
typically took place on dates of symbolic signiﬁcance in the precommunist
national calendar, such as dates of independence. One of the
ﬁrst large-scale demonstrations, for example, held on May 28, 1988, in
Armenia by the Association for National Self-Determination, marked
the sixtieth anniversary of precommunist Armenia’s date of independence
in 1918.81
Consistent with our claims, Karklins argues that such
“calendar demonstrations,” as they were called in the Baltics, rested on
a shared cultural knowledge that “facilitated mobilization by reducing
the need for formal communication.”82
Banned precommunist anthems,
nationalist songs, or songs from the wartime anticommunist partisans—known
to all—were also used to mobilize mass protests.83
The
movements also demanded revision of communist historical narratives
(surrounding the Molotov-Ribbentrop pact in the Baltic case) and the
protection of important national or religious sites (the Davitgaredzha
monastery in Georgia).84
In each of these countries, anticommunist opposition
was broad based, took the form of a nationalist movement, and
drew on precommunist myths, symbols, and rituals.
These movements channeled directly into political organizations that
defeated the Communist Party in the ﬁrst set of free elections in 1990.
In Georgia nationalist organizations such as the Chavchavadze society
coalesced into the Free Georgia movement, led by the nationalist
dissident Zviad Gamsakhurdia. The movement drew explicitly on
Georgia’s heroic pre-Soviet past (Chavchavadze was a prominent nineteenth-century
nationalist writer). Gamsakhurdia’s nationalist agenda
decried mixed marriages and even Soviet industrial advancements such
80
As these movements and the manner in which they drew on precommunist national symbols
and dates of signiﬁcance to mobilize the population are demonstrated in detail by Beissinger, we will
give only a brief overview here. See Mark R. Beissinger, Nationalist Mobilization and the Collapse of the
Soviet State (Cambridge: Cambridge University Press 2002), chap. 4.
81
Beissinger (fn. 80),187.
82
Rasma Karklins, Ethnopolitics and Transition to Democracy: The Collapse of the USSR in Latvia
(Washington, D.C.: Woodrow Wilson Center Press, 1994), 94–95, cited in Beissinger (fn. 80), 169.
83
In Estonia, for example, the “singing revolution” replicated the national singing festivals of the late
nineteenth century, singing precommunist songs and mobilizing as much as a third of the population into
demonstrations calling for independence; Beissinger (fn. 80), 172. Robert Person,“Resisting Hegemony:
Transformations in Estonian Identity under Soviet Rule” (Manuscript, Yale University, 2005).
84
Beissinger (fn. 80), 180.
THE GREAT DIVIDE 109
85
Zviad Gamsakhurdia, “Open Letter to Eduard Shevardnadze,” translated from the Russian by
the Zviad Gamsakhurdia Society in the Netherlands, April 19, 1992.
86
Beissinger (fn. 80), 194–98; Wilson (fn. 75).
110 WORLD POLITICS
as the “building of gigantic hydro-electric power stations and enterprises”
as “a manifestation of an ecological war against Georgia and
in the end its aim was the genocide of its people,” and it attacked the
importation of Soviet engineers to run the factories as part of a plan for
the forced assimilation of Georgia.85
In Armenia the nationalist movement to unite the Nagorno-Karabakh
region of Azerbaijan with Armenia repeatedly drew crowds in
the tens of thousands to the streets in the late 1990s. The de facto
leader of the Karabakh committee, Levon Ter-Petrosian, a nationalist
linguist born outside the USSR, drew on historical myths of a greater
Armenia to legitimate claims to Karabakh. With the ﬁrst free elections
in 1990,Ter-Petrosian and the Karabakh committee rode to power easily
and Ter-Petrosian became head of state. In all three Baltic states
the national movements rapidly developed political organizations with
broad popular appeal (ENIP, the Estonian Popular Front, and the defection
of a proindependence group from the ECP in Estonia, Sajudis in
Lithuania, and the Latvian Popular Front). Each of these swept the
communists from power in the Supreme Soviet elections in 1990. In
Ukraine, despite considerable repression from the hard-line communist
leadership in Kiev, massive nationalist demonstrations in the provincial
capitals of Galicia (Lvov, Ivano-Frankovsk, Ternopol) led to the formation
of Rukh, which took control of these local governments in the
1990 elections and took all of the elected seats from these regions to
the Supreme Soviet elections of 1990.86
In other Ukrainian provinces
the Communist Party safely and comfortably held on to power, even
in Transcarpathia, which had also been incorporated into the USSR
after World War II. It was only in those Ukrainian provinces that had
precommunist national schooling that the communists were voted out
of power. What was true for Ukraine was true for the Soviet Union as
a whole: where mass schooling was introduced under communism, the
communists held on to power in the elections of 1990.
Where no nationalist schooling took place, there was no pool of national
sentiment from which elites opposing the communist regimes
could draw. Even where prominent opposition ﬁgures existed, such as
Mihai Botez, Doina Cornea, and Mircea Dinescu in Romania, they
were unable to foment the kind of mass opposition to communist rule
that was the immediate cause of the communist collapse. Without the
groundswell of demands for their exit, Romanian communists were
THE GREAT DIVIDE 111
able to hold on to power as their counterpart regimes fell. Their internal
coup of 1989 kept them in power until the 1996 ﬁrst free elections,
which forced them out. Similarly, the communists in Albania held on
to power through the elections of April 1991 with over 56 percent of
the vote, leaving ofﬁce only in May 1992.
Bulgaria appears at ﬁrst to be an exception to the correlation between
nationalist schooling and the communist exit: despite the introduction
of Bulgarian national schooling (albeit one that was strongly
pro-Russian and pro-Soviet, given the Russian liberation of Bulgaria in
1905), the Communist Party formed the government after the ﬁrst free
elections in 1990, with 47 percent of the vote.87
It is important to note,
however, that the combined anticommunist opposition actually won
more votes, with 50 percent: the Union of Democratic Forces, the main
opposition movement, received 36 percent of the vote, the Bulgarian
National People’s Union (a throwback to prewar agrarian parties) received
8 percent, and the Turkish minority opposition party, 6 percent.
In short, nationalist schooling produced shared expectations and
standards for a legitimate government. It sustained both anticommunist
sentiment throughout the postwar era and the opposition that
arose (and organized where the communist regime made it possible).
The overwhelming and shared desirability of a “return to normalcy”
produced an elite consensus about the need for a new and noncommunist
regime, even if, as the cases of Slovakia, Croatia, Georgia, and Armenia
show, elite commitments to democracy itself were far shakier. As
Valerie Bunce points out, nationalism is “wanton” and can coexist with
both democratic and authoritarian regimes. It is the content of nationalism,
not its advocacy of uniting political and administrative units, that
determines whether it serves as a force for consolidating democracy or
for undermining liberal rule.88
V. CONCLUSION
To lengthen and deepen the causal chain behind the patterns of communist
exit from power, we argued that precommunist nationalist
schooling produced the shared memories and standards that made
popular acceptance of communist rule unlikely. Constant comparisons
to both the prewar era of independence and to neighboring noncom-
87
They then exited ofﬁce after their government collapsed in November 1991.
88
See Milada Anna Vachudová and Timothy Snyder, “Are Transitions Transitory? Two Models of
Political Change in East Central Europe since 1989,” East European Politics and Societies 11 (Winter
1997)
112 WORLD POLITICS
munist countries meant that where nationalist schooling existed, such
comparisons would strengthen anticommunist convictions. As a result,
decades of these comparisons under communism led to the rise of
widespread opposition to communist rule (whether highly organized
or not). In a much shorter causal link, this opposition then led to the
communist exit, both because communist rule had been discredited and
because the opposition provided a salient and far more legitimate governing
alternative. By contrast, where schooling was introduced under
communism, no such invidious comparisons were made. Rather, the
shared understanding was that communist rule brought modernity and
its attendant advancements.
The communist exit pinpoints which factors identiﬁed by earlier
modernization analyses changed popular values and expectations regarding
good governance. While education, economic development,
and urbanization were bundled together by these earlier analyses, we
show that it was mass schooling—and the national content it transmitted—that
inﬂuenced the degree to which communist rule was viewed
as legitimate (or not) and the subsequent mobilization against it. Other
factors may correlate with each other, but they do not explain the developments
in the communist world or the eventual trajectories of its
collapse.
The timing and content of schooling explain more variation than do
the alternative models, while also complementing them. The account
presented here elaborates the underlying mechanisms of geographic
proximity and prewar legacies and shows why the anticommunist opposition
arose in the ﬁrst place. The key to these extensions and to the
causal chain behind the communist exit is the reincorporation of the
microfoundations of individual and societal behavior: the shared understandings,
expectations, and standards for legitimate government.
These notions are why anticommunism was equated with nationalism
(even if not always with democracy). National identities may have also
helped to propel these countries through the troughs of painful market
reforms and confusing political change, generating popular acceptance
of the sacriﬁces necessary.This, then, may be one way in which “culture
matters”—not as a structural predisposition for particular behaviors or
as an afﬁnity for speciﬁc institutional solutions but as a historically created
set of widely shared standards for government origin, legitimacy,
and performance.
APPENDIX 1
LITERACY AND COMMUNIST EXIT
Literacy Rate % Years of
at Onset Precommunist Noncommunist Democratic
of Communist Curricular Seats in First Precommunist Statehood
Schooling Content Free Elections Urbanization Precomm ECE
Czech
Republic 99 National 84 25 20 1
Estonia 99 National 74.3 33 14 0
Hungary 96 National 83.2 42.5 4 1
Latvia 93 National 73 36 14 0
Slovakia 92 National 85 15 20 1
Slovenia 91 National 87 9.3 8 1
Poland 79 National 87 20 8 1
Lithuania 77 National 67.3 15 4 1
W. Ukraine
(Galicia) 70 National 95.7 23.2 8 0
Bulgaria 69 National 48 21 0 1
Croatia 68 National 75 9.3 8 1
Georgia 65 National 74 22 0 0
Romania 57 National 33 21 0 1
Armenia 56 National/Russian 89.8 19 0 0
Moldova 46 Romanian 54.4 16 0 0
E. Ukraine 46 Russian/Russophile 13.9 19 0 0
Russia 44 Russian 14 17 0 0
Belarus 42 W – Polish/Russian
E – Russian 21 19 0 0
Serbia 35 National 22.4 9.3 8 1
Macedonia 30 Bulgarian, Serbian 74.7 9.3 8 1
Bosnia- Bosniak/Yugoslav,
Herzegovina 28 Croat, Serb 79.2 9.3 8 1
Albania 20 Religious 32 13 5 1
Azerbaijan 20 Russian 22.2 28 0 0
Kazakhstan 18 Russian 5.6 9 0 0
Kyrgyz Republic 12 Russian 0 12 0 0
Uzbekistan 11 Russian, some
Turkic/Arabic 5 22 0 0
Turkmenistan 10 Russian 5 14 0 0
Tajikistan 4 Russian 5 10 0 0
114 WORLD POLITICS
APPENDIX 2: OPERATIONALIZATION
COMMUNIST EXIT
We use the share of parliamentary seats taken by noncommunist candidates
in the ﬁrst free elections. We measure seats because (1) in the
post-Soviet cases the share of the vote for different candidates is not
available and (2) the share of the seats is a more direct measure of
whether the communists were able to hold on to power, which is the
variable of interest. The results, however, are robust to both speciﬁcations
of the communist exit as seats and as share of the votes.
The ﬁrst free elections in the ECE cases were all held in 1990–91.
For the Yugoslav and Soviet cases, the ﬁrst free elections were held
while those federations still formally existed, so we use elections to the
republican bodies in each case. For the post-Soviet countries, the ﬁrst
free elections were the elections to the Republican Supreme Soviets in
the spring of 1990. The Communist Party had ceased to hold a legal
monopoly on power, the media were quite free and diverse, all residents
were allowed to vote, there were few restrictions placed on the nomination
of candidates or political mobilization, and, with the exception of
Russia, the rules and institutions were standard across all of the republics.
After spring of 1990 many of these countries never held another
free election, subsequently denied the vote to a substantial share of the
populations, or banned the Communist Party.
PRECOMMUNIST LITERACY
We use the precommunist literacy rate. With the exception of Georgia
and Armenia, this is simply the literacy rate at the time the communists
took power. For Georgia and Armenia we use the literacy rate in 1933,
the year that the Communist Party began to replace the precommunist
nationalist school curriculum.
NATIONAL CONTENT
In several cases the precommunist schooling did not cultivate titular
nationalism. Macedonians were educated in Serbo-Croatian and told
they were Serbs. Most schools in the Russian Empire were Russophone
and Russophile. For example, Russian was the language of schools in
Belarus and Ukraine, and students were taught that they were Russian.
For this reason, we also code a dichotomous variable to identify titular
national content of the precommunist education.
THE GREAT DIVIDE 115
PRECOMMUNIST URBANIZATION
Given the difﬁculties of calculating GDP, particularly at a provincial
level, we use urbanization—the share of the population living in cities
and towns—as a proxy variable to control for the level of economic
development at the onset of communism. Urbanization is strongly correlated
with industrialization: the correlation between urbanization and
industrialization is .885 across countries with over million inhabitants
in 1945.89
In the precommunist world the development of industry coexisted
with a large rural population that remained active in agriculture,
and urbanization rates signiﬁcantly lagged behind West European
rates.90
Some industries such as textiles in the Bohemian lands, were
also located in the countryside.91
Nonetheless, the urban centers were
the epitome of modern society, “the home and the symbol of industry,
commerce, and powerful banks.”92
YEARS OF DEMOCRATIC STATEHOOD
We follow Kitschelt and use the number of years of independent democracy
prior to the onset of communist rule.
DISTANCE FROM WEST
We replicate the coding from Kopstein and Reilly,93
measuring the distance
from the capitals to either Berlin or Vienna (whichever is closer.)
In the case of Ukraine, we used Kyiv as the capital of East Ukraine, and
Lviv as the capital of West Ukraine.
89
Thomas Wilkinson, “Urban Structure and Industrialization,” American Sociological Review 25
(June 1960), 357. Urbanization here is deﬁned as the proportion of the population residing in politically
deﬁned cities.
90
See Ivan Szelenyi, “Urban Development and Regional Management in Eastern Europe,” Theory
and Society 10 (March 1981), 180. Iván Berend and György Ránki, Economic Development in East-Central
Europe in the 19th
and 20th
Century (New York: Columbia University Press, 1974), 154.
91
Gale Stokes, “The Social Origins of East European Politics,” in Daniel Chirot, ed., The Origins
of Backwardness in Eastern Europe (Berkeley: University of California Press, 1989), 217.
92
Berend and Ránki (fn. 90), 310.
93
Kopstein and Reilly (fn. 5).