Comparative Opportunities

Introduction

In retrospect, the term Eastern Europe is kind of confusing. Interestingly for a phrase with a direction making up one of the two words, it has never really been a purely geographic term. The Czech Republic and Hungary are basically west of Finland and Greece, but the former are East European countries and the latter aren’t. Moreover, Moldova, Estonia, Latvia, and Lithuania—all pretty much due south of Finland and north of Romania and Bulgaria—were not part of Eastern Europe as long as they were part of the Soviet Union. Nor does the term identify some sort of imperial legacy: we can find countries that were part of the Hapsburg, Prussian, Ottoman, and Russian empires both inside and outside of Eastern Europe. Nor is it some sort of religious demarcation: there are Eastern European countries where the largest religious groups are Eastern Orthodox, Catholics, Protestants, Muslims, and even atheists.

Perhaps the only way we can make sense of the term Eastern Europe in the manner in which it has been used for the past 50 years is through the lens of Soviet communism. Simply put, Eastern Europe as an “ideal type” classification is best identified as those countries outside the Soviet Union that in the aftermath of World War II came under the rule of Soviet-style communist regimes. ¹ If we want to try to put a legal definition on it, we’d probably be forced to go through something like the Warsaw Pact. But of course even these types of definitions are nebulous. Did Yugoslavia cease to be part of Eastern Europe when Tito took the country down the path of non-aligned status? What about Ceauşescu’s Romania or Hoxha’s Albania? And can we legitimately speak of Eastern Europe today in any meaningful sense without Estonia, Latvia, and Lithuania? Furthermore, as recent developments in Ukraine continue to reinforce, such debates are unlikely to be resolved any time soon.

Nevertheless, in academic research the boundaries of “Eastern Europe”—whatever exactly that might have been—were historically rather rigid. To give a personal example, when I was a graduate student in the mid- to late 1990s, I was fortunate to be surrounded by—either as classmates or as visiting pre-docs or post-docs—a huge cohort of colleagues working on what we have come to call post-communist politics. Yet even at that time, almost every dissertation seemed be focused either on research from exclusively East European countries (Benoit, Botcheva, Ganev, Desai, Grzymała-Busse, Wittenberg, Zielinski), exclusively focused on Russia (Albats, Andrews, Frye, Giuliano, Gorenburg, Herrerra, McFarquhar, Treisman), or exclusively focused on ex-Soviet Republics (Darden, Hale, Jones-Loung, McMann). ²

Nor was this sort of pattern simply limited to PhD dissertations. In a 2002 Annual Review of Political Science article, ³ I coded every article written about elections and voting in a post-communist country in sixteen journals—eight general-interest political science journals and eight post-communist area studies journals—published between 1990 and 2000. Of the 101 articles on elections in post-communist countries that I found across these sixteen journals, a grand total of 3 compared elections from the former Soviet Union (not including the Baltics) and Eastern Europe. ⁴

With these points in mind, here’s the argument I would like to put forward concerning the future of “East European” studies in the context of political science. Regardless of whether “Eastern Europe” ever was a legitimate ideal type for scholarly inquiry, the boundary between communist regimes that were and were not formally part of the Soviet Union was a real one indeed in the academic community. As time passes, it will become harder and harder to a priori justify this classification without reference to a specific theoretical argument for which the classification is appropriate. Put another way, there will be studies now where it is more appropriate to group/compare Poland with Spain, not Albania. This will present a challenge to those of us who enjoy studying Eastern Europe for the myriad of reasons appearing daily on the wonderful “Why We Study Eastern Europe” Facebook page ⁵ established by Michael Bernhard, as we will no longer simply be able to justify comparing certain countries and not others because they are part of “Eastern Europe.” However, at the same time, this also represents an opportunity to those of us who enjoy studying politics in this region because it fundamentally increases the opportunity for all sorts of new comparative work that can (1) teach us new things about these societies in which we are interested and (2) provide new audiences for our research, all the while (3) allowing our work to speak to new debates in new ways.

Since a full-fledged literature review is far beyond the scope of a four-thousand-word symposium contribution, I will instead follow the direction of the special issue’s organizers and primarily focus on the possibilities that I have discovered in this regard through the evolution of my own research agenda. As a result, this will focus my contribution largely on the topic of the comparative study of East European political behavior. It will also entail leaving out many, many studies that could have been used to make similar points equally well, although I include—primarily in the footnotes—at least some illustrative examples of other studies that have utilized similar comparative approaches. I hope readers will find the approach more a useful parsimonious organizing principle than self-indulgent narcissism, but I leave that for others to decide.

In my own work, I began by first breaking the Eastern Europe versus former Soviet Union boundary, although I did this gradually in a series of stages, beginning first by including Russia in a comparison with several East European countries, ⁶ then the Baltics in a comparison with more East European countries, ⁷ and then finally “pan-post-communist” studies including all post-communist countries that met certain thresholds for free and fair elections. ⁸ The next step was to write articles that explicitly compared behavior in a small number of East European countries with individual non-post-communist countries. ⁹ The final—and most recent—step has been to embark on a large-scale comparative project with Grigore Pop-Eleches that has involved comparisons of political attitudes and behavior across eighty to ninety countries including most East European countries. ¹⁰ What is particularly intriguing about this last project is that the purpose of this study is to investigate the effect of communist-era legacies on post-communist political behavior. So in an interesting twist, the most “comparative” work I’ve embarked on that encompasses East European countries is explicitly designed to shed light on East European politics.

In the remaining sections, I expand a bit on what I have done and learned in each of these types of comparative analyses involving East European countries in studies that would not have been possible had we kept the old East European boundaries around our academic research.

Eastern Europe and the Former Soviet Union

As noted previously, when I was in graduate school, most of my colleagues were pursuing doctoral projects that continued to adhere to the Eastern Europe versus Russia/former Soviet Union boundary. ¹¹ I instead ended up writing a dissertation that examined the effect of economic conditions on election results in Russia, Poland, Hungary, the Czech Republic, and Slovakia, which eventually became a Cambridge University Press book. ¹² I wish in retrospect that I had a better story for how I somehow became convinced that I had to help break down what was going to become an archaic boundary in the future, but the truth of the matter is that I do not remember giving the matter much thought at all. Before I started my prospectus, I had been working for two years as a research assistant for Joel Hellman on work related to what would become his 1998 World Politics article “Winners Take All: The Politics of Partial Reform in Postcommunist Transitions,” which was of course explicitly comparative across the more politically competitive East European countries as opposed to the less competitive former Soviet republics. ¹³ Moreover, as I would later document in my 2002 Annual Review of Political Science article, Russian elections were by far the most popular subject for academic research on elections in post-communist countries. ¹⁴ Next most popular, however, were Poland, Hungary, and the Czech Republic, to which I added Slovakia (no. 9) because the first Czech elections in my study were actually Czechoslovakian elections. These were the countries people were studying; these were the elections getting attention from the scholars I was reading when I was a graduate student. So I did what we did in those pre-all-the-data-are-available-on-the-Internet days, and headed out to those countries to scope out data availability. While there, I started conducting interviews, learning about the countries in detail, establishing contacts at the state statistical offices, building up my data collections, etc.

As it turned out, with a bit of digging I could get the kind of data I needed (congruence between electoral districts and administrative units by which economic and demographic data were reported) in all five of these countries, so at that point it seemed completely arbitrary to exclude either Russia or the four East European cases simply because most previous studies would not have crossed these boundaries. As far as I was concerned, all five countries had seen communism collapse, had parties associated with post-communist reform, and had communist successor parties (these “types” of parties would play an important role in one set of my theoretical arguments that focused on parties’ relationship to the transition away from communism), so I could not come up with any compelling reason to start dropping cases. ¹⁵ Interestingly, though, every time I gave a talk based on this research, someone in the audience would ask the Russia versus Eastern Europe question. As a result, I knew that I would need to address the question in the book (i.e., whether my results were systematically different in Russia versus the four East European cases). I had a framework for incorporating these types of hypotheses (i.e., ones that suggested there might be different support for my primary economic voting hypotheses in some cases than others) that I called “Conditional Economic Voting Hypotheses.” ¹⁶ The problem was still that I had no good theoretical rationale for why I thought economic voting patterns should be different in Russia as opposed to Poland or Hungary (the too many time zones hypothesis?), so I actually ended up calling this section “intuitive conditional hypotheses.” ¹⁷

The next step I took in bridging the old East European versus former Soviet Union boundary was to include the three ex-Soviet Baltic republics (Latvia, Lithuania, and Estonia) with an otherwise East European set of comparative cases. This was hardly exceptional, but I mention it for two reasons. The first is that I had a clear theoretical rationale for doing so based on the topic of my analysis: the determinants of attitudes towards EU membership among citizens of the ten post-communist countries that would go on to join the EU in 2004 and 2007. ¹⁸ Given that research agenda, there was no conceivable reason to exclude Latvia, Lithuania, and Estonia from the analysis, and indeed Estonia as an über-market-reforming country (especially vis-à-vis low tax rates) gave us the opportunity to test an extra hypothesis in the latter piece.

The second reason I mention these sets of comparative studies is because it does call into question yet again how actually we draw the boundary of Eastern Europe. Former Soviet versus non–former Soviet is the most parsimonious line we can draw, but then that leaves the Baltics out of Eastern Europe. We can amend this definition to say “pre–World War II Soviet republics,” but if we are being honest that seems pretty arbitrary; do we really think that the “effect” of being part of the USSR came from events in the 1920s and 1930s but not in the late 1940s, 1950s, 1960s, 1970s, and 1980s? Especially for those of us studying post-communist political behavior, most post-communist citizens of the Baltic states were not even born in the 1920s or 1930s (or were young children). This again points to the usefulness of having a theoretical rationale for our explicit or implicit set of comparative cases (e.g., ex-communist countries being considered for EU membership in the first waves of expansion) rather than just falling back on “Eastern European countries,” which could be used to exclude or include the Baltics arbitrarily. ¹⁹

Of course, the ultimate breakdown of the former Soviet Union–Eastern Europe boundary in academic research is simply to move to “post-communist” countries as one’s unit of analysis. This is the approach I have taken in two coauthored pieces on aggregate-level voting patterns, one focused on turnout ²⁰ and one on electoral volatility. ²¹ In both cases, we simply set a threshold for minimally competitive elections based on an external source—in both cases Freedom House rankings of “Free” or “Partially Free” at the time of the election—and then included all of the former Soviet republics and East European post-communist countries that met the definition. In the turnout piece, we split the countries into subcategories, but it was in relation to their likelihood of joining the EU, not whether they were from Eastern Europe or the former Soviet Union. ²²

The extent to which the “post-communist” framework has continued to enjoy increased legitimacy as time has passed can be also seen in the literature that emerged in response to the Colored Revolutions. These events, which took place in Serbia, Ukraine, Georgia, and Kyrgyzstan (plus possibly Slovakia and Croatia depending on with whom you are talking), clearly did span the old former Soviet Union and Eastern Europe border. Yet authors did not hesitate to address these cases in a comparative framework. ²³

Although again not the purpose of this colloquium, it is worth pausing just a moment to note that the “post-communist” category is not without its issues either. In particular, there were of course ruling communist parties in countries in the Caribbean and Latin America, Africa, and Asia. None of these, however, suffered the kind of rupture with communism that we saw in Eastern Europe and the former Soviet Union, although that’s not to say they couldn’t in the future (Cuba? North Korea?). In the case of others, it is hard to say exactly where in the “post” communist spectrum they are currently (read: China and Vietnam). Throw in the preeminence of the Soviet Union in dictating policy among most of the Warsaw Pact countries of Eastern Europe throughout the latter half of the twentieth century (although there was of course variation here, and especially so in Yugoslavia), and the line between the European/Eurasian post-communist countries and those in the rest of the world seems justifiable for the most part. ²⁴

Eastern Europe and Non–Post-Communist Comparisons

Perhaps even more exciting than breaking down the boundary between Eastern Europe and the former Soviet Union has been the opportunity to situate East European countries in comparative studies with countries outside of the ex-communist world. The excitement here is generated by the chance to show that the countries in which we are interested can shed light on theoretical topics of interest to the broader political science community, even for people with no inherent interest in Eastern Europe. In a day and age where funding and hiring opportunities are growing scarcer, ²⁵ being able to argue that an East Europeanist hire is someone who can engage with the rest of an academic department is something that should not be taken lightly.

I have used this approach—comparing post-communist countries with non post-communist countries—in four different parts of my research agenda, the last of which I will address separately in the next section. The first example is work I touched on briefly in the previous section related to electoral volatility. ²⁶ Here, we were interested in contrasting the determinants of electoral volatility in new democracies and old democracies. ²⁷ Eastern and Western Europe seemed a perfect set of cases that would allow us to hold a great deal of political and cultural history constant while varying the length of democratic experience. ²⁸

A second example is part of my work with Ted Brader on the determinants of variation in the strength of partisanship. In a 2012 Comparative Politics article, ²⁹ we wanted to explicitly examine the effect of age of democracy and stability of party systems on the strength of partisan cueing effects (i.e., how much more likely an individual is to support a policy proposal if she is told “her party” proposed it). We developed a series of experiments to test these cueing effects in multiparty systems, ³⁰ and then wanted to run them in an older democracy with a stable party system, a new democracy with a relatively stable party system, and a new democracy with an unstable party system. We were doing this in the mid-2000s, so Great Britain, Hungary, and Poland, respectively, fit our needs perfectly. This is a good example of study that could not have been conducted only in Western Europe or Eastern Europe, but was possible through comparative analysis. ³¹

In a follow up study, ³² we wanted to see whether partisan cues varied based on party-level characteristics (e.g., whether the party was in the opposition or not and the clarity of the party’s ideological position). We again used our experiments from Great Britain, Hungary, and Poland, but here the fact that we had countries from the East and West played a different role. We of course could have run such an analysis using only East European or West European countries, but by including cases from both categories we provided additional evidence of the robustness of our results (i.e., the finding was not peculiar to either countries with new party systems or old party systems) and, ideally, would be able to appeal to a larger audience of people interested in both West and East European politics.

Combining an East European country with an established democracy also helped us in a more methodologically oriented article designed to introduce a new way to measure sociodemographic “cross-pressures” faced by individuals. ³³ The method was primarily developed using data from the United States, which as a two-party system is fairly atypical internationally, and especially so vis-à-vis Europe. So we wanted to demonstrate that our method would also work in a party system with many parties. What better data to use than the 2001 Polish National Election Study, which gave us a viable six-party system to study? In a context where East European cases can be seamlessly included in comparative studies with countries outside of the region, this approach allowed us to simultaneously demonstrate the robustness of our methodological approach and show readers who might never have been interested in East European politics that data from Poland could prove useful in their own research agenda.

A third area in which I have recently applied an East versus West European comparative research design is in the study of the effect of corruption on voting behavior. In a 2013 Electoral Studies article with Marko Klašnja, ³⁴ our motivation was to examine how the effect of corruption on voting behavior varied across a “high corruption” and “low corruption” country. Again, this seemed to be a perfect context in which to design a study using countries from Western Europe and Eastern Europe. We ended up running our experiment in Sweden—tied for 3rd on Transparency International’s most recent Corruption Perceptions Index of most clean countries—and Moldova, which was tied for 102nd on the same list. ³⁵ The results were illuminating. In Sweden, our “low corruption” country, voters punished a hypothetical mayor running for reelection for corruption regardless of the state of the economy. In Moldova, however, voters punished the hypothetical mayor for corruption only when the economy was also bad. When economic conditions had improved, however, Moldovan voters appeared less concerned about corruption. While it is of course possible we could have drawn the same conclusions from a high and low corruption country within Eastern Europe, the comparison of Western and Eastern Europe allowed us to get a much wider range of variation in corruption: the “most clean” Eastern European country in 2013 was Poland at 40th. ³⁶

Large-Scale Comparative Analysis

Just as the ultimate extension of the break down of the Eastern Europe versus former Soviet Union boundary is comparative studies including all of the post-communist countries, the ultimate extension of the break down of the boundary between studying post-communist countries and countries elsewhere in the world are large-scale comparative analyses without any geographic boundaries. This is exactly the empirical strategy that Grigore Pop-Eleches and I are employing in our current project on the effect of communist legacies on post-communist political behavior and attitudes. ³⁷

Here’s the basic logic. We want to understand the extent to which the legacy of communism has affected political behavior and attitudes in post-communist countries. The starting ground for such a study, therefore, needs to be examining the difference between attitudes and behavior in post-communist countries and in other countries. If we cannot document such a difference, then there is nothing for a communist legacy to “explain.” So we have been relying on large cross-national survey projects such as the World Values Survey, which allows us to look at attitudes and behavior in more than ninety countries, including most of the post-communist countries. We can then systematically document how attitudes or behavior differ in post-communist countries. In turn, we can then attempt to measure the extent to which communist legacies can account for these differences using new methodological approaches we have developed.

In the context of this article, this approach is important for two reasons. First, it illustrates another “opportunity” for using the countries we study and know well in comparative analysis. But perhaps much more importantly, it demonstrates how embedding data from East European countries in large-N comparative analyses does not have to involve abandoning post-communism–specific research questions to the study of topics developed in other corners of the political science universe, thus somehow implying that our questions are less important than “generalist” questions. Instead, it shows that we can use large-scale comparative analysis to answer questions that are crucially important to understanding the politics of Eastern Europe. So it is an opportunity to both simultaneously deepen our understanding of our region of interest and do so in a way that engages with research from the rest of the world. ³⁸

Conclusions

I want to close this essay by noting how, in a sense, the “comparative opportunities” that I have identified for scholars of East European politics have come full circle to where we started. Originally, we studied Eastern Europe because we want to learn about Eastern Europe. Other countries, even including Russia and the other republics that made up the former Soviet Union, were seen as being too different to offer real analytical leverage for understanding the politics of Eastern Europe.

Today, it is much harder to justify why a study of politics should be limited to Eastern Europe (or should only include cases drawn from the set of Eastern European cases) than it was pre-1989. Studies of new members of the EU can be justified. Studies of new democracies can be justified. Studies of competitive authoritarian regimes can be justified. ³⁹ Studies of countries with post-communist economic reforms can be justified. Studies of countries with colored revolutions or in which formally federalist regimes have split into their constituent parts can be justified. The list goes on and on, but none of these perfectly overlap with what we have commonly referred to as Eastern Europe. Some of these categories exclude some East European countries; others include non-East European post-communist countries; and others include countries we would not even call post-communist. So this is not to say we can no longer conduct research on only East European countries, but merely that we need to be sure to find theoretically justified reason for including (or selecting from) the particular countries we study that goes beyond simply the fact that they are East European.

At the same time, the challenge of justifying why we study Eastern Europe has also yielded the opportunity to include East European countries in comparative studies with countries outside of Eastern Europe. These types of studies can give us the opportunity to bring more data to bear on our hypotheses, to provide additional robustness test of our findings, to show the broader relevance of our questions and finding, and to engage the discipline more broadly in general research questions such as the nature of partisanship or voting. But—and here’s where the full circle comes in—it can also give us the tools to answer new questions about politics and political behavior in Eastern Europe itself. For scholars who remain interested in Eastern Europe for Eastern Europe’s sake, it would be foolish to avoid either the challenges posed or the opportunities presented by the developments of the past 25 years.