Contesting war memories: Parties and voters in contemporary Southeast Europe

Abstract

Over the past four decades, there has been a proliferation of interest in the causes, consequences, and dynamics of contestation over collective memories across a variety of fields. Unfortunately, collective memories—particularly those of traumatic experiences of violence such as wars and revolutions—have been largely absent from party politics research. Using data collected in an expert survey on the policy positions and ideological orientations of all relevant political parties, as well as an extensive survey of more than ten thousand voters in the six post-conflict countries of Southeast Europe, we demonstrate that collective memories of war are not only subjects of historiographical contestation but are also significant sources of ideological and policy differentiation among political parties, as well as one of the strongest determinants of voter choice. Our analysis shows that collective memories are politically contested and that party politics research would benefit from taking them seriously.

Keywords

collective memory war memory postwar politics voter choice Southeast Europe

“Who controls the past controls the future. Who controls the present controls the past.”

George Orwell, 1984

The study of collective memories has developed into a rich field crossing the boundaries between sociology, political science, and history over the past several decades. It has remained, however, largely detached from the study of party politics (Malka et al., 2019). This is unfortunate. Keeping collective memories on the sidelines of electoral politics research, and treating them as purely epiphenomenal, risks not capturing the essence of political competition in many polities where debates over recent and more distant pasts often dominate public discourse (Olick, 2016). Collective memories are in many ways societies’ connective tissues (Hite, 2012). They strike at the core of people’s identities (Alexander, 2004; Gillis, 1994). This is particularly the case when it comes to national identities and collective memories of violent and traumatic events such as revolutions or wars. “War made the state, and the state made war,” Charles Tilly famously remarked (1975: 42). Following Said’s (2010) argumentation, one could transform this observation into “memory of war created the nation, and the nation created the memory of war.”

This is not to say that national collective memories are uniform or homogenous—quite the contrary. All collective traumatic events are followed by contestations, challenges, and struggles over their remembrance and commemoration (Edkins, 2003). This memory contestation takes many shapes and forms and comes from a variety of directions—from center and periphery, from within and without, and from above and below (Olick and Robbins, 1998). Nevertheless, it is ultimately political and is fought over in the electoral arena. Its public prominence may move with the ebb and flow of social and political priorities, but its staying power is often long lasting and can span generations, as has been the case with the “politics of history” and the “historical consciousness problem” in post-World War II Germany and Japan (Hashimoto, 2015; Wolfrum, 1999), or with the debates over the historical legacies of the Civil War that continuously seem to wash over American politics. What is important to note is that these contestations over memories of traumatic conflicts—although to some extent limited by the historical events themselves (Kubik and Bernhard, 2014)—are shaped by contemporary considerations about what is yet to come. As we suggested by this article’s epigraph, collective memories of the past are fought over in the present, in order to shape the society’s future.

In this article, we examine whether political parties in the post-conflict polities of Southeast Europe differentiate themselves based on value stances toward the memories of the ended conflicts, and whether voters make their electoral choices based on these different value stances of the parties. We wish to expose the extent to which electoral competition in societies marred by recent violent conflict is founded upon considerations of how that conflict should be remembered. To this end, we analyze data collected in an expert survey on the policy positions and ideological orientations of all relevant political parties, as well as an extensive survey of more than ten thousand voters, in Bosnia and Herzegovina, Croatia, Kosovo, Montenegro, North Macedonia, and Serbia. These countries are good objects of study for two reasons. First, they have functioning, if flawed, democratic political marketplaces of ideas, policies, and parties, and their elections over the past two decades have been largely free and fair. And second, questions of collective war memories have maintained their public prominence to this day despite temporal distance from the regional wars of the 1990s.

Those regional wars of the 1990s were representative of the larger number of conflicts that erupted as the systemic constraints of the Cold War collapsed. All six countries under study were part of the Yugoslav federation, whose protracted break up resulted in a string of violent conflicts in: Slovenia (1991), Croatia (1991–1995), Bosnia and Herzegovina (1992–1995), Kosovo (1998–1999), and North Macedonia (2001). These conflicts took the lives of more than 130,000 people and led to the forcible—and in many cases permanent—displacement of more than four million people. The causes of the Yugoslav wars were complex, but in essence, they were nationalist wars fought for control of territory. Their logic was perhaps best summed up by the economist Gligorov (2011): “Why should I be a minority in your state when you can be a minority in mine?” In addition to devastating human loss and extensive damage to physical infrastructure, the conflicts also poisoned inter-ethnic and inter-state relations in the region with slow recovery partly due to the flawed process of transitional justice. Despite many efforts at reconciliation, memories of the 1990s conflicts continue to be a nearly constant presence and burden in regional politics to this day, with political entrepreneurs using the recent past to gain electoral advantage.

Our analysis convincingly shows that political parties in contemporary Southeast Europe indeed take different positions related to war memory and these positions are one of the principal determining factors of their overall ideological orientation, independent of factors that led to the regional wars in the 1990s, such as nationalism or attitudes toward the rights of ethnic minorities. What is more important, our results show that views on war memory are also of critical importance in voters’ decision-making on which party to support. Voters to a significant extent base their electoral choices on the level of their agreement with the political parties’ stance toward how recent wars should be remembered, even when we control for a number of other policy positions, as well as voter and party characteristics, including voters’ own war experiences. We argue this is clear evidence that political competition in post-conflict Southeast Europe—even more than two decades after the wars that plagued this region—is defined by voters and parties’ divisions on war memory. Considering everything we know about postwar societies in a variety of geographic and temporal contexts, we believe the results of our analysis can help us better understand the impact of war on postwar political competition far beyond the borders of contemporary Southeast Europe. The Yugoslav wars of the 1990s may have belonged to a particular class of post-Cold War conflicts, but the trauma suffered by the local populations, as well as the politicization of recent war memory, are virtually universal to all societies that experienced similar events. Simply put, collective memories of violent conflicts are politically contested, and party politics research would benefit from taking them seriously.

Collective war memories and political contestation

The study of collective memories has been a “nonparadigmatic, transdisciplinary, centerless enterprise” without agreement among scholars even on some of the field’s basic terminology such as “collective memory” (Olick and Robbins, 1998: 106). If we trace the field to what is generally considered its foundation in the scholarship of Maurice Halbwachs, we can find the roots of the concept of collective memory in the Durkheimian idea of “collective representations”—symbols or meanings that constitute the character of a social group (Olick et al., 2011). Though obviously closely related, collective memories should be understood as distinctive from individual memories, history, and historiography (Olick, 2016). Instead, they should be thought of as shared corpora of remembrances and understandings of historical information by social groups that strike at the core of the groups’ meaning and identity. What is critical to understand here is that memory scholarship, in spite of its Durkheimian roots that are often critiqued for essentializing and reifying social groups, has always considered collective memories as politically and socially contested and constructed (Gillis, 1994; Olick et al., 2011).

This contestation is obviously constrained to a certain extent by historical events and the repertoire of interpretations that could be deemed as culturally credible or potentially valid (Kubik and Bernhard, 2014). Nevertheless, the selectiveness of collective memory according to present-time conditions of remembering the past has been recognized by memory scholars from the very beginnings of the field. As much as they are connected to the historical record, collective memories are not permanent but are instead representations of that historical record, which are subjectively (and often consciously) constructed to fit the needs of the present (Hashimoto, 2015). This is particularly the case when it comes to collective memories of traumatic social experiences of violent conflict. Collapses of colonial empires and dictatorships, ends of civil wars and international conflicts, and revolutions—all of these events place collective memories at the heart of public discourse and political debate (Zubrzycki and Woźny, 2020). They do so not only because political identities are often defined by violent conflict but also because different forms of remembering (and forgetting) can serve as the means of social recovery from trauma (Edkins, 2003).

In spite of their undeniable importance and the proliferation of interest in their causes, consequences, and dynamics of contestation/construction across a variety of fields over the past four decades, collective memories have been absent from party politics research, save for a recent contribution looking at discursive expressions of different memory modes associated with the Israeli Mapai party (Malka et al., 2019). Researchers have provided us with tremendously insightful case studies of collective war memory social contestation in a string of important geographic and historical contexts, such as Japan (Hashimoto, 2015), Poland (Hackmann, 2018), or Germany (Olick, 2016; Wolfrum, 1999). They have also improved our understanding of concepts of increasing social relevance such as transitional justice (Rigney, 2012). The study of the politics of memory, however, has been marked by scarcity of systematic theorizing based on larger numbers of cases and comparative approaches.

One notable exception to this trend has been the proposition of a theory of the politics of memory by Kubik and Bernhard (2014). Together with the contributors to their edited volume, they trace the politics of memory and commemoration of communism in post-communist Eastern Europe and find the nature of the “memory regimes” (fractured, pillarized, or unified) in the countries of the region to be dependent on the interactions among the different “mnemonic actors” (warriors, pluralists, abnegators, and prospectives). We find great value in their approach, but the question we are primarily concerned with is different. Our goal is to distill the potential political contestation over collective memory of a violent conflict to its foundational essence. We believe that essence to be centered on the question of victimhood and the acceptance by different groups of the suffering of others.

Indeed, memory scholarship has long recognized that the denial of the reality of suffering of others is at the core not only of rejection of responsibility but also of perpetuation of conflict (Alexander, 2004). Very often, individuals and communities perceive memories of victimhood as zero-sum games. As Assmann (2007: 16) has argued, “one remembers something in order to be better able to forget something else. When applied to the realm of national memory, this means that one recalls one’s own suffering in order to avoid being reminded of one’s own guilt.” Or, to turn things around, one rejects recalling others’ suffering in order to be reminded of one’s own. Making that leap of acceptance of others’ suffering, however, is key to finding and building lasting peace. As Dixon (2018) has shown in her study of narratives of guilt in Turkey and Japan, acknowledging harmful acts is the first step on the road toward finding closure of dark chapters in nations’ histories. In fact, as Lind (2008) has demonstrated in her analysis of the politics of states’ apologies, the mere avoidance of denial of past violence is a critical facilitator of reconciliation—even without the additional steps of apology, compensation, or commemoration of the other side’s victims. Acknowledging harmful acts committed against others, therefore, is not just a symbolic show of solidarity. It also has real impact on the diffusion of conflict, prospects for reconciliation, and all the political, social, and economic consequences that a lasting peace could bring. We do not deny that war memories are complex social concepts that encompass a number of dimensions. At this early stage of scholarship on collective war memories and party politics, however, we believe a focus on what we see as the essential question of acknowledgment of guilt and victimhood is important in order to gain the strongest possible empirical footing, with refinements and greater theoretical richness to come with further research.

Ultimately, broad social acknowledgment of harmful acts committed against others is usually determined by a string of political decisions made by those who have control over the government. As all political decisions, it is subject to contestation that political parties can choose to engage in. A long strand of scholarship on party politics has recognized the role of parties in fostering issue salience and even cleavage formation (Lijphart, 1999; Sartori, 1969). In this line of argumentation, parties are seen as political entrepreneurs who articulate a certain agenda in public space in search of support (Inglehart, 2015; Lijphart, 1999; Sartori, 1976). The salience of politicized issues is thus a function of the competition among parties, who highlight certain social divisions to secure an electoral advantage (Hlousek and Kopecek, 2016). Considering everything we know about political competition in postwar societies—particularly those in Southeast Europe of the past two decades—this argumentation is in our view very plausible when it comes to contestation over collective war memories. Indeed, (some) parties in postwar societies actively keep the memory of war alive by partaking in commemorations of war-related events and engaging in public discourse and activities loaded with meaning related to the war past so much so that it has become an integral part of the political culture even decades after the wars end (Ashplant et al., 2015; Chirot et al., 2014; Lazic, 2013). The politics in Southeast Europe is replete with such examples that continue to mar the local political landscape to this day: from parties in Croatia sparring over the role of war veterans in society or the commemorative practices in the town of Vukovar that was the sight of the biggest battle during the war, to politicians in Bosnia and Herzegovina, Serbia, and Montenegro continuously debating the wars’ causes or the genocidal nature of the massacres at Srebrenica.

To be clear, the events of the wars in the 1990s in Southeast Europe, with the destruction of property, the often permanent expulsion of people from their homes, the decades-long struggles to find out what happened to lost loved-ones and to bring perpetrators to justice have the potential to shape political decisions of voters regardless of what political parties do. We, therefore, do not argue that the memory of war would have been politically impotent if it were not for the actions of political elites nor do we argue that parties simply respond to the aspirations of the voting public. Rather, we believe political parties amplify war-related sentiments that are at their core very much real. In this way, a number of scholars have argued that “keeping the war alive” has been the strategic choice of many postwar political parties in Southeast Europe with the goal of obtaining an electoral advantage (Bieber, 2019; Gödl, 2007), just like ethnicity and national identity have been politicized and constructed by political players across a variety of polities (Linz and Stepan, 1996). We do not disagree.

The causes of the Yugoslav conflicts were obviously rooted in pre-existing cleavages centered around nationalism and rights of different ethnic groups in the states created on the ruins of the dissolved federation. Moreover, these conflicts and the very dissolution of the Yugoslav federation were precipitated by politicians who (ab)used the memories of past inter-ethnic strife. Nevertheless, we argue that the impact of collective war memories on the contemporary politics of the region cannot be reduced to these pre-existing lines of division. Instead, the wars and their memory have added something unique to the nature of political competition. We firmly believe that the wars of the 1990s fundamentally altered local social structures and created populations with real and objective differences rooted in their vastly different experiences of the war. We also, however, firmly believe that political parties in the region have actively worked to politicize the recent war past in an effort to mobilize supporters. The essence of that politicization has been the question of collective memory of the conflict’s perpetrators and victims, that is, to which extent the suffering of the members of the out-group is deserving of acknowledgment and commemoration. Our first task is to demonstrate that parties in contemporary Southeast Europe indeed have different standpoints when it comes to collective memory of war victimhood, as there can be no political contestation without parties taking different positions.

Our proposition, however, goes further than this and suggests that war memories strike at the core of the central left–right political divide in the party politics of postwar societies. The left–right continuum occupies a central place in politics, both for elites and the public; this has been the case for decades (Downs, 1957). The labels “left” and “right” serve as instruments with which people are able to orient themselves in a complex political world. In other words, the left–right spectrum is a taxonomic system that allows people to understand, store, and order political information. The content of the left–right continuum, therefore, can reveal what issues are most important in a given society in a given era (Inglehart and Klingemann, 1976: 244–245). Classic accounts argue that the left–right divide was related to the conflicts between workers and the owners of the means of production (Kriesi, 2010). However, left–right semantics are dynamic and can change meaning following societal transformations and pivotal events. One example is the emergence of materialist/post-materialist politics in Western societies following WWII (Inglehart, 1984). We argue here that war and its memory can have a similar transformative impact on what a society considers left and right. Obviously, contestation on war-related issues in postwar societies can be completely unrelated to other ideological differences, as was the case in Ireland, where the two main competing parties, Fianna Fáil and Fine Gael, represented the two sides of the Irish Civil War but differed little in terms of other issues (Gallagher, 1985). We believe this scenario, however, to be an exception.

There are four major theories on how societal changes impact the left–right continuum (Kitschelt and Hellemans, 1990; Knutsen, 1995). The first argues that old meanings of left and right will be discarded in favor of new ones. The second that the labels left and right become outmoded by an entirely new political language. The third predicts that new and old meanings will be merged into one ideological dimension. The fourth and final argues that new meanings to left and right will be added to the old ones, without replacing them or making them fade away. This last perspective has received most empirical support and has guided our own expectations. Considering the nature of wars and collective war memories in former Yugoslavia, we propose that positions on collective war memories help define ideologically what it means to be a leftwing or rightwing party in the region today. The distinction between the nationalist right and the more liberal center in Serbia, for example, could not be clearer than when it comes to their attitudes toward Serbia’s role in the wars of the 1990s. Similarly, the distinction between the nationalist right and the social-democratic left in Croatia could not be clearer than when it comes to their attitudes toward war veterans (Dolenec, 2017).

While conceptually clear, our ability to discern the independent impact of one issue dimension on the left–right scale as a whole is hampered by the fact that parties engage in “ideology-making,” in which they group issues together because of a superordinate value or because they deem issues to belong together (Converse, 1964). A whole body of literature has demonstrated high levels of positional constraint among elites (Burt and Neiman, 1987; Jennings, 1992; Lesschaeve, 2017). In Southeast Europe, this is likely to be the case for war memories and nationalism, ethnic minority rights, and the gal/tan dimension in general. Nevertheless, the degree to which war memories have an independent effect on parties’ left–right positions would suggest that the war and the views on it have shaped the nature of political competition in the region. Our empirical propositions reflect this view. We therefore first suggest that:

Parties take different positions on war memory. These positions are closely related to their general worldview and ideological orientation.

Whether or not parties take different positions on an issue is ultimately, however, irrelevant if the voters find that issue not salient enough in their decision-making on which party to support. Harking back to the information processing models of Zaller (1992) and Iyengar and Kinder (1987), individuals retain a large number of possible frames or considerations relevant to political choices and decisions in their memory. When asked to evaluate an object, voters sample from these considerations. For instance, voters can evaluate parties on the basis of economic performance or the fight against corruption. However, not every consideration is equally accessible, and it is these differences in accessibility that determine the manner in which voters judge political parties (Scheufele and Tewksbury, 2006). Considerations that are highly accessible are more likely to shape voters’ assessments (Belanger and Meguid, 2008; Green and Hobolt, 2008). Voters are, therefore, expected to prioritize opinion congruence, that is, the sharing of policy views with parties, on dimensions that are more accessible to them. In turn, accessibility is determined by the importance or salience of a policy dimension (Soroka, 2006). This has been demonstrated numerous times in non-war contexts (Lefkofridi et al., 2014; Walgrave and Lefevere, 2013) but arguably extends to Southeast Europe as well.

The central claim of this study is that the memories of the ended conflicts are potent and relevant in the political competition of the region; through the inherent importance, the war past still enjoys decades after the conflicts’ end and the deliberate efforts by parties to amplify this importance. Attitudes toward and views of the (recent) war past, thus, strike at the core of people’s value systems giving rise to a new ideological conflict or cleavage which is rooted in cultural divisions that have transcended objective structural and experiential causes (see Kriesi, 2010; Torcal and Mainwaring, 2003). As such, voters’ preferences for political parties likely depend on the extent to which their personal views of the memory of war victimhood correspond to the parties’ views of the memory of war victimhood if those views are important to voters. If this is indeed the case, which we believe it is, then it would follow that voters’ support for political parties is dependent on the congruence between theirs and parties’ views of the memory of war victimhood—regardless of their personal experiences of war violence. More specifically, voters with less conciliatory views of war victimhood should support political parties with similarly hardline positions, and voters with more conciliatory views of war victimhood should support parties with similarly broadminded positions. This is why we hypothesize that:

Voters’ level of support for political parties is dependent on the level of their agreement with parties’ positions on war memory.

Data and method

To test our propositions, we rely on an expert survey on the ideological character, policy positions, and behavior of political parties and a voter survey with more than ten thousand respondents, both of which we conducted concurrently between October and December 2018 in the six post-conflict countries of Southeast Europe. For the former, we contacted political scientists in relevant fields from all major universities in Southeast Europe, as well as scholars working on the region from abroad, and invited them to participate in an online survey. In total, 73 experts participated. The expert survey follows the standard set by the Chapel Hill Expert Survey (CHES) requiring a minimum of four experts before a country can be included in the dataset (Polk et al., 2017), with our average number of respondents per country being 11. The party list (presented in the appendix) includes all major parties in each country, as well as a number of smaller parties that are nevertheless considered relevant to the country’s party politics. While expert surveys are certainly not without their problems (Budge, 2000), they have been found to produce similarly valid estimates of party positions on issue dimensions, when compared to the manual coding of party statements and elite surveys, regardless of parties’ general left–right positions (Hooghe et al., 2010; Marks et al., 2007; Ruedin and Morales, 2019; Steenbergen and Marks, 2007). In other words, expert placements of parties are made in their own right and are not simple deductions based on other political reputations.

In addition to party positions on war memory and the parties’ general left–right orientation, our expert survey captures information on five other subject areas: gal/tan positions, nationalism, economic policy, military policy, and ethnic minorities. Wherever possible, we employed the same formulations as the ones used by CHES and a comparison between the items that overlap with the CHES indicates a high degree of consistency.¹ With an average standard deviation of 1.9 per party and question, the precision of our measures is comparable to other expert surveys. We further complemented expert survey data with information about the parties’ parliamentary and government representation, as well as their age. When it comes to parties’ positions toward war memory, we decided to capture it with the question: “How do you estimate the position of each party toward acknowledging harmful acts committed against other groups during (the conflict relevant for the county in question)?” As argued above, we believe this question strikes at the core of the way war memory is understood and politicized, not only in the post-conflict societies of Southeast Europe. Expert survey data allows us to determine to which extent parties’ left–right positioning is related to their war memory stance, as opposed to their stances in other policy areas, and to provide an answer to our first hypothesis. Table 1 gives an overview of the descriptives for the left–right orientation, war memory, and the five issue/policy areas for the parties in our sample.

Table 1.

Descriptives of parties’ issue/policy positions.

	Mean	SD	Min	Max
Left–right position	5.72	1.59	2.33	9.07
War memory	4.68	2.15	0.86	8.83
Economic policy	4.90	1.06	1.67	8.33
Libertarian–authoritarian values	5.86	1.84	2.15	8.92
Nationalism	5.67	2.34	0.75	9.78
Minority rights	5.45	2.22	0.57	9.29
Military policy	5.50	1.93	1.38	9.53

The total size of the sample in our voter survey that was conducted on a dedicated online platform and mobile app, on the other hand, was 10,736, with 2718 respondents coming from Bosnia and Herzegovina, 2324 from Croatia, 1311 from Kosovo, 722 from Montenegro, 1460 from North Macedonia, and 2201 from Serbia.² To ensure representativeness of our sample in terms of gender, age, education, and region, we apply survey weights in all our analyses (Ansolabehere and Rivers, 2013) based on the raking method (Kalton and Flores-Cervantes, 2003). The voter survey captured respondents’ views on memory of their country’s most recent armed conflict with a Likert scale question identical in substance to the one asked in the expert survey. In addition to the respondents’ socio-demographic characteristics (gender, age, education, income, ethnicity, and religiosity) and attitudes toward a number of social issues and policy areas, the survey also captured respondents’ political preferences and commitments through a string of propensity to vote questions. For each party relevant to the respondents’ polity, respondents were asked on a 0–10 scale how likely it was that they would vote for that party in the future. In our research design aimed at answering our second hypothesis, we follow the approach proposed by Van der Eijk and Franklin (1996) and make these propensity to vote (ptv) scores into the dependent variable. With ptv scores we create a stacked data matrix in which the unit of analysis is the respondent*party combination. In other words, there are multiple observations per respondent, one for each party included in the survey. In contrast to discrete choice models, which only measure actual party choice, this approach allows for a far more detailed view of how individual and party characteristics make a party more or less electorally attractive. In addition, it allows researchers to test hypotheses of party choice determinants across a large number of parties and countries instead of estimating a model for each individual party.

When examining the role of standpoints on issues or values in party choice, the standard approach is to use a measure of distance or congruence between the positions of respondents and those of political parties (De Vries et al., 2011; Lefkofridi et al., 2014). Congruence is measured as the absolute difference between the mean respondent score (rescaled to match the expert survey format) subtracted from 10 in order to obtain scores ranging from 0 (no agreement) to 10 (complete agreement). The principal congruence value of our interest is related, obviously, to the question on the acknowledgment of harmful acts committed against other groups as it helps us address our second hypothesis, that is, whether voters’ level of support for political parties is dependent on the level of their agreement with parties’ positions on war memory.

As a means to test the robustness of the relation between views on war memory and party preference, we account for the role played by other policy dimensions. Our voter survey includes multiple questions related to social issues and policy areas that we believe present the strongest possible test for the relationship between war memory and voter choice: nationalism, ethnic minority rights, military policy, libertarian–authoritarian values, and economic policy. For the first three dimensions, measures of congruence were calculated similarly to that for war memory. For economic policy and libertarian–authoritarian values, however, Cronbach’s alphas proved to be too low. They were nevertheless included in the analyses via another method, of which more below. Nationalism, ethnic minority rights, and to some extent military policy should be highlighted because they arguably capture the essence of the conflicts of the 1990s and are thus the strongest test of the independent effect of war memory on party preference. The precise formulation of all the statements in both the expert and voter survey can be found in the Appendix, and Table 2 gives an overview of the descriptives of the ptv scores and the congruence variables.³

Table 2.

Descriptives of the Propensity to vote score and congruence variables.

	Mean	SD	Min	Max
ptv score	1.88	2.83	0	10
War memory congruence	7.77	1.57	0.88	10
Nationalism congruence	7.69	1.52	0.56	10
Military policy congruence	7.64	1.79	0.47	10
Minority rights congruence	7.65	1.69	0.57	10

To control for the role played by economic and libertarian–authoritarian attitudes, as well as respondents’ social characteristics, linear regressions are estimated for each party separately before the data is stacked (Walczak et al., 2012). For instance, in the case of social class, the ptv scores of one party are regressed on a set of social class indicators. This process is visualized in Figure 1 (also see Van der Eijk et al., 2006). Based on these regressions, predicted ptv scores, or $\hat{Y}$ s , are calculated and used as independent variables. The logic behind this method is that if social class indicators are good predictors of voters’ party preferences, regressions will produce predicted values that come close to the actual scores. These control variables, therefore, are based on the predicted values of models that contain one or more explanatory factors.

Figure 1.

Creating the stacked dataset.

Here, we should particularly highlight that our analysis controls for respondents’ personal war experiences, captured as a composite variable operationalized as the $\hat{Y}$ s obtained in a regression where propensities to support parties are predicted by six indicators of voters’ personal war experiences: (1) being a war veteran, (2) having been in physical danger during the war, (3) having been forced to emigrate because of the war, (4) having close family or friends be killed during the war, (5) suffering from a war-induced physical disability, and (6) suffering from war-related trauma. While the first five variables are dummies, war trauma is the sum score of respondents’ answers to six yes–no questions measuring symptoms of war-related trauma (see Supplemental Appendix for the exact wording of the questions). Cronbach’s alpha exceeding 0.80 shows that the scale consistently measures war trauma. Our analysis also accounts for voters’ ethnicity, religiosity (based on membership of a religious or church association), education (based on dummies indicating the highest obtained degree), social class (based on income deciles and employment status), gender, age, political interest, and being a non-voter (based on whether the respondent would vote if elections were to take place today). The analysis does not control for the rural/urban divide. However, given that it does control for a host of ideological and structural factors (most notably the regional distribution within each country) that in totality closely mirror the differences between urban and rural populations in Southeast Europe, we are convinced that the inclusion of a covariate for this divide would not substantially affect the results. On the party level, we account for party age and government membership. The stacking procedure obviously inflates the number of observations (n = 115,337) and renders the data by definition nested in voters (n = 10,736) and political parties (n = 69), which raises concerns over the independence of errors (Rabe-Hesketh and Skrondal, 2008). As a result, we use multilevel linear regression with random intercepts for the respondent and party level as crossed upper levels. In addition, to account for country-level differences in ptv scores, all our models include country dummies. As such, the full model used to estimate the preference of respondent i for party k presented in the next section can be written as follows

\begin{array}{l} Y_{i k} = β_{0} \pm β_{War memory congruenc e_{i k}} \pm β_{Nationalism congruenc e_{i k}} \pm β_{Military policy congruenc e_{i k}} \pm β_{Minority rights congruenc e_{i k}} \\ \pm β_{L i b e r a t i o n - a u t h o r i t a r i a n v a l u e_{i k}} \pm β_{E c o n o m i c p o l i c y r e f e r e n c e s_{i k}} \pm β_{P e r s o n a l w a r e x p e r i e n c e_{i k}} \pm β_{Ethnicit y_{i k}} \pm β_{R e l e g i o s i t y_{i k}} \\ \pm β_{E d u c a t i o n l e v e l_{i k}} \pm β_{S o c i a l c l a s s_{i k}} \pm β_{G e n d e r_{i k}} \pm β_{a g e_{i k}} \pm β_{P o l i t i c a l i n t e r e s t_{i k}} \pm β_{N o n - v o t e r_{i k}} \pm β_{G o v e r n m e n t c o a l i t i o n_{i k}} \\ \pm β_{Y e a r o f f o u n d i n g_{i k}} \pm β_{H R d u m m y_{i k}} \pm β_{X K d u m m y_{i k}} \pm β_{M K d u m m y_{i k}} \pm β_{ME d u m m y_{i k}} \pm β_{RS d u m m y_{i k}} \pm ζ_{i} \pm ζ_{k} \pm ε_{i k} \end{array}

Results

As stated above, the first task in establishing whether parties’ stance toward war memory could be structuring the content of political competition is demonstrating that there actually exists a sufficient difference among political parties when it comes to their views of war memory. Figure 2 shows the violin plots of party positions on this issue in all six countries of the region (the scores for parties in Bosnia and Herzegovina being split into those for the Federation and the Republika Srpska), with party positions weighted based on their seats in parliament. In other words, the plots show the likelihood of encountering the party positions in the legislature. Only in Republika Srpska does the party landscape exhibit little to no diversity. Parties in the Serb entity of Bosnia and Herzegovina seem to be united in their unwillingness to acknowledge war crimes—in our view—unsurprising considering how Republika Srpska was created. In all other countries of the region—and in particular in Croatia, Montenegro, and Serbia—there are parties representing a wide range of positions when it comes to war memory, with the depicted level of support varying depending on their contemporary electoral fortunes.

Figure 2.

Distribution of party positions on war memory.

Our first hypothesis, obviously, went beyond merely stating that parties in the post-conflict countries of Southeast Europe had differing positions toward war memory. It proposed that war memory was related to the parties’ overall worldview and that it struck at the core of their positioning on the ideological left–right continuum. Figure 3 shows the bivariate relation between parties’ left–right position (x-axis) and their views on war memory (y-axis). War memory is indeed strongly related to parties’ overall left–right position. Parties opposed to acknowledging war crimes are consistently found on the right side of the spectrum. This pattern is confirmed even when we split up the graphs by country in the Supplemental Appendix, with the only exception possibly being the parties in North Macedonia.

Figure 3.

Parties’ left–right position and stance toward war memory.

The relation between parties’ left–right orientation and stance toward war memory is more formally tested in Table 3. Here, parties’ general left–right position is the dependent variable and parties’ standpoints on six areas—war memory, nationalism, minority rights, economic policy, libertarian–authoritarian values, and military policy—are the explanatory variables. We test the correspondence of parties’ stance on war memory with their left–right placement in three models: pared down Model 1 with only War memory as the explanatory variable; Model 2 which adds the parties’ stance on Economic policy and Libertarian–authoritarian values; and finally Model 3 which adds Nationalism, Minority rights, and Military policy. We opt for this technique because, as discussed above, we recognize that in the Southeast European context Nationalism, Minority rights, and Military policy could be conceptually and in practical terms very closely related to War memory. The wars of the 1990s were ultimately nationalist wars which were fought militarily over both territory and minority/majority status of different ethnic groups. Parties’ positions toward nationalism, minority rights, and the military were arguably decisively molded by what took place during that decade. Being so related to our issue of interest—and yet conceptually distinctive from it—we believe our three models present a useful way of establishing exactly what the nature of the relationship between war memory and left–right orientation really is.

Table 3.

Determinants of parties’ left–right positions.

	Model 1			Model 2			Model 3
	B	SE	Sig.	B	SE	Sig.	B	SE	Sig.
War memory	−0.56	0.06	***	−0.19	0.05	***	−0.09	0.05	†
Economic policy				0.52	0.08	***	0.52	0.08	***
Libertarian–authoritarian values				0.56	0.06	***	0.49	0.09	***
Nationalism							0.00	0.08
Minority rights							−0.14	0.05	**
Military policy							0.09	0.05	†
Constant	8.42	0.31	***	0.58	0.65		0.81	0.60
Country controls	Yes			Yes			Yes
N	69			69			69
Δ Adj. R²	51.42% (53.25%)			82.18% (84.02%)			84.57% (86.40%)

Note: †p < 0.1; *p < 0.05; **p < 0.01; ***p < 0.001; Δ Adj. R² indicates the improvement in explained variance when compared to the model with only country controls.

Model 1 shows that War memory has a strongly statistically significant (at the 0.001 level) and negative relationship with left–right orientation. Substantively, Model 1 suggests that a one standard deviation (equal to 2.15) move toward greater acknowledgment of war crimes committed by members of the in-group leads to a decrease of the party’s left–right score (i.e., a move to the left) equal to 1.2. Considering that the standard deviation of the Left–right position variable is equal to 1.59, this is a substantively large effect. What is equally important to note is that even this simple model does well, with more than 53% of the variation explained. The effect of War memory survives the inclusion of Economic policy and Libertarian–authoritarian values in Model 2, though it obviously becomes smaller. The effects of the two newly added variables are also highly statistically significant and in the expected direction. With only these three explanatory variables, this model does exceptionally well as it succeeds in explaining 84% of the variation. The inclusion of Nationalism, Minority rights, and Military policy in Model 3 does little to improve the model’s explanatory power. Even here, however, the variable War memory remains statistically significant. The diminishing effect of War memory due to the inclusion of other party positions should come as no surprise and is evidence of the ideology-making tendencies of political parties (Converse, 1964). The origins of the conflicts in the 1990s can be traced to the pursuit of national self-determination and the rights of ethnic groups. It is therefore only natural for parties’ positions on these two issues to be connected to their War memory views. However, the fact that war memory views retain an independent effect on parties’ left–right positions heavily suggests that the wars and their memory have added something unique to the nature of political competition. Cognizant of all the limitations of this methodological approach, we feel confident in concluding that the evidence supports our first hypothesis.

Our second hypothesis, on the other hand, suggested that voters’ congruence with parties’ stance on war memory was a significant factor in their decision whether to vote for a party or not. The results of our analysis are presented in Table 4. Model 1 includes only country dummies and War memory congruence. Models 2 and 3 add the congruence measures for other issues and socio-demographic control variables. The models thus become increasingly stringent tests of the effect of war memory congruence on party preferences. In the models, the effects of almost all variables, including War memory congruence, are highly significant. This is unsurprising given the large sample size. This is why statistical significance is not a useful metric for assessing support that Hypothesis 2 receives in the data. While a non-significant finding would have been evidence for a lack of support, statistical significance must be supplanted by a substantive effect size before we can conclude that the assertion is confirmed. Therefore, we plot the relation between War memory congruence and party preference in Figure 4. Keeping all other variables at their mean values, no congruence on war past views is estimated to result in a ptv score of 0.8, while perfect agreement in a score of 4.3. This demonstrates, in our view beyond doubt, that War memory congruence has a sizable effect especially when considering that we are controlling for a number of alternative explanations. Our analysis therefore shows that respondents in the postwar societies of Southeast Europe still assess parties with the question of war memory in mind, more than two decades after the conflicts had ended. This is true even when controlling for the voters’ views on other, arguably closely related, issues and even when controlling for voters’ personal war experiences. This indicates that the legacy of war is not merely an expression of another political fault line that predates the war but that it even may be a cleavage-like fault line in its own right. We see this as clear evidence in support of our second hypothesis.

Table 4.

Analyses of respondents’ party preferences.

	B	SE	Sig.	B	SE	Sig.	B	SE	Sig.
War memory congruence	0.83	0.02	***	0.41	0.02	***	0.38	0.02	***
Nationalism congruence	—		—	0.57	0.03	***	0.65	0.03	***
Military policy congruence	—		—	0.25	0.01	***	0.17	0.01	***
Minority rights congruence	—		—	0.31	0.01	***	0.27	0.01	***
Libertarian–authoritarian values	—		—	0.86	0.02	***	0.68	0.02	***
Economic policy preferences	—		—	0.56	0.02	***	0.49	0.02	***
Personal war experiences	—		—	—		—	0.52	0.04	***
Ethnicity	—		—	—		—	0.68	0.02	***
Religiosity	—		—	—		—	0.07	0.01	***
Education level	—		—	—		—	0.09	0.01	***
Social class	—		—	—		—	0.40	0.02	***
Gender	—		—	—		—	0.07	0.01	***
Age	—		—	—		—	0.29	0.02	***
Political interest	—		—	—		—	0.12	0.01	***
Non-voter	—		—	—		—	0.11	0.01	***
Government coalition	—		—	—		—	0.40	0.43
Year of founding	—	––	—	—	—	—	−0.02	0.02	—
Constant	−0.84	0.40	*	−1.23	0.42	***	1.82	3.70	—
Country controls	Yes			Yes			Yes
	Total/Respondents/Parties			Total/Respondents/Parties			Total/Respondents/Parties
n	115337/10736/69			115337/10736/69			115337/10736/69
Δ AIC	−1971			−9814.3			−14,652.6

Note: †p < 0.1; *p < 0.05; **p < 0.01; ***p < 0.001; Δ AIC compares the model to a model with only the country controls.

Figure 4.

Relation between war memory congruence and party preference.

Regarding our control variables, voter/party congruence on nationalism also exerts a particularly large effect on party preferences, increasing propensity to vote scores by 0.99 for a one standard deviation increase in congruence. The effects of one standard deviation increases in Military policy congruence (1.79) and Minority rights congruence (1.69) are, on the other hand, substantially smaller (0.3 and 0.42 respectively). Here, we should also comment on the set of our socio-demographic control variables that are not measuring congruence on various issues. Their coefficients cannot be interpreted substantively. They represent predicted ptv scores based on indicators of the listed concepts. As such, they are scaled according to the dependent variable, that is, a 10-point scale (Van der Brug, 2010: 593). The values of the coefficients thus indicate how well the—ptv scores can predict the actual ptv scores. Most interestingly, the coefficient 0.51 for Personal war experiences means that for every increase in—based on respondents’ six measures of personal war experiences, the actual ptv scores go up by 0.51. This is a sizeable effect that is surpassed only by the effects for Libertarian–authoritarian values and Ethnicity. For the other variables in the model, these coefficients are smaller, suggesting that they play a weaker role. This suggests that voters’ personal war experiences are a sizeable factor in their decision whether to vote for a party or not. Our further robustness tests, however, do not offer solid clues about the direction the effect of Personal war experiences takes.⁴ This could be the case for a number of reasons: from specificities of party systems in individual countries to the possibility that party preferences in the region can be a matter of identity that war experiences have generated that is conceptually distinct from policy preferences. This question is beyond the scope of this article but will be the subject of our further research.

Conclusions

In January 2020, the President of Serbia, Aleksandar Vučić, during his visit to Serb-populated areas of the Federation of Bosnia and Herzegovina, stated “There will be no ‘storms’ and no tractors in the future of the Serb people” (Tportal, 2020). Vučić obviously referred to Croatia’s 1995 Operation Storm that defeated the Serb Krajina statelet and led to the exodus of nearly 200,000 Serbs, many of them escaping violence on their tractors and other agricultural vehicles. Mere days following Vučić’s statement, at the commemoration of the Markale massacre in Sarajevo, proceedings were interrupted by a local man who demanded that the ceremony clearly states that the people killed by mortar shells on the Markale market died at the hands of the Bosnian Serb forces (Glamoč, 2020). Such episodes are standard in Southeast Europe, even decades after the disastrous wars of the 1990s. However, does the fact that war has a significant role in politicians’ public discourse or in the everyday social and political lives of the people mean that war memories structure the nature of political competition? Our article strongly suggests that they do.

We demonstrate that war memory in Southeast Europe is not only the subject of historiographical contestation (Kasapović, 2019) but is also one of the principal sources of ideological and policy differentiation among political parties, as well as one of the strongest determinants of voter choice. Voters choose which parties to support based on the level of their agreement with the parties’ positions on the way recent war past should be remembered. Parties in turn take different positions on war memory and these positions are significant determinants of their general ideological profile and placement on the left–right continuum. We believe these lessons from post-conflict societies of Southeast Europe are portable across geographic and time contexts. A growing body of research shows that traumatic divisions such as wars can remain embedded in social relations and political competition for decades after the violence ends in societies as diverse as Peru, El Salvador, Sri Lanka, and Sierra Leone (Wood, 2008). There is nothing contextually so specific about the experiences of war trauma in Southeast Europe that would disqualify the lessons from this region being applicable across a wide range of cases, though obviously this needs to be empirically tested. We argue that how political players conceptualize public memories of such traumatic events is the principal mechanism that keeps these events and the resulting divisions politically relevant—most often with disastrous consequences, regardless of context.

As Kubik and Bernhard (2014) have shown, the implications of contentious fractures in collective memory and their active politicization are nearly uniformly negative. They destabilize institutions and poison public discourse. They heighten political contention to the levels of debates over legitimacy, thus complicating coalition- and consensus-building, as well as governance. They thus result in more polarized forms of party pluralism and lower levels of inter-personal trust, which can have negative effects on the development of cultural and social capital and economic growth. Does this mean, however, that the societies of Southeast Europe, or other contemporary post-conflict societies, will remain prisoners of their past for decades to come? After all, as Lipset and Rokkan’s argument suggests, political divisions remain salient as long as the underlying issues have not been resolved (Rohrschneider and Whitefield, 2009), and in many post-conflict societies, memories of past wars do not seem to be resolved. Our analysis does not offer much cause for optimism. We firmly believe that the only way for the spiral of politicization of war memory to be broken in Southeast Europe is for the societies of the region to have a clear European future. As Kitschelt et al. (1999) have argued more than two decades ago, only new relations, institutions, and development may gradually help historical legacies recede into the past.

Supplemental Material

sj-pdf-1-ppq-10.1177_13540688211039378 – Supplemental Material for Contesting war memories: Parties and voters in contemporary Southeast Europe

Supplemental Material, sj-pdf-1-ppq-10.1177_13540688211039378 for Contesting war memories: Parties and voters in contemporary Southeast Europe by Josip Glaurdić, Christophe Lesschaeve and Michal Mochtak in Party Politics

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This article is supported by H2020 European Research Council (714589).

ORCID iD

Josip Glaurdić

Supplemental material

Supplemental material for this article is available online.

Notes

Author biographies

Josip Glaurdi\x{0107} is an Associate Professor of Political Science and Head of the Institute of Political Science at the University of Luxembourg. He is also Principal Investigator of the project Electoral Legacies of War (ELWar) funded by the European Research Council Starting Grant. He earned his PhD in political science from Yale University in 2009.

Christophe Lesschaeve is a postdoctoral researcher at the University of Luxembourg. He earned his PhD from the University of Antwerp in 2017.

Michal Mochtak is a postdoctoral researcher at the University of Luxembourg. He earned his PhD from Masaryk University in 2015.

References

Alexander

(2004) Toward a theory of cultural trauma. In: Alexander

Eyerman

Giesen

, et al. (eds) Cultural Trauma and Collective Identity. Berkeley, CA: University of California Press, pp. 1–30.

Ansolabehere

Rivers

(2013) Cooperative survey research. Annual Review of Political Science 16(1): 307–329.

Ashplant

Dawson

Roper

(2015) The Politics of War Memory and Commemoration. London, UK: Routledge.

Assmann

(2007) Europe: a community of memory? Bulletin of the GHI Washington 40: 11–25.

Bélanger

Meguid

(2008) Issue salience, issue ownership, and issue-based vote choice. Electoral Studies 27(3): 477–491.

Bieber

(2019) The Rise of Authoritarianism in the Western Balkans. London, UK: Palgrave Macmillan.

Budge

(2000) Expert judgements of party policy positions: uses and limitations in political research. European Journal of Political Research 37(1): 103–113.

Burt

Neiman

(1987) Elite belief consistency and the effect of position in the policy-making process. Western Political Quarterly 40(1): 121–136.

Chirot

Shin

G-W

Sneider

(2014) Confronting Memories of World War II: European and Asian Legacies. Seattle, WA: University of Washington Press.

10.

Converse

(1964) The nature of belief systems in mass publics. In: Apter

(ed) Ideology and its Discontent. New York, NY: Free Press of Glencoe, pp. 206–261.

11.

De Vries

Van der Brug

Van Egmond

, et al. (2011) Individual and contextual variation in EU issue voting: the role of political information. Electoral Studies 30(1): 16–28.

12.

Dixon

(2018) Dark Pasts: Changing the State’s Story in Turkey and Japan. Ithaca, NY: Cornell University Press.

13.

Dolenec

(2017) A soldier’s state? Veterans and the welfare regime in Croatia. Anali Hrvatskog Politološkog Društva 14(1): 55–77.

14.

Downs

(1957) An economic theory of political action in a democracy. Journal of Political Economy 65(2): 135–150.

15.

Edkins

(2003) Trauma and the Memory of Politics. Cambridge, UK: Cambridge University Press.

16.

Gallagher

(1985) Political Parties in the Republic of Ireland. Dublin, Ireland: Gill&Macmillan.

17.

Gillis

(1994) Memory and identity: the history of a relationship. In: Gillis

(ed) Commemorations: The Politics of National Identity. Princeton, NJ: Princeton University Press, pp. 3–24.

18.

Glamoč

(2020) Incident na komemoraciji povodom sjećanja na poginule na Markalama u Sarajevu. Oslobođenje. available at https://www.oslobodjenje.ba/vijesti/sarajevo/incident-na-komemoraciji-povodom-sjecanja-na-poginule-na-markalama-u-sarajevu-529187.

19.

Gligorov

(2011) What price nationalism? Peščanik. available at https://pescanik.net/what-price-nationalism.

20.

Gödl

(2007) Challenging the past: Serbian and Croatian aggressor-victim narratives. International Journal of Sociology 37(1): 43–57.

21.

Green

Hobolt

(2008) Owning the issue agenda: party strategies and vote choices in British elections. Electoral Studies 27(3): 460–476.

22.

Hackmann

(2018) Defending the “good name” of the Polish nation: politics of history as a battlefield in Poland, 2015–18. Journal of Genocide Research 20(4): 587–606.

23.

Hashimoto

(2015) The Long Defeat: Cultural Trauma, Memory, and Identity in Japan. New York; NY: Oxford University Press.

24.

Hite

(2012) Politics and the Art of Commemoration: Memorials to Struggle in Latin America and Spain. London, UK: Routledge.

25.

Hlousek

Kopecek

(2016) Origin, Ideology and Transformation of Political Parties: East-Central and Western Europe Compared. London, UK: Routledge.

26.

Hooghe

Bakker

Brigevich

, et al. (2010) Reliability and validity of the 2002 and 2006 chapel hill expert surveys on party positioning. European Journal of Political Research 49(5): 687–703.

27.

Inglehart

Klingemann

(1976) Party identification, ideological preference and the left–right dimension among western mass publics. In: Budge

Crewe

Farlie

(eds) Party Identification and Beyond: Representations of Voting and Party Competition. London, UK: Wiley, pp. 243–273.

28.

Inglehart

(1984) The changing structure of political cleavages in Western societies. In: Dalton

Flanagan

Beck

(eds). Electoral Change in Advanced Industrial Societies. Princeton, NJ: Princeton University Press, pp. 25–69.

29.

Inglehart

(2015) The Silent Revolution: Changing Values and Political Styles Among Western Publics. Princeton, NJ: Princeton University Press.

30.

Iyengar

Kinder

(1987) News that Matters: Television and American Opinion. Chicago, IL: University of Chicago Press.

31.

Jennings

(1992) Ideological thinking among mass publics and political elites. Public Opinion Quarterly 56(4): 419–441.

32.

Kalton

Flores-Cervantes

(2003) Weighting methods. Journal of Official Statistics 19(2): 81–97.

33.

Kasapović

(2019) Povijest, povijesni revizionizam i politike povijesti. Časopis za Suvremenu Povijest 51(3): 939–960.

34.

Kitschelt

Hellemans

(1990) The left–right semantics and the new politics cleavage. Comparative Political Studies 23(2): 210–238.

35.

Kitschelt

Mansfeldova

Markowski

, et al. (1999) Post-Communist Party Systems: Competition, Representation, and Inter-party Cooperation. Cambridge, UK: Cambridge University Press.

36.

Knutsen

(1995) Value orientations, political conflicts and left–right identification: a comparative study. European Journal of Political Research 28(1): 63–93.

37.

Kriesi

(2010) Restructuration of partisan politics and the emergence of a new cleavage based on values. West European Politics 33(3): 673–685.

38.

Kubik

Bernhard

(2014) A theory of politics of memory. In: Bernhard

Kubik

(eds) Twenty Years after Communism: The Politics of Memory and Commemoration. Oxford, UK: Oxford University Press.

39.

Lazic

(2013) Memory claims and memory constraints. (re)negotiating statehood and identities in Serbia. Nationalities Papers 41(6): 936–952.

40.

Lefkofridi

Wagner

Willmann

(2014) Left-authoritarians and policy representation in Western Europe: electoral choice across ideological dimensions. West European Politics 37(1): 65–90.

41.

Lesschaeve

(2017) The predictive power of the left–right self-placement scale for the policy positions of voters and parties. West European Politics 40(2): 357–377.

42.

Lijphart

(1999) Patterns of Democracy: Government Forms and Performance in Thirty-Six Countries. New Haven, CT: Yale University Press.

43.

Lind

(2008) Sorry States: Apologies in International Politics. Ithaca, NY: Cornell University Press.

44.

Linz

Stepan

(1996) Problems of Democratic Transition and Consolidation: Southern Europe, South America, and Post-Communist Europe. Baltimore, MD: Johns Hopkins University Press.

45.

Malka

Shenhav

Rahat

, et al. (2019) The collective memory of dominant parties in parliamentary discourse. Party Politics 27(3): pp. 489–500. Online first, 1–12.

46.

Marks

Hooghe

Steenbergen

, et al. (2007) Crossvalidating data on party positioning on European integration. Electoral Studies 26(1): 23–38.

47.

Olick

(2016) The Sins of the Fathers: Germany, Memory, Method. Chicago, IL: University of Chicago Press.

48.

Olick

Robbins

(1998) Social memory studies: from “collective memory” to the historical sociology of mnemonic practices. Annual Review of Sociology 24: 105–140.

49.

Olick

Vinitzky-Seroussi

Levy

(2011) The Collective Memory Reader. Oxford, UK: Oxford University Press.

50.

Polk

Rovny

Bakker

, et al. (2017) Explaining the salience of anti-elitism and reducing political corruption for political parties in Europe with the 2014 Chapel hill expert survey data. Research & Politics 1–9.

51.

Rabe-Hesketh

Skrondal

(2008) Multilevel and Longitudinal Modeling Using Stata. College Station, TX: Stata Press Publication.

52.

Rigney

(2012) Reconciliation and remembering: (How) does it work? Memory Studies 5(3): 251–258.

53.

Rohrschneider

Whitefield

(2009) Understanding cleavages in party systems. Comparative Political Studies 42(2): 280–313.

54.

Ruedin

Morales

(2019) Estimating party positions on immigration: assessing the reliability and validity of different methods. Party Politics 25(3): 303–314.

55.

Said

(2000) Invention, memory, and place. Critical Inquiry 26(2): 175–192.

56.

Sartori

(1969) From the sociology of politics to political sociology. In: Lipset

(ed) Politics and Social Sciences. Oxford, UK: Oxford University Press.

57.

Sartori

(1976) Parties and Party Systems: A Framework for Analysis. Cambridge, UK: Cambridge University Press.

58.

Scheufele

Tewksbury

(2006) Framing, agenda setting, and priming: the evolution of three media effects models. Journal of Communication 57(1): 9–20.

59.

Soroka

(2006) Good news and bad news: asymmetric responses to economic information. The Journal of Politics 68(2): 372–385.

60.

Steenbergen

Marks

(2007) Evaluating expert judgments. European Journal of Political Research 46(3): 347–366.

61.

Tilly

(1975) The Formation of National States in Western Europe. Princeton, NJ: Princeton University Press.

62.

Torcal

Mainwaring

(2003) The political recrafting of social bases of party competition: Chile, 1973–95. British Journal of Political Science 33(1): 55–84.

63.

Tportal (2020) Vučić u BiH: Nema više ‘Oluja' i traktora u srpskoj budućnosti. Zagreb, Croatia: Tportal. 29 January. available at: https://www.tportal.hr/vijesti/clanak/vucic-u-bih-nema-vise-oluja-i-traktora-u-srpskoj-buducnosti-20200129.

64.

Van der Brug

(2010) Structural and ideological voting in age cohorts. West European Politics 33(3): 586–607.

65.

Van der Eijk

Franklin

(1996) Choosing Europe? The European Electorate and National Politics in the Face of Union. Ann Arbor, MI: University of Michigan Press.

66.

Van der Eijk

van der Brug

Kroh

, et al. (2006) Rethinking the dependent variable in voting behavior: on the measurement and analysis of electoral utilities. Electoral Studies 25(3): 424–447.

67.

Walczak

van der Brug

de Vries

(2012) Long- and short-term determinants of party preferences: inter-generational differences in Western and East Central Europe. Electoral Studies 31(2): 273–284.

68.

Walgrave

Lefevere

(2013) Ideology, salience, and complexity: determinants of policy issue incongruence between voters and parties. Journal of Elections, Public Opinion & Parties 23(4): 456–483.

69.

Wolfrum

(1999) Geschichtspolitik in Der Bundesrepublic Deutschland: Der Weg Zur Bundesrepublikanischen Erinnerung 1948–1990. Darmstadt, Germany: WBG.

70.

Wood

(2008) The social processes of civil war: the wartime transformation of social networks. Annual Review of Political Science 11: 539–561.

71.

Zaller

(1992) The Nature and Origins of Mass Opinion. Cambridge, UK: Cambridge University Press.

72.

Zhang

Mildenberger

Howe

, et al (2018) Quota sampling using Facebook advertisements. Political Science Research and Methods 8(3): 558–564.

73.

Zubrzycki

Woźny

(2020) The comparative politics of collective memory. Annual Review of Sociology 46: 175–194.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.41 MB