Loving it but not feeling it yet? The state of European identity after the eastern enlargement

Abstract

The inclusion of 11 new member states from the former Eastern bloc constitutes a significant challenge to the European Union in various respects. Many worry that whatever tenuous ‘European identity’ existed prior to the eastern enlargement, it has now become so diluted that no meaningful European political community can form. We provide an empirical account of the state of European identity after the eastern enlargement through a comparative analysis of affective and cognitive European identity in the old and the new Central and Eastern European member states. Our empirical analyses indicate that while the overall levels of cognitive European identification in the East are indeed lower than in the West, citizens from new member states are just as attached to Europe as citizens from old member states. Most importantly, not only is there no discernable difference in cognitive identification among young Europeans in East and West, but the youngest in the East seem to be even more strongly attached to Europe than their peers in the West.

Keywords

European identity European Union post-communist enlargement

Introduction

The accession of 11 post-communist states to the European Union (EU) constitutes a significant challenge to the Union in many respects. Although hotly debated, previous research has provided empirical evidence for the existence of an emerging European identification among the citizens of EU-15¹ member states (Bruter, 2005; Citrin and Sides, 2004; Fligstein, 2008). The eastern enlargement, however, has brought the viability of a European demos into question (cf. Mach and Pozarlik, 2008). Some scholars have argued that the different historical developments in Europe after the Second World War, including the experience of socialist rule in the East and the transition away from it, have provided for fundamentally different political cultures and attitudes in East and West (Checkel and Katzenstein, 2009; Fuchs and Klingeman, 2006; Thomassen and Bäck, 2009). Thus, the increased social, economic, and political heterogeneity introduced by the biggest enlargements so far could put a strain on incipient European collective identity.

The identity question constitutes a central issue in the current processes of change within the EU, especially in relation to the debate on the democratic legitimacy of its institutions. These discussions have only intensified in the context of the Great Recession, as the economic problems within the Eurozone have tested the limits of solidarity among Europeans and have triggered a rise in nationalist discourse. There is a prevailing sense that without achieving a certain sense of community in Europe, the goal of establishing the EU as a democratic system with political authority over Europeans is bound to fail (Bruter, 2005; Cerutti, 2011; Fuchs and Klingeman, 2006; Herrmann and Brewer, 2004; Thomassen and Bäck, 2009). Thus, if European identification is key to the future of the European integration project, are there systematic differences in the level of European identification in Eastern and Western Europe? We seek to address this question through a comparative analysis of the attitudes of average citizens in 18 EU member states. Our main motivation is to provide an empirical account of the state of European identity after the eastern enlargement and, thus, contribute to a debate that has been marked by a conspicuous lack of empirical analyses.

We advance two main arguments. First, based on previous research and social identity theory, we argue that the European identity question should be approached by analyzing both its cognitive (i.e. seeing oneself as European) and its emotional or affective (i.e. attachment to Europe) dimensions (Citrin and Sides, 2004; Kuhn, 2012). By distinguishing these two dimensions, we are able to show that citizens of post-communist EU countries are less likely to perceive themselves as European than their Western counterparts, even though their attachment to Europe is deep-seated. This could be due to the fact that the division running through the heart of the continent during the Cold War denied citizens in the East their belonging to a modern, economically developed and a unified Europe, making it harder for them to self-identify as fully European. Second, we argue that such differences in European identification are related to political socialization and, thus, are marked by generational difference. While older citizens of post-communist Europe might indeed find it harder to embrace their newly found supranational European belonging, younger people in Central and Eastern Europe (CEE), who have been socialized as part of the European community and have enjoyed EU citizenship rights for most of their lives, are just as likely to identify as Europeans as their peers in the West. Thus, this generational effect on European identification in the East can be explained by the different early socialization of the two groups: older citizens were socialized in a communist system with no reference to a unified European political identity while younger ones were socialized during the transition away from communism when the notion of ‘returning to Europe’ was widely prevalent.

Our arguments have broad implications for the debate on the future of European identity. If East Europeans are unquestionably attached to Europe, and if the young among them are particularly likely to espouse European identity, then citizens of the new member states could contribute to stronger European identification in the future. In other words, not only has the eastern enlargement not hurt the chances for the emergence of a common European identity, but it might also provide a much needed élan to the slow process of supranational identity formation in Europe.

Our main findings support this conclusion. While the overall levels of cognitive identification with Europe in the East are indeed lower than in the West, citizens from new member states are just as attached to Europe as citizens from old member states. Most importantly, not only is there no discernable difference in cognitive identification among young Europeans in East and West, the youngest in the East seem to be even more strongly attached to Europe than their peers in the West.

Eastern enlargement and European identity

The possible effects of the eastern enlargement on European identity in the enlarged EU have been assessed from different perspectives. Some authors argue that this enlargement has had a negative effect on the sense of community within the Union, as the new member states exhibit lower levels of trust in other people (Thomassen and Bäck, 2009). Another source of preoccupation has been the possibility that citizens of Central and Eastern European countries may remain more attached to their nations because of recently regained national sovereignty, and, therefore, nationalism in this region might stand in opposition to the cosmopolitan citizenship and supranational identity implied by European integration (Liebich, 2010).

Others yet propose that the eastern enlargement has made it more difficult for a European demos to emerge due to different political cultures in East and West which, according to Fuchs and Klingeman (2006: 28), are caused by:

different traditions and historical events in the distant past, but also by socialization and experience in the opposing societal systems in which people in eastern and western Europe lived from the end of Second World War until the collapse of the communist states.

These differences materialize in unequal levels of civic engagement and trust in other people, which, according to Fuchs and Klingemann (2006), indicate divergent conceptualizations of a democratic community in Western and Central-Eastern Europe which further complicates political integration. Finally, Checkel and Katzenstein (2009: 14) suggest that the eastward enlargement has profoundly transformed European identification processes, making the emergence of a common identity more problematic, notably due to the importance attached to religion as a constitutive element of European belonging in the new member states.

While the different historical experiences and differential socialization in CEE member states might influence their citizens’ attitudes and European identification, we should not overstate their importance either. In fact, the process of eastern enlargement relied heavily on the assumed existence of a common identity in Eastern and Western Europe (Sjursen, 2002). EU accession was the final step in the long process of making a definitive break with the communist past and reunifying with the rest of the continent after almost five decades of division along ideological lines (Vachudova, 2005). As such, the desire to be considered European was of paramount importance for the overwhelming popular support for European integration in CEE countries as the catchphrase ‘return to Europe’ captured so well.

In sum, while there is some empirical evidence pointing to attitudinal differences between East and West due to post-communist legacies in the former, a shared identity constituted an important element of and justification for the process of enlargement. Therefore, we must explore empirically the possible differences in European identification between East and West and examine their sources.

European identity—The conceptual terrain

A primary contribution of this article is to provide a theoretically motivated empirical account of the state of European identity after EU’s eastward enlargement. Before we can expound our theory, we need to clarify the conceptual framework that guides our thinking. Many scholars believe that a European political community is functionally needed for the EU to be a proper democratic system of governance (Bruter, 2005; Cerutti, 2011; Fuchs, 2011; Fuchs and Klingeman, 2006; Herrmann and Brewer, 2004; Thomassen and Bäck, 2009). However, what exactly such a notion of community would entail, how we would delimit the idea of European identity, and what would be the best way to measure it remain much debated.

While some authors suggest that we should do away with the concept of identity altogether because it is too vague (Brubaker and Cooper, 2000), others argue that, although heavily contested, it remains useful if we adopt an empirical, bottom-up perspective (Bruter, 2005) and if we acknowledge its functional value for European integration (Fuchs, 2011). Social identity theory developed in social psychology has been the starting point for most theoretically driven empirical research on European identity to date (e.g. Citrin and Sides, 2004; Risse, 2010; Sanders et al., 2012). From this perspective, European identity is constituted by at least two basic elements: affective and cognitive.² The affective element of identity refers to the evaluation and emotional significance of group membership (Tajfel, 1981: 255), based on a favorable view of the group one belongs to (Tajfel and Turner, 1979). Thus, the affective dimension of identity refers to ‘identification with’ (Citrin and Sides, 2004) and, in the case of European identity, it taps into the emotional closeness or attachment that citizens have to Europe as a territorial entity, which implies a positive evaluation of such membership.

The second element of social identity refers to self-categorization which implies greater emphasis on the cognitive processes of social identification (Turner et al., 1987). The cognitive dimension of European identity, thus, refers to the self-perception as a member of a community, an ‘identification as’ (Citrin and Sides, 2004), which taps into the perceived similarity to the members of the group (Turner et al., 1987). In this sense, the emphasis of most of the research on European identity is not on those who hold an exclusively European identity. Rather, the focus is on the existence of identities which are inclusive of identification as European (cf. Hooghe and Marks, 2005). This approach is based on the assumption that people can hold multiple, non-conflicting identities (Herrmann and Brewer, 2004; Risse, 2010) and that there is no reason to presuppose an opposition between national and European identity (Bruter, 2005) or that national identity necessarily hinders support for the EU (Christin and Trechsel, 2002). Thus, the focus is on those citizens who hold some kind of layered or mixed identity, both as nationals and Europeans.

Although social identity theory offers useful conceptual tools for studying European identity, some difficulties remain not least so due to a deep theoretical confusion in the literature that conflates European identity on one hand with attachment to the EU as a political community on the other. Most importantly, the use of the term ‘European identity’ to refer to a collective identity which underpins the EU can be challenged on the grounds that the geographical and cultural notion of Europe is broader than the political community of the Union. Moreover, ‘European identity’ could actually constitute a cultural category which has nothing to do with top-down identity creation, with European identification predating EU accession (Schilde, 2014).

Previous research that has aimed to unpack this difficult relationship has often relied on theoretical clues from studies of nationalism and has focused on the criteria for inclusion in a community. In this sense, Bruter (2005:12) proposes to distinguish between two basic components of European political identity: a cultural one, where belonging is based on perceived shared culture, values, religion, and ethnicity; and a civic one, defined as identification with a political structure, institutions, rights, and rules of the political community—the EU. We find Bruter’s distinction useful and, given the available survey data, we focus on cultural European identity. However, the results of our analyses point to a strong relationship between cultural European identification (attachment to Europe and self-perception as European) and trust in the EU and perceived benefits from European integration. Therefore, we contribute to the conceptual debate on European versus EU identity by showing that cultural European identity is directly linked to a positive image of the EU.

Explaining European identity in Eastern and Western Europe

The existing literature provides a number of explanations of European identification. In their review of identity theory in the context of EU governance, Kaina and Karolewski (2013) provide a synthesis of the existing approaches and divide them into those that focus on contextual factors and those that study the predispositions of individuals. Among the individual-level factors, they include resources, experiences, and attitudes as principal determinants of European collective identity.

First, the importance of individual resources reflects the findings of research on European attitudes. Thus, we must take into account one’s socioeconomic status, education, political interest, and knowledge, as well as language skills and their social capital (Kaina and Karolewski, 2013). Previous empirical studies on European identity have found that proficiency in foreign languages (Fuss et al., 2004), higher levels of education (Fligstein, 2008; Kuhn, 2012), and political interest (Boomgaarden et al., 2011) are all associated with increased European identification. These findings are related to the cognitive mobilization model, which explains how better educated people are more able to connect to a remote political community and develop a European identity (Inglehart, 1970). Thus, we expect a positive effect of higher education and EU-specific knowledge on European identification.

As far as attitudinal determinants of European identity are concerned, we can distinguish between evaluative and affective factors. Evaluative attitudes are characterized by instrumental rationality including perceptions of benefits from integration (personal and national) and general trust in European institutions. This set of determinants can be derived directly from social identity theory, which assumes that a basic tenet of identity formation is the acquisition of positive self-perception (Tajfel and Turner, 1979). Therefore, more positive evaluations of the EU should strengthen European identification.

In terms of affective attitudinal factors, the strength and meaning of national identity is assumed to play a crucial role in determining European identification. If constructed as inclusive, national identity may reinforce supranational identification, as the multiple identities theory posits (Díez Medrano and Gutiérrez, 2001; Kuhn, 2012). Exclusive national identity may, on the other hand, constitute an important obstacle to support for the EU (Hooghe and Marks, 2005) and make the emergence of a supranational identification more difficult. Thus, if national identity is conceived in more exclusive terms, based on kinship, ethnicity, and religion, it should affect negatively identification with Europe.

Finally, personal experiences constitute important elements which affect European identification and they are related to the socialization dimension of the general European attitude formation model. Transnational experiences seem to play an especially positive role, in line with some of the early theories of integration (Deutsch et al., 1957), which assumed that community formation would stem from increased transnational contacts between elites and citizens. Such positive impact of transnational contacts on European identification has been documented in more recent empirical research (Fligstein, 2008; Kuhn, 2012; Recchi and Favell, 2009; Sigalas, 2010; Stöckel, 2015). Moreover, the assumption that transnational contacts between Europeans foster European identification underlies several European polices such as the Erasmus program for student exchange in higher education.³

Previous research on European identity and its determinants has concentrated almost exclusively on Western and Southern Europe, while scholarship on EU attitudes in Central and Eastern Europe has focused mostly on the issue of support (Guerra, 2013a). The articles by Schilde (2014) and Sanders et al. (2012) are the notable exceptions. In her recent article, Schilde (2014) tests existing explanatory frameworks of European identification with data from CEE candidate countries. Sanders et al. (2012) focus on European citizenship as a multidimensional concept after the enlargement. The latter find that while European identification is indeed lower in CEE countries, the general structure of attitudes and the underlying cognitive processes are quite similar in East and West (Sanders et al., 2012: 229).

Theoretical framework

Our theoretical framework builds on the existing literature and advances a new theoretical proposition on European identity based on socialization and age effects in Eastern Europe. By applying the two-dimensional concept of European identity (i.e. cognitive and affective), we can better understand why EU citizens in the East are on average less likely to see themselves as European than their Western counterparts. We argue that the main reason for this is that citizens of post-communist Europe have not yet internalized their self-perception as European due to their short socialization as members of a common European political community. Social identity theory and the existing literature on collective identification with the EU points to the importance of ‘primary socialization’ that occurs in the early years of one’s life when Europeanizing discourses and symbols are internalized (Recchi, 2014). European identity, like national identity, develops and becomes crystallized during mid-adolescence and remains rather stable among adults. A recent study using panel data on adolescents and adults in Belgium confirms this logic (Verhaegen, 2015).

The socialization argument implies a generational divide in cognitive identification with Europe in the East because many East Europeans experienced their primary socialization under the communist system with no symbolic or discursive reference to a shared common European identity with Western Europe. This is in line with research that shows that older generations in the East are also more nostalgic about the communist past (Rose and Carnaghan, 1995). In contrast, younger people in the East who grew up during the transition away from communism were exposed to the Europeanizing discourse that accompanied the process of reuniting with Europe through EU enlargement (Vachudova, 2005). This suggests that while older people might find it more difficult to embrace a supranational identity, younger people should find identification as European more natural. These theoretical expectations regarding the cognitive dimension of European identification are summarized in H1a and H1b.

H1a: On average, EU citizens from the East are less likely to identify as European (cognitive identification) than EU citizens from the West.

H1b: While identification as European (cognitive identification) is on average lower in the East than in the West, there is an interaction effect between age and cognitive European identification in the East such that younger people are more likely to identify as European than older people.

It is important to note, however, that we are not making a case for the effect of the length of EU membership for European identification. This issue has been explored in previous research with mixed results (Bruter, 2005; Risse, 2010; Sanders et al, 2012). Our argument simply takes a central insight from social identity theory—that social identities are crystalized early on in one’s life and remain relatively stable afterwards—to its logical conclusion. If we are right that younger Eastern Europeans were socialized as members of a European political community while growing up, they should perceive themselves as Europeans much more readily than their parents and grandparents who were denied that socialization opportunity in their formative years. In a sense, we could be witnessing a process of ‘catching up’ of European identification in Eastern Europe. In the West, it is entirely plausible that the postwar generation that grew up in the wake of the horrors of Second World War was more European than newer generations which cannot relate to Europe with the same urgency. Thus, the focus is not on the length of membership in the EU but on the specific context of socialization of each cohort.

In contrast, attachment to Europe in the East (i.e. affective identity) is as deeply seated as it is in the West because membership in the EU was a widely popular political goal throughout the 1990s, and there was a general elite consensus in Central Eastern Europe on this issue. Although the EU did not represent the primary political community of reference in CEE countries, it was often seen as an alternative to the ill-performing national institutions of the young democracies in the East (Ilonszki, 2009) and Europeanization as the process that would return Eastern Europe to the family of modern European countries. Given the importance of positive evaluations of the in-group for affective identity, there is no reason to expect that citizens in the East would be less attached to Europe than their Western counterparts.

It is less clear, however, how the generational divide would play out in this dimension of European identity. On the one hand, it is plausible that young people in the East would be more attached to Europe than older people who lived under communist systems predicated on an ideological divide running through the heart of Europe. On the other hand, social identity theory implies that socialization during formative years is not as important for affective as it is for cognitive identity. Thus, given the widespread support for European integration across all demographic strata, we do not have a clear prediction about the effect of age on affective identity in the East.

These expectations about the affective dimension of European identity in the East are summarized in H2a and H2b.

H2a: EU citizens from the East are no less likely to be attached to Europe (affective identification) than EU citizens from the West.

H2b: Different age cohorts of EU citizens from the East are equally likely to be attached to Europe (affective identification).

As mentioned in the introduction, several authors presuppose the existence of an essential difference between the old and the new CEE member states of the EU. However, the potentially divisive effect of the eastward enlargement on the prospects of European political identity must be empirically tested. The next section of this article analyzes the empirical evidence regarding European identity in post-2004 Europe.

Data and empirical analyses

Most of the empirical analyses on attitudes towards European integration rely on Eurobarometer surveys (Bruter, 2005; Citrin and Sides, 2004; Fligstein, 2008; Fuchs, 2011). Instead, we use the 2007 and 2009 IntUne elite and mass surveys to explore the link between elites and mass European identity in the aftermath of the eastward enlargement. However, the main drawback of using IntUne, as opposed to Eurobarometer or European Social Survey, is their relatively poorer coverage in terms of countries (for details see the online appendix and Figure 1).

Figure 1.

European identity compared—Public opinion. Affective European identity: Share of responses ‘Somewhat attached’ and ‘Very attached’ to Europe. Cognitive European identity: Share of responses ‘Nationality and European’, ‘European and nationality’, or ‘European only’ to the question: ‘Do you see yourself as …?’. Data: IntUne Mass Survey Wave 2, 2009.

Measuring European identity

For the purpose of this article, European identity is understood in its affective and cognitive dimensions. The affective dimension is operationalized as attachment to Europe on a territorial level. The survey question we use to operationalize affective European identity is the following: People feel different degrees of attachment to their town or village, to their region, to their country and to Europe. What about you? Are you very attached, somewhat attached, not very attached, or not at all attached to Europe? For easier interpretation, we have dichotomized this variable to take values of 0 (not very attached, or not at all attached) and 1 (very attached, somewhat attached).⁴ Previous research has often treated attachment to Europe as proxy for European identity (e.g. Citrin and Sides, 2004; Deutsch, 2006), and some authors even argue that the former is the most reliable measure of the concept (Sinnott, 2006).

The cognitive dimension of European identity refers to the self-perception as European and we use the following survey question to tap into this dimension: Do you see yourself as …? (NATIONALITY) only, (NATIONALITY) and European, European and (NATIONALITY), European only. We code respondents as having inclusive cognitive European identity if they see themselves as European—either solely or with a double (national and European or vice versa) identification. This operationalization has often been employed by other scholars (Citrin and Sides 2004; Fligstein, 2008; Kuhn, 2012) and taps into the degree of Europeanization of identities (Risse, 2010).

Independent variables of interest

The most important independent variables for the purposes of this analysis are Age and East. Age has six categories that correspond to different age cohorts, where higher values indicate older cohorts. East is a binary variable that takes the value of 1 if the respondent comes from Eastern Europe and 0 if they come from Western Europe.

Controls

As discussed previously, the existing literature has identified a number of individual-level factors that affect European identification. In terms of resources, the cognitive mobilization theory indicates that better educated people are more capable of relating to and identifying with a remote political community, so we control for educational attainment (Education). Along similar lines, those who have a better understanding of the EU tend to be more favorably predisposed towards it and are more likely to hold a European identity, thus we control for EU-specific knowledge about membership as a proxy for the depth of knowledge about the EU (EU Knowledge).

To account for the effect of transnational experiences and direct interactions with other Europeans, we control for the number of trips an individual has made to another EU country in the preceding year (Visit EU). In terms of attitudinal predictors, positive perceptions of one’s community constitute a key element of social identification. Arguments about the importance of instrumental rationality and utilitarian considerations have dominated the literature on support for the EU. Given that such considerations can be sociotropic or egocentric in nature, we control both for (a) whether the respondent believes that their country has benefited from EU membership (sociotropic—Country benefit) and (b) whether the respondent feels that people like them have benefited from EU membership (egocentric—Personal benefit). Similarly, the extent to which one has diffuse support for the EU should also affect identification with Europe. We thus control for trust in the EU (Trust EU).

We also account for affective factors. How national identity is conceived should impact identification with Europe. Namely, if national identity is conceived in exclusive terms and is based on religion, kinship, and ethnicity then it should negatively affect identification with Europe. Thus, we created a new variable with the mean score of the survey items asking respondents to indicate how important (1—not at all important to 4—very important) (a) being a Christian, (b) being born in the country, and (c) having parents of one’s nationality are for the meaning of one’s national identity (Ethnic ID).⁵ Given that some scholars have raised the issues of religiosity in the East as an impediment to European identity, we control for the frequency of religious attendance (Religiosity). We control also for the standard left-right ideological self-placement (Left-Right) and the sex of the respondents (Female = 1).

In addition to individual-level determinants of European identity, we need to take into account the way the national context shapes identity. Political elites define and articulate national versions of European identity and, consequently, shape the degree of Europeanization of identities among the national mass public (Bruter, 2005; Hooghe and Marks, 2005; Risse, 2010) even if elite cues might be less important in the East due to lower levels of political trust in elites (Guerra, 2013b). Therefore, to account for the potential effect of elite cues, we include a country-level aggregate measure with the percentage of elites who claim to be attached to Europe (EU attached). To account for the different levels of economic development and general economic conditions in East and West, we control for GDP per capita, GDP Growth, and Unemployment. We also account for whether respondents live in a country that is a net recipient of EU funds (Net recipient) and for time effects associated with the wave of the survey (Wave).

Results

We begin by exploring the data in broad strokes using the most recent 2009 IntUne surveys. Figure 1 presents both dimensions of European identity for the average citizens in new and old member states.

The first interesting thing to note is that, on average, citizens of new member states are more attached to the EU (Affective European identity) than their Western counterparts.⁶ Specifically, almost 70% of citizens from the East say that they are somewhat or very attached to Europe, while 64% of West Europeans say the same. While there are significant differences between the CEE countries—Hungary being the country where citizens express the strongest attachment to Europe, and in Estonia the weakest—in all six countries more than 60% of the public claim to be attached to Europe. This finding is in line with H2a and suggests that citizens in the East are, on average, no less likely to be attached to Europe than citizens in the West.

Regarding the second dimension of European identity, the cognitive identity, the inverse is true; on average, only 51% of citizens from Central Eastern Europe think of themselves as Europeans, while that figure is 66% in the old member states. Furthermore, in each CEE country the share of citizens who perceive themselves as European is consistently smaller than the share of those who are attached to Europe—the biggest gaps between affective and cognitive European identity occur in Hungary and Bulgaria. This finding brings us to our central point: while citizens of CEE countries express overall strong attachment to Europe, they still have difficulties seeing themselves as European.

Interestingly, in old member states affective and cognitive European identity are on equal footing and the gap between the two dimensions is rather small. This is especially true for the founding member states such as Italy and Germany. In these countries, it is natural for those who are attached to Europe as a political community to also see themselves as Europeans. Such is not the case for the citizens of the new member states though, which points to the necessity of a more careful study of European identification in East and West.⁷

To summarize, when we analyze European identity as part of multiple political identities, there is no sign of the end of supranational identification (see also Recchi and Salamońska, 2014). To the contrary, it is clear that supranational belonging remains quite widespread in post-2004 EU. There is, however, an important difference in the cognitive and affective dimensions of identification between the old member states and the CEE new members. Namely, while citizens of CEE countries feel quite attached to Europe (affective dimension), this does not necessarily translate into considering themselves European (cognitive dimension) to the same extent as it does in the old member states.

To get a better grip on the causal dynamics underlying European identification, we turn to regression analysis to see whether the differences between East and West hold when controlling for a number of individual and country-level variables. We employ multilevel logistic modeling with two levels (individual and national) to test our hypotheses. In order to utilize all the data available from the IntUne surveys, we have combined the surveys from the 2007 and 2009 waves, giving us 29 country-level observations. As a robustness check, and to include all the individual-level data available, we have conducted additional analyses without EU Attached (elites) because there are no elite surveys for Estonia in the 2009 and for Slovenia in 2007 and 2009. This gives us 32 country-level observations. The results are virtually the same and we report them in the online appendix. As far as the modeling is concerned, likelihood-ratio tests suggest that models with random slopes for the interacted Age variable provide a better fit (i.e. have the lowest log likelihood), so all the results reported below are from models that use varying intercepts and varying slopes for Age.

For clearer interpretation of the regression results, we have standardized all the non-dichotomous independent variables by subtracting their mean and dividing them by two standard deviations, thus rendering all coefficients roughly comparable with each other, including those of the untransformed binary predictors (Gelman, 2008). The advantage of standardizing the independent variables is that the magnitudes of the coefficients are directly comparable because they present the change on the logit scale for the dependent variable as each independent variable goes from a low to a high value.

Figure 2 presents the ropeladder plot with the regression results with 95% confidence intervals for Cognitive European ID. Since we expect age to be a moderating factor, we include a cross-level interaction term Age*East to capture this effect. Figure 3 presents (a) the adjusted predicted probabilities for cognitive European ID for East and West over different age cohorts holding all other variables at their means, and (b) the marginal effects which test whether these predicted probabilities are statistically significantly different (for detailed tables, see the online appendix). Given that our main theoretical claim implies an interaction effect between Age and East, we begin by discussing the results from Figure 3.

Figure 2.

Cognitive European identity—multi-level logistic regression results.

Figure 3.

Cognitive European identity—predicted probabilities. (a) Predicted probabilities for cognitive European ID for East and West over different age cohorts and (b) the marginal effects which test whether these predicted probabilities are statistically significantly different. Crossing the zero line indicates that the differences are not statistically significant.

A number of interesting findings emerge. First, as can be seen from Figure 3(a), there is clear interaction effect between age and being from a CEE new member state, but age has the opposite effect in East and West.⁸ Namely, while older people are more likely than younger people to identify as European in the West, the opposite is true in the East—people in older cohorts in the East are significantly less likely to hold a cognitive European identity than their peers in the same cohorts in the West. The effect of the interaction term is substantively very large. For instance, Eastern European citizens who belong to the 55–65 age cohort are about 40% less likely to see themselves as European than their counterparts from Western Europe. These findings provide strong support for H1a and H1b.

To see whether the differences in cognitive European identity are statistically significant in East and West over different age cohorts, we present a marginal effects graph (Figure 3(b)) where crossing the zero line indicates that the differences are not statistically significant. As can be seen from Figure 3(b), the probability of having a cognitive European identity is statistically indistinguishable in East and West for those under the age of 25. This result is consistent with the literature that emphasizes the effect of ‘primary socialization’ in the early years of one’s life for the formation of European identity (Recchi, 2014). Members of the under 25 cohort in the East are the ones that grew up during the post-communist period and were steeped in the idea of a ‘return to Europe’, so it is not surprising that cognitive European identity is the strongest in this cohort. In short, even after controlling for a host of individual- and country-level factors, cognitive European identity is on average lower in the East than in the West (H1a supported), with younger people in the East having stronger cognitive European identity than older people (H1b supported).

As far as Affective European ID is concerned, Figure 4 presents the regression results with standardized variables using a ropeladder plots with 95% confidence intervals. Just as before, Figure 5 presents the predicted probabilities (a) and the marginal effects (b) of the impact of East and West on attachment to Europe for different age cohorts holding all else constant.

Figure 4.

Affective European identity—multi-level logistic regression results.

Figure 5.

Affective European identity—predicted probabilities. (a) Predicted probabilities for affective European ID for East and West over different age cohorts and (b) the marginal effects which test whether these predicted probabilities are statistically significantly different. Crossing the zero line indicates that the differences are not statistically significant.

The results are quite revealing. As we can see from Figure 5(a), citizens from the East are, on average, more likely to be attached to Europe than citizens from the West for all age cohorts and this is in line with H2a. However, this relationship is statistically significant only for the youngest cohort (i.e. those under the age of 25), and there is no statistically significant difference in the levels of attachment to Europe in East and West for those who are 25 or older (Figure 5(b)). Substantively speaking, those who belong to the youngest cohort (16–24) in the East are about 17% more likely to be attached to Europe than youngsters from the same cohort in the West. As with cognitive identity, age is positively related to attachment to Europe in the West, with those under 25 being about 18% less likely to be attached to Europe than those who are 65 and older. Contrary to what we expected (H2b), older people in the East are also more likely to be attached to Europe than younger people although the difference between the youngest cohort (<25) and the oldest one (65+) is only 6% meaning that there is a much more uniform high attachment to Europe in the East than in the West across different age cohorts. This observation becomes clearer if one looks at the slopes of the predicted probabilities for East and West in Figure 5(a)—the flatter slope for East indicates that age has a much smaller effect on attachment to Europe than in the West.

In sum, there is no evidence to suggest that citizens from the new member states of the EU are less likely to be attached to Europe than citizens from older member states (H2a supported). If anything, the evidence points to higher levels of attachment to Europe among the youngest in the East. Surprisingly, older East Europeans seem to be more attached to Europe than younger ones (H2b rejected), but the age effect is rather small. When taken as a whole, the results suggest very high attachment to Europe across the board in the East with over 70% of respondents in each age cohort indicating attachment to Europe.

As for the rest of the variables, the results are largely consistent with the existing studies. In line with the cognitive mobilization thesis, the more educated (Education) and the more knowledgeable about the EU (EU knowledge) are more likely to hold both stronger affective and cognitive European identity than the less educated and the uninformed. Interestingly, those to the left of the ideological spectrum are less likely to be attached to Europe or identify as European than those to the right; however, this effect is substantively small and only significant for the cognitive dimension of European identity. While women are more likely to be attached to Europe than men, the latter are more likely to see themselves as European than the former.

Holding a strong ethnic national identity (Ethnic ID) is not related to affective European identity (Figure 4), but it is strongly and negatively related to cognitive European identity (Figure 2). Unsurprisingly, this suggests that an inclusive, Europeanized identity is less compatible with an ethnic conception of national identity. Similarly, Religiosity has no effect on Attached to Europe, but there is a positive and statistically significant relationship between Religiosity and identification as European suggesting that higher levels of religiosity in some of the CEE countries could, in fact, be conducive to cognitive European identity. Although it is beyond the scope of this article, we should note that this finding speaks directly to the literature on the value ‘content’ of European identity and the central role that Christianity might play in the construction of European identity (Galland and Lemel, 2007). However, the substantive effect of regular attendance of religious services is very small—those that attend religious services more than once a week are only about 5% more likely to identify as European than those who never attend such services. So, the importance of religiosity for European identity should not be overstated either.

Furthermore, our results support the contact hypothesis as can be seen by the statistically significant relationship between Visit EU and both dimensions of European identity. In other words, those who get to travel frequently to other EU countries and who interact with other Europeans are more likely to hold a European identity than those who are less mobile. Utilitarian considerations seem to play a crucial role in forging a European identity. Those who believe that the EU is beneficial to them (Personal benefit from EU) or to their country (Country benefit from EU) are more likely to be attached to Europe and identify as European. For both dimensions of identity, egocentric considerations seem to play a larger role than sociotropic ones, as can be seen by the relatively large size of the effects of Personal benefit from EU in Figures 2 and 4.

Diffuse support for the EU, as measured by Trust EU, is positively, and statistically significantly, related to both dimensions of identity and this variable has one the largest substantive impact in the two regression analyses. Going from a low value to a high value of Trust EU is associated with a 22% higher probability of being attached to the EU and a 19% higher probability of identifying with Europe. This is a major finding that contributes to the debate about the distinction between civic (i.e. EU) and cultural European identification. Although the direction of causality is likely complicated, this finding shows that there is a strong relationship between the EU as a political institution on the one hand and identification with a European community on the other.

To turn to the country-level factors, elite cues seem to play an important role in forging a European identity; elite attachment to the EU (EU attached—elites) is positively related to both dimensions of European identity. Furthermore, there is no statistically significant relationship between being from a country that is net recipient of EU funds (Net recipient) and holding a European identity. The effects of the macroeconomic variables on European identity are less straightforward. While GDP per capita is negatively related to both dimensions of European identity, this effect is statistically significant only for cognitive European identity suggesting that, all else equal, citizens of richer European countries are less likely to see themselves as European. More short-term macroeconomic factors such as unemployment and GDP growth do not seem to have an impact on cognitive identification, but they matter for affective identification. What is less clear is why higher levels of unemployment and GDP growth are associated with less attachment to Europe, as they seem to be pointing in different directions. Further research is needed to explore the link between short-term macroeconomic performance and the more long-term process of European identification.

Conclusion

In this article, we have explored whether there are systematic differences in the levels of European identification in CEE and Western Europe. Our findings suggest that citizens from new member states remain attached to Europe but, consistent with other studies, they do not consider themselves European to the same extent as citizens in the West.

Such findings could validate the fears of those who anticipated a negative effect of the eastward enlargement on the European political community and a greater resistance to supranational identifications in Eastern Europe. However, our empirical analyses paint a brighter picture of the future of cognitive European identity. Not only are younger people in the new member states significantly more likely to identify as Europeans than older people in these countries, there is actually no discernable difference in cognitive identification among the youngest cohorts in old and new member states. We argue that such generational divide in cognitive European identification in the East is due to the fact that older East Europeans experienced their primary socialization under communist systems with no symbolic or discursive reference to a shared European identity with Western Europe. In contrast, younger people in the East who grew up during the transition away from communism were exposed to the Europeanizing discourse that accompanied the process of reuniting with Europe through EU enlargement. Such exposure facilitated identification as Europeans, because this dimension of European identity is based on perceived similarity, and the effects are the strongest for the youngest cohorts because national and supranational identities are crystallized in adolescence.

This brings us to the broader question of identity change. When thinking about the emerging European identity, we must remember that drastic identity transformations do not occur later on in one’s lifecycle. Rather, social identities, European identity included, crystallize in the early stages of life and remain relatively stable thereafter. Therefore, change could only materialize with generational replacement.

Empirical evidence of such generational effects for European identity is difficult to come by due to data limitations. However, our analysis has dynamic implications for identity change in Central and Eastern Europe based on generational effects. If social identity theory is correct about the crucial role of ‘primary socialization’ that occurs in the early years of one’s life when Europeanizing discourses and symbols are internalized, this does not bode well for European identity in the West. Our findings indicate that, all else constant, with intergenerational replacement, fewer Western Europeans will hold a European identity. Intergenerational replacement will have the opposite effect in the East, however, where younger people have a stronger cognitive European identity than older people. This implies that, contrary to concerns about its deleterious effect on European identity, the eastern enlargements might end up reinvigorating supranational identity formation in Europe. That said, European identification is a complex and an underdeveloped phenomenon, and its future far from predictable. There is no question that the eastern enlargement has added heterogeneity to the European community, which could end up undermining overall European identification over time. Furthermore, present-day economic difficulties in the EU could affect the identification processes among the young cohorts in the enlarged EU due to their disproportionate exposure to unemployment and other economic woes.

Footnotes

Author contributions

The authors contributed equally to this work.

Acknowledgments

We thank the editors of European Union Politics and four anonymous reviewers for their constructive comments. We also thank Samer Al-Saber, Christine Marshall, Jeff Rose, Anelise Shrout and Caroline Weist for helpful comments on an early draft as part of an interdisciplinary writing group at Davidson College.

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: Aleksandra Sojka would like to acknowledge the financial support received for the development of this research from the Spanish Ministry of Education (FPU grant), the University of Granada and the Andalusian Research Group in Political Science (SEJ-113).

Notes

References

Boomgaarden

Schuck

Elenbaas

(2011) Mapping EU attitudes: Conceptual and empirical dimensions of Euroscepticism and EU support. European Union Politics 12(2): 241–266.

Brubaker

Cooper

(2000) Beyond ‘identity’. Theory and Society 29(1): 1–47.

Bruter

(2005) Citizens of Europe? The Emergence of a Mass European Identity, Houndmills: Palgrave Macmillan.

Cerutti

(2011) How not to (mis)understand political identity in the European Union. In: Cerutti

Lucarelli

Schmidt

(eds) Debating Political Identity and Legitimacy in the European Union, New York: Routledge, pp. 3–15.

Checkel

Katzenstein

(2009) The politicization of European identities. In: Checkel

Katzenstein

(eds) European Identity, Cambridge: Cambridge University Press, pp. 1–25.

Christin

Trechsel

(2002) Joining the EU? Explaining public opinion in Switzerland. European Union Politics 3(4): 415–443.

Citrin

Sides

(2004) Can Europe exist without Europeans? Problems of identity in a multinational community. In: Hermann

(ed.) Advances in Political Psychology, New York: Pergamon Press, pp. 41–70.

Deutsch

(2006) Legitimacy and identity in the European Union: Findings from the old member states. In: Karolewski

Kaina

(eds) European Identity: Theoretical Perspectives and Empirical Insights, Berlin: Lit Verlag, pp. 149–178.

Deutsch

Burrell

Kann

(1957) Political Community and the North Atlantic Area: International Organization in the Light of Historical Experience, Princeton: Princeton University Press.

10.

Díez Medrano

Gutiérrez

(2001) Nested identities: National and European identity in Spain. Ethnic and Racial Studies 24(5): 753–778.

11.

Fligstein

(2008) Euroclash: The EU, European Identity, and the Future of Europe, Oxford: Oxford University Press.

12.

Fuchs

(2011) Cultural diversity, European identity and legitimacy of the EU: A theoretical framework. In: Fuchs

Klingemann

H-D

(eds) Cultural Diversity, European Identity and the Legitimacy of the EU, Glos: Edward Elgar Publishing, pp. 27–51.

13.

Fuchs

Klingeman

H-D

(2006) Democratic communities in Europe. A comparison between East and West. In: Klingemann

H-D

Fuchs

Zielonka

(eds) Democracy and Political Culture in Eastern Europe, London: Routledge, pp. 27–67.

14.

Fuss

García-Albacete

Rodriguez-Monter

(2004) The role of language skills and foreign country experiences in the development of European identity. Sociológia 36(3): 273–292.

15.

Galland

Lemel

(2007) Valeurs et cultures en Europe, Paris: La Découverte.

16.

Gelman

(2008) Scaling regression inputs by dividing by two standard deviations. Statistics in Medicine 27(15): 2865–2873.

17.

Guerra

(2013a) Central and Eastern European Attitudes in the Face of Union, Basingstoke: Palgrave Macmillan.

18.

Guerra

(2013b) Does familiarity breed contempt? Determinants of public support for European integration and opposition to it before and after accession. Journal of Common Market Studies 51(1): 38–50.

19.

Herrmann

Brewer

(2004) Identities and institutions: Becoming European in the EU. In: Herrmann

Risse

Brewer

(eds) Transnational Identities: Becoming European in the EU, Oxford: Rowman & Littlefield, pp. 1–22.

20.

Hooghe

Marks

(2005) Calculation, community and cues: Public opinion on European integration. European Union Politics 6(4): 419–443.

21.

Ilonszki

(2009) National discontent and EU support in Central and Eastern Europe. Europe-Asia Studies 61(6): 1041–1057.

22.

Inglehart

(1970) Cognitive mobilization and European identity. Comparative Politics 3(1): 45–70.

23.

Kaina V and Karolewski IP (2013) EU governance and European identity. Living Reviews in European Governance 8.

24.

Kuhn

(2012) Why educational exchange programmes miss their mark: Cross-border mobility, education and European identity. Journal of Common Market Studies 50(6): 994–1010.

25.

Liebich

(2010) Is there (still) an East-West divide in the conception of citizenship in Europe? In: Liebich

Bauböck

(eds) Is there (still) an East-West Divide in the Conception of Citizenship in Europe?, Florence: EUI Working Papers, pp. 1–4.

26.

Mach Z and Pozarlik G (2008) Collective Identity Formation in the Process of EU Enlargement. Defeating the Inclusive Paradigm of a European Democracy? RECON Online Working Paper. Oslo: ARENA Centre for European Studies.

27.

Recchi

(2014) Pathways to European identity formation: A tale of two models. Innovation: The European Journal of Social Science Research 27(2): 119–133.

28.

Recchi E and Favell A (eds) (2009) Pioneers of European Integration: Citizenship and Mobility in the EU. Glos: Edward Elgar.

29.

Recchi

Salamońska

(2014) Keeping the European faith. Collective identity before and after the Euro-crisis. PArtecipazione e COnflitto 7(3): 509–531.

30.

Risse

(2010) A Community of Europeans? Transnational Identities and Public Spheres, Ithaca: Cornell University Press.

31.

Rose

Carnaghan

(1995) Generational effects on attitudes to communist regimes: A comparative analysis. Post-Soviet Affairs 11(1): 28–56.

32.

Sanders D, Bellucci P, Toka G, et al. (eds) (2012) The Europeanization of National Polities? Citizenship and Support in a Post-Enlargement Union. Oxford: Oxford University Press.

33.

Schilde

(2014) Who are the Europeans? European identity outside of European integration. Journal of Common Market Studies 52(3): 650–667.

34.

Sigalas

(2010) Cross-border mobility and European identity: The effectiveness of intergroup contact during the ERASMUS year abroad. European Union Politics 11(2): 241–265.

35.

Sinnott

(2006) An evaluation of the measurement of national, subnational and supranational identity in crossnational surveys. International Journal of Public Opinion Research 18(2): 211–223.

36.

Sjursen

(2002) Why expand? The question of legitimacy and justification in the EU’s enlargement policy. Journal of Common Market Studies 40(3): 491–513.

37.

Stöckel F (2015) Contact and community: The role of social interactions for a political identity. Political Psychology. Epub ahead of print 26 August 2015. DOI: 10.1111/pops.12295.

38.

Tajfel

(1981) Human Groups and Social Categories: Studies in Social Psychology, Cambridge: Cambridge University Press.

39.

Tajfel

Turner

(1979) An integrative theory of intergroup conflict. In: Austin

Worchel

(eds) The Social Psychology of Intergroup Relations, Monterey: Brooks/Cole Publishing, pp. 33–47.

40.

Thomassen

Bäck

(2009) European citizenship and identity after the enlargement. In: Thomassen

(ed.) The Legitimacy of the European Union After Enlargement, Oxford: Oxford University Press, pp. 184–207.

41.

Turner

Hogg

Oakes

(1987) Rediscovering the Social Group: A Self-Categorization Theory, Blackwell: Oxford.

42.

Vachudova

(2005) Europe Undivided: Democracy, Leverage, and Integration after Communism, Oxford: Oxford University Press.

43.

Verhaegen

(2015) The Development of European Identity. A Study of Individual-Level Development Processes, Leuven: KU Leuven.

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

0.42 MB

5.29 MB