http://eiop.or.at/eiop/texte/2004-011a.htm
: 29.7.2004| This paper's comments page | Send your comment! to this paper |
| European and National Identities in EU's Old and New Member States: Ethnic, Civic, Instrumental and Symbolic Components(*) |
| Antonia M. Ruiz Jiménez, Jaroslaw Józef Górniak, Ankica Kosic, Paszkal Kiss, Maren Kandulla |
| European Integration online Papers (EIoP) Vol. 8
(2004) N° 11; http://eiop.or.at/eiop/texte/2004-011a.htm |
| Date of Publication in
: 29.7.2004 |
| | Abstract |
Back to homepage |
PDF | | This paper's comments page | Send your comment! to this paper | |
1
Research into European identity has evolved recently in two quite different directions, with very few contact between the two strands.(1) On the one hand, we find a number of theoretical and normative approaches which focus on the definition of the concept of European identity, but which have rarely been subjected to empirical testing with attitudinal data (Orchard 2002). On the other, there are a number of empirical approaches, mainly developed by sociologists and social psychologists, but which lack the necessary theoretical and methodological underpinning.(2) The analysis of European identity presented in this article cuts a middle path between these two dominant approaches to the subject. We introduce a number of existing theories about the potential emergence of a European identity and test them by using the most appropriate empirical data to answer the questions each poses. Our contribution to the field lies not, therefore, in the development of new theories or the presentation of unexpected empirical findings, but rather in the way this analysis empirically tests existing theories. Such a test is, of course, necessary if our knowledge of European identity is to go beyond the mere “history of a concept and a discourse” (Stråth 2002: 288).
2
Greatly simplifying existing theories about the emergence of a European identity, we have identified three main tendencies in the literature, distinguished by the different possible sources of European identification they defend. We have labelled these three approaches the “cultural”, “instrumental”, and “civic” theories. A general common preoccupation, within these theories, is the relation between citizens’ European and national identities. This has not been the object of systematic empirical research either. When at the end of the 1980s the EU launched a number of policies aimed at creating a European identity, the member states responded by incorporating into the Maastricht Treaty a clause stating that the Union should respect the member states’ respective national identities (article F, point 1). This reaction, along with the introduction of the principle of subsidiarity and the rejection of the word “federal”, revealed that, in line with the “cultural” theory’s conception of identities, many member states saw the creation of a European identity as a potential threat to their own national identities and their citizens’ national loyalties. Indeed, in the early 1990s national identity was used by the political elites to justify their defence of independent statehood and sovereignty. Due to the close links between national identity and national independence, many experts have argued that the European integration process could be seen as a threat to, and incompatible with, national identity (Höjelid 2001), and hence difficult to achieve.
Along with these pessimistic visions of the creation of European identity, we also find optimists who do not conceive European identity in zero-sum terms or see it as incompatible with national identity (Cowles, Caporaso and Risse 2001; Kersbergen 1997; Weiler 1999). Some authors believe that the increasing globalisation of communications and the economy is weakening popular interest in national, domestic concerns to the benefit of stronger identification with issues beyond national borders (Cerutti 1992). Advocates of such interpretations would highlight the growing importance of values with relatively universal appeal, such as human rights, the protection of civil liberties, social rights, and democracy. Other analysts conclude that EU integration actually strengthens the nation-state, making it more effective and strengthening governments’ capacity to tackle global and multidimensional problems which require complex coordination, regulatory and enforcement institutions etc. (Milward 2000; Moracscik 1998).
Those who remain sceptical about the potential development of a European identity reject the notion that citizens can possess more than one identity simultaneously, whereas other authors have noted that people can hold multiple identities, each deriving from the same source of human attachment (conceptualised as kind of “concentric circles” with different levels of intensity), or feel simultaneously attached to multiple identities based on different subjective factors of identification (Weiler 1999: 345; Kersbergen 1997: 11). Although little empirical research has been carried out into these questions, there is some evidence to support the hypothesis that attachments to multiple identities draw on different sources (Cowles, Caporaso and Risse 2001: 234).
The group of scholars that we identify with the “cultural” theory argue that if a European identity were to emerge, it would not, and should not, be based on the same elements (such as a common cultural heritage, language, myths, symbols and emotional bonds) which form the foundations of national identities (see Smith 1992, 1995, 1999; Østerud 1999). According to Østerud (1999), an EU identity could be established in two ways. One would emulate “classical” nation-building of the type seen during the 19th century, but now developing at the European level and including an attempt to construct a stronger European identity. However, such a process might be more problematic than in the case of nation states, due to a number of cultural, linguistic, economical and geopolitical factors. Cultural diversity and structural differences are deeply rooted and, as a result, expectations of the emergence of a European identity should be regarded as rather utopian. While it may be possible for such an identity to emerge, this would certainly take a long time and would ultimately lead to the substitution of national identities.
3
From this theory we can develop a number of hypotheses for empirical test:
For a second group of scholars, who we have identified as proponents of the “instrumental” theory, European identities would be consciously decided on, and are thus based on calculated individual self-interest. A numbers of authors argue that instrumental factors may play an important role in defining and strengthening individuals’ sense of identity (Brass 1979; Cinnerella 1997). The perception of the potential gains or losses that might result from membership of a given social group may influence peoples’ identification with it. The more people consider their country to lack sufficient decision-making power within the international context, the more strongly they identify with Europe. From this instrumental perspective, the better the citizen’s evaluation of the results of European policies (compared to the results of policies pursued by national government), the more likely s/he is to feel European. This cost-benefit understanding of identities implies that low welfare spending, poor economic performance, and low-quality national-level democracy would all make citizens more likely to feel more European (Gabel 1998; Eichemberg and Dalton 1993; Gabel and Palmer 1995; Kaltenthaler and Anderson 2001; Olsen 1996; Sánchez-Cuenca 2000; Fernández-Albertos and Sánchez-Cuenca 2001).(3)
There is a number of empirical hypotheses implied, some of which we will be able to test with the Eurobarometer 57.2 data:
4
Finally, a third line of analysis, which we have labelled the “civic” theory, stresses the development of identities around agreement over rules for peaceful political co-existence, shared cultural norms, and common beliefs (Mancini 1998:8; Weiler 1999: 346; Kersbergen 1997). From this perspective, the substance of EU membership (and European identity) lies in a commitment to the shared values of the Union as expressed in its constituent documents, a commitment to the duties and rights of a civic society covering specific areas of public life, a commitment to membership of a polity which promotes the direct opposite of classic ethno-nationalism, that is, human characteristics which transcend the differences at the heart of organic ethno-culturalism (Weiler, Haltern and Mayer 1995: 23). Drawing on deliberative theories of democracy and Habermasian conceptions of communicative rationality, the emergence of a European public sphere would be crucial for the emergence of a European identity (Eriksen and Fossum 2001).
The hypotheses to be derived from this stream are the following:
We have used these debates to select the different items and dimensions to be examined in this article. We provide empirical information relevant to each of the debates and hypotheses, which are then brought together and discussed in more detail in the conclusions. Our main findings are that European and national identities are compatible because: a) they are conceptually different and b) of a different order, involving different notions of loyalty. Eurobarometer studies have shown that they are more or less compatible, but the question of why this is so, and how and why they differ, has not been studied. While national identities were largely constructed during the specific historical, socio-economic and political circumstances of the era of nation-state building, European identity is still under construction involving, and presupposing the existence of national identities without replacing them. We also argue that national identities, which are deeply rooted in history, impose inertias and dynamics that affect the way European identities are seen and understood in individual countries.
This analysis is based on Eurobarometer data. In recent years many researchers have become increasingly dissatisfied with the way these surveys poll attitudes towards the EU, and particularly with the quantitative indicators used to measured identities. We have contributed to this debate by designing special new questions to measure national and European identities for Eurobarometer 57.2, and then examining the results in this analysis. These new questions will enable us to discover whether the two identities are similar (and can be measured using the same kind of indicators) or different (which would call for different instruments to measure each identity).
The data analysed here come from national, probability (multistage) samples from the EU member states, Hungary, Poland and the Czech Republic.(4) Samples were taken of the resident population aged 15 or over in each country. Details on the participants in the Standard Eurobarometer 57.2 are given in the table below.(5)
5
Before proceeding with the analysis, we eliminated non-nationals from the country samples. The analysis refers, therefore, exclusively to citizens of the countries in question. We applied weighting for the member states and accession countries, so each sample totals 1,000 respondents, thereby making the same contribution to the aggregate analysis. When carrying out some analyses we also centered the data. Finally, we have imputed missing values.(6) Although we found only a small percentage of missing values for individual questions, for the questionnaire as a whole we lost about half our respondents on these grounds. Attribution was, therefore, recommendable. We have rounded the values obtained through attribution to match the previous categories.
The Standard Eurobarometer 57.2 is particularly well-suited to test the theoretical debates discussed above. It contains one battery of questions measuring closeness to different in-groups and out-groups (including the nation, the EU, Europe and Central and Eastern Europe) and two batteries of questions measuring the dimensions held to be most relevant for identification with the nation and Europe. These last two sets of questions have not figured in previous or subsequent Eurobarometers or similar studies and will help us establish whether both types of identities can be, in fact, measured through just one or several questions put exactly the same way, as is standard practice in Eurobarometer studies.
We use the indicator “closeness to different groups of people” to measure identity. We consider that it has advantages over the indicators used in other Eurobarometers to assess European identity, such as “citizenship”, “support for European integration” and “pride in being European”. Since one of our main goals is to distinguish different possible sources of European and national identities, a useful definition of identity could begin with David Easton’s concept of political community (1965). He provides a definition of national identity which is also applicable to European identity: it is open, in the sense that it does not refer to the sources of this identity, and it is easy to operationalise for cross-country comparisons. Translating Easton’s definition of national identity to European identity we may define European identity: “as a we-feeling or a ‘sense of community’ that consists of the feeling of belonging together as a group which, because it shares a political structure, also share a political fate” (Easton 1965: 185). The question of the relative importance of different elements as the source of European and national identification, and whether they constitute a single or several dimensions, is a matter of empirical investigation that we will analyse in this article. “Closeness” is, therefore, the indicator we will use to measure identity. “Citizenship” can be understood as being more closely related to rights and duties (and the “civic” theory), and “support for integration” as more closely connected to economic (as well as perhaps political) costs and benefits (and the “instrumental” theory). In this sense, “closeness” is a more neutral notion. This is not meant to imply that any of the possible components of identity mentioned in the previous section are more important than others, and therefore it is better than citizenship and support for integration to measure identity as defined by Easton. “Pride”, on the other hand, could be defined as the positive effect that results from feelings of identity (not identity itself). The question on closeness that we will use includes level of closeness to some “in-groups” (those living in the same city, region, and nation), to EU citizens, fellow Europeans and people living in Central and Eastern Europe. It also included nine “out-groups” (both internal and external).(7) Six of those nine “out-group” are common to all countries; the other three represent the largest immigrant groups in each.
6
The two batteries of questions measuring the elements perceived as most significant for identification with the nation and Europe include a set of 14 items, each related to different components of identity mentioned in the section above. We have not established a predetermined and closed theoretical classification of these items as belonging to the different “civic”, “instrumental” and “civic” concepts of identity. How nationals of different countries categorize them is a question for empirical research. Nonetheless, the inclusion of these items in the questionnaire responded to the following hypothesised grouping of items. In the case of national identities, common culture, customs and traditions, common language, common ancestry and common history and destiny are related to the “cultural” conception of identity; common rights and duties and a common political and legal systems are linked to the “civic” theory; a common system of social security/welfare is associated with the “instrumental” theory about identities, as are, probably, national economy, national army and common borders.(8) We have also included some other items which, although not manifestly related to any of these three theories, could be termed “affective-symbolic” components of identity; these include national independence and sovereignty, national pride, national character and national symbols.
We suggest a similar grouping of items with respect to European identity. Common civilisation, membership of a European society with many languages and cultures, a sense of common ancestry and common history and destiny pertain to the “cultural” theory as applied to a European identity; an emerging common political and legal systems and common rights and duties are more closely related to the “civic” conception of European identity; and a common system of social protection and the right to freedom of movement and residence are linked to the “instrumental” vision of identity, as are, probably, the future European defence system, common borders and a single currency. (9) We also included pride in being European, in the sovereingty of the EU or in a set of European Union symbols as more closely related to an “affective-symbolic” conception of identities.
In these last two batteries of questions the respondent had the option to answer “I do not feel national/European”. In these cases, the interviewer did not proceed with the remaining questions in each battery. Except in Great Britain, this option was not offered as a prior filter, but only recorded if the respondent spontaneously mentioned the idea. This poses a problem for comparative analysis, since we found that as many as 62 percent of British respondents did in fact state that they did not feel European. Since we attributed the missing values, this could distort the validity of the results for the Great Britain. However, in order to test for this we ran some of the analyses shown below first with, and then without, imputation for the British case, and obtained virtually the same results on both occasions. Therefore, even in the British case we opted to attribute the missing values, as doing so does not seriously disrupt the analysis.
One shortcoming of our analysis is the fact that it is essentially static, while both individual and collective identities are in fact dynamic, changing over time. In the light of this, we have opted to compare and analyse the differences between countries. This kind of analysis can be considered dynamic to the extent that it shows differences among countries. Nonetheless, the reader should remain aware that the data simply constitutes a snapshot in time and depicts identities as being more stable than they really are.(10)
7
The percentage of people expressing a sense of attachment to their respective nations remains quite high, even if we also find evidence of an emerging European identity. If we compare the attachment to both identities, it would appear that national loyalties have yet to be transferred to this new European identity.
Let us begin our analysis by considering whether there has been any transfer of national identities to the supranational or European level as the hypotheses derived from the “cultural” (number four) and “instrumental” (number four) theories would suggest. The answer to this question can be gleaned from the results given in Tables 2 and 3, which shows the percentage of people who feel close to the nation (Table 2) and Europe (Table 3).(11)
In the case of national identities (Table 2), only if we confine our analysis to the figures of those who feel “very close” to the nation can we speak in terms of an erosion of national identities between 1995 and 2002.(12) If we consider those who feel both “close” and “very close”, we find no evidence of such a tendency; when both responses are included, national identities have weakened only in Germany (East), Austria and the Czech Republic, but even in these countries the vast majority of the population remains attached the nation, with figures ranging from 71.8 percent in the Czech Republic to 81.3 percent in Austria.
Even more obviously, whatever the extent of the weakening of national identities, loyalties have not been transferred from the nation to Europe (Carey 2002: 388). The common pattern found across all the countries in the study is a decline in feelings of closeness to Europe (table 3). This is the case whether we consider the percentage of the “very close” or the percentage of both the “very close” plus the “close”.(13) This suggests that other factors must be responsible for the weakening of political attachments at both the national and European levels; the nature of these, however, is a topic which lies beyond the scope of this article.
Regarding the hypotheses derived from the three theories above, the data in these tables do not confirm that the emergence of European identities weakens national identities as implied by the “cultural” and “instrumental” theory, both of which present attachment to different levels of identity as zero-sum games (as if a person has a fixed quantity of identity to be split among different levels of attachment or kinds of identities, so which ever quantity of identity is put in one level or identity must be subtracted from the other -hypotheses number four in the “cultural” and “instrumental” theories-). Although these data are too broad to confirm hypothesis number two on the “civic” theory, they tend to back the idea that national and European identities are compatible. However, at this stage we cannot say if this is so because the European identity is mainly based on “civic” elements as the “civic” theory suggest (hypothesis number two).
8
Regardless of whether national and European identities have increased or decreased in recent years, we are still left with the task of ascertaining how they are defined and whether or not they are compatible. Our findings suggest that both national and European identities are inclusive; that is, we find a relatively high proportion of respondents with dual identities (ie. citizens who feel close to both their nation and Europe at the same time) in all the countries. The two identities are inclusive (or compatible), but not of equal intensity. Attachment to national identities is still stronger than attachment to European identities, even in those countries where citizens with dual identities outnumber those with exclusively national identity. Hence, these findings confirm that, as suggested above, the emergence of a European identity does not threaten attachment to national identities.
While we have little empirical data on the compatibility of national and European identities, it does seem that there is some variation among countries in this regard: national and European identities seem to be less compatible in the United Kingdom and Sweden than in France, Luxembourg and Italy (Goldmann 2002: 296).(14) Goldmann and Gilland (2001) relate this finding to the differing importance of what they call the “civic” as opposed to the “ethnic” dimension of identities in those countries. Our own empirical research has produced similar conclusions in some, but not all respects.
Let us consider the question of the compatibility (inclusiveness) or incompatibility (exclusiveness) of national and European identities. To that end we have calculated, for each country, the proportion of the population expressing simultaneous attachment (closeness) to both the nation and Europe, as well as various measures of the strengh of their attachments to each of these identities. These data are included in Table 4.
The first point to be highlighted from this table is that national and European identities do not, in fact, seem to be incompatible (that is, exclusive). The percentage of respondents expressing dual identities (those who feel close to both their nation and Europe, shown in the first column) is quite high in all the countries. Nonetheless, this figure ranges from 36 and 40 percent of the total in Great Britain and Greece, respectively, to 64 and 61 percent in Italy and Spain. Looking at the same data from a different perspective (subtracting the percentage of people with exclusively national identities from the percentage of dual identities holders, shown in the second column), exclusively national identities are still held by a significant number of citizens of Great Britain and Greece, above all, but are also common in Germany (East and West) and the Czech Republic.(15) Therefore, from the first two columns in Table 4 we could conclude that national and European identities are compatible to some extend in all countries (rejecting hypotheses number four from the “cultural” and “instrumental” theories, and backing hypothesis number two from the “civic” theory). But the second column does also pose a question: is the variation in the net percentage of population with dual identities caused by the different importance of “cultural”, “instrumental” and “civic” elements of attachment in each country? We shall answer this last question in the following section. But before, and equally significantly, even in those countries where more citizens have dual than exclusively national identities, attachment to national identities is still stronger than attachment to European identities (the median value of subtracting median identification with the EU from median identification with the nation is always favourable to the nation, shown in the third column).(16) In other words, there is no correlation between a higher proportion of the population with dual identities and a relatively lower attachment to the nation. This finding tends to bear out the theory of concentric circles of attachments to different identities. The nation would be the inner, more substantial/solid/stronger circle of identification, while Europe would constitute a second, outer circle of identification.
9
In terms of the hypotheses outlined before, the data would back partially hypothesis number three from the “cultural” theory (European identities are weak in comparison to national identities, although we cannot yet assure if that would the consequence of the European identity lacking “cultural” foundations) and hypothesis number two from the “civic” theory (European and national identities are compatible, although we cannot yet assure, either, if that would be consequence of the European identity being based in “civic” elements). However, the evidence in the table would reject hypothesis number four and five from the “cultural” theory and hypothesis number four from the “instrumental” theory, since the emergence of a European identity does not weaken the strong attachment to the nation. In fact, stronger attachment to the nation is a feature common to all the countries (see that all the figures in the third column in Table 4 are positive). Furthermore, the strength of the bonds to the nation is similar in Great Britain (0.758), where exclusive national identities are predominant, and in Spain (0.718), where dual identities holders outnumber exclusive national identity holders. The same is true regarding West Germany (0.715) and the Czech Republic (0.748), or Greece (1.099) and Hungary (1.043).
On this basis, and substantiating our earlier point regarding the absence of any transfer of loyalties from the nation to the supranational or European level, we can confirm that the emergence of a European identity does not threaten attachment to national identities. In fact the proportion of the population holding exclusively European identities (feeling “very close” or “quite close” to Europe but “not very close” or “not at all close” to their nation) is quite low, ranging from between 9 and 8 percent in Great Britain and West Germany and 1 percent in Hungary and Poland.
Having established the apparent compatibility of national and European identities, we should ask whether this compatibility is a result of the two identities having different meanings? Grosso modo, we have found that the nation is mainly defined in “cultural” terms, in contrast to attachment to Europe, which draws above all on “instrumental” considerations. This distinction may help account for their compatibility and, on the other hand, explain why loyalties are not transferred from the national to the supranational level. However, these different meanings are not entirely consistent across countries, in that “cultural” factors appear to be more important for citizens in the Central and, above all, Eastern European countries, while “instrumental” considerations are more important for citizens in Western and Southern member states. The cases of Hungary and the Czech Republic, in particular, suggest that even if attachments to multiple identities derive from the same “cultural” sources, national and European identities may still be compatible. Finally, we find that it is the “affective-symbolic” attachment to the nation that seems to hinder the development of a European identity, while “civic-based” attachment to both the nation and Europe favours the inclusiveness of these two identities.
Table 5 lists the five items identified as the most important grounds for national and European identification in each country. We have included some data from Table 4 which should also be taken into account in this context (net percentage of dual identities holders and comparative attachment to national and European identities). This table makes it possible to begin exploring the dimensions of identification, and hence to gain a deeper understanding of the compatibility or otherwise of national and European identities.(17)
10
A first, vague impression is that attachment to the nation is largely defined in “cultural” terms (language and culture happens to be mentioned among the five most important items for national identity in virtually all the countries), while attachment to Europe is mainly based on “instrumental” considerations (the common currency -economy- and the right of free movement and residence is a recurrent element among the five most important items of European identity in virtually all countries). This is even more clearly shown in Table 6 and 7, which are simpler summaries from table 5 to make it more easily understandable.
Table 6 shows the five items that were most frequently mentioned among the five most important for national identification (we ranked the items according to their average importance and selected the five highest averages in each country). Some of these items were selected in all countries, while others were only relevant for a few or even a single country. The table also shows the number of times the item ranked first in the list of items. All countries mentioned a common language among the five most important elements for national identification, eight mentioned common culture, customs and traditions and common ancestry and seven referred to a common history and a common destiny. All the first-placed items mentioned correspond to the “cultural” theory of identities. Therefore, we can conclude that national identity is largely defined in “cultural” (or ethno-cultural) terms, therefore baking strongly hypothesis number one of the “cultural” theory (national identities are mainly based on “cultural” elements). Table 7 provides the same information, but in this case for European identities. The right to freedom of movement and residence in any part of the EU and a single currency were mentioned in nine and eight countries respectively. Other items such as language and culture or civilisation were also identified as important. That is, the most frequently mentioned items include those that belong to the “instrumental” conception of identity, while others corresponding to the “cultural” theory are also mentioned. This finding supports partially the hypothesis number two of the “instrumental” theory, since European identities seem to be based mainly in instrumental considerations. Nonetheless, the role played by “cultural” elements cannot be considered minor, especially in Central and Eastern European member states, where they rank higher than instrumental considerations. However, the important finding here, if we take the results shown in Tables 6 and 7 together, is the varying degrees of emphasis placed on instrumental items. In fact, this constitutes the main difference between national- and European-level identities. This plainly reject hypothesis number one from the “instrumental” theory (European and national identities are both based on “instrumental” considerations) and the “civic” theory (European identities are based mainly in “civic” elements). The evidence is debatable regarding hypothesis two from the “cultural” theory, since “cultural” elements seem to play also a role in European identities together with “instrumental” considerations. These findings might strengthen the hypothesis that it is possible to have attachments to multiple identities as long as such attachments derive from different sources. That is, national and European identities would be compatible insofar as attachment to the nation is based on “cultural” elements, and to Europe on “instrumental” considerations.
While this pattern applies in all the countries, we find that the two identities are more compatible in some countries than in others. This can be seen coming back to Table 5 and introducing in the analysis the information extracted from Table 4.
11
Looking at the net percentage of the population with dual identities,(18) it can be seen that it varies in different countries, even when they share similar sources of national (“cultural” elements) and European identification (“instrumental” elements). Take Great Britain and Italy as an example, “cultural” elements are important for the national identity of citizens in both countries (both have language and culture among the five most important items for national identification), while “instrumental” elements are equally important in both countries for the European identity of their citizens (common currency and the right of free movement and residence are mentioned among the five most important items for European identity), however exclusive national identity holders outnumber dual identities holders in Great Britain by 28 points, while in Italy the situation is exactly the contrary: dual identity holders outnumber exclusively national identity holders by exactly 28 points.
Moreover, Eastern European countries (and Central European countries to a certain extent) diverge from the general pattern described here, in that “cultural” (as opposed “instrumental”) considerations are important both for their European and national identifications.(19) The cases of Hungary and the Czech Republic, in particular, suggest that even if attachments to multiple identities derive from the same “cultural” sources, national and European identities may still be compatible, therefore opposing hypotheses number two and five from the “cultural” theory (that is, European identities would no be based on “cultural” elements, and citizens with strong national identities would not develop European identities). Furthermore, if we take into account that in both countries dual identity holders (as measured by the net inclusiveness) outnumber exclusive national identity holders by eight points and that the bonds to the nation (as measured by the comparative attachment) are still stronger that the attachment to Europe we can also reject the hypothesis number four from the “cultural” theory. That is, the emergence of a European identity based in “cultural” elements similar to those of national identity does no endanger or weaken national identities and loyalties.
What, therefore, makes identities compatible or incompatible? We can rule out differences in the relative importance of “cultural” elements of attachment to the nation, an explanation suggested by Goldmann and Gilland (2001), since this is a pattern which is more or less common to all the countries while the net percentage of dual identity holders varies considerably among them. Accordingly, one way of attempting to answer this question is by focusing on the countries with the highest and lowest level of compatibility between the two types of identities. The countries where identities are least compatible are Great Britain and Greece, while the countries in which the two identities are most compatible are Italy and Spain. What have each of these two pairs of countries in common which makes them different from the rest? Great Britain and Greece are the only two countries where pride is mentioned among the five most important items for national identification. Great Britain is the only country that mentions sovereignty, and where we find the lowest level of compatibility between national and European identification. Italy and Spain are the only two countries in which rights and duties figure among the five most important items for both national and European identification. This suggests that a strong feeling of national pride and attachment to national sovereignty may hinder the development of a European identity, but it does not presupose that the national identity in those countries is more “culturally” based. National pride and attachment to national sovereignty may also be linked to other economic, cultural or even sporting considerations. That is, this finding does not provide support for hypothesis number five from the “cultural” theory.
12
Furthermore, the relation between national pride and European identities is not explored in any of the three theories posed above. However there might be some interesting correlations: a recent investigation by Belot (2003) points exactly in the same direction regarding the importance of national pride (how it is constructed and the elements it lays on) to explain the European identity of young British and French people. Finally, our findings in this last part of the section back the hypothesis number two from the “civic” theory, since in those countries (Spain and Portugal) that give importance to civic elements both at the national and European identity level the measure of compatibility is the highest among the countries studied.(20)
Above we raised the question of the consistency across countries of the different meanings of national and European identities. In this section, however, we will examine the difference between countries in a little more detail. In the previous section we focused on the most important elements for national and European identification. However, what is the relative importance of these items when compared to other elements of identification? Which emerge as the most important components in each country when all of them are analysed at the same time? To answer these questions we first ran optimal scaling with each of the variables related to identities included in the Standard Eurobarometer 57.2. Then, in line with the theoretical concepts discussed in the previous section and confirmed by factor analysis, we constructed a group of parcels representing the indicated aspects of the attitudes under consideration. We identified subgroups and then computed categories by totalling individual items. Based on these indexes, we created simpler indicators based on thirdtiles.(21) We then ran optimal scaling using these parcels, thereby synthesising the information in order to cast light on the various debates about national and European identities.
In Figure 1 we have plotted countries and closeness to different in-groups and out-groups (either internal or external). We see a strong dominant, but trivial, first dimension (see Table 8) that mainly distinguishes between overall patterns of sentiments of strong (in the right hand side) versus mild (in the left hand side) closeness to different groups. On this first dimension, Hungarians show the strongest tendency to feel close to all groups, while, at the other extreme, citizens of the UK and Greece show weaker feelings of this kind. The second dimension is more substantial, although weaker.
This second dimension shows variation in attitudes toward groups and also reveals interesting differences between items and countries. We see that closeness to village, region and nation are quite strongly correlated and cluster together in the upper extreme of the dimension; at the bottom extreme we find closeness to out-groups. Closeness to European groups falls in between these two factors. From this distribution of the item it can be concluded that national and European identities go more or less together, since they are correlated. That is, this analysis again confirms that both kinds of identities are more or less compatible.(22) Closeness to out-group and national in-groups, in contrast, show a much weaker correlation. On this second dimension, we find that Hungary and Spain are the two countries in which the distinction between closeness to national in-groups and out-groups is strongest. Spaniards and Hungarians are also more likely to feel close to national in-groups in comparison to other countries.
13
At the other extreme we can see that Great Britain is the country in which citizens are most likely not to differentiate between national in-groups and European or out-groups, with, as mentioned above, weak feelings of closeness to all categories.(23) Other countries fall somewhere in the middle of this second dimension.
In Figure 2 we calculated the following parcels of items: closeness to village, region and nation went into one factor termed natin (closeness to national in-groups); closeness to the EU citizens, Europeans fellows and people living in Central and Eastern Europe were grouped together and labelled Europe (closeness to European groups); finally we computed all other items as components of a third factor named eurout (closeness to out-groups). A + sign indicates strong closeness (it was assigned to those in the top thirdtile), and a – sign indicates mild closeness (it was assigned to those not belonging to the bottom thirdtile). As can be seen in the graph showing these parcels, this analysis does not modify the results of the analysis, reproducing the same distribution of the countries on the different dimensions.
In Figure 3 we have included our countries and the items from the battery of questions dealing with the components of national identity. As in the previous graphs, we find a first dominant but trivial dimension, simply reflecting the overall pattern of strong (in the right hand side) or mild (in the left hand side) expressions of agreement with the different items. This first dimension runs from the strongest agreement with all items on the left hand side to milder agreement on the right hand. This highlights in particular the differences between Greece and Poland, with the strongest overall agreement, on the one hand, and Germany and Great Britain on the other. The first dimension accounts for substantially more variance than the second, but the latter distinguishes more clearly among countries (see Table 9).
This second dimension groups the elements of national identification along the axis, on which indicators measuring ethnic-cultural aspects cluster together at the top. At the bottom we find another cluster of civic-instrumental indicators, while the symbolic and affective elements are located towards the middle. The distribution of countries along this second dimension mainly serves to differentiate respondents in Hungary and the Czech Republic, who are more likely to accentuate “cultural” backgrounds in their national identification, from those in Spain, who are more likely to highlight “instrumental” considerations as foundations for national identity.
Based on this loading of items(24) and the theories about components of national identities outlined above, we have calculated packages of items that summarise the information. These have been plotted in a new graph. Common culture, customs and traditions, common language, common ancestry and common history and destiny were classified as ethno-cultural components of national identification and labelled as nethnic. A national army, common borders and national independence and sovereignty were grouped together as reflecting a sovereignty-based conception of national identities; we labelled this nsouver. A third bundle, termed nsymbol, comprised feelings of national pride, national character and national symbols. Finally, elements pertaining to an instrumental and civic understanding of national identification were combined together and labelled nintrum. This package includes: common political and legal system; common rights and duties; a common welfare system; and national economy. The result of the analysis is shown in Figure 4.
14
The first dimension reflects overall patterns of agreement with parcels of national identification; the second, more substantial dimension, distinguishes between ethno-cultural elements of national identification and all the other packages. On this dimension Hungary is still the country whose citizens are most likely to stress ethno-cultural elements of identification with the nation; towards the bottom of the dimension we find the Czech Republic and Germany, which are at a considerable distance from Hungary. At the top of the axis we found that Great Britain, Spain, Poland, Greece, Austria and Italy are all more driven by non-ethno-cultural aspects of national identification. Since most of the countries fall more or less in the middle of the second axis we can conclude that both ethnic and non-ethnic elements are important for national identification in most countries.
Figure 5 shows our countries of analysis together with the battery of questions dealing with the configuration of European identities. We can see that the first dimension reproduces what has been already seen in earlier graphs. This dimension represents the overall tendency for strong (in the right hand side) as opposed to mild (in the left hand side) agreement with all the items. As noted above, this dimension is strongest and dominates the second (see Table 10). It serves to distinguish between Germany and Great Britain, with the weakest agreement with all the items of European identification, at in the left extreme, from newcomers (Eastern European Countries, all in the right hand side of the dimension; the only exception is Austria), which are more likely to agree with European identification as such overall items.
The second dimension, although weaker, is more useful to classify countries. On the vertical axis we can see the ethno-cultural elements grouped together at the top. All other elements of European identification form a second cluster toward the middle of the second dimension. We could label this ethno-cultural versus non ethno-cultural elements of European identification. The location of countries along this dimension mainly highlights that newcomers (Eastern European countries) are more prone to agree with the importance of ethno-cultural elements for European identification. At the other extreme we find Italy and Spain, whose citizens largely stress instrumental reasons for their European identification. Since Italy is characterised by marked internal, North-South economic inequality (north/south), and because of the many scandals involving Italian politicians, Italian citizens attach more importance to instrumental aspects of European identity, perhaps in the hope that membership of the EU will improve their economic and political situation. Greece appears in an outlying position (quite far from any other country or group of countries). This mainly appears to reflect the lack of importance of culture as a component of Greeks’ European identification, explained by the fact that Greeks consider that culture pertains to the realm of their national identities.
Just as we did for national identification, we have created bundles of elements of European identification and included them in a new graph (Figure 6); these bundles have been compiled in accordance with the loading of items as seen in the graph above(25) and the theories about components of European identities outlined earlier in this paper.
15
The first parcel includes ethno-cultural components: common civilisation, membership of a European society with many languages and cultures, common ancestry and common history and destiny. This has been labelled eethic. We grouped the EU institutions and emerging common political and legal systems, common rights and duties, common system of social protection, the right to free movement and residence in any part of the EU and the single European currency as part of a second package named einstrum, which comprises both instrumental and civic grounds for identifying with Europe. An emerging EU defence system, common borders and sovereignty of the EU formed a third factor, bringing together the sovereignty elements of European identification. Finally a feeling of pride in being European and a set of EU symbols were combined together as elements of pride in European identification (esymbol).
Optimal scaling reproduces the same dimensions we saw before. Nevertheless, the position of countries is rather different on the second substantive dimension. Symbolic elements appear to be more important for the definition of Europe (either in positive or negative terms) in Poland, Italy and Great Britain than in the other countries.
In Figure 7 we try to summarise the analysis in terms of the theoretical dimensions or parcels of items. For that purpose we have combined all the dimensions and countries together. As a result, in this graph we can see similarities as well as differences among countries; these can be thought of as configuring individual country profiles. Greece and Britain have quite unique profiles, which are, moreover, the direct opposite to each other. The British are reluctant to agree with the question, feel less distant to out-groups, and are the least euro-enthusiastic citizens of all the countries studied; but they also feel distant from national symbolic values, ethno-cultural items, etc., generally showing less enthusiasm for everything. The Greeks, on the other hand, are more enthusiastic about all items and also more inclined towards elements of national identity. Germans and Austrians tend to form a cluster, which gradually expands to incorporate other countries in Central and Southern Europe, characterised by a middle position between national and European identification.
Finally, a correspondence analysis of all items and countries also produces interesting results, although the dimensions are more difficult to see and explain (figure 8). In this graph the vertical axis reflects national instrumental versus national symbolic identifications. The horizontal axis shows the difference between European identification based on culture and civilisation (“cultural” theory) versus instrumental identification. To better capture the different countries’ profiles, we will focus on the contrast between countries on the extremes of the graph. The variables which are mainly responsible for the differences among countries are (1) closeness to Central and Eastern Europe and (2) common civilisation and culture for the feeling of being European, both of which are most important for the Eastern and Central European countries. Citizens in current member states, on the other hand, feel European for largely instrumental reasons. Spain, Germany and Austria identify with Europe primarily on civic and institutional grounds. Greece constitutes a somewhat exceptional case. Greeks attach considerable importance to symbolic aspects of both their national and European identifies. Towards the bottom right of the graph and opening out to the left we find the symbolic and sovereignty dimensions of national and European identification. Symbols of national identification and pride are important for Greeks above all. Great Britain is close to Greece, although British respondents are more prone to mention instrumental components of European identity.
16
At the beginning of this article we discussed a number of theories and hypotheses regarding the possible sources of European identification and the relation between European and national identities. We will now discuss the plausibility and validity of these theories in the light of the empirical data and analysis presented above.
The “cultural” theory, and the hypotheses we derived from it, finds only partial support from our analysis. Regarding the first hypothesis, our data confirm that attachment to national identities is based on “cultural” elements in all the countries analysed (Tables 5 and 6). However, we must reject hypothesis number two, regarding the low importance of “cultural” elements for the emergence and development of an European identity. “Cultural” elements are not absent from the notion of European identity, and are mentioned among the five most important items in most of the countries analysed (in Table 7, “language” is mentioned by nine out of ten countries, and a “common European civilization” by seven of them). We have found also some support for hypothesis number three, that is, attachment to the nation remains stronger than attachment to Europe (Table 4, third column). However, European identities are not weaker than national identities because they lack “cultural” basis (as just mentioned above, and as part of hypothesis number three suggests). Our findings does not support hypothesis four either, since we observe countries (Hungary and the Czech Republic, in Table 5) that base both their national and European identity on “cultural” elements, in which dual identities holders outnumber exclusive national identity holders, and in which the bonds to the nation, though, continue to be stronger than bonds to Europe. Finally, regarding the hypothesis number five, the fact that the “cultural” strong attachment to national identities (shown in Table 4) has not prevented the emergence, albeit to a greater or lesser extent, of European identities in all the countries analysed (Tables 3 and 4) does not support it.
Summarizing, it should be remembered that authors such as Smith (1999) and Østerud (1999) hypothesised that it would be extremely difficult for a European identity to emerge, given a) the strength of national “cultural” identities and b) the simultaneous lack of European “cultural” elements shared by all Europeans. However, not only has a European identity emerged in all countries, as measured by the percentage of dual identity holders (in Table 4), but in most of them this European identity does also include “cultural” elements (among the five most relevant), in a similar vein to national identities (Tables 5 and 7). Hence, on the one hand we find that national identities are still stronger and primarily “cultural”; but on the other hand we find that “cultural” attachment to a European identity is also relevant (above all in Eastern and Central European countries, but also in other members states). So there is also much more common “cultural” ground among the European countries on which to built up a European identity than this theory would suggest.(26)
As for the “instrumental” theory of identity, here our findings have again provided only partial confirmation for it, and then only at the European level. Regarding the first hypothesis, that both national and European identities are based on instrumental consideration, we have to reject it partially. It is true that European identities are primarily “instrumental”, but this is not the case with national identities, which are mainly “cultural”, as we just said above (and can be gathered from Tables 5, 6 and 7).
17
The second hypothesis is also partially rejected, since European identities are not merely “instrumental”. We have seen that “cultural” elements also have a place among the most important elements that define an European identity (Tables 5 and 7). Again, it is not true that citizens with weaker national identities develop stronger European identities (hypothesis number three) or that in those countries with the higher percentage of citizens holding dual, or European, identities holders the attachment to the nation is weaker than in countries with higher percentage of exclusive national identity holders (hypothesis number four). It is clear from the data in Table 4 that the bonds to the nation have remained stronger than the bonds to Europe even in those countries where the large majority of citizens are dual identities holders. Therefore, that a citizens develop an European identity does not weaken his or her attachment to the nation.
In short, it must be remembered that, in this theoretical approach, people are seen as ration actors who would choose between a national or European identity depending on their perceived costs and benefits. However, we have seen that, in broad terms, national identities are mainly based on “cultural” components, as opposed to the “instrumental” considerations that are more relevant for European identity. This does not mean that “instrumental” elements are of no importance for national identities, as exemplified by the cases of Italy or Spain. However, it does show that these “instrumental” considerations are much more important at the European level. An important inference from this finding is that it would appear unlikely that citizens will transfer their loyalties from the nation to Europe based solely on “instrumental” considerations, either at the national or European level, since their attachment to each identity is based on different considerations. It is more likely that the percentage of dual identities might increase in line with the perceptions of the effective functioning of European institutions, but without eroding citizens’ identification with their nation.
The fact that European identities are based mainly on “instrumental” considerations has further important implications. It may favour the possibility of the EU being able to create European identities by intensifying the perceived (economic or political) benefits of membership. This is the vision defended by advocates of the “instrumental” theory. However, it should be noted that in countries which stand out for their strong sense of national pride, such as Greece or Great Britain, European identification might actually weaken as the perception that the EU is working effectively intensifies. Although this hypothesis has yet to be tested, we have found that in these two countries, both of which mentioned pride among the five most important components of national identification, identities are less compatible even though, as in other countries, “instrumental” considerations were mentioned as grounds for European identification. Accordingly, the perception that the EU performs better than the nation state could be perceived as a threat to citizens’ national pride (in order for national pride to remain high citizens must believe that their own country functions better than the EU).
In terms of the “civic” theory of identification, our analysis suggests that we are unlikely to see the emergence of a European identity based primarily on civic considerations. We have to reject the first hypothesis, regarding an European “civic” identity. According to our data, only in three out of nine countries did rights and duties figure among the five most important items for citizens’ European identifications (Table 7). Nevertheless the cases of Italy and Spain tend to back the hypothesis number two, that is, those two countries show that the coexistence of “civic” elements of attachment both to the nation and Europe favours the compatibility of national and European identities (Table 5).
18
The implication, from this second finding, is that it might be possible for European identities (dual identities) to spread even in those countries with strong national pride, increasing the perception of shared norms and values among all Europeans and particularly among nationals in each country and Europeans. It also back the idea that is will be possible to construct a strong civic European demos in the future.
Apart from discussing these theories, we have also considered a number of other questions. First, the different analyses presented here clearly show that national and European identities are compatible. This finding is not particularly surprising, but a number of further points should be highlighted. First, and as mentioned above, our data does not support the idea of a transfer of identity from the national to the European level. We hypothesised that these two identities are compatible because they are of different order and endowed with different meanings. Different statistical analyses have confirmed that this is indeed the case: attachment to national identity is largely “cultural”, while attachment to a European identity is primarily “instrumental”. Nonetheless, while this is a pattern common to all countries, the level of compatibility of the two kinds of identities varies considerably. Our hypothesis on this point it that the particular configuration of national identities in each country gives rise to distinctive dynamics and historical inertias that have an impact on the emergence and configuration of a European identity. Although we do no have the time or the space to develop this argument here, we will mention our four “extreme” cases-study. In the case of Great Britain and Greece, the way they have constructed their national identity based on strong feelings of national pride that relate to their role as world powers in the past may create a sense of “humiliation” in the need to cooperate with others (which in some cases had been ancestral enemies) to keep an status, this need being a constant remainder that they have ceased to be the great powers (political, economic or cultural) that they were. In the case of Spain and Italy, their contemporary national identity was projected to the future, into the EU, not toward the past. Among other factors, in Spain this can be attributed to the experience of the Civil War and the dictatorship (the same might well be true for the case of Germany); in Italy, to the corruption problems that swept the country in the 1990’s. Having troubles to identify with their immediate past, they turned to their integration into the EU as a reason for national pride. Therefore, in these cases integration is a remainder that they can do better.
We were also interested in how these two identities relate to each other. Are they of the same order or not? Our analysis seems to suggest that they are perceived as being of a different level, national identity constituting an inner circle and European identity an outer circle.
As a general conclusion, linking all the findings above, we could conclude that national and European identity are compatible because they are seen as identities of a different level, bearing different meanings. For advocates of more Europe, and for those European politicians interested in forging a European identity to serve as one of the legitimatising foundations of the EU, this finding could have both positive and negative implications. The good news it that the EU could swell the ranks of the citizens with dual identity by further strengthening the performance of the European institutions and the benefits they bring (or rather the public’s perception of both).(27) The fact that Europeans continue to feel, primarily and in the first place, nationals of their respective countries need not be bad news. In fact, if this point were made clear to national governments it might facilitate the emergence of the type of European identity that is currently resisted by some member states, concerned that it might erode their sovereignty (and the loyalty of their citizens).
19
As a final consideration related to the methodology used here, we have confirmed that national and European identities are realities of a different order and bear different meanings. It is mistaken, therefore, to address these two identities in public opinion polls through the same kind of questions, or even mixing both identities in the same questions as if they were alternatives of the same nature, making interviewers artificially select one among a group of possible identities. Questions of the type: in the future do you see yourself as more Spanish than European or more European than Spanish, are completely mistaken as this research should have shown.
Belot, C. 2003. “‘We Europeans?’ from European political community to a European demos, from a European identity to national identities and backwards: some new developments towards assessing EU social legitimacy”. Proceeding of the ECPR Joint Sessions of Workshops (Edinburgh 2003).
Brass, P. 1979. “Elite groups, symbol manipulation and ethnic identity among the Muslim of South Asia”, in D. Taylo and M. Yapp (eds.), Political Identity of South Asia. London: Curzon Press: 35-37.
Carey, S. 2002. “Undivided Loyalties. Is National Identity an Obstacle to European Integration?”. European Union Politics, 3 (4): 387-413.
Cinnirella, M. 1997. “Towards a European identity? Interactions between the national and European social identities manifested by university students in Britain and Italy”. British Journal of Social Psychology, 36 (1): 19-31.
Cerutti, F. 1992. “Can There Be a SupraNational Identity?”. Philosophy and Social Criticism, 18 (2): 147-62.
Cowles, M. G., J. Caparaso and T. Risse, eds. 2001. Transforming Europe. Europeanization and domestic change . Ithaca, New York: Cornell University Press.
Deflem, S. and F. C. Pampel. 1996. “The Myth of Postnational Identity: Popular Support for European Unification”. Social Forces 75(1): 119-143.
Easton D. 1965. A Systems Analysis of Political Life. New York: John Wiley&Sons
Eichnberg, R. C. and R. J. Dalton. 1993. “European and the European Community: the Dynamics of Public Support for European Integration”. International Organisation 47: 507-534.
Eriksen, E. O. and J. E. Fossum. 2001. “Democracy through Strong Publics in the European Union?”. ARENA Working Paper 01/16. http://www.sv.uio.no/arena/publications/wp01_16.htm
Fernández-Albertos, J. and I. Sánchez-Cuenca. 2001. “Factores políticos y económicos en el apoyo a la integración Europea”. Unpublished manuscript (Fundación Juan March, Madrid).
Gabel, M. J. 1998. Interest and the Integration. Market Liberalization, Public Opinion, and European Union. Ann Arbor: University of Michigan Press.
Gabel, M. J. and H. D. Palmer. 1995. “Understanding Variation in Public Support for the European Integration”. European Journal of Political Research 27: 3-19.
Gabel, M. and G. Whitten. 1997. “Economic Conditions, Economic Perceptions, and Public Support for European Integration”. Political Behaviour 19(1): 81-96.
Giesen, G. y T. Risse. 2000. “When Europe hits home: europeanization and national public discourses”. Unpublished research proposal.
Goldmann, K. 1998. Skilda identiteter gemensam demokrati I EU ett demokratiprojekt? Stockholm: Fritzes.
Goldmann, K. 2002. “Internationalisation and the nation-state: Four issues and three non-issues”. European Journal of Political Research 41: 281-305.
Goldmann, K and K. Gilland, eds. 2001. Nationality and internationalization: The national view of the nation in four EU countries. Stockholm: Department of Political Science, Stockholm University.
Höjelid, S. 2001. “European Integration and the Idea of European Identity: Obstacles and Possibilites”. ECPR Joint Sessions Grenoble 2001. Workshop 19: Identity Politics.
Kaltenthaler, K. C. and C. Anderson. 2001. “European and Their Money: Explaining Public Support for the European Currency”. European Journal of Political Research 40(2): 347-361.
Kersbergen, K. V. 1997. “Double Allegiance in European Integration: Publics, Nation-States, and Social Policy”. European University Institute Working Paper 97/15. http://www.iue.it/RSCAS/WP-Texts/97_15.html
Mancini, G. F. 1998. “Europe: The Case for Statehood”. Harvard Jean Monnet Working Paper 6/98. http://www.jeanmonnetprogram.org/papers/98/98-6-.html
Millward, A. S. (1992). The European Rescue of the Nation State. London: Routledge.
Moravscik, A (1998) The Choice for Europe: Social Purpose and State power from Messina to Maastricht. Cornell University Press.
Olsen, J. P. 1996. “Europeanization and Nation-State Dynamics”. ARENA s æ rtykk 96/3: 245-285.
Orchard, V. 2002. “Culture as Opposed to What” Cultural Belonging in the Context of National and European Identity”. European Journal of Social Theory, 5: 419-433.
Risse, T. 2001. “A European identity? Europeanization and the Evolution of Nation-State Identities”. In Cowles, Maria Green, James Caparaso and Tomas Risse (eds), Transforming Europe. Europeanization and Domestic Change. Ithaca, New York: Cornell University Press.
Sánchez-Cuenca, I. 2000. “The Political Basis of Support for European Integration”. European Union Politics 1(2): 147-171.
Smith, A. D. 1992. “National identity and the idea of European Unity”. International Affairs 68(1): 55-76.
Smith, A. D. 1995. Nations and Nationalism in a Global Era. Cambridge, UK: Blackwell Publishers.
Smith, A. D. 1999. Myths and Memories of the Nation. Oxford, UK: Oxford University Press.
Smith, T. W. and L. Jarkko. 2001. National Pride in Cross-National Perspective. National Opinion Research Center / University of Chicago, April 2001.
Stråth, B. 2002. “A European identity. To the Historical Limits of a Concept”. European Journal of Social Theory, 5: 387-401.
Weiler, J. H. H., U. Haltern and F. Mayer. 1995. “European Democracy and its Critique. Five Uneasy Pieces. European University Institute Working Paper 1995/11. http://www.iue.it/RSCAS/WP-Texts/95_11.pdf
Weiler, J. H. H. 1996. “The Selling of Europe: The Discourse of European Citizenship in the IGC 1996”. Harvard Jean Monnet Working Paper 3/96. http://www.jeanmonnetprogram.org/papers/96/9603.html
Weiler, J. H. H. 1999. The Constitution of Europe. “Do the New Clothes Have an Emperor?” and other essays on European Integration. Cambridge: Cambridge University Press.
(*) A first draft of this article was presented at the ECPR Joint Sessions of Workshops held in Edinburgh from 28 March to 2 April 2003: Workshop 2, “European culture(s) and European Integration”. We would like to thank the participants in that workshop, José I. Torreblanca and the anonymous reviewers of EIoP for their valuable comments and criticism. We also thanks the European Commission V Framework Program for their financial support (EURONAT project, contract no. HPSE-CT2001-00044).
(1) See the review of existing studies produced by Geetha Garib and Michael Braum for the European project PIONEUR (Working paper 4 – July 2003). http://www.obets.ua.es/pioneur/documentos_public.php
(2) For an up-to-date survey and review of the main approaches, see the literature review by Stephen Gibson for the project YOUTH AND EUROPEAN IDENTITY. http://www.sociology.ed.ac.uk/youth/docs/Gibsons_lit_rev.pdf
(3) The authors in this group are more empirically driven than those from the other two schools discussed here. Moreover, some of them focus on the variable “support for the European integration”, which is quite different from a “European identity”. Nonetheless, these two variables can also be treated as measuring different aspect of the same reality: they are, in fact, highly correlated. It is, therefore, reasonable, to assume that any findings about the reasons for support for European integration may also tell us something about why people hold a European identity.
(4) The inclusion of the last three countries was exceptional, and in these cases the questionnaire only incorporated socio-demographic information and questions related to national and European identities.
(5) Further details on the samples are given in the Standard Eurobarometer 57.2 codebook (http://europa.eu.int/comm/public_opinion/; http://www.nsd.uib.no/cessda/europe.html )
(6) This was done before centralizing the data.
(7) We use the term “internal out-groups” to refer to national minorities which are considered to be different by the dominant national group. We define “external out-groups” as those groups coming from different countries, such as immigrants.
(8) These items could have also a significant affective-symbolic dimension.
(9) These items could also have an important affective-symbolic dimension. More particularly, a single European currency may have a strong symbolic value “because a country’s money is a symbol of its sovereignty. Support for EMU and the euro, provides, therefore, a crucial test case for whether and why European citizens may be willing to transfer power from the nation state to European institutions, and it has important implications for the future direction of institution-building within the European Union” (Kaltenthaler and Anderson 2001: 141).
(10) This is also a methodological limitation deriving from the techniques and kind of data used to measure identities through the administration of a single questionnaire. As social scientists we were confronted by the well-known trade-off between more detailed and sophisticated case studies or simplified comparison among countries.
(11) These two tables (2, 3)compare data from 1995 and 2002. The selection of year and survey has been forced by the fact that only the ISSP 1995 included questions on closeness to the nation and Europe similar enough to those in the Eurobarometer 57.2. However, the reader should be aware that small differences in the wording of questions existed regarding closeness to Europe. Nonetheless we the comparison is interesting enough to have included it here.
(12) It should be noted that the number of people feeling “very close” to the nation has increased in Germany and Spain.
(13) The only exception is if we compare the percentage of those very close and close to the Europe in Great Britain in 1995, and the percentage of those very close and close to the European Union in 2002.
(14) See also Deflem and Pampel (1996); Carey (2002) and Risse (2001)
(15) See the column labelled “net percentage of dual identities” in Table 4.
(16) See the column headed “comparative attachment to national and European identities”. Independently of the proportion of the population holding exclusively national or dual identities, this column measures the relative strength of attachment to both identities. It is the median value of subtracting median identification with the EU from median identification with the nation. A positive value indicates that attachment to the nation is stronger than attachment to the EU.
(17) This analysis does not, however, allow us to compare the profiles of different countries. We can see from the Table that ethno-cultural elements are important for national identification in all countries, but when the profiles are compared, taking into account all the elements of identification, it can be seen that the relative importance of this set of elements varies from one country to another. In the following section we present the results of an optimal scaling analysis. This graph shows that although ethno-cultural elements are important for all countries, as shown in Table 5, they are relatively more important in Hungary, the Czech Republic and East Germany than in Spain, Poland and Greece, for example. Hence, analysis based on these findings cannot substitute optimal scaling or correspondence analysis, since the latter reveal features of individual countries that the findings presented cannot.
(18) It should be remembered that this measure compares the percentage of dual identity holders (national and European at the same time) to the percentage of those holding exclusively national identities.
(19) Lack of experience of European institutions, and the different wording of questions (which were logically phrased in the future tense) may account for these differences. However, it should be noted that albeit to a lesser extent, Austria and Germany show a similar pattern, suggesting the existence of a structural rather than a circumstantial explanation for these variations.
(20) The importance given to “civic” elements depend, however, on internal dynamic in each country: among others, the memory of the past dictatorship in Spain, and corruption scandals in Italy.
(21) We use this term as analogous to the terms percentile, but dividing the population in three equal parts instead of one hundred parts. Those belonging to the top thirdtile group received a 1 into the category under consideration.
(22) The optimal scaling analysis that we have performed is based on correlation among items. Countries are placed as supplementary variables and have, therefore, no influence on the configuration of the space.
(23) The specific position of this country on the graph also reflects British closeness to North American citizens and the Commonwealth in comparison to European groups.
(24) Tested by factor analysis.
(25) Tested by factor analysis.
(26) This is also strongly supported by the findings of qualitative interviews carried on with lay people in those same countries.
(27) While the working of the European institutions is the same across countries the perception about them and the percentage of dual identities holders varies.