The Swiss: A Political Nation?

Authors


Abstract

Abstract: Proponents of constitutional patriotism advocate the promotion of a political national identity. Whether a national identity founded on common political values is viable, however, remains a matter of dispute. The Swiss nation is one of the few empirical cases cited by those who argue that a political national identity can be a sufficient guarantor of institutional stability and social coherence; indeed, Switzerland is widely regarded as a prototype of the political nation. The aim of our analysis is to empirically test this claim. In doing so, we rely on classical typologies of national identity. We propose to use a more finely graded typology than the usual dichotomies, however – one which allows for combined types – and to focus on individual perceptions of national identity at the mass- and elite-levels. The empirical analysis of whether the Swiss nation can correctly be classified as a political nation draws on data from mass and MP surveys. Our conclusion is that the Swiss imagine their nation in both political and cultural terms, and the essence of their cultural identity is Swiss linguistic pluralism and regional diversity.

Introduction

Switzerland is widely perceived as a prototypical political nation (e.g. Habermas 1994 [1990], 642; Shabani 2002, 426f.; Theiler 2004, 644). Swiss identity, common wisdom claims, is based on common appreciation of political values such as direct democracy, neutrality and federalism. Yet the country’s main dimension of political conflict in recent years has been ‘cultural demarcation’ (Kriesi et al. 2008). How is it possible in a prototypically political nation that the most significant political fights are about whether or not the nation should be culturally open? Our paper tries to solve this puzzle by proposing that the narrative of the Swiss as a political nation falls short. It neglects, we argue, the fact that the Swiss also imagine themselves in cultural terms.

The question of whether or not Switzerland is a political nation has important implications for ongoing debates about the prospect of a European nation and the future of national identity in immigration societies. At the centre of this debate stands the question of whether ‘constitutional patriotism’ is sufficient for state integrity and national solidarity (e.g. Habermas 1998a, 117) or whether a common cultural identity is necessary (e.g. Huntington 2004, 19). According to Habermas, in multi-cultural societies a political national identity that does not rely on a common language or ethnicity can provide an alternative means of national solidarity. A national identity based on nationally specific interpretations of constitutional principles can take the place originally occupied by cultural or ethnic nationalism (e.g. Cronin 2003).

Critics of this proposition question whether a purely political national identity can become the object of attachment and identification; they also ask whether nationally specific interpretations of constitutional principles do not presuppose a common national culture (see for instance the discussions in Markell 2000; Müller 2006; Baumeister 2007). This debate raises the empirical question of whether instances of purely political national identity in fact exist or whether identification with constitutional principles always occurs in tandem with attachment to elements of cultural or ethnic national identity. The case of Switzerland is central to this debate, as the Swiss nation is often cited as an example of a well-functioning political nation in a multicultural setting.

In order to make an empirical test of whether the Swiss nation can be considered political or not, one must rely on a distinction between political and other national identities. The classical distinctions between political and cultural (as well as civic and ethnic) nations have recently been criticised on several grounds. In our theoretical chapter we will elaborate on these criticisms, and explain how we have adapted the classical typologies in such a way as to make the categories less ambiguous and to allow us to conceptualise various imaginations of national identity within the same nation. We conclude that, with these amendments, the classical typologies remain useful in empirically testing claims about Swiss national identity.

Empirically, this paper contributes to the literature on Swiss national identity in two ways. First, we complement the existing empirical evidence with analysis of survey data. Most studies in this field have analysed the emergence of Swiss national identity using historical analysis (e.g. Kohn 1956; Deutsch 1976; Froidevaux 1997; Stojanovic 2003). The few empirical studies that rely on representative surveys to appraise popular conceptions of national identity take the political character of Swiss national identity as given and use national identity as an independent variable to explain electoral results or attitudes toward European integration (e.g. Kriesi 2002; Christin and Trechsel 2002). A qualitative analysis of interviews with Swiss local politicians has shown that decision-makers are proud not only of Swiss political characteristics, e.g. ‘direct democracy’, but also of cultural aspects of national identity such as ‘cultural diversity’ (Helbling 2008, 119–122). With our empirical analysis we assess whether the conventional wisdom of an overwhelmingly political national identity is valid or whether cultural and ethnic elements play an equally important role.

Second, we consider perceptions of national identity among the political elite alongside those of the masses. Studies to date have focused either on the role of the elite in the creation of national identity during the formation of the Swiss nation-state (e.g. Im Hof, 1991; Marchal 1992) or on popular conceptions of national identity expressed through participation in specific political rituals (e.g. Bendix 1992; Torriani 2002). A systematic account of elite and mass conceptions of Swiss national identity, however, is lacking. In order to fill this gap, we analyse data from the Swiss Eurobarometer 2003 and a new survey of members of parliament from the year 2010.

Our main finding is that Switzerland’s elite and its masses embrace a similar conception of national identity, one which is both political and cultural. The cultural dimension of Swiss identity is, however, not linguistic or religious monism of the sort often associated with cultural nationalism. Instead, the essence of Swiss cultural identity is pride in linguistic pluralism and regional diversity. To be sure, this does not mean that the Swiss embrace multiculturalism – cultural pluralism is neatly restricted to the native cultures. The result – that Swiss national identity is based on political features and cultural pluralism – suggests that the Swiss are not a purely political nation. Rather, the case of Switzerland provides a feasible model for the construction of cultural national identities in culturally diverse polities such as the European Union.

A mono-national state (almost) by definition

Debates about countries being national or multinational, and minorities being ethnic or national, often devolve into debates on terminological rather than substantial theoretical and empirical issues. As a consequence, many political scientists have simply despaired of offering explicit understandings of ‘nation’ and ‘national identity’ (Greenfeld 1999, 45). For the sake of clarity, we will nevertheless circumscribe our use of these ‘essentially contested’ concepts, keeping in mind that such ‘definitions’ necessarily remain to some degree ambiguous.1 A nation, for instance, is understood as a group of people who perceive themselves as belonging to a certain state. This means that a nation is an ‘imagined community’ (Anderson 1983) because it is a community of subjective self-identification and – in contrast to other ‘imagined communities’– it imagines itself in relation to a state (e.g. Miller 1995). For purposes of classification, therefore, it does not matter whether this state actually exists or merely constitutes a political goal the community strives for.

National identity contains the dimensions of salience and content. While the salience of the identity refers to the degree to which individuals identify with the nation, the content of the identity refers to the quality of this imagined community.2 Consequently, in order to grasp the content of national identity, we focus on the mental representations of individuals rather than on political institutions. Our focus is not on citizenship regimes; these have been shown to be influenced by the content of national identity (Brubaker 1992), but should not be equated with the latter.

Understanding national identity as the self-perception of belonging to a state, we arrive at an understanding of the Swiss as one nation. Although this is the dominant view today among Swiss historians and political scientists, arguments for understanding Switzerland as a multinational state can be traced back to the 19th century, when the importance of cultural features (e.g. a common language) as defining characteristics of a nation emerged (Kohn 1956, 90). Since then the idea of the Swiss nation as a multinational state has persisted (see Deutsch 1976, 63; Abizadeh 2002; Bhattacharyya 2007). Kymlicka (1995, 18), for example, labels Switzerland the ‘most multinational country’ because the different language regions do not share the same language (and, therefore, the same culture).3

The classification of Switzerland as a multinational state, however, is often based on the assumption that a common language constitutes the basis of a nation and therefore ‘multilingual’ can simply be equated with ‘multinational’ (for a critical discussion see Dardanelli forthcoming). Other authors emphasize that language can be more than a cultural element and claim that a common language is crucial for a constitutional democracy because it facilitates communication among citizens in the political process (Ipperciel 2007). Thus Switzerland is defined as ‘multinational’ not ipso facto because it is multilingual, but because the different linguistic groups do not share a common public space.

Countering the argument that Switzerland is a multi-national country, several authors defend the notion that the Swiss case represents a “true nation-state” (Hobsbawm 1992, 65; see also Mill 1977 [1865], 546; Renan 1896 [1882], 893; McRoberts 2001; Stojanovic 2003; Dardanelli forthcoming). Historical analysis shows that since the foundation of the Helvetic Republic in 1789 Switzerland has followed a rather typical process of nation-building in which language has played a subordinated role (Stojanovic 2003). Confronted with the ‘cultural’ refutation of the existence of a Swiss nation in the 19th century, Swiss elites and intellectuals such as Gottfried Keller accentuated ‘love of freedom’ and an attachment to the territory (mainly the Alps) as distinctive Swiss characteristics, in lieu of common ancestors or language (Kohn 1956, 93). These symbols and myths helped to form both the idea and the official realization of a single Swiss nation despite language divisions (e.g. Kriesi and Trechsel 2008; Im Hof 1991; Marchal 1992), and they were prevalent not only among intellectuals but also among ordinary citizens (Stojanovic 2003).

Evidence from recent surveys has repeatedly shown that the Swiss linguistic communities neither perceive themselves as distinct ‘nations’ nor identify themselves primarily with their linguistic groups, but identify with Switzerland as a whole (e.g. Kriesi et al. 1996; Stojanovic 2003). The ISSP survey data presented here confirms this evidence. Table 1 shows that a large plurality of the population, 45.9%, is strongly attached to the Swiss nation. This attachment does not vary widely across language groups. The difference is only 3.4% between French and German speakers and 6.3% between Italian and German speakers. In all language groups the share of respondents that feel at least ‘close’ to Switzerland is over 85%. This result confirms our reasoning: if a nation is understood as a group that imagines itself as congruent to a state, there is only one Swiss national identity shared by the large majority of the Swiss citizens.

Table 1.   Salience of identities among Swiss language groups
 French speakersItalian speakersGerman speakersTotal
  1. Notes: Only Swiss citizens, respondents with non-national language as first language were coded according to second language, Romansh speakers (N = 2) also.

  2. Source: Swiss Eurobarometer 2003.

How close do you feel to Switzerland?
Very close43.440.546.845.9
Close42.846.047.946.8
Not very close10.410.85.26.4
Not close at all3.52.70.10.9
N17337725935

Categorising national identities

The relation between state and nation is not imagined as accidental; they are perceived to be related through characteristics that make it possible to distinguish one particular nation from others. These characteristics need not be objective markers like language and religious denomination, but can be purely subjective (common will, values, beliefs, etc.). Furthermore, it does not matter whether or how the imagined characteristics are related to historical reality. They can be based upon ‘invented traditions’ (Hobsbawm and Ranger 1992), ‘myths of common descent’ (Smith 1986) and the like. They do, however, need to contain an image of the content of national identity which makes their particular nation unique and justifies its right to its ‘own’ state.

There is a long tradition in political science of categorising the content of these national identities into different types (Brubaker 1999). In particular, the distinctions between political vs. cultural and civic vs. ethnic have remained in wide use until today (Nieguth 1999).4 In the civic vs. ethnic distinction, the ‘civic’ category includes political and cultural values, while the ‘ethnic’ category refers to a narrow understanding of ethnicity – essentially, perceptions of common descent. In the political vs. cultural distinction, on the other hand, the ‘political’ category is rather limited while ‘cultural’ refers to a broad set of features that range from everyday culture to imaginations of common ancestry. Both dichotomies have been criticised for generating one category that is too narrow (the ethnic and political categories) and one that is too broad (the civic and cultural categories) to meaningfully describe variation across nations. The critique that the ethnic and political categories are too narrow is based on the observation that their cells are notably under-populated (see Brubaker 1999, 60–61). Whether or not this criticism is valid depends largely on the assumption that a category that contains very few cases is theoretically not relevant. In light of the importance of the question of whether a political national identity is possible, we think that this assumption is questionable, and this criticism invalid.

The corollary criticism that the civic and cultural categories are too broad and need refinement is more serious. The classical typologies might be combined (without doing much harm to their original meanings) by combining the overlapping ethnic vs. civic and cultural vs. political dichotomies, creating three types of national identity. These can be distinguished as: a political, a cultural, and an ethnic identity (see also Kymlicka 1999; Nieguth 1999; Nielsen 1999). As depicted in Figure 1 this is achieved by dividing the civic category (in the ethnic vs. civic distinction) into political and a cultural categories, and by dividing the cultural category (in the cultural vs. political dichotomy) into ethnic and cultural categories. This division only makes sense, of course, if the ‘cultural’ category in the new typology is used in a narrow sense and does not include ethnicity. This means that what we call here the cultural national identity refers to cultural characteristics which are perceived not as inherited but as acquirable. These include language and religion, but also common values such as a ‘Protestant work ethic,’ sincerity, generosity, hospitality, etc. if they are imagined by a majority to be characteristic of the national community. Although such a cultural national identity asks a lot of those who would assimilate to it, assimilation is not by definition impossible. Note that in the context of culturally diverse nations such a cultural national characteristic might also be pride in the specific national cultural pluralism. In such cases, assimilation means to adapt to one of the national cultures and embrace the idea of cultural pluralism.

Figure 1.

 Types of national identity

This is different from the ethnic national identity. In the ethnic national conception, the nation is perceived as a group of people sharing the same ancestors, which means that national identity is inherited and cannot be acquired by newcomers. This is not to say that ethnic groups are in fact essential categories, merely that people share an understanding of them as such (Gil-White 1999). Special emphasis is placed on phenotypes and other markers commonly perceived as related to hereditary descent (Chandra 2006). Although it is true that ethnicity is a socially constructed category and therefore malleable, individuals that do not belong to the ethnic community cannot assimilate to it by their own efforts. Assimilation is possible only if the society at large redefines the boundaries of its ethnic categories.

A political national identity refers to ‘constitutional patriotism’. It indicates that people imagine the nation as a community sharing the same political values, produced and reinforced by participating in the same political process. These values can refer to the political system of a country, its political culture or its role in the world. The content of such a national identity consists of a “distinctive interpretation of those constitutional principles that are equally embodied in other republican constitutions – such as popular sovereignty and human rights – in light of its own national history” (Habermas 1998b, 408).

Having adapted the dichotomous categorisations into a triptych does not mean that we have addressed all the criticisms raised against such typologies. There remain at least three serious issues.5 First, it has been argued that national identity always contains ethnic and cultural elements. Jenkins (1997, 15), for instance, argues that “the national … [is] to be understood as historically and contextually specific social construction on the basic ethnic theme” (see also Helbling 2008, 34f.). According to this narrow understanding of nation, which requires more than the shared belief of belonging to the same state, a (purely) political national identity is ruled out by definition. But even if we argue that nation and ethnic group should conceptually be separated, it remains an open question whether a social group can imagine itself as belonging to one state without perceiving itself as possessing a common cultural and/or ethnic identity.

Second, it has been argued that every national identity contains various elements at the same time. The criticism is that there is always variation within nations concerning the content of national identity. Although individuals identify themselves with the same nation, it does not follow that they share the same image of that nation’s common characteristics (Gellner 1983; Keane 1994; MacCormick 1996; Bonikowski 2011). This observation might be regarded as a critique of the undervaluation of heterogeneity within collective actors under these typologies. This pitfall has been widely acknowledged, and confirmed in numerous studies of nationalist discourses, frameworks, etc. of sub-national actors. It is not always clear whether these studies implicitly suggest that, being aware of heterogeneity at the individual level, we should not aggregate at the state-wide level, or whether they are merely intended as corrective or complementary to macro-sociological categories. But even if one believes in the merits of macro-sociological analysis, the problem of aggregating individuals’ conceptions of the nation remains a serious one.

A straightforward solution to the aggregation problem is to think of nations as belonging more or less to a certain type. If the intention, however, is to inform ideal-type theory it may in some cases make more sense to aggregate into clear-cut types. Such is the case in our reference to the Habermasian proposition of ‘constitutional patriotism’. Because the theory of constitutional patriotism raises the question of the existence of political nations, we need an aggregation rule that tells us when an empirical case is pure enough to be classified as a political nation. As a conservative rule, we might therefore consider as macro-level elements of the national identity only those elements that are the common denominators of individuals’ perceptions. In other words, we should understand the national identity as only those features of a nation’s common identity on which a large majority of its members agree.

The reality of heterogeneous perceptions of national identity at the individual level also raises the question of whether political, cultural and ethnic conceptions of nationhood are mutually exclusive or mutually reinforcing (Janmaat 2006, 51). The empirical evidence from representative surveys (Hjerm 1998a; 1998b; Jones and Smith 2001; Janmaat 2006) suggests that citizens do make a distinction between an amalgam of political and cultural features of national identity (the ‘civic’ category) on the one hand and a perception of their nation as an ethnic community on the other. Although the survey items divide two underlying dimensions in a very similar way across countries, they are not necessarily substitutes; indeed, the emergence of two dimensions in the empirical analysis might be due to common agreement on one set of features (e.g. the political and cultural elements), and corresponding disagreement regarding other features (e.g. ethnicity).

The fact that distinctive conceptions of the content of national identity coexist within the nation (i.e. are not mutually exclusive) supports the plausibility of combinations of types of national identity. These are categorized by combining the three types of national identity into four new types: ethno-cultural, ethno-political, cultural-political and encompassing. Altogether, this produces seven types of national identity, along with the possibility that national identity is not salient at all (Figure 1). In addition to the pure types (consisting of only one category), the ethno-cultural category is congruent with what has been described as the cultural type in the cultural vs. political distinction; the cultural-political category can be understood as the civic type in the civic vs. ethnic distinction.

An additional third option emerges: the ethno-political type. This is a particularly interesting type for the Swiss case because cultural diversity might lead one to hypothesise that the political content of the national identity is accompanied by an image of a people with common ancestors but no common culture. Finally, the encompassing type would provide a model for those who maintain that citizens do not make distinctions about the content of national identity. Such theorists might consider important everything that could be seen as a distinction between the members of one nation and those of another.

The third serious criticism of the classical typologies comes from social constructivists, who point out that both the salience and the content of (national) identity are fluid and contextual.6 According to Huddy (2001, 147), this perception is often derived from social identity theory. Whether or not observations pertaining to social identities can be applied to political identities remains a matter of dispute. In his review of the crossover from social identity theory to political identity theory, Huddy (2001, 147ff.) suggests not only that the fluidity of social identities has been overestimated, but also that political identities are considerably more stable (see also Hjerm 2003, 416). Empirically, the notion of ‘fluidity’ is often further contradicted by observation of remarkable stability in the answers to survey questions over time, a result to which we will return in the empirical section.

In this context, it is also interesting to note that instrumentalists often assume that elites’ identities are more stable than those of the mass of citizens.7 According to this perspective, the identities of the masses are situational and contextual to political struggles among the elites, who alone possess stable preferences. If in fact the preferences of elites provide a relatively stable foundation from which to influence the preferences of the masses, it makes sense to pay particular attention to the conceptions of national identity at the elite level, and derive from them statements about national identity beyond the immediate context.

In this chapter we have amended the classical typology in such a way as to make the categories less ambiguous, allow for all combinations of types and include a category for those that do not identify with a nation at all. Furthermore, we have pointed to the potential benefits of looking at both the masses and the elites. These steps are, of course, not enough to counter entirely the criticism that this kind of deductive research imposes categories on the actors rather than identifying actual cognitive schemes used in everyday life, a valid point to which we will return in the conclusion. They do, however, allow for a potentially greater diversity of perceptions of national identity than the classical typologies used to date. At the same time, they allow us to make use of surveys with representative sampling in order to investigate national identity for the Swiss citizenry as a whole. This will be our aim in the next section.

Swiss national identity

The classical understanding of Swiss national identity is a political one, based on a common appreciation of neutrality, federalism and direct democracy (Hilty 1875, 29; Linder 2005, 30ff.; Kriesi and Trechsel 2008, 11). According to this view, it is an attachment to these fundamental political institutions and values that unites Swiss citizens, and therefore comprises the core of Swiss national identity. But the argument that a purely political national identity does not rise to the necessary pitch of emotional attachment has also been raised in the Swiss case.

Centlivres and Schnapper (1991, 157f.), for example, argue that Helvetic citizenship is congruent with a voluntaristic political conception of national identity, while at the cantonal and communal level an ethnic idea of the community, in which individuals share the same culture, is predominant (see also Kriesi 2007). Others maintain that national identity need not necessarily contain a cultural dimension, but that ‘politically connoted’ invented myths about glorious ancestors can be tied to a wholly rational political identity (Renan 1889; Kriesi and Trechsel 2008).

To ordain myths as political, however, is problematic. Myths (and other articulations of political ritual and national symbolism) are most accurately understood as forms that transmit an image of national identity; their meanings are therefore dependent on contextualization (Zimmer 2003).8 An example of this is the accentuation of the alpine landscape as a distinctive characteristic of Swiss national identity, an area for the projection of myths, memories and virtues (Kaufmann and Zimmer 1998, 486).

The Alps provide meaning as the place where in 1291 the three founding members of the confederation (Eidgenossen) came together as equally free men to confirm their alliance against foreign invasion. In this context, the Alps are associated with the tradition of direct democracy and therefore convey a clear political meaning. This discourse, linking direct democracy to the heritage of general assemblies (Landsgemeinde) in the Alpine cantons, was also a feature of the democratic movement in the second half of the 19th century (Kriesi and Trechsel 2008, 11).

Furthermore, the Alps have also been considered the cradle of the ‘homo alpinus helveticus’, the supposed true ancestor of the Swiss race (e.g. Mottier 2000; 2008; Kreis 1992). In this context, the Swiss mountain landscape refers to the origins of the Swiss population and, hence, to an ethnic understanding of national identity. In the beginning of the 20th century this narrative was prevalent until the Swiss state distanced itself from it after the Second World War (Mottier 2000).

Finally, the Alps, and especially the alpine community, have been related to common Swiss values such as diligence, humility and order (Berthoud 2001). Mountains were alleged to have a “purifying effect on human beings” (Zimmer 1998, 654f.). In contrast to the political interpretation of the landscape, values are not projected onto but rather derived from the Alps; schoolbooks, for example, were often used to transmit ‘Alpine’ values (Helbling 1994). Although the landscape is perceived to have an influence on its inhabitants, this interpretation cannot properly be related to ethnicity because these values are not based upon common descent but rather are acquired, and therefore reflect a cultural conception of national identity.

The studies cited above and the example of the alpine landscape emphasise that the same symbols can be interpreted in different ways and that significant intra-national variation in such interpretations is possible. From just these examples, it seems that political elements alone do not comprise Swiss national identity – rather, cultural and ethnic elements register as well. Yet despite these insights, the Swiss nation is usually classified as a political nation. In order to take into account variance in understandings of the same national identity, we propose to rely on individuals’ actual conceptions of the nation and aggregate them to classify the Swiss case.

To test the hypothesis that the Swiss have internalised a political image of national identity, we use two different data sources. For the conceptions of national identity among the mass of citizens we rely on the data collected in the Swiss Eurobarometer 2003; the views of the political elite are measured with data from the Comparative MP Survey9, conducted among members of parliament (MPs) in the year 2010.10

The Swiss Eurobarometer 2003 includes questions from the International Social Survey Program as well as some questions about specifically Swiss characteristics such as Swiss political institutions. The sample contains 1037 respondents out of which 942 Swiss citizens were selected.11 In contrast to most MP surveys, the Comparative MP Survey was conducted not only at the federal but also at the regional level. A larger sample of the political elite was therefore surveyed, with the advantageous effect of generating a large N (496 MPs filled out the questionnaire). The fact that the low response rate for Switzerland (18%) is still somewhat higher than the rate for the same survey conducted in other countries reflects the reality that non-responses are a general problem of elite surveys. The fact that in Switzerland the representatives from all regional (i.e. cantonal) parliaments were included instead of just a selection thereof (as in other surveys from the PartiRep project) precluded the possibility of face-to-face interviews. Furthermore, although the response rate is by most measures low, the fact that 18% of our clearly delimited population participated in the survey allows for very efficient weighting, rendering the sample largely representative. In order to correct for the non-response bias, the sample was weighted with information on MPs’ party affiliations and cantonal affiliations.12

Both surveys contain items that are usually associated with the political, cultural or ethnic types of national identity.13 The questions on the political identity typically refer to direct democracy and neutrality, while the question about the importance of Swiss ancestry clearly relates to an ethnic understanding. Measuring the cultural content of Swiss national identity is more difficult, especially with the Swiss Eurobarometer 2003, because no one of the nation’s high cultures with its corresponding language can be considered the Swiss national culture. Instead, the cultural and linguistic pluralism of the country is itself perceived as characteristic of the Swiss national culture.

This emphasis on linguistic diversity as a constitutive element of the Swiss identity has become increasingly relevant over time. Although awareness of Swiss linguistic diversity led to the introduction of the language paragraph in the Swiss constitution of 1848, it played only a minor role in national identity at that time (Godel and Acklin Muji 2004, 118; Linder 2005, 29). Since the 1930s, however, linguistic diversity as an element of national identity has become increasingly important, culminating in the public discourse about the renewal of the constitutional language paragraph in 1996 (Coray 2004, 248, 382; see also Koller 1998/9, 73; Demont-Heinrich 2005, 71). Consequently, linguistic diversity has recently been emphasized as an important element of Swiss identity by several authors (e.g. Stevenson 1990, 130; Steinberg 1996, 128; Koller 2000, 563; Barbour 2002, 162; Kägi-Diener 2008, 92ff.).

Furthermore, multiple surveys underline the importance of linguistic and cultural diversity to the Swiss. In a survey conducted in the 1970s among army recruits about Switzerland’s mission in the world, the “most frequent choice of German-Swiss respondents was the example of linguistic and cultural coexistence to the world” (McRae 1998, 110).14 A representative study among Swiss citizens in 1997 found that 72% of the respondents considered linguistic diversity an important Swiss characteristic, comparable to direct democracy (75%) and neutrality (59%) (Suter 1998, 139). The importance of linguistic diversity has also been confirmed in a recent survey (Gfs 2009; for similar findings among Swiss local politicians see Helbling 2008, 119).15

In the MP survey, the importance of linguistic and (more generally) cultural diversity is clearly recognized in questions on the value of linguistic, cultural, cantonal and regional diversity. Cultural (and in particular linguistic) diversity is less directly measured in the Swiss Eurobarometer 2003. The item ‘to be able to speak an official language’ probably comes closest but leaves some room open for interpretation (see Hjerm 1998a, 459; Jones and Smith 2001, 49; Janmaat 2006, 56).

On the one hand, the ability to speak a national language can be understood as a fundamental condition for participation in the political process (Bauböck 1996; Erk 2003). If language is framed as an indicator of integration into the political community, it should be classified as a political element of national identity. On the other hand, linguistic competence may also be seen as an ascribed characteristic by relating the national language to national culture, as is usually the case in the literature on nationalism (e.g. Anderson 1983; Kymlicka 1995). In this sense, the command of a national language can also be interpreted as a means to access culture, history and tradition and hence a cultural aspect of national identity. Watts (1996, 136f.), for example, highlights the role of language in Swiss national identity to create and propagate social reality in the form of myths promoting national unity. If the item is understood in this context, it can be attributed to the cultural dimension of national identity.

The difficulty of attributing any one item to entire categories, as in the case of language, calls for the combination of multiple items. In order to infer from the items the underlying dimensions of the content of national identity (as perceived by the respondents), we use exploratory factor analysis.16 The results of the factor analysis should allow us to determine whether our refined typology accurately describes the types of national identity that exist within the Swiss nation, or whether Swiss citizens distinguish along lines other than those we propose.

As Table 2 shows, Swiss citizens distinguish between three different underlying dimensions of national identity. On the first factor, the importance of having Swiss ancestry in order to be ‘truly’ Swiss loads highest, followed by items that refer to birth in Switzerland, length of time living in the country, professing a Christian religion, and having Swiss citizenship. This factor relates quite nicely to what we call the ethnic type of national identity because elements that are largely determined through descent load on this factor. The fact that the item on religion also loads on this factor, however, suggests that this category is somewhat broader than the ethnic category in our typology, or that religion is perceived as concomitant with birth. The loading of the last item on this dimension, ‘having Swiss citizenship’, can be explained by Switzerland’s citizenship regime, an ethnic-assimilation type which sets high barriers for immigrants to access the political community (Helbling and Kriesi 2004; Koopmans et al. 2005; Waldrauch 2007; for different results, cf. Tilley et al. 2008; Jones and Smith 2001).

Table 2.   Factor analysis for content of national identity among Swiss citizens
VariableFactor 1 (ethnic)Factor 2 (political)Factor 3 (cultural)
  1. Notes: Principal-component analysis with promax rotation. Cut-off point: 0.4. Original wording of questions: 1Some people say the following things are important for being truly Swiss. Others say not.

  2. 2Would you say that you are proud that Switzerland features these three political institutions?

  3. Source: Swiss Eurobarometer 2003.

Have Swiss ancestry10.84  
Born in Switzerland10.82  
Spend most time of life in Switzerland10.71  
Being Christian10.71  
Have the Swiss citizenship10.64  
Importance of federalism2 0.87 
Importance of direct democracy2 0.84 
Importance of neutrality2 0.70 
Respect institutions & laws of Switzerland1  0.83
Able to speak one of the national language1  0.80
% of variance explained0.745  
Eigenvalue3.651.481.13
N821  

The second factor includes items that refer to the importance of political institutions to Swiss national identity. On this factor, the item about pride in federalism displays the highest factor loading, followed by pride in direct democracy and neutrality, respectively. This factor can clearly be associated with the political type of national identity.

On the third factor load the questions on the importance of proficiency in a national language and respect for Swiss laws and institutions to qualifying as ‘truly’ Swiss. As discussed above, it is not entirely clear whether these items are best understood as referring to the national culture or reflecting the importance of being able to participate in the national political community. The item ‘respect institutions and laws’ is problematic, in that it does not directly relate to the content of national identity (see above). The fact that it loads together with the language item suggests that it is related not only to compliance with regulations, but also to cultural norms to which foreigners must conform. Swiss federal law, for instance, requires foreigners “to be familiar with the Swiss way of life, customs and manners” in order to become Swiss citizens (see Koopmans et al. 2005, 53). This cultural interpretation is also supported by the fact that these items do not load with those referring to the political content of Swiss identity.

The factor analysis for the members of parliament paints a different picture. As Table 3 shows, factor analysis of this data generates only one factor. This means that the Swiss parliamentarians make little distinction between the different contents of national identity and perceive Swiss identity along the lines of our encompassing type. In other words, the parliamentarians mainly vary in the importance they attach to the aggregate aspects of Swiss identity, not in the relative importance of different features of Swiss identity.

Table 3.   Factor analysis for content of Swiss identity among members of parliament
VariableFactor 1 (only factor)
  1. Notes: Principal-component analysis with promax rotation. 1Question: How important do you think are the following aspects of Swiss identity?

  2. Source: Comparative MP Survey 2010.

Ancestry10.77
Traditions and customs0.76
Neutrality0.68
Cantonal and regional diversity0.68
Federalism0.64
Linguistic and cultural diversity0.51
Direct democracy0.45
% of variance explained0.423
Eigenvalue2.96
N486

Hence, while the data suggesting distinctions among the Swiss citizenry between the three types of national identity seems to validate those who insist on classical typologies of national identity, the lack of such distinctions made by the political elite tells another story. The two results, however, may seem more different on paper than they are in reality. In order to judge whether the distinctions between contents of national identity are significant in quantitative terms, we need to analyse how salient and widespread the identification with each type of national identity is. This is particularly important for the ‘common denominator’ understanding of national identity because observation of the existence of different types of national identity tells us little about their prevalence. It is possible, for instance, for two distinct dimensions of national identity to be perceived as so important that they should each be considered elements of the national identity. Consequently, we seek to analyse the distribution of the factor scores and interpret them in absolute terms. Such an analysis is made possible by calculating factor scores for each respondent (and dimension) consisting of the mean score of all items that load on a particular dimension (see Di Stefano et al. 2009, 3). Since the same scale was used for all items, the factor scores can furthermore be interpreted in close relation to the original scale.17

Analyzing the factor scores for the citizen survey, we find that the cultural type of national identity is the most widely held (see Table 4). The mean factor score for the cultural dimension is about 3.4, which indicates that the prevalence of cultural perceptions of national identity is on average between very important and fairly important. The low standard deviation further indicates that there is near-unanimity in the importance attached to regional as well as linguistic and cultural diversity for Swiss national identity. Somewhat lower, but in absolute terms still high, is the mean value for the political dimension of national identity (3.1), showing that direct democracy, federalism and neutrality should also be considered elements of Swiss national identity. Only on the ethnic dimension do citizens score clearly lower (2.7) and the relatively high standard deviation shows that there is little consensus on this dimension.

Table 4.   Factor scores for identity types among citizens and members of parliament
Identity dimensionNMeanStd. Dev.
  1. Notes:1Two respondents could not be attributed to one of the three language groups.

  2. Sources: Swiss Eurobarometer 2003 and Comparative MP Survey 2010.

Citizens:
All1
 Ethnic identity (factor 1)8212.660.72
 Political identity (factor 2)8213.080.63
 Cultural identity (factor 3)8213.390.50
 German speakers
 Ethnic identity (factor 1)6402.700.72
 Political identity (factor 2)6403.110.62
 Cultural identity (factor 3)6403.420.47
 French speakers
 Ethnic identity (factor 1)1452.470.72
 Political identity (factor 2)1452.960.64
 Cultural identity (factor 3)1453.280.57
 Italian speakers
 Ethnic identity (factor 1)342.760.70
 Political identity (factor 2)343.010.48
 Cultural identity (factor 3)343.370.51
Members of parliament:
 All
 Encompassing identity (factor 1)4963.200.48
German speakers
 Encompassing identity (factor 1)3933.210.49
French speakers
 Encompassing identity (factor 1)853.140.45
Italian speakers
 Encompassing identity (factor 1)183.250.42

That citizens distinguish between cultural and political features of national identity (see Table 2), but value both as fairly-to-very important (see Table 4), is no contradiction. It rather shows that our ability to distinguish between a political and a cultural dimension of national identity is mainly the result of differences in the valorisation of these features as very or fairly important. A large majority of Swiss citizens value both cultural and political elements of national identity, from which we infer that a large majority of Swiss citizens have internalized a political-cultural type of national identity. The fact that the variance is not equal among all dimensions of identity, but that there is much more disagreement on the ethnic dimension, shows that people not only differ in the content of their national identity, but also make qualitative distinctions between elements of the content of Swiss identity. Because the descriptive statistics do not fully reveal whether the same individuals who score high on the political factor also consider the cultural dimension equally important, we construct different types of national identity as proposed in Figure 1.18

Table 5 shows that a majority of respondents identify themselves with either a political-cultural (36%) or an encompassing (31%) understanding of national identity. These findings support our hypothesis that Swiss citizens do not have a purely political national identity, but also place great value on the cultural aspects of national identity. Furthermore, this result (that the political-cultural type is the common denominator for Swiss citizens) is unaltered when the distribution of national identity types is considered across different language regions. Political as well as cultural components of national identity are dominant across language regions for a majority of citizens. This is not to say that there are no differences between the language groups. Tables 4 and 5, for instance, illustrate that French-speaking citizens generally place less emphasis on the ethnic element of national identity than German and Italian speakers.

Table 5.   Distribution of national identity types among citizens per language region (in percent)
National identityAll citizensGerman-speakingFrench-speakingItalian-speaking
  1. Notes: To construct different types of national identity, we dichotomise the respondents’ answers on each item. Responses indicating below the value of ‘3’ were recoded into 0, the others into 1. Then, we combine these dummies to classify respondents into the resulting eight ideal types of national identity (see also Figure 1). Individuals in the political-cultural category, for example, score high on the cultural and political but low on the ethnic dimension.

  2. Source: Swiss Eurobarometer 2003.

Non-nationalists3.73.06.92.9
Ethnic0.50.50.70.0
Cultural20.820.621.420.6
Political2.82.54.12.9
Ethno-cultural4.95.03.58.8
Ethno-political0.70.61.40.0
Political-cultural35.934.742.829.4
Encompassing30.733.119.335.3
Total100100100100
N82164014534

Further examination of the data reveals the expected patterns in individual-level characteristics (see Table 6). Respondents belonging to the encompassing type, for instance, are (compared to those in the political-cultural type) older (56), have spent less time in education (10 years) and position themselves more to the right on the left-right axis (5.9) than those belonging to the political-cultural one.19

Table 6.   Characteristics of citizens by types of national identity
 Left-rightGender (% of men)AgeYears of educationN
  1. Notes: Values refer to means, standard deviations are in parentheses.

  2. Source: Swiss Eurobarometer 2003.

Supra/subnationalists4.0 (1.5)43.342.4 (13.3)11.1 (4.0)30
Political4.5 (1.6)52.250.2 (14.1)12.0 (3.0)23
Ethnic5.3 (1.0)50.055.5 (20.2)14.8 (4.9)4
Cultural4.1 (1.8)48.046.8 (14.7)12.4 (3.7)171
Ethno-cultural4.3 (1.8)32.556.0 (18.3)10.1 (2.2)40
Ethno-political5.3 (1.8)33.348.7 (17.5)9.5 (1.2)6
Political-cultural5.3 (1.8)49.248.4 (17.4)11.9 (3.7)295
Encompassing5.9 (1.9)52.456.3 (17.2)10.3 (2.8)252
Overall Mean5.1 (1.9)51.250.7 (17.1)11.4 (3.5) 
Total N749821821821 

As shown above, the political elite made less of a distinction than their fellow citizens between the importance of ethnic aspects of national identity and that of political and cultural aspects. As the analysis of the factor scores further shows, most parliamentarians on average find all the features listed fairly important, and as the rather low standard deviation shows, there is considerable agreement on their importance for Swiss national identity (see Table 4). The result that the political elite ascribes greater value to the ethnic dimension of national identity than other citizens comes as a surprise, contradicting two recent studies which arrive the opposite conclusion. Further consideration of this result is merited.20

Elites’ variations across language groups are again very modest. The different features of national identity are on average considered to be at least fairly important. The only notable difference is between German speakers and French speakers: the former place somewhat more importance on the various aspects of national identity than the latter.

The break down of average scores by item and party affiliation shows that there is strong consensus regarding the association of direct democracy with national identity (Mean = 3.8, SD = 0.4), in contrast to the importance of ancestry (Mean = 2.6, SD = 0.9) (see Table 7). Unsurprisingly, the average score of SVP members on this item (Mean = 3.5, SD = 0.7) is considerably higher than that of the SPS and Greens (Mean = 2.1; SD = 0.8), with the members of the conservative parties, (CVP, BDP, EVP) falling in between. Interestingly, even the Liberals embrace cultural elements of national identity.

Table 7.   Average score on items of national identity of MPs by party
 SVPSPSLiberalsCVPGreensBDPEVPOthersOverall mean
  1. Notes: Ranking according to number of MPs in the national parliament.

  2. Source: Comparative MP Survey 2010.

Ancestry3.5 (0.7)2.0 (0.8)2.7 (0.8)2.9 (0.7)2.1 (0.8)3.0 (0.6)2.9 (0.7)2.6 (0.8)2.6 (0.9)
Traditions and customs3.4 (0.7)2.4 (0.8)2.9 (0.8)3.0 (0.7)2.3 (0.7)3.1 (0.7)2.8 (0.5)2.8 (0.8)2.8 (0.8)
Neutrality3.8 (0.4)2.5 (0.9)3.2 (0.7)3.2 (0.7)2.9 (0.9)3.4 (0.5)3.3 (0.7)3.0 (0.8)3.1 (0.9)
Cantonal and regional diversity3.6 (0.6)3.0 (0.7)3.6 (0.6)3.5 (0.6)3.1 (0.8)3.5 (0.7)3.4 (0.6)3.1 (0.7)3.4 (0.7)
Federalism3.4 (0.8)2.7 (0.8)3.3 (0.8)3.3 (0.7)2.8 (0.9)3.2 (0.8)2.9 (0.7)3.1 (0.8)3.1 (0.8)
Linguistic and cultural diversity3.4 (0.7)3.1 (0.7)3.2 (0.7)3.1 (0.8)3.2 (0.7)3.6 (0.5)3.3 (0.6)3.2 (0.7)3.2 (0.7)
Direct democracy4.0 (0.1)3.6 (0.5)3.8 (0.5)3.8 (0.4)3.7 (0.5)3.9 (0.3)3.9 (0.2)3.9 (0.3)3.8 (0.4)
Total N81110947555111941486

In summarizing the results, we want to emphasize that a large majority of both the citizenry and the political elites perceive national identity not only in political terms, but also in cultural ones. This picture does not change if we compare the distinct language regions. The data shows that the Swiss, often considered a prototypical political nation, should be considered instead a political-cultural nation. This result is in line with comparative research showing that political and cultural features of national identity tend to adhere, and are not separately distinguished by the citizens (Hjerm 1998a, 1998b; Jones and Smith 2001; Tilley et al. 2008). But while the data used in comparative research was not particularly suited to the qualification of the political content of national identity, and its results were consequently rather tentative, we were able to replicate the same results for a particularly ‘hard case’ with more appropriate data. Our empirical analysis, furthermore, went beyond these studies because we based our interpretations on mass and elite surveys and provided a solution for the analysis of factor scores.

Concerning the categorization of national identities, our analysis – at least in the case of Switzerland – has also shown that the pertinent question might not be whether citizens perceive the national community more in cultural-political or in ethnic terms. While we demonstrate a broad consensus on the importance of both political and cultural elements, it remains open to debate whether ethnicity constitutes an important element of national identity or not. A consequence of this result is that future research should focus on the distinction between encompassing and cultural-political types of national identity, i.e. the distinction between national identities that respectively include or exclude ethnicity as a constitutive element, rather than on the cultural vs. political and civic vs. ethnic distinctions.

Conclusion

Our goal has been to answer the question of whether the Swiss should be considered a political nation. What seemed to be a simple question ultimately required a considerable number of conceptual and empirical steps. The first step consisted of making explicit our understanding of national identity and establishing that, according to such an understanding, Switzerland constitutes a mono-national state. Next we developed a more nuanced typology of national identities, which incorporated recently raised criticisms and allowed us to empirically test the notion of the Swiss political nation. While this typology is more sensitive to the plurality of perceptions of national identity than former ones, we cannot be sure that citizens and politicians actually apply the categories placed at their disposition in surveys. This uncertainty suggests the need for qualitative research that tries to infer the categories that constitute peoples’ everyday conception of the Swiss nation.

This method-inherent limitation of our approach, however, does not invalidate our result: the Swiss are not primarily a political nation. A large majority of Swiss citizens and politicians imagine the Swiss nation in cultural terms as well. The cultural aspect of the Swiss national identity consists of the perception that the Swiss are a culturally diverse community – a paradigmatic case of unity in plurality. This Swiss cultural pluralism, however, should not be confused with multiculturalism, i.e. openness to all cultures. Swiss cultural pluralism is exclusively limited to native cultures and is characterized by strong delineation of immigrant cultures.

If Swiss national identity in the early 21st century is based on imagined communities that are both political and cultural in nature, then advocates of constitutional patriotism have lost an important exhibit in their case for the feasibility of a purely political nation. Of course, this is not to say that the idea of constitutional patriotism is not worth propagating. Still, our findings raise the question of whether the larger political nation under construction – the European Union – should put more emphasis on the cultural dimension. This is arguably a difficult task in such a culturally fragmented community. But as Switzerland shows, a nation can establish cultural pluralism as the essence of its cultural identity.

Footnotes

  • 1

     We follow the argument of Davis (2005, in particular chapters 2 and 3) that it is impossible to give clear-cut definitions as this would require the definition of all concepts used in the definition, leading to an infinite regress of definitions.

  • 2

     What we call salience of national identity is often understood among psychologists as strength of nationalism or nationalist sentiment. It is sometimes distinguished from patriotism, which, in contrast to nationalism, is understood as an identification that allows for critical reflection upon one’s nation. These understandings are furthermore distinct from a sociological one, which understands nationalism as an ideology advocating that the state and the nation should be congruent (e.g. Gellner 1983, 3).

  • 3

    Kymlicka (1995, 11) defines a nation as ‘a historical community, more or less institutionally complete, occupying a given territory or homeland, sharing a distinct language and culture’ and uses this term as synonymous with culture.

  • 4

     Among the more recent references, Greenfeld (1992) on the political vs. cultural distinction and Smith (1991) on the civic vs. ethnic distinction have probably been the most influential.

  • 5

     We do not address the criticism that the typologies are ‘normative’ (among those see Yack 1999; Brown 1999; Kuzio 2002).

  • 6

     For a discussion of ‘weak’ social constructivist notions of identity against ‘hard’ (allegedly) essentialist notions of identity see Brubaker and Cooper (2000, 6–14). See also Brubaker (1996, 56) for a typical critique of ‘hard’ notions of ethnic identity.

  • 7

     For reviews on instrumental approaches to nationalism see for instance Brubaker (1998, 289–292) and Breuilly (1994, 409–411).

  • 8

     Here we disagree also with authors who classify these symbols a priori as elements of the ethnic conception of national identity (e.g. Froidevaux 1997).

  • 9

     The Comparative MP Survey is part of the PartiRep (Participation & Representation) research project, an Interuniversity Attraction Pole funded by the Belgian Science Policy, which analyses changing patterns of participation and representation in modern democracies. The survey was carried out in 15 countries and 60 parliaments between 2009 and 2010 (for further information see http://www.partirep.eu).

  • 10

     The time gap of seven years might call the comparability of the two surveys into question. Because we believe that perceptions of national identity among the respondents, and in particular among the political elite (see the discussion above), remain rather stable, we believe that general comparisons can be made.

  • 11

     The response rate of the Swiss Eurobarometer 2003 is 30.2 %. The survey includes 432 male and 462 female respondents with an average age of 50.7 years (SD = 17.1) and an average full-time education of 11.4 years (SD = 3.5; see also Table 6).

  • 12

     An explorative analysis showed that party affiliation in centre-right parties (CVP, Liberals, SVP) and identification with the Swiss Francophone language community are particularly strong predictors of non-response. To account for the ‘party affiliation’ bias, the major parties with more than 2% of all MPs at the cantonal and national level have been selected (SVP, Liberals, CVP, SPS, Greens, BDP, EVP). Weight has been calculated by dividing the proportion of MPs per party to all MPs that filled out the questionnaire (sample proportion) by the proportion of all MPs per party to all MPs (population proportion). Based on the new distribution of MPs weighted by party affiliation, the ‘cantonal’ weight has been calculated accordingly. Both weights have been multiplied to obtain the final weighting variable.

  • 13

     See Tables 2 and 3 for the exact wording of the original questions.

  • 14

     French-speaking Swiss, however, more often described a humanitarian or democratic mission (McRae 1998, 110). This finding seems to indicate that the importance of linguistic diversity among citizens is mainly a concern of German-speaking Swiss.

  • 15

     When respondents were asked about elements of pride in Swiss politics, ‘coexistence of different linguistic groups’ ranked third after ‘independence’ and ‘neutrality’ with 88% of the respondents being ‘at least proud’ of linguistic diversity (51%‘very proud’ and 37%‘quite proud’).

  • 16

     Principal-component factor analysis with promax rotation, which allows dimensions to be correlated, has been used. We excluded the item ‘feel nationality’ because it cannot be clearly assigned to a particular type of identity (see Heath et al. 2009, 308).

  • 17

     We assign each item to one dimension by applying the cut-off value of 0.4. The scale contains the same options for all items: 1 = not at all, 2 = not very, 3 = fairly, 4 = very.

  • 18

     For the construction of different types of national identity see the note in Table 5.

  • 19

     Because only 53% of all respondents indicate party identification we do not report the results according to party affiliation here (but see Appendix 1). The data shows that over 55% of the respondents feeling close to the SVP have an encompassing idea of national identity compared to 41% those feeling close to the Liberals or the CVP. The majority of the SPS sympathisers can be found within the cultural and the political-cultural type of national identity.

  • 20

    De Cillia et al. (2009, 169) have found that culture and ethnicity in Austria are less important in the official discourse than in the semi-official and quasi-private discourses. A study on American national identity in the context of 9/11 finds that “the elite efforts appear to be mobilizing the masses away from ethno-cultural definitions of the national identity rather than toward them” (Schildkraut 2002, 518).

Acknowledgements

Oliver Strijbis was supported by the National Centre for Competence in Research – Challenges to Democracy in the 21st Century. For helpful comments on earlier drafts we would like to thank Marc Helbling, Nenad Stojanovic, Romain Lachat, Eva Green, as well as two anonymous reviewers. We would also like to thank Patrick Lengg for his assistance with the PartiRep survey.

Appendices

Appendix 1:

Distribution of citizens regarding feeling close to a specific party by type of national identity

 SVPLiberalsCVPSPSOthersNATotal
  1. Notes: Respondents for whom information on feeling close to a specific party not available are included in the column ‘NA’.

  2. Source: Swiss Eurobarometer 2003.

Supra/subnationalists3541830
3.5%4.1%5.0%4.2%
Political131421223
1.6%3.5%2.5%3.3%2.5%2.8%
Ethnic11114
1.6%1.2%0.8%0.2%
Cultural1103382594171
1.6%11.8%7.5%30.9%31.3%21.8%
Ethno-cultural2131212140
3.3%1.2%7.5%9.8%1.3%4.9%
Ethno-political1146
1.6%1.2%0.9%
Political-cultural2131164230155295
34.4%36.5%40.0%34.1%37.5%35.9%
Encompassing3435172117127252
55.7%41.2%42.5%17.1%21.3%29.4%
Total N61854012380432821
7.4%10.4%4.9%15.0%9.7%52.6%

Beatrice Eugster is a Ph.D. candidate at the Institute of Political Science at the University of St.Gallen. Her primary research interests include comparative welfare state analysis, migration and citizenship. Address for correspondence: Beatrice Eugster, Institute of Political Science, University of St.Gallen, Rosenbergstrasse 51, 9000 St. Gallen, Switzerland. Phone: +41 71 224 39 80; E-mail: beatrice.eugster@unisg.ch

Oliver Strijbis is a postdoctoral research fellow at the Institute of Political Science at the University of Hamburg. His research focuses on ethnic and national identity, elections, and social movements. Address for correspondence: Oliver Strijbis, Institute of Political Science, University of Hamburg, Allende-Platz 1, 20146 Hamburg, Germany. Phone: +49 40 42 838 32 33; E-mail: oliver.strijbis@uni-hamburg.de

Ancillary