Scholarly article on topic 'Nationalistic attitudes and voting for the radical right in Europe'

Nationalistic attitudes and voting for the radical right in Europe Academic research paper on "Political Science"

0
0
Share paper
Academic journal
European Union Politics
OECD Field of science
Keywords
{""}

Academic research paper on topic "Nationalistic attitudes and voting for the radical right in Europe"

Regular Research Article

Nationalistic attitudes and voting for the radical right in Europe

European Union Politics 0(0) 1-21

sagepub.co.uk/journalsPermissions.nav

DOI: 10.1177/1465116516678932

© The Author(s) 2016 ,eprints and permissions:

eup.sagepub.com

Marcel Lubbers

Department of Sociology, Radboud University, Nijmegen, The Netherlands

Marcel Coenders

Department of Cultural Diversity and Youth, Utrecht University, Utrecht, The Netherlands

Abstract

Voting for radical right-wing parties has been associated most strongly with national identity threats. In Europe, this has been framed by the radical right in terms of mass-migration and European integration, or other politicians bargaining away national interests. Perhaps surprisingly given the radical right's nationalist ideology, nationalistic attitudes are hardly included in empirical research on the voting behaviour. In this contribution, we test to what extent various dimensions of nationalistic attitudes affect radical right voting, next to the earlier and new assessed effects of perceived ethnic threat, social distance to Muslims, Euroscepticism and political distrust. The findings show that national identification, national pride and an ethnic conception of nationhood are additional explanations of radical right voting. National identification's effect on radical right voting is found to be stronger when populations on average perceive stronger ethnic threat.

Keywords

Nationalism, national pride, radical right, voting behaviour Introduction

It has been often assessed that the radical right in contemporary Europe is nationalist in its ideology. Mudde (2007) describes the ideologist position as one

Corresponding author:

Marcel Lubbers, Department of Sociology, Radboud University, Thomas van Aquinostraat 6, 6525GD Nijmegen, the Netherlands. Email: m.lubbers@maw.ru.nl

of 'nativism', implying that radical right-wing parties propose to keep the nation state population native, and that non-native elements are threatening the homogeneity of the nation state. In somewhat different terms, Rydgren (2007) labelled the radical right-wing ideology as one of'ethno-pluralism'. In this ideology, differences between cultures are emphasized, and to preserve perceived national unique features, different cultures should stay apart and should not mix, because that would lead to extinguish of the unique cultural characteristics. At the one hand, the uniqueness of the nation-state is stressed, and at the other hand, the threats against which the nation-state should be protected. Radical right-wing programs mostly open with questions on and perceived threats to national identity. The Flemish Interest party proclaims that it is 'a Flemish-nationalist party... implying that the cultural identity of our people are decisive for the creation and government of the state'. The German NDP (Nationaldemokratische) claims that national identity means that 'Germany should remain the country of the Germans, and should, in places where this no longer is the case, become it again'. And the Dutch Party for Freedom claimed in the first paragraph of the election program from 2010 that 'The Dutch do not have their equals. We are born from a fight for freedom. Our ancestors have turned this delta marsh in a country that is envied by the whole world'.

Empirical research on the predictors of a radical right-wing vote hardly addressed the role of the nationalist part of the ideology. It focused much more on the role of attitudes towards issues that are claimed to be related to threats to the nation. First, researchers studied the role of attitudes towards migrants and integration (e.g. Ivarsflaten, 2005, 2008; Norris, 2005; Rydgren, 2008; Van der Brug et al., 2000) and second, studies turned to the role of euro-scepticism (e.g. Werts et al. 2012). These ideologically related explanations turned out to exist next to political protest explanations such as political distrust or political cynicism (Ivarsflaten, 2008; Werts et al., 2012). Measurements of nationalistic attitudes to explain radical right-wing voting have hardly been included, possibly because most of the international studies rely on cross-national datasets such as the European Social Surveys and Eurobarometers in which measures on nationalism lack. In the country case studies where it was included, only a specific dimension of nationalistic attitudes - i.e. national pride - was studied (Billiet and De Witte, 1995) or it was not controlled for the relevant other predictors of radical right voting (Meuleman and Lubbers, 2013). These studies showed that nationalistic attitudes had at the most a small effect on voting for the radical right. We address the relation between nationalistic attitudes and a vote for the radical right further in this contribution. We question whether nationalistic attitudes affect the vote for the radical right next to attitudes that have previously turned out to be decisive in the vote for the parties, i.e. perceptions of migrant threat, euro-scepticism and political distrust. With this approach, we will provide insight into the role of identification with the nation (in comparison to identification with Europe or sub-national geographical units) in voting for the radical right and will discuss whether this identification is a sufficient or merely a necessary condition. Relating to other studies that study the consequences of national identification and nationalist attitudes

(e.g. De Figueiredo and Elkins, 2003; Gibson, 2006), we include measurements for multiple dimensions of nationalistic attitudes. We demonstrate whether conceptions of nationhood that exclude out-groups based on origin affect voting for the radical right (Rydgren, 2007). With that focus, the contribution shows whether nationalistic attitudes are the underlying explanations of the effects of perceptions of migrant threat and euro-scepticism on radical right-voting (Werts et al., 2012). Moreover, we test the relationship in a cross-national comparative setting, providing evidence whether a context in which a public opinion is more favourable towards the radical right, national identification is more likely to be positively related to radical right voting. The results indicate that nationalistic attitudes affect radical right-wing voting over and beyond ethnic threat, social distance to Muslims, Euroscepticism and political dissatisfaction, though its effect is modest.

Expectations

The literature has regarded the importance of unfavourable attitudes towards migrants (e.g. Van der Brug et al., 2000) and euro-scepticism in radical right voting as support for the idea that people vote because of a nationalist ideology (e.g. Werts et al., 2012). When migrants or the European Union (EU) are perceived as an economic, cultural or political threat to the national interests, voters would favour a party that claims to serve the natives' in-group interests. The question remains unanswered whether the radical right mobilizes support because of its appeal to people who identify strongly with the nation and are patriotic or only to people who perceive that the national interests are under threat from immigrants and further European integration. Building on Social Identity Theory (Tajfel, 1982; Tajfel and Turner, 1979), it is claimed that the nation can be one of the relevant groups that people identify with (e.g. Smith, 2007). When the nation is a more relevant group for someone's identity it is to be expected that the nationalist ideology of the radical right, referring to the relevance of that group membership, is supported stronger than when people do not identify with the nation at all. From this perspective, it seems that national identification is a necessary condition to support parties that proclaim a nationalist ideology, which should lead to the finding that among people that do not identify with the nation, voting for the nationalist parties is absent. We expect that people who do not attach importance to the nation and do not identify with the nation will be unlikely to vote for a party with a nationalist agenda.

Hypothesis la: People without a positive identification with the nation do not vote for

the radical right.

But will stronger identification with the nation also have the consequence of stronger support for a nationalist party? Radical right parties define an out-group in order to construct a national in-group. Bruter and Harrison (2011) describe this as the negative conception of identity. Due to the nationalist program of radical right

parties, we expect that stronger national identification induces more interest in and support for nationalist ideologies, and consequently a larger likelihood to vote radical right.

Hypothesis 1b: The stronger the identification with the nation, the stronger the support for the radical right.

However, in line with previous research that showed that national identity is not inherently related to an out-group bias (Brewer, 2001; De Figueiredo and Elkins, 2003; Gibson, 2006), the relation between national identity and radical right voting might be more complicated. Although nationalistic attitudes and perceptions of threat to the nation are correlated, love for the country and even perceived superiority of the country do not necessarily have to result in perceptions of migrant threat and euro-scepticism (Coenders and Scheepers, 2003). Associations between nationalistic attitudes and anti-immigration attitudes are far from perfect. And even though European identification is positively associated with positive attitudes towards migration (Curtis, 2014; De Vreese and Boomgaarden, 2005), others have shown that the identities are not mutually exclusive. Citrin and Sides (2004) provide evidence that a (small) proportion of EU citizens hold a dual identity, attaching importance to being a country's national and European. In accordance with these findings, theorists in the Social Identity Theory tradition have stressed that stronger in-group identification does not has to be accompanied with negative attitudes towards out-groups (Brewer, 1999; Brown, 2000). Brewer argued that identification is a necessary condition but not a sufficient condition for intolerance. The radical right plays both cards in its rhetoric and campaigns: it departs from a nationalist ideology, stressing the relevance of national identification and national pride, but it does so by framing that its definition of the national identity is under threat. To shed more light on this issue, we also address other aspects of nationalistic attitudes (national pride and conceptions of nationhood) as well as the national context in which national identity may be more strongly related to radical right voting.

Firstly, we take into account pride is one's nation as another aspect of nationalistic attitudes (Ariely, 2012; Solt, 2011). Identification does not necessarily have an evaluative component, whereas pride expresses an explicit positive evaluation of the nation. We therefore expect that stronger pride in the country is likely to increase the orientation to parties that address a nationalist ideology.

Hypothesis 2: The more national pride, the more likely the vote for the radical right.

However, here national pride - in the absence of negative identity construction -may be unrelated to voting for the parties that are built around a nationalist ideology. De Figueiredo and Elkins (2003) showed for the US that patriotism was not related to levels of prejudice. A similar finding was reported by Gibson (2006), showing that stronger sympathy for the in-group in South-Africa did not

give rise to intolerance. It showed that love for one's nation is widespread and that intolerance is just as likely among people with and without pride in the nation. Again we propose that love for the nation is a necessary condition for voting for a nationalist party. The radical right strongly relies on nationalist symbols to stress its love for the nation. People without any love for the nation will therefore be unlikely to vote for the nationalist party.

Hypothesis 2a: People without national pride do not vote for the radical right.

Secondly, the literature on nationalistic attitudes has further differentiated between attitudinal dimensions (Dekker et al., 2003; Hjerm and Schnabel, 2010). This differentiation can improve our understanding of the relation between nationalistic attitudes and radical right voting, and address the question whether these are inherently interrelated. One relevant distinction is made between civic and ethnic conceptions of nationhood (Ford et al., 2011; Kunovich, 2009; Reeskens and Hooghe, 2010; Rydgren, 2007). It focuses on how the nation - as an imagined community (Anderson, 1983) - is understood. These conceptions contain different views on what characterizes the national in-group and the extent to which national group membership is defined in a way that excludes migrants. In an ethnic definition of nationhood, the nation is defined in terms of shared ancestral origins and cultural homogeneity and distinctiveness. Membership of the national community is denied to anyone who is not a member of the native majority group. This ethnic conception of nationhood reflects the jus sanguinis (right of blood) principle of nation-building, according to which citizenship is based on ancestry (Brubaker, 1992). Here, the basis may be found why migrants and the EU are more likely to be seen as a threat; by the definition of what nationhood constitutes, people from other cultural backgrounds are excluded. In contrast, in a civic definition of nationhood, membership of the nation does not depend on descent, but on the willingness to accept the basis rules. Civic nationalism focuses on commitment to society and participation in society (Smith, 2001). The civic conception of nationhood reflects the jus soli (right of ground) principle of nation-building, according to which citizenship is based on living within the national territory.

Ethnic and civic conceptions of nationhood are not mutually exclusive; people may support both types of nationhood (Janmaat, 2006), but it is generally thought that an ethnic conception of the nation is stronger associated with exclusion of migrants than a civic conception of the nation (Kunovich, 2009). As Bruter and Harrison (2011) identify, radical right parties define the nation by referring to what the majority shows to have in common, such as history and language. People who do not share these common features are then excluded from membership. We hence focus on the criteria that people use to define the nation.

Hypothesis 3a-c: People who attach more value to (a) common national ancestry, to

(b) being born in the country and to (c) be able to speak the national language for

being truly a member of the nation, are more likely to vote for the radical right.

As ancestry is the most exclusive criterion to define nationhood (Rydgren, 2007), we would expect that this particularly holds for those who attach more value to common national ancestry.

National public opinion

The discussion on civic and ethnic conceptions of the nationhood has often been linked to different national traditions in dominant ideas of nationhood and official citizenship criteria. Based on citizenship regime traditions, France and Germany have often been regarded as ideal types of respectively jus soli and jus sanguinis countries (Brubaker, 1992). However, subjective conceptions of nationhood among the general public do not necessarily reflect the official citizenship criteria that are applied by the state. Furthermore, instead of departing from a dichotomy of ideal types of nations, we think that populations are affected by the general public opinion in a country on the conceptions of the nation. We argue that the extent to which nationalistic public opinions prevail in a country - regarding national pride and ethnic conceptions of nationhood - affect the relationship between national identification and voting for radical right parties. Since studies into the relation between national identification and intolerance have shown that the association is at the most weak, social identity theorists have stressed the conditionality under which national identification affects intolerance (Brewer, 2001). Gibson (2006) states that political and social salience of intergroup tension form such a moderating context. Weldon (2006) showed that laws on citizenship acquisition are associated to intolerance among the population. Also, Pherson et al. (2009) found that the relationship between individuals' level of national identification and antiimmigrant prejudice is moderated by the extent to which people within a country endorse specific definitions of the nation. Hence, to take the role of nationalism in radical right voting a step further, we also investigate to what extent differences between countries in the dominant nationalistic public opinion play a role in the voting for the radical right. It can be expected that national identification is stronger related to support for parties with a nationalistic ideology when people are encouraged by a public opinion that expresses a strong ethnic perception of the nationhood.

Hypothesis 4: The more value the population of a country attaches to a common ancestry as definition of nationhood, the stronger the positive effect of national identification on the likelihood to vote for radical right parties.

Finally, we anticipate a similar role of the dominant public opinion on migration and the EU. Previous studies showed that anti-immigration attitudes and euro-scepticism are important explanations of radical right voting. In national contexts where the dominant public opinion is euro-sceptic and critical towards migration, EU integration and migration are more likely perceived as threats to the nation. As the radical right contrasts the positive aspects of the own nation with the negative

features of perceived threats to the nation, we expect that in countries where EU integration and migration are stronger perceived as problems, national identification will be stronger related to voting for the radical right. Given the anti-Islamic rhetoric of many radical right parties in particular and negative public climate towards Muslims, we also focus on the role of the dominant public opinion towards Muslims.

Hypothesis 5a-c: The more unfavourable the public opinion climate towards (a) migrants, (b) Muslims or (c) towards the EU, the stronger the positive effect of national identification on the likelihood to vote for radical right parties.

Data and operationalization

Data were derived from the fourth wave of the European Values Study 2008-2010 (EVS, 2011). All European countries participated in this survey. Since we also test for interaction effects between nationalistic attitudes and the dominant public opinion within a country, we included data only from countries with at least 20 radical right-wing voters. The attitude towards the EU was not asked in Norway. Therefore, we also dropped this country from the analysis. This left us with 20 countries (see Table 1).

Radical right-wing voting: We classified parties according to the standards in the literature, discussing which parties belong to the party-family (Immerzeel, 2015; Mudde, 2007; Rydgren, 2008). Table 1 provides the table with the included countries, parties and the percentage of voters in the data.1

The question on voting behaviour tapped which party respondents had voted for in the last national elections. Our dependent variable is dichotomous, with the value 1 referring to a radical right-wing party vote, and the value 0 referring to a vote for another party. People that did not vote, or reported that they did not know which party they had voted for were excluded from the analyses. The percentage that claimed not to vote equals almost 25%, which is an underestimation of non-voting. Another 9% did not know which party to vote for and around 7% did not provide an answer.

Nationalistic attitudes: National identification was measured with a question in which respondents were presented five geographical units (locality or town, region, country, Europe, the world) and were asked 'which of the geographical groups would you say you belong to first of all?' and as a follow-up question 'and secondly?'. We distinguished between those who mentioned the country as their first choice (i.e. strong national identification), as their second choice (moderate national identification) and those who did not mention their country (weak national identification). Although the question triggers respondents to mention a geographical group even when they identify to none of these groups, we would anticipate that most of these

Table 1. Parties classified as Radical Right and their share of voters in the European Value Survey.

Country Acronym Party of votes

Austria FPO Austrian Freedom Party; Alliance for the Future of Austria 18.7

Belgium: Flanders VB Flemish Interest 13.8

Bulgaria ATAKA National Attack 9.1

Croatia HSP; HDSSB Croatian Party of Rights; Croatian Democratic Alliance of Slavonia and Baranja 7.0

Denmark DF Danish People's Party 8.7

Finland PerusS True Finns 12.3

France FN; MPF National Front; Movement for France 4.7

Germany Rep; NPD Republicans; National Democratic Party of Germany 2.1

Greece LAOS Popular Orthodox Rally 3.4

Hungary JOBBIK Movement for a better Hungary 3.1

Italy LN; MS-FT; NF Northern League; Three-coloured Flame; National Front 10.1

Latvia LNNK; VL For Fatherland and Freedom; All for Latvia 10.0

Netherlands PVV; TON Party for Freedom; Pride in the Netherlands 10.5

Poland LPR United Poland; League of Polish Families 3.5

Romania PRM Party for Greater Romania 6.0

Serbia SRS Serbian Radical Party 5.9

Slovak Republic SNS Slovak National Party 9.3

Slovenia SNS Slovenian National Party 10.9

Switzerland SVP Swiss People's Party 21.0

Turkey MHP Nationalist Movement Party 7.7

respondents would then have missings and mentioned nothing. This was however not the case. Figure 1 shows that national identification is much more relevant for European citizens than European identification. It also shows that national identification is particularly strong in Finland, Greece, Bulgaria and Turkey.

National pride was measured with the question 'How proud are you to be a country citizen?', with four answer categories, ranging from not proud at all, to very proud. Ethnic nationalism was measured by a question introduced by the instruction 'Some people say the following things are important for being truly [nationality]. Others say they are not important. How important do you think each of the following is?'. Respondents rated the importance of various criteria from 1 (very important) to 4 (not important at all). Ethnic nationalism was indicated

Figure 1. National and European belonging by country. Source: European Value Survey, 2008.

4 3.5 -3 - — 2.5 -■ 2 -■ 1.5 -■ 1 -■ 0.5 -■ il li II

ssss/ss/s/ / ^r ■ Ancestry ■ Being born in countr y ■ Language □ Respecting law

Figure 2. Importance of national ancestry, being born in the country, speaking the language and respecting political institutions and laws for being fully country's national.

by the importance attached to having national ancestry. The other criteria people could rate are 'being born in the country' and 'to be able to speak the national language(s)'. We also control for the criterion 'to respect country's political institutions and laws'. Figure 2 shows the differences between the included countries in how the populations attach importance to these criteria for being truly a member of the national group. In general, people attach more importance to language and respecting the country's political institutions and laws. In Eastern European countries, the population makes a less clear distinction. In Bulgaria, Croatia,

Greece, Poland, Romania, Serbia and Turkey, respondents value ancestry and being born in the country as relevant as speaking the national language and respecting country's political institutions and laws. The correlations between the three aspects of nationalistic attitudes (identification, pride and ethnic nationalism) are in general modest. The strongest correlation exists between national pride and ethnic nationalism (r = 0.27). All other correlations are positive and significant as well, but are weaker in strength.

Other socio-political attitudes: In this study, we control for other attitudes that previously have been found to be central in explaining radical right support (Werts et al., 2012). Regarding individual perceptions, first, perceived migrant threat was measured by five bipolar items ('immigrants take jobs away from natives in a country' versus 'immigrants do not take jobs away from natives in a country'; and in the same bipolar manner: 'a country's cultural life is [not] undermined by immigrants'; 'immigrants [do not] make crime problems worse'; 'immigrants are [not] a strain on a country's welfare system' and 'in the future the proportion of immigrants will [not] become a threat to society') (Cronbach's alpha = .87). We also include a measurement on the attitude towards Muslims. Although this measure as a separate motive for voters to choose the radical right has not been tested empirically very often, we think it is relevant since many of the radical right parties demarcate the in-group by opposing it to Muslims in European countries. Respondents were asked about their social distance towards a variety of groups with the question whether they would not like to have them as neighbours. When Muslims were mentioned we coded that as social distance towards Muslims. Third, we included euro-scepticism, measured with five items referring to fears related to the building of the EU ('A loss of social security'; 'A loss of national identity and culture'; 'Our country paying more and more to the EU'; 'A loss of power in the world for the country' and 'A loss of jobs in the country') (Cronbach's alpha = 0.89). Next, authoritarian tendencies were measured with three items asking respondents about the preferred political system ('having a strong leader who does not have to bother with parliament and elections', 'having experts, not government, make decisions according to what they think is best for the country' and 'having the army rule the country'). The four answer categories ranged from very good to very bad, with higher scores referring to more authoritarian tendencies (Cronbach's alpha = 0.58). Finally, we included political distrust, measured by the items 'confidence in parliament' and 'confidence in political parties' (r = 0.54). All attitudes with at least semi-interval measurement level were Z-standardized, resulting in scales with means of zero and standard deviations of one. The online appendix contains the descriptive statistics.

Structural background characteristics: The highest completed educational level was measured with the international comparable ISCED (International Standard Classification of Education) scheme. We applied the ISCED 2-digit code and grouped smaller categories together, resulting in 10 categories from 0 'pre-primary

education or no education' to 9 'second stage of tertiary education'.2 With regard to social class, we distinguished between people presently employed and non-employed people. For those currently employed, we applied the nominal class typology of Erikson, Goldthorpe and Portocarero (EGP, 1979). For those currently not employed, we distinguished between unemployed people and a residual category consisting of retired people, people working in their own household and looking after their children, students, disabled people, other social positions and employed people whose EGP score was missing.

Finally, we included sex, age, migrant status and church attendance as control variables in the analyses. With regard to age, a random sample of the adult population was drawn in each country, except for Finland (18-74 years of age). To prevent outliers on the age variable, we recoded the lower and upper values to respectively 18 and 90 years of age. The variable migrant status distinguishes between respondents whose parents were born in the country of residence and first and second generation migrants. Church attendance ranged from attending '(practically) never' (1) to 'more than once a week' (7).

Macro level characteristics: We include the average public opinion in a country regarding national pride, an ethnic conception of nationhood (regarding the relevance of having country's ancestry), migrant threat, distance toward Muslims and euro-scepticism. This was calculated by aggregating the individual level attitudes to the country level. We control for two macro-level characteristics that have been included often in previous research: the level of unemployment and the percentage of foreign-born population. These figures have been derived from Eurostat (2015a, 2015b). Eurostat did not provide the unemployment figures for Switzerland and Serbia. These figures were derived from the Wordbank. It were also the countries with lowest (3.4%) and highest (13.6%) unemployment rates, respectively. The foreign-born population figures are available for most countries from 2009 onwards. For France, Latvia and Austria figures from 1 January 2008 were already available, for the other countries for 2009. Exceptions are Bulgaria (2011), Slovakia (2010) and Switzerland (2011), for which the figures had become later available. For Turkey and Serbia, we relied on UN statistics. The foreign-born population was largest in Switzerland (24.7%) and lowest in Romania (0.8%).

Methods and results

To account for clustering within countries, we use multilevel analysis. It estimates the variance between countries in the extent of radical right voting. To test whether nationalistic attitudes are associated with radical right voting, we first build a model not controlled for the other attitudes. We then add the attitudes that have been found to affect radical right voting in previous research, to find out whether the effects of nationalistic attitudes remain. In a final model, we include the random slopes and the cross-level interactions, to test to what extent the effect of national identification is dependent on the national public opinion.

12 10 8 6 4 2 0

■ National identification

12 10 8 6 4

Not at Not Quite Strong

all very proud proud proud

I National pride

12 10 8 6 4 2 0

I National ancestry importance

Figure 3. Bivariate associations between nationalistic attitudes (national identification, national pride and national ancestry importance) and vote share for the radical right.

We first show descriptive findings providing evidence whether under the condition of absence of national identification and absence of patriotism, the vote for the radical right is absent as well, and what the bivariate association is between the nationalistic attitudes and the vote for the radical right. Figure 3 shows that the associations between three presented nationalistic attitudes and voting for the radical right are positive. With stronger identification, more national pride and more importance attached to national ancestry, support for the radical right is stronger. The association does not seem to be very strong though, in particular not where it concerns national identification. Here we see that also among people who do not mention the nation to be important, a share does support the radical right, refuting hypothesis 1a. Radical right voting also takes place among people with no love for the nation, contradicting our hypothesis 2a and contrasting our expectations that national identification and love for the country are necessary conditions to vote for the radical right. In line with hypotheses 1b and 2, we do find overall positive associations between national identification and pride and support for the radical right.

Table 2 presents the results of the analyses, showing to what extent the attitudes are related to radical right voting under control of each other and other characteristics. Model 0 shows that there is quite some variance between the European countries in the level of voting for the radical right, which is in line with actual election turnouts. In Model 1, we included the nationalistic attitudes together with the sociological background characteristics. National identification is not related to voting for the radical right, although its effect is almost significant. A stronger identification with the nation, as compared to other geographical units amongst which the EU, does hardly increase the likelihood of a vote for the radical right. National pride has the anticipated effect, although also this effect size is modest and it is significant only at p < .05. An ethnic definition of nationhood, by highlighting

Table 2. Unstandardized parameter estimates from logistic multi-level models on Radical Right-wing voting in 20 countries.

Model 0 Model 1 Model 2

b SE b SE b SE

Individual perceptions

National identification (z) .055 .029 .062* .030

National pride (z) .091* .033 .099** .033

Relevance of national .228** .040 112** .041

ancestry (z)

Relevance of being born in the .135** .038 .025 .039

country (z)

Relevance of ability to speak .186** .038 127** .038

the national language (z)

Relevance of respecting national -.094** .031 -.064* .032

institutions and law (z)

Perceived migrant threat (z) .449** .037

Social distance to Muslims .481** .067

Euro-scepticism (z) 2||** .036

Authoritarianism (z) .116** .034

Political distrust (z) .275** .033

Macro-level characteristics (controls)

Level of unemployment — 277** .119 -.358** .128

Percentage of foreign born .194 .107 .166 .116

Individual characteristics (controls)

Educational level -.067** .015 -.023 .015

Social position (higher controllers = ref.)

Lower controllers .154 .137 .114 .140

Routine non-manual .318 .167 .262 .170

Lower sales service .291 .170 .|8| .173

Self-employed people .385* .166 .298 .170

Skilled manual workers, .350* .151 .215 .154

supervisors

Unskilled manual workers .469** .149 .313* .153

Unemployed people .133 .172 -.009 .174

Other .158 .129 .084 .132

Migrant status

No migrant (ref.)

Second generation -.166 .127 -.098 .129

First generation -.210 .163 -.123 .166

(continued)

Table 2. Continued

Model 0 Model 1 Model 2

b SE b SE b SE

Church attendance -.083** .017 - .064** .017

Gender: Male (female = ref.) .358** .060 .366** .062

Age -.012** .002 - .014** .002

Intercept -2.313 0.130 -1.669 .216 -2.013 .224

ct2u0 - level 2 variance 0.324 0.108 0.240 .081 0.284 .096

(countries)

CT2e - level 1 variance 1.00 1.00 1.00

(respondents)

(z), standardized independent variable. N = 15,613; **p < .01, *p < .05 (two-sided).

the relevance of national ancestry to become a truly national, affects the likelihood to vote for the radical right strongly. From the other criteria that people value relevant to be a truly national, 'being born in the country' also affects radical right voting positively. The same holds for the aspect of language; voters who attach more relevance to understanding the national language in order to be seen as a truly national, are more likely to vote for the radical right. Attaching more importance to 'respecting political institutions and laws' as a civic definition of nationhood reduces the likelihood to vote for the radical right. For the background characteristics, this model replicates findings from earlier research, showing that lower educated, manual workers and men are more likely to vote for the radical right. It also shows that the older one is, the less likely it is to vote for the radical right. People who attend church more often are less likely to vote for the radical right. On the macro-level, we do find, as some previous studies found as well, that the unemployment level in a country is negatively associated with the likelihood to vote for the radical right. The effect from the percentage of foreign-born people is positive but (just) not significant.

In Model 2, we added the other socio-political attitudes. The question now is whether the effects of national pride and the relevance of ancestry, being born in the country and language as criteria for nationhood remain significant when the major other explanations of radical right voting are included, related to threat perceptions, authoritarianism and political distrust. Model 2 shows that perceived migrant threat, social distance to Muslims, euro-scepticism, authoritarianism and political distrust have the effects as anticipated, of which migrant threat is the strongest predictor. Our interest goes to the effects of nationalistic attitudes. Under control of all the socio-political attitudes, national identification is now (just) significantly related to voting for the radical right, which is in line with

hypothesis 1b. We found that the effect of national pride remains (and is even somewhat larger than in Model 1), corroborating hypothesis 2. Hypothesis 3a is also not refuted. The effect of the relevance people attach to ancestry is markedly smaller in Model 3 than in Model 2, indicating its strong relation with the other attitudes included, but its effect remains significant. The relevance of being born in a country for nationhood is fully explained by the included attitudes, refuting hypothesis 3b. This does not hold for the relevance people attach to language, that effect remains, supporting H3c.

Comparing the effect sizes, however, it shows that perceived migrant threat and social distance to Muslims have the strongest contribution in explaining radical right-wing voting. As is often argued, political distrust is a good runner-up. The role of the nationalistic attitudes is smaller and more comparable to the role of authoritarianism, just somewhat smaller than the effect of Euroscepticism. Concluding we can state that nationalistic attitudes do play a role in voting for the radical right, and we also showed that national pride and ethnic conceptions of nationhood are more important than the level of national identification in itself, but the direct contribution of nationalistic attitudes in explaining radical right voting is modest.

Finally, we investigated the effect of the national public opinion climate on the relation between national identification and voting for the radical right. Overall, national identification turned out to have a (small) effect on radical right voting. We also anticipated that when people identify with the nation strongly, they will turn to radical right parties in particular when the public opinion in a country expresses a stronger ethnic perception of the nation. We therefore included a random slope of national identification to model the variance in the effect of national identification between countries. We tried to explain this variance, by including an interaction between national identification and the public opinion on the relevance of national ancestry for nationhood. First, we have to conclude that the effect of this ethnic nationalistic public opinion is insignificant. The degree to which the general public in a country adheres to a more ethnic exclusive definition of nationhood is not associated with a larger likelihood to vote for the radical right. Second, the effect of national identification is not dependent on this aggregate ethnic perception of the nationhood in a country, refuting Hypothesis 4. The finding that the effect of national identification varies across countries is hence not due to the differences between the countries in the public opinion regarding the ethnic conception of the nationhood. We also tested whether the effect of national identification is stronger in countries with a public opinion that is more unfavourable towards migrants (level of migrant threat), or where there is on average more social distance to Muslims and a more euro-sceptic public climate. The interactions between national identification and public opinion on Muslims and the EU were not significant (Models 3b and 3c; refuting H5b and H5c). However, as shown in Table 3 (Model 3a), we do find that the effect of national identification is stronger in countries where the public perceives more migrant threat (supporting hypothesis 5a). With a one standard deviation increase in average public opinion

Table 3. Unstandardized parameter estimates from logistic multi-level models on radical right-wing voting in 20 countries.

.268 .032 1.00

Individual perceptions

National identification (z) .075 .046 .073 .047 .076 .051

Country-level characteristics

Public opinion on migrant threat —.133 .138

Public opinion on migrant threat x .107* .047

national identification Public social distance to Muslims —.046 .128

Public social distance to Muslims x —.092 .047

national identification

Public opinion on euro-scepticism —.160 .121

Public opinion on euro-scepticism x .047 .051

national identification Intercept

ct2u0 - level 2 variance (countries) .298 .100 .290 .098

Random slope of national identification .023 .013 .024 .013 CT2e - level 1 variance (respondents) 1.00 1.00

.091 .016

Note: Controlled for all individual level characteristics and national level of unemployment and percentage of foreign born as displayed in Table 2. N = 15,613; *p < .05 (two-sided).

Probability to vote radical right 0 16

........-""■"■M

2 -1.5 -1 -0.5 0 0.5 1 1.5 2 National identification .........Linear (Min. Public opinion migrant threat) Linear (Max. Public opinion migrant threat)

Figure 4. Predicted probabilities to vote radical right as a function of national identification, under conditions of a public opinion with low levels of perceptions of migrant threat (dashed line) and a public opinion of high levels of perceptions of migrant threat (solid line).

on migrant threat, the effect of national identification increases with .107. This means that in countries where on average migrants are stronger perceived as a threat, a solider sense of belonging to the nation is more strongly associated with the vote for the radical right. The reverse holds for countries with a public opinion expressing less migrant threat: there it is estimated that national identification lowers the likelihood to vote for the radical right. Figure 4 expresses this relation between national identification and support for the radical right in the different contexts of public opinion. It shows that national identification is not inherently related to a stronger likelihood to vote for the radical right. Only in countries where perceptions of migrant threat are widespread among the general public, a stronger sense of national identification is related to a higher likelihood to vote for the radical right

Conclusions and discussion

Radical right-wing parties campaign most strongly against immigration and Islam, and are known for their anti-EU stances and anti-establishment rhetoric. From that perspective, it is no surprise that previous research found repeatedly that antimigration attitudes, euro-scepticism and political cynicism are the strongest explanations of radical right voting. However, both radical right-wing parties themselves and social science theorists have emphasized the role of nationalist ideology. In the nationalist ideology, the nation with its ascribed characteristics is regarded as the key unit to structure socio-cultural life, and is opposed to larger units as the EU. The national characteristics should, according to the nationalist ideology of the radical right, be preserved and protected. Identity and pride in the nation, symbolized by abundant presence of the flag in campaign material, are at the core of the programs. Moreover, most of the radical right parties take exclusionist stances in their conceptions of nationhood, mostly favouring an ethnic conception. We therefore question to what extent voters with nationalistic attitudes were more in favour of voting for the radical right - surprisingly hardly studied in previous research. We find that people who identify stronger with the nation are somewhat more likely to vote for the radical right, but the effect is small. It is questioned to what extent people perceive themselves to belong to the nation, in competition with other (more local or international) geographical areas. People who picked the nation as most important to belong to are found to be more likely to vote for the radical right. Sometimes national pride is considered as a part of national identification as well; our results indicate a weak correlation though between identifying stronger with the nation and national pride. We test its effect here separately and find a (weak) association with radical right voting; the more pride in the nation, the more likely voters are to vote for the radical right. Finally, we distinguish between various conceptions of nationhood. As anticipated, our findings indicate that the more people value ancestry as relevant for nationhood, the more likely they are to vote for the radical right. In line with radical right parties' rhetoric about a nation's shared ancestry and culture, the importance attached to speaking the national

language was also positively related to support for the radical right. These associations between national identification, national pride and the relevance of ancestry and language on the one hand, and radical right voting on the other hand, remained when controlled for what turned out the strongest predictors of radical right voting: perceived migrant threat, social distance to Muslims, political cynicism and euro-scepticism.

Our results show that firstly, certain nationalistic attitudes are relevant in explaining radical right voting, but in line with earlier research, national identification and love for the country are hardly related to expressions of intolerance. Secondly, we find that it is not so much the degree of national identification that matters, but more the extent to which people attach pride to their nation and their ethnic or civic conception of nationhood. Thirdly, although on average the role of nationalistic attitudes is modest, national identity is more relevant in countries with a more negative public climate towards immigrants. Since European populations identify stronger with the nation than for example with Europe, changes in public opinion on migration may cause a larger shift in support for the radical right than might be thought based solely on the larger share of the population perceiving stronger migrant threats.

The role of nationalistic attitudes is complicated. Radical right supporters generally have exclusionist conceptions of the nation, but not always identify first and foremost with the nation. In that perspective they may be seen to have a negative conception of identity (Bruter and Harrison, 2011). Often they attach more importance to the region or local communities and do not express pride in the nation. Perhaps because they long to a nostalgic construction of the nation: they are proud of what the nation once was, not of the current state of the nation (Meuleman and Lubbers, 2013). From the ideologies of nativism (Mudde, 2007) and multi-pluralism (Rydgren, 2007) in which it is stated that the radical right ideologies depart from preserving the unique characteristics of the nation and protect it to possible threats, in particular the part describing the threats turn out the most relevant in explaining radical right voting. Threat from migration and Muslims and threats from the EU, next to political distrust, remain the most important predictors of voting for the radical right. For radical right voters, nationhood is strongly defined in terms of ancestry and speaking the country's language. But this latter criterion resonates among a much broader public than among the radical right's electorate only.

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

1. The data are from 2008. This implies that the Greek Golden Dawn was not relevant yet. For the UK, the British National Party and the United Kingdom Independence Party (UKIP) could not be derived since these were merged in the category 'other'. For Belgium, we only included the Flemish region and with that excluding the Belgian Front National.

2. The categories are 'pre-primary education or no education' (0), 'primary education or first stage of basic education' (1), 'lower secondary or second stage of basic education' (2), 'pre-vocational or pre-technical (upper) secondary education' (3), 'vocational or technical (upper) secondary education' (4), 'general (upper) secondary education '(5), 'post-secondary non-tertiary education' (6), 'first stage of tertiary education (vocational)' (7), 'first stage of tertiary education (general)' (8) and 'second stage of tertiary education' (9). We treated educational level as an interval variable since there were no relevant deviations from linearity.

References

Anderson B (1983) Imagined Communities: Reflections on the Origin and Spread of Nationalism. London: Verso.

Ariely G (2012) Globalisation and the decline of national identity? An exploration across sixty-three countries. Nations and Nationalism 18(3): 461-482.

Billiet J and Witte H (1995) Attitudinal dispositions to vote for a 'new' extreme right-wing party: The case of'Vlaams Blok'. European Journal of Political Research 27(2): 181-202.

Brewer MB (1999) The psychology of prejudice: Ingroup love or outgroup hate? Journal of Social Issues 55(3): 429-444.

Brewer MB (2001) Ingroup identification and intergroup conflict. In: Ashmore RD, Jussim L and Wilder D (eds) Social Identity, Intergroup Conflict, and Conflict Reduction. New York: Oxford University Press, pp. 17-41.

Brown R (2000) Social identity theory: Past achievements, current problems and future challenges. European Journal of Social Psychology 30(6): 745-778.

Brubaker R (1992) Citizenship and Nationhood in France and Germany. Cambridge: Harvard University Press.

Bruter M and Harrison S (2011) Mapping Extreme Right Ideology. An Empirical Geography of the European Extreme Right. Basingstoke: Palgrave Macmillan.

Citrin J and Sides J (2004) More than nationals: How identity choice matters in the new Europe. In: Herrmann RK, Risse T and Brewer MB (eds) Transnational Identities: Becoming European in the EU. Lanham, Maryland: Rowman & Littlefield Publishers, pp. 161-185.

Coenders M and Scheepers P (2003) The effect of education on nationalism and ethnic exclusionism: An international comparison. Political psychology 24(2): 313-343.

Curtis KA (2014) Inclusive versus exclusive: A cross-national comparison of the effects of subnational, national, and supranational identity. European Union Politics 15(4): 521-546.

De Figueiredo RJ and Elkins Z (2003) Are patriots bigots? An inquiry into the vices of ingroup pride. American Journal of Political Science 47(1): 171-188.

Dekker H, Malova D and Hoogendoorn S (2003) Nationalism and its explanations. Political Psychology 24: 345-376.

De Vreese CH and Boomgaarden HG (2005) Projecting EU referendums fear of immigration and support for European integration. European Union Politics 6(1): 59-82.

Erikson R, Goldthorpe JH and Portocarero L (1979) Intergenerational class mobility in three Western European societies: England, France and Sweden. British Journal of Sociology 30(4): 415-441.

Eurostat (2015a) Total unemployment rate. Available at: http://ec.europa.eu/eurostat/ (accessed 10 November 2015).

Eurostat (2015b) Population on 1 January by five year age group, sex and country of birth [migr_pop3ctb]. Available at: http://ec.europa.eu/eurostat/ (accessed 10 November 2015).

EVS (2011) European values study 2008: Integrated dataset (EVS 2008). GESIS Data Archive, Cologne. ZA4800 Data file version 3.0.0. DOI: 10.4232/1.11004.

Ford R, Tilley J and Heath A (2011) Land of my fathers? Economic development, ethnic division and ethnic national identity. Sociological Research Online 16(2): 8-20.

Gibson JL (2006) Do strong group identities fuel intolerance? Evidence from the South African case. Political Psychology 27(5): 665-705.

Hjerm M and Schnabel A (2010) Mobilizing national sentiments: which factors affect nationalist sentiments in Europe? Social Science Research 39(4): 527-539.

Immerzeel T (2015) Voting for a Change: The Democratic Lure of Populist Radical Right Parties in Voting Behavior. Utrecht: UU.

Ivarsflaten E (2005) The vulnerable populist right parties: No economic realignment fuelling their electoral success. European Journal of Political Research 44(3): 465-492.

Ivarsflaten E (2008) What unites right-wing populists in Western Europe? Re-examining grievance mobilization models in seven successful cases. Comparative Political Studies 41(1): 3-23.

Janmaat JG (2006) Popular conceptions of nationhood in old and new European member states: Partial support for the ethnic-civic framework. Ethnic and Racial Studies 29(1): 50-78.

Kunovich RM (2009) The sources and consequences of national identification. American Sociological Review 74(4): 573-593.

Meuleman R and Lubbers M (2013) Manifestations of nationalist attitudes: domestic music listening, participation in national celebrations, and far right voting. European Sociological Review 29(6): 1214-1225.

Mudde C (2007) Populist Radical Right Parties. Cambridge: Cambridge University Press.

Norris P (2005) Radical Right: Voters and Parties in the Electoral Market. Cambridge: Cambridge University Press.

Pherson S, Vignoles VL and Brown R (2009) National identification and anti-immigrant prejudice: Individual and contextual effects of national definitions. Social Psychology Quarterly 72(1): 24-38.

Reeskens T and Hooghe M (2010) Beyond the civic-ethnic dichotomy: Investigating the structure of citizenship concepts across thirty-three countries. Nations and Nationalism 16(4): 579-597.

Rydgren J (2007) The sociology of the radical right. Annual Review of Sociology 33: 241-262.

Rydgren J (2008) Immigration sceptics, xenophobes or racists? Radical right-wing voting in six West European countries. European Journal of Political Research 47(6): 737-765.

Smith AD (2001) Nationalism: Theory, Ideology, History. Cambridge, UK: Polity Pres.

Smith TW (2007) Social identity and socio-demographic structure. International Journal of Public Opinion Research 19(3): 380-390.

Solt F (2011) Diversionary nationalism: economic inequality and the formation of national pride. The Journal of Politics 73(3): 821-830.

Tajfel H (1982) Social psychology of intergroup relations. Annual Review of Psychology 33(1): 1-39.

Tajfel H and Turner J (1979) An integrative theory of intergroup conflict. In: Austin W and Worchel S (eds) The Social Psychology of Intergroup Relations. Monterrey: Brooks/Cole, pp. 33-47.

Van der Brug W, Fennema M and Tillie J (2000) Anti-immigrant parties in Europe: Ideological or protest vote? European Journal of Political Research 37(1): 77-102.

Weldon SA (2006) The institutional context of tolerance for ethnic minorities: A comparative, multilevel analysis of Western Europe. American Journal of Political Science 50(2): 331-349.

Werts H, Scheepers P and Lubbers M (2012) Euro-scepticism and radical right-wing voting in Europe, 2002-2008: Social cleavages, socio-political attitudes and contextual characteristics determining voting for the radical right. European Union Politics 14(2): 183-205.