Religion, Values and Politics: The Effect of Religiosity on Attitudes Towards Immigrants and Muslims

Abstract

The European Values Study demonstrates a significant impact of religiosity on political attitudes. As an example, the authors investigate the effect of religiosity on attitudes towards immigrants and Muslims, as they are crucial for liberal democracy and a focal point of conflicts about values, politics, and religion. In three steps, the authors reflect on the complexity of this effect. First, selected theories about secularisation, individualisation, and pluralisation of religion are used to outline the development of religion in Europe and its role in the political arena. Second, the effect of religiosity in relation to sociodemographic factors is analysed. A cluster analysis forming socioreligious types offers a deeper insight, quantitatively and qualitatively, of the effect of religiosity on attitudes towards immigrants and Muslims in selected countries. Third, the results are theoretically interpreted. The authors demonstrate that religiosity is neither an independent influencing factor nor one that directly affects attitudes towards immigrants and Muslims. Rather, across Europe, the effect of religiosity is inseparably connected with sociodemographic variables such as ‘age’ and ‘size of town’ and depends on country-specific factors such as political discourses on migration, and religious historical constellations.

1 Purpose, Objectives, and Context

The effect of religiosity on political attitudes and values as recorded in the European Values Study (EVS) is this volume’s focus. In our contribution, we investigate whether religiosity affects political attitudes directly by utilising in-depth random samples based on the EVS data. With this approach, we wish to encourage a more interdisciplinary and multi-perspective interpretation of quantitative data. We consider it necessary to set up this broader heuristic framework because our ‘spot checks’ will demonstrate that universal arguments based on sociological theories of religion are becoming increasingly less adequate for explaining the diverse developments of religion in Europe.

The starting point of our considerations is the concept of ‘embedded religion’ as it is discussed in sociology (Giddens 1991; Madsen 2009), religious and cultural anthropology (Taylor 2007), religious studies (Coe 2012–2013), and practical theology (Zulehner 1989). According to Giddens (1991), the process of modernity leads to a disembedding of social institutions; for example, social relationships break away from their local context and individuals enter a reflexive relationship with institutions. This process also affects religion. Since the Reformation, religion has been disconnecting from local ties and membership of small-scale social groups has been eroded. Consequently, the natural shaping of everyday life through a religiously structured culture – shaped through rituals, symbols, and daily life practices – has gradually declined. Individuals change their relationship with (church) institutions while religious affiliations and religious world views decrease (Taylor 2007).

However, this process runs dialectically (Giddens 1991). While the social orders of ancient and medieval religions erode, new constellations arise. New socioreligious orders, new institutionalisations, new functions, and semantics of religion are emerging. This dialectic can also be observed today as religiosity increasingly loses its position as a relevant area of life, while subjective religiosities are simultaneously taking on new functions and meanings in transforming cultural and political contexts – independently of the traditional churches but induced by political and media discourses.

For our purposes, the dialectic concept of ‘embedding’ and ‘disembedding’ of religion is hermeneutically important because it argues that religiosity is not a ‘pure’ or even primordial social reality but is inextricably linked with sociocultural, political, and historical contexts. This is also assumed by Coe (2012–2013): ‘Along with internal diversity and change over time, another fundamental tenet of a religious studies approach is to recognise the ways that religions are embedded in human cultures and not isolated in a discrete private sphere.’ As values are at the core of human cultures, religiosity is always deeply connected with them and dependent on other variables.

Finally, a practical-theological approach suggests that the vitality of (not only) Christian religiosity requires corresponding social ‘structures of plausibility’; for example, without being embedded in religiously structured, everyday cultures and religious communities, religiosity as a lived practice ‘evaporates’. This is based on the social structure of religiosity in modern societies, which are characterized by uncertainty and find themselves under a ‘heretical imperative’ forcing people to freely and consciously choose their faith. (Zulehner 1989, following the sociology of knowledge of Peter L. Berger and Thomas Luckmann).

The socioreligious developments documented by the EVS since the 1980s substantiate these theories, in particular for the religiosities in Western Europe that have traditionally been structured in a Christian ecclesiastic form. After a continuous erosion of institutional ties and affiliations, we can witness a constant decline of religious practice (prayer and attending church services), denominational self-image, and belief in God, as these are increasingly less socially embedded and supported (Polak and Schachinger 2011). This development was also clearly evident in the EVS 2017 (Pickel and Pickel, Chap. 5, this volume). After the continuous erosion of traditional church affiliation and forms of practice, belief in God – relatively stable in Europe for decades – is declining in many countries, even in those that have traditionally been dominantly Catholic. This can be seen in Table 6.1.

Table 6.1 ‘Do you believe in God: yes’ (v57) between 1990 and 2017, in per cent (EVS 1990–2017)

However, the need for an explanation arises when examining the noticeable country-specific differences and the ‘outliers’ that demonstrate increased belief in God between 2008 and 2017 (East Germany, Czechia, Sweden, Finland, Greece). These heterogeneous developments are also evident in other religious indicators (Pickel and Pickel, Chap. 5, this volume). Based on the theory of the dialectic between the embedding and disembedding of religion, the question arises as to whether the intensity of religiosity in the participating countries depends on other factors that shape religiosity in a manner specific to these countries. Are there any country-specific social, cultural, and political formations that can revitalise religiosity or belief in God? Can religious belief that has been decoupled from the Protestant churches be revived through new religious forms of practice? Does the political discourse of the right extremist populist party ‘Alternative für Deutschland (AfD)’ in what was East Germany, with its recourse to the ‘Christian Occident’, also affect the level of subjective religious self-assessment? Do the political references to ‘Christian values’ promoted in many European countries in migration policy discourses also change religious self-assessments?

These questions cannot be answered directly by the EVS. Nevertheless, they become relevant, reflecting the empirical findings presented by Susanne and Gert Pickel (Chap. 5, this volume). These authors prove the significant effect of religiosity on political values and attitudes relevant for liberal democracies. They show:

Prejudice and collective defence, such as those exercised against migrants and people who do not share established traditional values, provide a bridge for proximity to right-wing populists, but also a bridge to religion. Religious ideas work in two directions here: while socially engaged and thinking believers are pro-democratic and against prejudice, fundamentalist and dogmatic believers develop an elective affinity with right-wing beliefs that are anti-pluralist and then also anti-democratic. (Pickel and Pickel, Chap. 5, this volume: 194)

Thus, religious attitudes can strengthen pro-democratic attitudes but also promote beliefs that contradict the ideas of liberal democracy.

The connection between the so-called ‘normative-religious dimension’ and attitudes that are incompatible with a liberal understanding of democracy was already proven by the EVS 2008. According to Arts and Halman (2011) persons who have a better-than-average marked religious self-assessment, e.g. they believe in God, go to church regularly and pray often, show a significantly higher trust in authoritative institutions such as the army and the Church; moreover, for these respondents, authority takes precedence over autonomy in their educational values, they are intolerant towards ethnic or social minorities such as immigrants and homosexuals, hold cultural homogeneity as a high value, prefer materialistic over post-materialistic objectives, are opposed to employed mothers and have a solid regional dominating identity compared to a European or even global self-image.

This connection has taken on a politically precarious significance, at least since 9/11. After the terrorist attacks in Madrid (2004) and London (2005), not only has the threat of Islamist political extremism arrived in Europe, but religion, pronounced socially and politically irrelevant by academia since the 1970s, has returned to the stage of politics. In the wake of these terrorist attacks, Islam in particular has become a defining issue in politics – especially noticeable in the policies on migration, security, and integration of several political parties and governments in Europe, such as those in Austria, Germany, and Switzerland (Mattes 2016). In the context of political processes and interests, the focus of religion goes far beyond adaptation to the new challenges of religiously plural societies. Muslims have become synonymous with undesirable immigrants: ‘Populations which are hostile to migrants tend to identify migrants with Muslims (and Muslims with migrants)’ (Pickel 2018: 35). A universalised Christian religion has become a renewed feature of collective identity and is used as a symbolic demarcation to construct in- and out-groups (Mattes 2016). Furthermore, the so-called ‘refugee crisis’ in 2015 was used to accelerate the anti-migration dynamics that had already existed during the EU enlargement and resulted in a ‘religionisation’ of migration policies, not least through political framing and a media representation of Muslims as terrorists: ‘The refugee crisis has contributed to a coagulation of position against Muslims’ (Pickel 2018: 35). This resulted in a majority of European people viewing all refugees as Muslims: ‘Islam is regarded as the most threatening religion, with at least one-third of the population feeling this way’ (Pickel 2018: 28).

Consequently, many social, economic, political, ethnic, and cultural conflicts throughout Europe are interpreted as fault line conflicts (Huntington 1996) between Christian and Islamic cultures and values (Mattes 2016). Right-wing populist parties in particular have successfully served this interpretive scheme for years. Thus, the process of secularisation, which in Europe primarily affects the Christian churches, is accompanied by the development of a front between ‘opponents and proponents of Islam’ (Pickel 2018: 35), with both non-religious and religious people to be found on either side. Religion has become an authoritative line of difference in political discourses, serving as a distinction in political conflicts and becoming the subject of interdisciplinary discourses (Polak and Seewann 2019; Kiesel and Lutz 2015; Eberhardt and Bultmann 2019).

Is the effect of religiosity indeed as strong as it is assumed in these discourses, however? Is religiosity the central source of political attitudes? In the following paragraphs, we explore these questions by way of example. We test the hypothesis that sociodemographic factors and country-specific contexts have a significant effect on whether strong religiosity affects political attitudes and, if so, how. We assume that political discourses on religion also have a significant influence on the concrete shaping of religiosity. Although this influence cannot be tested directly with the EVS data, it will be included in our considerations.

The thesis that the effect of religiosity on political attitudes also depends on other factors has already been proven. For example, religious attitudes only lead to problematic political mindsets when combined with authoritarianism (Canetti-Nisim 2004). Religious identity markers, in turn, have a highly heterogeneous impact on political conflicts; for example, they can strengthen or weaken them, depending on the interpretation and practice of a religious self-concept (Werkner and Hidalgo 2016). The religiosity of religious movements emerges as a growing factor of influence in the context of politics against economic inequality (Ekrem and Birol 2012). Among Christians in the West, religiosity has recently functioned more and more as a cultural identity marker to segregate and exclude immigrants and Muslims (Pew Research Center 2018) and has a regional and heterogeneous (and diminishing) effect on voting behaviour (Pickel 2012). In combination with social engagement, it has a positive influence on social capital (Pickel and Gladkich 2012). Accordingly, a religious self-assessment as such is not sufficient to explain political attitudes. It is a variable dependent on other factors.

In testing our hyoptheses, we focus on the question of how religiosity affects attitudes towards immigrants and Muslims. Based on EVS data from 2008, Stefanie Doebler (2015) found that perceiving religion as important and attending church regularly are more strongly related to racial intolerance in highly religious countries and in countries that have legacies of political instability, violence, a low GDP, and low migration rates. According to her findings, a high religious self-assessment has the most influential impact on racial intolerance, whereas religious practice has no relevant effect. However, according to our findings of the effect of sociodemographic factors, her results will have to be put into perspective.

Our focus on attitudes towards immigrants and Muslims is also relevant because these attitudes have become a core issue of serious political conflicts and reflect attitudes towards religious and cultural diversity. The recognition of diversity, however, is at the core of liberal democracies (Müller 2017). Pursuant to Article 2 of the Treaty of Lisbon, it is also included among those values on which the European Union is founded. Therefore, attitudes towards immigrants and Muslims do not merely concern moral aspects of tolerance but are at the heart of a policy of recognising diversity as an essential core of liberal democracies (Honneth 2010; Taylor 1992).

Attitudes towards immigrants and Muslims are not independent values, however. Rosenberger and Seeber (2011), for example, demonstrated that the adoption of antipathetic attitudes towards minority groups in Western European democracies depends neither on the size of the respective minority within a country nor on the strength of a right-wing populist party. Rather, the adoption by mainstream parties of right-wing populist and xenophobic motives in their political argumentation is decisive. Since the claim of a conflict between Christian and Islamic values plays a central role in these argumentations, again questions arise. Is it primarily the effect of religiosity that increases the rejection of immigrants and Muslims? Is the adoption of right-wing populist and xenophobic rhetoric by mainstream parties sufficient to explain the rejection of immigrants and Muslims?

To answer our questions, we structure our paper into three sections:

• (a) The first section discusses theories of secularisation and individualisation to ask whether these are sufficient to explain the progressive decline of religiosity in Europe. Furthermore, we seek for an explanation for the paradox that religiosities are eroding across Europe while religion is gaining importance on the level of political discourses – not least in the context of migration. We assume that adding a theoretical pluralisation approach to the interpretation of the data could provide part of the answer here. Finally, we present a theoretical model of the phenomenon of the religionisation of politics and the politicisation of religion.

• (b) In the second section, we examine the effect of religiosity, gender, age, income, and the size of town on attitudes towards immigrants and Muslims as an example to test our assumptions by using the EVS data (1). Examining the effect of political self-positioning on attitudes towards immigrants and Muslims in selected countries opens a relativising perspective on the effect of religiosity (2). To explore this further, a cluster analysis comprising socioreligious types (3) and testing the attitudes of these socioreligious types towards immigrants and Muslims, including exemplary country evaluations (4) will finally document the heterogeneous effect of religiosity in Europe.

• (c) In the third section, we interpret our findings and hope to sketch a differentiated picture of the effect of religiosity on political attitudes using the example of the rejection of immigrants and Muslims. We will make an argument for the need for further interdisciplinary research, as the results cannot be interpreted by theories of the sociology of religion alone.

2 Theoretical Framework

2.1 The Development of Religion in Europe from the Perspectives of Theories of the Sociology of Religion

The transformation of ‘religion’ in Europe has been discussed controversially in the sociology of religion over many decades. In their contribution, Susanne and Gert Pickel (Chap. 5, this volume) present secularisation theories, individualisation theories, and the market model of religion. We question the first two and then add pluralisation theory approaches. The market model seems of less importance, as in our view it ignores too much the cultural, historical, and political impacts on religiosity, that is its embeddedness. Regardless of contemporary economic tendencies in the socioreligious field, religion is never just a commodity that is consumed.

• (a) Theories of secularisation

Classical theories of secularisation interpret secularisation as a consequence of modernisation processes (Pickel and Pickel, Chap. 5, this volume). In this process, religiosity does not necessarily lose significance for individuals, but it does lose relevance for society. However, because of a lack of structural and institutional anchoring in society, the loss of societal significance also leads to a decline in individual religiosity as a structuring factor for everyday life in successive generations. This development can also be explained with the model of (dis)embedded religion. According to Charles Taylor (2007), secularisation in this sense has had emancipatory consequences in the West since the Reformation, leading to the Enlightenment, the emergence of a scientific world view, greater individual freedom, and the pluralisation of life arrangements.

However, classical secularisation theories tend to overlook the dialectics of the process: the losses to which the process of disembedding leads can simultaneously force processes of re-embedding religious norms and practises in new contexts and formats, not least political ones. Religiosity can then regain social and political significance despite an individual decline in relevance. This is the main argument of the political scientist Olivier Roy (2008) when he explains global religious fundamentalism. He demonstrates that the ‘uprooting’ of religion from cultural contexts need not necessarily result in its disappearance but can even lead to the strengthening of religion. Lacking an embedding in religious traditions and institutions, such a deculturalised religiosity can then be used to serve political functions and interests. This restructuring process is currently most clearly visible in Islamist fundamentalism. But such developments can also be observed among Christians in highly modernised societies in the West: in the politically ambitious Christian fundamentalism of right-wing Christian movements and international networks (Wäckerlig 2019); in political neo-nationalisms that refer to the Christian heritage in Eastern Europe, such as those of Hungary, Poland or Russia (Höhne and Meireis 2020); or, to a lesser extent, in the political recourse to Christian values and identities in Western European countries such as Austria and Germany (Rausch and Varga 2020).

Simultaneously, secularised societies are confronted with the incorporated religiosities of a growing number of immigrants – especially Muslims and Christians from the Middle East, but also Orthodox Christians from Eastern Europe and Catholic immigrants from the global South. Religious immigrants are usually more traditional and conservative regarding their values and attitudes toward family, gender, and politics (Höllinger and Polak 2019). If highly religious, they lean towards vague attitudes towards democracy (Norris and Inglehart 2002) – like highly religious people do in general (Arts and Halman 2011). They disturb the secular understanding of religion, especially in Northern and Western European societies, where traditional religiosities are considered to have been overcome as pre-modern relics (Schreiter 2008). Much more culturally embedded in a religiously structured everyday life, migratory religiosity can arouse an abstract concept of religiosity reduced to cognitive convictions. While despite its traditional character, the religiosity of migrants does not necessarily have to be incompatible with modernity, immigrants sometimes reject a Western secularised religious self-assessment because, from their perspective, it leads to the self-dissolution of faith (Polak 2017). The more self-confidently immigrants then claim public space and relevance for practising their religion, the more sharply visible the lines of conflict around the understanding of secularity and its relationship to religion become.

This calls into question the classical paradigm of secularisation. This paradigm is based on an indissoluble contradiction between a modern and a religious view of the world. However, this contradiction does not seem to exist in such a radical form among many immigrants (and autochthonous persons), who are quite capable of combining their religious way of life with secular norms, values, and principles (Berger 2014).

On a global level too, the social significance of religiosity and religion in secular societies, such as the United States or India, contradicts the classical secularisation thesis (Reder and Casanova 2010; Casanova 2019). In the course of the academic controversies about the so-called ‘return of religion’ around the millennium, Peter L. Berger (1999) revised his theoretical secularisation approach and proposed a global desecularisation. Furthermore, Jürgen Habermas brought the term ‘post-secularity’ into the academic discourse (Renner 2017), characterising the need for liberal societies to reflect their own secularist biases as part of their self-definition and to make the surplus of religious resources fruitful for a secular society. Social and cultural studies also adjusted to the persistence of religious communities and institutions in society and now speak of ‘multiple modernities’ (Eisenstadt 2002) and ‘path-dependent secularisation’ (Pickel 2009). Moreover, secularisation can adopt heterogeneous political functions in different societies (Wohlrab-Saar and Burchardt 2011). It can be: (1) in the service of individual freedoms and rights (as, for example, in the United States); (2) in the service of balancing or pacifying religious difference (as, for example, in India or the Netherlands); (3) in the service of societal or national integration and development (as, for example, in France); or (4) in the service of the independently developing functional areas of society (as, for example, in early modern Western societies).

Finally, case studies in the history of religion and qualitative empirical studies question the universalistic paradigm of secularisation. For example, they argue that the causes of massive church departures in Germany from the second half of the twentieth century onward were not the result of modernisation processes but had social, political, economic, and cultural reasons (Krech 2013). On the other hand, secular school subjects such as ‘Lebenskunde’ (life studies) in Germany were introduced in Protestant rather than in Catholic-dominated states (Schröder 2020), which leads to the conclusion that the decline of religiosity cannot be explained by modernisation but by the removal of the compulsion for religious participation in school (Schröder 2020). Therefore, approaches based on the history of religion cast doubt on the secularisation theory of a universal rationalisation of the world. Secularisation is not a necessary consequence of modernity: ‘Secularisation is struggling, not destiny’ (Schlerka 2016: 132). It is a historically contingent phenomenon, a temporally conditioned episode with heterogeneous causes (Lehmann 2004).

Generally, a shift from theoretical secularisation models to empirical case studies can be observed in the sociological research on religion in recent years. These also include religious, historical, biographical, socio-cultural, and political aspects and developments. Therefore, Norris and Inglehart (2011: 106) state:

To summarise, in postindustrial nations no empirical support that we examined could explain why some rich nations are far more religious than others, and the study failed to establish a significant link between patterns of religious behaviour and the indicators of religious pluralism, religious freedom, and the perceived functions of the church. But, of course, this still leaves us with the question that we considered at the start of the chapter: why are some societies such as the United States and Ireland persistently more religious in their habits and beliefs than comparable Western nations sharing a Christian cultural heritage?

In addition to modernisation processes, cultural value changes are also transforming religious attitudes. Inglehart (2021) attributes the worldwide decline in religion, which he has observed since 2007, primarily to the attitudes of the younger population regarding pro-fertility norms. Since 2007, an overwhelming majority (43 out of 49) of the searched countries – especially high-income countries – became less religious. Almost all high-income societies have recently reached a tipping point where the balance shifts from the dominance of religiously prescribed pro-fertility norms (concerning gender equality, abortion, homosexuality, divorce) to the dominance of individual-choice norms. This value change forces the decline of religiosity, which is losing its influence on norms (Halman and Sieben, Chap. 4, this volume).

Additionally, the most recent international study of the sociology of religion by Detlef Pollack and Gergely Rosta (2022) documents a historically unprecedented secularisation in the religious strongholds of Europe and the United States. The study attributes this development to the rapidly dwindling belief in God and the preceding decline in communal religious rituals, the increasing value of self-determination, enjoyment of life and self-realisation, and a greater range of leisure opportunities. Moreover, as in Poland, too close a relationship between religious communities and political interests results in an increasing loss of liberal or already distant believers in particular.

Therefore, the transformation of religiosity cannot be explained solely by secularisation and modernisation processes. These theories must also consider changes in politics, intellectual history, ethics, and culture that are not just a result of but also an independent source of modernisation. The numerous relativisations of secularisation theories consequently call the universality of secularisation into question and make it necessary to proceed in an interdisciplinary manner when evaluating quantitative data.

• (b) Theories of individualisation

These theories explain the decline of religiosity through a changed relationship between the person and social institutions. Based on the right to personal freedom, people are less willing to submit to institutions (especially hierarchical institutions), and they make individual demands on them. This leads to a decrease in ties to religious communities, the loss of relevance of institutionalised religion for personal life, and the internalisation and privatisation of religiosity. However, the most important representatives of this position (Thomas Luckmann, Grace Davie, and Danièle Hervieu-Léger) assume that individual religiosity remains constant despite the decrease in ties to religious communities. In contrast, we assume that from a certain point in this erosion personal religiosity also ‘evaporates’, as documented by the EVS between 1990 and 2017. From a social science perspective – differing from a theological one – there is no proof that a religiously interpreted relationship to a transcendent reality must be an anthropological constant.

This thesis is supported, for example, by the research of Heiner Meulemann (2018). Based on the EVS data for Western Europe, he proves that the decline in membership of a religious institution is followed, with a time lag, by a decline in religious practice and religious belief towards a diffused religiosity. To Meulemann, this is a contradiction in terms, since, as a product of secularisation, such religiosity does not take over faith and its forms of expression but retains a religious self-assessment without content. Consequently, it does not grow but continuously decreases. Accordingly, diffused religiosity is a transitional stage between religiosity and non-religiosity (Meulemann 2018). Simultaneously, Christian traditions such as rituals for framing biographical transitions are in great demand in Western European societies, increasingly for cultural reasons but unsupported by individual faith (Meulemann 2018). Also, EVS results for Austria support the hypothesis of a gradual disappearance of religiosity and demonstrate that three-quarters of respondents now hold a purely immanentist world view, according to which the meaning of life lies in life itself and the laws of nature (Polak and Seewann 2019). In this context, belief in God appears to have become an abstract idea instead of a reality to believe in.

These developments are closely related to the fundamental change in the status and meaning of religion in society. According to Charles Taylor (2007), a modern world view forces religious people to legitimise the meaning and benefits of religion. Religion has thus come under suspicion for being irrational and pre-modern, especially in Western Europe. Unlike the situation in a society saturated with religion, non-believers no longer need to justify their world view, but conversely, believers are accountable. The associated power relations and cognitive dissonances consequently promote first the privatisation and second the disappearance of religiosity. The accountability of religious people also established itself in the second half of the twentieth century in the Soviet-dominated countries of Europe. Declared a ‘private matter’ in the initial phase of communism and socialism, religion was understood as an irrational counterpart to a scientifically explainable world and would die away in the course of social and economic progress (Schuster 2017). This view was also adopted by socialist thinkers in Western Europe and still plays a relevant role in contemporary and political discourses.

Approaches through individualisation theories thus provide plausible explanations for the transformation of religiosity in Europe, particularly in the transitional phase of diffused religiosity. They show possible points of contact that can serve interests other than genuine religious ones connecting with the ‘remnants’ of religiosity and taking over political functions. However, it is doubtful whether the ‘amount’ of religiosity will be preserved and whether and how the associated phenomena can be defined as religiosity.

• (c) Theories of pluralisation

The previous considerations have reflected the strengths and weaknesses of the theories of secularisation and individualisation. Now we want to add theories of pluralisation, as they take into account the politics and history of religion for the interpretation of data and offer explanations for the disparity and heterogeneity of the contradictory developments in the socioreligious field in Europe.

Based on the sociology of knowldge approach that focuses on the reciprocal relationship between the social construction of reality and the constitution of religiosity, Peter L. Berger (2014) moved to such an interpretation model, which is based on the concept of multiple modernities. Thus, the existence of increasingly religiously and ideologically plural societies and the heterogeneous political discourses and practices responding to these developments also affect subjective religiosities. According to Berger (2014), the paradigm of the pluralisation of religion therefore allows for a better understanding of the connection between individual and political components. Heterogeneous pluralisation processes lead to a multiplication of actors in the socioreligious field. Globalisation, mobility, migration, and digital technologies enable new kinds of contacts, relationships, and forms of communication that dynamise the development of new and hybrid forms of religiosities. Simultaneously, they open numerous new and politically explosive scenarios of conflict, as religious pluralisation also increases the transformation of religious semantics and dynamises new alliances with logics and interests other than religious ones, and particularly with those of a political nature. Religious pluralisation, therefore, confronts the whole world with the challenge of the ‘two pluralisms’ as the central global peace policy challenge, which a twofold dialogue must answer: both the dialogue between religiously diverse people and the dialogue between religious and non-religious people within societies. To complicate this issue, Berger additionally highlights his observation that most religious people also see themselves as secular, which underlines the necessity to develop theories beyond a strict separation between religion and secularity. Also, Robert Wuthnow (2007) sees the development of a reflected religious pluralism as the central challenge of the present. This confronts secular societies and religious communities alike with the task of structural and substantive transformation, the development of a new socioreligious order, and a public debate about the place and status of religion in society.

Unfortunately, the conservative concept of religiosity of the EVS and the respective national samples, with their low representation of religiously diverse groupings, makes it impossible to do justice to this approach in our analysis. But the challenges of religious pluralisation with simultaneous erosion of traditional religiosity in the context of political discourses must be in the hermeneutic background of our considerations. A theoretical pluralisation perspective can contribute to a better understanding of the apparent paradoxes in the development of religiosity in Europe, these being the coexistence of the loss of relevance of religiosity, the increasing political significance of religion, and the contradictory effect on political attitudes.

2.2 Religion, Politics, and Values

Social sciences, religious studies, theology, and Islamic studies have been noting a newly forming, tense relationship between religion and politics due to global political developments for some time (Fox 2018). Depending on the historical and cultural contexts in the United States, Europe, and the Middle East, this relationship can present in heterogeneous forms (Eberhardt and Bultmann 2019). Religion shows an ambivalent double face: on the one hand, religious actors can mediate in political conflicts and promote dialogue and reconciliation between opposing parties, and thus represent a central source for containing and preventing political, socio-economic, ethnic, and cultural tensions (Czada et al. 2012; Weingardt 2007). On the other hand, truth claims and identity patterns force ‘friend-foe constructions’ and promote intolerance, violence, and armed conflicts, as can be seen not least in religiously motivated terror or the Islamic State (Werkner and Hidalgo 2016). Conversely, political contexts also shape religious attitudes and religious communities (Lehmann 2019). Whether religiosity is an independent variable, which political interests can instrumentalise, or an intervening one, is a subject of academic dispute (Werkner and Hidalgo 2016). In any case, the connection with political contexts is evident.

This is impressively demonstrated in the volume Transformations of Religiosity by Gert Pickel and Kornelia Sammet (2012). The religious revival in Romania after 1989 was initially linked intricately to an economic crisis and social instability, and in the following phase to a growing association with national feelings. In Croatia, the decline of religiosity was combined with the growing importance of religion in the political sphere. The respective role of the church in the political sphere makes it possible to understand why church membership was on the rise in Russia, Romania, and Bulgaria after the fall of communism while it was declining in secularised East Germany and Czechia. In the secularised Western European societies, in turn, religiosity can and does promote social capital. A strengthened significance of religion in politics is also possible, as studies on religion and politics in Germany document (Pickel and Hidalgo 2013; Pickel and Liedhegener 2016): the institutional loss of importance of the churches does not exclude their relevance in civil society, as can be seen in their contribution to the controversial debates on religious freedom, biopolitics, euthanasia and protection of life, or the integration of Islam in Germany.

For our focus on attitudes towards immigrants and Muslims, the concentration on migration and Islam in political discourses on religion is of particular interest, as it affects the connection between religiosity and political attitudes. Regarding this hypothesis, Ivanescu (2010) speaks of ‘religionisation of politics’ and the ‘politication of religion’, for immigrants and Muslims represent not only the socially ‘other’ but also confront the secular nation-state with the challenge of cultural and religious plurality and thus with its self-image. This self-image is built on a national language and culture as well as on national institutions. Migration and Islam discourses are thus an expression of the ‘struggle for individual and collective national identity, be it in terms of secularism, democracy or citizenship’ (Ivanescu 2010: 309). Religion becomes a political issue.

This two-way process can also have an impact on the values of the autochthonous population. The religiosity of immigrants and Muslims appearing in the public sphere can challenge the values that have been agreed upon in a secular society, regarding, for example, attitudes towards family, gender, democratic and legal participation, or socially accepted religiosity. Conversely, religious immigrants and Muslims claim political interests based on their religiosity – particularly when they become increasingly self-confident in their struggle to have their identities and values recognised. Conflicts and power struggles are the result of and are fought out based on the distinctive feature ‘religion’ – sometimes even when it comes to ‘secular’ questions, such as legal, socio-economic, or political participation.

In the secularised societies of Western Europe, the postulate of a strict separation between politics and religion is thus called into question, as it proves to be a relative construct. This leads to power struggles for interpretative sovereignty and to European and Christian cultures and values being brought into opposition with Islamic cultures and values (Ivanescu 2010). In the countries of Eastern Europe, which are still in the process of nation-building, migration in turn calls the concept of the nation into question; here, religiosity is brought into play as a national identity marker (Pickel and Sammet 2012).

The so-called integration policy debates in Austria, Germany, and Switzerland provide another example of the religionisation of politics or the politicisation of religion. Astrid Mattes (2016) proves how Islam did not become a focal point of academic and political discourse in the aftermath of 9/11 or terrorist attacks but was problematised throughout this period mainly by the Christian Democratic parties to reformulate a universal collective identity oriented towards the Christian heritage, in an attempt to disguise a lack of political vision.

Finally, questions of values are also negotiated when the discourse on Islam in politics and academia is conducted against the backdrop of a normative concept of modernity and in the name of emancipation and progress (Opratko 2019). In these discourses, religiosity is viewed as unenlightened and irrational across the board. In the long run, this assessment applies not only to Islam but consequently to religion itself. This kind of liberal-secular matrix (Amir-Moazami 2018) is contrasted on the other side by a transnational, transatlantic fundamentalist Christian movement that fights the values of liberal democracies with reference to an alleged threat of ‘Islamisation’ (Wäckerlig 2019).

Value conflicts connected with religion are also evident in the European controversies over migration, asylum, and refugee policies. In their associated crisis narratives, the political opponents base their policies on ‘European values’ (Weymans, Chap. 3, this volume). While international organisations and civil society actors refer to the values on which human rights are based, right-wing parties refer to the protection of Europe’s Christian identity and values (Goździak et al. 2020).

All the above-mentioned developments are likely to be reflected in attitudes towards immigrants and Muslims. Since religion is an essential reference point in political and value discourses, the question arises as to whether religiosity affects these attitudes and, if so, how. If, as we assume, religiosity influences the rejection or recognition of immigrants and Muslims only in connection with sociodemographic factors, religiously formatted value conflicts must be interpreted multi-perspectively in the contemporary political context in which the struggle for recognition of identity, culture, and values is also inseparably intertwined with the struggle for social and socio-economic participation and equitable distribution (Fraser and Honneth 2003).

3 Effect of Religiosity on Attitudes Towards Immigrants and Muslims

In the following section, we will document the heterogeneous effect of religiosity on attitudes towards immigrants and Muslims in four steps.¹

3.1 The Effect of Religious Self-Assessment and Sociodemographic Variables on the Rejection of Immigrants and Muslims

First, we assume that the effect of religiosity relates to the sociodemographic variables of gender, age, income, and size of town. We do this by looking at the responses to the following EVS 2017 items: ‘I would not like to have immigrants²/Muslims as neighbours.’ In a first overview, the following results emerge:

• (a) Gender and religious self-assessment

A simple mean value analysis of the EVS 2017 data shows no significant differences between men and women for all countries surveyed on the question of whether someone rejects having immigrants and Muslims as neighbours. Across Europe, between 22% and 24% agree with this statement. When we extend the subdivision by gender, including self-assessment as a ‘religious person’, ‘not a religious person’, or ‘a convinced atheist’, there are still no significant differences within the respective groups of religious self-assessments across Europe in the rejection of having immigrants and Muslims as neighbours.

However, apparent differences within the genders can be seen between the respective groups of religious self-assessment (Table 6.2). Religious persons significantly more often reject having immigrants and Muslims as neighbours. While 23% of the interviewed women with a religious self-assessment agree with this statement concerning immigrants, only 15% of the female convinced atheists agree with it (Muslims: 23% and 16%). This effect of religiosity is also evident among men: 23% of religious men would not want to live next to immigrants, while only 17% of atheist men would not want to (Muslims: 24% and 19%).

Table 6.2 ‘I would not like to have immigrants/Muslims as neighbours’ (v24/v28), relationship between sex and religious self-assessment; also differences between age groups (6 intervals), in per cent (EVS 2017)

• (b) Age and religious self-assessment

Mean value analysis for all participating countries shows that age affects whether someone likes to have immigrants or Muslims as neighbours (Fig. 6.1).

Fig. 6.1

‘I would not like to have immigrants/Muslims as neighbours’ (v24/v28), differences between age groups (3 intervals), in per cent (EVS 2017)

A detailed evaluation according to age cohorts (Table 6.2) shows that the age group of 65+ years showing a significantly higher rejection than the other age groups: More than a quarter of respondents disapprove of having Muslims as neighbours. Among immigrants, who are rejected as neighbours by 25%, the difference between the 65+ and the other age groups is the same.

When we take the religious self-assessment of the age cohorts into account (Fig. 6.2), it becomes apparent that the rejection of Muslims increases with age (likewise with immigrants, which we do not depict graphically here). Once again, the group of 65+ shows the highest values for rejection. In all age cohorts, religious people show the highest rejection values compared with non-religious people or convinced atheists. It is also noticeable that differences regarding religious self-assessment are clearly less pronounced in the middle-aged cohorts than in the youngest and oldest groups.

Fig. 6.2

‘I would not like to have Muslims as neighbours’ (v28), relationship between religious self-assessment and age group (6 intervals), in per cent (EVS 2017)

• (c) Income and religious self-assessment

Figure 6.3 illustrates, that income³ also plays a central role in relation to the rejection of immigrants and Muslims as neighbours.⁴ Again, a religious self-assessment has a positive effect; however, it decreases with increasing net household income.

Fig. 6.3

‘I would not like to have immigrants/Muslims as neighbours’ (v24/v28), relationship between religious self-assessment and net household income, in per cent (EVS 2017)

Concerning the rejection of having immigrants and Muslims as neighbours, gender again plays virtually no role in relation to income (Fig. 6.4), with a tendency towards middle-income people showing a small gender-related difference in rejecting Muslims as neighbors.

Fig. 6.4

‘I would not like to have immigrants/Muslims as neighbours’ (v24/v28), relationship between sex and net household income, in per cent (EVS 2017)

The combination of age and net household income again shows the effect of age (Fig. 6.5): it is the people over 45 with a low net household income who are most against having Muslims⁵ as neighbours. Among the younger cohorts, income again plays a lesser role.

Fig. 6.5

‘I would not like to have Muslims as neighbours’ (v28), relationship between age (6 intervals) and net household income, in per cent (EVS 2017)

If we now look at the relationship between net household income and the size of town (Fig. 6.6), it can be seen that the latter has a clear effect on whether or not someone likes to have Muslims⁶ as neighbours. The rejection is more pronounced in all income groups in rural areas than it is in urban areas. Simultaneously, it increases in all groups in large cities, where low and middle groups in particular almost equalise.

Fig. 6.6

‘I would not like to have Muslims as neighbours’ (v28), relationship between net household income and size of town, in per cent (EVS 2017)

• (d) Size of town and religious self-assessment

In the next step, we take a closer look at the effect of the size of town on the rejection of having Muslims as neighbours (Fig. 6.7). A differentiated picture emerges:

Fig. 6.7

‘I would not like to have Muslims as neighbours’ (v28), relationship between size of town and age group (6 intervals), in per cent (EVS 2017)

Rejection is highest in small towns with up to 5,000 inhabitants, then decreases as the size of town increases up to 500,000 inhabitants (except for the 65+ group), and then increases again from a size of town of over 500,000 inhabitants (except the age group between 25 and 34 years and again for the 65+ group). Thus, although people in urban areas are generally less hostile towards Muslims than are people from small rural towns, the urban area as a social conflict zone seems to force the rejection compared with smaller medium-sized residential areas, where people usually learn to know each other more easily and managing diversity thus might be easier.

Concerning immigrants, the picture is slightly different (Fig. 6.8). In the metropolitan area (over 500,000 inhabitants), unlike the rejection of Muslims, no increasing rejection of having immigrants as neighbours is perceptible – except for a slight increase among 45- to 54-year-olds. Instead, it can be observed that rejection is greatest in rural areas, but thereafter the numbers remain relatively constant in each age group.

Fig. 6.8

‘I would not like to have immigrants as neighbours’ (v24), relationship between size of town and age group (6 intervals), in per cent (EVS 2017)

Combining the size of town with the religious self-assessment (Fig. 6.9), a comparison across European countries shows that religious people are more hostile towards Muslims than are non-religious people and people who describe themselves as convinced atheists, regardless of the size of town.

Fig. 6.9

‘I would not like to have Muslims as neighbours’ (v28), relationship between size of town and religious self-assessment, in per cent (EVS 2017)

This difference is particularly visible in large cities with more than 100,000 inhabitants. Nevertheless, the rejection of having Muslims as neighbours is highest in village settings.

The results are different again when it comes to the question of not liking to have immigrants as neighbours (Fig. 6.10). Convinced atheists have the fewest reservations, but in villages up to 5,000 and cities with more than 500,000 inhabitants there are no relevant differences between religious and non-religious people. The effect of religiosity obviously plays a minor role or no role here, and the rejection of immigrants as neighbours is probably fed by other factors, such as for instance negative experiences, xenophobia, or political attitudes.

Fig. 6.10

‘I would not like to have immigrants as neighbours’ (v24), relationship between size of town and religious self-assessment, in per cent (EVS 2017)

• (e) Summary

The results of our first, very general overview support our claim that religiosity does have a significant effect on the rejection of having immigrants and Muslims as neighbours. However, it unfolds its effect in a highly heterogeneous manner. Social factors play a decisive role alongside religious motivation. Gender is not a distinguishing feature. It only leads to increased rejection of immigrants and Muslims in combination with a religious self-assessment. Age has a strong effect. While in the older groups the rejection of immigrants and migrants may be related to the difficulties of getting used to living in religiously diverse societies, the results from the younger generation could reflect experiences of conflict in everyday life but also the difficulties of young people – less religious than older people – in asserting their own religious identity in increasingly secular societies, which can lead to the use of religiosity as an identity marker. Furthermore, people of higher age and lower income show higher scores in their rejection of having immigrants and Muslims as neighbours. Living in a rural area also fosters negative attitudes. Here, a difference becomes apparent regarding attitudes towards immigrants and Muslims. Religiosity shows an effect on the rejection of Muslims as neighbours regardless of the size of the town, whereas immigrants seem to be rejected less and probably for motivations other than religious ones. These results suggest that the effect of religiosity may depend on the social structures of coexistence: the socially and religiously more homogeneous rural area is associated with higher rejection. In contrast, social mixing and thus coexistence is more natural in urban areas, but religiosity becomes more of a distinguishing feature.

Overall, these heterogeneous correlations show a complex interaction between religious and social motivation and point to further factors that come into play. Whether a religious self-assessment is the primary cause, how a religious self-assessment connects with social factors forcing rejection, or, conversely, whether religiosity takes on an ideological function or other influencing factors come into play, must remain open here.

3.2 The Effect of Political Self-Positioning on the Rejection of Immigrants and Muslims

One exemplary influencing factor is examined in the next step: the effect of political self-positioning on attitudes towards immigrants and Muslims. For this purpose, we compare Czechia, Hungary, and Austria – three countries in which migration policy with a strong anti-Islamic discourse has played a key role in political discourses in the aftermath of the refugee crisis in 2015. This has been compounded by political disputes over the admission of mainly Muslim civil war refugees from Iraq and Syria. At the same time, these countries differ in terms of their religious composition and regarding their share of immigrant and Muslim population: Czechia dominantly atheist, low immigrant and Muslim share; Hungary dominantly Catholic but with greater Christian diversity, low immigrant and Muslim share; Austria dominantly Catholic, high immigrant and Muslim share.

In a first step, we look at the development of political self-positioning in these countries between 2008 and 2017 (Figs. 6.11, 6.12, and 6.13):

Fig. 6.11

Political self-positioning in Czechia 2008 and 2017 between 1 (left) and 10 (right), in per cent (EVS 2008 and 2017)

Fig. 6.12

Political self-positioning in Hungary 2008 and 2017 between 1 (left) and 10 (right), in per cent (EVS 2008 and 2017)

Fig. 6.13

Political self-positioning in Austria 2008 and 2017 between 1 (left) and 10 (right), in per cent (EVS 2008 and 2017)

In Hungary, a clear shift to the right can be observed in the political self-assessment of respondents during this period. In Czechia, an apparent strengthening of the political centre is noticable in the comparison period. For Austria, an increase in the area of ‘centre-right’ (6–8) is to be noted, but at the same time a strong decrease in the positioning as ‘right’.

If we now look at the developments of religious self-assessment in the respective countries (Pickel and Pickel, Chap. 5, this volume: 194) and the rejection of having immigrants and Muslims as neighbours (Fig. 6.14), the latter increases significantly in Hungary, while there is no significant increase in religious self-assessment. In Austria, too, self-assessment as a religious person has changed little in the comparison period, but agreement with the statement that one does not like to have immigrants and Muslims as neighbours has fallen.

Fig. 6.14

‘I would not like to have immigrants/Muslims as neighbours’ (v24/v28), in Hungary, Czechia, and Austria, in per cent (EVS 2008 and 2017)

Comparing these trends, political developments and the associated discourses seem to have a stronger impact on the rejection of immigrants and Muslims than does religious self-assessment, which remains mostly stable. This can be illustrated by the example of Austria: while in 1990 only 14% and in 1999 only 15% agreed with the statement that they did not like to have Muslims as neighbours, the figure rose massively to 31% in 2008 but fell again to 20% in 2017. In Hungary and Czechia, agreement to these statements increased immensely throughout this period. In all three countries, a massive politicisation of the question of accepting refugees took place in the comparison period, which was characterised by anti-migrant and anti-Muslim views. But in Austria, the governing conservative Austrian People’s Party (ÖVP) adopted anti-migrant and anti-Muslim discourses from the right-wing populist Freedom Party of Austria (FPÖ) and established a restrictive migration and refugee policy. A high acceptance level of this policy by most of the population can probably explain the decrease, as the policy made people feel protected from unlimited immigration. At the same time, Austrian people are experienced in living in a society shaped by migration and diversity, especially in cities. In contrast, a lack of such experiences in Hungary and Czechia may strengthen the impact of an anti-migration and anti-Muslim political narrative that portrays the danger of an Islamic invasion and Muslims as religious fanatics.

However, the effect of religiosity is negligible. Political factors that change in the short term probably have more impact on attitudes towards immigrants and Muslims than do religious attitudes, which change more slowly.

These findings refute Stefanie Doebler’s (2015) thesis that religiosity is the decisive factor for strengthening the rejection of immigrants and Muslims. On the one hand, the rejection of having immigrants as neighbours is increasing in two of the selected countries despite different levels of religiosity. On the other hand, the different political self-assessments alone do not explain the growing rejection. A more likely assumption is that, above all, the impact of media and political discourses must be taken into account.

Another country comparison reinforces the assumption that religiosity cannot be the decisive factor in the rejection of having Muslims as neighbours. With its strongly secularised population, Czechia records a massive increase. But the same tendency can be seen in pluralistic Christian Hungary and the Catholic-dominated countries of Poland and Slovakia. In contrast, secular France remains largely stable, while in Germany, Austria, and Great Britain – three secularised countries, but with a relatively high religious diversity – rejection is falling. These country-specific differences give the impression that concrete experiences of living together in cultural and religious diversity in secular contexts may play a central role in addition to political discourses. Experiences of living in diverse religious neighbourhoods – such as in Great Britain – can probably mitigate the effect of anti-migrant and anti-Muslim political discourses.

3.3 A Socioreligious Typology (Cluster Analysis)

In the next step, we will use cluster analysis to present a socioreligious typology. After demonstrating that sociodemographic variables, in addition to a religious self-assessment, have an effect on the rejection of having immigrants and Muslims as neighbours across all European countries, we include variables measuring both religiosity and sociodemographics. With this cluster analysis, we identify groups that differ significantly regarding these influencing factors. We will then ask again whether these groups differ in their attitudes towards immigrants and Muslims and, if so, how.

• (a) Clustering and distributions in detail

For this purpose, we compute a cluster analysis using Ward’s linkage method and Euclidian distance as the heterogeneity measure, which aims to identify groups that are as homogeneous as possible. We first include religiosity in the cluster formation. In order to get a more differentiated picture, we supplement the variable on religious self-assessment with further variables that the EVS makes available for measuring religiosity: belief in God, the importance of God in one’s own life, attendance at church services, praying outside of church services, and belonging to a religious denomination. This expansion should give us differentiated distinctions regarding the effect of religiosity, especially in relation to the intensity of religiosity – for example, whether a religious self-assessment is embedded in practical and institutional contexts or stands rather abstractly on its own. The distribution among the five clusters (Table 6.3) reveals different degrees of intensity of religiosity, with the intensity of religious self-assessment once again differing clearly from the religious practice associated with it.

Table 6.3 Distribution within the clusters formed in relation to questions about religion and religious practice, in per cent (EVS 2017)

Table 6.4 presents the distribution within the clusters in terms of religious denomination. It shows clear denominational differences. Thus, persons of a Catholic denomination are distributed much more evenly among the groups than are persons of a Protestant denomination, who are found more in the less religious groups. Muslims, on the other hand, are relatively evenly distributed in all clusters.

Table 6.4 Distribution within the clusters formed in relation to the religious denomination (only people belonging to a denomination), in per cent (EVS 2017)

Furthermore, we include the variables of gender, age (subdivided into three age cohorts of under 30, 30–49 and 50+ years) and size of town in the cluster formation (Table 6.5).⁷ Clusters with higher religiosity tend to be female and dominated by older persons. Regarding size of town, people within more religious clusters live primarily in villages or small towns, whereas people in less religious clusters live in bigger cities and metropolises.

Table 6.5 Distribution within the clusters formed in relation to sex, age groups (3 intervals), and size of town, in per cent (EVS 2017)

Since we assume that country-specific constellations also have an impact on the effect of religiosity, we now examine the distribution of the clusters within ten selected countries. For this purpose, we selected two each with a similar character in religious terms: two majority Catholic countries (Austria and Italy), two majority Protestant countries (Finland and Sweden), two majority Orthodox countries (Romania and Bulgaria), two majority mixed countries (Germany and Hungary), and two majority secularised countries (France and Czechia).

Figure 6.15 shows how the five clusters are divided within each country:

Fig. 6.15

Cluster distribution within countries, in per cent (EVS 2017)

The differences between the countries are visible. There are countries dominated by C1 (high religiosity, older, rural), such as Romania and Italy, and countries with a majority of C5 (low religiousity, younger, urban), such as Czechia and Sweden. In Romania and Sweden, one cluster each is dominant (C1 and C5), whereas in Austria and Hungary, the clusters are relatively equal in size. Accordingly, heterogeneous influences of religiosity are to be expected.

• (b) Description of clusters

In the following summary, we describe the five socioreligious types found in the ten selected European countries.

Cluster 1: The Small-Town Active Highly Religious (C1) – 27%

The members of cluster 1 are ‘highly religious’: almost all persons in this cluster (96%) describe themselves as a ‘religious person’; likewise, the question about belief in God is affirmed almost without exception (99%). With a mean value of 9.13⁸ concerning the importance of God in one’s own life, God also occupies an important position. Those from this cluster who attend a church service at least once a week represent 41%, while 94% pray at least once a week in addition to prayers during the church service. Likewise, almost all cluster members belong to a religious denomination (95%), distributed mainly among the Roman Catholic Church (46%) and Orthodox churches (30%) (Protestant churches 14%, Muslim community 5%). At 66%, women dominate this cluster. In addition, a high average age of 54.5 years can be observed, while those under 30 years of age, at only 12%, represent a clear minority compared with 62% of those over 50 years of age. This socioreligious type is mainly found in Italy (20% in cluster 1) and Romania (24% in cluster 1) and dwells in settlements with fewer than 5,000 inhabitants (20%), in small towns between 5,000 and 20,000 inhabitants (24%), and in smaller cities of between 20,000 and 100,000 inhabitants (21%).

Cluster 2: The Rural Part-Time Active Highly Religious (C2) – 16%

The members of cluster 2 are also characterised by a high belief in God (93%), and self-assessment as a religious person (85%) is also observed. However, the importance of God in one’s own life (M = 7.69) is lower than among the members of cluster 1. Clear differences between cluster 2 and cluster 1 also become visible in religious practice: only 16% attend a church service at least once a week and ‘only’ 49% pray at least once a week outside the church service. A quarter (26%) even states that they never attend religious services. The distribution regarding denominational affiliation is similar to that of cluster 1: Roman Catholic 55%, Protestant 15%, and Muslim 6%. Those belonging to a religious organisation represent 82%, while 18% do not belong to any religious organisation.

The average age of 52.7 years can also be classified as high in this cluster, where again the over 50-year-olds old make up the absolute majority of the cluster members, at 56%. Under 30-year-olds represent a clear minority, at 12%. The female share also predominates in cluster 2 (56%). Another special characteristic of this cluster is that almost half of all cluster members (46%) live in settlements with fewer than 5,000 inhabitants. In terms of country distribution, the members of this cluster are found in Italy (18%), Austria (13%), Romania (13%), Germany (12%), and Bulgaria (10%).

Cluster 3: The Urban Church Members (C3) – 13%

Cluster 3 shows a significant decrease in religiosity among its members, though 70% still belong to a religious denomination and 86% believe in God. However, God’s importance for their own lives (M = 5.81) plays a much smaller role. Only 65% see themselves as a religious person. Religious practice is low in this cluster. Only 3% attend church services at least once a week, 31% do not attend at all, 14% pray at least once a week, and 25% do not pray at all. Those cluster members who are members of a religious organisation are mainly found in the Roman Catholic Church (51%) and in a Protestant church (30%). In terms of age structure, cluster 3 differs significantly from clusters 1 and 2, with an average age of 51.3 years and a share of 16% of those under 30 years of age. In terms of size of town, the members of cluster 3 predominantly live in urban or metropolitan areas (20,000–100,000: 25%; 100,000–500,000: 22%; over 500,000 inhabitants: 26%). The ratio of female (51%) to male (49%) is balanced. The majority of cluster 3 members live in France (14%), Germany (13%), Italy (13%), Austria (11%), Hungary (11%), and Finland (10%).

Cluster 4: The Rural Passive Part-Time Believers (C4) – 19%

People belonging to cluster 4 can be characterised as passive part-time believers. The cluster shows a slight ‘female surplus’ (53%). About half (49%) describe themselves as religious and 16% describe themselves as atheist. Worship is either attended only on certain holidays (45%) or not at all (48%); 43% do not pray at all. The belief in God is still predominant among the majority (55%), but the importance of God in one’s own life can be classified as less important, with a mean of 4.1. The denominational affiliation is mainly distributed among the Roman Catholic Church (49%), Protestant (27%) and Orthodox churches (19%), and Muslim (3%). Only slightly more than half (56%) of the persons belong to a religious organisation.

A characteristic feature of the village active part-time believers is the relatively high proportion of people under 30 (19%). Nevertheless, people over 50 years of age make up the majority in this cluster (54% – average age 50.1 years). Slightly more than two-thirds of the persons in cluster 4 live in villages with a maximum of 20,000 inhabitants (under 5,000: 40%; 5,000–20,000: 28%). This group of people is found mainly in France (16%), Germany (15%), Czechia (12%), and Austria (11%).

Cluster 5: The Urban Seculars (C5) – 25%

The majority of urban seculars are male (56%) and do not belong to any religious denomination (71%). The number of those who identify themselves as religious (9%) is marginal compared with those who identify themselves as non-religious (64%) and as atheists (27%) in this cluster. Only 16% believe in God, but this faith has virtually no significance (M = 1.59) for their own lives. Nevertheless, religious services are attended – mainly on certain holidays, however (69%). Only 0.3% say they never attend religious services, while 86% never pray. Of the 29% who belong to a religious denomination, the majority are found in a Protestant church (57%) (Roman Catholic 30%; Orthodox 10%; Muslim 1%). Cluster 5 has the lowest average age (48.6 years) of all the clusters, and for the first time, the 50+ generation is no longer the absolute majority in this cluster (under 30: 16%; 30–49: 37%; 50+: 47%).

Members of this cluster are rare in the village setting (8%). They are distributed relatively evenly among cities of 20,000 inhabitants or more (20,000–100,000: 29%; 100,000–500,000: 25%; 500,000 and more: 26%) and are found mainly in Sweden (16%), Germany (16%), France (15%), and Czechia (15%).

The composition of the clusters clearly documents the degree to which sociodemographic variables affect religiosity. Thus, the highly religious are mainly found among older women in rural regions, while the decline in religiosity is found mainly in the middle and especially younger generations and in urban areas, as well as in Western or already long-secularised countries with a high proportion of urban areas.

3.4 Attitudes Towards Immigrants and Muslims Based on the Socioreligious Typology

In the next step, we ask which attitudes towards immigrants and Muslims can be found among the five socioreligious types and we evaluate them with regard to age, gender, and size of town.

• (a) Attitudes towards immigrants in relation to age, gender, and size of town

To this end, we first evaluate the EVS’s exemplary questions assigned to the topic area of immigrants for all five clusters by means of a mean value calculation.

On the question of whether immigrants are taking jobs away from local people, there are significant differences between the clusters regarding the respective age cohorts (Fig. 6.16). While the youngest cluster of urban seculars (C5) generally agrees least with this statement, the 50+ generation clusters show significantly higher agreement. In C2 in particular, there is a significantly higher agreement with this statement across all age groups. In C4, the middle-aged cohort shows higher agreement. In the competition for jobs, age and spatial-structural contexts are probably more important than religiosity.

Fig. 6.16

‘Immigrants take jobs away from local people’ (v185), relationship between age groups (3 intervals) and cluster (EVS 2017)

The response to the statement that ‘Immigrants make crime problems worse’ shows a different picture (Fig. 6.17). In four out of five clusters, the agreement increases with age. The influence of age is clearest in cluster 1. At the same time, all five clusters are close to each other. Obviously, people distinguish between labour market issues and crime problems.

Fig. 6.17

‘Immigrants make crime problems worse’ (v186), relationship between age groups (3 intervals) and cluster (EVS 2017)

As can be seen from Fig. 6.18, there are no significant variations within a cluster concerning gender. However, clearly noticeable differences can be observed between the genders among the clusters (due to age).

Fig. 6.18

‘Immigrants take jobs away from local people’ (v185), relationship between cluster and sex (EVS 2017)

The respective gender differences are also never more than a maximum of 0.2 points for all other Q52 statements. Gender does not lead to any notable differences in attitudes towards immigrants.

As expected, the size of town has an enormous effect on attitudes towards immigrants (Fig. 6.19 and Table 6.6). The rejection decreases with an increase in the size of town. The larger the size of town, the less immigrants are seen as a threat in the labour market or as a strain on a country’s welfare system. In small and medium-sized towns, rejection increases within the clusters.

Fig. 6.19

‘Immigrants take jobs away from local people’ (v185), relationship between cluster and size of town (EVS 2017)

Table 6.6 ‘Immigrants are a strain on a country’s welfare system’ (v187), relationship between cluster and size of town (average: 1 = fully agree; 10 = fully disagree; EVS 2017)

The comparison of the rejection level shows that the respondents distinguish between different issues. Job competition is perceived as a much lower area of conflict than are crime and social welfare. This is remarkable, as concern about one’s job can affect personal life more than the other two factors, which are more abstract issues on which perception depends much more on ideological interpretations and discourses.

• (b) Attitudes towards immigrants, religiosity, and country

In the previous evaluations, there are no relevant differences between the highly religious and the less religious clusters. Religiosity obviously plays a minor role in a more complex approach compared with sociodemographic factors.

But if we now look at the rejection of immigrants within the selected countries, clear but highly contradictory differences become apparent. We present the highly religious cluster 1 (Fig. 6.20), using the example of agreement with the statement ‘immigrants make crime problems worse’.

Fig. 6.20

Cluster 1: ‘immigrants make crime problems worse’ (v186) (EVS 2017)

The two countries influenced by Catholicism show similar and, by international comparisons, relatively high approval ratings. However, the highest rejection of immigrants can be found in Bulgaria, Hungary, Czechia, and Austria – countries that are strongly characterised by problem-oriented migration discourses but of a quite different religious character.

Conversely, those two countries where secularisation is most advanced show extremely different means: in Czechia, the highly religious respondents strongly agree with this statement (M = 2.85), whereas in France, this statement is rejected (M = 6.35). In France, religious identity has less public importance compared with egalitarianism and ideological laicism, which are highly valued. By contrast, in Czechia, religious identity issues are far more relevant in a minority situation, and secularisation has other historical origins and functions than it has in France. Rich and secular Protestant Sweden, on the other hand, is similar in its approval rate to poor, highly religious Orthodox Romania.

These perplexing and contradictory results demonstrate that classical theories of the sociology of religion are insufficient to exactly identify the influence of religiosity. The influence of political discourses, the level of wealth and demographic composition, the historical and political culture in dealing with immigrants, the influence of religious communities on politics, and many more factors must be explored. Religiosity and sociodemographic variables do affect attitudes towards immigrants, but the country-specific situation has a strong and somehow independent effect.

• (c) Desired social distance to minority groups

Finally, we look at the clusters with regard to the question of the rejection of having minorities as neighbours (Fig. 6.21). In addition to immigrants and Muslims, we also include people of a different race, homosexuals, and Jews, as religious motives could also be evident here. The picture remains complex and contradictory.

Fig. 6.21

Clusters 1–5: ‘I would not like to have as neighbours’ (v22, v24, v26, v28, v29), in per cent (EVS 2017)

The high rejection of immigrants as neighbours is evident in all clusters. Although it is lower in the urban clusters (C3 and C5), it is the leading rank in comparison with other groups. The two highly religious clusters (C1 and C2) especially reject homosexuals as neighbours. The rejection of Jews as neighbours is high, especially among the two older and highly religious clusters. The comparatively low rejection of people of a different race may be due to the politically desired rejection of racism or heterogenous understandings of ‘race’ in Europe. This group also plays a minor role in political discourse. The rejection of Muslims runs through all clusters. Only the urban church members (C3) show a lower reservation here.

Using the example of the rejection of having Muslims as neighbours, the effect of sociodemographic variables within the clusters is now examined. An analysis according to age, size of town, and gender once again shows effects (Table 6.7).

Table 6.7 ‘I would not like to have Muslims as neighbours’ (v28), relationship between cluster and age groups (3 intervals), size of town, and country, in per cent (EVS 2017)

It is mainly the 50+ generation within the small-town active highly religious who would not like to have Muslims as neighbours. A similar picture emerges for the 30- to 49-year-olds compared with the under 30-year-olds in cluster 5 of the urban seculars. The high rejection of Muslims in cluster 1 might also be connected with the high age of cluster 1.

Differences of up to 15% in the rejection of Muslims as neighbours can be seen across clusters between villages with fewer than 5,000 inhabitants and large cities with over 500,000 inhabitants. The size of town seems to be more decisive than is religiosity.

Regarding gender there are no significant differences.

The clearest differences again occur in the country comparison and reinforce the doubt about a direct influence of religiosity on the rejection of having Muslims as neighbours.

In Austria, rejection is highest among the village passive part-time believers (C4), followed by the highly religious cluster 1 which is level with the low-religious cluster 5. In Bulgaria and France, rejection is lowest among the highly religious clusters compared with the other clusters. In a country comparison, rejection is generally highest in Czechia. With C5, the urban, non-religious cluster shows the highest rejection of having Muslims as neighbours, while the highly religious cluster 1 agrees least with this statement. A comparison of countries with similar religious backgrounds reveals the same contradictory situation: Bulgaria has a significantly lower approval rate than Romania, despite the strong Christian Orthodox character in both countries. In contrast to Romania, however, Bulgaria has a large Muslim community that has existed in the country for centuries, probably explaining the higher acceptance level. Other examples are the two secular countries of France and Czechia. While among the small-town active highly religious in France only about 5% disapprove of having Muslims as neighbours, about 45% agree with this statement in Czechia.

4 Summary

Our analysis attempted to contribute to a more detailed view of the empirically verifiable effects of religiosity on political attitudes. The results raise awareness of the close connection between this effect and sociodemographic variables and country-specific contexts. The more detail with which one investigates these relationships, the more difficult it becomes to interpret them solely with theories of the sociology of religion. As we argued that attitudes towards immigrants and Muslims are key issues for the acceptance of liberal democratic values, we focused on the effect of religiosity on them. We demonstrated that religiosity is not an independent variable but has complex connections to age, income, size of town, and in some cases to gender too across all European countries. Additionally, the respective country context influences the effect of religiosity, but in a heterogeneous and contradictory way. These results raise more questions than answers.

They can be better understood using the theoretical models presented in the first part of the chapter. Religiosity is always embedded in social, cultural, political, economic, and historical contexts. It unfolds its specific effects only in connection with other factors. It can change functions and meaning. The increasing secularisation across Europe verified in this volume demonstrates processes of disembedding of religiosity and the effects of sociodemographic factors. Religiosity is increasingly losing influence in daily life, and it seems that other factors are becoming more decisive in relation to attitudes towards immigrants and Muslims, such as country-specific political interests and discourses, religionisation of politics, politicisation of religion, and historical contexts. Therefore, when it comes to the rejection of immigrants – because they take jobs away, increase problems with crime, or are a strain on a country’s welfare system – religiosity seems to play a marginal role. Political framings of Islam and Muslims and migration policies seem to have a stronger impact on attitudes towards Muslims than does religiosity, as is the case in Austria, Hungary, and Czechia. Historical traditions of religious coexistence, such as those in Bulgaria, or the reality of diverse religious societies or urban areas, such as in Germany, Austria, and Great Britain, seem to decrease rejection of immigrants and Muslims. In secularised France, with a long tradition of privatising religion and an ideological secularisation concept, the effect of religiosity on the rejection of immigrants and Muslims is significantly lower than in countries that are still Catholic dominated such as Austria and Italy.

These rather diverse effects of religiosity render questionable the premise of the individualisation theory according to which religiosity is an anthropological constant. Rather, it must be assumed that all measured forms of religiosity may disappear in some countries and thus lose their influence on values and attitudes. This can be seen in the example of the rejection of homosexuals as neighbours, which is deeply connected with religious attitudes and, therefore, much less pronounced among low-religious urban seculars and Northern European countries. However, a secularised world view does not immunise people against rejecting Muslims, as the cluster analysis for Czechia impressively shows. There, obviously anti-Muslim discourses also fall on fertile ground among non-religious people.

In general, universally oriented theories such as individualisation and secularisation no longer seem to fully explain the dynamics of the development of religiosity in Europe, as they do not take the contemporary macro-transformations of social and political impacts on a decreasing relevance of religion sufficiently into account. A theoretical pluralisation approach, which focuses on the interplay between the individual and politics, seems more appropriate, but subsequently requires interdisciplinary and qualitative case studies.

If, for example, we consider theories on the religionisation of politics or the politicisation of religion, it is hard to prove their effect on religiosities with the EVS data. However, such an effect cannot be directly confirmed at the level of individual religiosities in Europe. Political attitudes such as the rejection of immigrants and Muslims are connected to religiosity, but dependent on sociodemographic variables as proven across all European countries. Simultaneously, the effect differs across countries because of country-specific contexts such as the role of religion in a state or political discourse: sometimes there seems to be nearly no effect, such as in France; sometimes religiosity can presumably be activated by political discourses, such as in Czechia or Hungary. This raises the question of the mutual influences between sociodemographic effects and country-specific circumstances on religiosity: to what extent are these interdependent?

To make further qualified statements, the effect of political attitudes (party affiliation, political preferences concerning specific topics) would have to be researched separately. Additionally, an interdisciplinary discourse analysis on the role of religion in state and political discourse would have to be added, and the effect of such discourses on values and attitudes would need to be made measurable. Considering our few results, one could ask whether short-term political and media debates have a much stronger effect on attitudes towards immigrants and Muslims than do religious attitudes. These can be instrumentalised for political interests but with different impacts, as the example of Bulgaria shows in comparison to Czechia, Romania, or Great Britain. To prove this on a country-specific basis would require political science presuppositions and interpretations of the history of religion, which we cannot provide here.

However, our contribution does reveal that religiosity at the level of individuals plays a vital role in the field of religious and migration policy conflicts, though it is by no means either the only or main factor. For example, the distinction and line of conflict between the values of an autochthonous, Christian population and an immigrant, Muslim population, which are repeatedly claimed in political discourses, cannot be substantiated, as the distribution of Muslims among the five clusters shows. Conflicts would instead be ignited based on the differences that show up along the cluster differences: between the high and the low religious, between the generations, between the rural periphery and the urban centres, and finally between groups of different income levels (see also Bréchon, Chap. 8, this volume). It is likely that religiosity in these more socially, politically, and economically structured conflicts can be used – as a trigger or through reinterpretation by either political or religious actors, and with diverse effects, both increasing or decreasing conflict. The effect of sociodemographic factors on religiosity, in any case, points to the fact that social, political, and economic conflicts can be turned into religious ones, especially among highly religious people. However, this does not happen in the same way in every European country.

Considering the strong effect of sociodemographic variables, it also seems necessary to include socio-economic issues while investigating the effect of religiosity on democratic attitudes, such as the acceptance of immigrants and Muslims. In a demographically ageing Europe, generational conflicts and conflicts between rural and urban regions and between European regions because of unequal distribution of power and resources and geo-historical differences are presumably more important than religiosity, which is losing relevance in everyday life. Religiosity, however, can become a source of interpreting these conflicts for both political and religious interests, and can result in religious identity politics. Taking the dynamics of secularisation into account, religiosity may intensify and fuel these conflicts, but it cannot (any longer) trigger them solely. The significance and effect of religiosity as such at the level of the population seem to become less important than suggested by the discourses on the importance of religion in the political arena. Nevertheless, religiosity can be reinterpreted and reactivated by these discourses (Roy 2008). Consequently, investigating the impact of religiosity and religious identity politics on democratic values should be intertwined with, for example, questions of distribution policy (Fraser and Honneth 2003) and its effect on religiosity.

The effect of religiosity on political attitudes would now have to be researched in detail using other political attitudes. Attitudes towards immigrants and Muslims are only one of many possible examples to prove our thesis that religiosity has highly heterogeneous effects on other values because of its nature as ‘embedded religion’. In our view, the more differentiated the analysis, the more questionable the thesis of a direct and immediate connection between religiosity and political attitudes becomes. The discrepancies demonstrated in our few examples make it clear that there is no one-dimensional, linear connection, but that a multi-perspective approach is necessary for interpretation. A complex mixture of socioreligious and socio-cultural conditions in history and in the present; political interests and discourses; the role of religious communities; the historical and current relationship between the state and religious communities; the culture in dealing with immigration; the level of integration of immigrants and Muslims; and the Islam-related narratives in a country, etc. must be consulted to enable one to determine the effect of religiosity in concrete terms more precisely. Therefore, quantitative analyses should be supplemented with other disciplines of research on religion.

Religiosity is not an independent factor, but only unfolds its effect in specific and concrete contexts. However, the erosion of religion at the level of populations does not mean that religion becomes irrelevant as a political factor.